ASIAN INFORMAL WORKERS: PROTECTING WORKERS IN A NEW ERA OF CAPITALISM (Routledge Studies in the Growth Economies of Asia)

Citation preview

Asian Informal Workers

. . . a very interesting book which breaks new ground in examining the interconnection between child labour and women’s home-based work . . . essential reading for anyone working on child labour, women’s work, and informal work. The policy recommendations are well-argued and innovative . . . It finds a way through the dilemma that attempts to outlaw child labour in home-based work will not work and can harm those they seek to help. a book that Routledge can be proud of. (Prof. Diane Elson, University of Essex) . . . This book is an empirically grounded and policy focussed contribution to debates about how to enhance human security in conditions of flexibility and change. (Bob Deacon, Professor of Global Social Policy, Sheffield University) This book is a wide-ranging survey of the nature and extent of home work in Asia. When industrialization increased pace in the region after decolonization, there was an expectation that the process would result in most of the labour force being employed in formal sector industrial jobs. The reality, after over a half-century of development, is that even in the fast-growing economies in Asia, formal employment has grown rather slowly and that most non-agricultural employment growth has occurred in the informal economy. At the same time, there has been a feminization of informal workers and growth in subcontracted homework. This book draws on surveys carried out in five Asian countries – two low-income (India and Pakistan) and three middle-income (Indonesia, Thailand and the Philippines) – where subcontracted production, usually by women and children working out of home, is now widespread. Home-based work is the source of income diversification for poor families, but is also the source of exploitation of vulnerable workers and child labour as firms attempt to contain costs. This book examines the social protection needs of these workers, and argues for public action to promote such work and protect such workers as a possible new labour-intensive growth strategy in developing countries. Santosh Mehrotra is a Senior Advisor to the Planning Commission, Government of India, New Delhi. His books include Development With a Human Face (edited with R. Jolly, 1997); Eliminating Human Poverty: Macro economic Policies for Equitable Growth and The Economics of Elementary Education in India. Mario Biggeri is senior lecturer in development economics at the University of Florence, Italy. His research interests include economies of transition, international aid, clusters and informal activities, rural development, and child labour and child’s capabilities, and he has published widely in these areas.

Routledge Studies in the Growth Economies of Asia

1 The Changing Capital Markets of East Asia Edited by Ky Cao 2 Financial Reform in China Edited by On Kit Tam 3 Women and Industrialization in Asia Edited by Susan Horton 4 Japan’s Trade Policy Action or reaction? Yumiko Mikanagi 5 The Japanese Election System Three analytical perspectives Junichiro Wada 6 The Economics of the Latecomers Catching-up, technology transfer and institutions in Germany, Japan and South Korea Jang-Sup Shin 7 Industrialization in Malaysia Import substitution and infant industry performance Rokiah Alavi

8 Economic Development in Twentieth Century East Asia The international context Edited by Aiko Ikeo 9 The Politics of Economic Development in Indonesia Contending perspectives Edited by Ian Chalmers and Vedi Hadiz 10 Studies in the Economic History of the Pacific Rim Edited by Sally M. Miller, A. J. H. Latham and Dennis O. Flynn 11 Workers and the State in New Order Indonesia Vedi R. Hadiz 12 The Japanese Foreign Exchange Market Beate Reszat 13 Exchange Rate Policies in Emerging Asian Countries Edited by Stefan Collignon, Jean Pisani-Ferry and Yung Chul Park 14 Chinese Firms and Technology in the Reform Era Yizheng Shi

15 Japanese Views on Economic Development Diverse paths to the market Kenichi Ohno and Izumi Ohno

23 Technology, Competitiveness and the State Malaysia’s industrial technology policies Edited by K. S. Jomo and Greg Felker

16 Technological Capabilities and Export Success in Asia Edited by Dieter Ernst, Tom Ganiatsos and Lynn Mytelka

24 Corporatism and Korean Capitalism Edited by Dennis L. McNamara

17 Trade and Investment in China The European experience Edited by Roger Strange, Jim Slater and Limin Wang

26 Capital and Labour in Japan The functions of two factor markets Toshiaki Tachibanaki and Atsuhiro Taki

18 Technology and Innovation in Japan Policy and management for the 21st century Edited by Martin Hemmert and Christian Oberländer

27 Asia Pacific Dynamism 1550–2000 Edited by A. J. H. Latham and Heita Kawakatsu

19 Trade Policy Issues in Asian Development Prema-Chandra Athukorala 20 Economic Integration in the Asia Pacific Region Ippei Yamazawa 21 Japan’s War Economy Edited by Erich Pauer 22 Industrial Technology Development in Malaysia Industry and firm studies Edited by K. S. Jomo, Greg Felker and Rajah Rasiah

25 Japanese Science Samuel Coleman

28 The Political Economy of Development and Environment in Korea Jae-Yong Chung and Richard J Kirkby 29 Japanese Economics and Economists since 1945 Edited by Aiko Ikeo 30 China’s Entry into the World Trade Organisation Edited by Peter Drysdale and Ligang Song 31 Hong Kong as an International Financial Centre Emergence and development 1945–1965 Catherine R. Schenk

32 Impediments to Trade in Services: Measurement and Policy Implication Edited by Christoper Findlay and Tony Warren

39 Law and Labour Market Regulation in East Asia Edited by Sean Cooney, Tim Lindsey, Richard Mitchell and Ying Zhu

33 The Japanese Industrial Economy Late development and cultural causation Ian Inkster

40 The Economy of the Philippines Elites, inequalities and economic restructuring Peter Krinks

34 China and the Long March to Global Trade The accession of China to the World Trade Organization Edited by Alan S. Alexandroff, Sylvia Ostry and Rafael Gomez 35 Capitalist Development and Economism in East Asia The rise of Hong Kong, Singapore, Taiwan, and South Korea Kui-Wai Li 36 Women and Work in Globalizing Asia Edited by Dong-Sook S. Gills and Nicola Piper 37 Financial Markets and Policies in East Asia Gordon de Brouwer 38 Developmentalism and Dependency in Southeast Asia The case of the automotive industry Jason P. Abbott

41 China’s Third Economic Transformation The rise of the private economy Edited by Ross Garnaut and Ligang Song 42 The Vietnamese Economy Awakening the dormant dragon Edited by Binh Tran-Nam and Chi Do Pham 43 Restructuring Korea Inc. Jang-Sup Shin and Ha-Joon Chang 44 Development and Structural Change in the Asia-Pacific Globalising miracles or end of a model? Edited by Martin Andersson and Christer Gunnarsson 45 State Collaboration and Development Strategies in China The case of the China–Singapore Suzhou Industrial Park (1992–2002) Alexius Pereira 46 Capital and Knowledge in Asia Changing power relations Edited by Heidi Dahles and Otto van den Muijzenberg

47 Southeast Asian Paper Tigers? From miracle to debacle and beyond Edited by Jomo K. S.

55 Understanding Japanese Saving Does population aging matter? Robert Dekle

48 Manufacturing Competitiveness in Asia How internationally competitive national firms and industries developed in East Asia Edited by Jomo K. S.

56 The Rise and Fall of the East Asian Growth System, 1951–2000 International competitiveness and rapid economic growth Xiaoming Huang

49 The Korean Economy at the Crossroads Edited by MoonJoong Tcha and Chung-Sok Suh

57 Service Industries and Asia-Pacific Cities New development trajectories Edited by P. W. Daniels, K. C. Ho and T. A. Hutton

50 Ethnic Business Chinese capitalism in Southeast Asia Edited by Jomo K. S. and Brian C. Folk 51 Exchange Rate Regimes in East Asia Edited by Gordon De Brouwer and Masahiro Kawai 52 Financial Governance in East Asia Policy dialogue, surveillance and cooperation Edited by Gordon De Brouwer and Yunjong Wang 53 Designing Financial Systems in East Asia and Japan Edited by Joseph P.H. Fan, Masaharu Hanazaki and Juro Teranishi 54 State Competence and Economic Growth in Japan Yoshiro Miwa

58 Unemployment in Asia Edited by John Benson and Ying Zhu 59 Risk Management and Innovation in Japan, Britain and the USA Edited by Ruth Taplin 60 Japan’s Development Aid to China The long-running foreign policy of engagement Tsukasa Takamine 61 Chinese Capitalism and the Modernist Vision Satyananda J. Gabriel 62 Japanese Telecommunications Edited by Ruth Taplin and Masako Wakui 63 East Asia, Globalization and the New Economy F. Gerard Adams

64 China as a World Factory Edited by Kevin Honglin Zhang 65 China’s State Owned Enterprise Reforms An industrial and CEO approach Juan Antonio Fernandez and Leila Fernandez-Stembridge 66 China and India A tale of two economies Dilip K. Das

67 Innovation and Business Partnering in Japan, Europe and the United States Edited by Ruth Taplin 68 Asian Informal Workers Global risks, local protection Edited by Santosh Mehrotra and Mario Biggeri

Asian Informal Workers Global risks, local protection

Edited by Santosh Mehrotra and Mario Biggeri

First published 2007 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN Simultaneously published in the USA and Canada by Routledge 270 Madison Ave, New York, NY 10016 Routledge is an imprint of the Taylor & Francis Group, an informa business

This edition published in the Taylor & Francis e-Library, 2007. “To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.” © 2007 Santosh Mehrotra and Mario Biggeri, selection and editorial matter; the contributors, their own chapters All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data A catalog record for this book has been requested ISBN 0-203-96653-8 Master e-book ISBN

ISBN10: 0–415–38275–0 (hbk) ISBN10: 0–203–96653–8 (ebk) ISBN13: 978–0–415–38275–5 (hbk) ISBN13: 978–0–203–96653–2 (ebk)

To Sushma and Pia and to Elena, Andrea and Pietro Pio

Contents

List of figures List of tables Notes on contributors Preface Acknowledgements

xvii xix xxv xxix xxxi

PART I

A cross-country analysis of industrial outwork in Asia

1

1

3

The empirical context and a theoretical framework SANTOSH MEHROTRA AND MARIO BIGGERI

1.1 Background: the informal economy and industrial outwork 6 1.2 Homework – a driver of poverty, or a trigger of human development? 13 1.3 The capabilities of homeworkers – a micro-level analysis of homeworker households 21 1.4 Final remarks 26 Notes 27 2

Research methodology SANTOSH MEHROTRA AND MARIO BIGGERI

2.1 Research plan 33 2.2 Research methodology: instruments and data collection 34 2.3 Profile of selected sectors 39 Annex 2.1: Questionnaire guidelines for homeworkers (and control group) 44 Notes 60

32

xii Contents 3

Subcontracting and homework in the value chain

62

SANTOSH MEHROTRA AND MARIO BIGGERI

3.1 The value chain 64 3.2 Forms of insecurity of homeworkers in the value chain in the five Asian countries 67 3.3 Final remarks 78 Notes 80 4

Homeworkers

82

SANTOSH MEHROTRA AND MARIO BIGGERI

4.1 The social and economic profile of the households 83 4.2 Homeworkers a cross-country analysis 92 4.3 Health and economic issues related to women homeworkers: some empirical evidence for India and Pakistan 113 4.4 Concluding remarks 118 Notes 120 5

Child labour in homeworker households

123

SANTOSH MEHROTRA AND MARIO BIGGERI

5.1 Homeworker households and child labour 124 5.2 Understanding children’s work in homeworker households 127 5.3 The determinants of children’s work status and hours worked: an empirical analysis for India, Pakistan and Indonesia 154 5.4 Concluding remarks 162 Annex 5.1: definition of child activities 164 Notes 166 PART II

The Country Studies 6

Subcontracted homework in India: a case study of three sectors RATNA M. SUDARSHAN, SHANTA VENKATARAMAN AND LAVEESH BHANDARI

6.1 6.2 6.3 6.4 6.5 6.6

The informal sector and informal employment in India 175 Research method 176 Subcontracted manufacturing 179 Income and poverty in homework 188 The homeworker 191 Children in homework 196

171 173

Contents xiii 6.7 From research to action 202 Notes 208 References 209 7

Hazardous subcontracted homework in Pakistan

210

SHAHRUKH RAFI KHAN, SABA GUL KHATTAK AND SAJID KAZMI

7.1 7.2 7.3 7.4 7.5 7.6 7.7

8

The informal sector in Pakistan 210 Homework in Pakistan 213 Research method 214 Homework sectors 218 Exploitation of homeworkers 225 The impact of homework: a quantitative analysis 229 Conclusions and policy recommendations 236 Annex 7.1: Description of homework environments in fieldwork reports 240 Annex 7.2: Case studies 241 Notes 243 Bibliography 247

Women and children homeworkers in Indonesia

250

M. OEY-GARDINER, E. SULEEMAN, I. TJANDRANINGSIH, W. HARTANTO AND H. WIJAYA

8.1 8.2 8.3 8.4 8.5

9

Formal and informal in the Indonesian labour force 251 Research method 260 Sectors 262 Homeworkers in the sectors studied 270 Conclusions and policy recommendations 279 Annex 8.1: Case studies: homeworkers’ earnings 282 Annex 8.2: Results from FGDs with children about the workplace 283 Notes 283 References 287

Subcontracted homework by women and children in the Philippines R. ROSARIO DEL ROSARIO, R. PINEDA-OFRENEO AND PATAMABA

9.1 9.2 9.3 9.4

The informal labour force and homeworkers 291 Research method 296 Homework sectors 297 Profile of homeworkers 304

290

xiv Contents 9.5 Conclusions and recommendations 310 Annex 9.1: Case studies 315 Notes 317 Bibliography 319 10 Subcontracted homework in Thailand

322

NARUMON ARUNOTAI, NAPAT GORDON, RATANA JARUBENJA, NITAYA KATLEERADAPAN, NARONG PETCHPRASERT AND AMARA PONGSAPICH

10.1 Thailand’s industrialisation process and the informal sector 323 10.2 The emergence of Home Based Work in Thailand – some features 324 10.3 Research method 329 10.4 Homework sectors 332 10.5 The socio-economic situation of homeworkers 340 10.6. Initiatives in favour of homeworkers 349 10.7. Conclusions and policy recommendations 350 Annex 10.1: Case studies 351 Notes 354 References 357 PART III

Policy implications

359

11 Upgrading informal micro- and small enterprises through clusters: towards a policy agenda

361

SANTOSH MEHROTRA AND MARIO BIGGERI

11.1 Background 361 11.2 Informal activities, clusters and local systems of development 365 11.3 Clustering as a process of development: dirt roads, low roads and high roads 368 11.4 Empirical evidence on the potential of clusters 372 11.5 Policies to promote clusters of small enterprises and homeworkers’ activities 381 11.6 Concluding remarks 393 Annex 11.1: International agencies’ efforts at promoting clusters 394 Notes 397

Contents xv 12 Extending social insurance to informal wage workers

400

SANTOSH MEHROTRA AND MARIO BIGGERI

12.1 Understanding the historical evolution of social protection in rich countries and its relevance for Asian countries 402 12.2 Government and non-governmental initiatives for homeworkers in South and South East Asia 408 12.3 Implications of this study for social protection policy 413 12.4 Concluding remarks 427 Annex 12.1: ILO Convention on Home Work No. 177, Ratification of Relevant Conventions and Conventions 428 Annex 12.2: Ratification of ILO Conventions 438 Annex 12.3: Organisations for/of homeworkers 443 Annex 12.4: Main international voluntary certifications, corporate social responsibility (SA 8000) and the Global Compact initiative 445 Notes 450 References Index

453 469

Figures

1.1 Relationship between women’s achievements (Gender Development Index) and proportion of children (aged 10–14) not working 1.2 Household human development cycle without policy interventions: vicious circle from generation to generation 1.3 Household human development cycle with policy interventions: virtuous circle from generation to generation 1.4 Factors influencing household’s human development level 2.1 Countries and locations of selected clusters 2.2 Time line of the research 3.1 Example of input–output structure of a value chain 3.2 Value chain map – share of female workforce 3.3 Value chain map – incense stick making in India 3.4 Value chain map – carpet weaving in Pakistan 3.5 Production (a) and market (b) chain maps and involving intermediaries in rattan industry in Indonesia 3.6 Subcontracting chain in okra production in the Philippines 3.7 Relationship between foreign ordering agency, manufacturer, contractor, subcontractor and workers in Thailand leather sector 4.1 Double burden for homeworker women in India 5.1 Double burden for female children (5–14) in Pakistani homeworker households: hours per week in homework and hh chores 6.1 Areas covered for homework study in India 6.2 Value chain for Rs 100 worth to the consumer 7.1 Carpet weaving 8.1 Production chain involving intermediaries in the rattan industry 8.2 Marketing chain 9.1 Subcontracting chain in fashion accessories 10.1 Relationship between foreign ordering agency, manufacturer, contractor, subcontractor and workers 10.2 Value chain, saa paper sheet and products 11.1 Types of clusters 11.2 Clusters and human development strategic routes from dirt roads to high roads

22 23 23 24 32 33 65 68 70 71 71 72 73 98 150 174 187 222 267 269 303 334 338 367 368

Tables

1.1 2.1 2.2

2.3 3.1 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9

4.10 4.11 4.12 4.13 4.14 4.15 4.16 4.17

Informal sector employment – selected Asian countries Number of focus group discussions, case studies and households surveyed by country Surveys of homeworker households in five Asian countries: sectors, location, number of FGDs, case studies and households surveyed Market for selected products of homeworkers Homeworkers contract – verbal or non-verbal? Characteristics of the households surveyed House ownership in homeworker households Home, homework households by type of wall and roof materials Electricity in homeworker households Source of water in homeworker households Toilet facilities inside homeworker households Durable goods in homeworker households Socio-economic indicators for women (1999) Share of women working in homework (hw) households by sector and in non-homeworker households (control group, CG) Average hours worked in homework by women Educational level of women homeworker (hw) household Women homeworkers (hw) with health problems due to work Membership of organisation Ranking of problems by women homeworkers, by sector Ranking of priorities by women homeworkers, by sector and possible interventions by the government Summary statistics Determinants of women homeworkers, health status: Poisson regression model

10 34

40 42 75 85 86 86 88 89 90 91 93

94 97 100 101 105 108 110 112 115

xx

Tables

4.18

5.1 5.2 5.3 5.4 5.5 5.6 5.7 5.8 5.9 5.10 6.1 6.2

6.3 6.4 6.5 6.6 6.7 6.8 6.9a 6.9b 6.10 6.11 6.12 6.13a 6.13b

Determinants of women homeworkers, earnings (productivity in value per hour): OLS regression – robust standard errors Share of children working by age group in homeworker (hw) households Share of children working by age group in non-homeworker (non-hw) households (control group) Work and study status of child by age and sex (%) – homeworker (hw) households Work (as homeworkers) and study status of child by age and sex (%) – homeworker (hw) households Reasons for not sending children to school-homeworker and non-hw household Children working in homework: average hours worked by children per day by age group and sex Children working in homework: average hours worked by children per week by work and study status Health of children – homework (hw) households Determinants of child status: results of a multinomial logit regression Determinants of child working hours: results of a Heckman selection model regression India: contribution of the organised and unorganised sectors to net domestic product and employment Surveys of homeworkers households in India: sectors, location, number of FGDs, case studies and households surveyed Per cent of homework (hw) in non-agricultural informal enterprises and sectoral composition of hw Impact of macroeconomic policy-disaggregated by sector Market for finished products: domestic markets and export markets for the three sectors The value chain for commodity worth Rs 100 to consumers Composition of income of homeworker (hw) households Estimated proportion (%) of population below poverty line Work status of females (adult women and girls) – homeworker (hw) households Work status of females (adult women and girls) – non-hw households Family members who help women in homework Women literacy rates Frequency (%) of full-time participation in homework by household members Status of children by gender – homeworker (hw) households Status of children by gender – non-hw households

117 128 130 133 136 140 143 144 148 155 160 176

178 180 182 184 187 189 190 192 193 194 194 195 198 199

Tables xxi 6.14 6.15 7.1

7.2 7.3

7.4 7.5

7.6 7.7 7.8 7.9 7.10 7.11 7.12 7.13 7.14 7.15

8.1 8.2 8.3 8.4 8.5 8.6

School enrolment status of children (5–14) by gender Reasons for not attending school in homework (hw) households Pakistan: percentage distribution of persons (10 years and above) empoyed in the informal sector by occupation, gender and region Percentage distribution of employed persons (10 years and above) by employment status, gender and region Surveys of homeworker households in Pakistan: sectors, location, number of FGDs, case studies and households surveyed Details of activities by sectors and actors Daily earnings of homeworkers as a percentage of subcontractor earnings and as a percentage of what consumers pay Average hours worked and income by sector Percentage of male household heads not working by sector State of indebtedness by sector Mean household size and number of children per household Mean hours spent by children on homework and other activities Socio-economic profile of homeworker and non-homeworker households Monthly expenditure patterns (Rs) of homeworker and non-homeworker households Percentage responding affirmative regarding easy access and affordability of health facilities The main reasons why children doing homework and CG children were not in school Time allocation to homework (hw) and other activities by gender for children doing hw and children not doing hw Indonesia: share of working adult population 15 years and over, workforce and unemployed, 1990–99 Distribution of formal and informal workers by sector of economic activity, 1990–99 Share of female/male workers by sector, 1990–99 Numbers and distribution of child workers Distribution of child workers by economic activity and sector, 1990–99 Surveys on homeworker households in Indonesia: sectors, location, number of FGDs, case studies and households surveyed

200 202

212 212

216 219

221 222 225 227 228 228 230 231 234 234

236 252 253 255 256 258

261

xxii Tables 8.7 8.8 8.9 8.10 8.11 8.12 8.13 8.14 9.1 9.2 9.3 9.4 9.5 9.6

9.7a 9.7b 9.8 9.9

9.10 9.11 9.12 9.13 9.14 10.1 10.2

10.3

Homeworkers: some descriptive statistics by sector Share of children working by age group in homeworker and CG households Women workers: main reasons for doing homework Distribution of home and CG workers by level of education Average weekly working hours for homeworkers and CG workers based on gender Distribution of households by share of total household income derived from homework activities Average monthly equivalent wages for homeworker Percentage of children aged 6
claiming health problems during the preceding month Philippines: labour force participation rate and employment status, 1990–99 (in thousands) Share of formal and informal sectors, 1980–2000 Distribution of formal and informal employment by sector, 1989–93 Distribution of home-based workers by economic activity and location, 1993 Distribution of home-based workers by primary activity and by region, 1993 Surveys on homeworker households in the Philippines: sectors, location, number of FGDs, case studies and households surveyed Working hours per day full time by seasons Home décor – working hours by gender and season Characteristics of the CG and homework (hw) households Average annual household income by sector (hw and CG households) compared average to household income at regional level Percentage of female workers among homework (hw) and CG households Education attainment by level of school and by sector Share of children working by age group in homework (hw) and CG households Percentage of children working by gender and age group Percentage of workers that are members of the PATAMABA network Workers in labour force, 1988 and 1994 Thailand: share of workers by sector, distribution of formal/ informal sector workers by industry and distribution of informal workers in non-agriculture, 1994 Distribution of workforce and homeworkers by location and sector, year 1998

264 265 270 272 272 273 274 276 291 291 292 294 295

298 300 300 304

305 306 307 308 309 310 324

325 325

Tables xxiii 10.4 10.5 10.6

10.7 10.8 10.9 10.10 10.11 10.12 10.13a 10.13b 10.14 10.15a 10.15b 11.1 12.1

Distribution of homeworkers by sector Characteristics of labour force in the informal sector compared to those in the formal sector Surveys on homeworker households in Thailand: sectors, location, number of FGDs, case studies and households surveyed Value of exports, 1995–98 Summary of production characteristics of the three sectors Average income from homework (hw) and other sources per sector Percentage of households in which children are in school, by educational level Health conditions Socio-economic conditions in homeworker households Average monthly income and expenditure per capita Average income per household per month Average monthly household income from homework Percentage households in which children help with chores Children’s help with household chores Principal activities, by component and type of PIP Characteristics of social security systems: sources of revenue and types of benefits

326 329

330 333 336 341 342 343 344 344 344 347 348 348 395 405

Contributors

Narumon Arunotai has a PhD training in Anthropology. She has been with Chulalongkorn University Social Research Institute (CUSRI) for ten years, teaching Development Theory and conducting research on varions social issues. Laveesh Bhandari completed his PhD in Economics from Boston University, USA in 1996. He was a Consultant for Manhattan Funds, New York working on the valuation methodologies of derivative securities. He joined the National Council of Applied Economic Research (NCAER), New Delhi, as a Senior Economist in 1997. At NCAER he led research teams for various studies in the Industry and Infrastructure division. He was also the Managing Editor of the Journal of Emerging Market Finance. Mario Biggeri is Associate Professor in Development Economics at the Department of Economics, University of Florence, Italy, where he teaches Development Economics, Economics of Transition, International Aid and Development and Cluster Development in Developing Countries. He is ViceCoordinator of the PhD program in Politics and Economics for Developing Countries. He was educated at the University of Florence (BA), University of Reading (UK) (MSc), and at the University of Siena (Italy) where he did his doctorate (PhD). His research interests include clusters of small and medium enterprises and informal activities, child labour and child’s capabilities, international aid, rural development, and economies of transition, and he has published widely in these areas. He is co-author of a book on International Aid and co-editor of two other books. Papers have been published in Journal of International Development, Journal of Chinese and Business Studies, World Development, Journal of Human Development and other journals. He has been a Consultant for UNICEF Innocenti Research Centre (IRC) (2000–02), for the ILO/UNICEF/World Bank Project Understanding Children’s Work (UCW) Project (2003), for the L’Institute – Institute for Industrial Development Policy and for the Istituto Agronomico per l’Oltremare (IAO) [Italian Ministry of Foreign Affairs] on food security and international aid. Khun Napat Gordon was a Researcher at CUSRI and retired from CUSRI a few years ago.

xxvi Contributors W. Hartanto was at the Biro Demography dan Tenaga Kerja, Badan Pusat Statistik, Jakarta. Ratana Jarubenja is a researcher at the Chulalongkorn University Social Research Institute, Bangkok, Thailand. The late Nitaya Katleeradapan was a Researcher at CUSRI and passed away a couple of years ago. Sajid Kazmi is a Consultant for the Sustainable Development Policy Institute. Shahrukh Rafi Khan served as Executive Director of the Sustainable Development Policy Institute in Islamabad from 1996 to 2002. Khan has taught economics at the University of Utah, the State University of New York, and Vassar College. Visiting Professor of Economics at Mount Holyoke College, University of Utah, Department of Economics. Saba Gul Khattak Executive Director at SDPI, holds a PhD degree in Political Science and specializes in development studies, women’s studies and peace studies. She works on issues of women’s economic empowerment, governance, peace, security and violence and focuses on women and child workers, refugees, and other marginalized communities. She actively contributes to dialogues on these themes at different fora in Pakistan and abroad. Santosh Mehrotra is a human development economist educated at the New School for Social Research, New York, and the University of Cambridge (PhD). He was Associate Professor in Jawaharlal Nehru University, New Delhi, before moving to the United Nations in 1991. For the past 15 years he has worked on the human impact of macro-economic policy. He led UNICEF's research programme on economic and social policy on developing countries at the Innocenti Research Centre, Florence, 1999–2002. He was cheif economist of the global Human Development Report, 2002–04, and UNDP’s Regional Economic Adviser for Poverty for the Asia region, 2005–06. His books include: India and the Soviet Union. Trade and Technology Transfer, Cambridge University Press, 1990; Development with a Human Face. Experiences in Social Achievement and Economic Growth (with Sir Richard Jolly), Clarendon Press, Oxford, 1997; Le Developpement a Visage Humain, Economica, Paris, 2001; Universalizing Elementary Education in India. Uncaging the Tiger Economy, Oxford University Press, 2005 (coauthored); The Economics of Elementary Education in India, ed., Sage Publishers, 2006; and Eliminating Human Poverty. Macro-economic and Social Policies for Equitable Growth (with E. Delamonica), Zed Press, London, 2007. He is currently advisor to the Planning Commission, Government of India, and is involved in writing India's 11th Five Year Plan (2007–11). Mayling Oey-Gardiner, a Sociologist and Demographer, became in 2001 the first female professor at Faculty of Economics at the University of Indonesia. She led the Indonesia research team. Rosalinda Ofreneo is a Professor of the Women’s Studies at the College of Social Work and Community Development of the University of the Philippines and is one of the leading figures of the Philippine women’s movement.

Contributors xxvii PATAMABA is a leading organization of women home-based workers in the Philippines, which participated in the research, and its leader, Ms Gula, was a particularly important contributor to the survey. Narong Petchprasert teaches Economics at Chulalongkorn University. Amara Pongsapich was the Dean of Political Science Department after serving as Director of the Chulalongkorn University Social Research Institute (CUSRI) for many years, which carried out the Thailand country study. Rosario del Rosario is a Professor at the College of Social Work and Community Development of the University of the Philippines, and led the country research team in the Philippines for this study. Ratna M. Sudarshan joined the Indian Social Studies Trust as Director in August 2003. Earlier she was a Principal Economist at the National Council of Applied Economic Research, New Delhi. She has researched on gender and informal economy concerns, education and literacy, and has a strong interest in the policy process and in developing research-activist linkages. She has a Master’s degree in Economics from the University of Delhi and the University of Cambridge. Evelyn Suleeman was a Researcher at Insan Hitawasana Sejahtera (IHS), a social science research and consultant firm located in Jakarta, that carried out the Indonesia country study. Indrasari Tjandraningsih is Research Coordinator at the Akatiga Foundation, Jakarta, and has written extensively on labour issues in Indonesia. Shanta Venkataraman is an Economist at the NCAER with extensive experience in field research and survey design, implementation and data analysis. She has co-ordinated the data collection for the India study reported in this book. Hesti Wijaya was a Researcher at the MWPRI, an advocacy organization for home-based workers in Indonesia, which is linked to Homenet International.

Preface

There has been an increasing informalisation of the non-agricultural labour force in developing countries over the last half-century of development. Simultaneously there has been a feminisation of that very large segment of the labour force. One aspect of these dual phenomena is the growth of subcontracted homework in both manufacturing as well as services, and to some extent in agriculture. This book draws on small-scale surveys in at least three sectors carried out in five Asian countries – two low-income (India, Pakistan) and three middle-income countries (Indonesia, Thailand, Philippines) – where industrial outwork and informal sector production, usually by women working out of home, are widespread. The central issues of this study are at least four: understanding industrial/ agricultural outwork to homeworkers; the factors that foster the capabilities of homeworker households; if homework can ‘under some conditions’ foster the development of the local economic system; and the policies to be implemented at different levels to promote local economic development as well as human development of such households. Industrial outwork has a dual and contradictory character: on the one hand, as a source of income diversification for poor families and the emergence of microenterprises, and on the other, a source of exploitation of vulnerable workers as firms attempt to contain costs. Accordingly, this book examines the social protection needs of these workers, and also argues for public action to promote such work and protect such workers as part of a potential new labour-intensive growth strategy in these and other developing countries. It also examines the phenomenon of child labour in homework in the same countries, and discusses what can be done to minimise such work. Child labour is widespread in industrial outwork in the informal sector in most developing countries. This form of child labour will not attract the penal provisions of a country’s laws banning child labour. The book examines the incidence of child work in such households, the child’s schooling, reasons why children are working, their work conditions, their health, and also gender issues. Based on focus group discussions with children, it also attempts to articulate the voices of children working. Policy implications are drawn accordingly. Part I uses the survey data for all five countries and presents a cross-country analysis of the data. Part II presents the country case studies. Chapter 1 first

xxx

Preface

discusses the empirical context, by examining the growing informalisation of the non-agricultural labour force, and then presents a theoretical framework within which to locate homework. Chapter 2 reports on the methodology adopted by the surveys, and profiles the sectors chosen for study in the five countries. Chapter 3 analyses the value chain in the products under study in the five countries, as well as the forms of insecurity it results in. Chapter 4 does a cross-country analysis of the social and economic characteristics of the homeworker households, and reports on the findings on women workers and their households. Chapter 5 does a cross-country analysis of child work within the homeworker household. Part II (Chapters 6–10) is devoted to each of the country cases: India, Pakistan, Indonesia, the Philippines and Thailand. Part III discusses the implications for government policies of the entire set of studies – first growth-oriented policies (Chapter 11) and then social protection policies (Chapter 12). Within the conceptual framework of cluster theory, Chapter 11 examines the policies that can promote employment and the growth of the local economic system linking micro-enterprises and small firms to a longer value chain; it also discusses the Chinese and Italian experiences with the growth of clusters. Chapter 12, while using a historical approach to draw lessons for developing countries from the evolution of social security systems in the now industrialised countries, presents a plan for social insurance for informal workers in developing countries. Santosh Mehrotra and Mario Biggeri

Acknowledgements

Papers for this book were presented at UNICEF’s International Conference on Home Based Work by Women and Children in Asia, Bangkok, 6–8 December 2001, the conference on Child Labour in South Asia, Jawaharlal Nehru University, New Delhi, 15–17 October 2001, and a World Bank conference on Child Labour in Oslo in May 2002. We are grateful to M. Bellandi, A. Giommi, L. Guarcello, P. D. Falorsi, S. Falorsi, B. Meyers, J. Micklewright, Ravi Srivastava, M. Suhrcke and F. Volpi for their comments. C. Vizcaino provided excellent research support through much of 2002. Marco Corsi helped in the initial stages of the research project in 2000. We acknowledge with gratitude the help of S. Mariani in getting the database in order. Without the assistance of Yvette Verna this project would have taken much longer to complete; we are deeply indebted for her caring support, and the warmth of her personality. My years in Florence were constantly warmed by her presence, and her smile. Many international agencies and several NGOs contributed to the study. We are grateful to the Innocenti Research Centre, UNICEF, Florence (Italy), for its institutional support over the three years (1999–2002) that Mehrotra was there; the research project was conceived and designed with its institutional backing. The research was made possible by the financial support of the Norwegian government (through its support to the Child Labour programme of UNICEF, New York). We are grateful to Alec Fyfe for his encouragement to undertake this work. We are extremely grateful to the local affiliates of Homenet International (a network of NGOs working for, and with, home workers in the five selected countries in Asia) for collaborating with us on this research. The moral support we received during the course of this project from Women in Informal Employment Globalizing and Organizing (WIEGO), started in 1997, proved invaluable to us. We would particularly like to thank Renana Jhabvala (Self-Employed Women’s Association), Rakawin Lee (formerly of Homenet Thailand, and ILO, Bangkok), Hesti Wijaya (MWPRI, or Indonesian affiliate of Homenet) and Lucy Lazo (formerly of PATAMABA, Philippines, and later UNIFEM, Bangkok). Most important of all, this book would not have been possible without the cooperation of the authors of the country studies. We thank them for their enthusiastic involvement in bringing the studies to fruition.

Part I

A cross-country analysis of industrial outwork in Asia

1

The empirical context and a theoretical framework Santosh Mehrotra and Mario Biggeri

There are at least two major differences between the conditions under which developing countries are industrializing today, and the conditions faced by the now-industrialized countries over a century ago. First, there have been massive changes in technology. The techniques of industrial production in developing countries being used now, usually borrowed from industrialized countries with a rather different factor endowment, means that the employment elasticity of manufacturing output is much lower than it was over a century ago. The technologies imported by developing countries were the product of a second Industrial Revolution in the advanced capitalist countries. In most developing countries, in the phase of import-substituting industrialization, import and capital-intensive products and inappropriate technologies entrenched a situation wherein formal sector manufacturing output has grown, but manufacturing employment has not grown commensurately (Stewart, 1974). The employment growth in the formal sector is insufficient to absorb even the growth in the labour force. In fact, as population has grown and the agriculture sector sheds its surplus labour, the slow growth of manufacturing employment has contributed to a growth of employment in the informal economy in urban areas. The result is not what Lewis (1954) had anticipated: that the formal, modern, industrial economy will absorb the labour in the traditional, rural, agricultural economy (the ‘dual economy’), but there has been the emergence of a differentiated informal economy.1 This development is central to framing the argument in the rest of this book – and hence we devote some discussion in this chapter and elsewhere in the book to it. Indeed, although the literature produced by Lewis and his successors captured very important characteristics of the traditional (informal) sector (Lewis, 1954, p. 141) and the dynamics of the labour market of developing countries (and later of the urban–rural migration phenomena (Harris and Todaro, 1970) – the dualistic models devised overlooked other relevant aspects of the informal economy. In particular dualistic models identified the function of the traditional (informal) sector in the economic system merely as a passive one with the labour force seen as reserve army, as a reservoir from which the expanding modern sector draws labour. Furthermore, in the dualistic models the informal economy was depicted as an ‘homogeneous’ entity disregarding the peculiarities and different dynamics within it (to some extent connected to the specific rural/urban location). Indeed,

4

Santosh Mehrotra and Mario Biggeri

if on the one hand, the informal economy is a symptom of economic dysfunction and seen just as a low-income job creator, on the other it can be seen as an opportunity for the development process. Considering the diversity within the informal activities the dichotomized classification – formal versus informal activities – is not enough and tends to hide more than it reveals. In particular, instead of considering the informal economy as an undifferentiated whole there is the possibility of distinguishing among a more productive or dynamic-advanced component (often related to micro and small enterprises and accompanied by features such as dynamism, flexibility, relatively higher capital accumulation, entrepreneurial creativity and initiative, export links), and a non-advanced traditional part which more closely fits the customary image of the informal economy (Ranis and Stewart, 1995).2 The overlooking of these aspects of informal activities contributed to inappropriate analyses and to wrong policy planning and implementation. Moving to the second difference, we argue there is a great difference in respect of the demographic pressure faced by developing countries in the phase of their first industrial revolution, compared to the now-industrialized countries at an earlier comparable phase of development. During the first century after the Industrial Revolution began, the total population of Europe grew only from 185 million in 1800 to about 400 million in 1900. However, over the next century (1900–2000), when developing countries were attempting their own industrial revolution, populations multiplied several-fold.3 The prospects for the formal sector of industry or services to absorb the growing labour force is very limited indeed. Under the circumstances, growth in the informal economy, and with it homebased economic activity, seems inevitable. These are the structural factors, endogenous to a developing economy, that have led to the rise of informal employment. However, there have also been contingent factors that have driven its rise in the last two decades of the twentieth century. Since 1980 the structural adjustment and stabilization programme have resulted in downsizing in public enterprises and in government bureaucracies in first developing countries, and then in the 1990s in the erstwhile centrally planned economies. Thus, many public sector employees who were earlier in the formal economy have been forced to join the informal economy. These newly unemployed public sector workers, both in transition and developing economies, are in fact engaged in informal work. Another international force that has led to the growth of informal employment and the informal economy is increasing global integration and competition in the 1990s. To improve their global competitiveness, capital investing globally has increasingly shifted production to countries with lower labour costs or adjusted employment practices in rich countries to more informal arrangements. In other words, in recent years the expansion of the informal economy can be linked not only to the capacity of formal firms to absorb labour but also to their willingness to do so (Portes et al., 1989; Chen et al., 2002). This increased willingness has manifested itself in the increased international role of subcontracting through commodity chains. The post-Second World War period saw the development of essentially two types of international subcontracting: producer-driven commodity chains, and buyer-driven commodity chains, with

Empirical context and a theoretical framework 5 the difference lying in the location of the key barriers to entry (Gereffi, 1994b). In producer-driven chains, large transnational companies play a central role in coordinating a production network, including backward and forward linkages. Such chains operate in capital and technology-intensive commodities for example, automobiles, aircraft, semi-conductors, electrical machinery (e.g. the Ford Escort is manufactured and assembled in 15 countries). There is little scope for subcontracting to households in such industries. Buyer-driven commoditychains, however, are usually operated by large retailers and brand-name merchandisers. Such large retailers as K-Mart, Gap and Wal-Mart would play a central role. Such chains operate in labour-intensive consumer goods like garments, footwear, toys and houseware. These now extend to fruits, vegetables, non-timber forest products, and many more. The production of such goods is usually located in developing countries, while the high-value activities (e.g. design and marketing) are in industrialized countries. Often such chains extend all the way to the industrial outworker based at home. The trend of global manufacturing is to move from producer-driven to buyer-driven chains in international subcontracting (Gereffi, 1994b). In fact, Wal-Mart, the world’s biggest retailer, has driven this model, buying products from 65,000 suppliers worldwide and selling to over 138 million consumers every week through its 1300 stores in 10 countries (Oxfam, 2004a).4 This book is concerned with one aspect of those employed in the informal economy – industrial outwork – which accounts for a very significant part of the informal economy in most developing countries. The central concerns of this study are at least four: the understanding of homeworkers’ activities; the factors that foster the capabilities of homeworker households (especially of the women and children); if homework can ‘under some conditions’ foster human development at the level of the local economic system; and the policies to be implemented at different levels to promote employment and social protection for such workers. Section 1.1 of this chapter is devoted to this macro-economic issue, examining the significance of the informal economy in total employment – and is largely empirical in content. The second section, however examines the meso-economic phenomena of clusters of both homeworkers as well as industrial units, that is, Small and Medium Enterprises (SMEs), and how they could emerge as a dynamic engine of growth. This dynamic was demonstrated in the Italian ‘miracle’ in the twentieth century and more recently by the success of Chinese Township and Village Enterprises (TVEs). Hence Section 1.2 focuses on the characteristics of homeworkers and their positive and negative aspects, both from an individual micro-economic as well as from a meso-economic perspective. This discussion is part of a broader framework of cluster theory. From a dynamic point of view, we are interested in exploring the potential for clusters of homework activities developing into micro-enterprises, if not small enterprises, as a possible path for fostering economic and human development of the local system of production. These arguments will be analysed in detail in Chapter 11. Section 1.3 examines the factors influencing the human development (HD) level of any household (including those of homeworkers), and changes in the HD level of household members, as well as the role of synergies that can impact

6

Santosh Mehrotra and Mario Biggeri

that level through public action. It also discusses factors, both endogenous and exogenous to the household, that are important in understanding the dynamic of inter-generational transfer of human capabilities (or poverty) in poor households engaged in the informal economy. This conceptual discussion in Sections 1.2 and 1.3 frames the discussion in the rest of Parts 1 and 3 of the book. The chapter ends with some final remarks (Section 1.4).

1.1

Background: the informal economy and industrial outwork

Chen et al. (2002) have very usefully suggested that those in the informal economy can be classified according to the following employment status categories: 1

Non-wage workers: ●

Employers, including – –

●

Self-employed, including – – –

2

Owners of informal enterprises Owner operators of informal enterprise

Heads of family business Own-account workers Unpaid family workers

Wage workers: ● ● ● ● ● ●

Employees of informal enterprises Domestic workers Casual workers without a fixed employer Homeworkers (also called industrial outworkers) Temporary and part-time workers Unregistered workers.

The fact that some people hold multiple statuses complicates classification. However, Chen et al. (2002) rightly emphasize that there is a need ‘at all times to distinguish between how policies and regulations affect the self-employed (and their enterprises or economic activities), as opposed to how they affect informal wage-workers (and their employers)’ (p. 6). For statistical purposes, the International Conference of Labour Statisticians (ICLS) in 1993 determined to use a definition of ‘informal sector’ based on production units or enterprises rather than on employment status. The 1993 ICLS adopted an international statistical definition of the informal sector that was later included in the revised International System of National Accounts (1993 SNA) based on enterprises.5 The 1993 ICLS defined the informal sector as all unregistered or incorporated enterprises below a certain size, including micro-enterprises owned by informal employers who hire one or more employees on a continuing

Empirical context and a theoretical framework 7 basis; and own-account operations owned by individuals who may employ contributing family workers and employees on an occasional basis.6 In terms of size, informal employment comprises one-half to three-fourths of non-agricultural employment in developing countries: 65 per cent in Asia; 51 per cent in Latin America; 48 per cent of non-agricultural employment in North Africa, but as high as 72 per cent in sub-Saharan Africa.7 Some countries include informal employment in agriculture in their estimates of informal employment. In these countries the inclusion of informal employment in agriculture increases remarkably the share of informal employment: from 83 percent of non-agricultural employment to 93 per cent of total employment in India; from 55 to 62 per cent in Mexico; and from 28 to 34 per cent in South Africa.8 In terms of the two-fold classification above – self-employment in informal enterprises (small, unregistered) and wage employment in informal jobs (without secure contracts, worker benefits, or social protection) – self-employment has a greater share of informal employment (outside of agriculture) than wage employment. Self-employment accounts for 59 per cent of informal employment in Asia, 60 per cent in Latin America, 62 per cent in North Africa and as much as 70 per cent in sub-Saharan Africa. Wage employment is also significant in developing countries, accounting for 30 to 40 per cent of informal employment (outside of agriculture). As we noted above, these included casual day labourers/part-time workers, domestic workers, undeclared workers, and industrial outworkers (notably homeworkers). In fact, home-based workers and street vendors are two of the largest sub-groups of the informal workforce – with home-based workers the more numerous, but street vendors the more visible of the two. Taken together they account for an estimated 10–25 per cent of the non-agricultural workforce in developing countries and over 5 per cent of the total workforce in developed countries (ILO, 2002a). Informal employment is normally a larger source of employment for women than for men in the developing world; 60 per cent or more of women workers are in informal employment outside of agriculture.9 Although women’s labour force participation rates are lower than men’s, the limited data available point to the importance of women in home-based work (and street-vending10) in developing countries: 30–80 per cent of all home-based workers (including both selfemployed and homeworkers); and 80 per cent or more of homeworkers (industrial outworkers who work at home) are women. The term ‘home-based worker’ is used to refer to the broad category of workers who carry out remunerative work within their homes or in the surrounding grounds.11 The term ‘homeworker’ is used to refer to a sub-set of home-based workers: industrial outworkers who carry out paid work from their home, for firms/ businesses or their intermediaries, typically on a piece-rate. It is important to distinguish between the two, since homeworkers often face more problems of exploitation and isolation while the self-employed face more problems of exclusion. This book is devoted mainly to an analysis of homeworkers in five Asian countries. Homeworkers occupy a grey intermediate space between the fully independent self-employed and fully dependent paid employees and they are often forced by

8

Santosh Mehrotra and Mario Biggeri

circumstances to work for low wages without secure contracts or fringe benefits and to cover some production costs and associated risks – including, buying or renting and maintaining equipment; providing workspace and paying for utility costs; and buying some inputs – often without help from their employers. This implies that their net remuneration may be significantly less than indicated by the piece-rates that they are paid (ILO, 2002a, p. 44). Historically, most homework involved manual work in labour-intensive activities: especially in textiles, garments, and footwear manufacturing industries. Manufacturing and assembly usually involves sewing, packing and routine assembly. It has also involved artisan production, such as weaving, basket-making, embroidery and carpet-making. It is these kind of activities which are the subject of this book. Increasingly, homework also involves activities in the service and commercial sectors: clerical work in data processing, telecommunication, and telemarketing, but also highly skilled professional and technical consulting.12 Not surprisingly, in developed countries an increasing number of homeworkers are in services and commerce, and less in manufacturing.13 However, manufacturing is still important for homeworkers in rich countries. Thus, the garment industry in California is one of the largest in the United States, accounting for 160,000 workers, mostly immigrant women (with 15,000–20,000 garment workers in San Francisco and about 140,000 in Los Angeles).14 Recent research examined the condition of British homeworkers in the global supply chain. They are paid on average less than the minimum wage per hour, receive no sick, holiday, or maternity pay, they are made redundant without notice or compensation, they are not subject to adequate safety checks. Ninety per cent of these workers are women, 50 per cent of which are from ethnic groups, principally from Pakistan and Bangladeshi communities but also from Indian and Turkish groups (Oxfam, 2004b). In historical perspective, one should note that homework has been widely practised in industrialized countries going back to the industrial revolution. Homework in garment, textile and artificial flower production industries cannot be separated from that revolution. Homework survived largely due to the low wages paid to male workers in factories, the reserve army of labour available in agriculture, and the lack of alternative work for women. Homework was undertaken largely by poor, married women with children, who could thus support the family income – reasons not dissimilar to those relevant in developing countries today. Around 1900, criticisms of such work began, and finally prevailed among trade unions, since it was argued that they reduced employment for men in factories, weakened their bargaining position and detracted from women’s domestic ‘responsibilities’.15 In today’s developing countries, similar industrial outwork is being replicated – just on a much larger scale, affecting a much larger population. Thus, there are many reasons why we should be concerned about homeworkers: first, their significance in the growing phenomena of informal employment; second, the fact that they have none of the employment-based benefits or protection; third, the fact that working from home means they are usually isolated and face difficulties in organizing themselves; fourth, their work is associated with low pay. Finally, we

Empirical context and a theoretical framework 9 should be concerned because women are over-represented among them.16 In fact, it is in response to this growing phenomena and as a result of the efforts of advocates for homeworkers, there is a now an ILO Convention (no. 177), 1996, that sets out minimum standards for pay and working conditions for homeworkers that can form the basis for national laws and policies. A related Recommendation (no. 184) presents a whole programme of possible actions to improve the conditions of homeworkers. As of September 2004 only four countries have ratified the Convention: Finland (1998), Ireland (1999), Albania and the Netherlands (2002).17 Informal sector, subcontracting and homework in South and South East Asia Since the geographical focus of this book is on Asia, we turn now to some features of the informal sector, subcontracting and homework in South and South East Asia – so as to place our research on India, Pakistan, Indonesia, the Philippines, and Thailand in perspective. On account of the relatively fast growth of manufacturing in South East Asia, informal employment has grown in absolute terms. In other words, perhaps the pull factors from the formal manufacturing sectors have helped expand the informal economy in South East Asia. In South Asia the slow increase in agricultural output has led to farm incomes remaining stagnant, with surplus labour being pushed out to non-farm activities, whether in productive or non-productive activities – essentially as a survival strategy, to make a living. In either case, urban-based, informal activities in manufacturing have grown in both South and South East Asia, while also spreading to rural areas. Women who were hitherto non-earning members of the household, or engaged in traditional activities on their own account, now take on new types of work on a subcontract basis. Informal sector employment has grown in most Asian countries.18 In Pakistan, in rural areas, 73 per cent of all economically active females, and 68 per cent of such males, were in the informal sector; in urban areas, 61 per cent of all working women, and 64 per cent of such men, were in the informal sector (Khan et al., 2001). In India, as much as 80.3 per cent of the non-agricultural labour force is in the informal sector (with the remaining 19.7 per cent in the formal sector). A national sample survey in 1990–2000 in India showed that home-based workers represented 35.9 per cent of employment in the non-agricultural informal sector in 1999. The shares of informal employment, though still large in South East Asia, are not as significant as in India. Half of the Indonesian and Thai nonagricultural labour force is in the informal sector, while in the Philippines the share reaches 63.8 per cent (Table 1.1a–e). Home-based work seems to be prevalent in both urban as well as rural areas. As much as 59 per cent of India’s homeworkers are in manufacturing. What is significant is that of all homeworkers in manufacturing 75 per cent are in rural areas (rural areas is also where 56.4 per cent of all homeworkers in all sectors are located) (Sudarshan et al., 2001). Also in the Philippines, where homeworkers are 9 per cent of the informal sector labour force (14 per cent of non-agricultural

Table 1.1 Informal sector employment (million and %) – selected Asian countries Millions % of labour force (a) India (1993–94) Agricultural labour force Non-agricultural labour force Non-agricultural formal labour force Non-agricultural informal labour force Total labour force

264.4 124.6 24.6 100.1 389.0

68.0 32.0 19.7 80.3 100.0

Proportion of non-agricultural workers (10 years and above) formal and informal sector by gender and rural urban areas

(b) Pakistan (1999–2000) Total non-agricultural workers Formal Informal

(c) Indonesia (1999) Agricultural labour force Informal labour force Non-agricultural labour force Formal labour force Informal labour force Total labour force

Rural

Urban

Female Male

Female Male

100.0 26.9 73.1

100.0 100.0 32.4 39.3 67.6 60.7

100.0 35.9 64.1

Labour force*

% of labour force

38.4 31.6 50.4 25.5 24.9 88.8

43.2 35.6 56.8 28.7 28.0 100.0 Informal sector employment % of total employment

(d) Philippines All sectors Agriculture, fishery, forestry Non-agricultural Industry Mining and quarrying Manufacturing Electricity, gas, water Construction Services Wholesale, retail trade Transportation, storage, communication Financing, insurance, real estate Community, social, personal services

1989

1993

76.9 97.9 59.5 66.2 60.3 52.4 0.27 85.0 68.2 78.9 81.0 29.6 40.9

81.1 98.5 63.8 70.6 60.4 62.3 2.12 88.8 64.8 88.0 82.1 38.9 37.7

Empirical context and a theoretical framework 11 Table 1.1 Continued

(e) Thailand (1994) Agricultural labour force Non-agricultural labour force Formal labour force Informal labour force Total labour force

Millions

% of labour force

17.96 14.13 6.86 7.26 32.10

56.0 44.0 48.6 51.4 100.0

Sources: (a) Based on National Sample Survey 50th round; (b) Government of Pakistan, Labour Force Survey 1999–2000 (2001, p. 21); (c) Sakernas 1999; (d) Country report. Modified Residual Approach; (e) The 1994 Formal and Informal Labour Force Market Survey, National Statistical Office. Note * Working Labour force.

workforce (ILO, 2002a, p. 47), more than half of homeworkers are rural dwellers (56.9 per cent). Almost all home-based work in the Philippines is in the manufacturing sector (and this is concentrated in urban areas); mainly home-based work in services is found in rural areas (Rosario del Rosario, 2001). Also in Thailand most of the homeworkers (which represent 2 per cent of the non-agricultural workforce 80 per cent of which are women) are in manufacturing (95.4 per cent) (Arunotai et al., 2001). Thus, homeworkers are found mainly in manufacturing and services,19 but also in agriculture (as we will see later), and in both rural and urban areas. The ‘rise’ of subcontracting There are many features that separate the formal from the informal economies, but at least in manufacturing there is one strong link between the two especially in Asia and Latin America20 – the phenomenon of subcontracting (to smaller firms which in turn subcontract to homeworkers).21 We summarize here some of the reasons for the increase of subcontracting in the value chain at national and international level. One of the most notable changes in industrial production has been the reduction in some of the traditional costs of production such as transportation and communication costs.22 This has enabled production to be farmed out while control is still maintained by the principal firm contracting out the work. Furthermore, another reason for the increasing subcontracting are the international policies implemented through international financial institutions with increasing trade and financial liberalization23 and flexibilization of the labour market. ‘Throughout the 1980s and 1990s, the IMF and the World Bank recommended and required, through loan conditionality, that governments make their labour laws more “flexible” ’ (OXFAM, 2004a, p. 42).

12

Santosh Mehrotra and Mario Biggeri

‘Labour market flexibility’ is another term for weakening labour laws (including facilitation of ‘flexible’ contracts, limiting collective bargaining, increase overtime hours, cut overtime pay, reduce gender-sensitive benefits) (OXFAM, 2004a, p. 42).24 Governments under pressure from local and foreign investors and from IMF and World Bank loan conditions have too often allowed labour standards to be defined by the demands of supply chain flexibility: easier hiring and firing, more short-term contracts, fewer benefits, and longer periods of overtime. ‘In this view, the process of “informalization” can be seen as an international move on the part of the governments to “weaken the rights of workers and unions . . . with the acquiescence of the state in the interest of renewed economic growth” ’ (Rakowski, 1994, p. 504; Delahanty, 1999). Thus, the trends in the international economy have tended to increase the role of subcontracting. As mentioned earlier, there is evidence that both informalization of the labour force, as well as its feminization, have grown in recent decades (Carr and Chen, 1999, 2001). While perhaps it is less easy to establish empirically that subcontracting has grown, there are sound reasons for suggesting that subcontracting has indeed grown. Conceptually, there is some fragmentary evidence that subcontracting involving industrial outworkers should have increased with increased export production in developing countries (Carr et al., 2000; Kaplinsky et al., 2001; World Commission on the Social Dimension of Globalization, 2004). The most successful exporters among developing countries have been those engaged in labour-intensive manufacturing in the last three to four decades. In many export-oriented product groups, low labour costs have provided the cost advantage to developing countries – garments have been an increasingly important example, but not the only one. Labour intensity is high in such products. High labour costs in Europe, Japan and North America have driven firms in industrialized countries to outsource manufacturing of consumer non-durables to South East and South Asia, as well as Latin America. Thus there is reason to believe that subcontracting should have grown with export-orientation, and with it, homework. Historically, export-oriented labour-intensive manufacturing grew most rapidly first in the East Asian newly industrializing countries (NICs) – Taiwan, Korea (also Singapore, Hong Kong) – in the 1970s. The same pattern of growth was rapidly emulated by the South East Asian economies – Malaysia, Thailand and Indonesia – from the late 1970s onwards. Thus the Thai government adopted a policy of industrialization for export with special emphasis on labour-intensive industries during the fourth national development plan (1976–81), like Indonesia and the Philippines. Therefore, exports expanded so rapidly from these countries that in 1980 and 1990 exports accounted for 24 and 34 per cent of Thai GDP respectively, 25 (data of 1982) and 26 per cent of Indonesia’s,25 23 and 28 per cent of that in the Philippines. Even more remarkably by 2000, those shares had risen to 67 per cent for Thailand, 39 per cent for Indonesia, and 56 per cent for the Philippines (World Bank, 2002). Buyer-driven subcontracting chains in footwear, garment-making and textile industries were directly involved in the expansion of these labour-intensive manufactures. In other words, with the growth of export-orientation in policies subcontracting is likely to rise, including subcontracting to homework.26

Empirical context and a theoretical framework 13 In fact, the tremendous success of first the East Asian, and then the South East Asian countries in labour-intensive manufactured exports, based on their labour cost advantage, saw many other countries adopting similar commercial procedures in turn. For the high- and middle-income countries of Latin America it has been used as a strategy to defend against competition from Asian products, just as in East and South East Asia subcontracting has been an integral part of a strategy of export-oriented industrialization based on labour-intensive products. This global competition to cut costs and to avoid tax determines the unclear location of profits since multinational companies locate geographically using a ‘predatory practice’. The competition between costly labour and cheap labour and between tax conditions (and the distribution of tax-monitoring) among states, ends up ultimately under-mining the workers at the expense of social protection. Thus, in recent years, some Latin America countries followed a defensive strategy to remain competitive, as cost-cutting in formal enterprises constrained formal workers to become informal workers, thus losing their social protection. South and South East Asia had no protection at all in the informal economy. In Europe, the growing outsourcing and informalization of the labour market driven by trade and financial liberalization are eroding not only wages, but also social protection in countries that were thought immune on these issues, and workers’ right are again on the agenda of trade unions. European Union countries have gradually changed their laws on labour to reduce the cost of labour and to increase its flexibility (Vielle and Waltery, 2004); at the same time in some sectors informalization is increasing, as in Latin America. If the issues of social protection and labour income are not addressed with some regulation at international level the prospect for social protection of workers and human development in general looks like ‘a race to the bottom’ (a phrase popularized by Dani Rodrik, 1997). Indeed, in a world of mobile capital and immobile labour, the globalization of the economy can make tax-based redistribution or social protection problematic (Cornia, 2003). Similarly, developing countries aiming at attracting foreign capital are under pressure to reduce tax rates. Indeed, the main problem concerning globalisation is that the participation in the global product markets and the geographical dispersal of economic activity has not led to a concomitant spread in social and economic benefits for those newly integrated populations.27 (Kaplinsky et al., 2001)

1.2

Homework – a driver of poverty, or a trigger of human development?

Home-based work – at the end of the subcontracting chain – presents a dual character. On the one hand, it may trigger an increase in household income and hence provide a means for capability expansion. On the other hand, it may work as a constraint on human development and the capabilities of household members, as it leads, for instance, to child labour and keeps children from school, and may thus perpetuate a system of inter-generational transfer of poverty.

14

Santosh Mehrotra and Mario Biggeri

On the positive side, home-based work offers several advantages to families. Above all, it offers employment and hence an opportunity to enhance and diversify their income; it also saves workers travel time and they can do other activities in addition to home-based work. For men, other such activities usually include another economic activity (e.g. farming in rural areas or periodic wage work), and for women it normally implies the performance of their reproductive and domestic role. Home-based workers can gain specific skills in producing goods at home, increasing the human capital available at household level and at local level. The work and the experience can eventually trigger the entrepreneurial capacities of some workers/subcontractors and the home-based activities can be the start-up of a small enterprise – though perhaps only for men, since women’s lives are more constrained by their role in social reproduction (Paugl and Tinker, 1997). Homeworkers, or industrial or agricultural outworkers, are mainly located within a local system of production but unconnected, except for the interaction with the intermediaries, from its larger structural organization (e.g. a cluster of SMEs) they usually stand at the end of a value chain. A value chain has a producer segment and a distribution segment (see Chapter 3 for examples of such value chains). The distribution segment has the final consumer at one end, and the producing ‘firm’ at the other. The production segment has the producer firm at one end, and the homeworker at the end of a chain. The producer firm is the node in the middle. The production part of the value-chain involves a number of contractors, sub-contractors, and at its end the home-based workers in certain sectors and tiny micro-enterprises in others. What connects this chain is a system of sub-contracting – which forms the link between the formal and the informal parts of the economy. Depending on the number of intermediaries in any given sub-contracting chain, the links between the homeworker and the lead firm for which they work are often obscure. In long complex chains of intermediaries, bargaining for higher wages is complicated by the distance between the homeworker and the lead firm and the ambiguity over who is responsible for providing higher wages. (ILO 2002a, p. 43) Subcontracting is often beneficial to the local and national economy – as hinted at above. It offers the prospects of strengthening inter-industry linkages; enhancing the entrepreneurial skills of subcontractors; taking industry to rural areas; increasing industrial competitiveness; spreading technology down to less skilled workers; and generating employment. It also favours networking and institutional changes. There are several reasons why subcontracting is practised in manufacturing, and why it may be growing in most countries.28 Subcontracting by firms to homeworkers offers employers several advantages. First, firms can recruit from a much larger area than would be the case if hiring was limited to areas which are within commuting distance. Second, they can hire workers in accordance with variations in demand. Third, they minimize the risk of unionization. Circumventing safety nets, labour rights and safety in the work

Empirical context and a theoretical framework 15 place all help employers to save costs. A characteristic of homeworkers is their ‘isolation’ and the lack of information. Work done by homeworkers may also be less costly to the employers, since costs such as rent, power, water, tools and so on, are the responsibility of the workers, not the employers. Finally, part of the entrepreneurial risk is passed to homeworkers. These advantages – at a micro-level for the worker household and at the macrolevel for the industry and local economy – can mask severe disadvantages for the homeworker. In conditions of excess supply of labour, piece-rates (the normal form of payment in homeworker) can be low, and thus homeworkers’ share in the value chains extremely low. In many cases piece-rates are low despite the fact that homeworkers, in some cases, are highly skilled. Work conditions can be very exploitative especially if there are few alternative income-earning opportunities in the area or if work is available only as bonded labour. The exploitation of the homeworkers by local employers is just a first step in exploitation through the global/domestic value chain. The problem for homeworkers arises from their low market access and their lack of contact with the final consumer. The lack of unionization can also be an important source of the vulnerability of homeworkers’ families; it is an issue we discuss later in Chapter 4. Furthermore, the activities are often dangerous in terms of health in the first place for homeworkers, and in the second instance, for other members of the household since the activity is done in the home. Children, as analysed in Chapter 5, are often engaged in homeworker activities to respond to low price per piece and in order to generate additional income for the household. This implies that many children do not go to school, while others, such as part time workers, register negative effects on educational attainment (Mehrotra and Biggeri, 2002a). We argue that the fixed costs of finding a job are drastically reduced for the children if a household is involved in a small family business (especially if residing in an area where the labour market is very slack). Therefore, children in homeworker households, given the same conditions, have a higher probability than other children to be in the ‘only working’ or ‘working-and-studying’ categories (see Chapter 5). This is because there are fixed costs associated with sending a child to work outside the home that would offset the returns to that work. The fact that labour is available at home reduces fixed costs in finding a job for children, changing the opportunity cost of the parents and thus the decision to send children to school and/or to work. The same can be argued for women (see Chapter 4). Indeed, on account of the lower fixed costs they tend to work more, and since they are at home and involved in household chores, they are subject to a double burden. In terms of social protection, such workers in the informal sector tend to be by far the most vulnerable. Vulnerability – the probability that a shock will result in a decline in well-being – is largely a function of a household’s asset endowment (physical and human capital) and insurance mechanisms. The lack of insurance mechanisms can have serious consequences for income growth at household, hence at a micro level. Thus, if children are seen as a substitute for old-age

16

Santosh Mehrotra and Mario Biggeri

security, measures to provide a pension after retirement, and some form of social security in the interim, would greatly reduce the perceived benefit stream from additional children. The logic here is similar to the effect of improved and affordable health services on the survival chances of children. If child mortality were to fall on account of more effective health services, the long-term effect on fertility decline would have similar benefits in terms of economic growth as social security/old age pensions would. The combined effect of social security and health expenditures would cause a behaviour change comparable to that experienced by the now industrialized countries in the 100 years between 1850 and 1950.29 That behaviour change underlay the demographic transition in Europe and North America over the same period. (We return to this issue of the combined effect of health/education expenditures on the one hand and social protection on the other in both the empirical analysis and policy implications in later chapters.) Homeworkers and the theory of SMEs clusters So far we have discussed the pros and cons for individual firms and individual households of the subcontracting and industrial outwork system. The homeworker households producing the same type of products are often concentrated in a specific geographical location. In homework, as for SMEs, there is a sort of natural clustering. We may say that quite often the homework clusters are within a SMEs cluster, the size of which can vary greatly. This may offer some opportunities for policy-makers – just as we saw above that homework and subcontracting both had benefits as well as costs. The challenge for policy is to ensure that the benefits outweigh the costs at the meso level – for the local economy, and thereby the national economy. At least in theory it is possible to suggest that it is possible for the benefits to outweigh the costs. Clustering is a natural strategy for firms to reduce transaction costs and to capture positive externalities and synergies. This facilitates the clustering of homeworkers. Indeed, homeworkers are ‘naturally’ clustered because intermediaries and principal firms need to reduce the transaction costs connected to subcontracting, throughout the input and raw material distribution, the output collection, the contract enforcement and the dissemination of information (when no other means of communication is available). Moreover, specialization in one type of product improves the skills and the productivity of the local labour force. Repeated transactions reinforce the networking and the social capital. This can become one of the keys for development.30 Clustering – according to Small and Medium Enterprise Theory – is one of the conditions to develop a local system of production based on SMEs. As the literature suggests, a cluster of enterprises can have different starting points (UNCTAD 1994; Mead and Liedholm, 1998; Schmitz and Nadvi, 1999). Clusters often arise spontaneously, and are connected to lower transaction costs and positive external economies. For instance a cluster can be generated by an agglomeration of traditional artisan activities in a specific sector and location, or by the presence of a larger enterprise that subcontracts part of the production stages to smaller enterprises.

Empirical context and a theoretical framework 17 Collective efficiency is a term coined to identify the results of the key relationship among SMEs, a relationship of co-operation and competition (Schmitz, 1995). As the cluster develops, small and medium enterprises can develop with them through spill-over in the local system of production (Volpi, 2002). This development has a number of positive externalities, which pushes the process forward. The positive externalities are generated in the local system through super-specialization of enterprises and workers, specialization in services linked to the production process, capital goods production linked to products produced in the cluster, equipment sharing, and so on. In a cluster, the skills acquired by workers and the specialization of enterprises in the production process are important for the success of a local economic system.31As we are going to discuss in Chapter 11, clustering of SMEs facilitates and is facilitated by collective actions at enterprise level with the formation of producer associations. Collective action at the local level – through associations – can focus the attention of the central government on the cluster in terms of social, trade and industrial policies. As SME theory emphasizes, the development of a local system through SMEs needs not only collective action, but also the support of local and central governments. The local policy makers’ intervention is fundamental to upgrading the human capital and infrastructure that are external to a single enterprise. The fact is that these crucial factors are under-supplied by the enterprises since the single enterprise tends to under-invest, for instance, in managerial, administrative and marketing skills and in employees’ skill improvement and training. But the existence of several firms producing the same product means the labour force in the area can move from one enterprise to another and infrastructure can be used by everyone. What is an external factor for the enterprise is an internal factor for the local collectivity and for economic development. Without public and collective action the development through SMEs is limited (Becattini, 1990) and the synergy between income growth, poverty reduction and the quality of human capital cannot work properly. For instance, a credit system can trigger innovation in equipment and thus productivity. We believe that to improve the homeworkers’ household conditions and, at the same time, to develop homeworker activities, similar public interventions are needed. What happens to wages and employment in a cluster as opposed to non-clusters is an important empirical question. Sandee (2002) provides evidence from an Indonesian furniture cluster in rural Java, that fed both domestic and export markets, where employment levels rose from around 8,000 in 1989 to nearly 44,000 in 1998. Employment growth is particularly evident from the relatively mature clusters. Thus, Schmitz (1998) describing the footwear export cluster of Sinos Valley, Brazil, reports employment rising from 153,000 in 1991 to just over 170,000 in 1996. Further dramatic evidence comes from Bair and Gereffi (2001) who report employment levels increasing by 300 per cent in the 1990s in the blue jeans cluster of Torreon, Mexico. Employment levels in Torreon went from 12,000 in 1993 to 75,000 in 2000. A study of the Tiruppur cotton knitwear cluster in southern India, another ‘mature’ cluster, indicated employment doubling

18

Santosh Mehrotra and Mario Biggeri

during the 1980s from 20,000 to 40,000 and now to well over 200,000 (Cawthorne, 1995; Singh, 2003). Income growth for workers and small entrepreneurs is limited, but the issue is not just of wage levels and wage growth. Even if wage levels are not rising clearly in clusters, what is more critical is whether they are falling behind wage levels in non-clustered alternative employment that such wage workers could take on. Thus, the key point is relative wages. Both Schmitz (1995, 1999) and Nadvi (1999) argue that while wage levels were low within Brazil’s Sinos Valley shoe cluster and Pakistan’s Sialkot surgical instrument cluster, they were better than regional average wage levels. In Jepara, Indonesia, wage levels for skilled furniture craftsmen were, according to Sandee (2002, p. 197) ‘significantly higher than average provincial wage levels’. Similarly, Visser (1999), in his assessment of the counterfactual for the Gamarra garment cluster in Peru, also shows that wage levels in the cluster were above those in ‘similar’ non-clustered firms. He reports that average monthly pay per worker was ‘30 per cent higher in the cluster than elsewhere in the city’, although he also noted that workers in the cluster tended to work longer hours (Visser, 1999, p. 1559). Moreover, these differences in favour of clustered enterprises were most significant for the subgroup of very small firms. That is, for the very small firms, clustering advantages were especially critical. Visser’s findings from Peru are also backed up by more recent analysis undertaken by Nadvi and Barrientos (2004) on Italian data for employment and income growth in industrial district and non-district settings. On the question whether clustering leads to more rapidly increasing employment and better and faster rising wages, the Italian data is good for two simple reasons. First, the Italian experience has driven much of the research in the developing world. Second, Italy is one of the few countries where wage and employment data is available at the level of clusters. The Italian evidence shows is that although manufacturing employment in the Italian clusters as a whole declined by 2.2 per cent during the 1990s, manufacturing employment levels fell more sharply in noncluster settings (of around 10 per cent). Furthermore, salaries, for both white- and blue-collar workers, were higher in cluster settings than outside clusters. Salary gains, again for both white- and blue-collar workers, engaged in similar activities were greater during the latter part of the 1990s in clusters. The Italian data supports the view that clusters can generate improved incomes and employment and point to a ‘high road’ growth trajectory (Pyke and Sengenberger, 1992). SME theory on clustering indicates that a local system can have two paths of development: a low road or a high road (Pyke et al., 1990; Pyke, 1992; Pyke and Senberger, 1992). The low road is positive but not dynamic, due to the limited interaction and specialization (especially vertical) among enterprises in the local system. In the low road there is limited action by local government, there are few institutional changes and little co-operation among enterprises. In the high road different actors of the local system co-operate and compete for the efficiency of the system, and for its innovative and competitive evolution. There is an involvement of associations of producers, associations of labourers, of higher educational

Empirical context and a theoretical framework 19 institutions and local governments. Thus, in rural and urban Italy, home-based household activities played an active role in local development, and in many parts of the country small activities were able to take the high road.32 In some areas small manufacturing enterprises and home based workers related to them led to the emergence of industrial clusters, and in a few cases, due to the right institutional and non-institutional conditions, these matured into industrial districts of SMEs which contributed to the ‘Italian Miracle’. The areas are together called the ‘Third Italy’ that is, not the North, where large industries are concentrated, and not the South, which is predominantly agricultural. ‘Third Italy’ is characterized by a large number of clusters – mainly in north-east Italy (Veneto) and central Italy (Emilia-Romagna, Toscana, Marche, Umbria). It is often believed that industrial districts result from an entirely spontaneous process of economic development (localized external economies). On the basis of in-depth study of several Italian industrial districts it was found that the competitiveness and the dynamism of industrial districts’ firms are dependent upon social co-operation as discussed in detail in Chapter 11. The human development of cluster’s stakeholders is a necessary condition for the high road development. Furthermore, as some development economics pioneers underline, the economic growth of a country is not only strongly related to the increase in productivity33 – which is given by the passage of labour force from low productivity sectors to higher productivity sectors and by the increase in productivity in the same ‘sectors’ – but also by the other side of the coin that is, the demand side (also within the cluster) (Volpi, 2002, p. 293).34 The key to the high road to development of a local system is the continuous upgrading of human capital, and the triggering of institutional arrangements and networks at local level, as well as the capability to undertake collective action for the socio-economic development of the area. Unfortunately, a third road, that we may call a ‘dirt road’, is perhaps currently in existence in many developing countries.35 This is the case of a cluster that involves industrial outworkers without giving the workers any social protection (as in the case of countries under study in this chapter). The development of homeworker clusters is connected to the growing phenomena of subcontracting in a national and/or international value chain. Without public and collective action, this could often turn out to be the lowest way, a way in which the workers (‘reserve army of labour’, in a Marxian sense) are at the lowest level of social protection accompanied by exploitation. Ideally, a cluster of homeworker households can evolve within and with a local system of development (see Chapter 11). For instance, often there is potential for upward mobility: of the homeworkers into becoming subcontractors, and subcontractors evolving into entrepreneurs. Given their role in social reproduction, it is less likely that women would be able to graduate into subcontractors but at least the male household members of those engaged in home-based work could. The possibility to evolve from this ‘dirt road’ to the low road exists, but without joint action (collective actions by the exploited workers in their collective interest) and the support of public intervention this evolution is almost impossible.

20

Santosh Mehrotra and Mario Biggeri

However, what is the evidence on joint action in clusters in developing countries? Again, Nadvi and Barrientos (2004) point out that with the exception of some of the Indonesian findings, the general observations from studies of incipient clusters suggest that beyond a division of labour within the cluster (more pronounced in sectors like garments where such divisibility is feasible), and some backward and forward vertical linkages, horizontal collaboration between enterprises and at the cluster level is rare (Altenburg and Meyer-Stamer, 1999; Schmitz and Nadvi, 1999). Low barriers to entry, limited skill bases, extensive local competition, low trust within clusters despite the often strong presence of common social identities, can result in what Altenburg and Meyer-Stamer (1999) refer to as ‘poor contract enforcement’. This limits the potential gains that clustering could bring about, in terms of growth and pro-poor impacts. Stronger evidence on cluster-wide institutional joint action was observed by Nadvi (1999), in the context of the Sialkot surgical instrument cluster in Pakistan, and by Kennedy (1999) for the tanning cluster of the Palar Valley in Tamil Nadu, India. Compliance with global quality assurance standards, a necessary requirement for exports to leading global markets by the Sialkot cluster, came about through the catalytic role of the local trade association in channelling new know-how on quality management practices to the cluster. Through this process, Nadvi found that the vast majority of SMEs in the cluster could comply with international standards over a relatively short period of time. Had the association not taken on this function, most small firms would have closed given that the United States and the European Union (EU) accounted for over 90 per cent of the cluster’s sales.36 A further example of how clusters can promote collective responses to external threats comes from the response of Sialkot’s export-oriented sports goods cluster regarding the presence of child labour in manufacturing units within the cluster (Nadvi, 2003). Faced with the loss of key export markets, local firms through collective institutions (such as the local Chamber of Commerce) entered into an agreement with international bodies, including the International Labour Organization (ILO) and United Nations Children’s Fund (UNICEF), as well as leading global buyers. This resulted in an ILO monitoring programme of the cluster and a social development strategy, based on education and income generation, for child workers and their households. Thus, local joint action resulted in direct gains for the cluster, employment and cluster exports rose, while immediate poverty concerns for many of the more vulnerable members of the cluster’s labour force began to be addressed (Dei Ottati, 2002; Nadvi, 2003). Joint action is neither an obvious outcome of clusters, nor is it easy to achieve. The evidence that emerges from cluster studies suggests that joint action is less common in incipient clusters than it is in more mature clusters. Even in mature clusters, joint action is far from uniform. In the face of global pressures, ties with external actors begin to supersede local linkages (Brazil’s Sinos Valley is an example, see Schmitz, 1999). But in many cases, local cooperation can assist local small enterprises access markets, overcome constraints and confront vulnerabilities that they face in local and global markets. Often where cooperation does occur, it is strengthened by local social capital, common ties of community and identity that can foster cooperation and generate trust.

Empirical context and a theoretical framework 21 From this meso-level analysis of clusters, we will now move to a discussion of capabilities at an individual level . Although the development of a part of the industrial system through SMEs in a local system of development could be a very important step for poverty reduction since it creates new job opportunities in a higher productivity sector, and may raise wages, it may not become the entry point for development of human capabilities. From the viewpoint of human capability, the achievement of the positive externalities of clusters is dependent on at least three other fundamental factors: environmental sustainability, social protection and worker organizations. In the medium term there is no sense in income-generating activities producing many negative externalities dangerous for health and causing environmental degradation, since they will negatively influence the system of production. Social protection and basic social services are the platform for long-term development of the local system. Workers’ organization or ‘voice’ is the key to these institutional changes.

1.3

The capabilities of homeworkers – a micro-level analysis of homeworker households

The main practical reason for linking the meso-level analysis in the preceding section to the capabilities of individual workers (and non-workers) is to demonstrate the interaction between two type of synergies. The synergy between two sets of interventions can significantly enhance human capabilities and promote economic growth (Taylor et al., 1997; Mehrotra, 2004; Mehrotra and Delamonica, 2007). The first synergy is between interventions within basic social services (bss) – basic education, basic health, water and sanitation, and nutrition which result in achieved functionings. Interventions in health, nutrition, water and sanitation, fertility control and education complement each other. This increases the impact of any one from investments in any other. For example, the improved health status of a child improves her ability to learn, just as improved nutritional status does. Similarly, reduced family size improves the chances that a poor family will be able to afford education for all the children rather than merely the boys, and so on. In turn, educational inputs have an impact on all types of human capabilities (or human development outcomes). For instance, the outcome of the first synergy is shown by the positive slope (Figure 1.1) of the relationship between women’s achievements, the (Gender Development Index,37) and the proportion of children (aged between 10 and 14) not working. The reduction in working children in households can have the effect of stemming the inter-generational transfer of poverty in such households. The second synergy is between income increase, its better dispersal, and health and education outcomes.38 A continuous improvement in health and education indicators may be unachievable in the absence of income growth, just as sustained growth would be impossible without at least a minimally educated and healthy workforce. At the same time, at the macro-economic level it is critical to promote economic growth of the kind that improves the income distribution in favour of the poorest. This is the essence of the second synergy – the interaction among income-poverty reduction, the quality of human functionings at the aggregate societal level, and economic growth.39

22

Santosh Mehrotra and Mario Biggeri

Proportion of children (10–14) not working

1.00 0.95 y = –1.0695x2 + 2.0436x + 0.0332 R2 = 0.8314

0.90 0.85 0.80

y = 0.6607x + 0.4415 R2 = 0.774

0.75 0.70 0.65 0.60 0.55 0.50

1.00

0.90

0.80

0.70

0.60

0.50

0.40

0.30

0.20

0.45

Gender Development Index

Figure 1.1 Relationship between women’s achievements (Gender Development Index) and proportion of children (aged 10–14) not working. Sources: UNDP (2003), WDI (2003). Note Data for 143 countries, 2001 data for child labour.

The notion of synergy can be succinctly expressed as the enhanced impact changes in an independent variable has on the growth rate of a dependent variable, given the presence of a third variable. This leads to important, and often overlooked, inter-related effects in terms of policy. The impact of a policy intervention (e.g. land reform which directly reduces income poverty) on another variable (say, economic growth) crucially depends on the level of a third variable (e.g. good health and education status). Both these synergies operate at the macro, as well as the micro levels (i.e. that of the individual). At household level the policy implications of the effects of the two synergies in the context of the poor homeworkers is well captured in Figures 1.2 and 1.3. Poverty (capability deprivation) is passed on from one generation to future ones without external intervention to foster the two synergies. For instance, from generation to generation children in poor families are merely ‘born to work’ (Figure 1.2). As we will show later, very often in homeworker households the children are also working, quite often at the expense of their schooling. A household in that position needs help in improving its human capital or endowment (HE) and its

Little public investment in child health, education, family planning

Working children, poor home and working environment

1st gen. hh HD level = = Born to work 2nd gen. hh HD level

New households formed, same HD level, same expectations of children (same fertility), voiceless (no participation)

No external factors influence

Children become illiterate adults

Low job income, few assets, same opportunities

Figure 1.2 Household human development cycle without policy interventions (collective action or interventions among which public policy interventions): vicious circle from generation to generation.

Policy interventions for Synergy 1 Child education Family planning Health, nutrition, water and sanitation Young-adult literacy

Investment in children, health and education, household sanitation

No external factors influence Increase in the number of studying / studying and working children Better house and working environment

2nd gen. hh HD level > 1st gen. hh HD level New household formed, higher HD, behaviour change, change in expectations of children, increase in participation

+ Children become literate adults

Higher work income, some assets, more opportunities, more incentives, different allocation of resources

Policy interventions for Synergy 2 Infrastructure and services human capital training, services, credit, savings promotion, information dissemination, demonstration centres, collective action to reduce exploitation, old age pension, child care, social protection.

Figure 1.3 Household human development cycle with policy interventions: virtuous circle from generation to generation.

24

Santosh Mehrotra and Mario Biggeri +

Household human development level

High level of HD

+

Set of achieved functionings Positive external factors

Negative external factors

–

Low achieved functionings: multidimensional poverty

Figure 1.4 Factors influencing household’s human development level.

economic endowment (EE) (assets and income), and for this reason policy interventions are necessary. In Figure 1.3, with public action on several fronts simultaneously, the effect of synergies is captured. On the one hand, exogenous action to improve child health and education levels, and on the other hand, action to increase economic opportunities and strengthen the bargaining power of the homeworker, serve to turn the vicious cycle into a virtuous circle. And the bargaining position of homeworkers is facilitated by collective action and bottom-up institutional changes which reinforce the second synergy. We try to represent the combined effect of the dual synergies in Figure 1.4. Suppose that each household is positioned on a graduated scale in terms of their level of human development – a reasonably realistic assumption. The household’s human development level, or achieved functionings, relative to other households depends on the family’s endowments such as the level of education, health status, the resources available (income, assets), and so on; this capability set of the household is constructed generation to generation. It is the socio-economic history of the household that has determined that position and thus the vulnerability of the household, demonstrated in the working child.40 The socio-economic record of the household is determined by endogenous factors, which contributed to the human development level of the family. But there are other factors that can influence dramatically its position and are exogenous to the household. These external conversion factors may be policy-related (e.g. political commitment to the poor, economic policies), structural (e.g. labour force surplus, resource endowment, institutions, infrastructure, market structure including an exploitation chain), or events of an entirely random nature for the household (e.g. economic shocks,

Empirical context and a theoretical framework 25 natural disasters, civil wars, ethnic conflicts). Policies to promote human development and reduce child labour need to take several (not all) of these factors into consideration. Human development is a cumulative effort not just for the household but for the collectivity at local and national level (Sen, 2000). Some external factors or shocks which influence the household’s position, are exogenous to the household but are endogenous to the collectivity and thus can be influenced by collective action. For the second synergy, economic growth promotes human development only if the poor benefit from it; if the poor benefit, in terms of income increases and improved level of functionings, it would put into motion a virtuous circle of growth and equity. Therefore, in order to trigger the second synergy (and reduce child labour), a real increase in the income of the poor is a necessary condition. However, the growth of per capita income on average is no guarantee of human development, since it may result from enclave type development or merely natural resource extraction and export. From this follows that policies – to be relevant for poverty and child labour reduction – have to promote those sectors where the income of the poor majority comes from. In other words, policies have to focus also on the informal sector (and within that the homeworker sub-sector) – since the vast majority of the poor are engaged in informal activities (especially, but not only, if one includes those self-employed in agriculture). In any case, the fact that most of the labour force is in the informal economy has clear implications for the quality of economic growth (see also Cornia, 2006). The human development level of a household (HDhh) has two main dimensions which are closely related to each other: one, human endowments (HE) or human capital (e.g. education, health, both goals and means, skills, talent), and two, economic endowments (EE) (or instruments), such as the ownership of assets which can enable the household to generate incomes. Besides, there are the external conversion factors (e.g. infrastructure, as well as public policies and institutions), noted above, exogenous to the household which impact on both HE and EE, ct. Finally, there are also exogenous shocks vt, that can affect (positively or negatively) both. In other words, HDhht  f [(HEt
EEt), ct, vt] where time t  1, . . . , n. Changes in the household’s HD will thus have a path that can be highly volatile, and marked by uncertainty. A household, starting from its initial level of human development can experience movements in different directions, or maintain the same level of HD, depending on the HE and EE. In a graph with HE on the x-axis and EE on the y-axis, a right upward movement signifies an improvement in HD while a shock may reduce HD drastically. For instance, consider a household A0 (first generation) which experiences a movement upward in HD level, as it accumulates human capital or acquires an asset (an additional source of income) or both. Therefore the second generation starts from point A1, at a higher HD level. This underlines the fact that the HD

26

Santosh Mehrotra and Mario Biggeri

level is improved by HE and EE which work together to built the household HD. The variables x and m are the factors/determinants of HE and EE respectively. So, HEt  g (xt, EEt-1) and EEt  l (mt, HEt-1), thus HDhht  f [HE (g (xt, EEt-1)); EEt (l (mt, HEt-1)), ct, vt] Without external intervention the household will remain at t
1 at best in a status quo A0, as also the future generations of the household. In other words, with limited human capital, and few if any assets , the poor usually only have their raw labour power to offer in the market to make a living. Under the circumstances, without external intervention (collective and/or public action) a poor household may remain at best in a status quo of human capabilities and there will be an inter-generational transfer of poverty. A household needs interventions – public policy and collective action – for improving HE and EE. In the third part of the book we are going to analyse in detail the effect of the two synergies on the local system of development (Chapter 11) and on social protection (Chapter 12).

1.4

Final remarks

In other words, in economies with a high share of the labour force in informal sector employment, the triggering of the second synergy may well require a special policy package to promote not just the HE of the household, but also the EE. Given that a high proportion of those engaged in the informal economy are either self-employed in household enterprises, or dependent workers in a value chain of enterprises small and large, triggering the second synergy would involve active support for micro-enterprises. In South East Asia this task will be easier on account of the relatively equal distribution of human capital in the population; in South Asia, with much worse health and education indicators, the effectiveness of the strategy will be compromised by the unequal distribution of human capital through the population.41 But the prediction of cluster theory is that, given the right conditions in terms of human capital, it is possible for a dynamic to be set in motion that creates new, productive jobs, thereby increasing incomes for those in the lowest income deciles. To encourage this dynamic should be the aim of a policy directed at SMEs plus micro-enterprises. However, this dynamic could only be set in motion – and the high road in terms of cluster theory be achieved – if there were better social protection of those employed in the informal economy. Given the significance of those employed in the informal economy in all developing regions, there is a case for better social protection in order to promote economic growth – not just a moral case grounded in social justice. In other words, the integration of economic and social policy is dependent upon social protection for these vulnerable workers (Chapter 11).

Empirical context and a theoretical framework 27 Almost all countries today have some policies for SMEs, since they are seen as engines of job creation. However, Japan, Korea and Taiwan made particular efforts to support SMEs, while at the same time supporting large firms using capital-intensive techniques.42 Taiwan vigorously promoted small-scale industries, through establishment of industrial parks and districts with financial and technical support, as well as agriculture and rural industries. Since 1978 China has adopted a similar strategy, involving the dramatic growth of TVEs, with outcomes characterized by remarkable growth and poverty reduction.43 This book suggests that this strategy may need to be supplemented now in the rest of developing countries – especially South East as well as South Asia – with a ‘SME plus micro-enterprises’ strategy. The latter would take into account the changed realities of these economies since the East Asian trio (Japan, Korea, Taiwan) managed their evolution. In most developing countries, employment in manufacturing has grown more slowly than output. The employment elasticity of manufacturing in developing countries has been consistently low. This was less the case in South East Asia (Indonesia, Thailand, Philippines) than in South Asia, since the former adopted a strategy reliant more upon the exports of labour-intensive manufactures. In both, without explicitly incorporating micro-enterprises, including the chain of contracting under which a significant share of manufacturing currently takes place in the emergent market economies of Asia, a mere SMEoriented strategy may fail in the changed environment faced by these countries.

Notes 1 Breman (1996) and Chen et al. (2002), prefer to use the term ‘informal economy’ for the informal sector for the following reasons. First, the informal and formal parts of the economy are inter-linked (as we discuss in this chapter and in Chapter 3), and referring to them as two sectors can be misleading. Second, it is unhelpful to use the term ‘sector’ as a classificatory device for both the formality/informality of work status and for industry groups or commodity chains – and sector is better reserved for use with the latter. For work status, workers all over the world increasingly face degrees of informality (as we discuss further in Chapter 3) (ILO, 2004a). However, we argue (see also note 6) that the definition of informal sector can be rather useful. The informal sector is a sub-set of the informal economy. We may define informal sector as composed by the whole set of those activities not governed by legal institutions and laws in force, and/or those activities organized in ways different from the ones characterizing the economic activities typical of a ‘modern’ economy (Biggeri, 2004). The informal sector (which differs form illegal activities) covers ‘everything’ from small-scale manufacturing and repair, to trade, transport and construction and services. In urban areas, informal activities range from street vendors to small manufacturing entities and, in rural areas, small enterprises engaged in the production, transformation and sale of farm and non-farm (e.g. handicrafts and services) products would be considered as informal sector activities (Bangasser, 2000). Following this definition, the non-modern agricultural sector (subsistence and semi-subsistence or small household farming) is also considered part of the informal sector (see also Ray, 1998, p. 348; United Nations, 1996). This gives a more realistic perspective of informal activities also and of the structure of developing, especially low-income economies. 2 From the policy point of view, another main shortcoming of the dualistic models is probably the insufficient attention to institutional and some non-institutional dimensions overlooking their importance in economic development and in particular in labour

28

3 4 5

6

7 8

9

10 11 12 13

Santosh Mehrotra and Mario Biggeri transfer (Biggeri, 2004). Thus, the labour transfer process was hypothesized as characterized by a perfect mobility of labour from the traditional to the modern sector which does not reflect reality. From 62 million in 1900 in Latin America to 480 million in 1995; from 118 million in Africa to 732 million; and from 937 million in Asia to 3458 million (Ray, 1998). In fact, Walmart is a very large part of the export success of China, accounting for a signficant share of its foreign-enterprise driven export boom within this first decade of the twenty-first century. One of the new characteristics of the System of National Accounts (SNA) 1993, approved by the UN Economic and Social Council (at the recommendation of the 15th International Conference of Labour Statisticians and the UN Statistical Commission), is the recommendation to introduce, where relevant, a sub-classification of the households sector, including a distinction between the formal and informal sectors. This would make it possible to quantify the contribution of the informal sector to the national economy (Hussmanns, 1997). The informal sector was considered a subsector of the SNA institutional sector ‘household’. In accordance with the SNA 1993, household enterprises – as contrasted with corporations and quasi-corporations – are defined as production units which are not constituted as separate legal entities independently of the households or household members that own them, and for which no complete set of accounts (including balance sheets of assets and liabilities) are available which would permit a clear distinction of the production activities of the enterprises from the other activities of their owners. Within the household enterprises a distinction was made between enterprises of employers and own-account enterprises. But the ICLS also recognized that an enterprise-based definition would not capture all dimensions of informal employment. The conference recommended that further work was needed on the employment-based dimensions of informality. Chen et al. (2002) point out (see note 1) that an enterprise-based definition of the informal economy might be useful in statistical terms (in calculating national accounts), while an employmentbased approach is useful in policy terms (who to target and how to finance social protection policies). However we think that the former definition can be very useful as well especially to determine policy for local system of development. See also Ferran (1998). This is based on Charmes (1998, 2000, 2001). For Asia he was using data only for India, Indonesia, Pakistan, Philippines, Thailand, Iran and Turkey. Another ratio that tells a powerful story is the trend over three decades in selfemployment in non-agricultural employment by region. In the world as a whole, that share rose from 22.5 per cent in the 1970s to 27.3 per cent (1980s) to 30.9 per cent (1990s). It was pretty much stable in developed regions (11 per cent, 12.9 per cent, and 11.5 per cent in the 1970s, 1980s and 1990s respectively); but it has exploded in the developing regions, especially in the 1990s: Africa 27 per cent, 46.2 per cent and 47.7 per cent; Latin America 28 per cent, 28.8 per cent and 41.8 per cent and Asia 28.1 per cent. 26.2 per cent and 32.7 per cent (ILO, 2004a). In Latin America, 58 per cent of women in comparison to 48 per cent of men nonagricultural workers are informally employed, and in sub-Saharan Africa 84 per cent of women against 63 per cent of men. In Asia, the share of women and men non-agricultural workers in informal employment is roughly equivalent. See also (Gammage and Mehra, 1999). Apparently 30–90 per cent street vendors are women (ILO, 2002a). It does not refer to either unpaid housework or paid domestic work. See also Pedrero (1998). This is due to information technology and telecommunication up-grading which give rise home-based work opportunities (Felstead and Jewson, 2000) especially in developed countries but also in some developing countries. In fact, in the mid-1990s, homeworkers – persons who work more than half of their working hours at or from their home – represented between 4 and 7 per cent of the total workforce in 8 out of 12 European countries (Charmes, 2002).

Empirical context and a theoretical framework 29 14 Lora Jo Foo (2002) from Sweatshop Watch, California, noted that 61 per cent of the LA garment factories violate minimum wage and overtime laws and it was estimated in 1998 that employers owed workers $73 million in back wages to garment workers. 15 See Boris and Prugl (1996) and Prugl (1999) for an interesting discussion. 16 A growing body of literature has indicated this since the late 1980s. The first set of studies were commissioned by the ILO for Latin America (see note 19). 17 The Commission of the European Community recommended ratification on 27 May 1998. 18 According to the estimates of Charmes (ILO, 2002a, p. 19) (period 1994/2000) informal employment in non-agricultural employment in India is 83 per cent (86 female and 83 male), 78 per cent (77 female and 78 male) in Indonesia, 72 per cent (73 female and 71 male) in the Philippines and 51 per cent (54 female and 59 male) in Thailand. 19 The phenomenon is widespread in the middle-income countries of Latin America. See for example, a set of studies for the ILO in eight countries – Argentina, Brazil, Chile, Guatemala, Costa Rica, El Salvador, Paraguay and Peru (Heikel, 2000; Henriquez et al., 2000; Jelin et al., 2000; Lavinas et al., 2000; Tomei, 2000; Verdera, 2000). 20 In only a few African economies subcontracting from formal to informal sector enterprises is found widely (McCormick, 1999; Xaba et al., 2002). 21 Different forms of subcontracting can be identified: component subcontracting, activity subcontracting, assembly subcontracting, product subcontracting. Then we may have vertical subcontracting (i.e. inter-firm relationship) between two firms operating at two different stages of the chain and horizontal subcontracting between two firms operating at the same stage of the production chain (Unni et al., 1999; Unni, 2000). 22 The costs of international phone calls fell strongly. However, the real change is now occurring with internet use. Internet-based software systems made real-time information exchange possible, and enabled just-in-time production and delivery co-ordination between producers and retailers on an international scale (OXFAM, 2004a, p. 33). Cheaper transport with sea freight costs falling almost 70 per cent between the early 1980s and the mid-1990s, and with significant growth in airfreight services, the delivery costs for distant producers have fallen dramatically (World Bank, 2003). 23 For example, tariff reductions: between 1980 and 1998, average tariffs on manufactured products fell from 10 per cent to 5 per cent in industrial countries and from around 25 per cent to 13 per cent in developing countries, cutting the cost of trade in goods. Also, foreign investment incentives: export processing zones proliferated in the 1990s in developing countries, offering tax exemptions and investment allowances in order to attract local and foreign investors to produce goods for export (World Bank, 2003a). 24

For any single country, this advice from international financial institutions may seem necessary to stay competitive with other cheap and ‘flexible’ countries. But the advice to ‘flexibilise’ has been given systematically to many developing countries. The result? Countries are still in tough competition over labour costs, but all at lower levels of labour protection. Workers’ rights are simply being traded away. (OXFAM, 2004a, p. 43)

25 In Indonesia there was a sharp reduction in the early 1980s. In 1980 exports accounted for 34 per cent of GDP. Due to the oil crisis, this share was reduced to 29 per cent in 1981 and to 25 per cent in 1982. 26 It is noticeable that the export share of GDP differs sharply between the middle-income countries of South East Asia in our sample, and the low-income countries of South Asia where we have studied the phenomenon of homework. As we saw above, the former have a relatively high share of exports in GDP. By contrast both Pakistan (15 per cent) and India (11 per cent) have much lower shares. However, in export-oriented product groups the phenomenon of subcontracting is likely to have grown in South Asia

30

27 28

29

30

31

32 33

34

Santosh Mehrotra and Mario Biggeri as well. In addition, the need to cut costs has resulted in the growth of subcontracted manufacturing in South Asia for the domestic market. Nayyar defines globalization ‘as a process associated with increasing economic openness, growing economic interdependence and deepening economic integration in the world economy’ (Nayyar, 2003, p. 65). For India, a UNIFEM study suggests that subcontracting has been on the increase since the late 1970s. It notes that it was not significant around 1970, but from 1978 it became widespread among large factories with a share of 21 per cent (Bajaj, 2000; UNIFEM, 2000a). See also (Kurringa, 1999). Cigno and Rosati (1992) find that more than three-fourths of the fertility decline in Italy between 1930 and 1985 can be attributed to the extension of pension coverage. They also find a close relationship between the two variables in Germany, the United Kingdom and the United States (Cigno and Rosati, 1996). For similar evidence for developing countries, see Chernichovsky (1982), Nugent and Gillaspy (1983), Entwisle and Winegarden (1984) and Jensen (1990). However, one has to be careful since such specialization can also perpetuate a lowlevel equilibrium trap. For instance, when certain types of specialization becomes a form of a ‘caste’ system or be based on debt bondage, it can be a serious constraint to human development (Harris-White, 2003). Hence, policy at local level has to ensure that national level policy-makers are not oblivious to such local impacts. Marshall (1920) discussed the relevance of clusters of micro, small and medium size enterprises in the industrialization process of a country. Three economic concepts underlying this process of industrialization (parallel to those for medium–large size enterprises) are the effects of externalities (Volpi, 2002), joint actions and of economies of scale in the local system of production. Clusters can evolve into more complex structures as industrial districts (see Chapter 11). For instance, consider the case of Carpi, a famous knitwear industrial district in Italy. The roots of its development lie in the skills acquired by homeworkers in the sector of straw weaving. See Chapter 11 for details. As Reinert (2003) emphasizes, economic development and human development can occur at the same time only if a country is able to transmit part of the gain in productivity to an increase in real wages and in social protection of workers. This turns economic growth into human development, and workers became part of the ‘virtuous circle of development’ by influencing positively the demand for products; in other words, as if a country is following a ‘Marshall Plan’ of development. Reinert underlines that Japan, Italy and Korea and most of developed countries followed this route in their development process. Further two other important characteristics in the industrialisation process are: (i) to liberalize (open completely) the economy only after a certain level (that enables the country to benefit from free trade) of development occurred. This is how most of the industrialized countries have built their industrial success (Chang, 2002); (ii) the capacity to select and to develop Schumpeterian (innovative) sectors within the economy (rather than the Malthusian) (Reinert, 2003).

35 However, in many developing countries (e.g. China, and as we discuss below, parts of East Asia) the high road has also been followed. On China, see Bellandi and Biggeri (2005), Di Tommaso and Bellandi (2006). 36 Similarly, in the Palar Valley, pressures to meet environmental standards in leather processing called for the setting up of common effluent treatment plants. As Kennedy (1999) notes, local tanneries had to cooperate for survival, forming common plants through collaborative arrangements, monitoring problems of free riding in the management of treatment plants. As a result of this local joint action, a number of tanneries have expanded while the common treatment plants have emerged as key local institutions for collective organization.

Empirical context and a theoretical framework 31 37 The Human development index measures the average achievement in a country in three basic dimensions of human development: health measured by life expectancy at birth; knowledge, by adult literacy rate (two-thirds weight), and the combined primary, secondary and tertiary gross enrolment ratio (with one-third weight); and income, as measured by GDP per capita (PPP US$). The Gender Development Index adjusts the average achievement to reflect the inequalities between men and women for these same three dimensions. 38 Note that the first synergy is really a sub-set of the second synergy, but it is critical to distinguish between the two conceptually, since in reality, the first set of synergies in (being a goal per se) can be set in motion without necessarily the second set of synergies actually being in place. However, in the long run, there is a strong risk of the first set of synergies ‘running out of steam’ in the absence of the second set of synergies. 39 For a detailed elaboration of this dual synergy model, see Mehrotra and Delamonica (2007). For the capability approach see Sen (1999), Nussbaum (2000), Alkire (2002) and Robeyns (2005). 40 The vulnerability is connected to poverty and could be chronic or transient; reducing the vulnerability will reduce the probability that the child works (Lieten, 2004a,b). 41 That is why we argue in later chapters, especially Chapter 4, for the need for rapidly universalising access to quality schooling and health care services in South Asia – in much the same way that East Asia, and many (if not all) of the selected South East Asian countries have succeeded in doing. 42 It should be noted, of course, that the protection of the small-holder agriculture and small businesses in general in these economies would be considered by neo-liberals to be market-distorting. Korea protects the small-holder farms, not only by trade protection but by restricting the size of individual farms, banning absentee ownership and numerous other measures that is, in addition to the initial conditions of equality in assets created by land reform. 43 Chinese development is strongly related to the two synergies at work which brought economic growth and human development since the 1980s. Since 1978, the people under the poverty line fell from 250 million (almost all rural dwellers lacking adequate food and clothing) to 37 million (34 million of which are in rural areas) in 1999. In the same period the child labour (10–14 years old) incidence fell from 30.5 per cent in 1980 to 8.6 per cent in 1999. The reduction of poverty and child labour were attained mainly in the early 1980s when the rural reforms became the cornerstone of the whole development strategy. There were three main reasons for this success. The first was that the socio-economic bases for long term development were laid and all the elements for the first synergy were present. Indeed, this included basic industry and general infrastructure, but also improved irrigation schemes and extension services in agriculture. Health care and education systems were accessible and affordable (especially in rural areas) to all citizens. The second reason was the reforms and the institutional changes brought with them a gradual transition to the socialist market system. The institutional changes gave the right incentives at both levels – micro (household level) and macro level (through decentralization) – and promoted rapid economic growth. The last reason was the reforms in rural areas: the growth strategy was targeted to sectors which were the income source of the majority of the poor that is, agriculture and non-farm activities. Agriculture became – together with the rural SMEs or TVEs (mainly collective at that time) – the leading sector of Chinese socio-economic development and economic growth (Biggeri et al., 1999). The bottom-up reforms which created the household responsibility system (and the dis-aggregation of the Commune system) were fundamental to Chinese farmer incentives and welfare improvement. Maintaining the right to land use was the most important way of preserving farmers’ entitlements and it remains an important safety net for rural dwellers.

2

Research methodology Santosh Mehrotra and Mario Biggeri

Although the informal sector accounts for the majority of employment in the non-agricultural sector in the developing world it is widely recognised in the literature that there is very limited information about it and even scarcer information on homework activities (Chen et al., 1999). This is why ad hoc surveys and several focus group discussions and case studies were carried out to examine the work and conditions of women in home-based economic activities and child labour in these activities. In fact, as far as we know, there is nothing in the homeworker literature of the 1990s that specifically addresses the issue of child labour in homeworker households.1 Considering the importance of homeworkers in Asia2 it was decided to focus on three countries in South East Asia (Indonesia, Philippines and Thailand) and two countries in South Asia (India and Pakistan) (see Figure 2.1 (map)). This chapter describes the research plan (Section 2.1), the research methodology and the survey design used in this study (Section 2.2) and profiles the sectors

Afghanistan Pakistan

Nepal Bhutan India

Bangladesh Burma Laos Thailand Cambodia

Philippines

Sri Lanka Malaysia Indonesia Papua New Guinea

Figure 2.1 Countries and locations of selected clusters. Note Small circles denote exact locations of survey site.

Research methodology 33 analysed by country (Section 2.3). In Annex 2.1 the questionnaire that guided data collection in these countries is presented; the questionnaire was not exactly the same in every country, as country researchers added some questions, and made other questions more country- or sector-specific. However, this questionnaire provided the core elements of the information sought from homeworker households.

2.1

Research plan

A National Advisory Committee was created in each country involved in the study. Each committee consisted of the following stakeholders: UNICEF country offices/representatives, ILO, UNIFEM, Government’s Ministry of Labour, NGOs and other civil non-profit associations. An academic research group for the field study and the report was selected in each country. Terms of references were prepared for the surveys (e.g. common questionnaire, survey design), the focus group discussions (FGDs), the tabulation of the results and for the national report contents. One international meeting of the researchers and stakeholders in December 2000 (in New Delhi) discussed the questionnaire and research design after a pilot test within each country had been carried out of the questionnaire and a second meeting of a similar group after the reports had been drafted, and before they were finalised in December 2001 (in Bangkok). Several meetings were organised including a national workshop in each country at the end of the research (Figure 2.2). More details about the advisory committees and the composition and activities of the research groups are reported in Part II of the book. Both quantitative and qualitative methods were adopted for each national study on subcontracted homework by women and children. A summary of the sample for data collected through the ad hoc surveys and on the number of FGDs and case studies are reported in Table 2.1. The data collected during the surveys – once inserted3 and processed4 – were used for the cross-country analysis in Chapters 4 and 5, as well as to test some hypotheses through econometric analysis.

2000

Research preparation Pilot

2001

2002

2003

Field work Data collection

First meeting

Data processing Delhi workshop

Data analysis and report writing Reports editing and finalisation Country workshop Bangkok conference

Figure 2.2 Time line of the research.

Volume’s writing, editing and finalisation

34

Santosh Mehrotra and Mario Biggeri

Table 2.1 Number of focus group discussions, case studies and households surveyed by country Country

Households surveyed

Total womena

Total childrenb

Number of FGDsc

Number of CS

India Pakistan Indonesia Philippines Thailand All

603 397 300 173 399 1872

1021 596 505 296 —d 2418e

801 1245 487 357 289 3179

7 8 6 6 7 34

9 8 9 8 11 45

Source: UNICEF survey. Notes a Women here refers to all females equal to, or greater than, 15 years old. b Children from 5 to 14 (i.e. children under five are not counted in this table). c Considering separately the one for women and the one for children. d Data were not comparable for Thailand (women equal to, or greater than, 18 years old). e Excluding Thailand. FGDs  focus group discussions; CS = case studies.

2.2

Research methodology: instruments and data collection

The surveys: the questionnaire, sample design and data collection The quantitative method involved an ad hoc household survey, based on a guideline questionnaire for homeworkers (and control group), common to all the countries and designed by us (Annex 2.1).5 The units of the statistical population surveyed are the households engaged in homework. This population is active in the informal economy and thus very often ‘invisible’ to official statistics. It would thus be impossible to prepare a list of all households engaged in homework (including in a given sector for the entire country). Also for this reason the design of the sampling had to follow a specific method taking into account the information already available. We had information a priori that the homeworker households involved in manufacturing are generally clustered on the basis of the type of goods being produced. A second characteristic, very important for the sample design, is the degree of socio-economic homogeneity of homeworker households which is very high within each sector/cluster. This emerges in the literature (Khattak and Sayeed, 2000) and also in our FGDs (e.g. in respect of income, size) (Arunotai et al., 2001; Khan et al., 2001; Oey-Gardiner et al., 2001; Rosario del Rosario 2001; Sudarshan et al., 2001). The homogeneity of the homeworker households (in terms of economic characteristics) within each cluster and in each location is very strong (even the non-homeworker households share similar economic characteristics, for example, on average they had income only a bit higher than homeworker households).6 The reduced variability diminishes the importance of the

Research methodology 35 dimension of the sample size and increases the importance of the selection method. Furthermore, according to previous studies in each homeworker household there is at least one woman working full time or part time. Indeed, the share of homeworkers’ households with a woman involved is often over 95 per cent in developing countries and women represent over 90 per cent of the adult homeworkers in developed countries (Oxfam, 2004b). For these reasons the design of the sampling had to follow a method taking into account the information already available. Data collection was based on a multistage sample model with three stages.7 The first was to choose a large grouping, easily identifiable as the statistical population of households engaged in homework. The secondary sampling units – the sector/clusters – were selected by national experts, while the third sampling units – the households – were selected randomly. The problems to be solved in the design of the sample at national or regional level were the identification of three/four manufacturing sectors involving homework and the choice of the specific clusters to be surveyed. Considering that there are hundreds (or even thousands) of clusters of homework activities in each country – even if a list of these was available (and this is not the case) – a pure random choice is not the right way to proceed. This is the reason for the small number of clusters selected for the ad hoc micro-surveys in each country. For this reason the best way to proceed is to ask expert ‘privileged observers’ to identify the sectors/clusters to be surveyed (Fabbris, 1990). The sector/cluster selection probability is connected to the ‘probability’ of the experts correctly identifying those specific sector/clusters for the survey. Further, in each country the researchers sought advice from more experts from different institutions (local agencies, NGOs, workers’ association, researchers, government authorities) representing different categories of interest in order to compensate for an eventual bias in ‘privileged observer’ selection. The dimension of the sample n was chosen in each country according to the confidence interval based on a priori information on economically active children between the age of 5 and 14, a level of confidence of 95 per cent and an error margin of 2 per cent. In other words, the confidence interval tells us that we are 95 per cent confident that the value of the share of economically active children – between the age of 5 and 14 – would be within an interval of 2 per cent centred on the sample value for the country (i.e. for the selected clusters/sectors) and within an interval of 4 per cent for each sector. For example, for India, the sample size at national level of 439 has a margin of error of 2 per cent, and 110 at sector/cluster level has a margin of error of 4 per cent. It is obtained by the following formula: n

( p * (1  p)) * z2 e2

where n is the dimension of the sample, p is the value we know a priori (the share nationally of economically active children, which is 4.8 per cent in India).

36

Santosh Mehrotra and Mario Biggeri

This a priori information was taken from the Human Development of India household survey carried out in 1994 (national representative and based on 34,398 households). Z is the confidence value and e is the error we accept (to stay within an error  2 points in share terms). Therefore: n

(0.048 * (1  0.048)) * 1.962 0.022

 439

For Pakistan, using the same errors the sample sizes at sectoral level and national level are 75 and 302 respectively. The a priori information was from the Child Labour Survey in 1996 (conducted by FBS, ILO and IPEC); data (3.25 per cent) is for urban areas only. For Indonesia (72 and 289, respectively), the Philippines (51 and 203) and Thailand (111 and 444), for the a priori information we used data from the World Bank, 2002 (WDI, 2001) for children aged 10–14 and estimated for the age-group 5–14. In India,8 the first step was to choose the manufacturing sectors and the second to decide the cluster locations to be surveyed. According to a priori information (in this case a nationally representative sample survey, carried out by the National Sample Survey Organization in 1999–2000), there are 30 million homeworkers in India – and assuming the average national fertility rate applies to such households, there are about 150 million people involved. These workers are concentrated (around 70 per cent) in three sectors: bidi,9 incense stick (agarbathi) and garment (among which zardosi10) manufacturing. These sectors were thus chosen to be part of the sample design. The sectors/clusters and the locations were identified through an interaction with ‘privileged observers’ such as the Ministry of Labour, representatives of Self-Employed Women’s Association (SEWA) (an NGO engaged on a national scale with homeworkers in different clusters/sectors), representatives of the Social Security Association of India and of UNICEF India office. The Ministry of Labour is engaged in the formulation of a National Policy on Homeworkers and for this purpose has compiled a list of sectors in which homework is known to be significant. This list provided the starting point. The second step was to identify the locations of the clusters. The clusters chosen by the aforementioned experts are in different/distant states (Uttar Pradesh, Karnataka, Tamil Nadu, Madhya Pradesh) of the country and this helps to capture the different characteristics among Indian States. For Pakistan the selection of the sector/cluster (carpet weaving, incense stick-making, shrimp-peeling and sack-stitching) locations took into account previous studies that emphasised similar characteristics of homeworker households in Pakistan’s urban areas. The experts picked Karachi, Pakistan’s largest industrial city by far, because most homework activities take place there. Also, the close links of the research institute commissioned to conduct the research11 with PILER (Pakistan Institute of Labour Education Research), engaged in action research/training based in Karachi, facilitated the fieldwork. The sectors chosen would be representative of (urban) slums areas. The field team, after discussions with PILER project leaders and the Aurat (Women’s) Foundation in Pakistan, carried out the sector/cluster identification.

Research methodology 37 In Indonesia the choice of sector/cluster (rattan furniture, ceramic pottery and batik printing on garments) and locations for the ad hoc survey was again based on a priori information and expert collaboration. The choice of clusters of homeworkers in the wider labour market was based on the National Labour Force Survey data collected by the Central Board of Statistics (CBS) and in particular the results of the latest Economic Census of 1996. One member of the team of consultants12 was a member of the CBS. Experts from different institutions and independent researchers were involved as well. Among the experts there was a representative from MWPRI (Mitra Wanita Pekerja Rumahan Indonesia or National Friends of Women Home-Workers).13 The sectors selected were batik in the province of Central Java, and rattan furniture and ceramics in the province of West Java. Once the regencies (administrative area) were identified, two clusters in different districts were chosen for each sector. The unit for data collection, as for the other countries, was the household. Identification of potential households was delegated to the staff of local statistical offices at the regency level. In the Philippines14 the locations of sectors/clusters were chosen following the recommendation of UNICEF Manila and the NGO, Patamaba, involved in the research. The selected locations were Luzon and Visayas. The choice of sectors/ clusters was based on the involvement of subcontracting and production for domestic and export markets. The selected sectors were pyrotechnics production in Bulacan, okra production and packaging in Tarlac, home decor (christmas lights and christmas balls) production in Rizal, and fashion accessories production in Cebu. The Tarlac and Bulacan sites were rural, while Rizal and the site for fashion accessories in Cebu were at the outskirts of the two major Philippine cities – Metro Manila and Cebu City respectively. The sample was equally distributed between urban and rural areas. In Thailand15 the sectors/clusters locations were chosen with the collaboration of the field research team, the Office of Homeworkers (Ministry of Labour and Social Welfare) and Homenet Thailand (network of NGOs working to promote research and development for homeworkers). The survey covered the central, north and north east regions of the country. The criteria for the selection of the homework sectors/clusters (saa paper, leather craft and hybrid seeds production) were the involvement of subcontracting in the production process and the production both for domestic and foreign markets. The survey was conducted in both rural and urban areas, though the majority of the households surveyed were in rural areas. In the micro-surveys in each cluster/sector the homeworker households (households with at least one family member working mainly in the sector) were chosen randomly.16 The portion of the homeworker households interviewed was a large part (often more than 50 per cent) of the homeworker population of each cluster (very close to a micro-census in some of the clusters surveyed). The outcome of this sampling method can be considered representative at different levels in each country for the statistical population of homeworker households. For India, the samples obtained should be close to representative for

38 Santosh Mehrotra and Mario Biggeri homeworker households for the three selected sectors at national level; for Pakistan for the sectors selected in Karachi, but extendable – given the a priori information (Khattak and Sayeed, 2000) – to other urban areas of the country; and for Indonesia for the selected sectors and for the West and Central provinces of Java. The data collected for India (national level), Pakistan (urban areas) and Indonesia (West and Central provinces of Java) are used in the micro-econometric analysis in support of the theoretical discussion on inter-generational transfer of poverty and in order to draw policy implications later. The households surveyed in the Philippines and Thailand, on the other hand, are not enough (they were less than those requested according to the sample design) and thus we consider the error too high to use the data set for econometric analysis. Since the survey method was conceived to collect relevant social and economic data (especially about women and children) on homeworker households only, it presented two limitations. It did not allow for comparison with non-homeworker households and it did not measure the number of children working as homeworkers and the share of homeworkers in child labour at national level. Therefore, in order that at least comparisons could be made a control group (CG) of households in the same geographical area, not engaged in home-based work (with no family member working in any home-based activity) was included in each sample. The CG consisted of households chosen randomly in the same neighbourhood (or cluster location) as homeworker households. If the area was rural then households from the same/neighbouring village were included; if urban, the same neighbourhood. As already mentioned, the survey data shows they had roughly the same income level. The sectors/clusters locations selected, the sample size and the number of households interviewed are presented in Table 2.2. When in the text or in the tables we refer for each country to the aggregate ‘all’ (given by the sum of the sectors) we imply that the value is an estimate for the aforementioned levels. Furthermore, if certain information was not collected in the survey or the data were not comparable, it is so indicated with a dash. Focus group discussions, case studies and in-depth interviews Qualitative methods – FGDs, case studies and in-depth interviews – were used for each sector/cluster, parallel to the quantitative survey. In the five countries, a total of 40 FGDs were held with women and children working in the cluster/sectors under study. The discussions focused on general problems faced by homeworkers, ranking of potential interventions, and perceptions regarding the work of children. The main objective was to try to get feedback on proposed government plans and programmes. Since the household survey also asked such questions, the FGDs would also provide some validation of the survey. Around the same number of case studies was carried out in the sectors studied through in-depth interviews so as to capture sensitive issues – such as, among others, health impact on women, debt bondage, empowerment of women – that would help in developing policy recommendations. Case studies included male

Research methodology 39 and female workers, children and subcontractors located in rural and urban areas. One question which the case studies sought to answer is the relation between a homeworker and the subcontractor. In-depth interviews were conducted in the cluster/network for a small sample of contractors, subcontractors, employers and entrepreneurs. This was done in order to identify the meso-micro linkages and the value chain within each sector, and in particular, to understand the market structure, price mark-ups at different stages of production (i.e. starting with homeworkers and ending with final consumer) and the potential impact of macro- and trade policy on homeworkers.

2.3

Profile of selected sectors

This section presents the sector profiles covered for the homework and CG households, the city and region where homework activities take place, location of the sectors (i.e. urban or rural), the type of activity involved and the dynamics of the size of the market for the homework products. The three to four sectors examined in each country are listed in Table 2.2. The sectors in India are incense sticks (agarbathi) in Bangalore district (in the state of Karnataka); bidi-making (tobacco leaves rolled to make cigarettes) in Indore (in the state of Madhya Pradesh) and Vellore district (in Tamil Nadu); and zari/zardosi (embroidery on garments) in Lucknow district (in the state of Uttar Pradesh). In incense, zardosi and bidi (in Tamil Nadu) the sample covered rural, peri-urban and urban areas, but was limited to an urban area in bidi in Madhya Pradesh. Of the 452 homeworker households, there were 225 rural and 227 urban households in the sample in India. The CG consisted of 151 households. The market for all three product groups is external as well as domestic (Table 2.3). In Pakistan four sectors were chosen, all located in the major city port of Karachi, on the Arabian Sea coast. There were 303 homeworker households in the sample, and 94 in the CG – all located in urban areas. The four sectors were incense stick making (the same as in India), carpet weaving, sack stitching and prawn peeling. The market for carpets, prawns and incense sticks is predominantly external, and to some extent domestic; for sacks it is both. In terms of the ethnic composition of the groups engaged in homework, three were immigrant communities (Burmese Muslims from Myanmar and Bengali Muslims from former East Pakistan), and in the fourth sector (sack stitching) the community had migrated to Karachi from the Indian state of Gujarat at the time of independence. In Indonesia three sectors were selected – batik printing on garments, rattan furniture and ceramic pottery. There were 210 homeworker households in the sample, and 90 in the CG. The site for the batik sector was in Central Java – an area long known for Indonesia’s famous batik work – in three villages, located 10–15 km from the sub-district capital. While the general area is known as a batik centre, most men in the villages selected rely on fishing and farming for their livelihood. Batik-making at home is essentially women’s work. Farming, fishing and batik skills are passed on from generation to generation, and learned from experience rather than formal education or training institutions.

Sub-total

Batik

Rattan

Sub-total Indonesia Pottery

Carpet weaving Sack stitching Prawn peeling

Sub-total Pakistan Incense stick making

Zardosi

Bidi (TN)

India Incense stick making (agarbathi) Bidi (MP
TN) Bidi (MP)

Country/sector

Purwakarta, Plered/Sukatani District, West Java Cirebon, Weru District, West Java Pekalongan, Wiradesa/ Tirta District, Central Java

Orangi, Unit 12C, Karachi Korangi Town, Karachi Godhra, Karachi Machar Colony (Mohammadi Colony), Karachi

Bangalore District, Karnataka MP
TN Indore District, Madhya Pradesh North Arcot Ambedkar District, Tamil Nadu Lucknow District, Uttar Pradesh

Location

300

100

100

100

397

99 100 99

99

603

201

210

70

70

70

303

77 75 74

77

452

148

75

151 76

201 101 100

153

201

90

30

30

30

94

22 25 25

22

151

53

25

50 25

48

163

30

70

63

303

77 75 74

77

227

58

15

91 76

78

hw

CG

Total

hw

Urban

Households surveyed

51

14

30

7

94

22 25 25

22

60

12

—

25 25

23

CG

47

40

—

7

—

— — —

—

225

90

60

60 —

75

hw

Rural

39

16

—

23

—

— — —

—

91

41

25

25 —

25

CG

6

2

2

2

8

2 2 2

2

7

2

—

2 —

3

Total

—

—

—

—

8

2 2 2

2

—

—

—

— —

—

Urban

6

2

2

2

—

— — —

—

—

—

—

— —

—

Rural

Number of FGDs*

9

3

3

3

8

5 3 —

—

9

4

—

3 —

2

Number of CS

Table 2.2 Surveys of homeworker households in five Asian countries: sectors, location, number of FGDs, case studies and households surveyed

Photho District, Province, Angthong, Central, Song District, Phrae Province, Northern Ratburana District, Bangkok Province, Central, Samrong District, Samutprakarn province, Central Sakon Nakhon and Kalasin Province, Northeast

Rizal Province, Luzon Kalayaan in Angono, Rizal Province, Luzon San Vincente in Angono, Rizal Province, Luzon San Vincente in Angono, Rizal Province, Luzon Sta. Maria, Bulacan, Luzon Sierra, Concepcion, Tarlac Province, Luzon San Roque, Talisay, Cebu Province, Visayas

101 305 1405

399 1872

102

102

135

30

35 34

10

20

36 6

131

136

132

173

41

44 43

—

—

45 —

94 467

30

34

30

38

11

9 9

—

—

9 —

98 857

—

98

—

66

30

— —

—

—

36 —

Notes Dash indicates that certain information was not collected in the survey or data were not comparable. FGDs  focus group discussions. CS  case studies. hw  Homeworkers. CG  control group or non-hw household. * Considering separately the one for women and the one for children.

Source: UNICEF survey.

Hybrid seeds production Sub-total Total

Leather crafts

Sub-total Thailand Paper products

Fashion accessories

Pyrotechnics Okra

xmas balls

xmas lights

Philippines Home décor Metalcraft

28 253

—

28

—

20

11

— —

—

—

9 —

207 548

101

4

102

69

35 34

—

—

— —

66 214

30

6

30

18

9 9

—

—

— —

7 34

3

2

2

6

1

1 1

—

1

1 1

— 11

—

—

—

3

1

— —

—

1

1

7 10

3

2

2

3

—

1 1

—

—

1 —

11 45

4

4

3

8

2

2 2

2

42

Santosh Mehrotra and Mario Biggeri Table 2.3 Market for selected products of homeworkers Sector India Incense stick making Bidi (MP
TN) Zardosi Pakistan Incense stick making Carpet weaving Sack stitching Prawn peeling Indonesia Pottery Rattan Batik Philippines Home décor Pyrotechnics Okra Fashion accessories Thailand Paper products Leather craft Hybrid seeds

Foreign market

Domestic market

Regional local market

X X X

X X X

X X X

x X X X

X X x x

x x x x

x X x

X X X

X X X

X

X X

X x

X X

X

X

X X X

X X x

x x

Source: UNICEF survey and country reports. Notes X  relevant; x  minimal.

The rattan sites selected were an urban centre (Tegalwangi on the northern coast of Java) and two smaller villages (in Cirebon district). Tegalwangi is an urban centre associated with the rattan furniture industry, while the villages are largely agricultural with rice fields. As the terms of trade for agricultural products, especially rice, continue to fall, villagers prefer to find income earning opportunities through off-farm work. The third site, chosen for the third sector – pottery – consisted of the village of Anjun (in district Purwakarta), which is known as West Java’s small-scale pottery industry centre. All three products are mainly produced for the domestic market, but they are also exported. In Thailand, of the three sectors chosen, two were in manufacturing – paper products (saa paper, artificial flowers, etc.) and leather crafts – and one in agriculture, hybrid seeds production. There were 305 homeworker households in the sample, and 94 in the CG. All three product groups have a strong external market, but with a domestic market as well. Two villages were selected in northern Thailand for paper products. Most villagers are rice farmers; families also have their own mixed fruit garden, and during farming periods homework is not undertaken.

Research methodology 43 Hybrid seed production, an agricultural activity, is carried out in the field rather than at home. Nevertheless, it is part of the informal economy and is carried out on family farms. Hybrid seeds are used to grow water melons, chile peppers, cantaloupe and several vegetables. However, the seeds from these high-yielding crops are sterile, so farmers have to buy from seed companies in order to produce the new round of crops. The production of the crop and of seeds is mostly for export. In the Philippines there were four kinds of activities selected for assessing homework – home décor commodities (specifically Christmas balls and lights; metalcraft, for example, wrought iron baskets); fashion accessories (e.g. wood bead necklaces); pyrotechnics (or firecrackers); and okra vegetable production. There were 135 homeworker households in the sample, and 38 in the CG. Only fireworks are produced for the domestic market; the other product groups are mainly for export; okra production is only for export. All utilise the subcontracted labour of homeworkers. Okra growing is unique because, like hybrid seed production in Thailand, it does not involve the usual ‘in-house’ production. For all countries taken together, almost all product groups chosen could be regarded as manufacturing activities; only three involved the processing of agricultural products: prawn peeling in Pakistan, okra production in Philippines, and hybrid seed production in Thailand (all three for export). In fact, much of homework in Asia involves manufacturing activities. However, this does not mean that homework is confined to urban areas. In fact, the subcontracting of such manufacturing activities extends to the rural household. Hence, rural areas were part of the sample in India and Indonesia (though not in Pakistan, where all locations are in Karachi). In this sense it holds out the prospect of bringing rural areas into the fold of manufacturing activities. At the same time, agricultural activities are getting drawn into the globalised value chain and the putting out system. In terms of the size and dynamism of the market, of the three sectors chosen in India, only bidi-making seemed to be a declining industry; and in Indonesia rattan furniture was also regarded as a declining industry. However, all other activities in the other countries seemed to be producing for growing markets, regardless of whether they were traditional activities (e.g. pottery, leather, zardosi) or more non-traditional ones (e.g. okra production, hybrid seed production, prawns). While most were producing for the export market, the domestic market was also targeted. In our studies, there are several sectors involving foreign-firm buyer-driven chains involving international subcontracting – in manufacturing (e.g. leather goods in Thailand and carpets in Pakistan) and in agriculture (hybrid seed production in Thailand and okra production in the Philippines). Even for domesticfirm led chains, the products involved have a large export market, and this market has grown in the 1980s and 1990s. In the India study it is clear that at least two of the products – incense sticks and embroidered garments (zardosi) – are being produced for a large export market. The study suggests that subcontracting is, therefore, likely to have grown in the products where the market has been

44

Santosh Mehrotra and Mario Biggeri

growing. In Pakistan, too, carpets, incense and shrimps are three export-oriented products and their growth has lead to greater subcontracting.

Annex 2.1: Questionnaire guidelines for homeworkers (and control group)

Identification codes Country study, place, cluster/sector, household id and serial number of each member within the same household.

Section (i): Personal information All the questions not marked with ‘C’ should be addressed to adult homeworkers (Men and Women) Information on all the members of the household are required General (collect information on each member of the household)

1 ●

Name of the interviewee

●

Sex

●

Marital Status (married, unmarried, divorced, separated, widowed, other)

●

What is your age? (Child 1, Child 2, Child 3)

●

Ethnic and social identity (social resources – subcultures, kinship ties, common language, same-place identities – play a crucial role in shaping the regional economic networks and commodity chains).

Be of a local origin?

●

In order to succeed in becoming an homeworker in the local industry, is it important to:

Belong to the leading social community?

●

Education degree, Highest school Class Attended

Can you read, read and write. Formal education in years. Which kind of school did you attend/have you been attending (government/ private one).

Research methodology 45

Section (ii): Information on the H’hold’s activity 1

General ●

Number of people residing in the house (total number, women over 14 years, men over 14 years, girls less than 14 years, children less than 14 years).

●

Total number of earning members residing in the house.

●

Number of people who contribute to household income on a regular basis.

2

H’hold’s production’s ●

Monthly average income of the household in the last 6 months (in general, not only from homework).

●

Monthly average expenditure of the household in the last 6 months.

●

Have you taken any loan in the last 6 months?

●

How much debt has your household incurred?

●

From whom is the loan contracted?

Relatives Neighbours Professional money lender Bank Welfare society (NGOs, etc.) Employer Others

●

Industry of activity (describe fully).

●

Loss/Net profit in 2000.

Good Reasonable

46

Santosh Mehrotra and Mario Biggeri Nil Loss ●

Capital. Fixed capital. What is the current value of the assets the household owns for this activity? Machinery/Tools and other. Working capital (incl.stock and cash). Total capital.

●

What were the main sources of funds for your capital when you started this homework?

Government assistance

Bank loan International and NGO finance Private loan Another business (selling crops/friends-relatives/wages job of self or household member/other sources) ●

3

What was the household’s income in the last 12 months from all sources? (HW/Own Account-Home based/other business/wages and salaries/farming-fishingforestry/benefits from social security/other sources TOTAL household income.

Housing and living conditions ●

Ownership status of the household dwelling.

Owned Provided free by employer Rented from private owner (amount paid per month) Rented from Government/ public ownership (amount paid per month) Subsidies by employer (private or Government/public ownership) (amount paid per month)

Research methodology 47

Section (iii): Work related questions 1

General ●

Nature of work:

Factory Domestic service Helper in small shop, repair shop, etc.

●

How old were you when you began to work?

●

Job Training for this activity?

Yes (if yes, No. of weeks of training and type – with teacher or form family/friends) No

●

What is the main reason you do this work?

Traditional family activity Good for family-group activity Encouraged by government No other work available Adds to family income Good return-low capital Good market for products To collect money for your own use Because of the unemployment/ death/disability/illness of some other earning member of the household Other (describe)

●

What is the main disadvantage?

Little social impact Makes the house messy Less time for family Very tiring Low income Health problems Long hours None Other describe

48

Santosh Mehrotra and Mario Biggeri

2

Time organisation (includes time allocation questions) ●

How many days you work in a week?

●

Numbers of years in this activity

●

Are you doing this work all-year long?

●

How many hours are you working each day?

●

During what time and how many hours do you work?

Daytime for . . . hours per day

Evening time for . . . hours per evening Night time for . . . hours per night ●

Are your children going to school when you are working?

●

If not, who is taking care of them?

●

How many hours per day are you dedicating on a) Food preparation b) Household keeping c) Animal Husbandry d) Shopping e) Child care (including physical care, teaching, training, accompanying them to places) f) g) h) i)

Adult care Social life Community services Personal Care (including sleep and related activities, eating, personal hygiene and health, resting, religious practices, j) Mass media use

●

Do you shift regularly to other kind of works during the year?

Research methodology 49 3

Payment ●

How are you paid for your work?

On Piece rate On Time basis

●

If paid on piece rate, what is the rate?

●

How many pieces does the H’hold do in a day?

●

What is your payment arrangement? In kind/other (describe).

$ . . . ..in cash indicate the pay period: $ . . . ..in kind (estimates of all payments in kind) Total earnings: $ . . . .in-cash and in kind.

●

What was the amount paid to you for the latest pay period?

●

If paid on time basis, what is the frequency?

Monthly Weekly Daily

●

Are you paid on time?

●

How many days after delivery are you normally paid?

Up to 15 days Up to 30 days Up to 60 days Over 60 days

●

4

Are you engaged in this work with other members of the household?

Agreements with contractor ●

Is your contract with the contractor oral/ written/no contract?

●

If verbal, what is the content of the contract? (depending of nature of activity, this question should be developed by surveyor further).

50

Santosh Mehrotra and Mario Biggeri

5

Organisation of the production ●

What are the materials used for this work and who supplies these materials?

●

How do you get your raw materials?

Contractor deliver You collect form contractor You buy them You grow or gather them Other (describe)

●

In case you get yours orders through middlemen, where in the subcontracting chain are you?

One layer between you and the contractor

Two layers between you and the contractor More than two layers between you and the contractor Others (specify) ●

From how many sub-contractors do you get your work/order?

1

2 3 or more ●

What happens in case of faulty work?

Product rejected – cost borne by you Product rejected – cost borne by employer Product repaired

Research methodology 51 6

Market Information ●

How do you dispose of your finished work?

Contractors collect You deliver to contractor Other (describe)

●

Specify to whom you sell your products.

Direct to the consumer Direct to the retailer Direct to the wholesaler Through an independent sale representative To an export agent Through a consortium with other manufacturers Other (specify)

●

Where are the outlets at which your products can be bought? (in %)

Local area

Rest of state Rest of country Abroad ●

When working on HB activity do other members of household help?

(if yes, No. of adults – full time or part time No. of children – full/part time. Average number of hrs per week, per adult – full/part time – per child – full/part time)

●

How much of your income you keep for personal expenses?

52

Santosh Mehrotra and Mario Biggeri

7

Women working for someone else, or independently (self-employed) for earning (money) in cash or in kind ●

Do you give a part of all of your earnings to your parents/guardians/h’hold or other relatives you usually reside with?

Yes, all direct through the employer

Yes, all by self Yes, part through the employer Yes, part by self No Other ●

How is your relationship with your employer?

Good Bad Indifferent

●

If bad, give the main reason:

Wants too much work done Wants work done for long hours Pays poorly Does not pay on time Abuses physically Abuses verbally Other (specify)

●

Referring to the latest most recent payment, what is the approximate amount you were paid by your employer? (give amount and the period for which this amount was paid).

Research methodology 53 8

Homeworkers women’s organisation ●

Are you a member of any labour or trade association?

●

If no, are you interested in joining such an association?

●

Are you entitled to any employee social benefits from your work?

●

If yes

Yes No

Yes No Annual leave Sick leave Maternity leave Health care benefits Employment injury benefits Occupational disease benefit Retirement pension Benefits for dependants

Section (iv): CHILD specific section 1

Education ●

Does the child go to school?

(Child 5 years old) (Child 6 years old)

(TO PARENT)

(Child 7 years old) (Child 8 years old) (Child 9 years old) (Child 10 years old) (Child 11 years old) (Child 12years old) (Child 13 years old) (Child 14 years old) (Child 15 years old)

54

Santosh Mehrotra and Mario Biggeri ●

Reason for not sending children to school: (TO PARENT)

Cannot afford it The child is working Disagree with sending the child to school The school quality is bad

●

Do you value the education of boys and girls? (TO PARENT)

Equally More for boys More for girls

●

2

If attending school or training institution on a full time or part time, but also working, does your work affect your regular attendance or studies?

Yes No

Questions related to children working for someone other than parent ●

●

Does any child in the family work for someone else other than own parents or guardians?

Yes No

How is your relationship with your employer? C

Good Bad Indifferent

●

If bad, give the main reason: C

Wants too much work done Wants work done for long hours Pays poorly Does not pay on time Abuses physically Abuses verbally Other (specify)

●

Referring to the most recent payment, what is the approximate amount you were paid by your employer? (give amount and the period for which this amount was paid).

Research methodology 55 3

Children working for someone else, or independently (self-employed) for earning (money) in cash or in kind

●

●

Do you give a part of all of your earnings to your parents/guardians or other relatives you usually reside with? C

Yes No

Do you save any part of your earnings? C

Yes regularly Yes occasionally No

4

Questions related to children living away from a household (parents’ or guardians’ household)

●

Is there any child (or children) of this household who does (do) not usually live or reside here?

Yes (give number, names, sex, age and where they live or reside now).

Do you know where they live or reside at present? What does (name) do where he/ she is now?

Working independently/ as self employed Working for someone Other (specify)

No ●

Does (name) send money home to this household?

Yes

No

5

Children non economic activities: housekeeping activities/household chores Note: These are personal services of a domestic nature provided by unpaid household child members and, as such, are considered as non economic which, therefore, are outside the production boundary of the System of National Accounts. They include preparing and serving meals, making, mending, washing and pressing clothes, shopping, caring for siblings or sick and infirm persons in the household, cleaning and maintaining of the household dwelling, using, cleaning, serving and repairing household durables, transporting of household members or their goods, etc.

56

Santosh Mehrotra and Mario Biggeri ●

Have you been engaged in housekeeping activities or household chores in own parents’/guardians’ home on a regular basis during last week? (C)

Yes

Less than three hours each day 3–4 hours each day 5–6-hours each day 9 hours or more each day

No ●

●

●

6

Why? Are you supplanting your mother in children care and household duties? Does she work at home or outside? What is the time spent by mother on these tasks? (C) What do you do when your mother is working? (C) Who takes care about your when your mother is working? (C)

Perception of parents/guardians or relatives with whom the working child usually resides ●

What does (name) do for fun when not working?

●

If (name) is working, what is the main reason for letting him/her work?

To supplement household income To pay outstanding debt under contractual arrangement To assist/help in household enterprise Education/training programme is not suitable Education/training institutions are too far Other (specify)

●

If (name) stops working, what will happen?

Household living standards decline Household cannot afford to live

Research methodology 57 Household enterprise cannot operate fully and other labour not affordable Other (specify) ●

If given a choice, what would you refer (name) to do in the future

Going to school full time Working for income full-time Helping full-time in household enterprise or business Working full time in household chores or housekeeping Going to school part time and working part time for income Part time in household enterprise or business Part time in household chores or housekeeping Complete education/training and start to work Find a better job/work than the present one Other (specify)

Section (v): HEALTH

●

Evaluate the workplace for each of the following as Good-Fair-Poor

Space Noise Light Dust Smell

58

Santosh Mehrotra and Mario Biggeri

Air quality Temperature Hazards ●

How paid work impacts your health?

●

Does the respondents have any health problems as a result of this activity?

Yes (give type) No

●

If yes, what was/were the nature of your illnesses/injuries?

General (fever, cold, etc.) Eye infection Ear infection Skin problem Breathing problem Stiff neck Back problem Anaemia Other (specify)

●

Referring to the most serious injury/illness, how serious was the injury/illness?

Did not need any medical treatment Medically treated and released immediately Stopped work temporarily Hospitalised Prevented work permanently Other (specify)

Research methodology 59

Section (vi): Questions for subcontractors/employers

Explain the reason for recruiting

Possibility to recruit from a much larger area

Homeworkers

Possibility to hire workers in accordance with variations in the demand Minimization of the risk of unionization Freedom to vary the volume of the production Greater opportunity to vary the nature of the work Reduction of costs Greater flexibility in responding to the fluctuations and irregularity of the market

Do you normally recruit homeworkers on a written or verbal contract base? How do you calculate the rate of remuneration?

After consultations with the representative organizations of the workers Own methods (specify) Other

Which kind of work are you performing? Do you keep a register of all HWers to whom you give work? Is there any kind of registration with the competent authority of the workers you recruit? Are you aware/take in consideration the national laws concerning minimum labour standard conditions for workers (age, working hours, wages)?

Yes, regarding age Yes, regarding wages Yes, regarding working hours Other No

60

Santosh Mehrotra and Mario Biggeri

Are the workers paid for costs of production (over and above the piece/time rate)?

Use of power Use of water Use of equipment Other costs

Notes 1 For instance, none of the ILO studies mentioned in Chapter 1 (note 19) nor a recent book (Balakrishnan, 2002) raises the issue. 2 A study on home-based work by ILO in Latin America had already been undertaken see note 19, Chapter 1 and www.wiego.org. 3 This phase was carried out by the country teams. 4 This phase was carried out by Stefano Mariani in Florence, Italy, for the cross-country analysis contained in Part I of this book. He contributed in getting the database in order. 5 For instance, in India the questionnaire had the following sections: (i) Household characteristics: this section covered basic information about the household along with their social and ethnic background. Data on age, sex, marital status, educational status, occupation was collected for each person in the household. (ii) Economic profile of the household: including average monthly expenditure and income of the household. (iii) Information related to homework: including details on the type of homework, hours of work, days worked per week, months worked in a year, average wages earned, expenditure on raw materials, sources of funds, nature of contract, etc. (iv) Time allocation and work organisation. (v) Homeworkers women’s organisation: intended to capture benefits received from employer/contractor, and general awareness of worker issues. (vi) Child related: including extent of contribution of work by children and its effect on their education and household economy. (vii) Health: especially health problems suffered due to work. (viii) Perceptions: relating to policy interventions and to child work. The homeworkers in the sample were defined to include subcontracted workers, working from their home, or self-employed workers, working from their home. The categories are not mutually exclusive, and no effort was made to distinguish between ‘subcontracted workers’ and ‘occasionally subcontracted workers’. The majority of workers surveyed was found to be subcontracted workers; the same workers might be self-employed part of their time. 6 In order to test statistically the homogeneity, we did an ex-post t-test on the income of the homeworker households for three countries (India, Pakistan and Indonesia) by sector/cluster and by including all sectors/clusters together. We found that the mean is not statistically different (significant at 5 per cent). This is true also among the control group or CG households. Then, between homeworker households and CG households we found for India that by sector/cluster and by including all sectors/clusters together, the income mean is not statistically different. 7 By sampling in stages, you can reduce costs and still obtain a reliable sample frame (McCormick and Schmitz, 2002, p. 182). 8 The National Council of Applied Economic Research, New Delhi, was responsible for the study in India. 9 Tobacco leaves rolled to make cigarettes, consumed largely by the urban poor and in rural areas.

Research methodology 61 10 Embroidery on garments, with gold thread, involving skills usually passed on from generation to generation. 11 The research institute responsible for the research was Sustainable Policy Development Institute, Islamabad. 12 An academic consulting firm in Jakarta was commissioned to conduct the study for Indonesia. 13 The researcher was also involved in an ILO-DANIDA project in the early 1990s and in charge of a World Bank funded study on homeworkers in Indonesia. MWPRI is a member of Homenet International, which is also active in Thailand and Philippines. Homenet International’s national affiliates in the countries studied were involved in the research design and process. Homenet International, headed by Ela Bhatt, founder of SEWA, was the leading force behind the birth of the ILO Convention on Homeworkers, 1996. 14 The University of Philippines (Departments of Social Work and also Economics) and the NGO Patamaba (an affiliate of Homenet International) were commissioned to carry out the research. 15 The Chulalongkorn University Social Research Institute (CUSRI), Bangkok, were commissioned to carry out the study in Thailand. 16 The dimension of the sample was chosen in each country according to the confidence interval based on a priori information as described before.

3

Subcontracting and homework in the value chain Santosh Mehrotra and Mario Biggeri

There is much that distinguishes the formal from informal activity/enterprise. Normally formal sector activity is characterised by capitalist enterprise, the informal by family ownership; barriers to entry as against ease of entry; capital intensity as against labour-intensive technology; formal training as opposed to skills being acquired outside of formal education; within the ambit of government policy as against being outside government control; often unionised versus normally being non-unionised; regulated markets versus highly unregulated markets for labour, capital and products (Lund and Srinivas, 2000). But in many industries, it is subcontracting that provides the link in the chain between the formal and the informal parts of the production process (Horn, 2004). Homeworkers form a significant part of the informal sector in most developing economies, but are often linked to the formal economy through a value chain. In the wake of globalisation there has been a restructuring of production and distribution in many industries characterised by outsourcing and subcontracting, and increasingly involving global commodity chains (Gereffi, 1999). This is one reason for an increase in employment in the informal economy, as we noted in Chapter 1. Global trade and investment patterns tend to privilege capital, particularly firms that can move quickly and easily across borders, and to disadvantage labour, especially lower skilled workers that cannot migrate easily or at all. Moreover, globalisation tends to privilege large firms who can capture new markets quickly and easily, and disadvantages small and micro entrepreneurs who face difficulties gaining knowledge of, and access to, emerging markets (Rodrik, 1997). But globalisation can also offer opportunities (Gereffi and Kaplinski, 2001) to those in the informal economy in the nature of new jobs for wage workers or new markets for the self-employed. Therefore, the expansion of demand (considering the limited domestic market) and contribution to technological upgrading are the two major opportunities offered to local producers. Local producers, or better, the local system of production can learn from leaders (buyers and producers) of the chain. In this way it is possible to upgrade the cluster (Giuliani et al., 2005). In this case the relationships among actors also involves coordination or governance which are needed to realise the benefits in terms of local development and in non-market relationships.1 However, for the opportunities to improve the level of human development of those who work in the

Subcontracting and homework in the value chain 63 informal economy, public and private collective action is needed (as we mentioned in Chapter 1 and as we will discuss in detail in Chapter 11). International forces are not the only ones driving subcontracting. There are, in most developing economies, domestic forces as well. As we noted in Chapter 1, one determinant of subcontracting could be transaction costs; when transaction costs are high, firms will decide to produce in-house. Young industries with specialised needs and input supply unreliability tend to be vertically integrated. On the other hand, mature industries, in which products are standardised, will be characterised by firms that engage in outsourcing. For example, early Indian industrial growth tended to be highly vertically integrated, but increasingly resorted to outsourcing.2 Domestically, the labour market is another determinant of subcontracting. Outsourcing occurs when governance costs of integrated production rise because of poor labour relations. Unions perceive that the threat to strike would be weakened if management has the option to augment production from outsourcing. Lifting of labour regulations – which have typically happened with market-friendly policies – have tended to lead to a worsening of workers’ security. The paradox is that economic growth may be accompanied by a rise – not a decline – in human security. In fact, increased economic growth (i.e. GDP per capita) has in many parts of the world, not led to corresponding improvements of real freedom or basic work security.3 The ILO has identified seven forms of security associated with work, which could be pursued by governments, employers, unions and others. They are: ●

●

●

●

●

●

●

Labour market security. Adequate employment opportunities, through state-guaranteed full employment; Employment security. Protection against arbitrary dismissal, regulations on hiring and firing, imposition of costs on employers for failing to adhere to rules, etc.; Job security. A niche designated as an occupation or ‘career’, barriers to skill dilution, craft boundaries, job qualifications, craft unions, etc.; Work security. Protection against accidents and illness at work, through safety and health regulations, and limits on working time, unsociable hours, night work for women, etc.; Skill reproduction security. Opportunities to gain and retain skills, through apprenticeships, employment training, etc.; Income security. Protection of income through minimum wage machinery, wage indexation, comprehensive social security, progressive taxation to reduce inequality and to supplement those with low incomes, etc.; Representation security. Protection of collective voice in the labour market, through independent trade unions and employer associations incorporated economically and politically into the state, with the right to strike, etc. (ILO, 2004a).

It will be obvious in the course of this chapter and the rest of the book that informal workers studied here do not have access to any of these seven forms of security. Our primary claim is that the workers in these informal income-earning

64

Santosh Mehrotra and Mario Biggeri

activities are entitled to receive at least some of these basic securities. In the short run pragmatic factors (e.g. the benefits to employers that comes from the informality of the employer–homeworker relationship) may present a barrier to the extension of many forms of security, but some forms of basic security are not inconsistent with the informality of that relationship.4 The objective of this chapter is to present a cross-country analysis of subcontracting and the value chain in homework in the five countries examined in Asia and examine the implications for worker security. Section 3.1 first presents a framework to analyse the value chain in such sectors. Section 3.2 provides examples of the value chain from the five countries, goes on to examine the relationship of homeworkers to the subcontractors, forms of worker insecurity resulting from the value chain, as well as the price mark-ups in the chain for the sectors selected. Section 3.3 summarises the arguments of the chapter emphasising the possible policy implications.

3.1

The value chain

The concept of global commodity chains was introduced into the literature by Gereffi during the mid-1990s.5 Value chain is the set of value-adding activities through which a product passes from the design to the consumption stages. The worth of the product increases at each point of the process, hence the term value chain. These commodity chains can be local, national, regional or global with increased globalisation. A value chain is ‘global’ when different activities are carried out in different countries (McCormick and Schmitz, 2002). In other words The value chain describes the full range of activities which are required to bring a product or service from conception, through the different phases of production (involving a combination of physical transformation and the input of various producer services), delivery to final consumers, and final disposal after use. (Kaplinsky et al., 2001, p. 4) Our concern in this book is mainly with national value chains, even though many of the products under study are exported. Traditionally, homeworkers were analysed in their local context, but as we saw local conditions are increasingly affected by global forces. Value chain analysis helps us to understand the connection to other actors in the chain. The unequal distribution of revenue across the value chain in each sector indicates the existence of economic exploitation in the labour process. Value chain analysis can help to identify winners and losers resulting from the globalisation of product markets, and can help to find ways of spreading the gains of globalisation (Kaplinsky, 2000; McCormick and Schmitz, 2002). It also helps to identify the market power, profits share, the governance (Giuliani et al., 2004) and structure of the networks and informal/formal links. In general the value chain takes the shape as described in Figure 3.1. The diagram indicates the different steps of the value chain. The figure represents

Subcontracting and homework in the value chain 65

Design

Production

Small enterprises

Middleman

Raw materials or inputs

Homeworkers

Distribution

Wholesalers

Retailers

Consumers

Figure 3.1 Example of input–output structure of a value chain.

a typical simplified structure of a value chain. Indeed, if information is available it can be represented also as a map, giving all sorts of information regarding the chain, such as network structure, geographical spread, power relationships – control and governance. This can be represented graphically by using, for instance, the share of chain value added, share of chain sales, share of chain profits, relative rate of profit, share of chain buying power (Kaplinsky et al., 2001; McCormick and Schmitz, 2002). In other words, chain mapping can be used to complement the value added information with other relevant aspects of the chain, such as indicating a control over a key technology or a distinctive competence and holder of chain ‘market identity’ (e.g. brandname) (Kaplinsky et al., 2001, p. 66). A significant segment of the value chain often is in the informal economy. The link, between working in the informal economy and being poor, is strong, and stronger for women than for men. This is partly because a higher share of women than men work in the informal economy, and partly because women have lower representation in the higher income employment status in the informal economy (employer and self-employed) and higher representation in the lower income statuses (casual wage worker and industrial outworker) (Carr and Chen, 2001).

66

Santosh Mehrotra and Mario Biggeri

Sectors where buyer-driven value chains have grown Subcontracting chains offered opportunities for women in the 1990s – especially in the garments sector.6 The number of developing country garment export bases were being continually expanded over the 1990s and earlier in order to by-pass the import ceilings in rich countries mandated by quotas against previously successful apparel exporters – a phenomenon that is not true for footwear, since the latter are not subject to quotas (Gereffi, 1994a). Thus garment exports began first from the East Asian countries and then shifted to South East Asian and Latin American and South Asian locations. However, there is now a reverse trend (away from this dispersal towards greater geographical concentration), driven by technological changes: the bar code enables retailers to stock smaller inventories and change orders more frequently, resulting in demands for quick and timely supply of goods (or the ‘just in time’ inventory system). The products also change according to retail buying seasons, with leading apparel firms having six buying seasons in the year. As a result subcontractors who supply garments have to be located closer to the main markets in Europe and North America, and even domestically within these main markets. The just-intime delivery system resulted in an increase in homeworking in countries such as Turkey, Morocco, Mexico and Guatemala, and is beginning to threaten the largescale garment sector in Asia (Carr and Chen, 2001). The increasing competition of course means lower prices for consumers in European and North American markets, as corporations bid down labour costs in developing countries (Bairoch and KozulWright, 1996; Joekes, 1999). The end of the Multilateral Fibre Arrangement (MFA) in 2005 ending quotas means further dislocations for a host of factory and industrial outworkers in countries in Asia dependent upon garment exports (e.g. Bangladesh). The Philippines case demonstrates the upside as well as the downside of globalisation involving a system of production characterised by subcontracting. Women were the first to get jobs in the garment sector in the period of export expansion, but they were also the first to lose them when the Asian crisis broke in mid-1997. By early 1998 over 200 garment factories had closed in the Philippines with an instant fall in income for factory and homeworkers. Women had to seek alternatives such as horticulture – and there has been a rise (as we discuss later) in alternative homeworker activities, such as the kind discussed in this book (fashion accessories, home décor). Another group of products involving value chains is non-traditional agricultural exports – primarily fruits and vegetables and cut flowers aimed at the European, North American and increasingly the Japanese markets (as in the case of okra production in the Philippines intended for Japan, hybrid seed production in Thailand, discussed in this book). In Chile, for instance, there are 300,000 temporary workers in the agricultural sector picking fruit, of whom over 50 per cent are women. There are only 50,000 permanent workers, 95 per cent of whom are men. Women put in more than 60 hours a week during the season, but one in three earned less than the minimum wage (Oxfam, 2004a). In fact, for seedless grapes in 1993–94, producers accounted for 11 per cent of costs, while exporters, importers and northern retailers accounted for 28.26, and 35 per cent respectively (Barrientos et al., 1999).

Subcontracting and homework in the value chain 67 Another group of products with an international market and involving value chains are non-timber forest products – rattan (which we examine in the case of Indonesia), saa paper7 (which we examine in Thailand), shea butter, medicinal plants, gum arabic, brazil and other edible nuts which produce essential oils used for cooking and skin care. However, how much value is added in the developing country is an important factor in how much the original producers/collectors benefit from the product. Thus, one study in Burkina Faso estimated that shea butter was sold to consumers in Europe at 84 times the price paid to local women for the raw material. However, rattan in Indonesia is made into furniture domestically, and then exported – just as the saa paper is in Thailand. Figure 3.2 reports the share of female workforce within three different chains. The way the value chain is organised and controlled depends on many factors, including the network and ownership structure of the firms engaged in the production process. McCormick and Schmitz (2002) identify four types of relatioships between firms in the chain: (a) a market relationship; (b) balanced network; (c) directed network; and (d) hierarchy (see Figure 3.2). Analysis of relationships along the chain can help to identify the leverage points for policy. For instance, the price mark-up and the value-added at each step of the chain from the raw material stage to the consumer gives us the information to identify the exploited, as well as the margins for action (e.g. association of producers and co-operatives). The link between formal and informal sector activities helps us to identify the opportunity for enlarging social protection into the informal sector and the space for collective action or state intervention. Finally, it is possible by looking at the conduct of exporters and importers (from developed countries) to implement codes and standards stipulated by the international community.8 Such public opinion pressure can indirectly help to gradually influence the behaviour of importers or producers in general. Homeworkers are mainly unorganised, and the lack of joint action and collective action can make them vulnerable in the face of powerful employers. The employers/ subcontractors can take advantage of workers’ limited information and training and limited employment alternatives in different ways, squeezing the workers’ margins by paying low piece-rates, by perpetuating unwritten contracts, by giving no social protection, by delaying payment. These power relationships often lead to, or could be manifested in debt bondage, unsafe and hazardous work and child labour. The policy response to these problems has to be found at local level and accompanied by national and international policies.

3.2

Forms of insecurity of homeworkers in the value chain in the five Asian countries

The ILO has, since 2003, moved to a status-based or ‘job-based’ concept of informality – as opposed to an enterprise-based concept (see Chapter 1).9 It recognises the reality that globally workers face degrees of informality, the most formal having multiple dimensions of protection, the least formal none at all.

68

Santosh Mehrotra and Mario Biggeri

Chain 1 National market

Retail

60%

Whole sale

10%

Production

Chain 2 US market

Chain 3 German market

Unknown

75%

78%

85%

90%

Homeworkers 98%

Inputs

Suppliers of yarn and cloth

Suppliers of accessories and packaging material

Equipment suppliers

38%

20%

5%

Arm’s length market relationship Directed network

Balanced network Hierarchy (subsidiary)

Figure 3.2 Value chain map – share of female workforce (%). Source: McCormick and Schmitz, 2002, p. 92.

Presenting informality as a continuum, ILO (2004a) defines labour informality empirically in terms of five criteria: ●

●

Regularity status. A value of 1 is given if a person is in regular wage labour, whether full time or part time, or in registered self-employment, 0 otherwise. Contract status. A value of 1 if the person has a written employment contract (more than 12 months), 0 otherwise.

Subcontracting and homework in the value chain 69 ●

●

●

Workplace status. A value of 1 if the person works in or around a fixed workplace, be it an enterprise, factory, office or shop, 0 otherwise. Employment protection status. A value of 1 if the worker is protected against arbitrary dismissal or entitled to severance pay, 0 otherwise. Social protection status. A value of 1 if the worker is entitled to paid medical care, whether paid by the employer or by medical insurance, 0 otherwise (ILO, 2004a).

As we will discuss shortly, the homework examined in this book suffers from a very high degree of informality on most counts – and consequent insecurity. We now turn to a discussion of the value chain in the sectors studied. We examine in particular the relationship between the subcontractor and the homeworker by country. The questionnaire sought information as to whether the homeworkers had verbal or written agreements, whether the subcontractor delayed payments for work or engaged in other malpractices, and whether workers frequently switched between subcontractors or had stable, long-term relationships with them. We also investigated the question of the share of homeworkers in the final value (or price) of the product they produced. To avoid repetition with the chapters in Part II of the book, we describe here the value chain in just one sector (incense sticks) in one country (India), though the value chain is usually similar for most product groups considered in this research: 1 2

3 4 5

The value chain begins with the homeworker. Incense stick rollers are usually women who operate mainly from their home. The raw incense stick manufacturer (subcontractor/contractor) supplies raw material to the homeworkers as well as collecting the finished product from them. The perfumed incense sticks manufacturer perfumes and packages the sticks. The dealer or stockist distributes the perfumed sticks to wholesalers or semi-wholesalers (semi-wholesalers also do some retailing). The retailer sells the final product.

This value chain, of course, is valid for the domestic market, but if the product is exported then there is an international element to the chain that starts beyond the wholesaler (no. 4 above), since the wholesaler is dealing with the exporter or the domestic agent of the international buyer. We should clarify upfront that the set of studies in this book do not examine the foreign part of the value chain in the case of products that were exported.10 However, the domestic part of the value chain is indeed very much the focus of this chapter, and its consequences for the work conditions for homeworker families. In our studies, there are several sectors involving buyer-driven chains in international subcontracting – in manufacturing (e.g. leather goods in Thailand and carpets in Pakistan) and in agriculture (hybrid seed production in Thailand and okra production in the Philippines). Here we report some chain maps for some selected sectors that are analysed further in Part II of the book. These diagrams help us to visualise the place of the homeworkers within the chain.

70

Santosh Mehrotra and Mario Biggeri

Clearly what emerges is not only their level of isolation and weakness within the chain but also their distance from the consumers, especially when international consumers are involved all of which contribute to worker insecurity. Here we present graphically examples of chains from each of our countries. In India it is clear that at least two of the products – incense sticks (Figure 3.3) and embroidered garments (zardosi) – have a large export market, and this market has grown in the 1980s and 1990s. In Pakistan, too, carpets, incense and shrimps are three export products and their growth has lead to greater subcontracting. Figure 3.4 shows that carpet weaving involves several different stages of production that are open to specialisation and to vertical and horizontal subcontracting. Similarly, Figure 3.5 (on rattan in Indonesia) shows a typical example of an international chain involving local producer units at different levels. The presence of homeworkers increases substantially the level of flexibility of supply in a buyer-driven chain. Finally, in the Philippines okra production involves a value chain (see Figure 3.6) serving other countries (especially Japan); thus introducing into the local context a growing activity in agriculture involving a value chain (Oxfam, 2004a). Another international value chain is presented in Figure 3.7 – involving in the production of leather shoes and bags in Thailand. Since these are fashion

Homeworkers

Subcontractor

Factory

Wholesaler

Local consumers International traders

Retailers

International consumers

Figure 3.3 Value chain map – incense stick making in India. Source: India Country Study Report, Sudarshan et al., 2001. Note Double arrow suggests a two-way link.

Homeworkers

Washing

Subcontractor Tucking

Lachi tani

Cone making Wholesaler/exporter

Local consumers International traders

International consumers

Figure 3.4 Value chain map – carpet weaving in Pakistan. Source: Pakistan Country Study Report, Khan et al., 2001. Note Double arrow suggests a two-way link. (a) Large/medium – scale unit

Subcontractor 1

Subcontractor 2

(b)

Homeworkers C Exporter/buyer (for export market)

O N S

Homeworkers

Intermediary

Employer

Broker/ intermediary

U M E R

Figure 3.5 Production (a) and market (b) chain maps and involving intermediaries in rattan industry in Indonesia. Source: Pakistan Country Study Report, Oey-Gardiner et al., 2001.

72

Santosh Mehrotra and Mario Biggeri

Wholesaler/retailers

Wholesalers in Japan, for example Wari, Masuko, Matsuda, Ishima

Japanese Management Company, for example ANT Associates Management Company

Exporters, for example Agcrop, Agri (Japanese)

Lead subcontractor

Subcontractor farmers (share of net income = 50%) hw

Daily paid sorting house manager

Daily paid overseers

Daily paid workers Planters, weeders, pickers, haulers

Daily paid applicators of fertilizer/pesticide

Daily paid irrigation pump worker

Piece rate sorters, quality controllers/ packers

Figure 3.6 Subcontracting chain in okra production in the Philippines. Source: Philippines Country Study Report, Rosario del Rosario et al., 2001.

products, mass production of a few designs is a risk-prone strategy in a fashion conscious market characterised by shifting demand. Manufacturers usually do the cutting and patterning work themselves, while the human skills required entail subcontracting sewing work to homeworkers. Thai female workers are generally skilful sewers able to do refined craft work and the wage is lower than in neighbouring countries (e.g. Hongkong, Taiwan, Malaysia, Singapore, Korea). But Thailand cannot compete with these countries in terms of leather craft design.

Subcontracting and homework in the value chain 73

Foreign ordering agency

1 Manufacturer

Workers

2 Manufacturer

First subcontractor

Workers

Second subcontractor

Workers

3 Non-manufacturing contractor

Frist subcontractor

Homeworkers

Figure 3.7 Relationship between foreign ordering agency, manufacturer, contractor, subcontractor and workers in Thailand leather sector. Source: Thailand Country Study Report, Arunotai et al., 2001.

Hence, the industry must accept orders from brand-name companies and produce under license. Many brand-name leather products – Pierre Cardin, Gucci, Tiptop, Jacob – are produced in Thailand.11 Worker insecurity in the value chain Worker insecurity is manifested in the value chain in many forms: verbal contracts; delayed payments; general maltreatment; low piece-rates, resulting in low share in value added. Despite the multiple forms of exploitation, the relationship between the homeworker and the subcontractor remains a stable one (an issue we return to at the end of the section). If one can broadly distinguish between two types of hired labour, casual and long term (hired on a 12-month plus contract), then homeworkers are more like casual workers in that they are paid on a piece-rate,12 but they are unlike casual workers in that they usually have longer term stable relationships with subcontractors, which are far from permanent. In fact, they have none of the benefits that usually come with formal contracts, including being paid on a time basis (e.g. an eight-hour day), rather than a piece-rate basis. Another feature of homework (e.g. zardosi in India, batik work in Indonesia) was its seasonality of output based on varying demand, since firms often subcontract out production at times of high season. Although clustered, homeworker households tend to work in isolation. If a homeworker is not a member of an association her isolation is as extreme as her vulnerability. The lack of contact with the final consumer is another weak point

74

Santosh Mehrotra and Mario Biggeri

for homeworkers; this characteristic they share with a factory worker, but then the latter does not work in isolation. As already emphasised, knowing how and by whom a chain is governed helps to understand the distribution gains among firms along the chain and how possibly to intervene to obtain social protection since firms are vulnerable to public opinion in their home countries (McCormick and Schmitz, 2002). Verbal contracts, not written The insecurity of the home-based workers is well captured by the fact that like most informal sector activities, very few of them in our five countries have written contracts with their subcontractors (Table 3.1). In India all contracts are unwritten. In Pakistan the share is close to 90 per cent. In Indonesia the conditions of work are bound by verbal contracts in over 93 per cent of the cases in two of the three sectors, and 83 per cent of the cases in the third (rattan). In the Philippines over 90 per cent in all sectors had only verbal agreement. In Thailand, a similar situation prevails in two of the three sectors. The homeworker and subcontractors are often neighbours and acquaintances (or even kin), and work relations are based on mutual trust. However, hybrid seed production is unique among all five countries in having written contracts in three-quarters of the cases; the relationship here is highly formalised with a variety of conditions laid down. This is the case because hybrid seed production by farmers is undertaken on the basis of a sale (of seeds)–purchase relationship.13 In hybrid seeds, where seed companies are involved, a written contract is common between farmers and seed companies. The latter specify the terms and conditions of contracted farming and the quality of seeds which farmers have to deliver as a final product. It is important here – in the context of the ILO Convention’s definition of homework (see Chapter 1) – that the contract states that seed companies are buyers and not employers; that farmers are sellers and not employees, and that companies are not bound by labour protection laws in this arrangement. On account of this de jure buyer–seller relationship between the farmer and company, the farmer cannot be strictly seen as a homeworker; de facto the farmer is really an employee of the company, yet the company precludes the possibility of being recognised as such since the written contract states they are buyer and seller. However, by and large the seed farmers (homeworkers) are paid on a piece-rate – in this case by kilogram of seed. In Indonesia most homeworkers stated, as we noted earlier, that their arrangements with their contractors are either oral or they made no special arrangements. This is not surprising, since in the villages studied, homeworkers and their contractors were usually relatives or neighbours – a situation that prevailed in the South Asian sectors as well.

Delays in payment In India, in two sectors, in over nine out of ten cases there are no delays in payment; but in one sector (zardosi) there are delays in nearly four out of ten cases.

Subcontracting and homework in the value chain 75 Table 3.1 Homeworkers contract – verbal or non-verbal? (%) Sector

Type of contract Written

India Incense stick making Bidi (MP + TN) Bidi (MP) Bidi (TN) Zardosi All Pakistan Incense stick making Carpet weaving Sack stitching Prawn peeling All Indonesia Pottery Rattan Batik All Philippines* Home décor Pyrotechnics Okra Fashion accessories All Thailand** Paper products Leather crafts Hybrid seeds production All

Not written

Total

0 0 0 0 0 0

100.0 100.0 100.0 100.0 100.0 100.0

100.0 100.0 100.0 100.0 100.0 100.0

3.9 10.4 22.7 10.8 11.9

96.1 89.6 77.3 89.2 88.1

100.0 100.0 100.0 100.0 100.0

7.0 17.1 4.6 9.6

93.0 82.9 95.4 90.4

100.0 100.0 100.0 100.0

9.3 5.0 0.0 7.7 5.8

90.7 95.0 100.0 92.3 94.2

100.0 100.0 100.0 100.0 100.0

1.0 1.0 76.2 25.9

99.0 99.0 23.8 74.1

100.0 100.0 100.0 100.0

Source: UNICEF survey. Notes * Did you sign any contract with person who ordered products from you the first time? ** Source: Country report.

In Indonesia, as regards delays in payment, almost all workers seemed to be paid in full in the batik and rattan sectors; in pottery three-quarters said they were paid in full, while a quarter noted delays in the last payment. In Thailand, in all three sectors, about a quarter of workers interviewed stated that there were delays in payments received for work performed. Malpractices by subcontractor A third of the women in Pakistan felt that the subcontractors engaged in malpractices – even though they usually came from the same community or

76

Santosh Mehrotra and Mario Biggeri

neighbourhood. They supplied poor quality materials, miscounted, measured wrongly or arbitrarily rejected products. It was still interesting that nearly half (47 per cent) of the women felt that they had a good working relationship with the subcontractor, 30 per cent felt that it was indifferent and only 23 per cent thought it was bad. In India, the survey finds that homeworkers tend to remain with the same subcontractor, even in the absence of written agreements or benefit payments or delayed payments. The inertia in seeking alternatives is accounted for by the cost of change (lack of alternative or excess supply of labour), debt bondage, delayed payments and contractors preferring to keep workers isolated. All these aspects contribute to raising the opportunity costs of change. Homeworkers’ share in final product price What proportion of the final value of the product goes to the actual producer? We get some idea about this from data on India, Pakistan and Thailand. In India we find that between the homeworker at one end of the spectrum, and the retailer at the other end, there are usually four or five intermediaries. The number of intermediaries and the bargaining strength of the homeworker has implications for the share of the final consumer price that accrues to the homeworker. There may also be sector-specific technology-related factors that determine the homeworkers’ earnings including the time involvement, and the cost of the material compared to labour cost. For a commodity that costs Rs 100 to a consumer, the homeworker receives Rs 15 in zardosi; Rs 17 in bidi, but only Rs 2.3 in incense sticks (the latter is said to be so low because the cost of perfuming the incense stick is high). The retailer obtains high shares in zardosi – as with most high-valued garments and clothing products. In bidi and incense sticks, manufacturers obtain high shares, primarily because brand-names are important in these industries. The fact that bidi-workers are organised may be partly responsible for the relatively higher share of homeworkers in bidi, compared to their share in the other two product groups. In Pakistan (as in India), the value chain analysis suggests that the heavy reliance of homeworkers on the subcontractors limits their earnings. In Pakistan for incense sticks (a sector in common with India), the homeworkers’ household gets Rs 5 for making 1000 agarbathis; the retailer sells one thousand agarbathis for Rs 6000. In other words, the worker gets 0.083 per cent of what a consumer pays. For carpets, the homeworker apparently earns 18 per cent of what the final consumer pays for one square foot of carpet. In prawn peeling, a home-based worker’s share is 2.5 per cent of what a consumer pays in the domestic market. This gives an idea of the different products but for a correct comparison the hours of work per unit of product should be considered. In Thailand in saa paper making, the homeworker may be receiving some 20 per cent of what the consumer pays for a product (e.g. small notebook) in a low-end market (e.g. flea market), but only 5 per cent of the consumer price in a high-end market (e.g. for foreign tourists).

Subcontracting and homework in the value chain 77 Homeworkers in the Philippines are paid on a piece-rate basis or on the basis of the activity performed. As in the other countries, workers receive a negligible share of the final price of the product. For example in the pyrotechnic sector while distributors earn almost 50 per cent of the final price, homeworkers receive 4 per cent of the final price. Stability of relationship between subcontractors and homeworkers – a means of control Despite the exploitation by the intermediaries and the insecurities of the homeworker, several of the country studies point out that the relationship between each homeworker and subcontractor was quite stable, and lasted a reasonably long time. One can wonder what might account for the stability of the relationship between parties that have no formal written contract. Although subcontractors are just above the homeworkers in the value chain, it is important to point out also that intermediaries are often skilful organisers and coordinators. In a certain sense they are the corner stone of the subcontracting system. The employers use them to reduce transaction costs and form a cluster (network) of homeworkers. They are also used to train the homeworkers on specific skills or just to transmit the directives (from the principal firm) related to the production. In almost all the studied sectors, that both homeworker and sub-contractor belonged to the same community, and were often neighbours (e.g. Pakistan, the Philippines, Indonesia), if not kin was relevant to explain the control. The relationship between homeworkers and sub-contractors in the Philippines is mainly based on patronage and kinship ties. In some cases this favoured homeworkers. For example in the fashion accessories sector, thanks to kinship ties, homeworkers had access to a credit line. In Thailand the means of control exercised by the company over the hybrid seed producing families was to appoint a ‘co-ordinator family’, which was one of the seed-producing families. But it did not seem as though this family was in a conflictual relationship with the rest of the families; the role was essentially liasing between the producer households and the contractor. In Indonesia, the case studies found that by and large homeworkers maintain ties with mainly one contractor. In rattan and pottery sectors, however, it is possible to work for two contractors (though less so in batik). The contractors usually use the intermediaries/subcontractors to maintain close ties with the workers. In India and in Pakistan, debt bondage (especially in carpets and incense sticks) seemed to be a means of maintaining control by contractors over the workers – so that the latter do not switch between contractors. If the contractor is the only available source of credit, the earnings from homework are used as a form of repayment – another insidious form of control, which in some ways is intrinsic to the way the subcontracting is organised. The provision of raw materials at home and the collection of the output by the subcontractor, often claimed by the homeworker as an advantage of home-based work, also, however, precludes

78

Santosh Mehrotra and Mario Biggeri

the development of inter-worker relations, blunts the prospects of unionisation, ensures ignorance of the market, and keeps workers from diversifying sources of work. Thus, in the relationship between the contractor and the homeworker, several types of forms of control emerge, which also account for the stability of the relationship between the contractors/subcontractors and the homeworkers. The forms of control – apart from the low level of education of workers, the diffused poverty and the limited alternative employment – can persist due to one or more of the following reasons. 1

2

3 4

Limited information available even at the local level and ‘negative’ social capital: homeworkers and subcontractors belonging to the same community, and often being neighbours, if not kin (e.g. kinship ties and the belief ‘that contractors were not much better off than them’, as many respondents in our studies noted). The role of the middleman was essentially liasing between the producer households and the principal firm, which implies that homeworkers have no contact either with the main contractor or with the final consumer. The isolation from other workers since the subcontractor provides raw materials at home and then collects the output. Other forms of control are even stronger. In certain cases, for example debt bondage – so that the latter do not switch between contractors.

Policy has to take into account this stability of the relationship on the one hand, while trying to minimise its exploitative elements (low piece-rates, the low share in value, delayed payments and physical abuse in certain cases). In fact, one of the vexed questions in this regard is: how does one identify the specific firm for which the homeworker works? The current national and international standards for measuring ‘status of employment’ treat the intermediary – the contractor – who supplies raw materials and receives the finished goods against payment for the work done as the ‘employer’. However, in the context of the value chain, it is less than clear which firm should be considered as the employer of the homeworker: the intermediary that directly places work orders to the homeworker, the supplier that puts out work to the intermediary, the manufacturer that outsources goods from the supplier, or the retailer that sells the goods. One could argue that the intermediaries can potentially assume a central and positive role in the dynamics of the local system of development when they shift from an exploitative to a facilitator role.

3.3

Final remarks

Both domestic and international forces have tended to increase the role of subcontracting and commodity chains, especially of buyer-driven commodity chains. Such chains have grown particularly for garments, for non-traditional agricultural products and also for non-timber forest products. All these kinds of products are examined in this book.

Subcontracting and homework in the value chain 79 Value chain analysis helps us to identify winners and losers, and can help to find ways of spreading the gains of globalisation.14 The price mark-ups that we analysed here seem to suggest that in the majority of products examined here the losers are definitely the homeworkers considering the hours worked and the level of income. There are several forms of insecurity that result from the isolation of the homeworker in the value chain. Verbal agreements about production, delays in payment, arbitrarily rejected products – any of these could be used. Yet the most striking fact about the production process is that in all countries in most sectors despite the malpractices, homeworker households reported a stability of the relationship between the worker on the one hand, and the subcontractors and contractors on the other. There were several means of control that made this stability possible. One was simply that there was surplus labour usually available, and in the absence of alternative employment for the homeworker the bargaining position of the worker was quite limited. Ties of kinship was yet another. Indebtedness to the subcontractors was another. In terms of the scale of informality/formality of work status, we had identified five criteria to determine the degree of informality. The homeworker comes out as ‘highly informal’ on the scale. As regards regularity of status, the homeworkers are part-time wage labour, but on account of the seasonality of work that charcterises outwork, the stable relationship we have just described does not necessarily amount to a relationship for the worker of work security. Even more, the method of payment – piece-rate – undermines both the regularity as well as income security. Hence, we have to give a value of 0 for this criteria (where 0 defines informality, and 1 defines formality). For contract status, we have already seen that for most homeworkers there is no written employment contract, let alone a contract lasting more than 12 months. Hence a value of 0 is appropriate. For employment protection status, the homeworker is not entitled to either severance pay or protected against arbitrary dismissal. So a value of 0 on this criteria as well. As regards paid medical care, there is none – hence a value of 0 again on the social protection status. The outworker does indeed have a fixed workplace – the home (though not enterprise, factory, office or shop, as listed in the ILO criteria). The homeworkers – most of them – would get a value of 1 here. The method of simply adding the five elements means homework be regarded as highly informal – on a scale of 0 to 5, with 5 being totally formal (all criteria met) and 0 being totally informal. And of course, the more informal the work status, the higher the insecurities that the worker and her family suffer from. In order to spread the gains of globalisation and reduce insecurity and marginalisation of the workers, policies at different level, as describe in Chapter 11, are necessary. As explained in Chapter 1, two-sided interventions – on the productive side and the social protection/basic social services side – are needed because they can be considered part of the local system of development. Four main actors (as we shall see in Chapter 12) can be involved in action for change. These are (1) International community and UN agencies through the creation of international codes and standards; (2) Government level (including Local government) to create national legislation; (3) Cluster level and household (including MBOP and

80

Santosh Mehrotra and Mario Biggeri

trade unions) collective actions; (4) Actions by consumers. However, given the diversity of sectoral activities, and the physical isolation of the workers from each other, the task is a challenging one; the one source of hope, however, is that most of these activities tend to be clustered, and that may ease the task slightly. The central task will involve reducing to some extent the degree of vulnerability and insecurity of the worker, and that will be difficult to do without these workers having some form of social insurance. That is a task for which the critical responsibility rests with the national government.

Notes 1 According to Humphrey and Schmitz three types of governance exists: network (co-operation between enterprises with more or less equal power); quasi-hierarchy – independent enterprises with a leader in the chain defining the rules and others subordinated; and hierarchy – an enterprise owned by other enterprises (Humphrey and Schmitz, 2000; Giuliani et al., 2004). 2 In certain countries state policy may lead to an environment where outsourcing is encouraged. For example, in India, where the state had laid down a restriction on large firms on expansion of existing production capacity, it becomes incumbent on large firms to subcontract production to smaller firms (who may further subcontract to households). 3 ILO (2004a) rightly make a distinction between ‘work’ and ‘labour’. ‘Human beings, men and women equally, live and develop through their working lives, variously combining income earning activities with reproductive activity, ideally caring for their family, neighbouring and reaching out to the broader community and society . . . Work seen in this rich sense is much more than conveyed by the narrow notion of labour, associated with wages and a position of subordination tin the productive system’ (p. 14). 4 See Dasgupta (2001) on conceptual and statistical issues related to employment security in the informal sector. 5 Several scholars have contributed to value chain analysis. Gereffi’s contribution (of 1994a) has enabled important advances to be made in the analytical and normative usage of the value chain concept, particularly because of its focus on the power relations which are imbedded in value chain analysis and in particular the governance issues or the process of co-ordinating activities in a particular chain (Kaplinsky et al., 2001). In addition, Porter had earlier distinguished two important elements of modern value chain analysis (Porter, 1985). He drew the distinction between different stages of the process of supply (inbound logistics, operations, outbound logistics, marketing and sales and after sales service), the transformation of these inputs into outputs (production, logistics, quality and continuous improvement processes), and the support services the firm marshals to accomplish this task (strategic planning, human resource management, technology development and procurement). Porter complements this discussion of intra-link functions with the concept of the multi-linked value chain itself, which he refers to as the value system. 6 One of the most researched chains is fashion garments (one reason, in fact, why we consciously excluded garments from the sectors that were going to be studied in the five Asian countries). See for instance Rao and Husain, 1987; Delahanty, 1999; Unni et al., 1999; Baden, 2001 and Balakrishnan, 2002. For Africa see McCormick (2000). 7 Gathering the raw material is difficult as the largest concentration of saa trees (mulberry) are to be found deep in the forests of North Thailand where the bark is collected by villagers and hill-tribe people. Traditionally the paper has been used for Buddhist scripts, temple decorations at festival times, umbrellas, fans and kite making. Today it is also used for a wide variety of domestic and industrial purposes including

Subcontracting and homework in the value chain 81

8

9 10 11

12 13 14

paper towels in hospitals, wrapping and gift paper, wallpaper, lampshades, bookmarks and greetings cards. Oxfam (2004a) argues that companies increasingly hold up their ‘codes of conduct’ to assure the public that they care about labour standards down the chain. But their farm and factory audits focus on documenting the labour problems that exist, ‘without asking why those problems persist . . . But one root cause, long overlooked, is the pressure of retailers’ and brand name companies’ own supply-chain purchasing practices, undermining the very labour standards that they claim to support’ (p. 7). This shift happened at the 17th International Conference of Labour Statistics (ILO, 2004a). This was because this set of studies was focused on the conditions of work of homeworkers and their children, rather than on the external value chain of the product. Three groups of leather workers were selected: (i) individual family homeworkers in slum communities around Bangkok; (ii) homeworkers as organised groups with support from government agencies (in Pathumthani province); and (iii) small unregistered factory doing subcontracted work (with workers paid by the month). Not all casual workers need be paid on a piece-rate; they could be paid on a daily rate, depending upon the nature of the activity. Since these are genetically modified plants, the farmer must buy the seeds each year from the ‘selling’ (or subcontracting) company, for which the farmer produces. The value chains are becoming increasingly global and more difficult to map since some relevant information is difficult to obtain (Carr and Chen, 2002); however this type of approach is still at an early stage and there are margins for improvement (McCormick and Schmitz, 2002).

4

Homeworkers Santosh Mehrotra and Mario Biggeri

In Chapter 1 we noted that there is a low road and high road to the development of local systems through micro-, small and medium enterprises. The low road is characterised by limited interaction and specialisation (especially vertical) among enterprises in the local system, limited action by the local government, limited institutional development, little cooperation between enterprises, and so on. The high road is characterised by different actors in the local system cooperating, innovating, and competing, thus producing a collective efficiency of the system; this process along the high road also features the involvement of associations of producers, labour and local government. Some of the industrial clusters in the low road and most of the high road clusters can bring to the local system human development and poverty reduction (Nadvi and Barrientos, 2004). However, as we argue in Chapter 11, the experience of most of the clusters in the selected Asian developing countries involving homework does not display even the characteristics of the low road to development. For this reason we called it the ‘dirt road’, which brings no human development. This is primarily because the subcontracting process, while offering work to those hitherto not in the labour force, offers very little else. The subcontracting is simply driven by the desire of firms to cut costs to the barest minimum. Meanwhile, there is very little government action to ensure social protection for the workers.1 In fact, policies, as we discuss in Chapter 11, cannot stop at the level of homeworkers. Indeed, the situation is more complex since homeworkers’ clusters are often part of a wider local system of production for example, a larger cluster composed by Small and Medium Enterprises (SMEs). The objectives of this chapter are to examine the social and economic conditions of the women homeworkers and to analyse the differences in the characteristics of homeworker households and the control group in the five countries. Homeworker households have some characteristics which can have a negative or positive impact on the well-being of women. The next section (Section 4.1) presents the main economic and demographic features of homeworker households: the house, which is also the place of work; its characteristics such as ownership of house and durable goods within it; and the basic services available. In the second section (Section 4.2) we examine gender issues since we think that the women, who represent the majority of homeworkers, can be agents for enhancing

Homeworkers 83 the human development of the household. In particular, this section focuses on the main features of homework, on the conditions of work, on their problems and priorities and on their participation in joint or collective action. Health issues are analysed in Section 4.3 and econometric estimates are used to understand the factors influencing the health status of women workers. Another round of econometric analysis is devoted to estimating the factors that may improve the homeworkers’ incomes. At the end of the chapter (in Section 4.4), we summarise the main findings.

4.1

The social and economic profile of the households

Families below the poverty line The survey results highlighted important differences in terms of socio-economic level of homeworker households between South Asia and South East Asia. In South Asia homeworking is a survival activity to stave off destitution. This is reflected in the higher share of homeworker households living below the poverty line compared to national level. In India, more homeworker households are below the poverty line than the average for the population of that state.2 The same trend is present in Pakistan where 60 per cent of homeworker households were living below the poverty line, which is higher than the percentage of households recorded as poor in urban Sindh (in 1996–97). In South East Asia, the relatively higher social and economic level of homeworker households is reflected in their income level being often above the poverty line. In Indonesia this was the case for homeworker households in two of three sectors (rattan, pottery), even though the level of income is only slightly above the poverty line. Similarly, in Thailand, one of three sectors (leather) and in the Philippines one of four sectors (home décor) have an income level well above the national and regional poverty lines. These findings suggest that homework – as well as other informal sector activities such as micro enterprises – can be a defence strategy against poverty and vulnerability, as well as a source of income diversification. However, survey results also showed that overall the social and economic level of homeworker households was slightly lower than the Control Group (CG) households. First, per capita monthly expenditure is higher for the CG households compared to homeworker households in all countries (Table 4.1). In addition, the lower socio-economic level of homeworker households is reflected in the larger proportion of homeworker households that devote more than 75 per cent of household expenditure to food compared to the share of CG households. The importance of the income generated by homeworker activities would depend on the degree of diversification of family income. The homework income in a household involved in homework can be a primary source of income or a secondary source. In many countries homework appears to be a secondary source of income, though that is not uniformly true. For instance, in India, on average across the three sectors, just over a third of all income in homeworker households comes from homework. In urban areas, however, in the zardosi and bidi sectors,

84

Santosh Mehrotra and Mario Biggeri

homework accounted for more than half of the household income.3 In Thailand this share increases to 61 per cent and in hybrid seeds sector to 80 per cent; in other words, it is definitely a primary source of income here. Size of the household The average size of household members is very high in Pakistan (7.6 members) (Table 4.1). In the Philippines it is 6.1, followed by Indonesia 5.5 and India 5.2, while it is much lower for Thailand (under 4 members). The countries are in the same rank order if ranked by the number of children per household. The demographic transition is at different stages in the five countries and this has important implications since it can influence child participation in homework. In general the CG households are smaller in size. The home (workplace) and utilities The house represents an important asset for homeworker households. Indeed the majority of homeworker households own their house. This was the case for 85 per cent of homeworker households in Indonesia, 71 per cent in India and close to 70 per cent in Pakistan (Table 4.2). Given the dual function of the house (home as well as place of work), the physical conditions as well as access to basic services have a double importance. However, a significant minority in both India and Pakistan do not own their house. Table 4.3 presents the conditions/type of the house by country. The type of the house varies accordingly to the rural/urban location. In India although houses are in poor condition more than 50 per cent have walls made of bricks/wood and another significant share are of cement. The roof is mainly made with wood, tiles or stones, or terraced with cement slabs. In Pakistan, the houses are mainly made with mud and bricks (i.e. kutcha/pukka). In Indonesia the walls are of bricks and the roofs made mainly with tiles. In Thailand the walls are of bricks (especially in urban areas) and wood (especially in rural areas) while the roof may be made of corrugated iron sheets or tiles. For the homeworker the house is a place of residence and a place of work. The access to services contributes to the welfare of the family, and for homeworker households, often to the income generating activity. This implies that improvement in housing conditions and services could directly impact productivity. The type of services available in homeworker households depends on location (piped water in urban areas and wells in rural areas). In general the services are much better in South East Asian countries compared to the countries of South Asia. Electricity is quite widely available in the houses of homeworkers. This is not surprising since homeworkers’ production is often connected to the availability of light in the place of work (Table 4.4). The share of homeworker households with electricity is very high in Thailand and Indonesia. Pakistan and India also have a high share of houses reached by electricity.

Share of hh exp. on food 75%

1.9 1.5 1.6 2.0 1.8

4.5 4.8 4.1 4.5 4.5

1.8 2.3 2.2 2.1

2.5 2.7 3.2 2.2 2.6

0.8 0.8 1.1 0.9

4.8 4.7 4.9 5.9 5.2

7.6 7.3 8.5 6.9 7.6

5.1 6.3 5.2 5.5

5.9 6.8 6.8 5.0 6.1

4.0 3.7 4.1 3.9

1185.4 2179.2 1027.1 1445.4

1043.5 446.8 97.2 405.6 527.4

76777.8 51210.0 50019.1 58743.1

546.9 551.7 602.4 629.8 580.0

389.1 510.5 278.4 237.6 331.5

— — — —

— — — — —

44.3 74.3 28.6 49.0

45.7 54.5 66.2 40.5 51.9

34.0 45.2 77.1 54.0 49.3

3.6 3.9 4.3 3.9

5.9 5.6 6.0 4.8 5.9

5.1 5.7 5.2 5.3

5.9 6.1 6.7 7.2 6.5

4.5 4.5 4.3 4.8 4.6

0.5 0.9 1.1 0.9

2.2 2.4 2.5 2.4 2.4

2.3 2.0 1.9 2.1

3.1 3.6 3.7 3.8 3.6

1.9 0.8 0.7 1.7 1.4

Number of children

Av. hh size

Per/cap. monthly hh expenditure

Av. hh size

Number of children

Non-hw households

Homeworker (hw) households

Note The per capita monthly expenditure are expressed in local currency.

Source: UNICEF survey.

India Incense stick making Bidi (MP) Bidi (TN) Zardosi All Pakistan Incense stick making Carpet weaving Sack stitching Prawn peeling All Indonesia Pottery Rattan Batik All Philippines Home décor Pyrotechnics Okra Fashion accessories All Thailand Paper products Leather crafts Hybrid seeds production All

Country/ sector

Table 4.1 Characteristics of the households surveyed

1590.8 2012.6 1031.6 1559.5

— — — — —

76562.1 58792.4 68907.1 67741.3

657.2 734.4 757.6 662.9 703.9

393.2 478.0 520.3 237.3 369.9

Per/cap. monthly hh expenditure

— — — —

— — — — —

43.3 56.7 16.7 38.9

55.6 50.0 52.4 21.7 43.9

17.0 29.2 31.8 63.5 37.9

Share of hh exp. on food 75%

Table 4.2 House ownership in homeworker households (%) Country/sector India Incense stick making Bidi (MP) Bidi (TN) Zardosi All Pakistan Incense stick making Carpet weaving Sack stitching Prawn peeling All Indonesia Pottery Rattan Batik All

Yes

No

Total

60.6 62.3 81.3 81.8 71.2

39.4 37.7 18.7 18.2 28.8

100.0 100.0 100.0 100.0 100.0

74.0 71.4 82.7 50.0 69.6

26.0 28.6 17.3 50.0 30.4

100.0 100.0 100.0 100.0 100.0

88.6 78.6 90.0 85.7

11.4 21.4 10.0 14.3

100.0 100.0 100.0 100.0

Source: UNICEF survey.

Table 4.3 Home, homework households by type of wall and roof materials (%) Type of wall of the house

India Incense stick making Bidi (MP) Bidi (TN) Zardosi All

Mud

Wood or stone brick

Cemented

Others

Total

9.0 14.3 37.3 16.2 16.9

49.0 46.7 40.0 76.4 56.0

28.4 37.7 22.7 6.1 21.8

13.5 1.3 0.0 1.3 5.3

100.0 100.0 100.0 100.0 100.0

Type of roof of the house

Incense stick making Bidi (MP) Bidi (TN) Zardosi All

Thatch

Wood or tiles or stores

Terraced or cement

Total

20.7 0.0 57.3 10.8 20.0

16.1 81.8 21.3 53.4 40.2

63.2 18.2 21.4 35.8 39.8

100.0 100.0 100.0 100.0 100.0 (Table 4.3 continued)

Homeworkers 87 Table 4.3 Continued

Pakistan Incense stick making Carpet weaving Sack stitching Prawn peeling All

Indonesia Pottery Rattan Batik All

Pottery Rattan Batik All

Thailand Paper products Leather crafts Hybrid seeds production All

Paper products Leather crafts Hybrid seeds production All

Kutcha (mud)

Pukka (Brick)

Kutcha/pukka

Total

9.1 32.5 0.0 14.9 14.2

32.5 20.8 46.7 6.8 26.7

58.4 46.8 53.3 78.4 59.1

100.0 100.0 100.0 100.0 100.0

Brick

Wood

Bamboo

Total

78.6 97.1 75.7 83.8

1.4 0.0 1.4 1.0

20.0 2.9 22.9 15.2

100.0 100.0 100.0 100.0

Concrete

Wood

Tiles

Zinc asbestos

Others

Total

0.0 0.0 1.4 0.5

0.0 1.4 0.0 0.5

98.6 82.9 98.6 93.3

1.4 2.9 0.0 1.4

0.0 12.9 0.0 4.3

100.0 100.0 100.0 100.0

Brick

Wood

Bamboo

Brick and wood

Total

24.5 77.5 19.8

55.9 15.7 47.5

0.0 1.0 0.0

19.6 5.9 32.7

100.0 100.0 100.0

40.7

39.7

0.3

19.3

100.0

Corrugated iron sheets

Tiles

Corrugate iron sheets and hatched

Total

62.4 12.7 94.1

37.6 86.3 5.0

0.0 1.0 1.0

100.0 100.0 100.0

56.3

43.1

0.7

100.0

Source: UNICEF survey.

For water facilities the situation is less homogenous, and depends on the location (urban or rural) (Table 4.5a–d). In India most of the houses (60 per cent) have piped water outside the house. In urban areas, households tend to also have water inside the house. In rural areas the problem is larger. Water facilities in the house or close to the house reduces the time needed to fetch water necessary for drinking and cooking. This is a household chore typically performed by children,

88

Santosh Mehrotra and Mario Biggeri Table 4.4 Electricity in homeworker households (%) Country/sector India Incense stick making Bidi (MP) Bidi (TN) Zardosi All Pakistan Incense stick making Carpet weaving Sack stitching Prawn peeling All Indonesia Pottery Rattan Batik All Thailand Paper products Leather Crafts Hybrid seeds production All

Yes

No

Total

69.7 92.2 68.0 85.1 78.2

30.3 7.8 32.0 14.9 21.8

100.0 100.0 100.0 100.0 100.0

85.7 84.4 98.7 67.6 84.2

14.3 15.6 1.3 32.4 15.8

100.0 100.0 100.0 100.0 100.0

98.6 91.4 98.6 96.2

1.4 8.6 1.4 3.8

100.0 100.0 100.0 100.0

99.2 99.3 100.0 99.5

0.8 0.7 0.0 0.5

100.0 100.0 100.0 100.0

Source: UNICEF survey.

usually girls, and women. In Pakistan, in the households surveyed the source of water differs for different slums; only two of them are reached by water and the households have a tap inside. In the other two they have to buy the water from a vendor. In Indonesia the water comes mainly from wells, while in Thailand piped water is very common (83 per cent). The availability of toilet facilities varies among countries and by urban/rural location (Table 4.6). The urban areas tend to have more facilities. For instance, in Pakistan, more than 90 per cent of the households surveyed in the slums have a toilet facility inside the house (nearly 60 per cent of which have a septic tank, the rest have a flush facility) – the highest share for all five countries examined here (perhaps because all the households were in urban areas). Not surprisingly, the lowest share of households with toilets inside the homes was in India. South Asia has the lowest indicators for any region in the world in respect of safe sanitation. Ownership of durable goods It is also interesting to examine possession of durable goods (Table 4.7). This is another dimension of poverty, since the absence of durable goods underlines the

6.8 5.9 34.4 15.5

Water from artesian well

Piped water

92.4 91.2 65.6 83.2

11.4 1.4 17.1 10.0

Pump

Piped

0.0 0.0 4.3 1.4

19.5 2.6 26.7 0.0 12.2

Neighbourhood pump

Tap in the household

70.1 7.8 73.3 1.4 38.3

7.1 37.7 2.7 37.8 21.5

Piped inside the house

67.1 57.1 96.0 42.6 62.2

Note * Main source for drinking water.

Source: UNICEF survey.

Thailand Paper products Leather crafts Hybrid seeds production All

Indonesia Pottery Rattan Batik All

Pakistan* Incense stick making Carpet weaving Sack stitching Prawn peeling All

India Incense stick making Bidi (MP) Bidi (TN) Zardosi All

Pipe outside the house

Table 4.5 Source of water in homeworker households (%)

0.0 2.2 0.0 0.8

Buy water from other sources

70.0 41.4 58.6 56.7

Protected well

3.9 88.3 0.0 98.6 47.5

Purchased from supplier

15.5 1.3 0.0 17.6 11.2

Handpumps

0.8 0.7 0.0 0.5

Rain and piped water

17.2 55.7 11.4 28.1

Unprotected well

6.5 1.3 0.0 0.0 2.0

Others

0.0 0.0 1.3 1.4 0.7

Dug wells

100.0 100.0 100.0 100.0

Total

0.0 0.0 7.2 2.4

Protected spring

100.0 100.0 100.0 100.0 100.0

Total

9.7 0.0 0.0 0.0 3.3

Tanker truck

1.4 0.0 1.2 1.0

Unprotected spring

0.6 2.6 0.0 0.6 0.9

Ponds

0.0 1.5 0.0 0.4

Others

0.0 1.3 0.0 0.0 0.2

Others

100.0 100.0 100.0 100.0

Total

100.0 100.0 100.0 100.0 100.0

Total

90

Santosh Mehrotra and Mario Biggeri Table 4.6 Toilet facilities inside homeworker households (%) Country/sector India Incense stick making Bidi (MP) Bidi (TN) Zardosi All Pakistan Incense stick making Carpet weaving Sack stitching Prawn peeling All Indonesia Pottery Rattan Batik All Thailand Paper products Leather crafts Hybrid seeds production All

Yes

No

Total

39.4 71.4 17.3 90.5 57.8

60.6 28.6 82.7 9.5 42.2

100.0 100.0 100.0 100.0 100.0

97.4 96.1 100.0 77.0 92.7

2.6 3.9 0.0 23.0 7.3

100.0 100.0 100.0 100.0 100.0

60.0 78.6 42.9 60.5

40.0 21.4 57.1 39.5

100.0 100.0 100.0 100.0

54.9 88.5 44.2 65.2

45.1 11.5 55.8 34.8

100.0 100.0 100.0 100.0

Source: UNICEF survey.

low potential demand of these families. Generally there is a higher ownership of durable goods in the South East Asian countries. If nearly 60 per cent of the homeworker households in Indonesia and Thailand have a radio the share drops to 19 per cent in India and 9 per cent in Pakistan. The possession of a refrigerator is very rare in all countries. In Pakistan only 1 per cent of the households have a refrigerator, in India 1.6 per cent and in Indonesia 3 per cent. Not surprisingly, 77 per cent of the homeworker households surveyed in Thailand have a refrigerator. This is a sign that such kind of activities can generate a level of income which pushes the family above the poverty line and possibly beyond a state of vulnerability and form part of the demand which can foster the local system of development. The presence of a TV in the house is very high in Thailand (93 per cent) and high in Indonesia (60 per cent). However, also in India and Pakistan, considering the level of poverty, the share is quite high in some sectors, underlining that a TV is a popular durable good and a status symbol. Surprisingly, more households have TVs than radios in all countries by and large. Finally, the ownership of a means of transport follows the same pattern. The share is high in Thailand and in Indonesia and very low in Pakistan. Bicycles,

Source: UNICEF survey.

Indonesia Pottery Rattan Batik All Thailand Paper products Leather crafts Hybrid seeds production All

India Incense stick making Bidi (MP) Bidi (TN) Zardosi All Pakistan Incense stick making Carpet weaving Sack stitching Prawn peeling All

Country/ sector

38.6 45.7 41.4 41.9

39.4 30.1 51.1

40.1

60.6 69.9 48.9

59.9

87.0 93.5 88.0 94.6 90.8

13.0 6.5 12.0 5.4 9.2

61.4 54.3 58.6 58.1

84.4 93.4 80.0 71.6 81.0

15.6 6.6 20.0 28.4 19.0

77.9

83.3 82.4 67.9

5.7 2.9 0.0 2.9

0.0 0.0 2.7 1.4 1.0

0.0 5.2 3.0 0.7 1.6

22.1

16.7 17.6 32.1

94.3 97.1 100.0 97.1

100.0 100.0 97.3 98.6 99.0

100.0 94.8 97.0 99.3 98.4

No

Yes

Yes

No

Refrigerator

Radio

Table 4.7 Durable goods in homeworker households (%)

93.2

95.5 87.5 96.9

64.3 67.1 48.6 60.0

27.3 10.4 52.0 13.5 25.7

38.3 84.4 23.9 41.9 45.1

Yes

TV

6.8

4.5 12.5 3.1

35.7 32.9 51.4 40.0

72.7 89.6 48.0 86.5 74.3

61.7 15.6 76.1 58.1 54.9

No

77.9

86.4 80.9 66.4

100.0 82.9 94.3 92.4

— — — — —

— — — — —

Yes

22.1

13.6 19.1 33.6

0.0 17.1 5.7 7.6

— — — — —

— — — — —

No

Gas stove

81.8 27.3 44.9 54.4 57.7

No

44.6

43.9 24.3 66.4

18.6 57.1 78.6 51.4

55.4

56.1 75.7 33.6

81.4 42.9 21.4 48.6

1.3 98.7 2.6 97.4 16.0 84.0 1.4 98.6 5.3 94.7 bicycle/boat

18.2 72.7 55.1 45.6 42.3

Yes

Bicycle

96.1 88.3 98.6 100.0 96.4

No

—

— — —

—

— — —

0.0 100.0 0.0 100.0 1.3 98.7 0.0 100.0 0.3 99.7 motorcycle/ speedboat 1.4 98.6 17.1 82.9 7.1 92.9 8.6 91.4

3.9 11.7 1.4 0.0 3.6

Yes

Motorcycle

99.4 98.7 100.0 100.0 99.6

No

—

— — —

0.0 0.0 0.0 0.0

—

— — —

100.0 100.0 100.0 100.0

0.0 100.0 0.0 100.0 2.7 97.3 1.4 98.6 1.0 99.0 car/ship

0.6 1.3 0.0 0.0 0.4

Yes

Cars/jeeps/vans

92

Santosh Mehrotra and Mario Biggeri

for instance, are present in more than 50 per cent of the Indonesian households surveyed, and in 45 per cent in Thailand and 42 per cent in India, while only 5.3 per cent of the homeworker households in Pakistan own a bike.

4.2

Homeworkers a cross-country analysis

In view of the two synergies mentioned in Chapter 1, women at household or micro level have a dual role (often burden) in household human development. Women’s contribution to the first synergy within the household is fundamental.4 For instance, we specified earlier how the education of a girl child not only makes her an agent of her own well-being, but also that of her entire family. Furthermore, the education of a girl also equips her to take on a non-agricultural economic activity, where productivity is greater than in agriculture.5 Women’s contribution to the second synergy is evident through their participation in homeworker income-generating activities, and from the fact that a predominant share of the total labour force is in the informal sector, which is increasingly becoming feminised. From the subcontractor’s point of view, homework is advantageous as it saves the costs of space and of other production inputs such as electricity and water. From the homeworker’s point of view, homework enables women to work and generate income while continuing to devote at least part of their time to household and care-related activities. However, what is an advantage for the contractor can be a disadvantage for the home dwellers. First, the involvement of women in homework limits the time to look after children. If the woman’s work and income increases, the impression from national accounts statistics (if the work is captured in the statistics) is that the household’s standard of living has gone up, even though the neglect of the child may lead to her falling ill. The facts, however, may be otherwise. If the women have to work excessive hours to get a modest increase in income, fatigue could ultimately lead to illness and loss of the capacity to work. Second, homeworker activity reduces considerably the availability of space and worsens the already poor conditions of the house. And third, the use of toxic materials affects the home environment. This can cause deterioration in the health of adults and children in the household even if they are not involved directly in the production activity. We return to these issues later. In this section, in order to understand better the main social and economic characteristics of women in homeworker households, we analyse their participation in the labour force (in particular homework) and the intensity of this participation (looking at the hours worked per day). Given the importance of the first synergy, we also analyse their human endowment in terms of their education level and health status. Where data permits, the information on women in homeworker households is compared with women in the CG, and at national level (using World Bank World Development Indicators Data). Finally, drawing upon the focus group discussions (FGDs) and case studies, we are going to consider the women homeworkers’ opinions about the problems encountered and the priorities regarding their job.

Homeworkers 93 Women as homeworkers As is well known, in official statistics a large proportion of women workers are not accounted for. The main reason is that women tend to be engaged in informal activities and thus are ‘invisible’ workers (also in terms of GDP or income generated). In the five countries studied the female labour force recorded is quite low, with an even lower rate of participation in South Asia (Table 4.8). In other words, female participation in the work force would be much higher if the informal activities were included. It is in informal activities that female workers are prevalent. In particular, from the UNICEF survey (Table 4.9) the ‘invisible’ becomes clearly ‘visible’ showing women as workers in homeworker households; this is also true in terms of income generated. For instance in the surveys in homeworker households the workers are mainly female. In homeworker households in India the share is between 47 and 61 per cent varying by sector, while in Indonesia from 48 to 53 per cent (and both are higher than for the relevant CG); a similar range is found in Philippines sectors (42–48 per cent). Certainly, women’s rate of participation in the labour force in homeworker households is much higher than the female share of the total labour force in the country (see col.1, Table 4.8). It is also interesting to note from a methodological point of view, that the CG in our survey in the three countries (India, Indonesia, the Philippines) mentioned above have a similar rate of participation to the one reported in the World Development Indicators (WDI) data set. We also found that in almost all homeworker households there is at least an adult woman working as a homeworker. In India the share for the members of the family engaged in homework is over 98 per cent. Similar results are found in the other countries and in other studies, including in developed countries (see for instance McCormick and Schmitz, 2002, p. 92 and for Great Britain Oxfam, 2004b). Furthermore, as already mentioned in Chapter 1, we argue that women of homeworker households tend to work more. Indeed, the fixed costs of finding a job are drastically reduced for the household members if a household is involved in a small family business (especially if residing in an area where the labour Table 4.8 Socio-economic indicators for women (1999) Country

Labour force, female (% of total labour force)

Labour force activity rate, female (% of female population ages 15–64)

Illiteracy rate, adult female (% of female ages 15 and above)

Fertility rate poorest quintile (birth per women)

India Pakistan Indonesia Philippines Thailand

32.2 28.1 40.6 37.7 46.3

29.4 21.7 38.9 31.9 55.8

45.0* 70.0 18.7 5.1 6.5

4.1 5.1 3.3 6.5 —

Source: World Bank, World Development Indicators (2001). Note * Data refers to India Census (2002).

India Incense stick making Bidi (MP
TN) Bidi (MP) Bidi (TN) Zardosi All Pakistan* Incense stick making Carpet weaving Sack stitching Prawn peeling All

Sector

215

231 117 114 357 803

124

102 183 93 502

369

437 222 215 457 1263

—

— — — —

— — — —

—

209 111 100 282 672

181

88 153 87 427

99

204 107 97 274 644

166

— — — —

—

47.8 50.0 46.5 61.7 53.2

49.1

Total Total Women Women Women workers women working working working in hw as % of total workers

Hw households

— — — —

—

90.5 94.9 87.7 79.0 83.7

84.2

Women working as % of total women

86.3 83.6 93.5 85.1

79.8

88.3 91.5 85.1 76.8 80.2

77.2

Women working in hw as % of women

— — — —

—

97.6 96.4 97.0 97.2 95.8

91.7

— — — —

—

86 47 39 84 242

72

22 25 25 94

22

86 46 40 67 218

65

— — — —

—

20 15 5 12 50

18

Women Total Total Women working workers women working in hw as % of women working

Non-Hw households

— — — —

—

23.3 31.9 12.8 14.3 20.7

25.0

Women working as % of total workers

Table 4.9 Share of women working in homeworker (hw) households by sector and in non-homeworker households (control group, CG)

— — — —

—

23.3 32.6 12.5 17.9 22.9

27.7

Women working as % of total women

120 141 109 370

74 52 49 61 236

— — —

—

203 221 174 598

146 98 106 87 437

253 233 221

707

—

— — —

65 42 44 42 193

97 115 93 305

—

— — —

54 40 39 39 172

83 103 83 269

—

— — —

44.5 42.9 41.5 48.3 44.2

47.8 52.0 53.4 51.0

—

— — —

87.8 80.8 89.8 68.9 81.8

80.8 81.6 85.3 82.4

—

— — —

73.0 76.9 79.6 63.9 72.9

69.2 73.0 76.1 72.7

—

— — —

83.1 95.2 88.6 92.9 89.1

85.6 89.6 89.2 88.2

201

65 72 64

19 16 22 33 90

59 62 72 193

—

— — —

15 10 13 22 60

38 54 43 135

—

— — —

10 9 9 12 40

21 21 33 75

—

— — —

52.6 56.3 40.9 36.4 44.4

35.6 33.9 45.8 38.9

Note Women here refers to all females equal to, or greater than, 15 years of age; for Thailand the data refers to all adults equal to, or greater than, 18 years old.

Source: UNICEF survey.

Indonesia Pottery Rattan Batik All Philippines Home décor Pyrotechnics Okra Fashion accessories All Thailand Paper products Leather crafts Hybrid seeds production All —

— — —

66.7 90.0 69.2 54.5 66.7

55.3 38.9 76.7 55.6

96 Santosh Mehrotra and Mario Biggeri market is very slack). These fixed costs are related to the availability of jobs in the area, to the social networks (or ‘social capital’, if you will) of the household within the community which gets it jobs, but also to the indivisibility of external work contracts (i.e. homework can be shared among household members, which an external work contract cannot by virtue of its being performed outside the home). In some places if social norms would prevent women from working away from home, homework is the only ‘solution’. The feminisation of homeworker activities is clear if we compare Tables 4.8 and 4.9. The share of working women in total female population is very impressive in all countries. In India, in homeworkers households 84 per cent of females work (as a percentage of all women 15 years or older) of which 96 per cent are involved in a homeworker activity (Table 4.9). In India, for instance, more than 98 per cent of homeworker households have at least one woman involved fully or part time in homework. The CG, or non-homeworker households, recorded a much lower share of working women. In sectors in Indonesia and the Philippines again the total share of women working in homeworker households is high, over 80 per cent, of which more than 88 per cent are involved in homework. As in India, the share of the working women in total women is lower for the CG households in these countries. There is an inter-generational element in the transfer of work and skills between mothers and children. Since homework is mainly a female activity girl children are more involved in helping their mother, as confirmed by the surveys. The share of daughters involved in the sector depends also on the type of product, and social norms that shape the institutional framework. In some specific sectors daughters account for 90 per cent of the household members who help women in homework. Moreover, full-time homework has two main consequences for women: one is that they are unable to spend as much time with the children as before, affecting their caring capabilities. Thus, 71 per cent of Pakistani women homeworkers interviewed said that they are unable to take care of their children as well as before they did homework. The second is that the older girls in the family are the ones who start to undertake the care responsibilities of the child, consequently affecting their own schooling. We come back to this second issue in Chapter 5. Hours worked The hours worked by women in homework varies according to season and if the homework activity is a primary or a secondary source of income. In order to generate higher incomes in South East Asia, the women work long hours – just as in India and Pakistan. On an average in India, women worked almost nine hours a day during the peak season (Tables 4.10). One sector surveyed (zardosi) is affected by sharp seasonality in demand, where in the lean season working hours may fall to just over 5 hours per day. In Pakistan, hours worked per day were in the range of 6.6–7.6 hours. In Thailand too, hours worked were reasonably high: between 8 (hybrid seeds) and 10 hours (leather crafts). In the Philippines the

Table 4.10 Average hours worked in homework by women Sector (a) India Incense stick making Bidi (MP) Bidi (TN) Zardosi All

Per day peak seasons

Per day lean seasons

7.8 9.8 10.6 8.4 8.7

5.9 — 7.0 5.3 4.6

Per day (b) Pakistan Incense stick Carpet weaving Sack stitching Prawn peeling All

7.6 6.9 6.6 6.7 7.0 Per day

(c) Indonesia* Pottery Rattan Batik All

(d) Philippines Home décor Pyrotechnics Okra Fashion accessories

(e) Thailand Paper product Leather crafts Hybrid seeds All

5.3 5.4 5.0 5.2 Per day

Per day high season

Per day lean season

15.0 — 8.0 15.0

18.0 18.0 10.0 18.0

3–6 6 1–3 3–6

Per day

% hw working seven days

9.1 9.9 8.7 9.2

78.4 34.3 95.0 69.2

Sources: UNICEF survey; (d) UNICEF, Focus Group Discussions. Notes Women here refers to all females equal to, or greater than, 15 years old. * Considering 6 days a week.

98

Santosh Mehrotra and Mario Biggeri 120 110 100

Hours of household chores

90 80 70 60 50 40 30 20 10 0 0

10

20

30

40

50

60

70

80

90

100 110

120

Hours worked during a week

Figure 4.1 Double burden for homeworker women in India.

hours per day reach 18 in three or four sectors, but can vary considerably for reasons of seasonality. Thus, in some sub-sectors the hours worked per day and per week are very high considering the other household chores of women. For instance, in Thailand most of the women seem to be working seven days a week. What we get here is a pattern of work that suggests that these are full-time, rather than part-time workers. Yet they have none of the benefits that full-time workers in the formal sector enjoy. It is often the seasonality of the work which determines their status as part-time workers. The woman’s dual role in the household’s human development is often transformed into a double burden. Indeed, to the long hours of homework should be added the hours of household chores as reported in Figure 4.1.6 The data are for Indian women. This figure clearly highlights the burden homeworker women have to carry in India, as in other countries. The total number of hours devoted to work and household chores (given by the diagonals in the figure), underline the double burden of women and consequently they imply a reduction in their capabilities and well-being.

Homeworkers 99 Educational level The educational level of women is of importance for both synergies to be triggered at the household level. Although there are major differences among countries, the education level of homeworkers is much higher in South East Asian countries. In South Asia the homeworker women are often illiterate, and their education is lower than the CG in India (Table 4.11a–e). Thus in India, half of the agarbathi makers, a third of bidi workers and over two-third of zardosi workers had no education at all. In Pakistan over 94 per cent of all homeworkers had never attended school; the share was lower in the CG of women. In South East Asia the situation was quite different. In Indonesia over fourfifths of all homeworkers had completed five years of primary school (61 per cent in batik, 84 per cent in rattan and 95 per cent in pottery). In Thailand, for instance, only a small proportion of the female homeworkers had no education. In the CG 27 per cent of the women have an education higher than primary, compared to 15 per cent of homeworker women. In the Philippines 78 per cent of women have at least primary education. The homeworker women are thus disadvantaged compared to other local women both in South Asia and in South East Asia. There is evidence from elsewhere that the least educated are the least likely to be aware of advantages that could be gained from union membership.7 Ironically, the least educated and vulnerable groups that include women are the most in need of collective voice to articulate their multiple insecurities. Health problems Women working in homework often suffer from personal health problems which are work-related (Table 4.12a–h). Health problems undermine the two synergies by reducing women’s capability to both work and take care of children. Furthermore, when the illness is due to homework (e.g. toxic materials in incense sticks or pyrotechnics), all the family members are likely to be affected by their home environment, including children. There is evidence in the South Asian cases that women do not use health facilities. In homework the share of women suffering from different health problems is much higher than in the CG. In India across the three sectors studied, half of the women felt that they experienced at least one work-related disability. In other words, about 48 per cent of women working in homework reported health problems due to work. In all sectors some 30–71 per cent of the women working in homework seemed to face homework-related health problems. Most women reported problems that were clearly homework related – shoulder pain and backache were the most commonly cited problems. For example, the process of making incense sticks (in both India and Pakistan) is hazardous in several ways. Saw dust mixed with various colours and toxic chemicals is used to make a paste. The paste, normally handled with bare hands, leads to discolouring and injury to the skin. Workers inhaled the dust and toxins causing irritation in the upper respiratory tract that eventually resulted in asthma in many cases. In none of the cases was any kind of treatment resorted to, despite awareness of the health problem.

1.0 4.4 0.0 1.7

79.2 73.9 98.6 83.4

Upper primary school

18.1 19.2 35.4 14.8 21.0

12.5 14.5 1.4 9.8

Lower secondary school

44.4 44.2 50.0 50.8 47.2

Less than primary Primary school

23.3 18.4 4.6 15.9

Junior high school

100.0 100.0 100.0 100.0

Total

15.8 22.2 23.7 8.1 14.4

From 1 to 5

5.2 2.9 0.0 3.0

Upper secondary

Completed secondary level and vocational 37.5 34.6 14.6 31.1 30.5

10.8 5.0 0.9 5.7

Senior high school

13.0 26.5 12.3 9.8 13.4

From 6 to 8

hw Households

Source: UNICEF survey. Note Women here refers to all females equal to, or greater than, 15 years old.

Paper products Leather craft Hybrid seeds All

Sector

No school

0.0 0.0 0.0 3.3 0.9

Home décor Pyrotechnics Okra Fashion accessories All

(e) Thailand

No school

Sector

(d) Philippines

60.8 59.6 55.0 58.6

5.0 16.3 39.4 19.5

Pottery Rattan Batik All

Yes some education 6.5 1.3 13.3 2.7

59.1 35.0 43.0 76.5 61.0

No school

No or less than Primary school primary

No education 93.5 98.7 86.7 97.3

68.4 43.6 55.3 78.4 67.4

No or less than primary

Sector

(c) Indonesia

Sector Incense stick Carpet weaving Sack stitching Prawn peeling

(b) Pakistan

Sector Incense stick making Bidi (MP) Bidi (TN) Zardosi All

(a) India

2.1 4.3 0.0 2.1

Vocational 100.0 100.0 100.0 100.0

Total

100.0 100.0 100.0 100.0 100.0

Total

100.0 100.0 100.0 100.0

Total

0.0 0.9 0.0 0.0 0.1

Over 13

Completed tertiary level 0.0 1.9 0.0 0.0 0.4

0.0 0.7 0.0 0.3

Academy university

12.1 15.4 21.1 5.6 11.0

From 9 to 12

At least primary school 99.0 95.6 100.0 98.3

At least primary school 81.9 80.8 64.6 82.0 78.1

At least primary education 95.0 83.7 60.6 80.5

100.0 100.0 100.0 100.0 100.0

Total 31.6 56.4 44.7 21.6 32.6

At least primary education

Table 4.11 Educational level of women homeworker (hw) households (per cent)

Paper products Leather craft Hybrid seeds All

7.9 13.0 18.6 13.3

Pottery Rattan Batik All

0.0 17.4 0.0 7.1

No school

No or less than primary

52.3 43.5 10.0 82.1 51.8

40.0 37.0 7.5 79.1 45.4

75.0 47.8 84.6 66.1

Upper primary school

57.9 55.6 62.8 58.5

20.0 37.0 22.5 4.5 19.3

From 6 to 8 21.5 4.3 40.0 6.0 16.5

25.0 13.0 0.0 14.3

0.0 4.4 15.4 5.4

0.0 17.4 0.0 7.1

Lower secondary Upper secondary Vocational school

100.0 100.0 100.0 100.0

Total

At least primary education 92.1 87.0 81.4 86.7

1.5 0.0 17.5 0.0 3.7

From 9 to 12 Over 13

Junior high Senior high school Total school 100.0 18.4 15.8 100.0 13.0 18.5 100.0 4.7 14.0 100.0 11.9 16.3

16.9 21.7 12.5 10.4 15.1

No school From 1 to 5

Primary school

No or less than primary

Sector

Incense stick making Bidi (MP) Bidi (TN) Zardosi All

Sector

Non-hw Households

100.0 100.0 100.0 100.0 100.0

Total

47.7 56.5 90.0 17.9 48.2

At least primary education

Pain in some limb

18.2 31.2 5.3 10.8 16.5

Incense stick Carpet weaving Sack stitching Prawn peeling All

63.6 45.5 33.3 48.6 47.9

15.6 5.2 5.3 5.4 7.9

Pain in joints

14.3 9.1 8.0 8.1 9.9

Eye infection, itching

12.1 7.9 20.6

1.1 0.0 21.0

General (fever, cold, etc.)

30.3

46.7

Watery eyes Poor vision

Sector

(b) Pakistan Incense stick Carpet weaving Sack stitching Prawn peeling All

Sector

(a) India Incense stick making Bidi (MP) Bidi (TN) Zardosi

Sector

2.6 1.3 — 1.4 1.3

Swlling knees

— — 4.0 — 1.0

Ear infection

12.1 42.7 24.9

13.9

Shoulder pain

Swelling in other parts of body 1.3 2.6 2.7 — 1.7

28.6 18.2 18.7 12.2 19.5

Watering eyes

58.2 0.0 28.0

9.7

Back pain

9.9 6.7 6.6

16.4

22.1 19.5 18.7 8.1 17.2

Pain in chest

18.2 18.2 17.3 8.1 15.5

13.0 19.5 12.0 6.8 12.9

stomach belly pain

67.5 35.1 36.0 31.1 42.6

Difficult Cough to see, eye strain

58.2 18.0 8.9

6.1

33.8 26.0 22.7 23.0 26.4

General body pains

7.8 5.2 2.7 2.7 4.6

Asthma

1.1 7.9 2.7

15.8 6.6 3.4 5.8

40.0

Others

6.5 13.0 10.7 9.5 9.9

Dizzyness nauser-

2.6 1.3 1.3 31.1 8.9

Skin problem

Indigestion Swelling at Gynaecothe back logical of palms problems

Table 4.12 Women homeworkers (hw) with health problems due to work (%)

2.6 6.5 1.3 33.8 10.9

Blisters

19.5 5.2 22.7 5.4 13.2

— 5.2 1.3 35.1 10.2

Skin cracking discolourisation

10.4 9.1 1.3 1.4 5.6

Breathing Stiff neck problem

7.8 14.3 — 1.4 5.9

— — 14.7 14.9 7.3

Exhaustion

80.5 67.5 30.7 6.8 46.9

Anemia

24.7 26.0 22.7 35.1 27.1

Others (specify)

22.1 11.7 9.3 6.8 12.5

Pain in legs

(Table 4.12 continued)

Muscular

79.2 57.1 61.3 47.3 61.4

Back pains

4.1 1.9 0.0 6.6 3.3

No disability

4.6 6.6 23.8

0.0 0.0 0.0 0.0

Ear infection

2.1 1.7 1.1 1.6

Skin problem

3.1 0.9 1.1 1.6

Breathing problem

0.0 1.6 0.0

Eye infection

1.4 0.0 11.5 0.0 2.9

64.0 71.1 44.9 30.8 48.2

At least one disability

8.1 19.2 42.3 11.5 18.8

2.7 0.0 1.9 0.0 1.3

100.0 100.0 100.0 100.0 100.0

Total

4.6 4.9 19.0

Skin problem

5.4 3.8 0.0 6.6 4.2

Anaemia Dizziness Allergy Got High burned by blood chemicals pressure

General mild fever, common cold, tired, etc.

2.7 3.8 26.9 23.0 13.4

(f) India, at least one disability Incense stick making 36.0 Bidi (MP) 28.9 Bidi (TN) 55.1 Zardosi 69.2 All 51.8

(e) Thailand Paper product Leather craft Hybrid seeds

Sector

(d) Philippines Home décor Pyrotechnics Okra Fashion accessories All

Sector

1.0 0.9 2.2 1.3

Eye infection

Blurring of eyes

General (fever, cold, etc.)

(c) Indonesia* Pottery 22.7 Rattan 26.1 Batik 31.2 All 26.6

Sector

Table 4.12 Continued

4.1 7.0 6.5 5.9

2.7 0.0 0.0 0.0 0.8

4.6 4.9 0.0

Breathing, pulmonary problem

2.7 3.8 0.0 0.0 1.7

0.0 1.7 0.0 0.7

Stiff neck

0.0 7.7 5.8 0.0 2.9

6.2 9.8 4.8

Stiff neck

0.0 1.9 0.0 0.0 0.4

Asthma Muscle ache

Waist stomach aches

Difficulty Bladder in urinating problems

19.6 9.6 8.6 12.5

Coughing

40.0 29.5 61.9

33.0 33.9 38.7 35.1

Dizzy

0.0 0.0 1.1 0.3

Work accident

35.1 61.5 48.1 42.6 45.6

6.8 9.6 19.2 0.0 8.4

4.6 1.6 0.0

Other

4.1 40.4 42.3 29.5 26.8

14.9 1.9 0.0 1.6 5.4

Others, specify

23.8 47.6 18.2 28.0

Others

10.8 21.3 81.0

More than one

Back and Rheumatism Headache lower backache

0.0 0.9 0.0 0.3

Shoulder pain

Back, arm, leg problem

33.8 59.6 48.1 52.5 47.3

Body aches

3.1 4.3 0.0 2.6

Back problem

No

100.0 100.0 100.0 100.0 100.0

Notes Women here refers to all females equal to, or greater than, 15 years old; Only share of those giving affermative answer is shown in the tables. * Any health problems last month.

Source: UNICEF survey.

(h) Philippines, Have you gotten sick because of work? Home décor 26.5 73.5 Pyrotechnics 30.0 70.0 Okra 41.7 58.3 Fashion accessories 30.8 69.2 All 31.7 68.3

Yes

Total

Total

Sector

Yes

100.0 100.0 100.0 100.0 100.0

No

(g) Pakistan, Do you complain of hw related health problems? Incense stick 0.0 100.0 Carpet weaving 7.8 92.2 Sack stitching 9.3 90.7 Prawn peeling 6.8 93.2 All 5.9 94.1

Sector

104

Santosh Mehrotra and Mario Biggeri

Sudarshan et al. (2001) suggest that it may be difficult to improve the health status of the workers without a holistic approach to the problem. Sector-specific problems like shoulder pain in the case of zardosi, blistering on the palms because of rolling incense sticks and inhaling tobacco dust while rolling bidis resulting in bronchial problems, poor vision due to the toxic materials used in incense stick making, can often be eliminated by introducing innovative technology. For example, an improved frame or appropriate seat to minimise or prevent shoulder pain in the case of zardosi, providing gloves to be worn while rolling the incense sticks, or a nasal filter to prevent bronchial problems are suggested. The issue is: who is to provide these, and how? Without some joint or collective action by workers who organise themselves, neither public action nor any initiative by their employers is likely. In general it would be important to have a bio-chemical analysis of the materials used in the homework activities (e.g. in incense stick making) since it would have useful policy implications. In Pakistan the urban slums where these communities lived were highly polluted and without adequate social/physical infrastructure. They suffered many ailments like respiratory diseases, pains in the muscles and joints, and serious skin irritations and allergies. In many cases, despite the health impact of the hazardous work, they did not seek any medical attention because of unaffordable or poor health facilities. Often they relied on traditional medicine or medicines that had expired, disbursed by unqualified persons. The proportion of homeworker women currently suffering a disease/ailment is incredibly high. Nearly all the women working in homework seemed to be ailing. In South East Asia the impact of homework on the health of women is less dramatic than in South Asian countries. For instance, in Indonesia and Philippines only a third of women suffer from health problems. A higher share is recorded in the hybrid seeds sector in Thailand and pyrotechnics in the Philippines. Membership of organisations: vulnerability and associations of homeworkers, producers and social protection Information was also collected about the participation of women in associations or social networks (Table 4.13a–e). Such membership is important to arrive at collective answers for common problems and for suggesting possible interventions to the government. If a homeworker is not a member of an association, her isolation is as extreme as her vulnerability. The lack of contact with the final consumer is another weak point of homeworkers. The vulnerability of the homeworkers is well captured by the fact that like most informal sector activities, almost none of them have written contracts with their subcontractors. Provision of raw materials at home and collection of the output, often claimed by the homeworker as the advantage of homework, also precludes the development of inter-worker relations, blunts the prospects of unionisation, ensures ignorance of the market, and keeps workers from diversifying sources of work. Policy has to take into account this stability of the relationship on the one hand, while trying to minimise its exploitative elements (low piece-rates, the low share in value, delayed payments and physical

Table 4.13 Membership of organisation Sector

Are you a member of any labour or trade association? Total women Yes

(a) India Incense stick making Bidi (MP) Bidi (TN) Zardosi All Sector

(b) Pakistan Incense stick Carpet weaving Sack stitching Prawn peeling All Sector

(c) Indonesia Pottery Rattan Batik All Sector

(d) Philippines Home décor Pyrotechnics Okra Fashion accessories All

Total women hwers

No

Total

Yes

No

Total

0.9

99.1

100.0

1.2

98.8

100.0

77.8 23.7 1.4 15.6

22.2 76.3 98.6 84.4

100.0 100.0 100.0 100.0

75.7 21.6 1.5 16.8

24.3 78.4 98.5 83.2

100.0 100.0 100.0 100.0

Have you ever negotiated a better rate via collective action (working together with fellow workers)?

Do you access the services of the organisation?

Would you be willing to engage in collective action to negotiate wages rates/better condition if any organisation were to assist you?

Yes

No

Total

Yes

No

Total

Yes

No

Total

14.3 6.5 16.0 13.5 12.5

85.7 93.5 84.0 86.5 87.5

100.0 100.0 100.0 100.0 100.0

1.3 0.0 10.7 1.4 3.3

98.7 100.0 89.3 98.6 96.7

100.0 100.0 100.0 100.0 100.0

46.8 49.4 53.3 62.2 52.8

53.2 50.6 46.7 37.8 47.2

100.0 100.0 100.0 100.0 100.0

Is there an organisation which can help to get better pay ?

Are you a member of worker’s Organisation ?

Saving/Arisan

Yes

Yes

No

0.0 100.0 1.3 98.7 0.0 100.0 0.4 99.6

Total 100.0 100.0 100.0 100.0

0.0 0.0 0.0 0.0

No

Total

Yes

No

Total

100.0 100.0 100.0 100.0

100.0 100.0 100.0 100.0

21.4 2.9 11.4 11.9

78.6 97.1 88.6 88.1

100.0 100.0 100.0 100.0

Patamba member Yes

No

Total

33.3 2.6 54.3 63.6

66.7 97.4 45.7 36.4

100.0 100.0 100.0 100.0

36.8

63.2

100.0 (Table 4.13 continued)

106

Santosh Mehrotra and Mario Biggeri

Table 4.13 Continued Sector

(e) Thailand Paper product Leather crafts Hybrid seeds All

Are you a member of any labour or trade association? Yes

No

Total

94.8 87.1 98.6 93.6

5.2 12.9 1.4 6.4

100.0 100.0 100.0 100.0

Source: UNICEF survey.

abuse in certain cases). From the answers to the questionnaire and the discussion during FGDs it emerges that gender adds a further source of discrimination for women to that horizontal inequality based on ethnic groups, religion or caste (Nazfinger et al., 2000). Infact, membership of such a disadvantaged group and being woman brings a double horizontal inequality. When discrimination is sharp due to an informal institutional framework women suffer from constraints on their capabilities and lack of agency. There are different types of associations such as association of producers and membership organisation of the poor. Some of them offer social protection, in terms of health, welfare and education services, others offer services for production including collective price bargaining or both, that is, access to services for production and for social protection. The membership is important to arrive at collective answers for common problems and for suggesting possible interventions to the government. In India, only few of the incense-stick makers and zardosi workers are members of an organisation. Nearly a quarter of the bidi (Tamil Nadu, TN) workers are members, while as much as three-fourths of the bidi (Madhya Pradesh, MP) workers are members. Indeed, as members of an association, they participate in welfare funds (see also Sinha, 2001). Although none of the countries surveyed have social security mechanisms implemented by the government with widespread application in the informal sector, there is at least one important exception. An important initiative that has considerably benefited bidi homeworkers in Tamil Nadu, is the tripartite body of the bidi Welfare Fund. This fund was created in 1976 and is administered departmentally through a tax levied on the production of bidi exports. The fund’s goal is to provide medical care, education for children, housing, water supply and recreational facilities for bidi workers. The proposed national policy on home-based workers formulated by the government of India’s Ministry of Labour is advocating the widespread use of such welfare funds. It is important to emphasise that in India among the adult women workers, an important difference was observed in the health-seeking behaviour of bidi workers and others. None of the incense stick makers, and less than 5 per cent of the zardosi workers, either stopped work temporarily or permanently or even sought treatment when health problems arose. In contrast, over 40 per cent of bidi (MP),

Homeworkers 107 and over 80 per cent of bidi (TN) workers, did so. This may have to do with the availability of health facilities for bidi workers as part of the Welfare Fund facilities. In Pakistan, differently from India, hardly any homeworkers are members of an organisation. However, some of them receive services such as (health, welfare, education) while others participate in collective action to negotiate better piece-rates; more than half the women in Pakistan are willing to participate in collective action. In Indonesia, none of the sectors selected had a women’s collective,8 though some of them are involved in micro-finance schemes (arisan). In the Philippines and Thailand (where too the local network of Homenet is active), a significant proportion of workers were members of a women’s collective – certainly a much higher proportion than homeworkers in India (with the exception of bidi (MP)) or Pakistan. According to women during FGDs, homework – as well as informal sector activities such as micro enterprises – can be an important opportunity not only as a line of defence against poverty and vulnerability, but also of income diversification. In addition to generating improved welfare, they can lead to the evolution of entrepreneurial abilities and small enterprises. We underline the importance of membership of an association for the human development of the household and of women homeworkers by fostering their human and economic endowments. Perception of problems The women have a clear idea of the problems affecting their households and their homework activities as well as the priorities for improving their welfare. This has emerged clearly in the FGDs and from the survey, where the women were asked about their problems and their priorities for possible government intervention. The perception and the ranking of problems obviously depend on the level of development of the socio-economic system. If the local infrastructure is underdeveloped, the family’s attention is rightly focused on survival and homework is seen as a survival strategy. This is the case in India (Table 4.14a–d). In India, according to women homeworkers, the most serious problem was that of low piece-rates. Long hours of work and the lack of benefits, with low wages, were seen as the three most important problems. Lack of alternative employment is also cited as a major problem. In Pakistan, health problems seemed to figure at the top of the list of problems in all sectors. The tiring nature of the work was ranked as the second most serious problem in all sectors, followed by low income. In the Philippines, the low/irregular pay was seen by homeworkers as the most serious problem. Lack of credit/capital to sustain production and the limited number of orders were ranked as the next two most important ones. In Thailand, a significant proportion of homeworkers felt that they had no problem. However, the women felt the health impact of the work was serious; irregular supply of work was also cited as a problem in two of three sectors.9 If the level of socio-economic development of the household and of the local areas is a bit higher, the families concentrate more on issues related to their business as micro entrepreneurs. For the priorities, the ranking depends again on the household and local infrastructure, but is focused on how to improve the homework income-generating activity (Table 4.15a–e).

—

2 3

1**

1 1

None

6 7 6 — 6

Sector

(b) Pakistan Incense stick Carpet making Sack stitching Prawn peeling All

Sector

— —

1** 1*

5 4 5 4 5

Less time for family

3 5

—

— —

—

Space

2 2 2 2 2

Very tiring

4 —

—

— —

—

Seasonality

4 3 3 5 4

Low income

3 2

—

3^ 3^

—

Lack of benefits

Low Lack of Few Many Orders Pollution in Hazards of Disorder irregular capital/ orders rejects are neighbour- work to and violence pay credit to cancelled hood health at home sustain production

3 5 3 3 3

Makes the house messy

—

Long hours

—

Low wages

(a) India Incense stick making Bidi (MP) Bidi (TN) Pernambut Vill. Bidi (TN) Jeevanagar Vill. Zardosi – Urban Zardosi – Rural

Sector

Table 4.14 Ranking of problems by women homeworkers, by sector

— 6 — — 7

— —

3

2# 4

2

Lack of alternative empl.

7 — 7 6 7

Violence in the community (incidence of rape and incest) and order 1 drug addiction

— 8 — — 8

Take up limited Others living space

— —

4

— 2

—

Children neglected

Lack of Work is social Seasonal protection and services

Long hours

— —

5

— —

—

Health problem

Lack of Added alternative burden employment requirements imposed by licensing

1 1 1 1 1

Health problems

— 4

—

— —

—

Delayed payment

Dangers from inadequate sanitation/ many mosquitoes

— —

—

— —

1

Credit

No problem

1 1 1

Sector

(d) Thailand Paper product Leather sraft Hybrid seeds

3 3 — 3

5 — — 5

7 — 3

No equipment/ Tools

4 — — 4

2 2 7

Health impact

6 — — 6

Notes a What are the main disadvantages of this work? * High rejection rate. ** Lack of work. # Unemployment among men. ^ Low awareness – facilities exist. b What is the main disadvantage of hw? c What are the problems that you experience in the work you do now?

Sources: UNICEF survey; UNICEF, Focus Group Discussions.

2 2 8 2

(c) Philippines Home dècor 1 Pyrotechnics 1 Okra 1 Fashion 1 accessories

3 3 8

Inconsistent work supplied

7 4 4 7

8 7 7 8

— 7 —

Limited work quota

— 5 3 —

5 6 4

Low price, low pay, late payment

— 6 — —

— — 2 —

5 — 6

4 5 4

Work available More only than one seasonally

— 8 5 —

— — 6 —

— — 9 —

Table 4.15 Ranking of priorities by women homeworkers, by sector and possible interventions by the government Sector

Work place

(a) India, ranking of priorities Incense — stick Bidi (MP) — Bidi (TN) — Pernambut Vill. Bidi (TN) — Jeevanagar Vill. Zardosi – Urban 1 Zardosi – Rural — Sector

Pension/ provident fund/bonus/ insurance

Social security

Min Emp

Altern.

Loan facility

Schools

Training

Health facilities

Work security

4

—

—

2

5

—

3

1

1 —

— —

3* —

— —

— —

— 1

2 —

— —

2

—

1

—

—

—

3

—

2 1

3 2

4 3

5 4

6 5

— —

— —

— —

Loans

Minimum wages/ increase in wages

Free/ subsidised schooling/ good scholarship

(b) India, possible interventions by the government (%) Incense stick 2.3 63.0 12.7 making Bidi (MP) 1.8 10.1 29.4 Bidi (TN) 5.4 3.8 13.6 Zardosi 42.5 25.9 5.2 All 14.9 26.9 13.5 Sector

Technical and marketing our training

Credit, financial assistance

Quality of education be improved to get good jobs

Evening school/ boarding school

Free/ subsidised medical treatment

0.6

0.0

0.0

0.0

0.0 15.8 2.1 5.2

0.0 13.6 0.0 3.8

0.0 7.6 0.0 2.1

20.2 6.5 5.7 6.8

Financial assistance to build the house

Assitance with forming movements organizations

Framing better policy/ laws

Others (specify)

Free education fof children

(c) Pakistan, kinds of assistance needed for hbw from government and/other organisations to improve your welfare (%) Incense stick 8.4 25.2 6.5 2.8 13.1 0.9 7.5 Carpet weaving 7.8 23.3 7.8 0.0 7.8 1.1 15.6 Sack stitching 12.5 5.7 15.9 1.1 3.4 3.4 11.4 Prawn peeling 5.6 4.4 20.0 0.0 2.2 0.0 16.7 All 8.5 15.2 12.3 1.1 6.9 1.3 12.5 Sector

Ensure a steady flow of orders

Increase piece rates and lower prices of commodities

(d) Philippines, ranking of priorities Home décor 1 2 Pyrotechnics — — Okra 1 2 Fashion 1 2 accessories Sector

Market skill training

Specialised skill training

(e) Thailand, assistance needed (%) Paper product 2.1 11.5 Leather craft 2.9 5.7 Hybrid seeds — 10

Have more access to credit

Improve educational facilities and support for children

Organise to raise piecerates through dialogue workers/ subcontractors

Prevent undue competition from foreign – made firecrackers

3 2 — 3

4 — 3 4

— 1 — —

— 3 — —

Accounting skill training

Help with credit line

Help with marketing

Help with forming org.

Help with legal and welfare protection

Others

— 1.4 —

2.1 8.6 14.3

45 2 21

6.3 8.6 4.3

4.2 24.3 7.1

46.9 38.6 60

(Table 4.15 continued)

Provision for housing hw

Employment Regular in activities employment other than hw

Encouragement and recognition of the hw

Pre school with supplementary feeding

Steps against the middleman

Other facilities

No response

Total

12.7

7.5

0.0

0.0

0.0

0.0

0.0

1.2

100.0

18.3 10.3 0.0 9.3

13.8 9.2 0.0 6.8

0.0 0.0 10.9 3.2

0.0 0.0 2.6 0.8

0.0 10.3 0.0 2.9

0.0 0.0 0.5 0.0

6.4 0.5 2.1 1.8

0.0 3.3 2.6 2.0

100.0 100.0 100.0 100.0

Provide employment

Provide safe drinking water and improve sewage facilities

Provide sewing machines

Provide health facilities goods

Govt. does nothing

Well paid hw and market

Provide electricity

Don’t know

Noting

Total

12.1 11.1 12.5 6.7 10.7

2.8 0.0 2.3 7.8 3.2

0.9 2.2 1.1 0.0 1.1

0.9 1.1 3.4 4.4 2.4

0.0 1.1 0.0 1.1 0.5

0.9 3.3 4.5 10.0 4.5

0.0 0.0 0.0 2.2 0.5

1.9 8.9 13.6 13.3 9.1

15.9 8.9 9.1 5.6 10.1

100.0 100.0 100.0 100.0 100.0

Producers and workers be trained in safe and modern ways through the Pyrotechnics Regulatory Board

Alternative skills training, coupled w/ credit and market access

Make representations through producers’ associations

Gender sensitivity training and other community based interventions

More roads, schools, health services through community action at LGU level

Improve health and sanitation

Prevent drug addiction and violence in the community, especially violence against women

— 4 — —

— 5 — —

— 6 — —

— 7 — —

— 8 — —

— — 4 —

— — 5 —

Sources: UNICEF survey; UNICEF, Focus Group Discussions. Note * Including jobs for men.

Table 4.16 Summary statistics Variable India WHS AGE EDU NONEDU NCHILD Y_PC Y_H H_YEAR EXPWK ORGBIDI NONORGBIDI E_BEN HOUSEOWN HOUSESUR ELECTR UPPER NONUPPER URBAN RURAL ZARDOSI Pakistan WHS AGE EDU NONEDU NCHILD EXP_PC Y_H H_YEAR EXPWK ORGACC NONORGACC CLACTION

Definition

Mean

Std. Dev.

Number of health problems age  1 if woman is literate  1 if woman is illiterate Number of children Household income per capita Income per hour in homework Total hours worked in homework in a year Years of work experience in homework  1 if she is a member of an organisation or of the bidi sub-sector  1 if she is not a member of an organisation or of the bidi sub-sector  1 if she is receiving benefits from employer  1 if home is owned by the household  1 if home surroundings are in bad condition  1 if house has electricity  1 if household is upper caste  1 if household is not in the upper caste  1 if household is urban  1 if household is rural  1 if household is in the zardosi sub-sector

1.27598 28.44804 0.3867121 0.6132879 1.144804 4857.147 3.978155 1640.752

1.578055 10.75012 0.4874121 0.4874121 1.43618 3076.207 2.153705 579.3295

13.43441

10.96366

0.3117547

0.4636053

0.6882453

0.4636053

0.3679727

0.4826652

0.7444634

0.4365344

0.2947189

0.4563051

0.8449744 0.2078365 0.7921635

0.3622377 0.4061052 0.4061052

0.4957411 0.5042589 0.4207836

0.5004083 0.5004083 0.4941059

Number of health problems age  1 if woman is literate  1 if woman is illiterate Number of children household expenditure per capita Income per hour in homework Total hours worked in homework in a year Years of work experience in homework  1 if she has access to the services of an organisation  1 if she has no access to the services of an organisation  1 if she engages in collective action for better price

4.458746 32.31023 0.0594059 0.9405941 4.481848 621.8683 3.213025 2142.612

1.895651 7.521897 0.2367739 0.2367739 1.702793 285.8823 3.778316 980.0412

11.5636

7.326395

0.0330033

0.1789407

0.9669967

0.1789407

0.1254125

0.331734

Homeworkers 113 Table 4.16 Continued Variable

Definition

Mean

Std. Dev.

E_BEN

 1 if she is receiving employers benefits  1 if home is owned by the household  1 if household’s home is in bad condition  1 if house has electricity  1 if household is in the incense (agarbathi) sub-sector  1 if household is in the shrimp peeling sub-sector

0.0330033

0.1789407

0.6963696

0.4605856

0.7326733

0.4432964

0.8415842 0.2541254

0.3657345 0.4360887

0.2442244

0.4303371

HOUSEOWN HOUSECON ELECTR INCENSE PRAWNPE

Views regarding priority interventions The workers were also asked about their views on the priority interventions that the government should take to help them. In India, social security and regular work were seen as the important interventions; minimum wage and health facilities were also identified as priorities. Possible government interventions are (ranked in the following descending order): loans, pensions, provident fund, insurance and finally, minimum wages. In Pakistan, credit/financial assistance were seen as priorities; free education for children and alternative employment were also cited as high priorities. In the Philippines, the workers felt that the most important thing was to ensure a steady flow of orders; a higher piece-rate and access to credit were also seen as important potential interventions. In Thailand, marketing and skill training was seen as the most important intervention, followed by help through a credit line, and help with legal and welfare protection. In the two sub-sections that follow, we focus on the factors influencing the health problems and remuneration of women homeworkers.

4.3

Health and economic issues related to women homeworkers: some empirical evidence for India and Pakistan

In this section we present some empirical evidence on factors that affect the health and influence the income of women homeworkers. In particular, we are interested to test the role of human endowments and collective action. The data utilised in the analysis are from the India and Pakistan surveys.10 The definitions and the main characteristics of the variables used in the empirical analysis are described in Table 4.16.

114

Santosh Mehrotra and Mario Biggeri

The health of women homeworkers The dependent variable, chosen for the determinants of the health of women working as homeworkers, is the number of health problems which is considered as proxy for a woman’s health status (WHS). Women were asked if they had health problems due to homework. We found no reason or incentive for the respondents to misrepresent their perception of their state of health in India and Pakistan. In Pakistan the questionnaire section on health was very detailed and submitted by a doctor. Different kinds of diseases were recorded as reported in Table 4.12. The dependent variable for each woman was built on the number of diseases and varies from 0 (no health problems) up to 7 (the maximum of health problems recorded either in India and Pakistan). When the dependent variable consists typically of nonnegative count data we can obtain reliable estimates using Poisson regression model. The model specifies that each yi is drawn from a Poisson distribution with parameter
i which is related to the regressors, xi.11 It is worth noting that under relevant assumptions regarding the dependent variable the ordered logit model could be used as well. The results of the ordered logit are not reported simply to avoid duplication since they are very close to the Poisson regression model results reported here (Mehrotra and Biggeri, 2002b).12 A woman’s health is a basic human capability (Nussbaum, 2000). It may depend upon various factors such as personal characteristics, household characteristics, human endowments and economic endowments, type of activities involved, status of local sanitation, access to services and institutional factors. The independent variables we consider are thus the following: woman’s age (AGE), number of children (NCHILD), literacy dummy (EDU), hours worked in a year (H_YEAR), type of job (i.e. dangerous/non-dangerous) (sector), membership and access to services of an organisation (ORGACC in Pakistan and ORGBIDI in India, that is, either a member of an organisation or in the organised bidi sub-sector), household income (or expenditure) per capita (Y_PC; EXP_PC), household assets (HOUSEOWN), environmental conditions of the surroundings of the house and house conditions (HOUSESUR and HOUSECON) and other institutional aspects such as employers benefits (E-BEN) and membership of specific castes (UPPER in India), or urban location (URBAN in India). Therefore: WHS  f [AGE, NCHILD, EDU, H_YEAR, sector, ORGACC (or ORGBIDI in India), Y_PC (or EXP_PC), HOUSEOWN, HOUSESUR (and/or HOUSECON), E-BEN (plus UPPER and URBAN)]. A Poisson regression model is used for estimation of the coefficients for each country separately (Table 4.17). The estimates are carried out using the Huber/White/Sandwich estimator of variance to obtain robust variance estimates. The incidence ratios are estimated and reported in Table 4.17. Although with some difference the results obtained are quite consistent between India and Pakistan. The most relevant results of the research are the positive effect on health status of education and membership access to services,

0.0000245 0.0031089 0.1071063 0.0000839 0.0154271 0.1998077 0.0726089 0.0518521 0.0586414 0.04243 0.2558807

2.73 0.39 3.05 0.85 0.19 1.68 0.56 0.39 1.95 7.53 2.42

1.16 2.46 1.97 1.14 1.21 3.30 3.71 1.33 2.74 2.97 2.55 4.55

z

0.006 0.700 0.002 0.396 0.851 0.094 0.576 0.699 0.052 0.000 0.015

0.246 0.014 0.049 0.256 0.228 0.001 0.000 0.183 0.006 0.003 0.011 0.000

P|z|

0.000019 0.0072921 0.1165218 0.0002356 0.033136 0.056602 0.1016544 0.0815457 0.0007762 0.2362855 0.1179037

0.0001152 0.0048944 0.5363707 0.0000932 0.0273372 0.7266297 0.1829672 0.1217106 0.229094 0.4026081 1.120938

0.0000659 0.9999042 0.0024281 0.0215347 0.0012312 0.4181301 0.0000139 0.0000523 0.3768873 1.154396 0.1307411 0.5142642 0.0559348 0.1809813 0.0649769 0.3403622 0.1006374 0.6071433 0.1073847 0.5237824 0.1308205 1.002417 1.971807 0.7845611

[95% Conf Interval]

1.000067 0.9988018 1.386034 0.9999288 0.9971048 1.39796 1.041494 1.020285 1.12093 1.376366

0.0000825 1.012053 1.233284 1.000019 0.1374167 1.380579 1.12576 1.147623 1.424599 1.371059 1.762298

IRR. Err.Std.

0.0000245 0.0031051 0.1484529 0.0000839 0.0153825 0.2793231 0.0756217 0.0529039 0.0657329 0.0583992

0.9997426 0.004933 0.1311643 0.0000169 0.9141763 0.1350749 0.035912 0.1186696 0.1840768 0.145642 0.3918471

Robust

Notes India: Number of obs  444, Wald 2 (11)  90.52, Prob  2  0.0000, Log pseudo-likelihood  743.42772, Pseudo R2  0.0743. Pakistan: Number of obs  294, Wald 2 (10)  99.02, Prob  2  0.0000, Log pseudo-likelihood  602.38645, Pseudo R2  0.0409.

0.0000671 .0011989 0.3264463 .0000712 .0028994 0.3350139 0.0406564 0.0200825 0.1141589 0.3194468 0.6194206

Pakistan H_YEAR AGE NONEDU EXP_PC NCHILD NONORGAC E_BEN HOUSEOWN HOUSECON INCENSE cons

Robust Err.Std.

0.0000825 0.0048742 0.1063537 0.0000169 0.1190377 0.0978393 0.0319002 0.1034047 0.1292131 0.1062259 0.2223501 0.3028744

Coef.

India H_YEAR 0.0000958 AGE 0.0119814 NONEDU 0.2096806 Y_PC 0.0000192 E_BEN 0.1435777 NONORGBIDI 0.3225027 NCHILD 0.1184581 HOUSEOWN 0.1376927 HOUSESUR 0.3538903 URBAN 0.3155835 NONUPPER 0.5666186 cons 1.378184

HEASUM

Table 4.17 Determinants of women homeworkers, health status: Poisson regression model

1.000019 0.9927344 1.123582 0.9997644 0.967407 0.9449701 0.9033417 0.9216906 0.9992241 1.266536

1.000066 1.002431 1.001232 0.9999861 1.45774 1.139673 1.057529 0.9370891 1.105876 1.113362 1.139763

1.000115 1.004906 1.70979 1.000093 1.027714 2.068099 1.200775 1.129427 1.25746 1.495721

0.0002575 1.021768 1.519118 1.000052 0.0897319 1.672408 1.198393 1.405457 1.835181 1.688402 2.724859

[95% Conf Interval]

116

Santosh Mehrotra and Mario Biggeri

controlling for other relevant variables. The illiteracy of the women negatively influences health status, increasing the possibility of being sick. This is in line with the logic of the first synergy.13 In particular, we found poor health status significantly affected by illiteracy of the women EDU, the non-access to an organisation or to services NONORGAC (in India the membership of bidi sector as well NONORGBIDI), the environment in terms of unhealthy surrounding HOUSESUR and bad conditions of the house HOUSECON as well as by the ‘dangerous’ sectors as INCENSE for Pakistan14 and for India the URBAN location and the non membership in an upper caste (NONUPPER). All these variables have high incidence rate ratios (IRR). For instance in India we estimate not being a member of an organisation or part of bidi sub-sector (NONORGBIDI, that is, not receiving any social protection) increases the average number of health problems and in particular as emphasised by the IRR this is 38 per cent greater than for members of such organisations (39 per cent in Pakistan) controlling for the other variables. The same can be said for non-literacy with a IRR of 23 per cent in India and 38 per cent in Pakistan. The variables with the largest value of IRR include HOUSESUR, URBAN and NONUPPER in India and HOUSECON and INCENSE in Pakistan. The unhygienic environmental conditions of the surroundings of the house influence the health status negatively and strongly.15 This is obviously a problem that needs a collective response. Urban location of the households also negatively impacts health, perhaps on account of the congested slum areas in which homeworkers live. Also the focus group discussions highlighted the fact that incense stick making is dangerous for the health. A woman involved in that activity has a greater possibility of being sick. There is also a significant and negative effect of the number of hours worked in a year in Pakistan on health status. The more hours a woman works increases the possibility of being sick in Pakistan. Given the conditions and the scarce access to health centres women are probably working even when they have health problems. In India this independent variable is insignificant and with a negative sign. Then, the higher is the age (positive and significant) of the woman the higher is the probability of health problems in India although the coefficients and the IRR are very low. In Pakistan this variable is non-significant. These different results are probably due to the age range of the Pakistani women surveyed, between 15 and 55 years while in India the age is up to 65.16 The number of children a woman has also influences her health negatively in India (the level of the IRR is low). In Pakistan this variable is insignificant. Both in India and in Pakistan the effect of income per capita (expenditure per capita in Pakistan) as well as the proprietorship of the house are non-significant. It appears from the results that the second synergy seems to be not relevant at these levels of incomes and assets. The employers’ benefits (other than wages) given to the women are found insignificant, but we also know that the workers receive very little non-wage benefits.

Homeworkers 117 Table 4.18 Determinants of women homeworkers, earnings (productivity in value per hour): OLS regression – robust standard errors LNY_H

Coef.

Indiaa LNAGE 0.1342614 EDU 0.107561 LNEXPWK 0.0154471 LNHEASUM 0.0016151 ORGBIDI 0.1731382 E_BEN 0.224919 ELECTR 0.1863376 ZARDO 0.313275 RURAL 0.1055355 cons 0.2636352 Pakistanb LNAGE 0.2276596 EDU 0.7766068 LNEXPWK 0.0290051 LNHEASUM 0.0322365 CLACTION 0.4835748 E_BEN 0.4017346 ELECTR 0.2352908 PRAWN 1.585801 cons 0.185485

Robust Err. Std.

t

P|t|

[95% Conf. Interval]

0.0982046 0.0642953 0.0060508 0.0026628 0.061693 0.0622849 0.0824981 0.0824745 0.0674076 0.3483711

1.37 1.67 2.55 0.61 2.81 3.61 2.26 3.80 1.57 0.76

0.172 0.095 0.011 0.544 0.005 0.000 0.024 0.000 0.118 0.450

0.0587411 0.0187991 0.0035553 0.0036182 0.0518923 0.1025099 0.0242034 0.151187 0.0269413 0.4210217

0.5346369 0.3250192 0.0364008 0.0137488 0.2600774 0.4416156 0.5834764 0.4702188 1.772542

0.43 2.39 0.80 2.34 1.86 0.91 0.40 3.37 0.10

0.671 0.018 0.426 0.020 0.064 0.364 0.687 0.001 0.917

1.27986 0.824541 0.1369476 1.416266 0.0426341 0.1006443 0.0592949 0.005178 0.0282746 0.9954242 0.4673939 1.270863 0.9130291 1.383611 0.6603795 2.511223 3.302995 3.673965

0.3272639 0.2339211 0.0273389 0.0068483 0.2943841 0.3473281 0.3484719 0.475363 0.2380122 0.948292

Notes a Number of obs  455, F (9, 445)  5.55, Prob  F  0.0000, R2  0.0984, Root MSE  0.57198. b Number of obs  303, F( 8, 294)  2.89, Prob  F  0.0041, R2  0.0419, Root MSE  3.5327.

Women homeworkers’ earnings (productivity in value) and collective action Having examined the determinants of women’s health status (as one side of the synergies), we also explore the determinants of their earnings. This is done in order to understand if ‘alternative’ factors affect (other than capital and labour, the traditional neoclassical factors) women’s productivity in value and thus income and wealth. For comparability purposes, the dependent variable used is the income per hour of work Y_H. The productivity of a woman is theoretically related to various factors such as her age, the level of education (formal education), the years of experience (informal training) in the case of skill-involving processes, the personal health condition, membership of an organisation to pursue joint actions and input availability. The starting functional form utilised is an unconstrained Cobb–Douglas production function. Natural logarithms are used to transform and linearise the functional relationship – the transformation reduces the overall variability of the

118

Santosh Mehrotra and Mario Biggeri

data and thus the heteroskedasticity at the cross section level. The productivity function basic model becomes: ln (Y_Hi)  C
1 ln (AGEi)
2 D (EDUi)
3 ln (EXPWKi)
4 ln (WHSi)
5 D (ORGBIDIi)
6 D (E_BENi)
7 D (ELECTRi)
8 D (sectori) As reported in Table 4.16, AGE is the age of the woman, EDU is a dummy for her education/literacy, EXPWK are her years of experience/work in the activity, WHS is the woman’s health status, ORGBIDI is a dummy for membership of an organisation in India while CLACTION is a dummy for participation in collective action to obtain better price conditions for Pakistan. Then, we introduced E_BEN as a dummy for receiving employers benefits and ELECTRD is the dummy for electricity in the house, and ACTD is a sector activity dummy. The data utilised in the analysis are analysed for the two countries separately. Ordinary Least Square (OLS) estimates are carried out using the Huber/White/Sandwich estimator of variance to obtain robust variance estimates. The results are reported in Table 4.18. In both countries education and joint or collective action have an important role in determining earnings. The results indicate that the education of the woman as well as her membership in a women’s association of homeworkers (and in the bidi sector for India) influences the earnings positively and quite heavily. Joint actions (collective action for obtaining better prices in Pakistan) seem to produce positive effects in the productivity expressed in value per hour. The age of the women is insignificant.17 Both in India and Pakistan specific comments are needed. In India work experience, the employers’ benefits and having electricity as input positively affects the productivity as expected. The impact of health status is negative (as predicted) but insignificant. The women workers in zardosi have a higher productivity. In Pakistan the rest of the results indicate that the health status is negative and significant: more serious is the woman’s illness the lower is her productivity. The years of experience, the input available (electricity), and the employers’ benefits, although positive, are non-significant. Women workers in prawn peeling have a higher productivity.

4.4

Concluding remarks

We had noted in Chapter 1 that while it is impossible to establish quantitatively that subcontracting and homework have been expanding in the countries under review (since we do not have baseline data for earlier years), there is prima facie evidence to suggest that the phenomenon has been growing. Further we had noted in Chapter 3 in the value chain analysis that the homeworkers were working under conditions which contributed to their insecurity – with workers obtaining a small share of the total price paid by the consumer of the product they produce.

Homeworkers 119 In this chapter we have noted that they work long hours, especially in the high season, for low piece-rates (with delays in payment in many cases). The insecurity arises essentially due to the fact that the workers are isolated, and even though they live and work in a ‘cluster’ there is little collective action. In fact, the relationship between the employers/subcontractors and the homeworkers is a stable one – characteristic of a long-term stable workforce for a single employer – yet there is exploitation by the higher levels of the value chain and still no social protection. Second, we found that most of the homeworker households in the clusters in South Asia live below the poverty line, though in the clusters examined in South East Asia, that is the case for only some of the sectors. Most of the clusters in South East Asia have incomes above the national poverty line. This suggests that it is possible for homework to be the launch pad for upward mobility. With the requisite support from government in terms of both social protection as well promotive action, the inter-generational transfer of poverty and the low-capability equilibrium trap can be broken. It is relevant to remember that, as we discuss in Chapter 11, the homework clusters are often part of a larger system of production. Third, the women of homeworker households are more likely to work in homework. This finding is confirmed by comparing the data with the control group. Since they are also involved in household chores the double burden is likely to negatively impact their well-being and capabilities. Fourth, [the feminisation of the work has important implications for the gender dimension of a household’s human development cycle from generation to generation. Girls can play a key role in breaking the inter-generational transfer of poverty. Often in income-generating activities male children tend to follow in the father’s footsteps, while the female children those of the mother. Since homework is mainly a female activity girl-children are more involved in helping their mother, as confirmed by the surveys.] In some specific sectors daughters account for 90 per cent of the household members who help women in homework. If the mother is engaged in homework, the older girls in the family are the ones who start to assume the care responsibilities for other children, affecting their own schooling. In other words, there is a strong case here for providing communitybased child care, so that the older girls can be freed from this care-giving responsibility. Alternative child care would enable the girls to go to school, and if necessary, work part-time. Fifth, the low educational level of the homeworkers, and the health-related problems faced by them, indicated that without public interventions, the first synergy (between basic services) cannot be triggered in the case of these families. This finding is confirmed by the econometric results, examining the determinants of health status in India and Pakistan. In particular, it is important to underline the positive role of education in health status, and the negative effect of dangerous activities such as incense. Finally, collective action – or joint action as in the case of Pakistan – by the homeworkers will be needed if their earnings are to rise. This was confirmed by the econometric analysis results for India and Pakistan. Collective action and

120

Santosh Mehrotra and Mario Biggeri

organisational membership have a positive influence on women’s earnings. This calls for government support to relevant collective actions. The education and work experience have a positive influence on women’s earnings. Poor health status influences earnings/productivity negatively. Hence public action by the state is required. In conclusion, we are arguing that homework can promote human development of a household, but also foster local economic development, through an increase in the human capital of the local labour force and through the development of micro-enterprises and SMEs. Subcontracted home manufacturing work is a phenomenon to be found in both rural and urban areas in Asia – in contrast to the situation in Latin America (the other region in the developing world where subcontracting has become widespread).18 The fact that – driven by cost considerations – industrial and agricultural outwork has grown with a view to capturing export as well as domestic markets, and spread even to rural areas, seems to reinforce our suggestion in Chapter 1 on cluster theory. A cluster is a group of producing businesses in the same sector of production located around the same place. A cluster may also include suppliers of raw materials and specialised machinery, as well as traders dealing in the final product. Clusters of homeworkers develop because employers and intermediaries need to reduce transaction costs associated with input and raw material distribution, output collection, contract enforcement and the dissemination of information. The theory predicts that either the presence of a specialised skill in a particular geographical area, or simply the presence of a concentrated body of unskilled surplus labour can provide the basis for subcontracting growth to homeworkers. In brief, the challenge for policy is to minimise the vulnerability of workers in homework, while supporting the elements that ensure efficiency in production. In other words, the challenge is to limit the trade off between efficiency and equity. What needs to be done to at least reach the low road of local system development in the selected countries is encouragement by the government of some co-operative action by homeworkers. In addition, there is considerable scope for government action in a number of spheres (an issue we return to in Chapters 11 and 12).

Notes 1 As regards the state performing a role in the promotion of these economic activities, only in Thailand has the State played a promotive role (primarily because the sector was perceived as a foreign-exchange earner through exports). See Chapters 10 and 12. 2 The exception is agarbathi makers in rural areas, who happen to be located near the large industrial city of Bangalore, where wage employment is better paid, so the menfolk have been able to find better employment, thus increasing the homeworker household’s total family income. 3 In urban areas, homework accounted for 54 per cent of the household income among zardosi workers and for 70 per cent of the income among bidi workers in Tamil Nadu. 4 According to Moser the woman has a ‘triple role’: the reproductive role (household), the productive role (work) and the community managing role (community networking) (Richey, 2000). See also UNIFEM (2000b).

Homeworkers 121 5 Elsewhere, we have demonstrated that in developing countries that achieved high health and education indicators even at low levels of income, women not only achieved parity in education levels early, but also had high participation rates in the labour force. In other words, women were agents of change in terms of both our synergies. See Mehrotra and Jolly (1997) and UNDP (2003) for a further discussion. 6 Since homeworker women spend much time at home in isolation, if they do not get organised with other women, they cannot build solidarity and trigger community management. Participation is fostered by community-based organisations as we discuss later in this chapter. 7 Thus, ILO (2004a) notes, on the basis of worker surveys conducted in many countries: ‘Another worrying feature in South Africa, which is approximately replicated in other countries, is that the least educated are the least likely to have an attitude towards unions . . . making them the least likely to join or to participate. Women are far more likely to be unaware’ (p. 253). 8 In Indonesia, homeworkers in other parts of the country are indeed members of the local network of Homenet International. 9 No information is available for Indonesia. 10 As mentioned in the methodology (Chapter 2) the results are valid at state level for India and for urban slum areas in Pakistan. Indonesian data were not used because the question asked was too general, not specifying if the health problem were related to the homeworker activity.  y 11 The primary equation of the model is the following: Prob (Yi  yi)  e
i
i i/yi! , yi = 0, 1, 2, ..., 7 and ln
i  ␤'xi. (Poisson maximum-likelihood regression). 12 Ordered logit models are used to estimate the relationships between an ordinal dependent variable and a set of independent variables. Applying this type of model we are assuming (for the dependent variable) that the more the number of health problems indicated in her response, the unhealthier we consider the woman worker to be. We are making no judgement here about the intensity of a disease, or how the functioning of a worker is affected by any particular disease afflicting her; only that more the number of aliments afflicting her, the more unhealthy she is. 13 As reported later in Chapter 5 the higher the education of the mother the lower the conditional probability of her child working. 14 This is not tested in India because of the variable ORGBIDI. If INCENSE is used (instead of ORGBIDI) it has a negative effect on health status, as in Pakistan. 15 The interviewer had to make a judgement about whether the surroundings of the worker’s home were: dirty, average or clean. The dummy is constructed such that dirt is 1, while the other two are 0. 16 In Pakistan the questionnaire was administered only to one woman per family working in homework; in India, information was obtained from all the women homeworkers in the family. It is the latter phenomenon that explains the larger age range in India. Age squared (non linearity) is found in both countries to be non-significant and was thus omitted. 17 In both countries the interaction between health and education are found positive but not significant. According to FGDs and case studies this could be due to the vulnerability and the necessity to work even if in bad health conditions seems to be a particularly strong argument in Pakistan. 18 In recent studies in Latin American countries, homework is shown to be a predominantly urban phenomenon. A second contrast between Asia and Latin America is in respect of the sectoral distribution of homework: in Latin America it prevails in the services sector (personal and repair services, the promotion or sale of goods and services like insurance, credit cards, courses, travel) in all the six countries reviewed, while most of the Asia studies (including ours) suggests that the majority of homeworkers are in the manufacturing of goods. A final area of contrast is that in Latin

122

Santosh Mehrotra and Mario Biggeri

America men’s participation in industrial homeworking is much more significant than women’s (with the sole exception of the garment industry); women are the majority of workers in low-productivity, low-paid service activities in Latin America (Tomei, 2000). The fact that workers in Asia are mainly involved in production of goods in both rural and urban areas may well provide a good basis for pursuing a growth strategy based on public action to support micro-enterprises.

5

Child labour in homeworker households Santosh Mehrotra and Mario Biggeri

The literature on child labour burgeoned in the last decade of the twentieth century, as negotiations for the Uruguay Round of trade were closing in the early 1990s. International trade negotiators brought child labour on to the trade agenda as a ‘new issue’, linking labour standards to further market access in industrialised countries. The international funding of research for child labour expanded. International organisations created and expanded new programmes for addressing the data requirements in the area. Position papers were hurriedly prepared by international organisations, and evaluations of current programme activities were conducted to determine whether child labour needs were being addressed. Programmes and projects to address the problem multiplied, as did funds from bilateral and multilateral agencies.1 Too often in the child labour literature that has grown in the aftermath of this new concern for a rather old problem, the tendency is to address the issue as a stand-alone problem. The link was usually drawn to general poverty as a cause and child labour as the consequence of poverty. However, before long the discourse descended into ever greater refinement of issues of child labour itself, with scant attention to the forces at work in the macro-economy which characterised the expansion of the informal sector, and within it, the emergence of new forms of child labour in the development process. Most child work in developing countries is in the informal economy in household enterprises, either working for the family (of which the child is a member) as unpaid family labour, or as a paid employee outside the home in another, often unregistered household enterprise (for the definitions of child labour and child work see the Annex 5.1 at the end of the chapter). The focus in this chapter is on the phenomenon of child work and child labour within homework in manufacturing in the informal sector in the five Asian countries studied. Homework in manufacturing can easily lead to the employment of child labour, especially when the head of the household procures the raw materials from the contractor or the employer and performs the work at his home with family labour. There are two prima facie reasons to believe that such work may be on the increase. First this form of child labour will not incur the penal provisions of a country’s laws banning child labour. Both India and Pakistan instituted legislation banning child labour within the last couple of decades, but homework is not

124

Santosh Mehrotra and Mario Biggeri

covered by such legislation. Second, one possible result of the increasing national and international attention to child labour is to drive it underground, that is, away from the visible factories or workshops to home-based economic activity.2 Where a migrant labour force moves into urban areas, children – who were earlier working on farms as family labour – now begin to participate with their parents in such informal activities in manufacturing. Manufacturing activities could also be subcontracted to households still based in rural areas. Where the returns to schooling are likely to be high, as in a rapidly growing economy, there is less of a tendency for such informal manufacturing work, whether home based or not, to be undertaken by children. Under such circumstances, parents prefer their children to go to school. However, in economies – such as India and Pakistan – where economic growth has been less rapid, the returns to school are perceived to be lower, and the opportunity cost of schooling perceived to be higher, the incidence of participation of children in such work would tend to be higher. Even in rapidly growing economies (e.g. in Thailand, Indonesia), school-going children are engaged in home-based manufacturing, together with their family, after schools hours.3 Data on the scope and magnitude of child labour4 is quite limited in most countries,5 and information about the scale of home-based child work is even scarcer. This makes home-based productive activities of women, and especially of children, ‘invisible’, at least for policy-makers. This chapter, and the country studies in Part II, attempt to fill a gap in the child labour literature by focusing exclusively on child labour in homework activity. Section 5.1 presents a framework for the analysis of child labour in homeworker household. The lack of human capital and economic endowments among households where children work is likely to lead to an inter-generational transfer of human poverty and to child labour, as we discussed in Chapter 1. Section 5.2 goes on to present the findings related to child work in homework for the countries examined. Section 5.3 attempts an empirical analysis of the determinants of child labour status and of hours worked, using the raw data from the surveys in India, Pakistan and Indonesia. The concluding section summarises the findings.

5.1

Homeworker households and child labour

The child labour issue has to be addressed by starting with the child and her household, her household’s ‘history’, and the local external environment. The phenomenon of child labour has also to be located within a view of inter-generational transfer of poverty and its impact on child labour status.6 The static view has to give way to a dynamic view of child labour. In other words we argue that this dynamic view, as also a more holistic view of the child labourer within her family context, could be a first step in understanding the child labour phenomena in its complexity, consistent with more pro-active development policies in favour of the poor. The starting point is that each child in poor households in developing countries has often a ‘burden’ to carry from her birth. A child has a strategic role in these

Child labour in homeworker households 125 types of households. She has almost the same responsibility as an adult in the household’s survival and in case of necessity has to contribute to family income. This is similar to the experience accumulated by her parents during their lives as children and passed on from generation to generation. This reflects the absence of available choices, and underlines Sen’s notion of ‘development as freedom’ (Sen, 1999). In other words the child becomes a victim of inter-generational transfer of poverty, caught in a vicious circle, often impossible to overcome by the households themselves. Thus, the child’s activity status (only working, or working and studying, or only studying, or neither working nor studying) is a determinant of the human development of the household from generation to generation, as we discussed in Chapter 1.7 It is the parents who decide about the child’s daily life, about the future, about work, chores and schooling. Following the literature, the status of the child could be one of four possible situations: only working, only studying, working and studying and neither working nor studying. The parents’ decisions on the activity status of the child can be analysed in terms of opportunity costs for the parents. Therefore, as we analyse in Section 5.3, the child activity status (CS) depends on many factors, including the household’s demographic characteristics, its factor endowments in terms of human capital, assets and income, as well as its culture and religious beliefs and the institutions that form the boundaries of social life as well as external factors. Homeworker households present some specific characteristics. Indeed, we argue that industrial and agricultural outwork has a specific impact on children and women. In respect of children we assert that, in general, households incur fixed costs both in sending their children to work and to school. For instance, some of the fixed costs are related to the travel costs (both direct and indirect) that the child should bear to reach a place where she/he can be employed or she/he can attend school (plus the usual fixed costs such as books, meals, stationary, uniform, etc.). Other relevant fixed costs are the transaction costs in which parents are involved (travelling for finding a job or queuing for daily jobs). In the case of work these fixed costs are related to the availability of jobs in the area, to the social networks (or ‘social capital’, if you will) of parents within the community which gets them and the children jobs. The fixed costs of finding work outside the home arise from the indivisibility of external work contracts due to time scheduling and constraints arising from the nature of work (i.e. homework activities can be shared among household members, which an external work contract cannot by virtue of its being performed outside the home). The cost may also take the form of parental fear that the child may be maltreated, beaten or exploited by an employer when the child works away from home. In the case of girls, parents are particularly concerned about their security, especially as they grow older (as the literature on the life of domestic female servants has emphasised). Another important point which plays an important role in the decision whether to send a child to school, to work and study, to work, or to be ‘idle’ is that the returns to external work and to school are often very low.

126

Santosh Mehrotra and Mario Biggeri

We argue that the fixed costs of finding a job are drastically reduced for the children if a household is involved in a small family business (especially if residing in an area where the labour market is very slack). Therefore, we argue that children in homeworker households, given the same conditions, have a higher probability than other children to be in the ‘only working’ or ‘working-and-studying’ categories. This is because there are fixed costs associated with sending a child to work outside the home that would offset the returns to that work. The fact that labour is available at home reduces fixed costs in finding a job for children, changing the opportunity cost of the parents and thus the decision to send children to school and/or to work. Furthermore, the opportunity cost of sending female children to school is higher, since being at home they can be involved in household chores. Observe that the returns to work are influenced by the age and by the sex of the child. As age increases the returns to (manual) work increase as well, and in general the returns of male children tend to be higher than for female. The returns and the fixed costs to work are influenced also by the labour market and by the institutional framework. The age of the child also affects the fixed costs for schooling since as the child moves beyond primary education, attending school becomes more expensive. The returns to education and the fixed costs to attending school are influenced by the quality of the school and its relevance in the development of the local economic system. Hence, we argue that children in homeworker households, given the same conditions, have a higher probability than other children to be in the ‘only working’ or ‘working-and-studying’ categories. This is because there are fixed costs associated with sending a child to work and these should offset the returns. The fact that labour is available at home reduces fixed costs in finding a job for children, changing the opportunity cost of the parents and thus determining the decision to send children to school and/or to work. Therefore, we later argue also that child labour can be reduced in homeworker households with preventive measures that affect the opportunity costs and thus parents’ decisions on the status of the child. Such measures include measures affecting the child’s incentive to go to school but also collective actions of homeworkers and government interventions to increase social protection and adult incomes. We stress the generation to generation transfer of poverty (or its reversal) because some policies which may seem as an incentive to increase child labour temporarily actually help to reduce household poverty (or increase the household human development) and child labour reduction in the medium run. Indeed, the increase in the assets (e.g. land, a new machinery) of the family, which may generate a temporary increase in child labour, is an important strategy to reduce child labour. Cigno et al. (2002) argue that land redistribution may increase the incidence of child labour, since smallholder plots are intensively farmed, and the intensity of labour input rises with the size of the farm up to a certain farm size. As they put it in proposition (iv): Land redistribution could increase the incidence of child labor because it would increase the productivity of labour in households that receive

Child labour in homeworker households 127 additional land and reduce it in those from which land is taken away. This would increase the probability of work for children in lower-income households and reduce it for children in higher-income households where parents are unlikely to make their children work anyway. (Cigno et al., 2002, p. 59) If it is true that marginal value of child labour increases in the short period, the proposition (iv) and the related policy implications are wrong, since it analyses the problem in a static context. Apart from the fact that lower income households are likely to make their children work anyway (and that quite often the parents work their own land and the child is sent to other farms to work), the proposition does not take into account the human development of the household as well as the dynamic context in which the family operates. Indeed, in a dynamic context, such land redistribution can, all other things remaining equal, increase family human development through income and through new options (simply by the value of the land added, which could be rented) and later, if better or modern inputs can be bought, lead to a child labour reduction on the family farm; just as increasing income may lead to falling child morbidity and a higher likelihood that the child goes to school while family labour is replaced by hired labour. In the same way, policies directed towards increasing the income of homeworker households – although they may increase the marginal value of child labour temporarily – are a strategy to reduce child labour, especially if accompanied by other complementary policies.

5.2 Understanding children’s work in homeworker households In this section we discuss the incidence of child work in homework and control group (CG) households, the schooling of children in these households, the reasons why children are working/not in school, the contribution of children in terms of hours of work, their work conditions, their health status and how gender affects their time use. Finally, we narrate the views of children working – as they emerged from the focus group discussions (FGDs). This comparative discussion is based on the analysis of raw data collected during the country studies. Incidence of child work In India (as elsewhere, with very few exceptions) 15 is the legal minimum working age. A quarter of all children aged 5–14 in homeworker households work in the homework activity. The share is much smaller among the younger children (5–10 years), than among the older ones (11–14 years) – 13 per cent as against 44 per cent on average across all sectors studied (Table 5.1).8 In other words, the children start getting involved with homework as they grow older. The incidence of child work in the homeworker households is much greater than in the CG (non-homeworkers) households (Table 5.2). In fact, few children

71

84 44 40 111 266

87

75

111

50

323

136

101 46 55 128 365

170

185

142

186

683

—

—

—

—

—

20 16 4 27 56

9

—

—

—

—

—

41 26 15 79 145

25

II

I

II

Tot.

I

Children working

Total children

India Incense stick 207 making Bidi (MP
TN) 185 Bidi (MP) 90 Bidi (TN) 95 Zardosi 239 All 631 Pakistan Incense stick 257 making Carpet 260 weaving Sack 253 stitching Prawn 236 peeling All 1006

Country/ sector

6.6

—

—

—

—

—

19.8 34.8 7.3 21.1 15.3

I

—

—

—

—

—

48.8 59.1 37.5 71.2 54.5

35.2

II

% of children working

—

—

—

—

—

80.2 65.2 92.7 78.9 84.7

93.4

I

—

—

—

—

—

51.2 40.9 62.5 28.8 45.5

64.8

II

% of children not working

88

63

79

68

20 16 4 21 48

7

298

I

261

36

98

60

67

35 22 13 71 117

11

II

Number of children working in hw

55.6

52.5

63.6

53.5

52.5

29.7 42.2 17.9 38.5 26.1

8.7

Tot. 5.1

43.6

47.3

44.4

42.7

40.0

19.8 34.8 7.3 16.4 13.2

I

% of children working in hw

80.8

72.0

88.3

80.0

77.0

41.7 50.0 32.5 64.0 44.0

15.5

II

Table 5.1 Share of children working by age group in homeworker (hw) households (I  5–10; II  11–14)

44.4

47.5

36.4

46.5

47.5

70.3 57.8 82.1 61.5 73.9

91.3

Tot.

56.5

52.7

55.6

57.9

60.0

80.2 65.2 92.7 83.6 86.8

94.9

I

% of children not working in hw

19.2

28.0

11.7

20.0

23.0

58.3 50.0 67.5 36.0 56.0

84.5

II

—

—

—

—

—

90.2 90.5 89.5 86.8 82.1

52.9

Tot.

77.8

—

—

—

—

—

100.0 100.0 100.0 77.8 85.7

I

% of children working in hw of total children working

—

—

—

—

—

85.4 84.6 86.7 89.9 80.7

44.0

II

55 79 70 204

37 39 41 39

156

42

44 62

148

107 124 108 339

69 66 70 74

279

67

62 99

228

76

18 37

25

123

32 27 29 35

52 45 38 135

—

— —

—

34

6 13 11 4

8 2 3 13

—

— —

—

66

18 17 14 17

19 12 12 43

Note The children under five are not counted in this table.

Source: UNICEF survey.

Indonesia Pottery Rattan Batik All Philippines Home décor Pyrotechnics Okra Fashion accessories All Thailand Paper product Leather crafts Hybrid seeds production All —

— —

—

21.8

16.2 33.3 26.8 10.3

14.5 2.5 4.3 6.4

—

— —

—

53.7

56.3 63.0 48.3 48.6

36.5 26.7 31.6 31.9

—

— —

—

78.2

83.8 66.7 73.2 89.7

85.5 97.5 95.7 93.6

—

— —

—

46.3

43.8 37.0 51.7 51.4

63.5 73.4 68.4 68.1

11

4 4

3

31

4 13 10 4

7 2 2 11

24

4 5

15

63

16 17 13 17

13 9 7 29

15.4

12.9 9.1

26.9

33.7

29.0 45.5 32.9 28.4

18.7 8.9 8.3 11.8

7.4

9.1 6.5

7.1

19.9

10.8 33.3 24.4 10.3

12.7 2.5 2.9 5.4

31.6

22.2 13.5

60.0

51.2

50.0 63.0 44.8 48.6

25.0 20.0 18.4 21.5

84.6

87.1 90.9

73.1

66.3

71.0 54.5 67.1 71.6

81.3 91.1 91.7 88.2

92.6

90.9 93.5

92.9

80.1

89.2 66.7 75.6 89.7

87.3 97.5 97.1 94.6

68.4

77.8 86.5

40.0

48.8

50.0 37.0 55.2 51.4

75.0 80.0 81.6 78.5

—

— —

—

94.0

83.3 100.0 92.0 100.0

74.1 78.6 60.0 71.4

—

— —

—

91.2

66.7 100.0 90.9 100.0

87.5 100.0 66.7 84.6

—

— —

—

95.5

88.9 100.0 92.9 100.0

68.4 75.0 58.3 67.4

130

Santosh Mehrotra and Mario Biggeri

Table 5.2 Share of children working by age group in non-homeworker (non-hw) households (control group) (I  5–10; II  11–14) Country/ sector

Non-hw households Total children

India Incense stick making Bidi (MP
TN) Bidi (MP) Bidi (TN) Zardosi All Pakistan Incense stick making Carpet weaving Sack stitching Prawn peeling All Indonesia Pottery Rattan Batik All Philippines Home décor Pyrotechnics Okra Fashion accessories All Thailand Paper product Leather crafts Hybrid seeds production All

Children working

Tot.

I

II

I

II

66 35 19 16 69 170

46 15 10 5 48 109

20 20 9 11 21 61

1 0 0 0 0 1

2 0 0 0 0 2

43 61 70 65 239

32 46 46 46 170

11 15 24 19 69

— — — — —

— — — — —

56 47 45 148

33 28 28 89

23 19 17 59

1 0 0 1

3 2 1 6

17 19 16 26 78

11 10 10 10 41

6 9 6 16 37

0 0 0 1 1

0 0 1 4 5

14 26 21 61

5 17 15 37

9 9 6 24

— — — —

— — — —

% of children working I

% of children not working II

I

II

10.0 0.0 0.0 0.0 0.0 3.3

97.8 100.0 100.0 100.0 100.0 99.1

90.0 100.0 100.0 100.0 100.0 96.7

— — — — —

— — — — —

— — — — —

3.0 0.0 0.0 1.1

13.0 10.5 5.9 10.2

97.0 100.0 100.0 98.9

87.0 89.5 94.1 89.8

0.0 0.0 0.0 10.0 2.4

0.0 0.0 16.7 25.0 13.5

100.0 100.0 100.0 90.0 97.6

100.0 100.0 83.3 75.0 86.5

— — — —

— — — —

— — — —

2.2 0.0 0.0 0.0 0.0 0.9 — — — — —

— — — —

Source: UNICEF survey. Note The children under five are not counted in this table.

in the CG households are working. As much as four-fifths of the children working in homeworker households are involved in homework. This suggests that the fact that the parents are engaged in homework increases the likelihood that the children will be engaged in it as well and it follows the theoretical predictions.

Child labour in homeworker households 131 In households where the children were working as homeworkers, the family was asked what effect there would be on the household if the child stops working. In India, over a quarter (28 per cent) of the households felt that the household enterprise’s capacity to produce would not operate fully (i.e. would remain underutilised). As many as 57.5 per cent of the households felt that their living standards would decline; 8 per cent felt that the household cannot afford to survive without the child’s contribution to household income.9 In Pakistan the household size was larger than in other countries, and the number of children in each household was larger. Not surprisingly, a much higher proportion of children in homeworker households are working in Pakistan compared to the sectors in India – across all sectors the share is 56 per cent. The share of younger children working (44 per cent) is the highest for all the countries surveyed for this study, and the share does not vary much between sectors. As many as four-fifths of all older children are working (i.e. a much higher proportion than in India), with very little variation across sectors.10 In Indonesia there were 339 children between 5 and 14, of which 56 (18 per cent) were working – over 70 per cent of the working children in homeworker households were engaged in the home-based manufacturing activity; the rest were doing other work. By comparison, under 10 per cent of children were working in the CG households (which had 148 children). The share of working children in the CG is low – as in India and the Philippines – consistent with the national average for work participation rates of older children in Indonesia (3–4 per cent in recent years, or about 6–8 million children in the country). If the child were to stop working, in the household’s view, this would not have catastrophic consequences, unlike the answers we obtained in India. Thus, only 13 per cent of households felt that if the child stopped working, the household’s standard of living would decline, and the same proportion felt that the household production would operate at less than full capacity. As in India, among homeworker households, the incidence of child labour below age 10 is very low. In the Philippines a third of all children in homeworker households are working in homework – much higher than in the other two South East Asian countries. In fact over a fifth of the younger children and half of the older children are engaged in homework in the Philippines. Also as in other countries, these are much higher shares than in the CG households. As in other countries, nearly all of the working children in homeworker households are engaged in homework. The proportion of children engaged in all kinds of work is one of the highest for our sample of countries, and comparable to that prevailing in India for both the younger and older age groups. In Thailand only 6 of the 228 children were found to be working full time as homeworkers. However, about 15 per cent of children in homeworker households from the three sectors are helping in homework. Most of these children are not at primary school age (6–12), but attending school at the upper primary and secondary levels. The work of children was on a voluntary and irregular basis. Parents gave top priority to their study and did not allow children to help with homework during exams or when school homework had to be done.

132

Santosh Mehrotra and Mario Biggeri

Thus in the countries studied, while 56 per cent of 5–14 year olds are working in homework in Pakistan, 26 per cent in India, but only 15 per cent in Thailand, and 12 per cent in Indonesia. However, in the Philippines a third of homeworker household children were in homework. When compared with the incidence of child labour on a national scale in India, which in 1999 was 12.5 per cent (for 10–14 year olds),11 child labour in homeworker households is much higher. The same observation would hold for Pakistan, where the child labour incidence on a national basis is 16 per cent. The same applies to Indonesia, where 8 per cent of children in the country were supposed to be working.12 The incidence of child work in our CGs is very similar to the national incidence of child labour. Child schooling As already mentioned, the child labour literature refers to four categories of activity status of children: 1 2 3 4

only working only studying working and studying, and neither working nor studying.

Work (whether home based or outside the home) here refers to work other than household chores. Table 5.3 (a–c)13 shows the distribution of child status for ages 6–10 and 11–14.14 In India among the younger children, three-quarters are attending school (as opposed to just enrolled) in homeworker households. Among older (or upper-primary level children), the share of children attending school fell15 (to 59 per cent i.e. S
SW) – which is exactly as might be expected from the national trend (Table 5.3a). This is simply the converse of the phenomena we saw in respect of participation in homework – more of the older children worked than the younger children; the latter were more likely to be attending school. Comparing across sectors studied in India, school attendance is highest among children of bidi worker households. Dropout was lowest too. The higher share of children of bidi workers in school may be due to the activities of the Bidi Workers Welfare Fund, which provides scholarships.16 Of all young girls in India in homeworker households, two-thirds are studying full time, the same proportion as all young boys. Similarly, there is no gender discrimination in respect of schooling between older boys and girls. However, it is in the area of work that the gender difference shows up. While 17 per cent of the younger boys are working, 21 per cent of the younger girls are doing so.17 The proportion (48 per cent) of all older boys working is also much lower than for older girls (60 per cent, see Table 5.3a–c). In Pakistan among all the countries studied, the proportion of children in school was the lowest. The attendance rate for children of homeworker households is much

Table 5.3a–c Work and study status of child by age and sex (%) – homeworker (hw) households (a) India Sector

Age 6–10

Age 11–14

W

S

SW

N

W

S

SW

N

3.9 2.4 5.4 0.0 13.8 7.1

78.4 70.6 48.6 87.5 48.6 65.2

4.9 21.2 37.8 8.3 10.1 11.5

12.7 5.9 8.1 4.2 27.5 16.2

23.9 14.3 13.6 15.0 48.6 31.2

47.9 48.8 38.6 60.0 18.9 36.1

11.3 34.5 45.5 22.5 22.5 23.3

16.9 2.4 2.3 2.5 9.9 9.4

4.7 0.0 0.0 0.0 9.6 5.0

79.7 63.6 42.1 80.0 48.1 65.0

4.7 31.8 52.6 16.0 3.8 11.9

10.9 4.5 5.3 4.0 38.5 18.1

30.6 18.2 20.8 15.0 32.6 27.0

50.0 50.0 41.7 60.0 26.1 41.3

5.6 29.5 37.5 20.0 26.1 21.4

13.9 2.3 0.0 5.0 15.2 10.3

2.6 4.9 11.1 0.0 17.5 9.6

76.3 78.0 55.6 95.7 49.1 65.4

5.3 9.8 22.2 0.0 15.8 11.0

15.8 7.3 11.1 4.3 17.5 14.0

17.1 10.0 5.0 15.0 60.0 35.0

45.7 47.5 35.0 60.0 13.8 31.4

17.1 40.0 55.0 25.0 20.0 25.0

20.0 2.5 5.0 0.0 6.2 8.6

Non-hw households (control group) Total Incense stick making 2.6 87.2 Bidi (MP
TN) 0.0 100.0 Bidi (MP) 0.0 100.0 Bidi (TN) 0.0 100.0 Zardosi 2.6 82.1 All 2.2 87.1

5.1 0.0 0.0 0.0 0.0 2.2

5.1 0.0 0.0 0.0 15.4 8.6

5.0 0.0 0.0 0.0 9.5 4.9

60.0 95.0 88.9 100.0 28.6 60.7

10.0 0.0 0.0 0.0 0.0 3.3

25.0 5.0 11.1 0.0 61.9 31.1

Total Incense stick making Bidi (MP
TN) Bidi (MP) Bidi (TN) Zardosi All Male Incense stick making Bidi (MP
TN) Bidi (MP) Bidi (TN) Zardosi All Female Incense stick making Bidi (MP
TN) Bidi (MP) Bidi (TN) Zardosi All

(b) Indonesia Sector

Age 7–12 W

Total Pottery Rattan Batik All Male Pottery Rattan Batik All

Age 13–15

S

SW

0.0 0.0 1.5 0.5

62.3 86.3 85.3 79.4

30.2 4.1 10.3 13.4

0.0 0.0 0.0 0.0

63.3 89.5 90.9 82.2

30.0 2.6 6.1 11.9

N

W

S

SW

N

7.5 9.6 2.9 6.7

13.3 30.4 26.9 23.6

53.3 25.0 42.3 38.6

26.7 25.0 7.7 22.0

6.7 19.6 23.1 15.7

6.7 7.9 3.0 5.9

17.6 40.0 20.0 28.1

52.9 32.0 60.0 45.6

29.4 4.0 0.0 10.5

0.0 24.0 20.0 15.8

(Table 5.3 continued)

134

Santosh Mehrotra and Mario Biggeri

Table 5.3a–c Continued Sector

Age 7–12 W

S

Female Pottery 0.0 60.9 Rattan 0.0 82.9 Batik 2.9 80.0 All 1.1 76.3 Non-hw households (control group) Total Pottery 0.0 88.9 Rattan 0.0 93.9 Batik 0.0 94.1 All 0.0 92.2 (c) Philippines Sector

SW

N

W

S

SW

N

30.4 5.7 14.3 15.1

8.7 11.4 2.9 7.5

10.7 22.6 36.4 20.0

53.6 19.4 18.2 32.9

25.0 41.9 18.2 31.4

10.7 16.1 27.3 15.7

5.6 0.0 2.9 2.9

5.6 6.1 2.9 4.9

11.8 16.7 15.4 14.6

76.5 72.2 61.5 70.8

11.8 0.0 15.4 8.3

0.0 11.1 7.7 6.3

Age 6–12 W

Total Home décor Pyrotechnics Okra Fashion accessories All Male Home décor Pyrotechnics Okra Fashion accessories All Female Home décor Pyrotechnics Okra Fashion accessories All

Age 13–15

Age 13–14

S

SW

N

W

S

SW

N

0.0 2.3 2.4 0.0 1.2

74.4 59.1 63.4 66.7 65.7

20.5 36.4 29.3 16.7 25.9

5.1 2.3 4.9 16.7 7.2

8.3 11.1 8.7 19.2 12.1

33.3 33.3 34.8 42.3 36.3

58.3 55.6 43.5 34.6 47.3

0.0 0.0 13.0 3.8 4.4

0.0 4.2 0.0 0.0 1.2

66.7 54.2 70.0 75.0 65.9

27.8 37.5 20.0 5.0 23.2

5.6 4.2 10.0 20.0 9.8

6.7 22.2 8.3 15.4 12.2

26.7 33.3 25.0 46.2 32.7

66.7 44.4 50.0 30.8 49.0

0.0 0.0 16.7 7.7 6.1

0.0 0.0 4.8 0.0 1.2

81.0 65.0 57.1 59.1 65.5

14.3 35.0 38.1 27.3 28.6

4.8 0.0 0.0 13.6 4.8

11.1 0.0 9.1 23.1 11.9

44.4 33.3 45.5 38.5 40.5

44.4 66.7 36.4 38.5 45.2

0.0 0.0 9.1 0.0 2.4

Source: UNICEF survey. Notes Ages follows UNESCO (2001a) guidelines. W  go to work. S  go to school. SW  go to school and work. N  Neither go to school and work.

lower than the national gross enrolment ratio for Pakistan (99 per cent for boys and 69 per cent for girls at primary level) (UNICEF, 2001). Only a fifth of all children in homeworker households were in school in contrast to 35 per cent of children from the CG households. Most of the older children were actually working.

Child labour in homeworker households 135 Table 5.4a–d shows a much stronger degree of feminisation of homework in Pakistan than in India (although there is feminisation of homework in India as well). Thus, over half of all girls in the younger age groups are working in homework, while only a third of the boys are. In the older age group 95 per cent of girls are working in homework, while half of the boys are. Over a third of the older boys go to school, but only a fifth of the older girls do. Most of the children in Pakistan went to madrasas (Koranic schools).18 In all the South East Asia countries, the schooling status of children in homeworker households was very different to the South Asian clusters. In Indonesia as many as 93 per cent of children between 7 and 12 are in school among homeworker households (Table 5.3b). This is hardly surprising in a country where primary net enrolment rates are 94 per cent for both boys and girls (UNICEF, 2001a). We know that the dropout from school increases sharply after primary level. Thus for homeworker households, enrolment rates drop to 61 per cent for the 13–15 year olds (Table 5.4a–c). The feminisation of homework is evident in Indonesia as well. While over two-fifths of the older girls are working, under a fifth of the boys are working in the older age group (Table 5.4a–c). A higher proportion of the girls combine schooling with work, but a much higher proportion of the boys are only in school. Thus children of homeworkers are slightly less likely to attend school than children in non-homeworker households, and slightly more likely to drop out after primary level than in non-homeworker households. In general non-homeworker households are somewhat better off economically. From the gender differences in child status, it is clear there is a feminisation of homework. In Thailand most children attend school. Only six children among the homeworker households are in full time homework. These children had completed schooling to grade 6 or 9. The children who are working part time in homework, helping their families, are also going to school. In fact, 81 per cent of all school age children in homeworker households, and 78 per cent in CG households were in school.19 When parents were asked to which level you can afford/support your children’s education, one-fifth stated they can support their education through college level, and another fifth stated that it is up to the children themselves. It was found that the numbers of children continuing with their education beyond compulsory level is on the rise since the 1997 economic crisis and the government’s policy on ‘expanding opportunity’ for higher education. In the Philippines most of the younger children (in the 6–12 age group) were attending school (ranging from 84 to 96 per cent, varying by sector). Most of the older children (in 13–14 age group) were also in school (between 77 and 91 per cent). There appeared to be no significant difference in this respect between the homeworker households and the CG households. There is also no significant difference between the incidence of absences from school among children in households with and without homework.20 However, according to FGDs part of the children in the community have stopped schooling for different reasons including economic hardship, difficulties in working and studying at the

Table 5.4 Work (as homeworkers) and study status of child by age and sex (%) – homeworker households (a) India Sector

Total Incense stick making Bidi (MP
TN) Bidi (MP) Bidi (TN) Zardosi All Male Incense stick making Bidi (MP
TN) Bidi (MP) Bidi (TN) Zardosi All Female Incense stick making Bidi (MP
TN) Bidi (MP) Bidi (TN) Zardosi All (b) Pakistan Sector

Total Incense stick making Carpet weaving Sack stitching Prawn peeling All Male Incense stick making Carpet weaving Sack stitching Prawn peeling All Female Incense stick making Carpet weaving Sack stitching Prawn peeling All

Age 6–10

Age 11–14

W

S

SW

N

W

S

SW

N

2.0 2.4 5.4 0.0 10.1 5.1

78.4 70.6 48.6 87.5 50.5 65.9

4.9 21.2 37.8 8.3 8.3 10.8

14.7 5.9 8.1 4.2 31.2 18.2

9.9 8.3 4.5 12.5 43.2 23.3

53.5 50.0 38.6 62.5 20.7 38.7

5.6 33.3 45.5 20.0 20.7 20.7

31.0 8.3 11.4 5.0 15.3 17.3

1.6 0.0 0.0 0.0 1.9 1.3

79.7 63.6 42.1 80.0 51.9 66.3

4.7 31.8 52.6 16.0 0.0 10.6

14.1 4.5 5.3 4.0 46.2 21.9

5.6 11.4 8.3 15.0 21.7 13.5

55.6 50.0 41.7 60.0 30.4 44.4

0.0 29.5 37.5 20.0 21.7 18.3

38.9 9.1 12.5 5.0 26.1 23.8

2.6 4.9 11.1 0.0 17.5 9.6

76.3 78.0 55.6 95.7 49.1 65.4

5.3 9.8 22.2 0.0 15.8 11.0

15.8 7.3 11.1 4.3 17.5 14.0

14.3 5.0 0.0 10.0 58.5 32.1

51.4 50.0 35.0 65.0 13.8 33.6

11.4 37.5 55.0 20.0 20.0 22.9

22.9 7.5 10.0 5.0 7.7 11.4

SW

N

Age 5–10

Age 11–14

W

S

SW

N

W

S

27.6 38.4 19.0 46.8 34.0

8.8 5.8 15.5 1.6 7.4

12.4 3.7 25.4 0.5 9.4

51.2 52.1 40.1 51.1 49.1

57.5 72.0 50.5 70.0 60.4

9.2 5.3 3.6 8.0 6.2

19.5 8.0 37.8 2.0 20.4

13.8 14.7 8.1 20.0 13.0

17.7 26.2 10.7 35.2 23.1

13.9 9.5 14.7 3.3 10.0

2.5 6.0 28.0 1.1 8.8

65.8 58.3 46.7 60.4 58.1

12.0 42.3 33.3 42.1 32.1

32.0 11.5 10.3 10.5 15.6

12.0 7.7 41.0 0.0 19.3

44.0 38.5 15.4 47.4 33.0

36.3 49.5 28.4 57.9 44.4

4.4 3.0 16.4 0.0 5.1

20.9 2.0 22.4 0.0 10.2

38.5 45.5 32.8 42.1 40.4

75.8 87.8 59.7 87.1 74.8

0.0 2.0 0.0 6.5 1.4

22.6 8.2 36.1 3.2 21.0

1.6 2.0 4.2 3.2 2.8

(Table 5.4 continued)

Table 5.4 Continued (c) Indonesia Sector

Age 7–12 W

Total Pottery Rattan Batik All Male Pottery Rattan Batik All Female Pottery Rattan Batik All (d) Philippines Sector

S

SW

0.0 0.0 0.0 0.0

69.8 86.3 86.8 82.0

22.6 4.1 8.8 10.8

0.0 0.0 0.0 0.0

70.0 89.5 93.9 85.1

0.0 0.0 0.0 0.0

69.6 82.9 80.0 78.5

N

W

S

SW

N

7.5 9.6 4.4 7.2

8.9 19.6 7.7 13.4

64.4 25.0 42.3 42.5

15.6 25.0 7.7 18.1

11.1 30.4 42.3 26.0

23.3 2.6 3.0 8.9

6.7 7.9 3.0 5.9

11.8 20.0 6.7 14.0

70.6 32.0 60.0 50.9

11.8 4.0 0.0 5.3

5.9 44.0 33.3 29.8

21.7 5.7 14.3 12.9

8.7 11.4 5.7 8.6

7.1 19.4 9.1 12.9

60.7 19.4 18.2 35.7

17.9 41.9 18.2 28.6

14.3 19.4 54.5 22.9

Age 6–12 W

Total Home décor Pyrotechnics Okra Fashion accessories All Male Home décor Pyrotechnics Okra Fashion accessories All Female Home décor Pyrotechnics Okra Fashion accessories All

Age 13–15

Age 13–14

S

SW

N

W

S

SW

N

0.0 2.3 2.4 0.0 1.2

79.5 59.1 68.3 66.7 68.1

15.4 36.4 24.4 16.7 23.5

5.1 2.3 4.9 16.7 7.2

8.3 11.1 8.7 19.2 12.1

41.7 33.3 34.8 42.3 38.5

50.0 55.6 43.5 34.6 45.1

0.0 0.0 13.0 3.8 4.4

0.0 4.2 0.0 0.0 1.2

77.8 54.2 75.0 75.0 69.5

16.7 37.5 15.0 5.0 19.5

5.6 4.2 10.0 20.0 9.8

6.7 22.2 8.3 15.4 12.2

40.0 33.3 25.0 46.2 36.7

53.3 44.4 50.0 30.8 44.9

0.0 0.0 16.7 7.7 6.1

0.0 0.0 4.8 0.0 1.2

81.0 65.0 61.9 59.1 66.7

14.3 35.0 33.3 27.3 27.4

4.8 0.0 0.0 13.6 4.8

11.1 0.0 9.1 23.1 11.9

44.4 33.3 45.5 38.5 40.5

44.4 66.7 36.4 38.5 45.2

0.0 0.0 9.1 0.0 2.4

Source: UNICEF survey. Notes Ages follows UNESCO (2001a) guidelines. S  go to school. W  got to work as homeworkers. SW  go to school and work as homeworkers. N  neither go to school and work as homeworkers.

138

Santosh Mehrotra and Mario Biggeri

same time, family crisis or the distance to the school. There are almost no gender differences in schooling for either the younger or older children. The children’s working status can be examined by age. We can observe that the four statuses have a different path based on the age of children. For instance, ‘work only’ tends to increase, while ‘study only’ tends to decrease in the total share for each additional year. In Section 5.3 an empirical analysis is done to explore the determinants of child status in relation to work and study. ‘Neither studying nor working’: what are they doing? In homeworker households the share of the group of children ‘neither studying nor working’ can be lower since, children can work easily in homework activities especially female children (Table 5.3). However there is a pull factor as well since usually a lower share of children from homework households attend school – presumably to be ready to work in peak season. In the child labour literature, the ‘neither’ category is usually construed as those children who spend most of their time in household chores or the care of siblings (UNESCO, 2001b). This child status is quite ambiguous since it includes children really ‘idle’, children that have never been to school or that drop out from school or children who are disabled, temporarily looking for a job or doing intensive household chores (Biggeri et al., 2003). From the point of view of multidimensional well-being of children, household chores, although not accounted as child labour, can affect children’s capabilities significantly in terms of capabilities deprivation (Biggeri, 2003a). Several observations are possible about this category of children in India. First, about 16 per cent of the younger children, and less than a tenth of older children, are in the ‘neither’ category (Table 5.3). These children are not working (in either homework or for employers outside the home) nor going to school. Clearly fewer of the older children are in the neither category, since they start working in homework as they grow older. Second, there is not a sharp gender difference among the children in the ‘neither’ category; only a slightly higher proportion of boys than girls are in this category. This is plausible in this case since girls are more likely to work as homeworkers. The third observation is that within the ‘neither’ category the vast majority are reported as ‘not doing anything’ (84 per cent of young girls, 77 per cent of young boys, 63 per cent of older girls and 40 per cent of older boys). However, this is misleading, since the India survey also provides information about the time allocation of children in the ‘neither’ category, that is, those who are not engaged in work, homework or otherwise. Their time allocation, outside of sleeping and eating hours, are as follows: averaged across all sectors, roughly two hours are spent on assisting in ‘food preparation’; another hour or so goes towards ‘housekeeping’ work; and another half hour each is spent on animal husbandry, fetching drinking water, shopping and child care. The rest of the time is spent between a series of miscellaneous activities: fuel collection, fodder collection, socialising, personal care or watching television (if available).

Child labour in homeworker households 139 In Pakistan, half of the younger children (more boys than girls) seem to be in the ‘neither’ category – larger than in either of our other two countries (Table 5.4). There are two explanations for this apparent phenomenon. First, the younger age group is defined as starting at a lower age – 5 years – as compared to 6 in India and 7 in Indonesia, since this is the official starting age for school in these countries. We find in fact, that while the official starting age for schooling is lower in Pakistan, more children tend to start around 6 rather than 5 years of age. Second, some of these children are working for an employer outside the home – particularly the boys (fewer young girls than boys are in the ‘neither’ category). Among the older children the ‘neither’ category drops sharply in significance (to 13 per cent overall, almost entirely boys), since most of the children are in homework. Actually, since the Pakistan data only tell us about children working in homework (and not about those working outside the home), it is highly likely that some of the ‘neither’ category children are actually working away from home. As in India and Indonesia, the rest would be engaged in household chores. In Indonesia, as in India, fewer (a sixth) older children in homeworker households are in the neither category than the younger (less than a tenth) (Table 5.3). For these children, their time is allocated between shopping, ironing, cooking, washing utensils (17 per cent of respondents in ‘neither’ category), sweeping (21 per cent), mopping (10 per cent), dusting (11 per cent), laundry (11 per cent) and child-minding (19 per cent). In other words, most of these children, most of the time, appear to be engaged in household chores and sibling care.21 In the Philippines the proportion of children in the neither category is under 7 per cent. The data does not allow us to estimate this share for Thailand. Reasons why children are working/not in school In India the question asked was not ‘why is the child working’ but ‘why is the child not attending school’. Of those children not enrolled in school, the reasons for their non-enrolment were recorded.22 In all sectors a very high response was recorded against ‘school or studies not interesting’ as the reason for not attending school (Table 5.5a–d). Prima facie, this raises questions about the quality and relevance of schooling. Another important reason given (a fifth of the respondents) is that they cannot afford to send their children to school (Table 5.5a–d). In Pakistan the responses were mostly indirectly associated with poverty. According to the women interviewed, most of the children were working at home (88 per cent of the responses) to supplement family income. Both mothers and children were asked the reasons why children were not in school. In over threequarters of the cases, the answer of children and of mothers was that ‘school was too expensive’ – for both homework and CG households (Table 5.5a–d). Two-fifths of both boys and girls in homeworker households not enrolled in school indicated a desire to attend a school full time if given a choice, while others preferred combining school with work or play (24 per cent of boys and girls) (Table 5.5a–d).

School is of poor quality

School is too far

Hw households

School is too expensive

57.1 83.3 22.2 45.5 49.2

4.8 8.3 0.0 19.1 14.7

(b) Pakistan, why is your child not in school? Incense stick 1.8 1.8 85.5 making Carpet weaving 0.0 3.1 89.2 Sack stiching 0.0 0.0 79.5 Prawn peeling 0.0 1.6 92.2 All 0.4 1.8 87.4

Sector

54.3

8.7

1.8 6.2 2.6 0.0 2.7

0.0 2.6 3.1 1.8

6.5 4.8 0.0 11.1 0.0 2.3

Need income from child’s work/ poverty

1.8

Child prefers to stay at home

4.8 0.0 11.1 1.8 1.7

0.0

1.5 15.4 3.1 5.8

7.3

Others

4.8 0.0 11.1 0.0 0.6

0.0

100.0 100.0 100.0 100.0

100.0

Total

0.0 0.0 0.0 0.0 0.6

2.2

0.0 0.0 33.3 9.8

0.0

School is too far

0.0 0.0 0.0 0.0 0.6

2.2

0.0 0.0 0.0 7.7 13.9

33.3

100.0 55.6 50.0 75.6

100.0

School is too expensive

0.0 11.1 0.0 2.4

0.0

School is worthless

Not-hw households

100.0 0.0 100.0 0.0 100.0 0.0 100.0 84.6 100.0 61.1

100.0

0.0

0.0 11.1 0.0 2.4

0.0

Child prefers to stay at home

100.0 100.0 0.0 3.8 22.2

66.7

0.0 22.2 16.7 9.8

0.0

Others

0.0 0.0 0.0 3.8 2.8

0.0

100.0 100.0 100.0 100.0

100.0

Total

100.0 100.0 100.0 100.0 100.0

100.0

Cannot Child School or Parents Total afford working studies not attitude outside interesting indifferent

Cannot Child afford working in hw

Child School or Parents Child’s Lot of Care of Others Total working studies not attitude health work sibling outside interesting indifferent reason at home

Not-hw households

Hw households

(a) India, reasons for leaving school Incense stick 23.9 2.2 making Bidi (MP
TN) 23.8 0.0 Bidi (MP) 8.3 0.0 Bidi (TN) 44.4 0.0 Zardosi 15.5 18.2 All 18.6 11.9

Sector

Table 5.5 Reasons for not sending children to school-homeworker (hw) and non-hw household (%)

6.3 20.0 0.0 9.5

10.5

51.2

To fund school needs

68.8 13.3 84.2 61.9

To earn own money and hh add. inc for food and edu exps

Source: UNICEF survey.

(d) Philippines Home décor Pyrotechnics Okra Fashion accessories All

Sector

1.2

0.0 0.0 0.0 4.8

Need to work or be employed

0.0 2.3 10.5 3.6

Can’t study

2.3

6.3 0.0 0.0 4.8

Help mother and others with the work

4.8 2.3 5.3 3.6

Don’t like school

8.1

12.5 6.7 0.0 14.3

25.0 28.6 25.0 26.7

3.5

0.0 3.3 10.5 0.0

Prefer this kind of work and find it easy

100.0 100.0 100.0 100.0

Total

To keep us occupied

33.3 43.2 31.6 38.1

Other

20.9

6.3 53.3 5.3 0.0

To increase the earning, spending or savings of child

25.0 0.0 25.0 13.3

No money

Work

Wants to play

Work

No money

Not-hw households

Hw households

(c) Indonesia, main reason for no schooling Pottery 9.5 28.6 23.8 Rattan 9.1 34.1 9.1 Batik 5.3 36.8 10.5 All 8.3 33.3 13.1

Sector

0.0 14.3 0.0 6.7

Don’t like school

1.2

0.0 0.0 0.0 4.8

To learn to be independent

0.0 14.3 0.0 6.7

Can’t Study

1.2

0.0 3.3 0.0 0.0

To make parents happy

50.0 42.9 50.0 46.7

Other

100.0

100.0 100.0 100.0 100.0

Total

100.0 100.0 100.0 100.0

Total

142

Santosh Mehrotra and Mario Biggeri

In Indonesia a higher proportion of children work in homeworker households than in CG households. The higher proportion of working children in homeworker households appears to be a function of having income earning opportunities available at home; ‘family tradition’ was given as an important reason for doing homework.23 It is not unusual that when both parents have an income-earning activity or business at home, all members of the family help, including children of all ages. The FGDs show that the primary reason for children working is to pay school fees and earn pocket money. Very few children work without going to school – work and school tend to go together. However, of those not attending school among homeworker households, a third stated that ‘lack of money’ was the main reason for not attending school (Table 5.5). Among CG households, 27 per cent of the responses stated ‘work’ as the main reason why the few older children (15 per cent of children in the age group 13–15) were not studying. No children from the younger age group were only working (Table 5.3). The children’s earnings from work, whether directly from employers or indirectly through their mothers, are usually under the full control of the children. They are free to spend it as they think fit – a situation quite different from that prevailing in the South Asian homeworker households surveyed. In the Philippines, in the FGD the children explained they were not forced to work. They worked to augment family income, to help parents produce faster, and add to family earnings rather than spend time playing. They also worked for some pocket money and to save for school supplies and clothes.24 When they were asked whether children should work, the answer was two-sided. On the one hand, they said children should not do heavy work, since they were young and their bodies were not ready for such work. Yet, they said, children should work so they could help their parents. In Thailand of the 305 households surveyed, only two households sent under-14-year-old children to work outside the household. In other words, for the vast majority of households in homework – two agrarian communities (leather products, hybrid seeds) and one peri-urban slum community (leather products) – work was merely a family-based activity, to be undertaken once school was over. Schooling was seen as a high priority, and as many as one-fifth of the parents said they can support their children through college education. Only 3 per cent said that they could afford to support their children only up to primary level.

Contribution of child work and its impact on schooling There are two issues in respect of the hours that children work: one relates to how it affects their schooling (assuming that they combine schooling with work), and the other how important the work is in terms of its contribution to the total hours worked by the family. The contribution of children to homework was estimated based on time spent on work. Table 5.6a–c presents the hours worked per day, based on a six-day week, for all children – whether they only work or work and study.

Child labour in homeworker households 143 In India the women worked on average 8 hours per day, six days a week. The younger children on average worked 2.9 hours per day, and the older children 4.3 hours (Table 5.6a–c). The average contribution for the three sectors taken together is over 13 per cent of the total number of hours worked by the household members on homework. The girls seem to be spending more time on homework between ages 11–14. Clearly, the older children spend more time at work than the younger ones. If the children are enrolled in school, the number of hours worked would ordinarily interfere with schoolwork (as about 20 hours a week of work is usually seen as being consistent with full-time schooling).25 Table 5.7a–c shows how many hours per week children are working, enabling us to differentiate between children who are studying and those who are not. In India most of the children who are working tend to spend more time at work than children who are studying and working. What is perhaps more important is that among young and old children who are at school, their hours of work are compatible with full-time schooling only in bidi (Tamil Nadu). In all other sectors, the hours of work are close to what may be termed as a ‘danger zone’, as in interfering with studies.

Table 5.6 Children working in homework: average hours worked by children per day by age group and sex (considering 6 days a week) Sector

(a) India Incense stick making Bidi (MP) Bidi (TN) Zardosi All

5–10

11–14

Total

Number of children

Female

Male

Total

Female

Male

Total

0.8

5.0

3.3

4.2

3.7

4.1

3.8

18

3.7 0.0 2.9 2.9

2.3 1.7 6.7 2.9

2.8 1.7 3.1 2.9

2.6 3.1 5.2 4.6

2.5 2.0 4.8 3.7

2.6 2.5 5.1 4.3

2.7 2.3 4.7 3.9

17 38 92 165

(b) Pakistan* Incense stick making Carpet weaving Sack stitching Prawn peeling All

4.6

2.8

4.2

6.7

6.0

6.7

5.8

77

5.7 3.4 4.7 4.7

3.0 5.0 3.7 3.5

5.4 3.6 4.3 4.5

7.6 4.4 5.8 6.1

6.1 3.7 5.2 5.0

7.3 4.3 5.7 5.9

6.5 4.1 4.9 5.3

77 75 74 303

(c) Indonesia Pottery Rattan Batik All

1.7 2.5 1.5 1.8

2.4 4.7 0.0 2.9

2.1 3.6 1.5 2.3

2.5 3.8 3.0 3.2

3.8 0.0 0.3 2.9

3.1 3.8 2.2 3.1

2.7 3.8 2.0 2.9

20 11 9 40

Source: UNICEF survey. Note * How many hours do you work during a day (day and night)?

144

Santosh Mehrotra and Mario Biggeri

One can be certain that over 20 hours of work would interfere with school achievement, since working over ‘half-time’ can be a risk factor. The direct effects could be exhaustion or the diversion of interest away from academic concerns. Also, finding a relationship between time in work and school achievement does not by itself suggest that work is the causal factor that impacts on achievement. The causal arrows may in fact run in the opposite direction.26 It is possible that the effects of work on learning achievement begin even at 15 hours of work outside the home.27 By this measure, Pakistani children are working far too much. In Pakistan women worked an average of about 7 hours on homework in all sectors. Compared Table 5.7 Children working in homework: average hours worked by children per week by work and study status (a) India Sector

6–10

11–14

W

SW

W

SW

Incense stick making Bidi (MP) Bidi (TN) Zardosi All

28.0 28.0 — 12.8 16.3

18.0 15.3 10.0 24.8 18.0

26.2 14.0 28.3 32.3 30.8

21.8 15.6 8.5 27.5 20.0

(b) Pakistan* Sector

5–10

11–14

W

SW

W

SW

Incense stick making Carpet weaving Sack stitching Prawn peeling All

26.4 32.9 22.0 26.5 27.0

24.0 30.0 21.0 24.0 24.8

41.1 45.2 26.8 34.9 37.0

35.4 28.5 24.3 18.0 26.6

(c) Indonesia Sector

7–12

Pottery Rattan Batik All

13–15

W

SW

W

SW

— — — —

12.6 21.5 6.3 12.5

24.0 27.6 37.1 31.3

17.8 23.0 11.8 18.3

Source: UNICEF survey. Notes Ages follows UNESCO (2001a) guidelines. W  go to work. SW  go to school and work. * How many hours do you work during a day (day and night)?

Child labour in homeworker households 145 to this, children worked on average about 5.3 hours per day on homework (Table 5.6b). They also worked a six-day week. The children contributed as much as 43 per cent of the total hours the family devotes to homework, assuming that there is only one child. Clearly the contribution of the children to family income is much greater in these sectors in Pakistan than in the three sectors in India. As in India, children in Pakistan who are only working have longer hours of work per week than children who study and work (see Table 5.7b). In all sectors, all children, young and older, work hours which are totally incompatible with full-time schooling. The children’s responses suggest that the work interfered with schoolwork for 40 per cent of the children in India and for 47 per cent in Pakistan. In Indonesia on average, female adult homeworkers work about 5.2 hours per day, males 6.3 hours. Again, the contribution of children appears significant – but only in the case of the older children (who work 3.1 hours a day) – accounting for at least half of adult hours worked. The younger Indonesian children were working fewer hours (2.3 hours) than homework children in South Asia. There was no evidence that the work interfered with schoolwork, only with playtime (according to the responses). Based on the responses of children, work interfered with schoolwork for 25 per cent of children. Table 5.7c shows that there are no young children in Indonesia who only work; all of them work and study. The young children who work and study do not put in so many hours of work that it is incompatible for them to study. However, in two of three sectors, the older children seem to be in the danger zone, and it is likely that their studies will be affected by the number of hours worked. In the Philippines children 5–18 years of age account for as much as 20 per cent of total household income for two activities (okra packaging in Tarlac, Christmas decoration production in Rizal), and between 8 and 12 per cent in the other homework activities. The contribution of children to the income of CG households is negligible given that only a few of them are economically active. In Thailand we have seen that the children worked irregularly after school hours on a voluntary basis, without seriously affecting their studies.28 In hybrid seed production, children contribute their labour in the evening after school, or during weekends. They helped in post-harvest activities like washing, deseeding (the mature fruit) and drying of seeds.29 In chapter products, children helped in light work like picking fresh flowers or cutting the petals for decorating the mulberry chapter; or with paper craft such as gluing or pasting coloured paper. Clearly their contribution was quite limited. During the high season among all sectors, children worked between 5 and 6 hours per day in homework activities. During the low season work fell to 1–2 hours per day, depending on the sector. As in the other countries, older children spend more time on homework. To summarise, the Pakistani children are therefore working the longest hours, hours that are incompatible with schooling – given that most of them are not in school. The Indian children work more than the Indonesian children. The Thai children are working the fewest hours. Second, at least in India and Pakistan,

146

Santosh Mehrotra and Mario Biggeri

the number of hours worked per week by working children is likely to have affected their studies. Third, it is not just that a higher proportion of girls is working in all countries, but the girls are working longer hours than boys – both in the younger as well as the older age group. Finally, the children do seem to be contributing significantly to the total number of hours worked by the family in homework. Work conditions and relationship with contractors/parents In all countries studied, the relationship between the working children and their parents and contractors is a complex one. In India and Pakistan there was clearly an effort by contractors to keep the work of children clandestine and avoid visibility. Conditions in some sectors involved debt bondage; but generally children said they worked because they felt a sense of identification with household needs. The latter is also the case in Indonesia, Thailand and the Philippines; but here the work seems lighter in most cases, and the relationship with contractors not overly exploitative. In India about four-fifths of all working children in homeworker households work at home. In most cases the relationship of children with contractors is not direct, being mediated by the parents. In bidi (Madhya Pradesh) the contractors discouraged children from coming to fetch raw materials or deliver the finished product. There is an official ban on the use of child labour, and contractors are careful to avoid visibility – a reason why the contractors never have direct agreements or arrangements with children, rather through parents. Although written contracts do not exist, the relationship with contractors is a long-term one, and as we noted in Chapter 4, there is a high level of stability in the entire structure of production. As markets are expanding in zardosi and incense sticks, the contractor has an incentive to retain the existing workers. In bidi the unionisation of workers ensures that although the market is shrinking, all card holders receive some work. In Pakistan the contractors generally paid late for the work done and in some cases did not pay at all. They had assisted only 1 per cent of the women and under 5 per cent of the children in the treatment of ailments that resulted from homework. Children paid a price for mistakes and in about a tenth of the cases were beaten by the contractor. In Pakistan the homeworkers generally believed that the presence of contractors was welcome, because there were hardly any other work options. They did not complain much about the contractors profiting at their expense or at their bad behaviour as the typical employer–employee relationship did not exist between them, and they were related in most cases. The homeworkers also believed that the contractors were not much better off than they were. As the contractors belonged to the same community, and were often relatives, people had not taken any collective action for increased wages against them. In some cases, households had taken advance payment and were indebted to the contractors (as in the case

Child labour in homeworker households 147 of the bidi sector in India). In carpet weaving, homework in the site selected in Karachi followed the pattern of bonded labour.30 In Pakistan, a very important question was put to the mother: do you feel your children are as well cared for as before despite your work schedule? As many as 71 per cent of women (average across all sectors) answered in the negative; the shares of women who felt this way were similar across all sectors. In Indonesia the children have no direct contact with the contractors. The child homeworkers are likely to work under the supervision of their mothers, suggesting the important role of mothers in socialising and passing on skills to children. Of the 41 children in homework, only 8 worked for the employer and were paid directly. However, when asked, children expressed a preference for not working with their parents, as working for the employer gave them market wage-rates.31 In the Philippines, the survey asked an open-ended question about the views of the children on homework. Only 28 per cent responded of which the majority felt that homework was ‘burdensome’, although they were not against homework per se; the other responses were similar in that they were ‘unhappy’ about working or ‘did not want to work’. This is consistent with the children’s response to a question on the changes they desired in subcontracting work. Over a third said they wanted higher pay or to have regular or more orders from the contractors. In Thailand, as in Indonesia, it appears that the work that children engage in is light.32 In other words, it appears that the use of child labour in home contract work is not exploitative, and almost always goes hand in hand with schooling. It is similar to the general situation in agriculture, where children work as unpaid family farm workers with other family members. The difference is that the work was for remuneration and not for own consumption. Health conditions In India across the three sectors studied, about 16 per cent of children working in homework reported health problems due to work (Table 5.8). In all sectors some 11 to 18 per cent of the children in homework seemed to face homework-related health problems. Most children in zardosi sector reported problems that were clearly homework-related – shoulder pain and backache were the most commonly cited problems (as with adult women). In none of the cases was any kind of treatment resorted to, despite awareness of the health problem. In Pakistan nearly all the children in homeworker households seemed to be ailing. The proportion, though high, for the children from non-homeworker households, was under two-thirds. If the response was that they were ailing, the reason ascribed by the household in the vast majority of the homeworker households was that it was homework-related. The survey inquired into the frequency of illness in the past six months; the mean number of times for children in homeworker households was 5.2 (or almost once every month) while it was 3.8 for the CG. Only 8.5 per cent of the children in the CG responded that they

148 Santosh Mehrotra and Mario Biggeri Table 5.8 Health of children – homework (hw) households (%) INDIA; Health of Children: Children facing health problems due to hw hw Households SECTOR

Incense stick making Bidi (MP+TN) Bidi (MP) Bidi (TN) Zardosi All

No 88.9 85.7 83.4 _

82.5 84.0

Yes 11.1 14.3 16.6 _ 17.5 16.0

Total 100.0 100.0 100.0 _

100.0 100.0

a: PAKISTAN, Health of Children: Does the child currently have a disease/ailment? Non-hw Households

hw Households SECTOR Incense stick making Carpet weaving Sack stitching Prawn peeling All

No

Yes

2.6 1.3 0.0 5.4 2.3

97.4 98.7 100.0 94.6 97.7

SECTOR Incense stick making Carpet weaving Sack stitching Prawn peeling All

Total 100.0 100.0 100.0 100.0 100.0

No

Yes

31.8 18.2 52.0 40.0 36.2

68.2 81.8 48.0 60.0 63.8

Total 100.0 100.0 100.0 100.0 100.0

b: PAKISTAN, Health of Children: Does the child currently have a disease/ailment? If Yes why?

SECTOR Incense stick making Carpet weaving Sack stitching Prawn peeling All

hw Households Congenital Community hw disease related 0 0 1.3 0 0.3

18.7 14.5 14.7 8.6 14.2

81.3 85.5 84 91.4 85.5

Total 100.0 100.0 100.0 100.0 100.0

SECTOR Incense stick making Carpet weaving Sack stitching Prawn peeling All

Non-hw Households Community Other disease 53.3 27.8 8.3 0 0

Total 100.0 100.0 100.0 100.0 100.0

46.7 72.2 91.7 100 100

INDONESIA, Health of Children: General health problems SECTOR Pottery Rattan Batik All

hw Households No Yes

82.8 83.8 83.3 83.3

17.2 16.2 16.7 16.7

Total 100.0 100.0 100.0 100.0

SECTOR Pottery Rattan Batik All

Non-hw Households No Yes

87.5 90.9 78.0 85.7

12.5 9.1 22.0 14.3

Total 100.0 100.0 100.0 100.0

PHILIPPINES, Have you become sick because of work?

SECTOR

Home décor Pyrotechnics Okra Fashion accessories All

hw Households No Yes 71.0 29.0 72.7 27.3 54.0 46.0 50.0 50.0 60.5 39.5

Total 100.0 100.0 100.0 100.0 100.0

Source: UNICEF survey

needed medical treatment but could not afford it, this was the response of 43 per cent of the children in homework. In Indonesia, to establish the impact of homework on the health of children, comparisons were made with three other groups of children: (a) children of homeworker households not involved in homework; (b) working children of non-homeworker households and (c) non-working children in non-homeworker households. The results suggest that homework cannot be ascribed as a cause of poor health since there is little difference in the patterns of health problems experienced in the preceding month by the various groups of children. The FGD found no major problems stemming from the children’s work. The most frequently heard complaint was stiffness from sitting in the same position for too long. These back pains, however, disappear after a rest, though the workers then return to their respective jobs. In the FGDs with children, they said they find more benefits than disadvantages from work; they were not able to define the disadvantages of their work.

Child labour in homeworker households 149 In Thailand since the children only did limited hours of light work, there was no real issue of health hazard found in the study. However, since children share the same living conditions as parents their health is indirectly affected. For instance, in the leather crafts sector they are exposed to the smell and dust from hides, chemical substances such as tannin used for softening leather, noise and risk from sewing tools. In the hybrid seed sector, children are exposed to chemical pesticide and fertiliser. This was the case especially for girls involved in artificial pollination. However, from the FGDs we learnt that those involved in the pollination were aware of the possible impact from hazardous chemicals and are not allowed to go near the vegetable garden after the crops have been sprayed. In the Philippines, in the homework sectors (fashion accessories, for example earrings/necklaces; home decor, for example Christmas lights/balls/metalcraft; okra production) there appears to be little or no health hazard related to work. Most of the children engaged in packing okra did complain of dizziness when asked what discomfort was most experienced at work. Most of them felt they deserved to be paid a better piece-rate for work. However, there was no doubt that the production of fireworks was hazardous to the health/safety of children, in much the same way as incense stick making was in India and Pakistan. Thus, in South East Asia, only in the pyrotechnic sector did the child (and other) workers seem to suffer from work-related health hazards. However, while the sectors in South Asia were by no means selected for their health hazards, in nearly all the activities in both regions there were discernible health impacts related to homework. Gender and time use of children By and large, there are gender differences in all countries in the nature of the work children are doing and time use outside of work. There are the beginnings here of the gender differences that emerge later, and the feminisation of homework that we commented on earlier. However, in all countries the mothers expressed a desire to see both their sons and daughters in school, though the girls in South Asia may be supported in school for fewer years than the boys. In both India and Pakistan among the children in homework households there was a higher share of girls engaged in homework than boys. In India a little over half the children in homeworker households are girls.33 In Pakistan boys spent significantly less time doing homework and chores than girls did and significantly more at play and on schoolwork. Other than shopping, girls are relied on more heavily than boys to perform household chores. A higher proportion of boys go shopping, since that requires leaving the house; girls are not encouraged to leave the house. These cultural traits are also reflected in the fact that if girls engage in an income-earning activity, homework is preferred to working outside the home. The female children often have a double burden (as we discussed for women in Chapter 4, Figure 4.1). Indeed, to the hours of homework should be added the hours of household chores. In Figure 5.1 we report the weekly hours of activities

150

Santosh Mehrotra and Mario Biggeri 100 90

Hours of household chores

80 70 60 50 40 30 20 10 0 0

10

20

30

40

50

60

70

80

90

100

Hours worked during a week

Figure 5.1 Double burden for female children (5–14) in Pakistani homeworker households: hours per week in homework and hh chores.

performed by female children (aged from 5 to 14) in Pakistan . These girls are engaged in homework where they work long hours (reported on the x-axis). However to these hours we should add the hours of household chores performed during the same week by the same girl. The result is the ‘double burden’ that the girl child faces. In India the time spent by working children outside working hours is allocated as follows: cooking 1.3 hours, care of sibling/sick 1.5 hours, maintaining the house and outside work 1 hour each. In Pakistan, the working children spent on average across the sectors, 2.2 hours doing chores, 0.9 hours doing homework, 1.3 hours at meals and 1.6 hours at play. In Indonesia, there is a gender bias in the employment of children in homework. The survey found girls more likely to be working. Of the 41 children engaged in homework, 28 were girls, 13 were boys. In the batik homework communities, boys are more likely to help their fathers who are fishermen or fruit sellers. Hence, it is unlikely that a boy will be found waxing batik patterns. These skills acquired by girls during childhood will be useful after marriage, when social norms require that married women stay at home and take care of the family, while having a skill which they can offer in the local labour market as a secondary

Child labour in homeworker households 151 source of income. However, both boys and girls were found in the other two homework industries: weaving rattan for drawers for chests, or shaping painting and packing pottery. If both boys and girls were in homework, there were no differences between boys and girls in the number of hours worked. More girls than boys, however, were engaged in helping in household chores. In the Philippines, the mothers were asked their preferences with respect to the gender of children for homework. The replies of the respondents were consistent with the actual pattern of the gender distribution of children in homework. Thus, male and female children were almost equally preferred in two activities (okra, Christmas balls), boys in two (metalcraft, Christmas lights) and girls in one (fashion accessories). As regards the women’s preferences on the gender of children expected to assist in household chores, most (61 per cent) preferred girls while only a quarter (28 per cent) expressed a preference for boys. Besides working, children among homeworker households spend nearly two hours per day on household chores. In Thailand, until a few decades ago, girls had much less opportunity for education than boys as boys could stay in the monastery and study with monks. Even after a formal education system was established, girls did not attend school as much on account of traditional views on female roles. However, the study notes that this trend was reversed a decade ago and parents value the education of boys and girls equally (61 per cent of responses). Some (8.5 per cent of parents) preferred to educate their girls, as daughters tend to maintain close ties with parents even after their marriage. They are often the ones who help support their parents in old age. In terms of time use, both boys and girls spent six hours in the day at school, two hours at play, one for homework and one or two hours helping with household chores. Children spend approximately a quarter of the day at school and about two hours on play and recreation. An hour is allocated for doing homework and reviewing school lessons. The distinction is found in the average time spent on helping with household chores. The gender dimension in homeworker household could be appreciated even more by looking at children’s working status by sex and age. Indeed, as female children get older they are involved more in homework, and children in the ‘neither’ category increase their household chores. Voices of children Focus group discussions were held not only with the women homeworkers, but also the child workers (for the case studies of children see the country chapters in Part II of the book). In India, in the zardosi sector (garment embroidery in Lucknow) the children spoke in the FGD about long working hours, at times as long as 12 hours a day. Those who were studying felt that work hampers their studies. They would therefore like their school hours to be arranged so that their working hours are taken care of. In bidi-making (Tamil Nadu) the younger children said they could devote more time to studies if they did not work. But at the same time, they said, it was not

152 Santosh Mehrotra and Mario Biggeri possible to give up bidi work; a minimum specified work, needing two hours each day, had to be done, failing which they were punished. A few of them, meanwhile, were struggling to prove themselves to be good students so that they would also be entitled to a scholarship (like some older children) from the welfare fund. From the bidi-making children in Madhya Pradesh, two points emerged. One, children started participating in rolling bidis at the age of 10, and stop going to school between 10 and 15 years of age. Since boys are expected to support their family, they said, their education is taken more seriously. Second, children were drawn into the occupation out of economic compulsion, death or loss of job of the main earner in the family. This observation underlines the need for some form of social insurance for homeworker households, which mitigates their vulnerability. It also underlines the importance of exogenous shocks to the household – emphasised in our theoretical framework in Chapter 1 – which lock families into an inter-generational transfer of the child labour phenomenon. In the incense stick sector, the girls said they had been engaged in agarbathi rolling from (on an average) age 5 onwards. The participants in both the rural and urban FGD were girls since boys normally do not engage in agarbathi rolling. None of the girls wished to continue with this occupation. Those who have discontinued schooling wanted to switch over to tailoring and embroidery, while girls still enrolled have ambitions of becoming a teacher. The girls in urban areas were more aware and ambitious since some of them said they wanted to become doctors and computer professionals. Most interestingly, the girls knew that their mothers never give the money earned from agarbathi rolling to the fathers. The girls approved of that since they felt that the fathers were irresponsible, and it is the mothers who ultimately take care of the household expenses. The voices of the children in homework in Pakistan cry out for attention. In carpet-weaving, parents had taken a loan against their children’s labour. In the FGD the children said they often worked in the factories while their mothers worked at home.34 The incidence of physical abuse, especially of children, was very high. When children were asked what they would prefer to do/be in the future, 85 per cent of them expressed the desire to go to school full time. Most of the children wanted to go to school but knew that their parents could not afford that or to give them some pocket money out of their contribution to household earnings. For this reason, some (12 per cent of the children) felt that they needed to work part time while going to school. Meanwhile, the children going to school complained that they could not concentrate on their school homework due to part-time work. In the incense stick sector women did not indicate that there was any physical abuse by subcontractors, but verbal abuse, especially of children, was routine. Most of the children went to madrasas, since regular schools were not seen to be affordable. In the prawn-shelling sector, children worked in the shed (‘warahs’) and at home. Contractors preferred that the children worked at the warah since it was easier for them to control the children, push them into working faster and monitor to ensure there is no stealing. For homework, children fetched the prawns from

Child labour in homeworker households 153 the contractor and brought them home. Starting from an early age, the children worked long hours. They reached the warahs often at 4 am and continued working until 11.30 am. Children worked similar hours at home because there are no storage facilities and delay would mean that the highly perishable product would spoil. The contractors and their aides physically abused the children. Since a prawn is quite delicate, its tail will break easily during shelling. Children said they could be severely beaten for such a mistake. In sack stitching, in most cases children fetched the material from the contractors. Since the children were in direct contact with the contractor, they were often exposed to verbal abuse, but no physical abuse was reported. Girls mentioned that only they did the homework because sewing is only for females. According to the mothers, girls were more easily controlled than the boys. Thus gender stereotyping was taking place at several levels within the household. In Indonesia, three FGDs with children were organised, one in each sector. In batik and pottery industries, most children were working together with friends rather than alone, which enabled them to chat freely. In rattan, the children work at home with their mothers. The children said that it is not possible for them to work in the rattan factory/workshop because they are not skilled yet and also because they have to go to school. Factory work is a full time job and they cannot go to school. If given the freedom to choose, the children were asked if they would choose school, work only, or school and work. They preferred school and work, since work gave them an income. They said that if their parents could afford further education, they would continue. If not, they would quit school. But the study found no gender bias in school attendance. In Thailand, two FGDs with children were organised – in saa paper and in the hybrid seed production sectors. The work involving children was light and limited.35 But they were allowed to help only at weekends and during school vacations. In the second FGD (hybrid seeds), all attended school. They did not regard themselves as workers. Some said they learnt the skill of pollination from parents and could earn some money by hiring themselves out as pollinators in the neighbours’ farm during weekends (earning 100 baht, or $2.10 at the then market exchange rates, as pocket money). The younger ones (6–11) helped their parents in simple activities at the post-harvest stage, thus earning about 20 baht a day as pocket money. In the Philippines, the children said that they would like to both go to school as well as work. As regards future aspirations, the most common response of 5–14-year-old children in homework was a desire for white-collar jobs. This is true for both boys and girls. The second most common reply was the desire to finish their studies, a goal that was strongest among girls. Among the 15–19-yearolds, getting a white-collar job becomes less important, and schooling more a priority. Girls in this age group also expressed an interest in working overseas (given that Filipinos have migrated abroad in large numbers to unskilled jobs). The children also said that their mothers work more than their fathers; fathers have leisure time, drink and make trouble; mothers do not.

154

Santosh Mehrotra and Mario Biggeri

5.3

The determinants of children’s work status and hours worked: an empirical analysis for India, Pakistan and Indonesia

The determinants of child status The objective of this sub-section is to provide some empirical evidence on the determinants of child labour. As we saw in Section 5.1, the status of the child – work only, study only, work and study or neither work nor study – depends upon various factors: CSt  f (CCt, HEhht, EEhht, vt) (t  1) The dependent variable, CS is the child activity status: the probability of being in one of the four states (relative to one such category, for example working only). The activity status of the child depends upon various factors: the child characteristics (CC); the human endowment of the household (HEhh); the economic endowment of the household (EEhh) and the external factors to household (v, not examined here) that may influence the status of the child. When the data consist of such choice-specific attributes, the multinomial logic model (conditional) is the most appropriate to understand how a factor may influence the CS since the conditional probability among two status gives the direction and the magnitude of the effect of a factor. Prob (Yi j)

e ␤zij e ␤zij

兺j

Where, j  1, 2, . . . , J for a total of J alternatives (in our case 4 alternatives). The data utilised in the analysis are from the surveys in India, Pakistan and Indonesia (for the variable characteristics see Table 4.16 in Chapter 4). The estimations are carried out for the homeworker households only. As explained in the Annex 2.1 the results can be extended at country level for homework in the three sectors on aggregate in India, for urban/slum areas in Pakistan, and for the Central and West provinces of Indonesia.36 A multinomial regression is used for estimation of the coefficients for each country separately (Table 5.9). The comparison category of child status is ‘only working’. In our view working (only) is the worst possible status for the child. We would prefer the child to be either studying and working, or studying full time – the latter being the best case scenario. The results regarding the neither category are not reported for their ambiguity connected to reasons expressed in Section 5.2. India We first compare the status of studying full time with working only: increasing age of the child and exogenous shocks the family may have faced are likely to push

Table 5.9 Determinants of child status: results of a multinomial logit regression (reference group: working only) Coef.

Std. err.

z

Pz

dy/dx

India, nobs  562; LR 2 (27)  223.68; Prob 2  0.000; Pseudo R2  0.164 Study only Age [of the child] 0.577 0.070 8.22 0.000*** 0.0752 Female [dummy for child’s 0.326 0.268 1.22 0.224 0.0313 gender, female 1] Edum d [dummy for the mother’s 0.660 0.305 2.16 0.030** 0.1008 education/literacy, yes 1] Age dependency ratio [((0–14)
0.050 0.172 0.29 0.773 0.0132 (61_)) / (15–60)] Income per capita [of household] 0.000 0.000 0.81 0.419 0.0000 Organ [dummy, organisation 0.572 0.456 1.25 0.210 0.0818 membership, yes 1] Upper cd [dummy for being upper 0.340 0.342 0.99 0.320 0.1370 caste, yes 1] Home owned [dummy, yes 1] 0.696 0.291 2.39 0.017** 0.0526 Exogenous shock (without father) 2.076 0.468 4.44 0.000*** 0.3494 [dummy, yes 1] Constant 6.832 0.953 7.17 0.000 Study and work Age [of the child] 0.247 0.078 3.18 0.001*** 0.0326 Female [dummy for child’s 0.072 0.310 0.23 0.817 0.0326 gender, female 1] Edum d [dummy for the 0.683 0.338 2.02 0.043** 0.0321 mother’s education/literacy, yes 1] Age dependency ratio 0.050 0.188 0.26 0.792 0.0198 [((0–14)
(61_)) / (15–60)] Income per capita [of household] 0.000 0.000 2.43 0.015** 0.0000 Organ [dummy, organisation 1.770 0.464 3.81 0.000*** 0.2469 membership, yes 1] Upper cd [dummy for being 0.768 0.382 2.01 0.045** 0.1814 upper caste, yes 1] Home owned [dummy, yes 1] 0.555 0.339 1.64 0.101 0.0078 Exogenous shock (without father) 0.771 0.473 1.63 0.103 0.0872 [dummy, yes 1] Constant 0.951 1.084 0.88 0.380 Pakistan, nobs  997; LR 2 (24)  424.36; Prob 2  0.000; Pseudo R2  0.179 Study only Age [of the child] 0.170 0.055 3.11 0.002*** 0.0010 Female [dummy for child’s 1.924 0.289 6.65 0.000*** 0.0865 gender, female 1] Edum d [dummy for the mother’s 1.109 0.472 2.35 0.019** 0.0921 education/literacy, yes 1] Age dependency ratio 0.198 0.131 1.51 0.130 0.0096 [((0–14)
(61_)) / (15–60)] Chronic poverty [dummy share of 0.580 0.281 2.07 0.039** 0.0331 food expenditure on tot. yes 1] Collective action [dummy, 0.829 0.359 2.31 0.021** 0.0414 collective action of hwers, yes 1] (Table 5.9 continued)

Table 5.9 Continued Coef. Home owned [dummy, yes 1] Loan burden [loan stock as a share of total expenditure] Constant Study and work Age [of the child] Female [dummy for child’s gender, female 1] Edum d [dummy for the mother’s education/literacy, yes 1] Age dependency ratio [((0–14)
(61_)) / (15–60)] Chronic poverty [dummy share of food expenditure on tot. yes 1] Collective action [dummy, collective action of hwers, yes 1] Home owned [dummy, yes 1] Loan burden [loan stock as a share of total expenditure] Constant

Std. err.

z

P z

dy/dx

0.167 0.028**

0.0118 0.0001

0.435 0.315 0.001 0.000

1.38 2.19

0.415 0.656

0.63

0.082 0.044 0.576 0.219

1.88 2.63

0.061* 0.009***

0.0289 0.0211

1.220 0.379

3.22

0.001***

0.1914

0.091 0.101

0.90

0.366

0.0038

0.149 0.214

0.70

0.485

0.0310

1.065 0.270

3.94

0.000***

0.1173

0.644 0.266 0.001 0.000

2.42 3.13

0.016** 0.0429 0.002*** 0.0001

4.06

0.000

2.373 0.584

0.527

Indonesia, nof obs  321; LR  (15)  131.55; Prob   0.000; Pseudo R2  0.195 Study only Age [of the child] 1.079 0.195 5.53 0.000*** 0.0800 Female [dummy for child’s 0.062 0.470 0.13 0.895 0.1100 gender, female 1] Edum d [dummy for the 0.903 0.518 1.74 0.082* 0.0500 mother’s education/literacy, yes 1] Age dependency ratio [((0–14)
0.749 0.377 1.98 0.047** 0.1700 (66_)) / (15–65)] Expenditure per capita 0.000 0.000 1.72 0.086* 0.0000 [of household] Constant 15.025 2.878 5.22 0.000 Study and work Age [of the child] 0.722 0.197 3.66 0.000*** 0.0400 Female [dummy for child’s 0.898 0.507 1.77 0.077* 0.1100 gender, female 1] Edum d [dummy for the mother’s 1.361 0.592 2.30 0.021** 0.0700 education/literacy, yes 1] Age dependency ratio [((0–14)
0.135 0.374 0.36 0.719 0.1000 (66_))/(15–65)] Expenditure per capita 0.000 0.000 1.76 0.078* 0.0000 [of household] Constant 7.772 2.917 2.66 0.008 2

2

Notes Significant at 1% (***), significant at 5% (**) and significant at 10% (*); dy/dx is for discrete change of dummy variable from 0 to 1.  Dummy for poverty based on the share of expenditure in food items on total expenditure (‘chronic poor’  75%  1, else  0).

Child labour in homeworker households 157 the family into employing the child in full-time work. As age increases, the probability of studying full time decreases relative to full-time work. The marginal effect is 7.5 per cent, that is, as age increases by one year, the probability of working increases (studying decreases) by 7.5 per cent. Second, if the child is without a father, the probability of the child working full time is rather high, increasing by as much as 35 per cent.37 On the other hand, the human capital endowment of the household seems to have a positive inter-generational effect, that is, having an educated mother increases the probability of the child studying full time by 10.1 per cent. Also, the ownership of a house by the homeworker household (i.e. the economic endowment of the household) is also associated with an increase in the probability of the child studying full time; the marginal effect of ownership is 5.3 per cent. Gender does not seem to affect the probability of studying, as the coefficient is non-significant (even though the sign suggests that girls have less probability of studying full time). The age-dependency ratio is non-significant too (but again the sign suggests that the number of dependants reduces the probability that the child is studying full time). The income per capita38 and membership of a homework organisation, although non-significant, show a positive sign.39 Comparing the status of working and studying with working only, age, education of parents, income per capita, organisational membership and being upper caste are significant determinants of the probability of ‘working and studying’ instead of working full time. Consistent with the findings in the preceding paragraphs, as the age of the child increases, the child is more likely to work only, rather than work and study. Annual increases in age increase the probability of the child being in full-time work by 3.3 per cent. The consistency of this result suggests that there is a case for scholarships for children as they graduate from primary school into junior secondary, so that they do not drop out. The remaining factors seem to favour the child studying and working, rather than only working. Thus, as we saw earlier, the education of the mother increases the probability (by 4.3 per cent) of the child working and studying, rather than being in full-time work. An increase in income per capita also increases the probability of the child working and studying, rather than only working. The marginal effect is, however, low. It is plausible to argue that the reason for this low magnitude is that the income range of homeworker households is rather narrow, and the households are homogeneously poor. The low marginal effect of income per capita is also found to hold for Pakistan (consumption) and Indonesia. The implication is that any collective or public action to increase the low piece-rates to homeworkers would help the children as well. In this context, the regression results suggest that collective action by homeworkers may be particularly important. The membership of a homeworker in a homeworkers’ organisation increases the probability of her child studying and working, rather than being in full-time work – which, by itself, is a remarkable finding. The marginal effect is high at 25 per cent.40 The position of the household in the social hierarchy also seems to matter. Being upper caste decreases the probability of the child working full time – by 18 per cent. On the other hand, being lower caste (i.e. backward caste,

158

Santosh Mehrotra and Mario Biggeri

scheduled caste, scheduled tribe and other) increases the probability of the child working full time to the same extent. House ownership (a marker of wealth) and the child lacking a father (a marker of vulnerability) are very close to being significant, and have the predicted sign. Home ownership, that is assets in the household, makes it conditionally probable that the child studies and works, rather than only works.41 Pakistan Comparing the conditional probability of study only with working as homeworker only, the age and gender of the child, the education of the mother, the share of food and loans in total family expenditure, and the participation of adult women workers in collective action have an effect on probability of the child being in school full time (as against working only). As in India, as the age of the child increases, the probability of the child studying full time decreases. If the child is female, the likelihood of the child being in school full time decreases (by almost 8.6 per cent). Also as in India, the literacy of the mother increases the probability that the child will be studying full time; in fact, the marginal effect of the mother’s literacy is strong, leading to a 9.2 per cent greater probability that the child will study full time rather than work full time. Finally, the participation in collective action (to obtain better piece-rates) by adult women increases the probability of the child studying full time, by 4.1 per cent. Non-participation has the opposite effect. Furthermore, if the homework family allocates over 75 per cent of its household expenditure to food (indicating how poor it is), and as the share of loan stock in family expenditure increases, it increases the probability of the child working full time (rather than studying full time). The age-dependency ratio and ownership of home are non-significant but with the predicted sign. Comparing the conditional probability of studying and working with working as homeworker only, the age and gender of the child, the education level of the mother, membership in a homework organisation, ownership of home, and share of loan stock in family expenditure seem to have an impact on the probability of studying and working. As the child’s age increases, the probability also increases of the child studying and working rather than working only.42 The girl child is highly likely to be working full time, rather than going to school and working. Also, the higher the share of loan stock in household expenditure the higher the probability that the child will only be working, rather than studying and working. On the contrary, the literacy of the mother increases the probability (by 19 per cent) that the child is not in full-time work, but is rather combining study with work. Very importantly, the participation in collective action by homeworkers by parents increases the probability (by nearly 12 per cent) of the child studying and working, rather than merely working. Asset ownership (in the form of ownership of home) favourably affects the probability of combining studying with work, and is unfavourable to the child working only; the marginal effect is 4.3 per cent. Age-dependency ratio and the share of food in family expenditure are non-significant.

Child labour in homeworker households 159 Indonesia Comparing the probability of studying only with working only, as in India and Pakistan, the probability of the child only studying decreases relative to ‘only working’ with age. As in India and Pakistan, the education level/literacy of the mother increases the probability that the child will only study, rather than only work; having an educated mother increases that probability by 5 per cent. In other words, both age and mother’s education level seem to be robust determinants of the work/study status of the child. Finally, the higher the expenditure per capita of the household the higher the probability that the child will only study, rather than only work. Thus, both in respect of human and economic endowments of the household, the results are similar to that in India and Pakistan. The more dependants there are in the homeworker’s family, the less likely that the child will only study rather than only work; the marginal effect is 16.6 per cent.43 Comparing the probability of working and studying with working only, as in Pakistan and India, age increases the probability that the child will work and study, rather than be in work only. The girl-child is more likely to work and study rather than work only (unlike Pakistan); being female increases this probability by 11 per cent. Again, as in India and Pakistan, the education of the mother increases the probability that the child is working and studying; the marginal effect is 7 per cent. Finally, the higher the household’s per capita expenditure the higher the probability that the child will study and work, rather than only work. To summarise, the age of the child, the mother’s education level, and per capita income/expenditure in the family are significant and robust determinants of the child’s activity status. Also, while in Indonesia the particular sectors examined did not have homeworkers engaging in collective action, in India and Pakistan the mother’s membership in an association is an important determinant of child activity status. The determinants of hours worked by children Having examined the determinants of the probability of studying/working and only studying (versus only working), we also explored the determinants of the hours of work put in by children that were working. We assume that hours worked by the child is a function of the following independent variables: age, sex, the fact that the child attends school, the education of the mother, the age-dependency ratio and the expenditure/income of the household. The Heckman selection model assumes that the dependent variable is not always observed. In our case the dependent variable, hours worked, is observed only if the child works (Rosati, 1999; Ray, 2000; Rosati and Rossi, 2001). The likelihood of working (i.e. if the dependent variable is observed) is a function of other variables. The estimates of Heckman regression models with selection are carried out using full maximum-likelihood. From the empirical results (Table 5.10) we can see that the estimates pass the Wald test, and for all the countries the test for  and 2 are both significantly different from zero. This clearly justifies the use of the Heckman selection equation.

Table 5.10 Determinants of child working hours: results of a Heckman selection model regression (with sample selection, full maximum-likelihood) Coef.

Std. err.

z

Pz

India, nobs  562; Censored obs  405, uncensored obs  157, Wald 2 (6)  41.69; Prob 2  0.000a Hours worked Age [of the child] 0.726 0.159 4.58 0.000*** Female [dummy for child’s 1.462 0.564 2.59 0.010*** gender, female 1] Attending school [dummy 1.297 0.551 2.35 0.019** for child, yes 1] Edum [mother’s level of education] 0.022 0.078 0.28 0.781 Age dependency ratio [((0–14)
0.140 0.284 0.49 0.621 (61)) / (15–60)] Income per capita [of household] 0.000 0.000 2.39 0.017** Constant 6.998 2.702 2.59 0.010 Select Location [urban slum, dummy, yes 1] 0.426 0.135 3.17 0.002*** Household surroundings [dummy: 0.317 0.135 2.35 0.019** dirty, yes 1] Age [of the child] 0.194 0.027 7.06 0.000*** Female [dummy for child’s gender, 0.458 0.122 3.75 0.000*** female 1] Attending school [dummy for 0.259 0.139 1.86 0.062* child, yes 1] Edum [mother’s level of education] 0.010 0.020 0.49 0.623 Age dependency ratio [((0–14)
0.025 0.078 0.32 0.749 (61)) / (15–60)] Income per capita [of household] 0.000 0.000 0.86 0.391 Constant 2.994 0.414 7.23 0.000 Athrho 1.464 0.415 3.52 0.000*** lnsigma 1.305 0.132 9.87 0.000***  (rho) 0.898 0.080  (sigma) 3.687 0.487
(lambda) 3.312 0.714 Pakistan, nobs  1001; Censored obs  702, uncensored obs  299, Wald 2 (6)  151.68; Prob 2  0.000b Hours worked Age [of the child] 0.930 0.089 10.41 0.000*** Female [dummy for child’s 3.814 0.496 7.69 0.000*** gender, female 1] Attending school [dummy 1.705 0.458 3.72 0.000*** for child, yes 1] Edum d [dummy for mother’s 0.403 0.781 0.52 0.606 education/literacy, yes 1] Age dependency ratio [((0–14)
0.042 0.178 0.24 0.813 (61))/ (15–60)] Chronic poverty [dummy, share 0.225 0.364 0.62 0.536 of food expenditure on tot. yes 1] Constant 10.795 1.307 8.26 0.000 Select Location of cluster 0.063 0.030 2.11 0.035**

Table 5.10 Continued Std. err.

z

Pz

0.000

0.000

0.87

0.382

0.223 0.963

0.018 0.100

12.12 9.65

0.000*** 0.000***

0.243

0.115

2.11

0.035**

0.125

0.200

0.62

0.108

0.044

2.43

0.015**

0.010

0.091

0.11

0.913

3.172 1.960 1.366 0.961 3.918 3.766

0.237 0.248 0.065 0.019 0.256 0.305

13.39 7.89 20.88

Coef. Loan per child [amount of loan/ children number 5–14] Age [of the child] Female [dummy for child’s gender, female 1] Attending school [dummy for child, yes 1] Edum d [dummy for mother’s education/literacy, yes 1] Age dependency ratio [((0–14)
(61)) / (15–60)] Chronic poverty [dummy, share of food expenditure on tot. yes 1] Constant Athrho lnsigma  (rho)  (sigma)
(lambda)

0.532

0.000 0.000*** 0.000***

Indonesia, nobs  321; Censored obs  237, uncensored obs  84, Wald 2 (6)  67.31; Prob 2  0.000c Hours worked Age [of the child] 0.926 0.163 5.67 0.000*** Female [dummy for child’s 0.915 0.647 1.41 0.158 gender, female 1] Attending school [dummy for 2.443 0.735 3.33 0.001*** child, yes 1] Edum d [dummy for mother’s 0.382 0.747 0.51 0.609 education/literacy, yes 1] Age dependency ratio [((0–14)
1.356 0.453 2.99 0.003*** (66)) / (15–65)] Expenditure per capita [of 0.000 0.000 0.50 0.616 household] Constant 11.908 2.884 4.13 0.000 Select Location of cluster 0.055 0.070 0.78 0.433 Language of parents [language 0.473 0.247 1.92 0.055** used, Indonesian dummy, yes 1] Age [of the child] 0.242 0.038 6.40 0.000*** Female [dummy for child’s gender, 0.148 0.163 0.90 0.366 female 1] Attending school [dummy for 0.344 0.204 1.69 0.091* child, yes 1] Edum d [dummy for mother’s 0.020 0.277 0.07 0.942 education/literacy, yes 1] Age dependency ratio [((0–14)
0.322 0.127 2.53 0.011** (66)) / (15–65)] Expenditure per capita [of 0.000 0.000 0.40 0.690 household] (Table 5.10 continued)

162

Santosh Mehrotra and Mario Biggeri

Table 5.10 Continued

Constant Athrho lnsigma  (rho)  (sigma)
(lambda)

Coef.

Std. err.

4.144 2.561 1.270 0.988 3.561 3.519

0.629 0.804 0.128 0.019 0.456 0.503

z 6.59 3.18 9.93

Pz 0.000 0.001*** 0.000***

Notes a LR test of indep. Eqns (  0): 2 (1)  5.63 Prob2  0.0176. b LR test of indep. Eqns (  0): 2 (1)  21.64 Prob2  0.0000. c LR test of indep. Eqns (  0): 2 (1)  13.28 Prob2  0.0003.  Dummy for poverty based on the share of expenditure in food items on total expenditure (‘chronic poor’  75%  1, else  0). Significant at 1% (***), significant at 5% (**) and significant at 10% (*).

The results suggest that hours worked increased with increasing age in all countries. Thus not only is the CS importantly affected by the child’s age, but so are the hours worked. While gender did not seem to be a robust determinant of a child’s activity status, being female meant that you worked longer hours than boys in India (in Indonesia the variable is positive but non-significant). This is quite consistent with the earlier analysis in this chapter suggesting the ‘feminisation’ of homework. If the child went to school, it reduced the number of hours worked in the two countries. Also, the hours worked by the child fell with an increase in the per capita income of the household. The effect of the education of the mother is negative (i.e. as it increases the probability is that children work lesser hours), however it is non-significant in the three countries.

5.4

Concluding remarks

We would like to summarise the main findings on child labour in subcontracted home manufacturing in Asia. Overall, there is a sharp distinction between the experience of children in homework in South Asia on the one hand and South East Asia (Indonesia, Philippines, Thailand) on the other. Second, there are some important contrasts and fewer similarities between the situation of Indian and Pakistani children. The percentage of children who are only working, rather than studying and working, is much higher in the South Asian cases than in the South East Asian ones. It is not as though the phenomenon of child work is not prominent in South East Asia. Even in Thailand, a significant proportion of homeworker households have children who are working (some in homework activities, others probably farming). But it is the phenomenon of quite widespread involvement of children in work that seems striking across the board in Asia, regardless of whether it is a low-income or middle-income country.

Child labour in homeworker households 163 There is a much greater participation in school among children in homeworker households in South East Asia as compared to South Asia, at all ages. The difference is not very striking at the early ages, but it is much greater for the older age group. There is a strong impression from the data that while children’s contribution to the family work is valued, schooling is valued even more. Perhaps the difference lies in the reasons for not going to school. In South Asia there is a clear perception that they ‘cannot afford’ to go to school, or that ‘school is too expensive’. In South East Asia parents are willing and able to support their children going to school. In terms of hours worked by children, by and large the number of hours appears to be close to or over the limit of what is practical to combine with schooling. However, in Pakistan the number of hours are around 50 per cent higher than in other countries – consistent with a very high proportion of children not being in school. Within South Asia, there is a sharp contrast between India and Pakistan in some respects. In Pakistan a much higher proportion of children is out of school. Also, a much higher share of children are only in work. Similar proportions of children in both countries are engaged in both work and school. The contrasts are particularly striking in the one sector that is common across the two countries – incense stick making (agarbathi). Nevertheless, the phenomenon of debt bondage exists to some extent in both countries. The analysis reveals several other major findings. First, child labour exists – though to a much lesser extent – even in countries where primary schooling is almost universal (i.e. the South East Asian countries), and where poverty is lower. Second, more of the older children in homeworker households tend to work than in CG households and that incidence is also higher than the national incidence for child labour in a similar age group. In general, children from homeworker households have a higher probability of working than the children from CG households. Third, there is clear evidence of the feminisation of homework from childhood, and female children have a double burden to carry out especially in South Asia. Thus, more boys are in school than girls, more girls are at work than boys and the hours worked are greater for girls than for boys. These are analogous trends to those for adult women (see Chapter 4). Fourth, we found that the contribution of children who worked in homeworker households was considerable in all countries except perhaps Thailand. Finally, except in Pakistan, most of the young children do not ‘only work’; they work and study or only study. However, the pull factor of work and the push factor of unaffordable (and possibly poor-quality) schooling combine to induce dropping out from school. We argued earlier that homework within the household reduces the fixed costs for children (and parents) of finding work outside the home, since it reduces the transportation costs, transaction costs and allows for a higher divisibility of work ‘contracts’ inside the household business. Finally the regression results show that, together with other determinants, as age increases children are more likely to work in homework. The mother’s education level and per capita income/expenditure or assets in the household

164

Santosh Mehrotra and Mario Biggeri

were important determinants of the CS. Public or collective action that increases piece-rates for workers holds out the prospect of improving the child’s well-being. Moreover, collective action plays a role in the reduction of children ‘only working’, as underlined by the econometric results for India and Pakistan. Finally, the hours that children work in both countries suggest that their ability to do school-related homework is likely to be affected. The problem becomes more severe as the age of the child increases and if the child is a girl, she works longer hours. Policy measures could be directed to diminish the child exclusion and drop out from school by reducing the fixed costs of attending school and by increasing the returns to school (by improving the quality of schools and making schooling more suitable for the local economic system).

Annex 5.1: definition of child activities Child labour is classified as children’s work which is of such a nature or intensity that it is detrimental to their schooling or harmful to their health and development. The concern is with children who are denied their childhood and a future, who work at too young an age, who work long hours for low wages, who work under conditions harmful to their health and to their physical and mental development, who are separated from their families, or who are deprived of education. Such child labour can create irreversible damage to the child and is in violation of international law and usually, national legislation. (ILO, 2002c, p. 3) In other words child labour refers only to negative or undesirable forms of work that should be eliminated in accordance with the Minimum Age Convention (no. 132) and especially the Elimination of Worst Forms of Child Labour Convention (no. 182) and Art. 32 of the UN Convention on the Rights of the Child. A recent report from the ILO’s International Programme on the Elimination of Child Labour (IPEC) specifies four groups of child workers (ILO, 2002b, pp. 29–36): 1 2

3

4

Children at work in economic activity. Children engaged in child labour. This group includes (a) all economically active children 5 to 11 years of age; (b) economically active children aged 12 to 14 years, except those doing light work only for less than 14 hours per week; and (c) children aged 15 to 17 years engaged in hazardous work. Children in hazardous work: work that is likely to harm health, safety or moral development of a child. In addition to children employed in mining, construction or other hazardous activities, this group includes all children below 18 years of age who work 43 hours or more per week. Children in unconditional worst forms of child labour, as defined by ILO Convention No. 182. This includes children in forced or bonded labour, armed conflict, prostitution and pornography, and illicit activities.

Child labour in homeworker households 165 Since the last is a sub-set of the third44 and according to the age classification45 and time involvement, they can be summarised as follows: Child work Child labour

At least one hour of economic activity per week. Ages 5–11: at least one hour of economic activity per week. Ages 12–14: at least 14 hours of economic activity per week. Hazardous work At least 43 hours of economic activity per week. Although these definitions are important in classifying children’s work, it is relevant to note that the Country Statistics empirical and theoretical literature employ the broader concept of children’s work rather than the narrower one of child labour, because of the difficulties in drawing a statistical line between acceptable forms of work, on the one hand, and harmful forms that need to be eliminated, on the other.

Age in years

18

15*

12**

Work excluded from minimum age legislation***

Light work

Non-hazardous work

Hazardous work

Unconditional worst forms of child labour

Shaded area = child labour for abolition Source: ILO (2004b). Notes * The minimum age for admission to employment of work is determined by national legislation and can be set at 14, 15 or 16 years. ** The minimum age at which light work is permissible can be set at 12 or 14 years. *** For example, household chores, work in family undertakings and work undertaken as part of education.

Therefore, almost all empirical and theoretical studies on child labour classify children activity status into four mutually exclusive categories. This implies that

166

Santosh Mehrotra and Mario Biggeri

all children can be assigned to the following categories: 1

Only working, that is, children who are economically active (at least one hour per week).46 Only studying, that is, children who go to school but do not work. Working and studying, that is, children who combine school attendance with work. Neither working nor studying or idle (including household chores).

2 3 4

These definitions do not take into account that household chores can deprive children from their capabilities as well as work or child labour (see Biggeri, 2003a).

Notes 1 The ILO launched in 1992, the International Programme on the Elimination of Child Labour (IPEC); the ILO’s Statistical Information and Monitoring Programme on Child Labour (SIMPOC) began in 1998 collecting data. United Nations Children’s Fund (UNICEF) was called in 1995 by the US Congress Sub Committee on Labour for hearings on exports of goods produced by child labour (attended by Santosh Mehrotra on behalf of UNICEF), in the aftermath of the introduction of a bill by Senator Harkin, attempting to ban imports of goods produced by child labour. UNICEF produced a global position on the issue, and later appointed an advisor on child labour issues at its headquarters in New York. The Dutch and Norwegian governments organised an international conference on child labour in 1997. The World Bank produced a position paper on child labour, and started a new research programme on the subject around the late 1990s. The Nordic donors, in particular, decided to support a new research programme in 2000 to be jointly run by the World Bank, ILO and UNICEF, at the Innocenti Research Centre, Florence, UNICEF’s research institute. Santosh Mehrotra, who was leading the developing countries’ research programme for the UNICEF research institute, was directly involved in supervising this programme. 2 For example, over the late 1990s there were newspaper reports in India that the banglemaking industry in Firozabad, Uttar Pradesh, has seen a shift of child work from the small workshop to homework. See also Lietew et al. (2004). 3 However, in the Philippines, the incidence of child work in homework in our sample is comparable to that in India (though not as high as in Pakistan). 4 According to ILO estimates ‘It is estimated that there were some 211 million children ages 5 to 14 at work in economic activity in the world in 2000. . . . About 73 million working children are less than 10 years old. . . . The total economically active child population 5–17 years old is estimated at 352 million children’ (ILO, 2002b, p.15). Children under 5 years of age, which are not included in these four groups, are generally considered too young to be working. See also ILO (2003) and UNICEF (2003). 5 The main sources of data on child labour are national surveys done with the support of the ILO (SIMPOC surveys), the World Bank (LSMS) and UNICEF (the Multiple Indicator Cluster Surveys) over the 1990s. Data from these surveys was collated on a common website (www.ucw-project.org) by an inter-agency project Understanding Children’s Work (UCW) based at the UNICEF Innocenti Research Centre, Florence. 6 See, among the others, Basu, 1999 and Cigno et al., 2002. 7 The factors affecting the human development of a child are: 1

Endogenous (to the household) factors affecting child’s human development: (a) Human endowment: education of mother and father; health status of father and mother; child growing up in single-parent home; demographic characteristics

Child labour in homeworker households 167

2

(size of household, its age structure); water, sanitation, environmental quality; institutions (tradition, culture and religion). (b) Economic endowment: income; assets. Exogenous (to the household) factors affecting child’s human development: (a) Basic services in the local area: safe water, sanitation, environmental quality; preventive and basic curative health services; schools, kindergarten (availability and quality); institutions (tradition, culture, religion); conflicts (of ethnic, religious or civil nature). (b) Economic endowment in the local area: local (national) economic situation; institutions to enable market access; national institutions (laws), and their effectiveness (ban on child labour, compulsory education); social insurance and assistance policies; redistribution policies.

8 Where tables do not contain the relevant information for a particular country, is because the survey did not collect information which was exactly comparable on a cross-country basis. 9 Nearly 7 per cent of the households had other answers which did not fall in these three categories. 10 In Pakistan the CG households were purposively chosen to be those where children were not working. Hence the comparison with the CG households would not hold. 11 All data on national child labour incidence for all five countries are based on the World Bank’s World Development Indicators (2002) data for 1999. 12 The share of working children in homeworker households is nearly 50 per cent in India in our sample in the same age-group, in Pakistan more than 75 per cent and in Indonesia 29 per cent. 13 In Tables 5.1 and 5.2 the younger age group is defined as 5–10, so that the numbers are comparable with Pakistan. In Tables 5.3 and 5.4 the age groups considered follow UNESCO information on school starting age. Hence, they are different for India, since the starting age for school is 6, not 5 (as in Pakistan). In Indonesia the age range is 7–15. 14 For Pakistan we have information only about children in homework, and not those children who might be working outside the home. This information is compared to India and Indonesia in Table 5.4. 15 That is, only studying plus those studying and working. 16 In Tamil Nadu, school attendance rates are normally much higher than in Madhya Pradesh; they may also be higher on account of the coverage of the Welfare Fund scholarship scheme. The coverage of the Welfare Fund benefits has been very uneven between states. 17 That is, only working plus those studying and working. 18 In madrasas they are taught the noorani quaida (basic religious textbook) and use Urdu textbooks. 19 The data is for 6–21 year olds. Unfortunately the Thai study does not disaggregate the data by age, and almost certainly, in the relevant primary and secondary school age, almost all children are in school. 20 Children in homework in two sectors (home décor and okra) earn relatively higher incomes, and the incidence of child labour and absences from school are higher; perhaps the opportunity cost of going to school is high, given their earnings from homework. 21 No information is available in respect of children in the ‘neither’ category for Thailand and the Philippines. 22 The reasons behind children doing homework are not dissimilar to the reasons behind non-enrolment. In the zardosi sector, one important reason for early induction into this work is the desire to pass on a traditional skill; the significant contribution of children’s work to total family hours worked reinforces the low attendance in school in this sector. 23 Similar reasons apply in traditional sectors in other countries: incense sticks (in India and Pakistan), garment embroidery (in India) and carpets (in Pakistan).

168

Santosh Mehrotra and Mario Biggeri

24 This was the FGD at the site of households engaged in home decor production (Rizal province). 25 However, the children engaged in zardosi work seem to be by far the worst off in terms of hours worked, but the younger boys in the incense sticks sector also work long hours. 26 For instance, Heady (2000) finds that work has a substantial effect on learning achievement in the key areas of reading and maths in Ghana, but notes that this may be because those children who work are innately less interested in academic achievement. 27 Heady (2000) finds that work has a substantive negative impact. 28 Data on hours worked in homework were not collected in either Thailand or the Philippines. 29 Some children learned the skill of artificial pollination so as to earn some pocket money. 30 Adult males take the loans and their children work these off by foregoing a part of their daily wage. 31 That children preferred not to work with their parents and preferred to be independent workers was also found in shoes, metal, embroidery and rattan industries a decade ago (Tjandraningsih and White, 1992). 32 In hybrid seed production, children are mainly engaged in post-harvest activities such as washing, deseeding and drying. In chapter products, children helped in such work as picking fresh flowers or cutting the petals for decorating the mulberry chapter. Some helped with chapter craft such as gluing or pasting coloured chapter. 33 In zardosi, three-fifths of the working children are girls. In bidi, over half the working children were girls, while in incense sticks they represent just under two-fifths. 34 Women complained of low piece-rates and delayed payments by the subcontractors; but they felt that big factory owners, rather than the contractors, were the ones profiting at their expense. 35 The girls said that they cut flower petals as decorations, and also cut, glued and pasted the coloured chapter. 36 For an explanation, see Chapter 2. 37 The religion of the household (not reported in the regression for comparison among countries) has a significant impact on the status of the child. In particular, we find that among the Hindu households it is more likely that the children will be studying (and working/studying) rather than only working. On the contrary we find that among Muslim households it is more likely for children working rather than studying (and working/studying). 38 We have inserted income directly in the function since the estimation of the real effect of income is not one of the purposes of the chapter. However, for instance, simulating throughout a bivariate probit for India, an increment of income per capita of 40 per cent increases the probability of the child’s ‘working and studying’ by 3.3 percentage points and ‘only studying’ by 0.3 percentage points. It reduces the probability of the child to be in the activity status ‘working only’ by 1.1 percentage points and in the ‘neither working nor studying’ category by 2.5 percentage points. 39 The sign for upper caste is negative, which is counter-intuitive. However, the coefficient is non-significant. 40 In India, we had an additional regressor (which we do not have in other countries): non-wage benefits by employers. We found that if the employer offers non-wage benefits (e.g. pension, health services) the probability of the child studying and working (rather than only working) increases by 10.6 per cent. 41 The effect of the following variables was non-significant: gender of the child and agedependency ratio. 42 This result could be connected to the fact that the age category used is between 5 and 14 years, while in these urban slums children actually only start to go to school around 6 or older.

Child labour in homeworker households 169 43 Gender is non-significant as a determinant of this child activity status. 44 An estimated 171 million children ages 5–17 were estimated to work in hazardous situations or conditions. There were about 8.4 million children involved in other worst forms of child labour as defined in ILO Convention (trafficking; forced and bonded labour; armed conflict; prostitution and pornography and illicit activities) (ILO, 2002b). 45 ‘Child age groups were broken down in two different ways. For the estimates on economic activity we applied the commonly used age brackets 5–9, 10–14 and 15–17. For the presentation of data on child labour and hazardous child work we cut the first two brackets in a different way, into 5–11 and 12–14’ (ILO, 2002b, p.29). 46 ‘In line with the international definition of employment, one hour of work during the reference week is sufficient for classifying a person as at work in economic activity during that week. Included in the classification are also individuals with a job but temporarily absent from work due to illness, vacation and other similar absences’ (ILO, 2002b, p.30).

Part II

The Country Studies

6

Subcontracted homework in India A case study of three sectors1 Ratna M. Sudarshan, Shanta Venkataraman and Laveesh Bhandari

The creation of new jobs in export oriented production in developing countries as a consequence of the processes of globalisation and liberalisation is fairly well documented, as is the fact that much of this work has gone to women. It is known that parts of the production process may be outsourced and that the linkages between different levels of production may not always be clearly visible. Very little is known about the lower end of the production chain, which includes homeworkers – who these workers are, and what the terms and conditions of their work are. This chapter attempts to fill this gap, in relation to a few selected sectors and locations. The primary motivation behind this study is to identify appropriate policy interventions. It must be clearly stated that it is difficult to apply standard techniques of sampling to this group. The reasons include the absence of a sampling frame. This chapter is on a study that drew a purposive sample from selected clusters of homeworkers in three sectors and four locations (see Figure 6.1). These are: ● ●

●

agarbathi or incense stick making in Bangalore city in Karnataka; bidi rolling, that is, rolling tobacco in tendu leaf, in Indore District in Madhya Pradesh and North Arcot and Ambedkar Districts in Tamil Nadu; zardosi or gold thread embroidery in Lucknow city in Uttar Pradesh.

Around 600 households were canvassed, of which 25 per cent was a control group (CG) – (for more details on the sample design and the core questionnaire see Section 6.3 and Chapter 2). Focus group discussions (FGDs) and case studies were also conducted. The data presented here should therefore be interpreted with a degree of caution. Although the estimates cannot be generalised to all homeworkers, data have been collected as carefully as possible and with a view to identifying the situation of homeworkers and production relations in the sectors selected. There has been a resurgence of interest in the informal sector in the last decade, partly stimulated by the increased informalisation of employment. While the ‘informal sector’ is an enterprise-concept, ‘informal employment’ is a worker-concept (see Notes 1, 2, 5, 6 and Section 1.1 in Chapter 1). Homeworkers are an example of informal employment; their ultimate employers could form part of either the formal

174

Sudarshan et al.

Lucknow

Indore

Bangalore

North Arcot

Figure 6.1 Areas covered for homework study in India.

or the informal sector. In this study we have examined the market structure to show the linkages between homeworkers and other agents in the production and marketing chain, and a value chain has been constructed to show the contribution of different agents. However, it has not been possible to establish clearly the extent to which the homeworkers contribution is placed in the formal or informal sector. The unravelling of such formal–informal linkages will require further study. Section 6.1 provides the context for this study in a brief description of the size and nature of the informal sector and of informal employment in India. Section 6.2 briefly presents the research design in India. Section 6.3 analyses the market structure and locates homework within this, along with the mechanisms through which this kind of subcontracting is able to remain stable and viable. Section 6.4 discusses the possible role of earnings from homework in alleviating poverty, and the alternative sources of income available to homeworkers. Section 6.5 attempts to characterise the typical homeworker in these sectors. Section 6.6 examines the role of children in homework. Finally Section 6.7 summarises the policy interventions and areas of concern that emerge from the study.

Subcontracted homework in India 175

6.1

The informal sector and informal employment in India

The informal sector in India, as elsewhere, may be increasing due to liberalisation policies including downsizing of government and privatisation of public sector enterprises. The terms, ‘formal sector’ and ‘informal sector’, have not been used in measurement, but data in India is available from official sources on the ‘organised’ and ‘unorganised’ sectors separately. The unorganised segment of the economy broadly covers all of the agricultural sector (except plantation crops and operations of the government irrigation system), minor minerals, unregistered manufacturing units and all units of non-manufacturing activities except those in the public, private corporate and cooperative sector. Data on work participation rates from the National Sample Survey (NSS), and on population (estimated or projected) from the Census, gives fairly reliable estimates of the total labour force. Employment in the organised sector is available from the Directorate General of Employment and Training (DGET), Ministry of Labour, on an annual basis. Employment in the unorganised sector is generally computed as the difference between these two. So described, the unorganised sector accounts for 93 per cent of total employment. It contributes around 60 per cent to total net value added, and 60 per cent of total household savings originates from the unorganised sector (or around 50 per cent of all savings) (Pradhan et al., 1999). Rough estimates suggest that around 30–40 per cent of exports is contributed by the unorganised sector (Ghatate, 1999). More recently it has been accepted that informal employment is distinct from employment in the unorganised (or informal) sector, since in addition to informal workers in the informal sector, there are informal workers in the formal sector (e.g. due to outsourcing). Much less data is available on the levels of informal employment in India. Using official data on the organised and unorganised sectors, it can be seen that there has been relative constancy in the share of the unorganised sector in total employment between 1972/73 and 1991 (at around 92.1 per cent of the total), with a slight increase since 1991. The growth rate of employment in the organised sector was 2.76 per cent per annum over the period 1972/73–1983, but fell to 1.4 per cent per annum over the period 1983–91, and was only 0.5 per cent per annum between 1991 and 1999/2000. The phenomenon of ‘jobless growth’ appears to be a good description of the organised sector in the 1990s. In the unorganised sector, the corresponding rates of growth have also fallen, from 2.79 per cent per annum in the 1970s, to 1.7 per cent in the 1980s. There has been a slight increase in the 1990s, to 1.9 per cent. Average productivity (in value) in the organised sector is much higher, being over eight times that in the unorganised sector (Table 6.1). An important feature of the informal sector in India is the predominant presence of women. Data shows that 91 per cent of women in the non-agricultural labour force are in the informal sector as opposed to 70 per cent in the case of men (UNIFEM, 2000). Women in informal employment are further concentrated at the low end of the spectrum, in low paying and insecure jobs. According to NSS and other data, women in the unorganised sector could represent ‘anywhere

176

Sudarshan et al.

Table 6.1 India: contribution of the organised and unorganised sectors to net domestic product (Rs 10 million) and employment

Organised sector Output Employment Avg. productivity Unorganised sector Output Employment Avg. productivity Total NDP

1993–94

1997–98 (at 1993–94 prices)*

Average growth rate per annum (%)

257,455 27.4 9396

356,987

9.7 0.7 9

39.5

440,537 361.6 1218 697,992

545,709

5.97 0.5 5 7.3

60.5

12659

1479 902,697

Percentage of NDP 1997–98

100

Source: Government of India, Central Statistical Organisation (1993–94 and 1997–98), New Delhi, India. Note * GDP deflator 1.366927.

between 20–25 per cent of total employment in urban areas and between 30–40 per cent of total employment in rural areas’.

6.2

Research method

Definitions used In this study, homeworkers have been defined to include: (a) subcontracted workers, working from their house (b) self-employed workers, working from their house. This decision was taken mainly because the categories did not appear to be mutually exclusive, and no attempt was made to distinguish between ‘subcontracted workers’ and ‘occasionally subcontracted workers’. The focus of the study is on subcontracted manufacturing, and to that extent it is subcontracted workers that are the object of study. Data has also been collected on unpaid family labour, since in most cases the unit of work was found to be the family or household. The large majority of the workers surveyed were found to be subcontracted workers; the same workers might be self-employed for part of their time. Survey design and data collection A total of 603 households spread more or less equally between the three sectors were surveyed. The bidi sector sample is further divided between the states of Tamil Nadu and Madhya Pradesh equally. Having identified the survey area,

Subcontracted homework in India 177 homeworker clusters were identified with the help of local agencies and experts.2 Having chosen the cluster to be surveyed, households were selected to meet the sample requirement. In each of the four areas selected for study, an attempt was made to include both rural and urban households. In Lucknow (zardosi) and Bangalore (agarbathi) the rural areas are more accurately described as peri-urban. In Indore (bidi), rural areas were not surveyed, and for this reason the Tamil Nadu sample for bidi workers is largely rural (for location, see map, and for details of the sample households, see Table 6.2). The total of 200 sample households in each sector were divided as follows: (i) households with at least one family member working mainly in home sector. This could be further divided between: (a) household units (both spouses could be engaged in homework) for bidi
zardosi (no men in agarbathi) (referred to as ‘family units’) (b) households in which only women were working for homework. However, in either case child/children involved in homework were included (referred to as ‘non-family units’). (ii) households with no family member working in home sector, defined as ‘control group’ households. The CG thus consists of households in the same neighbourhood and geographic location as homeworker households, and could be expected therefore to have roughly similar income levels (referred to as CG or non-homework). The two categories in the total sample were split in approximately a 3:1 ratio respectively. The household survey was designed on the basis of the core questionnaire (see Annex 2.1 in Chapter 2). Compared to the core questionnaire, the household survey is slightly different in terms of level of detailed information and structure. In order to adapt the questionnaire to the national context and to gain a deeper understanding, more detailed questions were added in some sections. For example, in the section about social identification, questions regarding caste and religion were added.3 In the same way, the questions regarding expenditure requested detailed information about expenditure on food items, non-food items4 and frequency of purchase. In terms of the structure, our questionnaire included two additional sections on ‘time allocation and organisation’ and ‘perceptions’. The section on time allocation consisted of detailed questions regarding peak and lean seasons of work. In the section about women’s organisations, the questionnaire went beyond questions regarding membership of labour or trade associations, and current social benefits. In addition, households were asked to determine the type of benefits they would like to receive from subcontractors.5 In order to determine the economic level of the household and work conditions and potential health hazards, interviewers were required to observe carefully and record specific aspects6 as well as to ask households about ownership of specific consumer durable goods.7 The section on perceptions consisted of three open questions directed to households about

603

201 452

148

75

151 76

153

151

53

25

50 25

48

227

58

15

91 76

78

60

12

—

25 25

23

CG

225

90

60

60 —

75

hw

Rural

Notes Dash indicates that certain information was not collected in the survey or data were not comparable. FGDs  focus group discussions. CS  case studies. hw  homeworkers. CG  control group or non-household. * Considering separately the one for women and the one for children.

Source: India Country Study Survey (2001).

Sub-total

Zardosi

100

201 101

MP
TN Indore District, Madhya Pradesh North Arcot Ambedkar District, Tamil Nadu Lucknow District, Uttar Pradesh

Bidi (TN)

201

Bangalore District, Karnataka

Incense stick making (agarbathi) Bidi (MP
TN) Bidi (MP)

hw

CG

Total

hw

Urban

Households surveyed

Location

Sector

91

41

25

25 —

25

CG

7

2

—

2 —

3

Total

—

—

—

— —

—

Urban

Number of FGDs*

—

—

—

— —

—

Rural

Table 6.2 Surveys on homeworkers households in India: sectors, location, number of FGDs, case studies and households surveyed

9

4

—

3 —

2

Number of CS

Subcontracted homework in India 179 facilities that should be provided by the government, possibilities for technological improvements in each activity, and suggestions on how to prevent negative effects on children’s education.8 Finally, the household questionnaire was supplemented by a short form to compile a profile of the villages surveyed in order to form an idea of the available infrastructure. Focus group discussions, case studies and in-depth interviews A total of 7 FGDs were held with women and children working in the home sector under study. Two FGDs each were held in the zardosi sector and bidi sector, and three in the agarbathi sector. The number of participants varied from 10 to 25 and were mostly women and girls9 who were not covered in the primary survey. The discussions focused on general problems faced by homeworkers, ranking of interventions and perceptions regarding the work of children. The main objective in conducting these discussions was to try to get feedback on proposed government plans and programmes. Since the household survey also asked such questions, the FGDs would also provide some validation of the survey. A total of nine case studies10 were carried out in the sectors studied through in-depth interviews so as to capture sensitive issues11 that would help in developing policy recommendations. Case studies included male and female workers, children, and subcontractors located in rural and urban areas.12 One question which the case studies sought to answer is the relation between a homeworker and the subcontractor. In order to identify the macro-micro linkages within each sector, and in particular, to understand the market structure, price mark-ups at different stages of production (i.e. starting with homeworkers and ending with final consumer) and the potential impact of macro and trade policy on homeworkers, a small sample of contractors/subcontractors/employers/entrepreneurs was interviewed in detail.

6.3

Subcontracted manufacturing

It is estimated, using data from the NSS 55th Round Survey, that there may be approximately 28 million homeworkers in the non-agricultural labour force, amounting to around 35 per cent of the total number of informal workers in non-agricultural enterprises (79.7 million) (NSSO, 2000). Whilst the overall informal sector is evenly divided among urban and rural areas, 43.6 per cent in the former and 56.4 per cent in the latter, homework is mostly located in rural areas (70 per cent). The sectoral composition of homework consists of 59 per cent in manufacturing, followed by 23.5 per cent in the trade and repair services sector. Other sectors include storage services, financial intermediation and construction (Table 6.3). Although as a proportion of the total workforce this amounts to just 7 per cent, the significance of homework is better understood in relation to individual sectors and the nature of subcontracting arrangements in a globalising economy.

180

Sudarshan et al.

Table 6.3 Per cent of homework (hw) in non-agricultural informal enterprises and sectoral composition of hw 1999–2000

Total informal sector Other (Non-hw sector) hw sector Sub-sectors (% of hw) Manufacturing Construction Trading and repair services Hotels and restaurants Transport, storage and communications Financial intermediation Real estate, renting and business activities Education Health and social work Other community, social and personal services (excluding domestic services)

Location Urban

Rural

100.0 64.1 35.9 100.0 59.0 0.1 23.5 2.0 4.1

43.6 48.0 30.0 100.0 25.0 100.0 34.0 39.0 39.0

56.4 52.0 70.0 100.0 75.0 0.0 66.0 61.0 61.0

0.4 1.0

75.0 70.0

25.0 30.0

1.8 1.6 6.5

62.0 29.0 28.0

38.0 71.0 72.0

100.0

100.0

100.0

Sources: National Sample Survey Organisation (2000). Non-agricultural enterprises in the informal sector in India 1999–2000. Key Results, NSS 55th Round (July 1999–June 2000).

The 55th Round NSS data reveals some interesting facts about the nature of subcontracting in the informal sector. Over 72 per cent of rural and 68 per cent of urban, non-agricultural informal enterprises in the manufacturing sector operate without a contract. 76 per cent in the rural and 75.8 per cent in the urban areas procure their own equipment. However, in the case of 90.8 per cent rural and 82.4 per cent urban enterprises, the raw material is supplied by the master enterprise/contractor; and in the case of 92.5 per cent rural, and 91.1 per cent urban enterprises, the design is specified by the master enterprise/contractor. According to a recent UNIFEM study, subcontracting has been on the increase since 1978: ‘Subcontracting was not a significant activity in 1970. From 1978 onwards, it became widespread among large factories with a share of 21 per cent.’ A positive association is seen between subcontracting intensity and labour intensity. A correlation is also observed between subcontracting and the growth of key export oriented sectors (UNIFEM, 2000). It must be noted that in the Indian context subcontracting has been the norm traditionally. Homework has been a feature of traditional economic activity, especially associated with artisan employment. The changing economic environment affects each productive activity in different ways, at different speeds, and in differing degrees. The economic environment changes due to many factors. Broadly, changes in technology and

Subcontracted homework in India 181 changes in policy regime would also be expected to affect the homeworkers. However, it is not clear whether recent changes in both technology and economic policy immediately affect the conditions that homeworkers operate under. The activities under consideration are essentially labour-intensive manual activities conducted by those who are extremely poor; even if new technologies were available they would not be expected to immediately percolate down to them. Policy changes, however, could have a more immediate impact as in the example of zardosi, involving the embroidery of gold thread on garments. Trade liberalisation in the 1990s, specifically of gold imports, would be expected to have an impact on the materials. Conversely, import liberalisation could also affect homeworkers negatively, if the imported commodities compete with their output. However, in the case of the three commodities considered, no negative impact has occurred. This is because all three commodities – zardosi, agarbathis, and bidis are as yet not imported into the country.13 Overall, few policies on labour, contracts and social security have been specifically implemented for the homework sector, and given current trends, not much can be expected in the near future. Bidi is the only sector in which specific labour policy has been established. The creation of labour unions and the implementation of a minimum wage have impacted positively on the conditions of workers in this sector. Nonetheless, changes in policies can also have an indirect impact on homework sectors. A monetary policy, such as a devaluation of the Rupee (Rs), has a positive impact by making homework products more attractive in the export markets. This is true for the agarbathi and zardosi sectors since the bidi market is essentially domestic. In the same way, policies aimed at achieving higher economic growth benefit homeworkers by increasing the size of the market that they service (Table 6.4). In general, a number of factors contribute to extensive subcontracting in India, including: ●

●

● ● ●

●

the importance of operating close to homes where the workers can also undertake other activities; the reluctance of contractors to hire regular employees and as a result meet many labour-related regulations; saving on costs of space required for the task assigned to the worker; no need for expensive capital equipment that has to be shared between workers; strong social networks that prevent individual workers from reneging on contracts, these also facilitate lower transactions costs; repeated transactions result in less need for formal contracts. Conditions are such that self-enforcing informal contracts are easy to create and administer. The costs of transactions are therefore also lower.

The sectors under study The three sectors studied here form part of the manufacturing sector in India, which in total contributes around 15 per cent to the Net Domestic Product. Of this

182

Sudarshan et al.

Table 6.4 Impact of macroeconomic policy-disaggregated by sector Policy

Zardosi

Agarbathi

Bidi

Imports

Positive due to gold imports, lower input costs

None – tobacco imports not likely to be allowed

Exports Exchange rate devaluation

None Positive

Not significant, could be marginally positive later when perfume imports become prevalent None Positive

Labour policy

No significant changes in labour policy No changes Not occurred No changes in administrative set up that makes enforcement possible Significantly positive – growth in market sizes attributable to policy reform

No significant changes in labour policy No changes Not occurred No changes in administrative set up that makes enforcement possible Significantly positive – growth in market sizes attributable to policy reform

Contract law Legal reform Minimum wages

Market size

Social security/ None safety net

None

None Positive, but exports are an insignificant share Positive – due to unionization No changes Not occurred Minimum wages have been fixed, coverage of poor hw Insignificant – tobacco consumption is reputed not to have grown much None except through organization and unionization of workers

total, 5 per cent comes from the unorganised sector and 10 per cent from the organised sector (National Accounts Statistics). All three of the selected sectors are ‘traditional’ in the skills required, with a large and established local and national market, and a growing export market (in the case of bidi, given the growing international and national opinion against tobacco use, future market demand is now an uncertain factor). Data is available on the exports of bidi and agarbathi, but appropriate data for zardosi are difficult to find partly because of the unorganised nature of the activity and partly because of the very large number and diverse kinds of final products in which zardosi embroidery is used. Home production in all three sectors is for both the domestic market and for export. While zardosi work has been traditionally undertaken at home, this is truer in urban areas. Extension to rural households probably started around 1980,

Subcontracted homework in India 183 and was a consequence of expanding markets. The more highly skilled workers are still to be found in urban areas (Jalees, 1989). Zardosi products, like the products of other crafts sectors, have seen an increase in demand. The main markets for these products are in the United States, Dubai/Middle East and Europe. There is also some evidence of artisans migrating, usually for a period of a few years, to the Middle East, and returning to set up small production centres largely catering to export demand. No direct comparison of the costs of production between home and other enterprises has been attempted, but it should be noted that both types co-exist. In the zardosi sector, the small centres or ‘karkhanas’ that have emerged in the last few years are unregistered and unorganised enterprises. They generally employ only males, usually young boys, in a kind of apprentice system. It is also to be noted that awards to zardosi master craftsmen have been given, so far, only to men, and the sector is popularly perceived as a male-intensive one. The advantages of being a zardosi worker – regular work in the centres, national recognition, training – are thus appropriated by men. Women in this sector are confined to homework, segregated both by the nature of the production relation and the cultural exclusion imposed on Muslim women. Expanding market demand requires more workers, and the use of women homeworkers has probably been a very cost effective way of getting skilled workers without the need for outlays in training or infrastructure. Agarbathi exports are also increasing, with important markets being in the United States, United Arab Emirates and Saudi Arabia. Over the period 1990–91 and 1998–99, the value of exports increased by 281 per cent. This work has spread from its point of origin in Tamil Nadu to many other states, and as in the case of zardosi, homework has been stimulated by increasing demand over the last two to three decades. However, it is difficult to date these developments more precisely. The first formal production of bidis is said to have been around 1902, and until the 1940s the production was mainly in factories, and mostly men were engaged in this work. After 1976, the shift to homework was rapid, and the composition of bidi workers changed from predominantly male to predominantly female. The emergence of homework in the bidi sector is closely related to the passing of the bidi and Cigar Workers (Conditions of Employment) Act 1966, and The Bidi Workers Welfare Fund Act 1976, as also the increasing unionisation of bidi workers. It is more difficult to organise homeworkers, as well as to implement the various provisions of the law (see Kumar, 2001). The Bidi Workers Welfare Fund Act 1976 is administered departmentally through a tax levied on the production or export of specified goods. In case of bidi, it is 50 paise/1000 bidis. The fund is meant basically to provide medical care, education of children, housing and water supply and recreational facilities. Given the bans against tobacco, the entire industry may be gradually phased out. Exports have gone down in the 1990s, the total decline between 1990–91 and 1998–99 being around 55 per cent. At present, exports are mainly to the United Arab Emirates, followed by the United States and Singapore (Table 6.5).

184

Sudarshan et al.

Table 6.5 Market for finished products: domestic markets and export markets for the three sectors (a) Market for finished products Sector Foreign market

Domestic market Local market

Agarbathi Bidi Zardosi

Yes Yes Yes

Not seasonal Not seasonal Seasonal

Year when started

Year as hw 1970s (as reported in Bangalore) 1976 c.1980s (rural areas around Lucknow)

Yes Yes (shrinking) Yes

(b) Domestic markets Sector Traditional yes/no Agarbathi

Yes

n.a. (originated in Tamil Nadu)

Bidi Zardosi

Yes Yes

1902 c.1400

(c) Export markets for the three sectors Sector Value of exports Value of exports April 1990–March April 1998– 1991 at 1998–99 March 1999 prices (Rs lakhs) (Rs lakhs)

Export trends: average per annum increase 1991–99

Agarbathi*

3,200

12,200

35%

Bidi

7,930

3,500

6.9%

Zardosi

n.a.

n.a.

n.a.

Export markets

USA (13%) UAE (7.6%) Saudi Arabia (4.8%) UAE (65%) USA (7.5%) Singapore (3.7%) USA, Dubai/Middle East Europe

Sources: Export Markets: Agarbathi, Bidi: Monthly Statistics of Foreign Trade of India (Annual no. 1998–99) part I, II. Directorate General of Commercial Intelligence and Statistics, Ministry of Commerce. Zardosi: from case studies/interviews with contractors. Note * Agarbathi and other odoriferous preparations which operate by burning.

The labour process and the value chain Workers operate in a subcontracting environment in agarbathi manufacture, zardosi work and bidi sectors. Basically, the manufacturers/distributors/contractors provide the basic raw material to the workers who undertake the manufacture of end or intermediate product. They provide their output to the subcontractor/agent who in turn supplies the contractor. The chain thus continues to the manufacturer to the distributor, and so on.

Subcontracted homework in India 185 Contracts in the three sectors are verbal, presumably enforceable because of strong social networks and limited mobility. While there are no formal barriers to entry, such networks clearly play a role in determining who enters the sector. The environment of production and marketing is such that children, men and women may all be engaged in homework. The three activities in question tend to have many differences in the kind of tasks involved. In agarbathi most of the work related to bathi rolling is done at home by women and girls. These women are in contact with agents whose shops are in the vicinity of their homes. The agents provide the raw materials to the women rollers and then collect the finished products. Women are paid Rs 10 for a bundle of 1,000 agarbathis. These agarbathis rolled at home are called raw bathis and the agent is known as the raw bathi manufacturer. The raw bathis are passed on to the factory where perfuming and complete packaging for the retail outlets take place. Next in the line is the distributor (or super stockist, as he is sometimes called) who distributes the products to the wholesale and/or the semi wholesale market. The wholesale/semi wholesale dealer passes on the products to the retailers. Bidi rolling involves a series of activities, including: cutting tendu leaves to size; soaking the leaves in water for a few hours; rolling the bidis by putting the requisite amount of tobacco within the tendu leaves; tying the edges of the bidis with yarn; tying up rolled bidis into bundles; placing on a wooden tray and drying it in the sun; checking size and quantity of tobacco; curing by putting trays of bidis into a furnace; labelling and packing. The first set of activities, or the rolling work, is largely done by homeworkers who provide the unfinished product. The manufacturer roasts and rolls the bidi at his premises in usually small enterprises. This task is often done by the manufacturer’s relatives as well as some hired employees and both men and women are involved. Finally, the zardosi work involves embroidering gold thread on garments such as saris, kurtas (long shirts), dupattas (scarves) and so on. Unlike bidis and incense sticks, the end product is not homogenous and the quality of embroidery is an important factor in the value of the end product. The amount of gold thread used on the garment is another factor. The workers have some degree of flexibility in deciding embroidery designs but do not have complete independence in the matter. As can be seen, the nature of tasks varies from lower skilled as in agarbathi making to higher skilled as in zardosi. In the same way, the kind of output varies from being of extremely low value in bidis to high valued as in the zardosi garments. Finally, the type of product ranges from being standardised bidis to highly heterogeneous zardosi garments. Markets also vary by sector, from being predominantly domestic, as for bidi production, to an international market for incense sticks and zardosi. However there are similarities in the way their production is administered – subcontracting to homeworkers. As is true for most consumer products whether manufactured in large or extremely small-scale factories, or by homeworkers, are finalised in the local

186

Sudarshan et al.

system of production (often a cluster). The bulk of the value is added at stages later than that of production. This is true for all three commodities considered here. It is difficult to calculate the increase to value when exact data on the costs involved is not available. The surveys conducted here could not gather data on values and costs from a large enough cross-section of homeworkers, contractors, manufacturers, distributors and exporters, because of their reluctance in reporting these figures. This reluctance is natural, as much of the activity takes place in the unorganised sector, where formal information and record keeping procedures are non-existent. Over and above this there is the natural disinclination to part with information to ‘outsiders’. However, from various sources it is possible to put together how the price value increases across the different stages in the process by calculating the various inputs (production) to the stage where the products are eventually purchased by the consumer. To estimate the increase in value, detailed data on the costs of manufacturing, hired labour, rentals, profit, and so forth are required. Such data are not available. However, there are some data on the costs of the major inputs as well as who bears these costs. The estimation of the value chain also runs into problems because the chain itself is not the same for different types of homework. Often there are no subcontractors, while in other cases manufacturers might also be the distributors. Lastly, different sub-classes of commodities might have different values due to varying size and quality. In other words, different product classes within the products being studied – large, medium and small bidis, different types of incense sticks, different amounts and quality of zardosi – lead to a situation where it is better to study the price value increases in percentage terms rather than in rupee value. The same exercise was conducted for the same product classes, but the results are reported in value terms. That is, the value chain reports the values at different stages in the production/distribution process for Rs 100 at the retail level (Figure 6.2). Though the resultant figures are not derived from any robust sampling, or in-depth data collection, they do provide some indication of the value chain. Note that the retailer obtains extremely high shares in zardosi. This is similar to the costs for most garments and clothing products with a high value – retail margins are extremely high in these industries. Bidi and agarbathi manufacturers obtain high shares. This is also not surprising given the importance of brands in these industries. In sum, though we cannot claim a high degree of robustness of the methodology used for calculation of these figures, they do not appear to be far off the mark. The relatively high shares for homeworkers for zardosi work is also not very surprising given the importance of quality. However, the same cannot be said for bidi. The fact that bidi workers considered here are organised appears to play a very important role in their relative shares. There is a third factor that may be at work, although this can only be speculated upon, and that is the involvement of men in both zardosi and bidi work, which may help to raise the returns; agarbathi work is exclusively done by women (Table 6.6).

100 90 Retailer

80

Wholesaler 70

Distributor Manufacturer

Per cent

60

Contractor 50

Subcontractor HW

40

Inputs

30 20 10 0 Agarbathi

Bidi

Zardosi

Figure 6.2 Value chain for Rs 100 worth to the consumer. Source: India Country Study Survey (2001).

Table 6.6 The value chain for commodity worth Rs 100 to consumers Commodity/item

Value in Rs Zardosi

Inputs Homework Subcontractor Contractor Manufacturer Distributor Wholesaler Retailer Total Average no of intermediaries

4.0** 15.3 1.9 1.6 17.1 60.0 100.0 4

Bidi 20.0 17.1 0.9 0.9 41.9 *** 19.3 *** 100.0 5

Agarbathi 1.9 2.3 2.5 59.8* 14.6 9.0 10.0 100.0 5

Source: India Country Study Survey (2001). Notes * For Agarbathi the manufacturers value includes the cost of perfume. This cost can be high depending upon the type of perfume used. However, data was not available for the same. ** This is for low value Zardosi work for scarves. Input values are likely to be significantly higher for silks and sarees. The values for workers are however expected to remain broadly similar. *** The margin of the distribution chain for Bidi (distributor, retailer and wholesaler) is all included under the wholesaler. The relevant break-down is not available.

188

Sudarshan et al.

6.4

Income and poverty in homework

Homework is one source of household income. In agarbathi the contribution is 22 per cent in rural and 26 per cent in urban households. In bidi (Madhya Pradesh) the contribution is 30 per cent in urban areas, and in bidi (Tamil Nadu) it is 46 per cent in rural and 71 per cent in urban areas. Zardosi contributes 47 per cent in rural and 54 per cent in urban households. Overall, homework contributes 35 per cent to household income, taking rural and urban areas together. Information on household income from different sources was collected for one year preceding the survey (Table 6.7). The various sources of income have been broadly categorised into the following six groups, namely: ● ● ● ● ● ●

Homework activities Agriculture wage Non-agriculture wage (other than homework) Self-employed (including petty trade, business and all own account enterprises) Salaried employment (including fixed salary on a contractual basis) All others (including remittances, pension, interest/dividends and others. Income from cultivation is merged with this category since few cases were reported).

This survey covered two sets of households. First, those households that have at least one of the spouses engaged in work at home for the sector under study and two, households that are not engaged in any kind of homework. The former is referred to as Homework (hw) households and the latter as Non-homework or CG households.14 In both homework and Non-homework households the proportion of household income from non-agriculture wages is high, accounting for almost 54 per cent in rural and 40 per cent in urban areas. Its relative importance is not the same across the different sectors. In the agarbathi sector two-thirds of the income is accounted for by non-agriculture wage earnings in homework households and for over 90 per cent in Non-homework households. The contribution is the lowest among urban bidi households in Tamil Nadu, at around 5 per cent. Self-employment plays a small role among the homework households, although it is more significant in urban areas than rural, and most so for Tamil Nadu urban bidi households, where it contributes almost 24 per cent of household income. Overall, the contribution is under 10 per cent. But it is the main source of income for the zardosi CG (89 per cent), and over 57 per cent for the Tamil Nadu bidi CG (mainly weavers). It is relatively unimportant for the agarbathi CG and the Madhya Pradesh bidi CG of households. Salaried income is of significance only in the bidi Madhya Pradesh urban group (27.6 per cent). It is also significant for the bidi Madhya Pradesh CG (55 per cent), and to a lesser degree the bidi Tamil Nadu CG (29 per cent). In North Arcot Ambedkar Nagar/Vellore (Tamil Nadu), one of the villages selected was well developed, with a polytechnic institute in the vicinity, which opened up job

Subcontracted homework in India 189 Table 6.7 Composition of income of homeworker (hw) households (%) Sector

hw

Rural hw households Agarbathi 22.3 Bidi (MP) — Bidi (TN) 45.9 Bidi (MP
TN) 45.9 Zardosi 46.8 All sectors 35.0 Urban hw households Agarbathi 26.4 Bidi (MP) 30.9 Bidi (TN) 70.8 Bidi (MP
TN) 35.1 Zardosi 54.3 All sectors 35.6 Non-hw or CG households Agarbathi — Bidi (MP) — Bidi (TN) — Bidi (MP
TN) — Zardosi — All sectors —

Nonagri-wage

Self-emp.

Salary

Others

All

70.7 — 40.3 40.3 36.4 53.6

3.8 — 9.3 9.3 7.3 6.1

3.2 — 1.3 1.3 6.5 3.7

— — 3.2 3.2 3.0 1.7

100.0 — 100.0 100.0 100.0 100.0

62.0 28.3 5.6 25.9 27.1 39.9

5.4 12.8 23.6 13.9 10.4 9.9

6.2 27.6 — 24.7 8.3 14.3

— 0.5 — 0.5 — 0.2

100.0 100.0 100.0 100.0 100.0 100.0

93.9 36.2 13.8 21.1 3.3 43.3

2.9 8.4 57.2 41.2 89.3 36.6

3.0 55.4 29.0 37.6 5.8 19.7

0.2 — — — 1.7 0.4

100.0 100.0 100.0 100.0 100.0 100.0

Source: India Country Study Survey (2001). Notes MP  Madhya Pradesh. TN  Tamil Nadu.

opportunities for drivers, gardeners, watchmen and so on. These salaried workers live in the village. The reason for the high level of salary earnings in Indore is that a majority of the men in the two clusters sampled were receiving a regular income as compensation by virtue of their displacement from the textile mills that had recently closed down in this area. For the households surveyed, homework thus makes a significant contribution to household income, although most households do have other sources of income. Indeed, for most sectors it is the first source of income. Household income data collected from the field have been used to determine the levels of impoverishment in different sectors. Official data on poverty estimates is available from the Planning Commission, Government of India, at the state level for rural and urban areas for the year 1993–94 (Government of India, 1993). Using these data, the estimates for the study areas – agarbathi in Bangalore, bidi in Indore, bidi in North Arcot Ambedkar Nagar and zardosi in Lucknow – can be compared with the estimates for the States of Karnataka, Madhya Pradesh, Tamil Nadu and Uttar Pradesh respectively. The levels of deprivation seem to be the highest in the zardosi sector in rural as well as urban areas with nearly 90 per cent of the sampled population below

190

Sudarshan et al.

Table 6.8 Estimated proportion (%) of population below poverty line Sector

Agarbathi Bidi (MP) Bidi (TN) Zardosi

Survey estimates (2000)

Planning Commission (GOI) (1993–94)

hw household

Non-hw household

Rural

Urban

State

Rural

Urban

Rural

Urban

16 — 52 90

49 57 87 88

9 — 12 83

16 61 — 81

29.8 — 32.5 42.3

40.2 48.7 4.4 36.0

Karnataka MP TN UP

Sources: India Country Study Survey (2001); Government of India (1994).

the poverty line. This sector is clearly in need of urgent attention from policymakers, since by comparison the population below the poverty line as per official estimates stands at 42.3 per cent in the rural and 36.1 per cent in urban areas in the state of Uttar Pradesh. While only tentative conclusions can be drawn, it can be hypothesised that homework has spread among poor households with some measure of skill or appropriate networks, and may have contributed to reducing the intensity of poverty among these households, although the evidence suggests that many still remain below the poverty line (Table 6.8). This income is earned with long hours at work; the average working hours per day of women in homework in the three sectors are 7.6 in rural and 8.1 in urban areas. In all three sectors, the average working day is longer in urban areas. The average working day for bidi workers is the highest, but the caveat is that work is less often available than is desirable, so workers probably try to do as much as possible when it is available. Seasonality affects zardosi workers, stemming largely from the nature of the domestic market. Zardosi products are demanded at times of ritual or celebration so that there is a peak in demand around festivals.15 The rest of the year, demand is low and erratic, although there is no seasonality in export demand. Clearly, homework is not ‘part time’, and women work on average almost 8 hours a day. Despite this kind of time input, there is no regulation or payment of benefits to workers. In so far as regularity of payments is concerned, the survey showed that while 98 per cent of the agarbathi workers and 89 per cent of bidi workers reported payment on time, only 63 per cent of zardosi workers did. It is possible that delayed payments are used as a form of control by the subcontractor to keep workers tied; the skill of zardosi workers, particularly in more high value items, is crucial to quality. Indebtedness and poverty are closely linked, and with homeworkers there seems to be some degree of debt bondage to contractors. This issue has been examined by looking at the sources of loans taken by homework both for initial investment and to meet recurring expenditures related to work. Agarbathi workers are largely reliant upon their own resources, with 91 per cent of initial investment and 95 per cent of recurring expenditure coming out of their own resources. In the case of bidi workers, self-financing plays a much smaller role. Of initial investment

Subcontracted homework in India 191 27 per cent is through ‘private loan’ and 23 per cent of recurring expenditure is from ‘employer/contractor’. What this means, in effect, is a fairly high degree of debt bondage, something that was confirmed by case studies. Zardosi workers are forced to raise initial investment from their own resources (81 per cent) but also 11.5 per cent from ‘others’ which includes moneylenders and other informal and formal sources. In the case of recurring expenditure, 95 per cent of workers state that this is self-financed. Thus, even for work-related, productive expenditures, none of these groups of homeworkers have access to credit from other than informal sources. Some special concerns regarding earnings from homework include the fact of reduction in bidi rolling work over time. Households with other sources of income may be able to compensate to some extent for such loss of income but those households which are exclusively dependent on bidi rolling – which constitutes perhaps the largest in terms of employment for home-based workers – will need to be helped to find alternative sources of income. Suitable programmes for alternate skill training and employment are needed. Intra-sector differences in bidi rolling were found to be substantive, including the levels of poverty, worker–contractor arrangements, and the extent to which workers were able to access benefits through the bidi Workers Welfare Fund. Differences were especially evident between the more remote areas in Tamil Nadu, and those closer to cities. For example, the rejection level in rural areas was reported as being as high as 20 per cent, but closer to 5 per cent in peri-urban areas. There is an additional layer of intermediaries in the more remote areas, and subcontractors have a monopoly over their defined areas. Zardosi workers felt that the level of insecurity they faced could be considerably reduced if, for example, they could be organised along the lines of the Self-employed Women’s Association (SEWA) in Lucknow, which has been able to help these workers to market chiken-kari work through their outlets. This marketing support has worked as an incentive for workers to maintain a certain minimum quality.

6.5

The homeworker

The typical homeworker is a married woman with children. Women seem to prefer this activity given the degree of control and flexible timings and the possibility to supplement the household’s income. The survey was designed to focus on women workers, so that it is not surprising to find that a high proportion of women in the households that have been surveyed, are working. However, the profile of workers within these households as well as information on the average hours worked confirms that the activities under study are heavily female intensive. In comparison with the CG households, there is a clear feminisation of the workforce in the home sector, in the sense that a much higher percentage of women is found to be working here than in the CG. Over four-fifths (87 per cent) of the persons involved in homework are females who constitute nearly 95 per cent of the total working women in these households (Tables 6.9a and b). In contrast, only 23 per cent of women are found to be working in CG households.16 Among the homework households surveyed,

176 94 97 191 257 624

Agarbathi Bidi (MP) Bidi (TN) Bidi (MP
TN) Zardosi All

81.9 80.3 85.1 82.7 72.0 77.7

% of women working of total women

91.5 95.7 91.7 93.7 98.0 94.9

% of women working in hw of total working women 12.6 4.4 7.4 5.9 22.2 14.2

% of children working of total children

34.6 25.0 85.8 63.6 75.5 62.2

% of children working in hw of total children 77.7 100.0 50.0 57.1 80.0 77.8

Girls working in hw as % of total children 49.3 54.1 53.8 54.0 55.1 53.1

All females working in hw as % of total female pop.

97.1 94.7 68.6 79.6 86.0 86.6

All females in hw as % of all persons working in hw

Note It is important to note, as in this case, that the data in this chapter often refers to different categories from the ones reported for country comparison in Part 1 of the book.

Source: India country study survey (2001).

Total women working

Sector

Table 6.9a Work status of females (adult women and girls) – homeworker (hw) households

Subcontracted homework in India 193 Table 6.9b Work status of females (adult women and girls) – non-hw households Sector

% of women working to total women

% of children working to total children

Agarbathi Bidi (MP) Bidi (TN) Bidi (MP
TN) Zardosi All

27.7 32.6 12.5 23.2 16.4 22.5

4.5 — — — — 1.8

Source: India Country Study Survey (2001).

zardosi households had the highest proportion of women adding up to 29 per cent of the sample, while in agarbathi the figure was 12 per cent and in bidi 11 per cent. Within homework households, the average hours put in by women exceeds that put in by men. Agarbathi is exclusively women’s work while data shows that zardosi and, to a lesser extent, bidi are both occupations that involve the whole household. The lowest female/male ratio is found among bidi workers in Tamil Nadu which may partly be explained by the fact that some kind of minimum wages are assured to the workers, attracting more male participation. Also, the card holders of a Welfare Fund for Bidi Workers are men in most cases and holding this card entitles them in the long run to higher benefits for pension, provident fund, bonus and so on, which are supposed to be directly related to output. Survey findings show that around 50 per cent of female workers felt that homework is tiring. The highest response came from bidi (67 per cent) followed by zardosi (56 per cent) and agarbathi (29 per cent). Overall around 42 per cent felt that the return from this work was low and that the long hours of work put in every day have an adverse impact on other responsibilities. According to the time reported the average working hours were 8.3, compared to the small amount of time – 0.2 to 0.7 of an hour – allocated to childcare. Each woman doing homework has both adult and child helpers. Adult men are rarely found in agarbathi work, but do participate in bidi and zardosi work. Overall, each zardosi woman worker has 2.21 adult and child helpers; although only 0.62 is male. In bidi however, over half the helpers appear to be male. Among children, there is a higher participation of girls (Table 6.10). Of the 14.2 per cent of the working children, about two-thirds (62.2 per cent) are engaged in homework. Overall, in all three sectors nearly 60 per cent of working children are engaged in homework. Of these, nearly 75 per cent are working in the zardosi sector and 80 per cent of these workers are girls. In the other sectors too, though the degree of participation of children in homework is relatively less, the participation of girls is prominent.

194

Sudarshan et al.

Table 6.10 Family members who help women in homework Sector

Agarbathi Bidi (MP) Bidi (TN) Bidi (MP
TN) Zardosi All

% of females Adult   15 to total M F All members (age 5 and over)

Children 5–10

80.8 44.8 58.9 53.9

0.06 0.75 0.89 0.62

0.10 0.06 0.19 0.12 0.05 — 0.11 0.05

71.9 60.7

0.45 1.04 1.49 0.01 0.57 0.82 1.39 0.06

0.65 0.62 0.63 0.83

0.71 1.32 1.52 1.45

M

F

Children 11–14 All

M

F

All

0.16 0.31 0.05 0.16

0.06 0.21 0.09 0.15

0.26 0.21 0.09 0.15

0.32 0.42 0.18 0.30

0.16 0.17 0.16 0.39 0.55 0.10 0.16 0.14 0.27 0.41

Source: India Country Study Survey (2001).

Table 6.11 Women literacy rates (%) Sector

Agarbathi Bidi (MP) Bidi (TN) Bidi (MP
TN) Zardosi All sectors

hw

Non-hw

Rural

Urban

Combined

47.8 — 62.6 62.6 44.9 50.2

51.6 62.8 62.5 62.8 48.0 54.8

49.8 62.8 62.6 62.7 46.2 52.5

58.3 74.1 94.4 84.1 50.4 63.7

Source: India Country Study Survey (2001).

Social networks play an important role in the introduction of homework activities to the family. In fact, given low levels of literacy it is not surprising to find that skills are acquired informally and at an early age, and are mainly learnt from other members of the household or neighbours. The level of skill required varies, with zardosi work being the most highly skilled of the three sectors under study. Zardosi work also appears to involve more households, with about 48 per cent gaining an introduction through the family and about 47 per cent through relatives at ages under 14. Findings from the household questionnaire show that the overall literacy rate is nearly a fifth less in homework households compared to non-homework households (53 per cent compared to 64 per cent) and is substantially less than the recent literacy estimates for the country as whole. This is consistently observed in each of the three sectors with the zardosi sector having the lowest literacy both in rural and urban areas (46 per cent – see Table 6.11). Thus it can be concluded that skill formation owes little to literacy or formal education.

Subcontracted homework in India 195 Literacy rates are found to be low among women in homework households, and the general levels of literacy among women are higher in the CG of households. Policy initiatives have to be more focused on improving the literacy of women homeworkers in a way that would enable them to educate and equip their children better. The survey found that the household environment of homework households seems to be relatively poorer than that of the CG households. Not only is the quality of shelter poor, but also the homework has to be managed within a very limited space. Lack of space can lead to material being spoiled. Lack of electricity is a special problem when the home is doubling as a workplace, and specially so when the work involved requires focused concentration, as with zardosi, or fine work. Most of those who claim to have a toilet in the house do not have a proper one. It may often be one corner of the house in the villages. The general health of homeworkers is influenced greatly by the hygiene and sanitary conditions of the household and its environment as well as by the type of activity each sector requires. Bidi workers in Indore reported the highest prevalence of health problems (43 per cent) followed by agarbathi (41 per cent) and zardosi workers (39 per cent). Prevalence rates are higher in urban areas than in rural areas and, according to the survey findings, women suffer over 90 per cent of all health problems faced by homeworkers. The nature of health problems varies. In agarbathi gynaecological problems and body ache among women was widely reported; in the bidi sector joint pain and diahorrea were reported by many; in zardosi, which requires long hours of concentrated embroidery work, possibly in poor light, common complaints are shoulder pain, watery eyes and poor eyesight. There are differences across sectors in the percentage reporting full-time participation. The lower level of full-time workers in bidi reflects the peculiar situation of an industry with a shrinking market, but where workers are organised enough to have a work-sharing agreement among all those who are registered workers and have identity cards (Table 6.12). The survey reveals that homeworkers generally stay with the same contractor. Despite the absence of written agreements, benefit payments or any other formal arrangements, there is therefore a high level of stability in the whole structure of production. This situation is made possible partly by expanding markets (in the case of zardosi and agarbathi) which create an incentive for the contractor to hold on to old workers, and in the case of bidi by the greater degree of unionisation so

Table 6.12 Frequency (%) of full-time participation in homework by household members Sectors

Children 5–14

Adult   15

All

Agarbathi Bidi Zardosi All

26.7 13.2 44.4 32.3

72.7 39.9 75.5 59.5

54.0 33.5 65.5 51.5

Source: India Country Study Survey (2001).

196

Sudarshan et al.

that despite the shrinking market all card holders are entitled to a share of the work available. But a large part of the explanation for the workers’ unwillingness or inability to change contractors lies in the range of controls employed by contractors. These include: High cost of change: both contractors and workers are aware of the general lack of alternatives, and the excess supply of labour in relation to demand. Debt bondage: the contractor is often the only available source of credit, and the earnings from work are often the only form of repayment. In Tamil Nadu, this has led to bonded labour. Debt bondage in a weaker form is found in all the sectors. Delayed payments: this seems widely prevalent among zardosi workers, where each piece may take a long time to complete and payment may not be made immediately after completion. Keeping workers isolated: providing raw materials at home and collecting the finished product is interpreted by the workers as an advantage of homework. It also prevents the emergence of worker solidarity and unionisation, and exposure to other ways of thinking and types of work. Importantly, it ensures ignorance about the market. Incentives: small incentives may be offered to buy loyalty. For example, it was reported that some schoolchildren in urban areas roll agarbathis to cover their school transport expenses. There is also a ‘positive’ form of control arising from the benefits that bidi workers can access through the bidi Workers Welfare Fund.

6.6

Children in homework

‘Child labour’ refers to any child doing paid or unpaid work in factories, workshops, establishments, mines and in the service sector such as domestic labour. There are several distinctions that have been made in the law and in the literature on child labour. Children (under 14) who are employed outside the home as full-time wage earners and in activities that are prohibited by the Child Labour (Prevention and Regulation) Act 1986 comprise the ‘hard core’ of child labour. A second group includes children in activities not covered under the Act. A third category is children working part time or full time in family economic enterprises, or in domestic activities: around 80 per cent of working children fall in this group and are not currently covered by legislation. The last category would include children who are in school but who work on an occasional basis as well as children who are not in school. The term ‘child worker’ captures this last category as well. Those who take a ‘rights’ perspective on the question, and argue that all children have the right to education and childhood, favour the wider definition. From this definition it follows that all children who are out of school should be seen as child labour. It is also argued that this is the only way of ensuring equal access to schooling to girls many of whom would otherwise be

Subcontracted homework in India 197 found at home helping in the care of siblings and various household chores (see for example, UNDP, 2001). Another view is that it is important to focus on the ‘worst forms of child labour’ and that therefore it is justifiable to concentrate on children who are in bonded labour, or working full time in hazardous industries. To a large extent this position reflects an understanding of what is feasible given the context of poverty in which child labour is found. Thus, the ILO and UNICEF have stressed the need to eliminate the more exploitative forms of child labour first. For example the ILO’s International Programme on the Elimination of Child Labour (IPEC) has identified as its priority target groups bonded child labour, children in hazardous working conditions and occupations, children who are particularly vulnerable that is, very young (under 12 years of age) and working girls (see Annex 5.1). Children are also among the most vulnerable group of the population. There is a commitment on the part of the government to focus on the improvement of the position of children in need of special protection and the elimination of all forms of child labour. One of the main initiatives to achieve this is the universalisation of primary education. According to the 1993–94, NSS data – carried out by the Central Statistical Organisation of the national government – the incidence of child labour in India is 6.2 per cent of children in the 5–14 age group. The incidence in rural areas (7.2 per cent) is much greater than in urban areas (3.1 per cent). The difference between boys (6.3 per cent) and girls (6.0 per cent) is however small. Incidence is much higher in the 10–14 age group (13.8 per cent for rural boys, 14.1 per cent for rural girls and 6.6 per cent for urban boys, 4.5 per cent for urban girls) as compared to the 5–9 age group (1.1 for rural boys, 1.4 for rural girls and 0.5 per cent for urban boys, 0.5 per cent for urban girls). In the present sample of 452 households in homework, a total of 631 children were recorded in the 5–14 age group. They accounted for 28 per cent of the total population in the study area with 15.8 per cent and 12.1 per cent in the 5–10 and 11–14 age groups respectively. Correspondingly, in CG households a total of 170 children in 151 households were recorded. Data on children can be presented in the following categories: (a) working children – those reported to be working full time; (b) students – children who are enrolled in school and are students, although some of them may be working part time; (c) other children – those who are neither studying nor working (in the terminology of Chapter 5) or ‘nowhere’ children as popularly described. As child labour activists emphasise, these children are generally doing household chores and represent potential child labour. If we use the definition of working children as children ‘not in school’ we find around 28 per cent of boys and 28 per cent of girls between 5 and 10 to be in this group, and 46 per cent of girls and 41 per cent of boys in the 11–14 age group.

198

Sudarshan et al.

Table 6.13a Status of children by gender – homework (hw) households Sector

% of children working

% of children working in ‘hw’

% of children currently enrolled (students)

Girls Boys Girls Boys Girls Age: 5–10 years Agarbathi Bidi (MP) Bidi (TN) Bidi (MP
TN) Zardosi All Age: 11–14 years Agarbathi Bidi (MP) Bidi (TN) Bidi (MP
TN) Zardosi All

% of Total children children neither working nor studying

Boys Girls Boys Girls Boys

1.9 4.6 — 2.1 6.0 3.5

3.7 1.9 — 4.6 — — — 2.1 11.5 6.0 5.1 3.5

1.2 74.1 74.3 24.1 — 81.8 91.7 13.6 — 100.0 92.7 — — 91.8 88.5 6.1 1.6 53.7 49.2 40.3 5.1 71.2 71.7 25.3

22.0 8.3 7.3 11.5 39.3 23.2

100 100 100 100 100 100

100 100 100 100 100 100

25.7 5.0 20.0 12.5 43.2 30.0

36.1 8.3 15.0 11.4 30.4 25.4

2.8 8.3 15.0 6.8 16.7 8.7

16.7 8.3 — 4.6 33.3 15.9

100 100 100 100 100 100

100 100 100 100 100 100

17.1 5.0 15.0 7.5 43.2 26.4

54.3 85.0 75.0 80.0 37.8 58.7

47.2 83.4 85.0 84.1 41.7 58.7

20.0 10.0 5.0 7.5 18.9 15.8

Source: India Country Study Survey (2001).

Contractors may prefer to avoid direct arrangements involving children due to the possibility of being penalised. They consequently use the parents as mediators and avoid hiring children for high visibility tasks such as delivery of finished goods or raw materials. The incidence of child labour varies across the three homework sectors (Table 6.13a). Results from the survey showed that the zardosi sector has the highest percentage of children aged 5–14 years old, working full time and the lowest percentage of children enrolled in school. The zardosi sector has the highest percentage out of school in the 11–14 age group, with 85 per cent of girls and 64 per cent of boys not attending. The highest percentage of enrolments is found in the bidi sector, with students aged 5–10 years old, representing 71.4 per cent for girls and 100 per cent for boys. This sector also has the lowest percentage of children working, reaching only 28.6 per cent for girls in the same age level. However, it must be noted that a substantial percentage of children enrolled in school is also engaged in work, albeit on a part-time basis. Such is the case of both zardosi and bidi. This group is very small in the agarbathi sector. The incidence of child labour is much lower across the CG households in the sectors (Table 6.13b). Additional survey results showed that of all working children, a little over half are girls. Among bidi households, the burden of work seems to have fallen evenly on boys and girls. In agarbathi households, the participation of girls at work represents 38 per cent of all working children. In zardosi, however, girls in homework are 60 per cent of all working children in these households.

Subcontracted homework in India 199 Table 6.13b Status of children by gender – non-hw households Sector/age

% of children working

% of children working in hw

% of children currently enrolled (students)

Girls Boys Girls Boys Girls Boys Age: 5–10 years Agarbathi Bidi (MP) Bidi (TN) Bidi (MP
TN) Zardosi All Age: 11–14 years Agarbathi Bidi (MP) Bidi (TN) Bidi (MP
TN) Zardosi All

% of Total children children neither working nor studying Girls Boys Girls Boys

— — — — — —

5.0 — — — — —

— — — — — —

— — — — — —

88.5 75.0 60.0 100.0 100.0 100.0 71.4 100.0 76.0 30.4 81.0 76.5

11.5 40.0 — 28.6 24.0 19.0

20.0 — — — 69.6 21.6

100 100 100 100 100 100

100 100 100 100 100 100

— — — — — —

25.0 — — — — 6.7

— — — — — —

— — — — — —

66.7 37.5 33.3 37.5 100 100.0 71.4 — 28.6 100 100.0 100.0 — — 100 100.0 100.0 100.0 — 100 20.0 36.4 80.0 63.6 100 61.2 53.3 38.7 40.0 100

100 100 100 100 100 100

Source: India Country Study Survey (2001).

The inclination and enthusiasm of children to attend school is clearly evident from the ratio of children ‘ever enrolled’ estimated from the survey. The gap between the ever enrolled and the currently enrolled is a measure of the percentage of drop outs from school. The percentage of children ever enrolled is higher in all sectors than the currently enrolled. Highest enrolments are among bidi workers, with 94 per cent and 95 per cent ever enrolled in bidi in Madhya Pradesh and Tamil Nadu respectively, and 87 and 86 per cent as currently enrolled. Among agarbathi workers, as against 81 per cent ever enrolled children, 66 per cent were currently enrolled. Among zardosi workers, while 70 per cent were in the ever enrolled category, only 44 per cent were currently enrolled. There are differences in the level of enrolment across sectors, but in all cases there is substantial drop out. High levels of drop out are equally a problem among the CG, with the exception of bidi in Tamil Nadu where no drop out was reported. The zardosi sector emerges as one with low enrolment and exceptionally high drop out compared both to other sectors and to the CG. The zardosi sector presents the highest substantial percentage with between 35 and 40 per cent of both boys and girls reported as not attending school because they are working. Case studies suggest that poor performance in school is often a reason for drop out, although it may be followed by engagement in economic activity (Table 6.14). Activists argue that getting children into school is the best way of countering the forces leading to child work and child labour. But it is an important assumption

200

Sudarshan et al.

Table 6.14 School enrolment status of children (5–14) by gender Sector

Girls % of children ever enrolled in school

hw households Agarbathi Bidi (MP) Bidi (TN) Bidi (MP
TN) Zardosi All Non-hw or CG households Agarbathi Bidi (MP) Bidi (TN) Bidi (MP
TN) Zardosi All

Boys % of children currently enrolled in school (students)

% of children ever enrolled in school

% of children currently enrolled in school (students)

80.9 92.9 100.0 86.6 72.7 81.9

66.3 83.3 80.0 86.5 30.8 61.9

81.4 95.8 89.6 92.7 66.4 79.8

66.1 87.5 84.1 86.4 43.5 65.7

89.5 100.0 100.0 100.0 80.0 87.6

81.6 100.0 100.0 87.5 60.0 74.2

78.6 91.7 100.0 94.7 73.5 80.2

64.3 71.4 100.0 89.5 58.8 67.9

Source: India Country Study Survey (2001).

of this argument that ‘school’ should refer to ‘regular, full-time school’. This survey did not probe into the type of school attended by children. However, this is clearly an important factor in determining the extent to which schooling is likely to open up new opportunities for children. Further research is needed into this question, in order to establish the relevance and the quality of the education being imparted. Investigators report that a majority of the children, especially girls, from zardosi worker households (these were exclusively Muslim households) who were reported as being in school, were in fact attending madrasas. A madrasa is traditionally a place for religious instruction only. It is often connected to a mosque and its purpose is to introduce young boys and girls into the correct forms of prayer. Towards this end, Arabic is taught, so that children are able to read the Koran. However, some attempts have been made to extend the teaching given in the madrasas to include basic arithmetic and Urdu. It is not known, however, how successful these efforts have been. But parents have chosen to treat attendance at a madrasa as similar to attendance in a regular school, in their response to investigators. In the zardosi sector, one reason behind early induction into this work is the desire to pass on a traditional skill. Presumably, each household would attempt to balance the gains, present or future, from sending children to school and compare this to the value of keeping them at home and assisting in home and other household work.

Subcontracted homework in India 201 In Tamil Nadu, many of the children of bidi worker households were found to be attending schools set up under the programme for elimination of child labour. The higher participation of children of bidi workers in schooling may be due to the activities of the bidi Workers Welfare Fund, through which scholarships are given; in addition there have been special efforts such as the CLASS17 programme of the Tamil Nadu government. However, according to information disclosed by children during the FGDs, time devoted for studies is definitely affected by their work in bidi. Moreover, they declared the risk of being punished if they did not comply with the required working hours. Case studies confirm that both these schemes have played an important role in the higher participation by bidi workers’ children in school. Of those children aged 6–14 who are not currently enrolled, the reasons for their non-enrolment were recorded. In 44 per cent of the cases it was reported that the children were not interested in studying. The second most important reason that was stated for not attending school was that these children were working outside home (21 per cent). It is not clear whether they were doing this out of poverty, or whether they were attracted to the work in preference to being in school. Another 19 per cent of the homework households reported that they could not afford to send children to school. This may not necessarily mean that they are unable to bear the expenses of education, direct or indirect, but could also mean that the opportunity cost of sending these children to school is high. This, for example, appeared to be the case in a remote village in Tamil Nadu, where women reported in the FGDs that given high rejection rates and subcontractor commission they needed their children’s help in rolling enough bidis to earn a minimum amount (Table 6.15). The reasons behind children doing homework are closely linked to the reasons behind non-enrolment. The significant contribution of children’s work confirms the economic gains to the household from their participation. An estimate can be made of the contribution of children to homework, on the time spent on work. That is, no allowance is made for differences in the actual work done, work intensity and so on. The numbers here simply provide an estimate of the proportion of total time spent by children on homework as a ratio to total time spent on this by the household. The average contribution, for the three sectors together stands at over 13 per cent, ranging from a low of 8 per cent in bidi to 17 per cent in zardosi work with agarbathi at 15.4 per cent. Such incidence of child labour in homework, and the suggestion that both subcontracted homework and child labour may have increased over the last decade, stands in contrast to the fact that at a macro level, the incidence of child labour has shown a steady decline in both rural and urban India. These data on trends offer some support to the view that child labour decreases with urbanisation, and with the growth of more technologically sophisticated industries. Data from the NSS 55th Round shows a further decline in child participation rates. Social consensus is probably the most important factor explaining the nature of interventions adopted by a society against child labour, as well as the degree of

202

Sudarshan et al.

Table 6.15 Reasons for not attending school in homework (hw) households Sector

1

Agarbathi Boys Girls Bidi (MP) Boys Girls Bidi (TN) Boys Girls Bidi (MP
TN) Boys Girls Zardosi Boys Girls All Boys Girls

19.4 28.6

2

3 3.2 9.5

4

9.7 —

45.2 47.6 60.0 100.0

5

6

7

— 4.8

22.6 9.5

100 100

— —

— —

100 100

20.0 —

— —

20.0 —

50.0 —

— —

— —

50.0 —

— 20.0

— 80.0*

100 100

36.4 —

— —

9.1 —

54.5 44.4

— 11.1

— 44.4

100 100

15.1 12.8

9.4 20.5

30.2 16.7

41.5 43.6

1.9 1.3

1.9 5.1

100 100

18.9 14.8

6.3 16.7

21.1 12.0

44.2 44.4

1.1 2.8

8.4 9.3

100 100

Source: India Country Study Survey (2001). Notes * Of this 40 per cent was due to work at home. Can’t afford – 1, working in ‘hw’ – 2, working outside home – 3, child not interested – 4, lack of interest of parents – 5, others – 6 (incl. work at home, care of sibling, etc.)

success achieved in trying to implement legislation. In India, there has been some support for the view that child work is not always bad, drawing upon the Gandhian belief that work and education need to be integrated. The principle of Nai Talim is that knowledge and work are both forms of the same thing. The distinction, opposition or conflict between knowledge and work disappears, therefore it is impossible to distinguish between a knowledge-process and a work-process.18 Profit seekers may have exploited this perception. In addition, the quality and relevance of education are important elements in trying to generate a wider and stronger social consensus against child work and child labour.

6.7

From research to action

This section summarises the main recommendations emerging from the study. Recognition of homework: data collection Data on homeworkers is limited. While the NSS 55th Round survey is an important first step in improving the database, there is a need to firmly institutionalise this concern so that regular data becomes available.

Subcontracted homework in India 203 Our survey confirms the large proportion of workers, including ‘helpers’, both adult and children, in households doing homework. Both agarbathi and zardosi are sectors where the market is growing, and the high incidence of homework is encouraged by the efficiency and viability of subcontracting in the Indian context. It can, therefore, be expected to increase, making the issue of appropriate data collection even more important. While national level data is best collected by the official statistical agencies, there is also need for more micro studies in order to understand the specific vulnerabilities of various groups, and the Draft National Policy on Homeworkers provides an appropriate framework for such studies. Better data would help in the ‘recognition’ of homeworkers by making them, and their contribution to production, visible. Towards the ‘formalisation’ of homework Homework may be attractive both to the worker and the employer, although for different reasons. But from the perspective of workers, homework in sectors such as those studied here has several shortcomings: ● ● ●

the failure to ensure a decent wage; failure to give any kind of work security, since contracts are usually verbal; stability in work arrangements is maintained through the use by contractors and subcontractors of a variety of mechanisms to exercise control over workers, including delayed payments, debt bondage and others.

These problems can be dealt with to a large extent if the work is legitimised, so that workers are entitled to minimum wages, if agreements between contractors and workers can be monitored, and if workers are guaranteed access to credit facilities. These changes may require new legislation, systems of monitoring and so on. But most important of all, they require an agreement or consensus between ‘employers’ – contractors, traders, retailers – and workers, and an agreement both on responsibility and entitlement. Access: gender sensitisation Homework has a high female intensity. In the sectors studied here, females as a percentage of all persons working in homework are 97 per cent in agarbathi, 80 per cent in bidi and 86 per cent in zardosi work. In zardosi, a large proportion of the households is female headed (29 per cent). Special efforts will be needed to recognise and counter existing gender biases. Although the interventions mentioned above are needed for all homeworkers, it is clear that in the present situation we will need to reach out to women workers and it is not enough to say that they have equal entitlement. Thus, in the case of savings and credit institutions, the manner in which the SEWA Bank collects money from women by going to the house daily if necessary, should be recognised in any proposed interventions.

204

Sudarshan et al.

In the same way training at a central location may attract only a few women and some way has to be found to make such a facility more widely accessible. The challenge here is to sensitise systems that are currently not geared to giving any special attention to gender issues, and to develop consensus between male and female participants in homework. For example, in zardosi work, although more stable employment is becoming available through increasing export demand in the karkhanas, this work is available only to men. Women remain home and face a seasonal demand for the product. It may be possible to facilitate the setting up of women’s centres, but this will not happen on its own. An additional emphasis should be put on increasing the level of education of women. As was mentioned above, results from the survey demonstrated that with an increase in levels of education the proportion of women working in home activities falls. Training and certification Developing a system of giving training and certification at medium levels of skill is essential if any serious effort is to be made to increase occupational mobility. In addition, systems must be put in place to give recognition to highly skilled workers. To make such changes, the co-operation of those in charge of formal training institutions is needed. Training also needs to be demand-driven rather than supply-driven. Social security and the formation of associations One of the main motivations behind this study was to identify the nature of social protection needed by homeworkers. Since no security is currently available to these workers, this means that systems must be put in place to guarantee the entire range of social security that is considered as a basic entitlement for all workers. This would include basic sickness, old age and maternity benefits, and for children, access to schooling and health. There has been considerable discussion on the best mechanisms to implement a policy of basic social security, and the two approaches that are generally discussed are citizenship entitlements, and worker entitlements. The latter may be implemented through tripartite bodies such as the bidi Workers Welfare Fund. One of the objectives of this study was to see what light the survey could throw on the actual working of this Fund. From our findings, certain differences are evident between bidi and other sectors: ●

●

●

the relatively high share of homeworkers’ earnings in bidi in the value chain, perhaps due to their being organised; the greater use of government facilities by bidi homeworkers and the difference in health seeking behaviour, which could be a result of entitlements through the Welfare Fund; the shrinking market may have led to a reduction of full-time participation among workers rather than a reduction in the total numbers employed in this work. This sharing of work is possible as a result of the registration of workers;

Subcontracted homework in India 205 ●

the proportion of children in school is significantly greater than in the other two sectors, and this could be the result of scholarships offered by the Welfare Fund. As a result, the contribution of children in this sector is substantially lower than in the other two sectors.

These differences could be used to argue that the fact of workers being organised and the existence of a Welfare Fund has helped to improve the situation of workers, a fact that is clearly evident from comparison with the other two sectors. It should also be noted that: ●

●

as the sector is more organised, there is a greater contribution by employers to recurring costs incurred by workers; however, this may imply a cumulative indebtedness to the employer/contractor, which is greater than in the other two sectors; the coverage of the Welfare Fund and of other programmes designed for bidi workers is uneven, and workers in remote areas in Tamil Nadu are found to be in a very vulnerable situation.

The survey therefore confirms that the availability of a Welfare Fund has been able to empower bidi workers, but that greater efforts are needed even in this sector. Health concerns Among all the workers surveyed, the incidence of health problems was found to be high, and access to facilities poor. However, the zardosi workers were found to be particularly vulnerable in this respect. Both women and children are aware of health problems, in particular eyesight related problems, but no treatment is sought. A more detailed study is needed to ascertain the problems and appropriate interventions. Apart from curative interventions, improvements in the environment both within and around the house would make a major impact on health. Investing in water and sanitation, electricity, transport and telecommunication facilities, would raise productivity and incomes, improve the functioning of the market by giving workers easier access to relevant information, but also undoubtedly impact upon their health status. Occupational health and technological improvements are closely linked, and this again is a question of priorities and resources. These inferences are not new and have often been reiterated. The difficult task seems to be to motivate, or to generate incentives for those who have control over resources. Child labour General policy recommendations for all three sectors These include the following: Compulsory education for all children under the age of 14. The objective must be to ensure that all children are in regular, full-time schooling. At present, many of

206

Sudarshan et al.

the children surveyed and reported to be in school were actually in non-formal schools, or in madrasas, as explained earlier. This kind of schooling may not allow the acquisition of adequate abilities to make mobility in occupation possible. Much effort needs to be made to improve quality, and increase relevance, and to ensure that neither children nor parents are alienated from the school environment. Given that the quality and relevance of education is an important factor in voluntary enrolment and low drop-out, community user groups (in this case parents) are an important means of ensuring that teacher attendance is regular, and schools keep to the stated schedule. However more specialised inputs will be needed to assess the relevance of the curriculum. To some extent poverty is a factor behind non-enrolment/high drop out. An intervention that is reported to have been very successful in Brazil, the BoscaEscola programme, could perhaps be experimented with. In this programme mothers were given a monetary incentive for sending children to school. The idea of giving small incentives to children to enable them to come to school may also be experimented. However, making the payment to the parent would be discriminating against those parents who have already made the effort to send children to school despite their poverty. Instead, small rewards to all children say for regular attendance, at the end of a week or a month, would be one way of recognising the poverty aspect while not linked to the decision to enrol the child. There is a need for more research to identify the specific health problems of children, their causes, and the reasons behind the inability of children/parents to obtain medical treatment. Both health and education facilities for children could be made an integral part of a welfare fund, to be set up either as a sector specific fund, or an area specific fund. To release girls from the burdens of housework will require systems of child care near the home to be developed, and for outside work rotas could be encouraged. The study shows that around 4 hours per day is needed for girls to complete the household tasks expected of them. However, this time is likely to be distributed over the day in such a manner that they are not free to be away for several hours, as a regular school would require. There have to be changes in the manner in which households organise their work, and this may be easier if a whole community decides to do certain tasks in a joint manner. Such co-operative efforts may become possible if adult women are organised into groups and try to deal with the child care problem collectively. Some specific sector concerns Bidi sector The participation of child labour in bidi rolling is high, as confirmed by official data as well as the findings of this survey. However, the distinctive factor about this and the manner in which child labour in this sector is different, is that a high proportion of children is enrolled in school and work part time in bidi rolling. Enrolment is high because of interventions such as the CLASS

Subcontracted homework in India 207 programme in Tamil Nadu and, more generally, the provision of scholarships to bidi workers through the bidi Workers Welfare Fund. Despite this general picture, bidi is also the only sector in which bonded labour was still encountered by this survey. Sector specific policy recommendations that emerge therefore include the following: better monitoring to see that existing schemes are availed of in remote areas (the evidence suggests that there was less awareness among bidi workers in Madhya Pradesh than in Tamil Nadu, and in rural Tamil Nadu than in urban Tamil Nadu). This may require more thinking as to how the resources of the Fund can be augmented as well. Greater vigilance is needed to prevent the emergence of bonded labour situations. Agarbathi sector This sector has a high participation rate from girls. The point is to address the existing roles and responsibilities of girls, and interventions that would relieve them of some duties. Zardosi sector Zardosi work is highly skilled, and passed on from older members of the household to younger ones. Children’s involvement has been at the cost both of their schooling and health. Since many children from this group were found to be attending madrasas, there is need for some further research to understand better the extent to which such religious schools may be able to impart basic education and/or newer skills. The health issues also need further study prior to active interventions. Credit and marketing: institutional support Homework may be subcontracted, or own account. In this study we have focused on the conditions of subcontracted homeworkers (and used the term homework for this kind of work). Within the present set of arrangements, workers have to rely on their own resources or turn to subcontractors for credit. De-linking production and trade from credit provision must rank very high in the priority of interventions needed. Among the groups surveyed, a degree of debt bondage is found everywhere, and is highest among the Tamil Nadu bidi workers. However, it is necessary to analyse carefully the best way of making credit available. The regulations that any banking system need to follow make it difficult to deal with the very short-term and fluid needs of these workers. A very flexible kind of arrangement is needed. Perhaps one needs to think in terms of micro insurance and not micro credit, since at such low levels of income a part of the demand for credit will necessarily be for consumption loans. Similarly, joint marketing arrangements could be developed. But to develop sustainable arrangements of this kind, we need to think of an organisation that would combine organising women workers and giving training support, as well as marketing support. Developing sustainable systems The above recommendations have not explicitly discussed whose responsibility it is to implement the interventions suggested. The role of the government is clearly

208

Sudarshan et al.

high in any discussion of official data, legislation or tripartite arrangements. Beyond that several possibilities exist. A welfare fund will be sustainable only with regular contributions from employers and traders, as well as workers. It is also important to explore the nature of community systems that already exist, to see how best one might build upon existing social networks. The study has found that social networks, in particular family relationships, play a role in gaining entry into these sectors. However, the possibility of building upon these social networks to create community level assets, child care arrangements or social security systems, needs further examination. This is important because many of the interventions discussed above have a strong public goods dimension. This includes an institutional framework that provides credit more widely and easily to small producers, improvement of the environment – sanitation, water, communal spaces – as well as health facilities, and universal schooling.

Notes 1 This chapter is based on a UNICEF sponsored study conducted at the Indian National Council of Applied Economic Research (NCAER). Full acknowledgements and detailed findings are available from Sudarshan et al., 2001. 2 For instance for the agarbathi sector, the Agarbathi Manufacturers Association was contacted. In the Tamil Nadu area, the UNICEF office at Chennai assisted in the identification of the sample (for details see Sudarshan et al., 2001). 3 Question 1.8: ‘Specify religion: Hindu, Muslim, Christian, Sikh, other’; Question 1.9: ‘Specify caste SC-1, ST-2, backward, upper caste, other’. 4 Question 2.1: Non-food items included were: ‘fuel, house rent, clothing, domestic help, all others’. 5 The choice of benefits in Question 5.7 were: ‘annual leave, sick leave, maternity leave, compensation for accident, disability, occupational disease, pension, bonus, provident fund, others’. A question regarding the ILO convention was also included in this section. Question 5.9: ‘Have you heard about the ILO Convention?’. 6 Examples of Question 7.1: Nature of the house: kutcha, semi pucca, pucca, combination; floor of the house: mud, wood, stone, brick, cement, other; roof: thatch, wood, tiles, stones, terraced, cement slabs; cooking media: smoke emitting stove, smokeless chulah, non-smoke emitting stove, other. 7 Question 7.2 included: metal utensils, time piece, fans, torch, buckets, bicycle, wristwatch, record player, transitor, radio, motorcycle, cars, sewing machines, TV and refrigerator. 8 Questions 8.1–8.3 were: ‘What kind of facilities do you think the government can provide for you?’. ‘Can you think of any technological improvement in your activity?’ and ‘If you feel that by participating in this activity the education of your children is being disturbed, what suggestion do you have to make in this regard?’. 9 In the zardosi sector the number of women and girl participants was 20–25. In the bidi sector in Tamil Nadu and Indore, the number of participants was 9. In the agarbathi sector there were 6 participants in the FGD held in rural areas, and 8 in FGD held in urban areas. 10 4 cases studies in the zardosi sector, 2 in the agarbathi sector and 3 in the bidi sector. 11 Sensitive issues included health impact on women in the agarbathi sector, debt bondage in the bidi sector, empowerment of women in Indore in the bidi sector, new entrants in the homework activity, among others. 12 Some of the locations of the case studies were an urban centre in Lucknow; a rural centre in the village of Behta; Pernambut village in Tamil Nadu; and slum neighbourhoods of Ragi Guda Temple.

Subcontracted homework in India 209 13 Or at least, not to any significant degree, although there have been reports of agarbathi imports from Malaysia. 14 Further, the homework households included both ‘family’ and ‘non-family’ units. ‘Family’ units are those households where both the spouses work mainly in the home sector and ‘non-family’ units are those households where at least one adult woman member of the household works for homework and the male (household head) is not mainly engaged in homework. In either case, other family members may or may not be involved in homework. 15 One of the spin-offs of this research project, carried out in collaboration with SEWA Lucknow, was that the survey provoked an increase in membership of SEWA Lucknow among the zardosi workers in that vicinity. 16 It is important to note that the data in this chapter often refers to different categories from the ones reported for country comparison in Part 1 of the book. 17 Child Labour Abolition Support Scheme, functioning in the Vellore district of Tamil Nadu since 1995. 18 Excerpt from Vinoba’s Thoughts on Education (http://64.87.47.218/Though/educ.htm# NAI TALIM).

References Data source: India Country Study Survey (2001). Ghatate, Vinayak N. (1999), Informal Sector Contribution to India’s Exports: A Quantitative Evaluation. Consultancy Report for SEWA, New Delhi: Indian Institute of Foreign Trade. Government of India (1993), ‘Report of the Expert Committee on Estimation of Proportion and Number of Poor’, New Delhi: Planning Commission, Government of India. Jalees, Farida (1989), Glittering Threads: A Socio economic Study of Women Zari Workers, New Delhi: SEWA Bharat. Kumar, Arun (2001), ‘Beedi Industry in India. Background Paper Prepared for the Workshop on “Women Workers: an agenda for the future” ’, organised by the Group on Women Workers and Child Labour, National Commission on Labour, March 2001. NSSO (2000), ‘Non Agricultural Enterprises in the Informal Sector in India 1999–2000’ NSS 55th Round (July 1999–June 2000). National Sample Survey Organization. Ministry of Statistics and Programme Implementation, Government of India. Pradhan, Basanta K., Roy P. K. and Saluja M. R. (1999), ‘Informal Sector in India: A Study of Household Savings Behaviour’, Contribution of the Informal Sector to the Economy, Report no 2. New Delhi: NCAER. Sudarshan, R. M., Banerjee, M., Bhandari, L. and Venkatraman, S. (2001), ‘India Country Study, “Outsourcing of Manufacturing to Households: Subcontracted Home based Work in India” ’ (unpublished paper). UNDP (2001), Child Labour and Education: UNDP India Office’s Perspective, mimeo. UNIFEM (2000), A Preliminary Study on the Productive Linkages of Indian Industry with Home based Women Workers through Subcontracting Systems in Manufacturing Sector. New Delhi: United Nations Development Fund for Women.

7

Hazardous subcontracted homework in Pakistan Shahrukh Rafi Khan, Saba Gul Khattak and Sajid Kazmi

Since the late 1980s, scholars have acknowledged the existence of homework as an important component of the informal sector. On the one hand, with this outsourcing system entrepreneurs reduce production costs and circumvent labour legislation. On the other hand, workers find an income source in activities that require low levels of skills and investment. The expansion of homework in Pakistan can be partly attributed to the continued slowdown in the manufacturing sector and the decline in real wages due to intensive structural adjustment programmes (SAPs). Increasing poverty levels and the increasing flow of immigrants and refugees have also contributed to the expansion of homework. This chapter examines the conditions of homeworkers in four areas of activity: the first three in manufacturing – incense stick making (agarbathi), carpet weaving, bag stitching (bori) – and the last in fisheries, prawn shelling. Karachi was selected for the study for being Pakistan’s largest multi-ethnic industrial city in which most homework takes place. Section 7.1 examines the extent of the informal sector in Pakistan based on official sources. Section 7.2 reviews the existing literature about the conditions of homeworkers in Pakistan. It also examines the existing government initiatives and legislation for the protection of homeworkers. Section 7.3 presents the methodology used in the field study including a brief description of the instruments. Section 7.4 describes the production processes and the description of the value chain of each sector. Section 7.5 analyses the degree of exploitation of homeworkers. Section 7.6 draws on the quantitative results from the survey to analyse the main distinctions between homework and control group (CG) households. Finally, Section 7.7 summarises the findings and presents policy recommendations aimed at improving homeworkers’ conditions.

7.1

The informal sector in Pakistan

The definition used in official statistics does not reflect the real extent and dimension of the informal sector in Pakistan. According to the Federal Bureau of Statistics (FBS) of the Government of Pakistan,1 the informal sector is viewed as comprising units ‘ – such as household enterprises, engaged in the production of goods and services with the primary objective of generating employment and income [for]

Hazardous subcontracted homework in Pakistan 211 the persons concerned, not necessarily with the deliberate intention of evading payment of taxes or other legislative or administrative provisions’. Furthermore, ‘the units typically operate at a low level of organisation, on a small scale, and with labour relations mostly based on casual employment’. Other features include the ownership of assets by ‘owners’ rather than workers, expenditures that are indistinguishable from household expenditures and entities that are not party to contracts or incur liabilities. Since the latter is not relevant for Pakistan, the first two are used to define the informal sector for Pakistan for statistical purposes as follows: ●

●

all household enterprises owned and operated by own-account workers, irrespective of the size of the enterprise, household enterprises owned and operated by employers with less than ten persons engaged, excluding agricultural or non-market production.

Most of the literature on the informal sector in Pakistan is constrained by the official definition mentioned earlier (Kemal and Mahmood, 1993, pp. 5–15). It is therefore not surprising to find data indicating that unskilled females form a negligible proportion of the workforce in the informal sector (Kemal and Mahmood, 1993, pp. 96–102).2 Another limitation of the operational definitions used in official data statistics, is the exclusion of homework investigated in this report. First, it would be difficult to categorise the work we investigate as ‘units’ or ‘enterprises’. We study a ‘putting-out’ system whereby contractors supply material and have households do the processing. Second, ‘evading payment of taxes or other legislative or administrative provisions’ is clearly cited as a reason for the existence of such work. However, the work is home based and the labour casual, so perhaps it does partly fit into the informal or ‘sub-informal sector’. In terms of the operational definitions, the first bullet point would in most cases include the homework we investigate in that it is home based and size is not the issue. However, this would be true only if a very loose definition of enterprise as ‘activity’ is adopted. Even so, the equipment, such as in the case of carpet weaving, is generally not owned by the workers as assumed even by the first operational definition. Regardless of the conceptual flaws mentioned, the Labour Force Survey 1999–2000 tells us a lot (Government of Pakistan, 1998a). The highest proportion of both female and male non-agricultural workers is in the informal sector. About three-fourths (73.1 per cent) of female workers in rural areas and three-fifths (60.7 per cent) of female workers in urban areas are in the informal sector. For male workers, the proportions are similar with 64.1 in urban areas and 67.6 per cent in rural areas working in the informal sector. Between 1996–97 and 2000–01, the urban female informal workers, as a percentage of total female workers, increased from 55.3 per cent to 60.7 per cent and there was a corresponding drop in formal sector workers. A similar relative percentage increase was also evident for male informal workers. The higher proportion of males than females in the urban areas working in the informal sector, suggests the exclusion of much activity that

212

Rafi Khan et al.

actually goes on in the informal sector. However, this becomes even more evident from Tables 7.1 and 7.2, in which the percentage distributions by gender rather than proportions are presented.3 The most striking finding in Table 7.1 is how few women allegedly work in the informal sector, that is, about 14 per cent in all occupations combined in both rural and urban areas.4 This appears unbelievable taking into account the findings of our investigation in which 92 per cent of all adult workers in homework are women and, even among children, most were girls. Finally, according to the Human Rights Commission of Pakistan (1999), unionised labour is decreasing since only 7 per cent labour is unionised which suggests that informal contract labour is taking precedence over formal employment. Table 7.1 Pakistan: percentage distribution of persons (10 years and above) employed in the informal sector by occupation, gender and region (1990–2000) Occupation groups

Rural Male

Legislators, senior officers and managers Service and sales workers/ shop-owners Craft and related trade workers Plant and machine operators and assemblers Elementary/unskilled occupations Other Total

Urban Female

Male

Total Female

4.36

0.19

6.31

0.13

10.99

2.12

0.02

2.40

0.04

4.58

6.18

0.65

7.52

0.70

15.05

1.72

0.02

1.53

0.01

3.28

11.80

2.47

3.96

0.62

18.85

33.36 59.54

8.14 11.49

4.74 26.46

1.01 2.51

47.25 100.00

Source: Drawn from Government of Pakistan, Labour Force Survey 1999–2000 (2001a, p. 137; 2001b).

Table 7.2 Percentage distribution of employed persons (10 years and above) by employment status, gender and region (1999–2000) Employment status

Employer Self-employed Unpaid family helpers Employee Total

Rural

Urban

Total

Female

Male

Female

Male

0.01 2.62 9.42 4.13 16.18

0.16 43.02 16.66 23.99 83.83

0.04 1.62 1.14 5.87 8.67

2.26 32.26 8.83 47.98 91.33

1.82 47.11 11.09 39.98 100.00

Source: Government of Pakistan Labour Force Survey 1999–2000 (2001a, pp. 147–149; 2001c). Note All rural workers, female and male, add up to 100%. Similarly, all urban workers, female and male, add upto 100%.

Hazardous subcontracted homework in Pakistan 213

7.2

Homework in Pakistan

Reviewing the literature briefly The systematic studies about homework in Pakistan in the past focused mainly on the carpet weaving and football stitching sectors in Punjab (SCF, 1997; Ahmed et al., 1998; Khattak and Sayeed, 2000). Other authors have focused on the informal sector and attempted to quantify the number of homeworkers. Citing a World Bank study, they estimate that about 0.75 million women are engaged in subcontracted homework in the urban areas (Awan and Khan, 1992, p. 3). The growth of home carpet weaving dates to the early 1980s. The usual practice was for a thekedar (contractor) to install one or two looms (frames) in the household on a loan basis. In turn, the household was required to weave and pay off the loan with labour. Indebtedness gave the thekedar leverage, which he exploited by paying low rates. Child labour has always been associated with this sector. According to a World Bank study, there are approximately 1.2 million children (Awan and Khan, 1992, p. 3), under 15 years of age, working in the carpet weaving industry. Children’s involvement in this activity responded to the household’s economic need and affected their educational development. Enrolment rates in Punjab were low, and fell even more with the expansion of child labour in carpet weaving. Over half of the children that were attending school, left to weave carpets. Exploitation practices against children in this sector were very frequent. These included excessive hours of work (9–10 hours per day) and a hazardous environment that tremendously affected children’s health. Researchers observed severe muscular-skeletal disorders and stunting and frequent cases of fingertip injuries and backaches (Awan and Khan, 1992). The study on the subcontracting of football stitching interviewed 326 households in Sialkot (SCF, 1997). Researchers attributed the expansion of homework to the increasing demand for footballs in the domestic and foreign markets. For large manufacturers outsourcing production to homework was more cost-effective than producing at excess capacity. In order to facilitate outsourcing, manufacturers set up stitching centres. These centres allowed employers to supervise, monitor and control the quality of the products manufactured by homeworkers. More than half (58 per cent) of total homeworkers engaged in this activity were women. Despite various incentives to induce women to go to stitching centres, including free transportation, free meals and segregated rooms, and higher rates, only 2 per cent of women attended. Women’s preference to work at home responded to the cultural resistance to female mobility and flexibility of combining domestic work with paid work (see also Chapter 4). Another study used a purposive sample of 821 women on a range of activities including garments and football stitching (Ahmed et al., 1998, p. 18). Findings revealed extreme poverty and deplorable living conditions among households.5 Indicators of low living conditions included premature ageing among women, high illiteracy levels among women, low education levels, large family sizes and the high incidence of child labour. Children joined homework at early ages. The only

214

Rafi Khan et al.

exceptions were made for daughters-in-law. Since their involvement in homework was considered a social embarrassment, they were usually assigned the burden of domestic chores. As in most homework cases, the excess demand for work provided contractors with leverage to exploit women and pay reduced rates. A study6 by Khattak and Sayeed (2000) on garment, plastics and textiles sectors confirmed many of the findings cited earlier. This study provided a useful account of the factors that forced women towards homework. Women generally preferred homework due to the implicit pressure of family members and neighbours and the cultural context. For most women working outside the home, taking public transportation and joining a union were unthinkable options given their subordinated status in society. Legal protection and home-based work The lack of legal protection and enforcement mechanism also contributes to the expansion of the informal sector, and specifically, homework in Pakistan. The government’s labour policies and laws favour industrialists and investors (Sayeed and Ali, 2000) rather than the well-being of workers especially women and children. Although no direct legislation exists to protect homeworkers, international conventions as well as the Constitution of Pakistan and other laws interpreted broadly can be applied towards such protection. Pakistan is a member of several International Labour Organisation (ILO) Conventions including the ILO Homeworkers Convention of 1996, the Maintenance of Social Security Rights Convention of 1982, and the Labour Statistics Convention of 1985. At the national level, the Pakistani Constitution (1973) provides general provisions that could be extended to homeworkers including Article 3 that states that ‘the state shall ensure the elimination of all forms of exploitation’. In addition, Pakistan inherited British legislation that bans child labour and provides factory employment standards.7 Regarding women’s conditions, the 1994 Task Force on Labour policy recommended: ‘ . . . the constitution of a special committee to study and identify the problems faced by working women and their needs and to suggest ways and means to evolve certain minimum standards to protect women workers and prevent their exploitation’. Similar recommendations were drafted during the Pakistan National Report for the Fourth World Conference on Women in Beijing and the Report of the Commission of Inquiry for Women. Despite the awareness in these and other similar government documents, the government’s inability to initiate effective implementation for multiple reasons remains one of the biggest hurdles to improving the conditions of homeworkers.

7.3

Research method

Survey design and data collection The selected city for the study was Karachi, Pakistan’s largest industrial city.8 The homework sectors were chosen according to the following criteria: (a) homework

Hazardous subcontracted homework in Pakistan 215 includes hazardous environment or activities with potential impact on health; (b) women and children are involved as homeworkers; (c) the value chain is significant for at least one of the sectors. The four interviewing teams were comprised of one man and one woman who had graduate degrees in the social science disciplines. In addition, a female doctor accompanied the teams for gathering precise information about health and nutrition status of the households. Besides conducting the survey each team was required to write a journal based on the daily observations.9 Sample The methodology consisted of an interview survey conducted with 303 households involved in homework spread equally among the four selected sectors. In order to establish a reference point, a simultaneous survey was conducted with a control group (CG) of 94 households across the four sectors (Table 7.3). The purposive sampling used for the data collection was based on a paired snowball method. Given the difficulties in identifying homework households, four teams were responsible for gathering reference information from four households engaged in homework. The CG was selected from the household’s proximity to homework households to ensure similar socio-economic conditions across the homework and control sub-samples. An additional criterion for the selection of CG was that no child labour of any kind was being practised in order to understand the causes of child involvement in homework. This was unique to Pakistan, as in no other country did this condition apply in the selection of CG households. Questionnaire The design was based on a core questionnaire. In order to collect detailed information from each respondent (women, children, household, owner), nine different types of questionnaires were designed. The respondent questionnaires included one for homework households and another for CG households. The latter consisted of the same questions, excluding the section on work-related questions. The questionnaires filled in by the interviewing team consisted of two sets of questions regarding working and living conditions (to be filled in by the interviewer) and health conditions (to be filled in by the doctor). In addition, specific detailed questions were added to each questionnaire in order to adapt it to the Pakistani context. The questionnaires directed to women consisted of seven sections including general information, education, level of empowerment, work-related questions, non-economic activities (housekeeping activities/household chores), homework organisation and nutrition. The information from work-related questions allowed us to understand and quantify the extent of exploitation of homeworkers in the value chain. The section on empowerment consisted of questions aimed at understanding the power of decision-making in the household on aspects such as mobility,10 household financial management,11 children’s education,12 marriage

Orangi, Unit 12C, Karachi Korangi Town, Karachi Godhra, Karachi Machar Colony (Mohammadi Colony), Karachi

Location

397

99 99 100 99 303

77 77 75 75 94

22 22 25 25 303

77 77 75 74

hw

94

22 22 25 25

CG

—

— — — —

hw

CG

Total

hw

Rural

Households surveyed Urban

Notes Dash indicates that certain information was not collected in the survey or data were not comparable. FGDs  focus group discussions. CS  case studies. hw  homeworkers. CG  control group or non-hw household. * Considering separately the one for women and the one for children.

Source: Pakistan Country Study (2001).

Sub-total

Pakistan Incense stick making Carpet weaving Sack stitching Prawn peeling

Country/sector

—

— — — —

CG

8

2 2 2 2

Total

8

2 2 2 2

Urban

Number of FGDs*

—

— — — —

Rural

Table 7.3 Surveys on homeworker households in Pakistan: sectors, location, number of FGDs, case studies and households surveyed

8

— 5 3 —

Number of CS

Hazardous subcontracted homework in Pakistan 217 and household expenditures. In the section about nutrition,13 women were required to select from a list of types of food divided into meals per day. The aim was to learn about women’s dietary habits and nutrition levels. Finally, the health section was incorporated into the women’s organisation section.14 This is not to say that all this information would not inform us about women’s economic empowerment, critical to the study as a justification for women’s work as opposed to drudgery. The questionnaire directed to children consisted of seven sections including schooling, work-related questions, time organisation, household chores, earnings, social/family and nutrition. The section on nutrition was the same as the section used in the women’s survey. The section on time organisation included questions regarding consequences of faulty work including options referring to abuses by contractors and family members.15 There was no section on health since the doctor from the interviewing team was responsible for gathering precise information on children’s health. A third questionnaire was designed for household respondents. This questionnaire was divided into seven sections and included information of household activity, and detailed questions about living conditions, ownership status, number of households, income generated from homework, level of expenditure and indebtedness. The fourth questionnaire consisted of information filled in by the interviewer on the working and living environment of workers.16 The fifth questionnaire included three questions specifically for ‘seth’/owners17 regarding reasons to recruit homeworkers, and information about the value chain. A sixth questionnaire was filled in by the doctor and was aimed at collecting detailed information about the health status of children and women. Focus group discussions, case studies and field reports A total of eight focus group discussions (FGDs) were conducted. Two FGDs were held in each sector, one for women and one for children. There were approximately 10–12 participants from the household respondents in each discussion. The main topics discussed in the children’s FGDs were schooling, relationship with the subcontractor, impact of homework on health and playtime. The topics in the women’s FGDs focused on children’s education, health, violence in the community, perceptions about the subcontractor and the type of work performed, ethnic issues, reasons for engaging in homework and the impact of homework on health. A total of eight case studies were conducted: five in the carpet weaving sector and three in the sack stitching sector of which seven were about women and one about men. Households were selected for more in-depth case studies if they were either significantly worse-off or better off than the average, that is, there was some interesting distinguishing factor. The information collected for the case studies included general demographic information, relationship between spouses and household members, environment in the community, health status and relationship with subcontractors (see Annex 7.2).

218

Rafi Khan et al.

In order to enrich the qualitative data, each of the three field teams was assigned a task of writing a field report at the end of the fieldwork in each location including Orangi, Korangi, Machar Colony and Godhra. The observations in each of the five reports included: extent of exploitation visible in the contractor/worker relationship,18 impact of homework on health of women and children and their overall well-being, and the impact of homework on women’s empowerment (see Annex 7.1 for examples of field reports by sectors).

7.4

Homework sectors

Urban setting Karachi is known for being a multi-ethnic city with a population estimated at over 12 million.19 It is estimated that there are 54 per cent Urdu-speaking Mohajirs, 14 per cent Punjabis, 9 per cent Pathans, 6 per cent Sindhis, 4 per cent Baluchi and 12 per cent belong to other ethnicities including Bengalis/Biharis and Burmese (Parveen and Ali, 1996, p. 140). Known also as mini-Pakistan, Karachi continues to serve as the political hub for both the left and the extreme right. Since the mid-1980s Karachi experienced a high incidence of violence due to ethnic and political conflicts. According to media reports, an average of 630 violent deaths (95 per cent male) per year were recorded in the city during the ten-year period starting 1990 (Hisam, 2000, p. 23). Over the years, violence has been more or less restricted to the low-income areas of Karachi. These areas are, understandably, concentrated around Karachi’s industrial estates and around the port/harbour. These localities have been populated through ethnic and communal bonds, which are encouraged by state policy due to a variety of reasons, both good and bad.20 Many of Karachi’s illegal migrants have also settled in these areas. According to Hasan (1997; 1999, p. 41) Karachi’s illegal immigrants and refugees include Bangladeshis, Sri Lankans, Burmese, Filipinos, Iranians and Afghans. Barring the latter two, a majority of the others work in the garment and fishing industries while some along with the Afghans work as domestic servants. The areas where the survey was carried out have been the hub of violence and curfews. Korangi and Orangi have suffered the worst forms of politically motivated ethnic violence as well as undergoing religious/sectarian violence. Incense stick making (agarbathi) The majority of workers in this sector were Burmese who came to Pakistan in the late 1960s and early 1970s due to the hostile treatment of Muslims in Burma. Some of them had residential plots leased to them by the government. Others lived in huts and paid bhatta (forced rent) to toughs of a local political party, the Muhajir Qaumi Movement (MQM).21 Bengalis were the other community engaged in this work. Agarbathi making is a traditional activity mostly performed by women (in Orangi). The Burmese community was very conservative and enforced strict purdah. Girls were prohibited from going outside the house after puberty, which is

Hazardous subcontracted homework in Pakistan 219 Table 7.4 Details of activities by sectors and actors Activities by actors

Homeworkers

Subcontractors

Activities by sector Incense stick making

Carpet weaving

Sack stiching

Prawn shelling

Mixing of sawdust with colour Making paste Making agarbatti Drying Making bundles Supplying inputs

Weaving

Stitching

Shelling

Supplying inputs

Supplying inputs

Sorting Organising activity (Networking with retailers and workers) Providing inputs

Cone making Lachi tani (frame)

Supplying inputs Sorting

Providing credit Collecting output Sorting/quality control Delivering output

Kanni kichai (cutting) Tucking Washing

Source: Khan et al. (2001), SDPI (2001). Note Activities marked to subcontractor in the carpet-weaving sector are done by artisans whose services are hired by the subcontractors.

roughly at 12. Some households considered it inappropriate for girls to leave the house after the age of seven. Value chain Homeworkers are at the low end of the value chain and are responsible for the production of agarbathi. The activities in the production include mixing of the sawdust, paste and colour; rolling of the agarbathi sticks, and packing in bundles of 1,000 (Annex 7.1).22 Homeworkers work collectively and produce approximately 4,000 agarbathi per day (four bundles per day) (Table 7.4). Incense sticks are sold both in domestic and foreign markets. Domestically, agarbathi is mostly used in religious places like mosques, graveyards or mandirs (Hindu temples) and also used for fragrance in houses. The value chain involves various actors between the production and final consumption including homeworkers, and subcontractors at factories, wholesalers and retailers. Subcontractors are the pivot of the value chain. They are responsible for collecting the material from the factory and distributing it among several homeworkers. Subcontractors are also required to identify the defective units.23 The final stage of production takes place in factories where fragrances are added

220

Rafi Khan et al.

to the incense sticks and these are packaged for wholesalers and retailers to sell to local consumers or through international traders for export. Within this value chain, the homeworkers receive barely a fraction of the total value for which the product is sold. Homeworkers received an average of Rs 5.5 for 1,000 agarbathi (one bundle), while subcontractors received Rs 6.524 and retailers Rs 6,000 for the same quantity (Table 7.5). Hence, the share of remuneration that homeworkers receive is 84.6 per cent of the subcontractor’s earnings and 0.19 per cent of the final consumer price. The average monthly income per capita of a subcontractor ranged from Rs 9,000 to Rs 12,000.25 The per capita monthly earning of homeworkers (including children) was Rs 115 that was the lowest among the four sectors (Table 7.6). Carpet weaving Pakistani carpets are very popular in foreign markets. Recently, exports have experienced a considerable growth. During the period of 2000–01, carpet exports grew by 8.9 per cent and reached an export value of US$177.47 million.26 The export value of 2002 is estimated in US$263 million and has a target of $300 million. The only period that experienced a drop was in 1995 after the murder of Iqbal Masih, a child carpet weaver of Pakistan.27 The major markets for Pakistani carpets are the United States,28 Europe and Japan.29 The communities in this sector were located in Korangi and the population was mainly Burmese. The fact that they were ‘newcomers’ and not part of the city meant that they had less access to city facilities and were less urbanised.30 However, they claimed to be Bengalis since this ethnic group was perceived as a higher status group among the illegal immigrant community. Value chain The value chain is similar to the agarbathi sector in which subcontractors are the link between homeworkers, in this case the carpet weavers, and wholesalers and exporters (see Figure 7.1). Subcontractors worked as agents for exporters/ wholesalers and ensured a regular flow of work. They collected materials from the exporters/wholesalers, although, in some cases, they used their own material. The subcontractor bought wool from the wholesaler and hired workers to make cones out of the wool for use by the homeworkers. A karegar (artisan) set up the lachi and tani (main horizontal and vertical chords that embody the design) along which workers wove the carpet. In the final stage of production, the extra threads are cut and the carpet is washed with chemicals to give shine (Table 7.4). Homeworkers work collectively in the weaving of carpets. Household members took turns to work on the looms and completed pheras (rounds, standard units for measuring a day’s work). The wage rate per phera varied from Rs 1.5 to Rs 3 and a household completed 30 to 40 pheras per day (Table 7.5). A carpet could engage a family up to three months for a payment of approximately Rs 1,600. The average hours worked by women and children is 6.9 hours per day. Children worked in

4,000 sticks

25 rounds

Incense stick

Carpet weaving Sack stitching Prawn shelling

5.5 per 1,000 sticks 1.5 per 1 round 32 per 100 sacks 10 per kg prawns

Unit rates (average)

110.0

55.7

37.5

22.0

Daily earnings of hw (23)

3,245

61

135

26

Daily earnings of subcontractor on units produced by hw per day

3.4

91.3

27.8

84.6

4 as % of 5

4,345

n.a.

206

4,000

Domestic consumer’s payment for units produced by hw per day

Note Final consumer means domestic consumer because exporters were reluctant to reveal international market prices.

Source: Khan et al. (2001), SDPI (2001).

11 kg

174 sacks

Units produced (average)

Sectors

2.5

n.a.

18.2

0.6

4 as % of 7

164.8

n.a.

31.9

200.0

US consumer’s payment for units produced by hw per day (US$)

Table 7.5 Daily earnings of homeworkers as a percentage of subcontractor earnings and as a percentage of what consumers pay

1.1

n.a.

2.0

0.2

4 as % of 9 after converting 4 into US$

222

Rafi Khan et al.

Table 7.6 Average hours worked and income by sector Sector

Mean hours worked (children)

Mean hours worked (women)

Mean daily income per person (Rs)

Average monthly income per person (Rs)

Incense stick making

5.82 (2.78) 6.49 (2.92) 4.15 (2.20) 4.85 (2.50) 5.34 (2.76)

7.64 (2.26) 6.92 (2.67) 6.59 (2.49) 6.68 (3.10) 6.95 (2.75)

25.60 (14.32) 38.55 (25.86) 55.63 (33.43) 113.24 (82.96) 57.73 (57.19)

115.0

Carpet weaving Sack stitching Prawn shelling Total

169.0 203.0 503.0 247.5*

Source: Khan et al. (2001), SDPI (2001). Notes Parentheses contain standard deviations. Households are identified by their primary activity. Twelve households were also engaged in a second activity and in those cases the combined income is reported. Question: How many hours do you work during a day (day and night)? * Average of monthly income per capita.

HBW

Washing Subcontractor Tucking Lachi tani

Cone making Wholesaler/exporter

Local consumers International traders

International consumers

Figure 7.1 Carpet weaving.

karkhanas (work places or factories) where they were supervised by subcontractors. The monthly average daily income per person was Rs 169 (Table 7.6). Carpet prices vary according to the number of knots per square foot, complexity of design and quality of the wool. It usually ranges from Rs 9,000 to 11,000.

Hazardous subcontracted homework in Pakistan 223 The average revenue for subcontractors is Rs 4,000. In foreign markets carpets are sold at a mark-up price that ranges from Rs 100,759 to Rs 118,540 (equivalent to US$1,700–2,000) which applies to a 5  8 carpet. The share received by homeworkers is insignificant.31 They earn approximately 28 per cent of the subcontractor’s earnings, and approximately 2 per cent of exporter’s earnings (Table 7.5). Sack (bori) stitching The production of bori is derived from the demand for onions and potatoes in foreign markets that use this type of material for exports. The main market for this product is the United Arab Emirates. Homeworkers engaged in this sector are located in Godhra and migrated from Indian Gujarat. Godhra was different from all the other localities. Rather than a concentration on one particular trade, people were engaged in numerous activities. For this reason, Gujratis are traditionally viewed as entrepreneurs. Other ethnic minorities in this sector included Punjabi, Sindhi and Baloch. Value chain There were not many tiers involved between homeworkers and final consumers. Subcontractors are the link between the low and high end of the value chain. They are responsible for providing the inputs to homeworkers and delivering final good to exporters. Homeworkers are at the end of the value chain and are responsible for stitching the bags which does not require a high level of skill32 (Table 7.4). The value chain in this sector is not as unequal as in the other sectors given the few levels of intermediaries and simple production process. Homeworkers are paid Rs 30–33 per hundred bags stitched and subcontractors receive Rs 35 for the same amount. Thus, homeworkers receive 91 per cent of the share per bag (Table 7.5) and there is a minimal difference of Rs 2–5 between price paid to homeworkers and to subcontractors. Nonetheless, there is a significant difference in the monthly income between homeworkers and subcontractors. While homeworkers’ monthly per capita income from homework was Rs 203 (Table 7.6), subcontractors earned an average income from Rs 7,200 to 12,000 per month. This is because each subcontractor received final products from 40 to 50 households. Homeworkers worked collectively in bag stitching and produced approximately 300–400 sacks per day33 with the assistance of other household members. Given the variable market demand for potato and onions, homeworkers were employed in other activities such as towel cloth, waste sorting and ribbon making. Prawn shelling Prawn shelling34 is a traditional activity inherited from past generations in Machar Colony. Seven types of prawn are caught in Karachi. The biggest sizes are exported to Japan and United States while smaller and lower quality prawns are

224

Rafi Khan et al.

sold in the domestic market. Karachi exporters asserted that they face unfair competition from Balochistan exporters because the latter receive government subsidies and produce at lower labour costs. In addition, multinational companies have affected enormously the production of local fishermen and provoked environmental problems. Licenses to multinational companies for deep-sea fishing have reduced the catch enormously. According to the Secretary General, Pakistan Fisher Folk Forum, fish landing at the fishery was 24,837 metric tonnes in October–December 1999 and it increased to 32,550 metric tonnes in the corresponding period in the year 2000, an increase of 7,713 metric tones. This was possible because the government had cancelled licenses for the multinationals upon protests from the fishermen. The government has again started issuing these licenses, and the situation has started worsening. Although these companies are not supposed to fish in the radius of 35 nautical miles from any harbour, this last condition is violated in collusion with government officials. Deep-sea fishing by foreign owned trawlers results in environmental problems, such as dumping of small dead fish in the sea, and also economic hardships for the local fisher folk, and naturally prawn shellers. A significant group of illegal Bengali immigrants lived in Machar Colony and engaged in prawn shelling. Most of them lived in extremely poor conditions and were settled in landfill areas. Value chain The value chain entails three levels including homeworkers, subcontractors and exporters. As in the other sectors, the subcontractor is the link between the high and low end of the production chain. Subcontractors collect prawns from the fishery, deliver them to homeworkers for shelling and classification and deliver the shelled prawns to exporters. It takes approximately 1–2 hours to shell between 10 and 15 gala (1 basket containing 4–5 kg raw prawn) per day35 (Table 7.4). As seen earlier, homeworkers receive a small share of the total revenue. Homeworkers’ wage represents a fourth of the wage received by subcontractors. Assuming that 1 kg shelled prawn is sold at Rs. 295, homeworker’s share is approximately 3.4 per cent of what a subcontractor earned and 2.5 per cent of what a consumer paid in the domestic market. In the foreign market, the same amount is sold at US$14.98; hence, the share of a home base worker is approximately 1.1 per cent (Table 7.5). This sector has a particular payment method aimed at facilitating the accounting for subcontractors. Workers are issued tokens with the contractor’s name.36 One token is equivalent to one gala, and if the subcontractor is reliable, it can also be used as currency in shops. Payments are made after a fortnight in return for a kilogram of shelled prawns and the token. The average monthly income per capita from prawn shelling was Rs 503 and is the highest among the four sectors (Table 7.6). Table 7.6 shows hours of work for all sectors. Women devote 6.7 hours to shelling per day, while children work 4.8 hours per day. Work is seasonal and the

Hazardous subcontracted homework in Pakistan 225 high season is from October to March. In some months, work is available for more than ten days, in slack months it drops to once a week. Homeworkers’ activities during slack seasons include carpentry and masonry. There were more than 200 subcontractors engaged in the prawn sector in the colony. In this cluster subcontractors provide workspaces or jheenga waras where 50–70 children in the 4–13 age group work. Due to the perishable nature of the product and the lack of storage facilities children are required to start working at early hours (4 am), just as the fishermen’s boats return from sea.

7.5

Exploitation of homeworkers

Homework is predominantly a female activity. While there was an average of 1.41 adult women per household doing homework, only 0.11 men per household surveyed were engaged in homework.37 Alternatively, in 90 per cent of the cases, women rather than men were engaged in homework – though this was one of the criteria for the selection of the households. Similarly, while an average of 1.4 girls per household were engaged in homework, there was an average of 0.51 boys per household engaged in it. The mean age at which women in the sample began work was 20 and, at the time of the survey, they had been working for an average of 12 years. Women’s predominant engagement in homework is a result of household economic constraints, male household member’s unemployment, illegal status and social pressure. As a reason for doing homework, ‘rising prices’ and ‘desire to supplement family income’ constituted a significant share of the responses; the rest pertained to paying off loans or illness, death, disability or unemployment of some wage-earning member of the household.38 In various cases, women were the main income earners because of the high level of unemployment among men (see case study 1, Annex 7.2). Such was the case for the agarbathi sector where 42 per cent of men in households were unemployed (Table 7.7). Among all sectors the percentage of unemployment was 22 per cent. The other reason that forced women into homework was illegal status that restricted their mobility.39 Social pressure played a role in women’s engagement in homework. Women were pressured by husbands to engage in this work40 and accepted the superior position Table 7.7 Percentage of male household heads not working by sector Sector

Percentage

Incense stick making Carpet weaving Sack stitching Prawn shelling Total

42.4 20.3 15.3 22.0 26.1

Source: Khan et al. (2001), SDPI (2001).

226

Rafi Khan et al.

of males as a religious mandate.41 The internalisation of these norms was reflected in the survey responses. In 74 per cent of the 303 cases, women responded that their decision to work was ‘of their own volition’42 and 84 per cent of the women stated they preferred homework instead of working outside the home. Although women earned a living for their households, their contribution was not appreciated or acknowledged. Homework involved approximately 7.6 hours per day for them, in addition to the regular household chores and responsibility for child care. Women were often subjected to abuse, physical and verbal, from subcontractors (case study 1, Annex 7.2). In some cases, women were subjected to a lesser degree of this type of exploitation thanks to the purdah practice. This was the case for the women in prawn shelling who had little direct contact with subcontractors.43 A similar situation was found for Bengali women. Since they practised very strict purdah, they were not allowed to have contact with males from the community other than their husbands. Nonetheless, they were not free from mistreatment since males in the household often abused them.44 Women from Sindhi tribes in the carpet weaving sector experienced similar situations and could be subjected to constant beatings and abuse from males in the household.45 An exception to this mistreatment were women from the sack stitching (Gujrat community) sector who enjoyed a higher mobility and a lesser degree of gender discrimination. There were exceptional cases where women were allowed to run shops46 (see case study 3, Annex 7.2). Indeed, women’s decisional power was higher than in the other sectors and more influential in aspects such as household finance. Women from this sector were able to save money from homework for children’s education. Relationship with subcontractor Besides being exploited in the value chain, homeworkers were subjected to other types of exploitation from subcontractors including delayed payments and malpractice (case study 1, Annex 7.2). In almost two-thirds of the cases (63 per cent), the contractor was late in paying for the work.47 In addition, a quarter of the respondents stated that there were occasions on which they were not paid at all. A significant proportion of women (31 per cent) felt that the contractors made work difficult for them by engaging in malpractice. These included supplying poor quality material (19 per cent), miscounting, wrongly measuring or arbitrarily rejecting products. In the bag stitching sector women complained of receiving insufficient provision of raw materials.48 As a consequence, women were unprotected and their situation was aggravated by the fact that only 12 per cent had a written contract. During FGDs, women mentioned being physically and verbally abused by subcontractors (case study 1, Annex 7.2). However, homeworkers did not complain much about the contractors profiting at their expense or bad behaviour. On the contrary, homeworkers welcomed the presence of subcontractors given the limited employment alternatives. Another aspect that prevented homeworkers from complaining was the existence of kinship ties and the belief that contractors

Hazardous subcontracted homework in Pakistan 227 Table 7.8 State of indebtedness by sector Sector

Total annual stock of debt (Rs)

Mean monthly household exp. (Rs)

Debt stock as percentage of annual exp.

Incense stick making

9,431.82 (22,973.28) 11,984.62 (14,283.67) 22,587.76 (28,871.99) 19,188.14 (26,185.95) 15,511.62 (21,789.08)

4,103.13 (1,683.58) 4,032.23 (1,523.99) 5,113.29 (2,327.80) 4,370.34 (1,551.56) 4,400.41 (1,841.46)

19.2

Carpet weaving Sack stitching Prawn shelling Total

24.8 36.8 36.6

—

Source: Khan et al. (2001), SDPI (2001). Note Parentheses contain standard deviations.

were not much better off than them. This could explain the finding that 46 per cent of the women responded that they had a good working relationship with the contractor49 (case study 3, Annex 7.2). Even so, less than half (37.2 per cent) cited verbal and physical abused as being the causes for the bad relationship with subcontractors. Too much work (10.9), poor pay (20.5), late payments (23.1) and malpractices (8.3) accounted for the other responses. A critical aspect in the subcontractor and homeworker relationship was the level of indebtedness (Table 7.8). In all the sectors, the total debt stock represented a high burden on the household expenditure and income. The highest percentage of debt stock of annual expenditure was found in the bag stitching sector (38 per cent), followed by the prawn shelling sector (36 per cent). Given the scarce resources and savings, and high interest rates, households remained highly indebted. This led to bonded labour, which was the case in the carpet weaving and agarbathi sector where adult males acquired loans to then work off payments through children’s labour (case study 1, Annex 7.2). Children in subcontracted homework The number of children per household was high (4.46) for various reasons including early marriage,50 polygamy51 and rejection of family planning since it was considered to be against Islam. Household sizes were large and included an average of more than seven members (Table 7.9). The majority of the children went to madrasa52 because they could not afford regular school or because schools were inaccessible. Some of these madrasas provided both religious and regular education, while others provided religious education only. Not all children benefited from education at the madrasa since there was strong gender discrimination in favour of boys. According to the women, most of the children were working at

228

Rafi Khan et al. Table 7.9 Mean household size and number of children per household Sector

Mean household size

Mean number of children per household

Incense stick making

7.82 (2.70) 6.78 (2.11) 8.00 (1.99) 8.41 (2.61) 7.79 (2.46)

4.55 (1.72) 4.42 (1.72) 4.82 (1.45) 4.17 (1.91) 4.46 (1.74)

Carpet weaving Sack stitching Prawn shelling Total

Source: Khan et al.(2001), SDPI (2001). Note Parentheses contain standard deviations. Different age category from Chapter 4.

Table 7.10 Mean hours spent by children on homework and other activities Sector

Homework

Chores

Homework school

Meals

Play

Incense stick making

4.10 (3.44) 4.73 (3.69) 2.00 (2.51) 1.74 (2.15) 3.17 (3.28)

2.62 (2.40) 1.74 (1.45) 2.03 (1.75) 2.20 (1.96) 2.15 (1.94)

0.94 (2.14) 0.79 (2.07) 1.69 (2.76) 0.18 (0.82) 0.90 (2.13)

1.48 (0.80) 1.31 (0.67) 1.25 (0.79) 1.31 (0.89) 1.34 (0.79)

1.09 (1.78) 1.32 (1.74) 1.47 (2.11) 2.64 (2.69) 1.62 (2.18)

Carpet weaving Sack stitching Prawn shelling Total

Source: Khan et al. (2001), SDPI (2001). Notes Parentheses contain standard deviations. Data are obtained through a different question that the one used for comparison in Chapter 4.

home (88 per cent of the responses) to supplement family income. Only 5 per cent of the responses suggested that the children were doing homework because the family preferred they stay at home.53 Children’s time among all sectors was mainly distributed between homework and domestic chores (approximately five hours per day) leaving scarce time for school activities and recreation (Table 7.10).54 Children reported being tired after the homework and unable to complete the school’s homework, hence invoking the ire of the teachers. For example in the carpet weaving sector, children’s time was mostly devoted to working in karkhanas (average hours of 6.49) and domestic chores (1.74 hours). In the bag stitching sector, children assisted their mothers55 4.1 hours per day.

Hazardous subcontracted homework in Pakistan 229 According to reports in FGDs, children would have preferred going to school full time, and some were even conscious that their work supported their schooling. Effect of homework on health Homeworkers were affected by the polluted and poor conditions where they worked (Annex 7.1) and by the exposure to chemicals involved in the production processes. For example, the high content of toxic chemicals in sawdust used in agarbathi resulted in discolouring and injury to the skin, and irritation in the upper respiratory track that eventually developed into asthma in chronic cases. In bag stitching, the dust from bori caused throat irritation, itching and infections on the hands. Sneezing, coughing, runny noses, flu, respiratory problems, neck and chest pains and watery eyes were also common among children. Other ailments included leucorrhoea for women, watering and weakening eyes and pain in the chest, joints, back and limbs. In the prawn shelling sector, the constant contact with iced water caused infections, coughs, breathing problems and skin infections. One of the most common ailments was muscular pain due to repetitive activities and long hours of work.56 Access to doctors and public health institutions was very limited. In addition, homeworkers did not have access to government health centres.57 As a result, homeworkers substituted traditional medicine with the application of local remedies (desi totkas).58 Collective action Women engaged in very little collective action to improve their work conditions. The only exception was found among women from carpet weaving who compelled contractors to increase wages by collectively negotiating higher rates. However, only 12.5 per cent of home-based workers (38 homeworkers) managed to negotiate a rate increase via collective action with other homeworkers. Nonetheless, there was willingness to engage in collective action (53 per cent of responses). At the time of the survey, there were virtually no organisations capitalising on this potential for collective action. Only two women mentioned an association with an organisation of any kind in a work-related context.

7.6

The impact of homework: a quantitative analysis

This section complements the findings reported in Section 7.5 that was largely based on qualitative information. It compares the conditions of homeworkers with households from the CG. Homeworkers’ poverty Homeworkers are below the poverty line. While the two groups (homeworker and the CG households) were similar with regard to mean expenditure, the mean

230

Rafi Khan et al.

Table 7.11 Socio-economic profile of homeworker and non-homeworker households Socio-economic profile

HHs where women were doing hw

HHs where women were not doing hw

Mean monthly expenditure (Rs) Mean household size (nos.)

4,408.36 (1,890.60) 7.57 (2.35) 622.16 (287.57) 6.78 (2.10)

4,575.02 (2,548.30) 6.52 (2.47) 739.30 (388.10) 9.14 (2.47)

Mean monthly per capita expenditure (Rs) Living condition index@

Source: Khan et al. (2001), SDPI (2001). Notes Parentheses contain standard deviations. These data are obtained through a different question than the one used for comparison among countries in Chapter 4. @ The living condition index ranges from a minimum possible of 1 for the poorest household to 18 for the most well off. Houses scored 0 for the worst living condition on various categories, 1 for a slightly better state, 2 for a lower middle class condition on a particular category and 3 for a mark of wealth. Only households possessing a VCR could have scored a 3 (two among hw households and three among the CG). Thus the poorest household lived in a kaccha house, had access to community water supply, did not have access to running water or electricity and possessed no durable assets like a radio, cassette player, fridge, TV, VCR or a source of transportation.

household size was significantly larger among the homework group, which resulted in a significantly lower per capita monthly expenditure among the latter (Table 7.11).59 The Social Policy and Development Center (SPDC) (2001, p. 29) reported a mean monthly per capita cut-off expenditure of Rs 610 for 1996–97, to define poverty in urban Sindh. Inflating this to 2001, and using the subsequent cut-off of Rs 751.40, 76.9 per cent of homework households and 64.9 per cent of non-homework households were below the poverty line.60 This compares very unfavourably with the 20 per cent of households in income poverty in urban Sindh (calculated using 1996–97 data, p. 30). Households not engaged in homework (CG) have significantly better conditions than homeworkers (Table 7.12). There is relatively greater prosperity among the CG, reflected in the lifestyle as judged by household facilities and assets, higher investment on health, higher investment on education and better nutrition levels. It is quite evident from the data that families not involved in homework had better health status. The doctor judged 80.8 per cent and 72.1 per cent respectively of the boys and girls to be anaemic among the children doing homework and this was the case for 65 and 67.3 per cent respectively of the children in the CG. These rates of anaemia are much higher than among the general urban population.61 Thus, both groups revealed a very high incidence of anaemia, although girls fared better in this regard.62 Anaemia is directly related to diet habits. Indeed, CG women revealed better food intake than that of women doing homework.

Hazardous subcontracted homework in Pakistan 231 Table 7.12 Monthly expenditure patterns (Rs) of homeworker and non-homeworker households Items

HHs doing hw Expenditures

Items of daily use House rent Health Schooling Transportation Gas Electricity Water Other Total

3,294.52 (1,643.59) 127.23 (283.79) 341.82 (345.36) 110.15 (328.42) 117.55 (216.05) 114.69 (132.11) 116.11 (151.86) 136.99 (238.96) 49.32 (156.32) 4,408.36 (1,838.63)

HHs not doing hw % of total 74.7 2.9 7.8 2.5 2.7 2.6 2.6 3.1 1.1 100.0

Expenditures 3,056.88 (1,867.20) 153.19 (347.87) 462.51 (1,037.31) 117.83 (212.38) 212.23 (395.70) 149.04 (265.06) 176.44 (248.74) 118.24 (168.62) 129.04 (724.87) 4,575.41 (2,543.30)

% of total 66.8 3.3 10.1 2.6 4.6 3.3 3.9 2.6 2.8 100.0

Source: Khan et al. (2001), SDPI (2001). Note Parentheses contain standard deviations.

While a fifth of the CG for women consume a cup of milk per day, only 7 per cent of the homework women did so. Both groups eat vegetables only a few times a week and fruit about once a week or less. The median consumption of meat for the homework group was about once a week (46 per cent), a fifth of the women in the CG ate meat a few times a week or more. Overall, the nutritional intake between both groups was poor. As reported in detail in the country study (Khan et al., 2001), the intake of milk, vegetables, fruit, pulses and meat was much below the average dietary frequency of these foods among the general urban population for persons five years and above for 1990–94.63 The main predictors of households likely to take on homework were household size and living conditions (inverse association).64 Part of the reason for the greater prosperity of the non-homework households was the slightly higher number of male earners per household (1.16) compared to 0.98 among homework households.65 Remuneration from homework is very low. The mean daily household income derived from homework was Rs 57.7 (or just over $1 at the market exchange rate in 2000). This was generally from one activity. There were only 12 households out

232

Rafi Khan et al.

of 303 in our sample engaged in 2 activities. This income amounts to about one-third of the Rs 172.1 earned by unskilled daily labour in November 1999 in Karachi.66 Women homeworkers Surprisingly, there seems to be less constraint on the mobility of women not doing homework. While two-fifths of the women doing homework said they are free to go outside the house when they need to, half the women not doing homework said this was the case. Once again, we were surprised to find that women doing homework did not have notably more say in household decision-making than non-homework households. In both the homework group and the CG a quarter of the women said they were the heads of the household. Only 28 per cent of the women not doing homework said they would do it should the opportunity present itself. The rest had no intention of taking on such work. The reason, according to a quarter of the CG women respondents, was that they had enough income. Twenty per cent cited various other disadvantages of homework, another 10 per cent mentioned health reasons and 6 per cent mentioned that it was dangerous for children. Autonomy can also be manifest in ability to control earned resources. Women retained 84.5 per cent of their earnings from homework. Some of them (9.5 per cent) managed to save and 12.5 per cent engaged in saving via a ‘committee’ system.67 Only three women said they were retaining income for a daughter’s dowry. About a fifth of the income (23 per cent) was retained for personal expenses and almost two-fifths of the responses suggested this was to support a personal habit including the chewing of tobacco or various local equivalents. The other significant categories on which there were high responses included food (11.5 per cent), health (13.2 per cent), children’s education (7 per cent) and other children’s needs (13 per cent). The level of indebtedness is higher among homeworkers compared to CG households. There were also other differences between the two groups in a substantive sense. Three-fourths of households doing homework had contracted a loan in the last six months compared to 46 per cent for those in the CG. The mean accumulated debt of Rs 15,511 was also higher compared to Rs 13,946, though not by much. The pattern of borrowing was quite similar, except that while 23 per cent in the homework category borrowed from the contractor, only 10.7 per cent of CG households borrowed from their employers. Fewer than 5 per cent of the loans in both cases were interest based and while 37.9 per cent of the homework households expected to pay off their debt based on this work, 84.6 per cent of the CG households expected to repay their debt from their savings. Most of the responses on the advantages of homework were actually disadvantages of alternatives. Only 3.4 per cent of the responses cited the lack of skills for alternative work as a reason for doing homework and only a fifth of the responses cited being able to care for the children. Three-fourths of the respondents felt they faced mobility problems or felt unsafe going out and hence preferred homework

Hazardous subcontracted homework in Pakistan 233 to other work, and cited family or neighbourhood pressure against work outside the house. The bulk of the responses concerning the disadvantages of homework pertained to the negative health impacts (34.7 per cent) or fatigue (28 per cent). An untidy house was another disadvantage (14 per cent of the responses). Only 9.6 per cent of the responses cited less time for the family as a disadvantage. Perhaps the pressure of survival and the family at least being in the same premise accounts for the low response on this count. Average working hours of women engaged in homework are greater than women in CG households. Women engaged in homework face the double burden of working hours and domestic chores. Women worked an average of 3.5 hours on homework alone. They worked on average a six-day week and two-thirds of them worked the whole year round. Of those who did not work the whole year around, 93 per cent said it was because work was not available the whole year around. A comparison of the time profile of women of homework and CG households indicates that women engaged in homework spend less time in leisure than women from CG households (female homeworkers spend 1.49 hours compared to 2.11 hours spent by CG women). In addition to the hours devoted to homework, women are required to work an additional 4.3 hours on domestic chores which is slightly less than CG women.68 Homework affects women’s health status. While nine-tenths of the women reported having health problems as a result of homework, 79 per cent of the women in the CG also reported having health problems. Surprisingly, proportionately fewer women doing homework complained about exhaustion. Women not doing homework had higher incidence of responses on general ailments. They suffered from, as expected, fewer pains in the back, limbs and joints. As in the case of the children, only 16.7 per cent of the women in the CG responded that they could not afford the needed treatment while this was the response for 53.9 per cent of the women doing homework.69 Both access to and the expense of health care were problems and more so for the homework group relative to the CG (Table 7.13). Homeworkers have more limited access due to lack of economic resources. For example, while 63.8 per cent of CG households said they could afford a private doctor, only 12.9 per cent could among homeworkers. This difference reflects the ability of the CG households to invest more in human capital partly because of their smaller household size and higher per capita income. Children Children engage in homework at early ages. For children, the average age of starting work was 7.5 years and they had been engaged in this work for an average of 3.6 years. Unlike the adults, most of the children (85 per cent) received training. The responses showed that the bulk of the training came from within the household (77 per cent), relatives/friends (5.6) or neighbours (7.5). Only 3.7 per cent of the responses indicated that the contractor provided any training.70

234

Rafi Khan et al.

Table 7.13 Percentage responding affirmative regarding easy access and affordability of health facilities (%) Medical facility

Government dispensary Private doctor Hakeem Homeopath Government hospital Private hospital

Access

Affordability

Women not doing hw

Women doing hw

Women not doing hw

Women doing hw

12.8 5.3 14.9 17.0 12.8 5.3

14.2 4.3 22.4 21.1 14.2 4.3

— 63.8 59.6 59.6 — 35.1

— 12.9 11.6 11.6 — 4.3

Source: Khan et al. (2001), SDPI (2001).

Table 7.14 The main reasons why children doing homework and CG children were not in school (%) Reasons for not being in school

School does not provide a better job School too far School too expensive Prefer to stay at home Family needs income from work/poverty Parents think school is worthless

Reponses of children doing homework

Responses of children not doing homework

Responses of mothers of children doing homework

Responses of mothers of children not doing homework

—

—

—

2.5 65.2 2.8 27.8

8.2 69.4 4.1 n.a.

1.4 68.9 1.4 19.4

9.8 75.6 2.4 n.a.

0.9

6.1

0.7

2.4

0.6

Source: Khan et al. (2001), SDPI (2001).

Homework hinders children’s education. Only a fifth of the children doing homework were in school in contrast to more than double that percentage (42.6) among children from CG. Responses as to why children were not in school from the two groups and their mothers are indicated in Table 7.14. As indicated earlier, poverty, or the need to supplement family income, was an issue as is evident from 28 per cent of the responses for children doing homework. Women doing homework, as a group, understated the importance of this factor relative to the children. Only 9 per cent of the children doing homework thought that this interfered in any way in their schooling, mostly because they got tired, or work left little time for schooling. Of the 42 children who dropped out from school, the majority (57 per cent) did so because parents did not have enough money for school.

Hazardous subcontracted homework in Pakistan 235 Another way of exploring this issue was asking children their reasons for engagement in homework. Over four-fifths of the responses (83 per cent) pertained to the need to add to family income because of poverty, inflation, loans, or the illness, disability or death of earning members in the family while only 1 per cent of responses indicated children worked because they did not like school. This then confirms the earlier findings of why children engage in homework and are not in school. Children were also asked what they would prefer to do if given a choice. Children not enrolled in school among those doing homework overwhelmingly indicated a desire to attend school full time (59.6 and 57.7 per cent responses among boys and girls respectively) and others indicated a preference for combining school with work or play (23.6 and 24.1 per cent responses among boys and girls respectively).71 There is a strong association between household prosperity and child schooling among homework households. The relationship is positive, significant and linear and, as such, we did not identify threshold effects. A living condition index (we estimated) and per capita household expenditure had positive and significant association, thus confirming the subjective responses of why children are not in school. By the same token, household indebtedness has a strong negative and significant association with the household enrolment ratio. Children engaged in homework face a higher burden of work. Children’s working hours were excessive and did not leave time to engage in other activities. Children reported working an average of 3.7 hours during the day and 1.6 hours after sunset on homework. They also worked six days a week and 58 per cent said they worked the whole year round.72 Most of the earnings from homework was turned over to the parents by both girls and boys (over 90 per cent). A quarter of the children retained some of the earnings, and nearly all who retained earnings spent it all. Often children were punished for faulty work. Three quarters of the children said that there were times when they did not perform the work well. The responses indicated that the consequences were scolding by the family (28.9 per cent) or the contractor (19.7), a fine (14.4), a beating by the family (12.2) or a beating by the contractor (7.8).73 Children engaged in homework devoted less time to recreation activities. Children in the CG participated less in games (particularly girls), watched more TV and spent more time with friends (Table 7.15). Most notable is that less than half of the responses for girls doing homework, compared to the CG, indicated spending time with friends. Also, there were three times more responses suggesting that girls doing homework had no time at all for play. Children engaged in homework suffer from more health problems and cannot afford treatment. As in the case of women workers, these ailments include pain in the back, limb, joints, blisters and cracking skin. While only 8.5 per cent of the children in the CG responded that they needed medical treatment but could not afford it, this was the response of 43 per cent of the children engaged in homework. Conversely, 73 per cent of the children in the CG got medical treatment and were released immediately, while this was the case for only 33 per cent of the

236

Rafi Khan et al. Table 7.15 Time allocation to homework (hw) and other activities by gender for children doing hw and children not doing hw (hours) Activities

hw Chores Play School work

Children doing hw

Children not doing hw

Boys

Girls

Boys

Girls

2.04 (2.42) 1.17 (1.48) 2.67 (2.45) 2.67 (2.45)

3.40 (3.39) 2.35 (1.96) 1.40 (2.06) 1.40 (2.06)

—

—

0.90 (1.48) 3.44 (2.71) 1.93 (2.75)

1.51 (1.60) 2.64 (2.14) 1.66 (2.43)

Source: Khan et al. (2001), SDPI (2001). Note Parentheses contain standard deviations.

homework group of children. Finally, we inquired about the frequency of illness in the past six months, and the mean number of times for homework was 5.2 times (almost once every month) while it was 3.8 times for the CG. Government assistance Homeworkers are pessimists about government assistance. The qualitative findings indicate that at least the Bengali and Burmese communities expected little from the state, since they did not view themselves as citizens. A question about expectations from the state puzzled them, and some responded that they wanted money. Others asked that a reasonable social and physical infrastructure, particularly education, healthcare and sanitation, should be provided. In addition, of the third who responded to the question of what kind of assistance they would like from the government or some other organisation, technical training (32 per cent), credit (27 per cent) and framing better policies (28 per cent) accounted for the bulk of responses. However, the low response rate on this question indicates the pessimism women felt about the likelihood of anything concrete being done for them or their children.

7.7

Conclusions and policy recommendations

Conclusions Homeworkers, like all informal workers, are deprived of protection and basic safety nets. In addition, the group of homeworkers among the four sectors studied indicate that the level of human development is low as reflected in the low rates of education, large household sizes, poor health and nutrition status and low income per capita. The situation is even more dramatic for those who have an illegal status. Such is the case of Bengali men and women.

Hazardous subcontracted homework in Pakistan 237 Homework is predominantly a female activity. Women’s predominant engagement in homework is a result of household economic constraints, male unemployment, illegal status and social pressure. Indeed, poverty and related factors were the main motivation for engaging in homework for women and children. Over a quarter of the male household heads (26 per cent) were unemployed and thus the household relied on the earnings of homework of the women and children. The value chain in each sector indicated the existence of economic exploitation defined as the unequal distribution of revenue across the value chain. The share of homeworkers relative to the revenue of the subcontractors, contractors and final distributors is extremely low ranging from 0.06 of retail revenue per unit in incense stick making to 18.2 per cent in carpet weaving. Besides being highly exploited in the value chain, homeworkers were subjected to other types of exploitation from subcontractors including delayed payments and malpractice. A critical aspect in the subcontractor and homeworker relationship was the level of indebtedness. In all the sectors, the total debt stock represented a high burden on the household expenditure and income. Bonded labour was going on in the carpet-weaving sector. Thus women and children tried to work off loans that the families had taken. As a consequence, contractors took full advantage of their economic power and, given the excess supply of labour or the lack of organising ability or community unity, prevented homeworkers from collectively getting better terms. The urban localities these communities lived in were highly polluted and deprived of adequate social and physical infrastructure. Moreover, the type of tasks involved in subcontracting resulted in deterioration of health conditions due to exposure to sawdust and chemical substances in agarbathi, and carpet fibres in the bori sectors. Also the constant stitching caused hand infections and in prawn shelling, constant contact with iced water provoked breathing problems. These communities were conservative in their religious interpretation of Islam and oppressive of women. In some cases women were shielded from physical and verbal abuse from subcontractors thanks to the practice of strict purdah. Nonetheless, in the household males severely restricted women’s mobility and made them responsible for all the household chores and childcare in addition to homework. As mentioned earlier, the engagement of children is a result of households’ financial needs. On average, children began to work at the age of seven and a half years of age and worked an average of 4.5 hours a day, 6 days a week. Children’s time among all sectors was mainly distributed between homework and domestic chores (approximately 5 hours per day) leaving little time for school activities and recreation. Exploitation of children and abuse from subcontractors were very common, especially in waras in the prawn-shelling sector. Homework affected children’s education. This is reflected in the low percentage of children enrolled in schools (20 per cent) which is half of non-homework household children (42.6 per cent). The CG was better off as judged by household facilities and assets, higher investment in health, higher investment in education and better nutrition levels.

238

Rafi Khan et al.

The earnings from homework did not empower women when judged in terms of mobility, greater say in household decision making or independence in spending earned income. Recommendations The welfare of homeworkers can be addressed as part of a government’s anti-poverty strategy. However, in order to be able to deal effectively with the problem, the government will need to know the magnitude of home-based work. In this regard, the FBS will need to work with a definition of the informal sector that also includes homework. This would be a move away from the current unit or enterprise orientation. The Household Income and Expenditure Survey or the Pakistan Integrated Household Survey, rather than the Labour Force Surveys, are the appropriate surveys to gather such information. Alternatively, the FBS could rely on registration boards. Once the magnitude of homework is ascertained, it would be possible to find appropriate solutions. Pakistan, as a signatory to the ILO Labour Statistics Convention, is obliged to collect, compile, and publish such labour statistics. Additional considerations for the government are the efficient utilisation of existing human resources given that Pakistan is in the process of downsizing the public sector. An alternative could be to engage the surplus staff in local government for the delivery of services at the grassroots level such as the homeworkers’ neighbourhoods. Regarding discrimination against women, there are two important issues to keep in mind. First, home visits by women government servants may be necessary in communities that observe strict purdah. Cultural change leading to more fair and humane treatment of women by men, such as was the case, to some extent, in the Gujrati community, is a slow process and often accompanies economic growth. However, in view of the extreme oppression of women, particularly of those sold into homework, the full force of existing legal protection should be available to them via state supported legal counselling. In the short run this could be complemented by civil society organisations. Second, since many of the communities undertaking homework are ‘illegal’, they would be unlikely to want to register. Thus, the state would have to agree to blanket ‘amnesty’ policies so that individuals living in the country would acquire rights as humans. The ‘illegal status’ they currently live in results in exploitation by contractors and the police. This sense of insecurity also seems to push them into extreme forms of religious practice to gain social acceptability. The following section provides specific policies aimed at improving homeworkers’ current conditions: Credit One component of the current poverty alleviation strategy is the provision of micro-credit via the Khushali (Well-being) Bank. Credit provision should be flexible about what enhances well-being. In view of the debt-induced

Hazardous subcontracted homework in Pakistan 239 oppression of homeworkers, a good use of the funds would be to buy out the debt of bonded labour and other workers indebted to contractors, and offer them loans on easy terms. Establishment of registration boards Provincial or district governments should set up registration boards to register all labour, contractors and employees engaged in homework. Besides being a source of information for the number and size of homework sectors, these boards could function as welfare funds. Welfare funds Our research has indicated that homeworkers get an extremely small percentage of the total value generated by their work. Interfering in the market place is dangerous and often results in negative anticipated outcomes. However, in view of the large margin, a levy should be imposed on the owners, collected by the relevant local government institution, and utilised to establish a welfare fund for each sector to finance several activities.74 The federal, provincial or district governments could provide matching grants for this fund. These funds could finance the following kind of activities: ● ● ●

● ● ●

Safety equipment. Safety training courses for workers (attendance mandatory for those registering). Courses on labour legislation to raise awareness among owners and contractors (attendance should be mandatory). Training courses for homeworkers. Provide schooling and health facilities. Legal counselling, particularly for women.

Promoting the above stated objectives would also be in line with Pakistan’s international commitments, since it is a signatory to the ILO Conventions on social security, worst form of child labour and equal remuneration. There is some evidence that educated women dealt with the contractors and the world with more confidence and commanded more respect, and hence there is enhanced importance one can attribute to the social investment in girl’s education. Collective action Both the quantitative and qualitative information showed that there is virtually no organisation working with women to assist them to better their pay and working conditions although they did some collective action in order to obtain better prices. Over half the women indicated a willingness to be part of some kind of collective action to improve their working conditions, if some kind of leadership and assistance were provided. Several factors like excess supply of labour and fear of job loss, lack of information and awareness, lack of mobility and relationship with the contractor hindered collective action. There is a role for civil society organisations like Pakistan Institute for Labor Economics Research (PILER) and the Aurat Foundation to mobilise workers. Since power is so unequal in the market equation, and since the margin at the higher ends of the production chain is so high, some redress is called for, initially in the form of district government mandated rates until collective bargaining is established. Over time, efforts of civil society organisations and information

240

Rafi Khan et al.

made available by registration boards may lead to the creation of workers associations that would enable them to secure decent rates.

Annex 7.1: Description of homework environments in fieldwork reports Incense stick making (agarbathi) In a dark and barely lit room, women and children of the family sit toiling in the sweltering heat. There is barely enough breathing space, since more often than not, it is the only room in the house. Thick agarbathi dust carpets the floor. A huge bundle of sticks lies on one side and a large bowl filled with thick black tar like paste (which includes the sawdust and toxic chemicals) sits in the middle of the floor. One can feel congested as heaviness that hangs in the air due to the heat, dust and humidity. On one dirty piece of cloth is piled the sawdust, while an infant lies crying next to it, since the mother prefers keeping him close as she works hard to meet the target set by the contractor. The woman making the agarbathi sits crouched on the floor with her hands covered in the black paste up to her knuckles. She takes a stick, puts the paste on it with her hands, and then rolls the agarbathi in the pink dust until it clings tightly to the paste. She then puts it on one side to let it dry. She repeats the same process, sitting in the same position for hours at a stretch.

Carpet weaving A khadder (loom) for carpet weaving is placed in one corner of the courtyard, and it takes up most of the space there. Sitting behind it, are the women and children of the household. They sit upright, with their backs stiff, and their hands move with amazing speed on the taut strings and threads of the khuddee. With great concentration, they weave the thread around the strings, giving life to a pattern built-in by the contractor, and then, with a sharp sickle like instrument, they push it down. The women’s finger joints are hard and swollen, betraying years of hard work that they have been doing at the khuddee. Their eyes are vacant, and their expressions completely empty as they speak from behind the khuddee, still weaving and not stopping for a second.

Prawn shelling In one small corner of the courtyard sit the women and young girls of the house with prawns scattered all around them. They sit crouched in a pool of freezing and muddy water. The ice filled water gets splashed around while they take prawns out from overflowing bowls. The ice is meant to ensure the prawns do not rot. They take the prawns out of the basket, one at a time, they put them in a tub of water, take them out, shell them and then again put them in water in another tub

Hazardous subcontracted homework in Pakistan 241 to preserve them. The whole house has a sickening stink, for an outsider, as the day’s work goes on. Scales and shells of prawn cover their hands and cause skin irritation from constant scratching. The cold water also freezes the hands. Sack stitching (bori) Sitting in a relaxed manner, the women who stitch sacks get together and indulge in friendly conversation as they work. They laugh and joke as they measure and stitch the boris. Their work is tedious and tiring. However, they try to make it less onerous by getting together and doing it jointly. Thus work has the elements of a social gathering. They scratch their hands often, since the fibres from the bori cause irritation and often they let out a cough and complain of flu due to fibre inhalation. Their rooms, however, are brightly lit and airy and they work in an organised manner. They pile up the boris in neat stacks as they work so that the house does not get messy. Often, the men join them to stitch a bori or two and, if nothing else, then just to share the day’s happenings with them.

Annex 7.2: Case studies Case study 1 Sector: carpet weaving Location: Korangi Haseena was 20 years old, a mother of two and lived in Korangi. Her husband was unemployed and uneducated. Carpet weaving was the sole means of income that they had for making ends meet. Haseena, her husband and her six-year-old daughter wove a carpet all day. She did not have any extended family to turn to and owed about Rs 6,000 rupees to the contractor. Despite weaving all day, she had not been able to pay off her debt for the past two years. The contractor visited her house after almost a week and, after counting the phairas (rounds) completed, gave her half the money and deducted the remaining half from her debt. Although two years had gone by, her debt remained virtually unchanged from the day she took it because of the interest. The contractor not only abused her verbally, but, time and again, he beat her little daughter. Once, he came to demand the money that she owed him and, while arguing, he started beating her up. Her husband tried to stop him, but the contractor dragged him out of the house and beat him up too. They did not report the incident to the police, since they do not trust the police. They had no idea who they could turn to for help or how to pay off the contractor’s debt. Haseena’s little daughter had swollen finger joints because of carpet weaving. She also complained of fatigue and dizziness. Haseena herself had skin infection with blisters all over her legs. She too complained of dizziness and irritation in the skin. Her husband suffered badly from a chronic cough and breathing problems. They did not get any medical assistance from the contractor. Haseena’s case

242

Rafi Khan et al.

is one of many. Contractors take undue advantage of these families because they are indebted. Lives go to waste trying to pay off an un-payable debt. Case study 2 Sector: sack stitching Location: Godhra Parveen was married at 13 to Habib, a contractor in a towel factory. She mentioned that she was kidnapped at 12 from her house in Bangladesh and was brought to Pakistan by a dallal, along with other girls. The dallal, herself a woman, sold Parveen for rupees Rs 18,000 to a Bengali family. She stayed with this family as a domestic servant for one year. After that, she got married to Habib, who already had married three times before, once in Bangladesh and twice in Pakistan. The two wives in Pakistan had divorced him, but he still sent money to his wife in Bangladesh every month. Her husband abused her physically and her neighbours know about their domestic fights. Since she was alone in Pakistan, he took advantage of this and beat her at will. She often escaped to the neighbours’ house, but returned to her own house because of her children. She even refused the interview money, because she feared that her husband might become suspicious. He never gave her money, and she did not know how much his income was. Parveen stitched sacks and earned Rs 30 per 100 sacks. She felt that it was not a very hard job, but was hazardous due to the jute fibres. Sewing sacks enabled her to supplement the family income. Case study 3 Sector: carpet weaving Location: Orangi Hasina, a 22-year-old Bengali woman living in Orangi, had a different attitude from the rest of the women living in that locality. She had matriculated from school and her husband was also earning a decent living. Besides that, she wove carpets at home to support her family. Hasina also did volunteer work at the nearby clinic as a nurse and as a tutor for the neighbourhood children. Her living standard was much better than the people living around her, and it was the outcome of the couple’s hard work and planning. Their only child was very well mannered and clean, unlike most of the children observed during the fieldwork. She was the only woman we met in that locality who practised birth control. Hasina and her husband were of the opinion that they will have their second child only when they can provide the child enough time, attention and with the necessities of life. Unlike most of the other people interviewed, she did not complain about the behaviour of the contractor. She thought the contractor always treated her with respect and gave her wages on time. However, she also noted her health problem due to homework. She was very open on most issues and criticised men very openly in the focus group discussion. According to her, women were also responsible for their misery, since they do not make enough effort to bring about

Hazardous subcontracted homework in Pakistan 243 a change. Hasina said that they themselves had given their husbands the freedom to ill-treat them. It was encouraging to meet a woman like her in that locality where religious extremism seems to determine every norm of life. Her different attitude towards life was probably the result of several factors including her family’s education, her exposure to the outside world and a supportive husband.

Notes 1 Government of Pakistan, 1998b, p. 3. 2 Similarly, the leading piece by Ahmed Viqar (1993b, pp. 3–4), in another volume edited by Ghayur (Ahmed 1993a), is influenced by an enterprise orientation due to data constraints, but the limitations of such a definition are acknowledged. The more promising sectoral approach of Hussein et al. (1991) is premised around defining what constitutes formal and informal in various sectors like finance, housing, manufacturing, education, health, housing and transport and then identifying the informal activities by sector. Ahmed Nigar (1993, pp. 63–83) is notable among other scholars like Shaheen and Mumtaz (n.d.) for bringing home-based piece-rate women workers into the ambit of the informal sector. However, she concedes that, given the ephemeral nature of such work, it is understandable that the Federal Bureau of Statistics is not documenting it (pp. 68–69). 3 Clearly a very literal definition of informal is being used as indicated by the inclusion of legislators, senior officers and managers in the informal sector. 4 Of those urban female workers who are employed (whether in the formal or informal economy), the Labour Force Survey data (Table 7.2) suggests that barely 1.14 per cent are described as unpaid family helpers. Of the rural women who are employed, most (9.42) are unpaid family helpers, they are presumably working on family farms. 5 240 from Faisalabad, 175 from Sialkot, 216 from Lahore and 190 from Multan. 6 This study was part of a broader cross-country set funded by The Asia Foundation. Other studies in this set are by Boonmathya et al. (1999) for Thailand, Jayaweera et al. (1999) for Sri Lanka, Joseph et al. (1999) for the Philippines and Unni and SEWA (1999) for India. 7 Mines Act 4 of 1923, The Merchant Shipping Act 1923, Factory Act of 1934 and amended in 1977, the Employment of Children Act, No. 26 of 1938 and 1991, the Road Transport Workers Ordinance, No. 28 of 1961, the Bonded Labour System (Abolition) Act of 1992. 8 As the industrial centre of Pakistan, it is important not only because it has the highest concentration of labour, but also because it has contributed to the evolution of the labour movement in Pakistan. Given the complex social and political backdrop of Karachi, it best represents the issues that Pakistan’s labour faces today. In addition, contacts from the Pakistan Institute of Labour and Education Research (PILER) facilitated the fieldwork. 9 The format for the qualitative field reports included an emphasis on the key themes of the study including health/hazard issues, women/child empowerment issues, exploitation/ poverty issues, economic/value chain issues and other insights. 10 Question W4 ‘Do you go outside the house on your own when you need to? And W5: “If Yes, do you need permission from your brother, parents, husband?” ’ 11 Examples of questions include W6: ‘Who keeps the earnings from homework (%)?’ and the optional answers of ‘husband/guardian, retained and others.’ Question W7: ‘Do you save money through the bachat committee (BC) system?’ W10 ‘How much income do you keep for yourself ?’. 12 W19: ‘Do you have a saying in decisions regarding: schooling, marriage, family matters?’ 13 Examples of questions are W78: ‘How much milk do you consume?’ and the optional answers of: never, one cup/glass, two cup/glass, more than three cup/glass, and W38: ‘What did you have yesterday for breakfast?’, with the optional answers of ‘roti, pratha, tea, milk, left over, others’.

244

Rafi Khan et al.

14 W76, ‘Do you complain of homework-related health problems?’ Optional answers consisted of a list of ailments including eye infection, ear infection, cough, asthma. 15 Question C22 and C23: ‘Has it ever happened that you did not perform your work well?’ ‘If Yes, what were the consequences?’ The optional answers were, ‘none, scolded by family, scolded by contractor, beaten by family, beaten by contractor, not allowed to play, not allowed to go to school, others’. 16 The questions included: ‘Nature of the house’ and ‘How would you evaluate the workspace in home for each of the following categories: space, light, dust, air, smell, temperature, noise.’ 17 Questions included were S1: ‘What are the main reasons for recruiting homeworkers?’ The optional answers were: ‘Possibility to recruit from a much larger area; possibility to hire workers in accordance with variations in demand; minimization of the risk of unionization; no labour problems; can pay labour less to beat the competition; no problems with government regulations as in the formal sector; freedom to vary the volume of the production; greater opportunity to vary the nature of the work; reduction of costs since most costs are born by household; greater flexibility in responding to the fluctuations and irregularity of the market; other.’ 18 Exploitation is used here in a loose sense including verbal and physical abuse, delayed payments and poor rates rather than in the Marxian sense of the extraction of surplus value. Of course, exploitation in the Marxian sense is also likely to accompany exploitation based on unequal power as defined above. 19 According to the 1998 census data, Karachi’s population was 9,802,134 persons. However, Sindh’s political parties contest this figure as being underestimated. There is broader consensus that Karachi’s population is over 12 million (Hasan, 1999, p. 40). 20 The 1947 Partition of India led to a vast influx of refugees from India that settled in the heart of the city and later on shifted to some of the new settlements and colonies that were constructed for them at the government’s behest. Pathan migrant workers came in the 1950s and 1960s under Ayub Khan’s policy of bringing in these workers to work in the industries of Karachi. They were also given the transport business as part of state patronage. In the 1960s Punjabi migrant workers also found their way into Karachi. Baluchi migrant workers had also been present in Karachi due to its proximity with their province. Sindhis, who were in a majority in the past, became a minority in their own provincial capital. Karachi’s ethnic mix thus represents inter- as well as intra-provincial politics in Pakistan. 21 The term Muhajir literally means immigrants. The bulk of this community immigrated at the time of the partition of the sub-continent in 1947. This community was a minority at the time of partition, but also the most educated and advanced, it shouldered the bulk of the administrative burden of the city, province and country via key positions in the civil service. As the other ethnic groups claimed their share of positions, the muhajirs felt sidelined and the rapid emergence of the MQM in the 1980s gave expression to this feeling of exclusion. 22 Rolling of agarbathi consists of mixing a chemical paste and cooking it on the fire using a large pot. Thin wooden sticks are dipped in the paste and then rolled on the sawdust. 23 If this was the case, subcontractors imposed fines on homeworkers for the faulty work or missing units. 24 The exchange rate of one rupee (Rs) for the dollar in early 2001, the time of the fieldwork, was Rs 62.15 for $1. 25 According to the subcontractors, the work had decreased due to excessive taxation (source: field report, SDPI). 26 Source: http://www.finance.gov.pk/summary/main.htm 27 For more details, see http://www.anti-slavery.org/global/india/ and Von Moltke et al. (1998). 28 Carpets from India, China and Pakistan account for 81 per cent of total US imports. 29 The domestic market for carpets is very limited.

Hazardous subcontracted homework in Pakistan 245 30 Korangi lacked basic public services including solid waste disposal, garbage dumps, electricity connections and health services. 31 Cost (Rs) Thread 7 lbs @ Rs 115/lbs 805 Cone 14 Lachi 125 Tani 150 Labour (homework wage) 3,500 Kanni kichai (cutting) 40 Washing 600 Transportation 200 Total cost 5,434 32 Stitching consists of pulling chords. 33 Workers spend 1–3 minutes in the production of each bag, and work approximately 6.7 hours per day. 34 Prawns ( jeengha) have a hard cover that conceals the edible flesh. 35 Prawn process has the following cost structure for the production of 2,640 kg. Cost (Rs) Raw prawn @ Rs 60/kg 158,400 Dalali (commission) @ Rs 2/kg 5,280 Tax 18,480 Mazdoori (wage for workers in fishery) 200 Transportation 800 Wage of agents at wara 900 Amount paid to workers 6,600 Total cost 190,660 36 The reason why tokens include contractor’s name is that homeworkers received tokens from different subcontractors. 37 On average, there was one (0.98) non-homework earner in a household in which homework was being done. The mean earning members of the household in the non-homework households were 1.16. 38 Only 1 per cent of the responses suggested that the women were doing this to earn spending money for themselves. 39 Police harassed them and solicited bribes on a monthly basis. However, the field team observed many men of this community sitting outside the houses or at the teashops (source: SDPI field report 2001; Khattak 2001). 40 Many women appreciated husbands who did not impose such work on their families (source: field report 2001; Jalal 1991). 41 ‘Women were unwilling to try to get work outside the house and believed homework to be meant for them and not for men’ (source: SDPI field report 2001). 42 Five per cent responded that it was a joint decision. Thus in only 11.5 per cent of the cases was this decision that of the husband’s or the family’s, without taking the woman’s views into account. 43 In these cases women were contacted through children and other male household members. 44 Bengali women were brought from Bangladesh and sold to Sindhi and Punjabi men who mistreated them, compelled them to work and physically abused them. See case study on sack stitching, Annex 7.2. 45 Some Bengali women mentioned that Punjabi men are known for being quite ‘liberal’ with their own wives and imposed less restrictions on women’s mobility than Bengali men, and are more likely to allow girls to attend school (source: FGD, SDPI 2001). 46 Greater mobility was only relative. Even in this community, working outside the house by women was frowned upon and an important part of the reason why they opted for homework.

246

Rafi Khan et al.

47 In most cases (53 per cent), the delay was up to two weeks, and in another two-fifths of the cases, it was more than two weeks but less than a month. 48 FGDs: Women felt that subcontractors cheated them by giving less thread (sutli). 49 Thirty-one per cent felt it was indifferent and only 23 per cent thought it was bad (source: UNICEF, SDPI survey 2001). 50 The variation between prawn shelling households at 8.41 and carpet weaving households at 6.78 is considerable, though both are very high compared to the national average or the CG of households. 51 Women again accepted this as a religious right of men, although many scholars point to various qualifications with regards to this institution. These include exceptional wartime conditions in which this practice was permitted, the ability to treat all partners equally and the ability to support them well. 52 Madrasas are religious schools were children are taught the noorani quaida (basic religious textbook) and Urdu textbooks. 53 About 3 per cent of the responses suggested that the child might have found an alternative way to supplement family income if other options were available. 54 On average, children played about one hour a day in the incense stick sector. 55 Children collected the materials from the contractor and started stitching after returning from school. The women joined them once the household chores were completed. Children were allowed to take breaks between work, although they worked long hours. 56 Sitting for two-hour stints caused the children’s feet to hurt, and the field team observed nosebleeds. Long hours at work for women resulted in body, muscular, joint, back and shoulder pains (source: SDPI field report 2001). 57 In the incense stick sector, there was a government institute called the Ojha Institute of Chest Diseases located in the area. The field team visited this institute twice but on neither occasion were they able to meet the doctors. 58 The practice of local remedies was common among the Agarbathi sector workers (source: field report, SDPI 2001). 59 The average household size in urban Sindh in the 1998 Population Census was 5.8, so that homework households are larger than the reference group (Government of Pakistan, Economic Survey, Statistical Appendix, 2001a, p. 132). 60 The conversion factor for inflating the cut-off poverty line was computed using the general consumer price index, Government of Pakistan (2001a, p. 79). 61 Government of Pakistan (1998c; 1998d; 1998e, p. 63) indicates anaemia among urban 5–14 year old girls and boys to be 40 and 32.5 per cent respectively. 62 Anaemia could have been even higher but for the high fish diet among most of the communities surveyed. 63 Government of Pakistan,1998a, p. 47. 64 We utilised a logit analysis to rigorously identify the predictors of homework. While these variables are highly significant and the model is successful in predicting who is likely to be engaged in homework, the overall fit of the model, as judged by the pseudo R bar squares, is low. 65 The F-statistic is 3.65 and it is significant at the 6 per cent level of probability. In both cases, the number of non-resident relatives contributing to household income is negligible (Annex 7.2, case study 3). 66 Government of Pakistan, Economic Survey, Statistical Bulletin (2000b, p. 144; 2000a). 67 The generic form this takes is that a group of individuals, with mutual trust, agrees to pool an equal amount each month and the lump sum accrues via rotation to each member of the group until all get the pooled funds once. 68 CG women spent nearly five hours on domestic chores (source: UNICEF; SDPI survey (2001)). 69 Conversely, 59.7 per cent of the women in the CG got the needed treatment and were released immediately, while this was the response for only 26.3 per cent of the

Hazardous subcontracted homework in Pakistan 247

70

71 72 73

74

women doing homework. About one-tenth of the men did homework for an average of two and a half hours per day. Two-thirds of them complained of health problems related to this work. Again, unlike for adults, the average training period lasted 8.2 weeks. The high standard deviation of 19.3 weeks suggests that many children are viewed as trainees for a long time. The frequency distribution showed that 84.4 per cent of the children received training for four weeks or less, but 6.6 per cent of the children were considered trainees for a period between 52 and 97 weeks. Among the children in the CG, 81 per cent of the boys’ and 85 per cent of the girls’ expressed a desire to attend school full time for those children not in school. Of those who did not work the whole year, 87 per cent said this was because work was not available and 4 per cent said they did other work. For children doing homework, based on F-tests, it is quite clear that boys spend significantly less time doing homework and chores than girls and significantly more on play and schoolwork. This gender pattern is also evident in the CG, although these children play more and spend less time doing chores. We explored the nature of chores further and report the results based on their responses. A similar fund has been successfully set up by the bidi homework sector in India. This activity falls in the domain of the newly created local government institutions.

Bibliography Ahmed, Nigar (1993b), ‘Women Home Based Piece Rate Workers: A Study’, in Ghayur, S. (ed.), ‘The Informal Sector in Pakistan: Problems and Policies’, The Informal Sector Study No. 3, Friedrich Ebert Shiftung, Islamabad. Ahmed, Nigar, Qaisrani, S. and Tahir, M. (1998), ‘Social Protection for Women Workers in the Informal Home-Based Sector in the Leather and Textile Industries’, Aurat Foundation, Lahore. Ahmed, Viqar (1993a), ‘The Informal Sector: Needs and Options’, in Ghayur, S. (ed.), ‘The Informal Sector in Pakistan: Problems and Policies’, The Informal Sector Study No. 3, Friedrich Ebert Shiftung, Islamabad. Awan, S. A. and Khan, A. (1992), ‘Child Labour in the Carpet Weaving Industry in the Punjab’, UNICEF, Lahore. Boonmathya, R., Praparpun, Y. and Leechanavanichpan, R. (1999), draft report on ‘The Situation of Women Subcontracted Workers in the Garment Industry in Bangkok Thailand’, mimeo. Data Source: Pakistan Country Study Survey (2001). Government of Pakistan (1998a), Labour Force Survey 1996–97 and 1999–2000, Federal Bureau of Statistics, Statistics Division, Islamabad and also other years. Government of Pakistan (1998b), Resolution of the 15th International Conference of Labor Statisticians. Government of Pakistan (1998c), National Health Survey of Pakistan 1990–94, Pakistan Medical Research Council, Islamabad. Government of Pakistan (1998d), Household Integrated Economic Survey 1996–97, Federal Bureau of Statistics, Statistics Division, Islamabad. Government of Pakistan (1998e), Compendium on Gender Statistics, Federal Bureau of Statistics, Islamabad. Government of Pakistan (2000a), Annual Report, State Bank of Pakistan, Islamabad. Government of Pakistan (2000b), Economic Survey, Statistical Bulletin. Government of Pakistan (2001a), Economic Survey, Statistical Appendix.

248

Rafi Khan et al.

Government of Pakistan (2001b), ‘Poverty in the 1990s’ Draft, Federal Bureau of Statistics, Islamabad. Government of Pakistan (2001c), Pakistan’s Economic Performance during July–Jan./Feb. 2000–01, Finance Division, Economic Advisor’s Wing, Islamabad. Government of Pakistan (2001d and previous years), Economic Survey 1999–2000, 2000–2001, Economic Advisors Wing, Finance Division, Islamabad. Hasan, A. (1997), ‘The Growth of a Metropolis’, in Khuhro, Hamida and Mooraj, Anwer (eds), Karachi. Megacity of Our Times, Karachi: Oxford University Press. Hasan, A. (1999), Understanding Karachi. Planning and Reform for the Future, Karachi: City Press. Hisam, Z. (2000), ‘Karachi 2000: Life at the Fringe’, The News International, 21 May 2000. http://ilolex.ilo.ch:1567/scripts/convde.pl?queryC157&query0
157 Human Rights Commission of Pakistan (1999), State of Human Rights in 1999, Lahore: HRCP. Hussein, M. et al. (1991), ‘Pakistan: Informal Sector Study’, Submitted to USAID/ Pakistan, Development Research and Management Services (Pvt.) Limited, Islamabad. ILO (1982), Maintenance of Social Security Rights Convention, ILO, Geneva. ILO (1985), Labour Statistics Convention, http://ilolex.ilo.ch:1567/scripts/convde.pl? queryC160&query0
160 ILO (1996), Home Work Convention, http://ilolex.ilo.ch: 1567/scripts/convde.pl? queryC177&query0
177 Jalal, A. (1991) ‘The Convenience of Subservience’, in Kandyoti, D. (ed.), Women, Islam and the State, London: MacMillan. Jayaweera, S., Sanmugam, T. and Rodrigo, C. (1999), Draft report on ‘Women in Subcontracting Industries – Sri Lanka’, Colombo, mimeo. Joseph, L. Y., Ofreneo, R. P. and Gula, L. (1999), draft report on ‘Subcontracted Women Workers in the Context of the Global Economy: Philippine’, mimeo. Kemal, A. R. and Mahmood, Z. (1993), ‘Labour Absorption in the Informal Sector and Economic Growth in Pakistan’, Friedrich Ebert Shiftung, Islamabad. Khan, S. R., Khattak, S. G. and Kazmi, S. (2001), Pakistan Country Study, ‘Hazardous Home based Subcontracted Work in Pakistan’ (unpublished paper). Khattak, S. G. (2001), ‘Subcontracted Work and Gender Relations: The Case of Pakistan’, in R. Balakrishnan (ed.), The Hidden Assembly Line: Gender Dynamics of Subcontracted Workers in a Global Economy, New York: Kumarian Press. Khattak, S. G. and Sayeed, A. (2000), ‘Subcontracted Women Workers in the World Economy: The Case of Pakistan’, SDPI Monograph Series #15, Islamabad. Parveen, F. and Ali, K. (1996), ‘Research in Action: Organizing Women Factory Workers in Pakistan’, in Chhachhi, Amrita and Pittin, Renee (eds), Confronting State, Capital and Patriarchy, the Hague, the Netherlands: Macmillan Press Ltd. and St. Martin’s Press, Inc in association with the Institute of Social Studies (ISS). Sayeed, A. and Ali, K. (2000), ‘Labour Market Policies and Institutions: A Framework for Social Dialogue’, Pakistan Institute of Labour Education and Research (PILER) Research Report No. 3, Karachi. SCF (Save the Children Fund) (1997), ‘Stitching Footballs, Voices of Children’, Islamabad, mimeo. SDPI (2001), Field Report. Shaheen, F. and Mumtaz, K. (n.d.) ‘Invisible Workers: Piecework Labour Among Women in Lahore’, Women’s Division, Government of Pakistan.

Hazardous subcontracted homework in Pakistan 249 SPDC (Social Policy and Development Center) (2001), Social Development in Pakistan, Annual Review 2000, Towards Poverty Reduction, Karachi: Oxford University Press. UNICEF/Government of Pakistan, 1990, ‘Discover the Working Child’, National Commission for Child Welfare and Development, Islamabad. Unni, J. and SEWA (Self Employed Women’s Association) (1999), draft report on ‘Subcontracted Women Workers in the Global Economy: Case of Garment Industry in India’, mimeo. Von Moltke, K., Kuik, O., Van Der Grisp, N., Salazar, C., Banuri, T., Mupimpila, C., Inman, L., Mesa Rolfas, N. and Jose De Los Santos, J. (1998), Global Product Chains: Northern Consumers, Southern Producers, and Sustainability, Environment and Trade No. 15, UNEP, Geneva.

8

Women and children homeworkers in Indonesia M. Oey-Gardiner, E. Suleeman, I. Tjandraningsih, W. Hartanto and H. Wijaya

There is limited knowledge and awareness about the contribution of homeworkers in Indonesia. Despite the various efforts1 to collect information about this segment of the informal sector, homeworkers remain invisible in the Indonesian labour force statistics (Korns, 1993a,b) and consequently in the design of policy. Their cause has not gone entirely unnoticed in Indonesia. For eight years (1988–96) the International Labour Organisation (ILO), with Danish International Development Agency funding, conducted a campaign for social protection of homeworkers (ILO 1993a,b). The project aimed at promoting social protection and raising awareness about the limitations of homework including the lack of access and control over the means of production and resources (BerarAwad, 1993). In addition, the Ministry of Manpower set up a monitoring scheme for the promotion of social protection of rural women.2 However, during the political, economic and social turmoil that Indonesia experienced in the mid-1990s including the East Asian financial crisis, the extended drought caused by El Nino, the considerable devaluation of the Rupiah (80 per cent) and the fall of the Suharto regime, the projects mentioned were not implemented and attention to the issue of homeworkers flagged. Since then the government has focused on poverty reduction strategies. Although some homeworkers have been beneficiaries of these programmes, understanding about homeworkers’ conditions and main needs is still limited. This chapter is based on the findings of extensive fieldwork conducted in Indonesia in Central and West Java in three sectors, batik, rattan furniture and pottery. The goal is to understand the dynamics of homework within the wider context of the labour market and the informal sector. Focusing on these three purposively identified sub-sectors, the study explores the relationship between homeworkers and subcontractors as well as the roles of women and children in homework (Oey-Gardiner et al., 2001). This chapter is divided into five sections. Section 8.1 discusses the structural changes in the labour force during the 1990s. By analysing this period, we will be able to highlight the ongoing trends and the effects of the recent shocks on both the population and labour force, especially the formal and the informal sectors. It also presents briefly the main findings about the conditions and characteristics of homeworkers in Indonesia based on existing literature. Section 8.2 describes the

Women and children homeworkers in Indonesia 251 methodology used in the field study and a brief description of the survey questionnaire. Section 8.3 introduces the three sectors studied by describing the geographic environment, production process, and the terms of agreement and nature of the relationship between homeworkers and subcontractors. This section includes the general characteristics of workers in each sector. In section 8.4, the longest, we discuss the results of the survey on the main conditions of homeworkers, comparing them with households from the control group (CG). Finally, section 8.5 summarises the findings and provides detailed recommendations for policy makers.

8.1

Formal and informal in the Indonesian labour force

Before turning to homeworkers it is important to understand the context of the labour market. The following analysis focuses on the 1990s, a period in which the Indonesian economy experienced major structural changes. This discussion is based on the National Labour Force Surveys (Sakernas series).3 During the last decade the Indonesian labour force experienced various structural changes. In some respects, such as the share of the workforce in the total adult population, distribution of the workforce in main economic activities, and the degree of urbanisation, past trends were maintained despite the financial crisis in 1997. At the same time, the crisis provoked shifts in the composition of the informal sector, female participation and status in the workforce and level of unemployment (ILO, 1998). At the beginning of the decade, 65 per cent of the adult population (15 years old and above) belonged to the workforce. However, the percentage of workers in total population declined from 65 to 63 per cent between 1990 and 1999 (Table 8.1). This was mainly accounted for by the sharp rise in unemployment (from 2 to 6 between 1997 and 1999) after the economic crisis.4 Between 1990 and 1999, there were slight changes in the share of workers in each economic activity.5 More workers were concentrated in trade and services,6 while the share of workers employed in agriculture decreased (Table 8.2). The share of workers in agriculture among total workforce declined from 55 to 41 between 1990 and 1997. The first decline was accompanied with the increasing importance of trade and services. A subsequent increase in the share of agriculture occurred over 1998 and 1999. The reason was that agriculture became a cushion for those who lost their income earning opportunities as a result of the crisis. Formal and informal workers followed the pattern of the total workforce described above. Thus, by 1999 the majority of formal workers was concentrated in services (37 per cent), manufacturing (22 per cent) and agriculture (20 per cent) (Table 8.2). The share of formal workers in agriculture declined until the crisis. Prior to the crisis (1997), the percentage of formal workers in agriculture was the lowest with the concentration of 16 per cent of total formal workers. However, after the crisis it rose to 19 per cent. Similarly, the number of informal workers in

252

Oey-Gardiner et al.

Table 8.1 Indonesia: share of working adult population 15 years and over, workforce and unemployed, 1990–99 (%) 1990 1991 1992 1993 1994 1996 1997 1998 1999 Population (million) Workers (million) % working Unemployed (million) % unemployed of total workforce Formal workers (million) Informal workers (million) Total (%) % formal workers % informal workers Total (%) Informal sector Self-employed Unpaid family workers Formal sector % urban Formal sector Informal sector Self-employed Unpaid family workers

113.5 115.6 118.2 120.8 125.3 131.9 135.1 138.6 141.1 73.4 74.2 76.2 77.0 80.0 83.9 85.4 87.7 88.8 64.7 64.2 64.5 63.8 63.9 63.6 63.2 63.3 62.9 1.9 2.0 2.1 2.2 3.7 4.3 1.9 5.1 6.0 2.5 2.6 2.7 2.8 4.4 4.9 2.1 5.5 6.4 21.4

22.7

23.4

24.9

27.6

29.9

31.7

30.3

31.9

52.0

51.5

52.8

52.1

52.5

54.0

53.7

57.3

56.9

100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 29.2 30.6 30.8 32.3 34.4 35.6 37.2 34.6 35.9 70.8 69.4 69.2 67.7 65.6 64.4 62.8 65.4 64.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 44.5 26.3

44.2 25.2

44.4 24.8

44.5 23.2

44.2 21.4

46.9 17.5

44.3 18.5

45.9 19.5

45.7 18.3

29.2

30.6

30.8

32.3

34.4

35.6

37.2

34.6

36.0

47.2

49.4

49.5

49.9

49.8

52.7

52.9

54.2

54.9

19.1 9.1

20.8 10.2

21.8 10.7

22.4 11.2

23.2 12.4

24.7 13.3

26.9 15.1

27.7 15.9

29.3 17.7

Source: Based on special tabulations from the BPS Sakernas series. Note No Sakernas was conducted in 1995 because of the Supas (Intercensal Population Survey), which, regrettably, produced rather different results on many variables.

agriculture that had been systematically declining until 1997 suddenly started increasing after the crisis. Table 8.1 shows that in the 1990s the overall trend was that the share of the total workforce in the formal sectors increased from 29 per cent in 1990 to 36 per cent in 1999. However, there is a discrete break in that trend around the time the economic crisis broke in 1997 (ILO, 1998). There was a secular decline in the share of the workforce between 1990 and 1997 in the informal sector (and corresponding increase in the formal sector’s share, see Table 8.1). However, post-1997 the informal sectors’ share rises sharply; and in fact, the total number of workers in the informal sector, which had been relatively stable until 1996 despite the expanding workforce, rises suddenly post-crisis. Nevertheless, the number of formal sector workers registered an impressive increase of about 10.5 million between 1990 and 1999 – but then the size of the

Women and children homeworkers in Indonesia 253 Table 8.2 Distribution of formal and informal workers by sector of economic activity, 1990–99 (%) 1990

1991

1992

1993

1994

1996

1997

1998

1999

Total Agriculture 55.2 53.4 53.0 50.0 45.6 43.5 40.7 45.0 43.2 Manufacturing 10.2 10.4 10.5 11.1 13.2 12.6 12.9 11.3 13.0 Trade 14.8 15.1 15.1 15.9 17.1 18.9 19.9 19.2 19.7 Services 16.0 16.7 17.0 18.1 18.3 19.4 20.3 19.6 19.2 Others 3.8 4.4 4.3 4.9 5.7 5.6 6.2 5.0 4.9 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Number (million) 73.4 74.2 76.2 77.0 80.0 83.9 85.4 87.7 88.8 Formal Agriculture 23.2 21.1 21.3 20.5 18.1 17.3 16.3 18.7 20.1 Manufacturing 20.2 21.5 20.9 20.6 23.7 21.2 21.6 20.3 22.0 Trade 5.9 6.3 6.8 6.7 7.2 9.2 9.6 9.4 9.9 Services 40.7 40.4 40.8 41.2 38.4 39.9 39.3 40.6 37.3 Others 9.9 10.8 10.2 10.9 12.6 12.3 13.2 11.0 10.8 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Number (million) 21.4 22.7 23.4 24.9 27.6 29.9 31.7 30.3 31.9 Informal Agriculture 68.4 67.6 67.1 64.1 60.1 58.0 55.2 58.8 56.2 Manufacturing 6.0 5.5 6.0 6.5 7.7 7.8 7.8 6.6 7.9 Trade 18.4 19.0 18.9 20.4 22.4 24.2 25.9 24.3 25.3 Services 5.8 6.3 6.4 7.1 7.7 8.1 9.1 8.5 9.0 Others 1.3 1.6 1.7 2.0 2.1 1.9 2.0 1.7 1.6 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Number (million) 52.0 51.5 52.8 52.1 52.5 54.0 53.7 57.3 56.9 Source: Based on special tabulations from the BPS Sakernas series. Note Services consist of transportation, finance and social and public services; while others include mining and quarrying, utilities, construction and others.

workforce increased by over 15 million, so the numbers of those in the informal sector also grew by about 5 million. Another important trend was the rapid urbanisation of the population and workforce. In 1990, 31 per cent of the population and 25 of the workforce were in urban areas. By 1999, the percentage of the population rose to 42 and the percentage of the workforce in urban areas increased to 36. By looking at the growth figures, it seems that the crisis did not have a large effect on urban workers growth.7 Indonesia’s recent urbanisation occurred mostly because of pull factors8 of formal sector employment creation particularly in industry and the services in urban centres (World Bank, 1993). At the beginning of the decade, the percentage of formal workers in urban centres was 47 per cent. By 1999, the percentage increased to 55 per cent. The informal sector, both the ‘self-employed’9 and ‘unpaid family workers’,10 went through urbanisation as well. The percentage of urbanised self-employed increased from 19 to 29 per cent, while the percentage of urbanised unpaid family workers rose from 9 to 18 per cent

254

Oey-Gardiner et al.

between 1990 and 1999. The larger percentage increases for informal workers suggest that urban areas were absorbing a larger proportion of the migrant (rural–urban) labour force. On the other hand, certain structural changes that the labour force was going through at the beginning of the decade were interrupted by the crisis. These included levels of unemployment and the composition of the informal labour force, and the participation and status of female workers. The low and constant levels of unemployment were interrupted by the crisis. As depicted in Table 8.1, unemployment levels at the beginning of the decade were constant at under 3 per cent. Although there were increases during the middle of the decade,11 the more rapid rise occurred between 1997 and 1998 when the share of unemployed rose from 2 to 6 per cent (Table 8.1). Prior to the crisis, the informal sector experienced a contraction and a subsequent increase in relative and absolute terms. The reason is that in response to the crisis, some of the lower paid formal workers found refuge in the informal sector. In addition, the informal sector accommodated additional household members, such as women and children who were forced to enter the workforce for the household’s survival. This leads to another important aspect of the Indonesian informal workforce: the shift of status from unpaid family workers to self-employed workers. As depicted in Table 8.1, the percentage of unpaid workers of the total workforce declined from 26 per cent to 18 per cent between 1990 and 1999. On the contrary, the percentage of self-employed in the total workforce increased from 44 to 46 per cent during the same period. This suggests that the family’s ability to provide a safety net fell at the same time as household incomes declined. Although the overall workforce, formal and informal, has remained predominantly male12 (Table 8.3), there have been structural changes within the female workforce in the informal sector. According to data from the 1980s and early 1990s, the formal sector appeared to be feminising.13 This reading was felt consistent with a ‘birds-eye view’ of export-oriented and labour intensive industrial locations established in many urban fringes, where the majority of workers were female. This has led to suggestions of relatively ‘greater’ access among females to paid employment as employees in these formal sector enterprises. A closer examination of the data for the 1990s, however, does not support the earlier findings, as the share of females remained fairly constant at around 31–32 per cent (Table 8.3). The trends of the 1990s suggest that, while increasing numbers of women gained access to paid employment in the formal sector, the formal sector labour market for men was similarly growing at more or less the same pace as for women. Yet, women are not only absolutely but also relatively under-represented in the formal sector, as the share of females, at around 30–32 per cent, is lower than the share in the overall workforce of around 38–39 per cent. Similarly, the overall share of females in the informal sector too remained fairly constant at around 42 per cent. When considered separately, however, both the self-employed and unpaid family workers appear to be feminising. Among selfemployed workers the share of females rose from 26 to 30 per cent and among unpaid family workers from 69 to 71 per cent. The explanation is directly related to the mobility pattern for women.

Women and children homeworkers in Indonesia 255 Table 8.3 Share of female/male workers by sector, 1990–99 (%) 1990

1991

1992

1993

1994

1996

1997

1998

1999

Total workforce Female Male Formal sector Female Male Informal sector Female Male

100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 38.7 38.2 38.6 38.5 38.6 38.1 37.9 38.5 38.2 61.3 61.8 61.4 61.5 61.4 61.9 62.1 61.5 61.8 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 30.7 30.7 31.7 31.4 30.3 30.2 30.4 31.7 31.9 69.3 69.3 68.3 68.6 69.7 69.8 69.6 68.3 68.1 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 42.0 41.5 41.7 41.8 43.0 42.5 42.4 42.1 41.7 58.0 58.5 58.3 58.2 57.0 57.5 57.6 57.9 58.3

Informal workers: self-employed Female Male

100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 26.3 73.7

25.5 74.5

26.0 74.0

26.7 73.3

29.0 71.0

31.3 68.7

29.5 70.5

29.3 70.7

29.8 70.2

Informal workers: 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 unpaid family workers Female 68.7 69.4 69.8 70.8 71.8 72.8 73.2 72.2 71.4 Male 31.3 30.6 30.2 29.2 28.2 27.2 26.8 27.8 28.6 Male Formal sector Informal sector Self-employed Unpaid farm workers Total Number (m) Female Formal sector Informal sector Self-employed Unpaid farm workers Total Number (m)

33.0 67.1 53.6 13.5

34.3 65.8 53.3 12.5

34.2 65.8 53.6 12.2

36.0 64.0 53.0 11.0

39.1 60.8 51.0 9.8

40.2 4 1.7 59.8 58.3 52.1 50.3 7.7 8.0

38.4 61.5 52.7 8.8

39.6 60.4 51.9 8.5

100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 45.0 45.9 46.8 47.4 49.1 51.9 53.0 53.9 54.9 23.1 76.8 30.2 46.6

24.6 75.4 29.6 45.8

25.2 74.7 29.9 44.8

26.4 73.6 30.9 42.7

27.0 73.0 33.2 39.8

28.2 71.9 38.5 33.4

29.8 70.2 34.5 35.7

28.5 71.5 34.9 36.6

30.0 70.0 35.8 34.2

100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 28.5 28.3 29.4 29.6 30.9 32.0 32.4 33.8 33.9

Source: Based on special tabulations from the BPS Sakernas series.

The most important distinction is the shift of status of female workers from unpaid family to self-employed workers. Until the crisis, there was an increasing feminisation trend among unpaid family workers. The share of females among all unpaid family workers rose steadily from 69 to 73 per cent between 1990 and 1997, to decline thereafter to 71 per cent in 1999 (Table 8.3). The latter decline occurred due to the involvement in other types of informal activities to contribute to the household. As a result, female labour shifted in the 1990s from unpaid family workers to self-employed workers (the share of women in all self-employed workers rose from 26 to 30 per cent).

256

Oey-Gardiner et al.

The distribution of the male and female workforce between formal and informal workers, the latter consisting of unpaid family workers and selfemployed categories is presented in Table 8.3. Half the males are self-employed, but only a third of the women are self-employed. Nearly 40 per cent of the men are in the formal economy, but only 30 per cent of women are. Hardly any men are unpaid family workers, while a third of women are. These patterns are closely related to social perceptions of the role of women in the home and labour market held by men as well as women themselves. Women are considered secondary income earners and play a subservient and complementary role to their husbands. These social perceptions are also present in the minds of enumerators. Hence in a household consisting of a husband and wife where both are working in the same household enterprise, husbands will be categorised as ‘self-employed’ while wives will be categorised as ‘unpaid family workers’. As we analyse later in detail, data on homeworkers are available through official statistics. In 1999 the share of those with post-primary education was 40.4 per cent for the adult population, 36.6 per cent for workers and in particular 55.6 per cent for the formal workers and 25.8 for the informal workers.

The involvement of children (less than 15 years) in the labour force The following analysis relies on national labour force surveys, which collect data on individuals of the age group 10–14 years old (see Table 8.4). The reason for

Table 8.4 Numbers and distribution of child workers (10–14)

Number of child workers (in millions) Males Females Urban Rural Total Among child workers (%) Females Urban Java Formal Informal

1990

1991

1992

1993

1994

1996

1997

1998

1999

1.44 0.97 0.20 2.21 2.41

1.33 0.86 0.22 1.97 2.19

1.40 0.90 0.22 2.08 2.30

1.32 0.84 0.23 1.93 2.16

1.17 0.83 0.25 1.75 2.00

1.08 0.72 0.22 1.58 1.80

0.97 0.68 0.22 1.42 1.64

1.05 0.68 0.26 1.46 1.73

0.86 0.58 0.22 1.21 1.44

40.2 8.4 50.4 10.4 89.6

39.2 10.2 53.2 13.7 86.3

39.3 9.7 48.9 10.5 89.5

38.8 10.7 47.1 13.2 86.8

41.4 12.5 47.3 14.0 86.0

40.1 12.4 45.8 14.5 85.5

Source: Based on special tabulations from the BPS Sakernas series.

41.3 13.4 38.2 13.7 86.3

39.4 15.2 44.4 11.6 88.4

40.0 15.5 43.3 13.2 86.8

Women and children homeworkers in Indonesia 257 collecting data on 10–14 year olds is that in a peasant society, many children start helping their parents in gainful employment at early ages although the legal minimum age for work is 15 years.14 During the El Nino drought and the financial crisis between 1997 and 1998, child labour increased (from 1.6 million in 1997 to 1.7 million in 1998) especially among the poor who involved children as a household survival strategy. As a percentage of the age group the decline was from 11 to 7 per cent between 1990 and 1997; it rose to 8 per cent in 1998 and declined again to 7 per cent by 1999. The overall decline of children’s participation in the workforce is due to rising proportions of children attending school, from 83 to 86 per cent during 1990 and 1999. The increase in enrolment can be attributed to the universal schooling programmes implemented by the government since 1970. Thus in quantitative terms, the fate of Indonesian children improved during the last decade of the twentieth century. An important characteristic of child workers is the higher proportion of boys. The gender composition among child workers is similar among adults where approximately 40 per cent of workers are female. Most working children (85 per cent) live in rural areas although they are increasingly migrating to urban areas. Urban child workers rose from 200,000 to 220,000 or from 8 to 15 per cent of all child workers between 1990 and 1999. At the same time, the number of children in rural areas declined from 2.21 million to 1.21 million during the same period. This trend is closely related to the declining role of agriculture in overall labour absorption, as the workforce is generally transforming away from agriculture. If at the beginning of the decade three quarters of working children were helping in agriculture (75 per cent), by the end of the decade only two-thirds were still absorbed in this sector (66 per cent – Table 8.5). Increasingly as children leave agriculture, they join other workers in manufacturing (whose share of child workers increased to 15 per cent from 9 per cent), and trade. There was an increase from 10 to 13 per cent of children in trade and from 4 to 5 in services between 1990 and 1999. It must be mentioned that these statistics are likely to underestimate urban child workers, particularly urban street children15 and children working on fishing platforms, known as jermal on the eastern cost of North Sumatra (Mboy et al., 1997). In addition, official data does not take into account the cases of children assisting in child minding and housekeeping activities. As the crisis unfolded and the El Nino aftermath demanded more able bodies at the household level to contribute to household survival during the difficult months, and manufacturers too appear to have had to switch and rely on less costly labour, children became an obvious target. Hence the share of children absorbed in formal sector manufacturing rose from 27 to 30 per cent between 1997 and 1998 before resuming the pattern of decline initiated earlier. By 1999, of children working in the formal sector 25 per cent remained in manufacturing. The shares of children absorbed in formal sector trade rose from 6 to 12 per cent and in non-formal sector trade from 10 to 14 per cent respectively.

258

Oey-Gardiner et al.

Table 8.5 Distribution of child workers by economic activity and sector, 1990–99 (%) 1990

1991

1992

1993

1994

1996

1997

1998

1999

Total Agriculture 75.3 71.3 76.7 72.3 67.4 67.7 64.4 69.3 65.7 Manufacturing 9.3 10.2 9.4 10.6 12.6 11.2 12.5 11.1 14.6 Trade 9.5 10.4 8.8 10.3 12.2 14.5 16.3 13.5 13.3 Services 4.1 6.4 4.0 5.1 5.9 4.4 4.6 4.8 4.8 Others 1.7 1.7 1.0 1.6 1.9 2.1 2.2 1.3 1.6 Total % of child 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 workers Formal Agriculture 25.8 22.0 29.3 21.7 24.8 29.5 29.6 28.4 35.8 Manufacturing 32.0 33.9 37.2 36.4 35.9 27.9 27.1 29.8 25.1 Trade 6.1 7.0 6.2 7.5 8.8 11.4 13.0 10.7 11.7 Services 28.8 31.7 23.8 28.2 25.7 23.3 22.6 26.7 20.5 Others 7.2 5.4 3.6 6.2 4.8 7.9 7.7 4.3 6.9 % of formal child 10.4 13.7 10.5 13.2 14.0 14.5 13.7 11.6 13.2 workers Informal Agriculture 81.1 79.1 82.4 80.0 74.4 74.2 70.0 74.8 70.2 Manufacturing 6.7 6.5 6.2 6.7 8.8 8.4 10.2 8.7 13.0 Trade 9.9 11.0 9.2 10.8 12.7 15.0 16.8 13.8 13.6 Services 1.2 2.4 1.5 1.5 2.6 1.2 1.6 1.8 2.4 Others 1.1 1.1 0.7 0.9 1.4 1.1 1.4 0.9 0.8 % of non-formal 89.6 86.3 89.5 86.8 86.0 85.5 86.3 88.4 86.8 child workers Informal Self-employed 7.2 7.6 7.0 7.7 8.8 16.4 10.8 12.0 14.2 Unpaid family 82.4 78.7 82.5 79.1 77.3 69.1 75.5 76.4 72.6 workers Source: Based on special tabulations from the BPS Sakernas series.

Another important reality regarding child labour is their increasing participation in the formal sector. In 1990, 10 per cent of working children were in the formal sector. By 1999, the share of children rose to 13 per cent. Simultaneously, the share of child workers in the informal sector declined over time (from 90 to 87 per cent), yet they still account for the majority (Table 8.5). Thus child workers too increasingly opt to work on their own as self-employed (rising from 7 to 14 per cent of all working children) rather than as unpaid family workers (declined from 82 to 73 per cent of total working children) (Table 8.5). In fact, displacement of child workers occurred not only away from agriculture. Even though a good proportion of child workers, mostly in rural areas, have no other option beyond agriculture, they do shift. When possible, children abandon unpaid family worker (which declined from 82 to 73 per cent) in preference for ‘formal’ sector work as paid employees (which increased from 10 to 13 per cent) or as self-employed (which rose from 7 to 14 per cent).

Women and children homeworkers in Indonesia 259 Existing literature about homeworkers There is limited information and data about homeworkers. First of all they are not included in the national surveys and there is no clear definition about this type of worker. As mentioned above, the concept of informal sector used in official data subdivides informal workers into self-employed and unpaid family workers. The definition of the ILO Convention No. 177 of 1996 determines receipt of payment as one of the criteria to identify homeworkers. In the work status classification those receiving payment should technically be classified as either self-employed (with or without help) or as employees. Most studies have applied the concept of unpaid family workers to homework. In addition, a number of studies on homeworkers conducted in Indonesia have used further variations of the concept. For example, the study carried out by Moedjiman (1992) defined homeworkers as ‘carrying out part of manufacturing processes or industrial work at home and self employed’. The results of these earlier studies and activities emphasised not only the invisibility of homeworkers and the consequences thereof, but also the difficulty in defining homeworkers. According to studies, homework has existed in Indonesia for a long time. Hardjono (1990) states that the system has existed since 1928 in the textile industry. Other authors suggest that despite the lack of official statistics, homework is a significant phenomenon in the labour market. In a study which was part of the ILO-DANIDA project in Tasikmalaya district, homeworkers represent an estimated one-fifth of total workers.16 Moreover, homework is especially important for women as they constitute about two-thirds of all homeworkers, and one-third of the total of female workers (Suhaimi, 1995). The Central Bureau of Statistics (CBS) staff (Suhaimi, 1995), have demonstrated that homework is not limited to women only. It is claimed, however, that the typical homeworker is a married woman since social norms dictate that after marriage women should stay at home. Earlier studies (Save the Children Indonesia, 1996a) have related homework to poverty, long working hours and exploitation (Wijaya and Santoso, 1993a). For instance, according to Wirutomo (1992) homeworkers could not even afford the minimum basic needs with their small earnings. Literature on homework has often portrayed it as a magnet for child labour and as a source of exploitation depriving many of them of access to basic education and playtime with their peers. Unfortunately, in Indonesia, as in many other countries, the amount of information on the scale and magnitude of child homeworkers is limited. The few existing studies on homework have often failed to make any special distinction on child labour. For example, none of the studies carried out under the ILO-DANIDA Home-Work Project in Indonesia provided any special discussion on home-based child workers.17 At the same time, studies on child workers18 have tended to focus more on children’s involvement in hazardous occupations and have generally not made any clear distinction between child homeworkers and other child workers, including those working outside the home. The amount of information on detailed characteristics and the patterns of their involvement is limited and often of a more qualitative nature.

260

Oey-Gardiner et al.

8.2

Research method

Survey design and data collection The selection of the homework sub-sectors and location were guided by the following criteria: (a) Sectors in which production is being carried out by homeworkers. (b) Women and children are involved. (c) The production is for export markets. In order to identify the sectors that fulfil the above criteria, areas of concentration of cottage and small-scale enterprises were selected. The selected sub-sectors and districts19 included batik in Pekalongan located in Central Java, and rattan and pottery located respectively in Cirebon and Purwakarta in the province of West Java. Two approaches, quantitative and qualitative, were adopted in data collection of homeworkers in the batik, pottery and rattan sectors. The quantitative approach consisted of a survey administered to 70 households in each sector. In addition, 30 households per sector that were not involved in homework were surveyed to serve as the CG (Table 8.6). The qualitative approach consisted of focus group discussions (FGDs) with women and children and the gathering of information through three case studies in each sub-sector. Sample District statistical office staff members were requested to list about 250 eligible households, in 2–3 villages located in either one or both districts identified by the Central office. It was from this list that interviewers selected 70 households with homeworkers and 30 control households. The eligibility is defined as follows: 1

2 3

The household has to have a female homeworker in the selected sub-sector (we learnt that almost all homeworkers’ households has a woman doing homework at least part time). The work is carried out in the place of the worker’s choice but not at the place of the work provider. There is a child between the ages of 6–15 years.

Given the difficulties in identifying the list of households a ‘snowballing’ technique was implemented in which households were identified on the basis of information obtained from a household fulfilling the set criteria. The definitions used for data collection units in this study, that is homework households, include the following characteristics: (a) they are paid piece-rate; or they may also be unpaid family members; (b) they are subcontracted by a work provider, who may be an actual entrepreneur or capital owner, a middle-person, or a buyer; (c) they do not hire further paid workers.

Anjun and Babakansari Tegalwangi, Kaliwadas and Pesalakan Tratebang and Wiradesa Bener and Dedirejo Pekalongan Pekalongan

Tirto

Cirebon Cirebon

Purwakarta

Wiradesa

Weru Sumber

Plered

300

100

100

100

210

70

70

70

90

30

30

30

163

30

70

63

51

14

30

7

CG

Urban CG hw

Notes Dash indicates that certain information was not collected in the survey or data were not comparable. FGDs  focus group discussions. CS = case studies. hw  homeworkers. CG  control group or non-hw household. * Considering separately the one for women and the one for children.

Central Java Central Java

West Java West Java

West Java

Province

Total hw

District

Village(s)

Subdistrict

Households surveyed

Location

Source: Indonesia Country Study Survey (2001).

Sub-total

Batik

Rattan

Pottery

Sector

47

40

—

7

hw

Rural

39

16

—

23

CG

6

2

2

2

Total

—

—

—

—

Urban

6

2

2

2

Rural

Number of FGDs*

Table 8.6 Surveys on homeworker households in Indonesia: sectors, location, number of FGDs, case studies and households surveyed

9

3

3

3

Number of CS

262

Oey-Gardiner et al.

Control group CG households are those that do not fulfil all the above criteria, even though some of them may include a member active in the sector but not a woman as homeworker.20 The CG was from the same physical location. For comparability the Indonesian team maintained the requirement of having a child between the ages of 6 and 15 years old for the CG. Questionnaire The questionnaire was designed on the basis of the UNICEF survey guidelines (Mehrotra, 2000). The survey was divided into five different sections in which questions directed specifically to children, homeworkers and non-homeworkers were included.

Focus group discussions and case studies Two FGDs in each sub-sector were held. The first group consisted of women homeworkers, and the second of the children of homeworkers. The objectives of the FGD were to ascertain perceptions, attitudes and opinions of women and children on homework. FGDs focused on three major issues; advantages or benefits of homework, disadvantages and or problems faced by the household and ideal interventions by governments (central and/or local), NGOs, and other members of civil society. The three case studies for each sector consisted of typical and non-unique cases including: (1) a day in the life of a homeworker’s household,21 (2) a profile of an intermediary’s household,22 and (3) production to marketing chain.23 The in-depth case studies contributed to the understanding of the dynamics and processes in the lives of homeworkers.

8.3

Sectors

In this section a description of the environment and production process is given for each sector.

Batik Batik production in Java has been known since the early twelfth century. At that time, only Javanese nobility wore batik tulis. By 1850 batik making became more commercialised since the Dutch colonial government imposed several steps to facilitate interregional textile marketing. In this way by 1950 and with the introduction of the copper stamp, batik became available for the masses. Pekalongan, a district in the province of Central Java, is known for being the centre of the batik industry in Indonesia along with Yogyakarta and Solo. The selected study villages (Wiradesa, Bener and Dadirejo) are situated in the interior, about 10–15 km from the sub-district capital. Even though these villages are located a fair distance from the main provincial road, they are connected by kabupaten (district) roads and are within relatively easy reach using public transportation – mainly becak (pedicab) or andong (horse-drawn cart).

Women and children homeworkers in Indonesia 263 Sanitation is acceptable. Few houses have private toilet facilities disposing human waste into septic tanks. Most households simply use the ground or visit local shrimp ponds, rice fields or the river. Water supply comes from the district drinking-water company (Perusahaan Daerah Air Minum or PDAM), wells, pools or from local shrimp ponds. Just as in most rural areas elsewhere in the country, educational qualifications of these villagers are limited. The majority has at the most completed primary education, with a very limited number of secondary school graduates. Despite the concentration of the batik industry, most men rely on fishing and farming for livelihoods and batik is therefore a female activity. Production Process Despite commercialisation of batik production, old patterns of production that rely on small workshops and homeworkers under subcontracting and outsourcing arrangements have prevailed. The production process of hand-painted silk batik consists of three stages24 and the level of skills required is not high. However, highly skilled artisans are required to do the most sophisticated drawings or batik tulis by hand. Batik can be done on silk, and on a variety of unbleached cotton. The activities in batik are mostly performed in-house (88 per cent). Other workplaces include terraces and yards (7 per cent).25 Fine hand-painted batik relies entirely on manual production techniques. No machinery is used. The industry relies on highly skilled artisans and homeworkers. The more expensive the material the more likely it may be used for batik tulis. An interesting phenomenon in the batik industry in Pekalongan is the extreme level of specialisation by community. Thus in the selected communities batik activities are limited to only waxing on silk and adding tassels at the end of a piece of cloth mostly for scarves and shawls. In the selected research sites they only produce expensive silk batik that sells for Rs 175,000 to Rs 600,00026 (equivalent to US$18.42 to US$63.1) depending on size and quality of the cloth as well as the design and designer. There is a gender division of labour in the batik industry where women do the ‘light’ work of nglowongi (waxing) and the few men involved are responsible for the ‘heavy’ work of nglorod (washing off the wax in hot water). Batik is essentially a woman’s activity (97 per cent of total homeworkers are female) (Table 8.7). As batik on silk is a highly skilled activity, mothers prevent children from being involved in the work. However, children are involved in the production (Table 8.8). An important characteristic of children’s involvement is the high predominance of girls. The few boys found in the batik sector were responsible for rolling the thread. Most of the boys assist in their fathers’ occupations, for example, fishing and trade.27 There are three basic organisations of batik production in Pekalongan. The first organisation is the factory or workshop-based unit where most of the production process is done in combination with the use of homeworkers. In the second type of organisation, intermediaries hire homeworkers for commission to produce batik at the workers’ respective homes and provide them with the raw materials.

264

Oey-Gardiner et al.

Table 8.7 Homeworkers: some descriptive statistics by sector Item

Batik

Pottery

Rattan

Total

Number of households Number of home workers Average number of home workers Number of female home workers % female home workers

70 94 1.3

70 173 2.4

70 147 2.1

210 414 2.0

113

297

91

93

97.0

54.0

77.0

72.0

Source: Indonesia Country Study Survey (2001).

In the third type, often referred to as outsourcing, homeworkers receive raw materials from intermediaries and work at home (Smyth, 1992). The study focused on the third type of organisation. Most batik is produced in small-scale businesses on a subcontracting basis. Intermediaries, locally called cablik, are mostly women who usually also act as homeworkers. Intermediaries determine the wages paid to homeworkers and deduct their commission, often referred to as ‘transport fees’. The subcontracting chain in this industry is not complex. Batik is distributed in both local and national markets. As opposed to intermediaries and homeworkers, small and medium businesses have direct access to consumers through stores, and indirect access through fashion designers or traders. The demand for batik is fairly stable. The peak months occur during lebaran28 when demand for new clothing usually rises. Relationship with the subcontractor The relationship with the subcontractor is fairly good. Only 2 per cent of homeworkers in this sector claimed to have had a bad experience with a subcontractor. Subcontracting is informal and based on oral agreements.29 The third type of organisation was found in the villages studied. Thus 99 per cent of subcontractors provided the main input (silk cloth) to homeworkers. Only a small percentage of homeworkers (13 per cent) kept a record of the batik orders, materials supplied and payments. The payment arrangement is cash by piece, although in some cases (10 per cent) homeworkers are paid by set. More than half of the workers are paid on delivery of work (51 per cent) or weekly (39 per cent). For example, workers receive Rs 15,000 (equivalent to US$1.57)30 for a 2 metre batik piece that takes about 6 days of work. The daily average is Rs 2,500 taking into account that the cost of raw materials and tools are deducted from the wage rate (net income per 2 metre batik is Rs 12,000) (Annex 8.1). Among all homeworkers only 10 per cent claimed to have bargained with subcontractors for better wages. Contrary to the popular belief that homeworkers are not eligible to benefits, 77 per cent of batik homeworkers claimed to have received a bonus from subcontractors for Idhul Fitri (day of celebration after completion of a month of

Tot

96 99 117 312

153 159

41 48 44 133 60 73

Sector Age

Batik Pottery Rattan All

Girls Boys

CG household Batik Pottery Rattan All Girls Boys

17 23 19 59 30 29

70 65

38 52 45 135

II

1 1 0 2 0 2

6 7

3 8 2 13

I

2 3 2 7 5 2

28 16

13 19 12 44

II

Children working

Source: Indonesia Country Study Survey (2001).

24 25 25 74 30 44

83 94

58 47 72 177

I

Total children

Country

4.2 4.0 0.0 2.7 0.0 4.5

7.2 7.4

5.2 17.0 2.8 7.3

I

11.8 13.0 10.5 11.9 16.7 6.9

40.0 24.6

34.2 36.5 26.7 32.6

II

% of children working

92.8 92.6

94.8 83.0 97.2 92.7

95.8 96.0 100.0 97.3 100.0 95.5

I

88.2 87.0 89.5 88.1 83.3 93.1

60.0 75.4

65.8 63.5 73.3 67.4

II

% of children not working

6 5

2 7 2 11

I

22 8

8 13 9 30

II

18.3 8.2

10.4 20.2 9.4 13.1

Tot

7.2 5.3

3.4 14.9 2.8 6.2

I

Number % of children of children working in hw working in hw

31.4 12.3

21.1 25.0 20.0 22.2

II

81.7 91.8

89.6 79.8 90.6 86.9

Tot

92.8 94.7

96.6 85.1 97.2 93.8

I

68.6 87.7

78.9 75.0 80.0 77.8

II

% of children not working in hw

Table 8.8 Share of children working by age group in homeworker and CG households (I  6–10; II  11–14)

82.4 56.5

62.5 74.1 78.6 71.9

Tot

100.0 71.4

66.7 87.5 100.0 84.6

I

% of children working in hw of total children working

78.6 50.0

61.5 68.4 75.0 68.2

II

266

Oey-Gardiner et al.

fasting required for Moslems).31 However, it was revealed in the interviews that this bonus was deducted from wages (Rs 1,000 deduction). Rattan The rattan industry consists of the production of furniture and basketry. It developed as a result of government policies on rattan exports since the 1980s. Earlier most of the rattan exports from Indonesia were entirely unprocessed. In 1979, the government imposed a series of export bans to prohibit all exports of wholly raw materials, including rattan in 1986, and semi-finished rattan goods in 1989. Thanks to the export ban, the rattan industry experienced considerable growth and diversified into three major productions: washed and sulphurised rattan, semi-finished rattan and finished rattan. Currently, Indonesia produces 80 per cent of the world’s rattan. Rattan products are mostly exported to Malaysia, Japan, Taiwan, Saudi Arabia and Holland. Rejected products and over-production are placed on the domestic market, and in general are of poorer quality than those produced for the export market. Cirebon is a district located in the province of West Java and it is a centre for the rattan furniture industry on the island. Tegalwangi, where rattan activity is mostly concentrated, is an urban centre that includes a number of business facilities such as a bank, public telecommunication facilities and numerous shops. Two smaller villages, Pasalakan and Kaliwadas, were also selected as research sites.32 Except for Tegalwangi, the other selected villages are largely agricultural with rice fields supported by a variety of irrigation systems and include technical, semi-technical and rain-fed fields. As the terms of trade for agricultural products, especially for rice, continues to decline, villagers increasingly prefer to find income earning opportunities through off-farm work. Hence there has been an increasing involvement in the rattan industry. These farming villages have basic social services.33 Education facilities include kindergartens, primary and secondary schools, although some of these are in poor condition. Education levels are similar to many rural areas elsewhere in the country with about two-thirds of adults having completed at least elementary school. However, similar to the situation in Pekalongan, higher education is not a prerequisite for much of the work available. Farming skills, as well as those in the rattan industry are mainly learned on the job. Production process The process is characterised by semi-machine production that involves large-scale industries and homeworkers that produce to orders. While small businesses and homeworkers are responsible for the production of semi-finished goods and components, large- and medium-sized industries manufacture the finished goods. Thus most of the production, from making frames to packaging and distribution, takes place in large industries. The subcontracting system consists of a long production chain. It involves large industries, with more than 100 employees that in turn subcontract medium-size businesses. Medium-size businesses of 20–99 workers have the same production

Women and children homeworkers in Indonesia 267

Large/medium-scale unit

Subcontractor 1

Subcontractor 2

Homeworkers

Figure 8.1 Production chain involving intermediaries in the rattan industry.

pattern as large industries, although they use both a centralised system in the factory, and decentralised system that subcontracts smaller businesses of 5–19 workers, and homeworkers. Only large industries have access to the market (buyers and/or exporters) (Figure 8.1). As in the case of batik, the division of labour is based on gender. In rattan work, age also plays a role in the division of labour. Heavy pieces of work, such as frame construction or weaving large-sized items are assigned to men while women weave small-sized items and perform finishing jobs. Thus men perform the most skilled and better-paid work. While married women work at home, single women prefer to work in factories. More than half (77 per cent) of homeworkers in this sector were women (Table 8.7). Children were also involved in homework; out of 14 working children, 11 were engaged in homework (Table 8.8). As opposed to batik, no gender segmentation was found among children. Both boys and girls participate in rattan production. Market fluctuations are a particular feature in the rattan industry. Christmas and New Year are the peak season and the rest of the year is called low season, meaning normal production. Domestically, the rattan industry is not growing as fast as the pottery or batik industry.34 Relationship with the subcontractor The relationship with the subcontractor is overall good and subcontracting agreements are informal and oral. Only 9 per cent of homeworkers complained about subcontractors’ behaviour due to delayed payments.35 Subcontractors are responsible for the provision of raw materials for weaving. As in the case of batik, only 11 per cent of homeworkers kept records of

268

Oey-Gardiner et al.

the production orders. In addition, rattan homeworkers were not punished for faulty or delayed work. Rattan workers are paid by piece (68 per cent) and by set (31 per cent) on a weekly basis (8 per cent) and in a lesser degree on delivery. The form of payment is in cash and wage rates depended on the type of piece. For example, homeworkers earned Rs 650 per drawer. In one day of work a homeworker can produce five drawers resulting in a daily income of Rs 3,250, and an average income wage (payment is weekly) of Rs 19,500 (equivalent to US$2.05) (Annex 8.1). Like the batik homeworkers, rattan producers accepted wage rates they were offered and only 17 per cent claimed to have bargained for better wages. Regarding benefits, homeworkers received a bonus for Idhul Fitri in form of a sarong or a T-shirt. In addition, rattan workers were eligible to receive advance payments. Pottery Pottery is known for being a traditional home industry in which skills are passed down the generations. Fifty per cent of homeworkers said they became involved in pottery due to family tradition. The village of Anjun, located in the sub-district of Plered in the district of Purwakarta, is known as the West Java small-scale pottery industry centre. Anjun is fairly urbanised and has access to basic economic and social facilities.36 Babakan Sari, a neighbouring village, was also chosen as a research site. Unlike the people in Pekalongan (batik) and Cirebon (rattan), most of the residents of Anjun, relied entirely on the pottery industry for their livelihood. They held a variety of positions including entrepreneurs, labourers, vendors or other professions supporting the pottery industry. Only a few are left to work in agriculture. Given the economic activities available, particularly in the pottery industry which relies on handed down skills learned on the job, parents and children see little benefit from investing in education beyond the elementary level. Of the three study sectors it is here that the level of product differentiation is highest. As basic raw materials for the pottery industry are the least expensive and practically no quality control is imposed, it is common to have children actively participate in the production process. Not surprisingly educational qualifications of the residents are slightly lower than in the other two sectors. According to the village report three-quarters have, at the most, completed primary schooling and about half are illiterate. As with the other two study sectors – batik and rattan – in the pottery producing communities sanitation is fairly acceptable.37 Production process There are a number of stages in the production process depending on the item.38 Homeworkers are involved in the preparation of the raw material39 and the creation of the pottery items. As no machinery is used, skill is indispensable in determining the quality of the product. The stage of shaping and drying is done on a subcontracting basis by home-based workers or in the

Women and children homeworkers in Indonesia 269

C Exporter/buyer (for export market)

O N S

Home workers

Intermediary

Employer

Broker/ intermediary

U M E R

Figure 8.2 Marketing chain.

factory,40 and in own houses (47 per cent).41 The final stage of glazing generally requires more sophisticated technology and is done in larger factories. Like in the batik and rattan sectors, the production process takes place in a gender-based job allocation system. Adult males make large pieces of pottery while women usually make small pieces (bon-bon) and youngsters often do finishing like sanding and painting of inexpensive pieces. The percentage of female workers is 54 per cent of all homeworkers involved in this activity (see Table 8.7). The involvement of children in this sector occurs at various stages of the production process of small items of limited value. Children participate in sanding, burning, re-sanding, glazing, varnishing or painting. The number of children involved in pottery was 27, out of the total of 57 children in all sectors. Of the 27 working children, 20 were engaged in homework (Table 8.8). Products are sold in regional and foreign markets. Most marketing is conducted through buyers who place orders for outlet stores owned by medium to small businesses or directly to customers (Figure 8.2). The domestic pottery industry has grown fairly rapidly during the last few years, as the products have become acceptable decorative accessories both inside and outside homes and public places. Overall, the pottery industry is booming as it has become an important decorative commodity. Relationship with the subcontractor In general, homeworkers are satisfied with the relationship with subcontractors.42 The work agreement is informal and oral. As opposed to the batik and rattan, pottery homeworkers are less dependent on employers for raw material supply since clay mix is available locally. In comparison with the other sectors, pottery homeworkers are keener to maintain records43 given that they are required to deal with more employers. Potters are paid by piece (88 per cent), by set (9 per cent) and in some cases by kodi (20 pieces). Homeworkers receive payments on delivery of work (50 per cent). In this sector, workers are paid irregularly (23 per cent) and on a weekly

270

Oey-Gardiner et al.

basis (13 per cent). As opposed to the other sectors, a higher proportion of potters (30 per cent) claimed to have bargained on wages with subcontractors. The benefits they received from subcontractors include a bonus for Idhul Fitri, in the form of food and medicine.44

8.4

Homeworkers in the sectors studied

This section aims at identifying the profile and conditions of homeworkers by comparing them with CG households. The description includes specific characteristics of homeworkers per sector based on the survey results. Reasons for engagement in homework Family tradition is most common in the pottery industry where 50 per cent of households claimed they engage in homework for this reason. Far fewer, only 18 per cent in both batik and rattan, claim continuing a family tradition as main reason for doing the work. Fairly high proportions among batik (33 per cent) and rattan (28 per cent) workers claimed to have involved in homework because they were unemployed earlier and to gain ‘pocket money’. The majority of responses (40 per cent) indicated that the reason for engaging in homework is that ‘no other work is available’, or for limited skill levels (Table 8.9).45 Moreover, homework is considered as a second job, or a pastime to be done in between domestic responsibilities. Indeed, women typically spend between two and

Table 8.9 Women workers: main reasons for doing homework (%) Main reasons Family tradition Preoccupation (instead of doing nothing) Get pocket money Can do housework Can take care of children No fixed working hours No transport needed Skills learning Forced by parents No other work available No other skills Others Total

Batik

Pottery

Rattan

Total

18 12

50 8

18 18

32 13

21 3 5 — — 1 1 15 6 17 100

— — 1 1 2 3 — 12 1 22 100

10 1 2 1 3 — 1 29 1 14 100

8 1 2 1 2 1 1 19 2 18 100

Source: Indonesia Country Study Survey (2001). Note The order of choices was mixed and has been rearranged for this analysis.

Women and children homeworkers in Indonesia 271 three hours on homework in the morning and afternoon, depending on their daily routines and household responsibilities as per household size and composition. In general, homeworkers seem fairly positive about their jobs, as is reflected in the majority response (57 per cent) that they do not desire to change jobs. In the case of children, most of the responses indicated that mothers encouraged them to become involved in homework to pay school fees or contribute to the household income. Only 2 out of 41 children claimed that they were forced by parents to become involved in homework. Mothers considered the early involvement of children as training for the future.46 Children do not perceive homework as disadvantageous. On the contrary, it seems from FGDs that children enjoyed working since they saw it as an opportunity to socialise with friends and earn money. Homeworker households and women Homework is mainly a feminine activity. Homeworkers are predominantly female (72 per cent of homeworkers among the three sectors are female) (Table 8.7) and of an average age of 30–44 years old. This conclusion supports an earlier study (Suhaimi, 1995) which found a higher proportion of females among homeworkers. Sex composition and age of homeworkers varies over the sectors. While batik and rattan are predominantly a female activity (97 and 77 per cent of homeworkers are female, respectively), pottery is rather a neutral gender activity with 46 per cent of male and 54 per cent of female homeworkers (Table 8.7). Regarding age levels, the average age in the pottery sector was lower with 54 per cent of women from age group of 20–44 years old.47 Among child workers there is a higher percentage of girls (82 per cent) who are homeworkers in comparison with boys among whom homeworkers constitute only 57 per cent (Table 8.8). As mentioned above, the nature of the work done by women in the three sectors is ‘lighter’ in comparison with men’s activities.48 In rattan and batik, men’s limited involvement in homework responds mostly to transitional employment periods49 with the exception of the pottery sector in which homework can be men’s main activity. Another important characteristic of the homeworkers’ profile is the marital status. According to the survey results, most homeworkers are married (73 per cent of homework households are married) which was also the case for adult households (15 years old and above) among the CG group. Education level among homework and CG households is the same, yet not among men and women. Earlier studies (Wirutomo, 1992; Wijaya and Santoso, 1993; Suhaimi, 1995) found that homeworkers have a low educational level. This study found no differences in the education level of homeworkers and the CG. The majority of households in both groups completed primary schooling (61 per cent of homeworkers and 65 per cent of households from CG) (Table 8.10). Since primary schooling appears sufficient for obtaining employment there is little incentive for continuing to secondary school. There are observable differences in terms of gender. Among older generations there is substantial educational difference between women and men. The highest

272

Oey-Gardiner et al.

Table 8.10 Distribution of home and CG workers by level of education (%) Education

Homeworkers

No school Primary school Secondary school

Control

Batik

Pottery

Rattan

Females

Males

Total

30 63 7

3 62 35

5 58 33

15 62 23

3 58 39

11 61 28

10 65 25

Source: Indonesia Country Study Survey (2001).

Table 8.11 Average weekly working hours for homeworkers and CG workers based on gender Sex

Females Males Total T-test F-test

Homeworkers Batik

Pottery

Rattan

29 4 28

30 39 34

31 37 32

Total homeworkers

Control

T-test

30 38 32 2.370* 3.815*

43 35 39 4.177*

5.720* 0.891 3.378*

Source: Indonesia Country Study Survey (2001). Note * Significant at   0.05.

proportion of workers (30 per cent) with no schooling was found among the batik workers who are mostly women. In addition, a higher proportion of men have completed secondary school (39 per cent) in comparison with the proportion of women (23 per cent). The study found a significant difference between the working hours of homeworkers and CG households. On average, homeworkers worked fewer hours than CG workers (32 compared to 39 hours/week).50 Significant differences were found among homeworkers by sector. Batik homeworkers work the least number of hours per week (28 hours) in comparison with potters (34 hours) and rattan (32 hours) workers. The gender difference in monthly working hours among homeworkers is statistically significant. In general, women work fewer hours in homework than men (30 compared to 38 hours per week) (Table 8.11). The reason is that women allocate more time for duties such as housework and taking care of children and husbands. In short, for female workers, homework is considered a side job. As discussed above, women work in between domestic chores (also see evidence in Annex 8.1). Thus working hours depend on household circumstances. For example, homeworkers with small children tend to work fewer hours. Homework earnings are a secondary source of income. With the exception of the pottery sector, homework is not a major source of household income.

Women and children homeworkers in Indonesia 273 Table 8.12 Distribution of households by shares of total household income derived from homework activities (%) Share of household earnings from home-based work in total household earning

Batik

Pottery

Rattan

Total

25% 25–50% 50–75% 75
Total

73 16 6 6 100

14 16 20 50 100

63 21 3 13 100

50 18 10 23 100

Source: Indonesia Country Study Survey (2001).

The share of income derived from homework is positively related to person-time allocation to this income earning activity. About half the households derive only 25 per cent of total household income from homework activities, half derive 25 per cent. About one-fourth (23 per cent) derive most (75–100 per cent) of their incomes from homework (Table 8.12). Among the three selected sectors, the rattan workers are the worst off as substitution for rattan products by plastics and wood is increasing and there are few alternative income earning opportunities for rattan workers in the areas studied. Potters have a more positive perspective given that the domestic demand for pottery is fairly stable. Homeworkers earn more than formal sector workers and are above the poverty line. The survey results refute the common perception that homeworkers are the poorest in the employment sector. Part of the explanation lies in the fact that not all homeworkers’ households rely on earning from homework only, particularly batik homeworkers (Table 8.12). As depicted in Table 8.13, the wages earned among all sectors are higher than the Regional Minimum Wages in 2000.51 The total average monthly wage earned in batik is Rs 198,807 (equivalent to US$20.93) which is higher than the monthly average wage earned in Central Java (Rs 185,000, equivalent to US$19.47). In pottery, the average monthly wage is Rs 274,195 (equivalent to US$28.86) and in rattan Rs 212,355 (equivalent to US$22.35) which are both higher than the West Java monthly wage of Rs 230,000 (equivalent to US$24.21). The total average wage earned by men in the three sectors is also higher than the respective regional monthly wage. This is also the case for female workers, with the exception of rattan homeworkers where the female wage (Rs 136,000, equivalent to US$14.31) is lower than the regional wage. Children, on the other hand, are poorly remunerated. In the case of the batik (Rs 82,930-equivalent to US$8.73) and rattan sectors (Rs 78,206 equivalent to US$8.23) both wages are below the regional minimal wage. Another indicator that suggests that homework is not necessarily linked to poverty is the comparison of homework monthly expenditure per capita with the monthly expenditure that defines the poverty line.52 As depicted in Table 8.13, the

274

Oey-Gardiner et al.

Table 8.13 Average monthly equivalent wages for homeworker (Rp.) Wages by sex and age, regional minimum wages and poverty lines

Total average Adult female Adult male Children Regional monthly minimum wages in 2000 Hw households monthly per capita household expenditures District level poverty lines based on monthly per capita household expenditures in 2000 Per cent population below the 2000 poverty line (%)

Central Java

West Java

Pekalongan

Purwakarta

Cirebon

Batik

Pottery

Rattan

198,807 198,807 –* 82,930 185,000

274,195 227,046 305,969 –**

212,355 136,594 492,672 78,206

125,611

114,046

230,000 71,730

Urban Rural Urban Rural Urban Rural 78,019 62,861 86,755 67,050 86,755 67,050 14

9

28

Sources: Indonesia Country Study Survey (2001) and CBS, special tabulations from the 2000 SUSENAS. Notes – Average monthly wages are estimated from value of last order(s), divided by the number of person hours allocated to produce the order, multiplied by 173 hours/month, minus raw materials expenses during the preceding month. * Based on 1 worker only found in the survey. ** All child home-based workers in pottery are unpaid family workers.

monthly household expenditure per capita in the batik sector is Rs 125,61153 (equivalent to US$13.22); in pottery is Rs 114,046 (equivalent to US$12); and Rs 71,730 (equivalent to US$7.5) in the rattan sector. In the three sectors, the expenditure level per capita is far greater than the urban and rural expenditure in its respective regency that is, Pekalongan in Central Java, and Purwakarta and Cirebon in West Java. Hence, all homework households among the three sectors are above the poverty line. Moreover, expenditure habits also indicated that homeworkers are not very poor, as they can still allocate some money for arisan and attend religious activities. Indeed, the poverty incidence in the three areas in 2000 was low with 14 per cent of the population living below the poverty line in the batik area of Pekalongan in Central Java, 9 per cent in the pottery area of Purwakarta in West Java, and 28 per cent in the rattan area of Cirebon in West Java. Sectoral comparisons showed that pottery homeworkers had the best economic level since they held the highest percentage of workers in the top quartile (36 per cent) compared to batik (24 per cent) and rattan (10 per cent) households. If measured in terms of the poorest quartile, the pattern is the same, that is, the percentage of potters’ households in Purwakarta is lowest at 11 per cent, with

Women and children homeworkers in Indonesia 275 batik households in Pekalongan in the middle at 24 per cent and rattan households in Cirebon much over-represented (50 per cent). Pottery is the sector where half the households derive three quarters of household income from homework. This study has shown that homework is neither the cause nor the consequence of poverty. We thus infer that homeworkers can be poor, not poor and even well off. The poor rattan homeworkers live in the most economically depressed area of the three sectors studied while the batik workers and potters who live and work in far more economically dynamic areas are not as poor and their earnings are even competitive with the regional minimum wages. Even though the households studied do not belong to the very poor, in many ways they still project the behaviour of a culture of poverty as reflected by the way they maintain the work place and deal with health issues. In non-monetary terms homework households are worse off than non-homework households. Non-monetary indicators showed that homeworkers are less prosperous compared to CG households.54 Based on number and type of durable goods,55 the CG households are better off. While 18 per cent of CG households have six to eight items of durable goods, only 7 per cent of homework households own a similar number of durable goods. The only similarity is in terms of the share of households owning a television (68 per cent of home-based households compared to 66 per cent of control households but the number of control households is much smaller). The homeworkers labour market applies strong gender-based discriminatory remuneration. In those sectors where women and men are absorbed in homework men earn substantially higher wages than women (Table 8.13). The reason being that in the informal sector the labour market is strongly segmented. The general stereotypical rule applies that men do heavy jobs and work on large objects while women do light and fine work and produce small and delicate objects. These different commodities are valued differently by the market, and usually those goods produced by men are higher priced than the goods produced by women. As a result men earn more than women for jobs completed in the same amount of time. Homework affects the health status of homeworkers. The evidence that homework is affecting homeworkers’ health comes from a comparison of results with CG households. Although most of the households suffer from common ailments (cough, flu and fever), the incidence of suffering was higher among homework households (Table 8.14). In addition, homeworkers are more vulnerable to work accidents and muscular pain due to repetitive work and long hours. Health ailments such as eye or ear infections, skin rash and respiratory problems originated from exposure to toxic substances. Another factor that aggravated the health status of homeworkers is the polluted environment. This was the case in the pottery sector and, to a lesser degree, in the batik and rattan sectors. Child labour The incidence of child labour is higher among homework households than in CG households. There was a total of 312 children between the ages of 6 and 14

276

Oey-Gardiner et al.

Table 8.14 Percentage of children aged 6
claiming health problems during the preceding month Type of health problem Number of respondents Flu (colds, fever) Headaches Coughing Eye infection Ear infection Skin rash Breathing problems Waist ache Stiff neck Back pains Shoulder pains Work accidents

Home workers

Non-home workers

Batik

Pottery

Rattan

Total

N-HW

94 30 35 7 3 0 2 1 4 0 0 0 2

173 22 28 23 1 1 3 2 6 0 3 0 0

147 24 29 8 1 0 1 1 9 2 4 1 0

414 24 30 14 1 0 2 1 7 1 3 0 0

248 20 22 12 0 0 2 1 4 0 4 0 0

156 16 22 8 1 0 1 0 3 2 1 1 0

Source: Indonesia Country Study Survey (2001). Note As the list of ailments is not mutually exclusive, respondents can give multiple answers.

in homeworkers’ households, of which 57 were gainfully employed – 41 of these working in the relevant homework activities. In comparison, there were only 9 children working among CG households among 133 children in the same age range (Table 8.8). However, even among homeworkers’ households, the incidence of child labour below the age of ten is very low. A small child working (other than at household, or other simple chores such as collecting wood or taking care of animals) seems to be relatively rare. One reason, at least among the home-based industries, may be that small children are not capable (skill or concentration) of contributing to production. They may participate informally, but more in a sense of educational play that contributes to the acquisition of skills, but not providing any economic contribution. The substantially higher incidence of working children among homeworkers’ households suggests that the existence of a viable homework activity provides an additional incentive to put children to work. The number of children involved in homework represents 72 per cent of children among homework households (Table 8.8). This phenomenon is found among the three sectors and in a higher degree for girls than for boys. Four out of every five working girls are homeworkers but only three out of every five working boys are homeworkers. The sector with the highest incidence of child labour was pottery with 27 per cent, followed by batik (16 per cent) and rattan (14 per cent) (Table 8.8). Since pottery making is a traditional family business it is more likely that children are employed.

Women and children homeworkers in Indonesia 277 However, even if more prevalent among homeworkers’ households, child labour is not universal. It is very rare for parents to involve children, especially young children, in gainful employment. But children who have completed primary school are more likely to join income-earning activities. Children who work outside home, prefer to work with subcontractors and use earnings from homework for pocket money. It is not common for children of homeworkers’ households to work outside the home. Of the 57 child workers only 16 (or 28 per cent) worked outside the home. The 16 children working outside the home are aged 8–14 years. There are 244 children aged 8–14 in the homeworkers’ households. These numbers indicate that only about 6 per cent of children from homeworkers’ households work outside the home. This proportion is comparable to the 7 per cent of children from the CG households who are in employment. Another issue is that the case studies suggest that it is children who work prefer to work for others (see Annex), not their parents. The FGDs found that children in all sectors studied are working for income. Most of the children work inside the home and prefer to work with subcontractors, rather than parents, to ensure remuneration. Most of the children expressed a preference for working in a group at the employer’s workplace (Annex 8.1). Remuneration from homework is mostly used to pay school fees or to earn pocket money. Children receive payments directly and are able to manage their own earnings. On average, they receive between Rs 400–500 per hour, which results in an average of Rs 1,000 (equivalent to US$0.1) to Rs 2,000 per day (equivalent to US$0.2). Given the characteristics of the working children, it is not surprising that the overall average number of weekly working hours is not very high.56 Taking all the children together the average is only 20 hours a week or about 3 hours a day. An attempt was made to understand the determinants of hours worked by controlling for several variables. However, results show that the differences are not statistically significant. Thus working children allocate similar numbers of hours to employment irrespective of whether they belong to homeworkers’ households or not, whether they are involved in homework or not, age (between 8 and 14 years), sector, gender, and even school attendance, and whether they are paid or not. There is no gender discrimination among working children. Children are introduced to homework at an early age despite the minimum working age of 15 years. In batik, girls start work at eight years, while in rattan and pottery they start at 7–9 years. Usually they start by assisting mothers in simple tasks and acquire more responsibilities with experience.57 Once they acquire experience, both boys and girls work an average of 20 hours per week (3 hours per day). This refutes the stereotypical view that girls work longer hours than boys. In the same way, study findings contradict the idea that only girls are obliged to undertake domestic chores. Boys are not exempted from assisting mothers in housework. In terms of the allocation of activities, involvement in homework or domestic chores does not seem to interfere with other activities, including schooling.

278

Oey-Gardiner et al.

Although many child homeworkers stated that their school activities are not being interrupted by their work, the work has, to some degree, reduced their enthusiasm in pursuing further education. During FGDs, children mentioned that if their parents were rich enough to pay their further education, they would continue. If not, they will quit school. In discussing the choice of activity between school only, work only or school and work, the FGD participants mostly choose school and work. In an economically weak household, the children prefer to work. All child participants refused the possibility of ‘school only’ as they firmly stated: ‘ . . . no, we want to keep working,because we are used to work and earning money. If we work, we can buy our daily needs such as hair accessories, veil, etc. by ourselves’. Finally, mothers also supported the combination of work and school for children. School enrolment is lower among children from homework households. School enrolment among children from homeworkers’ households of all age groups is 81 per cent compared to 93 per cent in the CG. Another way of looking at the same phenomenon is by focusing on the working children. Up to age 12 (elementary school ages are between 7 and 12 years) practically all working children, even in homeworkers’ households, attend school. However, after completion of elementary school, the share of children attending declines rapidly. There are a number of possible explanations. Parents do not consider investment in higher education worthwhile. Another reason is that local employment opportunities do not require any further schooling. Moreover, the school system is not attractive to parents or children.58 Another factor that prevents children from pursuing higher levels of education after elementary is the cost. Educational expenses at the lower secondary level are too high for many parents. Moreover, there is limited access to secondary schools. Consequently, parents are more likely to be faced with transport costs when sending their children to school. Very few children were only working. The few cases of ‘only work’ children, were children who were those beyond the elementary school age (13 and older). Even though it is recognised that attendance at school is important (Hodgkin and Newell, 1998, 439), children did not consider homework as an interruption of their schoolwork. The effect is felt to be greater on their playtime rather than schooling. This finding is equally true for child homeworkers and child workers from the CG.59 Mothers also confirm that they never let their children work for extended periods of time and without interruption. The results suggest that homework cannot be attributed as a cause of poor health as there is little difference in the patterns of health problems experienced during the preceding month by the various groups of children. Besides, the qualitative study found no major health complaints arising from children’s involvement with homework. The most frequent complaint was stiffness from sitting in the same position for too long. In all three sectors, children do their jobs while sitting on a small, low bench or on the floor. As a result, when they have to remain in that position for extended periods, their backs become stiff and sore. These back pains, however, disappear if they take a rest, after which they often return to their respective jobs.

Women and children homeworkers in Indonesia 279

8.5

Conclusions and policy recommendations

Conclusions There is limited information regarding the contribution of homework to the national economy. Moreover, homeworkers remain ‘invisible’ in official labour statistics. Homework in the sectors and locations studied in Indonesia is mostly a female activity although men are also involved in the stages of production that require more strength. The only sector where women and men were equally engaged was in pottery. Where men were involved in this sector this represented the households’ main economic activity and thus the main household source of income. In rattan and batik, women contributed to the household income, and earnings were considered a secondary source of income used for daily needs and pocket money in the case of children. Children are commonly involved in homework. However, no exploitation of children is likely and their average working hours do not seem to interfere with schoolwork. Contractor/subcontractor relations with homeworkers in the selected industries imply fairly congenial and informal relations. Communication is mostly oral and few rely on written records. The relationship between homeworkers and contractors is informal in nature. Orders are mostly given orally or without any specific agreement. Accessibility, mostly due to price, determines the provider of main raw materials. In both batik and rattan it is practically universal that employers provide the main input, the cloth and the rattan to be woven, while homeworkers are responsible for all other secondary inputs and tools. Potters are generally expected to provide their own raw materials as contractors or employers pay differentially depending on who provides the clay mix. An important additional criterion of homeworkers is that they are paid piecerates, where the concept of a piece depends on the sector. For instance, in batik a piece is a piece of cloth (varying in length from 2 to 4 metres). In pottery a piece is any piece of pottery or a set (some make statues of differing sizes). In rattan, a piece may be a set of drawers or a tray, while for furniture makers a ‘piece’ may be a set of chairs and a table. Child labour is far from universal. The involvement of children as homeworkers is a function of the availability of income earning opportunities, rather than due to poverty pressures as is often thought. Most children both work and also attend school. And, if they do only work, they have probably reached the age beyond elementary school age, which is 13. Work, or the income earning activities of the child workers studied, seems not to be in conflict with other childhood activities, such as attending school and playing. The time commitment to homework is flexible and limited and therefore does not interfere with study and play. Working children spend only a few hours a day on economic activities, usually after school and before play time. In two sectors, their household expenditures exceeded estimated poverty lines for their localities based on the National Socio-economic Survey data on household expenditures; in one sector it is lower. Homework is an important source of

280

Oey-Gardiner et al.

income and an important contributor to keeping households out of poverty. In some households homework is practically the only source of income while in others it is a secondary income. Household incomes and expenditures suggest that the communities of potters have the highest economic level. Potters’ communities are also the most vibrant as the demand for their products is very high while batik and rattan industries seem like sunset industries suffering from competition from substitutes. Another weakness for homeworkers, and an issue of general concern, is the low awareness of occupational health dangers. The workers complain about ailments which some do recognise as arising from the activities they do, such as pollution in the home caused by their activities. Yet no preventive measures have been taken. Homeworkers are not unique and instead share similarities with the other members of the communities in which they live. In general the homeworkers studied share similar socio-economic characters with their ‘neighbours’ with their surrounding community and even the wider Indonesian population. This is also true of children. Even though, relative to the wider national population, children in homeworkers’ households are over-represented as workers, parents claim that their first priority for their children is schooling, at least to completion of elementary schooling. Recommendations Data collection and survey design Despite past efforts, homeworkers continue to be invisible statistically. In order to facilitate the collection of homework data we recommend: 1

2

3

to maintain as far as possible the work status categories as applied up to the 1999 Sakernas (National Labour Force Survey) and other surveys collecting labour force information, for the sake of continuity,60 allowing measurement of the labour force, the workforce, and the separation between formal and non-formal sector workers; to add a question on place of work, which, when combined with the work status variable allows for the separation of home based and non-homeworkers among non-formal sector workers; to add another question on size of establishment,61 as measured in terms of number of workers (to be used in combination with place of work) which can serve as a control variable enabling identification of home-based work sector.

Government specific programme The government should develop specific programmes to support homework activities financially and technically. These programmes can be part of the social safety net programme, now being executed on a large scale by the government to alleviate poverty.

Women and children homeworkers in Indonesia 281 Another aspect that government could focus on is the relationship with suppliers by providing incentives to firms that outsource production to homeworkers by for example, facilitating access to relatively cheap loans to these firms. NGOs could also be involved with government institutions in assisting the development of small-scale enterprises. The areas requiring empowerment should focus on overcoming homeworkers’ weaknesses through the following programmes: ●

Provision of market information

The most important area for empowerment of homeworkers is familiarisation with the market. Homeworkers, like most informal sector workers, have no or limited access to market information. This is an area where UNICEF, NGOs and the Ministry of Trade could work. ●

The specific types of market information that homeworkers can benefit from are, for example: – –

– – – – –

– ●

levels of regional minimum wages; conditions of the local labour market as there are localities where skills are in short supply, which should theoretically raise wages. Such information can be very useful in raising their bargaining position vis-à-vis those placing orders with homeworkers; levels of wages earned by other homeworkers or paid by other contractors or intermediaries; levels of wages for different stages of the production process; reasons for differential wages; types and quality of products that are in high demand; ways or factors that enables homeworkers to demand higher levels of remuneration, such as: designing new products – as homeworkers usually take orders from employers, contractors or intermediaries, they may well benefit if they can propose ‘new’ products of their own design, their own creations, which, of course requires training in design; designing products demanded by the market at a given time in a given market; quality, colour, size, finishing, packaging and so on are all important factors in marketing.

Creation of workers’ associations

Very few of the homeworkers studied were aware and interested in joining workers organisations or associations and they indicated both dissatisfaction with past numerous government initiated organisations, and also low awareness of the value of numbers. In this regard, the call is for labour unions and NGOs working with workers such as Homenet to consider homeworkers who would benefit from

282

Oey-Gardiner et al.

membership in associations or unions to negotiate: – – – ●

better rates/levels of remuneration; compensation for overhead costs covering such items as secondary raw materials, tools and place of work; health insurance.

Provision of health services

Considering the current political turmoil, including the recent laws introducing decentralisation and regional autonomy, and moreover the low levels of awareness of occupational health risks, a longer term health programme for homeworkers is anticipated. When stability has returned, with assistance from the Ministry of Health and its representatives in local governments, NGOs such as Homenet can be mobilised to: – – –

provide information for homeworkers on occupational health hazards as a consequence of the specific jobs they carry out; provide information on how to deal with such health hazards resulting from the specific jobs they carry out; provide information on the possible dangers of self-prescribed medication using store bought factory produced chemical drugs, often painkillers, available over the counter.

Annex 8.1: Case studies: homeworkers’ earnings Batik sector Mrs Sar, a batik maker starts her day very early in the morning with household chores: she washes clothes, boils water and cooks rice, sweeps the floor and yard and buys groceries. At 8 am she starts making batik and stops at 11 am to prepare lunch and perform afternoon religious duties. At 1 pm she returns to work and stops at 3 pm. Her average working day is 5 hours. She doesn’t work on Friday since that is pay day. She does a sheet of 2 meter batik for Rs 15,000 per piece which has to be done in 6 days. On average she gets Rs 2,500 per day. Still to be deducted from her wage is the cost of tools: wax, canting, kerosene, which are sold in warung nearby, amounting to Rs 3,000 per sheet. Thus the net amount she receives is Rs 12,000. Payment is on delivery of finished sheet. Rattan sector Mrs Eci, a rattan weaver, starts weaving drawers at 7 am. At 8 am she stops to do the cooking and washing. At 10 am she starts weaving again until midday and then breaks for an hour for lunch and afternoon prayer. She starts weaving again from 1 pm till 4 pm. During 6 hours work she finishes 5 drawers at Rs 650 each

Women and children homeworkers in Indonesia 283 or Rs 3,250 per day. She works 6 days a week and Sunday is a holiday. Payment is made once a week. Her average income a week is Rs 19,500. Drawers’ framestructures, rattan and nails are supplied by an intermediary. She provides the hammer, scissors and su – a tool to punch holes in the frames, consisting of a wooden handle with a nail at one end, which she made herself. An intermediary fetches and delivers the finished work. Pottery sector Mrs. Min, a potter, begins work at 8 am and stops at midday. An hour’s break is taken for lunch and afternoon prayers. She starts again at 1pm until 4 pm. She produces 300 bonbons per day. She is paid Rs 30 per product so she can earn Rs 9,000 daily. Payment is on delivery of finished work. Usually, 30 per cent goes in advance to buy clay (costing Rs 3,000 per 25 kgs, enough for 400 bonbons). Payment may be given in advance, and must immediately be traded with a quantity of bonbons of equal worth. She works 7 days a week with no holiday. Tools for shaping clay must be provided by the homeworkers, as well as other necessities.

Annex 8.2: Results from FGDs with children about the workplace Working together with friends is preferred by children and seen to be the fun part of the job. As a group of young girls who participated in a FGD put it: ‘ . . . we prefer not to work at home because doing batik alone will make us sleepy and bored. Working together is fun, we can chat and discuss the Indian movie we watched last night on TV. There are always funny stories told by one of us which keeps us wide awake. By working together the work can be done faster . . . ’ Children in the pottery sector also preferred to work in a group and in most cases it can be done when they work at their employers’ place. The reasons for working with friends are similar to those of girls in the batik sector. In the case of pottery, working in a group also motivates them to work faster because they compete with each other to produce more. As one employer said, ‘When the children come and work together sometimes they are very noisy and they race to produce more than the other. The result is that I have to check the products carefully and ask them to work finer instead of faster . . . ’ In the rattan industry the children work at home with their mothers. It is not common for children to work at the employers’ place because the work is distributed by the employer to their workers.

Notes 1 See Wijaya and Santoso, 1993; Moedjiman, 1993; Yayasan Pengembangan Pedesaan, 1993 and Bina Swadaya, 1996. Another study ‘The Asian Crisis and Women Home-based

284

2 3 4 5 6 7 8 9 10 11 12 13 14

15

16 17 18

Oey-Gardiner et al.

Workers in Indonesia’ was funded by the World Bank in 2002. The study focuses on two sectors, garments and food. At the time of writing its results were not available. Other initiatives included research projects conducted by Yayasan Pengembangan Pedesaan (YPP), Merdeka University, Save the Children Indonesia and Bina Swadaya. The governmental organisation Badan Pusat Statistik (until 1999 called Central Bureau of Statistics or Biro Pusat Statistik), conducted the Sakernas series. The total population was 113.5 million and rose to 141.1 million between 1990 and 1999 (i.e. at 2.4 per cent per annum). During the same period, the total number of workers rose from 73.4 million to 88.8 million (at only 2.1 per cent per annum). In 1990, 55 per cent of workers were employed in agriculture, 10 per cent in manufacturing, 16 per cent in services and 15 in trade. By 1999 the percentage of workers in agriculture was 43, 13 in manufacturing, 19 in services and 20 in trade. On the other hand, the share of workers employed in trade and services increased from 15 and 16 per cent respectively to 20 and 19 per cent between 1990 and 1999. As the crisis first hit Indonesia in mid-1997, the effects were only recorded by the 1998 survey. Past analysis suggest urbanisation to have occurred in response to population pressures in rural areas and the attraction of the ‘bright lights’ in urban centres. A ‘self-employed worker’ is a person who works at her/his own risk with or without the assistance of his/her family members, temporary workers or employees (Wirosardjono, 1984). An ‘unpaid family worker’ is a person who works without pay in an economic enterprise operated by other members of the family, relatives or neighbours. The percentage of unemployed in 1994 was 4.4 per cent and in 1996 was 4.9 per cent. The share of males in the total workforce has risen marginally from 61.3 to 61.8 per cent, while the share of female workers fell correspondingly. The share of females in the formal sector rose from 29.9 per cent in 1986 to 31.0 per cent in 1988, 31.0 per cent in 1990 and 31.9 per cent in 1992 (source: special tabulations from the BPS Sakernas series). While ‘childhood’ appears to have been extended slightly in Indonesia, there continues to be ambivalence on the minimum age to enter employment. The earliest regulation, limiting children to work was introduced in 1925 by the Netherlands Indies Government as an ordinance of 17 December 1925, when the minimum age was set at 12 years. Thereafter minimum ages were raised and jobs became more specific. A Government ordinance of 27 February 1926 specified 14 years as the minimum age for work aboard ship. In 1930 underground employment was limited to those 16 years and over. Since Indonesia declared Independence in 1945, Indonesian Government rules and regulations as well as ILO Conventions tend to accept age 14 as the absolute lower limit for entrance into the labour market. Hazardous work or those activities which jeopardise the health, safety or morals of young persons, is limited for older youngsters of at least 18 years of age. The current applicable regulation is Indonesia’s ratification of the 1973 ILO Convention No. 138 as Act No. 20, 1999, which specifies that the general minimum age to enter employment is 15 years with caveats for hazardous work requiring a minimum age of 16 or 18 years. Studies on street children noted variations in familial relations. While some street children literally live on the streets only, others were sent into the streets by their parents. This latter group of street children does have a home address and therefore theoretically is included in the survey sampling population. The number of total workers for the district of Tasikmalaya for 1995 was extrapolated from the results of the 1980 and 1990 Population Censuses. For example, see Tjandraningsih and White (1992), Chotim (1994) and Mustain et al. (1999). For example, see Haryadi and Tjandraningsih (1995), Mulandar (ed.) (1996), and Irwanto et al. (1999).

Women and children homeworkers in Indonesia 285 19 Indonesia is divided into provinces (the number continues to vary, as increasingly regions want to become independent provinces). Provinces are further divided into districts, which are further subdivided into sub-districts, which are in turn subdivided into villages. 20 Other positions include, for instance, contractor, employee, self-employed, etc. 21 The objective of this case study is to obtain a dynamic picture of the life of a homeworker household by profiling the household as a setting for the type of work being carried out. The focus is on reasons for involvement in homework and relationships with their employer or provider. 22 As the selected subcontractor is the employer or provider of the selected homeworker in a particular sector, this case study focuses on their forward and backward linkages. These cases shall provide another perspective on homeworkers and inter-relationships between themselves and homeworkers as well as with higher level links in the production to marketing chain. 23 The third case in a sector focuses on the chain from production to marketing. By first obtaining an overview of the selected industry, the focus is on price mark-ups in the chain as shared between producers and contractors, including transaction costs incurred at each level. 24 In the first stage highly skilled artisans prepare drawing in wax on a piece of silk. This is then followed by applying wax or nglowongi and mopok, performed by women homeworkers at the employers’ home or in a factory. The third stage is nglorod or washing off the wax in hot water, a job which is usually done by men. 25 In general the type of product determines the place of work, inside or outside the house. In the case of batik workers they prefer to work inside for ‘fear of dirtying the cloths’. 26 At the time of the survey the exchange rate was Rs 9,500 to US$1. 27 Boys who help in waxing batik patterns are derided by peers (field report). 28 The end of the Ramadhan fasting month. 29 In the Indonesian social context it is preferred to rely on memory rather than written records which is the case in the three sectors. 30 Based on the exchange rate mentioned. 31 This is an important benefit for formal sector workers in the private sector (not in public service) that will go on strike if it is not provided. 32 These are mainly farming villages located about 10 kms from the town of Cirebon and 3–5 kms from Sumber, the sub-district capital where government offices and public facilities are concentrated. However, public transport is readily available to connect the villagers to the towns of Cirebon and Sumber, the sub-district capital. 33 A community health centre, situated in the neighbouring village, is within reach by public transport. While many houses have private bathrooms and toilet facilities, only some have access to individual septic tanks. There is no piped water and villagers generally rely on shallow wells for their daily needs. Religious activities take place in mosques, musholla and langgar. 34 The industry is no longer as vibrant as it used to be, however new innovations and modifications have been introduced in production as well as new raw materials and designs. 35 FGD with women rattan weavers: ‘We have experienced being paid by bon (receipt) because our employer has not been paid by the factory. It was disappointing to be paid by bon because we only received Rs 5,000 from the Rs 20,000 we were supposed to get as weekly income. When that happened we felt angry and not in the mood for work and very tired. We have to wait until next week to get the rest.’ 36 The facilities included business districts, post office, telecommunication facilities and other business facilities, schools, health centres and mosques. 37 Most houses already have private baths, wells or water pumps, few have private toilet facilities and others rely on public facilities. They still rely on ‘traditional’ ways and means, such as ponds.

286

Oey-Gardiner et al.

38 Items vary from small items, called bonbon to large pots and vases of up to two metres high and more than a metre in diameter. 39 The basic raw material for pottery is clay mixed with quartz and lime. As clay is locally available there are people who produce the mixture which is relatively inexpensive. Hence 85 per cent of the homeworkers obtain their own clay locally. Only 10 per cent of the workers claim that the contractor supplies the raw materials and 5 per cent get it through other means. 40 As baking requires a kiln, it has to be done in a reasonable size enterprise and better quality products obviously require larger and better kilns allowing higher temperatures. Some final finishing, when it requires only painting of small goods like the bonbon or statues, can be done at home or in small factories. 41 Unlike batik workers, the work location is less important for potters as clay does not require much care. 42 Twenty-eight per cent of pottery homeworkers reported having had a bad experience with the subcontractor and this is the highest among the three sectors. The fact that pottery homeworkers are more likely to have ‘bad experiences’ with contractors can be attributed to the larger number of subcontractors they have to deal with, and their stronger commitment to their work. 43 Thirty per cent of homeworkers kept records. This is the highest percentage among the three sectors. 44 Five per cent of workers claimed to have received food, and another Five received medicine. 45 Working at home enables the women to set their own working schedules. As shown in the examples of how women in each of the sectors spend their day, batik making, rattan weaving or pottery is done in between their domestic duties thereby fulfilling their role as wife and mother while simultaneously earning money. 46 In a FGD a mother in the batik sector suggested ‘that children will be able to make batik well when they grow up. Later on they began to enjoy the work even without supervision’. 47 In the rattan and batik sectors, women’s age group ranged from 30 to 40 years old. 48 For example in the batik sector men are required to wash off the wax and in rattan men construct the frames, while in the pottery sector men are responsible for making the larger pots or vases of one metre diameter and height. 49 Males in the batik sector are mostly fishermen. The only men involved in batik were actually looking for another job. 50 CG household members are involved in a variety of activities including teaching, working in factories, farming, fishing and also in trade. 51 To establish this assertion, monthly wages by sex and age were calculated for each sector. Monthly equivalent wages are estimated from information on last payment received and estimated hours worked to produce the order for which the payment is received. These results in hourly rates, which are then multiplied by 173 to obtain an equivalent working month applicable to formal sector workers following the Manpower Ministerial Decree no. 72/1984 requiring formal sector workers to be salaried workers paid for 30 days or 173 hours a month. Since formal sector wages are earned for labour only and providers of raw materials differ by sector, for greater comparability we then reduce the earlier derived monthly earnings estimates with expenses made for raw materials during the preceding month. The estimated wages were separated by gender, since remuneration levels differ among male and female workers. 52 In this case we rely on the poverty lines estimated from the 2000 SUSENAS (National Social-Economic Survey), conducted by Statistics Indonesia, based on household expenditures, which is collected every third year. For intervening years estimates are calculated by adjusting for prices changes in the food basket and non-food commodities. 53 This figure is based on the total sample of 70 homework households in batik. Of these 70 homework households, however, during the preceding month 3 had extremely high expenses. Hence, if these 3 were excluded then the monthly per capita household

Women and children homeworkers in Indonesia 287

54 55 56 57

58 59

60 61

expenditures for the remaining 67 home work batik households declines to Rs 92,878, or almost 20 per cent lower than expenditures in pottery households. CG households’ main occupations include teachers, fishermen, farmers, entrepreneurs, labourers, bus and becak (pedicab) drivers and labourers. Durable goods are: cupboard, stove, radio/tape, television, refrigerator, telephone, bicycle/boat, motorcycle/speedboat, car/boat. This comment is made in light of factory workers often having to work 60 hours a week or more in the factory, exclusive of their travel time, which often takes 1–1.5 hours in big cities like Jakarta and its surroundings. For example in the rattan sector, an eight-year-old child can start by participating in the weaving for a rattan drawer. At first, they help their mothers at home and learn how to punch holes at the ends of rattan drawers. Next they learn weaving. Their mothers only allow them to handle an entire drawer after they have memorised every step, including how to fix the pitrit (fine rattan used as binding to cover joints, including nails), and how to attach a drawer handle. One of the major daily newspapers, Kompas, has repeatedly emphasised the poor quality of the system. Even when completing the first cycle or elementary schooling, children are often found functionally illiterate and non-numerate. Playtime is usually after doing homework and before reading the Koran. In batik and pottery, children prefer working with their peers who allow them to chat while working. The children claim that they are free to play with their peers because there is no obligation to work continuously. In the past, the number of categories of industrial classification was suddenly increased from 5 to 10 resulting in not exactly comparable data and therefore confusing time series information. This is a valuable lesson and should be avoided in the future. So far CBS uses the following establishment size classification: home/cottage 4 workers; small 5–19 workers; medium 10–99 workers and large 100
workers.

References Berar-Awad, Azita (1993), ‘Foreword’, in Lazo, Lucita (ed.), Shadows to the Fore, Practical Actions for the Social Protection of Homeworkers in Indonesia, Bangkok: ILO. Bina Swadaya Yogyakarta (1996), ‘The Batik Homeworkers of Bayat’, in Lazo, Lucita (ed.), Shadows to the Fore, Practical Actions for the Social Protection of Homeworkers in Indonesia, Bangkok: ILO. Chotim, Erna Ermawati (1994), Subkontrak dan Implikasinya Terhadap Pekerja Perempuan, Kasus Industri Kecil Batik Pekalongan (Subcontracting and Its Implications for Women Workers, the Case of Small Scale Batik Industry in Pekalongan), Bandung: Akatiga, working paper No. 3. Hardjono, Joan (1990), ‘Developments in the Majalaya Textile Industry’, Project working paper series No. B-3, West Java Rural Nonfarm Sector Research Project. Haryadi, Dedi and Tjandraningsih, Indrasari (1995), Buruh Anak & Dinamika Industri Kecil (Child Laborers & Dynamics of Small-scale Industry), Bandung: Akatiga. Hodgkin, Rachel and Newell, Peter, UNICEF (1998), Implementation Handbook for the Convention on the Rights of the Child, prepared for UNICEF. ILO (1993a), From the Shadows to the Fore, Practical Actions for the Social Protection of Homeworkers in Indonesia, Regional Office for Asia and the Pacific, Bangkok: ILO. ILO (1993b), Practical Actions for the Social Protection of Homeworkers in Indonesia. October, Bangkok: ILO. ILO (1996), Out of the Shadows, Practical Actions for the Social Protection of Home workers in Indonesia, Regional Office for Asia and the Pacific, Bangkok: ILO.

288

Oey-Gardiner et al.

ILO (1998), ‘Employment Challenges of the Indonesian Economic Crisis’, Jakarta: ILO and UNDP. ILO (2000), Consultative Forum on a Strategy for Employment-Led Recovery and Reconstruction in Indonesia. Conclusions, Bangkok: ILO. Irwanto, Farid, dan Anwar, Muhammad, and Jeffry (1999), Anak yang Membutuhkan Perlindungan Khusus di Indonesia: Analisis Situasi (Children Requiring Special Protection in Indonesia: Situation Analysis), PKPM Unika Atma Jaya Jakarta, Departemen Sosial, and UNICEF. Korns, Alex (1993a), ‘Where are the Homeworkers in the Labour Force Survey? Indonesian Statistics for Manufacturing by Homebased Workers’ in Internationa Labour From the Shadows to the Fore, Practical Actions for the Social Protection of Homeworkers in Indonesia, Regional Office for Asia and the Pacific, Bangkok: ILO. Korns, Alex (1993b), Indonesian Statistics for Manufacturing by Home-based Workers in ILO, Practical Actions for the Social Protection of Homeworkers in Indonesia, Bangkok: ILO. Mboy, Nafsiah, Irwanto and Budhiharga Wiladi (September 1997), ‘Indonesian Experience with Child Labour: Looking for Best Practices’, draft paper for discussion. Sponsored by ILO-IPEC. Mehrotra, Santosh (2000), ‘Outsourcing of Manufacturing to Women and Children Homebased Workers’, Terms of Reference for the Unicef Study, mimeo, Unicef Innocenti Research Centre, Florence. Moedjiman, Muhamad (1992), ‘Proceedings of the Tripartite National Policy Workshop of Rural Women Homeworkers in Indonesia’, in Homeworkers of Southeast Asia: The Struggle for Social Protection in Indonesia, Bangkok: ILO. Regional Office for Asia and the Pacific, pp. 236–241. Moedjiman (1993), ‘Breaking through Statistical and Policy Invisibility: Pilot Monitoring Scheme for Homebased Workers’ in ILO, From the Shadows to the Fore, Practical Actions for the Social Protection of Homeworkers in Indonesia, Regional Office for Asia and the Pacific, Bangkok: ILO. Mulandar, Surya (ed.) (1996), Dehumanisasi Anak Marjinal: Berbagai Pengalaman Pember-dayaan, Bandung: Akatiga and Gugus Analisa. Mustain et al. (1999), Studi Kualitatif Tentang Pekerja Anak di Jawa Timur, Java Timor, UNICEF, Universitas Airlangga Surabaya: Airlangga University Press. Oey-Gardiner, M., Suleeman, E., Brodjonegoro, B., Tjandraningsih, I., Hartanto, W. and Wijaya, H. (2001), Indonesia Country Study Report, Women and Children Home based Workers in Selected Sectors of Indonesia (unpublished paper). Save the Children Indonesia (1996a), ‘The Urban Homeworkers of Jelambar Baru’, in Lazo, Lucita (ed.), Practical Actions for the Social Protection of Homeworkers in Indonesia, Bangkok: ILO. Save the Children Indonesia (1996b), ‘Occupational Health and Safety of Urban Homeworkers’, in Lazo Lucita (ed.), Practical Actions for the Social Protection of Homeworkers in Indonesia, Bangkok: ILO. Smyth, I. A. (1992), Growth and Differentiation in Rural Handicrafts Production in Selected Locations in West Java, The Hague: Institute for Social Studies and Akatiga Foundation. Suhaimi, Uzair (1995), ‘Home-based Workers in Tasikmalaya: Size and Characteristics’, Jakarta, an unpublished paper for the Biro Pusat Statistik and the ILO Regional Office of Bangkok: ILO. Tjandraningsih, Indrasari and Benyamin, White (1992), ‘Anak-anak desa dalam kerja upahan, Prisma, January, vol. 1, pp. 81–95.

Women and children homeworkers in Indonesia 289 Wijaya, Hesti and Santoso, Heru (1993a), ‘Village-based Action Research in East Java: Rural women Homeworkers in the Garments Industry’, in International Labour Organization, From the Shadows to the Fore, Practical Actions for the Social Protection of Homeworkers in Indonesia, Regional Office for Asia and the Pacific, Bangkok: ILO. Wijaya, Hesti and Santoso, Heru (1993b), Rural Women Homeworkers in the Garments Industry, in ILO, Practical Actions for the Social Protection of Homeworkers in Indonesia. Bangkok: ILO. Wirosardjono, Soetjipto (1984), ‘The Meaning, Limitations and Problems of the Informal Sector’, Prisma, The Indonesian Indicator, NO. 32, June, pp. 8–84. Wirutomo, Paulus (1992), ‘Homeworkers of Indonesia: Who are they, Where are they?’, in Homeworkers of Southeast Asia: The Struggle for Social Protection in Indonesia. Regional Office for Asia and the Pacific, Bangkok: ILO. World Bank (1993), The East Asian Miracle, Economic Growth and Public Policy. A World Bank Policy Research Report. Published for the World Bank, Oxford University Press. Yayasan Pengembangan Pedesaan (1993), ‘The Gondang Experiment’, in Lazo, Lucita (ed.), Practical Actions for the Social Protection of Homeworkers in Indonesia. Bangkok: ILO. Yayasan Pengembangan Pedesaan (1996), ‘The Gondang Experiment, Part II’ in Lazo, Lucita (ed.), Practical Actions for the Social Protection of Homeworkers In Indonesia, Bangkok: ILO

9

Subcontracted homework by women and children in the Philippines R. Rosario del Rosario,1 R. Pineda-Ofreneo and PATAMABA

Since the 1970s, the Philippines government has implemented policies aimed at shifting the export sector away from reliance on agricultural and natural resources towards labour-intensive manufacturing. As a result, the composition of exports was significantly altered and non-traditional goods replaced the traditional exports. This shift produced further changes in the Philippine economy. These include the adoption of outsourcing (i.e. subcontracting) to smaller production units and/or homeworkers, as a means of increasing cost-effectiveness. This practice became very popular among main export industries including textile, footwear and handicraft production. Homeworkers and workers from the small production units were usually in the informal labour market. As a result, they were deprived of basic safety nets and fair remuneration. Moreover given the ‘invisible’ nature of these activities, exposure to health risks and to unfair treatment by subcontractors was very common. This chapter summarises the findings of an extensive field study about the conditions of homeworkers in the Philippines conducted in the provinces of Luzon and Visayas. In particular it explores the conditions of women and children in four different sectors including pyrotechnics, okra production, home décor (Christmas balls and lights) and fashion accessories. The chapter is divided into five sections. Section 9.1 presents an overview of the informal sector based on official information and examines the emergence of the subcontracting system and its effect on the labour market. In addition, we present the results of the home-based work survey conducted in 1993 and insights from existing literature on homeworkers in the Philippines. Section 9.2 presents the methodology used in the field study including a description of the sample and survey design. Section 9.3 examines each of the selected sectors and includes a description of the subcontracting chain and organisation, production process and conditions of the homeworkers. Section 9.4 profiles the conditions of homeworkers and gives a comparative analysis of homeworkers with the control group (CG). Finally, Section 9.5 summarises the main findings and presents a list of detailed policy recommendations aimed at improving homeworkers’ conditions including both community-based local organisation and national government policies and interventions.

Homework by women and children in the Philippines 291

9.1

The informal labour force and homeworkers

According to official data, the Philippine labour force did not experience extreme structural changes during the 1990s and informal and formal workers are equally distributed in the labour market. Between 1990 and 1999 approximately 90 per cent of the total labour force remained employed (Table 9.1). On the other hand, data on the informal labour market suggest that the informal labour force experienced slight changes during the last two decades. Between 1980 and 2000, the share of informal workers in the total labour force decreased from 58 to 50 per cent (Table 9.2). In addition, the status of informal sector workers shifted from ‘unpaid family workers’ to ‘own account workers’. This is evident in the decreasing share of ‘unpaid family workers’ (from 21 to 12 per cent) and the slightly increasing share of ‘own account workers’ (from 36.9 to 37.4 per cent) between 1980 and 2000. Table 9.1 Philippines: labour force participation rate and employment status, 1990–99 (in thousands) Year

Labour force participation rate

Total labour force

Total employed

% employed

1990 1996 1997 1998 1999

64.5 65.8 65.5 66.0 65.8

24,525 29,637 30,265 31,278 32,000

22,532 27,442 27,888 28,262 29,003

91.9 92.6 92.1 90.4 90.6

Source: National Statistical Co-ordination Board. 2000 Philippine Statistical Yearbook. Table 11.2.

Table 9.2 Share of formal and informal sectors, 1980–2000 (%) Year

Percentage of informal workers in total labour force

Ownaccount workers

Unpaid family workers

Percentage of formal workers in total labour force

Total

1980 1984 1989 1997 1998 1999 2000

57.6 55.1 53.5 52.1 51.0 51.7 49.8

36.9 39.2 38.4 38.0 37.5 37.7 37.4

20.7 15.9 15.1 13.4 13.5 14.1 12.4

42.4 44.9 46.5 47.9 49.0 48.3 50.2

100 100 100 100 100 100 100

Source: Department of Labour and Employment, Yearbook of Labour Statistics and Current Labour Statistics, various periods. Notes ‘Own-account workers’ and ‘unpaid family workers’ are used as proxy indicators of the informal sector. ‘Own-account workers’ include ‘self-employed’ and ‘employers’.

292

Rosario et al.

However, different estimates suggest an increasing trend towards the informal labour force, as emerges from survey results that adopted the ‘residual approach’ method. This method is based on a comparison between the enterprise-based employment data from the Annual Survey of Establishments and the total household-based employment data (from the Labour Force Survey). According to this approach, in 1998 the informal sector accounted for 85–90 per cent of the total labour in the private sector, and from 80–91 per cent in the trade, services, and transport sectors (1993 estimates). Further studies (Chua, 1996 in Rosario del Rosario, 2001) reinforce the residual approach findings. According to Chua, in all economic activities, both agriculture and non-agriculture, the share of informal workers outnumbers the share of formal workers (Table 9.3). He sustains that in the 1990s the share of informal workers among all economic activities experienced a tendency to increase.2 As depicted in Table 9.3 several most important economic activities had a high concentration of informal workers. These sectors included agriculture (99 per cent), construction (89 per cent), wholesale, retail trade (88 per cent), and transportation (82 per cent). Although significant, the presence of informal workers was less predominant in finance and insurance (39 per cent) and community and social services (38 per cent).

Table 9.3 Distribution of formal and informal employment by sector, 1989–93 (%) 1989

1991

1993

Formal Informal Total Formal Informal Total Formal Informal Total All industries Agriculture, fishery, forestry Non-agricultural Industry Mining and quarrying Manufacturing Electricity, gas, water Construction Services Wholesale, retail trade Transportation, storage, communication Financing, insurance, real estate Community, social, personal services

23.13 2.11

76.87 97.89

100 100

19.10 1.87

80.90 98.13

100 100

18.88 1.50

81.12 98.50

100 100

40.46 33.84 39.73

59.54 66.16 60.27

100 100 100

33.14 35.13 43.26

66.86 64.87 56.74

100 100 100

36.24 29.38 39.61

63.76 70.62 60.39

100 100 100

47.59 99.73

52.41 0.27

100 100

40.82 95.96

59.18 4.04

100 100

37.73 97.88

62.27 2.12

100 100

14.99 31.82 21.07

85.01 68.18 78.93

100 100 100

14.05 32.24 8.54

85.95 67.76 91.46

100 100 100

11.20 35.24 11.96

88.80 64.76 88.04

100 100 100

18.96

81.04

100

15.94

84.06

100

17.88

82.12

100

70.43

29.57

100

60.86

39.14

100

61.15

38.85

100

59.13

40.87

100

51.85

48.15

100

62.28

37.72

100

Source: Adapted from Chua, 1996 in Rosario del Rosario (2001). Note * Based on the Modified Residual Approach.

Homework by women and children in the Philippines 293 Other studies revealed the key role of the informal sector in the economy by estimating its contribution to the national economy. According to the National Commission of Filipino Women, between 1987 and 1993, 45 per cent of GNP was produced in this sector. The emergence of the subcontracting system As already mentioned, during the 1970s, the Philippine government pursued policies aimed at promoting exports based on labour-intensive industries (Alba and Villamil, 2000; CRSS and CRC, 1996). As a result, the composition of exports was significantly altered. Non-traditional goods including semiconductors and electronic components, garments,3 handicrafts and furniture replaced the agricultural and resource-based traditional main export products including sugar, copra and copper. Subsequently the economic reforms aimed at outward-oriented economic policies and export-led growth implemented during the 1990s led to increasing demand from exporting firms. The increasing demand combined with the shift of main economic activities led to further changes in the production processes that is, subcontracting of homeworkers. Subcontracting was often adopted in industries that allowed for flexibility in the production process and did not include the use of machinery. These activities included the production of handicrafts, leather products, footwear, toys, houseware, fashion accessories and home décor. In contrast, outsourcing was not adopted in factory-based industrial activities such as the production of semiconductors and electronics, due to the required use of sophisticated machinery. The flexibility in labour arrangements through subcontracting was attractive for firms due to its cost-effectiveness. The ‘contract’ consisted of an informal agreement between the parties. By circumventing labour standards4 and subcontracting workers from depressed urban and rural areas, firms were able to reduce costs of production, the risk of losses due to fluctuations in production, and the risk of unionisation. In a study undertaken by Torres (1993) on Philippine establishments, the most frequently cited reason given by firms for employing non-regular workers was the fluctuating demand for products. The study also discovered that export-oriented firms tended to subcontract work more than those producing for the local market. This could be explained by the fact that production of export products is more vulnerable to fluctuations in demand (Esguerra and Esguerra, 2001). Another factor that contributed to the expansion of subcontracting came from the supply side. Given the low skills required in the activities mentioned, subcontracting represented an employment opportunity for those with low levels of education and for those temporarily unemployed (Diaz, undated). Homeworkers Home-based work first manifested itself as a fast growing phenomenon in the 1970s becoming most prominent in the fringes of the urban centre of Metropolitan Manila, as well as in the provinces of Bulacan, Rizal, Laguna, Cavite and Batangas. The trend continues to the present and has expanded to the growth centres of the Visayas (Cebu) and Mindanao (Davao).

294

Rosario et al. Table 9.4 Distribution of home-based workers by economic activity and location, 1993 (%) Home-based workers

Percent

Rural

Urban

Total

Sub-sectors (% of hw) Agriculture/fisheries Manufacturing Services

100.0 0.5 97.8 1.7

43.1 95.5 78.6* 3.3*

56.9 0.5 21.4* 96.7*

100 100 100 100

Source: NSO, 1993. Note * Adjusted estimate based on certain regions not entirely national.

The Philippines is one of the few countries that has conducted a national survey on homeworkers. According to the survey, the number of homeworkers increased from 1.2 to 2 million of homeworkers between 1993 and 1995 (of a total labour force in 1996 of 29 million in the Philippines). These were mostly concentrated in Central Luzon (89 per cent) (Region III) and to a lesser degree in Central Visayas (Region VII) while the highest proportion of subcontractees was concentrated in Metro Manila. Homeworkers were equally distributed between urban and rural areas with approximately 43 per cent of homeworkers in the urban areas and 57 per cent in rural areas (Table 9.4). The majority of homeworkers were employed in manufacturing (98 per cent) and to a lesser degree in the services (2 per cent) and agricultural (1 per cent) sectors (Table 9.4). In manufacturing, the highest concentration of homeworkers was found in the production of textiles and garments with 54 per cent of the total homeworkers. In agriculture (including fishing), homework seems to be product-specific.5 However, information on the nature and extent of subcontracted homework, especially in agriculture and services, is not as readily available as it is in the garments sector (Del Rosario, 2000). Other manufacturing production in which homeworkers are involved include beverages and tobacco, paper products, chemicals, non-metallic and basic metal, jewels, toys and stationery/office supplies (NSO, 1993)6 (for regional and sub-sectors distribution see Table 9.5). Indeed, the proliferation of homework contributed to the increased participation of females in the age group 25–44 in the labour market, and to some extent children in paid work. According to the National Statistical Office (NSO), approximately 79 per cent of homeworkers in the five regions covered by the survey were female. These were mostly concentrated in manufacturing production and wood product manufacturing. According to the NSO findings, children were also involved in these activities. Indeed, 2.1 per cent of all homeworkers belonged to the age group 14 years old and below. This means that approximately 2,400 children, in the age group 10 years old and younger and 22,783 children aged 10–14 were engaged in homework. Finally, while the majority of women and children were engaged in the production stages of homework, more male workers were engaged as subcontractors. Thus, there seems to be a gender division of labour.

Homework by women and children in the Philippines 295 Table 9.5 Distribution of home-based workers by primary activity and by region, 1993 (%) Activity

All areas

Metro Manila

Central Luzon

Southern Tagalog

Bucal

Central Visayas

Total Agri., fishery and forestry Food, beverage and tobacco Textile product mfg. Wood product mfg. Paper product mfg. Chemicals mfg. Non-metallic mfg. Basic metal mfg. Jewellery mfg. Toys/doll mfg. Stationery/office supply Other manufacturing Services Not reported

100.0 0.5 1.9 53.6 23.6 2.6 0.8 0.6 0.8 4.7 0.3 0.1 9.1 1.6 0.6

100.0 0.1 5.6 46.4 9.1 6.2 5.3 1.1 3.9 0.1 1.3 1.3 16.9 2.8 0.1

100.0 0.2 1.2 81.2 12.1 0.2 0.4 0.2 — 0.2 0.4 — 2.1 1.9 —

100.0 0.2 1.0 61.0 28.3 4.9 0.3 0.3 1.2 — 0.1 — 1.4 1.6 0.7

100.0 — 2.5 73.5 15.1 6.6 — — 0.5 0.1 — — 0.1 1.8 0.5

100.0 1.3 1.7 15.6 40.4 0.3 0.2 1.2 0.4 15.9 — — 22.1 1/0 1.4

Source: NSO, 1993. Note Details may not add up to total due to rounding.

Insights from recent research Recent studies show that the Asian financial crisis in 1997 had a negative effect on homeworkers.7 The specific impact depended on many factors, including employment status (whether subcontracted, self-employed or both), geographic location (whether rural or urban) and type of industry. The garment industry provides an example of the dramatic impact of the financial crisis. Before the crisis, the garment sector experienced a marked deceleration in growth and consequently less subcontracted work was being put out to traditional embroidery communities. Moreover, due to the introduction of computer embroidery machines in garment factories, outsourcing of embroidery became unnecessary. Thus, subcontracted homework was reduced to low-paying and time-consuming tasks. This was the common experience of homeworkers in Malibong Bata, Pandi and Bulacan. A similar result was found in a survey of 160 respondents from San Vicente, Angono, Rizal; Santo Angel, Sta. Cruz, Laguna; Balingasa, Quezon City; and Taal, Malolos and Bulacan. This study showed the increasing financial burden of home female subcontracted workers engaged in sewing, embroidery, papiermache and bag-making due to the crisis. Their economic situation had worsened due to stagnant piece-rates and declining job orders given the large pool of labour and the cutthroat competition in both global and local markets. In addition, women faced economic difficulties due to the increasing prices of basic goods and services. Female homeworkers found themselves burdened with the

296

Rosario et al.

responsibility of cutting down expenses, incurring debts and undertaking secondary paid work to increase their sources of income. Deterioration of physical and social infrastructure further compounded their problems due to constant flooding and also problems with peace and order combined with economic hardship (Pineda-Ofreneo et al., 2001). Policies on homeworkers In 1997, the Department of Labour and Employment (DOLE) responded to the abuses and ambiguities attendant to subcontracting by issuing the Department Order – D.O. No. 10 Series of 1997 – an amendment to the Labour Code. The amendment recognised the principle of ‘business flexibility but not at the expense of labour’. Besides recognising subcontracting as a legal labour arrangement, the code included relevant policies aimed at ensuring fair treatment for homeworkers and fair conditions of employment, wage rates, occupational safety and health standards, and the right to self-organisation, security of tenure, and social and welfare benefits. In addition, this code required contractors and subcontractors to register with the regional offices or the Bureau of Labour Relations of the DOLE. Unfortunately, when a new President (Macapagal-Arroyo), came to power in 2001, D.O. No. 10 was withdrawn in reaction to complaints from formal trade unions. A similar initiative emerged from PATAMABA, an NGO working with homeworkers in the Philippines. They originally supported the establishment of Code 10 and strongly advocated the implementation of Department Order 5 (D.O. No. 5) that included special provisions in favour of homeworkers and further details about their rights.8 In addition, PATAMABA persuaded the government to ratify the ILO Convention 177 (The Homework Convention 1996) and the accompanying Homework Recommendation 1996, which had similar provisions to Department Order No. 5. However, like the pre-existing Labour Code, this Department Order has not been implemented effectively.

9.2

Research method

Survey design and data collection The methodology used in the current survey consisted of qualitative and quantitative instruments. The qualitative data was collected through focus group discussions (FGDs) and case studies where children and women were involved. In each site, the research team conducted one FGD and two case studies. In addition, a member of the research team conducted a key informant interview in each of the selected sectors. The quantitative data was collected through a survey addressing a sample of households in the communities concerned. The selected communities were from Luzon and Visayas.9 The sectors studied were pyrotechnics production in Bulacan, okra production and packaging in Tarlac, home decor (Christmas lights and Christmas balls) production in Rizal,

Homework by women and children in the Philippines 297 and fashion accessories production in Cebu. The Tarlac and Bulacan sites were rural, while Rizal and the site for fashion accessories in Cebu were in the suburbs of the two major Philippine cities – Metro Manila and Cebu City respectively. Prior to the selection process, the concept of homework was redefined to fit the reality of the selected sectors. On the basis of the definition provided by the ILO Convention of 1996,10 the concept of ‘home’ was expanded to other places including: ●

●

any part of the entire area of a production community, including the employer’s workplace or home;11 ‘home’ is not a locality but a unit of hierarchical obligational kinship, including contractual relationships that share income.

The sample consisted of households in which women and children were involved in homework. In order to establish a reference point a CG of households that were not involved in homework yet were from the same communities was surveyed (Table 9.6). The survey was designed on the basis of the common core questionnaire. There were four different sets according to the type of respondent (women and children involved in homework and women and children from the CG). In the first two sets, the survey included specific questions directed to homeworkers. The second sets included general questions about the household. All the questionnaires included a section on general demographic characteristics (age, education, number of economically active members per household and marital status) of the household, that were completed by the interviewer.12 The second section requested information about economic activities such as number of economically active household members, and annual income. The third section included questions directed to homeworkers about reasons for involvement, benefits and disadvantages, and general conditions. The last section aimed at discovering parents’ perception about the children’s schooling and labour in the case of adult women, while the children’s survey included questions regarding their attitude towards homework and household duties in the case of the CG.

9.3 Homework sectors The selected sectors/clusters were located in depressed areas with high levels of unemployment. There was limited access to basic services such as secure housing, water and toilet facilities. Thus, health, safety and environmental hazards were endemic in the communities.13 Homeworkers were located at the lowest level of the value chain that entailed from three to six levels of intermediaries. At the highest level of production, contractors assigned a written order of production to subcontractors or job suppliers. Simultaneously, subcontractors hired homeworkers for the production of orders under a mutual trust agreement. Due to the informality of the contract, homeworkers were not entitled to basic social security, health benefits or market level wages.

Rizal Prov., Luzon Kalayaan in Angono, Rizal Prov., Luzon San Vincente in Angono, Rizal Prov., Luzon San Vincente in Angono, Rizal Prov., Luzon Sta. Maria, Bulacan, Luzon Sierra, Conception, Tarlac Prov., Luzon San Roque, Talisay, Cebu Prov., Visayas

Home décor Metalcraft

30 135

173

34

35

10

20

36 6

41

43

44

—

—

45 —

38

11

9

9

—

—

9 —

66

30

—

—

—

—

36 —

hw

CG

Total

hw

Urban

Households surveyed

20

11

—

—

—

—

9 —

CG

Notes Dash indicates that certain information was not collected in the survey or data were not comparable. FGDs  focus group discussions. CS  case studies. hw  homeworkers. CG  control group or non-hw household. * Considering separately the one for women and the one for children.

Source: Philippines Country Study Survey, 2001.

Fashion accessories Sub-total

Okra

Pyrotechnics

xmas balls

xmas lights

Location

Sector

69

—

34

35

—

—

— —

hw

Rural

18

—

9

9

—

—

— —

CG

6

1

1

1

—

1

1 1

Total

3

1

—

—

—

1

1

Urban

Number of FGDs

3

—

1

1

—

—

1 —

Rural

8

2

2

2

—

—

2 —

Number of CS

Table 9.6 Surveys on homeworker households in the Philippines: sectors, location, number of FGDs, case studies and households surveyed

Homework by women and children in the Philippines 299 The subcontractors fixed the level of income and payments. Homeworkers were paid on a piece-rate, according to the sector and the specific product, and on a weekly or monthly basis. Delayed payments were very common among all sectors. The only exception was the case of homeworkers from the okra sector who received payments on time. The nature of the relationship between the subcontractor and the homeworker varied according to the sector. In the majority of the sectors the relationship was based on patronage or kinship ties. In some cases this favoured homeworkers with special treatment and provision of benefits. For example in the fashion accessories sector, thanks to kinship ties, the homeworker had access to a credit line based on receivables. There were also cases where the relationship between homeworker and subcontractor was purely a business agreement. This was the case for Christmas decorations and metal crafts in the home décor sector, and in the category of migrant workers in pyrotechnics production. Currently there is an additional determinant in the relationship between homeworker and subcontractor, in favour of the former, that is, membership of the PATAMABA network. For example, thanks to support from a member of the PATAMABA network in the okra sector, workers were able to get fairer rates for their production.14 The details of the benefits of this network and the share of workers affiliated will be discussed in Section 9.4. Homework in the Philippines is mainly a female activity. Among all sectors women were predominantly involved as homeworkers and to a lesser degree as subcontractors. Children as well as men were also involved although to a lesser degree. In the majority of the sectors/clusters male households were rarely involved in homework activities. Men usually worked in low-paying occupations such as farming, street vendors, carpentry and construction. The only sector where men were involved was in pyrotechnic production. Pyrotechnics production The production of pyrotechnics is a traditional activity that takes place at a rural municipality in Bulacan province in the island of Luzon. Pyrotechnics are exclusively sold in the domestic market during the Christian New Year period, Chinese New Year celebrations and other festive occasions. The subcontracting chain starts with middlemen (buyers/distributors) from Bulacan and Laguna (in Luzon), and Mindanao that sell the product in the local market. They assign the production orders to small workshops/producers that in turn hire homeworkers. Homeworkers are mostly hired during the high season when workshops are unable to fill the orders themselves. During the off-season, small producers hire homeworkers for specific aspects of production such as folding and fuse covering. Since subcontractors do not provide the raw materials, homeworkers are forced to borrow from moneylenders or the raw material suppliers to acquire them.15 In the pyrotechnic sector there are two categories of homeworkers: local workers from the same community and migrant workers. Migrant workers are hired exclusively during the high season (September–December) and receive free room

300

Rosario et al.

and board for this period. Local workers, however, do not receive any special treatment. Only in exceptional cases when profits have been significant do homeworkers (migrant and local) receive a bonus. The production process is simple and does not require a high level of skill. It consists of the fuse preparation, filling triangular papers, and packaging. There is a distinct gender division of labour in the production process. Men are in charge of the dangerous tasks such as mixing the chemicals while women are responsible for the fuse preparation and folding. Children from the age group of 12–15 years old are also involved and work under the same gender pattern. While girls and children below 12 performed safer indoor stages of production such as folding, labelling and packaging, boys handled the more dangerous outdoors or factory component of the production process. As mentioned above, work is seasonal. The average number of hours that women and men work during high seasons, is 18 hours per day (Tables 9.7a and b) and 8–9 hours per day during off-season months (June–August). Children’s working hours also vary according to the seasons. On average, they work 6 hours per day, causing a negative impact on their studies. As in all the home-based work sectors, workers are paid on a piece-rate basis. However, the rate varies according to the production activity. For instance, women receive P. 5–P. 5.5 per 1,000 folded papers16 or P. 10 per bundle of fuse materials covered with paper. Men work exclusively with the chemical mixture and receive P. 75 per day – much higher than the average daily earnings of women Table 9.7a Working hours per day full time by seasons (number of hours)

Fashion accessories Pyrotechnic production Home decor Okra

Average working hrs per day/worker

Average high season working hrs per day/worker

Average lean season working hrs per day/worker

15

18 18 7 10

3–6 6 3 1–3

8

Source: Philippiness Country Study.

Table 9.7b Home décor – working hours by gender and season Sector

High season Male

Metal craft Christmas balls Christmas lights

Low season Female

Male

Female

Adult

Child

Adult

Child

Adult

Child

Adult

Child

7 6 3

6 4 6

8 7 9

4 5 5

2 2 1

1 2 3

3 2 2

1 1 2

Source: FGDs, Case studies, key informant interviews.

Homework by women and children in the Philippines 301 (i.e. P. 25 per day). During the high seasons earnings might reach an average of P. 50–100 per day. Hence, although homeworkers play the main role in the production chain, they receive an insignificant share of the final price. While distributors earn a net profit of P. 3,263 for a P. 6,000 5-star firecracker, homeworkers earn P. 255.17 Home décor (Rizal) Home décor includes the production of metal crafts, Christmas balls and lights that are mainly exported to the United States, Europe, Australia, Malaysia, Hong Kong and Japan. Some of these products, such as Christmas balls, are also produced for the domestic market (Rosario del Rosario, 1998). The subcontracting chain in home décor production consists of export companies, small community-based subcontractors and homeworkers. Export companies assign the production orders and sample designs to the small subcontractors who in turn outsource the production to workers in their own workshops, or to homeworkers. Subcontractors might be located in the same community as in the case for metal crafts, or in adjacent communities as was the case for Christmas balls.18 As opposed to the subcontracting chain in pyrotechnic production, export companies and subcontractors provide homeworkers with the raw materials. Female workers and children are involved in the production of home décor products. Male heads of households are commonly not employed in this sector and work outside the home as migrant workers in activities such as carpentry, construction or driving. The production is seasonal thus the hours worked vary according to the season. The average working hours are 7 hours per day although in lean seasons this could be only 2–3 hours per day (Table 9.7). During the high season men also engage in home décor production, and work approximately the same hours as women. The only sector where women worked more hours than men during the high season was in Christmas lights production19 (Table 9.7). As for children, boys and girls worked approximately the same hours (5–6 hours) during the high season and only 1–2 during the low season (Table 9.7). Children usually work in the workshops after school. The production processes of the three types of home décor are simple and do not require a high level of skill. In the three sectors, homeworkers receive the patterns and materials from subcontractors and undertake the production at home or in workshops. As expected, the share that homeworkers received of the final price is very small. This was the case for the three types of home décor products. The sector in which homeworkers received the highest share of the final price was the Christmas balls sector. PATAMABA workers engaged in Christmas balls production estimated the production cost per ball at P. 17.41.20 The final retail price set by the exporter is P. 25 per unit or P. 100 per three units. Hence the exporter’s profit was almost 50 per cent of the final price, while homeworkers received approximately only 25 per cent of the final price. Subcontractors’ earnings came from mark-up prices. In the case of metalcraft subcontractors, for instance, earnings ranged from P. 5–30,21 which accounted for 23 per cent of the final price.

302

Rosario et al.

Okra production The production of okra in the Philippines is seasonal and takes place in Tarlac. The production is exported exclusively to Japan. Subcontracting in this sector consists of six levels of intermediaries at a local and foreign level. In Japan the management export company contacts subcontractors/ wholesalers/retailers. These assign the production orders to Philippine subcontractors. In the Philippines, the main subcontractor22 might have additional intermediaries (a team leader, two overseers and several agricultural workers) down to the lowest end of the value chain where homeworkers are located. The main subcontractor provides homeworkers with a workplace or ‘sorting houses’, basic tools and raw materials. Basic tools of production include fertilisers, pesticides, a plough and water pump used in cultivation and other elements such as cartons, plastic and containers required for storage and transportation. Production of okra consists of sorting, quality control and packing and it involves all members of the household including women and children. The harvested okra is brought to the sorting house in crates for quality control, bagging and packing. Once picked and cleaned, these are delivered to the nearest town warehouse. The washing, sorting and packing of okra in Tarlac, is confined to women and girls. The more difficult and strenuous task of picking okra in the field is confined to adult females although children of both sexes are likewise involved in this task. Most of the homeworkers in this sector are women. Male workers are rarely involved although they outnumber females in the field-based cultivation of okra. However, field processes do not exclude females who are particularly involved in the picking stages. During the high season average working hours are 10, and 1–3 hours during the lean season (Table 9.7). During the lean season, farming families plant mainly rice and some vegetables. Others (mainly males) undertake contractual employment. Some young girls may seek employment elsewhere as domestic helpers. Wage levels vary depending on the process involved in okra production, but are subject to the quota system, except in the case of daily workers. For example, a homeworker responsible for ploughing receives P. 300 per day, while homeworkers in charge of applying fertilisers earn P. 100.23 Field workers, both adults and children, (i.e. weeders/pickers) receive P. 80 on a daily basis. Sorters are also paid on a daily basis (e.g. managers, quality controllers). In general, piece-rate workers complain about their low wages, which are not enough for daily subsistence. The only benefit workers are entitled to is the free provision of rejected okra.

Fashion accessories The production of fashion accessories takes place in Cebu. It is primarily a female activity where both female adults and girls are involved.24 The production is seasonal and mainly exported to Japan, Hong Kong, Europe, Taiwan, Korea and Canada. There are various levels in subcontracting for fashion accessories production. The exporter/buyer/contractor is often based in a factory that employs workers to

Homework by women and children in the Philippines 303

Retailers abroad

Final consumer abroad

(final consumer)

Importers abroad

Exporters/contractors (in the Philippines)

Quality controller

Purchasers

Expediters

Agents

Suppliers/job outers/subcontractors

Job outers/subcons/ homeworkers (in the community)

Hired daily workers

Women, girls, boys, men homeworkers

Retailers (local consumers)

Figure 9.1 Subcontracting chain in fashion accessories.

produce and pack products (Figure 9.1). Exporters/buyers/contractors employ quality controllers, purchasers, shippers and agents to deal directly with suppliers and acquire the raw materials. The agent contacts subcontractors and provides them with the raw materials, which are given to the homeworkers for production. There are many items of fashion accessories produced including buttons, hair clips, pendants and others. The materials used vary and consist of shells, wood, resin, and beads of shells or wood. The production process is simple and consists of cutting the raw wooden log or plank by using machine woodcutters. Then each small wooden piece is pierced through with a machine drill (which girls as young as eight have also been observed to operate). Once drilled, these are strung together with a plastic thread (usually by women and children, especially girls). Then the wooden pieces are polished all together by machine (usually operated by

304

Rosario et al.

males), until they are smooth and look like regular beads. The strands may be sold as such, or are first made into necklaces or bracelets by attaching closures to them (mainly the job of women and children). The workload depends upon various factors including the specific circumstances of the family or household (e.g. the number of children, their ages, the economic situation, whether the mother is a sub-subcontractor or not, whether the father is employed/semi-employed/unemployed, whether the household is extended or not, whether the children are going to school or not). The average working hours are 15 hours per day. During the high season working hours might increase to 18 while in the low season, homeworkers only work from 3 to 6 hours (Table 9.7). As in the other sectors, homeworkers are paid on piece-rate or according to the stage of production accomplished. In Cebu, payment for those handling machines varies, and it would appear that drillers (who are mostly females) get less than polishers (who are mostly males). Drillers earn an average of P. 200–300 a week, while polishers get about P. 1,000 per week.

9.4

Profile of homeworkers

Homeworkers’ households The following section identifies the conditions of homeworkers in the sectors studied and highlights the main differences and similarities found in comparison with households from the CG. Besides having larger household sizes, homework households also have a larger number of economically active members in comparison with CG households. On average the size of a homework household is 6.1 which is greater than the 5.9 members of CG households. Moreover, approximately 50 per cent of homework households have economically active members that contribute to the household. In the case of CG, only one-third of household members are economically active (Table 9.8).

Table 9.8 Characteristics of the CG and homeworker (hw) households Sector

Home décor Pyrotechnics Okra Fashion accessories All

hw

CG

Household Number Number of Household Number size of eonomically size of children active hh children members

Number of economically active hh members

5.9 6.8 6.8 5.0

2.5 2.7 3.2 2.2

4.2 3.1 3.3 3.1

5.9 5.6 6.0 4.8

2.2 2.4 2.5 2.4

2.2 2.2 2.4 3.1

6.1

2.6

3.4

5.9

2.4

2.5

Source: Philippines Country Study.

Homework by women and children in the Philippines 305 Table 9.9 Average annual household income by sector (hw and CG households) compared to average household income at regional level (P.) Sector

hw

CG

Regional*

Region

Home décor Pyrotechnics Okra Fashion accessories

119,921 48,654 54,490 49,263

107,139 70,761 136,188 90,037

161,963 151,449 151,449 91,520

Southern Tagalog Central Luzon Central Luzon Eastern Visayas

Source: Philippiness Country Study. Note * Family Income and Expenditures Survey.

The average annual income level of homework households is lower than the regional and CG income level. Despite having a greater number of household members working, the level of income in homework households was lower than that of CG households (Table 9.9). Only homework households from the home décor sector earned a higher annual income than the CG households’ annual income.25 Among the other sectors the difference in income level was larger. The largest difference was found among okra production workers whose annual income level was approximately a third of the annual income earned by CG households. Similarly, the income level of pyrotechnic homeworkers was almost half of the income level of CG households. Comparison of the income level of homework households with the regional level (Table 9.9) demonstrates more clearly the homeworkers’ adverse economic situation. First of all, households from all sectors/region, both homework and CG households, are below the average income level of their region. However, the gap between the income level of CG households and the regional income level is not as large as the gap between the latter and homework households. The level of income of pyrotechnic and okra homeworkers was barely a third of the regional income level. Hence, by and large, homeworkers from the sectors mentioned face the worst economic constraints. Homeworker households and women Homework in the Philippines is mainly a female activity. Half of the total number of workers were women of which 82 per cent were engaged in economic activities. Of those economically active, 90 per cent were engaged in homework (Table 9.10). On the contrary only 45 per cent of the total number of workers were women among CG households of which 67 per cent were economically active. The homework female workers aged between 35 and 44 years were 56 per cent of the total. The share becomes 71 per cent considering the age between 35 and 54 years. The age distribution pattern is quite similar in all 6 areas except for Bulacan where a significant proportion of homeworkers (23 per cent) belonged to the younger age group of 25–29. Almost all of the respondents were married and 88 per cent live with their husbands while the rest are either separated or are widowed.

306

Rosario et al.

Table 9.10 Percentage of female workers among homework (hw) and CG households (%) Sector

Home décor Pyrotechnics Okra Fashion accessories All

hw

CG

Women working as % of total workers

Women working as % of total women

Women working in hw as % of women

Women working in hw as % of women working

Women working as % of total workers

Women working as % of total women

44.5 42.9 41.5 48.3

87.8 80.8 89.8 68.9

73.0 76.9 79.6 63.9

83.1 95.2 88.6 92.9

52.6 56.3 40.9 36.4

66.7 90.0 69.2 54.5

44.2

81.8

72.9

89.1

44.4

66.7

Source: Philippines Country Study.

Entering the labour market can be considered a sign of women’s empowerment. However only a fifth of female homework respondents considered homework as a means to be ‘economically independent from the husband’. This may be indicative of the constraints that married female workers face with regard to their participation in the work market in general. During their early childbearing years, younger married females are confined to the care of infants and toddlers. Without a substitute for direct parental care, it is unlikely that they will participate in the labour market including in homework activities. On the other hand, as the children grow up, mothers attain greater flexibility in combining reproductive work (childcare and other household chores) with economic activities. Older children require less parental care compared to infants and toddlers and can also be assigned household chores including caring for their younger siblings and assisting the mother in homework activities. In the FGD women emphasised that they needed the homework. While all the women are saddled with domestic duties aside from their homework, all of them admit that without their income from homework, their family situation would be worse. They are glad to be engaged in economic activities even if this means an additional burden, especially during times of economic difficulties when they need to take on greater workloads. The most frequently mentioned reasons for engaging in homework were to ‘meet the needs of the family’, ‘to supplement the earnings of the husband or family’ and ‘to be able to earn while being close to home, children, and community’. There are additional factors that influence the household’s decision to engage in homework such as lack of employment opportunities. A significant proportion (69 per cent) of homeworkers in Bulacan who were engaged in the hazardous occupation of pyrotechnics production stated that there were ‘no other opportunities available’, or that pyrotechnic production was ‘the best opportunity available’ for them to earn some income.

Homework by women and children in the Philippines 307 Table 9.11 Education attainment by level of school and by sector Sector

No school Less Primary Completed Completed Total At least than school secondary tertiary primary primary level and level school* vocational

Home décor Pyrotechnics Okra Fashion accessories All

0.0 0.0 0.0 3.3

18.1 19.2 35.4 14.8

44.4 44.2 50.0 50.8

37.5 34.6 14.6 31.1

0.0 1.9 0.0 0.0

100 100 100 100

81.9 80.8 64.6 82.0

0.9

21.0

47.2

30.5

0.4

100

78.1

Source: Philippiness Country Study. Note * Sum of those who have completed primary secondary/vocational and tertiary level of school.

The majority (78 per cent) of homeworkers among all the sectors has completed at least primary level schooling (Table 9.11). However only a third has completed secondary school and a negligible proportion has been able to reach tertiary level of education. Among all the sectors, workers from the okra production had the highest percentage of workers with a level of education below primary (35 per cent). On the contrary, workers from the home décor production sector had the highest percentage of workers to have completed at least primary level education. Homeworker households have different patterns of expenditures when coping with economic difficulties. The comparison of expenditure patterns between homework and CG households in critical economic situations demonstrates that external shocks have greater detrimental effects on homework households. CG households cope with crises by reducing expenditures on food and clothing. Similarly homework households have a tendency to economise primarily by reducing expenditure on food. However, they also economise by pulling children out of school, reducing expenditure on education, and by increasing the workload of the family particularly that of the female members. Hence the major difference is that homework households are more prone to reduce expenditure on education. Child labour and gender issues According to the survey results, by and large the incidence of child labour is greater among homework households (Rosario del Rosario, 1994, 1996). Only 14 per cent of the total of CG children (age group 11–14 years old) were working, compared to 53 per cent of total children among homework households from the same age group (Table 9.12). Another main distinction is the predominance of child labour among children of older ages (11–14 years old) compared to children in the younger age group (5–10 years old). This was the case for both CG and homework households. An overwhelming majority of the respondents consider the education of their children a priority. About 94 per cent of the respondents expressed their preference

308

Rosario et al.

Table 9.12 Share of children working by age group in homework (hw) and CG households (I  5–10; II  11–14) Age group/ sector

Home décor Pyrotechnics Okra Fashion accessories All

Percentage of children working-hw households

Percentage of children working in CG households

Homeworker households

I

II

I

II

I

II

Tot.

I

II

16 33 27 10

56 63 48 49

0 0 0 10

0 0 17 25

4 13 10 4

16 17 13 17

29 45 33 28

11 33 24 10

50 63 45 49

22

54

2

14

31

63

34

20

51

Number of children working in hw

Percentage of children working in hw

Source: Philippines Country Study.

for children being able to go to school. Only a few respondents (8 per cent) stated they would rather have children as income earners and contributors to household needs. There seems to be no significant difference in the aspirations that parents have for their daughters. Compared to sons, a slightly lower percentage (59 per cent of the total number of respondents) said that they also wanted their daughters to have a good education. Despite the results of the FGDs where parents did not appear to be gender biased, girls engaged in economic activities are more subject than boys to the double burden of working and assisting in the household chores. Moreover, when the household experiences financial difficulties, girls are the first to be encouraged to quit school in order to reduce expenditure. This was mostly the case for girls in the fashion accessories sector. Besides being responsible for the household chores, they devoted half of the day to homework. As a result many of them interrupted their studies. Girls’ participation in economic activities varies according to the activity and the location. There seems to be a preference for girls in indoor activities such as in the production of fashion accessories, while there is a preference for boys to handle more dangerous outdoor work, such as the production of pyrotechnics. Table 9.13 shows that at the younger ages (6–11 years old) the share of girls (30 per cent) working in homework was slightly higher than the share of boys (25 per cent). In particular, girls’ presence, in both the younger and older age groups, was predominant in the fashion accessories sector. On the contrary, the presence of boys was higher in the pyrotechnic sector for both age groups. In the older age group, 62 per cent of boys were working while this was the case for 46 per cent of the girls (Table 9.13). Thus, girls’ engagement in homework decreased at older ages.

Homework by women and children in the Philippines 309 Table 9.13 Percentage of children working* by gender and age group Sector

Home décor Pyrotechnics Okra Fashion accessories All

Age 6–11

Age 12–14

Boys

Girls

Boys

Girls

27.8 41.7 20.0 5.0

14.3 35.0 42.9 27.3

73.3 66.7 58.3 46.2

55.6 66.7 45.5 61.5

24.4

29.8

61.2

57.1

Source: Philippiness Country Study. Note * Working and working and studying.

Homework affects health status of workers including children. Almost half of the respondents (46 per cent) felt that homework had harmful effects on their physical well-being. The most common ailments reported were lack of sleep, loss of weight, fatigue and body pain. There is strong evidence of a relationship between working conditions and the health of homeworkers. In the okra production sector, workers suffered from ailments derived from the type of work in the field. Ailments included headaches, blurred vision, dizziness and even lung irritation due to excessive exposure to the okra cover.26 Another concern was the deterioration of living conditions due to the introduction of homework activities. On the other hand, a third of the respondents (37 per cent) felt that homework had no bad effects whatsoever on the household. Homeworkers’ organisations Homeworkers’ organisations have a positive impact. Communities with organised groups have higher levels of awareness, solidarity and co-ordinate group action to solve common problems, especially those connected with livelihood. As mentioned above, membership of the PATAMABA network had a positive impact on the nature of the relationship with the subcontractor as well as on the degree of cohesion among homeworkers. The majority of workers from the okra and fashion accessories sectors were affiliated to the PATAMABA network (Table 9.14). Interestingly okra workers were the only group that received payments on time from subcontractors; this might be attributed to support from and organisation by the network. Overall a third of the total of homeworkers among the four sectors were members of the network. Another benefit that the PATAMABA network provided was access to capital and loans. For example, thanks to PATAMABA membership a female homeworker in the fashion accessories sector had access to capital and organised her own enterprise for fashion accessories production.27

310

Rosario et al. Table 9.14 Percentage of workers that are members of the PATAMABA network Sector

Member

Home décor Pyrotechnics Okra Fashion accessories All

33.3 2.6 54.3 63.6 36.8

Source: Philippines Country Study.

According to the opinions expressed during the FGDs homeworkers main concern was to have better working conditions. This implies having health and safety nets, access to basic social services and sanitation facilities. One of their main concerns was also the lack of training and skill improvement courses especially for women. In terms of children’s education, they emphasised the need for more educational opportunities and access to schools.

9.5

Conclusions and recommendations

Conclusions Subcontracting homework is characterised by being an informal verbal agreement between subcontractors and homeworkers. This is a commonly used arrangement among various types of industries in the Philippines. Through this system contractors are able to reduce production costs and can adapt easily to demand fluctuations. On the other hand, homeworkers are deprived of basic safety nets and fair remuneration given the informal nature of this type of arrangement. There are many unresolved health and safety issues, especially in firecracker production. Given the importance of some of these sectors to the national economy, the government has launched important initiatives in terms of legal amendments in favour of homeworkers and national surveys. However, so far little has been achieved. The subcontracting chain involves between three and six intermediaries. Given that homeworkers are located at the lowest end of the value chain the share of remuneration is insignificant. The main determinants of the relationship between subcontractor and homeworker are kinship ties and membership of PATAMABA that facilitated trust in the mutual agreement. In the cases where homeworkers were migrants, the relationship was purely a business agreement. With the exception of fashion accessories and pyrotechnic production all the other products are exported to foreign markets. All workers are paid by piece-rate or activity performed, and homeworkers spend an average of 18 hours per day in homework activities during peak seasons. In low seasons, working hours range

Homework by women and children in the Philippines 311 from one to six per day. In addition, homeworkers are subjected to delayed payments and insignificant shares of the final price at which the product is sold in the domestic and foreign markets. The communities in which homeworkers live had poor infrastructure and sanitary conditions as well as high unemployment levels, and no major distinction was found among homeworkers from rural and urban areas. On average the education level did not go further than elementary education, which enables them to be involved in homework activities that do not require a high level of skills. The study found a predominance of women in homework. FGDs supported this finding since female households prefer homework because it allows them to earn additional income while taking care of domestic chores. In addition, their involvement in homework provides additional income for the household although males are still providing the greater share. Few males were involved in homework and are mostly in charge of heavier duties such as mixing chemicals in the pyrotechnic sector and performing the heavy duties of the okra harvest. Children are required to assist their mothers in the tasks involved in homework. This involvement negatively affects children’s health and school enrolment. The share of working children was greater for older age groups, which implies that the engagement of children in homework might have a negative impact on their education. The income level of homework households was lower than the income level of CG households and much lower than the regional income level. Expenditure patterns in times of difficulties are different for CG and homework households. While homework households tend to pull children out of school, CG households survived by relying on additional work by male household members. Recommendations In consultation with NGOs, people’s organisations, and other civil society stakeholders, there is a need to adjust macro-economic policies and budgetary priorities in order to address poverty. Government agencies Through appropriate mechanisms, government agencies such as the National Anti-Poverty Commission, the National Economic and Development Authority (NEDA), DOLE and its training adjunct (TESDA), the Department of Trade and Industry (DTI), and the Department of Social Welfare and Development (DSWD) should provide specific budgetary allocations to programmes and projects that will benefit homeworkers (Rosario del Rosario, 2000). This will include provision for one-stop centres for all credit programmes, training, market-identification at the local level for all existing and potential entrepreneurs. Action on the legal front The government of the Philippines could review and monitor the implementation of laws, rules, and regulations on subcontracting and homework, and if need be,

312

Rosario et al.

promulgate new ones, with a view to ensuring workers’ rights and welfare. This would include the following: (a) Test on site the implementation of the instrument for securing homeworkers’ rights under subcontracting (i.e. Department Order No. 5 under the Labour Code on Employment of Homeworkers); (b) Help raise the piece-rates and improve working conditions of homeworkers through participatory time and motion studies and similar interventions. This would follow the example of Angono, Rizal, where PATAMABA workermembers engaged in Christmas ball production were able to negotiate for a higher piece-rate (from P. 8 to P. 10 per ball) in August 1998 with the assistance of the DOLE and the Rizal Informal Sector Coalition; (c) Push for a Magna Carta for Informal Sector Workers through DOLE, which has been given technical assistance by the UNDP for this initiative. A Magna Carta would cover not only those under subcontracts (dealt with by D.O. No. 5) but also the self-employed, who usually also come from the ranks of displaced and/or enterprising subcontracted workers. This has been proposed by the Informal Sector Coalition (ISC), of which PATAMABA is a member and aims to integrate all the laws and policies affecting the informal sector. Lately, PATAMABA has shifted strategy towards pushing for a Magna Carta for Homeworkers instead, since encompassing the interests of all informal sector workers through one law is now deemed to be unrealistic. (d) Immediately ratify ILO Convention 177 (The Homework Convention 1996) and the accompanying Homework Recommendation 1996 which has provisions similar to D.O. No. 5 but provides more clarity or goes even further in the areas of data collection, the rights to organise and to bargain collectively, remuneration, occupational safety and health, hours of work, rest periods, leave, social security and maternity protection, protection in case of termination of employment, and programmes related to homework. Social protection The government could also examine the possibility of inclusion of homeworkers into the social security system and in the provision of safety nets. The ideal situation would be coverage by the Social Security System (SSS) of homeworkers so that employer–employee relations can be established with the subcontractors/ manufacturers. Homeworkers’ organisations could act as conduits for collection of payments and availability of benefits, so that transportation and opportunity costs incurred for the payment of premiums can be minimised. Qualified homeworkers could also be part of the joint programme of the Philippine Health Insurance Corporation, with Local Government Units shouldering the premiums for the health insurance of those in need. At the very least, homeworkers could have access to safety net programmes, especially of the DOLE, meant to assist displaced workers and other workers in crisis.

Homework by women and children in the Philippines 313 The government could explore replication in the sectors included in the study (initially perhaps in okra production) of the Social Amelioration Fund mechanism being implemented by the DOLE through the Bureau of Rural Workers in sugarproducing areas. The Fund comes from a P. 10 levy per picul28 of sugar, and should cover cash bonuses for both mill and plantation workers, maternity and death benefits, and credit assistance for socio-economic projects. Contributions to the fund are placed in a joint account of the sugar producers and the DOLE, which monitors compliance through a reporting system. The government could also consider providing transition mechanisms (through the DSWD) to alternative sustainable livelihood for communities engaged in hazardous work, such as pyrotechnics production. Meanwhile, action is needed to ensure the health and safety of workers, upgrade technology for the viable enterprises, and regulate marketing for consumer protection through an inter-agency29 mechanism working closely with local governments. With data from the research, the government could respond to the problems and needs articulated by the women and children involved in pyrotechnics production, the hazards they face, their problems with the present licensing system, and their desire to move to safer means of livelihood if given the opportunity to manage the transition. Health and safety seminars, protective gear (masks) at low cost, and advice regarding the ergonomics of production is needed. This can be a coordinated effort between the Occupational Health and Safety Centre (OSHC) of DOLE and the local government units. The government could cooperate with associations of fireworks producers in professionalising the industry, upgrading its technology, raising the quality and reliability of the products through standardisation of the production process and proper labelling, ensuring decent work and eliminating child labour. Addressing child labour The government could examine the possibility of providing (through the DSWD); scholarships, textbooks and assistance to acquire school supplies for child workers who would otherwise not be able to go to school. Similarly, the Department of Education, Culture, and Sports (DECS) and local governments could establish special facilities for out-of-school child workers who want to continue their education. Such action would need to be combined with incentives for the establishment of the Barangay Council for the Protection of Children (BCPC) in the barangays where child labour is endemic. There is a need for a homeworker’s desk in the local government system composed of government/NGO focal points to conceptualise and oversee the following: (i) effective monitoring of and research on adult and child workers in general; (ii) Information, education and communication (IEC) materials and information campaign on the situation of adult and child labour, as well as the conduct of seminars on rights as homeworkers, women, girls and children; gender sensitivity and environmental consciousness;

314

Rosario et al.

(iii) programmes that will create safety nets for subcontractors and homeworkers (e.g. promotion of schemes to provide capital and markets for local production of manufactured products); provide training; set up cooperative enterprises; (vi) a counselling/referral system to provide paralegal and other assistance to women and children homeworkers who have complaints about work, violence against women, abuse and other forms of violence and threats to safety in their specific communities; (v) Monitor, through the local governments, the health and nutritional status of child workers and provide programmes to improve this, especially in areas where child labour receives relatively higher remuneration and therefore children work more intensively and for longer hours. The Labour Code of the Philippines, the Child and Youth Welfare Code, and specifically, Republic Acts30 all contain provisions for the minimum age of employment. However, the Republic Acts allow for loopholes especially with regard to child homework. They allow the employment of children below 15 years under specific conditions (e.g. that such employment neither endangers the life and safety, health and morals, and normal development of the child and that the parent or legal guardian makes sure the child is provided with primary or elementary education; and secures a work permit from the DOLE, and provided it is under parental supervision). The assumption behind allowing a child below 15 years to work under parental supervision is the idea that the home is always a safe place. However, from repeated observations and testimonials, it appears that industrial homework has transformed the home into a cramped, polluted and hazardous place under subcontracting. Furthermore, parental supervision cannot be presumed a guarantee against child abuse and exploitation, since parents themselves are often the employers. Thus, amendments to incorporate the Convention on the Rights of the Child (CRC) and ILO Convention No. 138 into the law must be pursued, since child labour is not just a problem concerning workers, but the whole of society. There is a need for awareness raising campaigns regarding the Convention on the Rights of the Child, the ILO Convention No. 59 on the minimum employable age, and ILO Convention No. 138. Similarly, there is a case for supporting existing programmes and projects on child labour.31 Community-based Organisations (PATAMABA, other NGOs and people’s organisation (POs)) There is scope for greater action by community based organisations (CBOs) as well as to conduct gender awareness seminars and other interventions towards shared household and family responsibilities and the prevention of family violence. This would involve the government (DSWD), NGOs,32 and local governments to facilitate the provision of counselling, refuge, and other forms of assistance to battered women and children. Several issues would need to be addressed including: health, safety and environmental hazards, initially through

Homework by women and children in the Philippines 315 a zero waste management programme, herbal medicine and Botica sa Baryo (Medicines in the Village) facilities, cleaning and greening and so on. This can be facilitated through networking with local government units (LGUs), as well as with health and environmental groups. CBOs could spearhead the formation and consolidation of Barangay Council for Protecting Children through an advocacy campaign targeted at Barangay and other local government officials. These councils could have special programmes to assist child workers in continuing their education and in safeguarding their health through networking with NGOs (such as ERDA Foundation and funding agencies such as AUSAID), which have an interest in these concerns. CBOs face a challenge to help homeworkers to bargain better with subcontractors using available data on subcontracting chains, registration and financial records. They could also explore the possibility of shortening the chains (thereby providing better piece-rates to the homeworkers by dispensing with unnecessary middle persons) through collective subcontracting arrangements whereby workerleaders deal directly with those who order. This would require the following actions: (a) Engage in dialogue with subcontractors and principals, or ask friendly subcontractors to mediate between principals (or employer) and homeworkers. The assumption in this approach is that these subcontractors are allies of the workers on the strength of their kinship or neighbourly relations with them. They also want to receive (and if possible increase) the piece-rates due them, and to be paid on time. (b) Facilitate membership in the SSS and the Philippine health insurance programme for those interested to be able to claim sickness, maternity, disability, retirement and death benefits. (c) Organise the homeworkers that are not yet organised (e.g. those in pyrotechnics production). Communities better organised are in a better position to address issues of child labour and homework. As women and children go through each stage of empowerment, they acquire more knowledge, skills, self-confidence, bargaining strength to produce and reproduce resources at increasing levels through their own initiative. Thereby, they are able to ‘gain ascendancy in the economic, social, and political arena’ (Lazo, 1993). This last point is fundamental, given the invisible ceiling that prevents poor women from moving up and getting their just share of the fruits of development.

Annex 9.1: Case studies Case study 1 – Emily (pyrotechnics production) Emily earns an income of P. 300–400 per week, which she allows to accumulate as savings until the month of September or October. The amount is allocated as capital for her own firecracker production during the boom period. She needs the capital to be able to move forward. This is why she, like many others, is unable to

316 Rosario et al. change from firecracker production to another activity – it promises brisk business, and loans are accessible and easy to secure even from friendly neighbours. In the hope of earning more in order to pay her loan and have some extra to cover expenses, Emily continued to produce firecrackers right up to New Year’s Eve. She and her children worked through the night during the last week of December to meet the 31 December delivery. Case study 2 – Josie (pyrotechnics production) Josie, a child homeworker earns an average of P. 15 for folding and covering 3,000 pieces of paper containers or fuses in nine hours of work a day. She says that by earning she can help her parents, as she no longer has to ask them for pocket money for school. Josie’s daily routine includes waking up at 5:30 in the morning, after which she wakes her mother up, who in turn, prepares their breakfast. However, Josie usually misses breakfast because she has to catch the jeep that takes her to the intermediate school, far from their community. In the afternoon, after arriving from school, she helps her mother to do the household chores. She cleans the house, washes dishes and clothes. During schooldays, Josie starts her work with firecrackers after dinner. She folds the paper containers from 6.30 to 8.30 pm in the evening, finishing around one thousand pieces for P. 6. She has been involved in firecracker production since she was eight. Usually, she works the whole day during weekends, and during the rare occasions when she is absent from school. If her mother is away, she looks after her younger brothers in addition to her usual household chores. Needless to say, she seldom has time to play – unlike her younger brothers. When she has work, she cannot go out with her friends to play, relax and socialise. Case study 3 – Nene (okra production) Nene (48 years old), became an agricultural worker at the age of 12, and has been employed as an okra producer since 1995. She believes she developed rheumatism and paralysis as a result of fatigue. Her work entailed picking okra, loading these in the basket around her waist, hauling heavy crates, and soaking her hands and feet in water the whole time that she is cleaning the okra. She experienced body aches, headaches, back and waist aches, blurring of the eyes, dizziness and frequent bladder discomfort as a result of her stressful work in the field. Her field work required a regular bent-over position under the sun. She says that picking okra also causes much itchiness and lung irritation, due to the okra’s stiff hairy cover. She eventually had to stop working because she became seriously ill. She was diagnosed to have a lung and liver ailment, for which she is still being treated. Case study 4 – Lita ( fashion accessories) Lead subcontractors, subcontractors, job outers and suppliers get more payment than the workers, by virtue of their ‘capital’ in terms of working cash, land access/ownership, ownership of machines and special know how.

Homework by women and children in the Philippines 317 Similarly, Lita, a job outer in fashion accessories production owns her house, has a food business, and can therefore benefit from loans by virtue of the house and activity, her membership in PATAMABA, and her political connections. Because she has access to capital, she is able to buy or rent machines, and hire workers whom she pays to do the necessary processes needed in producing the raw materials for making bracelets, necklaces and others. She very probably has the right skills for managing a business and the contacts with subcontractors and contractors. The fact that five of her female children are already in fashion accessories production, and can assist her in her work, is probably a plus factor. Case study 5 – Gracia ( fashion accessories) Gracia (Lita’s fifth child), dropped out of school after her third year of high school. She preferred to do subcontracted homework in fashion accessories rather than continue to go to school when the family’s financial resources had diminished. While studying, however, she was frequently absent because of the sleepless nights she spent working on urgent orders for fashion accessories. Gracia entered school at the age of seven. She started working when she was in Grade 1. It was at that time that her face was pierced with a stick, which left a scar that deepened over the years, and deformed her face. While in Grade II, she continued helping her mother at the same time as doing household chores. From Grade II–IV, she helped her mother make breakfast before going to school at 7.00 am. After school, from 1.00 to 1.30 she helped make fashion accessories again. At 1.30 she would go to school up to 5.00 pm. From 5.30 to 7.00 pm, and after supper from 8.00 to 9.00, she would again help her mother make fashion accessories. After she dropped out, Gracia’s involvement in fashion accessories production became independent of her mother’s. Her schedule and pay in producing fashion accessories varies. She might string beads into necklaces or pierce shell beads from 9 in the morning to 3 in the afternoon, or she might string wood or shell beads along a wire from 10 in the morning to 12 noon. She also assists her mother in her food business. She washes the family’s clothes everyday for about 2 hours, before starting to work; she also does the dishes for 20 minutes at noon; fetches water; takes care of the sick if there are any (as she did her father when he was sick); and does other people’s laundry on Saturdays and Sundays, before again working on fashion accessories.

Notes 1 Rosario together with the country research group (including UNICEF, PATAMABA, UP, CSWCD) has done and written The Philippines Country Study Report (2002). Santosh Mehrotra, Mario Biggeri, Carolina Vizcaino and Stefano Mariani contributed to this chapter. 2 The only exception occurred in the services sector, in which the share of informal workers decreased from 68 per cent to 65 per cent between 1989 and 1993. 3 The garment industry has emerged as the foremost employer and leading export industry of the country.

318

Rosario et al.

4 Without labour standards, an employment contract is sufficiently flexible to reduce the need for managers to renegotiate a contract for every situation that arises. 5 The stages of vegetable farming that involve homework are the final stages of production, such as the packing of already harvested and cleaned vegetables (like okra and string beans) (Boquiren, 1987). In fishing, homework is used at the stage of fish processing. One study pointed out that women and girls were in charge of scaling, splitting, drying, and salting fish for fish dealers in the community. 6 Other studies show the existence of subcontracting to homeworkers in the production of: footwear; handicraft (made of indigenous fibres, metal, shell, coconut, etc.); fashion accessories; pyrotechnics; gloves; fibre mats; doormats; boxes; yarn; twine; nipa (native fiber) roof shingles; brooms; indigenous cigars; stuffed toys; antique reproduction; furniture; food preparation (including duck de-feathering and dressing); agriculture (as in contract animal rearing, vegetables, fruits and cut flowers and in fishing processing); and services (such as in medical transcription). 7 According to the official survey results, there is a progressive decrease in the proportion of homework households describing their economic situation as ‘very easy’, ‘easy’ or ‘just right’. Correspondingly, as one moves from 1990 to 2000, the proportion of households who claimed that the economic situation of the household was ‘difficult’ increased progressively. 8 Among the salient provisions in the rules is the right to self-organisation and registration, standardisation of rates by time and piece produced, prohibition of any deduction from homeworkers earnings for materials lost, regulation of employment of minors as homeworkers, prohibition of homework in dangerous occupations, and assistance from the DOLE Regional office for certain benefits including registered homeworkers’ organisations, information on wages; skills training; entitlement to social benefits and facilitation of loans with government and non-government financial institutions; and information on availability of housing programmes. 9 The selected communities were Sta. Maria in the province of Bulacan (Luzon), Kalayaan and San Vicente in Angono, Province of Rizal (Luzon), Sierra, Concepcion, in the province of Tarlac (Luzon), and San Roque, Talisay, in the province of Cebu (Visayas). 10 Homework is defined as ‘work carried out by a person in his or her home or in other premises of his or her own choice, other than the workplace of the employer’. 11 For example, in a vegetable producing community, home can mean the roofed abode, the yard or the cultivated fields. Another example is in a fishing community (as in the fishing community of Gigante, Iloilo, Panay), home can be the inside or the immediate premises of the home, the beach, common fish drying areas, or the employer’s warehouse. 12 For example in the children’s survey the interviewer had to collect information regarding height and weight of the child, apparent health, and basic questions regarding schooling. 13 Hazards included exposure to firecrackers and dangerous chemicals, uncollected garbage, lack of toilet facilities, mosquito problems (connected with dengue), drug addiction, alcoholism and family violence. 14 Okra workers were unsatisfied with their wages. Yet thanks to collective demand, they were able to obtain an increase in wages from their subcontractor. This development could be construed as erosion of the traditional bases of patronage. On the other hand it could very well have been a result of traditional relationships intersecting with more emergent ones. The lead subcontractor was simultaneously the chief of the local government, kin and/or personal patron of many workers, and peer of another local government leader who was an active PATAMABA member. Thus, the increase in wages might be attributed to the PATAMABA member who supported the workers’ demands, and persuaded the local leader (source: The Philippines Country Study Survey (2001) and field informant and FGDs). 15 The minimum required amount to purchase production materials is P. 10,000 (equivalent to US$200). Given the depressed market, many earn just enough to pay back their

Homework by women and children in the Philippines 319

16 17 18 19 20 21

22 23

24

25 26 27 28 29 30 31

32

borrowed capital and exchange for rice and food. See case study 1, Annex 9.1) (source: The Philippines Country Study Survey (2001) and FGDs and field informants). During peak seasons the rate is higher P. 5.50. The total cost of production for a 5- star firecracker is P. 2,647 of which P. 1,978 is the cost of raw materials including chemicals and paper. Foreign companies use a local contractor who in turn assigns production to 26 subcontractors located in Angono, Antipolo, Cainta, Taytay, Binangonan and Paranaque. Women worked 9 hours and men worked 3 hours during the high season. Production cost includes the cost of raw materials (velvet cloth/lamai, pins, beads, cord and glue) amounting to P. 9.41, and labour (for scraping, drawing, covering, designing, beading, finishing and quality control) at the cost of P. 8. Cost prices for the production of a wrought iron basket; materials  P. 85, labour  P. 20, and subcontractors’ profit  P. 30. The total cost is P. 135. Cost prices for the production of a glass flower stick; materials  P. 15, labour  P. 10, and subcontractors’ profit  P. 5. The total cost is P. 30. The lead subcontractor is usually a local community official who is usually a farmer producing rice and vegetables on his own land. Wages per activity are: ploughing per day  P. 300; planting, fixing dikes/worker/day  P. 70; applying fertiliser per day  P. 100, hilling up per worker/day  P. 300, weeding per worker/day  P. 70; spraying pesticide/worker/ day  P. 100; picking, cleaning, hauling, putting in crates/worker/day  P. 70–80; overseeing (includes daily irrigation): team leader/day  P. 150, and team member/day  P. 100. All the households in the sample area reported having at least one female over 19 engaged in the production of fashion accessories, 47 per cent reported having at least one female between 5 and 14 years of age, and 27 per cent reported at least one female in the 15–18 age group. Annual income was greater only by the small amount of P. 15,000. See case study 3 in Annex 9.1. See case study 4 in Annex 9.1. One picul is equal to 133.3 lbs. DTI, DOLE and TESDA. (RA)s 7610 and 7658 ‘An Act Providing for Stronger Deterrence and Special Protection Against Child Abuse, Exploitation and Discrimination, Providing Penalties for its Violation, and for Other Purposes’. Examples of such programmes are: the Breaking Ground for Community Action on Child Labour (BGCACL) which is a programme within the umbrella of the National Child Labour Program Committee (NCLPC) located in the DOLE and supported by UNICEF; rescue of child workers in the community in which several NGOs like Kamalayan and Visayan Forum which collaborate with the DSWD and the Philippine National Police (PNP); Child Hotline Project called Bantay Batay (Child Watch), instituted by the media network ABS-CBN through the DSWD; DSWD’s programme for children who have run away or are homeless and abandoned, called Lingap Centre, of which there are 17 such nationwide; Working Youth Centres which provide information and facilitation for youth in terms of education and training, welfare assistance, and employment promotion and generation; skills training for out-of-school youth 15–24 years of age conducted by TESDA; the teaching of children’s rights at all levels in community-based action programmes on child labour. Women’s Legal Bureau, Aru-Kalakasan, Women’s Crisis Centre, Likhaan, Women LEAD.

Bibliography Alba, M. and Villamil, W. (2000), ‘The Impact of Globalization on Female Employment in the Formal Sector’. Manuscript submitted to the International Labor Office, Manila.

320

Rosario et al.

Boquiren, R. (1987), Case Experiences of Child Labor in the Benguet Vegetable Industry, Diliman, NSTA-UP, University of the Philippines. Bureau of Rural Workers (BRW) (1994), ‘Implementation of the “Action Research to Promote Organization Among Women Piece-rate Workers in the Philippines” ’, BRW, Department of Labor and Employment. Center for Research and Special Issues (CRSS) & Center for Research and Communications (CRC) (1994), Meeting the Industrial and Labor Challenges of Subcontracting. Data Source, The Philippines Country Study Survey (2001). Del Rosario, R. (1991), Employers of Child Workers: A Profile, Diliman, Quezon City: SSHRC-UP-ORC, University of the Philippines. Del Rosario, R. (1996), Child Labor in Southeast Asian Manufacturing Industries: Focus on the Garments Industry in the Philippines, A Country Report, ILO-IPEC, Manila, Philippines. Del Rosario, R. (1998), ‘Revisiting Laura, A Garment Factory in Rizal’ (A Study on Globalization), Philippines: UCWS, QC. Del Rosario, R. (2000), ‘Philippines – Status of Public Policy Initiatives on Contingent Labor’, in Contingent Employment of Women Workers in Japan, the Philippines and the United States, Workplace Issues, A Policy Research and Analysis Project, Philippine American Foundation, USA. Del Rosario, R. (ed.) (2001), The Philippines Country Study Report. Outsourcing of to Households: Subcontracted Manufacturing Home Based Work by Women and Children, (unpublished joint report, UNICEF, PATAMABA, UP, CSWCD). Del Rosario, R. and Bonga, M. (2000), Child Labor in the Philippines: A Review of Selected Studies and Policy Papers, U.P.-OVCRD/UNICEF/AusAid, Manila, Philippines. Department of Labour and Employment (DOLE) (1998), Omnibus Rules Implementing the Labour Code. Department Order – D.O. No. 10 Series of 1997 amending the rules implementing Books II I and VI of the Labor Code. Diaz, Mariel O. (undated), ‘Subcontracting and Women Homeworkers (A Review of Related Literature)’, CSWCD, UP. Esguerra, Nerissa and Emmanuel Esguerra (2001), ‘Globalization, Labor Markets, Trade Union Challenges and Responses’ (Philippines), An Eight Country Study for ICFTUAsia-Pacific, [draft]. FGDs and field informants (2000), The Philippines Country Study (1989). ILO (1996), Convention 177 (The Home Work Convention 1996). Labor Code (1989) section 1, Rule XIII, Conditions of Employment, Rules and regulations Implementing the Labor Code. Labor Code, Homework Recommendation 1996. Lazo, Lucita S. (1993), ‘Some Reflections on the Empowerment of Women’, in Lazo, Lucita (ed.) From the Shadows to the Fore – Practical Actions for the Social Protection of Homeworkers in the Philippines, International Labour Organization Regional Office for Asia and the Pacific, Bangkok: ILO. National Statistics Office (NSO) (1996), Final Report on Institutionalizing the Enumeration of Home workers in the National Statistical Collection System of the Philippines, Manila, Philippines. NSO (1993), Survey of Homeworkers (SOH), Philippines. Ofreneo, R. (1997), ‘Contract Labor in the Philippines’, preliminary notes, May 1997.

Homework by women and children in the Philippines 321 Pineda-Ofreneo, R., Lim, J. and Gula, L. (2002), ‘The View from Below: Impact of the Financial Crisis on Subcontracted Women in the Philippines’, in Balakrishnan, Radhika (ed.), The Hidden Assembly Line – Gender Dynamics of Subcontracted Work in a Global Economy, Connecticut: Kumarian Press, Inc. Torres, Carmela (1993), ‘External Labor Flexibility: The Philippine Experience’, in Philippine Labor Review, ILS-DOLE.

10 Subcontracted homework in Thailand Narumon Arunotai, Napat Gordon, Ratana Jarubenja, Nitaya Katleeradapan, Narong Petchprasert and Amara Pongsapich

The growth of homework in Thailand can be attributed to the development of the export industry sector. While the first three Development Plans in Thailand had focused on encouraging the growth of import substituting industries, during the Third Development Plan (1972–75), the Thai government shifted to an export strategy based on labour-intensive manufacturing. Manufacturing production provoked a shift of output and labour from agriculture to industry. The new industrial sectors absorbed the migrant labour from rural areas slowly. As a result, migrant workers from rural areas were absorbed in the informal economy, including homework (Arunotai, et al., 2001). The Thai government has recognised the key role of homework in the development of the export sector. This is reflected in a number of studies regarding homework, and various programmes for homeworkers. Studies so far have focused particularly on the conditions of workers engaged in the production of textiles, garments and shoes, and the situation of women – who are most typically homeworkers – and the health hazards associated with these activities. Moreover, the Thai government has been pro-active in creating special agencies, such as the Office of Home Based Workers. Nonetheless, given the complexity and ‘invisible’ nature of homework as well as the limited data, further study is still required. The purpose of this chapter is to examine the current situation and main needs of homeworkers in Thailand. The three sectors explored are located in different regions of the country and have not been studied before. The sectors are: saa paper production in the northern region of the country (Angthong and Prae), leather craft production in Bangkok and environs (Samutprakarn, Rachaburi and Pathumthani) and hybrid seeds production in the northeastern region of the country (Sakonnakorn and Kalasin). The three sectors involve homework subcontracting, production for export markets and a considerable number of homeworkers. Section 10.1 presents an overview of homework through an assessment of the industrialisation process in Thailand and its impact on the development of outsourcing. This section discusses the linkages between industrial development and the expansion of the informal sector. Section 10.2 presents an overview of the conditions of homeworkers based on the existing literature. Section 10.3 provides a detailed description about the methodology used in the field study. Section 10.4 examines each sector at macro and micro levels. The macro analysis includes the

Subcontracted homework in Thailand 323 links with foreign markets and the contribution to GDP. The micro analysis includes a description of the value chain along with the various production stages. Section 10.5 discusses the socio-economic implications of the subcontracting system and evaluates the conditions of workers across sectors, with a special emphasis on women. Section 10.6 presents existing government and NGO initiatives. Finally, Section 10.7 summarises the main areas of concern and presents a list of recommendations.

10.1

Thailand’s industrialisation process and the informal sector

Thailand’s industrialisation was a result of aggressive government initiatives to modernise the country through the creation of an export industry. The initiatives, part of the National Social and Economic Development Plans, had an important impact on Thailand’s economic growth. The first two Development Plans (1961–71) focused on the development of import substituting industries. These plans targeted an 8 per cent growth for the industrial sector and a 5 per cent growth for the agricultural sector. In the Third National Development Plan (1976–81), the government adopted policies aimed at creating an export market with particular emphasis on the labour-intensive industries. The Investment Promotion Act and the Board of Investment were established in 1977 as a means to promote foreign and domestic investment in new export sectors. The Act included investor-friendly measures including tax and duty exemptions, and loose safety standards and labour legislation. The targeted sectors were rice mills, spinning mills, basic machinery and garments. As a result, the non-agricultural sector experienced considerable growth. While in 1960 the share of the non-agricultural sector in GDP was 17 per cent, in 1980 it rose to 77 per cent and 89 per cent in 1995. The expansion of the non-agricultural sector, and specifically industry, is also reflected in the increasing number of manufacturing establishments. According to the Industrial Census of the National Statistics Office (NSO), by 1980 there was a total of 8,180 manufacturing establishments, including rice mills, machinery manufacturing, clothing and spinning mills and the registered number of workers was 887,000.1 By 1997, manufacturing establishments diversified into additional industries. New products included frozen food and beverages, textiles, clothing, leather products, plastic and basic metal products. Moreover, the number of workers engaged in industrial production increased to 2.4 million. The development of the export industrial sector based on labour-intensive manufacturing provoked a dramatic shift in the composition of total output in the Thai economy, but less dramatic change in the distribution of the labour force between agriculture and the non-agricultural sector. In 1960, the share of agriculture in GDP was 83 per cent and it employed 83 per cent of the workforce. By 1980, the share of agriculture in GDP fell dramatically to 23 per cent, yet 71 per cent of total workforce was still employed in this sector . However, a significant shift of labour occurred by 1995. In this year, the share of agriculture in output fell to 11 per cent but 53 per cent of the total workforce was employed in

324

Arunotai et al.

Table 10.1 Workers in labour force, 1988 and 1994 (in thousand) Labour force 1988

Labour force 1994

Percent Percent of labour of labour force 1988 force 1994

Formal sector 4,984.40 7,438.80 17.04 Informal 24,272.40 24,656.30 82.96 sector Total labour 29,256.80 32,095.10 100.00 force

Average Annual growth average 1988–94 (%) growth %

23.18 76.82

49.24 1.58

7.03 0.23

100.00

9.70

1.39

Source: 1994 Formal and Informal Labour Force Market Survey, NSO.

agriculture. Hence, the shift of labour to non-agricultural activities occurred at a much slower pace than the growth of non-agricultural output (Narong, 1991). The industrialisation process accelerated the rural–urban migration. Poor rural families, including children under 13, migrated to urban centres in search of income and employment opportunities.2 Nonetheless, the inability of the manufacturing sector to absorb the shifting workforce had an impact on the employment structure and generated the expansion of the informal sector. In Thailand the informal sector accounts for the majority of the workforce, both in rural and urban areas, accounting for 77 per cent of the total workforce (1994) (Table 10.1). Although official data show a declining trend, the informal sector still represents the majority of the workforce. While in 1988 it accounted for 83 per cent of the workforce, in 1994 it accounted for 77 per cent. Nonetheless, compared to the percentage of formal workforce in the same year (23 per cent), the informal sector still represents the majority of the workforce. The majority of workers are employed in agriculture (56 per cent), while 44 per cent are engaged in non-agricultural activities (Table 10.2). The high percentage in the agricultural sector is explained by the cyclical nature of this activity. Informal workers are the majority in various non-agricultural sub-sectors including commerce (78 per cent), transportation (67 per cent) and construction (51 per cent) (Table 10.2). The only sub-sector where the proportion of formal workers is greater than the proportion of informal workers is in services with 64 per cent. Informal non-agricultural workers are distributed among commerce (39 per cent), manufacturing (22 per cent) and services (19 per cent) (Table 10.2). More than half of the total informal workforce (89 per cent) is located in rural areas. Since homeworkers are part of the informal workforce, the distribution is similar with 79 per cent of the total number of homeworkers in rural areas (Table 10.3).

10.2

The emergence of Home Based Work in Thailand – some features

It can be assumed that the number of homeworkers has increased with the proliferation of informal manufacturing and service establishments. However,

Table 10.2 Thailand: share of workers by sector, distribution of formal/informal sector workers by industry and distribution of informal workers in non-agriculture, 1994 (in thousand)

Agriculture Non-Agriculture Manufacturing Construction Electricity, gas, water and sanitary services Commerce Transport Services Others Total (%) Manufacturing Construction Electricity, gas, water, and sanitary services Commerce Transport Services Others Total

Total workforce

Informal

Units

%

Units

%

17,960.4 (56%) 14,134.7 (44%) 3,851.0 1,698.1 184.9

100.0

17,386.7

96.8

573.7

3.2

100.0

7,269.6

51.4

6,865.1

48.6

100.0 100.0 100.0

1,619.3 863.2 4.2

42.0 50.8 2.3

2,231.7 834.9 180.7

58.0 49.2 97.7

3,617.8 858.0 3,859.9 65.0

100.0 100.0 100.0 100.0

2,814.4 572.3 1,379.1 17.1 100.0 22.3 11.9 0.1

77.8 66.7 35.7 26.3

803.4 285.7 2,480.8 47.9

22.2 33.3 64.3 73.7

76.8

7,438.8

23.2

32,095.1

100.0

38.7 7.9 19.0 0.2 24,656.3

Formal Units

Source: 1994 Formal and Informal Labour Force Market Survey, NSO.

Table 10.3 Distribution of workforce and homeworkers by location and sector, year 1998 (in thousand) and percentage

Rural % Urban % Total

Total

Forma1

Informal

Homeworkers*

26,509.2

4,512.3 60.7 2,926.6 39.3 7,438.9

21,996.9 89.2 2,659.5 10.8 24,656.4

247,561.2 79.4 64,228.7 20.6 311,790.0

5,585.8 32,095.0

Source: NSO, Formal and Informal Labour Force Market Survey. Note * NSO, 1999 Home Work Survey (workers age group 13 years old and over).

%

326

Arunotai et al.

there is no official data to support this. Recent government’s concerns about homeworkers’ conditions and NGOs advocacy campaigns have highlighted the need to quantify the scale of homeworkers. Consequently in 1998 the government launched the first3 national HBW survey using the conceptual framework established in 1996 in the ILO Convention of Home Based Work. In addition, it established the Office of Homeworkers to ensure the periodical gathering of data. This first attempt revealed that there were more than 300,000 people in the labour force (aged over 13 years) involved in HBW. Most of them (79 per cent) were in the rural or non-municipal area, while 21 per cent were in the urban or municipal area (Table 10.3). Those in the rural area were mostly agricultural workers. Due to the cyclical nature of agricultural work, these workers are seasonally unemployed, and HBW is their means of earning some income during that period of ‘unemployment’. Though their main occupation may be agricultural production, this workforce cannot depend on agricultural income alone. This is in congruence with NSO’s 1999 Homework Survey result that indicates that ‘comparison between areas showed that those in the municipal area (urban) who were engaged in HBW as their primary occupation, outnumbered those in nonmunicipal area (rural) in all the regions’. Thus, HBW is the primary occupation for those in urban areas, as opposed to workers in rural areas for whom HBW represents a secondary occupation (Benja, 1995). According to the NSO Survey (1999) the largest share of homeworkers are employed in manufacturing (95 per cent). Other activities in which homeworkers are involved to a lesser degree included construction, commerce and services (Table 10.4). This leads to the next main finding: the feminisation of HBW. More than half (74 per cent) of the total number of homeworkers were female. Women’s involvement occurs thanks to the flexibility that HBW offers in combining household and reproductive responsibilities with work. Most of the women are married and economically dependent on their spouses whereas only 30 per cent are unmarried Table 10.4 Distribution of homeworkers by sector Sector

Percentage

Percentage of workers that require support to establish an organised group*

Agriculture Manufacturing Construction Commerce Service Transportation and communications

0.08 95.44 0.01 3.30 1.13 0.04

94.72 26.26 N/A 23.43 23.61 N/A

Source: Thailand Country Study Survey (2001). Note * 1999 NSO survey.

Subcontracted homework in Thailand 327 and in the 13–24 years age group. Finally, the census demonstrated that there are 1,218 organised groups, involving two-thirds of the total number of homeworkers (200,000). They are mostly employed in manufacturing. Subcontracting workers or, Chang Mao4 in Thai, has existed since the preindustrialisation period. During this period, shop-owners subcontracted women for the production of garments such as uniforms, underwear and handkerchiefs sold in the domestic market. The form of payment was by dozen of finished pieces. During industrialisation Chang Mao evolved into a more sophisticated system. The new system entailed various levels of subcontractors including unregistered manufacturing establishments, retailers and homeworkers located at the bottom of the production chain. As opposed to the exclusive production of clothing, the ‘sophisticated’ subcontracting system was used for the production of textiles, garments, plastics, shoes and handicrafts intended for both the domestic and foreign markets. Currently HBW in Thailand is undergoing a transformation with the emergence of organised groups of workers. Organised groups embrace a group leader that has a ‘managerial’ function and is responsible for the distribution and coordination of work, recruitment and quality control of production in favour of subcontractors. Nonetheless, thanks to NGO support, organised groups are useful in increasing the bargaining power through collective action. The emergence of HBW in Thailand could be explained by three different sets of factors. From the employer/contractor’s perspective, this subcontracting system enables the minimisation of costs of production and risk of unionisation. It also allows companies to adapt to market fluctuations with minimal investment. Therefore, it is cost-effective for companies to outsource production processes to unregistered manufacturing establishments and workers in the informal sector. From the worker’s perspective, HBW represents an alternative source of income, particularly in case of slack periods in agricultural production, and an attractive occupation for those confined to home.5 Homework is also preferred due to its minimal requirement of strenuous activities, physical strength or high level of skills. From the government’s perspective homework has a key role in the development of export industries based on labour-intensive activities. The lower costs of HBW make Thai exports very competitive and represent additional incentive to attract foreign investment into the country.6 In addition, from a welfare point of view, homework represents a provisional source of employment and income for the unemployed, rural migrants and the displaced during the economic crisis. In conditions of job retrenchment, HBW provides a cushion. Thailand’s economic crisis in 1997 forced various private companies to go through downsizing and even closure of small companies that caused higher levels of unemployment. Finally, since HBW generates income among rural households, it is also an attractive measure to reduce rural migration. Another determinant is Thailand’s social-cultural context. Thai society relies heavily on kinship based ties. Kinship ties are social links among people from the same community that involve mutual aid and assistance. With the introduction of the capitalist economy, kinship ties became the basis of patron–client relationships. Currently kin ties are also the basis of the subcontractor–homeworker relationship.

328

Arunotai et al.

Since both the subcontractor and homeworker belong to the same community, the contractual arrangement is based on mutual trust and formal written contracts are unnecessary. The characteristic of mutual assistance among female homeworkers has also favoured the introduction of HBW in Thailand. Finally, it must be mentioned that Thai society was already familiar with the subcontracting system under informal arrangements that is, Chang Mao. HBW has experienced various transformations through time: from traditional Chang Mao, to a sophisticated subcontracting system and to organised groups. Internal and external factors have contributed to the emergence of HBW in Thailand. Internal factors include the social-cultural context and the new needs of an industrial economy. External factors such as globalisation forces, trade liberalisation and the integration of international markets have also played a part in the expansion of HBW in Thailand. There is little information about the conditions of women engaged in Chang Mao. Most of the information regarding the conditions of homeworkers refers to the industrialisation period. Yet studies demonstrate that expansion of homework had some negative implications for homeworkers. During industrialisation, homeworkers were mostly migrants from rural areas. In urban areas, migrants engaged in informal activities that fitted their low level of skills and experience. As any other informal worker, homeworkers faced various risks including the lack of safety nets and legal protection (Table 10.5). Existing literature suggests that despite low pay and lack of social security, workers engaged in homework due to lack of better alternatives. Another study supported by the ILO was conducted by the Regional Research Network on the Status and Employment of Homeworkers in Indonesia, the Philippines and Thailand. Their conclusions were similar to the previous surveys and studies on the fact that women were dominant in numbers in home-based production and the subcontracting business (Pongsapich et al., 1991). Bargaining power against subcontractors was very low due to the excess supply of informal migrant workers. As a result, exploitation practices including unfair treatment, excessive work, delayed payments, low remuneration and hazardous work environments became common. Child labour is another example of the negative dimensions of HBW. Children were part of the rural–urban migration. Given the dire conditions of migrants, children experienced a shift of status from ‘unpaid family workers’7 to ‘informal low remunerated workers’. In this way, they became part of the ‘cheap labour force’8 and worked with their parents in informal manufacturing activities. After the government’s adoption of the investment operation plan, it was reported that children were recruited into the industrial sector (Bangkok Bank Report, 1979). The Ministry of Commerce reported that of 36,000 manufacturing establishments 5,000 exploited children of the age group of 12 years and below. According to official statistics, child labour (11–14 age group) increased from 702,119 children in 1970 to 1,024,200 in 1980 and to 1,179,500 in 1988. Children worked in small factories (less than 50 workers) operated by subcontractors in the textile, frozen food, basic metal, plastic products and glass industries. Approximately 14 per cent of manufacturing establishments in 1979 were using child labour (children under 12 years old) and 17 per cent of total workers of the 11–14 age group were engaged in manufacturing work.

Subcontracted homework in Thailand 329 Table 10.5 Characteristics of labour force in the informal sector compared to those in the formal sector

1 Rate of payment

2 Hours of work 3 Occupation Health and Safety (OHS) 4 Welfare

5 Skills attainment 6 Conditions of work

Labour force in the informal sector

Labour force in the formal sector

No minimum wage rate; wage depending on the amount and quantity of work No overtime payment No fixed working hours and period of working time No control or regulations of OHS. Individual worker responsible for his/her health and safety in work environment Employer/contractor need not provide welfare protection to the employees or contracted workers

Minimum wage determined by law with extra pay for overtime

No access to occupational skill training, depend on one’s own working experience Relatively flexible, can work at odd times or perform other activities at the same time

Fixed working time and hours There are laws to regulate employers in providing OHS requirements for the workers Employer required by law to provide welfare to workers who are also protected by Social Security Act Have access to on-the-job or in service training provided by employer Work under the rigid working hours with specific rest time

Source: Office for Home-Based Workers, Department of Labour Protection and Welfare.

Homeworkers’ conditions have recently improved, thanks to government policies including universal education and family planning initiatives. Since 1980, the Thai government extended compulsory education from 4 to 6 years in state schools. The success is reflected in the high primary enrolment rates for girls and boys9 and high completion levels until Por 6.10 As a consequence, child labour has decreased. According to the NSO HBW census, there are not any workers under 13 years.11 In the same way, the current low fertility rates and small size of households (four members) can be attributed to the family planning policies. High literacy levels,12 especially women’s, had important implications on human development overall. The latter is supported with fieldwork results presented in Section 10.5.

10.3

Research method

Survey design and data collection The methodology of the field study consisted of a survey covering a total of 399 households spread equally among the three sectors (Table 10.6). The survey

131 399

Prov., Northeast

136

305

101

102

102

94

30

34

30

98

—

98

—

28

—

28

—

CG

Notes Dash indicates that certain information was not collected in the survey or data were not comparable. FGDs  Focus Group Discussions. hw  homeworkers. CS  case studies. CG  Control Group or Non-hw household * Considering separately the one for women and the one for children.

Source: Thailand Country Study Survey (2001).

Hybrid seeds production Sub-total

Leather crafts

132

Photho Distr., Prov. Angthong, Central, Song Distr., Phrae Prov., Northern Ratburana Distr., Bangkok Prov., Central Samrong Distr., Samutprakarn Prov., Central Sakon Nakhon and Kalasin

Paper products

hw

CG

Total

hw

Urban

Households surveyed

Location

Sector

207

101

4

102

hw

Rural

66

30

6

30

CG

7

3

2

2

Total

—

—

—

—

Urban

Number of FGDs*

7

3

2

2

Rural

Table 10.6 Surveys on homeworker households in Thailand: sectors, location, number of FGDs, case studies and households surveyed

11

4

4

3

Number of CS

Subcontracted homework in Thailand 331 included 305 households engaged in homework, and 94 households with no such work. The latter were households residing in the same areas as homeworkers, and were used as a reference point or control group (CG). The criteria for the selection of homework households were: (a) household involvement of women and children, (b) the production processes include the use of chemicals that might have a potential impact on workers’ health and (c) products are exported.13 There were no specific criteria for the selection of CG households besides not having members involved in homework. The team of interviewers consisted of eight female and two male students with a graduate degree in social sciences. The surveys were conducted in workplaces including own home, business premises, unregistered mini factories and farm fields. Besides conducting the surveys, they were required to observe carefully and register the working and living environment. The majority of the households surveyed were in rural areas (68 per cent), given that a high proportion of homeworkers from the saa paper and hybrid seed sectors are located in these areas. In the urban areas the survey covered homeworker households engaged in the production of leather crafts. The survey covered the central, north and northeast regions of the country including the provinces of Angthong, Prae, Pathumthani, Ratchaburi, Samutprakarn, Bangkok, Sakonnakorn and Kalasin14 (Table 10.6). Questionnaire The questionnaire design was consistent with the core guidelines provided. Before the actual survey, a pre-test of 10 households was conducted in a slum community of 100 households.15 After the pre-test, it was decided to include questions related to the Thailand context. For example, given the phenomenon of rural–urban migration, households were requested to answer questions related to reasons for migration.16 The questionnaire consisted of 13 sections including: (1) personal information, (2) household information, (3) household economic information (income and expenditure), (4) work-related questions (first job, age on entering homework, skills training, etc.), (5) nature of homework, (6) remuneration/payment, (7) recruitment agreement, (8) production information (raw materials, faulty work), (9) market information (product market/sales), (10) homeworkers organisations, (11) work conditions and health information, (12) children in homework household and (13) questions related to children working for someone other than parents.17 Focus group discussions, case studies A total of six FGDs were conducted18 where women, men and children participated. The main topics discussed during FGDs were (a) the advantages and disadvantages of homework, (b) relationship with the contractor or subcontractor, (c) organisation and networking, prospect and constraints, (d) awareness of

332

Arunotai et al.

government policies and opinions related to protection, (e) work environment and health risks, (f) needs and priorities and (g) suggestions. A total of 11 case studies were conducted which involved homeworkers, non-homeworkers and contractors.19 The participants were mostly married women with pre-school children. Besides requesting general demographic information (age, education level, sex) from the participants, the main topics were the division of domestic responsibilities, household income and income sources, production processes including working hours and conditions, type of relationship with subcontractor, health risks, constraints and future aspirations.

10.4

Homework sectors

The following section describes the three selected sectors including the macro economic context, the value chain and the terms of contract (type of payment, working hours and level of wages). Leather crafts Product market The leather craft industry was introduced in Thailand as a form of agro-industry 60 years ago. At the beginning products were exclusively sold in the domestic market and consisted of simple crafts. In the 1970s the leather craft sector experienced a boom thanks to the investment of foreign brand-companies that outsourced production to Thai leather manufacturers. Thailand was known for its fine craftsmanship and cheap labour. According to some authors the existence of household industry production was another feature for which Thailand was known (Industry Economy Department, n.d. Sriburapa, 1986; Jarint, 1996). Foreign companies included famous brand names such as Jacob, Pierre Cardin, Gucci and Tiptop. Currently there are local brand factories,20 including Calvin and Rama, and subsidiaries of foreign brands that produce for the Thai market. The Thai government played an important role in the boom of the leather craft sector. Policies to promote investment in this export sector included special tax privileges21 and financial support to exporters (Rudeeruen, undated). According to a study on the leather industry (Economic Research of the Chulalongkorn University, 1999) the leather crafts exports can be divided into two periods. During the first period, (1980s) the sector experienced a boom thanks to reasons mentioned earlier. The peak year was 1989 when the export valued increased by 59 per cent from the previous year. During the second period that took place from 1991 to 1996, the sector went through stagnation due to insufficient skilled craft workers and competitive cheap labour from China, Vietnam, India and Indonesia. Basically foreign companies shifted orders to these countries given the lower costs of labour and raw materials. In 1998 the leather craft exports represented 42 per cent of the total value of leather exports (8,500 billion baht) (Table 10.7).22 The major export markets of Thailand’s leather goods are Hong Kong, the European Union and the United States. Demand and production are seasonal, and the highest consumption period

Subcontracted homework in Thailand 333 Table 10.7 Value of exports*, 1995–98 Main export markets (Leather)

1997 (million of baht)

1 Hong Kong 2 European Union 3 United States 4 Taiwan 5 ASEAN 6 United Arab Emirates 7 Japan 8 Australia Others Total

6,020.3 4,416.9 3,449.2 1,505.9 986.8 746.6 535.1 310.1 1,261.2 19,232.1

Markets (artificial flowers, leaves and plants)

1997 Percentage of (million total exports of baht) (1997) 34.47 62.5 1.45 2.6 1.41 2.6 1.16 2.1 6.85 12.4 1.78 3.2 0.96 1.7 0.56 1.0 1.29 2.3 0.12 0.2 50.06 90.8 5.09 9.2 55.15 100.0

1 United Staes 2 Japan 3 France 4 Great Britain 5 Italy 6 Portugal 7 Spain 8 Australia 9 Germany 10 United Arab Emirates Total Others Total

Percentage average exports 1995–98 28.84 26.12 16.97 7.40 5.22 3.60 2.80 1.54 7.51 100.00

Sources: Trade and Economic Information Centre, Department of Business Economic; Trade Statistics Centre with co-operation from Department of Customs. Note * In the trade system, the leather craft sector is classified as ‘wearing apparel including shoes and handbags’.

is Christmas (Jarint, 1996). The world market share for Thai leather exports is approximately 5 per cent, which is quite low compared to the other countries (Chayodom et al., 1999). The increasing demand from foreign markets and the high volume of sales in the international market of the brand name companies mentioned boosted the number of leather factories and mini shops in Thailand during the 1980s. There are currently 100 registered leather shoe factories in Thailand with a total of 32,000 workers on average; and approximately 115 leather bag factories with less than 100 workers.

334

Arunotai et al.

Foreign ordering agency

1 Manufacturer

Workers

2 Manufacturer

Workers

2nd subcontractor

1st subcontractor

Workers

3 Non-manufacturing contractor

1st subcontractor

HBWers

Figure 10.1 Relationship between foreign ordering agency, manufacturer, contractor, subcontractor and workers.

An important portion of leather crafts production takes place in formal manufacturing establishments. However, given the characteristics and requirements of fashion products, part of the production is also outsourced to homeworkers and informal manufacturing establishments. Fashion products such as leather bags and shoes face constant shifts of demand. Thus, mass production is not appropriate given the high costs of machine adaptation and investment required to modify products according to fashion trends. On the other hand, labour-intensive methods can be adapted to the changes in demand at considerably lower costs. For this reason, production of fashion leather products is usually outsourced to informal manufacturing establishments that use labour-intensive methods. The leather craft sector is characterised by a multi-stage production process that allows for various levels of subcontracting. The three levels of subcontracting in this sector are the following (see Figure 10.1): Large-scale manufacturers who own full production capacity factories; Small-scale manufacturers with limited capacity factories, who outsource part of the production to informal units or homeworkers; Non-manufacturing contractors that do not own factories nor production units. These contractors outsource production, originating from foreign orders or small-scale manufacturers, to other subcontractors or directly to homeworkers. The production process is the following. Foreign brand-name companies develop the styles and designs of the products. These companies assign production

Subcontracted homework in Thailand 335 under a license agreement to large and small manufacturers in Thailand. Manufacturers cut and trim the leather pieces and then outsource the ‘craft’ production to homeworkers. The latter are located at the low end of the production chain. They are responsible for the final product process that consists of pieceassembling, gluing, folding and colouring. These activities do not require a high level of skills. Labour process Almost all (99 per cent) of the workers are contracted informally under a verbal agreement (Table 10.8). Workers consider leather craft production as a full-time job given the ten hours they devote to this activity per day.23 The type of payment varies according to the subcontract arrangement. In the first type (individual families) workers are paid by piece. Since work is very irregular, homeworkers are usually engaged in other informal activities including collection of recyclable items, working in rice factories and pitting tamarind pods. In the second type of subcontract, organised groups are also paid by piece, yet apprentices that are under the control of a chairperson receive monthly payments. In the third type of subcontract, workers receive a monthly remuneration and skills are taken into account in the salary level. Novice workers earn approximately 3,000 baht while skilled craft workers receive a higher wage of 8,000–10,000 baht per month. The average monthly income from homework in the three sectors ranges from 2,000 to 3,000 baht, with the exception of skilled workers in ‘factory home’ that receive 8,000–10,000 baht per month (Table 10.8). In terms of income level, ‘home-factory’ workers are better off since they do not incur room and boarding expenses that are provided by subcontractors. Homeworkers that do not work in ‘home factories’ spend approximately 60 per cent of income on housing.24 Another expense that homeworkers incur is the investment on work tools and raw materials. The only exception is organised groups that avoid part of this expense by sharing sewing machines. Expenses regarding sickness and other domestic calamities are also covered by homeworkers. In the event subcontractors share kinship ties with workers, he/she provides extra bonus and cover expenses for sickness. However, this is not the case for most homeworkers that are migrants from different regions of the country and do not share kinship ties with subcontractors. Indeed, ‘individual family’ migrant workers are the most vulnerable group given their lack of organisation and isolation. The saa paper (artificial flowers and other handicrafts) Product market The production of saa paper handicrafts is a tradition that originated in the northern region of Thailand (specifically in Chiangmai, Lampang, Lampoon and Phrae). In the early days, saa paper was used as wrapping paper, traditional medicine scripture paper, firework fuses and for paper fans. Currently it is mostly used to produce artificial saa flowers and other types of handicrafts (notebooks, greeting cards, photo frames and home decorations) which are exported. Products are sold in the domestic market and prove to be very profitable

Table 10.8 Summary of production characteristics of the three sectors Saa paper (artificial flowers)

Leather crafts

Hybrid seeds

Yes Domestic market No (with the exception of petal pressing machine and cutting machine at (sub) contractors’ place) 0.10

Yes Export market Yes (machinery) (Poor hwers only use needle and thread)

No Export market Yes

1.50

0.05

USA, European Union, Japan

USA, European Union, China

Informal sub-sector Location urban/rural

Manufacturing Rural, mainly Northern region of Thailand

Skills Type of payment

Low By piece

Type of contract

Informal and mutual trust 28.4

Manufacturing Mostly in urban areas close to factories. All over Thailand, especially in Bangkok. Medium Monthly and by piece*** Informal, verbal

USA, Japan, Indonesia, Taiwan, China, Vietnam, Netherlands Agriculture Only rural. Northern and northeastern regions of Thailand.

Traditional Main market Use of technology

Share of total exports Main foreign markets

Percentage of workers that receive delayed payments Hwers invest on inputs Average monthly income from hw** Average working hours Seasonal

23.5

Low By kg Formal and written* 27.7

No (some hwers have to invest in saa pond) 1,000–2,000 baht

Yes (sewing machines)

Yes (land)

9,000 baht****

2,000–3,000 baht

8.45

10.85

9.1.0

No

No

Yes

Notes * The terms of the formal written contract consider farmers as ‘sellers’ and not ‘workers’. ** Survey results. Since monthly wage depends on various variables (season, type of hw, etc) these amounts are average monthly income from hw, from the highest percentage of workers. In the case of saa paper sector 31.4 per cent of hwers receive this range of monthly income from hw. In the cases of leather craft and hybrid seeds sector, the highest percentage of workers are 21.6 and 28.7 respectively. *** Depending on the location (mini shop or home) and status of the worker (apprentice or regular worker). **** Level of income depends on level of skills. Thus, this is only the case of skilled workers. Apprentices and novices earn approximately 3,000 baht per month.

Subcontracted homework in Thailand 337 since the raw material is available domestically and its price is not vulnerable to external shocks. However, the domestic market is small25 for artificial flowers compared to the foreign demand for these products. According to an estimate by the Department of Industrial Promotion, the export value of all saa handicrafts was 1,000 million baht in 1994. Currently the production of saa handicrafts represents 0.1 per cent of total value of exports (Table 10.8) and the main export markets are the United States (which takes 50 per cent of Thai exports), Europe (20 per cent) and Japan and other Asian countries (30 per cent) (Table 10.7). It is difficult to trace the exact value of exports due to the lack of registration of products made of this specific material. However, by looking at the export value of ‘artificial flowers’,26 it can be assumed that demand for saa products, and specifically artificial saa flowers, has decreased. The export value of artificial flowers in 1997 reached a value of US$55.15 million and decreased to US$38.42 million, showing a decrease of 17.1 per cent. The main markets for artificial flowers are the United States, taking 63 per cent of total exports of artificial flowers, followed by European countries including Italy, Portugal and France. Saa handicrafts production is an attractive source of employment given the low investment required, relatively simple work process, and fair market demand. It also represents a provisional source of income for the unemployed. Indeed, the Thai Farmers Bank Research Centre Report (2001) stated that during the past 4–5 years the number of artificial flower manufacturers increased due to the rise in unemployment levels. There are two stages in saa handicrafts production. The first stage is the production of raw material or saa sheet paper through the extraction of pulp from mulberry plant.27 The second is the production of handicrafts28 including artificial saa flowers. Skills required for both stages of production are low. Nonetheless, workers responsible for the design and production of handicrafts usually have more experience and artistic skills. Homeworkers are involved in the two stages and thus are at the low end of the production chain. The subcontractor is the link between producers and exporters. There could be additional intermediaries/subcontractors depending on the level of the markets. For instance, products directed to low-level markets are sold in flea markets and include only one intermediary. Higher level markets include department stores and foreign markets. In the latter case, there is more than one intermediary that do not buy the products directly from homeworkers (see Figure 10.2). Intermediaries located at the upper part of the value chain gain the highest revenue from the mark-up prices. On the contrary, low-end workers (homeworkers) receive an insignificant percentage of the total price. For instance, homeworkers are paid 2 baht per piece of saa handicraft for labour and skill as raw materials are provided by the (sub)contractor (Table 10.8), the handicraft they made is sold at 5 baht in a flea market, at 15 baht in a department store and at 60 baht in a tourist market. Labour process The subcontractor hires workers through an agreement based on mutual trust. Since workers and contractors usually come from the same

338

Arunotai et al. RAW MATERIAL PRODUCTION Saa paper sheet making 1.50–2.00 baht per sheet

HWer remuneration

HANDICRAFT PRODUCTION Handicraft making 2 baht per piece

HWers

per piece/sheet 1st intermediary

Whole sale price (10–20% mark-up)

Sheet 8 baht per sheet

2nd intermediary

Small note book 5 baht per piece

Contractor (capital investment + raw material advancement for HBWers)

(100–200% mark-up)

Sheet 15–20 baht per sheet

3rd intermediary

Small note book 15 baht per piece

Low-end market Flea market (Jatujak weekend market)

(100% mark-up)

Sheet 30–35 baht per sheet

Small note book 30 baht per piece

High-end market, Department store

3rd mark-up (100% mark-up)

Sheet N/A

Small note book 60 baht per piece

Market for foreign tourists (Khao San Road)

Figure 10.2 Value chain, saa paper sheet and products.

agrarian villages, the relationship is based on kinship ties. In many cases subcontractors also act as welfare benefactors and provide advance payments and loans at low interest rates to homeworkers. One of the subcontractors’ responsibilities is to provide raw materials and tools of work29 for the two stages of production. The form of payment is by piece and in general workers are punctually paid every two weeks. Saa paper work is a full-time job consisting of nine working hours per day.30 Nonetheless, saa paper production is considered a secondary activity in the

Subcontracted homework in Thailand 339 household economy since it is mostly performed by women (94 per cent of homeworkers are female) who are economically dependent on their husbands who are engaged in agricultural activities.31 The average income from homework ranges from 1,000 to 2,000 baht or less (Table 10.8). Hybrid seeds Product market This sector emerged as part of the national industrialisation initiative (1970–80) aimed at promoting product efficiency and attracting foreign investment in the agricultural sector. Hybrid seeds were first developed in the United States and Europe. Hybrid seeds production consists of the use of biotechnology to produce genetically modified seeds with specific characteristics including higher yields, larger trunks and greater resistance to climate changes. The pioneer investors and producers in Thailand were foreign companies from Taiwan, Japan, United States and The Netherlands. These companies established production centres in the north and northeastern region of the country (Sakolnakorn, Khonkaen and Kalasin), known for its good soil, climate and irrigation system. Given the success of this production several local companies invested in the business including Bangkok Seeds Industry and Siam Seeds (IFCT, 1983; 1986; Business Hybrid seed production, 1983). In a few years the hybrid seeds production expanded and became important in the Thai export sector. The export value of hybrid seeds rose from 293 million baht (equivalent to US$11.6 million) in 1995 to 899.8 million baht (equivalent to US$34.74 million) in 1997.32 According to the Office of Agriculture Economics, the quantity of vegetable seed exports increased by more than 300 per cent (from 477 tonnes to 1,853 tonnes) from 1983 to 1997. Currently this sector contributes 0.05 per cent to the total export value (Table 10.8). The major export markets are the United States, Taiwan, Japan, Indonesia, China, Hong Kong, The Netherlands and Vietnam. The four markets in Asia – Taiwan, Japan, Hong Kong and Indonesia – import approximately 45–50 per cent of the total hybrid seeds export volume. Compared to these countries, Thailand has a disadvantage in terms of designing skills (Jarint, 1996). Hybrid seeds production uses subcontracted work. The main distinctions with the other two sectors studied are that: (a) the contract farming system is not officially regarded as homework by the Ministry of Labour, (b) the productive unit is more than an individual. Usually production involves all family members (family unit) and (c) the workplace is the crop field as opposed to the home in the other two sectors. In general the required level of skills is low, although certain stages of production (e.g. artificial pollination) require more experience. Subcontracting in this sector includes three different levels. At the first level, hybrid seed companies subcontract Thai manufacturers. Hybrid seed companies can either be foreign companies that operate through local subsidiaries, or local companies. At the second level, manufacturers assign production to the seed production centres in each province. At the third level, production centres subcontract farmers in each village for seed plantation and cultivation. There are cases where

340

Arunotai et al.

further subcontracting may occur. This level consists of farmers’ subcontracting additional household members or neighbours for field assistance. The expansion of this sector has had both positive and negative economic effects. The positive effects are the provision of employment to rural workers and the contribution to the local and national economy. In terms of economic value, hybrid seeds have a greater value due to the resistance to diseases and pests, greater longevity and physical characteristics (shape and size) that are more highly valued in the market. The negative effects are related to the implications of all genetically modified organisms. The production of hybrid seeds raises important questions related to the threat to bio-diversity, possible genetic variations in adjacent species and potential economic disadvantages such as a trade monopoly. Another negative effect is the impact on traditional farmers. Traditional farmers do not have access to biotechnology. As a result their products are not as competitive and face market losses. Labour process Companies contract farmers on a formal basis through written contracts and the form of payment is by kilogram (Table 10.8). Although there is a formal written contract, companies are not required to provide the safety nets and legal protection entitled to formal workers. The reason is that farmers are defined as ‘sellers’ rather than ‘workers’. The hybrid seed sector is not organised along groups due to the high degree of dispersion among farmers. However, they work under the supervision of a group leader hired by the seed company. The group leader is responsible for recruiting farmers and assigning work. Family units, rather than only farmers, cultivate the crop. They cultivate the hybrid seeds in a section of their land leaving a portion for rice cultivation. Hybrid seed production is seasonal. On the contrary, rice production is considered the main activity all year round.33 If farmers are not landowners, they rent a neighbour’s land to plant the high yield crops.34 The cultivation process does not require a high level of skills.35 The average monthly income from homework is 2,000–3,000 baht (Table 10.8) and workers are busy full time with average daily working hours of 9.5 for male and 8.7 for female workers.

10.5

The socio-economic situation of homeworkers

The field research reveals important aspects about the current conditions of homeworkers in Thailand. This section presents the social implications of the subcontracting system and the main findings based on the survey results, FGDs and case studies. Homeworker households The overall situation of homeworkers is better than in other Asian countries. Homework, represents an ‘additional income source’ rather than ‘main income source’ as is the case of other Asian countries including India and Pakistan. It is considered an additional source of income by 44 per cent of homeworkers and the percentage of workers that engage in homework due to lack of employment is

Subcontracted homework in Thailand 341 Table 10.9 Average income from homework (hw) and other sources per sector (in baht) Hw household

Paper product Leather craft Hybrid seeds Total average

Average income per household

% from hw

Other sources (%)

7,117 13,064 6,466 8,856

32.9 67.3 80.5* 60.6

67.1 32.7 19.5 39.4

Source: Thailand Country Study Survey (2001). Note * During peak seasons.

very small (19 per cent). In addition, more than half of the women engaged in homework (63 per cent) responded that the main reason for engaging in the activity was to gain additional income. The fact that main income originates from different sources gives homeworkers bargaining power and it is less likely that there would be exploitation. Notwithstanding, homeworkers (‘individual families’) from the leather craft sector are the exception as in this case, the main income comes from homework (80 per cent) (Table 10.9). As for the hybrid seed sector, more than half of their income comes from homework (67 per cent) only during the peak seasons. Their weak bargaining power results in unfair treatment and exploitation practices by the subcontractor. Another important indicator of overall good conditions are the minor differences between homeworkers and CG households. Women in CG have not engaged in homework due to lack of time, which is mostly devoted to household responsibilities and childcare.36 As can be seen, the income level does not influence women’s decision. Similarities among the two groups include high literacy and enrolment rates and overall good nutrition levels and diet habits. A main difference among the two groups is schooling levels. The percentage of children that reach higher levels of school is higher among CG than homeworkers. Only 4.8 per cent of total homeworkers completed Mor 4–637 compared to 8.7 per cent in CG households (Table 10.10). One can conclude that early engagement in homework may be preventing youngsters from achieving higher levels of education. A second distinction is that CG households have better health status than homeworkers.38 An important aspect of homework in Thailand is that ailments originate from contact with toxic substances and the repetitive nature of work. Workers in the leather craft and saa paper industries are constantly exposed to glue smell while farmers in the hybrid seed industry are exposed to chemicals in fertilisers and pesticides (Table 10.11). In addition, all the activities require repetitive actions that cause muscular pain. For example, in the hybrid seed sector,

21.6 16.5 24.6 21.1

52.3 54.8 54.3 53.8

21.6 9.7 24.3 19.8

Note * Including young adults.

10.2 4.8 10.7 9.1

13.6 22.6 9.7 14.2

Elem. Lower Higher Vocational sec. sec.

Source: Thailand Country Study Survey (2001).

Paper products Leather craft Hybrid seeds Total

% of hh member in sch.

Homeworker

2.3 8.1 1.0 3.2

14.8 17.3 22.3 18.3

University % of hh member in sch. 25.0 65.2 41.4 45.6

50.0 4.3 13.8 19.1

6.3 8.7 20.7 13.2

18.7 13.0 17.2 20.6

Elem. Lower Higher Vocational sec. sec.

Non-homeworkers

Table 10.10 Percentage of households in which children* are in school, by educational level

— 8.7 6.9 5.9

University

Subcontracted homework in Thailand 343 Table 10.11 Health conditions

Health risks

Incidence of health problems (%)

Saa paper (artificial flowers)

Leather crafts

Hybrid seeds

Low Dust from materials Dyes and glue have chemical compound 59.7

High Chemical compound in glue and leather tanning 75.0

High Use of pesticides and chemical fertilisers 62.6

Sources: Thailand Country Study Survey (2001) and FGDs, Arunotai et al. (2001).

the pollination process requires long hours of sitting and leaning forward; in the leather sector workers suffer from stiff necks and sore eyes as a result of hours of concentration in manufacturing. Finally, women from the saa paper sector suffer muscular pains from long hours of sitting and assembling the artificial flowers. Nonetheless, the ailments suffered by Thai homeworkers can be considered minor compared to homeworkers from other Asian countries. An important aspect is that Thai workers are aware of the health risks and are likely to take the appropriate measures. The latter is reflected in their frequent use of government public hospitals. Survey results demonstrated that 50 and 64 per cent of homeworkers in the saa paper and hybrid seed sectors respectively, used government hospitals in cases of sickness. In the case of the leather sector, the percentage of workers is lower at 27 per cent as they prefer to buy medicine from drug stores. As was mentioned above, individual families engaged in leather craft production are the most vulnerable since the main income source comes from homework. The majority of workers (91 per cent) from this sector depend on homework income. Indeed, more than half (67 per cent) of their overall income comes from this activity (Table 10.12). At the same time, homeworkers face the direst conditions among the three sectors. As depicted in Tables 10.13a and b, leather craft homeworkers have the highest incidence of health problems, highest number of working hours, lowest conditions in terms of work environment and living conditions,39 and lowest economic status in terms of house ownership and household utilities. Further evidence of their worse condition is the lower share of children attending school. Only 64 per cent of children in this sector attended school, compared to 81 and 73 per cent in the hybrid seed and saa paper sectors respectively. FGDs support the present discussion. According to the discussion, the day-to-day survival need is the prime concern rather than defending workers’ rights.40 Finally, the highest percentage of migrants, the most vulnerable social group, works in this sector. On the contrary, homeworkers from the saa paper and hybrid seed sectors receive most of their revenue from other activities besides homework. In the case of the saa paper sector, homework contribution is 33 per cent. Their main income comes from

Table 10.12 Socio-economic conditions in homeworker households

Percentage of women in hh Percentage of children 0–4 years old in hh Percentage of children 5–10 years old in hh Percentage of children 11–14 years old in hh Average hw household size Literacy rates (%) Hw earners per hh Main reasons for doing hw (%) Percentage of hwers whose main activity is hw House ownership (%) Percentage of local workers Percentage of migrant workers Percentage of hw that have taken loans

Saa paper (artificial flowers)

Leather crafts

Hybrid seeds

94.1 20.7

68.6 24.0

69.3 19.3

32.1

34.1

36.2

19.1

4.0

21.6

3.9 97.1 1.5 Additional income 40.2 68.6

3.7 94.1 1.7 Additional income 33.3 91.2

4.1 99.0 1.9 Additional income 58.4 33.7

94.1 84.3 15.7 0.0

23.5 (migration) 6.0 94.0 0.0

96.0 91.1 8.9 75.2

Source: Thailand Country Study Survey (2001).

Table 10.13a Average monthly income and expenditure per capita (in baht) Homework

Non-homework

Average Income Expenditure Average Income Expenditure Poverty Line household per per capita household per per capita (per capita size capita size capita per month) Paper products 3.9 Leather craft 3.7 Hybrid seeds 4.1

1,825 3,531 1,577

1,185 2,179 1,027

3.6 3.9 4.3

2,402 3,408 1,865

1,468 2,121 1,082

Total average

2,271

1,445

3.9

2,587

1,559

3.9

911

Source: Thailand Country Study Survey (2001).

Table 10.13b Average income per household per month (in baht) Homework household

Non-hw household

Income difference in baht

Average size

Average income

Average expenditure

Average size

Average income

Average expenditure

Paper product Leather craft Hybrid seeds

3.9 3.7 4.1

7,117 13,064 6,466

4,632 8,063 4,211

3.6 3.9 4.3

8,647 13,292 8,019

5,727 7,849 4,436

1,530 228 1,553

Total average

3.9

8,856

5,637

3.9

10,089

6,082

1,233

Source: Thailand Country Study Survey (2001).

Subcontracted homework in Thailand 345 men’s agricultural activities (Case study 2, Annex 10.1). In the case of the hybrid seed sector, homework represents 81 per cent of monthly income (Table 10.9). Migrants are vulnerable and exposed to the highest risks. Migrant workers come from the north eastern region (56 per cent), central and eastern region (28 per cent), and northern region (9 per cent). They have usually completed a primary level of education41 and consider work experience in cities as a ‘certificate of life education’ (Banpasirichote, 1997, p. 61). Migrant worker conditions are the worst among all Thai homeworkers. First of all, the lack of kinship ties makes adaptation to the urban environment one of the most difficult challenges. Their employment search is more difficult due to their lack of a permanent residence and contractors’ requirement of a ‘House Registration Card’. If workers do not have a permanent address, employers require a guarantee letter from a neighbour or relative. However, given the lack of kinship ties workers have difficulty in getting the guarantee letter. An additional obstacle is the highly competitive labour market given the vast number of rural migrants. Another negative aspect for migrant homeworkers is the lack of access to government support and assistance. Despite being in urban centres, migrant workers are dispersed and difficult to reach. Moreover, the dispersion renders the organisation of groups more difficult. According to the NSO survey, there are 100,000 workers that are not members of any organised groups that live in slum areas and work in manufacturing establishments.42 The highest percentage of migrants is in the leather craft sector (94 per cent) (Table 10.12). Given the lack of opportunities, workers live in the slum or in ‘homeshops’ provided by employers where they share limited spaces with other people. According to declarations in the FGDs, migrant workers face the worst conditions.43 Workers in the other two sectors are mostly from the same community.44 Mutual ties play an important role in the relationship between subcontractors and homeworkers. The majority of homeworkers do not have formal contracts. Their agreements are based mainly on informal verbal agreement and mutual trust. This is the case for the saa paper sector where 91 per cent of workers work under mutual agreements (Table 10.7). It is also the case of the second level of subcontracting in the leather craft and hybrid sectors (i.e. contracts between workers from the same kinship ties in mini shops and contracts between farmers and field workers). The fact that subcontractors belong to the same community makes the nature of the relationship friendly and thus cases of exploitation are limited. Thanks to mutual or kinship ties homeworkers receive fair treatment and even benefits from subcontractors. The majority of homeworkers receive payments on time. Only one-fourth of homeworkers in the three sectors received delayed payments (Table 10.7).45 Moreover, the percentage of punctual payments is higher among workers that are subcontracted directly by private companies. Such are the cases of the leather craft and hybrid seeds sectors. Regarding benefits, in a case study from the saa paper sector, a subcontractor provided payment advances upon request for emergency reasons and loans at low interest rates for reasons unrelated to work (e.g. pay school fees, sickness of a household member). Borrowing to pay for household contingencies is very common among family relations.

346

Arunotai et al.

Another important consequence of mutual ties among homeworkers and subcontractors is the low likelihood of ‘debt bondage’. Hybrid seed workers are the most indebted group. However, loans are acquired from private institutions (e.g. Bank of Agriculture and Cooperatives) or cooperative saving groups rather than subcontractors. Loans are used to finance input costs including fertilisers and pesticides. In the urban areas (shoe-making in the slum community and in mini factories), relations between homeworkers and (sub)contractors are less cordial and more formal, which may have an impact on work conditions, payment and other welfare options. Homeworkers are above the national poverty line. Among all households (homeworkers and CG), the income per capita (from all sources) is higher than the national poverty line (911 baht monthly per capita). Homeworkers in the hybrid seed sector earn the lowest monthly income per capita (1,577 baht) followed by homeworkers from the saa paper sector that earn 1,825 baht (Tables 10.13a and b). Homeworkers in the leather craft sector earn the highest level of income per capita (3,531 baht), however they also face the highest living costs for living in urban areas. The level of expenditures from this sector is also the highest with a monthly expenditure of 2,179 baht per capita. More than half of the income (60 per cent) is spent on food, household contingencies and work tools. On the other hand, the expenditure level of the other sectors is approximately half of the expenditure level of workers from the leather crafts sector (approximately 1,000 baht). In addition, homeworkers from the saa paper and hybrid seed sectors are not required to invest on inputs. Homeworkers in the slum area (leather craft sector) are worse off as their livelihood mainly depends on daily earning. The difference between household average income from all sources between homeworkers and CG workers46 is not significant. The average monthly income of homeworkers is 8,856 baht that is lower than 10,089 baht earned by CG households (Tables 10.13 and b). Also the average monthly expenditure levels of homeworkers (5,637 baht) is slightly lower than the level of CG households (6,082 baht). The differences on expenditure per sector, show the same trend with the exception of the leather craft sector in which the level of expenditure of homework households (8,063 baht) is greater than the level of expenditure of CG households (7,849 baht). Regarding income from homework, wages depend on the season, quantity of work and skill level as the case of the leather crafts sector (Table 10.7). The average household income from homework ranges from 1,000 to 3,000 baht for 40 per cent of workers from all sectors (Table 10.14). An important aspect is that the longer the value chain, the lower the remuneration workers receive. This is especially the case of homeworkers in the saa paper sector that receives a small proportion of the value chain that entails two to three intermediaries. In the other sectors, the number of intermediaries varies from one to two and rather than intermediaries, these are mostly subcontractors. Homeworker households and women Women’s empowerment has generated improvement in the conditions of homeworkers. Women’s involvement is predominant in the three sectors studied.

Subcontracted homework in Thailand 347 Table 10.14 Average monthly household income from homework (hw) (in baht) Income

1–1000 1001– 2001– 3001– 4001– 5001– 7001– 9000 No Total 2000 3000 4000 5000 7000 9000 answer

Paper 28.4 products Leather 11.8 craft Hybrid 7.9 seeds All 16.1

31.4

14.7

8.8

2.9

5.9

—

2.0

5.9

100.0

12.7

15.7

13.7

3.9

7.8

2.9

21.6

9.8

100.0

16.8

28.7

7.9

9.9

6.9

4.0

7.9

9.9

100.0

20.3

19.7

10.2

5.6

6.9

2.3

10.5

8.5

100.0

Source: Thailand Country Study Survey (2001).

The highest participation occurs in saa paper production with 94 per cent of women, followed by 69 per cent in the hybrid seeds sector and 69 per cent in the leather crafts sector, of total number of workers per sector (Table 10.12). Empowerment of women, reflected in the high literacy level and management of income from homework, has important implications for homeworkers and the community. For instance, in a case study in the saa paper sector, a woman, Malee, after attending a training course, formed an organised group of saa paper producers and was able to offer employment to 60 female homeworkers from her village. This demonstrates that additional access to training and credit has a significant positive impact on communities. Another consequence of women’s empowerment is perceived in the small household size of the three sectors that consists of 3.9 members, of whom two are income earners (Table 10.12). This implies that fertility rates are low; and that there is a greater availability of resources to invest in children’s education.47 In addition, according to FGDs, women consider children’s education a priority with no gender preferences.48 In earlier days, girls had limited opportunities due to conservative views of women’s role. Currently the situation is different and mothers see education for both genders as a desirable option. The preference for schooling rather than work for children is reflected in children’s household activities. The involvement of children in domestic chores is low. Only 24 and 23 per cent of homework households in the saa paper and leather craft sectors respectively, require assistance from children (Tables 10.15a and b). Children in the hybrid seed sector tend to help more than children in the other sectors; girls help in chores more tham boys. This may be because parents in the hybrid seeds sector spend more time in the fields than at home and thus require assistance in sibling care. Children assist their parents on a voluntary basis for pocket money reasons and without disrupting school activities. An important aspect is that children’s exploitation in their own household was very rare. This finding is reinforced by a report on children’s hazardous work in Thailand which consider that the problem of children’s exploitation starts when they step out of their houses (Banpasirichote and Pongsapich, 1992). Children devote most of the time to school activities and leisure. Approximately a quarter of the day (six hours) is spent in school and the remaining time is spent on recreation.

348

Arunotai et al.

Table 10.15a Percentage households in which children help with chores Homeworkers

Non-homeworkers

Help Don’t No children Total help in school age

Help Don’t No children Total help in school age

Paper 29.4 products Leather craft 23.5 Hybrid seeds 50.5

24.5

46.1

100.0 30.0

20.0

50.0

100.0

24.5 28.7

52.0 20.8

100.0 35.3 100.0 50.0

20.6 30.0

44.1 20.0

100.0 100.0

Source: Thailand Country Study Survey (2001).

Table 10.15b Children’s help with household chores Homeworkers

Paper products Leather craft Hybrid seeds

Non-homeworkers

Girl

Boy

Total

Girl

Boy

Total

24.4 23.2 52.4

48.0 23.0 28.3

29.4 23.5 50.0

29.2 29.2 41.6

17.3 43.5 39.1

30.0 35.3 50.0

Source: Thailand Country Study Survey (2001).

Homeworkers’ organisations Group organisation proves to be beneficial, yet not all homeworkers are aware of government programmes. According to survey results, over half of homeworkers have a group leader. The latter are responsible for coordination between the contractor and workers, recruitment and distribution of work. Group leaders are very common in hybrid seeds given the coordination required with hybrid seed companies. This degree of organisation among homeworkers could be beneficial to the workers’ bargaining power and collective action. The highest levels of organisation were found among homeworkers from the saa paper sector. Workers from the leather craft and hybrid seed sectors were less aware about government initiatives.49 As opposed to other Asian states whose main concerns are associated with survival, Thai workers’ main concerns are related to homework efficiency and organisation. According to homeworkers, the policy priorities (ranked according to homeworkers prioritisation) should be provision of regular/constant work supply; higher remuneration; financial support for the poor, the elderly and the disabled; assistance with marketing products; provision of welfare protection; specialised training (technical skills); occupational support and the increase of access to credit and loans.

Subcontracted homework in Thailand 349

10.6

Initiatives in favour of homeworkers

Government initiatives The Thai government has been very active in the promotion of policies to protect homeworkers. On March 1998, a Committee on Development and Protection of Homeworkers was set up with a tripartite membership including the government, NGOs and the private sector. One of the most important achievements was the inclusion of homeworkers protection as part of the Social Development Policy Plan from 1996 to 2001. In 2001 the Ministry of Labour and Social Welfare allocated loans worth 50 million baht for 1,200 registered groups of homeworkers. Credit programmes consisted of a 20,000–500,000 baht loan with an interest rate of up to 5 per cent and a grace period of 2 years. Other government initiatives included the creation of homeworkers’ fairs aimed at facilitating contact between homework groups and consumers (Krungthep Thurakit Newspaper, 15 September 2001). Another important initiative was the establishment of the HBW office. The main goal is to gather and update information regarding contractors and design effective policy recommendations. Another interesting programme is the Social Investment Fund (SIF) sponsored by the World Bank. SIF provides funding for several purposes, one of which is to develop vocational training including craft skill development and various courses to organised groups. The following initiatives are directed specifically to the studied sectors. In the hybrid seed sector, the Ministry of Agriculture in conjunction with other Cooperatives has created a programme to minimise the negative effects on local traditional farmers. The programme consists of the distribution of seeds of various species to promote cultivation of alternative crops. In the saa paper sector, women have received special support and aid from government agencies. The Ministry of Interior, in conjunction with the Community Development Department, created an ‘income generation programme’ with the objective of introducing housewives to the production of crafts as an alternative source of income. Initiatives towards the improvement of homeworkers from the leather craft sector include the creation of special agencies that provide training, funding for necessary tools and machinery for organised groups of homeworkers. The agencies are the Community Development Department, the Public Welfare Department, the Industry Promotion Department and the Internal Trade Department. Other important initiatives Homenet, an NGO, is conducting an important initiative aimed at ensuring better conditions for homeworkers. During the 8th National Economic and Social Plan (1997–2001), Homenet, in collaboration with governmental agencies, proposed various polices for the protection of homeworkers. Their main achievement is the creation of group organisations to provide better access to training, funding and greater bargaining power to acquire legal protection and social security. Other efforts include the creation of workshops to promote skill

350

Arunotai et al.

development, and the creation of databases for the efficient management of homeworkers’ organisations.

10.7

Conclusions and policy recommendations

Homeworkers have played a fundamental role in the development of the industrial sector in Thailand. As a consequence, the government and other agencies have put special emphasis on understanding the conditions and extent of workers engaged in these activities. Despite the various initiatives aimed at improving work conditions, homeworkers are still vulnerable. The lack of legal protection deprives them of fair remuneration and healthy work environments. Workers are also deprived of safety measures. As a consequence they suffer from ailments caused by constant exposure to toxic substances used in the production processes. The majority of homeworkers in Thailand are female workers that have completed primary level schooling. They engaged in this activity to provide additional income for the household. Their involvement in homework affects the time devoted to childcare and other domestic chores, however, in communities based on kinship ties, women count on the assistance of other kin for these responsibilities. Children’s involvement is limited to domestic chores and to a lesser degree to homework activities. School attendance is high thanks to government policies on universal schooling and parental support. The most vulnerable group among homeworkers is rural migrants. The lack of kinship ties in urban centres hinders their adaptation. In addition, the lack of permanent residence, and dependence on income from homework makes them vulnerable to greater exploitation and poor work environments. Overall homeworkers have limited opportunities in the labour market due to the low level of skills and experience in the formal sector. This is a consequence of their lack of access to training. Recent government and NGO initiatives have partly solved this by promoting the organisation of groups and associations. However, due to the physical dispersion (i.e. in agricultural activities) and isolation in urban areas, a considerable portion of homeworkers have not been able to benefit from these programmes. In order to address the problems mentioned, we suggest the following recommendations. The list includes three kinds of interventions: government policy and planning, direct assistance and technical programmes. Government policies and planning Extension of legal protection for homeworkers Policy-makers should focus on the extension of legal protection to homeworkers. Thailand has several legal instruments that could be applied to the benefit of homeworkers. However, given the existing ‘legal bottlenecks’ due to the unclear definition of homeworkers’ informal status, these are not applicable. Only when these bottlenecks are resolved can important legislation such as the Labour Protection Act or the Civil and Commercial Codes and the Thai Labour Protection Law (1998) could be applied in favour of homeworkers. Inclusion of home-based workers in government bidding processes One of the main drawbacks that homeworkers face is uncertainty about availability of

Subcontracted homework in Thailand 351 work. In this regard, besides the promotion of legal protection, an effective measure could be the inclusion of homeworkers in bidding processes for government products (e.g. uniforms). Similar initiatives were proposed in the past. However, given the strict regulations for government contracting and purchasing these were not successful. Such was the case for the Bangkok Metropolitan Administration50 that attempted to subcontract work for the production of government uniforms.51 Awareness campaigns There should be a continuing effort in promoting public awareness about the important contribution of homework to national economy through the Office for Homeworkers. In addition, smaller provincial units of this office should be created at regional and provincial levels to extend direct assistance to homeworkers in all regions. Direct assistance to homeworkers As discussed ealier, the other main drawback that homeworkers in Thailand face is limited access to skills development. In this regard, we propose the following: promotion of training and skills development courses would ensure higher income and more opportunities in the labour market to homeworkers. Besides production skills, training should focus on the management of small businesses (business planning, accounting, marketing), quality control methods and appropriate use of technology in the production process. Promotion of homeworkers’ groups Ongoing initiatives aimed at promoting the organisation of groups have proven to be positive in increasing homeworkers’ bargaining power. Thus, additional resources from national agencies and other funding organisations should be directed to these types of programmes to be able to implement them at a national level. Incentives to businesses and industries There should be incentives to businesses and industries involved in subcontracted homework to provide fair remuneration and other forms of labour protection to homeworkers. Provision of credit Credit and market support should be extended to homeworkers. Greater access to credit will solve financial constraints in purchasing equipment and machinery. Creation of databases The creation of databases, including directories of homework organisations and contractors or subcontractors should be encouraged and supported. The NSO survey is an important resource. However, in order to ensure accurate information, definitions and conceptual framework should be re-established. In addition, crucial information such as gender and industry type should be included in the survey design. Future research A database will be useful to carry out longitudinal studies. Through this type of study, where the ‘working life’ of selected homeworkers are monitored for a long period of time, government officials and researchers will be able to measure the impact of policies implemented.

Annex 10.1: Case studies Case study 1: homeworker in the leather craft sector Malee aged 26 is a migrant from the northeast region. She and her husband have two children. The older one (8) lives with her mother in her home village and

352

Arunotai et al.

attends school. The younger one, who is still pre-school age (3 years old), lives with her. Malee began work at the age of 15 after she finished compulsory education at higher primary level (form 6). Her first employment was as an unskilled construction worker. She changed jobs about three times mainly in the manufacturing sector. She left the last job in a biscuit factory because she had her second baby and had to look after her full time. She had to depend on her husband who was a salaried worker in a flourmill earning about 9,000 baht a month. Malee began to take up homework when her younger daughter was 1 year old. She found the work through her neighbour who was a subcontractor who received contract work from a shoe factory in the neighbouring area. The subcontractor provided her with one-day training on how to sew the soles of the slippers together. She works about 9 hours a day, 6 hours during the daytime and 3 hours at night time. During the day she began work at 9 am after finishing all the housework and childcare. She would stop work at 4 pm to prepare dinner and care for the younger daughter. She resumes work from 9 pm to midnight after her daughter is put to bed. She works alone without any help, sewing the soles of the slippers together. Every day, she sews about 20–30 pairs with the rate of 1.50 baht per pair (small sizes). This means she earns an income of about 1,000–2,600 baht per month. Every day she delivers the finished slippers to the subcontractor and takes a new lot, which may vary in quantity depending on the supplies from the factory which again is dependent on the export orders. Malee is paid in cash every two weeks according to the amount she has delivered. Family finance In all, her family earns about 10,000 baht per month. House rent and food account for about 60 per cent of this. She sends home about 3,000 baht every 2–3 months to her mother as expenses for her older child. Relationship to the subcontractor Her relationship to the subcontractor was an informal and friendly one. Because she lives next door, Malee need not provide any document such as a copy of her identity card as a guarantee. The agreement was made verbally. The subcontractor will do the first quality control. If the slippers she made do not pass the quality control at the factory, she has to re-do them and send them back again. Malee sees the advantage of homework as an alternative source of employment for a housewife confined to the home. The only drawback in her mind was the irregularity of work. Case study 2: homeworker in the saa paper sector Napa aged 30 was born in a rice farming family. She completed her education to higher secondary level at the District school. Her first job was to work as a child minder at the Sub-District Day-Care Centre during the daytime. In the evening she found another job as an assistant in a dental clinic. She was employed and earned two incomes for several years until her contract with the Child-Care Centre ended and her dentist employer moved to Bangkok. She found another income source by doing contracted work (sewing shirts) at home earning about 60 baht per day.

Subcontracted homework in Thailand 353 She was recently married to a man from another district who then moved into her village and they have a one-year-old son. She and her husband were given a small plot of farm land of about 3 rai (about half a hectare) where they grow glutinous rice for their own consumption, and sweet corn as a cash crop. Her husband was a full-time farmer growing two crops a year. At peak periods additional labour had to be hired because Napa was engaged both in childcare and the paper making which she took up just after she settled down. Napa works under one of the lady contractors in the village who provides her with capital and raw materials for production, namely the mashed mulberry fibre, cash to buy flowers to decorate the paper and others. In one day she could produce about 100 sheets of plain mulberry paper or 70 sheets of decorated ones (with flowers and leaves). She gained 1 baht for plain paper and 1.50 baht for the decorated one. In all she earns about 100 baht a day. Homework and house chores Being a housewife, Napa finds herself having to shoulder most of the housework before she can begin work on the paper making at around 8 am. She has first to finish all the household chores such as cooking, feeding her son and washing clothes. She has to get up at around 5 am. During her working day, she would stop at intervals to (bottle) feed her son and give him food. Napa sees her homework as complementary to her child care duty. Although childcare may disrupt her work or reduce her productivity she sees it as necessary and as providing beneficial rest time. Last but not least, Napa was fortunate to have her old mother (70 years old) living with her who can help her by lessening the burdens of childcare and cooking. Once in a while especially at peak time of production she has to depend on the community child care centre. On normal days she works from 8 am with a few hours break at noon for her son until about 4 pm. She would spend evening hours looking for flowers for paper decorations around the neighbourhood. At night after dinner and putting her son to bed she would continue working sometimes with her mother, cutting and preparing the flowers for the next day’s work. Relationship to the contractor Napa, like other homeworkers in her group, finds herself very fortunate to work under the woman contractor who is also one of the community members and who lives only a few houses away. She respects the contractor as her elder kin. At times of need for cash contingencies she can always seek help or ask for advances from her contractor. Payment is made at the end of the second and fourth weeks of the month. Household finance Napa earns about 2,000–3,000 baht a month from her homework. Although the amount is not big she regards it as crucial to her family survival because it is the income she can earn almost every day as long as there are orders from the market. She uses her earnings to buy food for the family and on her son (powdered milk). Her husband on the other hand earns about 7,000–8,000 baht a year from selling sweet corn. However, she and her husband have to take a loan from the BAAC every year for the investment on the cash crop. It is Napa who keeps the money earned and manages it by depositing some in the bank.

354

Arunotai et al.

Case study 3: A non-homeworker Pirun, housewife, aged 31. Born in the northeast, she completed upper secondary education in Bangkok. She began work at the age of 19 as a clerk at a jewellery shop in Bangkok. She was in salaried jobs for 2 years before she got married at the age of 21. After she was married and had her first child she had to leave the job and became a full-time housewife living with her husband who is a construction subcontractor. At present she has 4 children, the youngest one aged 2 years. Pirun did take up home-based contracting jobs (pasting labels on bottles) for a while but found that it paid too little compared to the amount of time she had to spend and the fact that she then had no time to look after her young children, or for housework. Her husband also did not want her to continue with this work. She never again took up homework. Her main reason was that her family has six members and two of her children were pre-school age. Apart from child care she had to do the household chores of cleaning, washing and cooking all by herself. She began work early in the morning preparing food for all members and feeding the children. Clothes washing (manually) took a considerable time. She finished the work at around 1 to 2 pm and began cooking again at round 4 pm waiting for the school children and her husband to come home. She sought help from her sister-in-law who was a single woman and lived next door to help her with the child care while she was busy washing clothes or cooking. Pirun regarded her situation as one where it was not possible to do any homework and that the return was not worth the effort. Another reason was her lack of skills due to her long absence from income earning employment. Moreover, she could not cope with the pressures which tend to arise in homework especially during the peak periods.

Notes 1 This is only an estimate since unregistered manufacturing establishments were not counted in the official census. Unregistered manufacturing establishments are informal manufacturing units that engage less than ten workers. 2 According to the Survey of Migrant Population in Bangkok Metropolis (from 1976 to 1981), ‘economic motivation’ rather than ‘studies’ was the main reason why youngsters (age group of 10–19) migrated (Somsak, 1985). 3 Earlier initiatives for measuring homeworkers include the 1986 Report on Employment Situation of Homeworkers published by the Ministry of Interior, and a survey conducted by the Department of Labour Welfare and Protection in 1992. This survey was the first that quantified homeworkers at a national level. 4 The Thai word means contracted work or piece work done at home (Fujimake, 1995). 5 Those who are confined at home are usually housewives responsible for domestic chores and childcare, pregnant women, older women or disabled workers who cannot engage in formal employment or heavy work. 6 The Investment Promotion Act established in the 1970s provides incentives to foreign investors including the application of relaxed enforcement laws in terms of safety standards and labour protection. 7 In 1980, 82.8 per cent of children were working in the agricultural sector and did not complete primary school (grade 4 or grade 6, or before the age of 12 or 15) (NSO, 1980).

Subcontracted homework in Thailand 355 8 In the Board of Investment brochure (1978) there is a picture of working children with the caption: ‘Cheap Labour the Key Incentive for Industrial Investment in Thailand.’ 9 Primary enrolment rate is 98 per cent (1998). 10 Por 6 is the equivalent of sixth grade of primary school. 11 In the 1999 Labour Protection Act, a child labourer means a worker who is under 15 years. 12 The literacy level is 90 per cent for both men and women. 13 In the selection of households in the leather craft sector, an additional criterion was included. The types of subcontracts selected for the study were: (a) individual families, (b) organised groups and (c) households in ‘factory-homes’. In the first type, subcontractors from small manufacturers hired individual families that lived in urban areas in temporary shelters. In the second type, subcontractors hired households from organised groups such as the White House Shoes Group, and Khlong Jet Leather Bags Group in Pathumthani Province. Households from this category lived in outer urban areas and worked at home or at a shop under the supervision of a chairperson. In the third type, subcontractors hired workers and provide them with room and board in the so-called ‘factory-homes’. 14 These provinces were selected by the field research team and the Office of Homeworkers, Ministry of Labour and Social Welfare. 15 The pre-test consisted of a survey of ten households that lived in a slum community in Samutprakarn Province and were engaged in homework for the production of shoes. The respondents were mostly female of the age group 28–52 years old who had been engaged in homework for 5–12 years. The pre-test showed that the questionnaire was clear and straightforward. The final questionnaire interviews were conducted in late November and early December, 2000. 16 In Section 1 Personal information, Question 1.2 ‘(If not born here) Number of years residing at this place’, and Question 1.3 Reason(s) for moving/migrating here, 1.4 Language used at home. 17 Compared to the UNICEF guideline survey, the Thai questionnaire has seven additional sections. As seen earlier, these sections are subdivisions of topics mentioned in the UNICEF survey. For example, the sections of ‘remuneration’ and ‘recruitment agreement’ are topics included in the section on information of household activity and work-related questions respectively. A second survey was directed to non-homework households which excluded sections 5, 6, 7, 8, 9 and 10 and consist only of six sections. The time devoted to each survey was 25–40 minutes in the case of homeworkers and 15–25 minutes for CG households. 18 Two FGDs were conducted in the saa paper product sector of which one involved children and the second only adult workers. Of the three FGDs conducted in the hybrid seed sector, two included adult workers and involved children. Finally, only one FGD was conducted in the leather crafts sector and it involved adult workers. 19 Three case studies were conducted with homeworkers in saa paper products and four each in leather crafts and hybrid seed sectors. 20 These are located in Bangkok and other regions of the country (Seri, 1984; Nitaya and Ratana, 1999). 21 Leather manufacturers from central provinces near Bangkok are entitled to tax exemptions in the import of tanned leather, synthetic leather and machinery. In addition, all leather exporters are eligible to receive credit from the Bank of Thailand of Exporters and tax returns from the purchase of raw materials. 22 In 1998 (last quarter) the exchange rate was 36.69 baht equal to US$1. Source: EIU. 23 ‘Malee worked about 9 hours a day, 6 hours during day time and 3 during the evening.’ See Case study 1 Annex 10.1. 24 This was the case of Malee: ‘House and rent accounted for 60 percent of monthly income’ (source: Case study 1, see Annex10.1). 25 Domestic market is limited due to consumers’ preference for fresh flowers available throughout the year.

356

Arunotai et al.

26 Department of Customs defines ‘artificial flowers’ as products made of cloth, silk, plastic, paper and saa paper. 27 Saa paper pulp is extracted from mulberry or ‘por saa’ located in northern region of Thailand. The plant is basically crushed to create pulp/tissue that is then put into cement water tanks. It is then spread evenly with a nylon screen and left to sun or wind to dry. The dried product is the saa paper that is then transformed into handicrafts. 28 Production of flower parts consists of stretching and cutting wires, rolling coloured papers over flower stem, making pollens stems, gluing petals to stem and so on. 29 Each contract worker receives 2 kg of wet fibre paper a day to produce at least 100 paper sheets. 30 Production of saa paper requires long hours of standing and in particular lifting wooden frames of wet mulberry paper (Puangthip, 1994). 31 Selected communities of the field research were agrarian (Ongkarak and Ban Nuun). Farmers were involved in the production of tobacco, rice, corn as well as cattle and stock. In Case study 2 (Annex 10.1): ‘Napa’s husband cultivated the rice crop for their own consumption and sweet corn as cash crop.’ 32 The exchange rate in 1995 (last quarter) was 25.16 baht/US$. In 1997, the exchange rate (third quarter) was 25.9 baht/US$ (source: EIU, 1996 and 1997). 33 Two-thirds of farmers regarded rice farming as their main occupation despite their involvement in hybrid seed production. 34 In a case study from Ban Rai, the subsidiary of a foreign company subcontracted farmers for the production of different types of pepper. Farmers allocated their own land and sometimes rented neighbour’s land to plant these crops. 35 Cultivation consists of the preparation of wet sand buckets where seeds are placed for one week. This is the germination period. After one week, seeds are put in soil bags for one month to be then transplanted in the field, previously prepared. The next stage in the production is the artificial pollination that consists of pollen extraction from male flowers. These are placed in small round containers and then pollinated into female stems. The final stage is harvesting, which consists of the extraction of seeds from fruit resulting from the artificial pollination. The seeds are cleaned and dried and handed to the companies for quality control test. 36 For example, ‘Pirun did not engage in hbw because she did not have any time left after looking after her children and doing housework’ (see Annex 10.1, case study 3). In another case study, a woman called Manvika decided not to work at home because she required time to take care of her 5 month old baby, and her mother who was 77 years old. In addition, household chores took up much time. Source: case study. 37 Equivalent to the upper secondary level of education. 38 According to the survey results from the CG, Homeworkers are worse off. The average percentage of respondents that declared to have ‘no health problems’ was 73.2 per cent in the CG and only 34.2 per cent in the homeworkers group. 39 In order to determine the nature of the work environment, the survey defined poor work conditions in terms of space, noise and ventilation. Among the three sectors, the leather craft sector presented the poorest conditions in every criterion (especially ventilation). 40 Women’s perception is that ‘it is better than nothing since part of household daily expenses can be covered by their small earnings’ (source: FGD). 41 In Case study 1, Malee only reached primary level while Napa in the saa paper sector (Case study 2) reached secondary level and had a higher level of jobs before engaging in homework. 42 NSO Survey (1998). 43 A small group discussion among six homeworkers in an urban community in Bangkok revealed that migrant homeworkers find themselves to be the least fortunate and the most vulnerable. They had trouble finding jobs due to the lack of a registration card and permanent address. Ultimately they could become employed in the shoe factory thanks to a neighbour’s guarantee.

Subcontracted homework in Thailand 357 44 84 per cent and 91 per cent of workers are local in the saa paper and hybrid seed sector respectively. 45 The average percentage that receives payments on time was 68.1 per cent among the three sectors. 46 It should be noted that the average size of households for both homeworkers and CG is 3.9. 47 In a case study from the saa paper sector, Panee saved income from homework in a bank account and planned to use it for her children’s higher education: ‘Her aspiration is to send her 2 sons to highest education possible and allow them to be more than farmers.’ 48 According to the survey, 61 per cent of homeworkers valued equally the education of both girls and boys. 49 In a FGD, it was revealed that although the homeworkers resided in the urban centre close to Bangkok they had hardly any information or knowledge about government policies and programmes. Their exposure and contact to the relevant government offices and officials were minimal. 50 This programme was part of the Bangkok Metropolitan Administration’s special funds to alleviate the problem of unemployment in Bangkok. 51 Homeworkers were not eligible to participate in this bidding because requirements included availability of capital and high level of skills.

References Arunotai, N., Gordon, N., Hiranvorachat, P., Jarubenja, R., Katleeradapan, N., Petchprasert, N. and Pongsapich, A. (2002), Thailand Country Study Report. Outsourcing of Manufacturing to Households: Subcontracted Home Based Work in Thailand (unpublished paper). Bangkok Bank (1979), Biannual Report, Bangkok: Bangkok Bank. Bank of Thailand (2001), Economic and Financial Statistics, February (in Thai), Bangkok. Banpasirichote, C. (1997), The Situation of Child Labour in Thailand: An Overview, Bangkok: Chulalongkorn University Social Research Institute. Banpasirichote, Chantana and Pongsapich, A. (1992), Child Workers in Hazardous Work in Thailand, Bangkok: Chulalongkorn University Social Research Institute. Benja, Jirapatrapimol (1995), ‘Women and Home-based Industry (in Thai), Journal of Social Research, vol. 18, no. 2, pp. 45–54. Business of Hybrid Seed Production (1983) (in Thai), Bangkok: Business Research Section, Research and Planning Division. Chayodom Sabbhasri et al. (1999), ‘Case Study of Leather Industry and Leather Crafts’, The project on economic studies of 5 industrial goods in Thailand (in Thai), Bangkok: Economic Research Centre, Faculty of Economics, Chulalongkorn University. Data Source: Thailand Country Study Survey (2001). Department of Industrial Promotion, Ministry of Industry (1994), Pattterns of Thai Handicraft Products, Bangkok: DIP. Fujimake, Motoki (1995), Informal Economic Activity: General Thinking and Case Study in Bangkok. Faculty of Economics, Thammasart University, IFCT. IFCT (1983), Business of Hybrid Seed Production, Bangkok: Business Research Section, Research and Planning Division. IFCT (1986), Export Industry Report, Bangkok: Bangkok: Business Research Section, Research and Planning Division.

358

Arunotai et al.

International Labour Organization, Asian Regional Team for Employment (1996), A Policy Agenda for the Informal Sector in Thailand, Bangkok: ILO. Jarint, Jaroensriwatthanakul (1996), Prospect of and Potential for Exporting and Impacts of AFTA: Tanning and Leather Industry Tanning and Leather Industry, Thailand Development Research Institute. Narong, Petchprasert et al. (1991), Structural Analysis of Unemployment and Underemployment in Thailand (in Thai), Bangkok: Social Research Institute and Friedrich Ebert Stiftung Foundation. National Statistical Office (1980), Office of the Prime Minister, The 1980 Population and Housing Census of Thailand, Bangkok: National Statistical Office. National Statistical Office (1999), Office of the Prime Minister, The 1999 Home Work Survey, Bangkok: National Statistical Office. Nitaya, Katleeradabhan and Ratana, Jarubenja (1999), Child Workers in Samutprakarn, Bangkok: Social Research Institute. Pongsapich, Amara et al. (1991), Women Homeworkers in Thailand, Bangkok: Chulalongkorn University Social Research Institute. Puangthip, Tantirat (1994), Mulberry Paper Handicraft Products (in Thai), Bangkok: Industrial Promotion Department, Ministry of Industrial Works. Rudeeruen Ruengron-asa (Undated), Leather Crafts and Travelling Accessories in the United States (in Thai), Bangkok: Ministry of Commerce. Seri, Leelalai (1984), Child Labour Employment in Some Industries: Case Study in Bangkok and Samutprakarn, Bangkok: Faculty of Economics, Ramkamhaeng University. Somsak, Tambunlertchai et al. (1985), Change in the Industrial Structure and the Role of Small the Industrial Structure and the Role of Small and Medium Industries in Asian Countries: The Case of Thailand, NESDB. Sriburapa, Trisanawadee (1986), A Study on the Operations of the Leather Products Industry in Thailand (in Thai), Master of Commerce Thesis, Chulalongkorn University. Surachai, Leechanapaiboon (1996) Thailand Ready-made Hides Export, unpublished MA Thesis, Faculty of Economics, Chulalongkorn University. Thai Farmers Bank Research Centre, Economic Outlook (2001), Thai Farmers Bank Research Centre Co. Ltd. No. 864, 16 March. The Economist Intelligence Unit (EIU) (1996), Country Report Thailand, Main Report. The Economist Intelligence Unit (EIU) (1997), Country Report Thailand, Main Report. The World Bank (2000), World Development Indicators 2000, Washington, DC Wanchai, Janprasert (1998), Seeds Technology, Bangkok: Kasetsart University Press.

Part III

Policy implications

11 Upgrading informal micro- and small enterprises through clusters Towards a policy agenda Santosh Mehrotra and Mario Biggeri

11.1

Background

The development strategies adopted by many developing countries in the early post-independence period were generally guided by the classical theory of economic development in which emphasis was placed on capital accumulation for maximisation of GNP growth. According to the Harrod–Domar model (Harrod, 1939; Domar, 1941), and the initial experiences of Soviet Union and other countries, the increase of investment was central for economic growth and industrialisation. Capital appeared to be the ‘missing factor’ of developing countries, and this was translated from a policy perspective to an increase of investment in formal modern activities through private investment, public investment and by external capital (including FDI and/or international aid). It is well known that in this sticky model there was no space for the substitution of factors to relieve capital shortage (and generate labour intensive growth) nor to reallocate factors between sectors. Consequently, as we noted in Chapter 1, growth of output in the import-substituting industrialisation strategy was linked to production patterns and technology that were somewhat biased in favour of capital-intensive modes and employment creation was not a central part of this type of development strategy. It was assumed – and still is by the international financial institutions (especially the IMF) – that the rapid growth of GNP and investment in human capital would suffice to bring about both economic development and improvements in living standards for the poor (World Bank, 2000).1 Although capital remains central in the process of development many development economists pointed out that the underlying Harrod–Domar model is too simplistic a representation of the growth process, and many other factors besides capital accumulation affect development. Furthermore, past experience in most developing countries has shown that economic growth alone is not a sufficient condition for meeting broad-based human development goals.2 Economic growth was inadequate for addressing employment and poverty problems in the short to medium run, because the ‘trickle-down’ process failed to materialise (ILO, 1997).3 At best, whatever ‘development’ took place as a consequence of growth-oriented strategies, it was concentrated in the modern sector. Except in the three East Asian

362

Santosh Mehrotra and Mario Biggeri

late industrialisers (Japan, South Korea and Taiwan), employment grew slowly in the formal sectors of industry, but has grown in the informal economy (in both industry as well as services). This is especially true in low income and low-middle income countries where informal employment generates low incomes and is without social protection, and often involves subcontracting to informal micro- and small enterprises. In other words, the informal economy still employs the vast majority in these economies4 (see Table 1.1. in Chapter 1) and forms the core of the income earning activities of the poor. Furthermore, as we saw in the five countries studied (except in some of the South East Asian countries), formal sector employment in industry has remained an ‘enclave type’ of development. The underestimation of the role of the informal sector5 in developing countries too often contributed to inappropriate analyses (Kantor, 2000). The critical observation is that despite the large presence of informal activities, they are still neglected in industrial and development policies of many developing countries. If the strategy of the two synergies is to be successful – as explained in Chapter 1 the first synergy is between interventions within basic social services and the second is between income increase, its better dispersal, and health and education outcomes – economic growth has to be recorded in sectors where the incomes of the poor comes from. Furthermore, most of the Millennium Development Goals (MDGs), including the first goal, ‘Eradicate extreme poverty and hunger’, cannot be pursued without the passage of informal activities from a passive – or buffer – to an active role in the development process. We argue in this book that the growth and the development of informal and formal micro, small and medium enterprises and the development of local systems of development can be considered a direct approach to poverty reduction and human development, contributing at the same time to economic growth. The strategy, thus, should be to remove the constraints for the upgrading of informal activities into micro-enterprises and Small and Medium Enterprises (SMEs) without forgetting the demand side, domestic and international.6 As we have reported in Chapter 1, one of the most successful strategies – followed, with different paths, by some late industrialisers (e.g. Italy) and by some developing countries – is to develop the economy at local level through clusters of SMEs.7 Besides small firms can have a central role in local development not only because they generate more employment than large ones, but because healthy small enterprises in a cluster can foster the subcontracting system. Furthermore, Mead and Liedholm (1998), after reviewing the evidence, also argue that growthoriented small firms generally create employment of good quality, in that jobs created by expanding existing small firms are more productive than those that result from new small start-up firms. Moreover, they also show that returns per hour of family labour are substantially higher in firms with 2–5 workers (and even higher in those with 6–9 workers) than for those with one person working alone. Considering the varieties of clusters and the different endowments and institutions which characterise a locality, unfortunately there is no general recipe. Indeed, the industrial organisation can vary considerably – with informal activities that can

Towards a policy agenda 363 be well integrated to completely dis-articulated from the rest of the manufacturing sector; they can be part of a cluster of SMEs and be related (or not), at the same time, to large firms which can be national or foreign, or form part (or not) of a national and/or international value chain. We analyse homework and informal activities within a local economic system perspective – rather than the value chain perspective used in Chapter 38 – because both promotive action and social protection upgrading need to have solid roots in the local economic system to be effective. Like Giacomo Becattini (2004), a wellknown Italian industrial economist on industrial districts, we stress the need to change the coordinates of thinking ‘from per capita GDP to local well-being’. At least five observations are relevant in this context. First of all from a theoretical and practical point of view, informal activities (including homework) should be considered as forming part of a wider economic structure rather than treat them in isolation, with the tendency of considering these merely as urban activities marginal to the development process.9 Therefore, we affirm that these activities can become part of a process of development of the local economic system, and the upgrading of informal micro- and small enterprises should be one of the goals of industrial policy in developing economies. Even policy-makers generally recognise that clustering raises the cost-effectiveness of their support programmes and therefore propose to foster organised groups to develop cooperation among the occupants of clusters (Levitsky, 1996, p. 38). As we have seen in this book, our case studies are characterised by generic clusters of homeworkers and micro, small manufacturing firms, often dependent on the orders of larger industrial firms or traders sometime external to the cluster (i.e. delocalisation of production and value chains at national and international level). The second observation, related to the first, is that informal activities including homeworkers are studied separately from those for SMEs, although they may reside in the same local area and are often networked at least through subcontracting and via intermediaries (as discussed in the country studies). In the countless theoretical and empirical papers on SMEs and clusters in developing countries there still is a vacuum10 due (we think), to a mismatching of ‘interests’. On one side there is the ‘syndrome of studying success’. Clusters to be studied by economists and industrial economists need to be successful (at least once in their life) or need to present strong potential, so the support programmes of international agencies tend to select the ‘right’ clusters to demonstrate success and policy makers want to show successful cases. Thus, in the literature there is much evidence of these successful case studies. On the other hand, non-successful clusters, or unsuccessful parts of successful ones are analysed by those interested in poverty and in the reduction of exploitation of workers at local and international level. This lack of integration in research has its effect at policy level. Policy-makers now generally are not integrating informal activities development with the upgrading of clusters and of local systems of development, just as in an earlier phase of industrial growth they did not properly integrate the promotion of SME clusters with broader industrialisation strategies and objectives, despite having some ad hoc policies to promote small enterprises. Only some Membership-based

364

Santosh Mehrotra and Mario Biggeri

organisations of the poor (i.e. MBOPs defined more in general as community-based organisation – CBOs – for example SEWA, Homenet, PATAMABA) seem to have clearly understood the issue. They wish to transform informal activities and homeworkers subcontracting from a survival activity and a form of network exploitation to an opportunity for social and economic development. The third observation is that in planning for local system of development it is difficult to separate production policies from social protection policies (hence this chapter should be read in conjunction with Chapter 12 on social protection for informal workers). For instance, thinking about skilful homeworker-artisans like Pakistani women weaving carpets it is easy to understand that productivity is only one side of the story. As John Sender (2003, p. 419) writes about rural poverty and gender: If the poorest women now rely, and will increasingly depend in the future, upon wage incomes to survive and to escape from poverty, then it is not clear that fashionable policies providing them with training to make baskets, or offering them micro-credit to facilitate start-up enterprises in rural environments already over-supplied with similar and failing enterprises, are sensible. Aid agency policies now focus on ‘capacity building’ in all manner of ineffective, small-scale and corrupt decentralized organisations – NGOs, CBOs, Group Credit and other financial institutions, but any organisation that has a realistic prospect of increasing the political and economic bargaining power of the lowest-paid wage workers is shunned, or dismissed as potentially ‘market distorting’ and, ipso facto, harmful to the poor. There is, for example no support for, or even discussion of, the need to allocate resources to support the formation of trade unions by seasonal agricultural labourers. Nor is there support in the mainstream literature for effective legislation to monitor and enforce the rights of migrant domestic servants or women employed in garment sweat shops. Indeed, any state intervention that might create an ‘enabling environment’ for more effective struggle by such workers is likely to be rejected on the grounds that it might compromise the viability of small enterprises. In fact, the voluminous literature on poverty published by the aid bureaucracy and its consultants studiously avoids mentioning the specific organizations, the legislation, or institutions that have historically been most significant in defending the human rights and living standards of the poor in capitalist labour markets. At the same time, there is no discussion of the economic policies and the precise forms of state investment and intervention that might be capable of promoting the growth of female-wage labour-intensive enterprises. There is clear evidence that particular crops, agro-processing and other export industries employ relatively large numbers of women for wages. However, appropriate sectoral policies and industrial strategies are unlikely to be developed, if the most influential development economists insist that poverty can be reduced only when markets are deregulated and when states abandon their old-fashioned aspirations to formulate industrial policy. (Sender, 2003, p. 419)

Towards a policy agenda 365 The fourth observation is that there may be some constraints on industrial policy in today’s international environment. International Financial Institutions (IFIs) policies not only favour MNCs, but also reduce the scope for national government actions.11 Furthermore, as we wrote in Chapter 1, in the developing countries, governments under pressure from both local and foreign investors and from IMF and World Bank loan conditions, have too often allowed labour standards to be defined by the demands of supply chain flexibility: easier hiring and firing, more short-term contracts, fewer benefits and longer periods of overtime (Oxfam, 2004a). However, the claims about the death of industrial policy are premature, and despite WTO regulations, there is much scope still for countries to adopt industrial policies (Chang, 2002, 2003). Finally, it is important to avoid unrealistic assumptions about the development potential of clusters. That is, not all the clusters studied in this book including the sub-clusters of homeworkers can be upgraded in a virtuous development path of increasing production and social protection, but at least all of them can take the first steps to local development and poverty alleviation. In Section 11.2, we examine the phenomena of clustering in developing countries starting from informal activities. Although we are going to recall some of this literature theoretically and for practical examples, a literature review is beyond the scope of this chapter. In particular, we focus our attention on how it is possible to upgrade the clusters including informal activities and thus homeworkers. In Section 11.3, in order to advance the theoretical debate, we expand the theoretical framework presented in Chapter 1 based on the dirt road, low road and high road framework. Then, in Section 11.4, we discuss some experiences of successful clustering, especially Chinese rural SMEs or Township and Village Enterprises (TVEs). We also examine historically some Italian successful cases of passage from the Low Road 1 to the High Road 1 (see Section 11.3 for the meaning), famous for incorporating social protection as an asset rather than as a cost for the local system of development. In the last sections we concentrate on some best practices and on policies on how to promote clusters (Section 11.5), and micro-enterprises (Section 11.6).

11.2

Informal activities, clusters and local systems of development

Case studies and field research contributed to an understanding of informal activities and to advance the theoretical debate on their role in economic development, suggesting that the informal economy can be far from ‘stagnant’ and ‘non-dynamic’ (Ranis and Stewart, 1995) as it is characterised in the ‘modernisation view’. Indeed, the role of informal activities can go beyond the survival function: enhancing employment generation (in the labour intensive sector); poverty reduction (job income, income diversification); output creation; internal/domestic trade expansion opportunities; entrepreneurship formation; local human resource development; the transformation of savings into local investment; the use of appropriate technology in relation to local resources, and skills endowment and a production oriented to satisfy local demand (which can become also a constraint if not fostered, as we already highlighted).

366

Santosh Mehrotra and Mario Biggeri

In other words, on the one hand, the informal sector is a symptom of economic dysfunction and seen just as a low-income job creator, on the other it could be seen as an opportunity to implement the development process of the local area of production.12 We remind the reader that a common characteristic of the case studies reported is that at least a part of the production is exported through a value chain. This means that these homeworkers through subcontractors and intermediaries are part of the competitive producers’ network at international level. Another element – which emerges clearly in South and South East Asia and Latin America – is that subcontracting is a common link between formal and informal activities. Furthermore, there could be a dynamic contribution of informal activities to local economic development through the spin-off effects on local activities via consumption. Smallholders (the poor) are more likely to spend on local goods and services. In other words, the informal sector cannot be considered, as before by some economists, as the sand in which the multiplier effects are lost. This important conclusion emerged clearly also from recent studies even in rural areas (Delgado et al., 1998; FAO, 1998). From a policy perspective, the most relevant point, as already mentioned in Chapter 1, is that in activities such as micro-enterprises and SMEs (regardless of whether they are formal or not), there is a sort of natural clustering. Indeed, the gains from small-scale production enterprises13 for the local economic system can be amplified with agglomerations. A cluster – a spatial and sectoral concentration of enterprises – can have different starting points (Mead and Liedholm, 1998; Di Tommaso and Rabellotti, 1999, p. 11). For instance, it can be generated by an agglomeration of traditional artisan activities in a specific sector and location, or by the presence of a larger enterprise which subcontracts part of the production stages to smaller enterprises. Clustering is initially a natural strategy (or a spontaneous action) to reduce transaction costs and to capture positive external economies14 and synergies. Three economic concepts underlying this process of industrialisation are externalities, joint actions and economies of scale in the local system of production (Volpi, 2002). As the cluster develops, micro, small and medium enterprises develop within. This development has a number of positive external economies, which pushes the process forward. The positive external economies are generated in the local system not only through super-specialisation of enterprises (vertical, horizontal) and workers, but also by joint and collective action. Specialisation in services linked to the production process, capital goods production linked to products produced in the cluster, equipment sharing, and so on are the usual path of development of a cluster. In the evolution of a cluster of SMEs, the specialisation and the division of labour within firms results, over time, in the development of a pool of workers from semi-skilled workers to super-skilled. New enterprises emerge in the form of new suppliers, producers and traders. The interaction of these economic agents reduces transaction costs thus inducing further market entry in the cluster and at the same time increase trust among local stakeholders and help social cohesion. In particular, three factors need to be present in a locality to have a strong base

Towards a policy agenda

367

Urban

City

Small firm clusters

Rural

Territorial axis (local environment)

for a successful cluster: trust, entrepreneurship and artisanal competencies. Some of these factors are present only in some clusters of developing countries and others not, and unfortunately as we will see in Sections 11.5 and 11.6 they are not easy to build, especially the first, which is based on social cohesion and positive social capital. One of the basic concepts is that a cluster is characterised by a territory and by the actors and stakeholders within it. As underlined by Bellandi (2004), in a study on the organisation of industry in a territory, the processes of industrialisation in such a territory can be realised taking two different units of analysis. One is the industry, with its organisation and territorial character. The other is the locality, with its socio-economic character and evolutionary processes. In a territory you can find different industries with their territorial features, and different localities with their industrial features (see Bellandi and Sforzi, 2003). Possible combinations are represented in Figure 11.1. The interaction of these aspects and between actors implies that the cluster can assume different shapes of industrial organisational structure-from very simple based on artisan cluster in rural areas to high-tech cluster in urban areas and industrial districts (Bellandi, 2004). As indicated by Figure 11.1 there are different types of organisational structures within the term clusters. Clusters can evolve into more complex structures such as industrial district. In other words, the development of industrial districts is one possible evolution of a cluster (Becattini, 1987, 1990, 2000; Brusco, 1990; Bellandi and Russo, 1994; Bellandi, 2004). In the literature there are different classifications of clusters in developing and developed countries. They are often based on dichotomies15 different from the high/low roads described in Chapter 1 and that we analyse further in Section 11.3.

Industrial district

Big firm clusters

Rural local system Large enterprise Small enterprise Industrial organisation axis

Figure 11.1 Types of clusters. Sources: Bellandi and Sforzi (2003), Bellandi et al. (2006).

368

11.3

Santosh Mehrotra and Mario Biggeri

Clustering as a process of development: dirt roads, low roads and high roads

As we anticipated in Chapter 1, a cluster and its related local system of development can have two roads of development: a low road or a high road (Pyke et al., 1990; Pyke, 1992). We also advanced the debate on this dichotomy by arguing that high road/low road are not sufficient to depict fully the dynamic process of clustering and, thus, we introduced a dirt road to include the ignored dimension of social results. To examine better the different types of clusters around the world and to understand the possible routes or strategies of development that can be pursued to upgrade a dirt road cluster, in this section we are going to further develop this theoretical framework. Figure 11.2 describes the different typologies of clusters that can be found in developed and developing countries and strategic routes that they may follow in

Social outcomes BSS access, social integration and participation, social and environmental protection

No

Collective efficiency* High

Extremely rare Based on exploitation at all levels

High

(High Road 3)

(Low Road 3)

Low Cooperation and bad competition

High Road 1 Higher synergy and social cohesion

(High Road 2)

Dirt Road 3

Low Road 1

Dirt Road 1

Dirt Road 2

Low Road 2

Unsustainable in the market economy in the long run

Strategic route 1: the two synergies Alternative strategic route 2: SR dominant Alternative strategic route 3: CE dominant

For example, deviations

Figure 11.2 Clusters and human development strategic routes from dirt roads to high roads. Notes SR  social result; CE  collective efficiency; BSS  basic social services; Collective efficiency implies cooperation between cluster firms and absence of cut-throat competition (see also note 16).

Towards a policy agenda 369 their development. The framework is based on two dimensions: collective efficiency (the interaction of cooperation and competition (Schmitz, 1995)) and of social development. The nine cells in the matrix illustrate the combination of these two dimensions. First we examine the dimension of collective efficiency which is produced by both competition and cooperation between firms.16 An ‘healthy’ (positive, that is, non-disruptive) competitive environment gives the firms the right incentive to innovate. An ‘healthy’ cooperative environment or of collaboration (different from collusion) enables firms to pool resources when market conditions dictate it or when it is cost efficient to do so and to provide public services through collective action and public and private interaction. The other dimension introduced into the analysis is social development. In other words, if the goal of development is to move ‘from per capita GDP to local well-being’, those characteristics that influence positively human development, such as health, education (and Basic Social Services (BSS) in general), social insurance, equity, environmental protection and social environment or integration (a mixture of equity/cohesion) must be part of the analysis (ILO, 1999; Rabellotti, 1997, pp. 45–46; 1999). It is evident that even a cluster in the same road category can have a different level for each dimension. For instance, the Dirt Road 1, the worst form of cluster, has a very low level of cooperation (competition is supposedly high) along with very poor outcomes in social development. This Road is typical of the Pakistani clusters examined in Chapter 7.17 There is no horizontal cooperation since single micro-enterprises – such as homeworkers – are often clustered not only under the pressure of competition but also isolated. The development of clusters of this kind is often connected to the growing phenomena of subcontracting in a national and/or international value chain. Then, although most of the Indian clusters surveyed are in same situation (Chapter 6) others are closer to Dirt Road 2, as in the Bidi sector in Tamil Nadu since there is significant cooperation for social development (welfare funds) and hence higher social results than in Pakistani or the other Indian clusters analysed. Most of the East Asian clusters – examined in Indonesia, Philippines and Thailand (Chapters 8, 9 and 10) – are in a better position than the clusters selected in India and Pakistan in terms of collective efficiency and/or social results (this is also due to government interventions in social sectors and in promoting microenterprises and SMEs). Some clusters can be classified into Dirt Road 3 where collective efficiency is better than social results (e.g. Pyrotechnics, Home Décor, Okra, Fashion accessories in The Philippines and Hybrid seeds production in Thailand), and others into Dirt Road 2 where social results are better than collective efficiency (as Batik and Pottery in Indonesia and Paper products in Thailand). A few of the clusters studied (as Leather crafts in Thailand and Rattan in Indonesia) are close to Low Road 1. The dirt roads cluster is characterised often by micro-small enterprises in the informal sector with no or insufficient social protection. They can be a product of international competition, an expression of subcontracting (in the international

370

Santosh Mehrotra and Mario Biggeri

value chain) often involving foreign direct investment that escapes social/ environmental costs in their home countries,18 thus inducing the reduction of wage and labour rights and/or capturing negative externalities such as: poor environmental conditions, no social protection, hazardous occupations or processes and poor working conditions (e.g. poor lighting and ventilation, non-availability of safety devices, exposure to toxic substances and so on). The other squares of Figure 11.2 depict other types of clusters. The low roads, according to the different intensity of the two dimensions under analysis, can present medium collective efficiency combined with low social development (Low Road 3), as in most Chinese clusters, or the opposite (high social development plus medium collective efficiency) which is typical of declining clusters in developed countries (Low Road 2). Even the high roads clusters can be classified according to the level of both dimensions. The High Road 1 – on top right – is the best-case scenario. Clusters positioned in the High Road 1 are mainly from developed countries and rarely some developing countries. ‘This, therefore, is a self-reproducing model of development. And it concerns local development because the strategic factors consist of progressive relations between enterprises and local society’ (Bellandi and Sforzi, 2003, p. 213). Social progress, however, is usually the result of a conscious coordination among the local institutions: that is, the ‘high road’ to competitiveness is not the outcome of market mechanism, but of a combination of market and concerted collective action among representatives of the principal cluster (in the district) and the local establishment (Dei Ottati, 2002). It requires genuinely shared objectives, as well as a satisfactory distribution of income among the parties involved (fundamental for the renewal of the system) and a satisfactory standard of living, which is fundamental to promote cooperative labour and supplier relations. There are two cases that can be considered as extremes. The top left corner is the case of a cluster which for international or domestic competition and institutional mechanisms presents high collective efficiency but accompanied by a very low or no social development. This was an extremely rare case but is becoming more frequent under globalisation. For instance, cluster development in China (discussed later) is a typical example, where there are now some specialised towns (clusters) of this type and where the presence of temporary migrant workers exceeds (also ten times) the local population (Hirsch, 2005). International competition and local institutions which allow low income and minimal worker rights, induce a vicious circle and negative outcomes of the delocalisation of production (the race to the bottom with high profits for investors) (Biggeri, 2005). In this situation workers serve as a ‘reserve army of labour’, in a Marxian sense. The bottom right corner clusters cannot survive. Only strong government interventions can maintain them and this has to be done on political grounds (e.g. they are considered essential for the economy or the cost of a rapid dismissal is too high for the national or local society). This framework facilitates the understanding of the weakness of a cluster itself, and helps to explain the potential strategic routes to upgrade the cluster and the local systems. The transition from one stage of development to a better one is not an easy task, and retrogression can occur.

Towards a policy agenda 371 Therefore, we argue that the potential for evolving from a dirt road to a low road exists, but not without joint actions or collective actions carried out by the workers in their collective interest and the support of public intervention. According to this analysis there are three major and consistent strategic routes to upgrade the cluster positioned at Dirt Road 1 (see Figure 11.2). For instance, even simple but significant collective actions – for example, cooperation among producers (including homeworkers) – can upgrade the local system from Dirt Road 1 towards Dirt Road 3. This is the beginning of strategic route 3 that promotes collective efficiency at the ‘expense’ of social protection and social development, while collective actions or even joint actions (or policies by local or central Government) can improve the social result, moving from Dirt Road 1 towards Dirt Road 2. This is the strategic route 2 that enhances social development and poverty reduction but not fostering to the same extent collective efficiency (in our case cooperation since competition is assumed high). Strategic route 1 has the capacity to balance the two dimensions and can generate the second synergy in the local system of development and the possibility of pursuing High Road 1 of development depicted in the top right corner. Following this strategy route 1 through a gradual increase in collective efficiency and social development will mean the cluster has stronger social cohesion and positive social capital, and stakeholders share a common goal, that is, local well-being (Dei Ottati, 2002). The key to move and to remain on High Road 1 is the continuous upgrading of human capital, and the triggering of institutional arrangements and networks at local level, as well as the capability to undertake collective action for social and economic development of the area, and to respond to external shocks and recession periods. Sustainable SME development in local areas depends on their capacity to utilise the benefits of clustering, and in particular on spillover effects arising from the creation of positive external economies. However, this capacity is generated from factors which are often external to the small enterprise such as the human and physical infrastructure, which in turn are necessary conditions (but not sufficient) for generating positive external economies from SMEs sector, and longterm growth perspectives. Therefore, local policy-makers’ intervention is fundamental to upgrading the human capital and infrastructure which are external to a single enterprise. Without public and collective action the development through SMEs is limited (Becattini, 1990). Local collaboration therefore is always a result of ‘political’ action in the broad sense. In Italy, for instance, the three central institutions – workers unions, the artisans association and the employers’ association – are accompanied by ‘local establishment’ or intermediate institutions such as private or public agencies (e.g. service centres, chambers of commerce, banks, training agencies and local government). Among these the local government can assume a central role. The first and probably more obvious role is in defining the nature of economic planning (e.g. infrastructure, such as roads), regulatory powers (e.g. land use, licenses) and public services (e.g. technical education). The second is to indicate the objectives (shared development project) of the local community and thus to

372

Santosh Mehrotra and Mario Biggeri

mediate (e.g. in Italy social pacts, that are rules of behaviour that discourages the spread of destructive forms of competition) among the interests of the various actors in the cluster (Dei Ottati, 2002, p. 457). This logic, however, begs a fundamental question of agency. If local development requires effective local government, somehow the latter has to come about: then how can we ensure strong local government? Local governments in low- and middle-income countries are among the weakest of all levels of government. Most local governments in Asia, Africa and Latin America have weak human resource and financial capacity because these governments have in general been historically quite centralised. Centralised government was the inheritance post-colonial regimes got from their colonial masters in Asia and Africa; in Latin America, most post-Second World War governments tended in fact to be military juntas. Democracy is a 1990s phenomenon; similarly, decentralisation to local governments and municipalities has also only spread since the early 1990s. However, even though some functions have been devolved to local governments in the 1990s, they have limited independent sources of finance and have limited human resource capacity compared to higher levels of government (see for example UNDP, 2003; World Bank, 2004; Mehrotra, 2006). In other words, while our theoretical discussion in this section has assumed a strong and capable local government, central governments in developing countries remain highly suspicious of local governments. In fact, the empirical evidence is that rising levels of per capita income is correlated with high degrees of decentralisation, as well as improved indicators of governance, however defined. Given this chicken-and-egg situation, higher levels of government may have to substitute for local government action, given the current weakness of local government capacity in the low- and middle-income countries under discussion here. In Section 11.4 we present evidence of successful clusters. They show that while the role of local governments was important (especially in the successful case of China and Italy), much can be achieved even in the absence of such public action, as well as through other forms of collective action.

11.4

Empirical evidence on the potential of clusters

The general observations from studies of incipient clusters in India and Pakistan suggest that beyond a division of labour within the cluster, and some backward and forward vertical linkages, horizontal collaboration between enterprises (including homeworkers) at the cluster level is rare. Furthermore, in some sectors/cluster analysed low barriers to entry, limited skill bases, extensive local competition, low trust within clusters despite the often strong presence of common social identities, can result in what one can refer to as ‘poor contract enforcement’ and over exploitation of homeworkers.19 This limits the potential gains that clustering could bring about, in terms of growth and pro-poor impacts.20 In the five countries studied – despite some policies for SMEs and homeworkers – it is rare to find policies to promote clustering.

Towards a policy agenda 373 In this section we present some examples of collective actions in successful clusters in developing countries.21 Then, in order to understand the importance of local public interventions as one of the components of a successful clustering process, we present the example of Chinese rural SMEs or TVEs and finally, we examine historically some successful Italian cases of passage from Low Road 1 to High Road 1, famous for including social protection as an asset rather than as a cost for local development. The empirical evidence, in general, emphasises that clustering to be successful, needs participation of local communities, social environment, trust, joint actions, collective actions and public interventions (Rabellotti, 1997; Altenburg and MeyerStamer, 1999; Schmitz and Nadvi, 1999; Nadvi and Barrientos, 2004). Stronger evidence on cluster-wide institutional joint action was observed by Nadvi (1999), in the context of the Sialkot surgical instrument cluster in Pakistan, and by Kennedy (1999) for the tanning cluster of the Palar Valley in Tamil Nadu, India. Compliance with global quality assurance standards, a necessary requirement for exports to leading global markets by the Sialkot cluster, came about through the catalytic role of the local trade association in channelling new know-how on quality management practices to the cluster. Through this process, Nadvi found that the vast majority of SMEs in the cluster could comply with international standards over a relatively short period of time. Had the association not taken on this function, most small firms would have closed given that the United States and the European Union (EU) accounted for over 90 per cent of the cluster’s sales.22 A further example of how clusters can promote collective responses to external threats comes from the response of Sialkot’s export-oriented sports goods cluster regarding the presence of child labour in manufacturing units within the cluster (Nadvi, 2003). Faced with the loss of key export markets, local firms through collective institutions (such as the local Chamber of Commerce) entered into an agreement with international bodies, including the International Labour Organisation (ILO) and United Nations Children’s Fund (UNICEF), as well as leading global buyers. This resulted in an ILO monitoring programme of the cluster and a social development strategy, based on education and income generation, for child workers and their households. Thus, local joint action resulted in direct gains for the cluster, employment and cluster exports rose, while immediate poverty concerns for many of the more vulnerable members of the cluster’s labour force began to be addressed (Nadvi, 2003). Joint action is neither an obvious outcome of clusters, nor is it easy to achieve. The evidence that emerges from cluster studies suggests that joint action is less common in incipient clusters than it is in more mature clusters. Even in mature clusters, joint action is far from uniform. On one hand the ‘opening’ to global markets can enhance a cluster’s market performance23 and to upgrade but on the other hand with the opening a cluster has to face global pressures, and often ties with external actors begin to supersede local linkages (Brazil’s Sinos Valley is an example, see Schmitz, 1999).24 But in many cases, local cooperation and collective action can assist local small enterprises obtain access to markets, overcome constraints and confront vulnerabilities that they face in local and global markets.

374

Santosh Mehrotra and Mario Biggeri

Often where cooperation does occur, it is strengthened by local social capital, common ties of community and identity that can foster cooperation and generate trust. Many other clusters in the developing countries benefited from public interventions and in particular by the provision of public goods and services (among others see Rabellotti, 1997) – this is particularly the case in China since the economic reforms. China’s Township and Village Enterprises (TVEs) One of the most fascinating examples of cluster development is that of the TVEs in China between 1980 and 1993. The predecessor of the TVEs was the Commune and Brigade level enterprises of the pre-reform period. Those enterprises were renamed as TVEs in the 1980s and became the engine of growth and the driving force for market-oriented reform. New entry firms were the driving force behind China’s economic growth. But what is important is that most new entry Chinese firms were neither private firms nor state firms (i.e. national government firms), but local government firms.25 The extraordinary performance and development of the TVE sector are attributed to many economic, political and socio-economic factors and their roots can be traced back mainly to the rural institutional socio-economic reforms implemented by the Chinese Government after 1978. The new strategy consisted of changing some of those elements of the institutional framework of the ‘Maoist system’26 that reduced personal incentives and hampered the potential growth of the economy. The main characteristic of the reforms was their pragmatic attitude and experimental implementation. In particular, they did not follow a specific theoretical economic stream or a specific country model, but China went forward in transition to a ‘market socialist economy’ with its own pragmatic and gradual style. The agricultural reforms, the cornerstone of the rural economic development strategy, played an important role in removing economic stagnation in rural areas, and in particular in freeing-up human and capital resources for the development of the TVE sector. The increase of income of farmers and rural dwellers was fundamental to create the initial demand for the SMEs’ manufactured products (Biggeri et al., 1999). What this aspect of the Chinese case implies is that action to promote employment in small or micro-enterprises, or promote their clusters, is likely to lead to a pro-poor pattern of growth only if accompanied or preceded by other measures that generate demand for the products of clusters. Actions by local governments had an important role in the initial phase of the development of rural enterprises. As the Maoist system was modified to allow for local initiative and proper incentives were introduced, both the central and the local governments were left with an impressive array of policy instruments and political capacity. We should not undervalue this the experience of economic planning at People’s Commune level. In particular, local governments and village committees and the general organisational network (within the CCP) were able to

Towards a policy agenda 375 capture the new incentives and played an important role in the initial phase of development of rural enterprises, mobilising community capacities and local savings. In this sense the party network and the social cohesion (although often corrupt) was an important ingredient of success. The gradual decentralisation with economic reforms of the fiscal system, which allowed local governments’ revenues to be used in part to facilitate local socio-economic growth, was the major incentive for action. Once in control of their own revenues, local authorities had a greater incentive to develop their administrative areas, to increase their tax income, and to invest in new infrastructure and in new TVEs, diversifying the community business risk (Chang and Wang, 1994; Weitzman and Xu, 1994). Among local governments the erosion of the profits of State enterprises, the major source of local revenues, resulted in immediate fiscal distress which destabilised the distribution among regions and administrative levels, but also reduced bonuses and benefits, which in turn, hurt the reputations of local officials and thus supplied a strong incentive for local governments to push enterprise reform beyond even the limits set by the central governments and to initiate pragmatic experimental reform programmes (Sun and Ren, 1997). On the other side, the community members received direct and indirect benefits from collective TVEs such as job opportunities, job security, pension funds and communal welfare programmes in housing, health care, irrigation, road construction and other infrastructure (such as school and kindergarten). Collective TVEs (usually owned by local governments) could capitalise on the new market opportunities. They had readier access to credit (via bank or rural credit cooperative loans and investments of other collective TVEs), productive inputs and information, and possessed an advantage when applying for legal permits and arranging market linkages. In fact, as Yingyi (2003) argues: When the local government runs a business, ownership and control rights interact with government activities and generate two effects that are absent under private ownership. First, the local government has higher incentives in providing local public goods [like maintaining order, building roads, providing water systems, and implementing family planning] because its ownership rights give it access to the future revenue in a credible way. Second, anticipating this, the national government leave a bigger budget to the local government which optimally prey less on TVEs than private enterprises. This in turn makes the local government less worried about revenue confiscation and reduces TVE revenue hiding. Both effects improve efficiency. (p. 312) According to some research, private enterprises were often repudiated by party cadres. Therefore, it is argued that in some rural areas, especially in the 1980s, when private TVEs were formally allowed to exist, local governments interfered leading to market distortions becoming one of the main constraints to the

376

Santosh Mehrotra and Mario Biggeri

development of private TVEs. In other rural areas in the late 1980s and early 1990s, collective TVEs were often self-constrained in their development by the presence of a direct management or a very strong operational control by local governments and village committees. Chinese local governments were so pragmatic that without knowing anything about the theoretical importance of clustering, they followed ‘yi zhen yi pin’ policy, with one town or area specialising in one product and they were called Specialized Towns, that is, clustering, was occurring based on both an industrial as well as territorial specialisation. The year 1984 may be considered a landmark year for reforms in the rural enterprise sector: it signalled a new phase of development for such enterprises since individual private enterprises (with less than eight employees) were allowed to exist. A year later, in order to overcome the problems of obsolete technologies, and the lack of management training in the Chinese TVE sector, the State Science and Technology Commission began implementing the ‘Spark Plan’ through technological demonstration centres (Wang Zhonghui, 1993, pp. 23–24). These were required to increase investment in human capital, to master technology, and to teach entrepreneurs to manage effectively and efficiently. Once the institutional reforms took effect in 1980s, enterprises were given the opportunity to retain profits, thus providing the first stage of an incentive structure. The Central Government encouraged a high rate of re-investment of enterprise earnings through preferential tax treatment (especially for poor areas). The national government had required that the TVE after tax profits should be essentially used for two purposes: reinvestment and provision of local public goods. In 1992, 59 per cent of the after-tax profits of TVEs were reinvested, and 40 per cent were used for local public expenditures, a slightly higher proportion in each case than in 1985. This situation is so different from that which is normal in most other developing countries. In most developing countries, very little revenue is collected by local governments; even after revenue is collected, the government is often unable to spend it on local public goods in rural areas because of the political influence from urban elites. The linkages between TVEs and local government in China ensured the opposite. These incentives were later enhanced when TVEs assumed responsibility for their own profits and losses, and when profit retention for higher administrative level units was in part replaced by taxation. From 1988 private TVEs (with more than eight employees) were allowed to register and pay tax. Thereafter, as TVEs have increasingly had to rely on markets for their inputs and sale of outputs, competition between enterprises has intensified. Since the early 1980s most TVEs’ workers’ contracts embodied a higher degree of flexibility and employees put in a great deal of effort in order to keep their jobs and maintain their incomes. Chinese reforms have gradually created a domestic competitive environment. Furthermore, these market-oriented enterprises also benefited from the presence of state-owned enterprises heavy industries (e.g. for inputs) and at the same time enjoyed a smaller burden connected to welfare costs and a lower labour and land cost than urban enterprises.

Towards a policy agenda 377 Since collective TVEs remained poorly managed, in late 1980s another institutional reform was introduced. The contract system, following the agriculture Household Responsibility System (HRS) reform, was extended to collective TVEs in order to improve the management efficiency and increase the profitability of the enterprise (i.e. from direct management to out-contracting). This consisted of the opportunity to transfer – on the basis of a short-medium-term contract – the full or partial management and control of the collective TVEs to an individual or a group of private entrepreneurs. TVEs’ development was facilitated indirectly by the CCP policy of ‘leaving the land but not the village’ – which favoured investment on social infrastructure in small sized towns, reducing internal migration to big cities – and directly by the decentralisation and expansion of the Agricultural Bank of China and rural credit cooperatives since credit constraints are one of the main problems facing SMEs. In order to overcome the lack of incentives for managers of collective TVEs and to reduce the interference of local governments, further institutional reforms and specific policies were implemented at the beginning of the 1990s. The development path of TVEs has been influenced by two interdependent socio-economic aspects of Chinese economic development during the 1980s and the 1990s. First, the Chinese economy is characterised by strong regional differences (such as in natural resources, human capital and strategic economic location). These arise partly out of initial factor endowments and their resulting comparative advantages, and partly as a result of the implementation of different policies. The Dengist strategy of provincial development viewed these differences positively, and comparative advantages were developed at provincial and at county level, along with agricultural specialisation and diversification. Second, an international ‘open-door’ policy was pursued. This policy started in 1979 with the opening of the first four ‘Special Economic Zones’ in coastal areas, aimed at attracting foreign capital and technology in the coastal macro-region. The low labour costs of rural enterprises (comparative advantage), and their market orientation, made TVEs internationally competitive exporters of labourintensive products (Findlay et al., 1994, p. 175). The foreign direct investment (FDI) was an important factor in the growth of TVEs for the coastal region (see for instance Fusaku et al., 1994; Yang, 1997) involving various types of external economies and spillovers which are important once combined with specific factor endowments of the host province. In addition, joint ventures in the transfer of technology could be important since they can help rural firms – which have the advantages of cheaper land and labour, and fewer labour constraints – to overcome the constraints on lack of capital, modern technology, and management skills. As already mentioned, the protection of the Chinese economy implemented throughout this period gave to Chinese infant industries the opportunity to satisfy internal demand and protect it from the disruptive international competition. In East Europe and Russia the gap between internal demand for goods and supply was filled by imports. Nowadays, as it is pointed out by some experts, many TVEs will probably close as WTO rules will be introduced in China but many will be able to compete internationally. Empirical analysis has produced some evidence

378

Santosh Mehrotra and Mario Biggeri

for collective TVEs. It has been estimated that, between 1986 and 1993, collective TVEs’ provincial average labour productivity in value has been affected positively by human capital availability, infrastructure presence, credit policies and by clustering (Biggeri et al., 1999). We conclude this part on the Chinese experience with clusters by pointing out that in China, the human development perspective is rather scarce after 1990 with the dominance of the private enterprise over collective ones. The overexploitation of temporary migrant workers and their low human development reveals that this phenomenon is connected to the pressure of international markets on the one hand and to local institutional arrangements on migration on the other (Biggeri, 2005). In the same place you have high collective efficiency strong technological progress, and productivity increase mainly due to FDI, but at the same time very low social development (no trade unions are allowed and no social protection and almost no rights are given to temporary migrant workers (Hirsch, 2005).27 Italian clusters and homeworkers’ human development The most cited example of excellence in the cluster development process is the Italian case. As we noted in Chapter 1, the areas together called ‘Third Italy’ are characterised by a large number of clusters most of which are recognised as industrial districts – mainly in north east and central Italy. The Italian experience supports the view that clusters can generate improved incomes and employment and lead to High Road 1 (Pyke and Sengenberger, 1992) through strategic route 1 (see Figure 11.2) characterised by synergies in action. The point of departure was based initially on private collective actions. In some parts of Italy small manufacturing enterprises (with homeworkers working for them) led to the emergence of industrial clusters, and in some cases, due to the right institutional and non–institutional conditions and driven by an evolving set of private and public action, these matured into industrial districts of SMEs which contributed to Italian development (Becattini, 1990, 2000; Brusco, 1990; Bianchi et al., 2000). Its main characteristics are: specialisation of the district in a principal industry; inter-firm division of labour in different phases of production; inter-firm cooperation; local availability of skilled labour; local entrepreneurial dynamism; social, cultural and territorial embeddedness of the economic activities; and institutional ‘thickness’. Industrial districts are an economic, social and cultural phenomenon. In other words they are a combination of economic, social and institutional inter-dependencies fundamental to achieve economies of scale and scope, joint actions (vertical, horizontal and multilateral), collective efficiency, internalise external economies (specialisation, learning and creativity economies) and to share systemic conditions that have the nature of public goods (Becattini et al., 2003; Bellandi, 2004). We thus argue that to reach and maintain High Road 1 social outcomes in the cluster are as important as collective efficiency and that the synergies among these two dimensions foster development of the local system. Indeed, an in-depth

Towards a policy agenda 379 study of several Italian industrial districts found that the competitiveness and the dynamism of industrial districts’ firms are dependent upon social capital and cooperation (Dei Ottati, 2002). In some clusters, in rural and urban Italy, homeworkers’ activities played an active role in local development, and in many parts of the country these clusters became high road clusters. One competitive advantage of the productive system is the extreme form of flexibility since homeworkers guarantee labour in periods of production boom and get no orders when demand recedes. However, some of the problems encountered by Italian homeworkers were and are the same as those of homeworkers in developing countries.28 For instance, consider the case of Carpi, a famous knitwear industrial district in Italy. The roots of its development lie in the skills acquired by homeworkers in the sector of straw weaving. The small entrepreneurs used the middleman to enlarge production by subcontracting to the homeworkers. The joint actions, collective actions and public interventions and the evolution of institutional arrangements, and of the network created, were the keys to the local development of Carpi. The local government helped also on social development. For instance, at the end of the nineteenth century the municipality of Carpi reduced child labour by making education closer to the needs of the local system. They ensured that in kindergarten and elementary schools, in addition to regular education, the teachers were required to produce goods woven with straw. In that period, part time child labour was the norm. Then, after the end of the Second World War, the demand for straw weaving goods suddenly declined. The entrepreneurs were able to switch to knitwear, thanks to the skills of homeworkers. Gradually, the homeworkers become employees (or entrepreneurs) in SMEs, benefiting from social protection (Cigognetti and Pezzini, 1994). To understand the interaction among actors and stakeholders in local development equally interesting is the experience of two well-known industrial districts in Italy, Modena and Prato. First, it emerges that although these local systems of production are composed of SMEs the level of unionisation is even higher than in large industries – the exact opposite of the current situation in countries examined in this book. During the initial phase trade unions had a role mainly in the determination of the wage level and then the flexibility of working-time and production bonus. In Modena since the 1970s and 1980s the local governments (municipality, province and chambers of commerce) implemented not only favourable conditions to develop new enterprises in the local industrial area, but also the creation of a social consensus with coordinated interventions including local social services such as catering, transport, kindergarten, houses (public owned) for workers (to be added to national, almost free, social services such as health, sanitation and education) (Perulli, 1989). Trigilia (1989) found that in Prato social capital and cooperation had a fundamental role in reducing potential conflicts and reducing transaction costs. This also allowed for work flexibility (a fundamental competitive advantage of industrial districts) but at the same time social protection based on a mechanism

380

Santosh Mehrotra and Mario Biggeri

of periodic collective bargaining. This flexibility favoured local economic development and also, at household level, allows women and older persons to be productive homeworkers and have a ‘decent’ income. Further, homeworkers and subcontracters are incubators of entrepreneurship (Dei Ottati, 2002). Prato is a textile district and in 1985 the total employees in textile manufacturing were around 50,000 although falling consistently; and this labour force included at least 3,000 homeworkers, that is, 6 per cent of the labour force. Trigilia argues that in Prato we are in the presence of ‘manufacturing socialisation’ within the district in which there is total participation of the local community in local development, ensuring compensation flexibility (1989). The argument is so relevant to our purposes that it is better to elaborate on the evolution of the relationships among stakeholders. In 1945 the first agreement on wage increase was signed between the Industrial Employers Association of Prato (Associazione Industriali Pratesi) and the workers unions. In 1948 the same parties and the chambers of work signed the first agreement on tariffs of homeworkers. In the 1950s there were ten years of blackout in the relationship among stakeholders, driven by external economic and political forces. In 1961, however, there was an important agreement on homeworkers tariffs among the Industrial Employers Association and unions of workers. From 1961 the contracts and the tariffs have been revised periodically each two years. Going back to the content of the agreements we can observe that in the initial phase (1945–61) the agreements were mainly on wages and union rights. Then in the 1960s the agreement moved to involve funds for social interventions, in 1970s they also included safety in the place of work, and information sharing with workers on investment decisions, while in the 1980s on working time (and time flexibility of work) and on firing workers. The important experience of funds for interventions in the social area were the result of an agreement (territorial pact) between the Industrial Employers Association of Prato, the workers unions and the local government of Prato, and is managed by a committee composed of employers and workers associations (COGEFIS). It is financed by the enterprises (each has to contribute a total equal to 1 per cent of the wage bill). The fund is used to invest in kindergarten, transport and medicines, school grants and technical training courses. At the same time, the local government’s expenditure on social services (mainly schools) constitutes more than 50 per cent of total expenditure of the municipality of Prato and maintaining infrastructure and services to the enterprises. The interest in regularising the homeworkers and giving them social protection is not only a human goal, it is in the interest of the cluster and its social stability and provides a fertile environment for economic development. We have reviewed this experience with clusters in Italy in drawing lessons of policy for developing countries. However, one should note that Italy, and particularly northern Italy, is peculiar among industrialised countries on account of the continued dominance of small enterprises and traditional sectors in its manufacturing sector. What is most significant for other developing countries is the fact that the lack of internal economies of scale due to small average enterprise size is

Towards a policy agenda 381 compensated by external economies of scale created by the special industrial environment that is prevalent in these clusters. At the same time, most researchers agree that in Italy, the spontaneous clustering of enterprises preceded the creation of institutions to foster the exchange of information and technology. Some have rightly questioned the possibilities of creating a local productive system of small firms based on collective efficiency from scratch. Nevertheless, the empirical evidence we have cited from other countries, as well as from China, in this section, does confirm that there is plenty of action that governments, both national and local, can take to promote small and micro-enterprises.

11.5

Policies to promote clusters of small enterprises and homeworkers’ activities

The fact of being clustered does not automatically bring positive external economies in the productive system. Hence a policy agenda has to be outlined. The policy agenda has to be flexible since policy-makers need to tailor it according to the different circumstances of the country and the local social and economic system characteristics. This implies that policies valid for a certain cluster or valid ten years ago, in a different international environment, are not necessarily valid now for the hypothetical cluster under examination. International agencies such as UNIDO, UNDP and ILO have a wide experience in promoting clusters in developing countries and especially supporting private enterprises mainly by promoting services and marketing. For instance, the experience of UNDP programmes in Indonesia with the government of Indonesia for cottage industries is quite interesting and well documented. In these projects they fostered the so-called BIPIK programme. BIPIK stands for Proyek Bimbingan dan Pengembangan Industri Kecil khusus Golongan Ekonomi, or Project for Guidance and Development of Small Industry, especially that are operated by weak economic groups. The first aim was targeted group selection, then cluster organisation, reducing marketing problems, making investment and finance on the basis of results and cost-effectiveness of the project (Weijland, 1999). (For other programmes of international agencies’ interventions see Annex 11.1 at the end of the chapter.) ‘In the case of survival clusters, the deficiencies of the socioeconomic milieu are nevertheless intertwined with pronounced problems of inefficiency at the firm level. An adequate mix of general SME support and specific cluster policies is therefore necessary’ (Altenburg and Meyer-Stamer, 1999, p. 1698). Our analysis suggests that for upgrading the local economic system policies to promote a cluster’s collective efficiency have to be accompanied by policies for social development, since these two dimensions foster each other. Therefore, wherever possible, strategic route 1 – that is, balance between the two dimensions (see Section 11.3) – has to be pursued. In this section, we are concentrating a policy agenda for the productive side (the social dimension is examined in Chapter 12). The evolution from a casual cluster of micro-activities to an articulated cluster requires some conditions in place such as trust, entrepreneurship and competencies, as well as private joint action and public interventions and, obviously, time

382

Santosh Mehrotra and Mario Biggeri

to evolve. In particular, in low- and middle-income economies the development of a local system in the form of a cluster can be a slow and difficult process because of constraints due to low demand, poor infrastructure, inadequate physical and human resources and finances, and of course because entrepreneurship requires capacity building. For instance, problems found in clusters such as Dirt Road 1 could be opportunistic behaviour (as some Pakistan and Indian middleman), low trust, and low social cohesion; complicated government procedures and bureaucratic red-tape are further problems. Furthermore, since networking is scarcely developed in most of the cluster studies, firms fail to harness much of the potential of clustering. In this case it may be useful to link support for individual firms to cooperative behaviour (Altenburg and Meyer-Stamer, 1999). In other words, as we said, the possibility to evolve from the dirt roads to the low roads exists, but without joint action, collective actions of some of the stakeholders and the support of public intervention this evolution is almost impossible. Therefore, in the clusters examined in our case studies the first step is to start or to foster joint and collective actions; interventions of local government may follow these actions but it would be important if they could precede them since public institutions have an important role in promoting development opportunities and also in balancing the benefits and the costs for the stakeholders and the local area. Furthermore, local government can promote, in agreement with local entrepreneurs and other stakeholders, specialised public goods and services that support local enterprises and collective action (Bellandi and Sforzi, 2003) and, encouraging beneficial agreements among stakeholders, can facilitate networking. Indeed, it is important to remember that local government (and the State in several cases) may assume a central role since there are factors which are often external to the small enterprise such as human and physical infrastructure, which in turn are necessary conditions (but not sufficient) for generating positive external economies and long-term growth perspectives. Where the local government is too under-resourced or disorganised to take effective action, there is no alternative to the central (or in large countries the provincial) government taking the necessary action. Investments in basic infrastructure (like water, electricity, good roads, telephones, etc.) facilitates the development process of small enterprises and contributes to improving productivity and working conditions – but may have to be taken by a central (or provincial) government. Another important dimension that sectoral policies should deal with is the transfer of public resources (a central/provincial responsibility) and the removal of discriminatory practices (often, though not only, a local government responsibility). Policies and local institutions have always been essential (during the stages of takeoff, success and ongoing development and restructuring phase), and they should become ‘immersed’ in the economic-social relations of the local area (Bellandi and Sforzi, 2003). For instance, cluster policies may produce winners and losers (Nadvi and Barrientos, 2004), thus it is important that local government ensure, where feasible, that marginal groups are not weakened through this process in order to maintain or to foster community cohesion. This may require more explicit policy targeting of such firms, workers and other

Towards a policy agenda 383 stakeholders although a precondition to successful trageting is participatory planning. An important role – which is essential in developing countries where local government and State interventions are weak – is played by producers’ associations (entrepreneurs, artisan, workers) because of their possible role in fostering joint action for answering common problems. Common action can be fostered if networking facilities are implemented. Participation in group organisations for small enterprises is an important step to increase responsibility, both productive and social, and help to overcome constraints in the supply of larger orders and in technological upgrading and institutional capacity building (Romijin, 2001, p. 70). Given these considerations a policy agenda – to promote both the cluster and small enterprises within it – can be divided into three types of policies: macropolicies, institutional policies and direct policies to promote micro-enterprises and homeworkers’ activities. Macro-policies We have already anticipated at the beginning of this chapter that macro-policies can influence the development of the cluster and SMEs. In particular we have also highlighted importance of industrial policies and that they should be accompanied by trade policies, fiscal policies as well as labour and social protection policies (for social protection measures for homeworkers see Chapter 12). Industrial policies were a central reason for the success of industrialisation in the East Asian tigers – Japan, South Korea and Taiwan (Amsden, 1989; Rodrik, 2000; Chang, 2003). A key feature of their industrial policy was not only the governmental support for large enterprises, but also for their smaller enterprises. A relatively important vibrant small-scale business sector also existed which was the engine of employment and reduced overall income inequality (You and Yoon, 1995). Governments in almost all market economies intervene to a greater or smaller degree in the operation of their industries (including in the United States). What makes interventions by the three Asian tigers into an ‘industrial policy’ is that their interventions are generally coordinated and viewed as a coherent whole. Thus, what we are arguing for in this chapter is that cluster development with SMEs and micro-enterprises (including homeworkers) development should be part of such a broader industrial strategy. Trade policies and domestic demand We observed that in Thailand in particular the government seems to have played a strong role in promoting the goods produced through subcontracting. An important reason, the Thailand study tells us, is that exports generated by homeworkers were so significant that the government had no alternative but to take notice, and take promotive action. As reported in Chapters 6–10 most of the sectors studied involved the products being exported. For instance, according to one estimate (IIFT, 1999), the

384

Santosh Mehrotra and Mario Biggeri

contribution of the informal sector to India’s exports is roughly 40 per cent. Despite generating foreign exchange for the country, the claim of the wage workers in the informal economy in social protection is negligible. They are subsidising exports, but cannot claim any health, maternity, education, old age or retirement benefits.29 We saw in Chapter 1 that there is increasing evidence to suggest that export-oriented production on a subcontracted basis has been increasing over the last quarter century in Asian economies studied here. However, it is unfortunate that in today’s development discourse, ‘integration’ usually means integration into the world economy. Indeed, as Wade (2003) argues, in economies that are not internally integrated, with dense input–output linkages between sectors, wages are viewed merely as a cost, not as a source of demand. Domestic production is not well connected to domestic consumption, leaving exports as the main stimulus to economic growth. Cutting costs through downward pressure on wages becomes a necessity when integration is mainly external. For this reason we think that – with a balanced process of clustering (strategic route 1) – local demand can increase, for instance, if workers receive the benefits of growth in terms of social protection and higher wage, thus fostering local and national development. In other words, local protection needs greater internal integration and may require less openness and more readiness to protect domestic industry. However, protection from foreign competition has to be part of a broader industrial strategy to nurture the capabilities of domestic firms and raise the rate of domestic investment, in the context of a private enterprise, market-based economy. This last statement is particularly true for the larger South Asian economies under discussion here, than the South East Asian ones. In the early 1990s, analysis and policy prescriptions for SME development based on the cluster concept were extended to developing countries. While clusters of small- and medium-sized enterprises in developing countries did not necessarily have all the characteristics attributed to the Italian industrial districts, it was certainly possible to find agglomerations of small firms, inter-firm cooperation and division of labour and local institutions that nurtured their development. Nevertheless, more recent work on globalisation and the position of developing country producers within the global economy has cast doubt on the potential for sustained upgrading and income improvement in at least some developing country clusters. However, we should remember that the upgradation of dirt road clusters we analysed would be a gradual process that can take some decades (at least two or more). Thus the current globalisation may not help some emerging cluster, but may help others. One of the measures of success of industrial districts in Europe has been the success in export markets (Humphrey, 2003). The key question concerns the position of developing country firms and clusters within particular global divisions of labour. In some global value chains, firms, and even clusters of firms, may undertake only a limited range of functions. Like subcontracting firms, they may work to designs provided for them, using materials which are sourced by other firms. But in this case, the ‘other firms’ might be located thousands of miles away

Towards a policy agenda 385 (see Chapter 3 on value chain and subcontracting). The argument, therefore, is not about insertion per se into the global economy, but rather about the nature of this insertion. The point about industrial clusters is that while companies operate within a small geographical area, together they constitute a vertically integrated industry. It is the completeness of the local production system, comprising both production processes and the knowledge systems that support innovation within it. Even increasing investment by MNCs into industrial districts could be viewed as unproblematic. Crestanello (1996, pp. 76–77) argues that the arrival of MNCs in the Montebelluna sports shoes district and the Schio-Thene machinery manufacturing district only served to strengthen their competitiveness and has not undermined their internal cohesion. This outcome is particularly likely when firms come into the cluster as a means of strengthening their knowledge base rather than to secure cheap labour. However, globalisation has made the integrity of clusters much more complex. In this case, certain important activities, as we saw in our Thailand case study of leather products – the creation of designs from which the models are derived are done elsewhere. In this sense, the cluster is not ‘complete’. However, even without design there are advantages to clustering: specialised suppliers, labour and buyers, as well as the institutional fabric of training institutions, technical centres and so on (Humphrey, 2003). What linkages might exist between local firms and the global economy? The Italian industrial district literature emphasizes two main linkages: arm’slength market relationships and vertical integration (established through FDI in clusters). Arm’s-length market relationships occur when products are standardized, or easily customized to particular buyer requirements, or designed by the producer without co-ordination with specific buyers. The purchasers of such products are ‘design takers’: the design of the product is in the hands of the producer. In the case of finished products destined for consumers, the agents buying these products from clusters are most likely to be wholesalers, traders selling to a variety of customers and retailers (particularly small retailers or consortia of small retailers). By contrast, vertical integration involves direct co-ordination of activities within the firm. The most obvious form of this is through foreign direct investment into clusters. (p. 11) To upgrade a cluster an important option is to learn from higher level clusters engaged in the same product both in the national economy or in another country. Firms in developing countries may invest in developed country clusters, either in order to guarantee their position in these markets or in order to gain access to the knowledge base of other clusters. For example, some companies in the Sialkot surgical instruments cluster have established trading firms in the Tuttlingen cluster in order to facilitate access to German and global markets. Links can take many forms. Although it depends on governance, usually the cluster in developing countries would benefit from this connection in term of technology and

386

Santosh Mehrotra and Mario Biggeri

know-how while the advanced clusters will benefit in selling to it old machinery (or new machinery) in a sort of ‘cluster to cluster’ market cooperation and the opportunity to delocalise its production and sell services. However, since most of the advantages are for the follower most advanced clusters are reluctant to help followers (Di Tommaso and Bellandi, 2006). Cost of capital and credit The literature suggests that small firms face lower labour costs but higher capital costs than larger ones. The higher capital costs lead to a negative bias against small firms. The higher capital costs result from political pressures in favour of large enterprises, and also minimum thresholds for subsidised financial transactions in many countries. The higher capital costs may also result from higher risk that lenders face when lending to small firms, and the higher incidence of administrative costs. The lower labour costs facing small firms may be the result of exemptions, and lack of enforcements of labour regulations in small firms; they may also be due to lower average skill level of the labour force. Higher material input prices facing small firms may result from import quotas and other non-tariff barriers that give special access to imports to larger firms; they may also result from lack of bargaining power and absence of economies of scale (Haggblade et al., 1990). Clearly, if small firms face any of these problems, at least the policy-induced price differences among them will have to be addressed at the appropriate level of government. Government may intervene to simplify the procedures for small firms. In the absence of bank branches, alternative means have to be found for a credit system at local level for small enterprises. Institutional policies Fiscal reforms and tax incentives The development of the informal sector is often limited by formal constraints. Informal activities rarely receive services, finance, or infrastructure (roads, energy, education formation) from government, because of budget constraints or because ‘they do not pay tax’. Nevertheless, if tax holidays are often given to FDI and multinationals and infrastructural parks are built to induce foreign firms to invest, some incentives should be given, at least, initially to small and microenterprises. In this context, there could be also an increase in the incentives for the informal sector to embrace some relevant formal rules through the admission to infrastructural facilities, credit facilities and social benefits (for policies of registration, protection and promotion see Chapter 12). A simplified tax regime or differentiated tax rates for small enterprises have been used in some countries to encourage small firms to be included in the tax register. This may be more effective than the direct fight against tax evasion common to most informal enterprises. For example, ILO SEED shows in Peru that each additional unit of local currency of tax revenue from small firms through

Towards a policy agenda 387 improved enforcement is lost in administrative costs amounting to 75 per cent of that unit ILO, 1997 in (Veroera, 2000). Also, if subsidies are granted to large firms producing the same goods or services that small firms are producing, it reduces the latter’s competitiveness. On the other hand, there are countries where tax exemptions are available to small firms (e.g. a lower corporate tax rate in Mauritius), but this is contingent upon certification as a registered SME with the government’s small enterprise authority (Reinecke, 2002). The remarkable example of China’s fiscal decentralisation and its impact on TVE growth during 1980–93 also holds out an extremely instructive example. When there is a fiscal incentive to local governments to retain additional revenues collected, they have shown in China that they will do all in their power to support local business development. In fact, as we saw, TVEs, supported by local government, were a critical contributor to the economic growth of China for 15 years. However, one of the issues that remains unresolved is whether this local business development was made easier by the fact that these firms were mostly local government owned, rather than privately owned. In most developing countries, the fact that firms are likely to be privately owned will reduce the degree to which the tax benefits can be internalised by the local government. We noted in discussing how the high road became possible in Italy that the role of local government was fundamental to give a vision of development and to find institutional solutions for the supply of specific collective goods. However, the capacity of local governments in the developing countries of Asia we have been discussing in this book is unlikely to match these exacting requirements. The capacity of staff at central level is much better than at lower levels of government, and the requirements of local government staff capacity implicit in a high road of local cluster development may be rather high. Nevertheless, a good bureaucracy that can implement local level industrial policy is not as hard to construct as it is often believed to be. The Taiwanese bureaucracy in the 1950s was regarded as lacking in meritocracy and effectiveness – not surprising given the infamous corruption and incompetence of that bureaucracy in Mainland China before 1949. Similarly, it was only after the extensive civil service reform during the 1960s and 1970s that Korea was regarded as having a high-quality bureaucracy.30 Different is the case of Chinese local governments and policy-makers which in many cases have demonstrated very good skills in promoting their local economies. Removal of discriminatory regulation Regulations that formally apply to all enterprises may have the effect, even if unintended, of discriminating against the small firms. Uniform registration and reporting requirements may involve a higher cost for small firms relative to their turnover and resources.31 For instance, one way forward may be to replace demanding licence requirements by a simple registration procedure. This may encourage the formalisation of at least some informal enterprises. A licence usually requires an approval; registration, on the other hand, requires no approval.

388

Santosh Mehrotra and Mario Biggeri

In addition, to facilitate registration, several countries have set up centralised centres or ‘one-stop shops’, allowing potential and existing firms to all the needed information about existing regulations from one source (OECD, 1999 in Reinecke, 2002). Another example of bias against informal enterprises concerns the discriminations of urban regulations. Appropriate regulations and equitable allocation of urban space should be developed through a consultative process and negotiated settlements between all stakeholders (Chen, 2005).

Ensuring market access and market diversification For small firms, the way to survive in competitive conditions is market diversification. But the costs of market entry, especially export markets, is usually high, particularly for small firms. If clusters rely too heavily on one market, there are serious risks, as was demonstrated for the knitwear industry in Ludhiana (India) when the Soviet market collapsed after 1990. Market diversification can be supported by producers’ associations through market intelligence and by the government through support for participation in trade fairs, both national and international. This is occurring in many developing countries, for instance, many local government of ‘Specialised towns’ of Guangdong Province in China help their clusters through a permanent exhibition centre and trade fairs (Bellandi and Biggeri, 2005). A further aspect to keep in mind is that although the local system acquires skills in one type of product, it can eventually switch to other sectors that require similar skills.

Technological and management upgrading and working conditions Technological upgrading is a relevant issue within a cluster. Innovation can come from the individual firms and the benefit spillover in the local system of production. It is evident that centres of technological innovation organised by the associations of producers and the public intervention can help this upgrading. The process of collective technological innovation is based on networking since the exchange of goods and services among enterprises becomes an opportunity to exchange ideas and to innovate. Also FDI through subcontracting can help in this direction. Education policies and training would improve labour force skills and managerial capacities; demonstration and local research centres could be fundamental for technological upgrading. Local government in agreement with some of the stakeholders or viceversa can foster specific training courses and research and innovation centres. Another example is training courses for SME owners in management and business planning which include visits to more advanced firms or to fairs to show how production can be improved (Altenburg and Meyer-Stamer, 1999).

Towards a policy agenda 389 Direct policies to promote micro-enterprises and homeworkers’ activities: the role of joint and collective actions and public interventions The development of a part of the industrial system through micro- (less than five workers) and SMEs clusters could be a very important step for poverty reduction since it creates new job opportunities in a higher productivity sector, and often increases productivity and collective efficiency, but as already said it is not enough for local well-being. This policy has to be accompanied by other policies to upgrade productivity and efficiency of micro-firms, including homeworkers activities. Before going into details it is worth pointing out that although clustering policies may raise wages, it may not become the entry point for development of human capabilities. From the viewpoint of human capability, the achievement of the positive externalities of clusters is dependent on at least three other factors: environmental sustainability, social protection and worker organisations. As we will see in Chapter 12, social protection and basic social services (BSS) are the platform for long-term development of the local system. The voice of worker organisations is the key to these institutional changes. Although most clusters we analysed seem clearly on the Dirt Road 1, we also found a positive impact on human development of education and of access to BSS, and on production/worker earnings of workers’ organisations (see Chapter 4). This suggests that the inter-generational transfer of poverty and the low-capability equilibrium trap can be broken and that it is possible also for homework to be the launch pad for upward mobility for the households and for the local system of production. Membership based organization of the poor are a relevant starting point, but for further development of support to home-based producers, NGOs and local government become a pre-requisite for both social protection as well as promotive action.

Organisations of small producers and homeworkers As we explained in Chapter 1, from a policy perspective, the synergy between two sets of interventions can significantly enhance the quality of growth in favour of the poor. The first synergy involves interventions within the BSS – basic education, basic health, water and sanitation, and nutrition. The second synergy is the interaction among income-poverty reduction, health/education development, and economic growth – which is the defining feature of strategic route 1 (Section 11.3). We have learnt that homework may raise family incomes, in the absence of alternative employment. But the vulnerability of workers in the relationship is inefficient and inequitable as exhibited in the involvement of children in homework at the cost of schooling, the excessively long hours worked by women, especially young women (the ‘double burden’), the low piece-rates (with delays in payment in many cases), the unhygienic working conditions, the lack of pension

390

Santosh Mehrotra and Mario Biggeri

benefits – trapping families and the local system of production (i.e. the cluster) in a low-level of equilibrium, the poverty trap. The exploitation arises essentially due to their isolation, and even though they live and work in a ‘cluster’ there is still too little collective action. Therefore in the case of sub-clusters of homeworkers and micro-informal activities a central role can be assumed by MBOPs, and by local government (and if local governments are currently weak, then higher levels of government as appropriate to a country context). In the presence of structural disarticulation with the local economic system MBOPs can have a crucial role in production/trade upgrading, but also in the social protection and in empowerment of homeworkers, information (including awareness campaigns), increasing bargaining power for fair remuneration, social cohesion (including workers solidarity and organisation), and coordination of group action to solve common problems.32 Fair dialogue with contractors, legal protection, and social protection are the missions of associations such as the ones we encountered in the country studies (SEWA, PATAMABA and Homenet). The role of MBOPs (and community based organisations, CBOs) is central in the policy agenda for homeworkers and informal sector in general because such participation can generate private joint action, cooperative efforts and then collective actions. For informal workers such as homeworkers, these groups help to raise self-esteem, social cohesion and trust and their transformation from passive to active actors in local development. ‘A good example is financial services for micro-enterprises which build on group guarantees rather than tangible collateral’ (Altenburg and Meyer-Stamer, 1999). If the participation is large and the coalition is strong it can be possible to move to public collective action with the intervention of local government to provide public goods (at least basic productive infrastructure and BSS). In many countries artisan associations had a central role in the implementation of these public goods. Bottom-up institutional innovations should be encouraged, promoting and strengthening informal sector organisations and institutions for collective action. Pooling of community resources and collective action could be the way to reduce risk (sharing it in the community). Cooperatives can be fundamental in overcoming individual investment constraints. If there is scarce entrepreneurial capacity, because of scarce risk propensity, cooperatives, collective (village) and local government actions and investment could intervene. In sub-clusters of homeworkers as the ones we studied in Pakistan and in India where limited collaboration exists among workers the first step is to create such associations, which requires an understanding of the nature of the community system that already exists, to see how best one might build upon existing social networks. This means that there is not a unique response to institutional constraints which in different context may need different answers. Indeed, the final policy implication relates to the crucial promotive role of the organisation of workers/producers – an issue we return to in Chapter 12. Organisation of producers through associations and/or cooperatives is the only way of strengthening the bargaining position of homeworkers (perhaps with the help of an NGO). It also determines advantages (economies) in procuring raw materials, ensure fair piece-rates, explore markets, joint marketing arrangements

Towards a policy agenda 391 (as in some of Indian clusters studied) and with government support, organise credit and assist in design development (as in many of the South East Asian case studies). This is fundamental to trigger the second synergy which is crucial for the human development of the household and of the local system of development. However, such action would have to be supported by government, since the cooperative would face the combined resistance of both subcontractors as well as their employers (the principal firm).33 In the country studies there are some examples of joint action with the interventions of associations, such as SEWA in India, PATAMABA in the Philippines and Homenet in Indonesia and Thailand. Here we report the experience of the Fair Trade Group of Nepal (FTGN) as an interesting example of cooperation and collaboration (Biggeri, 2003b). The FTGN was formed by about ten associations of producers and other non profit organisation including NGOs and communitybased organisations. Although the Nepal civil war started in 1996, and Nepal is poorer in terms of GDP per capita than the countries under analysis, FTGN and its member associations were able to help and organise small family enterprises and homeworkers’ households in Kathmandu Valley and in some other parts of Nepal. Each association of FTGN is recognised by the Government and thus is formal. It provides services including promotion and technological upgrading, registration (informal) and some protection and with FTGN gives voice and services to poor homeworkers (mainly women). Each association (of artisan and craft producers) provides the following services: raw material procurement (often at a lower cost than in the market), training support, design development,34 some credit, market access in local shops (organised by association). They also help with access to international markets (EU and North America receives 80 per cent of their production) through fair trade procedures – FTGN sends the container jointly. In the last few years these associations are also trying to provide some sort of social protection to workers (including health for workers and grants for children for attending school) (Biggeri, 2003b). Access to services and information network Investments, for the provision of basic infrastructure (like water, electricity, good roads, telephones, etc.) facilitates the development process of small enterprises and contribute to improving productivity and working conditions of the informal and formal micro-enterprises and SMEs (for instance, knowledge of laws, rights and duties and regulations on safety in the work place). Other services are: technical and extension services – including the visit of technicians to the small-enterprise – quality control, credit (see later), services for administration and to registration. Then, relevant for the economic and social development of the area are counselling on safety equipment, and legal counselling for workers including homeworkers (mainly women). Most of these actions and services are better via collective action – primarily because micro-enterprises are often powerless against predatory government officials out to make a quick buck. Only some of these require public interventions.

392

Santosh Mehrotra and Mario Biggeri

In the case of informal micro-enterprise and the homeworkers the association of producers or the cooperatives could be an answer to this problem since they could buy a mobile telephone or better a user-friendly personal computer (as the ones developed in India for semi-illiterate people) that with Internet access can give information on prices of products. Access to telephony itself could enormously increase the networking capabilities of workers, as well as cut their transportation costs. In more advanced clusters – such as a few of the case studies in South East Asia – the services in the local economic system need to improve in quantity and quality such as banking facilities, specific infrastructure (e.g. centres for advice on design, quality control, packaging, legislation on products, for example for exporting in developed countries), and contract enforcement. Education and skills upgradation Our studies demonstrated low educational levels of homeworkers. Education policies and training would improve labour force skills (e.g. new techniques) and managerial capacities (including some rudimentary knowledge of accounting); demonstration and local research centres could be fundamental for technological upgrading, but also safety training course and courses to inform the workers about legal rights.35 Courses could be organised through collective action and should include certification of skills. In fact, one function that promotive services need to perform is to help homeworkers in declining industries (e.g. bidi in India) to diversify their skills and branch out into other activities that might be on the upswing, or where product demand is expanding. The promotive role of government would be incomplete without some support, especially in the form of training for the subcontractor community. Our studies talk of the poor quality of products, and rejection rates being high. They also mention the inability of subcontractors to meet the specifications and requirements set by the principal firm. This is hardly surprising considering that many of the subcontractors have no (formal) business schooling. In several sectors in the countries studied, it was reported that the subcontractors had in fact emerged from the group of homeworkers. In some cases they may have been encouraged by the principal firm to move into the subcontracting business. Training for such contractors could serve to create entrepreneurial talents for those already within the production chain. Credit and micro-finance Credit facilities and especially access to credit for micro-enterprises would be crucial. From the beginning the debate has been whether small loans should be provided by specialised micro-finance institutions or by formal sector banks with arms working on micro-finance. So far, microfinance reaches less than 13 per cent of the estimated 550 million working poor worldwide (UNIFEM, 2005). Much of the micro-finance has been extended so far by specialised micro-finance institutions.

Towards a policy agenda 393 Where banks have been involved (as in India), the commercial banks have not seen micro-finance as a commercial proposition. In fact, the public sector banks in India were not able to internalise lending to the poor as a viable activity but only as a social obligation – something that had to be done because the government wanted it done. The poor were not borrowers, they were seen as ‘beneficiaries’ (Thorat, 2005). Despite the massive expansion of rural branches of public sector banks in India (after the nationalisation of private banks in 1969), the access of the poor to banking remained limited for over two decades. The Reserve Bank of India, through the National Bank for Agriculture and Rural Development (NABARD), started in 1991 a pilot to give micro-credit to the rural poor by linking Self-Help Groups with banks. This was done because a 1981 a survey found that 36 per cent of the rural poor still utilised informal sources of credit – on exploitative terms (Prahalad, 2005). The programme has come a long way. More than 1.6 million Self Help Groups have been linked with 35,000 bank branches of 560 banks in 563 districts (i.e. India’s 607 districts) across all of India’s 30 states. As a result, about 24 million households have gained access to the formal banking system (Thorat, 2005).36 The reason for the success was that unlike micro-finance institutions, which had focused merely on credit, the banks (especially ICICI Bank) have first helped Self-Help Groups of the poor to generate savings, save them with the banks, and then use them to borrow for productive activities. The case that is now to be made is that these savings-credit activities of the poor have to be expanded to include pensions and social insurance mechanisms as well, as the SEWA Bank in India is attempting to do. According to the country studies it is important in some cases (e.g. in some dirt roads clusters) to de-link production and trade from the provision of credit by contractors in order to reduce debt bondage.37

11.6

Concluding remarks

The objective in this chapter was to spell out policies to promote both clustering (usually involving SMEs), as well as micro-enterprises of informal workers. The experience of Chinese TVEs, as well as Italian clusters, in the twentieth century offers hope for similar development in the twenty-first century in developing countries. It is not coincidental that in both successful cases of cluster development, the local government was a key player. Capable and effective local government may be a prerequisite for the success of clusters in promoting economic growth as well as employment. In the absence of effective local governments, higher levels of government may have to perform the functions of providing the public goods that were the key to the success of clusters in the two well-known cases. However, how effective such higher levels of government may be in performing this role remains to be seen. Eventually, there may be no alternative to the strong positive role that local governments were empowered to perform – both financially as well as institutionally. Meanwhile, much can be achieved in enhancing collective efficiency through the collective action by both federations of small enterprises and employers of informal workers, as well as membership- based organisations of the poor informal workers.

394

Santosh Mehrotra and Mario Biggeri

Annex 11.1: International agencies’ efforts at promoting clusters In this section we give some details of some of the main programmes used to promote clusters in developing countries by international agencies (we already introduced the UNDP approach in the text for Indonesia). UNIDO has experience in India since 1997. After selecting the cluster UNIDO’s cluster development agent analyses the cluster with the help and the participation of local actors. The creation of an environment of trust within the cluster is a clear relevant step. The following phase is the identification of the plan of action, the implementation of the activities, monitoring, and evaluation. The management of the project is that of the cluster development agent and then passed on to local actors (UNIDO, 1999). UNIDO has developed demand driven approaches; such an approach requires the identification in the clusters of a demand constraint rather than a supply problem. The triple C approach elaborated by Humphrey and Schmitz (1995, 1996) for UNIDO is focused on the demand side: customer oriented, collective and cumulative. A recent procedure to develop a cluster is the Participatory Appraisal of Competitive Advantage (PACA) (Meyer-Stamer, 2003). It is meant to give (the PACA team) a concrete aid to understand the needs of all the stakeholders to do research and analysis on how to develop the local economy in a rapid way. They offer services to the enterprises. At the end of 2004, 80 PACA projects were in place in 16 countries (Meyer-Stamer, 2004). Also the Business Development Service (BDS) Approach of ILO offer services to the enterprises. The services are usually training, marketing assistance, financial support, promotion of technological innovation and linkages among enterprises (ILO, 2005). Since 1998 there are Local Economic Development Agencies (LEDA) supported by ILO run on a voluntary basis and they pursue small local area development, social dialogue, credit assistance, marketing assistance for international market, response to poor and those marginalised by the society, social organisations promotion, and marketing assistance for exports. There are around 40 agencies that work in this field including UNDP/UNOPS, European Association of Development Agencies (EURADA), Italian Cooperation, ILO, International Liaison Services for Local Economic Development Agencies (ILS-LEDA). From the point of view of social development and poverty reduction a general criticism has been about the way they target the interventions since the beginning. These programmes are usually designed and applied only to successful cluster or to clusters that present strong potential of economic growth. In other words international agencies select the ‘right’ clusters to be certain of success. For instance (the selection criteria is similar to other programmes) the productive integration projects (PIPs) of the Inter-American Development Bank suggest that a number of critical factors should be taken into account in selecting project areas which should have the capacity to meet the administrative and accounting requirements imposed by the financing organisation and the following conditions should be

Towards a policy agenda 395 present: enterprise leadership, economically viable sector, minimum level of trust among enterprises, minimal geographic concentration of enterprises in the sector. This implies that none of the Dirt Road 1 and Dirt Road 2 clusters will be selected for a project. From the point of view of the international agency this is also reasonable. Indeed, the careful selection of projects increases the likelihood of success. We analyse below in detail to policies based on PIPs which are designed to promote competitiveness. They are based on systems of cooperation among enterprises as well as among enterprises and public and private institutions. The interest shown by the Inter-American Development Bank in this type of intervention has increased in recent years, as reflected in loans it has made to governments and in the operations of the Multilateral Investment Fund (MIF) and the Social Entrepreneurship Program (SEP). Although all these projects are based on associative schemes, they are differentiated by a wide variety of factors, the two most important of which refer to the degree of potential appropriability of the project benefits by entrepreneurs, and whether priority is given to a system of horizontal or vertical cooperation. (Dini et al., 2005) The design of each PIP involves three core elements: the specific objective, the project strategy and the institutional plan for project implementation. For a concrete example see Table 11.1. Table 11.1 Principal activities, by component and type of PIP Horizontal network

Vertical network

Cluster

Component 1: promoting cooperation among enterprises ● Analysis of Pilot initiatives; Reduction for example, of client the cluster ● Participative joint purchase enterprise of inputs inventories workshops on sector strategic planning ● Design of a cluster development strategy ● ●

●

●

Local context

●

●

●

Analysis of the local productive system Participative workshops on strategic planning Design of a development strategy

Analysis of enterprises SWOT analysis of potential groups Development of a collective project Analysis of the technicaleconomic feasibility of associative projects (Table 11.1 continued)

Table 11.1 Continued Horizontal network

Vertical network

Cluster

Component 2: facilitating access to markets ● Joint Access to Development of commercial large linkages between promotion enterprise small providers activities via procurement and purchasing catalogues, contracts enterprises participation through the Organisation in expositions, creation of and participation training of mechanisms in expositions sales staff, etc. for small ● Trade missions providers

Local context

Local image and quality standards campaign

Component 3: facilitating access to technology and business services ● Hiring of ● Development ● Assistance to technicians providers with of common for the group the support of technical ● Individualised large client services ● Generation of new advisory enterprises assistance, productive support counselling in services, such as productive laboratories, technology and training centres, etc. ● Creation of new ● management ● Procurement enterprises in of equipment productive ● for collective phases use important for development of the chain ● Technical assistance to small suppliers by the technical staff of client enterprises ● Training of specialised labour Standardisation of process technologies Coordination of joint production logistics Component 4: disseminating and developing institutional capacities ● Training of Creation of network technicians promoters specialising in ● Consolidation supplier of group development ● Strengthening of coordination mechanisms mechanisms for sector representation ●

Identification and dissemination of best practices Source: Dini et al. (2005, p. 7).

●

●

Negotiation with local technical support and training institutions to adapt their supply of services to the needs of enterprises Research projects for local enterprises Training of local labour

Training of promoters of local productive development Consolidation of local institutions

Towards a policy agenda 397

Notes 1 This policy, which tends to marginalise these activities from the development process, may be successful perhaps only when informal activities are a small fraction of the economic system. For a review see Alesina and Perrotti (1994). 2 Indeed, even in the presence of economic growth the gap between rich and poor has been increasing (Cornia, 2004), hence ‘the quality of growth matters’. 3 Although, as Peter Nolan says, we are in the presence of a ‘Global Business Revolution’ where the battlefield is dominated by Multinational Companies (MNCs) (Nolan, 2003), we are still dependent on the presence of SMEs for output expansion and labour force absorption. 4 As ILO studies affirmed in early 1990s ‘Contrary to earlier beliefs, the informal sector is not going to disappear spontaneously with economic growth. It is, on the contrary, likely to grow in the years to come, and with it the problems of urban poverty and congestion will also grow’ (ILO, 1991, p. 63). 5 The informal economy and informal sector are concepts strongly interrelated – see Chapter 1 for details. 6 As Lewis argues, ‘If 70 percent of the labour force consists of low productivity food farmers, with only a tiny surplus, the market for domestic manufactures is strictly limited. As the limits are approached, the pace of industrialisation can be maintained only by exporting manufactures . . .’ (Lewis, 1977, pp. 31–32). However, in this case although the GDP may increase substantially – it can be very difficult to diffuse the welfare and technology and most of the population would receive few benefits. Most of the population working in the informal sector would remain poor or stagnating with low productivity and disguised unemployment or would be exploited by globalised production chains without getting any benefit from growth. Here, we argue that in the process of clustering local demand can increase if workers receive the benefits of growth in terms of social protection and higher wage. 7 In 1896 Marshall affirmed the relevance of clusters of micro-enterprises and SMEs in the industrialisation process of a country Marshall (1920). 8 Networking through trade and the value chain can help to incentivise social protection in developing country clusters. However, as we discuss in Chapter 12 (Section 12.3), the process of protection has to be rooted endogenously to be really effective. 9 Furthermore, ‘When development planners speak of assisting the informal sector of developing economies they invariably are referring to operations that are located in urban areas’ (McLaughlin, 1990, p. 183). This overlooks the fact that if the problems of rural areas are not ‘addressed’ they would aggravate the urban ‘problem’. 10 Recent exceptions in the literature are Nadvi and Barrientos (2004) for UNIDO, ILO with some of SEED working papers (Kaplinsky and Readman, 2001), MCCormick and Schmitz (2002) and Altenburg and Meyer-Stamer (1999). Relevant works are those of active international associations of women such as WIEGO, Homenet and local association like SEWA (see the Annex 12.2 at the end of Chapter 12). For developed countries and especially Italy, see articles of the group of Florence-Becattini, Bellandi, Dei Ottati, Sforzi and Trigilia. 11 The successful use of selective industrial policies in the East Asian countries such as Japan, South Korea, and Taiwan is due to the judicious interventions of their governments – a mix of State interventions and market incentives (especially in relation to export markets) to promote a range of domestic industries (Chang and Grabel, 2004, p. 75). 12 It is possible to distinguish among a dynamic-advanced component (often related to micro and small enterprises and accompanied by features such as dynamism, flexibility, relatively higher accumulation, entrepreneurial creativity and initiative) and a non-advanced part. 13 At the firm level, micro-enterprises and SMEs (both in the formal and informal sector) have the advantage of flexibility and specialisation. Successful micro-enterprises and SMEs are able to react and respond quickly to changing demand and supply conditions

398

14 15

16

17

18 19 20 21 22

23 24

25

Santosh Mehrotra and Mario Biggeri

and product differentiation. Then, a micro and small enterprise embodies a stronger incentive system and at the same time reduces principal–agent costs connected to monitoring. A healthy SME sector can be the base for a competitive and innovative industrial sector and it can also complement large size industrial enterprises. ‘External economies are competitive advantages that an independent producer gains from embeddedness in a system of organised division of labour’ (Bellandi, 2004). Gulati (1997) divides clusters among modern which produce for metropolitan markets and export and artisanal that produce for local markets. Sandee (2002) distinguished between non-active cluster which produce for poor rural consumers and dynamic clusters in which enterprises are networked and can produce for international markets. Another relevant classification is the one of Altenburg and Meyer-Stamer (1999) that distinguish between three categories of clusters: survival clusters of micro- and smallscale enterprises, clusters with more advanced and differentiated mass producers, cluster of transnational corporations. Schmitz and Nadvi (1999) divide clusters between incipient and mature. The first are at the initial phase of development and use simple technologies and competencies and produce for domestic markets. The second produce for international markets and are vulnerable to international pressure. Giuliani et al. (2005) analyse and classify the upgrading of clusters in the value chain in terms of capacity to upgrade the technology (and on the basis of the number of technological gatekeepers) distinguishing between: basic cluster, low intermediate cluster, intermediate cluster, upper intermediate cluster and advanced cluster. Schmitz and other authors have criticised the original dichotomous classification; they underline how this distinction (low road, high road) does not depict the different degrees of cooperation and competition. For example, you can have a low road cluster with low cooperation but high competition and viceversa. We assume that most of the local systems of production are under competition at local and international level. Indeed, in Pakistan in the cluster analysed even the intermediaries (middleman) and traders have a negative role overexploiting homeworkers while in most of other countries their role is more positive since they can be useful in bringing labour demand and supply together. Industrial clusters as these are the product of fierce competition in the global market from MNCs and by other clusters (Giuliani et al., 2005). See also Cainelli and Zoboli (2004). This can be quite dramatic in some poor clusters of India and Pakistan and lead to labour bondage. See for instance Bianchi et al. (2000) and Rubini (2000). Similarly, in Chapter 1 we had discussed briefly empirical studies on the positive effects of clustering on wages and the reaction to external shocks. Similarly, in the Palar Valley, pressures to meet environmental standards in leather processing called for the setting up of common effluent treatment plants. As Kennedy (1999) notes, local tanneries had to cooperate for survival, forming common plants through collaborative arrangements, monitoring problems of free riding in the management of treatment plants. As a result of this local joint action, a number of tanneries have expanded while the common treatment plants have emerged as key local institutions for collective organisation. Markets opportunity can become very important when the internal demand is saturated. The opening is very much linked to the ability of local traders to access the global value chain in the hand of international buyers (Chen and Carr, 2002). There are strong and weak ties (Granovetter, 1973; Becattini, 1987, 2000; Putnam et al., 1993). Networking and strong institutions can influence trust positively. On the other hand strong ties and negative social capital can have negative impacts for the cluster. For ties and private and public networking and action see Evans (1996). For instance, private enterprise contributed less than 15 per cent of the national industrial output in 1993; in the same year local government firms contributed 42 per cent

Towards a policy agenda 399

26

27 28

29

of that output (Yingyi, 2003). By the late 1990s TVEs were being privatised, but China’s economic growth would not be the same without the early performance of TVEs. Despite its mistakes, during the Maoist period China achieved important results in economic development as well as in some important social issues. For instance, a base industrial sector was built up with the potential for further development and an agricultural infrastructure, such as irrigation schemes. The health care system and the education system were improved or introduced (especially in rural areas) and became accessible to all citizens. Also, at the end of the 1970s the ‘one child per family’ and affordable policy, introduced by Zhou Enlai in the early 1970s to stabilise the population, started to produce its first results. Indeed, the (coercive) population control partially reduced the labour supply. Out of 77 specialised towns under analysis 44 per cent of the labour force is composed of migrant workers (Bellandi and Biggeri, 2005). For a study on the effects of human development on different type of clusters with migrants see Hirsch (2005). On the role of homeworkers in Italian districts and non-districts firms see Omiccioli and Quintiliani (2000, 2004, pp. 198–207). According to their research – an ad hoc survey to compare district firms and non-district firms conducted in Italy in 1997 in manufacturing (700 enterprises) by the Bank of Italy – in district firms the recourse to subcontractors or homeworkers, in times of temporary production boom, increases 40.9 per cent, while in case of temporary falls of demand production the recourse to subcontractors or homeworkers declines by 43.6 per cent. Another interesting volume of essays is that edited by Bagnasco (1986). For homeworker women in the region of Valencia (Spain) see Picò and Sanchis (1986) and Benton (1989). In fact, Girl-children who drop out of schools to look after the younger siblings and attend to household chores while their mothers stitch garments or do embroidery or make agarbattis are never compensated for their labour and have no option of going to school or learning any skill and consequently turn into home-based workers when they start their families. Home-based workers subsidise capital costs of exports by working from their homes using their space, water, electricity and family time and yet for these workers and producers there exist no special rights or priority in housing schemes. (Singh, 2000, p. 11)

30 31 32 33 34 35 36 37

The girl is undermining her capabilities as a child (e.g. to be able to be educated) and compromising her future capabilities as adults (Biggeri et al., 2006). As Chang (2003) notes, Korea was sending its bureaucrats to Pakistan and the Philippines for extra training until the late 1960s. We cannot disagree with De Soto (1989) who in an influential study had argued that it took 289 days to register an informal enterprise in Peru and compliance costs were so big that they absorbed 80 per cent of a small firm’s profits. See for instance the results of the conference of WIEGO and SEWA (Chen et al., 2005). This is precisely the direction recommended in the Draft National Policy on Homeworkers of the Government of India Ministry of Labour 2000. For some products the association takes care of the final phase of the production. Apprenticeship is an important alternative strategy for acquiring skills and thus may be fostered by giving at the end of the apprenticeship period a grant to the owner and to the apprentice. However, about 60 per cent of the total Self-Help Groups linked to credit in the country are in the southern states of India. But in states which have a higher incidence of poverty the coverage is small. It is also important to understand that micro-credit is a facilitator that helps poverty reduction if the debtor can repay otherwise it can become a burden that is, other policies are needed for this ultra-poor category.

12 Extending social insurance to informal wage workers Santosh Mehrotra and Mario Biggeri

Since the eighteenth century, the rise of tax-based social spending1 [in rich countries] has been at the heart of government growth. It was social spending, not national defence, public transportation, or government enterprises, that accounted for most of the rise in governments’ taxing and spending as a share of GDP over the last two centuries. The increasing role of social spending in our lives has been linked to three other great social transformations: the transition to fuller democracy, the demographic transition toward fewer births and longer life, and the onset of sustained economic growth. Social spending’s share of national product derives its permanence from the likely permanence (we hope) of these three great transformations – that is, of democracy, of human longevity, and of prosperity. (Lindert, 2003)

The argument often made is that the low and low-middle income countries are too poor to be able to offer social security to its population. The welfare state came into being in the advanced capitalist countries during the twentieth century as their incomes rose rapidly, and that it was rising GDP that made it possible to increase social spending. However, what is interesting is that the rise of social spending went hand in hand with rising national incomes, it did not follow it. Social spending began rising after 1880. In the late nineteenth century about 1870, the average for general government expenditure in Europe as a per cent of GDP was around 10.8 per cent. During post-First World War it had risen to 19.6 per cent, and by pre-Second World War (1937) to 23.8 per cent (Tanzi and Schuknecht, 2000). The latter are also the shares of government expenditure in GDP in most developing countries under discussion here (closer to the low teens in sub-Saharan Africa, but around the low twenties in the rest of the developing countries) – and should rise with economic growth. There is, little in history to justify the argument that social security in the twenty-first century must remain the preserve of the advanced capitalist countries. In the richer countries, social security also grew with the spread of democracy (Lindert, 2003). Like advanced capitalist countries in the late nineteenth century, most developing countries have moved towards becoming democratic in the last quarter of the twentieth century. Most have also experienced sharp rises in

Social insurance for informal wage workers 401 longevity, which was another driver of the growth of social spending in rich countries. And as in the richer countries over a hundred years ago, developing countries are fairly well advanced in their demographic transition (except in Africa and Pakistan). Most also have a growing issue of ageing, although the problem is more serious in East Asia, and less so in Latin America and South Asia. Yet, all the Asian countries examined in this book have very weak social security systems. In all the South East Asian countries the public sector has given priority to social investments in basic health care and primary education, and paid little attention to social insurance or social assistance to maintain minimum levels of income. Social insurance coverage is available only within the formal sectors of the economy. Even among the developing regions, Latin America and the Middle East have higher levels of social spending (Lindert, 2003). Even in the middle-income countries of South East Asia, pension scheme coverage is only 15 per cent of the labour force in Indonesia (late 1990s), 28 per cent in the Philippines, and 7 per cent in Thailand.2 Of the East Asian countries only Thailand and Korea had unemployment insurance at the time of the East Asian crisis (expanded after the crisis), but the Philippines and Indonesia and Malaysia did not.3 In other words, the social insurance mechanisms were confined to those in the formal sector of the economy.4 For those in the informal economy, there was only some limited social assistance.5 India introduced a National Social Assistance Programme in 1995 for those in the informal sector, but the benefits are extremely small. In other words, the informal sector has almost no social insurance, and rather limited social assistance in Asia. Not only is social assistance and social insurance extremely limited in the low-income South Asian countries, what exists even in the middle-income countries of South East Asia studied in this book does not come close to the ILO Convention 102 on minimum standards relating to social security for formal sector workers. The Convention speaks of the following types of benefits as a minimum: medical care, sickness benefit, old-age benefit, invalidity benefit, survivor’s benefit, unemployment benefit and family benefit. In fact, a major issue around extending social protection to wage workers in the informal economy was best expressed in a 1991 report to the International Labour Conference. In his introduction, the Director-General of ILO (1991, p. 2) wrote: The dilemma, put simply, is whether to promote the informal sector as a provider of employment and incomes; or to seek to extend regulation and social protection to it and thereby possibly reduce its capacity to provide jobs and incomes for an ever-expanding labour force. It is interesting that this dilemma was also expressed to us by government observers in our workshops where we discussed the methodology, content and findings of the country studies. However, we do not see a dilemma here. We argued in Chapters 1 and 11 that social insurance for those in the informal economy can generate quality growth, not dis-equalising growth. We drew upon cluster theory to argue that case, and

402

Santosh Mehrotra and Mario Biggeri

suggested that state action on two fronts – social protection in the informal sectors and promotion of micro-enterprises – can generate a pattern of growth that not only creates employment, but promotes human development while generating profits for employers, an argument we develop further in Section 12.3 of this chapter. In Section 12.1 we discuss how the state came to play a predominant role in twentieth-century social provision in the now industrialised countries, and what lessons such experience might hold for extending social protection in contemporary developing economies. Section 12.2 examines the public and collective/ voluntary actions that have been taken in the five countries examined in this book in connection with homeworkers in the informal economy. Section 12.3 goes on to spell out the national and international actions that could or should be taken to ensure social insurance for homeworkers and informal economy wage workers generally; and thus promote the synergies at micro-level (i.e. at the level of human capabilities) as well as macro-level discussed in Chapter 1. It draws some policy implications from the preceding chapters on how human development outcomes for homeworkers can be improved through public and collective action.

12.1

Understanding the historical evolution of social protection in rich countries and its relevance for Asian countries

In the discussion among social scientists on the rise and evolution of welfare states two major ways of looking at the issue have emerged: one reflects systemic thinking, the other is focused on actors and their choices. In our discussion in this section, we shall do the same, examining systemic forces first, and then focus on actors (employees, employers and governments) and their choices. In the age of Adam Smith and the rise of classical economics, governments barely imposed any taxes at all, and there were hardly any social programmes either. Even through much of the nineteenth century, there were only two kinds of re-distributive tax-based social spending: poor relief and public schools. By 1880 social programmes had grown, and there is data for 21 European and North American countries. But the median for all social transfers (see Note 1 for definition) for these 21 countries was still only 0.29 per cent of GDP at current prices. However by 1990, the median had grown to 24 per cent of GDP. The century from 1880 to 1980 was the period that is seen as the rise of the state – and it was social spending that led to the rise of the state (Lindert, 2003; see quote at beginning of chapter). The increasing role of social spending after 1880 in the rich countries has been linked to three major social transformations: the transition to fuller democracy, the demographic transition towards fewer births and longer life, and the onset of sustained economic growth. While it is true that there is a positive correlation between level of income per capita and the share of social spending in GDP, the more important point of policy relevance for the Asian countries we have examined is that the growth of the welfare state in Europe and North America did no net damage to GDP

Social insurance for informal wage workers 403 per capita – which is why welfare states will not collapse. That the net national costs of social transfers, and of the taxes that finance them are essentially zero – this finding is of extreme significance to current day developing countries. Lindert (2003) points out that there are two general principles which explain why social spending has not weakened long-term economic growth. One is that high budget democracies show more care in choosing the design of taxes and transfers so as to avoid compromising growth. And second is that broad universalism in taxes and entitlements fosters growth better than the low-budget countries’ preference for strict means testing and complicated tax compromises. The welfare state package to which higher spending European countries are more committed includes social programme that make people more productive: public education is one such programme. A second example is their better fiscal and legal support for child care and parental leave. Their support for new parents’ career continuity cut their human capital losses from being forced off their career paths. For instance, such social investments in careers have given Sweden one of the world’s highest rates of relative earnings by women. Third, welfare states’ reliance on public health programmes have productivity benefits. It is not that the late-nineteenth- and early-twentieth-century discussion of social spending was driven by a belief in the positive relationship between economic growth and social provision (Lewis, undated). However, the hallmark of the period of the classic welfare state (from 1945 to the 1970s for Britain, and the end of the 1980s for many continental European countries) was to see the relationship between economic growth and social provision as positive. While early growth theorists (e.g. Roy Harrod in the 1940s) emphasised the growth of physical capital as basic to growth, later that discussion was extended to include human capital. The sequence of types of social programmes in advanced capitalist countries – which is of relevance again for contemporary developing countries – was as follows. Throughout the nineteenth century, spending on public education and poor relief rose. But while public spending was concentrated on public education, poor relief was largely financed from private and church charity. However, tax-based poor relief rose in England over 1795–1834, while 1834–80 is seen as the cutback era. The Reform of 1834 cut back English tax-based poor relief, but in the other European countries it rose to the English standard. Poor relief (or what today would be called social assistance) was the main form of social protection offered until around 1880. In fact, until 1880 there was almost no social insurance. Social transfers rose sharply over the half century 1880–1930, including work accident insurance and old age pensions. Unemployment insurance and unemployment benefits came last (along with housing subsidies) – halfway through the twentieth century. Bismarck’s legislation of the 1880s made Germany the pioneer in social insurance. It is of extreme significance for contemporary developing countries that the first insurance programmes – the Bismarckian programme of insuring accidents from 1881, sickness from 1883, and old age from 1889 – did not meet the modern definitions of social insurance, at least not in their earliest years. Unlike today,

404

Santosh Mehrotra and Mario Biggeri

German taxpayers contributed almost nothing to social insurance in the 1880s. Rather, the costs of insurance were borne by the workers themselves and by their employers.6 Similarly, in contemporary developing countries, our expectation is that there will be little or no interest in government to meet costs of social insurance for those in the informal economy from general government tax revenues (hence, we later propose a mechanism of financing which takes this fact of political economy as a given). Two insurance concepts have emerged in Europe during the twentieth century: the insurance concept and the redistribution concept (Broersma et al., 2003). Bismarck introduced the insurance concept in Germany in the last quarter of the nineteenth century. To Lord Beveridge is attributed the redistribution concept in Britain during the Second World War. The insurance concept is based on the idea of insuring workers against income loss in the case of unemployment, disability or retirement; it smoothes lifetime income. Both contributions and benefits depend on earnings, and most programmes are financed out of premiums and managed jointly by unions and employers. Redistribution programmes, on the other hand, are citizenship based, and cover all citizens (and not based on employment). Benefits are meant to relieve poverty, are not means-tested and provide only a guarantee of a minimum income. There is no connection here between contributions and benefits, because they are financed through general taxes. Public administrative bodies normally administer the programmes.7 The insurance principle dominates social security in Germany and to a lesser extent in the Netherlands, whereas the redistribution principle dominates social security in the United Kingdom, and to a lesser extent in Denmark and Sweden (Table 12.1). All European social security systems also provide some universal benefits: taxfinanced benefits for specific contingencies that do not require either a contribution or a means test. They can include child benefits and family support. Although the insurance principle provides relatively high benefits to a limited group, while the assistance principle grants relatively low benefits for a large group, overall spending as a percentage of GDP does not differ much between the two systems. However, as Table 12.1 shows, what does differ is the importance of specific sources of revenue for the social security system, as well as spending by type of benefit. While in Sweden and the United Kingdom the state (i.e. taxes) accounts for nearly 55 per cent of revenues, in Germany and the Netherlands, the state contributes less than 30 per cent. Insured persons themselves contribute the smallest proportion of all three contributors in Sweden, the United Kingdom and Denmark. The largest contributions in all other countries come from the insured persons and employers taken together – which of the two groups contributes more varies by country. As regards the composition of spending by type of benefits, social insurance accounts for the largest share for all countries. However, while in Denmark and the United Kingdom social assistance is the second largest category, in all other countries there is no pattern as regards the relative significance of social assistance and family allowances. In developing countries, while social assistance

Social insurance for informal wage workers 405 Table 12.1 Characteristics of social security systems: sources of revenue and types of benefits (per cent) Country

Sources of revenue Insured persons

Denmark 1970 1990 France 1980 1985 Germany 1960 1993 Netherlands 1960 1993 Sweden 1960 1993 United Kingdom 1959–60 1993–94

Employers

Spending by type of benefit State

Social insurance

Social assistance

Family allowance

14.1 4.6

8.9 5.1

75.8 87.8

69.4 62.6

18.5 28.6

6.6 3.3

21.0 23.0

53.4 50.3

24.1 23.7

62.6 65.7

15.8 13.2

10.6 11.2

25.9 36.8

44.4 32.3

25.0 27.6

66.7 81.7

6.6 6.4

2.0 2.6

40.9 46.0

40.3 18.0

12.2 21.4

64.4 77.5

4.8 2.2

13.0 4.8

20.5 1.0

11.0 43.0

66.9 56.1

71.3 83.8

12.2 1.5

10.7 12.8

20.0 14.0

17.9 22.8

58.7 54.7

71.0 57.8

10.8 26.2

5.4 5.6

Source: Broersma et al. (2003). Note Social insurance includes public health services.

is prevalent to a limited extent, social insurance is confined exclusively to the employees of the formal sector. In the advanced capitalist countries, the historical sequence of introduction of benefits was that work accidents were usually the first programmes to be legislated. There was relatively low employer resistance to them, since it abolished some unwanted bases for competition among firms as well as costly court procedures to determine fault, and also since legislation on these programmes often could be organised in terms of employers’ responsibility to insure their workers in some mutual or private company (Korpi and Mertens, 2004). Unemployment insurance generated the most employer resistance, and hence it was typically the last social insurance programme to be legislated. Employers’ feeling that unemployment insurance helps to maintain workers’ reservation wages, and thus to counteract downward pressures of unemployment on wage levels, made them resist.8 This history clearly indicates that in current day developing countries, there is little point in workers struggling for unemployment insurance at the current historical juncture – certainly not in low-income countries. In fact, given that unemployment insurance is not even available in the

406

Santosh Mehrotra and Mario Biggeri

formal sector of developing countries, there is no prospect of such insurance for wage employees in the informal economy. The earliest form of social transfers in Europe – poor relief – has already been replicated in developing countries in the form of cash transfers and food relief and public works programmes. Unemployment insurance, introduced the last in advanced capitalist countries, is available in only a few upper-middle income countries (e.g. Korea). Hence, social insurance is the only realistic way forward at the current historical conjuncture if social protection is to be extended to the informal economy in developing countries. The actors in the emergence of the welfare state Who were the actors in the emergence of the welfare state after 1880 in advanced capitalist economies? As Korpi and Mertens (2004) rightly points out, it was not the misery of the expanding industrial working class that caused concern among established state elites. They had learnt to live with that over centuries. The Worker Question, also termed the Social Question, arose on account of collective action by the industrial working class. The European revolutions of 1848 and those that followed over the next four years as well as the revolt leading to the Paris Commune in 1871 were telling events. The Communist League was founded in 1848, and its Manifesto began with the following: ‘A spectre is walking through Europe – the spectre of Communism’ (Marx and Engels, 1968). Marx’s Das Kapital was published in 1867. A first working class-based party was formed in Germany in 1863, and the International Working Men’s Association, later called the First International, a year later. After that socialist parties and also trade union confederations began to emerge also in other European countries. Clearly, this account suggests, that if the great hey days of social spending, and the emergence of social insurance (especially in Germany), was the result of collective action by the working class, nothing short of similar mobilisation will work for informal economy workers in contemporary developing countries. We have already argued in Chapters 4 and 5 that collective action has played a role in improving some conditions among homeworkers in the countries studied. However, we should not deceive ourselves that collective action by the working class will not cause a reaction and response. History teaches us something about that response. Korpi and Mertens (2004), researching the origins of the European welfare state, notes that there were four kinds of counter-responses. The first was repression by legislation or force to counteract workers’ organisational attempts. A second response was counter-mobilisation in labour markets by the creation of employers’ organisations and in some countries of employer-associated (‘yellow’) unions. Outside continental Europe, with the extension of suffrage to large segments of the working class, a third type of response was to use social insurance programmes to entice workers’ votes (Lindert, 2003). A fourth, and dominant response, was the development of the state corporatist strategy for counteracting working-class mobilisation. This strategy included social insurance programmes, and also included the formation of confessional (or religion-supported)

Social insurance for informal wage workers 407 cross-class parties and confessional unions. State corporatist institutions were going to encourage co-operation between employers and employees. Interestingly – from a contemporary developing country perspective – they were also going to identify and collectively strive on the basis of relatively narrow occupational communities, not broad-based class bases. It is on this same basis of narrow occupational categories of workers – on a product-group or trades basis – that we recommend (later in Section 12.3) that social insurance can be fostered in developing countries today. In fact, the state corporatist strategy came to be applied in many areas, one of the most important being social insurance.9 To establish who the main actors were in the emergence of social insurance, the timing of the rise of socialist parties, labour confederations and employers’ organisations has been studied. In 11 of the 12 European countries,10 the first social insurance laws were legislated after the foundation of a socialist party. What is also interesting is that an early left party preceded the early labourist union confederation in 11 of the 12 countries (with Britain being the exception, where trade unions came together to form the Labour Party).11 Thus socialist party activism led to the consolidation of unionism. In all European countries, early socialist parties also preceded the formation of national employers’ confederation. Thus, the dominant pattern seems to be that in Europe the early socialist party preceded the union confederation which in turn came before the employers’ confederation. The fact that the employers’ federations came last is not so surprising, since their representatives would already be present in national parliaments playing important roles even before the formation of such federations. Skocpol and Gretchen (1991) argued that the core programmes of modern welfare states were essentially ‘paternalist’, because they were initiated by men, principally for men, principally by men, working via the labour market and designed to maintain the income of the male breadwinner during periods of sickness or unemployment. This is consistent with the fact that women got the vote much later than men. The fact that women were excluded from the early state programmes of social protection in the form of social insurance meant that they remained the objects of charity for longer (Lewis, undated).12 Clearly, in contemporary developing economies, since feminisation of the workforce has accompanied an informalisation of the workforce, organisations of women will have to be more active and effective in the struggle for social insurance. Comparing the situation with contemporary developing countries in respect of political mobilisation of various groups, it is interesting that informal workers’ unions are few and far between in developing countries. Hence, they are very far from forming confederations. Meanwhile, the approach of formal sector unions in such countries towards informal sector workers has been ambiguous at best. At the same time, there is obviously little or no prospect for the employers’ of informal workers to be interested in their conditions. And as we see later, the governments in Asia have also been rather unconcerned. In addition, the international financial institutions have been encouraging greater labour flexibility globally, in both advanced capitalist as well as developing economies. Since the 1980s there has been an attack on the welfare state in

408

Santosh Mehrotra and Mario Biggeri

all OECD countries. What is remarkable, however, is that welfare pluralism has been encouraged over the last quarter century in the rich countries well after social insurance was universalised in those countries through public and collective action. Social spending increased to levels that made it over half of total government spending in most rich countries. But welfare pluralism is being foisted upon developing countries when a fraction of the population has access to social security. Combining labour flexibility with security would build a much stronger, more cohesive society. As the ILO’s Guy Standing asserts: A fear spreading around the world is that social progress was arrested in the late twentieth century, when flexibility came to mean insecurity, and when more people found themselves living isolated lives, without social responsibilities beyond themselves and their immediate family . . . If social groups become more detached from each other in terms of security and lifestyle, the sense of fraternity or community will be eroded . . . Justice requires that everybody should be provided with basic security (ILO, 1999)

12.2

Government and non-governmental initiatives for homeworkers in South and South East Asia

There are three major ways in which social security could be promoted in developing countries for informal sector workers and their dependents: through specially designed social insurance schemes, through social assistance13 and through the extension and reform of formal sector social insurance (Van Ginneken, 1996, 2003). However, as we have seen, the countries surveyed have very limited social security mechanisms of these kinds in the informal sector implemented by the government (with some exceptions). Here we will focus on policies targeting homeworkers. While governments cannot be said to have taken much action on behalf of homeworkers in the selected countries, some action has indeed been taken. In fact, as we discuss later, the Thai government has been quite proactive in this regard.14 India None of the countries under study here have ratified the ILO Convention on Homeworkers; in fact, none of the countries that have ratified it are from the developing world. India is a signatory country of the ILO Homework Convention 177 (1996). Although the government of India has not yet ratified the Convention, it has embarked on various initiatives towards that goal. According to the National Consultation on Home Based Workers launched by the Ministry of Labour in 2000, the government’s aim is to first create the necessary infrastructure and then be able to adopt the National Policy on Homework required by the ILO

Social insurance for informal wage workers 409 Convention. In the legal sphere, amendments to the existing legislation were proposed to develop legislation for unorganised workers during the Second National Commission of Labour. In 2004, when a new government came to power, it created a National Commission for Enterprises in the Unorganized Sector, with the mandate to review the social security system available for labour in the informal sector, and make recommendations for expanding their coverage. The Commission has drafted a bill that is intended to cover 300 million of the 370 million workers in the informal sector in India. To us, the coverage of the bill seems overly ambitious – and just like the plans of the previous government which had drafted a bill on homeworkers which did not make any progress, an excessively ambitious plan may never take off. Such fears have already been expressed by membership-based organisations of informal workers (SEWA). The bill intends to cover all workers in the unorganised sector with a monthly income of Rs 5,000 and below. The problem with such a definition is: how will the income of such workers be established, given that workers often derive their incomes from different sources (including informal wage work), and in rural areas this may partly be from owned land, and hence may yield no monetary income. The benefits proposed in the bill will include (a) health insurance; (b) maternity benefits; (c) life insurance and (d) old age pension for all eligible workers. The social security cover will be a contributory one. Contributions to be made by workers, employers (wherever identifiable), and the government shall be at the rate of one rupee per day per worker (i.e. Rs 3 per worker per day or Rs 1,095 per year). The first problem with such a proposal is that only for 17 per cent of informal workers can the employer–employee relationship be identified, according to National Sample Survey data. For the remaining 83 per cent of employers, since the employer is unidentifiable, the employer contribution is proposed in the bill to be met by government – a financial burden that the government may or may not be willing to bear. The burden in such cases is proposed to be shared between the central and state governments in the ratio of 3:1. For workers belonging to the ‘Below Poverty Line’ households, the contribution will be borne by the central government in toto – raising the cost to the central government further. In other words, the proposed bill is ambitious not only in coverage (300 million informal workers), but also in its scope. The Commission has estimated that the financial cost of the daily premiums for the Central and State governments together works out to 0.8 per cent of GDP in 2004–05. Taking account of administrative expenses, as well as expenses for capacity building, the upper bound of the public outlay on the social security plan will be around 1 per cent of GDP. The Commission makes the case that given the informal sector workers contribute 45 per cent of the national income, the financial contribution of the government is modest. In India, non-merit subsidies (i.e. essentially subsidies to the non-poor) amount to about 5 per cent of GDP (Srivastava, cited in Rao, 2005) – which are sustained because of the strong political lobby of a variety of interest groups (rich farmers, industrialists). However, it is unrealistic that given the fiscal deficit to GDP ratio of around 9 per cent in 2004–05 in India (for central and state governments

410

Santosh Mehrotra and Mario Biggeri

combined), the governments will be willing to take on yet another initiative for the poor (after a national rural employment guarantee scheme which was passed by Parliament in 2005, ensuring 100 days of employment a year to one member of the household in 150 districts of the country – a scheme that will cost about 1 per cent of GDP). Nevertheless, we should note that state governments in some southern states of India have been particularly solicitous of informal workers. For instance, Kerala state has a whole series of welfare funds, which are specific to products or trades – and 60 per cent of workers in the informal sector are covered by them. Another similar initiative that has been discussed in Chapter 6 in this book that has considerably benefited bidi homeworkers in Tamil Nadu, is the tripartite body of the Bidi Welfare Fund. This fund was created in 1976 and is administered departmentally through a tax levied on the production of exports. The fund’s goal is to provide medical care, education for children, housing, water supply and recreational facilities. The fund also serves as a forum where mechanisms for implementing a policy of basic social security are discussed. Another relevant issue is whether the invisibility of homeworkers could be reduced through counting them and identifying their problems through a survey. Until now, the government has not launched an official survey of homeworkers at a national level. However, it has included key questions in the 55th National Sample Survey15 that has revealed the overall extent of homeworkers at a national level. In addition, non-government projects have contributed to the gathering of data including a five-state survey (i.e. MP, UP, Rajasthan, Gujarat and Karnataka) conducted by SEWA. The most important initiative towards the protection of homeworkers in India has come from homeworkers themselves. The Self-Employed Women’s Association SEWA has been active in the organisation of self-employed women, including homeworkers over the last two decades. The aim of SEWA is to promote protection and rights of homeworkers. In addition, Indian homeworkers belong to the international alliance of organisations, Homenet South East Asia. This network advocates homeworkers’ rights and aims at influencing national policy and legislation in their favour. In fact, as we saw in Chapter 6, SEWA was a stakeholder in this research project in India. Pakistan No specific national programme towards homeworkers in Pakistan has been implemented until today. In the legislative sphere, although no direct legislation exists to protect home-based workers, international conventions as well as the Constitution of Pakistan,16 broadly interpreted, can be applied towards such protection. At the international level, Pakistan is a signatory of the following ILO Conventions: Convention 177 (1996a) on Homeworkers, Labour Statistics Convention (1985), and Maintenance of Social Security Rights Convention, 1982. Regarding data gathering, no specific national initiative exists. The existing data are mainly obtained from individual studies undertaken by international agencies and local researchers.17

Social insurance for informal wage workers 411 Thailand Since the implementation of the Fourth National Plan (1976–81),18 the government of Thailand has favoured outsourcing to homeworkers and advocated for their rights and well-being. This was basically due to advantages that the subcontracting arrangements provided to the government including the reduction of production costs, investment promotion, export earnings and the creation of sources of employment for those excluded from the formal labour market. For these reasons, the Thai government launched various research projects to identify the main needs and conditions of homeworkers. The first initiative was the carrying out and publication of the survey report on the Employment Situation of Homeworkers in 1986 conducted by the Ministry of Labour. The Department of Labour, Welfare and Protection conducted a similar project in three provinces in the North Eastern region of the country. In addition, in 1989 Thailand became part of the Regional Research Network on the Status of Homeworkers supported by the ILO.19 Individual studies that have contributed to the data collection include an in-depth study on homeworkers in the garment industry in Bangkok undertaken by an NGO in collaboration with local academic centres. At the institutional level, the government created a committee on Development and Protection of Homeworkers (Cabinet Resolution, 24 March 1998). This committee was based on a tripartite membership that consists of NGOs, private companies and government officials. The aim is to coordinate the various initiatives at a national level. Although the government has not yet ratified the ILO Homework Convention 177 (1996), it has launched important institutional initiatives. In 1998 the Office of Homeworkers was established under the Department of Welfare and Social Protection. The Office undertakes various activities including the preparation of regulation standards for the protection of homeworkers, coordination between the various agencies involved, and the coordination of the gathering and updating of data. One of the most effective initiatives in favour of homeworkers undertaken in Thailand was the creation of the Homenet network, itself part of Homenet International (of which SEWA in India is also a member). This network consists of NGOs, government agencies20 and homeworkers’ groups. The main goal of Homenet is to ensure the protection of homeworkers through the provision of safety nets, social welfare and compulsory registration. In addition, the network aims to ensure the participation of homeworkers in the national economy by providing them with technical support, skill development and training. Homenet has launched various campaigns aimed at raising awareness regarding homeworkers’ rights among entrepreneurs and government officials. Indonesia In Indonesia the ILO in collaboration with the Danish International Development Agency (DANIDA) implemented a project (between 1988 and 1996) aimed at promoting the social protection of homeworkers. This project was part of

412

Santosh Mehrotra and Mario Biggeri

a three-country programme with Thailand and the Philippines. Within the same project several studies were undertaken to determine the main needs and conditions of homeworkers. For example, the project ‘Rural Women Workers in the Outsourcing System’ revealed the lack of access and control over the means of production and resources of workers that led to poor working conditions. Regarding official data gathering, the Ministry of Manpower set up a monitoring scheme to gather data regularly. Currently the Central Statistical Board is considering adding a question regarding ‘place of work’ in the national labour force surveys, also known as Sakernas series. The Indonesian government has not implemented policies specifically towards homeworkers. However it has been very active on implementing policies towards small and medium enterprises which implicitly include household enterprises. The institution responsible for this programme is the Municipal Government of Jakarta. The policies include creation of credit schemes and promotion of capital access for small- and medium-scale enterprises in collaboration with local banks. The Philippines In the Philippines the most important national initiative towards homeworkers has been the undertaking of a National Survey on Homeworkers in 1993. In the legal sphere, the government of the Philippines attempted to regularise subcontracting agreements and provide legal protection to homeworkers. In 1997, the Department of Labour and Employment attempted to amend the Labour Code (i.e. Department Order No. 10) and include additional provisions to regularise the subcontracting agreements (Code 10). Unfortunately, due to the change of government the amendment was never implemented. Non-governmental initiatives include the advocacy campaign aimed at promoting the protection of homeworkers developed by the PATAMABA network. This organisation consists of a network of NGOs and homeworkers in the Philippines. PATAMABA originally supported the establishment of Code 10 and strongly advocated for the implementation of Department Order 5 (D.O. No. 5) that included special provisions in favour of homeworkers, including further details about the rights of homeworkers. In addition, PATAMABA is currently undertaking an advocacy campaign aimed at pressuring the government to ratify the ILO Convention 177. The lukewarm attitude of governments in Asia is not dissimilar to the situation in Latin America. However, in many Latin American countries including Peru, Brazil, Paraguay and Argentina, laws regarding homeworkers were made in the early 1990s. The law, with the exception of Chile,21 considers home-based work to be waged employment and provides equality of treatment in terms of social security and wage level rights. For example, the Peruvian labour legislation contains a number of long-standing and recent legal provisions protecting home-based work. The first embodied legislation dates back to 1918 and referred to the establishment of minimum remuneration for piece work to homeworkers in State enterprises (Article 27 of Act No. 2851).22

Social insurance for informal wage workers 413 Nonetheless, despite important progress in legislative terms in favour of homeworkers, most standards and regulations in Latin America are largely disregarded. Non-compliance is due to high administrative costs, and inappropriate monitoring schemes (Tomei, 2000). In other words, by and large, there has been limited government action in favour of homeworkers, both in Latin America as well as Asia. The only exception in Asia is perhaps Thailand. With this overview in mind of policies in place, we finally turn to some of the policy implications of our work in the area of social protection.

12.3

Implications of this study for social protection policy

We began in Chapter 1 by noting how important the informal economy has become in most developing countries, especially in Asia. It is a major contributor not only to employment but also to output – a fact also emphasised by Chapters 6–10. Just by way of example, the National Accounts Statistics of 1995 in India confirm that nearly 65 per cent of national income is contributed by the unorganised sector (Bhatt, 2000). Its contribution to the export earnings of many of the countries examined here has been considerable. Social protection is therefore the way in which the contribution of workers in the informal economy could be recognised, protected and even enhanced. What the preceding chapters have suggested very strongly is the dual character of subcontracted homework, at the micro-(household) level as well as at the macro-level. This dual character is contradictory: on the one hand, it is an important source of income for the homeworker households; on the other hand, the conditions of work, the low rates of pay, the close to poverty-line existence of the worker households, the health and child labour problems, all call out for much greater public intervention to protect the households. At the macro-level, forces are at work strongly encouraging the growth of subcontracting (as discussed in Chapter 1). If the synergies (discussed in Chapter 1) are to be realised, then public action needs to recognise both the efficiencies as well as inefficiencies of subcontracted homework. Subcontracted industrial outwork may well be an efficient alternative to factory employment – however, the beneficiaries of lower cost so far seem to have been the employers only; workers have not benefited. The objective is to move beyond the efficiency at the micro-level to an outcome that is efficient at the macro-level as well. Homework may raise family incomes in the absence of alternative employment. But the vulnerability of workers in the relationship is inefficient and inequitable at the macro-level – the involvement of children in homework at the cost of schooling, the excessively long hours worked by women, especially young women (the ‘double burden’), the low piece-rates, the unhygienic working conditions, the lack of pension benefits – keeping families trapped at a low-level of equilibrium, and in a poverty trap. This is particularly true in homework in the low-income countries, where schooling for all children, whether working or otherwise, is not a norm. In the middle-income countries, where schooling is a norm, the families engaged in homework are above the poverty line – perhaps only just above, and hence vulnerable.

414

Santosh Mehrotra and Mario Biggeri

National policies We first address the policies specific to child labour, and then go on to examine the policy implications for governments if they are to address the social protection needs of the households engaged in homework.

Policies specific to child labour For children homework activities in the household reduce the fixed costs of working since they reduce the transportation costs to the place of labour, transaction costs, and allow for a higher divisibility of work ‘contracts’ inside the household. All direct and indirect policies and actions (collective and noncollective) which reduce relatively the fixed costs to go to school help to reduce child labour and to increase school attendance. In other words, child labour can be reduced in homeworker households with preventive measures that affect the opportunity costs and thus parents’ decisions on the status of the child. Considering that child labour in homework has been quite invisible until now, legislation banning child labour in home-based work is clearly not the realistic way forward. The legislation in India and Pakistan banning child labour (in India the Child Labour Prevention and Regulation Act, 1986, and in Pakistan, the Employment of Children Act, 1991) applies principally to children under 14 years working outside the home in particular activities or industries, and does not include work on the family farm or home-based work. In fact, no law in India or Pakistan covers the employment of children in the agricultural sector or the informal economy (which usually employs less than ten workers). If the process of elimination of child labour in the now industrialised countries at a comparable stage of development is anything to go by, child labour laws (e.g. the 1833 Factory Act in Great Britain) did have a substantial effect on the amount of child labour and their conditions of work. However, they could operate effectively only in what is now called the formal sector of the economy, and ignored the informal sector where child labour conditions were much worse (Cunningham and Viazzo, 1996). We have emphasised in earlier chapters, in fact, that most child labour in contemporary developing economies is predominantly in the informal economy, including home-based work. The first prong of the strategy proposed here to reduce child labour is essentially a functional public school system. It is well established in the child labour literature that unless elementary schools of reasonable quality are functional, there is little likelihood of a decline in the incidence of child labour. An overwhelmingly important fact emerging from the studies is that the vast majority of children in the middle-income South East Asian countries examined are in school – whether they work or not. An ideal long-term goal is that all children are in school and do not work. However, in the medium term, that is not a realistic goal at least in India and Pakistan, or perhaps in Indonesia and the Philippines. To ensure that all children attend and complete elementary school (i.e. at least to the age of 14 or grade 8) is the responsibility of the state. Often the state is

Social insurance for informal wage workers 415 required by the country’s constitution to universalise elementary education. Usually the State has committed itself to that goal by ratifying the Convention on the Rights of the Child (1989). However, given current resource constraints, in many low-income countries of South Asia that goal may remain unachievable in the absence of official development assistance (ODA) (Delamonica et al., 2001). Yet, the difficulty is that while donor assistance seems to be plentifully available on issues related to child labour, ODA on basic education has been roughly constant through much of the 1990s (Mehrotra, 2001). In Chapter 5 we saw that it is fundamental to reduce the fixed costs of sending children to school and to improve quality. Our studies are suggesting that schooling is of poor quality particularly in India and Pakistan, even if accessible in urban areas, and we know from other literature that access itself is poor in rural areas (at least in India and Pakistan) (PROBE, 1998; Mehrotra et al., 2005). The emphasis needs to be on improving quality in urban areas, and improving both access and quality in rural areas in South Asia – and to make it affordable, since cost has been cited as an important disincentive. Most children in Pakistan (and in one sector in India) were going to Koranic schools, not regular schools, and hence receiving very little non-religious instruction. Girls in both South Asian countries needed special incentives to be sent to school. However, it is obvious that in the South East Asian country cases, despite the fact that children in homeworker households are working (as much as a fifth in Thailand and Indonesia) they are still going to school. Cultural differences do not account for much here in the case of the South East Asian countries: Thailand is predominantly Buddhist, Indonesia is largely Islamic, while the Philippines is largely Catholic. In all these countries public policy, however, is to achieve universalised schooling, whether in rural or urban areas, and remains affordable even in the wake of the economic crisis. If the synergy between interventions in the basic social services is to be triggered, access to universal schooling alone will not ensure that homeworker households send all their children to school. Our theoretical framework for enhancing human capabilities for the household members and reducing child labour suggested the need for simultaneous interventions in primary health care – as a means of directly reducing ill-health and indirectly as a means of reducing child mortality and thus inducing fertility decline through behavioural change. Our studies found that the largest proportion of children in the age group 6–14 that were working among homeworker households were from the largest households. Thus, the Pakistani households were the largest, and also had the highest proportion of children working; the South East Asian households were the smallest, and had the lowest proportion of children working; the Indian households were somewhere in the middle in respect of both variables. They also noted either the lack of health services or their un-affordability. Effective delivery of basic services (school, health services, safe water and sanitation) is a sine qua non, but not a sufficient condition for reducing child labour.23 The vulnerability of families working in the informal sector needs to be addressed, if exogenous shocks (e.g. the death of a father) are not to force children

416

Santosh Mehrotra and Mario Biggeri

into work at the expense of schooling. It is necessary to promote the well-being of children in home-based work through interventions which affect the families of which they are part, that is, to promote the second set of synergies (see Chapter 1). There is a strong case for either of two options: increasing public funding for the government health system, or for a universal health insurance mechanism. Both involve increased government financing. Both would attempt to reduce the current high levels of out-of-pocket spending by the poor – and both could prevent catastrophic expenditures being met out of pocket. We turn now to the more complicated issues of how to promote the well-being of homeworkers through interventions which promote the second set of synergies. Three major policy implications from the preceding chapters seem to emerge – and can be summarised in three words: registration, protection and promotion. Since we have already discussed the last item in Chapter 11, here we focus on the first two. Collecting data and registering the adult homeworkers A defining characteristic of work in the informal sector is its invisibility to the policy-maker (as we saw in Section 12.2), stemming partly from its immense diversity. Invisibility, however, is not its natural fate, since informal sector workers form the majority of those in the labour force defined as those between 15 and 65. Those under 15 are working mostly from the home, which does make them more invisible. Invisibility arises primarily from the fact that until recently, national statistical systems were not counting the informal sector, in either their household or enterprise surveys. That pattern has been broken in the last 15 years or so, with some 60 plus countries having conducted surveys focusing on the informal sector. However, within the informal sector the phenomenon of homework is not being counted. National sample surveys covering the home-based worker are a very limited and extremely recent phenomenon. Even the Indian survey of 1999 (on informal non-agricultural enterprises) only had a few questions about home-based work. Within the next five years this situation needs to change drastically if the invisibility of homeworkers within the informal sector is to be mitigated. However, relying on sample surveys is not sufficient; it is indeed an important tool for policy-making, and for advocates to engage in policy-dialogue with government policy-makers. What is equally, if not more, important is that all homeworkers are registered. For the worker’s well-being, this is of more immediate and direct importance. It is also consistent with the ILO Recommendation on Home Based Work (a companion to the Convention). This could involve the following: the issuing of identity cards (as, for example, happens in the bidi-making sector in India) and the creation of a registration board (e.g. as proposed in the Pakistan study). The latter would involve the registration of the subcontractors as well. Once the workers have an identity they can at least potentially claim some benefits. All workers, including those who work on a part-time basis, should be registered, given that for many homeworkers, this kind of workforce participation is undertaken largely by women who have other responsibilities as part of social reproduction.

Social insurance for informal wage workers 417 The registration of such workers is likely to be resisted by their ‘employers’. There is evidence from the bidi industry in India, where there exists the requirement of registration if workers are to avail of benefits of the Welfare Fund, that there is non-reporting of workers. Employers prefer a sale–purchase relationship between the homeworker and themselves, since that way they can avoid the responsibilities ensuing from a regular relationship. Of course, if the subcontractor himself is not registered, then the prospects for registration of the workers is correspondingly reduced. There is also the issue of multiple types of industrial work that such workers engage in: a woman may be an agricultural labourer in one season of the year, a leather craft worker in another, and a garment worker in another. In what sector of work should such a worker be registered? In one, both or neither? If registration brings material benefits, there is some risk of moral hazard if the worker is registered in both; at the very least, the worker needs to be registered in one sector. But since these records are meant to be computerised, and maintained locally, the risk of moral hazard can be substantially minimised if the worker is registered in both activities of work. In any case, the scale of the risk of moral hazard would depend upon the size of the potential benefits from registration. Dual registration is unlikely to be a major problem, given that our studies suggested that most homeworkers tend to have sTable relationships with a given contractor. Social insurance for informal economy workers The second policy implication emerges from the need for protection of all those engaged in the informal sector manufacturing activities. Social protection24 usually means one of two things: social assistance and social insurance. More than 2 billion – or about one-third – people in the world are not covered by any type of formal social protection, that is, neither by a contribution-based social insurance scheme nor by tax-financed social assistance. Most of these people are in the developing world. However, in developed countries only about 20 per cent remain outside the social security protection (Jenkins, 1993, pp. 3–20). Social assistance is usually provided to the poorest members of society, especially in a developing country. For instance, in 1995 India introduced a National Social Assistance Programme, which provides selected benefits to the nearly 30 per cent of the population below the poverty line, both workers and non-workers.25 Similarly, cash transfers are provided in Thailand and the Philippines to the poor (especially poor elderly and poor families with children). However, we have noted earlier that mechanisms for social insurance are all but non-existent for the majority of those who toil outside the formal sectors of the economy. We are defining social insurance here narrowly to exclude reimbursement and support in kind for health care and income maintenance of sickness. We are excluding these categories only because we have argued earlier that public health care services from general tax revenues should meet these needs universally – for the entire population.26 We exclude unemployment insurance since the share of those in informal employment is so large in the Asian countries we have examined that it is unrealistic to speak of such insurance on a wide

418

Santosh Mehrotra and Mario Biggeri

enough scale; let us recall that even in the advanced capitalist countries unemployment insurance was resisted most by employers and came last (see Section 12.1). We believe therefore, that a realistic programme of social insurance applicable to the informal economy wage workers, including homeworkers, in developing countries would be one which in the first instance only encompasses four categories of benefits: maternity benefits for women workers, disability and death benefits for all workers, old-age pension (or savings/provident fund scheme); and child care. In the short run, we would argue against a universal, citizenship-based scheme for all workers – even for these four categories of benefits. A universal scheme can only work if the government of the day was willing to use general government revenues to finance such a scheme; beneficiary contributions alone will not suffice to cover the costs of such a scheme, especially on a universal basis. The defining characteristic of wage work in the informal economy is low wages, and unstable income, since work availability is uncertain in nature and often seasonal in character. We have already seen in earlier chapters that incomes of households in industrial outwork is at (or not much above) the poverty level. If the benefits are largely dependent upon their own contributions, rather than government subsidies, then the benefits will be so small as to make it worthless for them to join such a scheme. Meanwhile, most governments in low and low-middle income countries would argue that their fiscal situation does not currently permit wholly government financed universal social insurance for its citizens (on that more later). Yet, the needs of wage workers for some minimal forms of social insurance are undeniable and urgent. Even in countries with high economic growth workers are increasingly in less secure employment, such as the self-employed, the casual and homeworkers. Hence the gradual extension of formal social security programme, that currently cover a fraction of the labour force, cannot be a solution for the social insurance needs of those in the informal economy. The labour force is expanding faster than formal sector employment. Besides, a defining characteristic of wage work in the informal economy is the precariousness of the family’s existence. Poverty is a static concept, but vulnerability is a dynamic one; those above the poverty line can easily slip into poverty. And as we noted in Chapter 1, many can suffer sudden and severe dispossession due to a variety of situations (illness or death of earning members of the family), or fluctuations in their surroundings (a drought or flood, the threat of which is never far in South Asia, or general recession as in the Asian economic crisis). Hence, we make a case here for a product-group or trades based social insurance mechanism, that is not primarily financed from beneficiary contributions. Sector and even product-group specific social insurance funds, financed mainly from a tax levied on the product, could be a significant way forward for all informal sector manufacturing activities. The point is that the tax on the product does not go into the general treasury, but is actually earmarked specifically for the purpose of creating such a social insurance fund. The same mechanism for protection that we propose here could also apply to agricultural products and

Social insurance for informal wage workers 419 hence the agricultural sector. However, the mechanism for those who work in the services would be trickier, and hence could be more difficult to implement. But even in services, such funds are not entirely inconceivable. They are already in place for film and cinema industry workers in India. Several state governments in India as well as the central government have several such funds – all in the informal sector. The Philippines already has such a welfare fund in agriculture for plantation workers. Such product or activity-based social insurance mechanisms can be an important precursor to the more universal citizenship-based social insurance mechanisms, characteristic of industrialised countries. We have already seen that in the case of bidi-making in India – an informal sector activity par excellence – a Welfare Fund has been in existence. Similar welfare funds exist for mica mines, for iron/manganese/chrome ore mines, building and other construction workers, and cinema workers. These funds place a levy on consumption or export of the products.27 There are essentially two types of funds in India: tax-based ones and contributory ones. The central government has created tax-based funds for six types of mines (mica, iron, manganese, chrome, limestone and dolomite), bidi workers, cinema workers, dock workers and construction workers. All these funds are based on a tax levied on the products produced, and then earmarked for the use of workers in that trade/product group. These funds were created by acts of Parliament, and then separate legislation was passed to impose the tax. However, the majority of these funds do not provide the key benefits that characterise social insurance – old-age pension, death and disability benefits, maternity benefits and child care. Instead, the Indian funds defray the costs of: medical facilities and hospitals; water supply and facilities for washing; educational facilities and scholarships; housing and recreation facilities; family planning services; transportation to and from place of work. The legislation also provides that the central government can use the funds to grant a loan or subsidy to a state government for other welfare schemes for workers in that trade. Several states in India have created funds that are contributory. Kerala has 20 such funds (for agricultural workers, autorickshaw drivers, cashew workers, coir workers, construction workers, transport workers and others). Similarly, Assam has one for plantation workers, as do the states of Gujarat, Maharashtra, Karnataka and Punjab. The experience with contributory funds is not encouraging (Jhabwala and Subramanya, 2000). Their coverage has been limited, and there have been difficulties in collecting the contributions. Krishnamoorthy and Nair (2001) report that the Tamil Nadu construction workers’ fund does offer group personal accident insurance and natural death and maternity assistance, but also gives assistance for education and marriage of children, and for funeral expenses. Given the limited benefits offered, the fund has accumulated large reserves. If tax-based funds are to be preferred, should the revenues come from general taxation or should there be some element of earmarking of revenues collected from taxes on the specific product produced by the informal sector workers? We don’t believe that there is any political constituency in any of our countries for a social insurance mechanism which would provide coverage to all workers, whether in the

420

Santosh Mehrotra and Mario Biggeri

formal or in the informal economy. In other words, citizenship-based entitlements of a universal nature financed from general taxation are out of the question, given the scale of poverty in most developing countries examined here. The demographic differences between Asian countries today and Europe in the early twentieth century are too great to allow for citizenship-based universal social insurance. In fact, even in Europe it was not until after the Second World War that universal citizenship-based social protection programmes became the norm. In other words, a social insurance fund receiving money from general government revenues is a political non-starter. The political economy of financing the social insurance fund we are proposing is critical to its creation in the first place, and its sustenance thereafter. Given the wide diversity of goods produced in the informal sector, and the workers’ consequent fragmentation and lack of organisation, and the large size of the informal sector workforce, it is unrealistic to expect that the government would be willing to finance such a large number of sector-specific funds from existing government revenues. The government may well be persuaded to provide some subsidies, but its most important role has to be to organise the creation, and the regulation, of such social insurance funds for a trade or product-group, and ensure a productbased levy is collected and reserved exclusively for the fund. The levy or tax has to be collected from the factories that subcontract the work or, where factories are not involved, the wholesalers responsible for marketing the product. Any products that are exported provide an additional opportunity for collecting the levy at the border. The levy has to be calibrated to meet the needs of the fund. For example, that on bidis in India is a bare Rs 0.50 (or one US cent) on 1,000 bidis, which is too low, and perhaps accounts for the fact that coverage by the fund is not universal for all bidi workers. The India study found that in many areas bidi workers were unaware of the existence of the fund. We anticipate that there will be resistance from employers’ to the imposition of taxes on products where the value chain involves fragmented informal economy workers, including homeworkers. However, the employers will be able to pass on the tax to consumers. Benefits Social insurance funds should provide the following benefits at a minimum:28 1 2 3 4

Specific health benefits,29 related to the nature of the homework, including maternity benefits; old-age pension; life insurance, that is, death and disability benefits; child care facilities.

These functions could only become operational if the social insurance fund registers the workers, contractors and subcontractors. But each of the functions is a critical element in a system of support for informal sector workers in a particular

Social insurance for informal wage workers 421 sector. The life insurance scheme would cover the family in the case of the death of a key breadwinner in the family. But disability or long-term sickness must also be covered by the social insurance fund. Catastrophic health expenditures in poor households make all the difference between living above or below the poverty line. Only specific health costs related to work-related sickness could be met from the fund, on the assumption that general health care is available through a public health system. Maternity benefits should include two elements: health care costs of maternity and a cash compensation for loss of income due to the suspension of work by women workers. However, the latter may need to be restricted to a limited number of pregnancies in the interest of promoting a small family norm. Given the large family size we noted in many of our selected Asian countries, this is important. Functional, affordable schools of reasonable quality offer an alternative to children who would otherwise work full time; welfare funds have been used to provide scholarships which make all the difference (e.g. as in the case of Bidi Welfare Fund in India). However, we would consider this an optional element in the benefit package, subject to user preference. Child care facilities on a community basis that can be organised through the auspices of a social insurance fund would enable mothers to work, enabling them to join the labour force from home, which may otherwise not be possible. Child care facilities can enable those girls to attend school who otherwise cannot because they have to look after younger siblings while the mother works. The old-age pension benefits would partially compensate for the ‘children for old-age security’ argument for high fertility. Old-age pension usually takes two forms: a provident fund and/or a pension paid on a recurrent basis on retirement. Given that wage workers in the informal economy are more likely to physically move or migrate during their work-life cycle, it is difficult to imagine how a pension payment scheme on ‘retirement’ might work in practice. Hence, an individual private account (or provident fund) could be considered. Such an account would involve savings by the individual member. To his/her personal savings could be added amounts at prescribed rates from social insurance funds’ own resources, and/or by the government – but not by the individual employer. This could act as an incentive for workers to save. Spread over a long duration of work life, the savings of even small amounts shall add up to a considerable sum. Individual employers of homeworkers (or other informal sector wage workers) should not be expected to contribute to the individual account of informal wage workers, otherwise it might serve as a disincentive for employers to take on additional workers: the classic dilemma of social protection in the informal sector, we discussed briefly in Section 12.1, would kick in. The precise nature of the benefits to be included in the portfolio of services of the fund will depend upon the specific situation. Ideally, the fund must offer only such services or benefits that it is financially capable of providing. For instance, it should not run schools or hospitals – that is naturally the responsibility of the state. However, ultimately, the decision should rest with the beneficiaries, since there are likely to be trade-offs between the number of functions and the size of benefits that can be paid out for each kind of benefit.

422

Santosh Mehrotra and Mario Biggeri

The case is often made for the need for such fragmented and dispersed workers to organise. In the absence of formal organisations, the Social Insurance Fund can, in the interim, serve as the means for bringing homeworkers together, sector by sector – giving them specific issues around which to pressure and lobby those who control their lives, whether it is employers or contractors or government officials. In other words, the state has to take proactive action to enable the collective voice of the homeworkers to be heard on specific issues of concern to them – around the institutional framework of the social insurance fund. Administration of social insurance funds If a multiplicity of product-group based or trades-based social insurance funds were to arise the administrative costs can turn out to be high.30 The administration of such funds can only be integrated if there are enough funds in a local area to need administrative consolidation. Administering such funds through existing government machinery integrated into the Labour Ministry would be the appropriate way to proceed. Instead of creating a separate machinery for each trade or product-group, there would be economies of scale if the management was concentrated in the Labour Ministry (or a separate board tailor-made for this purpose to manage a group of social insurance funds, as proposed in the Indian draft legislation). However, a monitoring mechanism would be needed for the supervision of the fund’s activities to ensure transparency and accountability. Representatives of each trade could provide that element of voice. We suggested that there is need for a system of registration of all workers (as provided for in the ILO Recommendation accompanying the Convention 177 on Homeworkers). The responsibility of registration should rest with the government ministry, rather than with the employer. Experience with the bidi industry in India has demonstrated that if identity cards to the workers are to be issued by the employers, the employers are not very responsive.31 Verification of who is a legitimate worker or not can be done on site, where peer pressure will prevent cheating, and also if subcontractors were required to be registered, then the latter could be used for purposes of verification. We have seen in our studies, that the relationship between subcontractors and homeworkers is not an arms-length one, but rather based on a stable relationship. Ultimately, as a multiplicity of trades-based social insurance funds emerge, they could later on be consolidated to all informal sector workers, or effectively made into a citizenship-based social insurance fund, deriving part funding from three sources: the contributors, the government and employers (see Table 12.1 for the sources of funding for social security in European countries). Existing institutions in the formal economy (public enterprises offering life insurance, health insurance) will need to be included in the actual provision of insurance facilities to wage workers in the informal economy through the management of the social insurance funds. In other words, the management of the funds will need to negotiate with insurance companies the terms of insurance cover provided for death, disability and maternity.

Social insurance for informal wage workers 423 International community A role for ILO The international actors that might play in role in supporting national action in the areas discussed earlier are mainly the ILO, the World Bank and international companies (the last through the enactment of codes of conduct). The core labour standards targeted for inclusion in the WTO since the Singapore Ministerial meeting (1996) of the Uruguay Round of multilateral trade negotiations – freedom of association, freedom of collective bargaining and organisation, freedom from compulsory or forced labour, a minimum wage of employment for children, and measures setting minimum standards in respect of conditions of work – do not extend to home-based workers. In other words, homeworkers were completely out of the jurisdiction of international labour standards – until the Convention on Homeworkers 1996 (No. 177). However, even that Convention has been ratified only by four countries. There is thus a strong case for domestic policy that addresses the interests of these workers (ILO, 1993, 2000; Hows, 1995). Maybe international labour standards will follow the lead of some domestic policy-makers! Informal workers are only rarely referred to explicitly in the actual texts of Conventions and Recommendations of the ILO, either to include or exclude their application to such workers. The only exceptions are Convention No. 150 (1978) and Recommendation No. 169 (1984). The need for extending protection to the informal sector has been expressed through the Recommendation: ‘While taking measures to increase employment opportunities and improve conditions of work in the informal sector, members should seek to facilitate its progressive integration into the national economy’ (Article 29). The gradual extension of the functions of the system of labour administration (including activities relating to conditions of work) to the informal sector was mentioned in the Convention No. 150 (in 1978): . . . each Member which ratifies this Convention shall promote the extension, by gradual stages if necessary, of the functions of the systems of labour administration to include activities . . . relating to the conditions of work and working life of appropriate categories of workers who are not, in law, employed persons, such as: . . . self-employed workers who do not engage outside help, occupied in the informal sector as understood in national practice. (Article 7) A number of fundamental labour standards should apply to workers independently of where they operate (Schlyter, 2002). These standards deal with human rights-related issues. The eight fundamental human rights Conventions are in the area of freedom of association and protection of the right to collective bargaining (Nos. 87 and 98), forced labour (Nos. 29 and 105), discrimination (Nos. 100 and 111), and child labour (Nos. 138 and 182). ILO member states seem to give these Conventions more importance since they have been ratified by over 100 states; five of the eight have been ratified by over 150 countries. The Director-General

424

Santosh Mehrotra and Mario Biggeri

of ILO, in his report to the International Labour Conference in 1991, emphasised that these Conventions are applicable to the informal sector: First, priority should be given to the full observance – in the informal sector and elsewhere – of ILO standards concerning fundamental human rights: freedom of association, freedom from forced labour and freedom from discrimination, as well as standards concerning the abolition of child labour. Practical measures will need to be taken in the countries concerned to ensure that national legislation is fully in accordance with these basic standards . . . National organizations of employers and workers should, in particular, be encouraged and assisted in their efforts to ensure that these standards are known, understood and applied in the informal sector. (emphasis added) There are two ILO Conventions that deal specifically with child labour: the Minimum Age Convention (No. 138) and the Worst Forms of Child Labour. Convention (No. 182). The Minimum Age for Work Convention covers all workers, but some articles in it allow countries to exclude limited categories of workers – but the country must do it at the time of ratification. None of the close to 115 countries that have ratified the Convention has excluded the informal sector. However, the national legislation in India on child labour does not prohibit home-based work (i.e. industrial outwork) by children, as we noted in Chapter 5. The Worst Forms of Child Labour Convention (No. 182) commits countries to eliminate all ‘practices similar to slavery, such as the sale and trafficking of children, debt bondage’. Within 4 years of its adoption in 2000, 116 countries have ratified it. However, we have noted in Chapter 3 as well as in the country papers on India and Pakistan that there is evidence of debt bondage in some sectors in some cases in these countries. The abolition of forced labour Conventions (Nos. 29 and 105), and the Worst Forms of Child Labour Convention (No. 182) require some forms of inspection of the informal sector. The Director-General of ILO, in his 1991 Report, noted: In most countries it would not be possible for labour inspection services or occupational health services – which are generally under-staffed or illequipped – to promote or enforce improved standards in the informal sector. But is it totally unrealistic to expect that more informal types of labour inspection and occupational health and safety counselling could be provided through community-based action, aided and stimulated by NGOs, voluntary agencies or by the government itself? Such community-based action assumes, of course, the existence of such collective bodies that might mobilise such action. And government inspection that barely functions (or has been completely co-opted by the employers) even in relation to the formal sector in most developing countries will not suddenly turn

Social insurance for informal wage workers 425 around and become concerned about these issues in the informal economy. Hence, membership-based organisations of the poor (community-based organisation) are a crucial part of the answer to mobilising the informal workers, as well as monitoring the social insurance funds we are proposing, apart from occupational health and safety counselling. The World Bank There may also be scope for some actions by the World Bank. Canagarajah and Sethuraman (2001) explore the possibility of designing new interventions – lending and non-lending services – by the Bank that address informal economy concerns. There has already been growing demand by NGOs for the Bank to expand its instruments so that it can work with partners other than the government in poverty reduction. However, it is unlikely that the Bank’s legal constraints and its operational practices can be changed easily, and this does not seem to be a feasible option in the short run. Hence, Canagarajah and Sethuraman (2001) tend to focus on non-lending instruments of the Bank. They call for a risk and vulnerability mapping and social protection inventory exercise for the informal sector. However, they admit that the World Bank’s Country Assistance Strategy (CAS) dialogue ‘is mostly oblivious to informal sector needs’. Clearly, there is much work for the Bank staff to be done internally before the CAS becomes less oblivious of the informal economy and informal wage workers in particular. There seems little mention of the informal economy in the Risk Management avatar of Bank’s Social Protection strategy (Holzmann and Jorgensen, 1999). International and national companies and codes Companies need to change their sourcing and purchasing practices, if their commitments to be socially responsible are to be more than empty words. Some corporations may be adopting socially responsible practices (World Commission on the Social Dimension of Globalization, 2004, pp. 91–95). Normally, corporations require that the manufacturers whom they contract work to sign agreements that bind them to pay legal wages and maintain safe working conditions. Corporations often state that they have done what they can, and it is not for them to monitor the practices of those who further subcontract out the work. Thus, Tesco and Sainsbury’s are required (through the UK Department of Trade and Industry’s responsible supplier relationship code) to treat suppliers fairly. But they point out that this code only affects their immediate suppliers, and that other suppliers down the chain are more difficult to monitor. There are two issues here. First, this position overlooks the narrow profit margins these corporations force the manufacturer to work with, which drives the effort down the chain to reduce labour costs. The main problem is the profit-making strategy driven by the corporations based in the North (Balakrishnan, 2002). Second, the argument depends upon how many players there are in the value chain. For instance, in wine and

426

Santosh Mehrotra and Mario Biggeri

deciduous fruit production in South Africa for the UK retailers there are only four in fruit and one in wine; under the circumstances, there is no reason why the supplier codes cannot be passed on (Oxfam, 2004a). Companies, both national as well as international, can hope to gain a competitive edge by gaining certifications (e.g. SA 8000) on grounds of their production practices and labour process (see for example, Annex 12.3). However, many companies’ inspections are often too quick, fail to involve workers and credible local organisations, and focus on a checklist of health and safety standards. Thus less visible problems – union restrictions, forced pregnancy testing, unpaid health benefits, and hidden homeworkers – are easily overlooked (Oxfam, 2004a, p. 56). These are obviously well-known to workers. We think, indeed, that the protection of local workers and their associations is the key to transform a trickle down policy to a bottom-up strategy rooted in the local system of development. Providing that the process of certification is fair, uncorrupted32 and with tight monitoring, the certification can be another instrument for a ‘better future’ and thus deserves attention. Unfortunately, up to now the results are still limited due to the small number of enterprises seeking certification (mainly large firms). Then, it is important to underline that it is relatively much easier for a larger firm to secure certification than for a small firm since the procedures for certification are not an easy task and are expensive (most of the experiences show that the benefits for the enterprise even in terms of profits are often higher than the costs).33 This last point clearly shows some limitations of this trickle down approach and its application especially in developing countries and it is evident that it leaves out the informal sector. Furthermore, the fact that these are mainly volunteer certifications brings the problems of incentives for the companies or enterprises. Incentives may come from enterprises linked in the value chain, but also possible profits due to increased demand for products with ethical work practices. Since far too few retailers and brands take labour standards seriously, the consumers’ opinions and choices are a key to finding more ‘inspired’ entrepreneurs. Most trade unions in developing countries are concerned with workers in the organized sector only, that is, regular salaried employment. Therefore, in most developing countries and in South Asia they cover only a small part of the labour force. However, as global competition intensifies, it has not only raised numbers of the unemployed and underemployed and of the informal ‘unorganised’ sector, but the ‘organised’ or formal economy is becoming more flexible and workers being retrenched as the value chain becomes global, with the consequence of trading away their rights (reduced social security, poor working conditions, fluctuating income, no registration, no job security). Trade unions in most countries, including South Asian ones, are now alive to the importance of international networking and within each country the importance of close relationships with the associations of workers of the informal sector. In this new global context it is in their own interest – it is not only an ideological matter – to engage with membership, based organizations of the poor in the informal economy. Initially, formal sector trade unions can start with

Social insurance for informal wage workers 427 activities to promote together code of conduts for decent jobs for themselves and also informal economy workers.

Northern consumers and codes Activism by consumers, especially in the North, where the product is exported from the South to the North may also play a role. Northern consumers and also local consumers can look at the history of the product (i.e. the processes): the fairness of the production process. Some international certifications and code standards are gaining ground (see Annex 12.3, which presents some of codes and certifications). They have some potential in limiting the predatory strategy at the Northern end of the value chain. The process of conversion has to start with the enterprises of the North and then gradually pass to developing countries in order to force the enterprises of the South to implement policies for social protection, fair wage, safety, environment care and so on. Otherwise, if wrongly applied it can be a bottleneck for most developing countries rather than a way to positively influence their human development. Changing some of the institutions is a way that can be pursued but requires time, specificity and positive economic and non-economic conditions (Chang, 2002; Chang and Green, 2003; Chang and Grabel, 2004; Khor, 2000, 2002).

12.4

Concluding remarks

The preceding discussion suggests that the scope for international action in future can be expanded, but at present it is limited at best. Hence, this book has tended to focus on actions necessary at the national end of the value chain. The social insurance scheme we are proposing provide the benefits required of the ILO’s minimum programme of social security: sickness benefit, health care, old-age pension, invalidity benefits, survivors’ benefit in the case of death, though not unemployment benefit and family benefit. Since we know that unemployment insurance is not available in the majority of low-income, and low-middle income countries, and was one of the last benefits to be granted even in rich countries, it is unrealistic to include unemployment insurance in our preferred benefits. Our concern, therefore, is very much focused on the social insurance elements of the minimum programme. Such a minimum programme of social insurance for wage workers in the informal economy can trigger a virtuous cycle within poor households, and reduce their vulnerability. If supplemented with targeted action promoting microenterprises in clusters – a policy conspicuous by its absence in these Asian countries so far – there is a possibility that the ‘dilemma of the informal sector’ would cease to be a dilemma. It could indeed set in motion employment-intensive growth that is genuinely pro-poor. The current dirt track of informal subcontracted work in a value chain may well turn into at least the low road to development, and ultimately even become the high road.

428

Santosh Mehrotra and Mario Biggeri

Annex 12.1: ILO Convention on Home Work No. 177, Ratification of Relevant Conventions and Conventions C177 Home Work Convention, 1996 Convention concerning Home Work (Note: Date of coming into force: 22/04/2000) Session of the Conference: 83, Date of adoption: 20/06/1996, Geneva The text of the ILO Convention given below becomes legally binding when ratified. The Convention was supplemented by a more detailed Recommendation which is not legally binding, but establishes guidelines for good practice. The General Conference of the International Labour Organization, Having been convened at Geneva by the Governing Body of the International Labour Office, and having met in its Eighty-third Session on 4 June 1996, and Recalling that many international labour Conventions and Recommendations laying down standards of general application concerning working conditions are applicable to homeworkers, and Noting that the particular conditions characterizing home work make it desirable to improve the applications of those Conventions and Recommendations to homeworkers, and to supplement them by standards which take into account the special characteristics of home work, and Having decided upon the adoption of certain proposals with regard to home work, which is the fourth item on the agenda of the session, and Having determined that these proposals shall take the form of an international Convention; adopts this twentieth day of June of the year one thousand nine hundred and ninety-six, the following Convention, which may be cited as the Home Work Convention, 1996: Article 1 For the purposes of this Convention: (a) the term ‘home work’ means work carried out by a person, to be referred to as a homeworker, ii(i) in his or her home or in other premises of his or her own choice, other than the workplace of the employer; i(ii) for remuneration; (iii) which results in a product or service as specified by the employer, irrespective of who provides the equipment, materials or other inputs used, unless this person has the degree of autonomy and of economic independence necessary to be considered an independent worker under national laws, regulations or court decisions; (b) persons with employee status do not become homeworkers within the meaning of this Convention simply by occasionally performing their work as employees at home, rather than in their usual workplaces;

Social insurance for informal wage workers 429 (c) the term ‘employer’ means a person, natural or legal, who, either directly or through an intermediary, whether or not intermediaries are provided for in national legislation, gives out home work in pursuance of his or her business activity. Article 2 This Convention applies to all persons carrying out home work within the meaning of Article 1. Article 3 Each Member which has ratified this Convention shall adopt, implement and periodically review a national policy on home work aimed at improving the situation of homeworkers, in consultation with the most representative organizations of employers and workers and, where they exist, with organizations concerned with homeworkers, and those of employers of homeworkers. Article 4 1 The national policy on home work shall promote, as far as possible, equality of treatment between homeworkers and other wage earners, taking into account the special characteristics of home work and, where appropriate, conditions applicable to the same or a similar type of work carried out in an enterprise. 2 Equality of treatment shall be promoted, in particular, in relation to: (a) the homeworkers’ right to establish or join organizations of their own choosing and to participate in the activities of such organizations; (b) protection against discrimination in employment and occupation; (c) protection in the field of occupational safety and health; (d) remuneration; (e) statutory social security protection; (f) access to training; (g) minimum age for admission to employment or work; and (h) maternity protection. Article 5 The national policy on home work shall be implemented by means of laws and regulations, collective agreements, arbitration awards or in any other appropriate manner consistent with national practice. Article 6 Appropriate measures shall be taken so that labour statistics include, to the extent possible, home work. Article 7 National laws and regulations on safety and health at work shall apply to home work, taking account of its special characteristics, and shall establish conditions

430

Santosh Mehrotra and Mario Biggeri

under which, certain types of work and the use of certain substances may be prohibited in home work for reasons of safety and health. Article 8 Where the use of intermediaries in home work is permitted, the respective responsibilities of employers and intermediaries shall be determined by laws and regulations or by court decisions, in accordance with national practice. Article 9 1 A system of inspection consistent with national law and practice shall ensure compliance with the laws and regulations applicable to home work. 2 Adequate remedies, including penalties where appropriate, in case of violation of these laws and regulations shall be provided for and effectively applied. Article 10 This Convention does not affect more favourable provisions applicable to homeworkers under other international labour Conventions. FINAL Article 11 The formal ratifications of this Convention shall be communicated to the Director-General of the International Labour Office for registration. Article 12 1 This Convention shall be binding only upon those Members of the International Labour Organization whose ratifications have been registered with the Director-General of the International Labour Office. 2 It shall come into force 12 months after the date on which the ratifications of two Members have been registered with the Director-General. 3 Thereafter, this Convention shall come into force for any Member 12 months after the date on which its ratification has been registered. Article 13 1 A Member which has ratified this Convention may denounce it after the expiration of ten years from the date on which the Convention first comes into force, by an act communicated to the Director-General of the International Labour Office for registration. Such denunciation shall not take effect until one year after the date on which it is registered. 2 Each Member which has ratified this Convention and which does not, within the year following the expiration of the period of ten years mentioned in the preceding paragraph, exercise the right of denunciation provided for in this Article, will be bound for another period of ten years and, thereafter, may

Social insurance for informal wage workers 431 denounce this Convention at the expiration of each period of ten years under the terms provided for in this Article. Article 14 1 The Director-General of the International Labour Office shall notify all Members of the International Labour Organization of the registration of all ratifications and denunciations communicated by the Members of the Organization. 2 When notifying the Members of the Organization of the registration of the second ratification, the Director-General shall draw the attention of the Members of the Organization to the date upon which the Convention shall come into force. Article 15 The Director-General of the International Labour Office shall communicate to the Secretary-General of the United Nations, for registration in accordance with article 102 of the Charter of the United Nations, full particulars of all ratifications and acts of denunciation registered by the Director-General in accordance with the provisions of the preceding Articles. Article 16 At such times as it may consider necessary, the Governing Body of the International Labour Office shall present to the General Conference a report on the working of this Convention and shall examine the desirability of placing on the agenda of the Conference the question of its revision in whole or in part. Article 17 1 Should the Conference adopt a new Convention revising this Convention in whole or in part, then, unless the new Convention otherwise provides (a) the ratification by a Member of the new revising Convention shall ipso jure involve the immediate denunciation of this Convention, notwithstanding the provisions of Article 13 above, if and when the new revising Convention shall have come into force; (b) as from the date when the new revising Convention comes into force, this Convention shall cease to be open to ratification by the Members. 2

This Convention shall in any case remain in force in its actual form and content for those Members which have ratified it but have not ratified the revising Convention.

Article 18 The English and French versions of the text of this Convention are equally authoritative.

432

Santosh Mehrotra and Mario Biggeri

ILO Recommendation on Home Work The General Conference of the International Labour Organization, Having been convened at Geneva by the Governing Body of the International Labour Office, and having met in its Eighty-third Session on 4 June 1996, and Recalling that many international labour Conventions and Recommendations laying down standards of general application concerning working conditions are applicable to homeworkers, and Noting that the particular conditions characterizing home work make it desirable to improve the application of those Conventions and Recommendations to homeworkers, and to supplement them by standards which take into account the special characteristics of home work, and Having decided upon the adoption of certain proposals with regard to home work, which is the fourth item on the agenda of the session, and Having determined that these proposals shall take the form of a Recommendation supplementing the Home Work Convention of 1996: adopts, this twentieth day of June of the year one thousand nine hundred and ninety-six, the following Recommendation, which may be cited as the Home Work Recommendation, 1996: I. Definitions and Scope of Application 1

For the purposes of this Recommendation: (a) the term ‘home work’ means work carried out by a person, to be referred to as a homeworker; ii(i) in his or her home or in other premises of his or her own choice, other than the workplace of the employer; i(ii) for remuneration; (iii) which results in a product or service as specified by the employer, irrespective of who provides the equipment, materials or other inputs used, unless this person has the degree of autonomy and of economic independence necessary to be considered an independent worker under national laws, regulations or court decisions; (b) persons with employee status do not become homeworkers within the meaning of this Recommendation simply by occasionally performing their work as employees at home, rather than at their usual workplaces; (c) the term “employer” means a person, natural or legal, who, either directly or through an intermediary, whether or not intermediaries are provided for in national legislation, gives out home work in pursuance of his or her business activity.

2

This Recommendation applies to all persons carrying out home work within the meaning of Paragraph 1.

II. General Provisions 3

(1) Each Member should, according to national law and practice, designate an authority or authorities entrusted with the formulation and implementation

Social insurance for informal wage workers 433 of the national policy on home work referred to in Article 3 of the Convention. (2) As far as possible, use should be made of tripartite bodies or organizations of employers and workers in the formulation and implementation of this national policy. (3) In the absence of organizations concerned with homeworkers or organizations of employers of homeworkers, the authority or authorities referred to in subparagraph (1) should make suitable arrangements to permit these workers and employers to express their opinions on this national policy on home work and on the measures adopted to implement it. 4

Detailed information, including data classified according to sex, on the extent and characteristics of home work should be compiled and kept up to date to serve as a basis for the national policy on home work and for the measures adopted to implement it. This information should be published and made publicly available.

5

(1) A homeworker should be kept informed of his or her specific conditions of employment in writing or in any other appropriate manner consistent with national law and practice. (2) This information should include, in particular: (a) the name and address of the employer and the intermediary, if any; (b) the scale or rate of remuneration and the methods of calculation; and (c) the type of work to be performed.

III. Supervision of Home Work 6

The competent authority at the national level and, where appropriate, at the regional, sectoral or local levels, should provide for registration of employers of homeworkers and of any intermediaries used by such employers. For this purpose, such authority should specify the information employers should submit or keep at the authority’s disposal.

7

(1) Employers should be required to notify the competent authority when they give out home work for the first time. (2) Employers should keep a register of all homeworkers, classified according to sex, to whom they give work. (3) Employers should also keep a record of work assigned to a homeworker which shows: (a) the time allocated; (b) the rate of remuneration; (c) costs incurred, if any, by the homeworker and the amount reimbursed in respect of them; (d) any deductions made in accordance with national laws and regulations; and

434

Santosh Mehrotra and Mario Biggeri (e) the gross remuneration due and the net remuneration paid, together with the date of payment. (4) A copy of the record referred to in subparagraph (3) should be provided to the homeworker.

8

9

In so far as it is compatible with national law and practice concerning respect for privacy, labour inspectors or other officials entrusted with enforcing provisions applicable to home work should be allowed to enter the parts of the home or other private premises in which the work is carried out. In cases of serious or repeated violations of the laws and regulations applicable to home work, appropriate measures should be taken, including the possible prohibition of giving out home work, in accordance with national law and practice.

IV. Minimum Age 10 National laws and regulations concerning minimum age for admission to employment or work should apply to home work. V. The Rights to Organize and To Bargain Collectively 11 Legislative or administrative restrictions or other obstacles to: (a) the exercise of the right of homeworkers to establish their own organizations or to join the workers’ organizations of their own choice and to participate in the activities of such organizations; and (b) the exercise of the right of organizations of homeworkers to join trade union federations or confederations, should be identified and eliminated. 12 Measures should be taken to encourage collective bargaining as a means of determining the terms and conditions of work of homeworkers. VI. Remuneration 13 Minimum rates of wages should be fixed for home work, in accordance with national law and practice. 14 (1) Rates of remuneration of homeworkers should be fixed preferably by collective bargaining, or in its absence, by: (a) decisions of the competent authority, after consulting the most representative organizations of employers and of workers as well as organizations concerned with homeworkers and those of employers of homeworkers, or where the latter organizations do not exist, representatives of homeworkers and of employers of homeworkers; or (b) other appropriate wage-fixing machinery at the national, sectoral or local levels.

Social insurance for informal wage workers 435 (2) Where rates of remuneration are not fixed by one of the means in subparagraph (1) above, they should be fixed by agreement between the homeworker and the employer. 15 For specified work paid by the piece, the rate of remuneration of a homeworker should be comparable to that received by a worker in the enterprise of the employer, or if there is no such worker, in another enterprise in the branch of activity and region concerned. 16 Homeworkers should receive compensation for: (a) costs incurred in connection with their work, such as those relating to the use of energy and water, communications and maintenance of machinery and equipment; and (b) time spent in maintaining machinery and equipment, changing tools, sorting, unpacking and packing, and other such operations. 17 (1) National laws and regulations concerning the protection of wages should apply to homeworkers. (2) National laws and regulations should ensure that pre-established criteria are set for deductions and should protect homeworkers against unjustified deductions for defective work or spoilt materials. (3) Homeworkers should be paid either on delivery of each completed work assignment or at regular intervals of not more than one month. 18 Where an intermediary is used, the intermediary and the employer should be made jointly and severally liable for payment of the remuneration due to homeworkers, in accordance with national law and practice. VII. Occupational Safety and Health 19 The competent authority should ensure the dissemination of guidelines concerning the safety and health regulations and precautions that employers and homeworkers are to observe. Where practicable, these guidelines should be translated into language understood by homeworkers. 20 Employers should be required to: (a) inform homeworkers of any hazards that are known or ought to be known to the employer associated with the work given to them and of the precautions to be taken, and provide them, where appropriate, with the necessary training; (b) ensure that machinery, tools or other equipment provided to homeworkers are equipped with appropriate safety devices and take reasonable steps to ensure that they are properly maintained; and (c) provide homeworkers free of charge with any necessary personal protective equipment.

436

Santosh Mehrotra and Mario Biggeri

21 Homeworkers should be required to: (a) comply with prescribed safety and health measures; (b) take reasonable care for their own safety and health and that of other persons who may be affected by their acts or omissions at work, including the proper use of materials, machinery, tools and other equipment placed at their disposal. 22 (1) A homeworker who refuses to carry out work which he or she has reasonable justification to believe presents an imminent and serious danger to his or her safety or health should be protected from undue consequences in a manner consistent with national conditions and practice. The homeworker should report the situation to the employer without delay. (2) In the event of an imminent and serious danger to the safety or health of a homeworker, his or her family or the public, as determined by a labour inspector or other public safety official, the continuation of home work should be prohibited until appropriate measures have been taken to remedy the situation. VIII. Hours of Work, Rest Periods and Leave 23 A deadline to complete a work assignment should not deprive a homeworker of the possibility to have daily and weekly rest comparable to that enjoyed by other workers. 24 National laws and regulations should establish the conditions under which homeworkers should be entitled to benefit, as other workers, from paid public holidays, annual holidays with pay and paid sick leave. IX. Social Security and Maternity Protection 25 Homeworkers should benefit from social security protection. This could be done by: (a) extending existing social security provisions to homeworkers; (b) adapting social security schemes to cover homeworkers; or (c) developing special schemes or funds for homeworkers. 26 National laws and regulations in the field of maternity protection should apply to homeworkers. X. Protection in Case of Termination of Employment 27 Homeworkers should benefit from the same protection as that provided to other workers with respect to termination of employment.

Social insurance for informal wage workers 437 XI. Resolution of Disputes 28 The competent authority should ensure that there are mechanisms for the resolution of disputes between a homeworker and an employer or any intermediary used by the employer. XII. Programmes Related to Home Work 29 (1) Each Member should, in cooperation with organizations of employers and workers, promote and support programmes which: (a) inform homeworkers of their rights and the kinds of assistance available to them; (b) raise awareness of homework-related issues among employers’ and workers’ organizations, non-governmental organizations and the public at large; (c) facilitate the organization of homeworkers in organizations of their own choosing, including cooperatives; (d) provide training to improve homeworkers’ skills (including non-traditional skills, leadership and negotiating skills), productivity, employment opportunities and income-earning capacity; (e) provide training which is carried out as close as practicable to the workers’ homes and does not require unnecessary formal qualifications; (f) improve homeworkers’ safety and health such as by facilitating their access to equipment, tools, raw materials and other essential materials that are safe and of good quality; (g) facilitate the creation of centres and networks for homeworkers in order to provide them with information and services and reduce their isolation; (h) facilitate access to credit, improved housing and childcare; and (i) promote recognition of home work as valid work experience. (2) Access to these programmes should be ensured to rural homeworkers. (3) Specific programmes should be adopted to eliminate child labour in home work. 30 Where practicable, information concerning the rights and protection of homeworkers and the obligations of employers towards homeworkers, as well as the programmes referred to in Paragraph 29, should be provided in languages understood by homeworkers. Convention no. 177 Country Ratification date Status (December 2005) C177 Home Work Convention, 1996 Finland 17:06:1998 ratified C177 Home Work Convention, 1996 Ireland 22:04:1999 ratified C177 Home Work Convention, 1996 Albania 24:07:2002 ratified C177 Home Work Convention, 1996 Netherlands 31:10:2002 ratified.

Annex 12.2: Ratification of ILO Conventions Fundamental ILO Conventions (December 2005) Selected conventions

Selected countries Brazil (90)

Freedom of Elimination of Elimination of association and forced and discrimination in collective compulsory labour respect of bargaining employment and occupation Conv. 87

Conv. 98

Abolition of child labour

Conv. 29 Conv. 105 Conv. 100 Conv. 111 Conv. 138 Conv. 182

01/02/1999 18/11/1952 25/04/1957 18/06/1965 25/04/1957 26/11/1965 28/06/2001 02/02/2000 02/11/1990

China (23)

28/04/1999 08/08/2002

France (na)

28/06/1951 26/10/1951 24/06/1937 18/12/1969 10/03/1953 28/05/1981 13/07/1990 11/09/2001

Germany (77)

20/03/1957 08/06/1956 13/06/1956 22/06/1959 08/06/1956 15/06/1961 08/04/1976 18/04/2002 30/11/1954 18/05/2000 25/09/1958 03/06/1960

India (41) Indonesia (17)

09/06/1998 15/07/1957 12/06/1950 07/06/1999 11/08/1958 07/06/1999 07/06/1999 28/03/2000

Italy (111)

13/05/1958 13/05/1958 18/06/1934 15/03/1968 08/06/1956 12/08/1963 28/07/1981 07/06/2000

Mexico (78)

01/04/1950

12/05/1934 01/06/1959 23/08/1952 11/09/1961

30/06/2000

Pakistan (34)

14/02/1951 26/05/1952 23/12/1957 15/02/1960 11/10/2001 24/01/1961

11/10/2001

Philippines (32)

29/12/1953 29/12/1953 15/07/2005 17/11/1960 29/12/1953 17/11/1960 04/06/1998 28/11/2000 26/02/1969 02/12/1969 08/02/1999

Thailand (14)

11/05/2004 16/02/2001

United Kingdom (86) 27/06/1949 30/06/1950 03/06/1931 30/12/1957 15/06/1971 08/06/1999 07/06/2000 22/03/2000 25/09/1991

United States (14)

07/10/1997 07/10/1997 24/06/2003 19/12/2000

Vietnam (16) Total (185)

02/12/1999

144

154

168

165

162

163

142

157

Ratification of ILO Conventions by countries studied India (December 2005) Convention

Ratification Date

Status

C1 Hours of Work (Industry) Convention, 1919 C2 Unemployment Convention, 1919

14/07/1921 14/07/1921

C4 Night Work (Women) Convention, 1919 C5 Minimum Age (Industry) Convention, 1919 C6 Night Work of Young Persons (Industry) Convention, 1919 C11 Right of Association (Agriculture) Convention, 1921 C14 Weekly Rest (Industry) Convention, 1921 C15 Minimum Age (Trimmers and Stokers) Convention, 1921

14/07/1921 09/09/1955 14/07/1921

Ratified Denounced on 16:04:1938 Ratified Ratified Ratified

11/05/1923

Ratified

11/05/1923 20/11/1922

Ratified Ratified

India (December 2005) Continued Convention

C16 Medical Examination of Young Persons (Sea) Convention, 1921 C18 Workmen’s Compensation (Occupational Diseases) Convention, 1925 C19 Equality of Treatment (Accident Compensation) Convention, 1925 C21 Inspection of Emigrants Convention, 1926 C22 Seamen’s Articles of Agreement Convention, 1926 C26 Minimum Wage-Fixing Machinery Convention, 1928 C27 Marking of Weight (Packages Transported by Vessels) Convention, 1929 C29 Forced Labour Convention, 1930 C32 Protection against Accidents (Dockers) Convention (Revised), 1932 C41 Night Work (Women) Convention (Revised), 1934 C42 Workmen’s Compensation (Occupational Diseases) Convention (Revised), 1934 C45 Underground Work (Women) Convention1935, C80 Final Articles Revision Convention, 1946 C81 Labour Inspection Convention, 1947 C88 Employment Service Convention, 1948 C89 Night Work (Women) Convention (Revised), 1948 C90 Night Work of Young Persons (Industry) Convention (Revised), 1948 C100 Equal Remuneration Convention, 1951 C105 Abolition of Forced Labour Convention1957, C107 Indigenous and Tribal Populations Convention, 1957 C108 Seafarers’ Identity Documents Convention, 1958 C111 Discrimination (Employment and Occupation) Convention, 1958 C115 Radiation Protection Convention, 1960 C116 Final Articles Revision Convention, 1961 C118 Equality of Treatment (Social Security) Convention, 1962 C122 Employment Policy Convention, 1964 C123 Minimum Age (Underground Work) Convention, 1965 C136 Benzene Convention, 1971 C141 Rural Workers’ Organisations Convention, 1975 C144 Tripartite Consultation (International Labour Standards) Convention, 1976 C147 Merchant Shipping (Minimum Standards) Convention, 1976 C160 Labour Statistics Convention, 1985 P89 Protocol of 1990 to the Night Work (Women) Convention (Revised), 1948

Ratification Date

Status

20/11/1922

Ratified

30/09/1927

Ratified

30/09/1927

Ratified

14/01/1928 31/10/1932

Ratified Ratified

10/01/1955

Ratified

07/09/1931

Ratified

30/11/1954 10/02/1947

Ratified Ratified

22/11/1935 13/01/1964

Denounced on 27.02:1950 Ratified

25/03/1938 17/11/1947 07/04/1949 24/06/1959 27/02/1950

Ratified Ratified Ratified Ratified Ratified

27/02/1950

Ratified

25/09/1958 18/05/2000 29/09/1958

Ratified Ratified Ratified

17/01/2005 03/06/1960

Ratified Ratified

17/11/1975 21/06/1962 19/08/1964

Ratified Ratified Ratified

17/11/1998 20/03/1975

Ratified Ratified

11/06/1991 18/08/1977 27/02/1978

Ratified Ratified Ratified

26/09/1996

Ratified

01/04/1992 21/11/2003

Ratified Ratified

Pakistan (December 2005) Convention

C1 Hours of Work (Industry) Convention, 1919 C4 Night Work (Women) Convention, 1919 C6 Night Work of Young Persons (Industry) Convention, 1919 C11 Right of Association (Agriculture) Convention, 1921 C14 Weekly Rest (Industry) Convention, 1921 C15 Minimum Age (Trimmers and Stokers) Convention, 1921 C16 Medical Examination of Young Persons (Sea) Convention, 1921 C18 Workmen’s Compensation (Occupational Diseases) Convention, 1925 C19 Equality of Treatment (Accident Compensation) Convention, 1925 C21 Inspection of Emigrants Convention, 1926 C22 Seamen’s Articles of Agreement Convention, 1926 C27 Marking of Weight (Packages Transported by Vessels) Convention, 1929 C29 Forced Labour Convention, 1930 C32 Protection against Accidents (Dockers) Convention (Revised), 1932 C41 Night Work (Women) Convention (Revised), 1934 C45 Underground Work (Women) Convention, 1935 C59 Minimum Age (Industry) Convention (Revised), 1937 C80 Final Articles Revision Convention, 1946 C81 Labour Inspection Convention, 1947 C87 Freedom of Association and Protection of the Right to Organise Convention, 1948 C89 Night Work (Women) Convention (Revised), 1948 C90 Night Work of Young Persons (Industry) Convention (Revised), 1948 C96 Fee-Charging Employment Agencies Convention (Revised), 1949 C98 Right to Organise and Collective Bargaining Convention, 1949 C100 Equal Remuneration Convention, 1951 C105 Abolition of Forced Labour Convention, 1957 C106 Weekly Rest (Commerce and Offices) Convention, 1957 C107 Indigenous and Tribal Populations Convention, 1957 C111 Discrimination (Employment and Occupation) Convention, 1958 C116 Final Articles Revision Convention, 1961 C118 Equality of Treatment (Social Security) Convention, 1962 C144 Tripartite Consultation (International Labour Standards) Convention, 1976 C159 Vocational Rehabilitation and Employment (Disabled Persons) Convention, 1983 C182 Worst Forms of Child Labour Convention, 1999

Ratification Date

Status

14/07/1921 14/07/1921 14/07/1921

Ratified Ratified Ratified

11/05/1923

Ratified

11/05/1923 20/11/1922

Ratified Ratified

20/11/1922

Ratified

30/09/1927

Ratified

30/09/1927

Ratified

14/01/1928 31/10/1932 07/09/1931

Ratified Ratified Ratified

23/12/1957 10/02/1947

Ratified Ratified

22/11/1935 25/03/1938 26/05/1955

Denounced on 14:02:1951 Ratified Ratified

25/03/1948 10/10/1953 14/02/1951

Ratified Ratified Ratified

14/02/1951 14/02/1951

Ratified Ratified

26/05/1952

Ratified

26/05/1952

Ratified

11/10/2001 15/02/1960 15/02/1960

Ratified Ratified Ratified

15/02/1960

Ratified

24/01/1961

Ratified

17/11/1967 27/03/1969

Ratified Ratified

25/10/1994

Ratified

25/10/1994

Ratified

11/10/2001

Ratified

Indonesia (December 2005) Convention

Ratification

C19 Equality of Treatment (Accident Compensation) Convention, 1925 C27 Marking of Weight (Packages Transported by Vessels) Convention, 1929 C29 Forced Labour Convention, 1930 C45 Underground Work (Women) Convention, 1935 C69 Certification of Ships’ Cooks Convention, 1946 C81 Labour Inspection Convention, 1947 C87 Freedom of Association and Protection of the Right to Organise Convention, 1948 C88 Employment Service Convention, 1948 C98 Right to Organise and Collective Bargaining Convention, 1949 C100 Equal Remuneration Convention, 1951 C105 Abolition of Forced Labour Convention, 1957 C106 Weekly Rest (Commerce and Offices) Convention, 1957 C111 Discrimination (Employment and Occupation) Convention, 1958 C120 Hygiene (Commerce and Offices) Convention, 1964 C138 Minimum Age Convention, 1973 C144 Tripartite Consultation (International Labour Standards) Convention, 1976 C182 Worst Forms of Child Labour Convention, 1999

Date

Status

12/06/1950

Ratified

12/06/1950

Ratified

12/06/1950 12/06/1950 30/03/1992 29/01/2004 09/06/1998

Ratified Ratified Ratified Ratified Ratified

08/08/2002 15/07/1957

Ratified Ratified

11/08/1958 07/06/1999 23/08/1972 07/06/1999

Ratified Ratified Ratified Ratified

13/06/1969 07/06/1999 17/10/1990

Ratified Ratified Ratified

28/03/2000

Ratified

Philippines (December 2005) Convention

Ratification Date

C17 Workmen’s Compensation (Accidents) 17/11/1960 Convention, 1925 C19 Equality of Treatment (Accident Compensation) 26/04/1994 Convention, 1925 C23 Repatriation of Seamen Convention, 1926 17/11/1960 C29 Forced Labour Convention, 1930 15/07/2005 C53 Officers’ Competency Certificates Convention, 1936 17/11/1960 C59 Minimum Age (Industry) Convention 17/11/1960 (Revised), 1937 C77 Medical Examination of Young Persons 17/11/1960 (Industry) Convention, 1946 C87 Freedom of Association and Protection of the 29/12/1953 Right to Organise Convention, 1948 C88 Employment Service Convention, 1948 29/12/1953 C89 Night Work (Women) Convention (Revised), 1948 29/12/1953 C90 Night Work of Young Persons (Industry) 29/12/1953 Convention (Revised), 1948

Status Ratified Ratified Ratified Ratified Ratified Denounced on 04:06:1998 Ratified Ratified Ratified Ratified Ratified (continued)

Philippines (December 2005) Continued Convention

C93 Wages, Hours of Work and Manning (Sea) Convention (Revised), 1949 C94 Labour Clauses (Public Contracts) Convention, 1949 C95 Protection of Wages Convention, 1949 C98 Right to Organise and Collective Bargaining Convention, 1949 C99 Minimum Wage Fixing Machinery (Agriculture) Convention, 1951 C100 Equal Remuneration Convention, 1951 C105 Abolition of Forced Labour Convention, 1957 C110 Plantations Convention, 1958 C111 Discrimination (Employment and Occupation) Convention, 1958 C118 Equality of Treatment (Social Security) Convention, 1962 C122 Employment Policy Convention, 1964 C138 Minimum Age Convention, 1973 C141 Rural Workers’ Organisations Convention, 1975 C144 Tripartite Consultation (International Labour Standards) Convention, 1976 C149 Nursing Personnel Convention, 1977 C157 Maintenance of Social Security Rights Convention, 1982 C159 Vocational Rehabilitation and Employment (Disabled Persons) Convention, 1983 C165 Social Security (Seafarers) Convention (Revised), 1987 C176 Safety and Health in Mines Convention, 1995 C179 Recruitment and Placement of Seafarers Convention, 1996 C182 Worst Forms of Child Labour Convention, 1999

Ratification Date

Status

29/12/1953

Ratified

29/12/1953 29/12/1953 29/12/1953

Ratified Ratified Ratified

29/12/1953

Ratified

29/12/1953 17/11/1960 10/10/1968 17/11/1960

Ratified Ratified Ratified Ratified

26/04/1994

Ratified

13/01/1976 04/06/1998 18/06/1979 10/06/1991

Ratified Ratified Ratified Ratified

18/06/1979 26/04/1994

Ratified Ratified

23/08/1991

Ratified

09/11/2004

Ratified

27/02/1998 13/03/1998

Ratified Ratified

28/11/2000

Ratified

05/04/1968 05/04/1968

Ratified Ratified

26/02/1969 05/12/1947 26/02/1969 08/02/1999 29/07/1964

Ratified Ratified Ratified Ratified Ratified

02/12/1969 24/09/1962 26/02/1969 05/04/1968

Ratified Ratified Ratified Denounced on 11:05:2004 Ratified Ratified Ratified

Thailand (December 2005) C14 Weekly Rest (Industry) Convention, 1921 C19 Equality of Treatment (Accident Compensation) Convention, 1925 C29 Forced Labour Convention, 1930 C80 Final Articles Revision Convention, 1946 C88 Employment Service Convention, 1948 C100 Equal Remuneration Convention, 1951 C104 Abolition of Penal Sanctions (Indigenous Workers) Convention, 1955 C105 Abolition of Forced Labour Convention, 1957 C116 Final Articles Revision Convention, 1961 C122 Employment Policy Convention, 1964 C123 Minimum Age (Underground Work) Convention, 1965 C127 Maximum Weight Convention, 1967 C138 Minimum Age Convention, 1973 C182 Worst Forms of Child Labour Convention, 1999

26/02/1969 11/05/2004 16/02/2001

Social insurance for informal wage workers 443

Annex 12.3: Organisations for/of homeworkers Women in Informal Employment Globalizing and Organizing (WIEGO) WIEGO is a worldwide coalition of institutions and individuals concerned with improving statistics, research, programmes and policies in support of women in the informal sector of the economy. Its name reflects two of its major concerns: Globalising – women workers in the informal sector are an integral part of the global economy and are affected by global trade and investment. Organising – women workers in the informal sector need to organise at local and international levels in order to respond effectively to the new opportunities – as well as the negative impacts – associated with global trade and investment. WIEGO grew out of earlier collaborations between the Self-Employed Women’s Association (SEWA), Harvard University, and the United Nations Development Fund for Women (UNIFEM). Website http://www.wiego.org/ Other non-governmental organisations The Clean Clothes Campaign (CCC) It aims to improve working conditions in the garment and sportswear industry. The CCC started in the Netherlands in 1990 and has national organisations across Europe. Website: http://www. cleanclothes.org/index.htm The Fair Wear Campaign It is a coalition of Churches, community organisations and unions, launched in Melbourne in December 1996 to assist groups and individuals to become involved in a range of actions that will draw attention and apply pressure upon government, retailers and manufacturers to make a difference to the working situation of outworkers. Website: http://fairwear.org.au/engine.php Homenet It is an international network of home-based workers and their organisations, set up to develop contacts between homeworker groups in different countries in 1994, and to coordinate international lobbying work, focusing in 1995 and 1996 on the International Labour Organisation. As well as international campaigning for the improvement of home-based workers’ conditions of work at national, regional and international levels; Homenet International collects and disseminates information on home-based work to members of the network and other interested organisations. It also assists in obtaining technical assistance for, and act as a channel for this assistance, to home-based workers. The network publishes a regular newsletter as part of the process of exchanging and disseminating information on home-based workers and their organisations. Including Mitra Wanita Pekerja Rumahan Indonesia (MWPRI) (Homenet Indonesia) and Homenet Thailand. Maquila Solidarity Network (MSN) It is a Canadian network promoting solidarity with groups in Mexico, Central America, and Asia organising in maquiladora factories and export processing zones to improve conditions and win a living wage. It supports groups in the North and South working together for employment with dignity, fair wages and working conditions, and healthy workplaces and communities. Website: http://www.maquilasolidarity.org/ resources/garment/

444

Santosh Mehrotra and Mario Biggeri

OXFAM International It is a development, relief, and campaigning organisation that works with others to find lasting solutions to poverty around the world. Website: http://www.oxfam.org.uk/ PATAMABA It is a non-stock, non-profit, people’s organisation founded to organise, consolidate, and expand the national network of informal workers, and to provide support for their personal, socio-political and economic well-being. It is composed of and led by home-based workers, construction and transportation workers, street vendors, waste recyclers and others working in the informal economy. It actively advocates and lobbies for proper legislation to protect the rights of the informal workers in the Philippines as workers and as women. They are affiliated to different trade unions and women’s organisations with similar aims. Website: http://www.homenetseasia.org/philippines/aboutus.html The Self-Employed Women’s Union (SEWA) This is a trade union registered in 1972. It is an organisation of poor, self-employed women workers. SEWA’s main goals are to organise women workers for full employment and self reliance. SEWA organises women to ensure that every family obtains full employment. By self-reliance women would be autonomous and self-reliant, individually and collectively, both economically and in terms of their decision-making ability. SEWA has an Academy which is a focal point for workers’ education and capacity-building and is also involved in collaborative and action-oriented research. SEWA also engages in literacy work and produces a newspaper and videos. It is a founder member of WIEGO. Website: http://www.sewa.org/ Women Working Worldwide It is a UK-based organisation which supports the struggles of women workers in the global economy through information exchange and international networking. It has conducted research and campaigning work on subcontracting in the garments industry and has also held a number of meetings and conferences related to this topic. Website: http://www.poptel.org. uk/women-ww/ International organisations The International Labour Organisation (ILO) It has kept the textiles, clothing and footwear sectors under review at international tripartite sectoral meetings, which provide a forum for discussion and consensus formation. The conclusions and resolutions they adopt provide guidance for policies in these sectors. The most recent reports for and of these meetings are included on the website: http://www.ilo.org/public/english/dialogue/sector/sectors/textile.htm ILO’s InFocus Programme on Boosting Employment through Small Enterprise Development (SEED) It was set up in January 2000. SEED’s mission is to help governments, social partners and communities unlock the potential for creating more and better jobs in the small enterprise sector. The small enterprise sector encompasses micro, small and medium-sized enterprises as well as cooperatives, homeworkers and other self-employed persons. It includes formal and informal enterprises, whether producing independently or operating as the smallest units in global production chains. For further details see: http://www.ilo.org/public/ english/employment/ent/sed/about/index.htm

Social insurance for informal wage workers 445 UNIFEM This organisation works to promote gender equality between women and men and to advance the status of women. In particular, it focuses on implementing the Beijing Platform for Action and other UN global commitments through political and economic empowerment of women, and through the full realisation of women’s human rights. It works with governments, NGOs, community and other organisations, as well as individuals. Website: http://www.unifem.org/ Research organisations The Institute of Development Studies, University of Sussex It is leading a major research programme on value chain analysis as part of its globalisation research programme. See the IDS globalisation website for further details on IDS work on value chain. Website: http://www.ids.ac.uk/ids/global/valchn.html The North South Institute (NSI) in Ottawa It has a long association with WIEGO as well as broader work on trade liberalisation and gender issues. An earlier sector study on the global garments industry was conducted for WIEGO by NSI. Website: http://www.nsi-ins.ca

Annex 12.4: Main international voluntary certifications, corporate social responsibility (SA 8000) and the Global Compact initiative Type of certification

Main volunteer certifications Laws of reference

Destination

Objects

QUALITY

UNI EN ISO 9001:2000

Manufacturing sector and services both in private and/or public sectors

ENVIRONMENT (ECO-AUDIT)

Reg. EMAS 1836/93 UNI EN ISO 14000/96

SAFETY

BS 8800/96 (OHSAS 18001/99)

Organisation of production, production plant, and services both private and/or public sectors Manufacturing sector and services both in private and/or public sectors

Guarantees to clients that the system of production is under control and able to produce constantly and trustable Guarantees to the collectivity that the enterprise prevents environmental damage Guarantees the adoption of correct criteria for health, sanitation and safety conditions in place of work

(Annex 12.4 continued)

446

Santosh Mehrotra and Mario Biggeri

Annex 12.4 Continued Type of certification

Main volunteer certifications Laws of reference

CORPORATE SOCIAL RESPONSIBILITY (ETHICS)

PRODUCT (ECOLABELLING) FORESTRY

SA 8000/97

ISO 26000 forthcoming in 2008 ISO 14040 – LCA Reg. CEE 880/92 FSC/93 EMAS/93 UNI EN ISO 14000/96 (ISOTR 14061/98) PEFC/99

Destination

Objects

Manufacturing sector and services both in private and/or public sectors

Guarantees a behaviour in conformity with the Universal Declaration of Human Rights (absence of child labour)

Mainly on furniture/wood manufacturing sector. Forest enterprises, trade agencies enterprises in furniture/wood manufacturing sector

Guarantees that a product satisfies the pre-requisites. Guarantees to consumers that the purchased inputs come from forests managed in environmentally sustainable manner

Note International Organization for Standardization or ISO is a network of the national standards institutes of 156 countries based on one member per country with a Central Secretariat in Geneva that coordinates the system. http://www.iso.org

Corporate social responsibility: the example of SA8000 This is one of the most well-known standards for the certification of social responsibility (Boiral, 2003). It has been created by a non-profit organisation CEPAA now SAI (social accountability international). SA8000 is based on international workplace norms in the International Labour Organisation (ILO) conventions and the UN’s Universal Declaration of Human Rights and the Convention on Rights of the Child. The main elements of SA8000 are: 1

2 3

Child Labour: No workers under the age of 15; minimum lowered to 14 for countries operating under the ILO Convention 138 developing-country exception; remediation of any child found to be working. Forced Labour: No forced labour, including prison or debt bondage labour; no lodging of deposits or identity papers by employers or outside recruiters. Health and Safety: Provide a safe and healthy work environment; take steps to prevent injuries; regular health and safety worker training; system to detect threats to health and safety; access to bathrooms and potable water.

Social insurance for informal wage workers 447 4

5

6 7

8

9

Freedom of Association and Right to Collective Bargaining: Respect the right to form and join trade unions and bargain collectively; where law prohibits these freedoms, facilitate parallel means of association and bargaining. Discrimination: No discrimination based on race, caste, origin, religion, disability, gender, sexual orientation, union or political affiliation or age; no sexual harassment. Discipline: No corporal punishment, mental or physical coercion or verbal abuse. Working Hours: Comply with the applicable law but, in any event, no more than 48 hours per week with at least one day off for every seven-day period; voluntary overtime paid at a premium rate and not to exceed 12 hours per week on a regular basis; overtime may be mandatory if part of a collective bargaining agreement. Compensation: Wages paid for a standard work week must meet the legal and industry standards and be sufficient to meet the basic needs of workers and their families; no disciplinary deductions. Management Systems: Facilities seeking to gain and maintain certification must go beyond simple compliance to integrate the standard into their management systems and practices.

Benefits of SA8000 Along with humane workplaces, the implementation of SA8000 offers more benefits to workers, companies and others: (a) Benefits for workers, trade unions and NGOs: Enhanced opportunities to organize trade unions and bargain collectively. A tool to educate workers about core labour rights. An opportunity to work directly with business on labour rights issues. A way to generate public awareness of companies committed to assuring humane working conditions. (b) Benefits for business: Drives company values into action. Enhances company and brand reputation. Improves employee recruitment, retention and productivity. Supports better supply chain management and performance. (c) Benefits for consumers and investors: Clear and credible assurance for ethical purchasing decisions. Identification of ethically made products and companies committed to ethical sourcing. Broad coverage of product categories and production geography.

448

Santosh Mehrotra and Mario Biggeri There are (as of 30 September 2005) 763 facilities (enterprises including MNCs, Cooperatives, NGOs) SA8000 certified in 47 different countries and 54 different industries with a total of 454,759 workers employed. Certified Facilities (by size) Workers employed

Number of facilities

% of total

1,000 251–1,000 51–250 1–50 missing info

110 194 232 224 3

14 25 29 29 0.4

Source: http://www.sa-intl.org/

For the implications for SMEs in developing countries see UNIDO (2002) The Global Compact initiative (UN) At the World Economic Forum, Davos, on 31 January 1999, UN Secretary-General Kofi A. Annan challenged world business leaders to ‘embrace and enact’ the Global Compact, both in their individual corporate practices and by supporting appropriate public policies. The Global Compact’s operational phase was launched at UN Headquarters in New York on 26 July 2000. It is based on the Universal Declaration of Human Rights including the Convention on Rights of the Child, the Declaration on Principles and Fundamental Rights of Workers and on the Rio Declaration on Development and Environmental sustainability. During the first Global Compact Leaders Summit, held on 24 June 2004 at UN Headquarters in New York, the Secretary-General announced the addition of a tenth principle against corruption (see later). The principles of the Global Compact Human Rights Principle 1: Business should support and respect the protection of internationally proclaimed human rights; and Principle 2: Make sure that they are not complicit in human rights abuses. Labour Principle 3: Businesses should uphold the freedom of association and the effective recognition of the right to collective bargaining; Principle 4: The elimination of all forms of forced and compulsory labour; Prinicple 5: The effective abolition of child labour; and Principle 6: The elimination of discrimination in respect of employment and occupation.

Social insurance for informal wage workers 449 Environment Principle 7: Businesses should support a precautionary approach to environmental challenges; Principle 8: Undertake initiatives to promote greater environmental responsibility; and Principle 9: Encourage the development and diffusion of environmentally friendly technologies. Anti-Corruption Principle 10: Business should work against all forms of corruption, including extortion and bribery. At the end of 2005 there were 2,727 facilities participated in the initiative, http://www.un.org/Depts/ptd/global.htm The network of the Global Compact It is relevant to emphasise that Global Compact is neither monitored nor certified by external agencies and that for a company to participate to the Global Compact is as simple as follows: 1

2

3

Sends a letter from the Chief Executive Officer (and where possible, endorsed by the board) to Secretary-General Kofi Annan expressing support for the Global Compact and its principles. Sets in motion changes to business operations so that the Global Compact and its principles become part of strategy, culture and day-to-day operations; Is expected to publicly advocate the Global Compact and its principles via communications vehicles such as press releases, speeches, etc.; and

Governments Civil society

Companies

UNEP

The global compact office

HCHR

UNDP

Local networks

Local networks

Advisory council

UNIDO ILO Labour organisations

Bussiness associations Academic institutions

Source: Bagnoli (2004).

450 4

Santosh Mehrotra and Mario Biggeri Is expected to publish in its annual report (or similar corporate report) a description of the ways in which it is supporting the Global Compact and its ten principles. This ‘Communication on Progress’ is an important tool to demonstrate implementation through public accountability.

We think the UN effort on Global Compact initiative is just another important signal. Indeed, under global compact it is fully up to the enterprise if the commitments to be socially responsible is more than empty words.

Notes 1 Lindert (2003) defines social spending as these kinds of tax-based government spending: basic assistance to poor families, or supplemental income; unemployment compensation; public non-contributory pensions, in which funds come from persons other than the recipient and his employer; public health expenditure; housing subsidies; and public expenditures on education. Social transfers are all of the earlier except public education expenditures. 2 All 7 per cent in Thailand was in the public sector and none in the private sector; in the Philippines 21 of the 28 per cent was in the private sector, the rest in the public sector; and in Indonesia, 10.6 per cent was in the private sector, and the remaining 4.9 per cent in the public sector (World Bank, 1999). 3 There were, however, some social assistance programmes: targeted distribution of imported rice to poor families in Indonesia and the Philippines; cash transfers in Thailand to elderly poor, needy families, poor children, and social funds in Thailand, Indonesia and the Philippines (World Bank, 1999). 4 In fact, the degree of government subsidy for social insurance for those employed in the formal sector is remarkable. For example, pension provisions in Thailand for private sector employees are both narrowly targeted and duplicative, with one-third of the newly introduced Social Security Scheme financed by the government budget (World Bank, 1999). 5 Meanwhile, budget allocations on a cash transfer programme for the elderly poor and for poor families reach a small share of those eligible and delivers a fraction of the value of the poverty line income (World Bank, 1999). 6 In the case of old-age and invalidity insurance, the state paid only 6 per cent of all insurance revenues as of 1891 and only 18 per cent in 1908 (Lindert, 2003). 7 A third alternative is the savings concept. Compulsory savings provide social protection for individuals, not only for retirement pensions but also for contingencies such as unemployment, disability and ill health. Benefits depend upon the accumulation of an individual’s compulsory savings. The typical case is the scheme in Singapore. 8 In most countries benefits are either flat rate or if earnings related, often have low ceilings. 9 Imperial Germany (created in 1870 by Bismarck) was the well-known case. Bismarck followed a two-pronged strategy to counter worker mobilisation. One of the prongs was repression based on the 1878 ‘Law on the Socially Dangerous Striving of Social Democracy’. The other was the social insurance programme, which was more a counter-attack than a yielding to popular demand (Korpi and Mertens, 2004). A confessional party was founded in 1870, a confessional union confederation in 1899. But a national employers’ federation did not arise until 1913. 10 The 12 countries are: Austria, Belgium, Denmark, Finland, France, Germany, Italy, Netherlands, Norway, Sweden, Switzerland and the United Kingdom. The exception was Finland.

Social insurance for informal wage workers 451 11 One factor for the British exception might be that the British electoral model with elections following the first-past-the-post rule had generated a system of two dominant parties, making it difficult for new parties to emerge. The same reversed pattern is also found in the one-time British colonies: Australia, Canada, New Zealand, and the United States, who share the British electoral model and two dominant parties (Korpi and Mertens 2004). 12 In Britain, the health insurance programme introduced in 1911 provided access to a general practitioner as well as a sickness benefit for the person insured. However, despite the fact that the main concern of the health authorities of the period centred on the high infant mortality rate, the female and child dependents of the male worker were not entitled to access to medical care (Lewis, undated). 13 Social insurance usually consists of measures to reduce risks that a worker is subject to through: old age pension, unemployment benefit, maternity benefit, life insurance (including disability) and health insurance. Social assistance usually consists of cash payments of various kinds (e.g. child allowance, food subsidies). 14 In the view of seminar participants in our Bangkok workshop (December 2001), an important reason for the Thai government being as solicitous as it has been for homeworkers is that the sectors where homework is prevalent are significant contributors to foreign exchange earnings. 15 A question on the place of work was canvassed in the 55th Round National Sample Survey in 1999–2000. In addition, it included the ‘unincorporated proprietary and partnership enterprises’ as informal sector enterprises. 16 Specifically the Article of the Pakistani Constitution deals with ‘elimination of all forms of exploitation and the gradual fulfilment of the fundamental principle, from each according to his ability to each according to his work’ (Government of Pakistan, 1973). 17 These studies include Awan and Khan (1992), Ahmad et al. (1998) and Khattak and Sayeed (2000). At an international level the World Bank conducted a study on female homeworkers. 18 Through this plan, the government favoured policies towards the industrialisation of the export sector with emphasis on labour-intensive industries. 19 This network consisted of other South East Asian countries including the Philippines and Indonesia. 20 The government agencies involved are the Ministry of Labour and Social Welfare, Ministry of Interior, Ministry of Agriculture Co-operatives, Ministry of Education, Ministry of Industry, Ministry of Commerce, Employers’ Organisation Council, Industrial Council and Provincial Trade Chambers. 21 Home-based work in Chile was first regulated in 1931 in the Labour Code. However, in 1981 the military government’s ‘Employment Plan’ restricted the regulation of employment relationships to basic aspects and eliminated the provision mentioned. Currently, Chile is one of the few Latin American countries where home-based work is not regulated and is treated as ‘the provision of services governed by civil law’ (Henrìquez et al., 2001). 22 Legislation in Peru evolved over time and homeworkers were even provided with social security rights through the 1980 Health Benefits Regime and the 1986 Peruvian Social Security Institute Pension Regime (Article 50, Constitution 1979). Homeworkers’ labour conditions deteriorated with the exclusion of Article 50 in the last Constitution of 1993. This represented a step backward in terms of homeworkers’ rights. However, thanks to the labour reform of 1991, homeworkers’ rights are still taken into account in Chapter IV of the Legislative Decree No. 728. This chapter sets down the legal framework and includes important provisions such as the requirement of a written agreement, registration and social rights including paid holiday, bonuses and compensation for length of service (Article 94) (Verdera, 2000).

452

Santosh Mehrotra and Mario Biggeri

23 See Mehrotra and Delamonica (2007) for a detailed discussion of the nature of interventions needed in the basic services for enhancing capabilities and realising the synergies we discuss here. 24 Social security is a broader notion than social protection. EUROSTAT (1995) rightly defines social security as: reimbursement and support in kind for health care; income maintenance and social support in case of sickness, disability, old age, death of a family member, family circumstances (pregnancy, childbirth, care for children and other family members) and unemployment; help towards the cost of housing and social support for social exclusion are not elsewhere classified. 25 They provide minimal assistance to poor households in the case of old age (after age 65 at Rs 75 or under $2 per month at current exchange rates), death of the male breadwinner, and maternity (Jain, 1998). 26 Whether this is done through a social insurance mechanism with primarily private provision or through a public health system is a matter of detail outside the scope of this book. For differing views on social insurance, see Gertler and Solon (2000). 27 The construction workers fund is financed by the contributions made by beneficiaries, a levy on all construction works at rates between 1 and 2 per cent of the construction cost incurred by an employer and non-mandatory grant/loans by the central/state government. These funds provide several kinds of similar benefits, though here we dwell specifically on those available to bidi workers. There are public health and sanitation and medical facilities for example, the Fund runs hospitals, a scheme for reservation of beds in tuberculosis hospitals, maternity benefits, and provides spectacles. Group life and disability insurance is provided, the premium for which is equally shared by the Bidi Workers Welfare Fund and the Social Security Fund of the government of India. There are educational schemes for example, financial assistance to purchase uniforms, slates, notebooks and textbooks; scholarships for children from class 5 onwards; a scholarship based on attendance in school by girls; and a lump sum given on passing board exams from class 10 onwards. The proposed national policy on home-based workers formulated by the government of India’s Ministry of labour is advocating the widespread use of such welfare funds. 28 Eventually the kind of facilities that are offered should be the choice of the workers themselves, since the higher the benefits the larger the tax that will have to be levied. Inevitably, the facilities will vary depending upon the country and sector, but we believe that the benefits we list should constitute a minimum. 29 Not general health care, which should be provided by the government health system, as we noted earlier. 30 The administrative costs of funds in India has not been low. It has varied from 7.9 per cent of total benefit expenditure in bidi, to 19.4 per cent in the dolomite mines fund (Subramanya, 2000). The administration of these funds is not integrated. 31 Thus Subramanya (2000) notes that the Ministry of Labour reported that bidi workers’ identity cards have been issued for about 2.7 million workers when the total number of workers is 4.3 million. 32 Some factories which rely on fooling inspectors – and when visits are quick, announced in advance, and conducted by foreigners, it is relatively easy (Oxfam, 2004a, pp. 62–63 for examples). 33 ‘I spent three years getting up to compliance with the SA 8000 standard’, said one factory owner in Thailand, ‘and then the customer who had asked for it in the first place left and went to China’ (quoted in Oxfam, 2004b, p. 57).

References

Ahmad, N., Qaisrani, S. and Tahir, M. (1998), Social Protection for Women Workers in the Informal Home-based Sector in the Leather and Textile Industries, Lahore: Aurat Publication and Information Service Foundation. Alesina, A. and Perotti, R. (1994), ‘The Political Economy of Growth: A Critical Survey of the Recent Literature’, World Bank Economic Review, vol. 8 (3), September, pp. 351–371. Alkire, S. (2002), Valuing Freedoms. Sen’s Capability Approach and Poverty Reduction, Oxford: Oxford University Press. Altenburg, T. and Meyer-Stamer, J. (1999), ‘How to Promote Clusters: Policy Experiences from Latin America’, World Development, vol. 27, no. 9, pp. 1693–1713. Amsden, A. (1989), Asia’s Next Giant, New York: Oxford University Press. Arunotai, N., Gordon, N., Hiranvorachat, P., Jarubenja, R., Katleeradapan, N., Petchprasert, N. and Pongsapich, A. (2001), Outsourcing of Manufacturing to Households: Subcontracted Home Based Work in Thailand, Country report, UNICEF Innocenti Research Centre, Florence (unpublished paper). Awan, S. A. and Khan, A. A. (1992), Child Labour in the Carpet Weaving Industry in Punjab, Lahore: UNICEF, Pakistan. Baden, S. (2001), Researching Homework and Value Chains in The Global Garments Industry: An Annotated Resource List and Binder, Prepared for Women in Informal Employment Globalizing and Organizing (WIEGO), Global Markets Program, January 2001. Bagnasco, A. (ed.) (1986), L’altra metà dell’economia: la ricerca internazionale sull’economia informale, Liguori, Napoli. Bagnoli, L. (2004), Quale responsabilità sociale per l’impresa?, Milano: FrancoAngeli. Bair, J. and Gereffi, G. (2001), ‘Local Clusters in Global Value Chains: The Causes and Consequences of Export Dynamism in Torreon’s Blue Jeans Industry’, World Development, vol. 29, no. 11, pp. 1885–1903. Bairoch, P. and Kozul-Wright, R. (1996), ‘Globalization Myths: Some Historical Reflections on Integration, Industrialization and Growth in the World Economy’, UNCTAD Discussion Papers No. 13, March, Geneva. Bajaj, M. (2000), ‘Invisible Workers, Visible Contributions: A Study of Homebased Women Workers in Five Sectors across South Asia’, Background Paper for Regional Policy Seminar, Nepal. Balakrishnan, R. (2002), The Hidden Assembly Line. Gender Dynamics of Subcontracted Work in a Global Economy, Bloom Field, USA: Kumarian Press, Inc. Balakrishnan, R. and Huang, M. (2000), ‘Flexible Workers-Hidden Employers: Gender and Subcontracting in the Global Economy’, Washington, DC: The Asian Foundation.

454

References

Bangasser, P. E. (2000), ‘The ILO and the Informal Sector: An Institutional History’, Employment Paper No. 2000/9, Employment Sector, International Labour Office, ILO, Geneva. Barrientos, S., Bee, A., Matear, A. and Vogel, I. (1999), Women and Agribusiness: Working Miracles in the Chilean Fruit Export Sector, London: Macmillan. Basu, K. (1999), ‘Child Labor: Cause, Consequence and Cure, with Remarks on International Labor Standards’, Journal of Economic Literature, vol. 37, pp. 1083–1119. Basu, K. and Van, P. H. (1998), ‘The Economics of Child Labour’, American Economic Review, vol. 83, pp. 412–27. Becattini, G. (1987), Mercato e forze locali: il distretto industriale, Bologna, Il Mulino. Becattini, G. (1990), ‘The Marshallian Industrial District as a Socio-Economic Notion’, in Pyke, F., Becattini, G. and Sengenberger, W. (eds), Industrial Districts and Inter-Firm Co-operation in Italy, Geneva: International Institute for Labour Studies. Becattini, G. (2000), Il distretto industriale. Un nuovo modo di interpretare il cambiamento economico, Torino: Rosenberg & Sellier. Becattini, G. (2004), Industrial Districts: A New Approach to Industrial Change, Cheltenham, UK: Edward Elgar Publishing Limited. Becattini, G., Bellandi, M., Dei Ottati, G. and Sforzi, F. (eds) (2003), From Industrial Districts to Local Development: An Itinerary Research, Cheltenham, UK: Edward Elgar Publishing Limited. Bellandi, M. (2004), ‘Chapter 5’, in Bellandi, M. and Di Tommaso, M. R. (eds), Industry, Clusters and Industrial Policy in Guangdong and their linkages with Some Selected Italian Industrial Regions, Report. L’institute, Institute for Industrial Development Policy. Bellandi, M. and Biggeri, M. (eds) (2005), La sfida industriale cinese vista dalla Toscana distrettuale, L’institute, Toscana Promozione, Consorzio Ferrara Ricerche, Firenze. Bellandi, M. and Russo, M. (eds) (1994), Distretti industriali e cambiamento economico locale, Torino: Rosemberg & Sellier. Bellandi, M. and Sforzi, F. (2003), ‘Multiple Path of Local Development’, in Becattini, G., Bellandi, M., Dei Ottati, G. and Sforzi, F. (Eds) (2003), From Industrial Districts to Local Development: An Itinerary Research, Edward Elgar Publishing Limited, Cheltenham, UK, pp. 210–226. Bellandi, M., Di Tommaso, M. and Rubini, L. (2006), ‘Aprendiendo del debate internacional sobre clusters y distritos: variables relevantes para la investigación empírica en China’, chap. 8 in Eleonora Di Maria and Stefano Micelli (editores) Sistemas locales en las redes internacionales de producción, Roma: Aracne, pp. 233–269. Benerìa, L. and Rodàn, M. (1987), The Crossroads of Class and Gender: Industrial Homework, Subcontracting, and Household Dynamics in Mexico City, Chicago and London: The University of Chicago Press. Benton, L. A. (1989), ‘Homework and Industrial Development: Gender Roles and Restructuring in the Spanish Shoe Industry’, World Development, vol. 17, no. 2, pp. 255–266. Bhatt, G. (2000), ‘Organizing Knowledge in the Knowledge Development Cycle’, Journal of Knowledge Management, vol. 4 (1), pp. 15–26. Bianchi, P., Di Tommaso, R. M. and Rubini, L. (eds) (2000), Le api audaci. Piccole imprese e dinamiche industriali in estremo oriente, Milano: FrancoAngeli. Biggeri, M. (2003a), ‘Children, Child Labour and the Capabilities Approach’, paper presented at the 3rd Conference on the Capability Approach: from Sustainable Development to Sustainable Freedom, 7–9 September 2003, Pavia University. Biggeri, M. (2003b), Report Nepal 2003, Associazione RAM, mimeo.

References 455 Biggeri, M. (2004), ‘Growth with Development: Informal Sector and Human Development in Low-Income Sub-Saharan Economies’, Studi e discussioni, No. 140, Dipartimento di Scienze Economiche, Università degli Studi di Firenze. Biggeri, M. (2005), ‘China in Perspective: From Economic “Miracle” to Human Development?’, paper presented at International Annual Conference on Development and Change Promoting Development in a Globalized World, ACDC, Neemrana, India, December. Biggeri, M., Gambelli, D. and Phillips, C. (1999), ‘Small and Medium Enterprise Theory: Evidence for Chinese TVEs’, Journal of International Development, vol. II, no. 2, March–April, pp. 197–219. Biggeri, M., Guarcello, L., Lyon, S. and Rosati, F. C. (2003), ‘The Puzzle of “idle” Children: Neither in School nor Performing Economic Activity: Evidence from Six Countries’, UCW WP, Understanding Children’s Work (UCW), Project, ILO/UNICEF/WORLD BANK. Biggeri, M., Libanora, R., Mariani, S. and Menchini, L. (2006), ‘Children Conceptualizing Their Capabilities: Results of the Survey During the First Children’s World Congress on Child Labour’, Journal of Human Development, vol. 7, no. 1, March, pp. 59–83. Boiral, O. (2003), ‘The Certification of Corporate Conduct: Issues and Prospects’, International Labour Review, vol. 139, no. 3, pp. 317–340. Boris, E. and Prugl, E. (1996), ‘Home-Based Producers in Development Discourse’, in Boris, E. and Prugl, E. (eds), Homeworkers in Global Perspective, New York: Routledge. Breman, J. (1996), Footloose Labour: Working in India’s Informal Economy, Cambridge, UK: Cambridge University Press. Broersma, L., den Butter, F. and Kock, U. (2003), ‘A Cointegration Model for Search Equilibrium Wage Formation’, Tinbergen Institute Discussion Papers 03-088/3, Tinbergen Institute. Brusco, S. (1990), ‘The Idea of Industrialised Districts: Its Genesis’, in Pyke, F., Becattini, G. and Sengenberger, W. (eds), Industrial Districts and Inter-Firm Co-operation in Italy, Geneva: International Institute for Labour Studies. Cainelli, G. and Zoboli, R. (eds) (2004), The Evolution of Industrial Districts: Change in Governance, Innovation and Internalisation of Local Capital in Italy, Heidelberg Physica Verlag. Canagarajah, S. and Sethuraman, S. V. (2001), ‘Social Protection and the Informal Sector in Developing Countries: Challenges and Opportunities’, Social Protection Discussion, Paper No. 0130, World Bank. Carr, M. and Chen, M. A. (1999), ‘Homebased Women Workers in South Asia: Creating an Enabling Environment’, Background Paper for Regional Policy Seminar. Carr, M. and Chen, M. A. (2001), Globalization and the Informal Economy: How Global Trade and Investment Impact on the Working Poor, WIEGO, May 2001. Carr, M. and Chen, M. A., (2002), ‘Globalisation and the Informal economy: how global trade and investment impact on the working poor’, Working paper on the informal economy, Employment Sector, ILO, 2002/1. International Labour Office Geneva. Carr, M., Chen, M. A. and Tate, J. (2000), ‘Globalization and Homebased Workers’, Feminist Economics, vol. 6, no. 3, pp. 123–142. Cawthorne, P. (1995), ‘Of Networks and Markets: The Rise and Rise of a South Indian Town, the Example of Tirupur’s Cotton Knitwear Industry’, World Development, vol. 23, no. 1, pp. 43–56. Chang, C., and Wang, Y. (1994). ‘The Nature of the Township-Village Enterprise’, Journal of Comparative Economics, vol. 19 (3), pp. 434–452.

456

References

Chang Ha-Joon (2002), Kicking Away the Ladder: Development Strategy in Historical Perspective, London: Anthem Press. Chang Ha-Joon (ed.) (2003), Rethinking Development Economics, London: Anthem Press. Chang Ha-Joon and Grabel, I. (2004), Reclaming Development: An Alternative Economic Policy Manual, Global Issues, London: Zed Books. Chang Ha-Joon and Green, D. (2003), ‘The Northern Agenda on Investment: Do as We say, Not as We Did’, South Centre, CAFOD, Gerafer. Charmes, J. (1998), Women Working in the Informal Sector in Africa: New Methods and New Data, Paris: ORSTOM. Charmes, J. (2000), ‘Size, Trends and Productivity of Women’s Work in the Informal Sector and in Old and New Forms of Informal Employment: An Outlook of Recent Empirical Evidence’, Saint Quentin en Yvelines, France, Centre of Economics and Ethics for Environment and Development (C3ED). Charmes, J. (2001), ‘Informal Sector, Poverty and Gender: A Review of Empirical Evidence’, Background Paper for the World Development Report 2001, University of Versailles-St. Quentin en Yvelines. Charmes, J. (2002), ‘Is Asking for “Place of Work” a Pertinent and Efficient Way to Better Measure and Understand the Category of Homeworkers, and More Generally, Outworkers in the Labour Force?’, in Measuring Place of Work, Geneva: ILO. Chen, M. A. (2005), ‘Rethinking the Informal Economy: Linkages with the Formal Economy and the Formal Regulatory Environment’, WIDER Research Paper, No. 2005/10, April. Chen, M. A. and Carr, M. (2002), ‘Globalization and the Informal Economy: How Global Trade and Investment Impact on the Working Poor’, Working Paper on the Informal Economy, No. 1. Geneva, ILO. Chen, M. A., Sebstad, J. and O’Connell, L. (1999), ‘Counting the Invisible Workforce: The Case of Homebased Workers’, World Development, vol. 27, no. 3, pp. 603–610. Chen, M. A., Jhabvala, R. and Lund, F. (2002), ‘Supporting Workers in the Informal Economy: A Policy Framework’, Working Paper on the Informal Economy, No. 2. Geneva, ILO. Chen, M. A., Jhabvala, R., Kanbur, R., Mirani, N., Osner, K. and Richards, C. (eds) (2005), Membership Based Organizations of the Poor: Reflections After an Exposure and Dialogue Program with SEWA in Gujarat, India, January 2005. Chernichovsky, D. (1982), ‘Fertility Behaviour in Developing Countries: An Investment Approach’, Reasearch in Population Economics, vol. 4, pp. 253–285. Cigno, A. and Rosati, F. C. (1992), ‘The Effects of Financial Markets and Social Security on Saving and Fertility Behaviour in Italy’, Journal of Population Economics, vol. 5, pp. 319–341. Cigno, A. and Rosati, F. C. (1996), ‘Jointly Determined Saving and Fertility Behaviour: Theory, and Estimates for Germany, Italy, UK, and USA’, European Economic Review, vol. 40, pp.1561–1589. Cigno, A. and Rosati, F. C. (2002), ‘Child Labor, Education, Fertility and Survival in Rural India’, Pacific Economic Review, vol. 7, pp. 1–15. Cigno, A., Rosati, F. C. and Tzannatos, Z. (2002), Child Labour Handbook, SP Discussion Paper No. 206, The World Bank. Cigognetti, L. and Pezzini, M. (1994), ‘Dalla lavorazione delle paglie all’industria delle maglie: la nascita del distretto industriale di Carpi’, in Bellandi, M. and Russo, M. (eds), Distretti industriali e cambiamento economico locale. A cura di, Torino: Rosemberg & Seller.

References 457 Cornia, G. A. (2001), ‘Social Funds in Stabilization and Adjustment Programmes: A Critique’, Development and Change, vol. 32, pp. 1–32, Institute of Social Studies, Oxford: Blackwell Publishers. Cornia, G. A. (2003), ‘Pace, Source, Components and Volatility of Change in the Modern Era, and its Relation to Behaviour and Health’, paper presented at the seminar on ‘Health and Social Upheavals’, 13 and 14 March 2003, University of Florence. Cornia, G. A. (ed.) (2004), Inequality, Growth, and Poverty in an Era of Liberalization and Globalization, WIDER Studies in Development Economics, Oxford University Press. Cornia, G. A. (2006), Pro-Poor Macroeconomics, Potential and Limitations, London: Palgrave Crestanello, P. (1996), ‘The Industrial Districts in Veneto: Changes and Tendencies’, in Cossentino, F., Pyke and Sengenberger, W. (eds), Local and Regional Response to Global Pressure: The Case of Italy and Its Industrial Districts, Geneva: International Institute for Labour Studies, ILO, pp. 67–96. Cunningham, H. and Viazzo, P. P. (1996), Child Labour in Historical Perspective 1800–1985. Case Studies from Europe, Japan and Colombia. UNICEF Innocenti Research Centre, Florence. Dasgupta, S. (2001), Employment Security: Conceptual and Statistical Issues, Geneva: ILO. Dei Ottati, G. (2002), ‘Social Concertation and Local Development: The Case of Industrial Districts’, European Planning Studies, vol. 10, no. 4, pp. 449–466. Delahanty, J. (1999), ‘A Common Thread: Issues for Women Workers in the Garment Sector’, Ottawa: The North-South Institute. www.wiego.org/publi4.ssi, Department for International Development. 2000. Eliminating World Poverty: Making Globalisation Work for the Poor. White Paper. www.dfid.gov.uk/ Delamonica, E., Mehrotra, S. and Vandemoortele, J. (2001), ‘Is EFA Affordable? Estimating the Global Minimum Cost of Education for All’, Innocenti Working Paper No. 87, UNICEF, Innocenti Research Centre, Florence. Delgado, C. L., Hopkins, J. and Kelly, V. A. (1998), ‘Agricultural Growth Linkages in SubSaharan Africa’, Research Report, No. 107, International Food Policy Research Institute. De Soto, H. (1989), The Other Path: The Invisible Revolution in the Third World, New York: Harper & Row Publishers. Di Tommaso, M. R. and Bellandi, M. (2006), Il Fiume delle Perle: Dimensione locale dello sviluppo industriale cinese e il confronto con l’Italia, Torino: Rosenberg & Sellier. Di Tommaso, M. R. and Rabellotti, R. (eds) (1999), Efficienza collettiva e sistemi d’imprese, Bologna: Il Mulino. Dini, M., Guaipatín, C. and Rivas, G. (2005), ‘Promoting Productive Integration in Latin America and the Caribbean: Suggestions for Project Preparation’, Technical Paper Series, Sustainable Development Department, Inter-American Development Bank, Washington, DC. Domar, D. (1941), ‘Capital Expansion, Rate of Growth and Employment’, Econometria, vol. 1, April, pp. 137–147. Entwisle, B. and Winegarden, C. R. (1984), ‘Fertility and Pension Programs in LDCs: A Model of Mutual Reinforcement’, Economic Development and Cultural Change, vol. 32, pp. 331–354. EUROSTAT (1995), ESSPROS Manual 1995, vol. 1, Luxembourg (unpublished). Evans, P. (1996), ‘Government Action, Social Capital and Development: Reviewing the Evidence on Synergy’, World Development, vol. 24 , no. 6, pp. 1119–1132. Fabbris, L. (1990), ‘Problemi statistici nella utilizzazione di dati rilevati presso testimoni priviligiati’, Atti del Seminario di studio, Rilevazioni per campione delle opinioni degli italiani, Bressanone, 13 settembre, pp. 89–115.

458

References

FAO (1998), The State of Food and Agriculture, Rome: FAO. Felstead, A. and Jewson, N. (2000), In Work, at Home, New York: Routledge. Ferran, L. (1998), ‘Note on Concepts and Classifications to Improve Statistics on Homebased Workers’, New York: United Nations Statistics Division. Presented at the Delhi Group Meeting on Informal Sector Statistics, Ankara, 28–30 April, 1998. Findlay, C., Watson, A. and Wu Harry, X. (eds) (1994), Rural Enterprises in China, London: Macmillan. Fusaku, K., Wall, D. and Wu, M. (1994), China’s Long March to an Open Economy, OECD, Development Centre of the Organisation for Economic Co-operation and Development, Paris. Gammage, R. and Mehra, S. (1999), ‘Trends, Countertrends, and Gaps in Women’s Employment’, World Development, vol. 27, no. 3, pp. 533–550. Gereffi, G. (1994a), ‘Capitalism, Development, and Global Commodity Chains’, in Sklair, Leslie (ed.), Capitalism and Development, London: Routledge, pp. 211–231. Gereffi, G. (1994b), ‘The Organization of Buyer-Driven Global Commodity Chains: How U. S. Retailers Shape Overseas Production Networks’, in Gereffi, G. and Korzeniewicz (eds), Commodity Chains and Global Capitalism, London: Praeger, pp. 95–122. Gereffi, G. (1999), ‘International Trade and Industrial Upgrading in the Apparel Commodity Chain’, Journal of International Economics, vol. 48, no. 1, pp. 37–70. Gereffi, G. and Kaplinsky, R. (2001), ‘The Value of Value Chains’, Special issue of IDS Bulletin, vol. 32. Gertler, P. and Solon, O. (2000), ‘Who Benefits from Social Health Insurance in Developing Countries?’, faculty.haas.berkeley.edu/gertler/working_papers/Philippines6.pdf, and Johannes Jutting, Health Insurance in Developing countries? www.zef.de/ zefnews/No7-5-2001-engl.pdf Giuliani, E., Pietrobelli, C. and Rabellotti, R. (2004), Upgrading in Global Value Chains: Lessons from Latin American Clusters, WP9/2004, PRIN. Giuliani, E., Pietrobelli, C. and Rabellotti, R. (2005), ‘Upgrading in Global Value Chains: Lessons from Latin American Clusters’, World Development, vol. 33, no. 4, pp. 527–547. Government of Pakistan (1973), The Constitution of Pakistan, Lahore: Publishers Emporium. Government of Pakistan (1998), Labour Force Survey, Islamabad: Federal Bureau of Statistics, Statistics Division. Granovetter, M. (1973), ‘The Strength of Weak Ties’, American Journal of Sociology, vol. 78, pp. 1360–1380. Gulati, M. (1997), ‘Restructuring and Modernisation of Small and Medium Enterprise Clusters in India’, Report, Small and Medium Enterprise Programme, UNIDO, Vienna. Haggblade, S., Liedhom, C. and Mead, D. (1990), ‘The Effect of Policy and Policy Reforms on Non-Agricultural Enterprises and Employment in Developing Countries: A Review of Past Experience’, in Stewart, F., Thomas, H. and de Wilde, T. (eds), The Other Policy: The Influence of Policies on Technology Choice and Small Enterprise Development, London: IT Publications. Harris, J. R. and Todaro, M. P. (1970), ‘Migration, Unemployment and Development: A Two-Sector Analysis’, American Economic Review, vol. 60, March, pp. 126–142. Harris-White, B. (2003), India Working: Essays on Society and Economy, Cambridge: Cambridge University Press. Harrod, R. F. (1939), ‘An Essay in Dynamic Theory’, Economic Journal, vol. 49, March, pp. 14–33.

References 459 Heady, C. (2000), ‘What is the Effect of Child Labour on Learning Achievement? Evidence from Ghana’, Innocenti Working Paper, No. 79, UNICEF Innocenti Research Centre, Florence. Heikel, M. V. (2000), ‘Home Workers in Paraguay’, SEED Working Paper No. 2, Series on Homeworkers in the Global Economy, Geneva: ILO. Henriquez, H., Riquelme, V., Gálvez, T. and Selamé, T. (2000), ‘Home Work in Chile: Past and Present Results of a National Survey’, SEED Working Paper No. 8, Series on Homeworkers in the Global Economy, Geneva: ILO. Hirsch, G. (2005), Internal Labour Migration in China: Social and Economic Implications for the Guangdong Province, Universita’ degli Studi di Firenze, Tesi di dottorato di Ricerca in Politica ed Economia dei Paesi in Via di Sviluppo. Holzmann, R. and Jorgensen, S. (1999), ‘Social Risk Management: A New Conceptual Framework for Social Protection and Beyond’, SP Discussion Paper No. 06, Human Development Network, Social Protection, Washington, DC: World Bank. Homenet (1998), ‘Organising for Rights: Special Issue on Consumer Campaigns and Homework’, Leeds, Homenet International. Homenet (1999a), http://www.homenetww.org.uk/pubs.html Homenet (1999b), ‘New Ways of Organising in the Informal Sector: Four Case Studies of Trade Union Activity’, Leeds, Homenet International. Homenet (1999c), ‘Using the ILO Convention on Homework’, Leeds, Homenet International. Homenet (2000), ‘Special Issue: Action Research on Homebased Work’, Leeds, Homenet International. Homenet Thailand (2002), ‘Impact of the Economic Crisis on Homeworkers in Thailand’, published by Homenet Thailand with support from the International Labour Office, Bangkok Office and East Asia Multidisciplinary Advisory Team, January. Horn, P. (2004), ‘Rethinking the Informal Economy: Linkages with the Formal Economy and the Formal Regulatory Environment’, paper presented at the EGDI and UNUWIDER Conference, Unlocking Human Potential: Linking the Informal and Formal Sectors, 17–18 September 2004, Helsinki , Finland. Humphrey, J. (2003), ‘Opportunities for SMEs in Developing Countries to Upgrade in a Global Economy’, SEED Working Paper, No. 43, Geneva: ILO. Humphrey, J. and Schmitz, H. (1995), Principles for Promoting Clusters and Networks of SMEs, Vienna: UNIDO. Humphrey, J. and Schmitz, H. (1996), ‘The Triple C Approach to Local Industrial Development’, World Development, vol. 24, no. 12, pp. 1859–1877. Humphrey, J. and Schmitz, H. (2000), ‘Global Governance and Upgrading: Linking Industrial Cluster and Global Value Chain Research’, IDS Working Paper 120, Institute of Development Studies, Brighton. Hussmanns, R. (1997), Informal Sector: Statistical Definition and Survey Methods, paper presented to the United Nations Expert Group Meeting on Household Satellite Accounting, New York. Huws, U. (1995), ‘Introduction. Action Programmes for the Protection of Homeworkers: Ten Case Studies from Around the World’, Geneva, International Labour Organization, pp. 1–13. IIFT (1999), Data on Exports, Indian Institute of Foreign Trade, IIFT, New Delhi, India: IIFT. ILO (1991), The Dilemma of the Informal Sector, Report of the Director-General (Part 1) to the International Labour Conference 78th Session 1991, Geneva: ILO.

460

References

ILO (1993), ‘From the Shadows to the Fore: Practical Actions for the Social Protection of Homeworkers’, 3 Vols, Indonesia, the Philippines, Thailand, Bangkok, International Labour Organization, Regional Office for Asia and the Pacific. ILO (1996a), Convention on Home Work No. 177, Geneva: ILO. ILO (1996b), ‘Out of the Shadows: Practical Actions for the Protection of Homeworkers’, 3 Vols, Indonesia, the Philippines, Thailand, Bangkok, International Labour Organization, Regional Office for Asia and the Pacific. ILO, (1997), African Employment Report 1995, East Africa Multidisciplinary Advisory Team, Addis Ababa: ILO. ILO (1999), ‘Job Quality and Small Enterprise Development’, SEED Working Paper, n 4, Series on Job Quality in Micro and Small Enterprise Development, Geneva: ILO. ILO (2000), ‘Labour Practices in the Footwear, Leather, Textiles and Clothing Industries’, Geneva, International Labour Organization, Sectoral Activities Programme. ILO (2002a), Women and Men in the Informal Economy: A Statistical Picture, Geneva: ILO. ILO (2002b), Every Child Counts: New Global Estimates on Child Labour, Geneva, International Labour Office on the internet: http://www.ilo.org/public/english/standards/ ipec/simpoc/ ILO (2002c), Combating Child Labour: A Handbook for Labour Inspectors, International Programme on the Elimination of Child Labour (IPEC), InFocus Programme on Safety and Health at Work and the Environment (SafeWork), International Association of Labour Inspection (IALI), Geneva: ILO. ILO (2003) Investing in Every Child: An Economic Study of the Costs and Benefits of Eliminating Child Labour, Geneva, International Labour Office, on the internet: http:// www.ilo.org/public/english/standards/ipec/simpoc ILO (2004a), Economic Security for a Better World, Geneva: ILO. ILO (2004b), Child Labour: A Textbook for University Students, Geneva: ILO. ILO (2005), Business Development Services, Strategy Note, Geneva: ILO. Indonesia Sakernas Series (1999), Departemen Pendidikan dan Kebudayaan, Badan Penelitian dan Pengembangan Pendidikan dan Kebudayaan, Pusat Informatika (1999), Rangkuman Statistik Persekolahan. International Standard SAI (2001), Social Accountability 8000, Social Accountability International. International System of National Accounts (1993), SNA. International Conference of Labour Statisticians (ICLS). Jain, S. (1998), ‘Area-Based Social Insurance for the Unorganied Sector’, in van Ginneken, W. (ed.), Social Security for All Indian, Delhi: OUP. Jelin, H., Mercado, M. and Wyczykier, G. (2000), ‘Home Work in Argentina’, SEED Working Paper No.7, Series on Homeworkers in the Global Economy, Geneva, ILO. Jenkins, M. (1993), ‘Extending Social Protection to the Entire Population: Problems and Issues’, International Social Security Review, vol. 46, no. 2, pp. 3–20, Geneva: International Social Security Association. Jensen, E. R. (1990), ‘An Econometric Analysis of the Old-Age Security Motive for Childbearing’, International Economic Review, vol. 31, pp. 953–968. Jhabwala, R. and Subramanya, R. K. (2000), The Unorganised Sector, New Delhi, India: Sage. Joekes, S. (1999), ‘A Gender-Analytical Perspective on Trade and Sustainable Development’, in UNCTAD, Trade, Sustainable Development and Gender, New York and Geneva: United Nations. Kantor, P. (2000), ‘Estimating the Number and Economic Contribution of Homebased Garment Producers in Ahmedabad, India’, Working Paper No. 75, National Council for Applied Economic Research, New Delhi, India.

References 461 Kaplinsky, R., (2000), ‘Spreading the Gains from Globalisation: What Can Be Learned From Value Chain Analysis?’, Brighton, Institute of Development Studies. Kaplinsky, R. and Readman, J. (2001), Integrating SMEs in Global Value Chains: Towards Partnership for Development, Vienna: UNIDO. Kaplinsky, R., Morris, M. and Readman, J. (2001), ‘A Handbook for Value Chain Research’, mimeo. IDRC Institute of Development Studies, University of Sussex; School of Development Studies, University of Natal. Kennedy, L. (1999), ‘Co-operating for Survival: Tannery Pollution and Joint Action in the Palar Valley, India’, World Development, vol. 27, no. 9, pp. 1673–1692. Khan, S., Khattak, S. G. and Kazmi, S. (2001), Hazardous Home based Subcontracted Work in Pakistan: Study of Multi-Tiered Exploitation, Country report, UNICEF Innocenti Research Centre, Florence (unpublished paper). Khattak, S. and Sayeed, A. (2000), ‘Subcontracted Women Workers in the World Economy: The Case of Pakistan’, SDPI Monograph Series No.15, Islamabad. Khor, M. (2000), Globalisation and the South, TWN, Third World Network, Malaysia: Jutaprint. Khor, M. (2002), The WTO, the Post-Doha Agenda and the Future of the Trade System: A Development Perspective, TWN, Third World Network, Malaysia: Jutaprint. Knorringa, P. (1999), ‘Agra: An Old Cluster Facing the New Competition’, World Development, vol. 27, no. 9, 1587–1604. Korpi, T. and Mertens, A. (2004), ‘Training and Industrial Restructuring – Structural Change and Labour Mobility in West Germany and Sweden’, International Journal of Manpower, vol. 25 (1), pp. 90–103. Krishnamoorthy, V. and Nair, R. P. (2001), ‘The Welfare Fund for Construction Workers in Tamil Nadu’, ESS Paper No. 18, Social Security Policy and Development Branch, Geneva, ILO. Lavinas, L., Sorj, B., Linhares, L. and Jorge, A. (2000), ‘Home Work in Brazil: New Contractual Arrangements’, SEED Working Paper No.7, Series on Homeworkers in the Global Economy, Geneva: ILO. Levitsky, J. (1996) ‘Support Systems for SME’s in Developing Countries – A Review’, Paper Commissioned by the Small and Medium Industries Branch, Vienna, Austria: UNIDO. Lewis, W. A. (1954), ‘Economic Development with Unlimited Supplies of Labour’, The Manchester School of Economic and Social Studies, vol. 12, no. 2, pp. 139–191. Lewis, W. A. (1977), ‘On Economics, Agriculture, and the Political Economy’, Elmhirst Memorial Lecture, Nairobi, Kenya, and also in Dams, Theodor et al. (eds), International Association of Agricultural Economists, Decision-Making and Agriculture, Oxford, UK: Alden Press. Lieten, G. K. (ed.) (2004a), Working Children Around the World: Child Rights and Child Reality, IREWOC, Institute for Human Development, New Delhi, India: Institute for Human Development. Lieten, G. K. (ed.) (2004b), The Child Labour Problem: Issues and Solutions, IREWOC/ DCI, Amsterdam, The Netherlands: IREWOC. Lieten, G. K., Srivastava, R. and Thorat, S. (eds) (2004), Small Hands in South Asia. Child Labour in Perspective, New Delhi, India: IDPAD and Manohar Publishers and Distributors. Lindert, P. (2003), Growing Public Social Spending and Economic Growth Since the Eighteenth Century, Cambridge, UK: Cambridge University Press. Lora Jo Foo (2002), ‘Labour Law and the Informal Economy Working Gp Mtg’, Geneva, 15–16 June 2002, Meeting Report. Lund, F. and Srinivas, S. (2000), Learning from Experience: A Gender Approach to Social Protection for Workers in the Informal Economy, Social Security Department, ILO, and Women in Informal Employment Globalising and Organising (WIEGO).

462

References

Marshall, A. (1920), Principles of Economics, London: Macmillan. McCormick, D. (1999), ‘African Enterprise Clusters and Industrialisation: Theory and Reality’, World Development, vol. 27, no. 9, pp. 1531–1552. McCormick, D. (2000), ‘Value Chains, Production Networks and the Business System: Preliminary Thoughts with Examples from Kenya’s Textiles and Garment Industry’, Nairobi, Institute of Development Studies, University of Nairobi. McCormick, D. and Schmitz, H. (2002), Manual for Value Chain Research on Homeworkers in the Garment Industry, Institute of Development Studies, WIEGO, Sussex, England. McLaughlin, S. (1990), ‘Skill Training for the Informal Sector: Analysing the Success and Limitations of Support Programmes’, in Turnham, D., Salomé, B. and Schwarz, A. (eds), The Informal Sector Revisited, Development Centre Seminars, Paris: OECD. Mead, D. C. and Liedholm, C. (1998), ‘The Dynamics of Micro and Small Enterprises in Developing Countries’, World Development, vol. 26, no.1, pp. 61–76. Mehrotra, S. (2001), ‘The Rhetoric of International Development Targets and the Reality of Official Development Assistance’, Innocenti Working Paper No. 85, UNICEF Innocenti Research Centre, Florence. www.unicef.org/irc Mehrotra, S. (2006) ‘Governance and Basic Services: Ensuring Accountability in Service Delivery through Deep Democratic Decentralization,’Journal of International Development, vol. 18, no. 3. Mehrotra, S. and Biggeri, M. (2002a), ‘The Subterranean Child Labour Force: Subcontracted Home Based Manufacturing in Asia’, Innocenti Working Paper, No. 96, UNICEF Innocenti Research Centre, Florence. www.unicef.org/irc Mehrotra, S. and Biggeri, M. (2002b), ‘Social Protection in the Informal Economy: Homebased Women Workers and Outsourced Manufacturing in Asia’, Innocenti Working Paper, No. 97, UNICEF Innocenti Research Centre, Florence. www. unicef.org/irc Mehrotra, S. and Delamonica, E. (2007) Eliminating Human Poverty: Macro Economic and Social Policies for Equitable Growth, London: Zed Press. Mehrotra, S. and Jolly, R. (eds), (1997), Development with a Human Face. Experiences in Social Achievement and Economic Growth, Oxford: Clarendon Press. Mehrotra, S., Panchamukhi, P., Srivastava, Ranjana and Srivastava, R. (2005), Universalising Elementary Education in India: Uncaging the Tiger Economy, New Delhi: Oxford University Press. Meyer-Stamer, J. (2003), ‘Participatory Appraisal of Competitive Advantage (PACA): Launching Local Economic Development Initiatives’, Mesopartner Working Paper, No. 01/2003. Nadvi, K. (1999), ‘Collective Efficiency and Collective Failure: The Response of the Sialkot Surgical Instrument Cluster to Global Quality Pressures’, World Development, vol. 27, no. 9, pp. 1605–1626. Nadvi, K. (2003), ‘The Effect of Global Standards on Local Producers: A Pakistani case study’, in Hubert Schmitz (ed), Local Enterprises in the Global Economy: Issues of Governance and Upgrading, Cheltenham: Edward Elgar Press. Nadvi, K. and Barrientos, S. (2004), Industrial Clusters And Poverty Reduction: Towards a Methodology for Poverty and Social Impact Assessment of Cluster Development Initiatives, Vienna: UNIDO. Nayyar Deepak (2003), ‘Globalisation and Development’, in Chang Ha-Joon (ed.), Rethinking Development Economics, London: Anthem Press, pp. 61–83. Nazfiger, E. W., Stewart, F. and Vayrynen, R. (2000), War, Hunger and Displacement, The Origins of Humanitarian Emergencies, Oxford: Oxford University Press.

References 463 Nolan, P. (2003), ‘Industrial Policy in the Early 21st Century: The Challenge of the Global Business Revolution’, in Chang Ha-Joon (ed.), Rethinking Development Economics, London: Anthem Press, pp. 299–323. NSSO, (2000), ‘Non Agricultural Enterprises in the Informal Sector in India 1999–2000’, NSS 55th Round (July 1999–June 2000), National Sample Survey Organization, Ministry of Statistics and Programme Implementation, New Delhi: Government of India. Nugent, J. and Gillaspy, T. (1983), ‘Old Age Pension and Fertility in Rural Areas of Less Developed Countries: Some Evidence from Mexico’, Economic Development & Cultural Change, vol. 31 (4), pp. 809–829. Nussbaum, M. (2000), Women and Human Development. The Capabilities Approach, Cambridge: Cambridge University Press. Oey-Gardiner, M., Suleeman, E., Brodjonegoro, B., Tjandraningsih, I., Hartanto, W. and Wijaya, H. (2001), Women and Children Home based Workers in Selected Sectors of Indonesia, Country report, UNICEF Innocenti Research Centre, Florence (unpublished paper). Omiccioli, M. and Quintiliani, F. (2000), ‘Assetti imprenditoriali, organizzazione del lavoro e mobilità nei distretti industriali’, in Signorini, L. F. (a cura di), Lo sviluppo locale. Un’indagine della banca d’Italia sui distretti industriali, Rome: Donzelli Editore, pp. 339–358. Omiccioli, M. and Quintiliani, F. (2004), ‘Entrepreneurship, Labour Organisation and Labour Mobility in Industrial Districts’, in Cainelli, Giulio and Zoboli, Roberto (eds), The Evolution of Industrial Districts: Change in Governance, Innovation and Internalisation of Local Capital in Italy, Heidelberg: Physica Verlag, pp. 198–207. Oxfam (2004a), Trading Away Our Rights: Women Working in Global Supply Chains, Oxfam International, Oxford. Oxfam (2004b), ‘Made at Home: British Homeworkers in Global Supply Chains’, Oxfam Briefing Paper. Pedrero, M. (1998), Homeworkers and the Home-Based Workers, New York: United Nations Statistics Division. Originally presented at the Delhi Group Meeting on Informal Sector Statistics, Ankara, 28–30 April, 1998. Perulli, P. (1989), ‘Il distretto industriale di Modena’, in Regini, M. and Sabel, C. F. (a cura di), Strategie di impresa e riaggiustamento industriale, Bologna: Il Mulino, pp. 249–282. Picò, J. and Sanchis, E. (1986), ‘L’Economia sommersa in Spagna’, in Bagnasco, A. (ed.), L’ altra metà dell’economia: la ricerca internazionale sull’economia informale, Napoli: Liguori. Porter, M. E. (1985), Competitive Advantage: Creating and Sustaining Superior Performance, New York: The Free Press. Portes, A., Castells, M. and Benton, L. A. (eds) (1989), The Informal Economy: Studies in Advanced and Less Developed Countries, Baltimore and London: John Hopkins University Press. Prahalad, C. K. (2005), ‘Aid is Not the Answer’, Wall Street Journal, August 31, p. A8? PROBE (1999), Public Report on Basic Education, New Delhi: Oxford University Press. Prugl, E. (1999), The Global Construction of Gender: Home-Based Work in the Political Economy of the 20th Century, New York: Columbia University Press. Prugl, E. and Tinker, I. (1997), ‘Microentrepreneurs and Homeworkers: Convergent Categories’, World Development, vol. 25, no. 9, pp. 1471–1482. Putnam, R. D., Leonardi, R. and Nawetti, R. Y. (1993), Making Democracy Work: Civic Traditions in Modern Italy, Princeton, NJ, USA: Princeton Press.

464

References

Pyke, F. (1992), Industrial Development Through Small Firm Co-operation: Theory and Practice, Geneva: ILO. Pyke, F. and Sengenberger, W. (eds), (1992), Industrial Districts and Local Economic Regeneration, Geneva: International Institute of Labour Studies. Pyke, F., Becattini, G. and Sengenberger, W. (eds), (1990), Industrial Districts and InterFirm Co-operation in Italy, Geneva: International Institute for Labour Studies. Rabellotti, R. (1997), External Economies and cooperation in Industrial Districts: A Comparison of Italy and Mexico, London: Macmillan Press LTD. Rabellotti, R. (1999), ‘Recovery of a Mexican Cluster: Devaluation Bonanza or collective Efficiency’, World Development, vol. 27, no. 9, pp. 1571–1586. Raikes, P., Friis-Jensen, M. and Ponte, S. (2000), ‘Global Commodity Chain Analysis and the French Filière Approach’, Economy and Society, vol. 29, pp. 390–417. Rakowski, C. A. (1994), ‘Convergence and Divergence in the Informal Sector Debate: A Focus on Latin America’, World Development, vol. 22, no. 4, pp. 501–516. Ranis, G. and Stewart, F. (1995), ‘V-Goods and the Role of the Urban Informal Sector in Development’, Development Studies Working Paper, Centro Studi Luca D’Agliano – Queen Elizabeth House, No. 85. Rao, Govinda (2005), Mid Year Review of the Indian Economy, India International Centre, New Delhi. Rao, R. and Husain, S. (1987), ‘Invisible Hands: Women in Home-Based Production in the Garment Export Industry in Delhi’, in Menefee, A. Singh and Kelles-Viitanen, A. (eds), Invisible Hands: Women in Home Based Production, New Delhi: Sage, pp. 51–67. Ray, D. (1998), Development Economics, Princeton, NJ, USA: Princeton University Press. Ray, R. (2000), ‘Child Labor, Child Schooling and their Interaction with Adult Labor: Empirical Evidence from Peru and Pakistan’, The World Bank Economic Review, vol. 14, no. 2, pp. 347–367. Regini, M. and Sabel, C. F. (eds), Strategie di impresa e riaggiustamento industriale, Bologna: Il Mulino. Reinecke, G. (2002), ‘Small Enterprises, Big Challenges: A Literature Review on the Impact of the Policy Environment on the Creation and Improvement of Jobs within Small Enterprises’, SEED Working Paper, No. 23, Geneva, ILO. Reinert, E. S. (2003), ‘Increasing Poverty in a Marshall Plans and Morgentau Plans as Mechanisms of Polarization of World Incomes’, in Chang Ha-Joon (ed.), Rethinking Development Economics, London: Anthem Press, pp. 453–478. Richey, L. A. (2000), ‘Gender Equality and Foreign Aid’, in Tarp, F. (ed.), Foreign Aid and Development: Lessons Learnt and Directions for the Future, London: Routledge. Robeyns, I. (2005), ‘The Capability Approach: a Theoretical Survey’ Journal of Human Development, vol. 6 (1), pp. 93–114. Rodrik, D. (1997), Has Globalization Gone Too Far?, Washington, DC: Institute for International Economics. Rodrik, D. (2000), ‘Development Strategies for the Next Century’, a paper prepared for presentation at the conference on ‘Developing Economies in the 21st Century’, Institute for Developing Economies, Japan External Trade Organization, in Chiba, Japan, January 26–27. Romijin, H. (2001), ‘Technology Support for Small-scale Industry in Developing Countries: a Review of Concepts and Project Practices’, Oxford Development Studies, vol. 29 (1), 2001. Rosario del Rosario, R. (ed.) (2001), The Philippines: Outsourcing of Manufacturing to Households: Subcontracted Home Based Work by Women and Children, Country report,

References 465 UNICEF Innocenti Research Centre, Florence (unpublished, joint report, UNICEF, PATAMABA, UP, CSWCD). Rosati, F. C. (1999), ‘Child Labor’, WP, World Bank. Rosati, F. C. and Rossi, M. (2001), ‘Children’s Working Hours, School Enrolment and Human Capital Accumulation: Evidence from Pakistan and Nicaragua’, Understanding Children’s Work Paper. An inter-Agency Research Cooperation Project, UNICEF Innocenti Research Centre. Rosenstein-Rodan, P. (1943), ‘The Problem of Industrialization of Eastern and South Eastern Europe’, Economic Journal, vol. 53, pp. 202–211. Rostow, W. W. (1960), The Stages of Economic Growth: A Non-Communist Manifesto, Cambridge: Cambridge University Press. Rubini, L. (2000), ‘Piccole nel Sud-Est asiatico’, in Bianchi, P., Di Tommaso, M. R. and Rubini, L. (eds), Le api audaci. Piccole imprese e dinamiche industriali in estremo oriente, Milano: FrancoAngeli. Sandee, H. (2002), ‘The Impact of the Crisis on Small-Scale Enterprises in Java, Findings from Selected Case Studies’, in van Dijk and Sandee, H. (eds), Innovation and Small Enterprises in the Third World, Cheltenham, UK: Edward Elgar Press. Schlyter, C. (2002), ‘International Labour Standards and the Informal Sector: Developments and Dilemmas’, Employment Sector, Working Paper on the Informal Economy, No. 3, Geneva, Switzerland: ILO, Schmitz, H. (1995), ‘Collective Efficiency: Growth Path for Small-Scale Industry’, The Journal of Development Studies, vol. 31, no. 4, pp. 529–566. Schmitz, H. (1998), ‘Fostering Collective Efficiency’, Small Enterprise Development, vol. 9, no. 1, pp. 4–11. Schmitz, H. (1999), ‘Global Competition and Local Cooperation: Success and Failure in Sinos Valley, Brazil’, World Development, vol. 27, no. 9, pp. 1627–1650. Schmitz, H. and Nadvi, K. (1999), Clustering and industrialisation an ‘Introduction’, World Development, vol. 27, no. 9, pp. 1503–1514. Sen, A. (1999), Development as Freedom, Oxford: Oxford University Press. Sen, A. K. (2000), ‘A Decade of Human Development’, Journal of Human Development, vol. 1 (1), pp. 17–23. Sender, J. (2003), ‘Rural Poverty and Gender: Analytical Frameworks and Policy Proposals’ in Chang Ha-Joon (ed.), Rethinking Development Economics, London: Anthem Press, pp. 405–423. Singh Navsharan (2000), ‘Situating Home-based Work in the Webs of Macroscape’, Background Paper for Regional Policy Seminar. Singh, Nirvikar (2003), ‘Some Economic Consequences of India’s Institutions of Governance: A Conceptual Framework’, Santa Cruz Center for International Economics, Working Paper Series 1025, Santa Cruz: UC Center for International Economics. Sinha, S. (2001), Strength in Solidarity: Insurance for Women Workers in the Informal Economy, Ahmedabad, India: Self Employed Women’s Association (SEWA). Skocpol, T. and Gretchen, R. (1991), ‘Gender and the Origins of Modern Social Policies in Britain and the United States’, Studies in American Political Development, vol. 5, pp. 36–93. Smith, T. M. F. (1983), ‘On the Validity of Inferences from Non random Samples’, Journal of the Royal Statistical Society, Series A, vol. 146, pp. 394–403. Stewart, F. (1974), ‘Technology and Employment in LDCs’, in Edwards, E. O. (ed.), Employment in Developing Nations, New York: Columbia University Press.

466

References

Sudarshan, R., Banerjee, M., Bhandari, L. and Venkatraman, S. (2001), Outsourcing of Manufacturing to Households: Subcontracted Home based Work in India, Country report, UNICEF Innocenti Research Centre, Florence (unpublished paper). Sun, Shangqing and Ren, Xingzhou (1997), A Grand Market in China: Retrospect and Prospect, New World Press. Tanzi, V. and Schuknecht, L. (2000), Public Spending in the 20th Century: A Global Perspective. New York: Cambridge University Press. Taylor, L., Mehrotra, S. and Delamonica, E. (1997), ‘The Links between Economic Growth, Poverty Reduction and Social Development: Theory and Policy’, in Mehrotra and Jolly (eds), Development with a Human Face: Experiences in Social Achievement and Economic Growth, Oxford: Clarendon Press. Tewari, M. (1999), ‘Successful Adjustment in Indian Industry: the Case of Ludhiana’s Woolen Knitwear Cluster’, World Development, vol. 27, no. 9, pp. 1651–1672. Thorat, S. (2005), ‘Persistent Poverty—Why SC and ST stay chronically poor’, DFID Working Paper, UK. Tjandraningsih, I. and White, B. (1992), ‘Anak-anak desa dalam kerja upahan’, Prisma, January, pp. 81–95. Tomei, M. (2000), ‘Homework in Selected Latin American Countries: A Comparative View’, Working Paper, No. 1, International Labour Office, Geneva. Trigilia, C. (1989), ‘Il distretto industriale di Prato’, in Regini, M. and Sabel, C. F. (a cura di), Strategie di impresa e riaggiustamento industriale, Bologna: Il Mulino, pp. 283–333. UNCTAD (1994), Technological Dynamism in Industrial Districts: An Alternative Approach to Industrialization in Developing Countries, UN Conference on Trade and Development, Geneva. UNDP and Moscow University (2000), Human Development: A Textbook, Moscow: Human Rights Publisher. UNDP (2003), Human Development Report, New York: Oxford University Press. UNESCO (2001a), Statistical Yearbook, Paris. UNESCO (2001b), Education for All and Children Who are excluded, World Education Forum, Dakar Senegal, 26–28 April 2000, Paris, France: UNESCO. UNICEF (2001), State of the World’s Children, Oxford University Press. UNICEF (2003), State of the World’s Children, Oxford University Press. UNIDO (1999), ‘SME Cluster and Network Development in Developing Countries: The Experience of UNIDO’, PSD Technical Working Papers 2, Vienna: UNIDO. UNIDO (2002), Corporate Social Responsibility: Implications for Small and Medium Enterprises in Developing Countries, Vienna: UNIDO. UNIFEM (2000a), A Preliminary Study on the Productive Linkages of Indian Industry with Home based Women Workers through Subcontracting Systems in Manufacturing Sector, New Delhi: United Nations Development Fund for Women. UNIFEM (2000b), Progress of the World’s Women 2000, New York. UNIFEM (2005), Progress of the World’s Women 2005, Overview: Women, Work and Poverty, New York, www.unifem.org United Nations (1996), Informal sector Development in Africa. Office of the special coordinator for Africa and the least developed countries. Unni, J. (2000), Urban Informal Sector: Size and Income Generation Processes in Gujarat, Part, I, Delhi, India: National Council of Applied Economic Research. Unni, J., Bali, N. and Vyas, J. (1999), ‘Subcontraced Women Workers in the Global Economy: Case of Garment Industry in India’. Study report sponsored by the Women’s Economic and Legal Rights (WELR), Program of the Asia Foundation, Washingon, DC.

References 467 Van Dijk, M.P. and Rabellotti, R. (1999), Enterprise Clusters and Networks in Developing Countries, EADI Book Series 20. Van Ginneken, W. (1996), ‘Social Security for the Informal Sector: Issues, Options and Tasks Ahead’, Social Security Department, Working Paper, International Labour Office, Geneva. Van Ginneken W. (2003), ‘Extending Social Security: Policies for Developing Countries’, International Labour Review, vol. 142, no. 3, pp. 277–294. Verdera, F. (2000), ‘Homeworkers in Peru’, SEED Working Paper No. 3, Series on Homeworkers in the Global Economy, Geneva, ILO. Vielle, P. and Waltery, P. (2004), The Challenges of Flexibility for Social Protection. Interactions Between Labour Market and Social Protection, Brussels, 16 May 2002. Visser, E.-J. (1999), ‘A Comparison of Clustered and Dispersed Firms in the Small Scale Clothing Industry of Lima’, World Development, vol. 27, no. 9, pp. 1553–1570. Volpi, F. (2002), ‘Economie Esterne e Cluster. Una nota sui problemi dell’industrializzazione nei paesi in via di sviluppo’, Economia Politica, vol. XIX, no. 2. Wade, R. (2003), ‘Globalisation Plus’: Fostering Truly Integrated Economies’, in K. Marshall and O. Butzbach (eds), New Social Policy Agendas for Europe and Asia. Challenges, Experiences and Lessons, Washington, DC: World Bank. Weijland, H. (1999), ‘Microenterprise Clusters in Rural Indonesia: Industrial Seedbed and Policy Target’, World Development, vol. 27, no. 9, pp. 1515–1530. Weitzman, M. L. and Xu Chenggang (1994). ‘Chinese Township-Village Enterprises as Vaguely Defined Cooperatives’, Journal of Comparative Economics, vol. 18, pp. 121–145. WIEGO (Women in Informal Employment Globalizing and Organizing) (2000), http://www.wiego.org/main/index.html World Bank (1999), ‘Towards an East Asian Social Protection Strategy’, Human Development Unit, East Asia and the Pacific Region, September. World Bank (2000), World Development Report 2000/2001, Attacking Poverty, New York: Oxford University Press, OUP. World Bank (2002), World Development Indicators 2001, Washington, DC: World Bank. World Bank (2003a), Global Economic Prospects and the Developing Countries, Washington, DC: World Bank. World Bank (2003b), World Development Indicators 2003, Washington, DC: World Bank. World Commission on the Social Dimension of Globalization (2004), A Fair Globalization: Creating Opportunities For All, World Commission on the Social Dimension of Globalization, International Labour Office, Geneva. Xaba, J., Horn, P. and Motala, S. (2002), The Informal Sector in Sub-Saharan Africa’, Working Paper on the Informal Economy, Employment Sector 2002/10, International Labour Office, Geneva. Yang, Dali L. (1997), Beyond Beijing, Liberalization and the Regions in China, London, New York: Routledge. Yingyi Qian (2003), ‘How Reform Worked in China’, in Rodrik, Dani (ed.), In Search of Prosperity. Analytical Narratives on Economic Growth, Princeton, NJ, USA: Princeton University Press. You-II Lee and Yoon Hwan Shin (1995), ‘Korean Direct Investment in Southern Asia’, Journal of Contemporary Asia, vol. 25, no. 2. Zhonghui, Wang (1993), China’s Policies Towards Collective Rural Enterprises, Small Enterprises Development, vol. 4 (1), pp. 16–26.

Index

absence of written agreement 76, 195 advanced capitalist countries 3, 305, 400, 403, 405, 406, 418 Aurat (Women’s) Foundation 36, 239 Barangay Council for the Protection of Children (BCPC) 313 Basic social services (BSS) 21, 79, 266, 310, 362, 368, 369, 389, 415 Batik production in java 262, 263; Intermediaries called cablik 264; Nglorod (washing of wax in hot water) 263, 285 n.24; Nglowongi (waxing) 263, 285 n.24 below the poverty line 31 n.43, 83, 90, 119, 190, 229, 230, 246 n.60, 274, 407, 417, 421; Bidi Welfare Fund 106, 132, 183, 193, 204 Blue-jeans cluster of Torreon, Mexico 17 bonus from subcontrators 264 Bosca-Escola Programme, Brazil 206 buyer-driven chains 5, 43, 69, 70 capital accumulation 4, 361 Carpet weaving 36, 39, 70, 147, 152, 210, 211, 213, 217, 220, 226, 227, 228, 229, 237, 240, 241, 242, 246 n.50 Ceramic pottery 37, 39 Child activity status (CS) 125, 154, 159, 169; child characteristics (CC) 154; economic endowment (EEhh) 25, 26, 154; human endowment of the household (HEhh) 25, 26, 154 Child characteristics (CC) 154 Child labour 13, 20, 36, 28, 67, 124–125, 131–132, 136, 138–139, 143, 145–147, 151, 153, 157, 159, 161–165, 166 n.5, 167 nn.2, 11, 20, 169 n.44, 196, 198–199, 201, 202, 205–206, 209 n.17,

215, 239, 257–259, 277, 279, 313–315, 319 n.31, 328–329, 355 n.11, 373, 379, 415, 423–424, 437, 446, 448; activists 197; bonded 197; detriminants of 154; elimination of 164, 166 n.1, 200, 313; employment of 123; gender issues 307; in homeworker households 32, 123, 132, 139, 197, 201, 277, 414; incidence of 198, 201, 213, 275, 276, 307; reduction 25, 31 n.43, 126, 127 child schooling 132, 235 China’s Township and village enterprises (TVEs) 5, 27, 31 n.43, 373–376, 387, 393, 399 n.25; collective TVEs 375–378; ‘Spark Plan’ through technological demonstration centres 376; Specialized Towns 376 CLASS programme in Tamil Nadu 204, 206 Clean clothes Campaign (CCC) 443 Clusters 17–20, 30 n.31, 35, 36, 38, 60 n.6, 177, 297, 299, 361, 363, 365, 367–370, 372–374, 378–386, 388, 390–395, 397 nn.7, 8, 398 nn.15, 18, 19, 399 n.27, 427; cluster theory 5, 26, 120, 401; of homeworkers 5, 16, 29, 35, 37, 63, 82, 119, 120, 173, 363, 365; incipient 20, 372, 373; industrial 19, 82, 378; SMRs 16, 362, 389 COGEFIS 380 collective action 17, 19, 24–26, 63, 80, 83, 104, 107, 113, 117–120, 126, 146, 157–159, 164, 229, 239, 348, 366, 369–373, 378–379, 382, 383–393, 402, 406, 408 commodity chain 4, 27, 44, 62, 64, 78; see value chain [value-adding activities through which a product passes; from design to production]

470

Index

community-based organizations (CBOs) 121 n.6, 290, 314, 364, 425 contract status 68, 79 Control group (CG) of households [not engaged in home-based activity] 34, 38, 42, 60 n.6, 82, 83, 92, 119, 127, 173, 177, 210, 215, 251, 262, 290, 331 Convention on homework 61 n.13, 408, 423 Corporate social responsibility: Benefits of SA8000 447; Example of the SA8000 445, 446, 448; Social Accountabilty International (SAI) 446 Cotton knitwear cluster of Tiruppur, South India 17 credit facilities 203, 386, 392 Danish International Development Agency (DANIDA) 61 n.13, 259, 411 debt bondage 30 n.30, 38, 67, 76, 77, 78, 163, 191, 196, 203, 207, 208 n.11, 346, 393, 424, 446; to contractors 190, 196 degree of informality 69, 79 delayed payments 69, 73, 76, 78, 104, 168 n.34, 190, 196, 203, 226, 237, 244 n.18, 267, 299, 311, 328, 345 demands of supply chain flexibility 12, 365 Department of Education, Culture and Sports (DECS) 313 Department of Labour and Employment (DOLE) 296, 311, 312, 314, 318 n.8, 319 nn.29, 31 developing countries 3–5, 7–8, 12–13, 18–20, 27, 28 n.12, 29 nn.23, 24, 30 nn.29, 35, 35, 66, 82, 121 nn.5, 123, 166 nn.1, 173, 361–363, 365, 367–368, 372–374, 376, 379–381, 383–385, 387–388, 393–394, 400–401, 403, 405–408, 413, 418, 420, 424, 426–427, 448 different product classes 186 Directorate General of Employment and Training (DGET) 175 dirt road 82, 365, 368–369, 371, 382, 384, 389, 395 discrimination 106, 132, 226–227, 238, 277, 319 n.30, 388, 423–424, 429, 438, 447–448 diversification of family income 83 double burden 15, 98, 119, 149, 389, 413; of the girl child 150, 163, 308; of women homeworkers 98, 233

double horizontal equality 106 dual synergies 24 Economic endowments (EEhh) 25, 26, 107, 114, 124, 154, 187 employment protection status 69, 79 empowerment of women 38, 208 n.11, 347, 445 entrepreneurial risk 15 European Countries 28 n.13, 337, 403, 406, 407, 422; fiscal and legal support for child care and parental leave 404; insurance concept 403; public health programmes 403; redistribution programmes 404; welfare state packages 403 excellence in cluster development process 378 excess supply of labour 15, 76, 196, 237, 239 exploitation 7, 15, 19, 24, 64, 73, 77, 119, 214, 218, 226, 237, 238, 243 n.9, 244 n.18, 328, 341, 345, 350, 363, 364, 390, 451 n.16; of children 237, 279, 314, 319 n.30; of the homeworkers 15, 73, 77, 213, 215, 225, 372 export-oriented labour–intensive manufacturing 12 Fair Trade Group of Nepal (FTGN) 391 Fair Wear Campaign 443 family tradition 142, 268, 270 fashion accessories 37, 43, 66, 77, 149, 151, 290, 299, 302, 303, 308, 309, 310, 316,317, 318 n.6, 369 Federal Bureau of Statistics (FBS), Government of Pakistan 36, 210, 238 feminisation 12, 93, 96, 119, 135, 149, 162–163, 191, 255, 326, 407 fiscal and legal support for child care and parental leave 403 Focus group discussions (FGDs) 33, 34, 92, 106, 107, 121 n.17, 127, 149, 178, 179, 262, 296, 308, 310, 311, 318 n.14, 319 n.15, 331, 340, 345, 355 n.18; with child workers 135, 142, 148, 153, 179, 201, 217, 229, 260, 262, 271, 277, 278, 283; with women homeworkers 38, 107, 179, 201, 217, 226, 246 n.48, 260, 262, 347 football stitching 213 Footwear export cluster of Sinos Valley, Brazil 17

Index 471 formal economy 4, 62, 212, 401, 417, 422, 426; social insurance mechanisms 401 formal sector activity 62, 419; capitalist enterprise 62 future market demand 182 Gamarra garment cluster, Peru 18 Gender Development Index 21, 22, 31 n.37 Global Compact initiative (UN) 445, 448, 450 global integration and competition 4 globalization 12, 13, 30 n.27, 62, 79, 319, 320, 425 global product markets 13 global quality assurance standards 20, 272 global supply chain 8 government and non-governmental initiatives for homeworkers 408 health conditions 121 n.17, 147, 215, 237 health status of 21, 24, 83, 92, 99, 104, 113, 114, 116, 118, 119, 120, 121 n.14, 127, 166 n.71, 205, 217, 230, 233, 275, 309, 341 high road 18, 19, 26, 30 n.35, 82, 365, 368, 370, 371, 373, 378, 379, 387, 398 n.16 hired labour 127, 186; casual and long term 73 home-based work/workers 7, 9, 11, 13, 14, 19, 28 n.12, 38, 60 n.2, 74–75, 106, 191, 214, 229, 238, 268, 273, 280, 290, 293–295, 300, 350, 399 n.29, 410, 414, 416, 423–424, 443–444, 451 n.21, 452 n.27 home décor 37, 149, 168 n.24, 296, 335 Homenet 37, 61 n.13, 107, 121 n.8, 281, 282, 364, 390, 391, 397 n.10, 410, 411, 443, 444 Homework 8, 9, 12, 13, 16, 29 n.26, 32, 34–39, 42–43, 60 n.5, 64, 69, 73–74, 77, 82–84, 92–93, 96, 104, 107, 114, 118, 119, 120 n.3, 121 nn.6, 16, 18, 123–125, 127, 130–132, 135, 138–139, 142–144, 148–152, 158, 161, 163–164, 166 nn.2, 3, 167 nn.14, 20, 168 n.28, 174, 177, 181, 183, 185–186, 188, 190, 193–196, 198, 225–228, 231–232, 234–242, 243 n.11, 244 n.14, 245 nn.37, 41, 46, 59, 64, 247 nn.69, 73, 74, 250, 259, 260, 262, 267, 270,

272, 273–280, 285 n.21, 286 n.53, 287 n.59, 294, 297, 299, 301, 304–310, 312, 314, 317, 318 nn.5, 7, 8, 10, 322, 323, 326–328, 331, 335, 339, 340, 341, 343, 347–348, 350–351, 353, 355 nn.15, 17, 356 n.41, 357 n.47, 363, 389, 403, 411, 413–414, 420; hours worked by women in homework 96, 220, 389, 413; involved manual work labour-intensive activities 8; in manufacturing 123 Homework Based Work (HBW) 326, 327, 328, 329, 356 n.36 homeworker clusters 82 homeworker households 5, 15, 16, 19, 21, 22, 32–34, 36–39, 42–43, 60 n.6, 73, 79, 82–84, 90, 92–93, 96, 119, 123–127, 130–132, 134–136, 138–139, 142–143, 146–148, 150–154, 157, 159, 161–163, 167 n.12, 177, 271, 305, 307–308, 331, 340, 346, 413–415 Homeworkers (industrial outworkers who work at home) 5–8, 11, 14–17, 21–22, 24, 28 n.13, 30 n.32, 34–44, 53, 59, 60 n.5, 62, 64, 66–67, 69–73, 76–79, 81 n.11, 82–84, 87, 92–93, 96, 99, 104, 106–107, 113–114, 116, 118–120, 121 nn.8, 16, 18, 126–127, 131, 135, 138, 146–147, 151, 157–159, 173, 191, 213, 215, 217, 219–220, 223–226, 229–230, 232, 236, 239, 244 nn.17, 23, 245 n.36, 250, 251, 256–257, 259–260, 262–264, 266–283, 285 nn.22, 24, 286 nn.39, 42, 43, 290–291, 192–307, 309–315, 318 nn.6, 8, 322, 324, 326–328, 332–335, 337–341, 343, 345–353, 354 n.3, 355 nn.14, 17, 19, 356 nn.38, 43, 357 nn.48, 49, 51, 363–366, 369, 371–372, 378–381, 383, 389–392, 398 n.17, 399 n.28, 399 n.33, 402, 206, 408–411, 413, 416–418, 420–423, 428–430, 432–437, 443–444, 451 nn.14, 22 home (workplace) and access to basic services 84; electricity 297; water facilities 297 horizontal inequality 106 household income data 189 house ownership (A marker of wealth) 158, 343 human capital endowment (HE) 22, 92, 157 human development index 31 n.37

472

Index

Human development level of a household (HDhh) 25; see also economic endowments (EE); human capital or endowment (HE) Human rights commission 212 hybrid seeds production 37, 42, 322, 229, 269 import and capital-intensive products 3 inappropriate technologies 3 incense stick making (agarbati) 36, 39, 104, 116, 149, 163, 173, 210, 218, 237, 240 incentives 29 n.23, 31, 196, 205, 206, 213, 281, 313, 351, 354 n.6, 374–377, 386, 397 n.11, 415, 426 incidence of child work 127, 132, 166 n.3 incidence of physical abuse 152 income increase 21, 25, 92, 362 India: bidi, incense sticks (agarbati) and garment (zardosi) 36, 39, 43, 70, 83, 106–107, 152, 184–185, 194, 199, 204, 210, 219; Bidi Welfare Fund 106, 410, 421; Self-employed Women’s Association (SEWA) 36, 191, 410, 443; chiken-kari work 191 Indonesia: see Danish International Development Agency (DANIDA); International Labour Organization (ILO); Mitra Wanita Pekerja Rumahan Indonesia [National Friends of Women Home-workers] (MWPRI), rattan furniture, ceramic pottery and batik printing on garments Industrial Employers Association of Prato, Italy (Associazione Industriali Pratesi) 380 industrial outwork system 16 industrial production in developing countries 3 Industrial Revolution 3, 4, 8 informal contract labour 217 informal economy 3–6, 9, 13, 25–26, 27 n.1, 28 n.6, 34, 43, 62–63, 65, 123, 243 n.4, 322, 362, 365, 384, 397 n.5, 401–402, 404, 406, 413–414, 417–418, 420–422, 425–427, 444 informal employment 4, 7, 8, 9, 28 n.6, 173, 174, 175 informalization of labour force/market 12, 13 informal sector 32, 74, 175, 211

informal sector activity [family ownership], 173, 210, 419; see homeworkers; verbal contracts infrastructure 17, 24, 25, 31 n.43, 104, 107, 179, 183, 236, 237, 296, 311, 371, 375, 377–378, 380, 382, 386, 390–392, 399 n.26, 408 insurance concept 404 insurance mechanisms 15 integration of economic and social policy 4, 26, 30 n.27, 328, 363, 369, 384, 385, 394, 423 Intermeditiaries called cablik 2 international competition, 369, 370, 377 International conference of Labour Statisticians (ICLS) 6, 28 n.6; International financial institutions 11, 12; see IMF; World Bank International Labour Organization (ILO) 20, 29 n.19, 60 n.1, 61 n.13, 67, 166 nn.1, 4, 5; see also Convention on home work; International Programme on the elimination of Child Labour (IPEC) International Monetary Fund (IMF) 11, 12, 361, 365 International Programme on the elimination of Child Labour (IPEC) 164, 166 n.1, 197 international sub-contracting 4, 5; buyer-driven commodity chains 4, 5, 12, 69, 70, 78; producer-driven commodity chains 4, 5 International System of National Accounts 6 international trade negotiators 123 Italian clusters and homeworkers’ human development 18, 378, 393; see also COGEFIS excellence in cluster development process; Industrial Employers Association of Prato (Associazione Industriali Pratesi); knitwear industrial district, Capri, Italy keeping workers isolated 196 knitwear industrial district, Capri, Italy 30 n.32, 379 Labour force 3, 4, 7, 9, 12, 16, 17, 19, 20, 24–26, 37, 82, 92–93, 120, 121 n.5, 124, 175, 179, 211, 238, 243 n.4, 251, 254, 256, 280, 291–292, 294, 323–325, 328, 380, 388, 392, 397 nn.3, 6, 399 n.27, 401, 412, 416, 418, 421, 426

Index 473 Labour intensive consumer goods 5 labour market flexibility 12 labour market of developing countries 3 lack of alternative employment 8, 107, 232, 306 Leather crafts 42, 149, 331, 332, 345, 346, 355 nn.18, 19, 369 literacy rate 31 n.37, 194, 195 local collaboration 371 Local Economic Development Agencies (LEDA) 394 local economic system perspective 363 local institutions 30 n.37, 370, 382, 384, 398 n.22 Madrasas [traditionally a place for religious instruction] 135, 152, 167 n.18, 200, 206, 207, 227, 246 n.52 malpractices by subcontractor 69, 75, 79, 227 Maquila Solidarity Network (MSN) 443 market access 43, 123, 167 n.2, 388, 391 Membership-based organisations of the poor (MBOPs) 104, 106, 364, 390 migrant labour force 124 Millennium Development Goals (MDGs), 362 Mitra Wanita Pekerja Rumahan Indonesia [National Friends of Women Home-Workers] (MWPRI) 37, 61 n.13, 443 Muhajir Qaumi Movement (MQM) 218, 244 n.21 Multilateral Fibre Arrangements 66 Multilateral Investment Fund (MIF) 395 Multinational companies 13, 334, 397 n.3 narrow occupational communities 407 net domestic product 181 Nglorod (washing of wax in hot water) 263, 285 n.24 Nglowongi (waxing) 263, 285 n.24 Occupational Health and Technological improvements 205 Okra production 37, 43, 66, 69–70, 72, 149, 290, 296, 302, 305, 307, 309, 313, 316 Ordinary Least Square (OLS) 118 Organizations for/of homeworkers 21, 364, 424, 426, 429, 433–434, 437; see Clean clothes Campaign (CCC); Fair Wear Campaign; Homenet;

Maquila Solidarity Network (MSN); Mitra Wanita Pekerja Rumahan Indonesia (MWPRI); OXFAM International; PATAMABA; SelfEmployed Women’s Union (SEWA); United Nations Development Fund for Women (UNIFEM); Women in Informal Employment Globalizing and Organizing (WIEGO) outsource manufacturing of consumer non-durables 12 outsourcing 13, 62, 63, 80 n.2, 175, 210, 213, 263, 264, 290, 293, 295, 322, 411, 412 OXFAM International 444 Pakistan: see Aurat (Women’s) Foundation; carpet weaving, incense stick-making, shrimp peeling and sack-stitching; Pakistan Institute of labor Economics research (PILER); Social Policy and Development Center (SPDC) Pakistan Institute of labor Economics Research (PILER) 36, 239, 243 n.8 Participatory Appraisal of Competitive Advantage (PACA) 394 PATAMBA 37, 61 n.14, 105, 290, 296, 299 Philippines: see Barangay Council for the Protection of Children (BCPC); Department of Education, Culture and Sports (DECS); Department of Labour and Employment (DOLE); and fashion accessories; PATAMBA, NGO; pyrotechnics production, okra production, packing, home décor (Christmas lights/balls) poisson regression model 114 policy implication 16, 22, 38, 64, 104, 127, 402, 413–414, 416 potential health hazards 177 poverty reduction 17, 21, 27, 82, 250, 362, 365, 371, 389, 394, 399 n.37, 425 Prawn shelling 152, 210, 223, 224, 226, 227, 229, 237, 240, 246 n.50 priority interventions 113 Productive Integration Projects (PIPs) 394 profit seekers 202 public health programmes 403 Pyrotechnics production 37, 296, 299, 306, 313, 316

474

Index

quality management practices 20, 373 Questionnaire guidelines for Homeworkers and CG 44–60, 215, 297 Rattan furniture industry 42, 266 recreation activities 235 redistribution programmes 404 removal of discriminatory regulation 387 Saa paper [artificial flowers and other handicrafts] 37, 42, 67, 76, 153, 322, 331, 335, 341, 343, 346–349, 352, 355 nn.18, 19, 356 nn.26, 27, 30, 357 nn.44, 47 Sack(bori)stitching sector 217, 223 Scholarships for children 157, 452 n.27 sector specific policy recommendations 207 Security 16, 36, 63–64, 67, 69–70, 80, 106, 113, 125, 181, 203–204, 208, 214, 239, 296–297, 312, 328, 349, 375, 400–401, 404–405, 408–410, 412, 417–418, 420–422, 426, 436, 450 n.4, 450 n.22, 452 nn.24, 27 Self-Employed Women’s Union (SEWA) 36, 61 n.13, 191, 209 n.15, 243 n.6, 293, 364, 390–391, 397 n.10, 399 n.31, 409–411, 443–444 self-employment 7, 68, 188 self-financing 190 Self Help Groups 393, 399 n.36 Shoe cluster Sinos Valley, Brazil 18 Shrimp/prawn peeling 36, 44, 70, 113, 152, 210, 223, 224, 226, 227, 229, 237, 240, 246 n.50 size of household 84, 167 n.71(a), 329, 357 n.46; see poverty reduction Social Entrepreneurship Program (SEP) 395 social insurance 80, 152, 167 n.2(b) 369, 400–409, 411, 413, 415, 417–423, 425, 427, 431, 451 n.13, 452 n.26 Socialist party activism 407 Social Policy and Development Center (SPDC) 230 social protection 5, 7. 13, 15, 16, 19, 21, 26, 28 n.6, 30 n.33, 67, 69, 74, 79, 82,104, 106, 116, 118, 126, 250, 312, 362–365, 369–371, 373, 379–380, 383–384, 389–391, 397 nn.6, 8, 401–403, 406–407, 411, 413–414, 417, 425, 427, 450 n.7, 452 n.24 socials and economic conditions of women homeworkers 82

stability of relationship between subcontractors and homeworkers 77 Structural adjustment programmes (SAPs) 210 subcontracting chain 12, 13, 62, 66, 72, 74, 118, 264, 290, 299, 301, 303, 310, 315 Subcontracting system 11, 29 n.21, 30 n.28, 43, 63–64, 69, 72–75, 77, 79, 82, 174, 176, 186, 266, 290, 293, 323, 327, 328, 340, 362 supply chain flexibility 12, 365 Surgical instrument cluster, Sialkot, Pakistan 18, 20, 373 surplus labour 3, 9, 79, 120 Tanning cluster of Palar Valley, TN, India 20, 373 Tax-based funds 419 technological and management upgrading 386 technological innovation 388, 394 Thailand: see Homenet, Thailand; Homework Based Work (HBW); hybrid seeds production; leather crafts; saa paper toxic materials 92, 99 trade and financial liberalization 11, 13 trade policies and domestic demand 383 Trade unions 8, 13, 63, 80, 296, 364, 378, 379, 407, 426, 444, 447 Training and Certification 204 unemployment insurance 401, 403, 405, 406, 417, 418, 427 Unionisation 14, 15, 104, 146, 183, 195, 196, 244 n.17, 293, 327, 379 United Nations Children’s Fund (UNICEF) 20, 166 nn.1, 5, 197 United Nations Development Fund for Women (UNIFEM) 30 n.28, 33, 120 n.4, 175, 180, 392, 443-445 unorganized sector 409 Value chain 62, 64, 65, 66, 67, 69, 70, 73, 76, 79, 118, 174, 186, 219 verbal contracts/absence of written agreement 73, 74, 79, 195 wage-employment 7, 120 n.2 welfare state packages 403 white-collar jobs 153

Index 475 Women in Informal Employment Globalizing and Organizing (WIEGO) 60 n.2, 397 n.10, 399 n.32, 443–445 Women’s Health Status (WHS) 114, 118 worker-contractor arrangements 191 worker insecurity 64, 70, 73

working-class mobilization 406 World Bank 11, 12, 29 n.22, 29 n.23, 36, 61 n.13, 92, 166 n.1, 166 n.5, 167 n.11, 213, 253, 284 n.1, 349, 361, 365, 372, 423, 425, 450 nn.2, 3, 4, 5, 451 n.17 World Development Indicator (WDI) 93