Women, Work and the Japanese Economic Miracle: The case of the cotton textile industry, 1945-1975 (Routledgecurzon Studies in the Modern History of Asia)

Citation preview

Women, Work and the Japanese Economic Miracle

This book shows, how during the period of the Japanese economic miracle, a distinctive female employment system was developed alongside, and different from, the better known Japanese employment system which was applied to male employees. Women, Work and the Japanese Economic Miracle describes and analyses the place of female workers in the cotton textile industry, which was a crucially important industry with a large workforce. In presenting detailed data on such key issues as recruitment systems, management practices and the working experience of the women involved, it demonstrates the importance for Japan’s postwar economy of harnessing female labour during these years. Helen Macnaughtan is Handa Fellow in Japanese Business and Management at the School of Oriental and African Studies, London. Her areas of research interest include employment, labour management, gender and economic development in Japan.

RoutledgeCurzon Studies in the Modern History of Asia

1 The Police in Occupation Japan Control, corruption and resistance to reform Christopher Aldous 2 Chinese Workers A new history Jackie Sheehan 3 The Aftermath of Partition in South Asia Tai Yong Tan and Gyanesh Kudaisya 4 The Australia–Japan Political Alignment 1952 to the present Alan Rix 5 Japan and Singapore in the World Economy Japan’s economic advance into Singapore, 1870–1965 Shimizu Hiroshi and Hirakawa Hitoshi 6 The Triads as Business Yiu Kong Chu 7 Contemporary Taiwanese Cultural Nationalism A-chin Hsiau 8 Religion and Nationalism in India The case of the Punjab Harnik Deol 9 Japanese Industrialisation Historical and cultural perspectives Ian Inkster

10 War and Nationalism in China 1925–1945 Hans J. van de Ven 11 Hong Kong in Transition One country, two systems Edited by Robert Ash, Peter Ferdinand, Brian Hook and Robin Porter 12 Japan’s Postwar Economic Recovery and Anglo-Japanese Relations, 1948–1962 Noriko Yokoi 13 Japanese Army Stragglers and Memories of the War in Japan, 1950–1975 Beatrice Trefalt 14 Ending the Vietnam War The Vietnamese Communists’ perspective Ang Cheng Guan 15 The Development of the Japanese Nursing Profession Adopting and adapting Western influences Aya Takahashi 16 Women’s Suffrage in Asia Gender nationalism and democracy Louise Edwards and Mina Roces 17 The Anglo-Japanese Alliance, 1902–1922 Phillips Payson O’Brien 18 The United States and Cambodia, 1870–1969 From curiosity to confrontation Kenton Clymer 19 Capitalist Restructuring and the Pacific Rim Ravi Arvind Palat 20 The United States and Cambodia, 1969–2000 A troubled relationship Kenton Clymer 21 British Business in Post-Colonial Malaysia, 1957–70 ‘Neo-colonialism’ or ‘disengagement’? Nicholas J. White

22 The Rise and Decline of Thai Absolutism Kullada Kesboonchoo Mead 23 Russian Views of Japan, 1792–1913 An anthology of travel writing David N. Wells 24 The Internment of Western Civilians under the Japanese, 1941–1945 A patchwork of internment Bernice Archer 25 The British Empire and Tibet 1900–1922 Wendy Palace 26 Nationalism in Southeast Asia If the people are with us Nicholas Tarling 27 Women, Work and the Japanese Economic Miracle The case of the cotton textile industry, 1945–1975 Helen Macnaughtan 28 A Colonial Economy in Crisis Burma’s rice cultivators and the world depression in the 1930s Ian Brown

Women, Work and the Japanese Economic Miracle The case of the cotton textile industry, 1945–1975 Helen Macnaughtan

First published 2005 by RoutledgeCurzon 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN Simultaneously published in the USA and Canada by RoutledgeCurzon 270 Madison Ave, New York, NY 10016 This edition published in the Taylor & Francis e-Library, 2005. “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.” RoutledgeCurzon is an imprint of the Taylor & Francis Group © 2005 Helen Macnaughtan 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 has been requested ISBN 0-203-39085-7 Master e-book ISBN

ISBN 0-203-66964-9 (Adobe eReader Format) ISBN 0–415–32805–5 (Print Edition)

For Sam and Cora

Contents

List of illustrations Acknowledgements Note on Japanese terms Frequently used abbreviations 1

A historical legacy

xi xiii xv xvi 1

Female labour and textiles in prewar Japan 1 The textile industry’s postwar transition 9 The importance of women in textiles 15 A note on methodology 23 2

Shaping the female labour market

27

Textile workers and the Occupation authorities 27 Teething problems, unions and technology 40 3

Recruiting women workers

58

The postwar recruitment environment 58 Recruitment of young female workers 60 Recruitment problems for the industry 79 Recruitment of older female workers 89 4

Managing women workers The development of labour management strategies 102 Managing young female workers through the dormitory system 108 Finding alternative strategies for managing older female workers 126

102

x Contents 5

Educating and training women workers

133

Factory girls and schooling 133 Women and workplace training 149 6

Wages and welfare for women

159

Women and the wage system 160 Nominal, comparative and real wage trends 172 Welfare for women 177 Evaluation of wages and welfare 195 7

Japanese female textile workers in context

209

The textile industry as an employer of women workers 209 The impact of textiles on the national labour market 211 The everyday lives of female textile operatives 214 Appendix: fieldwork interviews and survey References Index

217 220 229

Illustrations

Figures 1.1 6.1

Breakdown of female textile workers by marital status, 1952 and 1972 Example of a typical (monthly) wage system

20 161

Tables 1.1 1.2 1.3 1.4 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 3.10 3.11 3.12 3.13 4.1 4.2 4.3 4.4 5.1 5.2

Females employed in manufacturing sector (main sectors) 18 Breakdown of female employment in textile manufacturing 18 Age breakdown of female textile workers, 1952 and 1972 20 Female textile workers in cotton companies by employment status 24 Supply trends for young females 62 Demand trends for young females 64 Employment of new high school leavers via PESO 65 Female new graduate placements via PESO (manufacturing) 66 Top six prefectures for supply of female junior high school leavers in the textile industry, 1956–68 76 Textiles’ employment of female school leavers by regional area of Japan, 1963 and 1971 77 ‘Sufficiency rate’ in JSA companies 81 Employment of female new graduates across JSA companies 82 Employment of female school leavers in the textile industry 84 Female school leavers employed by job type 87 Supply and demand for ‘regular’ workers by age 91 Employment of female seasonal workers under the ‘Hokkaido¯ Plan’ 98 The transition from young to older female labour in textiles 100 Example of an ideal labour management policy by age level for females 114–15 Example of daily routine for dormitory boarders 119 ‘Characteristics’ of older female workers 128 Management strategies for the employment of older women 130 Role of dormitory versus school 134 Example of a typical shift-work/education hours system 140

xii Illustrations 5.3 5.4 5.5

To¯yo¯bo¯ education curriculum subjects To¯yo¯bo¯ tertiary education Proportion of female employment in Big Ten spinning and integrated weaving departments 5.6 Example of a typical training schedule for new joiners 6.1 Average monthly wages per female regular employee 6.2 Average male–female wage differentials for textile workers by age 6.3 Average wages of female textile workers by age 6.4 Wage structure for female seasonal workers in the ‘Hokkaido¯ Plan’ 6.5 Starting salary for female junior high school graduates and wage-rise movements 6.6 Dormitory facilities 6.7 Motherhood protection provisions, 1947–74 6.8 Cost of living surveys for female textile workers A.1 Personal profile of fieldwork survey respondents

144 145 150 155 162 165 165 172 174 183 192–93 200 219

Plates 2.1

Women marching through the streets during the 1954 ¯ mi Kenshi strike O 42 2.2 Female textile workers dressed for a union rally, May Day 1958 45 2.3 Woman working at ‘automatic spooler’ machine 54 2.4 Woman working at ‘cotton comber’ machine 55 3.1 New recruits (wearing their school uniform) enter factory gates and are greeted by their seniors (senpai), April 1968 69 4.1 One day in dormitory life, 1958 120–21 5.1 Factory girls (wearing school uniform) in front of high school building at To¯yo¯bo¯’s Komatsushima Factory 142 5.2– New female employees receiving training on machinery use 5.3 and safety, 1971 153 6.1 A modern three-storey dormitory building for female employees 181 6.2 Club building and recreation area 181 6.3 Women performing morning exercises in company grounds 186

Acknowledgements

This book is based on my PhD thesis (2001) and fieldwork in Japan (1997– 98). As with any research, I am grateful for the support and assistance of many people during that time. I would like to thank Professors Abe Takeshi, Miyamoto Matao and Sugihara Kaoru at the Faculty of Economics at Osaka University for helping to arrange my fieldwork year in Japan and for providing personal introductions to vital contacts in the textile industry. Within the industry, I am grateful to all those who assisted me in finding documentation and who gave up their time to meet with me and take me to various places of interest, including several cotton textile factories. In particular, I would like to thank Mr Kataoka of Osaka Sen’i Gakuen, Mr Sudo of the Japan Spinners’ Association and managers at Unichika, Kurabo¯, Zensen, To¯yo¯bo¯ and To¯yo¯bo¯ Union. I owe a very special thank you to Mr Murakami of To¯yo¯bo¯, who met with me regularly throughout my fieldwork year and gave me access to the documentation room at To¯yo¯bo¯. He also arranged for factory visits, the opportunity to interview and distribute surveys to female textile workers and provided me with the To¯yo¯bo¯ photographs in this book. I am grateful to staff at To¯yo¯bo¯’s TMC and Nyu¯zen factories, Unichika’s Kaizuka factory and to those I met during a visit to Nishiwaki, all of whom provided such warm hospitality. Finally, I am extremely grateful to the many women workers who allowed me to interview them and who responded to my questionnaire survey (but whom I will not name in order to protect their confidentiality). Discovering their personal experiences and responses provided a very enjoyable part of my research and I hope that I have been successful in incorporating their voices within this book. I am very appreciative of financial support received from the Economic and Social Science Research Council, Economic History Society and Japan Foundation Endowment Committee. I would like to extend a special thank you to Professor Janet Hunter (London School of Economics), who throughout the course of this research always made herself readily available, gave invaluable support, advice and encouragement, as well as read through countless drafts. I am also grateful to the Department of Financial and

xiv Acknowledgements Management Studies, School of Oriental and African Studies (SOAS) for providing me with a postdoctoral fellowship, enabling me to complete the writing of this book. Helen Macnaughtan

Note on Japanese terms

Standard romanisation has been used for Japanese words, and macrons have been used above vowels to indicate that the vowel is long. Japanese words and phrases are given in italics, with the exception of people’s names, place names (e.g. Hokkaido¯) and company or institutional names (e.g. To¯yo¯bo¯). Japanese names are given in the Japanese order – family name followed by personal name – but have not been assigned vowel macrons. Translations from Japanese sources into English are, for the most part, my own.

Frequently used abbreviations

JSA MOL PESO SCAP Zensen

Japan Spinners’ Association (Nihon Bo¯seki Kyo¯kai) Ministry of Labour (Ro¯do¯sho) Public Employment Security Office (Ko¯kyo¯ Shokugyo¯ Anteisho) Supreme Commander for the Allied Powers National Federation of Textile Industry Workers’ Unions (Zenkoku Sen’i Sangyo¯ Ro¯do¯ Kumiai Do¯mei – Zensen Do¯mei)

1

A historical legacy

In the 1979 Japanese film Aa Nomugi To¯ge (Ah Nomugi Pass),1 a story is told of young women from impoverished rural families who make the long trek crossing the snow-covered Nomugi Pass in the Nagano Alps to work in a silk-reeling factory. It is a tale of melancholy and misery which depicts the exploitation of young Japanese women under early industrial advance, and is a narrative which has been woven with fact and fiction to remain both in the history books and in the minds of many Japanese today. The story of female textile operatives is one viewed as having taken place prior to the Second World War, playing an important role in Meiji and Taisho¯ economic and social development. However, it is also a story that continued and evolved into the postwar decades, providing a crucial base for Japan’s reconstruction and growth and shaping the lives and roles of a further generation of Japanese women. This book is the result of research into this postwar history and seeks to show how this story of women and textiles has important significance and much broader application than might be first imagined for Japan’s postwar development, the evolution of labour and employment practices and in particular the socio-economic role of women in contemporary Japan.

Female labour and textiles in prewar Japan There is little doubt that the Japanese textile industry was a leading sector of Meiji industrialisation. Beginning in the 1870s, the industry developed to become the world’s second largest producer and exporter of cotton goods by the 1930s. A distinctive characteristic of this industry was the high proportion of women among textile workers, consistently at levels of over 80 per cent throughout the prewar period. During the Tokugawa era, cottage industry had rapidly expanded within the context of it being supplementary income to agricultural households, and female labour had been crucial.

1 Directed by Yamamoto Satsuo and based on the 1968 book of the same title by Yamamoto Shigemi.

2 A historical legacy Women becoming migrant waged labour was consistent with the notion of earning auxiliary income for the family unit. However, what was different from proto-industry was that many now travelled great distances to enter factories and lived independently from their rural family homes, primarily in factory dormitories for several years. Female workers were now actively recruited as a cheap source of competitive labour. With the introduction of the first modern mills from 1873, girls were called upon by the Meiji government to ‘reel for the nation’. It was initially women from lower samurai or wealthier peasant families who entered factories, but by the late 1880s it was predominantly young women from indigent farming families who were called upon to migrate from rural villages and urged to work under the Meiji nation-building goals. The advantage of employing females was that they were considered to be more dextrous, docile, obedient and cheap. Japanese women rapidly came to represent the largest sector of the new industrial workforce and formed a crucial backbone to Japan’s industrial revolution. However, as young unmarried women they were also viewed as temporary and expendable labour. An early work focusing on females in the industry was Hosoi (1925). Hosoi worked in a spinning factory and recorded both his experiences and the working conditions of female factory girls, describing the misery of conditions from the 1880s to 1920s, criticising the factory system and calling for more social recognition of the female workers. His work portrayed characteristics of labour relations in the industry, which came to be almost traits of the industry, including the recruitment contracts between girls’ fathers and factories, the confinement of young girls in dormitories and the exploitation of young female labour. His treatise became so famous that the title Joko¯ Aishi (Pitiful History of Factory Girls) remains as a phrase encompassing the historical image of the industry even today. Hosoi set a tone of ‘dire labour conditions’ and ‘exploitation’, themes which were picked up by other prewar works, including his contemporary Sakura’s (1927) work on the miserable conditions of female silk workers in the Okaya Silk Mills. This overriding image of Joko¯ Aishi outlived the conditions themselves as they saw some improvement in the interwar and postwar periods. In general, traditional scholarship came to view factory girls as predominantly passive victims of the industry, an unskilled and cheap labour source ‘sold’ into the industry by the patriarchal heads of rural households. Labour conditions were certainly dire during the mid-Meiji period. Women in cotton spinning worked gruelling 12 hour shifts, continually on their feet at the machines, typically with three rest breaks totalling only an hour per shift.2 Sanitation in the factories and dormitories was poor, with

2 Typically, half of the girls began at 6:00 a.m., while the other half did the night shift beginning at 6:00 p.m. This was known as the double-shift system and remained in place until the night ban came into force in 1929.

A historical legacy 3 girls sharing grimy bathing facilities and often bedding. Occupational ailments were widespread, ranging through trachoma, bronchitis, acute weight loss, intestinal problems and tuberculosis. Early reports into labour conditions focused on this issue of the physical well-being of the workers, including Ishihara (1913), who presented research into workers’ health and illness in textile factories carried out prior to the promulgation of the Factory Law, focusing on both factory reports and follow-up surveys of those girls who had previously worked in factories and returned to their native villages. Government and industry concern was particularly focused on the issue of tuberculosis, and the concern that factory girls were transporting the disease back to the rural areas. The history of this issue is found in works such as Kagoyama (1970) (a reprint of Ishihara’s original work), Hunter (1993b) and Komatsu (2000). The recording of the everyday lives of factory girls has contributed to a detailed understanding of their history. Takase (1972, 1979) looked at the history of female workers in the Tomioka silk mill, based on the author’s survey of female workers’ gravestones, factory registers, prefectural histories and local historical newspaper reports. Yamamoto (1979, 1980) is a two-volume series recounting the history of the many young girls who crossed the famous Nomugi Pass to enter employment in regional silk factories. Tsurumi (1990) describes the working lives of factory girls (ko¯jo) during the period from 1872 to 1912. Focusing on aspects such as recruitment, working and living conditions within factory and dormitory compounds, and the working girls’ interaction with employers and their families back home, Tsurumi (1990) uses personal diaries, factory songs, letters and documentation from the period to tell the story from the perspective of the daily lives of the young female operatives. She tells how they worked for a few years to supplement the incomes of their families, who lived in poor rural areas, or to save up for a dowry. Although, as she herself agrees, such written accounts should be treated with caution, her analysis of the records provides a rich commentary on the daily conditions for female workers during this period.3 An excellent account of the prewar years is provided in Hunter (2003). Based on extensive research into the workings of the female textile labour market during the decades before the Second World War, the story of women and textiles is placed firmly within Japanese economic history. The issue of recruitment has been an important one in the story, serving to highlight both the practices of the industry and the historical link between the textile mills and the poor rural areas where recruitment took place. Hunter describes the methods employed by textile industries to recruit

3 Other recorded accounts include Tojo (1990), which looks at the registration of female silk operatives, Yamanouchi (1975), which is an autobiography of a young spinning operative and Hunter (1984b), which is based on the diary of a silk operative.

4 A historical legacy young female workers from the rural areas between the 1870s and 1930s. The heavy dependence on the recruitment of female workers was partly a result of the textile industries being traditional proto-industrial employment for females as well as the belief that women were more dextrous on spindles, but for the most part was based on the fact that young females were an extremely cheap source of labour. Competition to recruit girls became so severe that the dominant method of recruitment was for factories to employ a recruiting agent (boshu¯ ju¯jisha). Hunter notes that recruitment strategies often used deceptive tactics to hire girls and work conditions did not live up to promises made. Such policies, while advantageous to employers and to the economic development of the textile industry during this period, no doubt contributed to problems of high turnover and the continued low status of female workers. Saxonhouse (1976, 1977) and Saxonhouse and Kiyokawa (1985) used econometric methods to examine the high turnover in the industry, noting that the industry relied upon a labour force of young women who worked on average for no longer than two years. The high turnover in the industry was attributed to young girls ‘escaping’ from the factories because of unhappiness with poor working conditions. They noted, however, that the industry was able to improve upon the organisation, quality and productivity of the labour source, despite the continuance of high levels of absenteeism and turnover, through improvements in technology, accumulating worker experience and increasing levels of basic education. The use of female workers in Japan’s prewar textile industry received much international attention. Japan’s rival powers in the West persistently accused Japan of making its industrial advance based on social dumping, that is the exploitation of cheap (female) labour, thereby having an unfair wage–price advantage.4 The employment conditions of female textile operatives therefore became the focus of both national and international debate surrounding industrial work conditions and workplace reform. While traditional scholarship had highlighted the plight of women in textiles, particularly during the Meiji period, research into the Taisho¯ and early Sho¯wa period, particularly the interwar years, highlighted that worker consciousness and industrial labour movements led to some improvement in working conditions. Some revisionist arguments have also suggested that the conditions for female factory girls were not as bad as those of their counterparts, such as those who were forced to work in brothels or who remained behind in poor rural households, as well as relative to the conditions for female textile workers in other nations at similar periods of industrialisation. What is clear from the historiography is that some movement towards gradual improvement in labour conditions in the industry

4 Scholars commentating at the time, such as Utley (1931) and Nawa (1948), suggested that the rapid progress and success of the Japanese cotton textile industry could be attributed to the low wage costs of the predominantly female workforce.

A historical legacy 5 took place through pressure on the industry, legislation and partly as a result of worker activism. The involvement of women in labour disputes and early activism has therefore been an important focus of research. Strikes had been recorded as early as 1886, though they became more frequent and effective in the 1920s.5 Tsurumi (1995) notes the ‘small acts of everyday resistance’ utilised by the girls.6 Labour activism among women aimed at improving their working conditions, and gradual reform took place. The abolition of night work (between 11:00 p.m. and 5:00 a.m.) officially shortened working hours after 1929,7 though not necessarily per worker, and the textile industry played an influential role in the establishment of this early factory legislation (see Akamatsu 1977; Kojima 1983; Hunter 1989). Shimazu (1970) viewed factory women as having been integral to the existence of a proletarian labour movement in Japan, while Shibata (1975; Shibata and Yamada 1983) noted that the struggles of working women throughout Japan’s modern industrial history played an important role within the history of the grassroots labour movement. While the majority of female textile workers were employed out of economic necessity, it is believed that many used the opportunity to develop an ethos as workers and establish a degree of independence and self-fulfilment. Although scholars like Saxonhouse had noted the high turnover in the industry, attributing it to young girls fleeing from the factories, others commented that this turnover did not solely suggest that the girls left to return to the country and marry, as many actually moved between factories in attempts to gain skills and better working conditions, with many eventually marrying and remaining in urban areas. Such arguments pointed out that it was important that female operatives should not be seen purely as having been ‘expendable farm daughters’. Bernstein (1988) suggested that the earnings of factory girls provided indispensable support to the rural ie (household) and enabled rural society to persist alongside industrialisation.8 Although, as various traditional studies have noted, many workers were no doubt unhappy, a key point is that many were nevertheless highly motivated, and achieved subtle gains. Despite early coercive motivational

5 A strike by female silk reelers over wages and working hours at the Amamiya Silk Spinning Mill was the first recorded strike in modern Japanese history, and set the scene for several strikes in Yamanashi prefecture around the turn of the century. Later strikes over working conditions such as those at Fuji Boseki (in 1927) and Toyo Muslin (in 1930) achieved improved conditions. For details see Bernstein (1988) and Molony (1991). For details of the To¯yo¯ Muslin, Kanebo¯ and Kishiwada disputes, see Suzuki (1989), and for a collection of interviews of persons involved in several prewar labour disputes, see Watanabe (1980). A concise history of the labour strikes in the industry can be found in Ro¯do¯ So¯gi Cho¯sa Kai (1956), and for the silk industry, see Matsumoto (1991). 6 Tsurumi (1995) p. 18. 7 The ban on night work was a revision to the Factory Act 1911 and took effect on 1 July 1929. 8 See Bernstein (1988) pp. 54–77.

6 A historical legacy techniques, industrial employers gradually responded to women by setting up various factory incentive schemes such as attendance prizes, which motivated the young workers and gave them a sense of pride.9 Factories also established education and lessons within dormitories for the girls and they were better educated than their counterparts in other countries at a comparative stage of development. Sugihara (1986) argued against Saxonhouse and Kiyokawa’s (1985) suggestion that young female workers entered factory employment with limited objectives, which produced low levels of job attachment and high absentee rates. Sugihara (1986) noted that an exchange of social values occurred between the traditional and the modern as the female dekasegi (migrant workers) moved between industry and rural family households, which evolved from a traditional sense of loyalty to a modern sense of work commitment. Other research has also described the increased consciousness of the female textile workers and recognised not only their contribution to industry but also their influence upon socioeconomic structures of the time. Molony (1991) noted that the farm girls were not naive, having already played a valued role in rural by-employment, and often initiating the economic decision to become dekasegi and supplement the family income. The development of communal living in the dormitory system, though of a poor standard, gave women the chance to develop a degree of collective perception as workers. As women gained in strength as an organised labour force, concern for their conditions of employment and calls for legislation were not solely aimed at female textile workers, but at all women. The abolition of night work for women workers was closely tied to the preservation of motherhood and how this primary role for women should dictate their position in the labour force. In line with women being banned from political activity, legislation was enacted to preserve and prioritise women’s primary role as the linchpin of the domestic family system, rather than as economic producers serving the developing nation. Gender had become an allpervasive concept in the controversial debate over the abolition of night work. Supporters for the ban were predominantly the state and leading intellectuals, although a few women’s groups also participated. The major opponents were employers, who initially viewed female workers not strictly as a gender group but as an economically productive group. The central argument for protective legislation was that women’s ideal roles were as wives and mothers, and that even unmarried textile workers must be protected for their future reproductive roles.10 The 1929 revisions to the

9 See Sugihara (1986). Many factories employed regular attendance prizes, to which the response by many girls was ‘over-motivation’, that is, they worked even when sick. 10 Although concern was also expressed as to the health of male workers and their need to be protected for the sake of their families, the ban was never extended to include men, and it should be noted that organised male labour supported the ban precisely because they wished to curtail the power of female labour.

A historical legacy 7 Factory Act, Japan’s first legislation relating to the issue of maternity, played a significant role in creating a discourse on ‘motherhood protection’ (bosei hogo) which would shape feminist debate into the postwar period. In effect, the ban on night work implemented for the first time a legal division of labour by gender, which was to have implications for the postwar development of Japan’s modern employment structure. In general, traditional research on the prewar history of labour in the textile industry often criticised and questioned the industry’s decision to employ young female workers. The industry was seen as having viewed this labour force as being of a docile, subservient and easily malleable nature. This was deemed to be a result of the isolated rural areas from which they were recruited, the general social backwardness of these areas, and the social environment of the period which made them subservient to the control of their fathers under the patriarchal ie (household) system. Revisionist research has, however, suggested that the long history of female workers in the textile industry resulted in an increased consciousness of these women as workers and produced an experienced industrial labour force. At the very least, there is no doubt that women workers were active in prewar textile strikes and had some influence on improved labour conditions by the 1930s. Overall, the historiography indicates that the contribution of young female workers to the prewar textile industry was significant. While Koyama (1961) had stated that young factory girls were ‘largely ignorant’, ‘satisfied with poor pay’ and ‘unorganised’,11 Tsurumi suggested that by the 1920s female textile workers were conscious that they belonged to a group called factory women.12 Their status as victims was, as she pointed out, relative. Overall, the women do not appear to have thought that they had been singled out for special victimhood. Their stories stress that the kind of experiences they encountered were quite ordinary for women of their class and time. (Tsurumi 1995, p. 27) Moreover, she commented that: They often saw themselves as active, dynamic figures who had the power to make substantial contributions to their families’ welfare. They sometimes saw themselves as the major force behind production of wealth in the textile industries. (Tsurumi 1995, p. 33)

11 Koyama (1961) p. 13. 12 Tsurumi (1995) p. 30.

8 A historical legacy Both Ishii (1972) and Nakamura (1994) suggested that it was the dexterity and skill of young female workers and not efficient machinery or high quality raw materials which formed the backbone of the Japanese silk industry. Francks (1992) noted that by the interwar period there was a relatively skilled and high-productivity female labour force in the textile industry, but that it had not been a smooth or costless process. Tsurumi (1990) concluded that the contributions of young females in textiles to the industrialising economy were significant, although at high levels of personal, physical and social sacrifice. It is evident that several key themes emerge from the research on the prewar industry – the issue of recruitment, the issue of dormitories, the issue of economic exploitation (low wages) and the image of Joko¯ Aishi – all of which can be said to have left an important legacy for the postwar industry to deal with. It is clear that any continuation of the historiography into the postwar period has to take into account not only this ‘legacy’ of the prewar history, but also any socio-economic transition in the postwar years. This will be kept in mind as this book tells the story of women working in the textile industry during the postwar years. It is perhaps of little surprise, given the overriding importance of the textile industry within Japan’s prewar economy, that the bulk of the literature and research on female textile workers has, thus far, been mainly confined to the prewar years.13 However, the continuing story of women working within this industry after the Second World War is an equally important one. It is hoped that this study will not only depict this, but moreover present a picture of the historical process that led to particular roles for women and the emergence of an economic structure of female employment in postwar Japan. The findings of this analysis can then contribute to the reconsideration of Japan’s ‘economic miracle’ from a new perspective focusing on women workers and give rise to further debate on the implications of gender in economic development.

13 An exception is some coverage of the first decade following the end of the Second World War, by Shimazu (1953) examining female workers in large-scale cotton spinning factories and Kajinishi et al. (1955) detailing the silk industry. Two other key works were Seki (1954) and Shindo (1958), both of which were translated into English. While these two volumes provide a useful picture of the industry during these early postwar years, it must be noted that both authors were from the industry’s business community and a central agenda in their work was to provide a ‘correct knowledge and understanding’ of the industry for foreign observers and defend against the international attack on the Japanese cotton industry as having carried out ‘social dumping’. Neither work focuses on the specific dynamics of the female labour market in great depth. Other Japanese works that include references to the very early postwar period are sparse in nature, but include Hirosaki (1967), Yasui (1967), Saito (1969) and Shiozawa (1976).

A historical legacy 9

The textile industry’s postwar transition While this is a study of female labour rather than the textile industry itself, it is vital to introduce the structure of the industry and outline key postwar trends for the industry in order to provide a context within which to place this research on its employment of women. The textile industry referred to in this study is the manufacturing industry concerned with the production of natural textile mill products using natural fibres.14 Therefore, it does not include the synthetic fibre industry, which has a very different industrial and labour structure and manufacturing process, and which is usually classified as part of the chemical industry. Nor does it include the apparel industry, which produces various finished products made from fabrics. The natural textile industry is generally classified as comprising several sectors, based on different types of processing and materials. These are the spinning industry (bo¯sekigyo¯), which produces yarn and thread from predominantly raw cotton or wool; the weaving industry (orimonogyo¯), which produces rolls of fabric from thread; the knitted fabrics industry (meriyasu seizo¯gyo¯); the dyeing and finishing industry (senshoku seirigyo¯); the silk-reeling industry (seishigyo¯); and a range of other small miscellaneous natural textile industries producing products such as ropes, nets, lace and braid. This study is concerned with these natural textile industries, but mainly with the two largest sectors of spinning and weaving within that industry. The structure of the natural textile industry was characterised during the period by several features: a large number of competing firms; small average size of enterprises, though with large differences across scale within this ‘average’; relatively little product diversification; and ease of entry. There was therefore a substantial number of small-scale (often family-owned or family-run) enterprises, particularly in weaving, still in existence alongside the large-scale multi-mill (multi-process) firms during this period. In this respect it was still a labour-intensive industry, and conditions were liable to variation across the scale and range of enterprises. The central umbrella organisation in the industry was (and still is) the Nihon Bo¯seki Kyo¯kai (Japan Spinners’ Association), hereafter referred to as ‘JSA’.15 Despite its name, the JSA dealt with all sectors of the textile industry, and not just spinning. During the period, the JSA classified its affiliated companies into three main groups. The largest companies (o¯tekaisha) were known as the ‘Big Ten’ cotton spinning companies. They were To¯yo¯bo¯, Kanebo¯, Nichibo¯, Fujibo¯, Nisshinbo¯, Kurabo¯, Daiwabo¯,

14 The industry is classified as bo¯shokugyo¯ (spinning and weaving) in the Population Census of Japan (Kokusei Cho¯sa Ho¯koku) and as sen’i ko¯gyo¯ (fibre or textile manufacturing industry) in the MITI Census of Manufacturing Industries (Ko¯gyo¯ To¯ kei Hyo¯). It is commonly referred to as sen’i sangyo¯ (textile industry). 15 This was formed as Dai Nippon Bo¯seki Rengo¯kai (All Japan Spinners’ Federation) in 1882. It was re-formed as the current JSA in 1946: see JSA (1982).

10 A historical legacy Shikibo¯, Nittobo¯ and Kureha. The Big Ten were formed during the prewar period, and tended to carry out mass production of textile products as well as operate various divisions concentrating on different processes. They remained as the predominant textile firms during the postwar decades, though with some adjustments. To¯yo¯bo¯ acquired Kureha in 1966 (making it the ‘Big Nine’),16 and Nichibo¯ merged with a synthetic company Nippon Rayon to become Unichika in 1969. The next groups of companies were the Shinbo¯ (‘New’ spinning companies) and the Shinshinbo¯ (‘New-New’ spinning companies).17 Although some of the New companies were established prior to or during the Second World War, they were given the status of ‘newcomers’ or ‘new-newcomers’ based on the timing of their membership in the JSA. In comparison to the large-scale Big Ten companies, with their long prewar history and association with the JSA, they were seen as younger companies, and were predominantly smaller in scale when compared to the Big Ten. Under JSA affiliation, and under industry plans to unify the industry during this period, they essentially acquired many of the features of the Big Ten companies. Several New companies became as prominent, at least in terms of scale of production, as the Big Ten. Despite being in competition, the textile firms often shared information on labour policies, as well as technology, raw materials and markets, and came together for joint discussions, particularly under the industry association JSA, but also under the national textile union federation. While the structure of the textile industry was such that small-scale enterprises made up around 80 per cent of the total number of textile enterprises at the start of the period, it was large-scale enterprises that controlled production. According to the JSA, the Big Ten cotton companies together produced about 70 per cent of all exported cotton goods at the start of the period.18 Many of the large number of small enterprises at the other end of the industry specialised in weaving and dyeing, often through indirect subcontracting links to the big spinners via trading companies.19 Not all textile firms were members of the JSA, though a majority of spinning enterprises were. The JSA was essentially an industry (employer) organisation, where the Big Ten and others met to discuss policy. The association sought to promote co-operation (e.g. joint wage negotiations) among

16 While the Big Ten officially become the Big Nine after 1966, the ‘Big Ten’ is a term often used to describe these top cotton spinners, because of their long history dating from the prewar period. Therefore, this study will make use of this term to refer to the Big Nine or ¯ mi Kenshi later became included in the Big Ten following its growth Ten companies. (O during the late 1970s and 1980s.) 17 This study will refer to them as New and New-New using capital letters for ease of reference. 18 JSA (1954) p. 4. 19 See McNamara (1995) pp. 151–53 and Nakamura (1995) p. 168.

A historical legacy

11

members, and also filtered information between textile companies and government. In terms of spinning, members of the JSA (Big Ten, New and New-New firms) represented around 80 per cent of total natural and multifibre spinning capacity, while the remaining (small-scale) 20 per cent was found in sanchi (producing areas). Weaving was more predominant outside of the JSA (and in sanchi), because of the range of small-scale and familyrun weaving establishments. In general, therefore, small-scale firms were more likely to fall outside of the JSA umbrella. However, the power of the JSA as an industry association for medium and large scale enterprises, meant that the (labour) patterns and conditions established in the larger enterprises during these years also affected small firms outside the JSA. While there is little doubt as to the importance of the textile industry in Japan’s national economy prior to the Second World War, its role within the postwar economy requires some introduction. Although it gradually became a declining industry during the later years under study, the industry continued to play an important role in Japan’s economic development during the early postwar years under consideration. Although the years immediately after the Second World War will also be discussed, the two decades from 1955 to 1975 have been chosen as the key period of study for several reasons. First, this was the period of high economic growth, as well as growth in the numbers of women entering the Japanese labour force, and as a large employer of women, textiles offers an important industry study of these trends. Second, although the textile industry is in relative ‘decline’ during this period, it is still a significant industry within the early postwar Japanese economy, particularly in the 1950s and 1960s. Until the early 1950s, the industry was regaining its feet under reconstruction programmes, but the second heyday for the industry during the period from the mid-1950s to the mid-1970s allows examination of an industry which is still significant as an employer and producer within the manufacturing economy, but one that is requiring structural adjustment and undergoing transition during the high-speed growth era. Finally, from the late 1970s, the industry began substantially to rationalise its labour force under technological modernisation and at the same time began to both diversify away from textile production alone, as well as move some of its textile investment abroad. In many ways, the textile industry was in a much different position after the mid-1970s, as it applied for ‘depressed industry’ status. The Japanese cotton industry was brought to its knees by the end of the Second World War. Two-thirds of its prewar cotton spindles were scrapped by wartime administrators, and bombing and destruction of urban areas had caused a further loss of 20 per cent of spinning and 14 per cent of weaving capacity.20 Following the end of the war, the textile industry saw great

20 In 1937 the cotton industry had operated with 12,165,000 spindles; by the end of 1945 it had only 1,115,000 spindles in an operable condition. Moreover, supplies of raw materials

12 A historical legacy changes in both its internal structure and the economic environment within which it operated. As economic recovery and restructuring programmes took place during the Occupation, the industry was rebuilt and began to export again. With the first improvements in the general economic situation, a main focus within the industry became the improvement and installation of new machinery, and the amelioration of working conditions for labour. The period up to 1951 was a period of recovery to prewar productivity levels. Thereafter, labour productivity in the industry began to increase beyond prewar levels and moved into a new phase. Like many industries in the early postwar period, the Korean War (1950–53) gave a kick-start to the textile industry, as it increased exports and supplied United Nations (UN) forces. The subsequent end of the Korean War brought a sluggish domestic market and contributed to a brief recessionary climate in the industry in 1954–55. While the Korean War had brought economic revitalisation to the industry and led to a large increase in production workers, the first postwar curtailment of operations was enforced at the recommendation of the government in March 1952. A ‘human rights’ strike by labour in 1954 and a subsequent wage strike in 1955 made progress difficult for the industry. The curtailment of operations since 1952 was felt by the industry to have contributed to dissension between labour and management, particularly over the issues of wages and reductions in employee recruitment.21 However, the following year saw improved conditions and was noted as particularly remarkable for its general increase in manufacturing employment after the stagnation that had occurred in the national economy.22 In 1956, the curtailment of operations in the textile industry was lifted, and this signified the start of more prosperous conditions, and led to both an increase in production (particularly from 1959–60 onwards) and a rapid increase in labour. Demand for increased numbers and recruitment of new female employees was a feature of the good economic conditions during the next few years, as well as improvement in labour conditions, including working and living conditions, wages and welfare. However, the industry was beginning to comment on what it saw as the emerging problem of ‘labour instability’. The beginnings of a labour shortage of young workers was identified, as well as the concern that the industry was losing its appeal as an industry of desired employment within an increasingly competitive environment for female labour (see Chapter 3).23 During the late 1960s and early 1970s the industry made serious attempts to rationalise through the modernisation of equipment, however, despite improved technology and labour productivity, the industry was still were virtually non-existent and power and fuel supplies were restricted and unreliable (Walton 1958) p. 696. 21 Nenju¯ no Bo¯sekigyo¯ (1955) p. 81. 22 Nenju¯ no Bo¯sekigyo¯ (1956) p. 74. 23 Nenju¯ no Bo¯sekigyo¯ (1963) p. 77.

A historical legacy

13

recording difficulty not only in recruiting additional labour, but also in its attempts to replenish the workforce.24 The good economic conditions recorded in the 1960s, as the national demand for textile products increased especially for natural textiles, meant that the industry was once again a leading Japanese industry playing a vital role in economic growth. At the same time, however, increased competition (including competition for labour) from other manufacturing industries, particularly electronics, was affecting the industry. Economic shocks in 1971, with the Nixon Shock and the United States–Japan textiles issue, and in September 1973 with the global oil shock quickly led to somewhat depressed conditions within the industry and the curtailment of recruitment, particularly the hiring of temporary female workers. A depression cartel was enforced in the industry from the first half of 1975, although the absorption of the effects of the economic shocks was relatively quick, and with the abolition of the cartel later in the year, demand for production and labour again increased. However, just after the period under study, economic conditions for the industry worsened, and, in 1978, the natural textile industry petitioned successfully to be designated as a depressed industry.25 In many ways, conditions for the textile industry followed national economic cycles during these years, with temporary inflationary booms and recessions within an overall climate of rapid postwar economic growth. These decades were, however, a significant period of structural transition for the textile industry in terms of its overall importance within the national economy (internal factors), as well as within a changing global market for textile products (external factors). General trends included changing demand for textile products from consumer demands for quantity to demands for quality; a steady decline in the prices of natural fibre products; change in emphasis from overseas to domestic markets; and an important shift within textile production away from natural textiles to synthetic textiles. Changing technology and new methods of producing petrochemicals contributed to a growth in the use of chemicals as raw materials for synthetic fibres, such as polyester and acrylic. The production of synthetic non-cellulose fibres in Japan increased tenfold between 1960 and 1974. Much of the growth in synthetic textiles can be attributed to changing technology and consumer demand. This shift in the textile industry as a whole (natural vis-à-vis synthetic fibres) greatly affected the position of the natural fibres processing industries. A decline in export demand was a significant reason for the relative stagnation and decline of the natural textile industry by the mid-1970s. The increased technology and increasing competitiveness of the natural textile

24 Nenju¯ no Bo¯sekigyo¯ (1970) pp. 44–45. 25 See McNamara (1995) pp. 38 and 140–41 and Nakamura (1995) pp. 230–33. This legislation remained until 1983.

14 A historical legacy industries that developed in other emerging Asian economies led to a lower demand for Japanese textile imports by these countries. The volume of textile goods as a proportion of Japan’s total exports by value declined from 37.3 per cent in 1955 to 13.5 per cent in 1965, and to 5.3 per cent in 1975.26 In 1973, for the first time since the early 1950s, Japan became a net importer of textile products. At the same time, the textile industry’s importance within the national economy also declined relatively because of the rapid growth in other manufacturing industries in Japan during this period. The production of natural textile fibres was, however, still considered to be an important industry. Japan still produced around 25 per cent of total global cotton yarn production in the mid-1970s.27 While in many ways these postwar decades represent the restructuring of an industry seemingly in decline, scholars have commented that, unlike many of their Western counterparts, Japan’s large-scale textile companies have survived the restructuring of their declining industry, through corporate structural adjustment.28 McNamara (1995) stated that as ‘widely recognised leaders in their textile industry, the moguls (Big Ten) with their large workforces, extensive exposure with commercial banks, and close ties to Diet members were not likely to be sacrificed in any government-sponsored restructuring programme.’29 A general view is that the cotton textile industry was restructured and ‘phased out’ in Japan rather more successfully than it was in other nations, including the United Kingdom. It is certain that the flexibility of the industry’s predominantly female workforce played a key role in the ‘success’ of that ‘decline’. Survival has meant different strategies among the Big Ten, particularly after the period under study. To¯yo¯bo¯, Kanebo¯ and Unichika extended their operations during the period to synthetic fibres, and were listed as the top three (natural) spinners as well as in the top nine synthetic fibre producers in terms of assets and sales in the mid-1990s. Nisshinbo¯ owned just under 25 per cent of the total shares in To¯ho¯ Rayon, also one of the top nine synthetic fibre producers.30 Kanebo¯ also diversified into non-textile products during the 1960s and is now distinctive within Japan as a producer of cosmetics and pharmaceuticals. The remaining Big Ten firms have remained as prominent cotton spinners, with some diversification into polyester/cotton blend production as well as a wide range of non-textile products. Textile firms not

26 Nakamura (1995) p. 66 (table 3.3). 27 Park (1976). In terms of synthetic yarn production, Japan was also the world’s second largest producer behind the United States. 28 For studies of textiles’ structural adjustment during decline, see Ike (1980), Dore (1986) and McNamara (1995). The natural textile industry has declined since the 1970s not only in Japan, but also in Europe and the United States. See Toyne et al. (1984) and Singleton (1997). 29 McNamara (1995) p. 5. 30 McNamara (1995) pp. 8–10.

A historical legacy

15

only diversified their corporate portfolios within and outside of natural textile production, but also took production overseas, particularly during the 1970s. Product specialisation during the 1970s saw more sophisticated textile products with higher value-added content. Profits in the natural textile industry during the period, however, were generally modest, although quite variable, and not on par with other manufacturing industries.31 Restructuring and survival did not always allow for huge profits. Overall, the Big Ten and large-scale enterprises have remained both as prominent industrial firms within Japan and as global textile producers. In this sense, the ‘decline’ of the industry is not as simple as it appears, and the history of these companies as both prominent corporations and large employers of Japanese female workers deserves recognition. While various academic works have examined the textile industry during this period as a case study of Japanese structural economic adjustment and ability to shift industrial structure away from declining industries to new competitive ones, there has, however, been no detailed examination of the situation of labour and the position of women within the industry during this period.

The importance of women in textiles When considering the postwar trends of the industry, of utmost importance for this study is the fact that large numbers of female workers continued to be employed in this industry. The industry’s postwar years present an interesting period for study, not least because the industry continued to be a prominent employer of women throughout the early postwar decades, both in terms of sheer numbers and in the development of employment policies, but also because the textile industry can be representative of the large numbers of women working in production departments and factories during the high-speed growth years. It is therefore important that the industry’s historical significance as a key employer of female workers be recognised. Its labour and management policies were characterised not only by attempts to respond to changed circumstances but also by the burden of history and the difficulty associated with breaking free from earlier established patterns of operation. At the same time, its position can be placed within a broader context, and this study will argue that the industry took a leading role in formulating employment strategies for female workers (both prewar and postwar) and as such had a decisive influence on the segmented nature of the labour market that developed for women workers in Japan. While this book is essentially the study of women in one industry, it does embody many much broader issues. The historical development and evolution of a system of female employment in these early postwar decades was crucial in structuring the female labour market portrayed in much of the

31 Park (1976) p. 22 and McNamara (1995) pp. 72–73.

16 A historical legacy contemporary literature on women and work in Japan. In other words, the decades studied here witnessed the genesis of many of the issues and ‘problems’ that continue to characterise the patterns of women working in Japan today, and the textile industry was a key player in this evolution. The early use of temporary factory workers in textiles, for example, gradually developed into a system of ‘part-time’ female labour, the analysis of which has been the focus of much of the recent works on the female labour market in Japan. Within the historiography on Japanese women working, the 1980s and 1990s, particularly the post-EEOL era,32 have been a major focus of the literature. However, the postwar period prior to the 1980s remains a crucial period of study. This is the era when Japanese women worked within a new and rapidly evolving socio-economic environment and under new employment conditions provided by legislation such as the Labour Standards Law 1947. This earlier postwar history as well as the dynamics of how female employment operated at the micro (industry) level deserves more attention, and this book seeks to go some way to filling both of these voids in the current research. The decades 1955 to 1975 represent not only another (postwar) ‘miniheyday’ for the textile industry, but also a period of crucial restructuring of female labour within the Japanese postwar economy. Many supply and demand factors are at work, influencing the nature of female labour participation and creating a postwar pattern and structure of female labour that is both new and a product of the history of women working in the Japanese economy. At the same time, the social environment is under transition and interacts with the economic changes. Four main trends can be identified during these years: first, the increasing numbers of women entering the labour force; second, a shift in the industrial structure away from agriculture towards the manufacturing and service sectors; third, the changing socio-economic status of Japanese women; and fourth, the emergence of older married women in the female (non-agricultural) labour force. While many of these changes also occurred in other industrial societies, the speed at which these changes took place and emerged as enduring structures in these twenty years in Japan was astonishing. Such important changes in the Japanese female labour market indicate that any industry study like this one cannot stand alone and must be evaluated within the national context, and several key questions can already be asked of the industry. If increasing numbers of women were entering the labour market during these years, what were the effects on an industry that had long dominated the female labour market? If economic growth was creating increased demand and opportunities for women in the labour market, what effects did this competitive environment have on the textile industry? If older women were increasing their presence in the postwar

32 EEOL is the Equal Employment Opportunity Law 1986.

A historical legacy

17

labour market, what did this mean for a traditional employer of young female labour? As a prominent employer of female labour, what was the position of women within the industry and how did this change over the period? What was the industry’s response to all these socio-economic developments and how did the employment system for women evolve during these years? These are important questions that will be answered as the following chapters analyse the case of female textile operatives in depth. However, this chapter first provides an introduction to the demographics of women in the industry during these years. Textiles was the largest employer of women among the Japanese manufacturing industries during these years (see Table 1.1). Other prominent employers of female labour were the related apparel industry, the food industry and the rapidly growing electronics industry. Textiles’ proportional employment of women, as a percentage of total manufacturing employment, was just above 40 per cent in 1955, although this had declined to 15 per cent by 1975, due in part to some decline in numbers in textiles, but mainly as a result of a rise in the numbers of women employed across all the main manufacturing industries. Within textile manufacturing, around two-thirds of women were employed in the main spinning and weaving sectors (see Table 1.2). On average, 70 per cent of all workers in the textile industry were female during these years (though this declined to around 65 per cent in the 1970s), and this proportion of females within the textiles labour force was predominantly higher than that found in other manufacturing sectors. The related apparel industry showed a similar dominance of female workers, while all other manufacturing industries showed an increase in the proportion of women within their labour force during the period. The proportion of women employed in the food and electronics industries in particular showed an increase, growing from around one-third of the labour force in the mid1950s to around half of the labour force by the early 1970s.33 Thus, despite a relative decline in the number of women within textiles and the decline of the industry itself by the mid-1970s, overall, the industry can be seen as a continuing stronghold for female employment during this postwar period. In fact, by occupation classification, it was ranked as the third largest occupation for women during these years.34 The textile industry’s employment of women was therefore significant, both within the manufacturing sector and within the national female labour market.

33 For detailed analysis see Macnaughtan (2001) pp. 88–90. 34 The classification of ‘textile operative’ ranked third behind the classifications of ‘clerk’ and ‘service worker’ during the period. Despite the growth of the service sector, it was therefore still a prominent specific occupation for females in terms of numbers during the 1950s, 1960s and early 1970s, ranking ahead of other ‘female jobs’ such as nurses, teachers, hairdressers and telephone operators. The classification of ‘electronics operative’ grew from twelfth place in 1955 to fourth place just behind textile operative by 1970. Data from Kokusei Cho¯sa Ho¯koku. See also Burke (1979) p. 226.

1,653,732 2,651,954 3,590,911 4,356,365 4,053,195

679,103 824,667 910,132 826,061 609,342

(No.)

41.1 31.1 25.3 19.0 15.0

(%)

(%)

85,550 5.2 133,576 5.0 224,908 6.3 309,981 7.1 405,164 10.0

(No.)

Apparel

190,669 322,123 524,778 576,159 608,504

(No.)

Food

11.5 12.1 14.6 13.2 15.0

(%)

(%)

63,346 3.8 257,745 9.7 338,165 9.4 632,742 14.5 515,509 12.7

(No.)

Electronics

61,312 125,812 188,564 275,310 274,663

(No.)

Metals

3.7 4.7 5.3 6.3 6.8

(%)

679,103 824,667 910,132 826,061 609,342

100.0 100.0 100.0 100.0 100.0

67,494 37,676 28,158 20,651 13,747

9.9 4.6 3.1 2.5 2.3

181,699 268,639 265,017 233,060 137,916

(No.)

Spinning

26.8 32.6 29.1 28.2 22.6

(%) 300,263 326,080 347,666 269,749 193,950

(No.)

Weaving

44.2 39.5 38.2 32.7 31.8

(%)

(%)

42,269 6.2 70,203 8.5 122,307 13.4 155,564 18.8 145,217 23.8

(No.)

Knitted products

Notes Figures for 1955 and 1960 are only for establishments employing four or more persons. Figures for 1965 to 1975 are for all establishments.

Source: Kokusei Cho¯sa Ho¯koku (Population Census of Japan).

1955 1960 1965 1970 1975

(%)

(No.)

(No.)

(%)

Silk reeling

Total textiles

Table 1.2 Breakdown of female employment in textile manufacturing

Notes Figures for 1955 and 1960 are only for establishments employing four or more persons. Figures for 1965 to 1975 are for all establishments.

35,342 53,761 61,853 59,334 51,658

(No.)

5.2 6.5 6.8 7.2 8.5

(%)

Dyeing and finishing

2.4 3.9 3.8 4.9 4.9

(%) 21,332 37,131 44,530 57,361 48,192

(No.)

1.3 1.4 1.2 1.3 1.2

(%)

Iron and steel

14,965 16,846 17,414 14,928 12,069

(No.)

2.2 2.0 1.9 1.8 2.0

(%)

37,071 51,462 67,717 72,775 54,785

(No.)

5.5 6.3 7.4 8.8 9.0

(%)

Ropes and nets Misc. textiles

40,403 102,390 138,142 213,530 197,632

(No.)

Machinery

Sources: Kokusei Cho¯sa Ho¯koku (Population Census of Japan). Ko¯gyo¯ To¯kei Hyo¯ (Census of Manufacturing Industries).

1955 1960 1965 1970 1975

Total

Textiles

Table 1.1 Females employed in manufacturing sector (main sectors)

A historical legacy 19 The Japanese textile industry, particularly spinning and silk reeling, had a long tradition in the prewar period of employing a primary labour force of very young females, and this tradition continued into the early postwar period. Official industry data on the average age of female workers revealed that it was around 20 years of age from the late 1950s to early 1970s, with the textile industry employing a significantly larger proportion of the youngest female workers (15–19 years of age) within its female workforce during these decades compared with other manufacturing industries. However, from the mid to late 1960s, data on the Big Ten companies revealed that the oldest age group (40+ years of age) of female workers was showing an increase in proportional share – a significant turning point in the female labour structure, as the cotton textile industry (particularly the Big Ten) was traditionally the heaviest employer of young females across the industry.35 The average age of female workers differed across the scale of companies, with older women being much more prominent in small-scale operations (particularly weaving), while the predominance of young female production workers in spinning (particularly in the Big Ten) continued during the postwar period under study. A gradual increase in the proportion of middleaged or older (predominantly married) female workers in contrast to the traditional core labour force of young (unmarried) females during these years is an important trend overall (see Table 1.3). It is more important than official industry documentation reveals, particularly if one considers that the JSA data mostly did not include temporary workers, who were generally older females, suggesting that the proportion of older women would have had a greater ‘real’ impact on much of the official recorded data on the demographics of female textile operatives. The percentage of older married women in the textile industry’s female workforce increased during these years (see Figure 1.1) due to the effects of a shortage of young labour. In a 1972 survey by the Ministry of Labour, 53.4 per cent of textile companies surveyed were ‘positively thinking of using married females’ while 32.9 per cent retained a policy of ‘using unmarried females as central labour as much as possible’.36 This increase in the proportion of married females within the industry during the period is an important one. The transition from young females, with no dependants and fewer family responsibilities, to older married women with domestic responsibilities affected the institutions of the industry, particularly management policies and strategies. It also offers a case for examining at a grassroots industry level the implications for industry during these years of the national trend for increased participation of older married women in the paid sectors of the economy, in contrast to the prewar period. However, the under-reporting or un-reporting of information on temporary or non-regular labour (essentially older married women) is a common 35 For detailed analysis of age see Macnaughtan (2001) pp. 91–98. 36 MOL (1972) pp. 8–9.

20 A historical legacy Table 1.3 Age breakdown of female textile workers, 1952 and 1972

1952 1972

Average age (years)

15–18 (%)

15–25 (%)

25 and over (%)

30–39 (%)

40–49 (%)

19.6 29.3

40 11.8

90 48.7

10 51.3

17.0

27.7

Source: MOL (1952, 1972). Notes Data for 1952 are for cotton-spinning factories only; data for 1972 are for all textile factories. 1952

1972 Divorced/ widowed 7%

Married 6%

Not married 94%

Married 43%

Single 50%

Figure 1.1 Breakdown of female textile workers by marital status, 1952 and 1972. Source: Men Bo¯seki Ko¯jo¯ no Joshi Ro¯do¯sha: Jittai Cho¯sa Ho¯koku, Women and Minors Bureau, MOL (1952). Sen’i Ko¯gyo¯ ni okeru Fujin Ro¯do¯ Jittai Cho¯sa: Kekka Ho¯kokusho, Women and Minors Bureau, MOL (1972). Notes Data for 1952 are for cotton-spinning factories only; data for 1972 are for all textile factories.

feature during this period, and has important implications for all areas of female employment (e.g. recruitment strategies, management policies, wage and welfare structures), particularly as their employment within the industry expanded during the period. This gradual shift from young to older female workers and the growing importance of older workers within the textile workforce is a key focus in the following chapters. It is important as a central transition within the industry itself, and also as an example of the central shift in the composition of women workers within the Japanese economy during these years. While young female labour remained important for the textile industry during the period, older women were increasingly significant, affecting various employment features of the industry. Another prominent feature of female employment recorded in research on the prewar industry was the consistently high turnover and short

A historical legacy

21

employment service of female textile workers. This can be said to have continued as a key labour issue for textile employers and managers during the postwar period. The average length of female employment in the cotton industry had been extended considerably in the early 1950s following the prewar period.37 The JSA maintained that the change during this period was due to improved labour conditions and amelioration of conditions of employment in the cotton industry. However, length of service cannot necessarily be ascribed to better labour conditions after this period, as after 1955 conditions, particularly in the Big Ten, improved, but the average length of service dropped to three or four years service. By the early to mid1960s, the industry was greatly concerned at the high turnover and low length of service of female employees, which had been a feature of the industry since 1956. While an initial glance at the official leaving rates and length of service figures might have suggested a decline in numbers because of the declining industry during the period, the situation is not that simple. Women were not leaving the industry because they were being laid off, but were consistently moving on from the industry. In fact, the industry was greatly concerned at the declining numbers, even under labour rationalisation plans, and from the early 1960s focused major efforts on improving recruitment and improving benefits within the industry, particularly in an attempt to attract young females into the industry. This included improvements in welfare facilities, as well as opportunities for young females to attend school and gain further educational qualifications while working for the industry (see Chapters 5 and 6). From the early 1960s the JSA aimed to try and alleviate the high leaving rate through management strategies focusing on dormitorybased welfare, recreation and educational facilities.38 While a shorter length of service was a prominent feature of part-time, seasonal and temporary workers when compared with regular workers, onequarter of them were found to have worked for six or more years. Of these, 25.1 per cent were in large-scale companies, 34 per cent in medium-scale, and 41.2 per cent in small-scale, reflecting that those in small-scale companies generally remained in employment with the same company longer. Length of service was lowest in spinning with only 22 per cent working six or more years, while in other sectors the proportion of such workers was much higher, in some sectors at around 40 per cent.39 This indicates that the length of service data for the Big Ten recorded by the JSA was shorter than for the textile industry as a whole, particularly when compared to weaving and small-scale enterprises. A 1970 MOL survey cited

37 Shindo (1961) p. 29 noted that the average service period in 1955–56 of 5 years 3 months was almost three times as long as the average length of service in the prewar period. 38 Nenju¯ no Bo¯sekigyo¯ (1963) p. 81. 39 MOL (1972) p. 52.

22 A historical legacy the average length of service for older female workers at 13 years 3 months in one medium-scale weaving company.40 Like all data, length of service data must be read with caution, as the data are based predominantly on young regular workers and therefore do not take into account the situation of temporary and part-time workers, and do not consider the return of women to the textiles workforce. Even in Japan today, the calculation behind length of service information does not take into account females who leave the workforce and later return. This is primarily because length of service relates to wage and welfare calculations within Japanese companies. During these years females often worked for a few years until marriage and/ or childbirth, took a break for childcare, and later rejoined the workforce for several years. It is this pattern that defined the national ‘M-curve’ model of female employment that was dominant during this period. The dormitory system of accommodation for (young) female workers, which was established by the industry during the prewar period, also continued as a predominant feature of the industry, with 75–90 per cent of female workers recorded as residing in dormitories during the 1950s and 1960s (depending on sector or scale of industry).41 The proportion of workers residing in dormitories had effects on recruitment and management policies. Furthermore, dormitories were a significant welfare benefit provided by the industry and affected the relative value of wages as well as the working and living environment for females employed by the industry. In this sense, they were an important feature of the industry, as a recruitment tool, a management system, and a welfare benefit, and they are therefore discussed in great detail in the following chapters. In general, the trend for the industry is one of continued high rates of female workers residing in dormitories, particularly in spinning. However, some decline in proportion took place as the percentage of female workers commuting to work grew during the period, reflecting a rise in older married non-regular women workers who were less likely to be housed in female dormitories and more likely to commute from their family homes. This was not, however, always clear cut. Female seasonal workers utilised within the industry (see Chapter 3) were, for example, predominantly accommodated in dormitories during their seasonal period of employment.42 The increased use of non-regular female labour was another key trend during these years. A Ministry of Labour survey in 1972 commented that an

40 MOL (1970) pp. 80–81. 41 For detailed data on dormitory residence see Macnaughtan (2001) pp. 103–06. 42 JSA figures indicated that around 70–75 per cent of these female workers were in dormitories during the late 1970s. Figures are from JSA Ro¯mu-bu (Labour Department): 1977: 77.1 per cent; 1978: 79.1 per cent; 1980: 84.5 per cent; 1981: 69.9 per cent; 1982: 76.8 per cent. There are no figures on residence available for the earlier period on seasonal or temporary workers.

A historical legacy

23

average 68 per cent of factories surveyed were utilising temporary, seasonal or part-time labour, with 41.2 per cent of companies using part-time labour and 34.8 per cent using temporary and seasonal labour.43 The use of parttime, seasonal and temporary labour was found to be highest in large-scale companies and correspondingly in spinning factories. Further data, displayed in Table 1.4, indicate that by the 1970s around one-third of female employment in the JSA companies annually was comprised of temporary workers, a significant proportion of the female workforce to have been predominantly unreported or unrecognised in the industry information as well as research to date. Much of this demographic analysis points to the fact that there were increasingly two distinct principal groups of women within the female textiles workforce during these years – young unmarried women living in dormitories and older married commuting working mothers. The characteristics and segmentation of these two key groups of Japanese women workers will be examined in further detail throughout the remaining chapters.

A note on methodology This book is based on my doctoral thesis, which was a qualitative historical study of the employment of female labour in the textile industry, relying mainly on four types of official primary sources of original documentation: industry statistics and national labour data; industry qualitative reports and records; Ministry of Labour (Women’s Bureau) surveys; and individual textile company records and management reports. This work also employed an overlay of oral history based on the findings of personal research and fieldwork. The largest source of statistical and qualitative information to be found on labour in the textiles industry is that recorded by Nihon Bo¯seki Kyo¯kai (Japan Spinners’ Association or JSA). Another source is the Zensen Do¯mei (National Federation of Textile Industry Workers’ Unions or Zensen), a federated union covering industrial sector and corporate unions. National labour data have been compiled primarily from government sources, particularly from the Ministry of Labour and the Ministry of Trade and Industry. The Women’s Bureau of the Ministry of Labour was the source of valuable postwar surveys on female workers in textile factories. Finally, company records and management reports were gathered from the archives of selected textile companies. One problematic feature of the industry sources used for this study was their focus on large-scale cotton textile companies, while information on small to medium establishments proved to be less available. While this is in

43 MOL (1972) p. 10.

51.2 43.8 48.1 51.4 44.0 61.8 46.7 42.5

1969 1970 1971 1972 1973 1974 1975 1976

7.4 5.7 6.7 7.4 8.4 12.2 11.0 10.6

5.4

11.0 11.5 11.8 9.7 9.2 7.8 5.1 6.1

18.2 69.6 61.0 66.6 68.5 61.6 81.8 62.8 59.2

81.0 11.1 13.9 12.3 9.9 12.2 10.1 16.9 17.9

13.2 15.6 15.0 15.1 17.1 4.5 13.5 13.6

19.0 6.1 9.3 5.9 6.0 8.6 3.4 6.7 8.8

Part-time workers

0.5

0.2 0.2 0.5 0.5 0.2

Day labourers

Notes a Based on a survey of JSA-affiliated cotton companies. Figures are for workers employed throughout the calendar year. b Junior high school graduates. c Senior high school graduates.

Source: JSA (1979a) p. 796.

57.4

1967

Seasonal workers

Regular temporary

Regular Subtotal employees (prev. grads)

New JH gradsb

New SH gradsc

Temporary labour

Regular labour

30.4 39.0 33.4 31.5 38.4 18.2 37.1 40.8

19.0

Subtotal

Table 1.4 Female textile workers in cotton companies by employment status (regular and temporary employment) (%)a

100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0

100.0

Total

A historical legacy 25 many ways indicative of the structure of the industry itself, it has resulted in the study concentrating on the situation of women working in the large-scale enterprises, and particularly in cotton spinning. This leads to limitations to the research as a comprehensive textile industry study. However, fieldwork did provide the opportunity to visit a sanchi (textile-producing district), during which some information on smaller-scale establishments and weaving was obtained.44 Where possible this has been used for comparison with the situation of women working in large-scale factories to give more of a ‘feel’ for the situation of female operatives across the industry. Visits to industry archives also revealed that not as much original documentation on labour during this early postwar period existed as envisaged prior to fieldwork. In many cases original documentation was destroyed, particularly when the companies compiled their company histories during the 1970s or when they moved offices. This has resulted in intensive use of the limited amount of original industry documentation available. A dependence on the official industry sources as well as company management records could have led to the problem of a lack of the women’s voice in this story. Attempts to rectify this and provide a more balanced picture have been made in two main ways. First, use has been made of the direct surveys on women workers in the industry carried out by the Women and Minors Bureau of the Ministry of Labour, in particular two surveys carried out in 1952 and 1972 respectively. The material in these documents was gathered from observations during factory visits and following sample surveys of female operatives, and as such, the voices of women heavily overlay much of the recorded findings. Where possible this element of the findings has been incorporated into the study. The dates of these two surveys also allow for comparison of the situation of female workers in the industry near the beginning and end of the period under study. Second, during research fieldwork (1997–98), several personal interviews both with women who had worked in the industry during the period of study, and with (male) managers and industry leaders were conducted in order to record their memories, personal experience and opinions of the industry during that time. In addition, a questionnaire survey was distributed to 50 women who had worked in the industry during this time, asking them to respond to questions directly relating to the key issues covered in the

44 Information on one sanchi is available following a visit during fieldwork on 4 August 1998. The name of this sanchi is Banshu¯ and it is located in Nishiwaki-shi, Hyogo prefecture. This sanchi had 936 factories in 1955 and 1,463 factories in 1975 (postwar peak of 1,488 factories in 1973). The number of employees was recorded as 17,874 in 1963, but had declined to 6,967 by 1975 (breakdown by sex not available). Nishiwakishi Kyo¯iku Iinkai (1989) pp. 33–35. The sanchi labour union, Banshu¯ Ro¯do¯ Kumiai, was affiliated to Zensen.

26 A historical legacy chapters.45 While there are some inherent biases in the survey,46 it is felt that its utilisation within this research to make suggestions and to back up the main arguments provides an important contribution, particularly with regard to details that were not adequately recorded in the available official documentation. Overall, therefore, this research combines labour data, industry records, management reports and worker surveys and interviews in order to present a picture of Japanese women working in the cotton textile during these crucial early postwar decades.

45 This survey will be referred to as the ‘fieldwork survey’ throughout this book; 45 women responded to the survey distributed. 46 There are two main biases: first, the survey was distributed with the help of and through the networks of one main company, so the majority of the survey respondents are from the same company. However, it is felt to be representative of the situation in most Big Ten companies. Second, because of the available network channels for distribution the survey went primarily to women who have either worked continuously or who have rejoined the company, and have thus worked in the industry for a lengthy period of time (it therefore does not cover the situation of many who worked for only a few years and then left the industry).

2

Shaping the female labour market

Women have, from the onset, dominated the all-important textile industry. (SCAP 1946a, p. 93)

Textile workers and the Occupation authorities The reconstruction of the Japanese cotton textile industry after the Second World War took place during the Occupation years (1945–52), and it is fair to say that the Supreme Commander for the Allied Powers (SCAP) took a great interest in the labour conditions and employment practices in the industry. The textile industry’s employment history came under both scrutiny and criticism from the Occupation authorities for two key reasons – first, its heavy employment of women, particularly young girls, and second, for economic reasons, initially because of the dominant role it had played in the prewar economy, and soon after, in terms of its continued importance as an industry to lead postwar economic revival. SCAP focused attention on the system of employment of female workers that had taken shape in the industry during the prewar period, and set out to influence how the industry’s employment of women would evolve into the postwar years. In fact, SCAP clearly took the view that the textile industry offered a model, albeit a rather antithesis model, for how more general labour reform should be pursued. SCAP documentation recorded: Although there was no special legislation for textile labour, the overall labour legislation which applied equally to all industries included certain provisions especially designed to correct malpractices notably prevalent in the textile industry prior to the war. (SCAP 1951, p. 38) SCAP’s central agenda for Japan during its rule was a collection of reforms based on an overriding concern for democratisation. Democratisation was felt by the Allies to be a safeguard to any possible future revival of militarism in Japan. Within this, one of the major reforms was the democratisation

28 Shaping the female labour market of labour, with an aim to provide ‘modern’ labour legislation, essentially a labour movement and labour market modelled along Western standards. The dissolution of many of Japan’s prewar systems and practices and their replacement with new structures was a high priority for the Occupation authorities during the immediate postwar era. Labour reform programmes were viewed by the administration as critical to this movement towards a democratic Japan. An early report by a SCAP committee stated: Among the changes which are taking place in postwar Japan, the development and consolidation of a new dignity and status for labour is one of the most important. The success of the Occupation in achieving its long-range goal – a peaceful and democratic nation – depends to a large extent on developments in this field . . . [the advancement of labour] is one of the highroads toward the greater freedom and greater material well-being, for the common people of Japan, which alone can assure their lasting support of democratic institutions. (SCAP 1946a, p. 1) SCAP’s concern was both for a higher standard of living for labour, as well as for the implementation of an effective labour administrative machinery. A plethora of labour laws were established during the Occupation period,1 but two main components of labour reform are critical to this analysis of female labour in the textile industry. The first is protective legislation, particularly that concerned with the employment of women and minors, and the second is recruitment legislation, much of it embodied in the Employment Security Law 1947. While protective legislation had developed to a fairly detailed extent in the prewar period, SCAP felt that the enforcement of such legislation had not been adequate, and many regulations, including the Factory Act 1911, had been ‘largely suspended’ during the wartime economy.2 While laws which had lapsed during wartime were restored (at least on paper) soon after Japan’s surrender to the Allies, SCAP subsequently set out to broaden the scope of and raise the actual enforcement standards of existing labour law, as well as to implement revised and new legislation. The employment of women and minors was an area that came under critical review during the early postwar years. Commenting on the prewar labour situation, SCAP noted in 1946 that the ‘use of women and child labour in occupations harmful to them was commonplace.’3 Because the textile industry had been a heavy employer of both women and female

1 Trade Union Law 1945, Labour Standards Law 1947, Employment Security Law 1947, Unemployment Insurance Law 1947, Workman’s Accident Compensation Law 1947, Labour Relations Adjustment Law 1947, Health Insurance Law (established 1922) amended 1947, and the Welfare Pension Insurance Law (established 1941) amended 1947. 2 SCAP (1946a) p. 13. For discussion of Factory Act 1911, see Hunter (1989). 3 SCAP (1946a) p. 13.

Shaping the female labour market

29

minors in the prewar economy, labour conditions within the industry came under close scrutiny. General concerns over child labour included three main issues: first, the setting of minimum age legislation, which raised the minimum age limit for employment of a worker from its existing 12–14 years to the 15 year standard, in line with the Conventions of the International Labour Office;4 second, the implementation of maximum working-hour limits for minors (minors being those employed under 18 years of age);5 and third the general enforcement of higher standards and the extension of existing legislation across the broad range of (non-agricultural) industry, including clauses such as those in the Factory Act 1911, which banned children from night work and certain hazardous occupations. For the textile industry, which had already been governed by the Factory Act in the prewar period, this was nothing particularly new; however, the raising of the minimum age was to have an impact on the industry’s available labour supply. With the implementation of the Labour Standards Law in 1947, existing legislation was extended alongside new legislation in a series of protective clauses for women workers. These covered issues including working-hour limits, hazardous work, maternity and childcare leave, menstruation leave and equal pay for equal work.6 There is little doubt that awareness of the employment situation of women in the prewar textile industry had highlighted the need for SCAP’s labour reform to pay attention to women workers at this time. SCAP recorded: The special problems of women are important in a program of protective legislation both because of their numerical importance and because of special circumstances in Japanese industry. Many of them are drawn from the tiny farms by aggressive recruiting techniques for a term of temporary employment . . . their permanent return at the end of this period from the factory dormitory to the farm creates a huge labour turnover with green recruits constantly flowing in to supply the replacement demand. The particular need for adequate protection for this class of women and children, which constitutes such a large proportion of the factory workers in Japan, requires little elaboration. (SCAP 1946a, p. 93)

4 This also fell into line with the extension of compulsory education to 15 years of age in this period. 5 Work hours for minors (15–18 years) were not to exceed 8 hours per day and 48 hours per week, or be between the hours of 22.00 p.m. and 5.00 a.m. (Children under the age of 15 were permitted to be employed outside of school hours under certain circumstances, but only in non-manufacturing industries: Sugeno 1992, ch. 3, S.7, Ss.1.) 6 These clauses for women have been well documented by those studying the provision of the Labour Standards Law, and the later Equal Employment Opportunity Law of 1986. (See for example Akamatsu 1985; Sakurai 1987; Lam 1992.)

30 Shaping the female labour market In fact, concern for the employment situation of women, as well as their recently granted political rights and equal status within the constitution, gave a great deal of focus to women’s issues during this early postwar era. It also led, at the recommendation of SCAP, to the later establishment of a separate Women and Minors Bureau within the new Ministry of Labour. While attention to the implementation of labour law during the early postwar period has often focused on core legislation such as the Labour Standards Law 1947, the Labour Relations Adjustment Law 1946 and the Trade Union Law 1949,7 little scholarly attention has been paid to the implementation of legislation surrounding recruitment. The textile industry offers an important case study in this respect. SCAP’s main criticisms of the employment system in operation within textiles centred on four issues: first, the recruitment of young farm girls by recruiting agents, often under unjust contracts; second, the constant relocation of young rural workers to the factory sites for temporary employment; third, the payment of wages and advancement of wages to parents rather than to the individual worker; and fourth, the poor conditions as well as restrictions to worker freedom seen as inherent in the dormitory system. This chapter sets the scene for a discussion of these key issues, taken up in greater detail in the subsequent chapters. SCAP and recruitment The Occupation authorities were concerned with the pattern of recruitment heavily utilised by the industry in its prewar employment of young female operatives. The basis of the system was an emphasis on long-distance recruitment, relocation of young female workers, and employment within a factory-dormitory system. In the early prewar period, the cotton industry had initially utilised recruitment methods such as the introduction of new workers through the personal connections of current factory employees and the posting of recruitment notices on the gates of mills.8 However, as the industry’s demand for labour skyrocketed, serious efforts for the procurement of female labour were launched, and recruiting agents (boshu¯ ju¯jisha) were appointed by companies to travel the countryside and channel young female operatives into factories. This system, with its links between village recruiters and companies, had been associated with various problems, not least corrupt practices.9

7 For postwar labour legislation see Hanami, 1979; Sugeno, 1992. 8 See Hunter (1984a). 9 Some reform had been implemented in the 1920s, mainly in large-scale cotton companies, including attempts to license recruiters and union movements to protect female operatives from unfair recruitment practices. The practice of long-distance, short-term female labour recruitment, however, remained.

Shaping the female labour market

31

SCAP was keen to examine the issue of labour recruitment from a regional perspective. One of the provisions of the Employment Security Law 1947 was that Public Employment Security Office (PESO) branches were to be the linchpin of a new national recruitment system and were to endeavour to place workers in employment within local regional areas.10 The Ministry of Labour aimed to make the placement of workers outside of their local area of residence more difficult for employers by placing restrictions on accessibility. Recruitment by employers for local establishments which does not involve change of residence by the workers is unrestricted. Recruitment by employers or their agents from outside the locality requires permission from the Labor Minister who may modify the conditions for such recruitment. (SCAP 1947d, p. 172) The government also took action to dissolve established schemes, which until then had enabled industries to recruit widely throughout Japan. The long-standing agreement between the Bureau of Employment Security and the Ministry of Transportation providing for rail-road transportation at half-fares for workers recruited in distant areas was discontinued, thus eliminating a situation in which the Bureau, while promoting a policy of utilising local manpower before recruiting in other areas, was enabling the employers to import workers at reduced fares. The textile industry which recruits cheap labour in remote rural areas will be most affected by this change. (SCAP 1948a, p. 215) Soon after the implementation of the Employment Security Law, the Employment Security Bureau issued detailed guidelines on recruitment methods to all prefectures in April 1948. To ensure that local labour was employed before companies went further afield to recruit, recruitment goals were allocated to each prefecture. These included the detailing of the times of year when local and non-local recruitment could take place. In April companies were to recruit only at local PESO branches. In May they could pursue other forms of local recruitment activity, for example house to house visits (though banned for new school leavers; see Chapter 3). Not until June could they meet any deficits in their local prefectural labour goals by recruiting widely throughout Japan.11 In practice, however, industry could circumvent the regulations. Companies could effectively fix their labour

10 SCAP (1947d) p. 171. 11 SCAP (1948b) p. 219.

32 Shaping the female labour market requirements through established connections and networks well in advance of school graduation in April or prior to official recruitment deadlines (a practice which continues in Japan today). Moreover, textiles, as a special designated industry, could apply for permission to recruit widely because of its labour supply difficulties. Permission for direct recruitment by employers outside the labour market areas was to be given only for those critical industries where the employment bureaus could not obtain an adequate force and only if they complied with recruitment and employment regulations. (SCAP 1947b, p. 197) The 1947 policy guidelines sought to impose tough regulations on longdistance recruiting by companies in favour of prioritising local and adjacent prefectural recruitment. It was recognised, however, that without regulation this system would lead to conflict, and so a system of regional blocks was established. This was known as the ‘five-block system’.12 Within these five regional blocks covering Japan, national labour supply and demand was to be regionally co-ordinated, in effect first discouraging industries from going out of their local prefecture for recruitment, but then further discouraging them from recruiting outside of the immediate surrounding regional block of neighbouring prefectures. These regional regulations and the five-block system limited the freedom of the textile industry, which wanted to keep its wide recruitment areas and secure its traditional source of young female labour. This system had some initial success.13 However, as an industry designated for revival, the cotton industry could appeal to the Ministry of Labour to consider its special demand for labour. In fact, textile objections led to the subsequent abolition of the new five-block system in 1950.14 In this sense the aims of the new recruitment administration suffered an early failure with the establishment and subsequent abolition of an attempt to co-ordinate labour supply and demand regionally. It also indicates the somewhat conflicting role played by the textile industry during the period of reform. It was not only an industry that was held up by SCAP as a model for

12 Block A was Hokkaido¯ plus the five prefectures comprising To¯hoku. Block B was Fukushima prefecture, the seven prefectures making up Kanto¯, and the three prefectures of Yamanishi, Niigata and Nagano. Block C was the prefectures of Shizuoka, Aichi, Gifu, Toyama, Ishikawa and Fukui. Block D was the seven prefectures forming the Kinki region. Block E was the nineteen prefectures in total within the Chu¯goku, Shikoku and Kyu¯shu¯ regions. 13 During 1947–49 a marked increase in numbers employed locally (c. 58 per cent) by the industry was noted and compared to the prewar period (average of less than 30 per cent). 14 Under pressure from the industry, the five-block system was abandoned. ‘Out of area’ recruitment remained on the condition that the activities of recruitment staff were to be strictly regulated by PESO and the MOL (JSA 1979a).

Shaping the female labour market

33

reform, but also an industry designated by SCAP to lead economic reconstruction. SCAP and dormitories Closely associated with SCAP attempts to stop industry practice of recruiting labour across a wide regional base was efforts to halt the relocation of labour into factory or company-run dormitories. The SCAP authorities viewed this as a ‘feudal’ and ‘archaic’ practice. Commenting on the dormitory system, SCAP recorded: Life was extremely routinized . . . There were many limits to their [young girls’] freedom. It was not easy for them to leave the mill grounds outside working hours. There were restrictions about their receiving mail and reading material and even receiving visitors. Many of these conditions are still found in dormitories today . . . This system has long been criticised by students of the problem, some of whom would like to abolish it entirely. They have pointed to the serious abuses that are possible under the system . . . Even more important, in terms of the goals of the Occupation, is the fact that the dormitory system has been one of the many social patterns which have impeded the growth of democratic living and individual freedom. (SCAP 1946a, pp. 95–96) The aspect of worker freedom was an issue that was particularly taken up by SCAP. 15 In fact, there was a movement within SCAP at the beginning of the postwar period to ban the use of factory dormitories altogether. After much debate, however, it was conceded that the system had developed ‘naturally’ in Japan as a result of special social and economic circumstances, and that both the Japanese government as well as industry and labour leaders were in favour of setting up tighter controls within the system rather than abolishing it. SCAP noted: While the dormitory conditions are unsatisfactory, it is also true that the living conditions for many Japanese workers outside the dormitories are deplorable – in respect to fuel, food, clothing and housing . . . workers must travel long distances to their jobs in overcrowded transportation . . . on balance, it appears that under existing conditions, the dormitory system can be useful if properly operated. (SCAP 1946, pp. 95–96) 15 Existing legislation, particularly the ‘Regulations for Dormitories attached to Factories’ (1927) as a result of the Factory Act 1911, had dealt with various physical conditions and health and safety standards within dormitories, but had did not dealt with the ‘human problems’ of personal freedom for dormitory workers. The enforcement of regulations was also viewed as having left ‘much to be desired’ (SCAP 1946a) pp. 96–97.

34 Shaping the female labour market The continuation of the dormitory system into the postwar period therefore also became a subject of legislative reform. Conditions in dormitories became controlled to an extent by provisions within the Labour Standards Law 1947 (LSL), and the physical standards for their construction were concretely prescribed in the ‘Standards for Dormitories Attached to Undertakings’ (i.e. located within factory compounds) under this law. While the focus was mainly on physical standards for the operation of dormitories, Article 95 of the law did refer to the issue of worker freedom and required that enterprises could only draw up regulations governing the boarders’ daily time schedules (including sleep hours and staying overnight away from the dormitories etc), with the ‘majority consent’ of dormitory residents. SCAP’s recommendations for the operation of dormitories were not only to extend and tighten controls over the physical standards of dormitories, but also to establish definite limits on the authority of the employer to control matters involving life in the dormitories and the freedom accorded to workers. In this way, the dormitory system survived and persisted into the postwar period.16 It is fair to say, however, that, in order to overcome its prewar Joko¯ Aishi image, the industry itself was also by this time very keen to improve the standards of dormitories. A Dutch visitor to the textile industry in 1949 recorded: We must admit that the dormitories belonging to the large mills, which we visited, looked well-cared for. Evidently Japanese mill-managers attach considerable value to making this show a success . . . each mill excels in reciting to visitors the maximum number of details especially with respect to the condition of the dormitories, the number of hospital beds, shops and hair-dressing salons, the laundry capacity, the social clubs, the number of lessons in cookery, sewing, needlework, flowerarrangement, tea-ceremony, drawing and music, the diversity of outdoor sports such as baseball, basketball, tennis, ping-pong, fencing, archery, jujitsu and last but not least the films, plays and the mass excursions indispensable in Japan. An abundance of facts and figures sharply contrasting with the very brief and concise sketch of the capacity and production which one gets slipped into the hand as well. It is evident from all this that there is a wish to eradicate the legends of mill-hands living in a state of slavery . . . it must be admitted that the accumulation of labourers in the most important establishments leaves little room for

16 A more revisionist view now suggests that SCAP was not really as intent on banning dormitories as it appeared (rather it wanted to abolish long-distance recruitment), and instead used this ‘threat’ as a form of leverage to exert pressure on employers during negotiations with them to carry out their proposals for dormitory reform (Interview I, January 2003).

Shaping the female labour market

35

criticism, provided we recall the simple ways of living of the Japanese and the inbred group-solidarity of the people. (Kroese 1950, p. 80) Various socio-economic factors played a role in the development of the dormitory system in the postwar period. Improvements in the general standard of living nationally in Japan during the high-speed growth period certainly had an impact on basic expectations for living standards in dormitories as well. Prewar criticism of the industry, which Shindo (1961) notes had gone as far as regarding the dormitories ‘in the same light as compulsory labour camps’,17 prompted the cotton industry to commence positive efforts towards the improvement of dormitories, beginning with the JSA’s formulation of the ‘Measures for Democratic Improvements in the Dormitory System’ in 1947. For the industry, the pervasive image of Joko¯ Aishi was a clear motive for improving the dormitory system. The continued use of the dormitory system in the postwar period therefore not only brought the system to the attention of SCAP and placed it within more general movements for labour improvement and legislative reform, but also of course brought it to the attention of textile managers and industry umbrella organisations. Within the industry itself, dormitories were a major topic of discussion between employers and unions. This included a discussion on whether dormitories were to be officially considered a component of welfare or a component of labour conditions, given that they occupied such a central position. The argument for the inclusion of dormitories under labour conditions was based on the fact that the newly enacted Labour Standards Law regulated certain aspects of dormitories.18 The unions argued that, by coming under labour legislation, dormitories should also be considered as coming under labour conditions (i.e. governed by labour contracts). However, this theory lost ground in the debate, as clauses within the LSL were viewed as having been designed to eliminate the ‘harmful effects’ of the prewar dormitory system, and covered the general premise of dormitories rather than their actual utilisation (as private domains for individual living) and their detailed living conditions within.19 Dormitories therefore officially came to be viewed as a welfare facility rather than an element of labour conditions, and not intrinsically of the same nature as, for example, wages (see Chapter 6). The use of dormitories came under much scrutiny by the textile industry itself during these years. It was questioned how much were they to be, for example, for the protection and growth of young female workers as opposed to being necessary in order to carry out the shift-work system. Two important

17 Shindo (1961) p. 97. 18 Labour Standards Law, Article 10. 19 Shindo (1961) pp. 90–92.

36 Shaping the female labour market factors were therefore playing a role in the early years with regard to the improvement of dormitory standards. The first was legislation and the second was management strategies within the industry. While the industry was certainly now legally bound to maintain dormitories to a prescribed standard, legislation did not extensively cover the management (organisation) side of dormitories. The operation of dormitories, both as a welfare benefit and a management strategy for the organisation of workers, had to be reorganised on an ideological as well as on a practical level. In effect, the industry now had to determine how dormitories were to be maintained and managed without the elements of coercion and exploitation of workers that had characterised their prewar antecedents (see Chapter 4). SCAP and wages and welfare ‘Exploitation’ was a crucial aspect of the criticism that was levelled at the textile industry’s prewar patterns of employment, not least the issue of low wages for female textile operatives and the associated accusation of social dumping. Defence of low wage levels in the industry has focused on the fact that the traditional wage system in Japan is based on criteria such as gender, age and length of service, which in the case of young females all contribute to lower levels in the system of wage calculation. However, it is the expenditure on welfare – primarily dormitories, food and education for young females – which is an important component that must be taken into consideration when analysing the wage levels for young female workers in the industry. SCAP appears to have taken these factors into consideration when assessing the industry: In June 1949 wages in the textile industry were 51 percent of wages in manufacturing industries as a whole. By August 1950 the average wages in the textile industry had risen to 62 percent of the level for manufacturing as a whole. This average was weighted by the low income of young girls who predominated in the industry. However, these wages were supplemented by housing, light, fuel and food provided by employers free or at a nominal cost. The majority of workers in large plants lived in company-provided dormitories or houses. In the cotton spinning industry, for example, almost 72 percent of the workers lived in dormitories and seven percent in company-owned houses. Additional contributions were made by some of the larger textile firms in the form of welfare and educational facilities such as hospitals, recreational centres, libraries and other services. (SCAP 1951, pp. 47–48) Based on a 1949 visit to the industry, Kroese (1950) commented that a calculation of supplementary wage costs (i.e. welfare costs) borne by employers revealed that these extra costs amounted to on average almost 30 per cent of

Shaping the female labour market

37

wage costs, of which food (12 per cent) and dormitories (4.3 per cent) were a large proportion of that 30 per cent.20 Two early postwar works on the cotton textile industry suggested that the industry’s provision of welfare for female workers by the early to mid-1950s had important effects on their cost and standard of living. Shindo (1961) noted that in the mid-1950s the Engel’s coefficient in cost of living for female cotton workers was only 18–21 per cent, while the percentage of income spent on clothing expenses as well as that on savings and remittances stood as high as 21–26 per cent and 22–28 per cent respectively.21 The study further commented that through the utilisation of welfare facilities the average female cotton worker’s cost of living was held down at an estimated 60 per cent. Seki (1956), commenting on the high weight of savings and clothing expenditure in the early 1950s, suggested that, while some of these savings were built up by necessity for workers to transmit money to their homes, irrespective of the reason for savings, it was clear that savings could take place and offered grounds for evaluation that the wages paid were quite good.22 The welfare of workers was certainly of concern to Japanese unions during these early years after the war. The payment of family allowances was one of the first demands made by textile labour unions after their formal organisation at the beginning of the Occupation. Special payments bearing no relation to job performance by a worker were to take into account the ‘special needs of the family’. While these were mainly aimed at male workers as the family head, various concerns for worker well-being during this period improved welfare for all workers.23 Following the surrender, textile employers initially found the recruitment of labour to be difficult, because food and housing in urban areas, where most of the mills were located, was less available than in the rural areas where the majority of workers were recruited from.24 Food shortages and inadequate housing resulted in problems of high turnover and worker absenteeism. To overcome this, the government granted the textile industry extra food subsidies in order to entice rural workers to come back into the mainly urban textile factories. In an attempt to reduce the number of separations and induce recruitment the Government announced that special rice rations would be granted textile workers as of 1 November [1946] and transportation and distribution of those extra rations would be given priority. (SCAP 1946b, p. 173)

20 21 22 23

Kroese (1950) p. 79. Data are 1949 JSA statistics. Shindo (1961) p. 145. Seki (1956) p. 174. The economic role of women in the survival of families was particularly important at this time in view of unemployment, repatriation and male deaths during wartime. 24 SCAP (1951) p. 35.

38 Shaping the female labour market A variety of inducements and bonuses was also offered by textile employers to workers in an effort to increase production. In many ways, these Occupation years set a tone for how the textile industry would provide for the well-being of its workers. Like all industries in Japan, the cotton industry was subject to the various national social insurance laws instituted during the Occupation period, that focused on improved welfare benefits for workers.25 However, the industry also established many additional non-legal welfare facilities for its workforce during the postwar decades, not least in terms of dormitory and schooling facilities for young female workers (see Chapter 6). The impact of SCAP The position of the textile industry during this period of national labour reform is both interesting and important. On the one hand, it is clear that the industry’s prewar history and employment system was heavily criticised, and was the cause and ‘model’ for much of the reform. On the other hand, by the late 1940s textiles was viewed as an important industry for economic recovery in terms of its production and export potential. It was classified as one of the industries for revival under the direction of GHQ, and the new Ministry of Labour established a committee to deal specifically with the issue of textile labour. It was therefore clearly recognised by SCAP as a vital industry in the reconstruction of Japan, and was an industry where there was a national concern to supply required amounts of labour.26 This economic need to quickly get the industry up and running again often conflicted with SCAP’s more ‘moral’ objectives to reform labour conditions in the industry. It seems evident that recruitment and welfare (dormitories) systems operated by the industry vis-à-vis young female workers were a prime concern of SCAP during the Occupation years. While it was also concerned with general workplace standards and conditions, these two issues seem to have come under much scrutiny. In general, measures taken to reform and overhaul the various labour conditions, which had characterised the textile industry during the prewar period and brought the industry much international criticism, appear to have taken early effect during the Occupation period. Commentators on the industry at the time reported: Those who have watched from close quarters the rapid rise of the Japanese cotton industry in prewar days, will recall how at the time

25 Legislation included Health Insurance (1922), Welfare Annuity Insurance (1944), Unemployment Insurance (1947) and Labour Accident Compensation Insurance (1947). The costs were all borne to varying degrees by both government and enterprise. 26 SCAP reports noted the ‘special circumstances’ of the textile industry very early on. In the regular SCAP reports during the Occupation, there was a separate section on the textile industry, and the industry was also referred to often in SCAP reports on employment and recruitment legislation.

Shaping the female labour market

39

there were repeatedly complaints, especially from Western Europe, about the practices of dumping. Now it is not by any means an easy matter to expose such measures . . . The brand of slave-labour, which at one time the Japanese recruiting system so unmistakably bore has, at any rate, formally disappeared. (Kroese 1950, pp. 73 and 77) Resulting from postwar legislation, the once important feature of exceedingly cheap labour has become far less important. (Walton 1958, p. 704) The effects of SCAP reforms can only be assessed, however, by examining how practices in the industry took hold in the years following the Occupation. Commenting on SCAP’s labour legislation, for example, Kroese recorded in 1950: Of course everything is worked out perfectly smoothly on paper, but whether all these regulations are being properly enforced is not easy to ascertain. Only the proverbial discipline of the Japanese could make a proper observance of these laws possible, that is if the Japanese authorities issue instructions to that effect. And then the question presents itself as to what situation will arise when the American control is relaxed. Only at that time will it become clear to what extent the various ideas infiltrated by the foreigners out of the West have really become common-property and have come to stay in the Japanese labour world. (Kroese 1950, pp. 84–85) These key issues of recruitment, dormitories, wages, welfare and the (re)evolution of an employment system for young female workers in this industry during the years following the Occupation are therefore the subject of the following chapters of this book. As with most historical studies, various elements of both continuity and change will be revealed to be at work during these years. In many ways, leaving reform and reconstruction of the industry aside, the industry inherited a status quo in terms of its production and labour facilities; that is they were already established and geared for the utilisation of a core labour force that was young and female. As Chapter 3 will show, the industry continued, particularly during the 1950s and 1960s, to aim its main recruitment and employment programmes towards young female school leavers. It attempted to continue to recruit across a wide regional base, and to accommodate young females within factory-run dormitories. However, while this system of employment remained in many of its key elements, it also had to adapt to the various socio-economic changes of the period, not least changes that had been initiated during the Occupation years.

40 Shaping the female labour market In summary, it can be said that there were two central legacies for the textile industry in the early years following the Second World War: first, the legacy of established practice and an existing system of employment of young female operatives, and second, the associated legacy of poor employment conditions inherent within the Joko¯ Aishi image. Both influenced the reform of labour initiated and overseen by SCAP during the early postwar years, and had a bearing on how the industry itself would approach its employment of women workers during the following years.

Teething problems, unions and technology Despite the fact that textiles was a workplace for women, I don’t think that the wages and the conditions were good for women. We used to be called ‘Jo¯ko-san’ and there were many girls who came from far away. I don’t think the image was very good. But I appreciate that I have been able to keep working all these years. (Fieldwork survey respondent no. 20; employed 1960– age 15–) Following the Second World War, during the years of transition from recovery to economic growth, the textile industry encountered various ‘teething problems’ as it regained its footing. Problems surrounding recruitment and dormitories have already been introduced. Welfare issues and wage levels also created instability and unrest for the industry. A ‘human rights’ strike by labour in 1954 and a subsequent wage strike in 1955 made progress difficult for the industry during these years. The 1954 strike was the most serious labour dispute to occur in the textile industry in the postwar period, and was even the subject for a novel by the famous Japanese author Mishima.27 The O¯mi Kenshi strike ¯ mi Kenshi Spinning Co. and was The dispute began on 2 June 1954 at the O settled 106 days later on 16 September, after several unsuccessful attempts at mediation, including official conciliation attempts by the Ministry of Labour’s Central Labour Relations Board. The strike came to be seen, both within Japan and abroad, as a ‘human rights strike’ against the company’s working conditions and labour management. It turned out to be the last human rights strike, reminiscent of the activism taken by female textile operatives in the prewar period. The textile industry, even after the dispute was settled, feared any ‘international misunderstanding about the labour situation of our country’, and published an article in English to explain and clarify the situation.28 In many ways the Japan Spinners’ Association must

27 Mishima (1998). ¯ mi Kenshi strike written by the JSA in English). 28 JSA (1954) p. 4 (article on the O

Shaping the female labour market

41

have feared both domestic and international criticism of the industry’s labour conditions, because it had been such a problematic issue for the industry in the prewar period and lingered on as part of the Joko¯ Aishi legacy for the industry into the postwar period. In the 1954 article, the industry pointed out that the: ¯ mi antiquated labour–management relations as charged against the O Kenshi are far and remote from the normal state in other cotton mills in ¯ mi Kenshi, Japan, particularly in those large firms equal in size to the O and are confined to this particular company.29 ¯ mi Kenshi incident not only showed that various For the industry, the O practices of the past, albeit few remaining, still needed to be overcome, but was also no doubt influential in the industry’s efforts to co-ordinate labour conditions across the entire textile industry during subsequent years. ¯ mi Kenshi incident, predominantly female factory operatives In the O protested job injustices and brought 22 demands against the company, ranging from demands for recognition of their new labour union and dissolution of the old union (seen as an agent of management), to workplace, wage and welfare issues, and personal issues such as the freedom to go out during their non-working time and not have the company inspect their private mail. Government investigations revealed acts of: discriminatory treatment of workers; compulsory savings; dismissal without notice; disregard for restrictions on working hours, rest time, holidays and leave; mid-night work by female workers; dangerous work for juvenile workers; restriction of the freedom of private life in the dormitory; inadequate dormitory equipment and lack of safety and hygienic facilities; false wage ledgers; . . . opening of personal letters; interfering in school attendance; restriction of freedom of marriage; inspection of personal belongs; and compulsion of Buddhism.30 Intervention by both the industry and the government took place over several months until the company reached an agreement to grant worker demands. Government response included mediation with business people, official ministry conciliation, suspension of the company’s recruitment of workers, thorough investigation of the alleged unfair labour practices and general investigation into any alleged violations of the LSL and elementary human rights.31 During the dispute, industry umbrella organisations requested

¯ mi Kenshi was not included in the 29 JSA (1954) p. 4. While a large company at the time, O Big Ten. 30 JSA (1954) pp. 3–4. 31 JSA (1954) p. 1.

42 Shaping the female labour market ¯ mi Kenshi recognise the new workers’ union which had formed and that O dissolve the old union, which the company eventually did.32 During the dispute large numbers of female workers directly participated in strike activities. These included formal written complaints to the union; the blocking of factory gates by linking arms and swaying while singing union songs; marching with placards and running in zigzagging lines within factory grounds and in local streets; giving testimonies at union meetings and in later formal investigations; assisting with the production of strike notices and leaflets; and lining the streets outside the official hearings to protest against arriving members of management (Plate 2.1).33 In general, the more ‘violent’ activities during the strike were undertaken by the minority (in number) male workers, and included the breaking of managementerected barricades and company facilities (e.g. windows); the overturning of company vehicles; participation in a hunger strike; confrontation with both police and bo¯ryokudan (yakuza-linked ‘gangsters’ hired by management to threaten striking workers); and one public suicide (in protest) by a 19-yearold male worker.34 However, in general, female workers were the majority (as the majority labour force) in the main strike displays, as well as present on the periphery of other more confrontational displays. The persistence of strike action eventually brought a labour victory, not least because as the strike escalated it caught mounting public attention. One newspaper headline was ‘Mada ikite iru Joko¯ Aishi’ (The pitiful history of female factory

¯ mi Kenshi strike. Plate 2.1 Women marching through the streets during the 1954 O Photo courtesy of To¯yo¯bo¯.

32 JSA (1954) p. 5. 33 Zensen (1954). This source is actual film footage of the strike in 1954. It provided an invaluable insight into the nature of the strike activities. 34 Zensen (1954).

Shaping the female labour market

43

35

workers still lives). The strike led to large public support in the form of letters of support and cash donations, national media attention, words of support during a visit by a Japanese Member of Parliament, as well as international attention and support including a visit to the strike by an English Labour Party representative.36 Overall, the industry reported that the strike had resulted in the mobilisation of approximately 60,000 people and cost an estimated ¥3,000 million.37 Not only was this a significant strike in the history of the textile industry, but also it was important in terms of the history of the participation of female workers in such activity, as well as the level of public attention on the industry generated by the strike. It also provides us with an insight into the activism of the industry’s female labour force, and through it a sense of their consciousness as workers. In terms of the level of direct worker activism and strikes, however, the ¯ mi Kenshi dispute was somewhat of an exception during these postwar O years and in many ways signalled the end of an era. The main labour disputes in the textile industry in the postwar decades were demands for wage increases. These were generally successful and only a small percentage ¯ mi Kenshi strike was an inspirinvolved work stoppages.38 However, the O ation for a significant wage strike in the subsequent year (see Chapter 6). The newly organised labour unions launched a unified wage strike across the industry in 1955, and this heralded an era of annual wage demands, which like many other industries in Japan, developed into the ‘spring labour offensive’ (shunto¯) negotiations. The textile union and female workers It is evident that by the mid-1950s the textile unions were gaining a degree of strength and bargaining power. As was common practice in Japan, unions in the textile industry during the postwar period were organised within the major companies (enterprise unions). However, they also combined together to form industry unions, and were members at a national level within Zenkoku Sen’i Sangyo¯ Ro¯do¯ Kumiai Do¯mei (National Federation of Textile Industry Workers’ Unions), commonly termed ‘Zensen’. Zensen had several (process- or industry-based) divisions within its organisational structure: cotton spinning, wool, hemp, synthetic fibres and the ‘local’ textiles divisions. The ‘local’ division was comprised of the unions of smaller textile enterprises, which combined to form regional (prefectural) federations. Dore (1986) commented that Zensen was somewhat unusual as an industrial federation in Japan in that it exercised stronger control and guidance over its lower-echelon member unions than most federations, and 35 36 37 38

Zensen (1954). Zensen (1954). Zensen (1954). SCAP (1951) p. 50.

44 Shaping the female labour market because its organisational structure stretched further down the scale spectrum to include the regional unions organising small-scale enterprise workers. In this respect it was not purely dominated by the enterprise unions of the larger-scale firms.39 The cotton spinning division was further divided into two groups: the larger company unions composed of the Big Ten and some of the New spinning companies, and other (smaller) cotton spinning unions, generally with membership of fewer than 500 persons. Workers in the non-spinning affiliated sections (e.g. weaving, dyeing) of the large spinning companies belonged to the individual enterprise unions and therefore also to the cotton spinning division. However, for any issues arising with regard to the type of work they were engaged in, they also participated in activities of the other relevant textile divisions within Zensen.40 In general, those employees who were eligible for union membership were all employees excluding management (usually those above kacho¯ ‘section head’ level), medical and educational staff, and non-permanent workers (e.g. temporary, seasonal and part-time workers).41 This is a significant point, when considering that the proportion of temporary and part-time female workers increased in the industry from the 1960s, although some moves were made to incorporate these workers in unions from the 1970s. Zensen has one of the longest traditions and history of continuous union organisation in Japan, beginning in the 1920s. It was founded in its postwar form in July 1946, making the unions in the leading large-scale cotton spinning companies among the first to be organised in postwar Japan. As a federation, Zensen was one of the largest and most powerful industry-wide organisations during the period of study, and has generally been seen as a right-wing labour organisation, aligned to the Liberal Democratic Party (LDP) which continuously held political power during this time. By 1955, industry documentation reported that union activities were strong, with New as well as New-New spinning companies showing remarkably increased unionisation levels, and Zensen showing a progressively strengthening position overall.42 As scholars have commented,43 it often appeared that union–management consultation and discussion in the postwar textile industry placed overriding emphasis on dealing with wage disputes, while other issues were relegated to a position of secondary importance. Wage negotiations were certainly an

39 Dore (1986) pp. 188–89, 217. The MOL 1952 survey noted that, even by the early 1950s, almost all of the spinning factories surveyed had unions and labour contracts in operation, and were under the umbrella of Zensen. The exception was a few recently established mills. 40 Seki (1956) p. 166. 41 MOL (1952) p. 8. 42 Nenju¯ no Bo¯sekigyo¯ (1955) p. 83. 43 Dore (1986) p. 189.

Shaping the female labour market

45

important concern of the union (see Chapter 6); however, Zensen’s activities were wide-ranging and covered issues aside from wages, including maintenance and improvement of working conditions, improvement of dormitory conditions, retirement systems, labour contracts, and welfare and recreation. Zensen was also actively promoting specific policies for female workers during the years under study. In general, formal meetings between unions and management in the industry were held monthly (in accordance with the provisions of Article 65 of the LSL) in order to promote labour–management co-operation, as well as seek solutions to current problems. However, there also existed various other organs and channels of communication to deal with various labourrelated issues, including ‘non-working’ issues such as welfare and recreation. These channels of communication included various committees, such as health and safety committees, dormitory self-government contact committees (see Chapter 4), recreational club committees and liaison with educational staff. In this way, dormitories and other welfare facilities, aside from their direct functions, can be also seen as having acted as channels of communication within the industry. In this respect, Zensen was concerned not only with direct labour issues, such as wages and working conditions, but also with the general environment of ‘community life’ present within the textile industry’s employment patterns. As a result of the composition of labour present within the industry, Zensen was prominent during these decades as the labour organisation most active in organising female workers. It was particularly noted for having taken the lead in organising part-time female workers, who were for the most part either not prominent within or left out of most unions in Japan during these years (Plate 2.2). Dore (1986) commented that, because

Plate 2.2 Female textile workers dressed for a union rally, May Day 1958. Photo courtesy of To¯yo¯bo¯.

46 Shaping the female labour market Zensen had a high proportion of female members in its unions, it had often been seen as a weak or a reasonable union, depending on the current perspective, as well as viewed in the later years under study as a union making efforts to stave off the inevitable problems of a declining industry.44 A major concern of Zensen, during the later years of the period, particularly during the 1970s, was the fight against labour rationalisation. Overall, Zensen attempted during the period to involve itself with a wide range of labour issues, including wages, working conditions, accommodation facilities and health and welfare. It also carried out specific activities vis-à-vis female workers during these years. A history of Zensen’s union activities and policies towards women, published by the union itself, categorised the postwar period into four time periods and four phases of activity, covering 40 years from 1946 to 1985.45 These were as follows: 1

2

3

4

1946–55 ‘The awakening of workers’ rights, and the advance of dormitory democratisation.’ 1956–65 ‘The increase in working students, and the important move from dormitory policy to policy for women.’ 1966–75 ‘The increase in female officials, and system of special hearings at general meetings. Large results in the unification push for “motherhood protection”.’ 1976–85 ‘Ties with the “United Nations Decade for Women”, the fight together with other organisations in the aim for male–female equality.’

The first three phases fall directly within the period under study, and indicate that the major issues during those years were education for females working in the industry, dormitory and female-oriented policy, and the welfare programme known as motherhood protection (bosei hogo). While these are all discussed in detail in the following chapters, this section will outline the development of Zensen’s specific interests in female workers. A Women’s Policy Division (Fujin Taisakubu) within Zensen was established as a specialised department in July 1946. At that time the union had 67,000 members, 80 per cent of whom were female. By 1973, this had grown to 600,000 members, 60 per cent of whom were female.46 The aims of the Women’s Division were as follows: to have women participate positively in union activities; to raise women’s awareness through union activities; and 44 Dore (1986) p. 218. 45 Zensen (1996) pp. 44–54. 46 Zensen (1997) pp. 12–13.

Shaping the female labour market

47

to encourage a free and independent environment for women through educational and cultural activities, as well as the promotion of an ‘independent’ dormitory life.47 Although women were prominent within the union in terms of membership numbers, they were not always as actively prominent. A 1952 MOL survey commented on the relationship between female textile workers and unions at this early stage. Although a high unionisation rate for registered (permanent) female workers was recorded (96 per cent for women and 94 per cent for men) by the survey, with around 80 per cent of both sexes (members) attending general meetings, the report commented that females were generally not active participants at meetings.48 Various events committees also met, though less regularly, and again there was little participation seen by those women present at the meetings. Overall, it was felt by those conducting the survey that most women were in a union that was aimed at and run by male employees. Some lectures and discussion groups were held for (aimed at) females, but these were felt to be few. In terms of activity that was directed solely towards women, the survey commented that it was prominently dormitory-related problems and issues that were discussed. A central aim behind the establishment of a specific Women’s Division within Zensen was no doubt to improve this situation, and gradually raise the profile of women within the union as well as diversify union activities. In the early postwar years, the dormitory system was certainly a central issue for the union, and was felt to be one that was significant for female labour, given that the majority of workers at that time resided in dormitories. As early as 1947 Zensen had adopted a central resolution to ‘democratise’ dormitories, and in 1951 it formulated a ‘Dormitory Democratisation Plan’.49 As already noted, it initially attempted to push for dormitory issues to be an official labour condition, i.e. covered by labour contracts and subject to union–management negotiation. However, dormitories remained essentially an extra-legal (welfare) entity. Although not directly subject to collective agreements, the union continued to be very interested in their improvement and maintenance over the period. The settlement of dormitory issues took place in negotiations between enterprises (management) and boarders’ (workers) self-governing bodies instead, with the central aim of promoting ‘autonomous management of dormitories’ by the boarders themselves. In light of its unofficial relationship, union representatives would often attended various dormitory contact meetings, but only as observers. Zensen also held a series of nationwide dormitory seminars, to which voluntary attendance by self-government committee representatives was welcomed.50 47 48 49 50

Zensen (1997) p. 12. MOL (1952) p. 8. Zensen (1959a). Zensen (1997) pp. 12–13.

48 Shaping the female labour market In many ways, this illustrates the context within which general labour relations were conducted between unions and employers in the industry. While officially and legally there were labour agreements concluded between labour and management outlining basic employment conditions, the practical application of a working (and living) environment for labour and the actual contact between unions and management naturally differed in scope and character from the labour agreements. While there was some natural tension and hard bargaining between union and management over the years, in general it can be said that efforts for improvement and unification of conditions within the industry were viewed as desirable for, and requiring the support of, both unions and employers during this period. This was particularly true of the 1950s and 1960s. The unions were keen to be involved in any discussions and policies that sought to improve conditions for female workers within the industry, such as democratisation of dormitories and improvement of other welfare facilities, particularly in extensive attempts to overcome the inherited Joko¯ Aishi complex or image. Zensen was concerned with both external and internal factors affecting the situation of labour within the industry. In the 1960s Zensen commented that not only was a technological revolution beginning to occur during these years in the workplace, but also the increase in household appliances and domestic consumer spending nationally had created a consumption revolution. Concern was expressed that consumerism had led to a rise in prices, with labour having to work longer hours to raise income levels.51 The 1960s also brought a shortage of young (female) labour, which became a serious concern for the industry (see Chapter 3). Within textiles, young females (junior high school graduates) became known as Kin no Tamago (literally ‘Golden Egg’, i.e. a precious thing),52 and as their recruitment became difficult from the 1960s, employers invested in education systems, whereby young females working and studying in two shifts became the mainstay labour force. While, like dormitories, the education system was a welfare benefit, and not an object of direct union concern, Zensen was keen to monitor conditions and actively promoted liaison with educational staff. Working mothers also became an important issue for the Women’s Division during the mid to late 1960s, indicating the growth of older women within the textile labour structure as the shortage of young labour worsened. Whereas education and dormitory life remained important issues for young female workers, the union was particularly interested in the health and wellbeing of older women and working mothers within the textiles workplace. Some educational programmes were also extended to them, with Zensen promoting the establishment of lifestyle education (seikatsu gakko¯) for women in small to medium enterprises in the sanchi.53 51 Zensen (1996) p. 48. 52 Zensen (1996) p. 48. Several fieldwork interviewees also used this term (Interview B). 53 Zensen (1996) p. 53.

Shaping the female labour market

49

In 1971, Zensen held its first national women’s meeting. Previously, female union representatives had attended general national annual meetings, with three female national representatives elected at the 1965 annual conference; 400 people attended this 1971 meeting, the theme of which was ‘the 1970s working woman’.54 Central issues discussed were improvements in the ‘motherhood protection’ system (see Chapter 6) as well as male–female wage equality. Around 55 female representatives also attended the national Zensen meeting that year to discuss issues of equality and wages. Union activities targeting women saw a clear increase from 1971, no doubt influenced by the development of feminism and the women’s rights movements in other (predominantly Western) industrialised nations.55 Certainly, the voices of women within Zensen were increasingly being heard during this period. Overall, the textiles union made positive attempts to focus attention on and interact directly with women during the period under study. With female workers comprising the core labour force in textiles, the union was perhaps less able to ‘ignore’ women, as unions in other industries were often able to do during this period. Dore (1986) commented that, with young female workers being gradually replaced by older married women across many textile sectors by the 1970s, often as temporary and part-time labour with traditionally no union membership or employment guarantees, Zensen resolved at its 1973 annual conference to begin the organisation of part-time female workers.56 Zensen had, however, made efforts prior to the 1970s to understand and improve the position of non-permanent (predominantly female) workers within the industry, including a survey of such workers carried out in 1962.57 Furthermore, in 1970 the distribution division was set up within Zensen, and union membership was extended to retail and supermarket employees. The idea behind expanding Zensen’s coverage was to include all textile processes, from the processing of raw materials to the selling of finished goods. The extension of union membership to the retail and service sector, an area employing large numbers of female part-time workers, no doubt facilitated Zensen’s efforts to unionise these workers.58 As the economy changed, this division gradually became much larger than the traditional natural fibres divisions, and in this sense the identity of Zensen itself further evolved after the 1970s. The shop-floor environment and technology Women employed in the textile industry during these postwar decades were predominantly shift-workers. The typical shift system employed in the 54 55 56 57 58

Zensen (1997) p. 12. Zensen (1997) pp. 12–13. Dore (1986) p. 210. Zensen (1962b). See Ashida (1982).

50 Shaping the female labour market industry was a two-shift (per day) system (2-ko¯tai seido), allowing for around 16–19 hours of operation. Within this, particularly in large-scale enterprises, a day-only shift was often also operated, usually a maximum shift of around 8 hours. Under the two-shift system, workers were divided into two groups between the hours of 05:00 and 22:30. The early shift (sen-ban) usually worked from around 05:00 to 13:45, and the late shift (go-ban) from around 13:45 to 22:30 (i.e. 8 hours of work with a 45 minute break each). The shifts of the two groups were usually interchanged each week, with the shift change occurring after the weekend break. While the two-shift system was the predominant pattern in the industry, particularly in the large-scale cotton companies, as more part-time female workers entered employment over the period, the proportion of and variety of ‘day-only’ shifts showed an increase. In a MOL 1952 survey, for example, it was recorded that an extremely high proportion (90 per cent) of female workers in spinning and weaving within the cotton companies worked within the two-shift system. However, in a 1972 survey, 51.7 per cent of all textile companies surveyed were operating on a day-shift only basis.59 By the end of the period, some decline in the proportion working under the two-shift system can therefore be witnessed. However, the prominence and importance of this system, particularly for young female workers, continued as a feature of the industry during this period. Overall, women in the industry did not appear to question the shift system mode of operations in itself, and the 1972 MOL survey noted that 87.4 per cent of survey respondents said they were satisfied with the present shift system, while those who responded that they had problems with it were only 8.3 per cent.60 There were however minor complaints about its impact on their daily routines. A common response from my fieldwork survey was that the early shift was harder than the late shift, especially in winter, mainly because of the need to rise early. Several women remembered vividly their feelings when music or a bell was sounded to wake them for their shift.61 However, a few women responded that they did not mind the early shift because it gave them a sense of having more free time in the afternoons and evening, particularly for education, or for those married, for domestic responsibilities. Official hours of work in the textile industry were set by legislation, beginning in the prewar period. Until the postwar Labour Standards Law was enacted, 1929 factory legislation had played a large role in the protection of female industrial workers by banning night-time work, thereby reducing the number of hours in a day women could legally work. The 59 MOL (1972) p. 20. 60 MOL (1972) p. 64. 61 Around 65 per cent of respondents commented that the early shift was difficult and/or that it was hard to get up for the early shift. Two respondents commented that even now they can still remember exactly what music was played to wake them up.

Shaping the female labour market

51

enactment in July 1929 reduced hours in the industry to two shifts of 8.5 hours each. From April 1947 under the Labour Standards Law, hours were officially reduced to 8 hours per day (48 hours per week) and the shift system in the industry was adjusted accordingly to two shifts of 8 hours each. In 1957 negotiations began between Zensen and cotton companies over further reductions in the working hours of the industry. The result was that hours were further officially reduced by 15 minutes to 7.45 hours per day/shift, i.e. 46.5 hours per week. Worker opinion on work hours was fairly positive. In response to questions about work hours in the MOL 1972 survey, 62.9 per cent of female production workers answered that work hours were ‘suitable/reasonable’, while 17.9 per cent said they were ‘too long’. While 19.7 per cent said hours began ‘too early in the morning’, only 8.2 per cent said hours were ‘too late at night’.62 In response to questions on break-times, 52.9 per cent of females responded that breaks were ‘reasonable’. Of shift-workers, 35.6 per cent said breaks were too short, while 22 per cent of day workers said breaks were too short, and 32.6 per cent of shift-workers said they wanted small breaks outside of the main meal break.63 A common response from my fieldwork survey was that an extension of the break to 60 minutes instead of 45 minutes was seen as desirable, as well as separate short breaks in line with male workers.64 The push for a reduction in working hours by Zensen in 1957 was an important issue. In many ways, the question of shorter working hours can be said to have developed alongside the debate over the exploitation of young female workers by the industry, particularly in the prewar period. Previously, discussion and reductions in working hours had been carried out voluntarily by management, usually in line with production curtailment under economic slow-down, or through the legislative measures of government such as the prewar abolition of night work for females. However, in 1957 a claim for shorter hours was for the first time pushed through under labour union pressure, which the industry viewed as an ‘epoch-making’ event in its labour history.65 JSA documentation recorded that during this year it would ‘not have been an exaggeration to say’ that management had been engaged on a daily basis in negotiations with Zensen over the reduction of hours dispute.66 The Big Ten companies in particular had to deal with various worker strikes during 62 MOL (1972) p. 69. 63 MOL (1972) p. 71. 64 Around 40 per cent of respondents commented that the break was too short. Common complaints were that they felt 45 minutes was insufficient time to have both their main meal and a (work) break, particularly if the company dining room was several minutes’ walk from the factory. Several also commented on the fact that male workers were allowed short (‘cigarette’) breaks twice a day of 15 minutes each, but women were not eligible for these breaks (only toilet breaks outside the main meal break). 65 JSA (1958) p. 5. 66 JSA (1958) p. 4.

52 Shaping the female labour market the negotiations. The reduction to 7 hours and 45 minutes per shift came into force gradually between October 1957 and April 1958, however variance by scale was a large feature within the industry. It was primarily large-scale factories which actually implemented this change in working hours. The MOL 1972 survey recorded that the majority (70 per cent) of large-scale factories (employing 300 persons or more) operated on a 7 hour 45 minute shift or work day, but in small-scale factories the 8 hour shift was prominent. Therefore, while union pressure was successful in reducing hours, it was less effective in terms of implementation in smaller enterprises. However, compared to actual working hours in other manufacturing industries at the time, young female workers in large-scale textile enterprises worked very few overtime hours, not only because they worked under the shift system, but also because they attended educational lessons prior to or after their work shift.67 Another factor bearing an important relation to working hours and conditions was the holiday system. Like work hours, the standards for holidays and special leave were prescribed in the postwar period under the LSL, while implementation of holidays and leave within industry was generally based upon labour agreements formulated between labour and management. From the mid-1950s the textile industry, on average, maintained a system of six work days and one rest day per week, with around ten annual (statutory) holidays. The six-day working week was found to be still predominant in the majority of textile companies by the MOL 1972 survey, but in the mid-1970s, the industry was finalising the implementation of a five-day working week system in the industry.68 Other forms of leave granted in the industry included felicity (e.g. marriage) and mourning leave, and special leave pertaining to female workers covered under motherhood protection as detailed in the LSL (see Chapter 6). Working conditions are not of course determined by hours of work alone, but also affected by the working environment and, particularly in factory work, the type of equipment and machinery used. As Shindo (1961) noted, the work environment in textiles embraced factors that had a bearing upon the intensity of labour, including the modernisation of machinery, promotion of work safety, degree of improvement in workplace facilities such as lighting, ventilation, temperature and humidity control, and noise reduction.69 Many of these features of the workplace environment are directly linked to health and safety issues, the minimum standards for which were

67 Interview I. 68 Zensen had been struggling to push through the five-day week (shu¯kyu¯ futsuka sei), which was initially accepted by large-scale enterprises (Big Ten predominantly) in 1972, and finally implemented across the majority of the rest of the industry from 1976: JSA (1979) pp. 808–11. Women who responded in the fieldwork survey commented how much better working hours were with the change from a six-day to five-day working week. 69 Shindo (1961) p. 68.

Shaping the female labour market

53

regulated by legislation during the postwar period. Under the implementation of the Labour Standards Law and a subsequent Industrial Safety and Health Law, the industrial workplace was subject to a greater degree of legal requirements, and the working environment of the Japanese cotton industry saw improvements. With the increase of mechanisation during high economic growth, particularly during the 1960s, worker safety and the prevention of worker accidents was a prominent national issue. Some persistent problems in the textile workplace environment were, however, highlighted by the 1972 MOL survey. General observations by those who conducted the survey were that factors such as noise, heat and dust in factories were often quite severe, with conditions felt to be particularly severe in the silk, wool and dyeing industries because of the processes required, and worse in smaller-scale compared to large-scale companies. Specific complaints and observations of the cotton spinning and weaving industries recorded in the survey included problems associated with one person in charge of several machines; machines protruding into aisles making it difficult to pass; lack of space, and mess because of half-made goods everywhere; no space between machinery; poor lighting; bobbins overloaded; lack of protective equipment for face and hands when spraywashing cloth/fabrics; uneasiness expressed by workers using benzene in confined spaces to remove stains from cloth; finger and minor injuries from maximisation of use of the machinery and resulting broken needles and thread; very noisy machinery and lack of earplugs; severe dust and no masks; severe heat in summer (cases of over 40 degrees) and very cold in winter, i.e. severe temperature ranges often resulting in the workers catching colds; old equipment dating back to the Meiji period; instances of women having to work at machines not ‘fitted’ to their stature, because the jobs had previously employed males, leading to strain on arms and backs; and more general complaints as to the poor status of females relative to males despite their efficient work.70 Overall, the surveyors recorded that they would like to see these conditions improved upon.71 From these observations, it can be seen that even in the early 1970s, conditions in the industry still required improvement. However, it can also be said that conditions were very much based on scale (and sector) of enterprise. There was a great variance in environment and equipment by scale, and large-scale factories were operating to much higher standards. Respondents in the fieldwork survey indicated that the most common environmental problems were the temperature ranges in summer and winter and the cotton dust getting in their eyes and nose. A few respondents, 70 In general, the more specific and problematic complaints recorded in the survey were more numerous in the dyeing and processing industry, and included the breathing in of sulphuric acid, allergies to chemicals, slipping on wet floors, and general temperature ranges and noise levels: MOL (1972) pp. 47–49. 71 MOL (1972) pp. 47–49.

54 Shaping the female labour market however, pointed out that once the machines were running they kept the factory relatively warm in winter (compared to those who performed outside jobs such as agricultural work), and that air coolers in summer were not necessarily a good thing because the cotton had to be kept at a certain temperature to reduce thread breakage. With the exception of a couple of respondents, all mentioned that conditions improved greatly with the introduction of modern machinery, which in general reduced factors such as noise and dust, and improved the quality of output. The development of technology on the textile shop-floor therefore had an impact on the type of environment that the women worked in. The basic principles of spinning thread and weaving cloth are much the same now as they ever were. The principal difference over time has been the move from doing this with manually operated tools to the use of machinery (Plates 2.3 and 2.4). In general, the modernisation of technology and machinery in any industry has an impact not only on the workplace environment and nature of jobs performed, and thereby labour management and recruitment strategies, but also on the productivity of the industry and its labour requirements. Textiles was no exception to this during the period.72

Plate 2.3 Woman working at ‘automatic spooler’ machine. Photo courtesy of To¯yo¯bo¯. 72 The aim of this section is not to provide a thorough technical evaluation of textile machinery, which is beyond the scope of this study, but to provide an overview of the nature of the changes in technology, with a view to how these affected the textile labour force.

Shaping the female labour market

55

Plate 2.4 Woman working at ‘cotton comber’ machine. Photo courtesy of To¯yo¯bo¯.

At the end of the Second World War, the textile industry faced several problems – an initial labour shortage and a need to modernise equipment to reduce labour intensity, a need to improve operational speed and increase productivity, and strengthened foreign competition and pressure to increase the quality of production.73 To overcome these, after 1945, particularly from 1950, the Japanese spinning industry began to look towards further automation, in order to increase its operating speeds. It was in the 1960s, however, that a move to ‘large packaging’ machinery really took place. In 1960 the Continuous Automated Spinning (CAS) system was announced.74 In the late 1960s high-speed drawing frames and open-end spinning frames were developed and introduced into the industry.75 The effect of CAS technology was an unprecedented reduction in labour requirements. Further innovations that contributed to reduced labour requirements in spinning mills included automatic can transport devices for drawing frames, and automatic doffing devices for use on spinning frames.76 Generally speaking,

73 JSA (1979a) p. 199. 74 CAS implemented continuous automated cotton mixing, carding and drawing processes. Started by To¯yo¯bo¯ in October 1960: JSA (1979a) p. 200. 75 For details of all the technical machinery introduced, see To¯yo¯bo¯ (1980) ch. 4. 76 This meant that full cans of cotton thread were now transported automatically rather than manually by workers, while the doffer automated the task of replacing full bobbins with empty bobbins.

56 Shaping the female labour market the impact of this new technology into the industry can be seen as taking off from around 1967 onwards.77 Weaving technology did not progress as quickly in the industry as spinning technology did. This was partly due to the nature of the weaving process itself, and partly reflective of the structure of the weaving industry, with much of it carried out in small- to medium-scale enterprises, unlike largescale spinning. Many enterprises continued to use the system of group running of automatic looms, which had been introduced into the industry in the late 1920s and early 1930s.78 However, advancements in the 1960s and 1970s centred on the domestic development of two shuttle-less looms, the water-jet loom and the air-jet loom.79 An all-out move to jet and water looms did not really occur in the industry until around 1979. Overall, technology – the implementation of modern technology and machinery – and the rate of (labour) productivity growth in the industry can both be said to have increased significantly during the period. There was a general trend toward larger-scale enterprises in terms of equipment, and the volume of thread produced per spindle also rose, showing a 25 per cent increase between 1950 and 1975. The overall trend in the industry was therefore towards a greater scale of modern equipment and a greater production capacity from this equipment during the period. The proportion of processes that were modernised (implemented more modern machinery) increased from 12 per cent in 1960 to 52 per cent by 1975.80 All of this had an important effect upon required numbers of worker and labour productivity. While increased productivity was a result of improved technology and the implementation of automation, the consequence or result of productivity increases was a need for labour rationalisation. Plans for labour rationalisation naturally met with union and labour resistance. As early as 1957 Zensen had begun to examine the effects of labour rationalisation and the workplace environment after labour reductions.81 In 1962 and 1965 some industrial rationalisation measures took place, such as the closing down of factories, pay cuts and layoffs. However, the nature of rationalisation was

77 A five-year (1967–72) Textile Industry Reorganisation Programme called for the scrapping of outdated machinery and replacement with modernised equipment. Openend cotton spinning frames introduced in 1968 were expected to reduce personnel by 50 to 70 per cent and reduce production costs by 40 per cent. 78 Under this system, four looms with automatic shuttle changers were installed as a set. 79 The speed of a shuttle loom was limited because it had to mechanically move a heavy shuttle back and forth. Shuttle-less looms overcame this speed limitation. Instead of flying across on a shuttle, yarn was transported across with air or with drops of water. The first domestic jet looms were marketed in Japan in the mid-1960s while water-jet looms were introduced in the mid- to late 1970s. 80 JSA (1979a) pp. 204–05. 81 Zensen (1997) p. 47.

Shaping the female labour market

57

often such that the impact of labour reductions was felt predominantly by temporary female workers, who were not unionised.82 Layoffs for permanent employees were assisted by government schemes, and companies often laid off workers for only temporary periods or rotated employees for ‘morale reasons’.83 On the other hand, the modernisation of equipment and the introduction of increased (24 hour) operations also meant increased employment,84 and improved the shop-floor environment for workers. Despite labour-saving technological advances, the industry remained labour intensive (rather than capital intensive) during these years, and although modernisation meant lower labour requirements, it did not automatically solve labour problems. From the 1960s, the industry increasingly struggled to recruit and to fill labour quotas for registered full-time (young female) workers (see Chapter 3). In 1963 JSA documentation reported that the industry recognised that a trend of declining labour supply would continue and there was therefore a need to rationalise and become less labour intensive. In this sense, the labour shortage (in this case of young females) was in part pushing technological improvements. Moreover, labour productivity across the textile industry was still relatively low compared to other major industrial sectors, and the weaving sector was still particularly labour intensive. Further moves for technological modernisation, centring upon labour-saving devices and labour rationalisation were being considered by the industry in the 1970s.85 Overall, while the modernisation of equipment, improvements in technology, and decline in labour requirements were all clear trends during this period, at the same time labour insufficiency and recruitment problems for the industry would suggest that increased technology and productivity alone could not simply alleviate labour needs during these decades. The continued recruitment of women workers was therefore an ongoing issue of importance.

82 Shimazu (1970) looks at how ‘rationalisation’ developed in Japan and how women workers were often the ones most affected by technological innovations and rationalisation within that. He examines rationalisation activities across a range of industries including textiles. See also Shimazu (1965). 83 See Dore (1986) pp. 209–10. 84 Because of protective legislation, only male workers could work this third shift overnight. 85 Heavy investment in ‘state-of-the-art’ textile machinery that began during the 1970s (continuing after the period under study) resulted in a much more capital-intensive industry by the late 1980s early 1990s than it had been to date. See McNamara (1995) pp. 41–42.

3

Recruiting women workers

The postwar recruitment environment The ‘new’ recruitment system in operation in the textile industry from the 1950s had its roots in reforms implemented in the early postwar years, particularly those initiated and influenced by the occupation authorities. In July 1947, at the suggestion of SCAP, a circular entitled ‘Policy on Recruitment of Workers’ was issued by the Ministry of Labour’s Public Employment Security Bureau. It discussed the rationalisation of what the authorities viewed as old and embedded practices within recruitment, particularly those which had been in force in the textile industry. Then, in November 1947 the Employment Security Law was passed, announcing detailed regulations for recruitment activity and effectively abolishing the various methods used by the textile industry to date. With the promulgation of this law, a system of PESO branches was put in place, through which recruitment now had to be largely channelled. PESO was therefore designated as the new vehicle for labour recruitment in Japan. This system of public labour exchanges was not entirely new as labour exchanges had been operated by local governments under the direction of the Ministry of Welfare during the interwar years. During the Second World War, the system had become linked to the labour front for both the mobilisation of forces and the channelling of labour into wartime production.1 During the Occupation, the labour exchanges were re-established under the remit of the Ministry of Labour, and, mindful of their wartime image, were also given a name change from Kinro¯sho (Diligent Labour Office) to Ko¯kyo¯ Shokugyo¯ Anteisho (Public Employment Security Office). SCAP commented that this was ‘in order to emphasise the postwar shift in employment exchange activity from wartime labour mobilisation to voluntary consultation, recruitment and placement service for employers and job applicants.’2 SCAP was keen to move the image of the labour exchanges away from one of ‘control’ to one of ‘service’. It also wanted to remove recruitment

1 SCAP (1946a) p. 17. 2 SCAP (1947b) p. 181.

Recruiting women workers

59

activities from the control by industry itself that had occurred in the prewar period, particularly in textiles. This new legislation contained three specific components of recruitment reform, which were of concern to the industry. The first was the restrictions placed on the areas where recruitment could be carried out and the basic principle that workers should be recruited from areas within commuting distance from the place of employment. This issue of regional recruitment and relocation of workers heavily practiced by the prewar textile industry has already been introduced. The industry’s recruitment by region (and distance) in the postwar years will be examined in more detail in this chapter. A second issue in the legislation was a ban placed on the ‘direct recruitment’ of new school leavers (shinki gakusotsu-sha). This form of labour now had to be employed via PESO and meant that it was now illegal for the industry to directly recruit new school leavers – a group which had been, and still was, its prime source of labour. Third, workers other than new school leavers could be directly recruited, but only under explicitly detailed regulations issued by the Ministry of Labour. The legal clauses pertaining to the recruitment of new school leavers are particularly important to this study and require further explanation. In 1949, a revision of the legislation relating to new school leavers and their introduction to industry took place. This was the adjustment of the ‘PESOonly’ system to one whereby schools could opt to co-operate with PESO in the recruitment process. Under this new scheme, school graduates could be recruited in the following three ways: 1

2

3

A company could apply to its local PESO branch, which would then introduce job-seekers to the company. This first method was, in effect, a ‘PESO-led’ method. A junior or senior high school could co-operate with and be linked to PESO. Essentially, PESO would allow schools to ‘undertake a part of the Public Employment Security Offices’ functions with respect to employment placement and vocational guidance for such persons’.3 This can be viewed as more of a ‘school-led’ method. A high school could accept job offers directly from a company and introduce its students directly to the company, thereby essentially taking on full responsibilities as a recruitment office.4 With little direct PESO involvement, this could in effect become more of a ‘company-led’ method.

During the period under study, methods 1 and 2 were mainly utilised in the recruitment of junior high school graduates, while method 3 was more

3 Employment Security Law, Article 25–3. 4 Employment Security Law, Article 33–2.

60 Recruiting women workers common for senior high school graduates. The cotton industry mainly relied upon method 1,5 and was therefore effectively operating within a PESO-led system of recruitment. These changes to the legislation in 1949 formed a new system of recruitment, creating recruitment networks and channels for textile companies through both PESO branches and high schools, particularly junior high schools. For the textile industry, these changes signified that the key components of the recruitment system established during its prewar history were abolished by 1950. In its place were PESO-led regulations for the recruitment of the industry’s main source of labour, i.e. young female school leavers.

Recruitment of young female workers The labour market for young females Because the work required in the cotton factories had long been seen as consisting of simple repetitive operations, requiring manual dexterity rather than craftsmanship, it had long been viewed as a suitable industry for young women desiring short-term employment. This ideal, and this already established pattern of employment, continued as the industry’s main motive for selecting young females as the core labour force. Young labour was also seen as flexible by the industry, as young females had no dependants and could therefore travel and be relocated into the factory-dormitory system (although they did require training as they had no work experience). During the 1950s and early 1960s, therefore, the key labour force in the textile industry was still female junior high school graduates, as young labour was (initially) plentiful, low-waged and highly effective. The labour market for young female school leavers during these years was dominated by two key elements: first, the effects of increased education and concurrent diminished supply of young persons available to the labour market, and second, a consequent rise in the number of job-offers for young persons relative to the number of job-applicants. That is, a situation of increasingly reduced supply and high demand existed in the labour market within which the textile industry operated its central recruitment programme during these postwar decades. To observe this labour situation, this study will examine national rates of education for young females in Japan during the period, the subsequent effects on the labour market, and the consequences for industry. It will also provide an analysis of differences in the female labour market between junior high school leavers and senior high school leavers. It should be first noted that there were no major differences by sex in the young labour market in terms of proportional representation, particularly

5 JSA (1979a) p. 777.

Recruiting women workers 61 from the 1960s. Between 1950–70 the proportion of females among all junior high school leavers seeking work was a constant average of 48 per cent. Until the mid-1960s there were slightly more males than females proceeding on to senior high school, but by the late 1960s the difference had gone. The number of females not proceeding either to senior high school or paid work (i.e. doing housework or household agricultural labour) decreased, reaching the same level as males by 1964. A key change for both sexes, however, was a rise in the rate of high school graduates taking up non-agricultural work.6 Therefore, in terms of proportional (and numerical) representation, females were as important as males in the young labour market during this early postwar period and showed similar national trends. The cotton industry had a long history of recruiting and employing young females as its core labour force. In the interwar period the labour force was predominantly made up of young females who had completed the five years of compulsory elementary education, with some workers as young as 12 years of age.7 In the postwar period, however, the introduction of a new education system and the extension of the compulsory years of education to nine years (including six years of elementary school and three years of junior high school) meant that the main recruitment source for the industry became new female junior high school graduates (usually 15 years of age). This extension of compulsory education combined with minimum-age employment legislation immediately decreased the number of young persons available to the textile industry. Moreover, as increasing numbers of girls opted to progress beyond compulsory education, even the availability of female junior high school graduates steadily diminished over the period. Table 3.1 highlights this rapid change that occurred in post-compulsory education. While in 1955 only around half of female junior high school graduates were proceeding to further education, in 1960 the national rate of female junior high school graduates proceeding to senior high school exceeded 50 per cent for the first time, and from then on displayed a steady climb. The following year, 1961, also saw the lowest absolute number of junior high graduates nationally to date, due to the late 1940s decline in the birth rate.8 By the end of the period over 90 per cent of female graduates were going on to post-compulsory education. The proportion of female

6 By 1970 93 per cent of male and 97 per cent of female junior high school leavers were in non-agricultural employment: Kariya et al. (1997) p. 174. 7 In large factories, however, the age limit was generally 14 years of age with those as young as 12 years old few in number by the 1930s. 8 A sudden decline in the national birth rate in 1948 was due to the legalisation of abortion and the end of the baby boom that took place immediately after the end of the Second World War. The reduced numbers of those born after 1948 in turn reduced the numbers graduating junior high school in 1961 to an all-time low in absolute terms. The number of graduates was recorded at around 20 per cent lower than previous years: Nenju¯ no Bo¯sekigyo¯ (1961) p. 75.

62 Recruiting women workers Table 3.1 Supply trends for young females

1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975

To further education (%)

To labour market (%)

Junior high

Senior high

Junior high

Senior high

54.2 58.8 60.7 63.2 65.6 67.0 68.4 70.7 73.2 76.0 79.1 82.3 84.8 87.4 89.0 90.7

14.1 15.1 16.2 17.6 19.3 20.1 20.3 20.8 20.8 21.5 22.7 24.9 27.4 29.7 31.2 33.5

37.4 34.9 32.6 30.1 28.4 26.0 24.4 22.7 20.6 18.4 16.1 13.5 11.5 9.4 7.7 6.0

58.6 62.9 63.9 64.5 66.7 63.0 59.7 60.7 61.1 61.4 61.1 59.2 56.5 54.0 51.6 48.0

Source: JSA (1979a) p. 778. Notes Figures are proportion (%) of all female junior and senior high school leavers annually proceeding on to further education and proportion entering the workforce via PESO job placements. Figures are for March of each year.

senior high school leavers proceeding to further and higher education also showed an increase, from 14 per cent in 1960 to just under 34 per cent by 1975. This means that, in less than 20 years, the traditional source of female labour for the textiles industry (female junior high school leavers) had declined to a very small labour pool, and the next recruitable pool of young female labour (female senior high school leavers) was also diminishing. In fact, as Table 3.1 shows, the proportion of female junior high school leavers available to the labour market declined from around half in 1955 to only 6 per cent in 1975. The changing socio-economic environment in Japan during this period was a key factor in this trend. Whereas in the 1950s young females were still entering the labour force for economic reasons, increased economic prosperity during the 1960s as well as changed aspirations in the minds of young females in seeking employment took place.9 From 1960, a transition of the general (national) labour market from a ‘buyers’ market’ to a ‘sellers’ market’ was observable in Japan. In this year, the overall national job-offers to job-applicants ratio was 1.0 and rose

9 Kataoka (1993) p. 47.

Recruiting women workers 63 10

thereafter. However, if one focuses on school leavers, i.e. young labour, which was the centre of the labour shortage problem, the shortage had begun as early as 1955, with the ratio for junior high school graduates already at 1.1 (i.e. above equilibrium). This demand side of the labour market for new school leavers is shown in Table 3.2. The figures represent the number of job vacancies relative to the number of job-seekers through PESO offices nationally for junior and senior high school graduates. The growth in the number of job vacancies available for every one female junior high school graduate seeking employment increased from 1.1 in 1955, to 1.9 in 1960, 6.3 in 1971 and 6.5 in 1974. The demand for senior high school graduates started slightly later,11 and for females was apparent from 1960–61 onwards, increasing rapidly from 1.3 in 1961, to 2.9 in 1964 and to 5.5 in 1971. It is clear from the figures that education trends (and economic growth) created a major imbalance, as the ratio of job offers to numbers of young females desiring work was increasingly high over the period. A consequence of these labour supply and demand trends for young females was that the actual numbers of female junior high school graduates employed in textiles declined rapidly from the early 1960s onwards. The number of new female junior high school leavers employed in the JSA cotton companies declined from a peak of just under 108,000 in 1963 to around only 18,000 by 1975 (see Table 3.2).12 In contrast, the number of new female senior high school graduates in the cotton companies increased until 1967, reaching just over 21,000 females, after which they also showed a decline, down to around 12,500 by 1975. The textile industry identified this emerging problem of supply–demand imbalance and labour shortage early on, in the late 1950s and early 1960s. Less than a decade after this, it had developed into an extremely competitive labour market environment within which textile companies attempted to recruit. By 1969, the industry, noting the 4.8 job-offers to job-applicants ratio for female junior high school leavers alone, recorded that the number of junior high school graduates who applied for jobs through PESO agencies in that year had been recorded as 6 per cent fewer than the previous year, while demand had increased by 45 per cent over the same period. Causes cited by the industry for this trend included the reduced birth rate, as well as the fact that family heads were now earning more income, meaning children no longer had to leave school to work and earn income for the household.13

10 The job-offers to job-applicants ratio is the number of registered job-offers available visà-vis the number of registered job-applicants (job-seekers). It is recorded in government statistics as: 1.0, 1.1, 1.2 . . . 2.0, 2.1 . . . (etc.), i.e. 2.0 means two jobs available for every one job-seeker. 11 This no doubt reflects industries gradually changing their job-offers to target senior high school graduates as the supply of junior high school graduates declined, as well as the increasing progression to higher education near the end of the period. 12 The 1975 figures were particularly low, however, due to the effects of the oil shock. 13 JSA (1968) p. 72.

64 Recruiting women workers Table 3.2 Demand trends for young females

1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975

Job-offers to job-seekers ratio (%)

Numbers employed in textiles (via PESO)

Junior high

Senior high

Junior high

Senior high

1.1 1.1 1.1 1.2 1.2 1.9 2.7 2.9 2.7 3.8 3.9 3.1 3.7 4.5 4.8 5.4 6.3 5.2 5.7 6.5 5.7

1.0 1.0 1.0 0.9 1.0 1.0 1.3 1.9 1.9 2.9 2.6 2.0 2.3 3.1 3.9 4.8 5.5 2.5 2.6 3.1 2.7

81,729 64,703 80,327 107,838 103,583 98,960 81,883 68,447 64,129 58,906 49,486 46,074 38,290 30,756 28,010 18,355

6,250 6,558 7,600 10,277 9,426 16,598 21,073 21,128 16,474 15,763 13,272 13,502 11,735 18,454 16,199 12,455

Source: Shokugyo¯ Antei Gyo¯mu To¯kei. Notes Figures are based on ‘active’ applications and openings i.e. not including those cancelled. Figures are for March of each year. From 1972 method of calculation of the ratio for SH graduates changes and cannot be compared with previous years.

The industry also commented in 1971 that the rapid increase in the joboffers to job-applicants ratio was by more than the Ministry of Labour had foreseen.14 In many respects the industry was no doubt taken by surprise, but a labour shortage so soon after economic take-off, and forecasts of its increasing intensity, placed extreme pressure on textile employers to quickly adapt to both new legislative requirements and new labour market trends in an attempt to meet recruitment targets. There has been very little research into the ‘systematisation’ of labour recruitment in Japan during the early postwar period, and the crucial role played by textiles as a heavy employer of young labour within this system.15

14 Nenju¯ no Bo¯sekigyo¯ (1961) p. 53. 15 One exception is the analysis of recruitment by Kariya et al. (1997).

Recruiting women workers 65 The PESO system took off in the mid-1950s as the central agent for the (nonagricultural) recruitment of high school leavers. For the textile industry this essentially meant a key change in its recruitment of young female labour, from a company-run, recruiter-led system to a PESO-led system. Table 3.3 shows the rate of employment via PESO for new school leavers. While in 1953, only 28.3 per cent of all (male and female) new junior high school leavers were employed via PESO introduction, this had jumped to 65.8 per cent by 1961 and to 74.6 per cent by 1975. The employment of females within the PESO system was high. As Table 3.3 shows, the proportion of all female junior high school leavers employed via PESO rose from 64.7 per cent in 1960 to 88.4 per cent by 1974. Similarly, it rose from 56.8 per cent to 86.8 per cent for female senior high school leavers. The successful ‘reinvention’ and dominance of PESO in the postwar recruitment of young labour was a clear trend. An average of 60–70 per cent of new junior high school graduate job placements (male and female) were via PESO introduction during the 1960s and 1970s. This was compared with only 12 per cent of placements through the prewar employment offices in the late 1930s.16 Table 3.3 Employment of new high school leavers via PESO Job placement via PESO as proportion of . . . All JH leavers 1953

28.3

1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975

n.a. 65.8 62.9 60.1 62.0 66.1 62.8 65.2 67.3 70.2 72.6 n.a. n.a. 74.9 n.a. 74.6

All female JH

All female SH

64.7 69.1 67.2 65.3 67.8 72.6 70.5 73.4 75.7 79.5 82.2 84.9 85.1 85.8 88.4 86.1

56.8 60.4 62.3 56.4 55.1 74.6 55.1 52.4 52.8 53.4 54.7 86.5 86.0 85.6 86.8 86.9

Source: for JH and SH females: JSA (1979a) pp. 778–79. For all JH leavers: Kariya et al. (1997) p. 149. Notes Figures for ‘all JH leavers’ are for total male and female junior high school leavers. JH = junior high school; SH = senior high school. Figures are for March of each year.

16 Kariya et al. (1997) p. 128.

66 Recruiting women workers The manufacturing sector, particularly textiles, was very influential within this recruitment system, and in the young female labour market. From the mid-1950s to early 1960s, 75–80 per cent of all female junior high school leavers placed in jobs through PESO were going to manufacturing and on average 40 per cent were going to the textiles industry.17 The placement of new female school leavers in manufacturing for the latter half of the period is shown in Table 3.4. Of those female junior high school leavers entering manufacturing via PESO introduction, over 60 per cent were employed in the textile industry, making it dominant in the employment of this type of female labour among the manufacturing industries. However, the textile industry was much less dominant in the recruitment and employment of Table 3.4 Female new graduate placements via PESO (manufacturing) Junior high school females

1966 1967 1968 1969 1970 1971 1972 1973 1974 1975

Textiles

Food

Metals

Machinery (incl. electronics)

Iron and steel

60.4 56.5 59.6 60.6 58.5 64.4 67.8 66.8 67.0 67.6

7.4 6.3 5.8 4.9 4.7 5.0 6.6 5.6 5.1 8.4

1.7 1.6 1.8 1.5 1.7 1.7 1.5 1.3 1.1 1.3

15.7 19.9 18.8 18.8 20.7 15.4 11.8 15.1 15.1 9.9

0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1

Senior high school females

1966 1967 1968 1969 1970 1971 1972 1973 1974 1975

Textiles

Food

Metals

Machinery (incl. electronics)

Iron and steel

25.0 22.2 17.7 18.2 16.1 17.7 20.0 19.7 18.1 16.6

12.9 10.8 9.5 9.1 8.7 9.1 12.0 10.4 8.3 11.8

5.0 4.6 5.6 5.2 5.3 5.6 5.5 5.0 4.4 4.5

27.1 35.4 36.7 37.1 39.8 36.0 30.3 36.3 39.3 33.6

1.9 2.2 2.1 2.3 2.9 2.9 2.8 2.6 3.0 3.8

Source: Ro¯do¯ To¯kei Nenkan (Japan Labour Yearbook). Note Figures are for March of each year. 17 Kariya et al. (1997) p. 178. The placement of male junior high school leavers via PESO was also dominated by manufacturing during this period (around 70 per cent) but there was less concentration by specific industry (mainly spread over the metal, machinery and electrical industries, though with 20 per cent already in the retail sector in the 1950s).

Recruiting women workers 67 senior high school females, employing fewer than 20 per cent of those in manufacturing. The rapid growth of PESO during the 1950s and 1960s is followed by a subsequent relative decline in its proportional share of female recruitment from the late 1960s. This was the effect of an increased rate of education to senior high school level and the changing demographic composition of the female labour market. The gradual increase of senior rather than junior high school leavers (and later university leavers) entering the labour market was accompanied by a growth in direct graduate recruiting by companies via educational and other links rather than via PESO directly. In fact, during these years it can be said that gradually the role of PESO became more indirect, becoming more of a mediator and overseer between educational institutions and industry. The role of schools in recruitment While the role of schools is often difficult to pin down in the documentation, they are nevertheless a component within the recruitment network and process. As already noted, there were three ways to recruit new school leavers: through a PESO branch, through a school ‘under the guidance of and linked to PESO’, or through a school registered to ‘undertake part or all of PESO duties’.18 The majority of junior high schools (70 per cent in 1952) opted to be only linked to PESO, while senior high schools were more likely to act as PESO agents themselves.19 In either case, the degree to which schools were inextricably tied to the new PESO system meant that they became a crucial link in the textile industry’s recruitment strategy, particularly given the industry’s focus on new school leavers. While regulations sought to give priority in recruitment to companies in the same local area (prefecture) as a school, out-of-prefecture links also became well established. The system of recruitment and employment that evolved in Japan during these early postwar years followed the schedule of high school (and university) graduation. With graduation occurring in March, the general 18 JSA (1979b) pp. 7–8. This did at times create some confusion between the role of PESO branches and schools in the introduction of school leavers to industry. The industry complained about this in a report to the MOL in 1964: JSA (1979a) pp. 783–84. 19 Sources do not stipulate why junior high schools were more likely to undertake only part of the duties of recruitment, while senior high schools commonly opted to undertake full PESO responsibilities. However, several reasons can be suggested. Establishment of recruitment functions at a school – effectively becoming a ‘employment office within a school’ – no doubt required investment in terms of the training and/or appointment of specialist staff. Also, with the decline in proportion of junior high school leavers entering the labour market during this period, it was more likely that senior high schools would be acting as recruitment and employment advisors for their pupils. Over time (continuing in Japan today) it was at the university level that recruitment activities by educational institutions became most prominent, and a growth in the numbers of senior high school graduates recruited via PESO directly increased.

68 Recruiting women workers method used by the textile industry (and others) was to carry out recruitment activities a few months prior to March and to have a large annual intake of new female recruits in April. While there had been some school role in the prewar period, the industry had primarily used a system whereby recruiting agents had visited villages to directly recruit young girls from rural households. In the postwar period, however, house visits (katei ho¯mon) for the purpose of recruiting new school leavers were banned, and high school teachers can be said to have taken on many of the roles of (particularly rural) recruiting. Kariya et al. (1997) comment that teachers often acquired the nickname of ‘PESO officer’ during these years.20 In effect, schools and teachers became a link not only between PESO and companies, but also between rural and urban previously filled by recruiting agents. The flow of recruitment contact was from company to PESO to school and back again – with the company informing PESO of labour demands, PESO contacting and/or guiding schools and teachers who would supply several candidates, from which the company would ultimately make their selection via PESO. Kariya et al. (1997) note that it was felt that including schools within the recruitment link would foster a ‘customary’ and ‘natural relationship’.21 While PESO offices were charged with carrying out administrative duties, teachers were expected to show feelings, such as ‘understanding’, ‘sympathy’ and ‘sincerity’. Schools were viewed as more natural as agents of introduction, because they knew pupils personally and directly, and it was therefore easy for them to give career advice to students. In this sense, schools were entrusted not only with the role of education but also with the role of guiding subsequent employment. With the guidelines for links between schools and PESO established, schools (particularly junior high schools) therefore had a vital role to play within the context of this recruitment system. Even small-scale textile companies and sanchi generally had some sort of supply link and contact with high schools and local PESO agencies.22 Both high schools and PESO branches became venues for textiles’ public relations (PR) and for recruitment contacts and networks. Industry PR included several media over the years. A key recruitment tool used by the industry was the opportunity for young females to receive further education (schooling) while working (see Plate 3.1 and Chapter 5). Education was firmly linked to recruitment and employment programmes to provide incentives for young women to stay working in the industry longer. In 1967, a movie was made by the industry to celebrate 100 years of the spinning

20 Kariya et al. (1997) p. 130. 21 Kariya et al. (1997) pp. 140 and 152. 22 The sanchi in Nishiwaki, for example, had links with PESO branches and schools in areas of labour supply. From 1960 it implemented active and concrete PR campaigns in an attempt to decrease the difference in recruitment ‘success’ between large-scale and smallscale enterprises: Banshu¯ shoku Nyu¯ su (1960a) pp. 1–2.

Recruiting women workers 69

Plate 3.1 New recruits (wearing their school uniform) enter factory gates and are greeted by their seniors (senpai), April 1968. Photo courtesy of To¯yo¯bo¯.

industry’s history and was provided to schools and to PESO agencies as a PR vehicle for the industry. Information on the industry in the form of pamphlets and booklets was also regularly distributed by the industry.23 School visits (gakko¯ ho¯mon) by industry recruiting staff were, however, strictly regulated under PESO guidelines in terms of when they could take place in the school year and what kind of PR activities could be undertaken.24 As the supply of junior high school leavers diminished, particularly after the mid-1960s, the industry attempted to encourage female senior high school graduates (who tended to avoid working in factories and for textiles) into the industry. In 1971 the industry reported that steps towards the recruitment of senior high school graduates became easier due to the lifting of a ban on school visits, although it did not lead to major changes in the overall situation. There was an increasing trend reported towards ‘inprefecture’ industries having the advantage in securing graduate labour, and the textile industry responded by increasing school visits and associated recruitment activities.25 Recruitment efforts by the industry also included in 1975 the distribution of pamphlets to senior high school teachers with the aim of providing a ‘correct image’ of the textile industry. The 1975 pamphlet 23 JSA (1979a) p. 788. 24 JSA (1969) pp. 21–22. 25 Nenju¯ no Bo¯sekigyo¯ (1971) p. 53.

70 Recruiting women workers was entitled ‘Textile Industry Achieves Transformation to KnowledgeIntensive Industry’, and was distributed to 4,420 schools and 630 PESO offices nationally.26 Company recruiting offices also went directly to schools to help explain employment policy and the nature of the industry. The role of schools therefore became an important part of the recruitment process. The results of my fieldwork survey showed that the majority of respondents had been recruited through their school. When asked how they were recruited into the industry, the response was as follows: ● ● ● ● ●

via school or school teacher 67 per cent; company recruiter 20 per cent; via PESO directly 4 per cent; via personal acquaintance 6 per cent; stated that they ‘decided by themselves’ (but didn’t cite avenue of how they were eventually employed) 3 per cent.

In this respect, schools, particularly school teachers, and company recruiters who visited schools, were the central ‘direct’ recruitment networks for the textile industry during these years, while PESO branches and the MOL were ‘indirectly’ regulating this activity. The role of recruiters and local labour offices The textile industry’s recruitment strategy and activities during this period were both an ongoing response to the establishment of the new PESO system and a reflection of the recruiting agent system that had operated in the prewar industry. One of the aims of SCAP’s recruitment reform during the occupation had been the complete abolition of the use of recruiting agents. However, under pressure from industry, including textiles, the system in some respects remained, although under a revised format. Recruitment staff were strictly regulated under MOL guidelines and had to be company employees (i.e. no freelance agents). They had to carry identification cards and submit periodic reports of their activities to the MOL. Spot checks on recruiter activity in the textile industry were also carried out from 1948. Recruitment staff in the industry therefore came under considerable supervision by PESO officials from the late 1940s onwards.27 In June 1952 a recruitment registration system was established by the Ministry of Labour, making it mandatory for those engaged in employment activities relating to labour recruitment to register themselves with PESO agencies. The textile industry established a new recruiter system (chu¯zai-in seido),28 which was 26 Nenju¯ no Bo¯sekigyo¯ (1975) p. 87. 27 SCAP (1948c) pp. 290–91. 28 Literally means ‘staff posted to another area’; in this case the recruiting staff of textile companies posted to the local labour offices set up in regional supply areas.

Recruiting women workers 71 fully operational by 1963. Recruiters from textile companies also formed regional prefectural committees with the aim of establishing a strong bond among companies across the industry to create effective recruitment channels into the industry.29 This establishment of a unified system of recruitment activity by the industry was in many ways a response to the new national system of recruitment regulations set up by the government. The textile industry persisted in its main recruitment drive for both young female labour and labour from distant locations, even though both were now covered by strict recruitment regulations. In many ways, the system of recruiting young labour from remote regions had always required a considerable amount of industry input in terms of recruitment personnel as well as investment in strategy, methods and cost. However, it now demanded even more input to maintain its existence and construct a new network within the new legal framework. Textile companies needed agents and channels of recruitment with links to PESO branches, local regional contacts and local schools in strategic labour supply areas. Therefore, while legal reform and reorganisation in the industry meant that the prewar system of recruiting agents completely disappeared, effectively in their place, a system of local labour offices (ro¯mu shutcho¯-sho) with organised recruitment staff was gradually established by cotton companies. Local labour offices were set up by textile companies in key supply areas throughout Japan, even including those prefectures where there were no factories in the local area. The location of local labour offices was based on areas where, traditionally and currently, companies were able to recruit significant numbers of young female workers (see next section for key regions). Although mindful of the regulations that direct recruiting of new school leavers was banned, these local labour offices were essentially set up to do direct recruiting. They were established to carry out all activities surrounding direct recruitment including the ‘preservation of the supply area’, ‘administration’, ‘initial selections’, ‘surveys’ and ‘contact with households’.30 They were to co-operate with and give feedback to the government’s recruitment policies, and to promote the industry through advertising. They had to submit regular reports on their activity and plans to both PESO branches and to schools that they dealt with.31 In other words, a central role of the offices was to closely foster and maintain contact networks, particularly with local PESO branches and with local schools. The recruiting of young females by the industry was seen as having special needs. Because junior high school graduates were by nature young, they

29 The Big Ten companies formed the Bo¯seki Koyo¯ Antei Kyo¯gikai (Committee for Employment Stability in Spinning) while other textile companies created Ken Sen’i Ro¯mu Kyo¯gikai (Senro¯kyo¯) (Prefectural Committees for Textiles Labour) in each prefecture: JSA (1979b) p. 9; JSA (1979a) pp. 784–85. 30 Kariya et al. (1997) p. 193. 31 JSA (1969) pp. 9–13.

72 Recruiting women workers were also viewed as being somewhat ‘emotionally unstable’. Although their work within the company was often seen as ‘temporary’, their recruitment and relocation for several years of work in the industry was considered a sensitive matter compared to the employment of more mature workers for example.32 This is why the local labour offices and their recruitment personnel (chu¯zai-in) were viewed as a vital component in the recruitment network and in the process of obtaining a primary workforce of female junior high school graduates. Because the industry heavily employed this recruiter system it had to closely adhere to the new regulations. A Ministry of Labour record stated that the majority of persons being registered as recruiters between 1954 and 1958 under the new system, particularly those licensed as ‘out-of-prefecture recruiters’, were primarily female employees of textile companies.33 The local labour offices therefore became the postwar cotton industry’s new labour recruitment vehicle. They operated within a three-component based system comprising PESO agencies, local high schools and cotton companies. A system of liaison between local labour offices, PESO staff and school teachers was established, with personal contacts of a major significance. Kariya et al. (1997) point out that 50 per cent of recruiters employed in the local labour offices were above 55 years of age, and were often retired school teachers, PESO staff and local agricultural co-operative workers.34 Industry documentation commented that the maintenance of personal networks and the promotion of the industry’s image through PR were the two most important recruitment activities undertaken by the local labour offices.35 The development of this recruitment system and network was not without its inherent problems. During the early years each textile company was operating under different internal systems and had different ‘titles’ for recruiters, which often led to problems and misunderstandings. The rapid expansion of the system alongside growing problems in labour supply also brought associated problems of lack of training and insufficient experience of recruitment employees. The intense competition to recruit labour placed stress upon the networks and personal contacts to perform, leading to episodes of corruption. These fundamental problems resulted in pressure on the textile industry to implement concrete guidelines for recruiting activity. Following consultation with the Ministry of Labour in 1969 initiated by JSA companies, the industry carried out a reform of recruitment activity and established clear standards. These included full and complete registration of all recruiters, the creation of a recruiter guidebook demonstrating proper recruitment policies, and the implementation of recruiter training 32 33 34 35

JSA (1979a) p. 777. Kariya et al. (1997) p. 193. Kariya et al. (1997) p. 194. JSA (1979a) p. 800.

Recruiting women workers 73 36

meetings. The first of a series of training sessions for recruitment employees (chu¯zai-in kenshu¯kai) in JSA spinning companies was held in conjunction with the MOL in 31 prefectures in 1969.37 In the following year, the four other textile groups (wool, synthetics, silk and hemp) joined the programme, which ran until 1976. The need to standardise the industry’s recruitment system, have quality recruitment employees and continue the use of effective recruiter training sessions was stressed.38 A MOL representative attended these meetings to give guidance and assistance, which initially focused around topics of ‘labour recruitment’ and the ‘role of recruiters’ but later widened to include ‘new senior high school graduate recruitment and employment’ and the role of ‘labour managers’. While an overriding agenda throughout these years was the effective training of recruitment personnel within the industry, it can also be said that a key aim of both the MOL and the textile industry was to unify and establish specific procedures within the new postwar recruitment system as well as discuss labour supply problems. Ongoing labour supply pressures and dwindling numbers of traditionally key workers through the period certainly meant that the duties of the local labour offices and recruiters were not easy. Recruiting new school leavers was particularly difficult because it came under strict guidelines and supervision from PESO with direct recruiting activities by company staff banned. The extent of the often ‘informal’ personal connections and networks within the system made it a complex one to maintain. Recruitment remained an ongoing concern for the industry even during periods of economic slump. 39 The demands of the industry, the changing economy and the increasing labour shortage during the period put a strain on the very nature and components of the recruitment system in operation. By the late 1960s, a ‘recruitment war’ between textile industry firms had broken out.40 In November 1968 the Ministry of Labour sent a letter to the six textile groups (cotton spinning, synthetics, wool, hemp, silk and weaving) in which it criticised the ‘overdoing it’ by the industry’s local labour recruitment offices with regard to their recent recruitment activities aimed at new school leavers.41 Corrupt activities and links between local labour offices, PESO agencies and even school teachers in an attempt to gain adequate numbers 36 JSA (1979b) p. 20. This recruiting handbook, JSA (1969), has also been made use of as a source. 37 JSA (1979b) p. 20; Nenju¯ no Bo¯sekigyo¯ (1974) p. 61. 38 JSA (1979b) pp. 13 and 15. Recruitment staff who attended these training meetings received a card of attendance from the MOL, and a list of attendees was distributed annually to regional PESO branches. 39 The 1973 oil crisis and 1974 depression was a period of concern for local labour office recruitment staff, but due to the quick recovery of the economy after 1974, active recruitment was again back on the agenda: Nenju¯ no Bo¯sekigyo¯ (1975). 40 To¯yo¯bo¯ (1986) p. 399. 41 JSA (1979b) p. 19.

74 Recruiting women workers of young females for employment had become so conspicuous that they had become the object of social criticism. Events escalated such that it led to media attention and police investigations. The matter was even brought to the attention of the Diet during parliamentary session. The Ministry of Labour made a bid for complete guidance, but stopped short by demanding that the industry establish independent controls.42 The aggressive recruitment by the industry which had led to such strong public criticism and intervention meant that the industry was forced to introduce measures of self-discipline in order to continue with its established recruitment system using local labour offices.43 In the main these involved the industry’s tighter control over local labour offices and recruitment staff, but their wide deployment throughout Japan continued. Regional and long-distance recruitment In 1960 the Ministry of Labour carried out a survey on national labour demand and supply. The survey’s main conclusions were that the demand for school leavers and young labour would become increasingly pressured, and that, in response, there was a need to establish clear labour conditions and standards nationally.44 The ministry appealed to industry to continue to recruit only ‘within’ neighbouring regions (prefectures) where possible. Both the continual discouragement to recruit long distance and the increasing demand across all regions for young labour posed definite recruitment obstacles to the textile industry which had an established history of recruiting long distance. However, the textile industry’s power as a key industry and employer during these early years often conflicted with policy guidelines which sought to impose tough regulations on the longdistance recruitment of labour. Industry practice therefore continued to place pressure on policies designed to regulate regional and long-distant recruitment during these early years. The textile industry was not alone in its pattern of long-distance recruitment. The manufacturing sector in general had a history of recruiting widely. A high proportion of all out-of-prefecture junior high school labour employment was in the manufacturing sector. From the late 1950s to mid1970s 35–40 per cent of all junior high school graduates employed in manufacturing were taken out-of-prefecture (i.e. not locally), which was above the national average figures of 30–35 per cent of all junior high school graduates.45 Furthermore, most of the junior high school graduates employed away from their local areas into manufacturing were female and

42 43 44 45

To¯yo¯bo¯ (1986) p. 399. JSA (1979b) pp. 20–21. Nenju¯ no Bo¯sekigyo¯ (1960) p. 92. Kariya et al. (1997) pp. 146–47.

Recruiting women workers 75 there had been a sudden increase in this trend between 1953 and 1960. Both PESO and the textile industry were key players in this interregional movement of young female labour. Somewhat ironically, given the government’s regional recruitment aims, activities by PESO did play a large role in the out-of-prefecture recruitment of young labour. From 1960, 70 per cent of all junior high school graduates employed out-of-prefecture were via PESO placement.46 Despite the regional regulations and aims, the role and importance of textiles within the PESO system meant that the industry could effectively continue in its long-distance recruitment of young female labour, albeit with some significant changes. At the beginning of the period, the textiles industry could rely upon a recruitment stronghold within a handful of prefectures. Often this was based on the industry having established a strong recruitment base and supply link with the region throughout the prewar period. Table 3.5 shows the textile industry’s recruitment of female junior high school leavers by the top six regions (prefectures) between 1956 and 1968. In 1956, 53.5 per cent of all new female junior high school leavers recruited to textiles were taken from the industry’s then ‘top six’ recruiting prefectures. However, in 1968 the percentage taken from these exact same six prefectures had declined to 26.8 per cent.47 During these years, not only did the ‘top six’ regions drastically change, but also the proportion taken from a small top regional grouping declined (down to 36 per cent by 1968) to be replaced by a greater spread over a wider range and number of prefectures. This change can be further illustrated by the case of To¯yo¯bo¯. In 1954 To¯yo¯bo¯’s main cotton spinning factory had taken around 97 per cent of its female employees from the surrounding six prefectures. However, from 1965 there was a shift towards enlargement of the recruitment region, with the number of main prefectures from which recruitment took place increasing to exceed 20 prefectures nationally. By 1977, female employees originating from new recruitment constituencies (those introduced since 1963) had reached 86 per cent of the labour force.48 This trend was felt by the industry to be a significant change, indicating a relative decline in the importance of those regions within which textiles had maintained a long recruiting history, and a widening of overall regions within which textiles recruitment took place. It points to a loss of recruitment strongholds held by the industry, and increased recruitment competition for the industry during these years. The industry also cited the increasing trend of job-seekers heading to large cities and urban areas, which made recruitment more difficult.49

46 Kariya et al. (1997) p. 149. Contact from out-of-prefecture PESO offices and recruitment agents was also found to be significantly higher for females than for males (p. 178). 47 Kariya et al. (1997) pp. 194–95. 48 To¯yo¯bo¯ (1986) p. 400. 49 JSA (1979a) p. 783.

(%) 100.0 53.5

(No.) 17,466 9,337

Total grads. Total top six

(No.) 23,799 10,651

Kagoshima Nagano Niigata Shimane Miyazaki Shizuoka

Prefecture

1961

(%) 100.0 44.8

3,146 1,777 1,762 1,353 1,315 1,298

No.

(No.) 33,827 11,957

Kagoshima Miyazaki Niigata Shimane Ehime Shizuoka

Prefecture

1965

Notes Based on a survey by JSA. Recorded respondents were companies associated to JSA. Number of respondents (companies) were recorded as 96 (1956), 124 (1961), 132 (1965), 110 (1968).

Source: Kariya et al. (1997) p. 195.

2,314 1,896 1,751 1,185 1,183 1,008

Kagoshima Nagano Niigata Miyazaki Gifu Shizuoka

1 2 3 4 5 6

No.

Prefecture

Ranking

1956

(%) 100.0 35.3

3,905 2,019 1,712 1,464 1,458 1,399

No.

Table 3.5 Top six prefectures for supply of female junior high school leavers in the textile industry, 1956–68

(No.) 23,408 8,428

Kagoshima Miyazaki Kumamoto Hokkaido¯ Niigata Ehime

Prefecture

1968

(%) 100.0 36.0

2,582 1,354 1,263 1,250 1,052 927

No.

Recruiting women workers

77

The dominance of specific recruitment regions, and any changes over time, can be identified from Tables 3.5 and 3.6.50 The data indicate several features. First, Kyu¯shu¯ was a significant recruitment region for the industry. Around 30 per cent of all junior high school females employed by JSA companies in 1963 and 1971 were from Kyu¯shu¯, while just under one-quarter of senior high school females were from this region in 1971 (Table 3.6). Both Kagoshima and Miyazaki were the top recruitment prefectures within Kyu¯shu¯, joined also by Kumamoto by 1968 (see Table 3.5). Second, Hokkaido¯ grew in importance as an important labour supply area during the period, entering the top six after 1965 and representing over 17 per cent of all junior high school and over 13 per cent of all senior high school females employed by the cotton companies in 1971. Third, the textile industry was a significant employer of young females from Shikoku (particularly Ehime prefecture) and from Chu¯goku (particularly Shimane), each of these two regions comprising over 30 per cent of all employment of junior high school females within the region in 1971. In general, the figures indicate that the prominent labour supply regions were the most northern (Hokkaido¯, To¯hoku) and the most southern (Kyu¯shu¯, Chu¯goku) regions of Japan. Along with Shikoku and areas along the western coast of Japan (e.g. Niigata), they were also the most rural and isolated areas.51 Table 3.6 Textiles’ employment of female school leavers by regional area of Japan, 1963 and 1971 Junior high – 1963

Junior high – 1971

Senior high – 1971

Region

JSA as % Region as JSA as % Region as JSA as % Region as of all JH % of all of all JH % of all of all SH % of all employed JSA employed JSA employed JSA

Hokkaido¯ To¯hoku Kanto¯ Hokuriku Chu¯bu Kinki Chu¯goku Shikoku Kyu¯shu¯ Okinawa

0.2 4.8 0.2 9.8 10.6 2.6 13.9 12.9 12.5

0.1 8.6 0.7 11.7 23.7 4.5 12.2 10.1 28.4

18.9 18.1 2.0 18.1 13.8 6.9 30.9 31.8 25.8

9.1 17.3 1.8 7.3 9.4 3.5 8.6 9.8 31.0 2.1

0.7 0.9 0.1 1.2 0.4 0.5 1.5 1.3 1.1 3.0

6.1 13.5 1.7 11.0 7.7 8.3 15.3 7.7 24.1 4.4

Source: JSA survey.

50 Japan is divided into 10 main regional areas, comprising 49 prefectures (to-do¯-fu-ken) nationally. The 10 regions from north to south are: Hokkaido¯, To¯hoku, Kanto¯, Hokuriku, Chu¯bu, Kinki, Chu¯goku, Shikoku, Kyu¯shu¯ and Okinawa. 51 Nishiwaki sanchi used similar regional supply areas for the recruitment of female labour. The largest out-of-prefecture regions in terms of employee origin for this sanchi in 1960

78 Recruiting women workers Attempts to recruit labour from even the remotest regions of Japan are highlighted by the case of Okinawa. Okinawa had been a recruiting region for the textile industry in the prewar period, but was under US (and Ryu¯kyu¯) control until its formal return to Japan in 1972. The island had its own employment security office administration, but from the mid-1970s government efforts were stepped up to get the employment system in the mainland extended to Okinawa. As early as the mid-1960s, however, the textile industry was keen for Okinawa to become an established recruitment region again, and in 1968 the JSA had held negotiations with the Ryu¯kyu¯ government and had also set up an office in the capital Naha.52 While it does not appear to have been a significant labour supply area during the period under study (i.e. while under separate rule), the reoccurrence of the ‘Okinawa issue’ in the industry recruitment documentation, as well as a greater use of temporary labour from Hokkaido¯ (see later in chapter), reflected the industry’s growing need and desire to go further afield in its recruitment search for female labour.53 In effect, therefore, the textile industry continued its strategy of longdistance recruitment into the postwar period. Between 1960 and 1975 an average 70–80 per cent of all young female labour (junior and senior high school leavers) was recruited into the JSA companies from out-ofprefecture supply areas.54 However, the industry was forced to operate within the new (heavily regulated) PESO dynamic and also to diversify its recruitment bases nationally. This required a highly organised recruitment system and a high degree of input by the industry and individual companies. On the one hand, textiles could still rely to a certain extent on its (historical) ability to recruit from distant and remote rural regions. Although it was not the initial desire of young female workers or their parents to go out-ofprefecture for work, a persistent imbalance in regional supply–demand resulted in many doing so.55 The ability of the industry to place young female labour within a carefully co-ordinated residential and management structure no doubt aided the industry to recruit long distance. During the 1960s the Big Ten textile companies rapidly expanded their number of local labour offices across regions in a bid to secure labour supply channels. Overall, the textile industry was undoubtedly a huge player in terms of the regional supply and demand of young labour. In 1951, 64 per cent of all job

52 53 54 55

were Shikoku (Tokushima, Ehime and Ko¯chi) followed by Kyu¯shu¯ (Kagoshima and Kumamoto). It was planning to expand operations in To¯hoku from this year: Banshu¯ shoku Nyu¯ su (1960b) p. 4. JSA (1979b) pp. 29–30. From 1976 the recruiter system in Okinawa was established under the same system as that in mainland Japan: JSA (1979b) p. 30. JSA (1979a) p. 778. Kariya et al. (1997) p. 179.

Recruiting women workers 79 offers and 97.5 per cent (94 per cent for females) of all out-of-prefecture job offers for junior high school graduates were in textiles.56 Compromise had to be established between enforcement of the new regional regulations and aims, and the demands of an important industry which was a heavy user of female labour in Japan at the time. One of the duties entrusted to PESO upon its establishment was to ensure that demand–supply co-ordination was possible across regional areas. In many ways the textile industry both used and extended the PESO regulations and system to ultimately continue to recruit over a wide geographical area in the postwar period. The demands of industry and the changing economy, as well as the increased pressure on labour markets during the period, generated tension between the contrasting roles of the PESO administration. However, the shortage of young labour, which became a major concern after 1960, combined with the acceleration of the economy, placed even further pressure on the labour market, effectively forcing the textiles industry to recruit across diversified regions nationally in its quest to secure labour.57 In many ways, the continued custom of long-distance and out-ofprefecture recruitment of female labour secured textiles a significant share of the available female junior high school labour supply pool during this period. Their recruitment and employment structures were already geared to this type of labour and in a sense meant the industry could continue to rely on this young rural labour supply. On the other hand, the continuation of this core recruitment strategy, and their image as an employer of young rural females, may have meant they were handicapped in terms of competing for alternative (even senior high school females) and for urban-based labour. This issue of recruitment ‘performance’ greatly concerned the industry during these years.

Recruitment problems for the industry From the early 1960s the national shortage of young labour began to seriously affect the cotton industry’s established pattern of recruiting young female labour. An industry report commented as early as 1961 that the cotton spinning and weaving industry had been ‘experiencing increasing difficulty in recruiting the necessary labour force, especially young women’.58 It noted that in its practice of recruiting new female junior high school leavers to the cotton mills during March–April of each year,

56 Kariya et al. (1997) pp. 190–91. 57 The decline in junior high school leavers was also changing the nature of regional recruitment. From the late-1950s until the mid-1960s, junior high school graduates predominated in all out-of-prefecture labour migration nationally, but by the late 1960s the proportion of senior high school graduates was surpassing the share of junior high school graduates. 58 JSA (1961) p. 50.

80 Recruiting women workers management was having the ‘greatest difficulty’ in securing the necessary numbers to fill vacancies due to the ‘changed position of demand and supply’ in this labour source in recent years.59 The explanation for these recruitment difficulties was recorded by the industry as the result of increased rates of education beyond the compulsory level, the take-off of the economy, especially in the heavy industry and service sectors, and the drop in the national birth rate. By 1965 the weight of management concern in the industry was heavily centred on the problems associated with recruiting young female high school graduates. This lack in the required numbers of young females entering the workforce persisted for the industry throughout the years under study, even during periods of recession and layoff, and despite the industry’s attempt to reduce the number of workers required through equipment modernisation and technological upgrading programmes. It resulted in a fundamental labour problem, which the industry referred to as labour insufficiency (ro¯do¯ryoku fusoku). The sufficiency rate (ju¯soku ritsu) was a key term used in the industry documentation during this time.60 This rate referred to the number of regular workers recruited as a percentage of the workers wanted (required by the industry). The sufficiency rate declined for the industry over the period, and meant that the industry struggled to operate with an optimum required workforce. As early as 1960 the industry reported that, in an era of increasing competition, it was becoming difficult to acquire the quantity of labour required. The sufficiency rate for female junior high school graduates in the cotton industry was calculated at 83 per cent, but fell sharply in the following year to 66 per cent.61 Although the situation in 1961 was made worse by the record low number of school graduates in that year, concerns for future labour forecasts were considerable. Job vacancies were numerous at all companies but reportedly could not be filled. The industry felt that the supply–demand imbalance was leading both to a lower ‘quality’ workforce and to higher initial salaries.62 These were all general trends which continued and strengthened throughout the 1960s. Table 3.7 shows sufficiency rates in the JSA companies for available years between 1960 and 1973. In 1960, overall labour sufficiency in the cotton industry was already below optimum levels and calculated as an average of 59 60 61 62

JSA (1961) pp. 50 and 51. Nenju¯ no Bo¯sekigyo¯ (1961) p. 75; JSA (1979a) p. 782. JSA (1979a) p. 782. Nenju¯ no Bo¯sekigyo¯ (1961) p. 75. Industry documentation often referred to the ‘quality’ of the young labour force declining during these years. To¯yo¯bo¯ (1969a) pp. 23 and 40. In many cases it referred to the fact that with increasing rates of education at senior high school, the industry was no longer recruiting the ‘crème de la crème’ of junior high school females as it had been able to do previously, and was also having to recruit from a wider range of high schools including schools of a lower ‘rating’ than previously dealt with (Interview I and Interview M).

Recruiting women workers 81 Table 3.7 ‘Sufficiency rate’ in JSA companies (%) Big Ten 1960 1961 1963 1964 1965 1973

86.0 72.6 56.4 46.1 46.0

New companies

New-New companies

Total (av.)

34.6 28.1 28.1 28.0

72.1 65.0 53.0 41.8 37.8 36.0

{52.0} 43.4 35.0 35.4 35.0

Source: Nenju¯ no Bo¯sekigyo¯ (The Spinning Industry Yearly). Note The sufficiency rate refers to the number of regular workers recruited as a percentage of the workers wanted (required by the industry).

72.1 per cent.63 This declined over the period to only 36 per cent in 1973. The ability to fill labour requirements appeared to be significantly different by scale of company. In 1963, for example, the Big Ten companies were operating at just under 73 per cent labour sufficiency, but the rate in the New and New-New companies was much lower, at 43 per cent and 35 per cent respectively. In fact, the New and then the New-New companies operated at a greater level of labour insufficiency than the Big Ten throughout the period, reflecting that, by scale, small- and medium-scale companies experienced a greater degree of difficulty in their ability to employ labour. Labour insufficiency and the struggle to recruit young females was much higher in the sanchi enterprises than in large-scale corporations. In 1961 Nishiwaki sanchi reported that their job-offers to job-applications rate was around 4.0 for new female junior high school leavers, compared to a rate of 2.7 for JSA companies in the same year.64 It was the employment of young females that differed markedly by scale of operations across the industry. Although figures are not readily available for a concrete breakdown of female employment by scale, Table 3.8, showing breakdown by JSA-affiliated companies for available years, provides some indication. The figures suggest that the Big Ten companies employed more than half of all female junior high school graduates in the industry from the mid-1960s to mid-1970s, and around 70 per cent of all female senior high school graduates in the mid-1970s. While New cotton companies showed some increase in their proportional share of the employment of female junior high school graduates during the period (one-quarter

63 This was a JSA survey in 1960 of 124 spinners (representing almost 90 per cent of the entire spinning industry). The total number of workers required was 35,950, but they had to operate with only 25,924 (including new graduates), which represented 72.1 per cent of overall requirements: JSA (1961) p. 51. 64 Banshu¯ shoku Nyu¯ su (1961) p. 8.

55.9 57.3 45.9

1973 1974 1975

26.8 25.7 32.9

21.5 20.3 23.1 24.6

New companies

17.3 17.0 21.3

20.5 22.3 23.4 22.6

New-New companies

68.4 71.6 65.4

Big Ten

Senior high

Note Figures are for cotton spinning companies affiliated to the JSA.

Source: Nenju¯ no Bo¯sekigyo¯ (The Spinning Industry Yearly).

58.0 57.5 53.5 52.8

1963 1964 1965 1966

Big Ten

Junior high

18.6 18.2 18.0

New companies

Table 3.8 Employment of female new graduates across JSA companies (%)

13.0 10.3 16.6

New-New companies

57.8 59.6 49.4

Big Ten

25.5 24.4 30.2

New companies

Total female graduates

16.6 15.9 20.4

New-New companies

Recruiting women workers

83

of female junior high school leavers during the mid-1960s and 1970s), overall the New and New-New companies employed only a small proportion of female senior high school graduates. This suggests, not surprisingly, that large-scale companies within the industry had greater power and ability to secure female senior high school graduates than did small- and mediumscale companies. Large-scale companies had greater implementation of recruitment and management policies and were therefore able to attract higher numbers and a ‘better quality’ labour force. It will be shown in the next section how textile companies attempted to fill the ‘gaps’ in their labour requirements during these years. A feature that has become apparent from the analysis thus far is that the textile industry’s recruitment and employment of young females was dominated by junior high school leavers. In fact, the industry’s procurement of junior high school graduates as a proportion of national availability was strengthening relative to other industries during this period, despite the overall decline in numbers in this labour pool (see Table 3.9). The employment of female junior high school leavers across the textile industry as a proportion of all PESO placements was high, and increased from just under 40 per cent in 1960 to just over 50 per cent by the 1970s. Their employment within the JSA companies alone increased from 14 per cent to 24 per cent during the same period, meaning almost one in four of all female junior high school graduates nationally worked for the JSA cotton companies. The textile industry must have therefore held some attraction or ‘pull’ for young females seeking employment. The historical image of the industry as an employer of young female labour certainly played a part. Several of my fieldwork survey respondents, particularly those who entered the industry during the 1950s and 1960s, stated that the industry was well known for its employment of women.65 An article written by an industry manager in 1963 included a quote revealing the impression of a new young female employee of To¯yo¯bo¯. Kobayashi (1963) suggested that it showed the ‘gentle deep emotion of a young girl stepping into society after just leaving junior high school’. ‘Wow, it’s really nice’, in spite of myself, I had cried out. And that surprise had before long swelled into a dream and desire. A beautiful factory, kind factory workers, my room neatly in order and tidy, there’s

65 While the image of the industry has not always been positive, many of the women surveyed and interviewed said that the industry had good employment conditions for young girls. It could be suggested that the industry had a better image in rural areas than in urban areas. Textiles was seen as a ‘good job’ by many in the rural areas. Although, as one of the survey respondents pointed out, there were very few alternative employment opportunities for females in the early postwar years, another respondent said that she had had a choice between textiles and electronics, but that she and her family had felt that electronics was an ‘unknown’ industry at that time compared to textiles.

211,547 165,934 208,597 239,748 228,534 218,422 179,571 158,934 141,103 124,738 107,689 91,187 74,978 60,851 54,306 39,588 34,148

81,729 64,703 80,327 107,838 103,583 98,960 81,883 68,447 64,129 58,906 49,486 46,074 38,290 30,756 28,010 18,355 16,821

Note Figures are for March of each year.

Source: JSA (1979a) p. 778.

1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976

All job placements via PESO (A) (B)

38.6 39.0 38.5 45.0 45.3 45.3 45.6 43.1 45.4 47.2 46.0 50.5 51.1 50.5 51.6 46.4 49.3

30,074 23,799 26,433 25,009 31,514 33,827 27,095 22,503 23,408 23,476 20,225 18,837 17,230 14,689 13,395 8,748 8,406

14.2 14.3 12.7 10.4 13.8 15.5 15.1 14.2 16.6 18.8 18.8 20.7 23.0 24.1 24.7 22.1 24.6

C/A (%)

Employed in JSA companies

B/A (%) (C)

Employed in textiles

Junior high females

Table 3.9 Employment of female school leavers in the textile industry

144,065 169,058 190,757 169,968 151,845 263,983 249,294 247,605 251,537 238,726 229,928 346,146 319,883 307,218 300,277 277,293 261,519

All job placements via PESO (D)

Senior high females

6,250 6,558 7,600 10,277 9,426 16,598 21,073 21,128 16,474 15,763 13,272 13,502 11,735 18,454 16,199 12,455 13,967

(E) 4.3 3.9 4.0 6.0 6.2 6.3 8.5 8.5 6.5 6.6 5.8 3.9 3.7 6.0 5.4 4.5 5.3

E/D (%)

Employed in textiles

815 518 299 1,679 2,697 2,714 2,500 3,634 3,553 2,783 2,817 2,558 2,646 2,605 1,949 1,985

(F)

0.5 0.3 0.2 1.1 1.0 1.1 1.0 1.4 1.5 1.2 0.8 0.8 0.9 0.9 0.7 0.8

F/D (%)

Employed in JSA companies

Recruiting women workers

85

a school right next to the dormitory where I can study cooking which I really like, and there’s also an excellent hospital. When I received a guided tour of the large factory and saw all the machines lined up, in spite of myself my body had tightened up and the joy of working had swelled inside my chest. As I got into my futon thinking about the things my elder female fellow-workers (senpai) would be teaching me from now on and the interesting stories I would hear, of course I felt a little homesick. But, a new life is beginning. I made an oath in my heart that I must become a splendid member of society (shakai-jin). (Kobayashi 1963, p. 142) Despite the somewhat ‘dramatic’ nature of the excerpt, it is of interest. As Kobayashi comments, an important point from this excerpt is the young women’s apparent motives for joining the company. They appear to be not just to earn an income.66 The author also outlined a recent survey of new employees taken by To¯yo¯bo¯ at the time. When questioned on their motives for joining the company, the results were as follows: ● ●

● ● ● ● ● ● ● ● ●

to earn own salary 30 per cent; to acquire education (implied ‘becoming educated in skills and culture’) 28.6 per cent; to assist the family income/budget 12.8 per cent; because can attend school 10.3 per cent; want to work (shigoto ni yarigai ga aru) 5.9 per cent; to have lots of friends 3.5 per cent; to have a life away from home 3.2 per cent; to have an urban life 2.3 per cent; to enjoy sports 1.9 per cent; because there are many recreation facilities 1.3 per cent; other reasons 0.2 per cent.67

Therefore, while many were, of course, entering employment to earn income, at the same time, for many there was also the chance to gain some level of education, to develop personally, and to have a communal life within the company facilities away from home. Education was a particularly notable ‘pull’, with just under 40 per cent of respondents expressing some sort of interest in the educational opportunities offered by the company, thereby expressing a desire or interest to both work and study at the same time. The results of my fieldwork survey back this up. When asked why they wanted to work and why they chose textiles, the response was as follows: ●

education 20 per cent;

66 Kobayashi (1963) p. 142. 67 Kobayashi (1963) p. 143.

86 Recruiting women workers ●

● ● ● ● ● ●

living facilities (cited availability of dormitory or recreation facilities or chance to make friends) 19 per cent; because it was a large company 16 per cent; for financial reasons 12 per cent; because it was the best employment opportunity available 12 per cent; because it was nearby (near home) 6 per cent; other reason 10 per cent; no response 5 per cent.

These results would suggest that ‘positive’ features of the industry – educational opportunities, living facilities, the existence of large companies and the image of a good employment opportunity – played a significant part in attracting young females into at least the Big Ten companies within the industry. The 1972 MOL survey also briefly surveyed the ‘worker consciousness’ of workers. When surveyed on the main reason for working, the main response of unmarried women was to ‘progress as a person’; when asked until when they wanted to continue working, 24 per cent of unmarried women said ‘until marriage’ and 18.9 per cent said ‘until graduation’. Among those female workers going to school in the survey, 54.1 per cent responded that they ‘definitely wanted to continue working until graduation’.68 Overall, these surveys give clear indication of reasons for the attraction of the industry as a place of employment for young junior high school females, and its ability or success in the recruitment and employment of this group of young labour. However, there appears to be relatively less evidence of its ability to attract senior high school females. As noted already, the industry attempted to expand its main recruitment drive for young females away from the diminishing supply of junior high school females towards the larger labour pool of senior high school leavers. Senior high school females had been systematically introduced into the textiles workforce since 1955. Their numbers had increased until 1968–69, but then had declined after this (see Table 3.9, p. 84). While the industry recruited a significant proportion (40– 50 per cent) of all junior high school females nationally, it recruited less than 10 per cent of available senior high school females in the mid- to late 1960s, but on average employed only around 5 per cent of all available senior high school females. Table 3.9 indicates that the JSA companies employed less than 1 per cent of available senior high school females compared to almost one-quarter of all junior high school females. This is a key point given that the industry tried to move from the employment of junior high school females to senior high school females, particularly from the 1960s.69

68 MOL (1972) p. 84. 69 The success of attempts to integrate senior high school females into the labour force in To¯yo¯bo¯ was discussed in To¯yo¯bo¯ (1970a). The company noted that they had employed far fewer than forecast (p. 2).

Recruiting women workers 87 The textile industry was consistently unable to replace its shortfall of junior high school females with senior high school females. In 1971, for example, when the proportion of female junior high school graduates employed by textiles decreased by 6.9 per cent (due to declining numbers), the employment of female senior high school graduates only went up by 1.2 per cent. Despite the industry’s recruiting efforts, it had less ability or power to secure a significant proportional share of the female senior high school labour pool compared to junior high school females. This recruitment record vis-à-vis female senior high school leavers was a point of concern for the industry during these years. In some respects, textiles was not alone with regard to this issue. Manufacturing industries in general employed a far lower proportion of senior high than junior high females. As Table 3.10 shows, around 60 per cent of all female junior high school leavers were employed in production jobs, while around 75 per cent of senior high school graduates were employed in office and sales jobs, with on average only 10 per cent employed in production jobs. This suggests that either manufacturing and production industries were more likely to recruit younger ‘less-educated’ females, or that ‘moreeducated’ females had less of a desire to work in production jobs. In reality, it was probably a reinforcing combination of both. The textile industry commented that females who graduated from senior high school preferred to work in sales and in general office work rather than working for a factory, and that they tended to seek ‘more professional’ jobs, such as nurses and beauticians.70 There was a general feeling that female junior high school graduates, that is those with that level of education, were ‘suited’ for textile employment, whereas female senior high school graduates were suited for Table 3.10 Female school leavers employed by job type (%)

1960 1965 1970 1975 1976

Employed in production jobs (departments)

Senior high school females employed in

Junior high

Senior high

Office and sales jobs

54.4 59.7 66.6 60.7 61.9

7.2 8.1 15.1 10.5 10.9

Technical and Other types production jobs of jobs

1961

78.3

8.6

13.1

1968

74.8

13.1

12.1

1976

75.3

11.0

13.7

Source: calculated from JSA (1979a) p. 773. Note Figures are for March of each year. Despite the source, figures are for all industries (not just textiles).

70 Nenju¯ no Bo¯sekigyo¯ (1973) p. 67.

88 Recruiting women workers more skilled positions in other manufacturing sectors or in non-manufacturing industries such as the service sector. Both the image of the industry as an employer of the youngest females (i.e. straight out of compulsory education), and the gradually changing socio-economic status and options available to young women during the period, no doubt meant that in some respects textile jobs held little charm for female job seekers leaving senior high school.71 Dore (1986) also suggests that the labour shortage was exacerbated in textiles because of the ‘social definition of the industry as a declining industry’.72 These issues all point to a recruitment problem that the textile industry faced during this period, that of increased competition for young female labour. Competition from other industries in the context of a growing economy during these years was having a large effect on the textile industry’s ability to recruit young female labour, particularly in the case of senior high school leavers. While there still remained a high percentage of ‘new’ junior high school leavers in textiles, the increased demand and labour shortage meant that the industry began to employ a higher proportion of ‘previous leavers’ over the period.73 As competition, particularly from the electronics industry, increased from the mid- to late 1960s, the industry, relative to others, was showing signs of a greater recruitment performance struggle.74 Textiles showed especially high labour insufficiency rates during the 1960s and 1970s, while the sales sector and the light electrical industry showed the lowest levels.75 Newer industries were more likely to be able to attract senior high school leavers and often made it a policy to try and do so. Although the electronics companies were initially in competition with textiles in the recruiting of female junior high school graduates, in 1969 they shifted to predominantly employing female senior high school graduates and were able to recruit them in greater numbers than textiles. Electronics companies even came to textile company-run schools to recruit girls who would be graduating with their senior high school diploma.76 Having held such a dominant position in the female labour market for so long, the industry was perhaps finding it difficult to compete with other

71 The industry also commented that senior high school females tended to leave after an average of two years’ employment, i.e. after attending junior college or polytechnic schools while working, whereas the senior high school programme for junior high girls in the industry was longer (four years) and meant they generally worked longer in the industry while studying: JSA (1991) p. 41. 72 Dore (1986) p. 205. 73 Refers to those who left education after junior high school level but did not go straight into textiles employment, i.e. ‘second jobbers’. 74 Nenju¯ no Bo¯sekigyo¯ (1967) p. 77 noted the ‘blow’ dealt to the industry because of the increasing numbers of young females being attracted to the electronics industry. 75 Kariya et al. (1997) p. 180. 76 Interview M. The labour manager interviewed was previously in charge of company schools.

Recruiting women workers 89 (newer) industries, such as the electronics industry and the clerical/service sector. The shortage of young female labour, and the increased recruitment competition for this labour group, meant that the industry had to seek alternative sources of labour. The main alternative source of labour was older female workers.

Recruitment of older female workers It is much more difficult to analyse the recruitment of older female workers than that of young female workers in the textile industry for two main reasons. First, the industry documentation often did not include them as they were not regular workers, i.e. (young) new school leavers. Information on employment groups within which older females were particularly prominent, such as temporary or seasonal workers, was often unrecorded. Second, their recruitment and employment was much less systematic than that of young females. In part this is due to the fact that the industry was forced to systematise the recruitment of young females under the PESO regulations. In the main, however, it was because the recruitment drive by the industry focused on female school leavers during this period, and the recruitment and employment of older women was often supplementary in nature to this. In 1968 the industry commented that as a result of a shortage of young labour and associated recruitment difficulties there had been a ‘rapid increase of the older age group in the workforce’.77 As noted in the previous section, the industry’s initial response to the declining numbers of junior high school females was to diversify recruitment to include other young females, primarily ‘previous junior high leavers’ and senior high school females. However, this quickly proved to be problematic, with the industry unable to recruit in sufficient numbers. Other tactics included the modernisation of equipment and subsequent reduction in number of workers required, attempts to entice young workers to stay longer by tying length of service to educational opportunities (see Chapter 5) and the employment of men particularly in a third (night) shift of operations.78 In 1970 the industry also considered introducing foreign workers into the industry, but this did not materialise.79 It was ‘middle-aged’ and ‘older’

77 JSA (1968) p. 72. 78 In general, however, the industry never had a recruitment strategy that included the diversification of its workforce to include more than a minor proportion of male production workers: JSA (1979) p. 781. Details of the modernisation of equipment as well as the introduction of a third night shift are discussed in Chapter 2, while the ‘linking’ of service to education is discussed in Chapter 4. 79 This was not the textile industry alone. In September 1970, the Japan Federation of Employers’ Association (Nikkeiren) submitted a report to the MOL, including requests to consider the employment of foreign labour. However, after deliberation, the MOL

90 Recruiting women workers women in the labour population who gradually became the key alternative source of labour to young female workers. The labour market for older women The textile industry commented that it began to see limits in the supply of young labour via the PESO-regulated recruitment channels after 1964. The re-establishment in 1964 of PESO Clause 36 regarding direct recruiting (for non-new school leavers only), which had been suspended since 1960, meant that the industry could establish direct recruiting channels to secure older women workers. In 1967, when the industry reported that the rapid reduction in young females had left them in dire circumstances in terms of labour,80 the industry began to employ women from areas local to factories as part-time employees, thereby employing older women on essentially a ‘full-time’ and regular basis. In this year, the national demand for ‘general’ (male and female) labour, i.e. not including new school leavers, went just over demand–supply equilibrium for the first time. Previous to this there had been a labour surplus of older workers. Table 3.11 shows the national job-offers to job-applicants ratio for females broken down by age groupings. The following points can be noted. First, from around 1967 until the mid-1970s (i.e. until the oil crisis period) the ratio for all females rises above the equilibrium of 1.0 (see dark shading in Table 3.11), indicating that in this period of high economic growth combined with a young labour shortage there was an overall demand for female workers in the economy. Second, by age group the highest demand, not surprisingly, is for the youngest age group (below 20 years of age) as shown by the medium shading in the table. This category would have been dominated by ‘non-new’ school leavers and young ‘second jobbers’, including those leaving the textile industry after completion of the average several years’ work (as well as those leaving after gaining a senior high school qualification while working). However, perhaps the most important point is that between 1967 and 1973 the demand for older women, particularly those aged 30–44 years, increased significantly and in most cases was above supply–demand equilibrium (indicated by lightest shading in Table 3.11). It is clear that from the mid-1960s there was a rising demand for older female workers in the economy. While national statistics highlight the increased demand for older women in the labour force, PESO data also points to their increased employment in

response was that Japan’s labour force was sufficient in numbers if the ‘middle and older aged’ (chu¯ ko¯nen) groups in the labour population were taken into account. The MOL also said that the introduction of foreign labour would cause social problems: JSA (1979a) p. 797. 80 The term kyu¯ haku was used, literally meaning ‘be reduced to poverty’: JSA (1979a) p. 795.

0.7 0.8 1.0 0.5 1.0 1.3 1.4 1.7 1.6 1.2 1.6 2.3 1.1 0.7

0.7 0.8 0.8 0.6 0.8 1.0 1.0 1.3 1.2 1.0 1.3 1.8 0.9 0.6

0.9 1.1 2.5 1.8 1.9 2.3 2.6 3.7 4.2 3.2 4.4 6.5 3.5 2.9

19 and under

0.7 0.8 0.6 0.5 0.7 0.7 0.7 0.9 0.9 0.6 0.8 1.1 0.6 0.5

20–24 0.6 0.7 0.7 0.5 0.8 1.0 1.0 1.3 1.3 1.1 1.5 1.8 0.9 0.6

25–29

Females by age groupings

Note Figures are job-offers to job-seekers ratios.

Source: JSA (1979a) p. 774.

1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975

Total

Both sexes Females

Total

0.9 1.1 0.8 0.6 0.9 1.2 1.4 1.7 1.7 1.5 2.2 2.9 1.3 1.1

30–34 0.5 0.7 0.7 0.5 0.8 1.2 1.2 1.5 1.4 1.1 1.8 2.4 1.0 0.9

35–39

0.8

0.3 0.5

0.9} 0.8} 0.7} 1.1} 1.7} 0.7} 0.5}

0.6} 0.4} 0.5} 0.7}

40–44

0.6

0.3 0.3

45–50

Table 3.11 Supply and demand for ‘regular’ workers (i.e. excluding new school leavers) by age

0.2 0.3 0.3} 0.1} 0.2} 0.2} 0.3} 0.4} 0.4} 0.4} 0.6} 0.9} 0.4} 0.2}

50–54 0.1 0.1

55–59

0.1 0.0 0.2 0.1 0.1 0.1 0.1 0.2 0.2 0.2 0.2 0.4 0.2 0.1

60 and over

92 Recruiting women workers the textile industry. The overall proportion of females employed by textiles via PESO fell throughout the period, from 63 per cent in 1956 to 44 per cent in 1970, while the share via personal or direct recruiting connections rose from 22 per cent to 42 per cent respectively. Given the restrictions on directly recruiting new school leavers in place during the period, this indicates that new school graduates (young labour) were gradually being replaced by other (i.e. older) forms of female labour. The rate of school-led introductions also suddenly showed a rapid drop from a constant proportional share of 10 per cent up until 1968 to only 4 per cent in 1970, also reflecting a sharp move away from primarily young workers in the late 1960s.81 In the context of a national labour shortage, the government made serious attempts to encourage employers to expand their labour structures towards ‘alternative’ groups of the population, and from the mid-1960s onwards the middle-aged female labour force was mobilised nationally in Japan. Many entered production, factory and later service employment, under a range of employment status type. In general, the increase in part-time work for females was particularly notable, and although the textile industry was perhaps later than other industries in making use of this form of labour, even in this industry, where there was a long tradition of recruitment of young females, the recruitment of older women significantly increased during this period. While there was a longer history of older (regular) female workers in the weaving industry and in small-scale textile operations, the use of older female workers, particularly in the cotton spinning industry, fell mainly into the categories of part-time, day-only, temporary and seasonal workers. The example of To¯yo¯bo¯ highlights the dramatic increase in the recruitment and employment of older workers. The supply of new female junior high school graduates showed an increase in To¯yo¯bo¯ until 1963–64, but even during this time the employment of female junior high school graduates for the company, as well as for the industry as a whole, was recorded as insufficient. Each factory was coping with the annually intensifying labour shortage of junior high school females by gradually testing alternative methods and sources of labour. In To¯yo¯bo¯ the number of seasonal factory workers (commonly referred to as ‘special temporary workers’) suddenly jumped in number, from 400 in 1962 to over 1,000 in 1963 and reached over 3,000 in 1968.82 By 1970 JSA companies had actively developed and strengthened a tendency to use these alternative sources of female labour where possible. In this year, the employment of part-time females alone reached almost 10 per cent of the labour force, a record high for this postwar period.83 In 1971, special legislation promoting the employment of older workers (those aged 45+) by Japanese industry was established, and the industry recorded that 81 Kariya (1997) p. 182. 82 To¯yo¯bo¯ (1986) p. 399. 83 JSA (1979a) p. 797.

Recruiting women workers 93 the numbers of older females employed across the various textile sectors showed an increase. The legislation was later revised in October 1976, stipulating that those aged 55+ must comprise 6 per cent of a company’s employees. The industry noted that there were some cases, particularly in spinning, where the proportion of older workers had been below this rate, but that in small-scale companies where there were far fewer young school graduates, the proportion of older females was higher than this limit.84 Further revision of PESO legislation by the MOL, extending the time periods and application periods within which direct recruiting could take place, was also undertaken in 1971 and 1974. The textile industry commented that these reforms were in line with what they had requested.85 After 1973 the ratio of job vacancies to persons seeking work, which had steadily risen until then, suddenly dropped due to the economic uncertainty following the oil shock in November 1973. As Nakamura (1995) noted, after 1973 when the oil shock led to an energy supply crisis and almost all industries had to cut production, companies rationalised their labour force numbers and labour costs by first of all dismissing their part-time workers, a large majority of whom were (older) females.86 After cutting this part-time female labour force during 1973–74 however, many companies actively sought to rehire them after the crisis was over and the economy recovered. The textile industry also followed this general national response to the depression. As Dore (1986) commented, many textile companies said they had cut their numbers of part-time and temporary employees in 1974, although a few said they had maintained their recruitment of female parttimers because of the ongoing female graduate supply crisis.87 Although in many ways temporary in nature and susceptible to layoff, the increased recruitment of older groups of women was a clear trend during these years. Methods of recruiting older women The recruitment of older women in textiles often followed a traditional seasonal-based pattern. In general, the industry employed female junior high school graduates in spring (the Japanese graduation season). If labour requirements remained, female senior high school graduates and females who had graduated junior high school in previous years were employed as full-time workers. If the number of workers was still insufficient, as it 84 JSA (1979a) p. 797. In October 1976 a law promoting the employment of disabled persons (which had been legislation since July 1960) was also revised, increasing their rate of employment within enterprises. This was also part of the government’s strategy to encourage employers to employ wider groups of the population. 85 Permission for direct recruiting extended from six months to one year, with applications able to be made twice instead of once a year: JSA (1979a) p. 797; Nenju¯ no Bo¯sekigyo¯ (1973) pp. 69–70. 86 Nakamura (1995) pp. 222–24. 87 Dore (1986) p. 210.

94 Recruiting women workers increasingly became during this period, then other ‘temporary’ forms of labour were employed to make up the required workforce. There was also an established custom of utilising seasonal labour in the industry, especially during the period from January to March, prior to the annual intake of new graduates. Initially, this was due to the high turnover rate of young females and reduced numbers throughout the year as a certain proportion of the intake quit, gradually placing strain on employee numbers. In particular, many would quit after bonuses were paid in December, so January–March was often a period when extra temporary labour was required. Later, it was required more generally because of the shortage of young female labour from the outset of the recruitment year. Seasonal labour was therefore prominent, particularly during autumn and winter, when workers were recruited to maintain and ‘hold over’ the number of the workers until the next graduation season in spring. They were often referred to as hokan yo¯in (literally ‘complementary, essential workers’).88 There were three main reasons for the use of temporary workers in the textile industry during this postwar period. First, it was to make up for the increasing shortfall of young regular female workers over the period. Second, as the textile industry reorganised and later declined within the national economy during these years, temporary workers could be utilised to provide flexibility within the workforce. Third, the textile industry tried to operate during these years on a sufficient and fixed number of workers system, aimed at the most efficient use of technology and (minimum) labour. Temporary workers could alleviate any fluctuations in this model.89 The industry’s use of temporary labour began with the use of rinjiko¯ (temporary factory workers), a high proportion of whom were female. Although this type of worker had existed previously, a rapid growth in this category of worker occurred from the early 1960s.90 A survey of temporary workers across the industry undertaken by Zensen in 1962 showed that in 1960–61 the use of temporary workers was relatively low, though varied by scale. Large-scale companies reported employing on average 3–5 per cent of temporary workers (rinjiko¯ only) in their labour force, while small to medium-scale companies employed on average around 10 per cent.91 Zensen commented that the use of temporary workers, particularly in the large companies, was low compared to other manufacturing industries (average of 8 per cent across manufacturing sector) and particularly compared to heavy industries such as iron and steel, machinery and shipbuilding. However, the report commented that the low use was in part due to the nature of regular

88 JSA (1979a) p. 794. 89 Zensen (1962b) pp. 5 and 9. 90 The category of rinjiko¯ first emerged in Japan during the 1930s when the economy was booming. They disappeared during wartime, and re-emerged again in the 1950s: Azumi (1969) pp. 47–48. 91 Zensen (1962b) p. 5.

Recruiting women workers 95 workers in the textile industry. Because regular workers were young females who often worked for only a few years, in this respect they already functioned similarly to temporary workers.92 However, this situation rapidly changed from this point onward. In 1962 alone, there was only a 0.1 per cent increase in the number of regular workers, but an 18.8 per cent increase in the number of temporary workers.93 In the 1970s gradual improvement (including unionisation) of the employment position of these workers within companies took place. While it is difficult to pinpoint the real numbers of older female workers in the industry working as temporary forms of labour during this period, a feel for the methods employed in recruiting this type of worker can be gathered. It has already been noted that the proportion of women entering textile employment via personal connections and direct connections showed a rise during the period. Older women living local to factories could often be easily recruited through local advertising or through local acquaintances. While the PESO system was primarily a vehicle for the recruitment of young labour, older women and part-timers also came through the PESO system to a certain extent. Particularly when their recruitment was more systematic, such as ‘regular temporary’ or ‘part-time’ employees, PESO offices were used as recruitment contacts, often in conjunction with regional contacts such as local agricultural co-operative representatives, as introduction agents for older female workers.94 The temporary day-to-day employment of older females was often very ‘internal’. In times of extra demand, for example, housewives from company housing blocks within or near factory compounds (i.e. the wives of male textile workers) could often be called upon to do shifts.95 In 1967, industry documentation recorded that the lack of school graduates had become so serious that the industry had been forced to employ temporary and parttime workers as well as housewives living in company housing.96 This was not an entirely new development. In a Big Ten company video documenting the prewar history of the company, old film footage shows an infant nursery on factory grounds to allow older women and wives of male company employees to work factory shifts. In rural areas during the 1960s the employment of female day-labour was often quite haphazard. For example, an interview with a former labour manager revealed that company factories would run buses during the mornings and evenings which would take local routes. Those rural housewives who were free to work a day shift could just turn up at bus stops and be taken to the factory for the day.97

92 93 94 95 96 97

Zensen (1962b) p. 5. Zensen (1962b) p. 5. MOL (1970) p. 77. Nenju¯ no Bo¯sekigyo¯ (1967) p. 76. JSA (1979a) p. 797. Interview M.

96 Recruiting women workers A more permanent form of older female labour was the status of parttime worker (pa¯to). Although the practice of older women working nonregular hours as a non-regular employee was not new, the industry recorded that it officially introduced the formal status of ‘part-time worker’ from 1967. It also stated in 1971 that due to the serious lack of young female graduate labour, a prosperous economy and higher demand in recruitment, older women, particularly housewives, were employed in the industry as part-time workers, and sometimes full-time workers.98 More permanent part-timers were often previous ‘regular’ (experienced) workers. The fieldwork survey showed a strong link between working several years full-time (as regular employees) after leaving school, followed by a break for marriage and childbirth and a return to factory work again for several years as an older part-time employee. One method of seasonal recruitment used by the industry during this period was very systematised. In 1964 the textile industry initiated a special recruitment policy of employing female seasonal workers known as the ‘Hokkaido¯ Seasonal Labour Programme’ (Hokkaido¯ Joshi Kisetsu Ro¯mu-sha Ho¯shiki). The policy was utilised in the first two years by spinning companies only, but from 1966 expanded to all textile groups as a unified recruitment practice. The industry stated that it began to discuss ‘concrete’ policies of securing an alternative labour force (outside of young school leavers) during its implementation of this seasonal recruitment plan.99 In many ways, the successful deployment of older women within this seasonal recruitment programme prompted the industry to seek methods of securing this group of female workers within a relatively regular or stable employment pattern. The Hokkaido¯ programme was a response not only to the young labour shortage, but also to a Ministry of Labour suggestion to the industry that it alleviate its labour problems by utilising seasonal labour from Hokkaido¯. Hokkaido¯ was an area of concern to the MOL due to the high costs of providing unemployment benefits to unemployed labour during the non-farming season. The system, initially begun in Hokkaido¯, spread to other rural regions nationally. From 1966 the entire textile industry came to employ this method and to apply unified recruitment and employment conditions, as established by joint discussions within the industry.100 The main contents of these conditions were as follows: ●

the uniform establishment of a daily wage, starting allowance and experience allowance, with wages decided by considering a ‘balance’ with the wages of registered regular workers (see Chapter 6);

98 JSA (1979a) p. 795. 99 JSA (1979a) p. 794; Nenju¯ no Bo¯sekigyo¯ (1967) p. 76. 100 JSA (1979a) pp. 794–95; JSA (1979b) p. 27.

Recruiting women workers 97 ●

●

a regionally co-ordinated system, with a central aim of ‘assigned’ and ‘fair’ employment across the various regions where factories were located (whereby companies would be assigned specific numbers by region); seasonal period limits, whereby the method would be used annually from 1 October to 31 December (i.e. the workers would return in time to prepare for the farming season which would become busy from spring).

The reasons for the introduction of this method were cited in an industry report. First, due to the shortage of new female school graduates, the seasonal workers were to be a ‘complementary’ labour force. Second, the use of ‘idle’ labour in the non-farming season would give economic benefit to the Hokkaido¯ region. There was already an established tradition of dekasegi (migratory labour) in Hokkaido¯ (e.g. to inns and hotels) especially in periods of farming depression such as that caused by cold-weather damage. Third, for the Ministry of Labour, this would alleviate high unemployment benefit payments in Hokkaido¯, as well as aid the amelioration of regional pressure in the labour market.101 Table 3.12 shows figures for the employment of female workers under the Hokkaido¯ recruitment programme for available years from 1964. As can be seen, this method reached a peak in numbers in 1970, reaching over 4,000 women annually from this region alone during the mid- to late 1960s. However, it then appeared to show a subsequent decline. Although the decline in numbers from 1970 was reflecting changes in the economy which had begun with the Nixon shock in August 1971, the industry also cited the decline between 1970 and 1973 as being due to competition with other industries, which had launched similar programmes to secure older female workers and seasonal labour.102 The spread of the system to other areas would have also made Hokkaido¯ relatively less important. This increasing demand is also reflected in Table 3.12. In spinning alone, the job-offers to actual employment ratio was high, and increasing particularly rapidly after the late 1960s, reflecting the industry’s report that competition for even this type of labour was increasing. From the latter half of 1974, however, numbers declined due to the depression in textiles and the effects of the oil crisis. In total, 45 companies had planned for job-offers to 4,603 women in 1974, but in the end this was greatly reduced to only a single company using 25 women due to the oil shock. In 1975 the textile industry suggested to the Ministry of Labour that special steps (revisions) be taken in utilisation of the method from then on. This entailed making it more open plan with the application process to become direct recruiting rather than via PESO.103 101 JSA (1979b) p. 28. 102 Nenju¯ no Bo¯sekigyo¯ (1973) p. 69. 103 After 1974–75 the regional co-ordination of the system was abolished, but the unified/ standardised conditions for employment remained.

98 Recruiting women workers Table 3.12 Employment of female seasonal workers under the ‘Hokkaido¯ Plan’ Number of women Total

1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976

815 1,027 3,970 4,207 4,031 4,387 4,203 3,140 2,665 2,187

Number of companies

Demand and supply (spinning sector companies only)

Spinning Other JSA companies textile companies a only companies

No. job offers (A)

No. employed (B)

Ratio

815 1,027 1,665 1,723 1,484 1,692 1,645 1,314 1,082 848 25 b 270 300

2,245 3,208 4,400 6,599 6,628 6,727 8,165 8,144 7,366 8,200 105 1,045 1,194

815 1,027 1,665 1,723 1,484 1,692 1,645 1,314 1,083 848 25 270 300

2.8 3.1 2.6 3.8 4.5 4.0 5.0 6.2 6.8 9.7 4.2 3.9 4.0

2,305 2,484 2,547 2,695 2,558 1,826 1,583 1,339

31 47 127 160 166 164 173 161 140 141 13

(A/B)

Source: JSA (1979b) p. 29. Notes a Other textile companies refers to the six other fibre groups within the industry in addition to the cotton spinning sector. b Figure of 25 for 1974 is reported as ‘unclear’.

The demand and competition for this type of female labour was therefore increasing during the late 1960s and early 1970s. The recruitment of even seasonal and temporary female labour involved the relocation of company recruiting staff to rural areas for several weeks of the year. In these areas recruitment activity involved contact with PESO offices and staff, a liaison process with local agricultural co-operative representatives and negotiation and contact with rural households, often involving the traipsing from house to house in the region in an attempt to secure seasonal labour for the system.104 The recruitment of female labour during this time can be said to have been an intensive, exhausting, time-consuming and expensive process, even for supplementary forms of labour. In fact, the industry commented in 1968 that, ‘to our regret, even the employment of older workers is becoming difficult’.105 Given all the problems associated with the employment of young females during the period, as well as the complex and varied supplementary systems of alternative recruitment aimed at the employment of older women in the textile workforce, it must be asked why the industry did not attempt to place more emphasis on the recruitment of other forms of labour, particularly older female workers, as regular workers. 104 Interview M. 105 JSA (1968) p. 72.

Recruiting women workers 99 The transition to ‘older’ female labour In many ways, it can be said that the recruitment, employment and management policies of the industry (aimed at young female labour, with dormitory and educational systems – see next chapters) was lagging behind the actual structure of the labour force, which was gradually moving away from young to older women. This is not altogether surprising. The industry had a long history of employing young women, and many of its managerial policies and factory facilities were already set up for young female workers, and, it must be noted, had worked successfully in the past. During the 1950s when young labour was still abundant in Japan, these systems had been reconstructed and continued from the prewar period. In this sense, it can be suggested that there was probably some degree of path dependence, timelag and transition cost involved in changing from a core recruitment strategy aimed at young female school leavers. There was however some attempt to modify this in the mid-1960s. In response to the problems caused by the shortage of young labour, at least two of the Big Ten companies did attempt to set up factories based solely on the employment of middle-aged female labour. These test factories operated for several years in the 1960s, but in the end were deemed failures and operations ceased.106 The primary reason given for the failure of the experiment was the high ‘absenteeism’ of women due to their householdrelated duties and domestic responsibilities. It had been initially felt that this would be the cheapest method by which to operate a factory, due to there being no need to provide dormitory and education systems, thereby reducing the labour cost.107 However, absenteeism was found to be a serious problem and in the end the companies reverted to using young female workers where possible, despite all the problems surrounding their recruitment and employment. It should be noted that when the work of older women in the industry was cited in management reports, no problems with the work efficiency and ability of older women were noted, but only the problem of their ability to commit to regular employment (see Chapter 4). This experiment suggests that the industry was in effect not against the permanent employment of older female workers, but that social structures and the consciousness of women as workers versus their responsibility for the family and household meant it was difficult to use solely older women at this time. Those factories situated in rural areas also had the added complications of farming seasons when attempting more permanent employment of older rural women. The industry commented that it was not always easy to put in place recruitment and employment policy for the employment of

106 Interview L and Interview I. 107 The industry commented that by 1965 the cost of recruitment activities aimed at young female school leavers had greatly increased: Nenju¯ no Bo¯sekigyo¯ (1963) p. 77; Nenju¯ no Bo¯sekigyo¯ (1965) pp. 74–75.

100 Recruiting women workers ‘alternative’ forms of labour. Because it was primarily labour recruited from local areas, much of it depended on the area within which each factory was located and what kind of other businesses were in the same area.108 This early failure to secure older women as a permanent labour force no doubt helped lead to their intense use as supplementary workers and the formulation of a postwar pattern of part-time labour for older women within the Japanese economy. In many ways this period also reflects a gradual realisation of the need for older workers rather than an all-out direct effort to recruit them. Despite the problems associated with the employment of older women (in addition to the problems of employing younger women) industry statistics do show a definite transition from young to older female workers during this period. Table 1.4 in Chapter 1 showed the female employment structure broken down by regular and temporary status for the years 1967–75 in the JSA-associated cotton companies. Figures showed the significant use of temporary (older) women within the industry’s labour structures throughout each year. Although no earlier official statistics exist, industry documentation estimated figures for 1960, and these are shown in Table 3.13 and compared to the situation in 1976. While alternative forms of labour comprised only around 10 per cent of the labour structure in 1960, they were more than 40 per cent by 1976. It can be assumed that the situation previous to 1960 was even more marked. In light of the evidence presented in this section, and in the book so far, therefore, this period saw an important transition in the industry’s labour force away from a primary emphasis on young female labour to include diversified forms of older female labour as well. Until the late 1960s the employment of alternative forms of female labour had been undertaken by individual companies acting alone, but after this period the industry realised that a more unified approach was required to establish more concrete methods of recruitment and employment of older female workers, particularly labour management policies (see Chapter 4).109 Both employer attitudes and strong social domestic roles for older women contributed to their employment as non-regular workers. They had to juggle Table 3.13 The transition from young to older female labour in textiles (%)

Junior high school graduates Senior high school graduates Previous high school graduates Other (i.e. older females) Total Source: JSA (1979a) p. 776.

108 JSA (1979a) p. 798. 109 JSA (1979a) p. 798.

1960

1976

83 2 5 10 100

42 11 6 41 100

Recruiting women workers 101 their factory work with their childcare, housework and family obligations. In spite of such constraints, as the following two quotes depict, they often gave many years of factory service and were determined in their labour. I joined the company when I was 30 years old because my child was older and the factory was near my place (my neighbour was already working there). It wasn’t a particularly good job and the dusty air made me want to quit, but I couldn’t because I had taken the job against family opinion. It was difficult to handle both working and housework but I worked until retirement age. (Fieldwork survey respondent no. 30; employed 1968–98, 30–60 years of age) I don’t like the job, but I put up with it because the factory is near my house. Many times I have felt that it is too difficult, but when I come home I’m not able to say to my mother-in-law that I want to change my job. I think this is the reason I’ve kept on working. I think I’ll persevere a bit longer. (Fieldwork survey respondent no. 32; employed 1967– , age 24–) Overall, the use of less systematised and temporary recruitment programmes to employ this labour group highlights the fact that the economic contribution of older women during this period was significant, but that it was often hidden or not fully recognised. The substantial use of seasonal labour also indicates that women, particularly rural women, often had a triple burden during these years, not only the double burden of domestic duties and agricultural work, but also the added burden of supplementary factory (economic) work.

4

Managing women workers

Labour management is a management method for a firm to make its employees work effectively towards achieving group objectives. (Hazama 1997, p. 6)

The development of labour management strategies In many ways, the early postwar years for the textile industry were characterised by a ‘groping for a new direction’ in labour management, often based on trial and error.1 Various factors were at play including ongoing historical issues, legal changes, new labour relations and recruitment foundations, new management theories from abroad and the changing socio-economic environment. In some respects, early postwar textile managers were constrained both economically and psychologically by previous labour management institutions existing in the industry. At the same time, the changed postwar environment, as well as a desire to escape from the prewar image of the industry, brought pressure on managers to place renewed emphasis on managing workers. This resulted in the industry groping for new ways forward, firmly based on existing structures, yet expanding their aims and operation to address the changing environment. Before examining the various elements of the management style that was applied to female workers during these years, it is important to place labour management in the textile industry within a historical context.

1 Because the focus of this chapter is labour management, it has relied heavily on available internal company management reports collected during fieldwork, particularly those of To¯yo¯bo¯. In this respect, To¯yo¯bo¯ is a major case study within this chapter. However, other sources indicate similar issues for female labour management across the industry, and are used to back up the To¯yo¯bo¯ sources where possible. In this sense, the case of To¯yo¯bo¯ is felt to be representative of the main labour management issues facing the industry during these years, particularly in medium- to large-scale enterprises. While the various systems discussed would have been subject to some variation within different firms, the To¯yo¯bo¯ sources provide crucial information allowing examination of the labour management of female workers at a detailed company level.

Managing women workers 103 Labour management in historical perspective The development of early labour management ideologies in the prewar period was to a large extent constructed around ‘family’ based relationships. Hazama (1997) examined management concepts in prewar Japan, ranging from early boss-worker (oyabun-kobun – literally ‘parent–child’) relationships at an individual or small-scale business level, to later styles of managerial familism (keiei kazoku shugi) used by large-scale industry.2 As a large player in the industrial economy in the prewar years, the textile industry played a significant part in the formation of these labour management styles. Kanebo¯, for example, developed a management ideology known as large familism (dai kazoku shugi). The development of various forms of paternalistic management, at least in rhetorical terms, can be said to have been strong in Japanese prewar industry. While some researchers have suggested that Japan’s history of labour management can be seen as culturally unique, originating in Buddhist and native customs, it should be noted that foreign influence also played a role. Scientific management theories were introduced in the early twentieth century, particularly Taylorism, which was based on an analysis of individual jobs from a technical perspective and aimed at the most efficient job performance. This scientific based form of workplace operations (kagakuteki so¯gyo¯-ho¯) was introduced into the textile industry from around 1910.3 Other management ideologies were also developed by the large-scale textile enterprises. The management policies at Kurabo¯, for example, were ¯ hara Magosaburo¯) Christian-influenced strongly shaped by one man’s (O 4 idealism. It was not surprising, however, that family based styles of management were dominant, as much of the prewar economy was a familyunit based economy. Managerial familism was based on the concept of the ie (household), which in the prewar years held males and females in different regard in terms of status. Although women worked within industry, their employment was predominantly viewed as temporary and supplementary for the household income, and it was common for textile employers to exchange employment contracts with the male head of the household rather than the female worker.5 The style of employment and composition of labour force used by the prewar textile industry certainly created a distinct pattern. Because females were relocated often far away from their family homes into factorydormitory complexes, and because they were primarily young and unmarried, Hazama suggests that ‘their degree of dependence on the

2 Hazama (1997) p. 10. 3 Kanebo¯ (1988) p. 129. Frederick Winslow Taylor (1856–1915) developed this system of scientific management in the United States in the early 1900s (see Tsutsui 1998). 4 Kurabo¯ (1988) pp. 130–31. 5 Tsurumi (1990) pp. 126–27.

104 Managing women workers enterprise was extremely high, and they were immersed in familistic management from an early stage.’6 The enterprises took it upon themselves to train and rear workers from the time they were young (kogai yo¯sei), not only to secure skilled workers but also to create a labour force with a strong ‘corporate consciousness’. The textile industry also took it upon itself to ‘train’ female workers for their future married life by providing instruction in traditional female etiquette based arts, thereby promoting the ‘cultivation of womanly virtues’.7 This early establishment of significant company influence on an employee’s life, both working and non-working, had a large impact on the development of later corporate styles in Japan. The role that women played as workers in the prewar economy was also crucial. Until the 1930s it was female workers who were dominant in Japan’s industrial sector, and the management policies developed by the textile industry in particular were aimed at this group of worker.8 It was only as the heavy industries took off in the 1930s that the number of male industrial workers exceeded that of female workers in 1934. Hazama (1997) comments that from this period onwards one had no choice but to attach greater importance to labour management aimed at the head of household . . . a marked contrast to the period when labour management was mainly aimed at female workers who were regarded as inferior to men and were also only supplementing family income.9 However, because it had been women who were initially dominant in the industrial workforce, a management style essentially set up for their management established a precedent or blueprint for later management styles developed for male workers.10 The early management of women in the textile industry therefore had an influence not only on management styles that developed for males, but also on the form that the management of women would take in the postwar economy. Labour management in the postwar period While this discussion has pointed out that more formal theories of labour management and their implementation in practice were something that was not new, the postwar period was felt to bring a new era. One source even commented that, while in the prewar period the job of labour management

6 7 8 9 10

Hazama (1997) p. 63. Hazama (1997) pp. 92–107. See Sugihara (1996) ch. 12. Hazama (1997) p. 44. Witness the later development of corporate ideals such as the ‘company as family’ and the development of the corporate ‘salary man’ identity.

Managing women workers 105 had been often mainly to calculate wages, in the postwar period this had expanded to include labour management concepts.11 This would suggest that ideals of labour management began from scratch in the postwar period and developed centred around various fundamental new issues. While this is too simplistic a view, it does demonstrate the view and feeling among textile management that the postwar years signified a new period in labour management ideals and structure. In general, the postwar period brought a broader trend towards the development of management strategies in the industry. In the early postwar years, many management theories were brought in from the United States, in particular ‘human relations based labour management’ (ningen kankeiteki ro¯mu kanri-ron) and ‘organisation based theory’ (soshiki-ron) including the ‘line and staff system’.12 Textile managers researched and adapted these new systems to match industry circumstances, combining them with existing management methods developed by the industry in the prewar period and compiling new labour management handbooks for the industry. Two main factors were influencing employers as this process was carried out: first, the labour management history of the industry, and second, the contemporary environment. The issue of the exploitative management of female labour or Joko¯ Aishi in the prewar years was uppermost in the minds of managers as the industry’s labour strategies were re-evaluated after the war. A focus of the industry’s concern from the late 1940s onwards was how to overcome this inherited view of working conditions for the core female labour force. The inheritance of the Joko¯ Aishi concept had a very strong influence on the postwar image of the industry and became a basis for reforming ideas on the labour management of the female workforce. Kataoka (1993) noted that despite early postwar efforts to improve dormitories and to build schools, employers felt that the image of Joko¯ Aishi was still persisting.13 This was no doubt a factor that prompted not only a firm investment in improving existing conditions, but also the research into and implementation of various management theories from abroad after 1955. Contemporary factors influencing the development of labour management systems at this time were socio-economic in nature. These included the establishment of new national labour laws in Japan in the late 1940s, highspeed economic growth after 1955, subsequent labour shortage problems, the rise in education, as well as technological advances and associated labour

11 Kataoka (1993) p. 20. In many respects, this is a value judgement, based on postwar managers’ ambitions for the industry as well as a perception of what prewar managers had done. 12 To¯yo¯bo¯ (1974) p. 1. This was important in the development of ‘Quality Control’ (QC) circles within workplace management, which has received a fair amount of research attention in the postwar period (see for example Gordon 1998). 13 Kataoka (1993) p. 21.

106 Managing women workers rationalisation policies. While these have been discussed in the previous chapters, it is important to recognise that the changes in these years represented a very new environment for the textile industry within which new labour policies and management strategies had to be defined. The ‘gender factor’ played a large role in this process. The management of an industry where female workers dominated was felt to be very different from one where male workers predominated.14 The cotton industry, particularly spinning, was still geared to employ primarily young female labour, viewed as a workforce who would work for just a few years while they were ‘post-schooling’ and ‘pre-marriage’.15 Because of their youth they were described as ‘easily upset’.16 They were not to be organised within a lifetime employment system as males were.17 All of these ‘female’ characteristics influenced the development of labour management strategy, which for young female factory operatives was to be based on three main components: ●

●

●

welfare facilities (centring around dormitories) for the enrichment of their life; education to prepare them for their future life (not necessarily viewed as a ‘working’ life); in-house training to develop their ability as workers.

In the postwar years, their employment in large numbers straight from school within a structured annual system became known as shu¯dan shu¯gyo¯ (group employment) and played a large role in the organisation of this young workforce into a working and living community. Because women comprised the core labour force in the industry, the key management ideas that developed essentially became a system of ‘female labour management’. The changes during these years, discussed throughout this chapter, were responses to changed characteristics of the female workforce both by composition and consciousness, as well as a response by textile managers to changing socio-economic structures. There is little doubt that in the textile industry the dormitory and education systems were a central focus of the labour management strategy for 14 15 16 17

JSA (1969) p. 27. JSA (1969) p. 27. Comment made in JSA handbook: JSA (1969) p. 28. In the postwar period a distinctive ‘Japanese style of management’ emerged, particularly as Western researchers advocated its part in postwar economic growth. Japanese style management was defined by several concepts including a degree of ‘paternalism’ ‘managerial familism’ ‘groupism’, ‘lifetime employment’ and ‘seniority’. However, this style of management gradually came to encompass predominantly full-time ‘salary man’ male employees. While many of the components of male employment and management originated from precedents in the early management of females, it has been male workers that have been the prime focus of the research into ‘Japanese style management’.

Managing women workers 107 (young) female workers. A report by a labour manager in 1963 outlined how dormitories and education were to be seen not only as facilities (welfare), but also as an ideal method of labour management.18 The report noted that while welfare facilities, including dormitories and their management, were recent strategies in many industries and not just in textiles, such systems were not just about the investment of money in abundant facilities, but were to be extended to form consistent policies of warm-hearted and effective labour management.19 In this way, dormitories not only were to be effective systems of welfare, but also were expected to have an effect on human relations and the employment environment. Therefore, their management and administration was deemed to be very important. Dormitory management in itself was expected to respond to changing labour demand and supply circumstances and be an effective management tool in the positive recruitment of workers, as well as provide a strong component of human relations management, an idea of growing importance in this period. During the postwar years, an overriding central desire and focus of labour management was the creation of a stable labour force in the industry, and within this a fundamental aim to continue to use young females as the core workforce. It was felt that a unified management system within the textile industry as a whole was needed nationally in order to ensure this stability. The dormitory and education systems for female workers became a focus of this unified system. During these years, the development of labour management ideas and strategies to organise labour not only took place in large-scale cotton spinning enterprises, in conjunction with industry associations and unions, but also gradually saw a trickle-down effect to smaller-scale enterprises. The concept of labour management was an important issue for the industry during this period and was devised to work alongside the recruitment strategies discussed in Chapter 3. Labour management departments took on a central role within enterprises and were seen as employing the ‘elite’ staff.20 Various committees were formed as channels of discussion around core issues, such as dormitories, education and welfare, with committee members generally comprised of elite managers from different divisions within the industry. Such committees met regularly to discuss conditions and problems of labour management and inter-firm co-operation was often a feature.21 One report recorded that the central aim was to ‘manage the dormitory and education system by standing up to the criticisms of history and successfully incorporating the past, present and future’.22

18 19 20 21

Kobayashi (1963) pp. 138–39. Kobayashi (1963) pp. 137–38. Interview I. This chapter has made use of several available reports from such committees as primary sources, particularly those of To¯yo¯bo¯. 22 To¯yo¯bo¯ (1977) p. 15.

108 Managing women workers

Managing young female workers through the dormitory system Dormitories and the textiles workplace A key question that arises when considering the dormitory (and education) system is why the industry went to so much trouble to provide such extensive facilities, and how these aimed not only to organise workers but also to benefit the company. The dormitory system was felt by textile managers to offer a direct influence on the workplace, to contribute to industry aims and to have practical application to the workplace. The provision of nonworking facilities for employees could be viewed as somewhat of a detour in management strategy for the workplace. However, several features of the combined dormitory and education system were felt to offer sufficient benefit to the company and applicability to the shop-floor. First, a stable dormitory life had long been viewed as bringing a stability to the production process. The textile industry, particularly spinning, was a processing industry with a large dependence on overseas markets for raw materials as well as an inevitable severe competition for export markets, requiring a two-shift system of operations and a focus on high-volume production. The dormitory institution was based on the idea that dormitories were an important part of production management, which, if they were abolished, would make workplace management and the production process severely difficult. As long as factories continued in their present scale, location and operational structure, dormitories were felt to be inseparable from production. Further to this, it was felt that if dormitory management was carried out correctly, the industry could expect orderly and efficient labour operations as feedback. Given the transition to older (commuting) female workers that took place later in the period, it should be recognised that much of this argument suggests that the industry was continuing with an inherited status quo. Second, textile jobs such as spinning and weaving belonged to the category of ‘simple work’ and therefore, while females were viewed as suited for such jobs, it was also recognised that these jobs could become monotonous. Providing young women with a full dormitory and school life, it was hoped, would contribute to relieving the monotony and boredom of work and provide them with opportunities for ‘self-growth’.23 Furthermore, taking into account the characteristics of such jobs, it was difficult to measure the ability of an individual worker just by judging her performance in a small part of the production process. Production required good quality teamwork skills, and both communal dormitory life and educational development were seen as important in achieving those qualities. Third, dormitories and schools were recognised as the places for the development of human relations, as they intrinsically had a human relations based 23 Kobayashi (1963) pp. 146–47.

Managing women workers 109 administration system. The effectiveness of the administration of human relations within the dormitories and schools was therefore considered to have an important effect from a labour perspective on the human relations within the workplace, as these boarders were also workers. Of course, the shift system of production in operation in the textile industry also constituted in itself an effective method of organising and ‘managing’ the employment and work patterns of the core female workforce. It was the basis for calculation of work hours, the allocation of women within work teams and the arena for shop-floor training (see Chapter 5). The shift system naturally dictated daily life routines for female workers, not only work hours but also sleep patterns and meal times. It was also a key basis for the organisation of female workers into dormitory residence. However, it was dormitories in themselves that evolved to become a crucial tool of labour management strategy. The dormitory system was therefore a legacy of the industry’s long history of utilising young female labour, recruiting long distance and running the shift-system mode of production. In many ways, the continued prominence of dormitories in the postwar years may be viewed as management’s utilisation and continuance of something concretely in operation within the industry already, i.e. a degree of path dependence. However, this did not imply that it was a static system or one that was simple to pick up again and use after the Second World War. One company report noted that management had to continually try its utmost to reject ‘patchy’ and ‘glossed over’ policies with regard to dormitories and had to ‘endeavour to consider and measure the essential elements necessary for the adaptation and continuation of this system over time’.24 In this respect, despite its lengthy and customary use within the industry, the system required a high degree of maintenance and monitoring in order to be effective under postwar conditions. The industry’s continued employment of young female workers also determined the continuance of the dormitory system. The postwar industry had in various respects been faced with a status quo in its re-establishment during the early years, and had chosen to deal with it by continuing and strengthening the use of the dormitory system to house and effectively manage a core labour force of young women. Particularly in the first two decades after the Second World War, and particularly in large-scale companies, some 90 per cent of all female employees were resident in dormitories.25 As long as the central recruitment strategy during these years aimed at young female school leavers, this meant that the dormitory (and education) systems continued to be prominently placed within management strategy. Therefore, despite some changes, not least that the number of commuters gradually increased during these years, the Japanese cotton industry still felt the need to maintain the dormitory system essentially in its 24 Kobayashi (1963) p. 151. 25 Kobayashi (1963) p. 139.

110 Managing women workers traditional operational form. It is certain, however, that dormitories were also used as a means to integrate all young female employees into a specific management structure. The results of my fieldwork survey indicated that even young girls recruited from areas local to factories were required to enter dormitory residence.26 The JSA certainly viewed dormitories as a key focus of the management of labour within the industry.27 It encouraged textile companies to draw up clear management ideals and aims for the operation of dormitories. Using To¯yo¯bo¯ as an example, we can see that these included both tangible and less tangible aims.28 Dormitories were to be essentially a place for private group living and private life. It was therefore necessary to handle them as a ‘separate sphere’ not directly associated with employment. The company position was to be one of administrator and owner of the facilities, while dormitories were to operate through their own self-government. Dormitories were expected to be a ‘place of rest, development and growth for young people’ and a ‘communal society’. The company’s role was to supply co-operation and assistance to enable this to happen.29 Based on these essential ideals for the quality and character of dormitories, To¯yo¯bo¯ identified two central management goals: first, life in dormitories was to be ‘cheerful’, ‘comfortable’ and ‘abundant in content’, and second, cooperation was to be given towards development of the self-government system within dormitory operations. With these ideals in mind, the labour management of female workers in the industry can be examined through the day-to-day operation of the dormitory system from two main perspectives: first, the management of dormitories ‘from above’, i.e. the role of the company as administrators, and second, their management ‘from within’, i.e. the role of boarders and the so-called self-government system. Managing dormitories from above The case of To¯yo¯bo¯ is used here to provide an example of how the administration of dormitories operated at a detailed level.30 In general, the structure of dormitory administration was based on the idea that companies were to provide two levels of support to dormitories. These were classified as ‘material’ and ‘spiritual’. A range of dormitory administrators were required both to handle concrete administrative affairs and to provide 26 Several respondents from the fieldwork survey (9 out of 45) said they were recruited from the local area (same town or nearby town in prefecture) but said that they entered the dormitory. Three stated that this was ‘compulsory’ or ‘obligatory’ and one said that this was what ‘all the girls did’. 27 JSA (1969) p. 34. 28 An early To¯yo¯bo¯ plan for the management of dormitories was entitled ‘Essential Characteristics of Dormitory Life’: To¯yo¯bo¯ (1977). 29 Kobayashi (1963) p. 151. 30 Kobayashi (1963) pp. 147–64 is a main source here, except where cited otherwise.

Managing women workers 111 personal guidance to boarders. The structure of officials and administrators in a medium- to large-scale dormitory attached to a factory typically comprised several positions. Within the labour management division of the textile company, there would be a (male) dormitory section manager. A regular management-level employee, he would be responsible for overseeing and supervising the operation of the dormitory at the company level. It was usual to have one or two males with official and ultimate responsibility for dormitories and several female positions running the dormitories below them.31 Beneath this dormitory section manager there would typically be the (female) position of dormitory manager or matron. This was a live-in position and was the real dormitory manager from a practical perspective. Administrative support staff within the dormitory generally comprised two types: live-in (female) office administrators and cleaning or maintenance staff (usually comprised of older commuting men and women). Finally, forming close contact with all managers and administrators was the dormitory medical nurse. In terms of profile, dormitory matrons were generally a mix of young female university graduates and older (middle-aged) women. An example of their profile in To¯yo¯bo¯ was presented in Kobayashi (1963).32 First, by age, 24 matrons were in their twenties, 21 matrons in their thirties and 29 matrons in their forties. Second, by length of service, 25 matrons had under 5 years’ service, 27 matrons had 5 to 15 years’ service, and 22 matrons had served 16 years or more. The 1972 MOL survey also noted that the average age of matrons fell into two major age groupings; those in their twenties comprised 38.3 per cent and those in their forties comprised 26.1 per cent. This indicates that matrons were usually either fairly young (but still several years older than the majority of boarders) or middle-aged. These age groups were naturally thought to act as either older peers or older ‘sisters’ (senpai) or as surrogate mother figures for young female boarders. Depending on the scale of the dormitory, matrons might be day-only, shift-workers, or a combination, but they were primarily full-time positions.33 The role of older women 31 MOL (1952) p. 7 noted that male dormitory managers did not generally interfere in the self-government system of the dormitories. 32 Kobayashi (1963) p. 154. 33 The 1952 MOL survey noted that 80 per cent of (cotton spinning) dormitory administrators were full-time and female (including dormitory managers and matrons), and half of those not full-time were also employed as regular staff in company labour divisions: MOL (1952) p. 7. The 1972 MOL survey noted that 37.5 per cent of dormitories did not have an official matron, but 33.4 per cent had 1 matron, and 28.7 per cent had 2 or more matrons, MOL (1972) p. 41. Corresponding to the range of enterprises by scale inthis survey, this would indicate that the majority of medium- to large-scale dormitories had at least one matron or on-site female manager, while smaller-scale factories were less likely to have this full-time position. Observations during a fieldwork visit to a sanchi revealed that in many smaller family-run weaving businesses, the wife and/or grandmother would be responsible for looking after the young girls employed and housed onsite. In this sense, they had the role of ‘surrogate mother’ and ‘matron’ (Interview K).

112 Managing women workers in these non-production and female staffing positions was clearly important within the dormitory system, and it was possible for women who had worked several years in the industry as factory employees to be promoted to these positions. Office administrators, on the other hand, were primarily younger, mostly senior high school graduates who lived in among the other female boarders, and who could often move up to manager level over time if they did well. They were often those who had shown ability and initiative working in the factory initially. The duties of a matron were outlined as primarily twofold – those involving the dormitory facilities and those involving the people (boarders). The first included jobs associated with the maintenance and administrative use of dormitory buildings and equipment. They were expected to assess the living environment in order to provide boarders with both a cheerful and comfortable life and were to ensure the provision and maintenance of equipment at a suitable level to maintain that ideal. Provision of care was also required with regard to boarders’ health and physical needs. The latter duties were based on ideas of human relations management. They revolved around the care and guidance of boarders at the two levels of individual and group, including the need to co-operate with the self-government committees to ensure their smooth running and also to ensure a ‘smooth balance between individual and group life’. The matrons were also an important point of contact between the company (via the dormitory section manager) and the boarders (workers). A dormitory matron’s duties were therefore of two crucial types – administrative jobs and caring jobs. Because the dormitory boarder structure was comprised of many young females, one of the matron’s responsibilities was to develop the means of providing care and guidance for young boarders in her charge. On the other hand, because the aim of the dormitories was that they be run in the form of self-government, matrons were also encouraged to foster the independence of boarders as much as possible. In line with this goal, it was deemed to be often important that the exercise of direct administration (upbringing of boarders) take a back seat to a more indirect administration (encouraging boarder independence). In light of this, it was felt that matrons should therefore possess the fundamental qualities of friendship, caring and intelligence, be healthy and cheerful in nature, and have consulting skills. Overall, they had to have the ability to carry out the administration of a large household and ensure the smooth running of that household. This involved the upbringing of young personalities, the maintenance of the household environment, and co-operation with selfgovernment operations, the company and with schools. Females in general, and particularly older females, were felt to be very suited for this type of household management position.

Managing women workers 113 A crucial component of the labour management strategy for young women living in company dormitories was that it was to provide for their education in the skills necessary to manage their own private lives, and to promote their growth as ‘individuals within a society’. To achieve this, it became common in many large-scale enterprises to no longer think of female boarders in a unified bunch as young, but rather divided into age and length of service classifications. In To¯yo¯bo¯ this operated as follows: ‘Juniors’ included two groups – ‘first-term juniors’ (1–2 years’ service) and ‘secondterm juniors’ (3–4 years’ service); ‘seniors’ (5 or more years’ service) were viewed as adults (seijin), being 20+ years of age.34 Various guidance and education policies were formed aimed at what was essentially an attempt to manage the different age levels. During this period, many large-scale companies introduced a system whereby older or senior girls lived in housing separated from the main dormitory complex. This was set up in the form of small apartments or households where they could organise themselves within smaller living groups and have more freedom (e.g. freedom to entertain in the evenings).35 Zensen commented that, in general, there was a move towards making female dormitories more like apartments in form, particularly from the 1960s.36 In many ways, this was to prepare older girls for running a household and for marriage within the wider context of the industry’s responsibility for the ‘socialisation’ of young females. An example of the ideal labour management of young females by age level, formulated by the JSA, is shown in Table 4.1. It speaks for itself in terms of its detailed policy and guidelines on how textile companies could effectively organise and socialise young female employees through their residence within factory dormitories. Managing dormitories from within In 1947 the JSA, after carrying out thorough discussions on the management of dormitories, formulated a plan entitled ‘Measures for Democratic Improvements in the Dormitory System’. Mindful of the provisions of the Labour Standards Law, the aim of this programme was to set out new foundations for the operation of textile dormitories, whereby employees

34 To¯yo¯bo¯ (1969a) p. 14. 35 The term seijin-ryo¯ (adult dorm) or katei-ryo¯ (household dorm) is used in other industry sources, indicating the use of a similar system across the medium- to large-scale enterprises under their association: Zensen (1996) pp. 112–14. 36 Zensen (1996) p. 48.

17–18 years

16 years

15 years

Age

●

●

●

●

●

●

●

●

●

●

●

●

●

●

Become bored with factory and dorm life and want to quit for no thought-out reason Interest in appearance and fashion Interest in the opposite sex Think selfish is a good thing

Become used to factory life Somewhat selfish and still lack understanding of bigger things Still no real opinions of their own Youth/teenager years beginning Begin to like trends

Undeveloped socially Manners not good No opinions of their own Spend all their wages Quick to learn

Characteristics of age group

●

●

●

●

●

●

●

●

●

●

●

Encourage development of student and senpai life Instruction on how to behave with opposite sex Encourage thinking on purpose of life

Instruction on how to be senpai (older peer) Importance on community of dormitory life stressed Instruction on ‘destiny’ and ‘fortune’ Developing consciousness as students Teaching about society

Introduction to group life and to society Teaching distinction between right and wrong Teaching manners and etiquette

Management aims

Table 4.1 Example of an ideal labour management policy by age level for females

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

Lectures on society; use of slide shows, tapes etc. Emphasis on club activities Encouragement of developing hobbies Recognition that this is difficult age and providing encouragement in all areas

Encouragement of participating in club activities How to plan for rest of student/school life Group discussions on books and life topics Encouragement of reading and keeping of diary Budgeting meetings and instruction on how to save with purpose How to be kind to others Encouragement to write thoughts in company magazines

Instruction on etiquette (greetings, bowing, manners, time keeping etc.) Basic budgeting and guidance on how to shop Instruction on how to write letters and use of words Recommendations on what books to read

Management content/methods

●

●

●

●

●

●

●

●

●

●

●

Growing interest in marriage Develop understanding of what is important Growth of individual activities Want to move at own pace

Graduate from the factory education system Begin to think about future and marriage Become interested in accomplishments (gigei) Cliques grow Take on dorm group life Still lacking common sense Big interest in sex

Characteristics of age group

Source: Taken from JSA (1969) pp. 38–40.

20 years

19 years

Age

●

●

●

●

●

●

●

Adult life begins Encourage correct way of viewing marriage Advanced instruction on budgeting

Encourage them to think about poststudent life as a ‘new life’ Encourage pursuit of further qualifications Emphasise how to develop group life experience Meetings with medical and counselling staff

Management aims

●

●

●

●

●

●

●

●

●

Provide consultations on marriage Set up visits to talk with wives in company housing Set up meetings with previous female employees Instruction on Engels coefficient and budgeting Provide fullness of content in the household dorm

Encouragement of interest in junior college or specialist school courses Lectures on how to make best use of savings Visits to household to explain education to parents Place them in (adult) ‘household’ dorms (kateiryo¯)

Management content/methods

116 Managing women workers (boarders) would ultimately run a system of ‘self-government’. The role and policy of textile enterprises would be to co-operate with this selfgovernment structure to provide a ‘happy’, ‘congenial’ and ‘beneficial’ environment within the dormitories. General rules for self-government administration and a system of contact with self-government were drawn up. This policy was in recognition of not only the ‘democratic’ climate in Japan at the time, but also a recognition of the dormitories as a place where individuals had the right to have a private life, an area which had come under much prewar criticism. Self-government administration in dormitories was based on various groups and committees of boarders organised around specific types of administration or events. The main officials of self-government were the Self-Government Committee president and vice-president, the treasurer and the committee members. The 1952 MOL survey noted that those dormitories with no self-government system in operation were the exception, and in general self-government rules and official dormitory guidelines were decided upon separately. It was usual to have an annual general meeting for major resolutions and decisions, and several small committee meetings for simple day-to-day resolutions.37 Again, using To¯yo¯bo¯ as an example, the main organisational arms of the self-government system were typically as follows:38 General Meeting (AGM) Committee, Self-Government Committee, Standing Committee, Specialised Committee, Dorm Self-Government Committee and Boarders Committee. While the AGM and Self-Government Committee meetings were largely annual events, the Standing Committee met more frequently (on average monthly) to discuss daily life issues such as dormitory regulations, health and facilities. The Specialised Committee was for dealing with specific culture and sporting events and met as required. The Dorm Self-Government and Boarders Committees met less frequently and mainly at special times of the year, such as when preparations for the arrival of new graduates or boarders took place in March–April, or when special problems and events in daily life arose requiring discussion. While those in the posts of president, vice-president and treasurer were generally required to attend all of the various self-government committee meetings, a selection of representatives from among all boarders usually attended as committee members. All boarders were free to attend the large annual meetings, while for the more regular meetings a cross-section of all boarders might be required, e.g. a representative from each dormitory room or block might be chosen. The more specialised meetings were usually attended by those with a particularly interest in attending, i.e. those

37 MOL (1952) p. 7. 38 To¯yo¯bo¯ (1961b) pp. 13–15; Kobayashi (1963) pp. 159–60.

Managing women workers 117 with an interest in specific sports or cultural events would serve as regular representatives. The various official post-holders within the self-government committees were elected annually by ballot and all boarders were eligible to vote. The posts ran annually from March to March, and individuals could stand for re-election. Those opting to run for the various posts usually had to have been living in the dormitory for a certain period of time, and on average held their posts for some three years.39 Boarders paid small fees, usually monthly, towards the self-government committee, primarily to be allocated to various special events rather than the day-to-day operation of facilities, which were company funded.40 The company would also provide additional money for special cultural and sporting events. Although aimed to be self-run, typically the female dormitory administrators had some form of relationship with the selfgovernment organisations, dealing with any problems that occurred and acting as a point of liaison with the company. In addition to the internal administrative bodies already outlined, there was also the Self-Government Contact Committee. The structure of this committee included factory managerial representatives, the dormitory section manager, dormitory matrons, Self-Government Committee members and also a representative of the labour union who would be present to act as an observer only. This committee was the primary point of contact between the factory or company, the dormitory and the union, and on average met once a month. Discussion centred mainly around dormitory equipment and selfgovernment requests, though sometimes there might be specific requests from the company. In this sense, the dormitory section manager was not responsible to the residents despite the ‘democratic’ set up of the dormitories. The relationship between the self-government committees and the labour union was found to consist mostly of the discussion of economic issues, centring around facilities.41 Zensen held an annual national conference for representatives of dormitory self-government committees, and at each conference surveyed attendees on various issues. Two main issues at the 1961 conference were

39 To run for a post in To¯yo¯bo¯ one had to have been resident on average a minimum of 6 months: To¯yo¯bo¯ (1961b) pp. 11–12. A Zensen survey found that on average workers could be promoted to positions within the self-government committees after 2–3 years’ service in small- to medium-scale enterprises, and after 4–6 years’ service in large-scale enterprises: Zensen (1961) pp. 35 and 37; 60 per cent of the female survey respondents were self-government committee presidents (including vice-presidents) and on average had held their post for 3.4 years: Zensen (1961) pp. 38–39. 40 Token fees of around 50–80 yen per month recorded in To¯yo¯bo¯ (1961b) p. 18 and Kobayashi (1963) p. 150. These amounts related to less than 0.5 per cent of the average monthly basic (contractual) wages for female workers (if related to figures in Table 6.1 in Chapter 6). 41 MOL (1952) p. 7.

118 Managing women workers problems in the dormitories and future plans or hopes for self-government.42 When surveyed on what their committee members (boarders) wanted most, 47.1 per cent responded with issues relating to dormitory facilities, 44.7 per cent responded with self-government management issues, and 8.2 per cent responded with other issues.43 The main dormitory issues (desires) in 1961 were abolition of shared rooms, improvement of overall facilities, an increase in dormitory equipment and the provision of individual electrical sockets for personal (boarder) use. The main self-government issues were the need for better exchange with unions, desire for improvements in food, moves towards a more pleasant (tanoshii) dormitory life, and the problems of acquiring full-time dormitory staff. The prominent other issue was a desire for an increase in wages. An improvement in all these issues can be said to have taken place during the 1960s and 1970s and is discussed in detail in Chapter 6. The survey commented that it was difficult to design a unified dormitory policy across the industry, due to the range in size of dormitories (ranging from 15 to 3,000 boarders) and the range in composition of boarders (ranging from 15 to 50 years of age and from junior high school to university graduates). However, despite this, the survey concluded that, in general, unified efforts and movements for specific dormitory policies, particularly improvement of dormitory conditions and their management, had brought about a brighter and happier dormitory life and helped overcome the problems and image of Joko¯ Aishi that had been rooted in the industry’s dormitory history.44 The organisation of dormitories in terms of daily routine followed the two-shift system of factory production. Routine centred around the primary activities of work, meals, schooling and recreation, and at a general level was classified into ‘awake’ and ‘sleep’ hours. While the shift system naturally dominated routine, the self government body was influential in setting nonwork patterns. Sleep hours, for example, were generally set at eight hours per day, and this was seen as an important rule so as not to cause problems for others, due to the nature of communal living and ‘in-and-out’ daily movement of boarders commuting to and from shift-work.45 Dormitories were often organised, in terms of arrangement of rooms and building wings, in separate shift spaces and in age-classification groupings. These included

42 Zensen (1961) p. 34. This available document is the survey from the tenth annual conference held by Zensen in 1961. In that year 822 persons (both union and selfgovernment representatives) attended the conference, 68 per cent of whom were female. Attendees were from the five main ‘union groups’ (cotton, synthetics, wool, hemp and the ‘local’ unions), and 32.4 per cent of all attendees responded to the Zensen survey. The average age of female attendees was 24.4 years and the average length of service was seven years, both of which were reported as having increased in recent years. 43 Survey results of female respondents only: Zensen (1961) pp. 44–45. 44 Zensen (1961) p. 58. 45 Kobayashi (1963) p. 148.

Managing women workers 119 consideration for factors such as production/shift groupings, new-boarder groups, school-attendance groups, and junior-senior age groups. Rules on moving to and from work and noise-levels were established by the selfgovernment bodies. An example of how the dormitory routine managed the daily life of boarders is shown in Table 4.2 (see also Plate 4.1, pp. 120–21). In addition to daily routine, there were various annual and one-off dormitory events. These included company-organised events for all employees, company-organised events only for dormitory boarders, club-organised events and dormitory self-government organised events. The nature of events ranged from cultural, educational, sporting and recreational to workoriented health and safety activities. In addition to its practical operational role, the self-government system was viewed as a means of growth and development for young females. As members of the various selfgovernment committees the female boarders had to conduct meetings, make administrative decisions and have contact with outside groups in the organisation of events. The aim was that this experience would enhance their qualities as both workers and ‘social individuals’. During these years, this ideal was continually built upon in management plans. A 1969 management report stated that all the dormitory personnel positions were to be Table 4.2 Example of daily routine for dormitory boarders Wake-up

Room cleaning

Early shift

04.30

04.30

Breakfast Lunch Dinner

07.30 13.30 17.30

05.00–13.30

Late shift

07.30

07.30

Breakfast Lunch Dinner

08.15 12.45 18.45

13.30–22.00

Day-only shift

06.00

06.00

Breakfast Lunch Dinner

06.15 12.00 17.00

07.00–15.30

Early shift Late shift Day-only shift

Meals

Factory work

School (max. 3 hrs) Bath

Cooking/ facilities use

Sleep

14.30 09.30 16.30

20.00 12.00 20.00

20.30 23.30 22.00

18.00 22.00 18.00

Source: Kobayashi (1963) p. 169 Notes The use of dormitory equipment (e.g. ironing and laundry equipment) was often also scheduled for each shift within dormitory rules. Curfew times were set at 21.00 for the early shift (22.30 for the night before the weekly rest-day) and 22.30 for the late shift. If boarders wanted to stay away from the dormitory overnight, they had to make an application via their room leader or dormitory leader: To¯yo¯bo¯ (1961b: 12) p. 4.

Shot 1 4.15 am – wake up and morning ablutions.

Shot 2 Just before 5.00 am – off to the factory.

Shot 3 5.00 am to 1.30 pm – working the early shift.

Shot 4 2.00 pm – cleaning duties, then off to school.

Shot 5 5.00 pm – a group photo after lessons.

Shot 6 6.00 pm – dinner.

Plate 4.1 One day in dormitory life, 1958. Photos courtesy of To¯yo¯bo¯.

Managing women workers 121

Shot 7 7.00 pm – relaxing in the dormitory.

Shot 8 9.00 pm – sound asleep.

Plate 4.1 continued.

more clearly defined and laid out. In addition there were further plans to continue to emphasise and promote the increased consciousness of the girls to carry out self-government themselves.46 The dormitory system certainly played an important part in effectively organising and managing young female workers. It also became a large part of their lives. Many of my fieldwork interviewees and survey respondents recalled with fondness and nostalgia their time spent in textile dormitories. They spoke of the camaraderie, the friendships, the recreational events and the communal living. As one respondent reveals, dormitory life could touch and shape their future lives. I started working at 15 years of age and I’m still working for the same company. I have good memories of my life in the dormitory. In the dormitory, we had a schedule – we set the day for checking the equipment, times for cleaning communal areas, times for doing laundry etc. I do it in my house now too. I set a day for various tasks, for polishing glass etc. I was introduced to my husband by a colleague who worked parttime and we got married three years later when I was 24. It was as if I became a bride of the dormitory rather than my home. I worked only the day shift after I got married and brought up two children. Now I’m working as before because the children don’t need as much care and the factory is short of workers. I’m now a workplace leader and hope to work until retirement, even though the factory has become smaller because of the bad economy. (Fieldwork survey respondent no. 19; employed 1966– age 15–) 46 To¯yo¯bo¯ (1969a) p. 12.

122 Managing women workers Dormitories in transition Maintenance of the dormitory system naturally encountered the need for change during the period under study. The monitoring of dormitory operations and requirements was an ongoing part of discussions and reports by industry labour management committees, from which central themes and developments can be identified.47 Management reports generally examined how management of the dormitories had progressed to date, identified problems for attention and focused on implications and requirements for the future, particularly the need for adaptation in light of any changes in the female labour structure. In general, three main issues can be said to have emerged during the period: first, those relating to dormitory personnel and administrative problems; second, the issue of the declining supply of young female labour and diversification in the female workforce; and third, general socio-economic and environmental changes. These issues, and the industry response to these, highlight that the dormitory system may have been ‘inherited’ from the prewar decades, but it was certainly not static during these years. It also demonstrates the need to constantly re-evaluate the effectiveness of the labour management programme for young female workers. Several problems concerning dormitory management and administrative personnel were identified by To¯yo¯bo¯ in the ongoing revision of the dormitory system.48 There were found to be limits to effective decision-making within the operational structure, mainly resulting from an unclear assignment of responsibility across the various dormitory staff positions and self-government bodies. Dormitory matrons were found to be often caught in the middle between company expectations and boarder expectations. They were to be encouraged to administer from a position of ‘fairness’, in an attempt to resolve the current situation where they had to treat females as either boarders or workers (i.e., moving beyond the dilemma of being caught in the middle to having the ability to judge and deal with both sides).49 A review of staff positions found that they involved too much deskwork and odd jobs, leading to slow problem resolution. Moreover, managers stated difficulties in securing required numbers of dormitory personnel, as well as personnel with adequate ability. It was noted that the use of many junior college female graduates often meant that they left after only two to three years’ service and had to be replaced. This made it difficult to establish a degree of continued guidance and provide for a flow in dormitory operations. Finally, it was felt that effective group ‘life guidance’ for young girls was not taking place. Key solutions for these problems included strengthening the independence and autonomy of the self-government

47 Company reports include Kobayashi (1963), To¯yo¯bo¯ (1969a) and To¯yo¯bo¯ (1977). 48 To¯yo¯bo¯ (1969a) pp. 8–9. 49 To¯yo¯bo¯ (1977) p. 8.

Managing women workers 123 system, and the establishment of a new post within dormitories known as ‘life consultant’. This would not be a desk job and would be filled by somebody (ideally female) with counselling skills, with general experience and ability to provide advice and assistance, and with free status to resolve such problems. In many ways, this was a recognition that bringing in more skilled personnel for female workers within the dormitory management system could prove beneficial to operations, and this was introduced in To¯yo¯bo¯ from 1969. A second feature of management reports was the issue of changes in the female labour structure during this period. Under a labour surplus in the prewar and early postwar periods, the industry had developed its female labour management strategies based on young female workers living in dormitories. This was essentially a singular or unified (homogeneous) labour force and labour management system, at least in theory if not in practice. However, the diminished supply of young female labour and the diversification of the labour structure led to a recognition that there was a need to not only develop alternative systems of management but to also implement changes to the existing dormitory system. Overall primary aims for the management of female labour were set as securing the required labour force, sustaining that labour force, and the development of a labour force with ability. The dormitory system was still viewed as fundamental to these main aims, and revisions to the system were proposed. There was to be a move in emphasis away from the consideration of dormitories as a place of group living to one which also considered the needs of individuals. In light of this, management goals were to take into account a move from so-called ‘heteronomy to autonomy’, from ‘singular management to pluralist management’, and a movement from the ‘convenience of management’ to the ‘convenience of the residents’.50 The average age of dormitory boarders had risen by the early 1970s. While, 70 per cent of dormitory residents were still between the ages of 15 and 19 years, 30 per cent were now over 20 years of age. Suitable management methods relative to the varying ages of boarders were to be considered and included when making dormitory policy. It was recognised that, while the role of young women as a central labour force had declined relatively over the years, and subsequently the number of boarders had also declined, they were still being actively recruited. Therefore, dormitories, which were the centre of young female labour force management, were still deemed important. In order to invest, however, in labour stability and continuity, with the aim of developing a good quality workforce with ability, the dormitory system had to adapt to the demographic changes in the workforce. Overall, however, improvements in young female labour management in general were still viewed as centring around improvements to the dormitory system.

50 To¯yo¯bo¯ (1969a) pp. 3 and 5.

124 Managing women workers The effects of general socio-economic trends and change were also subjects for discussion during this period. It was felt that there had been a change in the social consciousness of young females, to which the management of education and dormitory systems must adapt.51 The rise in education and living standards as well as the improved status of women as workers and members of society were all considered. Both the dormitory and education systems as well as the theme of industry having a ‘social responsibility’ to young females in their (working) care had to reflect such changes. It was recorded that the methods and system implemented by the industry to date ‘had not been wrong, but required significant adaptation to the external environment’.52 Recent changes in the industrial environment, such as technological progress, reductions in workforce numbers and moves away from the labour-intensive nature of textiles were also to be taken into account when considering future policy. Finally, the cost of labour and the cost of management was changing. Indeed, the constant need for monitoring of management systems with reports and committees, and identification of ideas and devices to improve and make effective the process of management as an ‘institution’ was noted as a significant postwar trend. The response of textile managers to the range of issues outlined primarily centred upon adaptation of the dormitory system to take into account these changed conditions – the need for skilled dormitory personnel, the diversified structure of the female workforce and dormitory boarders, and the effects of socio-economic transition. The environment around the individual young boarder was noted as comprising the factory workplace, the dormitory, schools, family/household and society.53 A ‘perfect principle of management’ was deemed to be one which took into account all of these elements when considering future management policy. While mechanisms for this were felt to be in place under ‘perfect conditions’ within the system already, it was noted that any further slimmed down structure resulting from a decline in dormitory personnel and/or boarders could create the potential for halfway measures. That is, any weakening of the system would naturally lead to conflict between those guiding and those receiving guidance. Efforts were to be implemented to tone down excessive or overbearing strategies that might result in feelings of restriction for boarders. Rather, companies were to promote ‘freedom alongside sufficient guidance’, leading to better self-government.54 While elements of the system might appear restrictive or overbearing, it is also important to remember that it was not an easy task to effectively manage large numbers of young girls as well as provide guidance for their socialisation as ‘future adults’. However, as the following quote depicts, the 51 52 53 54

To¯yo¯bo¯ (1977) p. 1. To¯yo¯bo¯ (1969a) pp. 1–2. To¯yo¯bo¯ (1977) p. 4. To¯yo¯bo¯ (1977) p. 4.

Managing women workers 125 close working and living environment, and the care and indeed intervention of management, could install a strong work ethic in young female operatives. When I first started working I was very ambitious and this attracted the attention of my seniors (senpai), and they bullied me. It made me more ambitious and eventually they gave up and I won. This was a good memory and even now gives me courage to keep going. About six months after I joined the company I had decided to quit because I didn’t like the way my supervisors did things. I had already sent my luggage home. When I was finishing work one day I saw that my mother had come to the factory. Somebody had noticed my luggage was missing and that I was not as usual and had called my mother. This event made me feel that people cared about what I did and so I decided to keep working. (Fieldwork survey respondent no. 40; employed 1956–60, then 1963– age 15–19, then 22–)

Overall, it was felt that general efforts to improve the dormitory system, even in small-scale dormitories, would lead to autonomy for boarders and a more effective management system. Management and guidance were to be recognised as two important components of this approach. The direct management of boarders, a large proportion of whom were still made up of young girls under 20 years of age and from distant regions, meant that industry responsibility for their development and protection was still required. Management of the younger age groups would therefore constitute largely care and some advice, while for those over 18–19 years of age the emphasis was to be more on advice and only small amounts of guidance. While skilled personnel, particularly school teachers and dormitory matrons, were responsible for much of the guidance and care, the administering of advice could fall to other company personnel, including factory managers, self-government committees, nurses, older female workers and other company officials. An overall aim was that regardless of the numbers of dormitory staff and boarders, regardless of factory/dormitory scale, regardless of economic upswing or downturn, the essential nature and quality of the management system for young females was not to be compromised or affected.55 The overriding emphasis in current and future management strategy was therefore the continued promotion of the dormitory (and education) system of management. In those factories acquiring labour from outside of the prefecture and actively recruiting young female school leavers, dormitory and education management was to be a central, strengthened focus of

55 To¯yo¯bo¯ (1977) p. 7.

126 Managing women workers factory management. While it was recognised that changes meant both adaptation and variations to the system, the ‘mission’ and ‘ideals’ of the dormitory-focused system continued to be enforced by textile labour managers throughout the years under study.56 However, the rise in older and temporary female employees also meant that labour managers had to develop alternative strategies for these groups of workers.

Finding alternative strategies for managing older female workers As previously noted, it is difficult to analyse the position of older women within the labour management structures, mainly because industry documentation often did not include them as they were generally not employed as regular workers. Both the dormitory and education systems were established as central programmes for the management of young females rather than older females. Older women, at least in theory, were not felt to be conducive to the industry’s established mode of operations, which involved the relocation of workers into company dormitories. This was because older women were primarily married and, therefore, less flexible and less mobile, mainly desiring employment in close proximity to their permanent residence. In this respect, the textile industry initially took the view that it was difficult to incorporate older women into the current labour management structures and into shift-work operations. At the same time, because the dormitory system was already long established within the industry’s mode of operations, companies made little attempt to employ commuting older women as a regular workforce. The utilisation of older women in the industry during the period was therefore primarily in the form of temporary or part-time production labour, although there were some employed in more permanent nonproductive roles such as dormitory matrons and teachers in company-run schools. The industry itself commented that, in general, the textile industry, particularly the spinning sector, was late in its utilisation of older women and part-timers compared with other industries and that this was primarily due to the structure of the shift system of employment in factories. Overall, it appears that the shift system was seen as a barrier to the employment of older women, rather than, for example, a system that allowed for flexibility in work hours. Indeed, one response from my fieldwork survey firmly stated that the shift system gave her the opportunity to juggle her work hours and domestic (household and children) responsibilities.57 In many ways, the general view taken by textile employers towards non-regular workers was because older women as temporary workers had initially emerged as a

56 To¯yo¯bo¯ (1977) pp. 1–2. 57 Fieldwork survey respondent no. 40; employed 1956–60, then 1963–, age 15–19, then 22–.

Managing women workers 127 force to counteract economic buffers and cope with operational or labour fluctuations. By their very nature they were therefore seen as only temporary, and their integration within the management style established for regular workers was not seen as something desirable or necessary. As this study has shown, however, despite the industry’s long tradition as an employer of young females and a management and production style centred around their employment, there occurred a gradual movement in the structural composition of the female labour force to include large numbers of older women as temporary, part-time and seasonal female workers. While this did not significantly impact on the dormitory and education systems, which continued in operation for young females throughout the period, it did raise several issues. The diversification of the female labour force held important consequences for management. Because the employment of young workers and older workers was essentially very different in nature and composition, it required different management styles and approaches. In many ways the perception that older workers were not suitable for the textiles mode of labour management and operations had to be reconciled with the reality that their actual employment within the labour force was increasing significantly, and that they would have to become a recognised focus of labour management strategy. Labour management strategy in transition It was gradually acknowledged by the late 1960s that a new labour management structure was required to take into account both female dormitory boarders or workers and commuting permanent, temporary and part-time female workers (i.e. a widening range of female workers). In order to do this, managers felt that a system based on an individual’s value and ability was crucial. In many ways this was opposite to the traditional employment and management of young females, which was based on values such as age (i.e. employment at school leaving age) and involved a strong company role in nurturing, guiding and developing ability. In other respects, this movement towards older women could be encompassed within a general labour management strategy which was already being employed in the management of young females – the movement from a unified or singular management approach to one of plural management (by diverse age levels). In other words, this strategy could be extended to include movement from a traditionally singular system of management based on young females living in dormitories as regular employees, to a new system comprising many layers: young females versus middle-aged females, dormitory boarders versus commuters, and regular employees versus temporary and part-time employees. The positive utilisation of older women and housewives, while on the one hand requiring the development of a new labour management system, also required the creation of an environment within which older women could be incorporated into an essentially young workplace and utilised to effect.

128 Managing women workers There was a recognition that not only was the industry obliged to employ older female workers because of the decline in young workers available, but also housewives were being released from the home in increasing numbers with the aim of improving their standard of living and enriching their life and leisure time. However, the majority were not under formal labour contracts and, until the late 1960s, management policy towards them had not been sufficiently thought out.58 Forecasts for their increased use in the workforce signified that the management of such workers was likely to be an important issue for the future, and, like that of regular employees, required ideas and policy on employment, guidance, training and education. Temporary and part-time workers were noted (and classified) as being predominantly ‘housewives’ and it was therefore felt necessary to understand their different characteristics relative to young female workers. Some basic characteristics of older female workers were identified by To¯yo¯bo¯ in a labour management report, and are shown in Table 4.3. The identification of these characteristics highlights an important point. The ability to depend upon older women as regular workers appeared to be a significant concern of the industry during this period. This was primarily based upon the degree of importance placed on their socio-economic role within the domestic sphere. The 1972 MOL survey reported a higher incidence of married women compared to single women taking holidays in addition to the regular holidays. In general, these higher rates of worker absenteeism among married women compared to younger female workers were recorded as due to household and childcare duties. Chapter 3 noted that in many ways the industry felt that married women could not be relied upon as a sole source of labour, as the case of the 1960s test factories indicated. However, this image may well have developed into something of a stereotype in the postwar period and created a reinforcing cycle that permitted discrimination against female workers in the Japanese labour market. Indeed, the industry’s very Table 4.3 ‘Characteristics’ of older female workers Strong points

Weak points

●

Emotionally stable

●

●

Stable attitude to work

●

●

Quick assimilation into company

●

Source: To¯yo¯bo¯ (1969a) p. 27.

58 To¯yo¯bo¯ (1969a) p. 26.

Often absent due to housework responsibilities Absent for child-related activities (school events), and for social events (weddings, funerals etc.) Absent long term and in concentrated periods (during school holidays or the farming season)

Managing women workers 129 focus on dormitories and education as the core management strategies, both of which contained a range of lifestyle components that were far broader than the workplace skills required from a labour force, systematically reinforced the view that the real role of (even unmarried) women was not within the formal economy, except for temporary periods. In many ways the commitment of older women to work could have been capitalised upon to a greater degree and at an earlier stage. In its 1972 survey the MOL pointed out that while fewer than 10 per cent of young females responded that they would like to work ‘as long as possible’ (when asked until when they wanted to continue working), 70–80 per cent of married women responded so.59 Zensen certainly appeared keen to strengthen the position of older women as regular workers within the textile industry during the period, with the integration of ‘working mothers’ a firm union topic.60 In spite of these issues, and a certain degree of ‘reluctance’ on the part of textile employers to employ older women in more permanent ways, it is evident, however, that management strategies to deal with their integration into the textile labour force did evolve during the period. To give one example, To¯yo¯bo¯ proposed basic labour management policies and strategies to be implemented when integrating older women into the workforce, and these are presented in Table 4.4. While management strategies for young women often extended to preparing them for their future lives as wives and mothers, Table 4.4 highlights the problems faced by employers when their female workers had indeed reached this stage in their lives. While detailed information on the practical integration of older women within the textiles workplace is scarce, a 1970 MOL survey examined the labour management of older women in several manufacturing and service sector industries, including a weaving factory where older women constituted 42 per cent of the female workforce.61 This survey provides a good case study of how the integration and organisation of older females was undertaken at the workplace level. The study commented that the factory had already established a pattern of continuing employment of female workers after marriage, specifically those who could commute and lived locally, and this accounted for around 37 per cent of its female workforce being ‘older’. However, it noted that due to the shortage of young female labour the factory had lately begun employing older women with no previous experience through the local PESO branch.62 This employment of older women as part-time workers had begun in 1968. It was felt that if the numbers of part-time workers increased significantly, then the workplace

59 MOL (1972) p. 84. 60 Zensen (1997) p. 48. 61 MOL (1970) pp. 75–76. The factory, based in Hokuriku, operated 521 weaving machines and employed 248 women (76 per cent of total workforce), 103 of whom were older than 30 years of age. 62 MOL (1970) p. 77.

130 Managing women workers Table 4.4 Management strategies for the employment of older women Basic strategies ●

●

●

●

● ● ●

Establish a management system that will set and maintain various criteria for their employment, including their relationship to unions and establishment of a wage policy Give adequate consideration to their not feeling discriminated against in terms of work, welfare and pension status (i.e. they are not usually fully incorporated into these systems) Establish separate recruitment policies from regular employees, taking into account factors such as their previous employment experience, their family circumstances (because it has a major impact on their attendance abilities) and their health Involve their families in order to gain their understanding and co-operation (e.g. invite them to view factory) Provide introductory training to help them assimilate into the workplace If necessary, consider the establishment of extra facilities, e.g. a day nursery Consider the positive appointment of those with high ability or qualifications, into higher positions such as consultants for young female workers

Strategy content ●

●

●

●

●

●

●

In preparing work conditions and the workplace environment for older women, endeavour to provide jobs that feel ‘worthwhile’ Do the utmost to consider how to make housework and paid employment compatible Establish an employment system that will display and utilise the maximum aptitude of housewives as a labour force Create a work environment which will harmonise the qualities of housewives, who have rich life experiences, amongst young workers Raise their pride and selfconsciousness towards their work to ‘regular employee’ level (i.e. give them a feeling of being regular employees even though they may not officially be) Measure their advancement, make them realise that they are a key labour force, and eliminate their feeling that they are only ‘filling in their leisure-time’ Establish a management policy that will measure their future employment expansion

Source: To¯yo¯bo¯ (1969a) pp. 27–28.

environment would have to undergo significant changes, including the deployment of more varied and shorter worker shifts in order to try and fit work hours around a workers’ domestic needs.63 Although the factory worked on a two-shift system of production operating between the hours of 05.00 and 22.10, 68 per cent of older women were employed in day-only shifts, usually between the hours of 08.00 and 16.45. For the majority, it was reported that ‘domestic responsibilities’ made it impossible for them to be incorporated into the two-shift system. While production itself ran under the shift system, many of the ‘pre- and postweaving’ processes could be assigned predominantly to the day shift, particularly the checking of fabrics and the preparation of finished products

63 MOL (1970) p. 81. By 1970 part-time workers numbered 13, making them just over 5 per cent of the total female workforce.

Managing women workers 131 for transportation. The study also commented that management felt that the physical energy required for the weaving process directly was not suitable for those aged 45+ years. However, it also stated that those older women with many years of experience were very good in production support roles, as well as in non-production and supervisory positions.64 There were felt by management to be two main limitations to the employment of older women. The first was their household and domestic responsibilities, which meant they were frequently absent, particularly for events relating to local associations and children’s schooling. This meant that their placement in work teams was often problematic. The second limitation was the difficulty in having older women work directly with much younger women and the various problems of incompatibility, reflecting the ‘generation gap’, that arose.65 The factory under study commented that it had implemented a system whereby older and younger women did not always work directly together, with older women generally working in preparation and post-weaving divisions within the company. However, it also suggested that older women were often placed there as they were much more reliable in these jobs, as younger women were more liable to dislike or complain about these ‘more messy’ jobs.66 However, general efforts were also made to incorporate them into the workplace so as not to discriminate between them and younger regular workers. Attempts to give older workers, particularly part-timers, a sense that they belonged in the company and to integrate them (at least socially) with younger workers included encouraging them to take part in company trips and recreational events. A ‘contact committee’ was also established, and all workers were urged to meet once every two months for about an hour to discuss workplace issues and ways to improve human relations within the factory and boost morale.67 Overall, management felt that the increased use of older women, whether one ‘liked it or not’, was necessary due to the decline in young workers, and that older female workers should be thought of as people who could bring knowledge and skills. In this respect, it was necessary to plan for more positive ways to employ them and to ‘modernise’ workplace facilities to suit them.68 Overall, it does seem evident that, despite the overwhelming emphasis placed on the dormitory and education systems as the focus of labour management of the core (young) female workforce, a discrete labour management policy for older female workers did develop during these years. This was very much still in its primary stages and problematic in nature, not least with regard to the industry’s ‘ability’ to rely on older female 64 65 66 67 68

MOL (1970) p. 78. MOL (1970) pp. 78 and 82. MOL (1970) p. 79. MOL (1970) p. 83. MOL (1970) p. 83.

132 Managing women workers workers and integrate them into established workplace structures, as well as reconcile this with their domestic roles. However, in many ways, the features of their labour management highlighted in this section did develop to a greater extent after the period under study, as the trends in the female labour market identified in this research took even further hold.69

69 For example, in a fieldwork visit to a To¯yo¯bo¯ factory in Toyama on 3 July 1998, it was apparent that the labour force consisted predominantly of older women. Managers at the factory reported that they had only one young female still going through their education system (i.e. commuting to part-time senior high school) and that it was probable that this system would be wound down in the foreseeable future. This factory was, however, renowned for a longer than average length of service and a greater average proportion of older women than other To¯yo¯bo¯ spinning factories.

5

Educating and training women workers

Factory girls and schooling These [survey] questions have brought back many good memories of my youth. It has been 30 years since then and I feel that those days were very significant for me. I was 15 years old when I joined the factory and didn’t know anything. I got up at 4.30 a.m. and helped clean the room before work. It was hard, especially on cold winter mornings. The discipline and education were great advantages for a 15 year old, and I think it made me mature. I had no time to sit down while working and my legs became tired and it was hard, even though it was a simple job. I remember how happy I was when we won the competition against other dormitories for the Tanabata festival when I was a leader in our dormitory. After graduating from the company school I kept studying at a specialist school (semmon gakko¯). I learned everything from sewing and flower arrangement to the Japanese harp. (Fieldwork survey respondent no. 42; employed 1963–71, age 15–23) Education, dormitories and the workplace Cotton companies initially established dormitories in the prewar period to house female workers for many practical purposes focusing around production. However, as the functions of the dormitory system developed, they were not confined to securing labour for production alone, but also tied up with the concept of the employers’ fundamental ‘social responsibility’ for protecting and guiding young female workers, particularly those recruited from remoter regions.1 In this respect, dormitories had to provide for the growth and development of young workers both as a labour force and also as members of society (shakai-jin). It was felt to be the ‘moral obligation’ of industry to manage the extra-curricular life of young girls resident in factory dormitories in order to achieve this. The education of these young girls

1 Kobayashi (1963) p. 142.

134 Educating and training women workers therefore became a central focus of the extra-curricular activities provided for by textile companies. While education and dormitories were each distinctive systems in their own right, at the same time they were also intercepting and linked systems. In fact, they were often referred to as one unified system, or mixed together in the primary documentation and company reports and difficult to separate out. Their central point of interception was the notion of upbringing or industry’s responsibility to provide guidance, development and growth for the young female workers in its charge. Kobayashi (1963) noted the different role and aims of schools and dormitories in the postwar years, as shown in Table 5.1. Each system was meant to encompass different levels of individual development, group living, individual education and group education. It was deemed essential that dormitory managers and school teachers should co-operate across these levels. The young average age of female textile operatives led to a recognition by management that this was an important period in a young woman’s development and life, both physically and mentally. Any labour management system had to therefore offer various opportunities for education alongside and complementary to the living environment, offering not only safety, protection and care for young girls, but also the opportunity for ‘character building’. The education and dormitory systems were designed to interact and work together to achieve this development of the young female worker. This was in effect a sophisticated and comprehensive management system for what was during this period essentially a young and temporary workforce.2 While schooling was certainly viewed (like dormitory life) as being within the realms of private life, there was believed to be some interaction with the workplace. The idea was that there were three overlapping circles or spheres – comprising the dormitory, the school and the workplace – making up a complete labour management system for young female workers. Of primary importance was the industry’s aim of stabilisation of the workforce. At first glance, the education of workers is seemingly not directly related to the Table 5.1 Role of dormitory versus school School ● ● ●

Teacher–pupil relationship Public life Place for education and theories

Dormitory ● ● ●

Human relationships Private life Place for personal life guidance and practical application of theories

Source: Kobayashi (1963) p. 168.

2 ‘Temporary’ in the sense that on average they worked (or at least were expected to only work) several years for the company until, for example, graduation or marriage.

Educating and training women workers 135 workplace and, given the average short length of service of many female workers, one wonders at the benefit to the company of providing such facilities. Like the dormitory system, however, the education system was felt to have (albeit somewhat ‘indirect’ at times) an application to the workplace. The effects of making good use of the ‘leisure time’ of female workers by providing education and practical guidance were viewed to be considerable in that they matched the life cycle and hopes of the young women as well as filled the desires and requests of their families. Overall education was deemed to affect the quality or personal character of the female workers and thereby had effects on and benefited the industry’s quality of labour force, teamwork system and production process. There are three main reasons why an education system was developed for young female workers in the cotton industry during the postwar years. First, education was a tool for improving the image of the industry and overcoming the ‘inferiority complex’ associated with being a female textile worker (i.e. overcoming the Joko¯ Aishi image). Second, in response to a growing shortage of young labour due primarily to rising rates of education, the industry’s education system was a recruitment and management strategy to keep young women working in textiles for as long as possible, directly contributing to the key aim of ‘stabilising’ the workforce. Third, it developed within and alongside the dormitory system as a focused programme of protection, development and growth for young women. Young females (particularly new junior high school leavers) were known as Kin no Tamago (‘Golden Eggs’). In other words, they were a ‘precious thing’ (in part because of their youth and in part because of their declining supply) to be treasured, nurtured and held on to. The image of the industry, based on previously existing prewar labour conditions which had been heavily criticised, was perceived by many managers as still ‘lingering’ in the early postwar decades. One labour manager of the period cited an example of this image or complex, recording the situation of girls moving to and from factory compounds and not wanting to stand at bus stops with the name of the spinning company on them.3 The poor image of female textile operatives was still fresh in the minds of many Japanese in the years after the Second World War. However, even today, with the prewar development of the industry studied within the senior high school curriculum, the image of prewar and postwar labour conditions is often confused, and has meant that the image and phrase Joko¯ Aishi has remained a pervasive one in Japanese society. The urbanisation process as well as the growth of new manufacturing industries in the early postwar years also contributed to the poor image of the industry. While textile work was still seen as a ‘good job’ in the rural areas of Japan, newer industries such as the electronics industry had a better image and status in urban areas. In addition,

3 Kataoka (1993) p. 46.

136 Educating and training women workers textile girls, because of their predominantly rural background and education only to junior high school, were often looked down upon in urban areas as ‘poor’ girls working.4 The improvement of conditions and the strengthening of education to provide the opportunity for young girls to acquire further qualifications and education while working in the industry was a labour management strategy to directly tackle this lingering Joko¯ Aishi image. The issue of education appears strongly in the industry documentation after 1960, in many ways a response to the growing awareness of the young labour shortage. As noted in Chapter 3, increasing numbers of Japanese girls going on to senior high school education, combined with increasing numbers of job opportunities for females in other industries, created problems for the industry in its procurement of adequate workforce levels. A central attempt to secure young females recruits was the development of an educationbased strategy which would seek to match school attendance with length of service in the industry. While somewhat dependent on the job assigned, in general, it was felt that within two to three years female workers could reach a peak of experience and work ability (technical skill), given the nature and simplification of jobs in the industry. Therefore, particularly in the prewar years and at the beginning of the period under study when there was a labour surplus, employers had felt that if the average female operative worked for two to three years, then the company had gained sufficient profit and benefit from that operative (especially as the next young recruits were waiting in the wings to take their place). However, as it became increasingly difficult to secure young female workers, attempts to stabilise their employment and length of service were formulated, and education was utilised as a major tool within these. A central fundamental idea was the matching of a regular course of education (around three years) at cotton schools and later the matching of part-time senior high school attendance (four years) to a stable period of industry service. That is, in order to graduate with their educational qualification, the worker would be employed by the company for a stable three to four year period of time as they both studied and worked. Therefore, while the content of the schooling and education a young woman attained was not directly related to factory work, it was promoted with the aim of creating a stable and set period of employment within the industry.5 Development of the education system Education had existed in the prewar textile industry, but in the postwar years, strategies developed to extend it away from just ‘bridal training’ and the ‘use of free time’.6 Education was to increase the quality of workers by 4 This assessment is based on overall analysis of fieldwork interviews and surveys. 5 The term ‘education’ is used in this section to primarily discuss ‘schooling’ and is not to be confused with other types of education and workplace training, discussed in the next section.

Educating and training women workers 137 providing senior high school levels of education as well as the opportunity to attain further qualifications. This issue of a quality workforce was a key feature of management concern during these years. Industry documents commented on a trend of increasingly ‘lower quality’ junior high school graduates coming into the industry, and education was viewed as a means of improving the quality of the workforce available. From 1960 the industry began to think about extending its internal education system along the lines of the national school system, to include regular curriculum subjects. This was not without its problems because, as one industry report noted, in the textbook for the curriculum subject of Social Studies, for example, textile wages were ranked as the lowest of industry wages, and it was felt that if the girls read this it would add weight to the complex of being a female textile worker.7 However, this idea went ahead and the JSA’s Committee for Education was formed in the early 1960s. During the 1960s and 1970s education became a standard feature throughout the industry with the normal situation in medium- to large-scale enterprises being a factory with an attached dormitory and attached school. School facilities were expanded and the curriculum content developed along standard (national) senior high school lines. Although the industry was clearly implementing improvements to its educational system, it quickly came to be a matter of management concern that the standard and quality of education offered within the industry schools would still not be on par with a senior high school graduation qualification.8 This would present a problem in light of the rising pressure to compete in a labour pool where rapidly increasing numbers of young girls were moving into this ‘higher’ non-compulsory level of education. The industry’s response was to further develop factory schools to regular (parttime) senior high school standards. While initially school teachers were hired as regular company employees, and in some cases were dormitory staff at the same time, as the need for a better quality of education increased, these posts were divided. Qualified school teachers were hired or existing staff were provided with the opportunity to gain teaching qualifications. Many of the factory schools progressed during the period in terms of the quality and curriculum of education offered and were granted legal status as ‘non-regular senior high schools’ under the School Education Law.9 From

6 Based on ideas developed in the larger factories in the prewar period, education was developed in order to fill up the free time of girls living in the dormitories. This mainly revolved around traditional ‘bridal training’ systems of education, such as tea ceremony, sewing and flower arrangement, though there was some schooling too. 7 Kataoka (1993) p. 47. 8 Kataoka (1993) p. 47. 9 They could not be given ‘regular’ status because senior high school education in these schools, as well as public part-time high schools, took longer than the three years in regular high schools, although the curriculum was the same.

138 Educating and training women workers 1962, opportunities to commute to local (government) part-time senior high schools, and to undertake education by correspondence, were also developed across the industry. On a fieldwork visit to Nishiwaki sanchi, an area of small-scale textile establishments, it was reported that even small-scale family-run enterprises gave girls the opportunity to attend part-time senior high school or specialised schools while working during this period – though the take-up of this opportunity and actual attendance was lower than that at medium- to large-scale corporations.10 Furthermore, from 1967, due to the increase in the number of senior high school graduates working in the industry, textile companies started to develop partnerships with junior colleges (tanki daigaku) in order to give female workers the opportunity to gain higher education.11 At the same time, a policy was implemented of trying to retain in employment workers who graduated with a junior college education under this system, through their assignment to posts such as school teachers and dormitory leaders within the industry.12 Union interest in the education system no doubt also played a role in its improvement. Zensen had carried out a survey of dormitory boarders attending factory schools in 1962.13 When asked how they felt about attending school, an average 56 per cent of respondents replied that they did not particularly enjoy school but that they attended in order to improve themselves. Fewer than 5 per cent said they liked going to school and around 20 per cent said they disliked school but they couldn’t do anything about it (shikata nashi ni itteiru). When asked to comment on what they had heard about school before they began, and what the actual circumstances were like once they began, the main response was that there were substantial differences. A central response was that many had thought they could study the same subjects as a regular or part-time senior high school but that this was often not the case in the factory schools. Not all had been aware that everybody (i.e. not just those who wanted to) was ‘required’ to attend school. Other complaints were that the teachers were not external teachers (i.e. were company staff) and the schools were too close to the dormitory.14 An average 40 per cent of respondents thought that the school facilities were poor, 12 per cent thought they were good, 30 per cent said they thought they were average, and 12 per cent said they didn’t know (wakaranai). Moreover, 44 per cent said the lesson content was poor, while 38 per cent said it was average. Only 14 per cent of respondents said they thought that external teachers were poor, compared to 46 per cent for internal teachers (company staff). Common responses, when asked what the classroom environment

10 11 12 13

Interview K. JSA (1979a) pp. 788 and 800. To¯yo¯bo¯ (1970a) p. 13. The survey was of schools in six factories, and the number of survey respondents was 2,539: Zensen (1962a) p. 38. 14 Zensen (1962a) p. 41.

Educating and training women workers 139 was like, were that it was too cramped, too noisy, too cold or too hot. Less than 5 per cent responded that the classroom was spacious or had a ‘bright’ environment.15 When asked if they thought they would like to learn at a place outside the factory school, 67 per cent responded that they would, and the most popular choices were at a part-time high school, correspondence school and a technical or specialist school.16 Overall, Zensen commented that the survey had revealed that there were significant problems within the factory school system and that the system required direct management, improved policies, better teaching staff and more thought on the needs and desired schooling qualifications of the boarders or pupils.17 This no doubt placed pressure on the industry to reform and improve the education system in the way that it did during the 1960s. In general, the ideal of education as an integral part of management was being developed during this period. Company reports from the period can be used to illustrate the reasons for and the aim of development of the education system. A To¯yo¯bo¯ report outlined the management aims of the school system as a place of education and guidance for female workers during the 1960s and 1970s, focusing on three central points.18 First, from an industry perspective, the company was a place where ‘social responsibilities’ were carried out and the moral obligations of ‘upbringing’ and ‘developing’ young girls took place. This idea of the social development of young girls within industry care was a central component and aim behind the push for increased educational opportunities and raised graduation levels during these years. It was also a method whereby the industry could seek to secure and maintain a quality labour force. Second, from the workers’ perspective, the company was a place where they could have their ‘primary intentions fulfilled’, with the attainment of an educational qualification as well as the acquisition of ‘economic fortune’ – i.e. the ability to earn their own income, accumulate savings and at the same time further their education. Third, the concrete aims of education from a labour management perspective were to both develop the goals of the industry and increase the number of girls graduating through the system. The development of education during this period reflected not only the changing socio-economic environment of the time, but also that this was a period of new values and gradual awareness of the changing characteristics or nature of labour. Young females could no longer be taught simply to obey or to be prepared for marriage,19 but could expect to receive modern educational training and be expected, to a certain extent, to think for

15 16 17 18 19

Zensen (1962a) p. 44. Zensen (1962a) p. 45. Zensen (1962a) p. 46. To¯yo¯bo¯ (1977) pp. 5–6. For many years the education of females in Japan was tied with the ideal of developing the virtue of ryosai kenbo (good wife, wise mother) (see Uno 1992).

140 Educating and training women workers themselves. Education also offered the chance for young females to ‘become adults’ away from home.20 In many ways the development and diffusion of education and the postwar creation of a mass middle class in turn had an impact on in-company education. The industry’s education system moved away from the previous supplementary style of education developed in the prewar industry to a system designed to match current socio-economic trends. Therefore, this postwar period saw rapid developments in the educational opportunities offered to young female workers by the cotton industry, within a context of clear management strategies and aims. The education system in operation The shift system in operation in an average textile company’s workplace dictated the schedule of educational hours for those young women attending school while working. The schedule shown in Table 5.2, taken from the MOL 1972 survey of textile factories, is representative of the general pattern that was in operation. Schooling hours were dependent upon shift-work hours and therefore, like shift-work, followed a weekly rotation. For the most part, four years were needed to graduate from senior high school (based on four hours of schooling a day) and three years required to graduate from junior college (based on three hours of schooling a day). In both cases, this represented a year longer than under the normal (national) education system.21 The 1972 MOL survey stated that the spinning industry (with the highest continuing predominance of young female workers) had in recent years in ‘most’ companies set in place the provision that workers be able to attend school while working as a permanent feature of their job-offering policy.22 Of the companies surveyed operating a shift-work system, 71 per Table 5.2 Example of a typical shift-work/education hours system Shift hours (incl. 45-min meal break)

School hours

A: Senior high school (1) Early shift (2) Late shift

05.00–13.30 13.30–22.00

14.30–18.30 08.50–12.50

B: Junior college (1) Early shift (2) Late shift

05.00–13.30 13.30–22.00

14.50–18.00 09.00–12.10

Source: MOL (1972) p. 17.

20 To¯yo¯bo¯ (1968) p. 3. 21 MOL (1972) p. 16. 22 The report noted that in recent years the possibilities for shift-workers, particularly in spinning companies, to attend school (including part-time high school) had progressed rapidly: MOL (1972) pp. 14–15.

Educating and training women workers 141 cent provided the facility for shift-workers to commute to part-time high school, while 31 per cent provided the facility to commute to junior college. Of those companies providing the part-time high school facility, 31 per cent also provided a commuter bus. In general, educational opportunities and facilities were found to vary across companies, and, as might be expected, were more prominent in large-scale companies. Workers were also found to be more likely to take up the opportunity to attend school in large-scale companies. While no official industry figures are available on the school attendance of young female workers, the industry commented that, in general, attendance rates increased throughout the period. To¯yo¯bo¯ commented that by 1975 an average of 70 per cent of those who went into schooling completed the four years successfully.23 A slightly higher figure was given in a fieldwork interview with a labour manager at Unichika, who said that 80–90 per cent of female junior high school girls employed by the company stayed for four years and graduated senior high school, particularly in the earlier half of the period.24 The 1972 MOL survey, which included an examination of the situation of education for female workers across the industry, recorded that from a survey of 6,000 women, 32 per cent were attending part-time high school and 4.9 per cent were attending junior college (no mention of attendance at ‘in-house’ factory schools). This is felt to be significant considering that the proportion of senior high school leavers, secondjobbers and older women in the industry had increased by this time, and that this survey was comprised of women of all ages. In large-scale companies, in particular, school attendance was made a permanent and almost ‘compulsory’ feature for young girls, especially those who joined the workforce after graduating from junior high school. For example, within To¯yo¯bo¯ the policy was that all female workers under the age of 20 years were to attend education (Plate 5.1).25 While, under legislation covering dormitories, a company was not in theory allowed to force workers to attend education, management policy was such that they were very much encouraged to attend. In any case, many young girls cited education opportunities as a specific reason for entering textile employment during these years.26 Several quotes from my fieldwork survey reveal that studying while working was not always an easy task, but that it offered various opportunities and benefits.

23 The school dropout rate (taigaku-ritsu) was recorded as 30 per cent in To¯yo¯bo¯ (1975a) p. 15. 24 Interview M. 25 To¯yo¯bo¯ (1975a) p. 11. Confirmed by To¯yo¯bo¯ manager (Interview I). 26 Shindo (1961) p. 94 noted that the educational facilities offered by the industry were often cited by female workers as one of the major factors attracting them to employment with cotton companies. The fieldwork survey confirmed this, as noted in Chapter 4.

142 Educating and training women workers

Plate 5.1 Factory girls (wearing school uniform) in front of high school building at To¯yo¯bo¯’s Komatsushima Factory. Photo courtesy of To¯yo¯bo¯.

I wanted to graduate from senior high school while working. I couldn’t rely on my parents for this because I had lots of siblings and was raised in a rural area. I was ambitious and decided to get a senior high school qualification by myself. I was able to go to a prefectural senior high school near the factory, and the factory was located by the sea. So it was like a fantasy [akogare] to go there. (Fieldwork survey respondent no. 41; employed 1968–78, then 1990–93, age 16–26, then 38–41) It was very hard to study and work at the same time. Although now I am able to say that I would like to try it again, at the time I hated it. But I carried on, because it was interesting and I encouraged myself to get through each stage that I faced. (Fieldwork survey respondent no. 9; employed 1970– age 15–) In the late 1960s 70 per cent of people went on to senior high school, and so I was envious of those who could. However, I’ve worked for 33 years and I think I’ve had more experiences than other people. I experienced the hardship of working and studying at the same time with lots of friends and so we were able to enjoy it together. I appreciate now that I

Educating and training women workers 143 studied there rather than at a regular high school. I’m thankful that I have been healthy enough to work and to bring up children, and I’m grateful to my parents. Also, the relationships that I’ve formed with lots of people have kept my mind young, and I enjoy my life. (Fieldwork survey respondent no. 16; employed 1965– age 15–) The two Ministry of Labour surveys on female textile workers, done in 1952 and 1972, are useful to contrast the level of educational facilities and present their development over these two decades. In 1952, most cotton spinning factories were found to have some sort of school, but it was not comparable to a regular senior high school system. While administration fees were in general borne by the company, the workers paid for teaching materials and textbooks. Usual subjects were flower arranging and dressmaking (both Japanese and Western sewing).27 Lessons were taken outside of work hours and were on average two hours per day. Those companies operating dormitories usually had further ‘learning’ facilities in addition to schools, including recreation and sports facilities. A JSA report in 1954 also confirms that education hours for female workers were shorter at the beginning of the period than those noted later. In 1954 the average course of study in schools attached to factories extended for two to three years with school hours of about two hours per day. In the 1972 MOL survey, however, the large increase in number of girls commuting to part-time senior high schools outside of factory grounds was noted. The typical fees for commuting to part-time high school included annual application, entrance and equipment fees as well as monthly lesson fees comprising around 25,300 yen per student per year.28 These were generally all borne by the company (see Chapter 6). The organisation of the educational curriculum in the cotton industry schools, many of which were reported to match regular senior high schools in the level of facilities offered by the end of the period under study, generally followed a four-year structure. Subjects taught included the basic course (futsu¯-ka) offering Japanese, social studies, mathematics and home economics, as well as the home economy course (kasei-ka) offering ‘traditional’ lessons including sewing, music, callisthenics, tea ceremony and flower arrangement. In To¯yo¯bo¯, the education programme was divided into these two courses from 1964.29 General subjects included in these courses under the To¯yo¯bo¯ system are outlined in Table 5.3. The basic course generally followed the curriculum in regular senior high schools under the Department of Education (Monbusho) system, and was offered either at factory schools or under a system to commute to part-time high schools in the nearby community. The home economics course was usually ‘in-house’

27 MOL (1952) p. 6. 28 MOL (1972) pp. 18–19. 29 To¯yo¯bo¯ (1975a) p. 26.

144 Educating and training women workers Table 5.3 To¯yo¯bo¯ education curriculum subjects Basic course (futsu¯-ka)

Home economy course (kasei-ka)

Japanese Social studies Mathematics Home economics Science Music English Literature History Political economy Geography Callisthenics Recreational club activities

Japanese Social studies Etiquette Cooking Sewing House economy/budgeting Music (optional) Tea ceremony (optional) Flower arrangement (optional) Calligraphy (optional) Book-keeping (optional) Callisthenics Recreational club activities

Sources: To¯yo¯bo¯ (1968) pp. 5–8; To¯yo¯bo¯ (1969a) pp. 22–24.

education, held either in factory schools or in lesson rooms within dormitories, and was felt to contribute toward the ‘cultural refinement’ of female workers, based on traditional education set up by the industry in the prewar period. The results of the fieldwork survey indicate that home economics, sewing and flower arrangement were the most popular subjects studied. This no doubt indicates not only that all young women were encouraged to take lessons in these subjects, but also that the majority of girls would have opted to study a home economics course either at the factory schools or part-time senior high schools, a result not surprising for this period. Science, rather than what might be seen as ‘home science’, as a subject was clearly not as popular. However, a significant proportion of the girls also studied mathematics, history and foreign languages. This would indicate that a move away from less traditional (i.e. bridal training alone) education toward regular education (high school education as it is today) was taking place during these years. The regular course also increased in popularity during the later years. To¯yo¯bo¯ recorded that in 1971 the majority of pupils were enrolled in the household course (65.8 per cent) while only 34.2 per cent took the regular course. However, by 1975 the household course had declined to 48.3 per cent of pupils, while the regular course had increased to 51.2 per cent.30 The regular course was also forecast to increase to over 60 per cent of pupils within the next few years and plans were being made to further improve and increase the quality of the course.31 During the period, the teaching of more traditional skills such as sewing and cooking in the household course also underwent change, and was extended to cover

30 To¯yo¯bo¯ (1975a) p. 8. 31 To¯yo¯bo¯ (1975a) p. 17.

Educating and training women workers 145 courses on ‘household management’. These included additional aspects of study such as nutrition and budgeting, and in this respect can be said to have developed more in line with the subject of home economics in nature. Girls enrolled on the basic course or going to part-time high schools were often encouraged to attend these household management lessons or lectures as well. Further education after senior high school level was also established by the industry during the period. This not only was to attract more female senior high school leavers into textile employment, but also acted as an incentive for those workers finishing their senior high schooling to remain for a further few years. The opportunity for girls to acquire additional qualifications and stay longer in employment was a key management strategy developed, particularly in large-scale companies. In addition to senior high school graduation, qualifications that girls could study towards included dress-making qualifications, various teaching licenses, and national nursing qualifications.32 To¯yo¯bo¯ commented that it introduced a widening range of further education options, particularly from the late 1960s, and these are shown in Table 5.4. In terms of education facilities, the majority of spinning companies had inhouse schools, often referred to in the industry documentation as ko¯jo¯ gakuin (factory schools). While medium- to large-scale enterprises typically had their own schools, those in the sanchi were usually run by unions or local co-operatives rather than individual companies given the small scale of enterprises.33 The schools on factory grounds were usually adjacent to or close by the female dormitory, often connected by a corridor or covered walkway for ease of movement back and forth. Special educational sessions were often held in the company dormitories, in addition to the more formal school courses. These included special lectures on topics such as ‘budgeting and savings guidance’, ‘safety and emergency drills’, ‘health, sanitation and Table 5.4 To¯yo¯bo¯ tertiary education Junior college (tandai)

Specialist schools (semmon gakko¯)

Nursery/day care qualification Teaching qualification Nursing qualification Correspondence JC courses

Cooking school Dressmaking school Tea ceremony school Beauty school

Source: To¯yo¯bo¯ (1970a) p. 11.

32 To¯yo¯bo¯ (1969a) p. 23. 33 Records for Nishiwaki sanchi indicate the establishment of a school for workers within the companies affiliated to the local union as far back as 1909: Nishiwakishi Kyo¯iku Iinkai (1989) p. 53. Later the use of part-time schools in the local prefecture was more common (p. 22).

146 Educating and training women workers nutrition’, ‘discipline’, ‘essential rules for group living’, ‘common-sense skills’ and ‘daily life practical skills’.34 These sessions were usually carried out by dormitory and consultancy staff. In To¯yo¯bo¯, for example, the new post of ‘lifestyle consultant’ held guidance sessions on various practical topics. This aimed at providing the boarders with practical guidance for everyday life and the development of knowledge and skills, which they would need both during their time employed by the company and also for their life beyond.35 Company (and union) magazines and newsletters, as well as dormitory guidebooks, also played a part in educating, providing communication channels and ‘managing’ the morale of workers. In addition to regular school buildings and lesson rooms, further nearby buildings (dependent on the scale of the factory) often housed library, lecture, music, cooking, sewing or general home economic rooms. Educational equipment, in addition to basic equipment provided in the schools, ranged from collections of books housed in reading or library rooms, to sewing machines, flower arranging and tea ceremony equipment, radios, pianos and gramophones or record players. Club sports facilities were also common. Overall, it is evident that an increased range of education was provided by the industry for its young female workers. A clear trend is an increase in both content and scope in the curriculum and educational facilities in these years. It was felt that the varied and changing desires and quality of students required a more diversified educational policy with an agenda of variety. In this respect, it would be fair to suggest that some form of educational advancement or attainment was a clear benefit for the majority of young females working within the textile industry during this period. Evaluation of the education system In many ways the transformation of the education system and the changes made to the content and range of educational courses offered to students (workers) were a response to the changing socio-economic circumstances of the time. At first the textile industry depended mainly on its traditional household subjects, which had been established during the prewar period, but as more young Japanese girls progressed on to senior high school education, the industry attempted to compete with this trend in terms of the educational opportunities it offered. The JSA commented in 1969 that the school curriculum had completely changed within the industry during the previous ten years.36 While the success of the education (and dormitory) system is difficult to evaluate in concrete terms, several comments can be

34 To¯yo¯bo¯ (1969a) p. 5 and To¯yo¯bo¯ (1977) p. 6. Information and practical guidance was also provided to female boarders in dormitory handbooks, including To¯yo¯bo¯ (1961b). 35 To¯yo¯bo¯ (1969a) p. 11. 36 JSA (1969) p. 27.

Educating and training women workers 147 made. In general, industry managers felt that continuing and strengthening education (alongside dormitories) for young female workers was a positive management strategy during these years. The benefits, particularly in largescale companies, for employing this core female labour force within a clearly focused management programme for a defined period was felt to bring a stability to the industry’s labour force that had not existed previously. One former labour manager of the period also felt that the complex (low status) of the female factory girl in the industry was successfully overcome, as workers could graduate with the same education levels as those who had continued in the education system after junior high school.37 In fact, textile girls who had attained their senior high school education were sought after as recruits into other production jobs alongside regular senior high school recruits. In the electronics industry, for example, which had a policy of recruiting only senior high school leavers, they were highly sought after for their work experience. One labour manager told how electronics industry representatives would come to textile company-run schools to recruit for the industry.38 From a worker’s perspective, the education (and dormitory) system also offered substantial economic benefits (see Chapter 6) and opportunities for self-development. The position of the education and dormitory systems within the average company structure is also worthy of further comment. The labour divisions within companies were felt to employ the most elite staff during these years. One To¯yo¯bo¯ report commented that when selecting and training the core (male) labour management staff, the company always tried to place and rotate new university graduates within the division of the labour section dealing with dormitories and schools.39 This placing of the most educated male staff (recruits) in charge of the dormitory and school systems was felt to be important and reflected the core position of the two systems within the fundamental structure of labour management. Another To¯yo¯bo¯ report commented that, by the end of this period, various components of labour management that had been extended and developed successfully in the dormitory and education systems were gradually to be established and strengthened as fundamental labour management strategies within the factory and workplace itself.40 In other words, progress in the development of successful dormitory and education management, particularly that aimed at developing an individual’s personal skills, was felt to be usefully extended to the factory itself. This no doubt influenced the growth of training programmes for female workers in the industry during these years. With the decline in young workers over the period and the increased use

37 38 39 40

Fieldwork interview with JSA official (Interview L). Interview M. Kobayashi (1963) p. 153 (confirmed in Interview I). To¯yo¯bo¯ (1977) p. 15.

148 Educating and training women workers of older women, the proportion of the female labour force requiring dormitory housing and education saw a decline, particularly from the late 1960s.41 To¯yo¯bo¯ recorded that while 83 per cent of their female workforce were boarders in 1960, this had fallen to 73 per cent by 1975. The proportion of all boarders attending school fell from 70 per cent in 1969 to 59 per cent in 1975, making the proportion of ‘pupils’ among all female workers 58 per cent in 1969, but down to 43 per cent in 1975. In the main, education was aimed only at new ‘school graduate’ labour (i.e. young females) rather than ongoing (chu¯to) employees and older women. However, some companies made efforts to extend educational opportunities to older women. In the 1972 MOL survey, 10–20 per cent of the companies surveyed had some sort of educational schemes for ongoing employees.42 Unlike young school leavers, it was felt that the majority of older women did not require further education. This could be considered somewhat of a fallacy given that many older women in the earlier postwar decades may well have missed out on chunks of their formal education due to the disruption of the Second World War.43 There was therefore no firm educational policy for older women and part-time workers implemented by the industry. However, in some cases, educational programmes were extended to them, with Zensen promoting the opening of lifestyle education (seikatsu gakko¯) for women in small to medium enterprises in the sanchi.44 These served to provide them with extracurricular domestic skills such as cooking and sewing rather than more technical or broader skills. The provision of education for primarily young female workers not only highlights the continued emphasis on this core labour force but also illustrates once again that younger and older women in the industry were thought of as distinct groups – working students versus working mothers. Overall, the education system for young female workers that was implemented by textile companies during these years can be considered diverse and substantial. It expanded in scope and content during this time and, like the dormitory system, represented a considerable investment in young female workers. It no doubt had some impact on social values as the industry encouraged and endorsed the education of young rural girls. Moreover, by giving young women access to tertiary qualifications (teaching, nursing, beauty school) the industry was essentially preparing its employees for their next job, which can be seen as going beyond their mere social development as shakaijin. Dormitories and schooling often dominated the industry documentation on labour management issues, and in this respect provide us with an interesting and rich depiction of the daily working lives of young

41 42 43 44

To¯yo¯bo¯ (1975a) p. 3. MOL (1972) p. 10. See Broadbent (2003). Zensen (1996) p. 53.

Educating and training women workers 149 girls in this industry during these years, as well as highlight the considerable interest taken in non-working elements by textile employers. However, it is of course important to remember (though one could be forgiven for not, given labour management focus on dormitories and education) that women were employed by the industry to work on the factory floor. This also required investment in workplace training by textile employers.

Women and workplace training Employers in critical industries should be encouraged to study and correct causes of turnover. Methods of training skilled workers should be instituted in factories and workshops. (SCAP 1947b, p. 197) The nature of the job Women were employed to do a range of production jobs in the industry, the nature of which can be highlighted by an overview of the two main textile processes of spinning and weaving. There were three main process sections within spinning – preparation for spinning, spinning and finishing. Preparation involved getting the raw natural cotton (or wool) fibres, which were agricultural products, ready for the actual spinning process. There were three processes within this preparation: first, ‘mixing’ to mix together the various kinds of raw cotton and ‘scutching’ to unravel the compacted raw cotton and remove impurities, to produce loose cotton fibres in sheetshaped ‘laps’; second, ‘carding’ and ‘combing’ to comb the laps to remove shorter fibres and impurities and produce lengths of cotton slivers; and third, ‘drawing’ to stretch and combine slivers and reduce thickness irregularities, followed by ‘roving’ to twist the slivers into cone shapes. Spinning itself was also a three-part process: first, pulling the fibres and elongating and narrowing them lengthways, known as ‘drafting’; second, twisting the thread to strengthen it, known as ‘twisting’; and third, winding the thread, so it would not lose its twist, on to bobbins using spindles, known as ‘winding’. Finally, the finishing process involved ‘reeling’, ‘bundling’ and ‘baling’ the spun yarn into forms ready for weaving and/or transportation.45 The process of weaving also comprised three main segments – preparation for weaving, weaving and finishing. Preparation involved two main parts: preparing the ‘warp’ (threads stretched lengthways in the loom) by winding and sizing them onto the loom, and then preparing the ‘weft’ (threads to be woven across warp) by winding and spooling. Weaving itself consisted of three basic processes: first, ‘shedding’ which vertically separated and opened the warp yarns; second, ‘picking’ which inserted a weft yarn in the

45 JSA (1955) pp. 87–89.

150 Educating and training women workers open shed; and third, ‘beating’, the pushing of the inserted yarn against the warp yarns. Finishing involved inspecting the produced fabric, and then folding, stamping and baling ready for transportation. These basic processes in both spinning and weaving therefore defined the nature of the job for female factory operatives. An indication of the labour intensity of the various process sectors and the distribution of females employed in spinning and weaving jobs is illustrated in Table 5.5, which shows the distribution of female employment in spinning and integrated weaving departments in the Big Ten companies for 1960 and 1965. Because the companies were predominantly spinning with some degree of weaving, more women were employed in spinning than weaving. The more important figures are the ones that show the labour intensity within spinning and weaving respectively (columns 2 and 4). Within spinning, around two-thirds of women were employed in the actual spinning process itself (spinning and winding), while around 25 per cent were involved in the spinning preparation processes, and just under 10 per cent in spinning Table 5.5 Proportion of female employment in Big Ten spinning and integrated weaving departments

Spinning processes Mixing and carding Drawing and roving Spinning Winding Checking, finishing and baling Other Spinning subtotal Weaving processes Weaving preparation Weaving Checking and finishing Weaving subtotal Other processes Total

1960

1965

As a % of As a % of total spinning spinning or and weaving weaving respectively

As a % of As a % of total spinning spinning or and weaving weaving respectively

6.9 11.6 24.3 19.2

9.7 16.4 34.5 27.3

5.7 11.0 25.8 20.2

8.2 16.0 37.4 29.2

7.9 0.6 70.5

11.2 0.9 100.0

5.8 0.6 69.1

8.4 0.8 100.0

6.1 11.0 5.3 22.4

27.1 49.3 23.6 100.0

6.3 11.8 4.4 22.5

28.0 52.5 19.5 100.0

7.0

8.4

100.0

100.0

Source: Men Bo¯seki Ro¯do¯ To¯kei Shiryo¯ (1960, 1965). Notes Figures are for workers in direct production departments only. Figures for 1960 are end of June; figures for 1965 are end of December.

Educating and training women workers 151 finishing processes. Within weaving, just over half were involved in the actual weaving on looms, while just under 30 per cent were involved in the weaving preparation, and just over 20 per cent in the finishing and checking of woven fabric. It was common for males employed in production departments to be heavily employed in packaging and transportation jobs as well as in the overseeing of machinery. The MOL 1972 survey noted that in terms of posture required, a majority of 71.4 per cent of jobs across the textile industry required women to stand up, while only 13.7 per cent were sit-down jobs and 12.6 per cent were a combination of both standing and sitting. In spinning 95 per cent of jobs were performed standing up, while in weaving the figure was 88 per cent.46 The majority of women in the fieldwork survey reported that they stood to perform their work, including both those who stood to operate machinery and those who had to move around to check on machinery.47 The equipment carried or used by women, based on responses in the fieldwork survey, gives an indication of the more detailed movements involved in their jobs. Those involved in the preparation process or in cleaning commonly used brooms, brushes and air hoses. The use of small scissors, hand knotters, threaders and winders by those in spinning jobs shows that they were predominantly involved in threading the cotton thread onto equipment, keeping the thread in operation in the machines, and repairing any breakage. Prior to automation, the changing of bobbins was also a central task. Those in finishing and checking, particularly in weaving, used pickers, pencils and charcoal to mark and indicate any product faults. Just under half of all respondents from the fieldwork survey thought their job was average or OK (futsu¯) in terms of difficulty, while 36 per cent thought their job contained a degree of difficulty, and 17 per cent thought it was easy. Just over half said they worked in teams of various sizes, while the remainder said their job usually required them to work alone. Many reported that working with other girls their own age when they were young, and the strong friendships they made during that time, were happy memories for them, though a few said that relationships between juniors (ko¯hai) and seniors (senpai), i.e. different peer groups, were sometimes difficult. A central change highlighted by respondents was a change in the machinery in terms of technology – the introduction of faster machines, automation of many of the processes, improvements in production quality, and a reduction in the number of workers within the workplace. While most respondents thought this made the job and workplace better, a few expressed the view that mechanisation made it more difficult in terms of the skill required. In general, therefore, it was these types of simple or light spinning and weaving 46 MOL (1972) p. 58. 47 Only 2 out of 45 women said they mainly sat to perform their job: one woman worked in accounts and was not a production worker, the other was investigating and checking products.

152 Educating and training women workers production jobs within the shop-floor environment that female workers had to be introduced to and trained up for by the industry. Women and training The ideal of training for workers as a formal management process developed in the postwar period. From the early 1950s, cotton companies were among the first industries in Japan to establish structured and standardised training programmes such as Training Within Industry (TWI), introduced from the United States. TWI consisted of three main strands: job instruction (JI) (shigoto no oshie-kata), job method (JM) (kaizen no shikata) and job relations (JR) (hito no atsukaikata).48 In the textile industry, each strand formally consisted of ten hours of lecture-based instruction, and together formed a complete training system designed to implement the most effective and efficient methods of teaching the job to workers, improving the job, and motivating the workforce. In terms of training for female workers, the industry adopted various in-factory training programmes based on the different types of jobs, and aimed at the range of female employees from (young) new recruits up to women who had been there for three to five years (Plates 5.2 and 5.3). These on-the-job training (OJT) programmes became standard activities in the industry within a few years. There were two main types of training programmes used for female workers. These were new entrant training (shin-nyu¯shain kyo¯iku) and the more general TWI-based workplace training (shokuba-nai kunren).49 TWI was a component of the new entrant training, and was also used as the main component of promoting and training women to various ‘female leader’ positions within the workplace. It was also used in the training of those who became trainers or educators within the workplace themselves. It should be noted that in many cases older women already had production experience (previously when unmarried) and therefore did not require a lot of training upon employment. Those who were employed with no experience were given training using a TWI format (rather than new entrant training given to young female joiners). New entrant training became fairly detailed over the period, and while it was training for specific production jobs, it also involved broader aspects of induction and preliminary education of young workers into a work environment. Young workers were seen as becoming new ‘members of society’ (shakai-jin) on their first day of entering company employment.50 New entrant training therefore included learning about the company environment and mental preparation for work, as well as an induction to production processes. The first two were felt to be very important in making the 48 To¯yo¯bo¯ (1975b) p. 41. 49 Additional training for male workers included technical and management training programmes. To¯yo¯bo¯ (1975b) p. 6. 50 JSA (1969) p. 29.

Plates 5.2 and 5.3 New female employees receiving training on machinery use and safety, 1971. Photos courtesy of To¯yo¯bo¯.

154 Educating and training women workers transition from young school leavers to workers. The training programme usually lasted for four weeks, though in some companies, particularly largescale firms, this was followed by another four weeks and a follow-up review three months later. In small-scale companies, however, training programmes were often shorter than this average. The MOL 1972 survey recorded that basic job training was completed in about a two-week period, and the majority of on-the-job training was conducted within a time period of less than six months in all companies.51 Because the first four weeks of the standard training programme involved both adaptation to the workplace and living environment as well as the learning of processes, the integration into the job itself was gradual. In the first week, young females usually worked half days to learn job processes and received lectures for the other half of the day. In the second week they might be integrated into a day shift. Integration into the standard two-shift system was scheduled to take place either from the third or fourth week.52 The JSA set standards for new entrant training in the postwar period, based on what it believed to be the most efficient and easy to learn methods, and based on consultation with experts. These ideal standards included the assigning of one trainer for every three to five new entrants, the level believed to alleviate the anxiety of new entrants. The first four weeks were designed to be introductory training (do¯nyu¯ kunren), while the second four weeks were to include more skilled (proficient) training (shu¯juku kunren).53 The JSA also set general standards for attainment levels. For example, in spinning an ‘ideal’ standard was the situation whereby one worker was competent enough to look after 15 spinning machines. It was recommended that, on average, a period of six to nine months was necessary to achieve this.54 A typical example of the initial four-week training programme for new entrants in spinning issued as a guideline by the JSA is outlined in Table 5.6. In the large-scale textile companies, broader workplace training systems developed together with and alongside the school based education to offer a ‘complete’ education to employees. As Kobayashi (1963) noted, a system of practical education was set up for the purpose of educating and developing employees as members of the workplace and of society. In addition to the on-the-job training offered by TWI-based courses, broader training included lectures on health and safety in the workplace, home economy and budgeting and self-development.55 Both these ‘training within industry’ and 51 52 53 54 55

MOL (1972) pp. 12–13. JSA (1969) pp. 30–31. JSA (1969) p. 31. JSA (1969) p. 32. Kobayashi (1963) pp. 166–67 noted that the course for self-development in To¯yo¯bo¯ comprised the following over the years: Course One: ‘Introduction to New Life Themes’; Course Two: ‘Towards a Healthy and Sound Life’; Course Three: ‘Self-monitoring and Meditation: Re-Thinking of Self’; Course Four: ‘Household Life’ (topics included marriage, human relations and social responsibilities).

Educating and training women workers 155 Table 5.6 Example of a typical training schedule for new joiners Training schedule

Training content

Week 1 (Day training)

● ● ● ● ● ● ● ●

Welcome ceremony for new employees (nyu¯ sha-shiki) Introduction of staff Allocation of new entrants into teams Factory tour Rules and guidance lecture Introduction to processes (including introduction to materials and machinery, tying and cutting of thread) Entrance examination for school Recreation event

Week 2 (Day shift)

●

Training on machinery (introductory spinning processes; mainly mixing and spinning preparation; simple spinning operations)

Week 3 (Shift-work training)

●

Training on machinery (main spinning processes; changing of spindles, checking quality of thread; dealing with machinery trouble etc.) Inspection of all job processes

●

Week 4 (Shift-work training)

●

●

Training on machinery (further spinning processes; how to turn off machinery, use of knee-brakes etc.) Inspection of movement around machinery

Source: taken from JSA (1969) pp. 32–33.

‘practical skills’ courses aimed at improving health and sanitation conditions, promoting worker skills, developing human relations among workers, and developing individual skills and self-growth. These were all seen as factors contributing to a qualitative improvement of the labour force as both workers and individuals. Shindo (1961) commented that this extensive system was a postwar development. While training systems were maintained by some spinning companies in the prewar period, these were for the education of key (male) technical staff and for a minor proportion of all employees. The extension of more technical education within TWI to cover all employees, including the core force of female employees, progressed to a fairly high level and was felt to contribute to the steady increase in productivity within the industry during the early postwar period.56 While formal workplace training programmes were implemented by employers to develop a more skilled and efficient workforce on the shop-floor, training programmes were also implemented in order to develop and maintain a stable

56 Shindo (1961) pp. 108–09.

156 Educating and training women workers and quality workforce – an idea similar to the indirect benefits offered by dormitories and education to employers as well. Women and promotion In additional to training programmes, promotion opportunities for female workers were also developed in the postwar period, particularly in largescale cotton companies. The MOL 1972 survey recorded that 56.5 per cent of all the companies surveyed provided the opportunity for female workers to be promoted, while 38.6 per cent of companies did not. Among companies, the promotion opportunities for women were higher in medium and large companies, and correspondingly highest in the spinning and knitted goods sectors of the industry. The proportion of females at some sort of officer (yakuin) level was 6 per cent of all female employees surveyed by the MOL in 1972 (the corresponding figure for males was 22 per cent).57 Promotion levels for females were lower than those for males, reflecting the fact that the industry employed females in greater numbers as core production staff, as well as the system of assigning male jobs and female jobs, with more males in supervisory positions than females by job type. In addition, the general feature of promotion being based on length of service was not conducive to women who often moved in and out of the labour market and had a shorter length of service on average compared to males. While females who showed ability were able to be promoted off the shopfloor and into administrative positions, their proportion of all women in the industry was very low. Those females who were in non-production jobs were also rarely found in management positions or in supervisory positions supporting production areas. For example, the 1972 MOL survey commented that the proportion of health and safety administrative workers in textiles who were female was only 9.6 per cent on average, and lower in spinning at 7.8 per cent.58 My personal fieldwork survey indicated that the person directly supervising respondents (production workers) was more likely to be male than female.59 However, females were often found in supervisory positions outside of the factory, particularly within female dormitories – e.g. as dormitory matrons, school teachers, nurses and counsellors. JSA data available for 1960 and 1965 indicate that of all female non-production workers, 70 per cent were employed within dormitories or factory schools, with the next largest proportion employed in company medical clinics.60

57 58 59 60

MOL (1972) p. 14. MOL (1972) p. 39. Only 20 per cent of respondents said that they had been directly supervised by females. The proportion in medical clinics was 16 per cent in 1960 and 6 per cent in 1965: Men Bo¯seki Ro¯do¯ To¯kei Shiryo¯ (1960, 1965).

Educating and training women workers 157 In terms of promotion schemes for women within production departments, these grew considerably during these years. The various promotion positions or posts available to female workers included team leader (kumicho¯), patroller (mi-mawari), trainer (yo¯sei-kakari) or simply female workplace leader (joshi shokuba riidaa).61 The opportunity to train and be promoted to these various positions of female leader was on average after a worker had been employed for three to five years. Training consisted of three main parts following TWI practices: ● ● ●

how to teach job processes (10 hours); how to handle people (10 hours); training to educate new recruits (9–12 hours).

The training programme for female leader positions was established by the JSA after four years of surveys and research (1962–66). The JSA commented that it wanted to encourage leadership skills in female workers so that they could come close to supervisory positions.62 The main duties of female leaders involved the monitoring of production efficiency and quality, teamwork morale and human relations, and communication between ordinary production workers and factory management and supervisors.63 Females could also train to become new entrant trainers themselves (shinnyu¯shain yo¯sei-kakari). The opportunity to train for this post was generally offered after three to four years of employment. Applicants were chosen for their knowledge of the job, specific skills, manner, health and personality. Training was generally 9 hours (3 × 3 hours) and the lecture-based content included ‘the important responsibility of trainers’, ‘how to plan and lead training’ and ‘the right way to provide guidance and results of correct guidance’.64 Hazama (1997) suggests that the use of a range of statuses in the jobgrading or promotion system, including those for manufacturing workers such as team leaders (hancho¯) and group leaders (kumicho¯), resulted in numerous job ranks ‘much beyond what was considered necessary for functional rationality’ and not only created ‘job status for the purpose of managing the organisation, but also contained an honorary titular meaning which referred to the vertical relationship in social status’.65 While this may have been true to a degree, it can also be said that, overall, the extension of promotion opportunities to female production workers during these years

61 To¯yo¯bo¯ (1970a) p. 10 and JSA (1969) p. 29. 62 JSA (1969) p. 34. Strictly speaking, most supervisors (kantoku-sha) were male. The results of the fieldwork survey confirmed this. 63 JSA (1969) p. 33. 64 JSA (1969) pp. 29–30. 65 Hazama (1997) pp. 14–15.

158 Educating and training women workers was both a form of direct line management and an incentive for women to remain longer in the industry. In doing so, they could be rewarded with promotion opportunities and greater responsibility to varying degrees. Like the provision of dormitory and educational facilities, improved training and promotion contributed not only to a better skilled workforce for production purposes, but also fulfilled industry obligations to encourage the selfdevelopment of female employees.

6

Wages and welfare for women

The term welfare facilities inclusively denotes the various facilities established and operated, at the expense of the enterprises, for the increase of material as well as spiritual benefits to the workers and their families. (Shindo 1961, p. 90) Wages and welfare can be closely identified in character because they both provide material benefit to workers. It is this view that gave basis for the contention that company dormitories should, like wages, be covered by labour legislation and be subject to collective bargaining. However, the push by Zensen after the Second World War to have dormitories recognised as a ‘labour condition’ alongside wages failed, and dormitories remained a welfare facility.1 While wages and welfare must be clearly distinguished from each other, their combined importance during the period must also be evaluated, as they both had a direct influence upon the livelihood and standard of living of employees. Welfare benefits, as separate from but supplementary to cash wages, must therefore be taken into account when discussing the broader issue of wages for female workers in the textile industry. Any consideration of the cost of female labour in textiles must take into account the considerable non-wage costs that were involved, and recognise that the provision of various welfare facilities was an important factor in attracting female workers to textile employment – it was not just wage levels which were important. This study locates welfare considerations and wage payments to females within the context of the overall working environment, and casts doubt on the value of simple wage levels as an indicator both of the economic well-being of female textile workers and of the status of their work.

1 Shindo (1961) pp. 90–91 noted that a clear distinction between the utilisation of welfare facilities and the payment of wages meant that welfare facilities such as dormitories did not constitute a workplace condition, and therefore remained separate from the labour contract relation (emphasis is mine).

160 Wages and welfare for women

Women and the wage system In the prewar period the calculation of wages was done on a piecework basis. A female operative working in cotton spinning or weaving was paid wages according to the quantity and quality of the work she produced.2 In the postwar period, however, fixed wages became the norm. Only a few workers, mainly in independent weaving establishments or in knitted goods, continued to be paid under the piecework system. While initially the wage system was based on a daily wage (nikkyu¯), in the 1960s there was a general move towards a monthly wage (gekkyu¯), consistent with ideas of a standard wage (kihon chingin).3 The daily wage basis remained, however, as female production workers in textiles were mostly paid monthly, but based on a daily wage calculation system (a monthly wage system of calculation was rare). The wage system most representative of the industry was one based first on age and education. A starting salary system was adopted throughout the cotton industry, with female workers receiving a minimum starting salary when initially employed at the age of 15, which was then raised after six months of employment and then again at the end of the first year. At the age of 18, after three years of service, their monthly wage rose again to a minimum wage for 18 year olds, and was further raised every year after that. The wage structure in textile factories, particularly in medium to large-scale enterprises, operated under a complex system which took into account factors such as an individual’s qualifications and characteristics, personal attributes (including gender) and nature of job performed. Wages were calculated first using an individual wage, which was based on the starting salary levels already outlined as well as age, educational background and length of service. A basic wage was then calculated reflective of job duties and functions, and to this various other allowances were added, which for women mainly included shift-work allowances and ranking allowance (see Figure 6.1). Additional ‘living compensations’ took into account factors such as family circumstances, regional circumstances and any need for household allowances. The range of wage allowances during this period reflected a concern to increase living standards and alleviate daily living expenses, as well as a movement to achieve unity and equality across the industry. In general this can be viewed as having been part of a national pattern under high-speed economic growth in Japan. Industry documentation usually defined the basic wage as a total of the individual and basic wage shown in Figure 6.1 (i.e. determined by age, education, service years, ability, type of work, etc.). The JSA commented that in 1954 this represented about 50–70 per cent of the total wage. Incentive wages or piece-rate wages in the same year accounted for around

2 Tsurumi (1990) p. 148 notes that piecework wages were common for women whereas most male cotton workers were paid fixed wages according to wage classifications. 3 Interview I.

Wages and welfare for women 161

Individual wage (honninkyu¯ ) ● ●

●

➡

Age Educational background Length of service

Basic wage (kihonkyu¯ ) ● ●

➡

Duties Functions

Allowances

●

●

●

Gross pay

➡

Deductions

● ● ● ●

Welfare payments Food contribution Dormitory fee Union fee

Shift-work allowance Ranking allowance Living allowances

➡ ●

➡

Gross pay

●

●

Production incentives Overtime pay

Take-home pay

Tax

Figure 6.1 Example of a typical (monthly) wage system.

15–35 per cent of aggregate wages, while overtime pay and non-working pay (i.e. bonus payments) accounted for 2–3 per cent of the total wage packet. The living allowances, including family allowance, regional allowance and housing allowance represented about 10 per cent of the total but was noted as ‘gradually decreasing of late’.4 Table 6.1 provides an indication of the average monthly wages of female workers in the textile industry from 1955 to 1975. Wage data are broken down in these Ministry of Labour statistics by ‘contractual’ (individual/basic) wages and ‘special’ wages (including additional allowances and annual bonuses), and several initial comments can be made. First, the figures indicate that female wages saw a significant increase, at least in nominal terms, during the 20-year period. Second, the proportion of total female wages made up of special payments, including allowances and bonus payments, grew from around 10 per cent of the wage packet in the mid-1950s to over 20 per cent by the mid-1970s. Third, the increase in the wage figures points to a trend of rising wages, better wage conditions and effects on the ‘prosperity’ of workers throughout the period, subject to the circumstances of inflation. While the figures in Table 6.1 are average wages, various factors influenced the actual ‘spread’ of wages, including the various components presented in Figure 6.1 such as age and education, as well as factors such as gender, employment status, and scale and sector of enterprise. Wages differed by the scale of enterprise as well by product sector across the textile

4 JSA (1954) pp. 8–9. These figures are based on average wages in the Big Ten companies.

162 Wages and welfare for women Table 6.1 Average monthly wages per female regular employee (¥)

1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975

Total

Contractual

Special

% contractual

% special

7,493 8,126 8,254 8,063 8,893 10,057 11,384 13,288 14,613 16,126 18,109 19,713 21,774 25,289 29,944 35,914 40,928 47,837 61,451 72,457 78,157

6,816 7,154 7,179 7,149 7,605 8,538 9,514 11,241 12,192 13,513 15,321 16,657 18,202 20,878 24,337 28,571 32,902 37,989 45,385 56,796 63,415

677 972 1,075 914 1,288 1,519 1,870 2,047 2,421 2,613 2,788 3,056 3,572 4,411 5,607 7,343 8,026 9,848 16,066 15,661 14,742

91.0 88.0 87.0 88.7 85.5 84.9 83.6 84.6 83.4 83.8 84.6 84.5 83.6 82.6 81.3 79.6 80.4 79.4 73.9 78.4 81.1

9.0 12.0 13.0 11.3 14.5 15.1 16.4 15.4 16.6 16.2 15.4 15.5 16.4 17.4 18.7 20.4 19.6 20.6 26.1 21.6 18.9

Source: Ro¯do¯ To¯kei Nenpo¯. Notes Figures are for all textiles industries. Workers are regular workers only in companies of 30 or more workers.

industry. The MOL 1952 survey noted that the average cash wage for female workers in New companies established after the Second World War was on average 10 per cent lower than the wages in the established Big Ten companies.5 JSA figures for 1955 indicated that average hourly wages for females in the New and New-New companies were only 70 per cent of the hourly wages for those in the Big Ten,6 while MOL data for 1960 indicated that average wages for females in small- to medium-scale enterprises was around 15 per cent lower than those in large-scale enterprises. However, after 1960 the gap in average wages between small- to medium- and largescale enterprises appeared to show a decline. By 1965 wages in medium-scale enterprises were only around 1 per cent lower than those in large-scale enterprises, while in small-scale enterprises they were around 4 per cent lower.7 JSA figures for 1965 also indicated that wages in the New and New New companies were now only 1–3 per cent lower than those in the Big Ten.8 5 6 7 8

Wages as of November 1950, MOL (1952) p. 4. Men Bo¯seki Ro¯do¯ To¯kei Shiryo¯ (1955) p. 43. Chingin Ko¯zo¯ Kihon To¯kei Cho¯sa (1960, 1965); see Macnaughtan (2001) p. 282. Men Bo¯seki Ro¯do¯ To¯kei Shiryo¯ (1965) p. 36.

Wages and welfare for women 163 This narrowing of the gap in wages between small- to medium- and largescale enterprises was the result of efforts by the textile unions to make unified wage demands and set wage levels across the industry during the high-speed growth period. Industry documentation indicates that basic starting salaries for female junior high school leavers, following Zensen demands and industry negotiation, were set at the same levels both in the Big Ten companies and in companies within the Nishiwaki sanchi union after 1960.9 However, this analysis of the spread in wages is based on ‘minimum’ wage levels only. Real differences in the average total wage packet for females across the industry lay in the wages and additional allocations that were paid over and above the basic starting wage levels set by industry–union negotiations, with larger enterprises naturally more likely (and able) to make various payments in addition to the minimum wage levels than smaller enterprises. In essence this reflected the difference in what was known as the wage structure (chingin ko¯zo¯) and that referred to as the wage shape (chingin keitai). The wage structure was the fixed wage system – the starting salaries and wage levels based on age and education and set by wage negotiations. The wage shape was the form that wages took in individual companies after this – payments based on internal company decisions to make incentive or bonus payments and pay various allowances.10 The spread of wages differed not only by scale, but also by job type within the industry. JSA figures indicated that female workers in the integrated weaving departments of spinning companies earned slightly more on average than those in spinning departments.11 Wages for those in independent weaving enterprises were, however, lower than those in integrated weaving and spinning sectors, reflecting a difference in scale as well as industry sector. The MOL 1972 survey recorded that, in general, in sectors where there were more small-scale enterprises and less technology, such as weaving, wages were lower than those found in the large-scale cotton companies. Those in weaving, however, were found to earn comparatively more than those in ‘miscellaneous’ sectors such as braid and ropes.12 Wage differences also existed between production and non-production workers in the industry. On average, those females employed in administrative departments received a higher wage than those in production departments. However, when the difference was looked at by age, the youngest workers received a higher wage in production than those in office

9 The same figures of ¥10,000 in 1963, ¥17,000 in 1967 and ¥35,000 in 1972 were recorded as the minimum starting salaries for female junior high school leavers in JSA (1979a) p. 828 and Banshu¯ Ro¯do¯ Kumiai (1985) pp. 202–03. 10 Koike (1962) p. 79. 11 Menshi Bo¯seki Jijo¯ Sanko¯sho; see Macnaughtan (2001) p. 282. 12 MOL (1972) pp. 72–73.

164 Wages and welfare for women work.13 Wages by department (i.e. production workers versus administrative workers) revealed that female production workers received on average 20 per cent less than female administrative workers, although the gap can be seen to have narrowed somewhat over the period.14 While the different situations of production and non-production workers must be recognised, and include female dormitory staff and teachers within the industry as well as office workers, their low numbers in proportion to numbers of production workers can be said to not significantly distort the industry wage data overall. A key factor that did influence the structure and spread of female wages in the industry was, however, the issue of gender. Wage differentials Because of its high employment of females, the textile industry offers a good case for examining the dynamics of a ‘female wage system’ in operation in postwar Japan. The main individual wage throughout much of Japanese industry during the postwar period was based on a worker’s age, gender, educational background and length of service. Within this system, wages for women were heavily influenced by the fact that women moved in and out of the workforce during their working lives, and were also employed under a wider range of (regular and non-regular) worker status within the economy. There were substantial wage differentials by sex in the textile industry, as Table 6.2 shows, with female workers earning on average 47 per cent of male wages over the period.15 A large factor in this differential was age. The wage disparity increased by age grouping, with young female workers earning 80– 90 per cent of the male wage under the age of 20, but older female workers in their forties earning less than 40 per cent of the male wage. The MOL 1952 survey also recorded that there was very little male–female differential for those under the age of 18 with girls earning 98 per cent of the male wage, but this changed to 80 per cent at 20–25 years, 52 per cent at 35–39 years and 49 per cent at 40–49 years. Wage differentials were also tied to length of service. In the MOL 1952 survey, females earned on average 72 per cent of the male wage at a length of service of less than six months, but this changed to 51 per cent at 15–20 years of service.16 Table 6.3 shows a breakdown of average monthly wages for female workers by age group for the five-year benchmarks between 1960 and 1975. Initial wage differences are apparent between female minors and those

13 MOL (1952) p. 5. 14 Chingin Ko¯zo¯ Kihon To¯kei Cho¯sa; see Macnaughtan (2001) p. 282. 15 The MOL 1952 survey reported a similar figure. In cotton spinning factories the monthly male–female wage differential was on average 48 per cent, which was reported as not unusually large when compared with other industries, but noted as a significant feature for an industry based predominantly on female labour: MOL (1952) p. 5. 16 MOL (1952) p. 5.

45.7 47.5 47.8 48.0

94.6 93.8 91.2 90.8

87.5 83.6 80.8 84.3

18–19

71.8 70.4 71.7 73.1

20–24 56.2 53.8 54.9 58.5

25–29 42.6 44.6 45.6 47.5

30–34

40.2 42.2 43.8

35–39

8,641 15,900 30,000 68,300

6,850 13,700 25,800 59,300

17 and under

8,500 15,800 29,000 65,100

18–19

9,491 17,600 32,000 71,000

20–24

10,117 17,600 32,400 72,300

25–29

35–39

{9,570} 17,000 16,700 31,300 31,600 70,100 70,000

30–34

Notes Figures are average wages for workers in all textile industries. Figures are basic monthly wages (do not include bonuses and other allowances).

Source: Chingin Ko¯zo¯ Kihon To¯kei Cho¯sa.

1960 1965 1970 1975

Total (average)

Table 6.3 Average wages of female textile workers by age (¥)

Notes Figures are calculated from average wages for female workers in all textile industries. Source figures are basic monthly wages (do not include bonuses and other allowances).

Source: Chingin Ko¯zo¯ Kihon To¯kei Cho¯sa.

1960 1965 1970 1975

Average for 17 and all workers under

Table 6.2 Average male–female wage differentials for textile workers by age (% f:m)

45–49

42.2

45–49

{16,700} {31,500} 69,300 72,100

40–44

33.0 38.0 40.3 41.2

40–44

55–59

46.3

55–59

{8,730} {15,200} {30,500} 69,700 65,900

50–54

38.1 43.4 42.6

50–54

13,800 27,000 63,600

60 and over

47.8 53.7 59.1

60 and over

166 Wages and welfare for women above 18 years of age. The 1952 MOL survey reported that the average wage of female minors was around 87 per cent of the average normal female wage, with the move to the normal wage occurring once girls turned 18 or 19 years of age.17 This system was consistent in the industry throughout the period. In general, female wages increased with age, although after 30 years of age a somewhat declining or static trend could be seen, as shown in Table 6.3. This was in marked difference to male wages, which showed consistent and steady increase by age group (until around 50 years of age). This feature of the wage data was due to older women (re-)entering the textile workforce after marriage and child-rearing, particularly as non-regular workers. The MOL 1972 survey also found that wages by accumulated length of service in the industry generally showed a steady increase with service years, but began to drop after 30 years of service for women, while still increasing for men.18 Defence of these sex-based wage differentials in the cotton industry has historically been based on the assigning of different jobs (i.e. ‘male jobs’ and ‘female jobs’) within the industry. Scholars have cited the clear distinction of work assignments between male and female workers, with nearly all female workers in the cotton industry engaged in simple and light labour such as preparation for production, looking after machines and checking produced goods, while male production employees were generally engaged in such work as repair, maintenance and operation of machines (i.e. jobs demanding higher technical ability or supervisory roles). Defence of the differential has also cited length of service as a prominent factor. As women generally worked fewer years than males in the industry, they did not accumulate wage rises in accordance with length of service and seniority. The main male–female differential is, however, found in the rate and nature of wage-increases within the industry, and is not peculiar to the textile industry. It was (and still is) customary in Japan to weave into the wage system allowances for a worker’s need to support family dependants (based on ‘subsistence’ ideas). This created a wage disparity between male workers with dependants and young single female workers, who did not receive family allowances. In addition, because males were seen as the head of the family household, married female workers also did not generally receive such allowances. The MOL 1952 survey commented on this, noting that there were substantial wage differences for those who had a dependent family: those with a dependent family of two persons earned double the amount of those without a dependent family, and those with a dependent family of five earning three times the amount. However, the survey recorded that 88 per cent of female workers were classed as ‘not having a dependent family’. This figure of 88 per cent above is actually low – usually the majority

17 MOL (1952) p. 4. 18 MOL (1972) p. 73.

Wages and welfare for women 167 (i.e. close to 100 per cent) of female workers were classed as not having dependants – and no doubt reflects that male deaths in wartime were still affecting the financial situation of families in the early 1950s and the role of the women as economic providers. JSA figures for 1965 indicate that only 0.5 per cent of females in the Big Ten were classed as having dependants.19 This system all relates to the concept of the ‘family wage’. The payment of a family allowance was one of the first demands made by labour unions after their formal organisation during the Occupation years. Special payments, bearing no relation to job performance by a worker, were to take into account the needs of the family, and of the male worker as the family head. They were thus essentially gender-based wage allowances. Despite the various complexities of the wage system in the industry, therefore, wage disparity was based to a large extent on socio-economic ideals of male– female roles (gendered household roles), creating large differences in wages between the sexes. The textile industry can be said to have been no more ‘peculiar’ or ‘worse’ than other industries in Japan in this respect, and there is a large literature on discrimination in the female labour market, particularly discrimination centred around wages. A core issue in this debate has been how far the differential is a result of discrimination and how far the result of different labour and job attributes, and the quality of human capital. Given the additional ideals and values assigned to male–female roles both within the home and in the economy during this period in Japan and its direct impact on the separate gender-based wage structures, it is not a clear-cut debate.20 It should be emphasised, however, that Japan has been no different in this respect from other countries.21 Allowances and retirement Allowances for female textile operatives were mainly comprised of shiftwork allowances, ranking (female leader) allowances and overtime pay. Shift allowances were the most common, and 98 per cent of respondents in the fieldwork survey said that they were paid a shift allowance.22 In the MOL 1972 survey, 91.1 per cent of those companies operating on a shift-work

19 The corresponding figure for male workers was 80 per cent: Men Bo¯seki Ro¯do¯ To¯kei Shiryo¯ (1965). None of the fieldwork survey respondents stated that they received family allowances in their wage packet. 20 An in-depth discussion of this is beyond the scope of this book, but works on this include Yashiro (1981), Kawashima (1983) and Osawa (1994). 21 See Goldin (1990) and Date-Bah (1997). 22 In the survey 40 out of 45 respondents answered the question on allowances, of which 39 said they were paid shift allowances. The one respondent who said she was not paid a shift allowance worked in the factory (on the shift system) for only six months before being transferred to office work. She was paid a commuting allowance as she lived locally and was a senior high school graduate.

168 Wages and welfare for women system were found to be paying shift-work allowances. The allowances were found to be on average ¥80–100 per day, and represented about 20 per cent of the average ‘special wage’ payment for female workers in that year.23 Commuting allowances were common for day-only shift-workers. The payment of annual bonuses (kimatsu ichijikin) was a feature of the wage system established in Japan during the postwar period. In the textile industry, payments were usually made twice a year, once in summer and once in winter.24 The amount of the payment was calculated based not only on basic factors such as an individual worker’s age and length of service, but also on their ability and performance during the year. While average annual bonus payments were, like wages, often subject to union–management negotiations, the actual allocation (i.e. distribution or spread) of bonus payments was primarily at management’s discretion.25 JSA figures indicate that total average bonus payments, as a proportion of average monthly wages, increased from 1.5 months pay in 1955, to 3.8 in 1960, to 4.2 in 1968 and 4.8 in 1974.26 Another factor influencing additional payments within the wage system, and in particular differences between the male and female wage system, was the issue of retirement. At the beginning of the period, the retirement age was fixed at 55 years of age.27 However, during the period, particularly from the mid-1960s, the union made efforts to postpone the age of retirement. The retirement age was gradually extended during the late 1960s and early 1970s, and then finally fixed at 60 years of age in 1975. This official retirement age in the industry was, however, viewed as being more applicable to male than to female workers. Those women who did remain in the industry, or entered as older female workers, were often encouraged to retire at least

23 MOL (1972) pp. 5 and 37. Also recorded as ¥100 per day in the Big Ten in 1973, JSA (1979a) p. 837. (Calculation is based on ¥100 × 20 days per month = ¥2,000; this represents 20.3 per cent of the special payment figure of ¥9,848 recorded in 1972 in Table 6.1.) 24 In the fieldwork survey, 43 out of 45 respondents said that they received a bonus twice annually. The other 2 said they did receive bonuses but didn’t respond to the question of how often. 25 By age, younger female workers generally received larger bonuses than older women, with less bonus received by those after 25 years of age on average, mostly due to worker status. However, this feature is also reflective of the industry structure, i.e. large-scale enterprises (higher bonuses) versus small-scale enterprises (lower bonuses). Young regular female workers were far more likely to be in the large-scale firms, particularly as the shortage of young labour developed. 26 JSA (1962) pp. 445–48 and JSA (1979) pp. 852–55. 27 This refers to teinen taishoku seido ‘mandatory retirement system’, a system whereby a labour contract ends when a worker reaches a specified age. The usual mandatory age of retirement in the private sector was 55 years of age in the early postwar period, extended to 60 years of age during the high-speed growth period by most industries. During this period the Older Persons Employment Security Law also established a system with retirement at 60 years of age, which set a ‘social standard’ for employers to ‘endeavour’ to meet: Sugeno (1992) pp. 391–92.

Wages and welfare for women 169 five years earlier (in age) than men. For the most part, however, it was expected that female workers would leave the industry after several years of service, usually in their twenties (i.e. after graduation and/or when they married). A retirement allowance (taishokukin) scheme for regular workers existed in the textile industry, and was for those workers who served three years or more in the industry. The amount of the allowance was calculated according to length of service, and industry documentation records that the textile industry was encouraged to ‘seriously consider’ this system as an incentive to keep young female workers longer in the industry.28 However, for women who left and re-entered the workforce for marriage and/or childrearing breaks, the pension often applied only to the initial years of employment when they were full-time regular workers. Most women who re-entered the workforce, at least until the late 1960s, were employed under non-regular employee status, which was not covered under the retirement allowance scheme. Reports indicate that instead a special marriage pension (kekkon taishoku) system existed in the cotton industry for female workers. This comprised a lump sum payment upon their retirement from the industry workforce for marriage. When there had been an excess supply of female workers, the industry had created the system whereby the company would pay a special allowance to a female worker who resigned because of marriage (kekkon kasan seido). This was to encourage the replacement of older female workers by new (young) females,29 and in this respect can be viewed as the operation of a marriage bar. However, due to the declining supply and lack of female workers from the mid-1960s, the majority of textile companies abandoned the system around 1966–67 and resignation for marriage was considered as an ordinary resignation. However, during a revision of retirement allowances in general in 1969, the marriage pension was revived to serve as an incentive to keep young female workers longer in the industry (i.e. until they married).30 Wages for older women As with other issues surrounding the employment of various categories of non-regular workers, statistics and information on the wages of temporary workers in the textile industry during these years are sparse. However, the limited information available provides an indication of the wage levels achieved by non-regular female workers. In 1962 Zensen carried out a survey on temporary workers across the textile industry, recording

28 JSA (1979a) pp. 856–57. 29 JSA (1979a) pp. 857–58. 30 JSA (1979a) pp. 857–58. This would have tied in with efforts to link education attendance to length of service for young female workers.

170 Wages and welfare for women information on employment conditions including wages. The daily wages for a female temporary factory worker (rinjiko¯) in 1962 were recorded as ranging from ¥280 to ¥570 per day in cotton spinning, with an average of around ¥450 in spinning and ¥420 in weaving noted.31 The survey compared an average ‘minimum’ daily wage of ¥300 with the average wages of young regular workers in the industry. Based on 26 days of employment per month, a monthly wage of ¥7,800 was contrasted with the minimum starting wage of a 15-year-old junior high school graduate at ¥8,500, and the minimum wage of an 18-year-old regular worker (not taking into account length of service and other conditions) at ¥10,000. This indicated a minimum wage level for temporary workers at somewhere between 78 and 90 per cent of the minimum wage for a young regular female employee. The survey further noted that, despite the recent supply problems associated with regular (young) workers and the associated emergence of rising numbers of nonregular workers, the gap between regular and non-regular wage levels was large, especially when wage rises, bonus payments and pension payments were taken into consideration. While wages for young regular workers were fixed and annually reviewed, the wages for temporary workers were less fixed. Temporary workers therefore generally earned significantly less than young regular workers even though their average age was likely to be much higher. Zensen commented that there were often skilled workers among the category of temporary workers, making them at least as good as, if not sometimes better (older and more experienced) than regular (young and less experienced) workers. While there were found to be some companies using age and experience as the basis for setting temporary wage levels, the effect this had on wage levels in general was not significant, and in most cases the criteria for setting wage levels were not fixed. The survey cited the case of a regular worker who ‘retired’ and then soon after worked for the same company under a temporary worker status. Upon re-employment the worker was paid a reduced wage level, recorded as 70 per cent of her previous wage level.32 Many temporary workers did not receive allowances such as travel and housing allowances. A central problem that the union faced when seeking better working conditions and wages was that temporary workers ‘needed to work even though the salary was low’. Furthermore, because most temporary workers were older, their incorporation into (or bringing into line with) the regular (honko¯) employee system based on age and length of service would have meant a large wage rise for temporary workers which companies were reluctant to accept.33 31 Wages were lower on average in other sectors such as wool, hemp, synthetics, and as low as ¥200 per day in small enterprises in the wool dying and finishing sector: Zensen (1962b) p. 11. 32 Zensen (1962b) p. 11. 33 Zensen (1962b) p. 16.

Wages and welfare for women 171 A 1970 MOL survey of older women working in a weaving factory noted that the traditional wage system based on age and length of service did not take into account older women (re)entering the labour force. The factory studied indicated that while they tried to take a worker’s ability into account they could not ‘ignore’ the wage system based on seniority and service that had existed for a long time. The factory did, however, include a production incentive wage as a small proportion of the total wage packet. Wages for older women were based on a daily wage, 81 per cent of which was comprised of the basic daily wage, 4.8 per cent of which was production incentive wages and the remainder of which was other allowances, mainly shift allowance and attendance allowance.34 Older women with no previous experience were hired at the same starting salary as a female junior high school graduate, and were eligible for production incentive wages after six months of employment. Those older women with previous experience were eligible after two months, subject to management review. Bonus payments for older women were calculated on a sliding wage scale and allocated at management discretion, although they were very low compared to the bonus payments of regular workers. The hourly wage rate was recorded in the survey as ¥110–15, not including production incentives and allowance payments.35 This figure is much lower than that recorded by the Zensen survey in 1962 already outlined, particularly compared to daily wages in spinning. There is evidence to suggest, however, that wage levels for temporary workers showed some increase after the mid-1960s as more were employed in the industry to compensate for the lack of young regular female workers. The wages for the seasonal employment of women under the Hokkaido¯ scheme were highly structured. Under this special employment plan for seasonal female workers (see Chapter 3), a wage structure based on a daily wage calculation was employed (see Table 6.4). Workers were employed under unified wage conditions agreed across the industry. These included a basic daily wage, which saw substantial nominal increase during the period of operation, and an experience allowance, based on an additional daily amount for each year of experience. For example, in 1970 the allowance for workers with three years’ experience would have been ¥90 (3 × ¥30) per day. Also, an annual one-off start-up allowance was given as a relocation or bonus payment, which also saw a substantial increase in nominal value over the period. These daily wage rates (i.e. not including the various allowances and extra payments) were found to be fairly comparable to the average basic (contractual) wages for female workers shown in Table 6.1, indicating that, at least for a temporary period, female seasonal workers received good

34 MOL (1970) p. 80. 35 MOL (1970) p. 82.

172 Wages and welfare for women Table 6.4 Wage structure for female seasonal workers in the ‘Hokkaido¯ Plan’ (¥) Daily wage

1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978

630–650 630–650 650–680 800 810 900 1,200 1,300 1,500 2,000 2,500 3,000 3,300 3,300 3,500

Experience allowance

650–670 670–700 890 (incl.) 30 30 30 50 50 50 70 70 70

Starting allowance 3,000 3,000 3,000 3,000 3,000 5,000 10,000 10,000 10,000 20,000 20,000 20,000 20,000 20,000 20,000

Daily wage-increase from previous year (%) 0.0 5.5 18.5 1.3 11.1 33.3 8.3 15.4 33.3 25.0 20.0 10.0 0.0 6.1

Source: JSA (1979b) p. 29. Notes Daily wage based on wage for those with no experience. From 1970 experience allowance is daily allowance for each year of experience (prior to this, it is a one-off lump sum payment for the season). Between 1968 and 1974 companies might also pay additional allowances (e.g. for shift-work); however, information on the amount was not available. From 1975 this was generally included in the daily wage calculation.

average wages relative to basic female wage levels in the industry.36 It also indicates that the industry had to offer a fair wage level in order to secure even temporary female workers as the labour shortage of young female regular workers squeezed their labour requirements during these years.

Nominal, comparative and real wage trends A significant feature of the wage data for female textile workers was the remarkable increase that took place in nominal wages during these postwar years. The average monthly female wage in 1975 stood at ten times the nominal wage in 1955.37 This trend in textiles can be said to be representative of general wage movements during the high-speed growth period, which saw

36 Calculation is based on taking average daily wage figures from Table 6.4 and multiplying by 20 days per month which is roughly comparable to the contractual wage figures in Table 6.1. For example, in the three benchmark years of 1964, 1970 and 1975 this gives the following figures respectively: ¥13,000 compared to ¥13,513 in Table 6.1; ¥24,000 compared to ¥28,571; and ¥60,000 compared to ¥63,415. 37 See Macnaughtan (2001) p. 297.

Wages and welfare for women 173 substantial money rises. For female wages in the textile industry, a large portion of the nominal wage rise can be attributed to movements in the starting wage level, particularly that set for 15-year-old female junior high school graduates. Table 6.5 shows the level of initial wages in the Big Ten companies from 1961, and the wage rises that occurred. Table 6.5 also makes a brief comparison with the situation in the manufacturing industry in general. Significant features are that, first, wages saw a substantial increase in nominal value over the period (column A); second, the percentage wage rise that occurred (column B) appeared to be consistently above the annual percentage rise in the consumer price index (CPI: column C); and third, while prior to 1970 female junior high school graduate salaries in the Big Ten were above the manufacturing average, from 1970 they appear to have been below the average (compare columns A and D). Overall, however, the annual wage increase rate for this level in textiles was on par with or above the manufacturing average (compare columns B and E). The industry commented that, after 1955, particularly from 1960 onwards, starting salaries rose dramatically due to the emerging supply problems of younger labour. In 1965 the industry commented that the ‘most notable feature’ in the labour situation of Japan’s cotton industry was the labour shortage, which had become apparent around 1955 and thereafter increasingly acute, and that continually increasing wages were a reflection of this labour shortage.38 The report also noted that the rapidly increasing competition and demand for female labour in other sectors of the economy, particularly electrical appliances, precision machinery as well as wholesaling and retailing, were also putting pressure on wage levels. The industry also commented on the increasing power of labour unions during these years. It noted that union activities were being carried out on an annual basis, calling not only for improved cash wages, but also for improvements to welfare facilities, bonus payments, retirement allowances and workplace conditions.39 The industry reported that lengthy negotiations between labour and management, particularly after 1960, involving strikes in the Big Ten companies, had resulted in basic wage levels in both the Big Ten and in many other cotton firms being raised significantly. Basic nominal data for female textile workers alone, however, cannot show the entire wage picture. The wage levels in the industry during this period need to be placed both within a comparative manufacturing context, and analysed in terms of real wage value. The 1952 MOL survey noted that the average monthly wage for females in the textile industry as a whole was ¥5,381, while the manufacturing average was ¥5,080, suggesting that female wage levels in the industry were not comparatively low.40 Textile industry 38 JSA (1965) p. 77. 39 JSA (1965) p. 78. 40 MOL (1952) p. 4. Average hourly wages for female workers were ¥27.9 in spinning, ¥29.4 in all textiles and ¥27.3 in all manufacturing.

14.3 17.3 10.1 10.5 15.1 16.4 17.6 18.4 16.8 15.7 22.1 31.4

2,736 3,657 2,370 2,589 4,049 4,875 5,853 6,921 6,994 7,752 12,400 20,913 6,300

10,000 13,000 13,700 15,000 17,000 19,500 22,500 25,500 29,500 35,000 41,000 53,000 53,000

1,500 3,000 700 1,300 2,000 2,500 3,000 3,000 4,000 5,500 6,000 12,000

27.0

(B)

8,500

Source: JSA (1979a) p. 828.

1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975

(A)

Wage increase rate (%)

1,236 657 1,670 1,289 2,049 2,375 2,853 3,921 2,994 2,252 6,400 8,913 6,300

Amount of wage rise for registered worker 5.3 6.8 7.6 3.9 6.6 5.1 4.0 5.3 5.2 7.7 6.1 4.5 11.7 24.5 11.8

Rise in CPI from previous year (%) (C) 8,311 9,489 10,515 12,224 13,783 14,833 16,493 18,381 21,771 26,262 31,326 36,160 44,121 60,015 61,943

(D)

Amount (¥)

625 1,178 1,026 1,709 1,559 1,050 1,660 1,888 3,390 4,491 5,064 4,834 7,961 15,894 1,928

Rise from previous year (%)

(New) Amount (¥)

Amount (¥)

Rise from previous year (¥)

Starting salary

Annual wage increase

Starting salary

13.6 10.5 8.8 12.2 9.7 9.0 12.1 13.3 15.6 18.1 16.3 14.9 19.4 31.8 12.9

(E)

Wage increase rate (%)

2,322 1,316 1,195 1,542 1,379 2,191 2,631 3,291 3,429 4,549 4,305 4,997 6,690 12,781 13,295

Amount of wage rise for registered worker (F)

Annual wage increase

Manufacturing industry average

Big Ten cotton companies

Table 6.5 Starting salary for female junior high school graduates and wage-rise movements

Wages and welfare for women 175 records also suggested that in 1961 wage levels were high compared with other manufacturing industries, with cotton mill workers cited as better paid by 44 per cent for males and 14 per cent for females on the basis of total average wages including term-end bonuses.41 However, the 1972 MOL survey commented that average starting wages for female school leavers entering the textile industry were below the national average for all industries.42 This indicates that a decline in textile wage levels relative to the industrial average took place over the period. A comparison of average monthly wage data for female employees within the main manufacturing industries suggests that female wage levels in textiles were consistently below average manufacturing wage levels, at around 90 per cent throughout the period (dropping somewhat after 1973).43 This appears to have been in line with female wage levels in food manufacturing and above wage levels in the apparel industry. Wage levels in the electronics and metals industries were closer to the manufacturing average, while wages in the machinery and iron and steel industries showed wage levels above this manufacturing average. Overall, female wages in textiles appear to have been around 10 per cent lower than wage levels in its main rival for labour, the electronics industry. It can also be said that female wages in the light manufacturing industries (textiles, apparel and food) were at levels below those in the heavy industries (metals, machinery and iron and steel). As already noted, however, wage levels differ significantly by age, educational background and length of service. Average female wage data do not take into account different demographic compositions within each industry, which have an effect on average wage levels. The much higher employment of younger workers, particularly junior high school leavers, by the textile industry had the effect of lowering average wage figures comparative to other industries. Even by the end of the period, when the number of older females had increased within the industry, textiles showed a significantly higher employment of the very youngest females (15–19 years) within its workforce compared with the other industries. In terms of younger females, other industries, particularly electronics, employed a far greater proportion of females aged in their twenties compared with textiles. For example, textiles had the lowest proportional employment of females aged 25–29 among the main manufacturing industries. While its employment of older females was more in line with other industries (and its employment of females above 45 years was comparatively high), the textile industry’s heavy use of the very youngest females, who had the lowest basic wages set by age, 41 JSA (1961) pp. 51–52. 42 MOL (1972) pp. 36 and 186. 43 For detailed analysis see Macnaughtan (2001) pp. 300–04. Nominal wage levels for females in textiles and their relation to average manufacturing wage levels (expressed as a percentage of manufacturing wages) are shown on p. 302.

176 Wages and welfare for women education and service criteria, can be said to have had a significant effect on average wage levels.44 The continued ‘young’ feature of the female labour force, particularly regular workers, in textiles therefore had the effect of lowering average wages absolutely and comparatively. Large increases in nominal wages can often be the result of general inflation. While wages showed a tenfold increase in nominal terms from 1955 to 1975, there was only a threefold increase in real terms.45 However, a threefold increase is still significant, and until the mid-1970s wage increases outpaced general price rises, resulting in a considerable improvement in the real earnings of female textile workers during this time. While wages still rose in 1975, there were signs of a slight drop in the rate of increase in real wages, no doubt reflecting the relative decline of the industry by this time as well as the significant increase in the employment of older and non-regular female workers. Wage trends and the union Wage decisions, particularly annual wage increases, were the result of annual bargaining between industry management and industry unions, known as the spring labour offensive (shunto¯). While union calls for higher wages began early on during the Occupation years, it was in 1955 that the industry’s most serious wage dispute of the postwar period occurred. Opposition between union and management over the union’s call for higher wages in the Big Ten companies spanned 20 days, and developed into factory strikes involving 140,000 workers at its peak. Workers (including large groups of women) went on strike in October in front of the main JSA cotton company gates and marched with placards through the streets, ¯ mi Kenshi strike of the previous year (see Chapter 2). In reminiscent of the O fact, management was so worried about the power of labour following the ¯ mi Kenshi strike that it attempted to crack down on the strike activities O very quickly. However, much support was shown for labour, not only from Zensen but also from national labour and union committees.46 As the strike ¯ mi Kenshi strike, it came to intensified in both spirit and in length, like the O the attention of politicians and to the media and public, and was an economic blow to the industry, which was reportedly losing at least ¥600 million per day through the halt of production and exports. In the final days of the strike the difference between the union demands and the company settlement offers had in fact become so small that some of the press labelled the strike the ‘caramel strike’, as the amounts under negotiation were not much

44 See Macnaughtan (2001) p. 303. 45 For details of real wage calculation see Macnaughtan (2001) p. 305. 46 Zensen (1955). This source is actual film footage taken of the 1955 wage strike (now on video).

Wages and welfare for women 177 47

more than the cost of caramel sweets. However, for the unions it was the principle that was at stake by this time, and eventually management agreed to accept the union’s minimum demands for wage levels, and a compromise settlement was reached in November 1955. Zensen commented that, in retrospect, because the unions had held out in the strike in this crucial year they had set a precedent which moved into annual wage offensives. While the wage strike of 1955 was recorded as ‘something out of the ordinary’ in its events at that time, it was seen as a turning point for the industry in terms of making wage negotiations possible.48 Union moves to centralise wage-rise negotiations and make wages uniform (i.e. reduce disparity) across the different strands of the textile industry and across different scales of textile enterprise became a key feature of the period under study, and although not as colourful as the 1955 strike, were also not devoid of struggle. Overall, these efforts to standardise and unify the industry can be said to have been successful. An industry report in 1968 commented that wage rises had been very high in recent years, and the wage level of starting rates for young workers in particular had gone up remarkably, resulting in almost no difference in wage rates between different enterprises or across scales of operation.49 By the mid-1970s, however, high rates of national inflation as well as larger than expected wage rises in 1972 led the industry to comment that wage increases had moved away from common sense.50 The industry began to move much of its production overseas, and to a large extent this move was because of rising wages (and subsequent increased production costs). While the Japanese cotton textile industry had once enjoyed, and was criticised for, its wage advantage over other industrialised countries producing textiles, by the end of this period this was certainly no longer evident. In fact, rising wages had placed the industry at a significant disadvantage compared to the new Asian economies where the production of textiles was rising, and with which the Japanese industry competed in terms of exports.51 In addition to wages, increased welfare also meant that Japanese female textile operatives could no longer be seen as cheap labour in any absolute or relative sense.

Welfare for women Welfare facilities in the cotton industry originated from its long history of employing young female workers. The heavy use of young females,

47 48 49 50 51

Zensen (1955). Zensen (1955). JSA (1968) p. 72. Nenju¯ no Bo¯sekigyo¯ (1973) p. 57. Park (1976) p. 17 commented that wage levels for textile workers in Taiwan and South Korea were around one-third of their Japanese counterparts, and around one-fifth in Thailand and Indonesia.

178 Wages and welfare for women including minors, resulted in an accompanying natural need to maintain a large number of protective measures for the young girls within the care of the industry. In addition to this prewar legacy, the early postwar situation brought additional welfare needs. After the Second World War, there were problems generally in Japan with sustaining daily life, including a lack of housing and adequate levels of food for many of its people. The provision of welfare to improve the daily living standards of workers was a topic of discussion very early on during the Occupation period, discussed by both SCAP and by industry unions. During the postwar years, welfare facilities were introduced into nearly all industries, and were no longer to be considered peculiar to the textile industry alone. With the implementation of the Labour Standards Law, in particular, many industries in Japan began to pay serious attention to the establishment of welfare programmes for their workers. However, the cotton textile industry’s attention to welfare facilities had a long history, and in this respect it was a forerunner on welfare issues. Moreover, the industry’s intensive provision of welfare for young female workers made it somewhat unique, particularly with regard to facilities such as schools and education, which were not as widespread in other industries. While welfare did become a standard feature in many industries, the range of welfare for female workers and the subsequent increasing costs of welfare became a serious issue for the textile industry, particularly in the later years of the period. As the socio-economic situation changed, companies began to try to cut or reduce unnecessary or out-of-date welfare to avoid financial difficulty.52 However, overall, the extension of new and improved welfare facilities was a notable feature of the years under study. Welfare facilities in the postwar textile industry can be classified under the three general categories of accommodation, education and health. Accommodation facilities consisted primarily of boarding dormitories for single, predominantly female, workers, while company housing was provided for married male employees and their families. Because females were more numerous in the industry, female dormitories were viewed by the industry as being the core of welfare facilities, and together with additional attached facilities, including dining halls, hospitals, schools, shops and even beauty salons, they combined to create a welfare structure which provided for the daily lives of workers. While dormitories and education for female workers have already been discussed in previous chapters – in terms of their development, management and operation – the emphasis of the discussion in this chapter is on their nature and official classification as welfare facilities. Dormitories and education were not legal welfare requirements, although dormitories were partly covered by legislation if provided by industry. The content and scope of both dormitories and education as

52 JSA (1979a) pp. 863–65.

Wages and welfare for women 179 welfare facilities versus their expansion in the postwar years as ‘systems of labour management’ highlights both the complex nature of the two as well as their importance in the employment of female workers by the textile industry. Crucially, this chapter will also evaluate the cost of both dormitories and education, and the implications for the standard of living of female workers. It should be noted that, during these years, there were various national welfare insurance schemes (shakai hoken) in operation, towards which companies were legally obliged to make payments. Many of these were instituted in the immediate postwar years. As these welfare provisions were required by law, they were the same for all industries, and an analysis of these systems in detail is beyond the scope of this study. The main focus in this examination of welfare is additional ‘non-legal’ welfare provided by the cotton textile industry, as well as those legal welfare provisions specific to women workers.53 Dormitories as welfare Dormitories were integral to the history of the textile industry and over time they developed away from being primarily for production purposes to become socio-economic facilities for workers. In the postwar period, particularly from the early 1960s, a strong movement towards the modernisation of dormitories took place, and the industry made conscientious efforts to provide facilities in addition to the living quarters prescribed by law. These included health, cultural, sports, recreational and educational facilities. This was alongside moves to democratise dormitories, which provided internal systems for the guidance of boarders, personal counselling and selfgoverning ability of residents (see Chapter 4). The predominance and provision of dormitories for young unmarried workers were not unusual and was a feature of Japanese industry during these years. In a 1964 survey by Zensen, welfare data indicated that 93.7 per cent of all industries in Japan had ‘single’ dormitories (dormitories for unmarried workers), and there was a particularly high use (in terms of numbers of employees in dormitories) by the textiles, steel and transport industries. For married workers it was more common to provide company housing. In the same survey 87.4 per cent of industries had company housing, with particularly high provision seen in the manufacturing and finance industries.54 In the MOL 1972 survey, 60 per cent of all textile companies surveyed had female workers living in dormitories, and this was 53 Welfare provided by the textile industry above and beyond that required by law is referred to in this study as ‘non-legal welfare’, to differentiate it from ‘legal welfare’. This term was used by the industry in English-language publications, including To¯yo¯bo¯ (1955) p. 29. 54 Zensen (1964) pp. 46–47.

180 Wages and welfare for women as high as 95.8 per cent in large-scale companies, with the spinning sector especially high.55 At the time of the 1952 MOL survey all the textile companies surveyed were observed as having dormitories in line with the provisions of the Labour Standards Law. However, due to emergency production operations in place at that time, it was recorded that there were increasing cases of regulations slipping, as companies were unable to cater for the rapid increase in workers.56 By the time of the MOL 1972 survey, however, the proportion of current residents to available (full) dormitory capacity was as follows: 23.2 per cent of dormitories were 30 per cent full; 28.7 per cent were 40–50 per cent full; 38.9 per cent were 60–80 per cent full; and 8.5 per cent were 80 per cent or more full.57 Therefore, it can be said that the situation by the end of the period was such that space was available in the dormitories, compared with the situation in 1952 whem some overcrowding was present. This is not only reflective of the changed structure of female labour that occurred during this time, but also the result of improvements and modernisation of dormitory living standards. An industry report in 1969 noted that, although the large decline in labour numbers had given birth to more ‘elbow room’ in dormitories, the rises in employee income had resulted in an increase in the personal effects of boarders. It also commented that the Westernisation of the lives of young females and the increase in importance of an individual or independent life had resulted in a situation of fewer boarders but more individual rooms.58 Dormitories in the cotton industry were classified as ‘dormitories attached to the workplace’, which meant they were located within company/ factory grounds. Traditionally, textile dormitories had been wooden twostorey buildings within factory grounds. From the early 1960s there was an increase in concrete dormitory buildings with what the industry referred to at the time as ‘deluxe’ (i.e. ‘modern’) facilities, to the extent of including a music room and library (Plates 6.1 and 6.2).59 At this time, several threestorey complexes made of concrete had also been built and the number of these was on the increase.60 The 1952 MOL survey recorded that the majority of dormitories had a capacity of 1,000 persons or under, although there were some large dormitories which could house 2,000–3,000 persons.61 In the 1972 survey, of the 293 companies in the survey that had dormitories, 20.8 per cent accommodated fewer than 30 persons, 32.4 per cent accommodated 30–99 persons, 22.2 per cent accommodated 100–299 persons and 23.9

55 56 57 58 59 60 61

MOL (1972) p. 43. MOL (1952) p. 6. MOL (1972) p. 41. JSA (1969) p. 36. JSA (1969) p. 35. Kobayashi (1963) pp. 147–48. MOL (1952) p. 7.

Plate 6.1 A modern three-storey dormitory building for female employees. Photo courtesy of To¯yo¯bo¯.

Plate 6.2 Club building and recreation area. Photo courtesy of To¯yo¯bo¯.

182 Wages and welfare for women per cent could accommodate 300 or more persons.62 This would indicate that, generally speaking, a single factory dormitory was likely to accommodate a few hundred girls rather than thousands of girls, particularly by the end of the period. The Labour Standards Law detailed legal requirements for physical standards of dormitories, mostly in terms of building structure and size. Under this legislation, floor space in dormitory rooms was limited to up to 16 persons per room with a minimum requirement of 1.5 tatami mats (2.5 square metres) for one person.63 In the 1952 MOL survey the average size of rooms was recorded as 15–20 tatami mats, sleeping 10–12 persons per room, on three-layer futons, indicating rooms in line with the legal standards. Although most of the dormitories were at full capacity at the time of this survey, none of the dormitories were observed to be using the system of both (morning and evening) shifts sharing the same room, which had been a common complaint in the prewar industry. In most cases the different shifts were accommodated in different annexes of the building. By 1963 the average number of boarders per 15 tatami mat room was recorded as down to seven or eight girls in To¯yo¯bo¯ dormitories. In addition to (bed)rooms, standard facilities were recorded as bathing areas, laundry facilities, cooking areas and interview/study rooms.64 However, by 1969, the industry noted that the old style of dormitory rooms having 15 tatami mats and housing up to ten boarders had disappeared, and that now two or three persons per room was common. The industry also commented that the new direction in dormitories was towards individual rooms with Western beds. Across the textile industry, the 1972 MOL survey recorded that the majority of dormitory rooms (76.8 per cent) were traditional Japanese tatami mat rooms, with the number of persons per room predominantly between one and four persons, with five or more persons per room accounting for only 12.4 per cent of all industry dormitories.65 Moreover, Zensen recorded that a movement towards ‘adult’ or ‘household’ dormitory accommodation, particularly for women aged 18 and over, had required the establishment of new facilities from around 1968. Dormitories became like apartments in nature, particularly the popular ‘2DK’ format, comprising two rooms, a dining kitchen area, bathroom and own entrance (genkan). This type of apartment usually accommodated four women living together.66 While physical standards for dormitories were covered under legislation, many of the facilities, especially in terms of equipment, provided by dormitories were not. With the movement to modernise dormitories, a trend towards ‘fuller’ room interiors and contents was noted. This included, in 62 63 64 65 66

MOL (1972) p. 40. JSA (1954) p. 13. Kobayashi (1963) p. 148. MOL (1972) pp. 44–45. Zensen (1996) p. 114.

Wages and welfare for women 183 particular, the general provision of modern radiators or heaters, kotatsu (tables with in-built foot-warmers) and electric fans during the 1960s, a reflection of the quickly changing times in Japan.67 This was in contrast to the MOL 1952 survey which recorded that the heating of dormitories largely consisted of charcoal braziers (hibachi).68 The fieldwork survey asked those respondents who lived in dormitories to indicate what facilities (and live-in staff) were in their dormitories, and the results are presented in Table 6.6. Dormitory facilities were primarily operated at the expense of the industry, although some small contribution from workers was sometimes required. From the late 1960s and early 1970s some companies took to deducting ‘boarder expenses’. These were usually minimum contributions and were based more on ideas about teaching young females about paying ‘rent’ and budgeting, rather than on a need for the company to recover expenditure. The payment of subsidised levels of rent was a particular feature in adult apartment-based dormitories. Other small payments were required by the girls, such as annual dormitory self-government fees, which usually went towards recreational and cultural events. Overall, the trend towards more dormitory space per boarder and the Table 6.6 Dormitory facilities Dormitory equipment

(%)

Additional rooms and staff

(%)

Dormitory bath Washing area Laundry and ironing room Washing machine Dryer Kotatsu Brazier Air conditioning unit Kitchen Desks Muscial instruments (usually piano) Television Telephone (usually communal) Radio

94.7 86.8 89.5 57.9 18.4 84.2 28.9 2.6 71.1 63.2 47.4 71.1 60.5 36.8

Music room and/or lesson room Consultation room Sports club Sick room

65.8 76.3 55.3 55.3

Rooms organised by age levels Adult dormitory and/or adult rooms Matron Nurse Consultant or adviser Room leader Teacher Other staff

39.5 47.4 76.3 57.9 15.8 89.5 68.4 21.1

Source: fieldwork survey. Notes Based on 38 of the 45 surveys who responded to this question. Of the respondents, 10 per cent lived in the dormitories during the 1950s, 66 per cent during the 1960s and 36 per cent during the 1970s (includes some overlaps when a respondent lived for many years in the dormitory).

67 JSA (1969) p. 36. 68 MOL (1952) p. 7. A Zensen survey of dormitories across the industry also recorded that 48 per cent of female dormitories were still heated by hibachi: Zensen (1963) p. 39. In the fieldwork survey, 29 per cent said they had hibachi, and of those the majority (73 per cent) were living in the dormitories prior to 1963.

184 Wages and welfare for women modernisation of facilities and equipment was the result of the influence of several factors. First, modernisation and improvement of conditions in the dormitories were a clear attempt to overcome the Joko¯ Aishi image. Second, the trends in dormitory living during these years reflected the changing socio-economic climate and improved standards of living in Japan in general, including the rapid development of everyday consumer appliances. Third, the facilities had to take into account changing management strategies, particularly the new management style of organising young female workers by age groupings (juniors and seniors) and fostering their independence and personal development. The well-being of workers The textile industry had a history of providing various additional (non-legal) health and medical welfare facilities for labour. To some extent, these facilities were an attempt to break with the industry’s image as a ‘hotbed of tuberculosis’ in the prewar years.69 Tuberculosis was an important concern of the prewar industry, and it was frequently criticised as a cultivating bed for the spread of tuberculosis in Japan’s rural areas as female workers returned home from factories ill and spread the disease.70 The incidence of the disease peaked nationally in 1947–48, mainly due to general poverty and malnutrition, but thereafter effective immunisation reduced the disease.71 The role of the industry in spreading the disease has been disputed by some researchers. In any case, it is certain that the image of textiles being associated with tuberculosis spurred the industry to make efforts to eradicate the disease during the late prewar and early postwar periods. In the postwar period particularly, general progress in medicine, as well as an ‘early discovery-early treatment’ rule established by the industry, resulted in a rapid decline in the incidence of tuberculosis. Incidence of the disease in the spinning industry particularly was greatly reduced, and almost all cases were minor; workers were able to resume work after a short period of treatment. While emphasis was still placed on the issue of tuberculosis in the early postwar documentation, after 1955 it scarcely rated a mention. A major focus on worker health in the early postwar years was a result of the circumstances of poverty and poor health brought about by the war years. Over time the emphasis shifted from issues of ill health and the prevention of sickness, to the maintenance of worker well-being and to the

69 Shindo (1961) p. 80. 70 Seki (1956) pp. 166–67. For research on female textile operatives and tuberculosis, see Kagoyama (1970) and Hunter (1993b). 71 Hunter (1993b). Seki (1956) p. 167 noted that although the incidence did climb to a considerable extent between the 1920s and 1950s, this was believed to be due more to a thorough application of early diagnostic measures than to any actual increase in the disease.

Wages and welfare for women 185 ‘protection of motherhood’. Worker nutrition and diet were important issues for the industry during the immediate postwar period when food was scarce in Japan, particularly in urban areas. The government granted the textile industry extra food subsidies in order to entice rural workers to come back into the mainly urban textile factories.72 Early postwar research conducted by the industry looked into workers’ fatigue and calculated daily average calorie consumption needed by female mill workers in both summer and winter seasons. Food and nutrition were central components in the industry’s efforts to reduce illness, resulting in a yearly decline in the sickness rate during the early postwar years. A key development during the period under study was a move from the need to supplement daily subsistence and ensure adequate levels of food for workers, to one where taste and variety of food provided had to be considered alongside nutrition. Free or subsidised food for workers also came to be seen as a considerable expense. The industry was also concerned with adequate levels of rest for workers, complementary to nutrition. As noted, daily routines for young female workers living in dormitories were scheduled around production shifts. While these schedules had to co-ordinate work hours, education hours and regular meal times, they also had to consider the ‘non-awake’ hours. Dormitory self-government committees were responsible for considering dormitory rules, such as curfews and use of extra-curricular facilities, to ensure that the boarders had sufficient hours of sleep. A general aim was to provide a worker with 7–8 hours of sleep, and the results of the fieldwork survey indicate that the average amount of sleep was 6.7 hours a night. The main comments on levels of tiredness were that it was difficult to get up for the early shift and that girls were busy after work as well with study and with club activities. However, several mentioned that, as they were young at the time, they didn’t feel tired and in this respect sleep hours were sufficient.73 Recreation and physical exercise also became of increased importance during the postwar years (Plate 6.3). Zensen (1964) data recorded that 80 per cent of textile companies had an exercise area, 45.5 per cent had a gymnasium, and 72.7 per cent had an entertainment or recreation room.74 The industry trained ‘recreation leaders’ and provided not only a range of sporting and recreation events, but also courses aimed at group management or team-building exercises. These were not only for the purposes of physical exercise, but also aimed at facilitating and fostering the independence of young workers. Recreational facilities were usually managed by in-house clubs organised by workers, particularly within dormitories, and were subsidised to a large extent by management. The most common 72 SCAP (1946b) p. 173. 73 In the survey 38 of 45 (84.4 per cent) responded to questions about average sleep hours, with 39 per cent recording 6 hours of sleep, 37 per cent at 7 hours’ sleep, 16 per cent at more than 7 hours, and 8 per cent at less than 6 hours a night. 74 Zensen (1964) p. 51.

186 Wages and welfare for women

Plate 6.3 Women performing morning exercises in company grounds. Photo courtesy of To¯yo¯bo¯.

facilities were volleyball, tennis and table tennis areas. A range of recreational and sporting clubs and circles were established for young female workers around the school system and were similar to the kind of clubs run in regular senior high schools. To¯yo¯bo¯ commented that the most popular clubs were the film and cinema circle, the English language conversation circle and the volleyball club.75 Athletic meetings and sports competitions were often held in spring and autumn annually. Volleyball was a particularly popular sport for young female factory workers, as well as baseball. Sporting competitions, particularly among the Big Ten, were regularly held, and corporate female volley ball teams also competed in the national league. One of the Big Ten, Unichika, has had a particularly strong history in volleyball. Members of their female volleyball team competed in the Japan team at the 1964 Tokyo Olympics, taking the gold medal. In fact, Unichika and Big Ten companies often recruited girls specifically for their sporting skills.76 In addition to

75 To¯yo¯bo¯ (1970a) p. 12 . 76 Although employed as regular (factory) workers, in recent years these girls rarely actually work in production (based on conversation with manager during fieldwork visit to Unichika factory on 25 September 1998). Sports clubs and team-building exercises are still a major feature of corporations, especially large corporations in Japan today.

Wages and welfare for women 187 sporting events, recreational excursions and trips were common during these years. Zensen noted that 95.2 per cent of textile companies organised an annual recreational trip. This cost the textile industry on average ¥1,265 per person in 1964.77 Such trips around Japan, often to hot-springs (onsen), were significant events for young girls who had predominantly come from rural areas in Japan and had rarely experienced travel.78 Other facilities provided by the industry in addition to dormitory-centred or sporting facilities, particularly in the large-scale enterprises, included shops offering daily necessities and hair salons for female workers. These products and services were generally offered to workers at below-market prices. The industry commented in 1969 that the increase in married women would mean that the next generation of facilities would be day-care facilities and nurseries.79 Overall, the increasing importance of recreation in society in general, and the growing emphasis placed on individual worker well-being and standards of living, led to increased facilities of this nature being provided by industry.80 The well-being of young female workers in terms of their personal development was also a primary focus of management concern (see Chapter 4). This extended to preparation for their future life beyond the company, particularly their future roles within marriage. To¯yo¯bo¯, for example, even provided a marriage consultation service (kekkon so¯dan-sho) for young women. The company commented that because many of its young female workers came from distant rural areas and were employed within a workplace environment which was predominantly ‘female’, they were naturally concerned about how they would meet and find a future marriage partner. To¯yo¯bo¯ commented that information on this service and assurance to those worried about their ability to find a marriage partner by entering textile employment was included within recruitment policy.81 In this sense, company attention to worker well-being can be said to have sometimes extended beyond the physical, to the emotional.

77 Zensen (1964) p. 49. 78 Interview I. Many of the fieldwork survey respondents stated that recreation events and annual trips, particularly hiking and camping, were enjoyable memories of their dormitory life. Moreover, 4 of the 45 respondents cited the company’s club and sporting activities as direct reasons for their decision to enter employment, although sports was not a subject directly asked about in the survey. 79 It is not clear from the industry documentation to what level these type of facilities developed, however two of the fieldwork survey respondents said that the factory they worked for provided a nursery. 80 JSA (1969) pp. 42–43. 81 To¯yo¯bo¯ (1970a) p. 13. It should be noted that as more and more women entered paid employment during the postwar period, marriages between those meeting at the workplace (shokuba kekkon) as well as the role of workplace colleagues (particularly senior colleagues) as matchmaker (nako¯do) was not uncommon in Japan.

188 Wages and welfare for women Protective welfare for women Protective measures for female workers have played a large part in the history of women working in modern Japan. In the postwar period, protective legislation for women under the Labour Standards Law covered several issues: labour hours and holidays (Article 61); maternity leave (Article 65); childcare leave (Article 66); menstruation leave (Article 67); dangerous work (Article 63); and a ban on underground work (Article 64). Protective clauses relating specifically to the physiology of females, particularly menstruation, maternity and childcare leave, came under the title of ‘motherhood protection’ (bosei hogo). This does not mean that the measures applied only to women who were actually pregnant at the time. They applied to all women, based on the idea that all had the possibility of motherhood at some point in time. Zensen commented that because the majority of female textile workers were young, union movements for motherhood protection began in the early postwar years with more general protective welfare movements for young women as future mothers (sho¯rai no haha). These included early demands from 1946 for general improvements with regard to the nutrition, health and well-being of young females, particularly as dormitory boarders, and under the overall aim of ‘not allowing Joko¯ Aishi to reappear’. The union commented that in this sense the movement was for the protection of women (joshi hogo and fujin hogo) rather than just the protection of motherhood.82 However, with the specific protection clauses implemented under the Labour Standards Law in 1947, and following a visit to the MOL’s Women’s Bureau by Zensen female representatives in 1949 to discuss the legislation, moves to improve the implementation of these legal standards across the industry began in earnest.83 Several programmes were run by Zensen’s women’s division aimed at promoting protection specific to females across the textile industry during the period under study. From 1954 to 1959 research into protection, particularly menstruation leave, was incorporated among general demands for improved health and well-being for female workers, including the provision of nutrition consultations, health checks, research into the prevention of illnesses, improvements to food (particularly in winter) and implementation of rest areas in factories. From 1960 to 1970, union activities sought to further strengthen the overall health and well-being of female workers within motherhood protection aims, by focusing on raising women’s consciousness and awareness of health and protection issues. Zensen noted that young female workers were often put off by the image and terminology of motherhood protection (bosei hogo) as they thought it applied only to those who were pregnant or already had children, and the union was keen to 82 Zensen (1996) p. 128. 83 Zensen (1996) p. 129.

Wages and welfare for women 189 find alternative approaches which would educate and interest them in the range of health issues encompassed under ‘protection’.84 A major concern of Zensen during the 1960s was that the rate of women taking up menstruation leave was low, and it was often difficult to take such leave. A survey undertaken by the union in 1961–62 found that only 16.6 per cent said that it was easy to take menstruation leave, while 55.57 per cent said it was difficult to take.85 Many also responded that they did not want to take leave, with common reasons cited by young women in particular as ‘too embarrassing to take leave’ and that they ‘didn’t want to take it because others did not’. Responses from older women (particularly those in female leader positions) included requests for it to be easier to apply for leave, and to eliminate men from positions in control of the application process. However, the main problem was recorded to be economic, with many citing the fact that not only was it unpaid leave but also a shortage of workers often made it difficult for them to take time off. In many ways, the textile industry was examining this issue within a wider context during these years. From 1960, health issues relating to the increasing numbers of women in the workplace became prominent social issues in the context of high-speed economic growth and rapid technological developments. Zensen commented that it was concerned during these years not only about the effects of labour shortage and increased use of machinery on worker health, but also the growing difficulty for female workers to take not only menstruation leave but also their paid holiday allowances.86 Following its survey, Zensen began to push for unified policies relating to workers’ health across the industry from 1962. These included research into illnesses associated with worker fatigue, further improvements to company food and worker nutrition, recognition of LSL protective standards, education programmes for women and easing of the application process for menstruation leave, as well as a push for the implementation of paid menstruation leave.87 The union pushed for the latter under the reasoning that motherhood protection was a social issue and responsibility, and that female wages were much lower than male wages and also comprised fewer additional welfare payments.88 Between 1968 and 1970, Zensen continued to promote motherhood protection as an overall health issue, incorporating it 84 Zensen (1996) pp. 129–30. 85 Survey respondents were 17,000 women from across the various sectors of the textile industry. The low rate of taking menstruation leave was contrasted with survey results that reported that 47 per cent of respondents ‘suffered’ to varying degrees during menstruation and 16 per cent said they took some form of medication to relieve pain and discomfort: Zensen (1996) p. 132. 86 Zensen (1996) pp. 130–32. 87 Zensen (1996) p. 133. The union commented that at that time only a small proportion of textile companies offered paid menstruation leave, and this was only for regular (monthly salaried) and not for temporary (daily waged) workers. 88 Zensen (1996) p. 135.

190 Wages and welfare for women both within movements to improve workplace conditions and within aims for male–female equality, and under a central advertising slogan: ‘Do you know your motherhood protection rights? Are you protecting your health?’ (Shitte imasu ka bosei hogo no kenri, Mamotte imasu ka anata no kenko¯).89 In the 1970s, however, Zensen’s movement for motherhood protection issues took on renewed strength. As increasing numbers of older women entered employment, Zensen began to place increased emphasis on more direct maternity issues, including both legal provisions such as pre- and postchildbirth leave and childcare time, and non-legal issues such as the provision of workplace nurseries. This reflected national concerns of the time with regard to the welfare of working women, encompassed under the government’s passing of the Working Women’s Welfare Law in July 1972.90 Although Zensen recorded that this movement suffered setbacks during the 1973–74 oil shock and recession when many temporary female workers lost their jobs, the issue of working women, particularly older women, was their central concern from the 1970s onwards.91 In 1971 Zensen took the decision to make unified demands to management for the establishment of motherhood protection agreements across the industry, which included the extension of legal clauses as well as the establishment of non-legal provisions. These were gradually settled and established by textile companies during 1973 and 1974.92 Table 6.7 summarises the changes to motherhood protection provisions during that period. Overall, Zensen felt that these demands had been successful in establishing a detailed system of motherhood protection across the industry, particularly in light of the fact that the demands had been settled during a period of recession and rationalisation.93 Menstruation leave was an issue unique to Japan. Under the conditions of the law, menstruation leave could be demanded by ‘women for whom work during menstrual periods would be especially difficult’ and by ‘women whose duties were harmful to their physiology’.94 While some Japanese feminists argued that menstruation leave hindered the fight for equality in the workplace, others have argued that protective clauses were necessary and that better ‘protection’ for men needs to fall into line with that provided for women. Providing information to female workers in 1959, Zensen explained the provision of menstruation leave in Japan as being due to the lack of paid sick leave compared to other countries and to the poor work-

89 Zensen (1996) p. 137. 90 Zensen (1996) pp. 137–38. 91 The 1970s and 1980s became known as the women’s era (onna no jidai), which culminated in the Equal Employment Opportunity Law 1986 which contained various welfare provisions from the Working Women’s Welfare Law. 92 Zensen (1996) pp. 138–47; Nenju¯ no Bo¯sekigyo¯ (1974) p. 74. 93 Zensen (1996) p. 146. 94 Sugeno (1992) p. 313.

Wages and welfare for women 191 place conditions at the beginning of the postwar period when the legislation was introduced. It suggested that women everywhere sometimes had to take time off work during menstruation, but that it just was not explicitly recognised when better provision of sick leave covered such absence.95 The Ministry of Labour was very keen to observe the implementation and take-up of menstruation leave in its surveys on female textile workers, particularly in the later 1972 survey. It asked women to respond to a range of questions in its aim to determine trends. These included how many days on average were taken by those requesting leave, why and how often leave was taken, and differences in the rate of leave by industry sector, by job type and by worker age. It even asked workers about the regularity of their menstrual cycles.96 While the survey responses are of interest, such statistics establish very little about any impact that factory work may have had on the health of women. In many ways they simply demonstrate the intense level of interest taken in the well-being of women workers as well as the particular concern there was during these years for female physiology under the idea of protecting the functions of motherhood. However, Zensen was clearly also concerned at what it saw as the low rate of women taking up menstruation leave in the industry, compared to other industries at the time.97 It cited a 1960 Women’s Bureau (MOL) survey recording the rate of leave for female textile workers at only 3.7 per cent compared with an average of 34.4 per cent for all industries. Zensen noted this low rate, in contrast to the results of its own surveys, in which 90 per cent of female textile workers responded that menstrual leave ‘was necessary’ when asked. This led the union to express concern at the difficulties women faced in the industry when attempting to take menstrual leave, and to call for companies to make it easier and friendlier for women to do so. In addition to its formal request for full pay during menstruation leave (see Table 6.7), the union called for companies to consider several strategies. These included a call for firms to assign female staff as consultants and advisers on menstruation leave issues, the request that less authorisation by medical staff be required in order for women to take menstruation leave, the need for better education of female workers on their rights, the provision of rest rooms (spaces to lie down) within the workplace for women to use, and the facility for women to switch to lighter work during menstruation if required.98 The Japanese national health insurance scheme (kokumin kenko¯ hoken) system paid out for a woman during maternity (pre-birth and post-birth) leave where no collective workplace agreement was in place (see Table

95 Zensen (1959b) p. 198. In the revision of protective clauses during the implementation of the EEOL 1986 menstruation leave was incorporated into more general requested leave. 96 See Macnaughtan (2001) pp. 334–35. 97 Zensen (1962c) p. 34. 98 Zensen (1962c) p. 34.

Leave upon request for ‘women for whom work during menstrual periods would be especially difficult’; unpaid leave if no workplace agreement in place

Six weeks of pre-childbirth and six weeks of post-childbirth leave; unpaid by company if no workplace contract exists, but childbirth benefit payable under national health insurance

No legal provision until 1992 Child Care Leave Law

Thirty minutes’ break twice per day for up to one year after giving birth; unpaid if no workplace agreement in place

No legal provision

Menstruation leave

Maternity leave

Childcare leave

Childcare time

Morning sickness leave

Labour Standards Law 1947a

Table 6.7 Motherhood protection provisions, 1947–74

Total of 14 days’ leave; 100% pay

Forty-five minutes’ break twice per day, with ability to take total of ninety minutes at one time; 100% pay

Within three years after childbirth; three months at 60% pay

Eight weeks pre-birth and eight weeks post-birth leave; 60% pay

Two days’ leave per cycle; 100% pay

Zensen Demands 1971b

Up to 14 days’ leave; 60–100% pay

Failed as unified demand; remained at LSL standard (30 minutes twice per day) (mostly unpaid)

Within one year after childbirth; as paid leave in 60% companies

Eight weeks pre-birth and eight weeks post-birth leave; 60% pay (ten companies agreed 60–80% pay)

Two days leave (monthly); average of 60% of full pay

Demands Settlement 1973–74c

Employer obligation to transfer a pregnant woman to light work upon her request (including modification of working hours – not paid); general ‘ban on work harmful to maternal functions’

Change of work during pregnancy

Adjustment of working hours during pregnancy; up to 2 × 30 minutes per day modification allowance (i.e. reduction in hours) 100% pay

Once each 4 weeks at 4–7 months pregnancy; once each 2 weeks at 8–9 months; once a week at 9–10 months 100% pay

Zensen Demands 1971b

Remained at LSL standard (i.e. no detailed/fixed standards)

Once each 4 weeks at 4–7 months pregnancy; once each 2 weeks at 8–10 months; around 60% of companies agreed 100% pay

Demands Settlement 1973–74c

Source and notes a Sugeno (1992) pp. 307–14. b Zensen (1996) p. 140. c Nenju¯ no Bo¯sekigyo¯ (1974) p. 76; Zensen (1996) pp. 142–47. By November 1974 demands had been settled in 101 textile companies, with some variation. The table summarises the most common settlements (i.e. settlement in majority of companies). Not surprisingly, better settlements (longer time and more pay) were more common in medium- to large-scale enterprises.

Employers must ‘endeavour’ to give consideration so that women workers have necessary time off to receive health guidance and medical examinations during pregnancy; detailed under later Maternal and Child Health Law 1965 (not paid)

Hospital visit leave

Labour Standards Law 1947a

194 Wages and welfare for women 6.7).99 Agreed payment by companies above and in addition to the health insurance payments was found to be rare in the textile industry, a situation which was recorded as ‘not good’ (yoku nai) by Zensen in comparison to other industries. Citing a Women’s Bureau (MOL) survey on the topic, it noted that while 37.6 per cent of all industries were paying above the health insurance payments, only 7.2 per cent of the textile industry was.100 This was no doubt indicative of the fact that few married women were regular workers in the industry at this time. However, transition in the female workforce certainly prompted the industry to make its later push for paid maternity leave in the industry. Zensen also noted that one-off celebration payments upon the birth of a child were commonly paid to employees; however, these were viewed as having been token goodwill gestures only, rather than substantial maternity payments (they were also paid to male employees).101 The two types of leave for the purposes of childcare must be clearly differentiated (see Table 6.7). Childcare leave is for an intensive period of leave of up to one year following birth requested by a female employee. Childcare time was specifically leave in the form of work breaks to (breast)feed infants, when requested by nursing mothers. This was unpaid time unless a collective agreement existed in the company. The convenient time for such breaks was to be decided and worked out between the employee and the company. No records were available regarding the proportion of women taking up this entitlement.102 Companies were required to count this ‘on-leave period’ under motherhood protection legislation as periods when a worker was deemed to ‘report to work’, as the length of the working year (period of employment) affected length of service and pension systems. Zensen was certainly very concerned with the issues of motherhood protection and female worker well-being from the early 1960s onwards, but 99 The method of payment under this system consisted of several components – childbirth expenses (bunbenhi), birth allowance payments (shussan teatekin) and childcare allowance (ikuji teatekin) – and was calculated as follows: (1) for home birth: monthly salary × 50 per cent (bunbenhi) plus daily wage × 60 per cent × 84 days (shussan teatekin) plus ¥2000 (ikuji teatekin), or (2) for hospital birth: monthly salary × 50 per cent (bunbenhi) plus daily wage × 40 per cent × number of days in hospital plus daily wage × 60 per cent × remaining days at home (shussan teatekin) (number of days in hospital and number of days at home to total 84 days) plus ¥2,000 (ikuji teatekin): Zensen (1962c) p. 39. 100 Companies usually contributed into either a company-run health insurance scheme or into the national health insurance scheme for workers, so in this respect companies did contribute to maternity pay: Zensen (1962c) pp. 40–41. 101 Payments were usually higher for a first child and slightly lower for subsequent children. Average payments during the period were around ¥3,000–4,000 for the first child and ¥1,000–3,000 for subsequent children among the Big Ten and larger-scale companies, though were smaller (often around ¥500 per child) in small- to medium-scale companies: Zensen (1962c) pp. 54–63. 102 Zensen (1962d) pp. 18–27.

Wages and welfare for women 195 particularly in the 1970s. In its early surveys, it even reported data on the nature of births by female workers by different industry (e.g. proportion of natural, premature and difficult births), the breastfeeding ability of mothers, and data following the comparative progress of children born to female workers by industry.103 Zensen appeared keen to suggest that there were more birth difficulties for female textile workers compared to other industries, particularly in comparison to farming women and to women who worked within the household, although the statistics appear to be not that significantly different. However, the union may have felt that survey results of this nature offered grounds for educating employers on their social responsibilities to protect motherhood and urging them to implement better standards. While improvements clearly took place within the industry, with the implementation of better standards of motherhood protection particularly in the medium- to large-scale and Big Ten companies by the mid-1970s, in general, the textile industry appears not to have provided significant levels of ‘motherhood related’ welfare above that required by law during the period. The emphasis on non-legal welfare for females in the industry was much more on dormitory and education facilities, rather than maternity or childcare facilities. The interest in motherhood protection, particularly issues relating to maternity, grew later in the period as the industry recognised the increasing numbers of older women employees. The lesser emphasis placed on these issues relative to dormitories and education was no doubt due to the continued character of the industry as a heavy employer of young unmarried females, as well as a result of the general M-curve labour participation trend in Japan after the 1960s, when more older married women were employed within the economy, but typically those who had already dropped out of the labour market for a period of childbirth and infant rearing.

Evaluation of wages and welfare The analysis of welfare is important to an overall discussion of female wages, as it takes into account the various benefits in addition to their cash wages that female workers received by working in the industry, particularly the benefits of living in company dormitories. Indeed, it can be suggested that welfare acts to boost ‘real wages’ for females in the industry and provides a standard of living that cannot be assessed through an examination of wages alone. In the 1950s, several studies suggested that the provision of welfare for female workers by the industry had a significant effect on any evaluation of their wages.

103 Statistics from a March 1956 survey by Zensen compare the birth and post-birth circumstances for females across various industries: Zensen (1962c) pp. 64–66.

196 Wages and welfare for women Commenting on welfare provisions relative to wages, Seki (1956) noted that in 1952 the ratio of welfare expenses to cash wages was 11.5 per cent for manufacturing industries as a whole, but higher, at 16.8 per cent, for the cotton industry. He further noted that, in the case of cotton spinning companies, supplementary food payments by companies were of an amount almost identical with that of other non-legal welfare expenses. Overall, his study suggested that the welfare expenditure entailed in health services, factory schools, sports and recreational facilities in addition to food expenditure borne by the company amounted to approximately 20 per cent of the average pay in 1952.104 Looking at 1956 data, Shindo (1961) suggested that non-legal welfare expenses per worker per month amounted to more than ¥2,300 in the cotton spinning industry, exceeding the average of all manufacturing industries, which was ¥2,018. He also noted that the ratio of welfare-related expenses to wages stood at approximately 9 per cent in the manufacturing industry, but that the ratio was nearly twice as high in the cotton industry, at around 18 per cent.105 A 1953 survey by the Kansai Employers’ Association also showed that the ratio of non-legal welfare expenses borne by the cotton spinning industry amounted to around 20 per cent of the average wage, which was double that of other manufacturing industries. The Engel’s coefficient, including the deduction of food charges by the company, stood at 21 per cent, while the weight of net income expenditure on clothing and savings was high at 29 per cent and 28 per cent respectively.106 Such figures suggest that the industry’s welfare facilities were considerable relative to those of other industries in these early years. This analysis seeks to extend and examine this idea in order to determine the effects welfare had on the lives of female workers in terms of their average cost and standard of living during the postwar decades. Expenditure on welfare Expenditure on non-legal welfare for workers by the textile industry mainly included food, accommodation, transportation, non-legal insurance schemes (e.g. subsidies for mutual benefit society), medical facilities, education, sports and recreation expenditure. Welfare expenditure for female workers by the industry was mainly comprised of accommodation, food and education, and welfare expenditure by cotton companies in general was noted as being particularly high for those female workers residing in dormitories.107 An examination of the breakdown of welfare expenditure by the Big Ten cotton companies between 1955 and 1970 revealed that legal 104 Seki (1956) pp. 171–72. 105 Shindo (1961) pp. 95–96. 106 The Engel’s coefficient is the proportion of consumption expenditure spent on food: reported in To¯yo¯bo¯ (1955) pp. 5 and 30. 107 To¯yo¯bo¯ (1955) p. 29.

Wages and welfare for women 197 welfare accounted for around one-third of all welfare expenditure, while non-legal welfare facilities accounted for the largest share of welfare spending.108 With the exception of a significant increase in the proportion spent on the national welfare pension system, expenditure on legal welfare items can be said to have been fairly stable proportionally during this period. However, expenditure on non-legal welfare facilities showed variation. Expenditure on dormitories was the largest item of non-legal welfare expenditure, and increased over the period, comprising nearly one-quarter of all expenditure by 1967. Union figures for 1964 suggested that textile companies paid around 60 per cent of dormitory expenses (not including food), although this was much higher in large-scale and Big Ten corporations where the company bore the majority of expenditure. As noted earlier, towards the end of the period under study there was a decline in the number of young female workers in general within the industry, and a subsequent decline in the number of dormitory boarders. However, at the same time an increase in dormitory facilities and the growth of individual rooms and adult or apartment-style dormitories took place. Therefore, the trend toward fewer dormitory boarders did not simply imply less cost for the industry. Expenditure on food was in decline proportionally during the period. Food expenses included not only foodstuffs, but also labour, lighting and fuel costs for company dining halls (shokudo¯).109 Companies tended to pass on a greater share of food (i.e. meal) costs to employees over the period, which would account for its decline within company expenditure. At the beginning of the period, food for workers had been an important socioeconomic issue in Japan because of food shortages at the end of the Second World War. However with increased prosperity during the postwar period, this was a non-legal welfare provision that, while it remained, declined in relative importance and proportional expenditure. Expenditure on schooling showed an increase over the period, and represented nearly 10 per cent of all non-legal welfare expenditure by 1967. Like dormitories, education was a significant benefit for young female workers in the industry. MOL figures for 1972–73 indicated that the typical fees for commuting to a regular part-time senior high school included an annual application, entrance and equipment fees, as well as monthly lesson fees, comprising around ¥25,300 per student per year. These were generally all borne by the company.110 To¯yo¯bo¯ figures, however, indicated that total educational costs were much higher than this. In a 1975 company report it stated that the average educational costs borne by the company across all its factories amounted to ¥6,750 per student per month, giving an annual total of around ¥81,000 per student. These costs included the salaries paid to 108 For original data, see Macnaughtan (2001) pp. 340–41. 109 Zensen (1964) pp. 56–57. 110 MOL (1972) pp. 18–19.

198 Wages and welfare for women teachers at company run schools, which was noted to be the largest expense.111 To¯yo¯bo¯ also carried out two educational investment plans during the period, when it invested in improvements and upgrades to educational equipment. The first was between January 1970 and April 1972 at a cost of ¥58 million and the second was from May 1973 to October 1975 at a cost of ¥43 million.112 An analysis of the expenditure by the Big Ten cotton companies as a proportion of average monthly female wages in those companies indicates that non-legal welfare expenditure represented at least one-quarter of the average female wage between 1955 and 1970.113 While not every female employee would have benefited from the complete range of welfare expenditure, the majority of the welfare provisions would have been given to regular female employees (young full-time females). This therefore represented a significant benefit to workers relative to their cash wages, particularly given that wages were also rising significantly in real terms during these years. JSA documentation recorded that welfare expenditure by the Big Ten companies relative to labour costs (cash wages) was high compared with other manufacturing industries. It cited a 1970 survey by Nikkeiren which recorded average welfare expenditure by all manufacturing industries at 13.0 per cent of labour costs, and 17.6 per cent in the Big Ten.114 While average labour costs may have been comparatively lower in textiles, given the higher proportion of young female workers, the indication is that overall wage levels in textiles at this time can be thought of as being at least more in line with those in other industries if the substantial provision of welfare is taken into account. It must be asked what the incentive was for textile companies to spend and invest substantial amounts on welfare facilities for young female workers. The provision of dormitories goes back a long way in the industry’s history, and was presented as a sense of responsibility for the group predominating in the labour force – i.e. young females recruited from distant locations required accommodation within a ‘protective’ environment while working. However, the provision of dormitories and other welfare has been revealed by many who have researched the prewar period as being initiated by the

111 To¯yo¯bo¯ (1975a) pp. 15–16. It also commented that if the number of students per teacher fell below 22.5 students, then the school system was in the red financially. Presumably this was in terms of internal management budgets given that the system was company financed. 112 To¯yo¯bo¯ (1975a) pp. 26–27. These figures can be placed within some context by comparing them to the figures for average expenditure on education by the Big Ten. For example, in 1967 the average annual expenditure on education was c. ¥272 million (¥422 per student per month × 53,752 workers × 12 months). This makes the extra expenditure on equipment by To¯yo¯bo¯ of ¥101 million combined (43 + 58) roughly equivalent to about 40 per cent of one year’s average expenditure on schooling at that time. 113 For details of calculation see Macnaughtan (2001) pp. 339–44. 114 JSA (1970) p. 56.

Wages and welfare for women 199 industry in attempts to overcome labour shortage and high turnover through coercive means. Welfare ideologies of familism and paternalism were more rhetoric than reality until the interwar and postwar periods. The provision of better welfare for workers was encouraged by the Occupational authorities as efforts to kick-start the industry and fill its labour requirements in the early postwar years. However, the substantial extension and expansion of welfare provision and expenditure in the postwar period was both a strong recruitment mechanism, and a reflection of general socio-economic changes taking place in Japan during this period. The provision of non-legal welfare facilities was aimed at the shaping of a ‘full daily life’, providing an ‘internal society’ for workers that included the provision of facilities required not only for working life but also for non-working life, thereby improving overall standards of living. Cost of living for women One means of welfare assessment is an examination of the breakdown of female workers’ cost of living indices – i.e. what proportion of their monthly take-home pay they spent, and what items they spent it on. This analysis of their cost of living, and its relation to both welfare and wages, helps to provide an overall picture of the standard of living for females working in the industry during this period, and enables a further evaluation of real wages earned by female cotton workers. As already shown, the real wages of female textile operatives considerably increased during the period relative to the consumer price index. As well as annual wage rises, the payment of annual bonuses and the provision of retirement allowances also contributed to a raised wage level overall. However, the provision of welfare benefits by the textile industry also had an important effect on the ‘real earnings’ of female workers. There appears to be no cost-of-living data for female workers available for the postwar period under study, with the exception of a survey done by To¯yo¯bo¯ in 1955. The results of this survey as well as the results of my fieldwork survey are presented in Table 6.8. The results of the fieldwork survey cannot be considered adequate for a full analysis.115 However, given the lack of information available on the spending of wages by female textile workers, it can be used in combination with the results of the To¯yo¯bo¯ analysis to make various suggestions. The effects of welfare provision on the average expenditure by female workers on items such as accommodation, education, food, recreation and savings (shown in Table 6.8) provide an indication of their income power and financial well-being during this period. Table 6.8 indicates an important benefit that females in the industry had 115 Survey response is only 23 respondents out of 45 (51 per cent) and it did not pinpoint exact years. However the results can be taken as an indication of spending breakdown for the 1960s in particular.

7.3 28.9 4.9 2.0 3.2 4.6 3.6 5.1 2.3 17.3 10.8

90.0

Food Clothing Cultural Recreation Transport Social Furniture and utensils Daily necessities Miscellaneous Savings Remittances

Total Source: fieldwork survey results.

103.4

3.9 28.7 6.1

Other Savings Remittance Total

4.3 3.3 19.8 13.0 12.6 10.0 1.7

(Average)4 Household Accommodation Food Clothing Education Recreation Transportation

Fieldwork survey3

18.0 31.4 23.3

25.0 8.6 22.8 14.3 22.3 13.5 12.7

(Respondent based)5

21.7 91.3 26.1

17.4 39.1 87.0 91.3 56.5 73.9 13.0

Response rate6

Notes 1 Survey based on 3,341 female workers in To¯yo¯bo¯ dormitories. 2 Based on income for July 1955 (net income after deduction of social insurance fees, income tax, meal charges and membership fees). 3 Only 23 out of 45 completed surveys responded correctly to cost of living breakdown question. 4 Average = average of all 23 respondents. 5 Respondent based = average of only those who responded to each item (e.g. average savings rate for those who responded that they did save). 6 Response rate (%) = (no. of those who responded to each item/23 total respondents) * 100.

Source: To¯yo¯bo¯ (1955) p. 6.

NB: Survey includes average of 10% of income carried over to next month (hence total = 90%).

(% expenditure)2

(Item)

To¯yo¯bo¯ 1955 survey

Table 6.8 Cost of living surveys for female textile workers (monthly expenditure: %)1

Wages and welfare for women 201 from living in company dormitories. The proportion of their monthly expenditure spent on accommodation was very low compared with other expenditure, and implies the freeing up of income for other types of spending. The To¯yo¯bo¯ 1955 survey did not even include accommodation as an expenditure item for its survey of females in dormitories. The fieldwork survey showed that average expenditure was comparatively low for the categories of accommodation and household, while transportation was also very low, indicating that accommodation in dormitories also kept down commuting and travel expenses. Dormitory facilities were primarily operated at the company’s expense, although the practice of obtaining some small contribution from workers was implemented during the period, which would account for the small level of expenditure indicated by the fieldwork survey. Figures for To¯yo¯bo¯ in 1961 noted that small dormitory fees of around ¥60 per month were charged to boarders, while the MOL 1972 survey recorded that 94 per cent of companies with dormitories were deducting a contributory monthly dormitory fee ranging from ¥4,000 to 7,000 per month.116 With the move to the provision of adult apartments for young females (seniors), Zensen recorded that female residents in these paid ¥300 each per month for a 2DK apartment (usually four sharing), compared to the monthly rent of ¥15,000 of a 2DK apartment on the private market. They also contributed to the cost of utilities, although these were noted to be low, and they usually still had two meals per day within the company facilities.117 Overall, therefore, the provision of accommodation, the cost of which was borne primarily by the industry, had the effect of keeping expenditure on rent – usually a staple item of expenditure – very low for young females during these years. Education was a significant item of expenditure according to the results of the fieldwork survey. Again, education was not included in the To¯yo¯bo¯ 1955 survey, which suggests the increased importance of education during the period, and with it some commitment and contribution towards educational expenses by young females themselves. While the majority of education and administration fees were borne by the company, contributions from female students were typically for materials and equipment, including textbooks, clothing, school trips and equipment for lessons.118 To¯yo¯bo¯ documentation indicates that it charged its students a nominal monthly fee towards lessons

116 To¯yo¯bo¯ (1961a); MOL (1972) p. 45. 117 Zensen (1996) p. 114. 118 A manager at Unichika confirmed that on average 10–20 per cent of all educational expenditure (including fees, uniforms, texts etc.) was passed on to young female workers during the period under study. The company would loan a lump sum to cover these expenses and the girls would pay it back gradually from their monthly salaries over the course of the year. The reasoning behind this was that if the company paid all educational expenses then the girls would not study as hard. If, on the other hand, they had to contribute towards their education and pay back their educational loans then they would study and work hard to do so (Interview M).

202 Wages and welfare for women (jugyo¯ryo¯), and that this was ¥100 in 1956, ¥300 in 1970 and ¥500 from 1973.119 The MOL 1972 survey suggested that educational costs amounted to the following average expenses for female workers attending four years of part-time senior high school at this time: first year – ¥930 per month (¥11,159 per annum); second year – ¥1,006 p.m. (¥12,070 p.a.); third year – ¥891 p.m. (¥10,681 p.a.); and fourth year – ¥1,383 p.m. (¥16,597 p.a.).120 These figures can be considered as minimal amounts relative to the average wages of the time. The minimum starting salary for 15-year-old female junior high school graduates in the Big Ten companies for 1972 was ¥35,000 per month (¥41,000 for 18-year-olds with three years of service).121 If average monthly educational costs are taken as around ¥1,000, this represented around only 2.6 per cent of the average wage (not including extra payments and bonuses), although the fieldwork survey indicated that total expenditure on education was at higher levels than this. A survey conducted by Zensen in 1962 on education in company schools asked pupils what they thought of the educational expenses (gakushu¯-hi) that they paid – 42.8 per cent answered that they thought they were average, while 35.9 per cent thought they were expensive.122 In the same survey, average expenses (including materials and textbooks) were ¥9,299 for the first year, ¥14,828 for the second year and ¥27,907 for the third year, indicating levels higher than those recorded a decade later by the MOL.123 First year pupils were also asked in the survey whether they received any financial assistance towards their educational expenses from their families, and an average 81 per cent of them responded that they did not.124 On the one hand, this raises the issue that, unlike their non-working peers studying in regular senior high schools and funded by their parents, the majority of young female textile operatives had to contribute to their own educational expenses. On the other, it points to a significant fact. Young females working in the textile industry were receiving heavily subsidised education. This is highlighted by a comparison of the average contribution towards senior high school educational expenses paid by young female textile workers, with those borne by parents supporting children in full-time senior high school education. Zensen figures suggested that females in textiles paid an average ¥17,345 per year in 1962, while MOL 1972 figures suggested an average of ¥12,627 per year.125 By way of comparison, average figures for

119 120 121 122 123

To¯yo¯bo¯ (1975a) pp. 26–27. MOL (1972) p. 19. JSA (1979a) p. 836. Zensen (1962a) p. 41. Zensen (1962a) p. 43. Figures are average of all six factories surveyed for a standard three years of schooling. 124 Zensen (1962a) p. 44. 125 These figures are calculated by adding together the per annum expenses reportedly paid by female workers in the two surveys, then dividing by the number of years of education.

Wages and welfare for women 203 parents’ liability of educational expenses per senior high school student were ¥37,077 in 1962 and ¥73,155 in 1972.126 This indicates that females attending school while working for the industry received a senior high school level of education at well under half that of the average cost of such education, and in many cases even as low as under 20 per cent of the cost.127 Average food expenditure was indicated to be higher in the fieldwork survey than that recorded in the earlier To¯yo¯bo¯ survey. Although meal charges were already deducted from the To¯yo¯bo¯ survey, the difference between the two surveys points to the trend noted in industry documentation, that food was an expense that was gradually rationalised as a welfare benefit and increasingly passed on to workers as a direct cost. In contrast to expenditure on ‘staple items’ seen as clearly necessary to daily life, i.e. mainly accommodation and food,128 is expenditure on ‘luxury items’, particularly recreation, leisure and savings. Expenditure on luxury items in the 1955 survey – cultural, recreation, social, miscellaneous, savings and remittances – amounted to 41.9 per cent of all expenditure, while similar expenditure in the fieldwork survey – education, recreation, other, savings and remittances – amounted to 61.3 per cent of total expenditure. These levels would indicate not only the ability of young females in the industry to spend on leisure and to save, but also an increased power to do so during the period. Savings are an important indication of how well off a worker is. The savings rates are significant in both surveys presented in Table 6.8, although the proportion of remittances appears to have declined during the period. In general, the figures indicate that on average females saved around 28–30 per cent of their income during the period under study, and the fieldwork interview indicates that the proportion of females who could and did save was very high at over 90 per cent (of respondents). The majority of them (89 per cent) also indicated that they participated in company savings plans. The decline in remittances vis-à-vis personal savings during this period was significant, particularly in light of the situation in the prewar period when rural household incomes relied on regular remittances from daughters working in textile factories. By the MOL 1952 survey, however, it was found

126 Data taken from Nihon To¯kei Nenkan for each year. Data are based on a survey of parents’ educational expenses at public schools. 127 While it is clearly difficult to pinpoint exact figures given the limited data available, it is evident that they received subsidised education. Disparity in the figures no doubt also reflects differences across the scale of enterprises within textiles. Large-scale enterprises were able to make higher contributions towards education than small-scale enterprises. 128 The costs of ‘furniture and utensils’ and ‘daily necessaries’ in the 1955 survey can also be thought of as staple items similar to accommodation and food.

204 Wages and welfare for women that those contributing to the family budget were few at only 4 per cent. Although the motives for some working were cited as ‘to help the family’, this was felt to be overwhelmingly for the purpose of contributing to and saving for a ‘wedding trousseau’ and ‘marriage’. Based on this, those sending regular monthly remittances back to their families constituted 45 per cent, while the proportion who didn’t send any money back at all was only 12 per cent.129 A survey by To¯yo¯bo¯ of the savings aims of females recorded that: 39.4 per cent were saving for marriage, 32 per cent had not decided specifically what their savings were for, 25 per cent were saving for their own future, while 1.2 per cent were saving to help the family and another 2.4 per cent for other reasons.130 Kobayashi (1963) commented that the survey response indicated that a systematic savings plan of some form or another existed for the income that young females were earning for themselves, and the plans for these savings were overwhelmingly in preparation for their own future, although the course of this ‘future life’ was not yet firmly decided upon.131 As Chapter 5 noted, the ‘practical education’ of young females that took place, usually as part of their management within the dormitory and education systems, included lectures on household economics, budgeting and saving. To¯yo¯bo¯ commented that they gave female junior high school leavers a ‘model plan’, which had as its goal a total savings amount of ¥600,000 over the course of four years while working and studying for their senior high school qualification.132 It is apparent that the proportion of female wages spent on specific items is important, as it provides a general picture of how a female worker spent her wages during this period. It also indicates what items a worker had the power to spend available income on. The provision of welfare in the form of dormitories and subsidised food meant that the expenditure on staple items was held down, while the proportion of wages available to be assigned to more luxury items and in particular savings was high. Moreover, the opportunity to receive education at a nominal cost was also significant during the period. The prewar situation when young girls were employed as an extension of household labour had gone by the mid-1950s, with fewer contributing to household budgets, more working for themselves and many attracted by the opportunities to work and study at the same time. The results of the fieldwork study indicated that 20 per cent saved with their family in mind, 40 per cent for their marriage, and 48 per cent for their own future, and the majority of women felt that their wages gave them the ability

129 130 131 132

MOL (1952) p. 3. Kobayashi (1963) p. 143. Kobayashi (1963) p. 144. To¯yo¯bo¯ (1970a) p. 12. Taking the average total nominal wage for 1974 of ¥72,457 from Table 6.1, a figure of ¥600,000 would have been equivalent to around 8.3 months worth of wages.

Wages and welfare for women 205 133

to do so. The provision of welfare facilities for young female workers during these years clearly provided an economic benefit above and beyond cash wages. However, as the following two quotes reveal, the various facilities also enhanced their personal development as well as working lives. I joined this factory when I was 15 years old. I learned how to cook, how to be independent, and I graduated from correspondence high school. I got married to a work colleague and then lived in a company house. I had a baby but was able to work again after my child was one year of age because there was a nursery in the factory. (Fieldwork survey respondent no. 27; employed 1968–, age 15–) My family relied upon my mother’s income. I had two siblings and we needed money so I worked to help out my mother. I think I was very lucky. The environment overtook the hardships of the job. I learned a lot in the company school and my education was paid for. We were all studying hard in those days and the teacher was good and helpful. If I was reading a book the teacher would advise me on what would be good to read next and lend me books. My room leader was really nice and we would stay up late on Saturday nights cooking and chatting together. All of these people were taking care of us as if they were our mothers. (Fieldwork survey respondent no. 43; employed 1962–70, age 15–23) In the postwar period, the prewar theory that young girls from rural areas could be easily recruited and housed in company run dormitories under conditions dictated by employers, and taken advantage of as cheap and docile labour under the pretext of providing them with homelike facilities is less valid. Reports indicate that by the 1960s the industry was aware that the democratic education system had elevated and changed the awareness of Japanese females. The industry commented that, while there were still some females entering employment in the 1950s because of poor household finances, by the 1960s socio-economic change meant that the main aim of females seeking employment was to gain their own income, as well as develop their self-growth and further their education. Furthermore, the rises in base wages, the improvements to existing welfare and the investment in a comprehensive education system for young females during these

133 Around 95 per cent of all survey respondents indicated that they saved. Saving for ‘family’ may often have indicated for their own (future) family after marriage in many of the cases, as this was not differentiated in the survey. The results include overlaps, i.e. those who responded that they saved with more than one aim in mind. Respondents were also asked whether they thought their wage level was low, average or high: 36 per cent of those who answered said they thought it was low, 55 per cent thought it was average and 9 per cent thought it was high.

206 Wages and welfare for women decades meant that this group of workers could no longer be considered as a ‘cheap’ source of labour by the textile industry. Welfare for older women Like all the issues relating to older female workers examined in this study, it is difficult to gather information on welfare for older women. As with many of the labour systems in practice during these years, most welfare facilities set up in the textile industry were specifically aimed at young regular female workers, particularly dormitories, education and recreation. Company housing and loan schemes were for married male employees and their families. As older women tended to be non-regular workers, married and commuters, the majority of the welfare facilities were not generally applicable and suitable for them. However, as with young female workers, basic wage data may not provide a complete picture of the economic wellbeing of older female workers. A 1962 Zensen survey of the situation of temporary workers across the textile industry noted that many of the working (legal) regulations for regular (honko¯) workers did apply to temporary workers in the majority of cases. These included working hours, holidays, overtime pay, social security and access to some welfare equipment (such as medical or recreation facilities).134 These were the same for both regular and temporary workers as they were either regulated by law or by Zensen itself. However, the report noted that there were no regulations for wage levels, special payments, bonus payments and company retirement allowances for temporary workers. This created a significant difference in the wages and working conditions between regular and temporary workers. Furthermore, the union reported that the majority of temporary workers across the industry were not entitled to receive transportation fees and housing support fees. It is evident, therefore, that the main welfare facilities for female workers in the textile industry – dormitories and education – were established predominantly for young regular female workers. However, the main welfare systems in operation did sometimes extend to include older temporary female workers, particularly in large-scale companies. Those females who were seasonal workers under the Hokkaido¯ Seasonal Plan, for example, were relocated from rural regions in Hokkaido¯ to factory locations and did live temporarily in the dormitories alongside regular female workers, thereby benefiting from those facilities. They also received transportation (relocation) payments. Even temporary daily workers, such as the rural housewives described in Chapter 3, received small benefits such as free transportation by bus to and from the factories and a meal in the factory dining hall.

134 Zensen (1962b) p. 13.

Wages and welfare for women 207 During the period, however, as the problems of acquiring young female labour reduced labour supply and caused a rise in the employment of older workers and commuting workers, it was recognised that welfare for older women also required attention. Moreover, the increase in labour costs during these years also meant that companies had to adapt the welfare system in operation.135 This involved both further investment in the system as well as rationalisation of some of the facilities to meet the socio-economic changes of the time. In many companies regular temporary workers, particularly part-time workers (pa¯to), were extended all the rights of regular workers and Zensen was one of the first unions in Japan to unionise parttime workers in the early 1970s. As early as 1959, Zensen commented that a survey carried out on female health showed that working mothers in particular were tired and overworked in the workplace. Zensen called for moves to provide flexible and reduced hours and at the same time called for more temporary workers to be hired to overcome the labour shortage, reduce worker tiredness, and give workers in general the opportunity to take more holidays and breaks. It also commented that extra ‘service jobs’ such as serving tea to male workers impeded the work of females and urged this practice to stop.136 General interest in the health and welfare of older women increased during the period, particularly from the late 1960s. The type of welfare programmes benefiting older women versus younger women were clearly different in terms of worker needs. The weaving factory case study in the MOL 1970 survey of older women briefly looked at welfare facilities, and provides an example of how welfare needs were assessed for older female workers. The factory stated that its main welfare facilities for older women were a shop, a bath-house and a baby-feeding room. The shop was provided so that housewives (older women employees) could easily pick up daily necessities in their break times to take home with them. The bath had been installed so that women could bathe after their shift before going home; however, it was noted that many of the women actually brought their children along with them to use the bathing facilities. The baby-feeding room was 4.5 tatami mats in size and allowed women to take up their 30minute nursing breaks twice a day. The report stated that as the factory was now employing older women as regular employees, they were eligible to join the union after two months of employment, with a retirement age set at 48 years. It was possible, in some cases, for women to work for a further two years until the age of 50, although the retirement allowance was stopped at age 48 and the additional wages in those two years were not included in the

135 To¯yo¯bo¯ (1986) p. 407. 136 Zensen (1959b) p. 198.

208 Wages and welfare for women final retirement calculation.137 It was also reported that the firm was planning to provide a day nursery room by spring of 1970, and was thinking about whether that might be extended to provide a proper nursery or kindergarten operating around the two-shift system and providing support for working mothers.138 Although there is little documentation on factoryrun nurseries, it does seem evident from fieldwork interviews that many larger-scale factories did actually provide day nurseries – a substantial benefit for working mothers in the industry. In practice, there remained clear differences in the effects of welfare (and wages) on older female workers relative to younger workers, particularly those living in dormitories and receiving education. However, the socioeconomic conditions of the period should not be overlooked in this evaluation. The position of a married women within the idea of the ‘family wage’ was one primarily of financial dependency, and her main role was seen as within the household, regardless of whether she worked or not. While an older married woman might earn ‘supplementary’ income for herself or the household, the wage and welfare systems developed in Japan during this time meant that in theory she was covered by her husband’s income as the family head, which included full welfare benefits. Furthermore, clauses in the national tax and pension systems placed limits on the amount a married woman could earn over and above her husband’s income before incurring strong tax penalties. As the labour structure over these years changed to include more married and older women, their reasons for wanting to work and gain income no doubt changed for this population of women. However, socio-economic structures clearly served to keep their wages and welfare benefits low.

137 MOL (1970) pp. 80–81. At the time of the survey the average length of service for all older women was recorded as 13 years 3 months. This indicates (as noted in Chapter 1) that in many cases women worked for extensive periods, not always fully represented by industry length of service statistics. 138 MOL (1970) p. 79.

7

Japanese female textile workers in context

The textile industry as an employer of women workers What is striking from this study is that it shows the tremendous transition taking place in Japan from industrialisation and ‘catch-up’ to the problems of an industrialised country. At a macro-level, significant and rapid socioeconomic transition was taking place. The economy moved away from agriculture to manufacturing and tertiary industries, technology advanced rapidly, education for young people increased dramatically, urbanisation spread, and high economic growth was achieved. This study of employment in the textile industry has allowed an examination of how these socioeconomic changes affected employers and workers at an industry and microlevel, and how these changes affected the role of Japanese women in the economy specifically. It is evident from my research that the industry made substantial efforts to improve employment conditions during these decades. While improvements to labour conditions were prompted by the implementation of national legislative reform during the early postwar years – including regulation of work hours, controls on recruitment, standards for the workplace and dormitories, and the specific protection of women workers (motherhood) – it is also clear that the industry did much to provide an improved environment for textile workers over and beyond that required by law. The provision of a full non-working environment and opportunities for educational and personal development for young female workers was particularly notable. The rapid increase in socio-economic standards of living was captured by the continual rise in wages above CPI levels, the increase in bonus payments, employer investment in welfare facilities and the attention paid to labour management strategies. However, the transition taking place during these years was perhaps best epitomised by features such as the shift in industry medical facilities away from the eradication of disease to regular health checks for workers, the trend in the provision of subsidised food for workers away from subsistence and health to issues of taste and variety, and the move away from young girls providing financial support to their families to having the power to spend more on luxury items and save for themselves.

210 Japanese female textile workers in context The textile industry was clearly an industry that had a labour history. Its evolution in the prewar period generated structural rigidities and a degree of path dependence. For postwar textile employers, the cost of change was often deemed to be greater than sticking to what the industry had already established. However, perceptions of the prewar labour force also encouraged the industry to seek new ways forward and to endeavour to dispel the lingering Joko¯ Aishi image. Moreover, as a traditional employer of large numbers of women and among the first industries to utilise females within the paid economy, its response to new developments in the postwar period was in many ways crucial. The industry was greatly challenged by an increasingly limited supply of its key workforce of young females, an increasingly competitive environment of demand for female workers in general, and the need to adapt its existing employment structures away from young females alone to include older working women. The response of a key employer at this turning point in women’s history in Japan may therefore be seen as a crucial factor in prescribing subsequent economic roles for women. The activities of the textile industry during these years were not only a response to a new socio-economic environment and legal framework, but also a proactive approach to a continuing discourse on the role women were to play in the paid labour force. In fact, it is also fair to say that the new legal environment within which the industry had to operate – particularly the implementation of recruitment controls and protective legislation for women – was to a large extent set in place precisely because of the role the industry had taken in defining prewar employment for females. The ways in which women were employed in this early industry set precedents and patterns for their subsequent employment. The industry’s search for ways to effectively utilise young women for a maximum number of years before they acquired domestic responsibilities upon marriage, as well as its proactive utilisation of older married women within a widening range of temporary and supplementary employment statuses, has been highlighted in this study. Its continual discourse on the responsibility of industry to guide the social development of young females and prepare them for their future nonworking roles as wives and mothers, as well its ongoing evaluation of the ability of older women to commit to regular paid employment that would be compatible with socially defined gender roles, were also evident. In many ways, activities such as these led to mainstay features of the Japanese female labour market, such as young ‘office ladies’ (OLs), older pa¯to and the arrival of the age of housewife labour. The failure of the ‘test factories’ in the 1960s, which aimed to utilise older women as a reliable core workforce, suggested that both employers and Japanese society in general had an opportunity to address the sex-segmentation existent in the labour market and in the home. However, textile employers did not really come up with an answer on how to reconcile full-time employment with married domestic life for older women. This ‘failure’ to do so led to entrenchment and institutionalisation of a gender-defined employment system. In this sense, it could be suggested that

Japanese female textile workers in context 211 there was a missed opportunity or turning point for Japanese women’s employment at this time in history. The industry continued to set a blueprint for the employment of women within industry in general in Japan, and established employment structures for Japanese women that were constantly defined by their being female. Japanese women were, however, no doubt also better able to take up paid work because of the leading role played by earlier textile operatives. While increased education for women and increased economic demand for women clearly played a role, the history and collective experience of females working in the textile industry gradually impacted on social opinion and attitudes towards working women. The activities and activism of textile women also made the way easier for each following generation of female cohorts entering the labour market. In this sense, the development of the female labour market in textiles and beyond was not simply a question of oppression or economic needs, as interaction between the force of gender construction, often at a state level, and the grassroots achievements of working women has also played a vital role. Overall, the impact of the textile industry as an employer of females within Japanese economic history was pivotal. Just as the pattern for male employment within the industrial labour market was being defined to a large extent by the steel industry, so too was the pattern for female employment being shaped by the textile industry. Other industries looked to the textile industry as the model for establishing or diversifying females within their labour force. When the electronics industry began to employ women on their production lines from the late 1950s onwards, they competed with the textile industry for labour and sent recruitment staff to remote areas and relocated young females into factory dormitories along very similar lines to operations in textiles.1 While textiles was not always as representative of women’s work in Japanese industry to the degree that the example of electronics indicates, it did however embody issues relating to women’s work, and to employment in general, that were issues for the economy as a whole in this very rapidly changing environment. The impact and influence of the textile industry therefore extends beyond its internal employment structures, and can be located into a broader context of industrial relations and to developments in other industries.

The impact of textiles on the national labour market The application of this research on one industry to the national context should not be overlooked. Many of the employment structures and trends found in the textile industry during these years, and indeed often initiated by that industry, became standard patterns in other industries. The methods

1 Partner (1999) pp. 208–09.

212 Japanese female textile workers in context implemented in the recruitment of young labour that developed during these years remain in force in Japan today. The annual spring employment of large numbers of young people upon graduation, with direct recruitment networks between companies and educational institutions and the strict monitoring of the system by the Ministry of Labour, became an established feature during the decades of postwar high-speed economic growth. As noted in Chapter 3, the textile industry was a large player in the evolution of this system and was a forerunner in creating a system whereby young labour was recruited en masse (shu¯dan shu¯gyo¯) and relocated into industry. The industry was also influential in establishing a discourse on the economic value of women’s work relative to that of male workers. In many ways it capitalised on the fact that its core labour force was female. It established a ‘female’ wage structure and system that was based on the notion that females did not require wage levels as high as males because they were not the heads of household, despite the fact that many did make significant financial contributions to the household, particularly in the prewar and early postwar years. Textile labour unions were quick to institutionalise wage systems in the postwar years that paid additional household and dependent allowances to male workers. They were also influential in promoting the ideal retirement of women upon marriage through the promotion and payment of a marriage pension (kekkon taishoku). The industry can also be said to have established precedents for the ways in which women were employed alongside men, particularly in terms of the implementation of technology. Indeed, it was initially because the industry and technology allowed for and suited the ‘nimble fingers’ of women that they were employed in such vast numbers. Male workers generally took responsibility for more technical jobs, and there was a clear system of male–female job differentiation employed within the industry. The textile industry, and indeed Japan, was not alone in this. Advances in technology have often meant that women have been ‘allowed’ to advance into previously male domains of employment, while the mass spread and simplification of technology have often been deemed to require nimble fingers – from spinning and weaving factory girls, to transistor girls to female typing pools. The wider implementation of social insurance schemes following the war and moves to increase wage levels and achieve rising standards of living within a booming economy were central features of these postwar decades. The study of the textile industry’s response to this socio-economic climate revealed not only moves for better wages, but also the increased provision of non-working welfare benefits and facilities for workers. In fact, the early implementation of welfare for young females in the textile industry set a precedent for how the nature of welfare facilities would develop across many industries in the postwar period. The provision of dormitories for young workers, company housing for married male employees, company dining facilities, and the extension of a range of company-run recreational

Japanese female textile workers in context 213 activities first emerged in the textile industry, but spread to become commonplace in other industries.2 In many ways this provision of a complete and full non-working environment for workers (and their families) was behind the extension of a corporate family environment around which the postwar middle-class male worker or ‘salary man’ ideal developed. The early implementation of welfare and labour management strategies for female workers in the textile industry therefore influenced not only how the employment of women would continue and take shape into the postwar years but also how the employment of men would broadly be defined. This leads to a significant point. In many ways, the male lifetime employment system that developed during the postwar period of economic growth in Japan depended on non-lifetime employment for women and the flexibility of older women. The utilisation of women in the paid labour market, particularly in non-permanent employment, often reflects transition in an economy and/or specific sectors of industry at any one time. Female labour, particularly older females, were increasingly required by Japanese industry during these years because rapid growth was placing pressure on the supplies of regular and young labour. Moreover, females were vital for a period when a restructuring of the workforce and reorientation of sectors of the economy was required. In terms of textiles, technological innovation during this period was going some way toward reducing the need for large numbers of workers, but more importantly, the decline and restructuring of the industry in terms of its importance within the national economy perhaps meant that textile employers realised that they needed large numbers of labour only for x number of years. From this perspective the flexibility of young girls and older women who could work ‘temporarily’ for indefinite periods was adequate and indeed proactive. The successful restructuring of the Japanese economy during these years and the smooth phasing out of older industries and sudden emphasis on new industries like electronics was vital, and was no doubt one reason why flexible female labour was so crucial to Japanese industry. This brings us to the crucial question of how to evaluate the impact and contribution of women to the economic growth that took place during these years. While it could be unwise to overemphasise and over-indulge the image of the young female worker and the older working mother shaping the ‘economic miracle’ with their nimble fingers, it would be equally unwise to overlook their contribution. The place of women within Japan’s postwar economic development and employment practices was ignored or discounted by traditional economic history for many years. This study of their employment within the textile industry during these years revealed that their contribution and participation within the labour market was varied, diverse and often hidden, but indeed deserving of recognition. This

2 Gordon (1998) p. 96.

214 Japanese female textile workers in context book has argued that the shaping and evolution of a postwar employment system for women had historical significance in the textile industry as a prominent employer of female workers in Japan. Indeed the continuing changes in the social and economic environment following the years studied – the transition to a low growth-rate economy, the growth of the service industry, the shift to a higher education society, the ageing of the population, the micro-electronics revolution, and the increase in female part-time workers – has served only to further entrench the nature of the roles of women within industry that have been highlighted in this study. Indeed, as Japan looks to its economic future in the twenty-first century, the important history and contribution of its working women offer rich lessons in the ongoing (re)evaluation of Japanese industry, employment, institutions and socio-demographic trends.

The everyday lives of female textile operatives This economic history of female textile workers has presented in detail not only the contribution of women to the economy as industrial workers, but also the way in which economic roles for women were constantly redefined at the industry level within contradictory ideals of women belonging to the domestic sphere. As with the prewar research, if female textile workers are portrayed only from a traditional perspective as ‘victims’ of industrial development, then they are also seen as having been primarily passive recipients of economic demand and social change. This overlooks the fact they were active participants. Their interaction within a changing sexual division of labour and new legislative framework, as well as their strength as workers, provided a backbone for industrial progress and shaped Japan’s modern economic history. The case of the female textile workers not only highlights a consistent implementation and redefining of gender roles, but also illustrates how women, as a dynamic force, were able to gain a degree of strength and consciousness as workers within the often seemingly confining and contradictory nature of their employment and social status. The forces that determined why women entered textile employment during these years were not just economic in nature. It should be emphasised that women entered employment in textiles for a variety of reasons. They entered not only for economic reasons, but also for educational opportunities, for the chance to live away from home and be independent, for personal development, to be part of a community, and because they wanted to work. These opportunities were there in part because of the changes taking place in the economy, but also because of social changes that had fundamental implications for the lives and life cycles of working women. It should be remembered that the industry, and the various employment structures examined throughout these chapters, were a daily reality for thousands of working women during these years. These were women who

Japanese female textile workers in context 215 were not just working. They were studying while working when young, and they were working mothers juggling home life and work when older. Whatever the demands on their various roles as women, and whatever the hardships of working, textiles was an industry that provided females with gainful employment. Various quotes taken from my fieldwork survey and scattered throughout this book have put the voices of some of these women into this work. Their words have highlighted the influence the industry had on their daily lives, and provided an insight into their resilience and consciousness as workers. This research is part of their story – a story that filled me with admiration for their work and lives, and a story that begged to be recorded. It is only right that this book should end with their voices. My house was burnt to the ground during the war and life was very difficult. I had to work to have a livelihood and so I entered textile employment in 1948. I worked there for 40 years until retirement. During those years there was sex discrimination and men received higher wages and bonuses than females, and this still influences the pension. I would like to see these inequalities disappear in my lifetime. (Fieldwork survey respondent no. 1; employed 1948–88, age 16–56) When I joined, textile employment was the best job and paid good wages for women . . . I did seven different jobs over the years. The cooler wasn’t efficient enough for a large factory . . . the dust got into my nose and my eyes and I remember being worried that it was not good for my health. I have good memories of dormitory life and meeting to discuss the organisation of various activities . . . it was a good experience to understand that there were many kinds of people and many ways of thinking. The company asked me to keep working when I resigned, but I left to look after and help bring up two grandchildren. I think the company contributed to the Japanese economy . . . and I became confident through my job. (Fieldwork survey respondent no. 14; employed 1944–80, age 15–51) When I joined the company, people around me said I was silly (baka). Now they say that it’s good that I work for a large company. It was lonely at first. I was far from home and I wanted to go back. I couldn’t make friends initially. I was a member of the dorm self-government committee and it was hard work, but I enjoyed the company trips. I studied hard two hours a day Monday to Saturday in the company school. I lived every day to the full. I was able to keep working after marriage because the company had a nursery. I’m still working and I now have three children, so I’m grateful to the company. (Fieldwork survey respondent no. 28; employed 1965–, age 15–)

216 Japanese female textile workers in context I’ve been working for this company ever since I joined at 15 years of age. It is important for women to keep working even after they have a family. I think it’s healthy, and it’s important to have a plan for the day and be positive for its achievement. I want to keep managing myself and keep working. (Fieldwork survey respondent no. 35; employed 1960–, age 15–)

Appendix Fieldwork interviews and survey

Fieldwork interviews As agreed during fieldwork, the anonymity of all interviewees has been preserved. Interviews with female textile workers Interview A: individual interview with female employee conducted on 22 June 1998 at To¯yo¯bo¯, Head Office, Osaka. Interview B: group interview with four female employees conducted on 26 June 1998 at To¯yo¯bo¯’s TMC Factory, Hyogo. Interview C: individual interview with female employee conducted on 30 June 1998 at Kurabo¯ Head Office, Osaka. Interview D: group interview with five female employees conducted on 3 July 1998 at To¯yo¯bo¯’s Nyu¯zen Factory, Toyama. Interview E: individual interview with former female textile worker conducted on 24 July 1998 in Nishiwaki, Hyogo. Interview F: individual interview with female employee conducted on 6 August 1998 at Wacoal Head Office, Kyoto. Interview G: group interview with two female employees conducted on 20 August 1998 at Wacoal Head Office, Kyoto. Interview H: group interview with six female employees conducted on 25 September 1998 at Unichika’s Kaizuka Factory. Interviews and meetings with textile industry managers Interview I: series of regular meetings with former To¯yo¯bo¯ manager (OB) conducted between November 1997 and September 1998 at To¯yo¯bo¯ Head Office, Osaka (follow-up discussions in January 2003).

218 Appendix Interview J: meetings with Zensen officials conducted on 28 April 1998 at Zensen Headquarters, Tokyo. Interview K: meetings with sanchi employers and union members conducted on 4 August 1998 in Nishiwaki, Hyogo. Interview L: individual interview with JSA official conducted on 5 August 1998 at JSA Head Office, Osaka. Interview M: meeting with company manager conducted on 7 August 1998 at Unichika Head Office, Osaka.

The fieldwork survey Survey content The survey was a questionnaire in the Japanese language and was compiled by myself during the fieldwork year, and therefore any errors within it are my own. The survey consisted of six sections, comprising 65 questions in total, based on themes explored in the main chapters. The questions were a mixture of multiple choice and free response. The six sections were: 1 2 3 4 5 6

Personal history Working life and working conditions Residence and dormitories Education Work, wages and welfare Retirement and post-working life

At the end of the survey, space was also given for respondents to write freely with any further comments. The response to this provided many of the quotes presented throughout the book. I am grateful to Mr Shibata and Ms Tada of To¯yo¯bo¯ and Ms Okazaki of Kurabo¯ for their assistance with the distribution of the survey. Response to survey In total, 55 questionnaire surveys were distributed to women workers via personal connections at both To¯yo¯bo¯ and Kurabo¯ companies. The majority (50 surveys) went to women working in various To¯yo¯bo¯ factories. The surveys were distributed during August 1998 by company managers and staff, but the respondents replied in confidence by mailing their replies back to me in prepaid envelopes provided with the survey; 45 women responded to the survey between August and early October, giving a response rate of 82 per cent. Of the 45 respondents, 40 were from To¯yo¯bo¯ and 5 were from Kurabo¯. A brief overview of the personal profile of respondents is provided in Table A.1.

Table A.1 Personal profile of fieldwork survey respondents Profile detail

No. of respondents

Percentage of respondents

Age at joining company: 15 years old 16 years old 17–19 years old 20–30 years old

30 6 5 4

67 13 11 9

Period when joined company: 1945–54 1955–59 1960–64 1965–69 1970–74 1975–

4 6 11 14 8 2

9 13 24 31 18 4

Years of service (up to 1998): –4 yrs 5–19 yrs 20–29 yrs 30+ yrs

2 6 7 30

Average length of service for all respondents was: 23.4 yearsa

Textile sector where worked:b Spinning processes Weaving processes Other

25 10 5

63 25 13

Origins of respondents:c

69% of respondents were from agricultural families. 43% said that a female member of her family had previously worked for the industry

Notes a The high average length of service was noted as a bias of this survey in Chapter 1. Length of service was calculated up to and including 1998 when the survey took place. Many of the workers were still working at this time and presumably continued, indicating even higher lengths of service; 8 out of 45 respondents said they took a break from work (on average two years for either marriage or child-rearing) but then returned to work— this has not been accounted for in the total length of service figures (this does not include those who took only standard maternity leave). b In the survey 5 out of 45 respondents did not answer as to what kind of job they did. Therefore percentages are calculated from total of 40 respondents rather than 45. ‘Other’ included work in finishing departments and non-production jobs. c In the survey 6 out of 45 respondents did not answer the question on family background, so response based on 39 women.

References

Akamatsu, R. (1985). Danjo Koyo¯ Kikai Kinto¯ho¯ oyobi Kaisei Ro¯do¯ Kijunho¯ (The Equal Employment Opportunity Law and the Revised Labour Standards Law). Tokyo: Nihon Ro¯do¯ Kyo¯kai. Akamatsu, R. (ed.) (1977). Nihon Fujin Mondai Shiryo¯ Shu¯sei, Dai 3 Kan, Ro¯do¯ (A Collection of Materials on Japanese Women, Vol. 3, Labour). Tokyo: Gakuyo Shobo¯. Ashida, J. (1982). ‘Pa¯totaima¯ Soshikika no Genjo¯: Zensen Do¯mei no Torikumi ni tsuite (Actual Situation of the Organisation of Part-timers: Case of Textile Workers Union).’ Nihon Ro¯do¯ Kyo¯kai Zasshi, 24 (11): 4–13. Azumi, K. (1969). Higher Education and Business Recruitment in Japan. New York: Teachers College Press, Columbia University. Banshu¯ Ro¯do¯ Kumiai (1985). Yu¯ai to Shingi: Zensen Do¯mei Banshu¯ Ro¯do¯ Kumiai Rengo¯kai Ko¯ryo (Friendship and Faith: Principles of the Banshu¯ Labour Union). Nishiwaki-shi, Hyogo-ken: Banshu¯ Ro¯do¯ Kumiai. Banshu¯shoku Nyu¯su (1960a). Hyogo-ken Orimono Ko¯gyo¯ Kumiai Rengo¯kai (Hyogo Prefecture Weaving Industry Union). No. 6, September. Banshu¯shoku Nyu¯su (1960b). Hyogo-ken Orimono Ko¯gyo¯ Kumiai Rengo¯kai (Hyogo Prefecture Weaving Industry Union). No. 9, December. Banshu¯shoku Nyu¯su (1961). Hyogo-ken Orimono Ko¯gyo¯ Kumiai Rengo¯kai (Hyogo Prefecture Weaving Industry Union). No. 19, October. Bernstein, G. (1988). ‘Women in the Silk-Reeling Industry in Nineteenth Century Japan’, in Japan and the World: Essays on Japanese History and Politics in Honour of Ishida Takeshi, ed. G. Bernstein and H. Fukui. New York: St Martin’s Press. Broadbent, K. (2003). Women’s Employment in Japan: The Experience of Part-time Workers. London: RoutledgeCurzon. Burke, M. B. (1979). ‘Female Labour Patterns in Postwar Japan: 1955–1970’. University of Pennsylvania (PhD thesis). Chingin Ko¯zo¯ Kihon To¯kei Cho¯sa (Basic Statistical Survey of Wage Structure). Tokyo: Ministry of Labour. Date-Bah, E. (ed.) (1997). Promoting Gender Equality at Work. London and New York: Zed Books. Dore, R. (1986). Flexible Rigidities: Industrial Policy and Structural Adjustment in the Japanese Economy 1970–1980. London: Athlone Press. Francks, P. (1992). Japanese Economic Development: Theory and Practice. London: Routledge.

References 221 Goldin, C. (1990). Understanding the Gender Gap: An Economic History of American Women. Oxford: Oxford University Press. Gordon, A. (1998). The Wages of Affluence: Labor and Management in Postwar Japan. Cambridge, MA, Harvard University Press. Hanami, T. A. (1979). Labour Law and Industrial Relations in Japan. Amsterdam: Kluwer, Deventer. Hazama, H. (1997). The History of Labour Management in Japan. London: Macmillan. Hirosaki, S. (1967). Nihon Joshi Ro¯mu Kanri Shi: Shakai Keizai o Kicho¯ to suru (History of the Management of Japanese Female Labour: A Social Economic History). Tokyo: Keibundo. Hosoi, W. (1925). Joko¯ Aishi (Pitiful History of Factory Girls). Tokyo: Kaizo¯sha. Hunter, J. (1984a). ‘Recruitment in the Japanese Silk Reeling and Cotton Spinning Industries, 1870s–1930s’, in Proceedings of the British Association for Japanese Studies, Volume 9, ed. J. Chapman and D. Steeds. Sheffield: University of Sheffield. Hunter, J. (1984b). ‘Labour in the Japanese Silk Industry in the 1870s: The Tomioka Nikki of Wada Ei’, in Europe Interprets Japan, ed. G. Daniels. Tenterden, Kent: Paul Norbury. Hunter, J. (1989). ‘Factory Legislation and Employer Resistance: The Abolition of Night Work in the Cotton Spinning Industry’, in Japanese Management in Historical Perspective, ed. T. Yui and K. Nakagawa. Tokyo: University of Tokyo Press. Hunter, J. (ed.) (1993a). Japanese Women Working. London: Routledge. Hunter, J. (1993b). ‘Textile Factories, Tuberculosis and the Quality of Life in Industrialising Japan’, in Japanese Women Working, ed. J. Hunter. London: Routledge. Hunter, Janet (2003). Women and the Labour Market in Japan’s Industrialising Economy: The Textile Industry before the Pacific War. London: RoutledgeCurzon. Ike, B. (1980). ‘The Japanese Textile Industry: Structural Adjustment and Government Policy.’ Asian Survey, 20 (May): 532–51. Ishihara, O. (1913). ‘Joko¯ to Kekkaku (Factory Girls and Tuberculosis).’ Kokka Gakkai Zasshi, 322 (November). Ishii, K. (1972). Nihon Sanshigyo¯ Shi Bunseki (Analytical History of Japan’s Silk Reeling Industry). Tokyo: Tokyo Daigaku Shuppan Kai. JSA (1954). Labour Situation in the Japanese Cotton Spinning Industry – with ¯ mi Kenshi Strike. Osaka: Research Division, Japan Spinners’ Reference to the O Association. JSA (1955). Menka kara Orimono made (From Raw Cotton to Textile Fabrics). Osaka: Nihon Bo¯seki Kyo¯kai (Japan Spinners’ Association). JSA (1958). ‘News and Current Problems on the Japanese Cotton Industry.’ Nihon Bo¯seki Geppo (Monthly Report of the Japan Spinners’ Association), 137 (May): 2–7. JSA (1959). ‘News and Current Problems on the Japanese Cotton Industry: Dispute over Wages in Japanese Cotton Mills.’ Nihon Bo¯seki Geppo (Monthly Report of the Japan Spinners’ Association), 154 (October): 2–3. JSA (1961). ‘News and Current Problems on the Japanese Cotton Industry: Employment Trends in the Cotton Industry and Establishment of Textile High

222 References School.’ Nihon Bo¯seki Geppo (Monthly Report of the Japan Spinners’ Association), 178 (October): 50–53. JSA (1962). Sengo Bo¯seki Shi (Postwar History of the Spinning Industry). Osaka: Nihon Bo¯seki Kyo¯kai (Japan Spinners’ Association). JSA (1965). ‘Present Conditions and Problems of the Cotton Industry in Japan: Labour Situation.’ Nihon Bo¯seki Geppo (Monthly Report of the Japan Spinners’ Association), 223 (July): 76–78. JSA (1968). ‘News and Current Problems on the Japanese Spinning Industry: Labour Situation in the Japanese Textile Industry.’ Nihon Bo¯seki Geppo (Monthly Report of the Japan Spinners’ Association), 259 (July): 71–73. JSA (1969). Boshu¯ kankei Gyo¯mu Tanto¯sha: Gyo¯mu Techo¯ (Business Handbook for Recruitment Staff). Osaka: Nihon Bo¯seki Kyo¯kai (Japan Spinners’ Association). JSA (1970). Sho¯wa 45 Nenchu¯ no Bo¯sekigyo¯ (The Spining Industry during 1970). Osaka: Nihon Bo¯seki Kyo¯kai (Japan Spinners’ Association). JSA (1979a). Zoku Sengo Bo¯seki Shi (Postwar History of the Spinning Industry Continued). Osaka: Nihon Bo¯seki Kyo¯kai (Japan Spinners’ Association). JSA (1979b). Sen’i Sangyo¯ ni okeru Ro¯do¯sha Boshu¯ no Sho Mondai (Various Issues of Labour Recruitment in the Textiles Industry). Osaka: Nihon Bo¯seki Kyo¯kai Koyo¯ Tokubetsu Iinkai (Special Committee on Employment, Japan Spinners’ Association). JSA (1982). Bo¯kyo¯ Hyakunenshi (100 Year History of the Spinners’ Association). Osaka: Japan Spinners’ Association. JSA (1991). ‘Bo¯sekigyo¯ ni okeru Jinin Taisaku ni tsuite (Issues on Employment in the Spinning Industry).’ Nihon Bo¯seki Kyo¯kai Geppo¯ (Monthly Report of the Japan Cotton Spinners’ Association), 528–29 (February and March). Kagoyama, T. (1970). Joko¯ to Kekkaku (Factory Girls and Tuberculosis). Tokyo: Koseikan. Kajinishi, M., S. Tatewaki, et al. (1955). Seishi Ro¯do¯sha no Rekishi (History of Silk Workers). Tokyo: Iwanami Shoten. Kanebo¯ (1988). Kanebo¯ Hyakunenshi (100 Year History of Kanebo¯). Osaka: Kanebo¯ Kabushiki Kaisha. Kariya, T., S. Sugayama, et al. (1997). ‘Shinki Gakusotsu Ro¯do¯ Shijo¯ no Seidoka Katei ni kan suru Kenkyu¯: Sengo Nihon no Shokugyo¯ Antei Gyo¯sei to Ro¯do¯ Shijo¯ (Research on the Systemisation Process of the New School Leavers Labour Market: the Public Employment Security Administration and the Labour Market in Postwar Japan).’ Shakai Kagaku Kenkyu¯, 49 (2): 123–200. Kataoka, M. (1993). ‘Sengo no Bo¯seki ni okeru Ro¯mu Kanri Kyo¯iku Mondai (Problems of Labour Management and Education in the Postwar Spinning Industry).’ Nihon Bo¯seki Geppo¯ (Monthly Report of the Japan Spinners’ Association), 553–54 (March and April). Kawashima, Y. (1983). ‘Wage Differentials between Women and Men in Japan’. Stanford University (PhD thesis). Kobayashi, S. (1963). To¯yo¯bo¯ no Kishukusha Kanri (Management of To¯yo¯bo¯ Dormitories), To¯yo¯ Bo¯seki Ro¯mubu Fukubucho (Vice-director of Labour Department, To¯yo¯bo¯). Ko¯gyo¯ To¯kei Hyo¯ (Census of Manufacturing Industries). Tokyo: Research and Statistics Department, Ministry of International Trade and Industry. Koike, T. (1962). Nihon no Chingin Ko¯sho¯ (Wage Negotiations in Japan). Tokyo: Tokyo Daigaku Shuppankai.

References 223 Kojima, T. (1983). Hataraku Josei 100 nen no Ayumi – Dokyumento (Working Women: A Documentary of 100 Years of Progress). Tokyo: Kawade Shobo¯ Shinsha. Kokusei Cho¯sa Ho¯koku (Population Census of Japan). Tokyo: Bureau of Statistics, Office of the Prime Minister. Komatsu, Y. (2000). Kekkaku: Nihon Kindai Shi no Uragawa (Tuberculosis: The Underside of Japan’s Modern History). Osaka: Seifudo Shoten. Koyama, T. (1961). The Changing Social Position of Women in Japan. Geneva, Switzerland: UNESCO. Kroese, W. T. (1950). The Japanese Cotton Industry: Report of a Journey to Japan in the Autumn of 1949. Leiden: H. E. Stenfert Kroese’s Uitgeversmij N.V. Kurabo¯ (1988). Kurashiki Bo¯seki Hyakunenshi. Osaka: Kurashiki Bo¯seki Kabushiki Kaisha. Lam, A. (1992). Women and Japanese Management: Discrimination and Reform. London: Routledge. McNamara, D. (1995). Textiles and Industrial Transition in Japan. Ithaca, NY: Cornell University Press. Macnaughtan, H. (2001). ‘Female Labour in Japan’s Cotton Textile Industry, 1955– 1975’. LSE, University of London (PhD thesis). Matsumoto, E. (1991). Seishi Ro¯do¯ So¯gi no Kenkyu¯ (Research on Silk Reeling Labour Disputes). Tokyo: Hyaku Shobo¯. Men Bo¯seki Ro¯do¯ To¯kei Shiryo¯ (Collection of Cotton Spinning Labour Statistics) (annually 1955–70). Osaka: Nihon Bo¯seki Kyo¯kai (Japan Spinners’ Association). Menshi Bo¯seki Jijo¯ Sanko¯sho (Reference Book on the Circumstances of the Cotton Spinning Industry) (annually). Osaka: Nihon Bo¯seki Kyo¯kai (Japan Spinners’ Association). Mishima, Yukio (1998). Silk and Insight. Trans. from the Japanese Kinu to Meisatsu by Hiroaki Sato. Armonk, NY: M.E. Sharpe. MOL (1952). Men Bo¯seki Ko¯jo¯ no Joshi Ro¯do¯sha: Jittai Cho¯sa Ho¯koku (Female Workers in Cotton Spinning Factories: Report on the Survey of Conditions). Tokyo: Ro¯do¯sho Fujin Sho¯nen Kyoku (Women and Minors Bureau, Ministry of Labour). MOL (1970). Chu¯ko¯ Nenrei Fujin no Ro¯mu Kanri Jireishu¯ (Case Studies of Labour Management of Older Females). Tokyo: Ro¯do¯sho Fujin Sho¯nen Kyoku (Women and Minors Bureau, Ministry of Labour). MOL (1972). Sen’i Ko¯gyo¯ ni okeru Fujin Ro¯do¯ Jittai Cho¯sa: Kekka Ho¯kokusho (Survey of Conditions of Female Labour in Textile Factories: Report on Results). Tokyo: Ro¯do¯sho Fujin Sho¯nen Kyoku (Women and Minors Bureau, Ministry of Labour). Molony, B. (1991). ‘Activism among Women in the Taisho¯ Cotton Textile Industry’, in Recreating Japanese Women 1600–1945, ed. G. L. Bernstein. Berkeley, CA: University of California Press. Nakamura, M. (1994). Technology Change and Female Labour in Japan. Tokyo: United Nations University Press. Nakamura, T. (1995). The Postwar Japanese Economy. Tokyo: University of Tokyo Press. Nawa, T. (1948). Nihon Bo¯sekigyo¯ no Shiteki Bunseki (Historial Analysis of the Japanese Spinning Industry). Tokyo: Cho¯ryu¯sha.

224 References Nenju¯ no Bo¯sekigyo¯ (The Spinning Industry Yearly) (annually 1955–75). Osaka: Nihon Bo¯seki Kyo¯kai (Japan Spinners’ Association). Nihon Bo¯seki Kyo¯kai Geppo (Monthly Report of the Japan Spinners’ Association). Osaka: Nihon Bo¯seki Kyo¯kai (Japan Spinners’ Association). Nihon To¯kei Nenkan (Japan Statistical Yearbook). Tokyo: Bureau of Statistics, Office of the Prime Minister. Nishiwakishi Kyo¯iku Iinkai (1989). Do¯shizan – Nishiwaki-shi Kyo¯do¯ Shiryo¯kan Kiyo¯ (Doshizan – Bulletin of the Nishiwaki City Document Library). Nishiwakishi, Hyogo-ken: Nishiwakishi Kyo¯iku Iinkai. Osawa, M. (1994). ‘Japanese Style Employment Practices and Male–Female Wage Differentials.’ Japanese Economic Studies, 22 (5–6): 3–43. Park, K. C. (1976). ‘An Economic Analysis of Trends in Japanese Textile Industry, 1953–1973’. Catholic University of America, Washington, DC (PhD thesis). Partner, S. (1999). Assembled in Japan: Electrical Goods and the Making of the Japanese Consumer. Berkeley, CA: University of California Press. Ro¯do¯ So¯gi Cho¯sa Kai (1956). Sen’i Ro¯do¯ So¯gi to Kumiai Undo (Textile Labour Strikes and the Union Movement). Tokyo: Chuo¯ Ko¯ronsha. Ro¯do¯ To¯kei Nenkan (Japan Labour Yearbook) (annually). Tokyo: Ministry of Labour. Ro¯do¯ To¯kei Nenpo¯ (Year Book of Labour Statistics) (annually). Tokyo: Statistics and Information Department, Ministry of Labour. Ro¯do¯ryoku Cho¯sa (Labour Force Survey). Tokyo: Bureau of Statistics, Office of the Prime Minister. Saito, H. (1969). Fujin Ro¯do¯ – Sono Ro¯do¯ Kagakuteki Bunseki (Women’s Labour: Scientific Analysis of that Labour). Tokyo: Ro¯do¯ Kagaku Kenkyu¯jo¯ (Labour Science Research Institute). Sakura, T. (1927). Joko¯ Gyakutai Shi (History of the Ill-treatment of Female Factory Operatives). Shinano, Nagano: Mainichi Shinbunsha (reprint 1981). Sakurai, K. (1987). Bosei Hogo Undo Shi (History of Motherhood Protection Movements). Tokyo: Domesu Shuppan. Saxonhouse, G. (1976). ‘Country Girls and Communication among Competitors in the Japanese Cotton-Spinning Industry’, in Japanese Industrialisation and its Social Consequences, ed. H. Patrick. Berkeley, CA: University of California Press. Saxonhouse, G. (1977). ‘Productivity Change and Labour Absorption in Japanese Cotton Spinning 1891–1935.’ Quarterly Journal of Economics, 91 (2): 195–219. Saxonhouse, G. and Y. Kiyokawa (1985). ‘Supply and Demand for Quality Workers in Cotton Spinning in Japan and India’, in Japan and the Developing Countries: A Comparative Analysis, ed. K. Ohkawa and G. Ranis. Oxford: Blackwell. SCAP (1946a). Final Report of the Advisory Committee on Labour: Labour Policies and Programmes in Japan. Tokyo: Supreme Commander for the Allied Powers. SCAP (1946b). SCAP Directive No. 13. Tokyo: Supreme Commander for the Allied Powers. SCAP (1946c). SCAP Directive No. 14. Tokyo: Supreme Commander for the Allied Powers. SCAP (1947a). SCAP Directive No. 15. Tokyo: Supreme Commander for the Allied Powers. SCAP (1947b). SCAP Directive No. 19. Tokyo: Supreme Commander for the Allied Powers.

References 225 SCAP (1947c). SCAP Directive No. 22. Tokyo: Supreme Commander for the Allied Powers. SCAP (1947d). SCAP Directive No. 26. Tokyo: Supreme Commander for the Allied Powers. SCAP (1947e). SCAP Directive No. 27. Tokyo: Supreme Commander for the Allied Powers. SCAP (1948a). SCAP Directive No. 30. Tokyo: Supreme Commander for the Allied Powers. SCAP (1948b). SCAP Directive No. 31. Tokyo: Supreme Commander for the Allied Powers. SCAP (1948c). SCAP Directive No. 32. Tokyo: Supreme Commander for the Allied Powers. SCAP (1948d). SCAP Directive No. 34. Tokyo: Supreme Commander for the Allied Powers. SCAP (1948e). SCAP Directive No. 35. Tokyo: Supreme Commander for the Allied Powers. SCAP (1951). History of the Nonmilitary Activities of the Occupation of Japan 1945– 1951: Textile Industries. Tokyo: General Headquarters. Tokyo: Supreme Commander for the Allied Powers. Seki, K. (1954). Nihon Mengyo¯ Ron (The Cotton Industry of Japan). Tokyo: Tokyo Daigaku Shuppankai. Seki, K. (1956). The Cotton Industry of Japan. Tokyo: Japan Society for the Promotion of Science. Sen’i Nenkan (Textile Year Book) (annually). Osaka: Nihon Sen’i Kyo¯gi Kai (Japan Textiles Council). Shibata, E. (1975). Gendai no Fujin Ro¯do¯sha (Modern Women Workers). Tokyo: Gakushu no Tomosha. Shibata, E. and I. Yamada (1983). Danjo Byo¯do¯ no Kyo¯, Ashita: Nihongata Shakai to Josei Ro¯do¯ (Equal Rights, Today and Tomorrow: Japanese-style Society and Female Workers). Tokyo: Gakushu no Tomosha. Shimazu, C. (1953). Joshi Ro¯do¯sha: Sengo no Men Bo¯seki Ko¯jo¯ (Female Workers: Postwar Cotton Spinning Factories). Tokyo: Iwanami Shoten. Shimazu, C. (1965). Go¯rika to Fujin Ro¯do¯sha (Rationalisation and Women Workers). Tokyo: Ro¯do¯ Junpo¯sha. Shimazu, C. (1970). Fujin to Ro¯do¯ (Women and Labour). Tokyo: Shin Nihon Shuppansha. Shindo, T. (1958). Nippon Mengyo¯ Ro¯do¯ Ron (Labour in the Japanese Cotton Industry). Tokyo: Tokyo Daigaku Shuppankai. Shindo, T. (1961). Labour in the Japanese Cotton Industry. Tokyo: Japan Society for the Promotion of Science. Shiozawa, M. (1976). Kekkon Taishokugo no Watashitachi: Seishi Ro¯do¯sha no Sono Go (After Termination for Marriage: Silk Factory Workers). Tokyo: Iwanami Shoten. Shokugyo¯ Antei Gyo¯mu To¯kei (Statistics on Activities of the Public Employment Security Office). Tokyo: Bureau of Employment Security, Ministry of Labour. Singleton, J. (1997). The World Textile Industry. London: Routledge. Sugeno, K. (1992). Japanese Labor Law. Seattle, WA: University of Washington. Sugihara, K. (1986). ‘The Transformation of Social Values of Young Country Girls: Towards a Reinterpretation of the Japanese Migrant (dekasegi) Industrial

226 References Labour Force’, in Aspects of the Relationship between Agriculture and Industrialization in Japan, ed. J. Hunter. London: London School of Economics, STICERD. Sugihara, K. (1996). Ajia kan Bo¯eki no Keisei to Ko¯zo¯ (The Formation and Structure of Inter-Asian Trade). Tokyo: Mineruva Shobo¯. Suzuki, Y. (1989). Joko¯ to Ro¯do¯ So¯gi: 1930 nen Yo¯mo¯su So¯gi (Women Factory Workers and Labour Disputes: The Toyo Muslin Labour Dispute in 1930). Tokyo: Renga Shobo¯ Shinsha. Takase, T. (1972). Ikyo¯ ni Chitta Wakai Inochi: Kan-ei Tomioka Seishijo¯ Ko¯jo no Haka (Young Lives Fallen in a Strange Land: Graves of Silk Factory Girls in Tomioka). Tokyo: Taimatsusha. Takase, T. (1979). Kan-ei Tomioka Seishijo¯ Ko¯jo Shiryo¯ (Historical Materials on Factory Girls at Tomioka Silk Factory). Tokyo: Taimatsusha. Tojo, Y. (1990). Seishi Do¯mei no Joko¯ To¯roku Seido: Nihon Kindai no Henyo¯ to Joko¯ no ‘Jinkaku’ (Silk Reeling Union’s Registration System of Female Factory Operatives: Japan’s Modern Transformation and the ‘Character’ of Female Factory Operatives). Tokyo: Tokyo Daigaku Shuppankai. Toyne, B., et al. (1984). The Global Textile Industry. London: Allen & Unwin. To¯yo¯bo¯ (1955). Cotton Industrial Wages in Japan: To¯yo¯bo¯ Economic Series No. 2. Osaka: To¯yo¯bo¯, Institute for Economic Research. To¯yo¯bo¯ (1960). Jichi Iin no tame ni (For Self-Government Committee Members). Osaka: To¯yo¯bo¯. To¯yo¯bo¯ (1961a). Fukuri Ko¯sei Shisetsu no Atarashii Do¯ko¯ ni tsuite (New Trends in Welfare Facilities). Osaka: To¯yo¯bo¯. To¯yo¯bo¯ (1961b). Kiyakushu¯ Joshi Kishukusha (Collection of Rules for Female Dormitories). Osaka: To¯yo¯bo¯ Joshi Kishukusha Wakaba Ryo¯ Jichikai (SelfGovernment Committee of Wakaba Female Dormitory, To¯yo¯bo¯). To¯yo¯bo¯ (1968). Gakko¯ kankei Kitei (Collection of Rules Relating to Schools). Osaka: To¯yo¯bo¯ Ro¯muka (Labour Division, To¯yo¯bo¯). To¯yo¯bo¯ (1969a). Kongo no Joshi Kanri no Arikata (Future Management of Female Workers). Osaka: To¯yo¯bo¯ Joshi Kanri Semmon Iinkai (Specialist Committee for Management of Females, To¯yo¯bo¯). To¯yo¯bo¯ (1969b). Ko¯jo¯ Hyo¯jun Sagyo¯ Soshiki oyobi Shokumuna ni tsuite: Sagyo¯ Soshiki no Kaisetsu (Organisation of Factory Work Standards and Classification of Duties: Commentary on Work Organisation). Osaka: To¯yo¯ Bo¯seki Kabushiki Kaisha, Ro¯mubu Ro¯muka (Labour Division, Labour Department, To¯yo¯bo¯). To¯yo¯bo¯ (1970a). Ko¯sotsu Joshi Genjo¯ Do¯nyu ni tsuite (The Introduction of Female Senior High School Leavers to the Workplace). Osaka: To¯yo¯bo¯ Ko¯sotsu Joshi Genjo¯ Do¯nyu Semmon Iinkai (Specialist Committe for the Introduction of Female Senior High School Leavers to the Workplace, To¯yo¯bo¯). To¯yo¯bo¯ (1970b). Fukuri Ko¯sei Senmon Iinkai To¯shinsho (Report by the Special Committee for Welfare). Osaka: To¯yo¯bo¯. To¯yo¯bo¯ (1974). Shokuba Ro¯mu Kanri no Ju¯jitsu Kyo¯ka ni tsuite (Strengthening the Fullness of Workplace Labour Management). Osaka: To¯yo¯bo¯ Shokuba Ro¯mu Kanri Semmon Iinkai (Specialist Committee for Workplace Labour Management, To¯yo¯bo¯). To¯yo¯bo¯ (1975a). Kongo no Gakuin Unei no Kihonteki Ho¯ko¯ (Fundamental Direction of Future Management of Schools). Osaka: To¯yo¯bo¯ Gakuin Semmon Iinkai (Specialist Committee for Schools, To¯yo¯bo¯).

References 227 To¯yo¯bo¯ (1975b). Kyo¯iku Kunren Tanto¯sha Jitsumu Kenshu¯kai Shiryo¯ (Practical Training Meeting Documents for Education Training Staff). Osaka: To¯yo¯bo¯ Kabushiki Kaisha Chuo¯ Kenshu¯sho (To¯yo¯bo¯ Central Research Centre). To¯yo¯bo¯ (1977). Kongo no Joshi Ryo¯ Gakuin Unei no Ho¯ko¯ (Future Direction of Management of Female Dormitories and Schools). Osaka: To¯yo¯bo¯ Ro¯muka (To¯yo¯bo¯ Labour Division). To¯yo¯bo¯ (1978). To¯yo¯ Bo¯seki Ro¯do¯ Kumiai Shi (History of the To¯yo¯ Spinning Company Union). Osaka: Zensen Do¯mei To¯yo¯ Bo¯seki Ro¯do¯ Kumiai. To¯yo¯bo¯ (1980). Shin Sangyo¯ Shirizu 14: Sen’i Sangyo¯ (New Industries Series No. 14: Textiles Industry). Tokyo: To¯yo¯ Keizai Shinposha. To¯yo¯bo¯ (1986). Hyakunenshi To¯yo¯bo¯ (The 100 Year History of To¯yo¯bo¯). Osaka: To¯yo¯bo¯. Tsurumi, P. E. (1990). Factory Girls: Women in the Thread Mills of Meiji Japan. Princeton, NJ: Princeton University Press. Tsurumi, P. E. (1995). ‘Whose History is it Anyway? And Other Questions Historians Should Be Asking. In this Case about the Cotton and Silk Thread Factory Women of Meiji Japan.’ Japan Review, 6: 17–36. Tsutsui, W. (1998). Manufacturing Ideology: Scientific Management in TwentiethCentury Japan. Princeton, NJ: Princeton University Press. Uno, K. S. (1992). ‘The Death of “Good Wife, Wise Mother”?’, in Postwar Japan as History, ed. A. Gordon. Berkeley, CA: University of California Press. Utley, F. (1931). Lancashire and the Far East. London: George Allen & Unwin. Walton, A. D. (1958). ‘The Japanese Cotton Textile Industry, 1945–1955.’ Journal of the Textile Institute, 49 (12): 696–705. Watanabe, E. (1980). Tatakai ni Ikite: Senzen Fujin Ro¯do¯ Undo¯ e no Sho¯gen (Living to Fight: Testimonies of the Prewar Women’s Labour Movement). Tokyo: Domesu Shuppan. Yamamoto, S. (1979). Aa Nomugi To¯ge: Aru Seishi Ko¯jo Aishi, Shinpan (Ah the Nomugi Pass: The Sad Story of Silk Factory Girls, new edition). Tokyo: Asahi Shinbunsha. Yamamoto, S. (1980). Zoku, Aa Nomugi To¯ge (Ah Nomugi Pass Continued). Tokyo: Kadokawa Shoten. Yamanouchi, M. (1975). Yamanouchi Mina Jiden: Ju¯ni-sai no Bo¯seki Joko¯ kara no Sho¯gai (Autobiography of Mina Yamanouchi: A Life Story from a 12-Year-Old Female Spinning Factory Operative). Tokyo: Shinjuku Shobo¯. Yashiro, N. (1981). ‘Male–Female Wage Differentials: Rational Explanation.’ Japan Economic Studies, 9 (winter): 28–61. Yasui, J. (1967). Sen’i Ro¯shi Kankei no Shiteki Bunseki: Ro¯shi Kankei no Nihonteki Tokushu¯sei no Saikento¯ (Historical Analysis of Labour Relations in Textiles: A Re-examination of Special Characteristics of Japanese Labour Relations). Tokyo: Ochanomizu Shobo¯. Zensen (1954). Tachiagaru Joshi Ro¯do¯sha (Female Workers Rise to their Feet). Nihon Eiga Shinsha (video). Zensen (1955). Tatakai Nuita Nakamatachi (Those who Fought it Out). Zensen Do¯mei (video). Zensen (1959a). Kishukusha Minshuka Ho¯shinsho (Plan for the Democratisation of Dormitories). Tokyo: Zensen Do¯mei. Zensen (1959b). ‘Fujin no Kenko¯ Mondai ni tsuite (Women’s Health Issues).’ Sen’i Ro¯do¯ (Textile Labour), 133 (March): 1–198.

228 References Zensen (1961). ‘Jichikai Daihyo¯ wa Ko¯kaisha Kotaete iru: Do¯mei Daikai 10 Kishukusha Jichikai Daihyo¯sha Zenkoku Taikai Shussekisha no Ankeeto (How Self Government Committee Representatives are Responding: Survey of Respondents from the 10th Zensen Annual National Meeting for Dormitory Self-Government Committee Representatives).’ Sen’i Ro¯do¯ (Textile Labour), 101 (December): 34–58. Zensen (1962a). ‘Ko¯jo¯ Gakuen ni manabu Kishukusha Izumaisha no Jitsukan: Seito wa nani o kangaete iru ka (Reality of Dormitory Residents learning in Factory Schools: What are students thinking?).’ Sen’i Ro¯do¯ (Textile Labour), 109 (October): 37–53. Zensen (1962b). ‘Senii Sangyo¯ no Rinjiko¯: Sono jittai to taisaku (Temporary Workers in the Textile Industry: Conditions and Measures).’ Sen’i Ro¯do¯ (Textile Labour), 109 (October): 4–11. Zensen (1962c). ‘Seiri Kyu¯ka to Sanzen Sango Kyu¯ka no Ho¯teki Kaisetsu to Mondai ni tsuite (Legal Explanation of the Issues relating to Menstruation Leave and Pre- and Post- Birth Leave).’ Sen’i Ro¯do¯ (Textile Labour), 103 (March): 30–68. Zensen (1962d). ‘Sen’i Sangyo Kaku Kumiai betsu Ikuji Jikan no Kitei (Regulations on Childcare Time in Textile Unions).’ Sen’i Ro¯do¯ (Textile Labour), 106 (June): 18–27. Zensen (1963). ‘Kishukusha Jittai Cho¯sa no Shu¯kei ni tsuite (Summary of Survey on Conditions in Dormitories).’ Sen’i Ro¯do¯ (Textile Labour), 113 (March): 37–39. Zensen (1964). ‘Gyo¯shu betsu ni mita Fukuri Ko¯sei Shisetsu (A Look at Welfare Facilities by Industry).’ Sen’i Ro¯do¯ (Textile Labour), 125 (June): 46–57. Zensen (1996). Zensen Josei Undo¯shi: Kishukusha Minshuka kara Danjo Byo¯do¯ Koyo¯ e (History of Women’s Movement in Zensen: From Democratisation of Dormitories to Equal Employment). Tokyo: Zensen Do¯mei. Zensen (1997). Danjokyo¯ ni Shigoto to Katei ga Ryo¯ritsu dekiru Kankyo¯ Seibi e mukete (Towards Maintenance of a Male–Female Work Environment where Work and Home are Compatible). Okayama: Zensen Chuo¯ Kyo¯iku Senta¯ (Zensen Central Education Centre).

Index

absenteeism 99, 128, 131 activism see strikes age 19–20, 89–90, 92, 99–100, 106, 111, 113–15, 118, 123, 125–31, 134, 148, 160–61, 163–66, 168–71, 175–76, 184, 207–08, 213, 215 Big Ten 9–10, 14, 19, 44, 51, 78, 81–82, 86, 150, 162, 167, 173–74, 176, 186, 195–98 bosei hogo see motherhood protection child labour 28–29, 61, 164, 166, 177–78 contracts 30, 103, 128 dekasegi see migration demand (for textiles) 13 dormitories: ‘apartment-style’ dormitories 113, 182–83, 197, 201; dormitory routine 118–21, 185; dormitory staff 111–12, 117–18, 122–25, 134, 137, 146–47, 156, 164; facilities and equipment 34, 117–18, 145, 180–84; funding of 117, 183, 197, 201; as labour management 36, 106–27, 129, 134–35, 146–47, 179, 184; legislation 34–35, 180, 182; movement to ban 33–34; predominance of 22, 109, 148, 179–82; and production 108–09, 134; seasonal workers in 22; self-government of 47, 110, 112, 116–19, 121–22, 124, 179, 183, 185, 215; as welfare 35–36, 47, 107, 159, 178–84, 201; worker freedom in 34–35 economic miracle 8, 213 education (schooling) 21, 52, 68, 85–86,

106, 133–49, 178; attendance 136, 138, 140–41, 143, 148, 169; bridal training 104, 136, 139, 144; curriculum 137–38, 143–46; factory-run schools 136–45, 186, 196, 202, 205, 215; funding of 143, 197–98, 201–203; as labour management 106–07, 127, 134–35, 140, 146–47, 154, 179; national education trends 60–64, 67, 80, 124, 136; of older females 148; part-time high schools 136–39, 141–43, 145, 197, 202, 205; teaching staff 137–38, 156, 164, 198; tertiary education 138, 141, 145, 148; as welfare 48, 107, 178–79 electronics industry 13, 17, 88, 135, 147, 173, 175, 211, 213 Employment Security Law 28, 31, 58 Equal Employment Opportunity Law (EEOL) 16, 190–91 exploitation 1–2, 4, 8, 36, 51, 105 export 1, 10, 13–14, 38, 177 Factory Act 5–7, 28–29, 50–51 familism 103–04, 199, 213 food (and nutrition) 37, 185, 188–89, 196–97, 200–01, 203, 206, 209 gender 6, 8, 36, 106, 160–61, 164, 167, 208, 210–14 health and safety 52–53, 119, 145, 153–56, 178–79, 184, 189–91, 194, 196, 207, 209; see also well-being (of workers) Hokkaido¯ 77, 96–98, 171–72, 206 intensity (labour) 9, 52, 55, 57, 150

230 Index Japan Spinners’ Association (JSA) 9–11, 21, 23, 40, 110, 113, 146, 154, 157 joko¯ aishi 2, 8, 34–35, 40–42, 48, 105, 118, 135–36, 147, 184, 188, 210 Kanebo¯ 14, 103 Kurabo¯ 103 Kyu¯shu¯ 77 labour management: development of 102–07; of older females 126–32; of young females 108–26, 184 labour shortage 12, 19, 48, 57, 63–64, 73, 79, 88–89, 92, 94, 122–23, 136, 173, 189, 199, 207, 210; labour ‘insufficiency’ 80–81, 88 Labour Standards Law (LSL) 16, 29, 34–35, 45, 50–53, 113, 178, 180, 182, 188–89, 192–93 legislation (labour) 27–28, 30, 92–93, 179–80, 190–93, 209–10; see also Labour Standards Law (LSL) length of service 20–22, 156, 160, 164, 166, 169–71, 208 male employment 6, 51, 57, 89, 104, 155–56, 160, 164, 166–69, 178, 189, 206, 208, 211–13, 215 manufacturing industry 13–14, 17, 173–75, 179, 196, 198; female employment in 18, 66, 74, 87–88, 94 marital status 2, 19–20, 126, 128–29, 166, 169, 179, 187, 194, 204, 208, 210, 212 maternity 188, 190–95 M-curve 22, 166, 169, 195 menstruation 188–92 migration 2, 6, 97, 125; see also recruitment, long-distance Ministry of Labour (MOL) 23, 30–32, 38, 40, 58, 64, 70, 72, 74, 93, 96–97, 188, 191, 212 minors see child labour motherhood protection 6–7, 29, 46, 49, 52, 185, 188–95 New-New spinning companies (Shinshinbo¯) 10, 44, 81–83, 162 New spinning companies (Shinbo¯) 10, 44, 81–83, 162 night work 5–6, 50–51, 57, 89 nimble fingers 212–13 Nishiwaki see sanchi

Nisshinbo¯ 14 non-regular labour 19, 21–22, 89, 92, 96, 100, 126–28, 164, 169–70, 176, 206, 210, 213; and unions 44–45, 49, 56, 95; see also seasonal workers; temporary workers non-wage costs see welfare nurseries (child-care) 95, 187, 190, 207–08, 215 Occupation of Japan 27–28, 38–39, 167, 176, 178, 199; see also Supreme Commander for the Allied Powers (SCAP) Okinawa 78 ¯ mi Kenshi 40–43, 176 O oral history evidence 23, 25, 187, 208, 215–19 part-time workers 16, 23–24, 49–50, 90, 92–93, 96, 100, 129, 131, 207, 214; see also non-regular labour paternalism 103, 199 pension see retirement productivity (labour) 8, 12, 56–57 promotion 112, 152, 156–58, 167 Public Employment Security Office (PESO) 31, 58–60, 63, 65–73, 75, 78–79, 83, 89–90, 92–93, 95, 97–98, 129 rationalisation (labour) 12, 21, 46, 56–57, 190 rearing (of young females) 104, 108, 113, 119, 122–25, 127, 133–35, 139–40, 146, 148, 152, 154–55, 158, 179, 187, 198, 204–05, 210 recruiting agents (recruiters): postwar 70–74, 78, 98, 211; prewar 4, 30, 68, 71 recruitment: competition 12–13, 63, 73–74, 79, 88, 97, 210; of female school leavers 59–89, 212; and high schools 59–60, 67–70, 72, 92; legislation 30–32, 58–60, 70–71; long distance 30–32, 71, 74–79; of older females 89–101; prewar 3–4, 30, 68 remittances 200, 203–04; see also savings retirement 168, 169–70, 173, 199, 206–08, 212, 215 safety see health and safety sanchi 11, 25, 68, 81, 112, 138, 145, 148, 163

Index 231 savings 37, 145, 201, 203, 205, 209 schools see education (schooling) seasonal workers 22–24, 92–94, 96–98, 101, 171–72, 206 shift work system 2, 49–52, 109, 118, 126, 130, 140–41, 154–55, 160, 167–68, 171, 185, 208 Shikoku 77 silk 1–3, 8–9, 19, 53 sleep patterns 109, 118, 185 social dumping 4, 36, 39 spinning: process 9, 108, 149–52, 154; technology 54–55; wages 163, 170–71 sports (and recreation) 117, 119, 146, 179, 181, 183, 185–87, 196, 212–13 standard of living 37, 124, 159, 178–80, 195–96, 212; cost of living 199–205 strikes 5, 7, 12, 40–43, 51, 176, 211 Supreme Commander for the Allied Powers (SCAP) 27–40, 58, 70, 178 synthetic textiles 13–14 Taylorism 103 technology 12, 49, 52–57, 80, 151, 163, 166, 189, 212–13 temporary workers 16, 19; 22–24, 94–95, 170–72, 190, 206 textile industry: curtailment of operations 12, 51; decline 11, 13–15, 17, 88, 176, 213; production 10–11, 14; scale (of enterprises) 9–11, 21, 23, 52–53, 56, 81, 83, 93, 107, 111–12, 137–38, 141, 145, 147, 156, 161–63; 168, 177, 187, 195, 203; structural adjustment 11, 13–15, 213; structure 9 To¯yo¯bo¯ 14, 75, 83, 85, 92, 102, 110–11, 116–17, 122–23, 128–29, 139, 141, 143–48, 182, 185–87, 197–204 training 104, 106, 109, 147, 149, 152–58; new entrant training 152–55, 157 Training Within Industry (TWI) 152, 154–55, 157

tuberculosis 2, 184 turnover 4–5, 20, 94, 199 Unichika 10, 14, 186, 201 unions: and female workers 43–49; structure 43–44; and wages 167–68, 173, 176–77, 212; and welfare 178, 189; see also Zensen wage differentials 49, 164–67, 212 wage negotiations 10, 43–44, 168, 171, 173, 176–77; shunto¯ 43, 176–77; wage rises 161, 171–74, 176–77, 205 wages 159–77, 195–96, 198–99, 202, 204–208, 212; bonus payments 161, 168, 170–71, 173, 175, 199, 206, 215; ‘cheap’ female labour 2, 4, 36, 39, 177, 205–206; family allowances 37, 160–61, 166–67, 208, 212; of non-regular (older) labour 96, 169–72, 207; real wages 176, 195, 198–99, 209 weaving: process 9, 108, 131, 149–52; technology 56–57; wages 160, 163, 170–71 welfare 177–208; expenditure 36–37, 178, 185, 196–99, 201, 207; facilities 36, 106, 159, 173, 177–78, 187, 212; non-legal 38, 47, 179, 184, 190, 195–98, 209; for older women 206–08 well-being (of workers) 38, 184–87, 194, 201, 206 women workers: consciousness of 6–7, 43, 86, 99, 104, 121, 124, 180, 205, 214–15; demand for 90–92; flexibility of 14, 60, 94, 126, 213; proportion in textile industry 1, 18 working hours 2, 5, 50–52, 109, 130; holidays 52, 189 workplace conditions 2, 4, 52–54, 173, 190–91, 209 Zensen 23, 43–49, 51, 56, 94, 117–18, 129, 138–39, 148, 163, 169–71, 177, 182, 188–95, 202, 206–07