Democracy and the State in the New Southern Europe (Oxford Studies in Democratization)

  • 92 742 2
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up

Democracy and the State in the New Southern Europe (Oxford Studies in Democratization)

O X F O R D S T U D I E S I N D E M O C R AT I Z AT I O N Series editor: Laurence Whitehead ..................... D E

1,647 717 3MB

Pages 454 Page size 252 x 358.92 pts Year 2008

Report DMCA / Copyright

DOWNLOAD FILE

Recommend Papers

File loading please wait...
Citation preview

O X F O R D S T U D I E S I N D E M O C R AT I Z AT I O N Series editor: Laurence Whitehead

.....................

D E M O C R A C Y A N D T H E S TAT E I N T H E NEW SOUTHERN EUROPE

O X F O R D ST U D I E S I N D E M O C R AT I Z AT I O N Series editor: Laurence Whitehead

................ Oxford Studies in Democratization is a series for scholars and students of comparative politics and related disciplines. Volumes will concentrate on the comparative study of the democratization processes that accompanied the decline and termination of the cold war. The geographical focus of the series will primarily be Latin America, the Caribbean, Southern and Eastern Europe, and relevant experiences in Africa and Asia.

OTHER BOOKS IN THE SERIES Democracy and Diversity: Political Engineering in the Asia-Pacific Benjamin Reilly Democratization: Theory and Experience Laurence Whitehead International Democracy and the West: The Role of Governments, Civil Society, and Multilateral Business Richard Youngs The Architecture of Democracy: Constitutional Design, Conflict Management, and Democracy Edited by Andrew Reynolds Institutions and Democratic Citizenship Axel Hadenius The European Union and the Promotion of Democracy: Europe’s Mediterranean and Asian Policies Richard Youngs

Democracy and the State in the New Southern Europe ..................... Edited by

R I C H A R D G U N T H E R , P. N I K I F O R O S DIAMANDOUROS, AND DIMITRI A. SOTIROPOULOS

1

3 Great Clarendon Street, Oxford OX2 6DP Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide in Oxford New York Auckland Cape Town Dar es Salaam Hong Kong Karachi Kuala Lumpur Madrid Melbourne Mexico City Nairobi New Delhi Shanghai Taipei Toronto With offices in Argentina Austria Brazil Chile Czech Republic France Greece Guatemala Hungary Italy Japan Poland Portugal Singapore South Korea Switzerland Thailand Turkey Ukraine Vietnam Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries Published in the United States by Oxford University Press Inc., New York ß The Social Science Research Council 2006 The moral rights of the author have been asserted Database right Oxford University Press (maker) First published 2006 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this book in any other binding or cover and you must impose the same condition on any acquirer British Library Cataloguing in Publication Data Data available Library of Congress Cataloguing in Publication Data Democracy and the state in the new Southern Europe/edited by Richard Gunther, P. Nikiforos Diamandouros, and Dimitri A. Sotiropoulos. p. cm.—(The new Southern Europe) Includes bibliographical references and index. 1. Europe, Southern—Politics and government—20th century. 2. Democracy—Europe, Southern. I. Gunther, Richard. II. Diamandouros, Nikiforos P. III. Soteropoulos, Demetres A., 1960—JN94.A58.D45 2006 320.94—dc22 2006018946 Typeset by SPI Publisher Services, Pondicherry, India Printed in Great Britain on acid-free paper by Biddles Ltd., King’s Lynn, Norfolk ISBN 0–19–920281–8 978–0–19–920281–2 0–19–920282–6 (Pbk.) 978–0–19–920282–9 (Pbk.) 1 3 5 7 9 10 8 6 4 2

................ Preface and Acknowledgments ................ Richard Gunther and P. Nikiforos Diamandouros

Sponsored by the Subcommittee on the Nature and Consequences of Democracy in the New Southern Europe of the Joint Committee on Western Europe (1975–96) of the American Council of Learned Societies and the Social Science Research Council. This book was brought to publication with the generous assistance of the VolkswagenStiftung and the Mershon Center, Ohio State University. This book is the fourth in a projected five-volume series on The New Southern Europe. It is the product of an initiative directed by the Subcommittee on the Nature and Consequences of Democracy in Southern Europe (or, more simply, the Subcommittee on Southern Europe) of the American Council of Learned Societies (ACLS) and the Social Science Research Council (SSRC). The fundamental objective of this long-term project was to place social science studies of Greece, Italy, Portugal, and Spain within the mainstream of comparative research by presenting systematic analyses of political, social, economic, and cultural changes that dramatically transformed the region over the course of the final four decades of the twentieth century. Previously, lagging processes of socioeconomic development and interludes of authoritarian rule had effectively marginalized three of these four countries (Greece, Portugal, and Spain) from comparative social science research, leading many scholars (e.g. Arrighi 1985) to regard Southern Europe as a ‘semiperipheral’ region. The sociopolitical and economic transformations analyzed in the first three volumes in this series have progressively eroded away the various Southern European ‘exceptionalisms’ that had relegated the region to this marginalized status, particularly with regard to Greece, Portugal, and Spain. This book carries this comparative analysis further by extending it to transformations of the state and key sectors of public policy that have come to define the ‘modern social-welfare state’.

vi

Preface and Acknowledgments

The intellectual history of the various initiatives that eventually gave rise to this project is worth recounting briefly. (For a fuller account of the circumstances that gave rise to this project, see the Preface to Volume 1 of this series: Gunther, Diamandouros, and Puhle 1995: ix–xxx). The initial stimulus for the creation of a committee of social scientists interested in studying the evolution of Southern Europe was provided by the events in Greece and Portugal in 1974 that toppled the authoritarian regimes in those countries. Following a series of formal and informal meetings over the ensuing two years, a research group was formed, called the Committee on Southern Europe (COSE), whose purpose was to promote the systematic study of the region over the period from the end of the Napoleonic Wars to the present. Initial participants in this effort were Douglas A. Chalmers, P. Nikiforos Diamandouros, Giuseppe DiPalma, James R. Kurth, Juan J. Linz, Edward E. Malefakis, the late Joyce F. Riegelhaupt, and Philippe C. Schmitter. The eventual product of this initiative was the creation of an informal, interdisciplinary network of scholars interested both in Southern Europe as a region and in the particular historical or political processes it exemplifies best, such as rapid and profound socioeconomic transformation in late developing societies and successfully negotiated transitions to democracy. This coordinated research effort helped raise the visibility of the Southern European region in meetings of professional associations at the national level (in particular, the political science and sociological associations in Europe and the United States, including the Council of European Studies), the regional level (e.g. the European Consortium for Political Research, where a separate Southern European Study Group was set up in the early 1980s), and the international level (e.g. International Political Science Association and the International Sociological Association). The result was a significant accumulation of knowledge on various dimensions of the polities, economies, and cultures of Southern European societies, with significant spillover effects for both area studies and comparative analysis. This burgeoning activity came to happy fruition in 1987–8. Two informal meetings served as the immediate stimulus for a new round of activity: the first took place in the context of a conference on transition and consolidation in Latin America and Southern Europe, organized by Guillermo O’Donnell and Philippe Schmitter at the Kellogg Institute of the University of Notre Dame, in April 1987. There, a small group consisting of Maria Carrilho, P. Nikiforos Diamandouros, Richard Gunther, Juan Linz, and Philippe Schmitter explored available options for a new initiative and extensively deliberated over the requisites and content of its intellectual agenda. This was followed by a second crucial planning session, made possible through Schmitter’s

Preface and Acknowledgments

vii

energies and capacity to secure the required financial support, which brought together a larger number of European and American specialists on Southern Europe to the magnificent setting of the European University Institute, in San Domenico di Fiesole, near Florence, in September, 1987. The principal outcome of this meeting was a decision to approach the Joint Committee on Western Europe of the ACLS and the SSRC with the idea of their sponsoring this activity. The Joint Committee’s prior interest in democratization and regime transition in Southern Europe made it an appropriate vehicle for a larger undertaking centered on democratic consolidation. Endorsing the idea for such an initiative, the Joint Committee, in December 1987, proposed to the boards of the ACLS and SSRC the establishment of the Subcommittee on the Nature and Consequences of Democracy in Southern Europe, which, following approval, formally came into being in April 1988. Co-chaired by P. Nikiforos Diamandouros (University of Athens) and Richard Gunther (Ohio State University), the Subcommittee consisted of the following members, drawn from Greece, Italy, the Federal Republic of Germany, Portugal, Spain, the United Kingdom, and the United States: Manuel Villaverde Cabral (University of Lisbon), Maria Carrilho (Advanced Institute of Labor and Enterprise Sciences, Lisbon), Juan J. Linz (Yale University), Carmelo Liso´n Tolosana (University of MadridComplutense), Edward E. Malefakis (Columbia University), George Th. Mavrogordatos (University of Athens), Jose´ Ramo´n Montero (Autonomous University of Madrid), Leonardo Morlino (University of Florence), Gianfranco Pasquino (University of Bologna), Vı´ctor Pe´rezDı´az (Juan March Center for Advanced Study in the Social Sciences, Madrid), Hans-Ju¨rgen Puhle (University of Bielefeld, now University of Frankfurt am Main), Sidney Tarrow (Cornell University), and Loukas Tsoukalis (St. Antony’s College, Oxford, now University of Athens). Hans-Ju¨rgen Puhle played a particularly important role over the following decade, steering the Subcommittee’s project through the German foundation world, and providing sage advice at crucial intervals. The Subcommittee’s charge was dual: ‘(1) to engage in a systematic study of the nature of democratic consolidation in Greece, Portugal, Spain, and post-Fascist Italy, by exploring its cultural, economic, political, and social dimensions; and (2) to use the insights derived from this regional case study to contribute to the emergent, more general, theoretical debate concerning the properties of, and processes involved in, the consolidation of democracy’. The interrelated dimensions of change in the New Southern Europe which the Subcommittee was to explore were five: processes of democratic consolidation, the nature of

viii

Preface and Acknowledgments

democratic politics, economic and social relations in the region, the changing functions of the state, and the dynamics of cultural change. The final step toward the launching of this ambitious project came in May 1988, when the Stiftung Volkswagenwerk (now VolkswagenStiftung), underscoring its role as a preeminent European institution committed to the fostering of social science knowledge across national borders, agreed to provide generous support for a multiyear program of research, conferences, and publication activities designed to produce five collective volumes dealing with the interconnected dimensions of change mentioned above. With additional support from the Werner Reimers-Stiftung and the German Social Science Study Group on Spain and Portugal, the project was informally launched in Bad Homburg, Germany, in July 1989, where specific conference and publication plans were made. Volume 1 of the series (The Politics of Democratic Consolidation, edited by Richard Gunther, P. Nikiforos Diamandouros, and HansJu¨rgen Puhle) is the product of a conference held in Madrid at the Juan March Center for Advanced Study in the Social Sciences in July 1990, and of a follow-up authors’ meeting hosted by the Department of Sociology, University of Rome, in December of the same year. The second volume (Parties, Politics, and Democracy in the New Southern Europe, edited by Diamandouros and Gunther) was made possible by an initial meeting held in July 1991 at the European Cultural Center in Delphi, Greece, and a March 1992 follow-up authors’ conference in The Hague sponsored by the Institute Clingendael. Volume 3 (Economic Transformation, Democratization, and Integration Into the European Union: Southern Europe in Comparative Perspective, edited by Heather D. Gibson) also issued from a twin set of conferences: the first was held in Sintra, Portugal, in July 1992, while the April 1993 follow-up meeting was hosted by the Levi Foundation in Venice. The initial conference leading to Volume 4 (this book) was hosted by the Zentrum fu¨r interdisziplina¨re Forschung of the University of Bielefeld, Germany, in July 1993, and the authors’ meeting was held at Bog˘azic¸i University, Istanbul, in July 1994. Finally, Volume (forthcoming) 5 originated in two conferences held in Greece and Spain. The initial meeting was held in the magnificent setting of Elounda, in Eastern Crete, Greece, in July 1995. This meeting also featured the last formal meeting of the Subcommittee on Southern Europe. The authors’ conference for this volume was convened in Mallorca, Spain, in July 1996. It is hoped that these conferences and resulting published volumes will fulfill the Subcommittee’s mandate, which was to more fully integrate Southern Europe into the mainstream of comparative social

Preface and Acknowledgments

ix

science research, and to derive from the study of its democratization experience insights capable of contributing to the fast-growing body of theoretical literature on democratization.

Acknowledgments A more complete recognition of the numerous intellectual, organizational, financial, and administrative contributions to this project has been gratefully recorded in the Acknowledgments section of volume 1 (Gunther, Diamandouros, and Puhle 1995: xxxi–xxxiii). Here, we shall confine ourselves to thanking all those individuals and institutions who contributed to the birth of this volume. Foremost among these is Hans-Ju¨rgen Puhle, who played the central role in organizing the meeting at which initial drafts of these papers were presented and discussed, and in securing generous funding from the VolkswagenStiftung and the Center for Interdisciplinary Research of the University of Bielefeld. Kent Worcester, at that time program director at the Social Science Research Council, also provided invaluable assistance in making possible this successful conference, as well as the authors’ meeting the following year in Istanbul. We gratefully acknowledge the kind collaboration and generous support of the Department of Social Sciences, University of the Bosphorus, which hosted that follow-up meeting. We should also like to express our deep appreciation to Dr Helga Junkers of the VolkswagenStiftung for supporting our efforts and ensuring the continuous flow of the financial support that have made both this volume and the larger project of which it is a part possible. In addition to the always-insightful comments of the members of the Subcommittee who attended the Bielefeld meeting, the intellectual and scholarly criticisms and advice offered by our invited commentators also greatly strengthened this volume. These valued contributors include Thomas C. Bruneau, John Goldthorpe, John Laughlin, Wolfgang Merkel, Antonio Vitorino, and John Yfantopoulos, at the Bielefeld meeting, and, in Istanbul, Faruk Birtek, Giuseppe DiPalma, Gøsta Esping-Andersen, Claus Offe, Ilkay Sunar, and the late Vincent Wright. We are also most grateful to Lawrence Whitehead for his encouragement and inclusion of this volume into his series, Oxford Studies of Democratization, and to Dominic Byatt, chief editor at Oxford University Press, for his thoughtful suggestions and support for the publication of this volume.

x

Preface and Acknowledgments

Finally, we are extremely appreciative of the generous financial support from the VolkswagenStiftung and the Mershon Center of the Ohio State University, which made this and other volumes in the series on The New Southern Europe possible. Richard Gunther and P. Nikiforos Diamandouros Editors of The New Southern Europe series

................ Contents ................ List of Figures List of Tables Contributors 1. Introduction: Democracy and the State in the New Southern Europe P. Nikiforos Diamandouros, Richard Gunther, Dimitri A. Sotiropoulos, and Edward E. Malefakis 2. The Welfare State and Democracy: On the Development of Social Security in Southern Europe, 1960–90 Francis G. Castles 3. Social Policy, Democracy, and Citizenship in Southern Europe Marisol Garcı´a and Neovi Karakatsanis 4. Democratic Consolidation, Judicial Reform, and the Judicialization of Politics in Southern Europe Pedro C. Magalha˜es, Carlo Guarnieri, and Yorgos Kaminis 5. Old Problems and New Challenges: The Enduring and Changing Functions of Southern European State Bureaucracies Dimitri A. Sotiropoulos 6. Multilevel Governance and the Transformation of Regional Mobilization and Identity in Southern Europe, with Particular Attention to Catalonia and the Basque Country Iva´n Llamazares and Gary Marks 7. The Environment, Socioeconomic Transformation, and Political Change in the New Southern Europe Geoffrey Pridham and Jose´ Magone

xiv xv xvii

1

42

87

138

197

235

263

xii

Contents

8. Democratic and Economic Consolidation in Southern Europe Stavros B. Thomadakis

305

9. Conclusion Richard Gunther and P. Nikiforos Diamandouros

338

References Index

363 395

This page intentionally left blank

................ List of Figures ................ 2.1. Social security transfers as percentage of GDP 2.2. Southern European social security transfers as percentage of GDP

49 53

................ List of Tables ................ 2.1. 2.2. 2.3. 2.4. 2.5. 2.6. 2.7. 2.8. 2.9. 2.10. 3.1. 3.2. 3.3. 4.1.

4.2. 4.3. 4.4. 6.1. 6.2. 6.3. 7.1.

7.2.

7.3.

Social security development and democratic transitions OECD transfers and democratization OECD transfers and socioeconomic variables Growth of OECD transfers and socioeconomic variables, 1960–90 Transfers and socioeconomic variables OECD transfers and political and policy variables Transfers, party politics, and policy OECD transfers and family of nations Transfers expenditures OECD transfers, 1960 and 1990 Unemployment rates in Southern Europe, 1999 Young people residing at home in Spain Female labor force participation rates in Southern Europe The institutional outcomes of judicial reform in the Southern European democratic transitions: rules of appointment and supervision of the career judiciary Judicial review of legislation in Southern Europe The composition of constitutional courts in Southern Europe Abstract a priori review in Portugal Identities in the Basque Country Identities in Catalonia Percentages of respondents in the Basque Country and Catalonia with multiple identities Percentages of respondents regarding the environment and fighting pollution as an urgent problem and importance of environment in relation to economic and social policies Distribution of materialist- and post-materialist values in Southern Europe and other EU countries, 1986–7 and 2000 Percentage of environmental directives translated into national law, by country and year

58 60 62 64 65 68 69 75 77 81 96 108 113

156 162 164 169 254 254 255

271

272 299

xvi

List of Tables

7.4. Noncommunication cases, nonconformity cases, bad application of horizontal cases, 2000–1 and 2003 8.1. Economic performance, 1960–79 8.2. Government expenditures as percentage of GDP, 1973–9 8.3. Tax and social security receipts as percentage of GDP 8.4. Indices of economic performance, 1980–99 8.5. Fiscal structures 8.6. General composition of receipts, 1980–2002 8.7. Tax composition

300 310 314 317 323 328 329 332

................ Contributors ................ Francis G. Castles is professor of Social and Public Policy in the School of Social and Political Studies of the University of Edinburgh, and adjunct professor of Political Science in the Research School of Social Sciences at the Australian National University. He has published nearly 20 books and over 100 articles on comparative public and social policy, focusing on the determinants of welfare state and public expenditure outcomes in OECD countries. His recent books include Comparative Public Policy (1998), The Polity Welfare State Reader (coeditor, 2000), Reshaping Australian Institutions (ed. with Geoff Brennan 2002), The Future of the Welfare State: Crisis Myths and Crisis Realities (2004), and (with Herbert Obinger and Stephan Leibfried) Federalism and the Welfare State: New World and European Experiences (2005). P. Nikiforos Diamandouros is professor of Comparative Politics (on leave of absence) in the Department of Political Science and Public Administration, University of Athens. Since 2003, he has been the European Ombudsman, having served (between 1998 and 2003) as the first National Ombudsman of Greece. He has also taught at the State University of New York and the Juan March Center for Advanced Studies in the Social Sciences (Madrid), and held a research appointment at Columbia University. His recent works include The European Ombudsman: Origins, Establishment, Evolution (ed. 2005), Hoi aparches syngrotises synchronou kratous stin Ellada, 1821–1828 [The Origins of State-Building in Modern Greece, 1821–1828] (2002), Parties, Politics and Democracy in the New Southern Europe (ed. with Richard Gunther 2001), Politismikos dyismos kai politike allaghe stin Ellada tis Metapolitefses [Cultural Dualism and Political Change in Postauthoritarian Greece] (2001), and The Politics of Democratic Consolidation: Southern Europe in Comparative Perspective (1995). Marisol Garcı´a is associate professor of Sociology of the University of Barcelona. Her recent publications include Ciudadanı´a: justicia social, identidad y participacio´n (with Steven Lukes 1999), ‘Minimum Income Policies to Combat Poverty: Local Practices and Social Justice in the ‘‘European Social Model’’ ’ in Yuri Kazepov (ed.), Cities of Europe (2005), ‘Barcelona: Governing Coalitions, Visitors and the Changing City Center’ in Lily M. Hoffman, Susan S. Fainstein, and Dennis R. Judd (ed.), Cities and Visitors (with Nuria Claver 2003), ‘Why Some People Are More Likely to be on Social Assistance Than Others’ in Chiara Saraceno et al. (eds.), Social Assistance Dynamics in Western Europe (with Yuri Kazepov 2002), and ‘The Inclusive Power of Standard and Non-Standard Work’ in Rik van Berkel and Iver Hornemann (eds.), Active Social Policy in Europe (with Jan de Schampheleire 2002).

xviii

Contributors

Carlo Guarnieri is professor of Political Science at the University of Bologna, where he has also served as Dean of the Faculty of Political Science. He has written extensively on the relationship between courts and politics in democratic regimes. He is author of La giustizia in Italia (2001, published in Spanish as Judicialismo, 2003), Giustizia e politica I nodi della seconda Repubblica (2003), Los jueces y la polı´tica (with Patrizia Pederzoli 1999), The Power of Judges: A Comparative Study of Courts and Democracy (2002), L’indipendenza della magistratura (1981), Pubblico ministero e sistema politico (1984), and Politics in Italy 2005 (ed. with James Newell 2005). Richard Gunther is professor of Political Science at the Ohio State University. He has served as Executive Director of International Studies at OSU and as cochair (with Nikiforos Diamandouros) of the Subcommittee on Southern Europe of the American Council of Learned Societies and the Social Science Research Council. His recent publications include Democracy in Modern Spain (with Jose´ Ramo´n Montero and Joan Botella 2004), Political Parties: Old Concepts and New Challenges (ed. with Jose´ Ramo´n Montero and Juan Linz 2002), Political Parties and Democracy (ed. with Larry Diamond 2001), Parties, Politics and Democracy in the New Southern Europe (ed. with Nikiforos Diamandouros 2001), Democracy and the Media (ed. with Anthony Mughan 2000), The Politics of Democratic Consolidation (ed. with Nikiforos Diamandouros and Hans-Ju¨rgen Puhle 1995), and Elites and Democratic Consolidation in Latin America and Southern Europe (ed. with John Higley 1992). Yorgos Kaminis is assistant professor of Constitutional Law at the University of Athens. He has also worked as legal expert at the Scientific Department of the Greek Parliament, and as of 2003 holds the position of Greek Ombudsman. His publications include La transition constitutionnelle en Gre`ce et en Espagne (1993), Paranoma apodiktika mesa kai syntagmatiki katochirosi ton atomikon dikaiomaton [Illegally Obtained Evidence and Constitutional Guarantees of Individual Rights] (1998), and ‘He symmorfose tis Elladas stis apofaseis tou Evropaikou Dikastiriou Dikaiomaton tou Anthropou’ [‘Greece’s Compliance with the Decisions of the European Court of Human Rights’] in To Syntagma (2, 2003). Neovi Karakatsanis is associate professor of Political Science at Indiana University South Bend. Her publications include The Politics of Elite Transformation: The Consolidation of Greek Democracy in Theoretical Perspective (2001), articles in Armed Forces and Society, South European Society and Politics, The Greek Review of Social Research, and The Classical Bulletin, as well as a number of edited volumes. She is also an officer in the Modern Greek Studies Association, and the Research committee on Armed Forces and Society of the International Political Science Association. Iva´n Llamazares is associate professor of Political Science at the Universidad de Salamanca. His main interests include political parties and party systems in Europe and Latin America. He has published articles in Social Forces, South European Society and Politics, Revista de Estudios Polı´ticos and Revista de

Contributors

xix

Instituciones Europeas, as well as chapters in several edited volumes. He has edited (with Fernando Reinares) Aspectos polı´ticos y sociales de la integracio´n europea (1999). Pedro C. Magalha˜es is researcher at the Social Sciences Institute of the University of Lisbon. He is one of the coordinators of the Portuguese Election Study. His research interests include comparative judicial politics, judicial behavior, public opinion, and electoral behavior. His book publications include O parlamento portugueˆs: uma reforma necessa´ria (with Anto´nio de Ara´ujo, Cristina Leston-Bandeira, Andre´ Freire, and Marina Costa Lobo 2002), A abstenc¸a˜o eleitoral em Portugal (with Andre´ Freire 2002), Portugal a votos: as eleic¸o˜es legislativas de 2002 (ed. with Andre´ Freire and Marina Costa Lobo 2004), and Portugal: democracia y polı´tica (ed. with Anto´nio Barreto and Braulio Go´mez Fortes 2003). He has also published articles in Comparative Politics, West European Politics, International Journal of Public Opinion Research, South European Society and Politics, and the Journal of Southern Europe and the Balkans. Jose´ M. Magone is senior lecturer in European Politics at the University of Hull. He is a specialist in the politics of Southern Europe and the European Union. Among his publications are Iberian Trade Unionism: Democratization under the Impact of the European Union (2001), The Politics of Southern Europe: Integration into the European Union (2001), Contemporary Spanish Politics (2004), The Developing Place of Portugal in the European Union (2004), and The New World Architecture: The Role of the European Union in the Making of Global Governance (forthcoming 2006). Edward E. Malefakis is professor Emeritus of Modern European History at Columbia University. He has taught at Columbia University, the University of Michigan, and Northwestern University. He is the recipient of the herbert Baxter Adams Prize of the American Historical Association, of the Premio Internacional Nebrija for lifetime contributions to Spanish Studies, and of a Guggenheim Fellowship. His most important work is Agrarian Reform and Peasant Revolution in Spain: Origins of the Civil War (1970). He is also the author of La Guerra de Espan˜a 1936–1939 (1996), and has written extensively on various aspects of nineteenth- and twentieth-century Spanish History. Gary Marks is Burton Craige Distinguished professor of Political Science at the University of North Carolina, Chapel Hill, and Professor in the Chair in Multilevel Governance at the Free University, Amsterdam. His recent books include European Integration and Political Conflict (ed. with Marco Steenbergen 2004), Multi-level Governance and European Integration (with Liesbet Hooghe 2001), and It Didn’t Happen Here: Why Socialism Failed in the United States (with Seymour Martin Lipset 2000). Geoffrey Pridham is professor of European Politics at the University of Bristol. He has written widely on problems of democratic transition and consolidation, both theoretically and empirically, initially with respect to Southern

xx

Contributors

Europe and then increasingly Central and Eastern Europe. Among his recent publications are The Dynamics of Democratization: A Comparative Approach (2000) and Designing Democracy: EU Enlargement and Regime Change in Post-Communist Europe (2005). His current research concerns the continuing democratizing impacts of EU conditionality on post-communist politics, especially in Slovakia, Latvia, and Romania. Dimitri A. Sotiropoulos is associate professor of Political Science and Public Administration at the University of Athens. His most recent publications include The Unknown Civil Society (ed. with V. Voutsakis, J. Kallas, and N. Stylianidis 2004), ‘South European Bureaucracies in Comparative Perspective’ in West European Politics (2004), ‘The EU’s Impact on the Greek Welfare State’, in Journal of European Social Policy (2004), and ‘Positive and Negative Social Capital and the Uneven Development of Civil Society in Southeastern Europe’, in Southeast European and Black Sea Studies (2005). Stavros B. Thomadakis is professor of Financial Economics and Graduate Program Director in Applied Economics and Finance in the Department of Economics, University of Athens. He has also been Professor of Financial Economics at Baruch College and the City University of New York, and Visiting Professor at MIT’s Sloan School of Management. He has published many scholarly articles and books on banking, capital markets, regulation, and industrial economics, and has written extensively on the political economy and recent economic history of Greece. Between 1996 and 2004, he served as chair of the Capital Market Commission of Greece, and has recently been appointed chair of the Public Interest Oversight Board.

................ 1 ................

Introduction: Democracy and the State in the New Southern Europe P. Nikiforos Diamandouros, Richard Gunther, Dimitri A. Sotiropoulos, and Edward E. Malefakis

The full title of the research project sponsored by the Social Science Research Council and the American Council of Learned Societies, of which this volume is a product, is ‘The Nature and Consequences of Democracy in the New Southern Europe’. The selection of this title is fortuitous insofar as it nicely summarizes and combines the two principal foci of our collaborative research effort. The first of these involves democratization, and the second encompasses the much broader cluster of cultural, economic, political, and social-structural transformations that have moved Greece, Italy, Portugal, and Spain from the ‘semiperiphery’ of Europe to the mainstream of advanced, affluent, and democratic postindustrial countries of the European Union (EU). While the various volumes of the series focus primarily on one or another of these facets of sociopolitical transformation, they are all rooted in a recognition that these various dimensions of cultural, economic, social, and political change are closely interrelated in sometimes complex ways. Volume 1 of this series, The Politics of Democratic Consolidation, focused its analysis on the processes of democratic transition and the consolidation of new democratic regimes. In contrast with the dynamics of politics that had characterized the four countries of Southern Europe over the previous century-and-a-half, stable, consolidated democratic regimes have been established in each of these four countries. Accordingly, democratization served as a dependent variable in the analyses published in that volume. Once established, however, these consolidated democratic regimes began to function as an independent variable significantly affecting subsequent developments with

2

Introduction

regard to politics, economics, public policy, and the state—themes that, in turn, constitute the principal foci of the ensuing four volumes in this series on the New Southern Europe. The second volume in the series, Parties, Politics and Democracy in the New Southern Europe, analyzed in detail the development and evolution of political parties, the basic characteristics of electoral behavior, and the dynamics of democratic politics in Greece, Italy, Portugal, and Spain within the context of consolidated democratic regimes. Volume 3, Economic Transformation, Democratization and Integration into the European Union (edited by Heather Gibson), analyzed in detail the transformation of the Southern European economies, particularly with regard to the manner in which these economic changes related to the integration of all four Southern European countries into what is today the EU. In this fourth volume in the series, we explore the various ways in which Southern European states and several of their key public policy domains have been transformed in recent decades. Over the course of these collaborative studies, several recurring themes have been explored. The first and most basic is that Southern European ‘exceptionalism’ has been progressively overcome, and politics, economic relations, and fundamental features of social structure have come to resemble more closely those of other Western and Northern European countries. With regard to political regimes and the fundamental dynamics of politics, all four Southern European countries had political traditions that were largely devoid of extended periods of stable democratic governance. At various points in their political development since the mid nineteenth century, each of them had experienced considerable political instability, and had suffered under sometimes protracted periods of authoritarian rule. As Volume 1 demonstrated, however, the democratization processes that unfolded in Italy from the 1940s through the 1970s, and in Greece, Portugal, and Spain in the 1970s and early 1980s, appear to have eliminated all residual traces of this nondemocratic legacy, and the consolidation of democracy throughout the region is as solid as in other parts of Western and Northern Europe. In short, the political exceptionalism of the region has given way to a progressive approximation of more widespread European norms of democratic governance. The issue of democratic stability has been definitively resolved, for the first time in the region’s history, and the democratic regimes that were established in the aftermath of authoritarian interludes are now as fully legitimate as are the long-established democracies of the West. In the cases of Greece and Spain, democratic consolidation was achieved in a remarkably short period of time (between five and seven years). The revolutionary turmoil of the Portuguese revolution somewhat complicated these pro-

Introduction

3

cesses of political change, regarding both the completion of the transition to democracy and its consolidation. Nonetheless, by the late 1980s all questions concerning the legitimacy of Portuguese democracy were completely resolved. In the case of Italy, whose democratic transition was accompanied by the outbreak of the Cold War and its domestic ramifications, democratic consolidation required a full three decades. But by the end of these processes of regime transformation, all four countries took their places among the ranks of fully consolidated and stable democracies. Volume 2 found that in some ways there was also an approximation of key features of competitive democratic politics characteristic of Western and Northern Europe, but, at the same time, there was no convergence on a single model of Southern European politics. In fact, partisan institutions, electoral mobilization strategies, the dynamics of electoral competition, the basic techniques of campaigning, and patterns of mass level electoral behavior are about as diverse in Southern Europe as in other, more long-established democratic systems. This is particularly true with regard to the fundamental features of core political institutions—with the four countries spanning the full range from constitutional monarchy to parliamentary system to semipresidentialism, from centralism to political decentralization, from ‘majoritarianism’ to ‘consensualism’—and with a variety of electoral laws having variable impacts on the structure and dynamics of party systems and, in turn, on government formation. Volume 2 did identify one important similarity among these four cases: various manifestations of the exceptionalism of previous periods have progressively disappeared. In contrast with previous democratic and semidemocratic interludes, when parties of both left and right maintained antisystem or semiloyal orientations toward the regime, the party systems of Greece, Italy, Portugal, and Spain are devoid of antisystem or semiloyal parties (partial exceptions being the increasingly irrelevant Communist Party of Portugal and certain Basque nationalist parties in Spain). Similarly, while polarization had characterized partisan conflict in the past, often with regime-destabilizing consequences, competition among the major parties today is characterized by increasing moderation and centripetal drives that further enhance the stability of these democratic systems. Finally, party systems under several of the previous democratic or semidemocratic systems in these countries had been characterized by fragmentation, which in turn contributed to high levels of cabinet instability. Instead, there has been an increasing tendency toward majoritarianism in all four countries, combined with clear patterns of alternation in government between the major parties of the left and right (which, in the case of Italy, also contrasted with the centrist

4

Introduction

immobilism that characterized the four-and-a-half decades of the socalled ‘first republic’). One final significant change in the domain of partisan politics does not relate to Southern European exceptionalism, but, rather, to a transformation that can be observed in most Western European democracies. This entails the shift from mass-based parties (particularly of the socialist or Christian Democratic variety) to organizationally thin parties which have embraced the highly personalized catch-all electoral strategies that are often regarded as making up a syndrome of ‘the new campaign politics’ (Pasquino 1995). Overall, it is clear that substantial transformations of partisan politics have occurred in all four countries in the region. We also argued in Volume 2 that the transformation of partisan politics took place more quickly and has been more profound in Greece, Portugal, and Spain than in long-established democratic systems as a result of what we referred to as ‘leapfrogging’ processes of change. In these three countries, the establishment of new party organizations following the suspension of partisan politics under the preceding authoritarian regimes occurred in the absence of the institutionalized legacy of older forms of party organization and electoral behavior. The absence of these institutionalized legacies enabled these countries to pass over developmental stages and organizational forms of earlier eras that had become entrenched features of many parties in the longer-established democracies of Western and Northern Europe. Accordingly, and somewhat paradoxically, the new democracies of Southern Europe, unencumbered by these institutionalized legacies of the past, more quickly and thoroughly took on features of the late twentieth century version of ‘modernity’ than have many previously established democratic systems. In the case of Italy, however, this leapfrogging process of change occurred only in the 1990s. The party system of the first republic was founded in the late 1940s, when older mass-based party models were still dominant, and prior to the advent of television as the primary medium of communication between party elites and voters. Accordingly, parties based upon these older models were fully institutionalized and remained largely unchanged over the following four decades, despite substantial changes in Italian society (such as secularization and far-reaching economic development) that were occurring during this period. It was only after these parties were completely discredited by the mani pulite scandals and began to disintegrate that far-reaching change was possible. It was only after the electoral earthquakes of the early 1990s finally succeeded in erasing those established parties from the political scene that Italian partisan politics leapfrogged (albeit somewhat belatedly) to a full and complete embrace of new party models and the new campaign politics.

Introduction

5

The economies of Southern Europe provide even clearer evidence of leapfrogging insofar as the delayed processes of economic change in Greece, Portugal, and Spain transformed predominantly preindustrial (if not premodern) economies into postindustrial ones (with huge tertiary sectors), without ever having passed through the stage of full industrialization. In detailed analyses of the economies of the region, Volume 3 found that the economies of Southern Europe, which did exhibit strong intraregional similarities (particularly with regard to the predominance of inefficient parastate sectors), have had to confront the imperatives of increased efficiency and competitiveness posed by the challenges of European integration. These four countries, however, often adopted different approaches to these challenges, and have met with varying degrees of success. Again, these evolutionary processes can be characterized by the term ‘approximation’, rather than ‘convergence’, particularly since each of these four countries retains sectors that have resisted market-driven pressures toward greater efficiency and modernization. Social services and, more generally, the welfare state sector, as well as the public administration, the health sector, and education constitute salient examples of state subsystems where efficiency and modernization have only managed to make belated and, on the whole, weak inroads. The capacity of these sectors of the state to effectively resist change and to retard the impact of modernization rests on the ability of powerful social coalitions of diverse compositions, ranging from educated but nonpropertied middling strata to well-entrenched workers in heavily restricted professions, to defend privileges derived from premodern and highly regulated economies closely associated, as we see, with predemocratic and antidemocratic periods in the histories of these states. In this fourth volume of the series, we explore basic characteristics of the state in these four Southern European countries, as well as important features of their public policy outputs. Again, we begin with a picture of Southern European exceptionalism. In several ways, important sectors of the states of Southern Europe shared certain similarities that differentiated them from other Western and Northern European counterparts. While each of the four Southern European states had its own distinctive features, they all shared certain common characteristics best captured by the seemingly paradoxical terms, ‘heavy state’ and ‘weak state’. Similarly, some public policy outputs of these countries were strikingly different from those produced by more well-established democratic systems: in particular, the overall magnitude of public policy outputs (especially with regard to social welfare policies) was substantially smaller in Greece, Portugal, and Spain than was typical of other Western industrialized societies. In part, this was the

6

Introduction

result of highly inefficient structures of taxation, which provided inadequate volumes of revenue to support expanded state activities. As we see in the following chapters of this book, the Southern European state and its predominant patterns of public policy outputs have changed substantially over the past few decades. In sharp contrast with processes of change in the politics and economics of the countries in this region, however, we see no evidence of leapfrogging with regard to the transformation of Southern European state or its public policy outputs. While we observe that certain sectors or subsystems of the state have been more substantially transformed than others, we, as noted above, also see that institutionalized legacies inherited from the authoritarian and preauthoritarian periods of the past have served as important obstacles to change. To put it in the language of historical institutionalism, we see that change is path dependent. As a result, in these specific sectors the legacies of past exceptionalism persist to this day, and processes of change and reform remain incomplete.

The Theoretical Framework Over the last thirty years, studies of the state have moved in a tangential fashion. Researchers have touched on different aspects of state structures, functions, and transformations, but have rarely shared the same substantive focus or theoretical framework. In the 1980s and the 1990s, earlier debates between pluralists and Marxists, as well as among different currents of neo-Marxism on the nature and the functions of the state, gave way to diverging theoretical approaches. These shared little common ground which would allow for a debate to take place. While some of the new approaches emphasized more diffuse conceptions of state power (Foucault 1980), other approaches saw the state as one of the colonizing forces of the ‘lifeworld’ (Habermas 1987). At the same time, the call to ‘bring the state back in’ (Evans, Rueschemeyer, and Skocpol 1985) made students of the state sensitive to the autonomous function of state structures. The study of the state independent of conflicting social interests prepared the way for the renewed study of institutions in general (March and Olsen 1984, 1989). Institutionalist approaches, which have been popular among political scientists since at least the late 1980s (Hall and Taylor 1996), are not only about state institutions, such as the justice system and the public bureaucracy discussed in Chapters 4 and 5 of this volume, but also about other structures, such as labor markets, welfare regimes, or nongovernmental organizations. Historical institutionalism, in particular, claims that public policy

Introduction

7

outputs are the result of prior experience of institutional actors with related policies as well as the ideas and interests of those actors (Hall 1986, 1989; Peters 1996: 210). Past decisions and arrangements shape future trajectories in the sense that existing commitments may ‘lock-in’ the moves of policymakers (Pierson 1994). In other words, change is limited by path dependency. However, policy outputs, while constrained by past institutional choices, are not solely the result of institutions. Policy outputs reflect political processes and the influence of political forces, which are structured by existing institutional arrangements (Thelen and Steinmo 1992: 3; Rothstein 1996: 142). With regard to state institutions, comparative political research has moved in different directions, four of which stand out: the study of the determinants of public policy; new modes of governance (as opposed to government); changes in national states brought about by globalization and, as far as Western and Southern Europe are concerned, by Europeanization; and the distinction between ‘strong’ and weak states with regard to corresponding state capacities. In an effort to assess ‘what the state does’ rather than ‘what it is’, some analysts of public policy have studied ‘policy communities’ and, later on, ‘policy networks’ involved in policymaking and policy implementation (Rhodes and March 1992; Smith 1993: 76–85). Both the more cohesive ‘communities’ and the looser ‘networks’ do not only include state actors, but also professional associations, academic and research institutions, and other nonstate collective actors, as well as influential individuals such as experts and interest group representatives. On the other hand, there is a shift to more decentralized forms of state organization, such as the creation of regional authorities and multilevel governance in Southern Europe (see Llamazares and Marks, this volume), and an increasing transfer of competencies to supranational authorities, such as the EU. These parallel developments and other changes are reflected in the substitution of governance for government as a major topic of research (Pierre and Peters 2000). The substitution reflects the tendency for decision-making power to leak away from the central government—that is, from state elites—and to be dispersed among expanding circles of state and non-state actors. The state has been ‘hollowed out’ by granting powers and assigning public resources (such as funds and personnel) down to subnational authorities and up to supranational entities, as well as by privatizing parts of its assets. Its new role is not so much to intervene in social and economic relations, but to facilitate and set the stage for the development of such relations. Hence, the state enables rather than determines social action. It becomes a ‘regulatory state’ (Majone 1999).

8

Introduction

The emergence of hollowed-out states is to some extent the result of globalization and Europeanization. Academic interest in these trends has not left the study of the state unaffected. Because of the ascendency of neoliberalism and the pressures of globalization, there has been a debate about whether the state is in retreat (Strange 1996; Evans 1997; Mann 1997), particularly with regard to integration into the EU. Europeanization is seen as a wide-ranging process affecting national politics and policymaking, discourse, identities, political structures, and public policies (Featherstone and Radaelli 2003), and its effects are multiple and ongoing. We refer to some of these impacts toward the end of this chapter, although an adequate mapping of the effects of Europeanization on the states and democracies of Southern Europe would require another edited volume (such as Gibson 2001a, the third volume in this series). Finally, it is obvious that the empirical referents of the above theoretical approaches to the state are the national states of advanced industrial societies of the late twentieth century. In comparative historical studies of states in the West, as well as in the study of developing areas today, there is an emphasis on state strength and state weakness. State strength is understood in terms of state capacities, such as the differential capacity of states for policy innovation (Hall 1986; Weiss 1998: 24–40). Theoretically, strong states are those which have the capacity to impose and implement their objectives despite resistance from other powerful sectors of society (Skocpol 1985). Many indicators of state strength have been proposed (Holsti 1995: 330–4; Fukuyama 2004: 9–19). Conceptualizations of state strength are often used to measure the capacities not only of advanced industrial states but also of Third World states (Migdal 1988: 18–19; Evans 1995). In our study of Southern Europe, state strength refers to the capability to effect welfare reform (Castles in Chapter 2, and Garcı´a and Karakatsanis in Chapter 3 of this volume), to change the judicial system after the transition to democracy (Magalha˜es, Guarnieri, and Kaminis in Chapter 4), to democratize and modernize the public bureaucracy (Sotiropoulos in Chapter 5), to formulate and implement environmental policies (Pridham and Magone in Chapter 7), and to consolidate public finance (Thomadakis in Chapter 8). In the introduction of this volume, our theoretical framework is eclectic. It draws on old and new theories of the state and is probably more akin to historical institutionalist theories of the state. With regard to Southern Europe, we do not see the state in the manner of either postwar pluralist approaches, as an agency simply reacting to the influences of political parties and interest groups, or dogmatic Marxist approaches which reduce the state to an agency fulfilling the

Introduction

9

requirements of capitalist development. Finally, while we speak of ‘the state’, we do not conceive of it as a single entity. Rather, the state is a set of organizations and institutions, which, while guided by state elites, are placed at the intersection between the national and the international structures, and exercise supreme legal-bureaucratic authority within a defined territory (Katzenstein 1978; Krasner 1978; Evans, Rueschemeyer, and Skocpol 1985).

The Traditional Southern European State In order to fully appreciate the magnitude of state-structural and policy changes that have taken place in Southern Europe over the past three decades, and to better understand the difficulty and complexity of these transformations, let us turn our attention to a historical overview of the traditional characteristics and shortcomings of the state in Southern Europe. For the purposes of this analysis, which is informed by historical institutionalism, the temporal scope of this brief historical overview runs from the 1790s to the mid-1970s—that is, from the French Revolution and the Napoleonic wars to the nearly simultaneous fall of the Portuguese, Greek, and Spanish dictatorships. The 1790s initiate a profound, decades-long process of rupture with the ancien regime, to which the Industrial Revolution would soon contribute. The mid-1970s signal a moment when the accumulated changes of the post–World War II era had so profoundly altered socioeconomic structures and mental attitudes that fundamental conditions of Southern European life were transformed, and quite different societies from those which had existed only a few decades earlier came into being, in the process confronting the state with multiple challenges for reform, many of which constitute the main focus of this volume. Analytically, a convenient point of departure for our brief examination of the nature and dynamics of the traditional Southern European state is to point to its two enduring but also sharply contrasting structural characteristics, which, in one form or another, lasted well into the second half of the twentieth century, before becoming greatly eroded and undermined (but not completely erased) in the course of the last three to four decades. These characteristics are what we call the ‘heaviness’ and, at the same time, the ‘weakness’ of the traditional state in the region—characteristics which, despite very significant variation in the origins and subsequent state-building experiences observed in each of the four countries, were sufficiently pronounced as to allow us to speak of a shared pattern in an ideal typical sense.

10

Introduction

The ‘Heaviness’ of South European States While the particular attributes of each of these two structural characteristics will be more fully analyzed below, it is important to stress at this point that the notion of heaviness, as conceptually distinct from ‘strength’, implies both the tendency of the traditional Southern European state to be heavy-handed in its relations with society and also, especially in the twentieth century, frequently to resort to the creation of large and extensive structures, with often overlapping functions, in an attempt to cope with recurring problems that for a variety of reasons it failed to address efficaciously. The coexistence of heavy-handed behavior vis-a`-vis society and of new structures side by side with surviving old ones was what imparted to the traditional Southern European state its peculiar character of both heaviness and weakness. Taken together, these two structural characteristics contributed to the ‘impaired functionality’ of the traditional Southern European state, whose capacity to provide for an efficacious political and institutional framework capable of substantively contributing to the development of the societies under its jurisdiction remained limited and contested, and its legitimacy weak and fragile. This overall feature of impaired functionality should be seen as placing the traditional Southern European state near the middle of a conceptual range defined at one end by highly functional states—such as those which, over time, developed in Britain, France, Germany, Japan, Sweden, and the United States—and, at the opposite end, dysfunctional states such as those of late Imperial China, the Ottoman Empire, and much of Africa, Latin America, and the Middle East, whose articulation with their respective societies was highly problematic or negative, and whose legitimacy was very low and precarious. In short, the traditional Southern European state could be thought of as falling in the intermediate category of ‘semifunctional’ states. What were the attributes defining these two structural characteristics of the traditional Southern European state? Let us begin with the state-heavy dimension. One of its most salient features was the great, often extreme, centralization of state power which, over time, prevailed throughout the region. This was especially notable in the larger countries, Italy and Spain, and constituted a radical departure from their long and active traditions of strong regional autonomy, which had its roots in the distant past. In Italy, this tradition traced its origins to the independent city states of the medieval and early modern eras; but it also survived in different form even in those areas which fell under Spanish or Austrian control, due to the separate administrative status

Introduction

11

these regions were allowed to retain. In Spain, several independent kingdoms had existed in medieval times, and though united with Castile beginning in the 1470s, they had continued to enjoy great regional autonomy long afterward. Yet precisely because centrifugal forces were so strong in these two countries, almost all power was centralized in the state in the modern era. One decisive moment in Spain was in the 1710s, when the autonomous status of Catalonia, Arago´n, and the Valencian region was abrogated. A second was in the mid nineteenth century, when the two Carlist bids for greater regional powers (1833– 40 and 1873–6) were defeated, and the autonomy of the Basque country and Navarre was considerably reduced, though not entirely eliminated. The decisive moment in Italy was 1859–61 when, rather than calling a constituent assembly or otherwise negotiating with the various regions being incorporated into the newly formed nation-state, the Piedmontese monarchy simply imposed the laws, institutions, and administrative practices of Piedmont on the rest of the peninsula, without concessions to its many distinct governmental traditions. While neither nation went quite as far as France in wiping out the remnants of its historic territorial units, Spain was divided into fifty and Italy into sixty-nine provinces. None of these were given significant powers either of taxation or administration, and although provincial assemblies existed in each, they were basically advisory bodies to the regional executives (the governor generals in Spain, the prefects in Italy), who were appointed by the central state. Portugal and Greece also had highly centralized systems, with no significant powers being delegated to regional bodies, and with each of the newly created administrative districts under the control of stateappointed officials. The social and political costs were probably less important in these countries, however, since their smaller size meant that centripetal forces were weaker in them. Portugal, the oldest nationstate in Europe, was linguistically and culturally homogeneous from the very beginning, such that the final stage of state-building (the last wave of expulsion of the Moors from the most southern part of the country in the thirteenth century) effectively established a unified nation-state that, with the partial exception of the Azores, has never faced regionalist challenges. In Greece, on the other hand, a centralized model of statebuilding was adopted by an initially absolutist Bavarian monarchy selected by the Great Powers in the heyday of post-Napoleonic reaction to rule over a destitute peasant population, divided by intense and highly conflictual local and regional rivalries and united more by bonds of common religion (Eastern Orthodoxy) than of nationalism. Throughout the region, the jealous guarding of centralized authority did not stop at the regional level, but went on to reject strong local

12

Introduction

government. Spain’s medieval tradition of vigorous municipalities acting in alliance with the Crown had been extinguished under the Habsburgs and early Bourbons. Although one strand of nineteenth-century liberalism advocated greater municipal liberties, the rival moderado current, which became dominant in the 1840s, would remain in effect for more than a century, despite occasional attempts to alter it. Not until the democratic reforms of the post-Franco period did the municipal and regional levels of government regain significance in Spain. Italy granted its municipalities greater taxing powers and somewhat more significant spheres of separate competence, but as in Spain, the state appointed the mayors from among the elected municipal council members, and retained the right to unseat the councils at will. Secretaries of the communes were also appointed by Rome and, again as in Spain, ended up exercising considerable power because their civil servant status made them independent of the local factions. Similar arrangements, mutatis mutandis, prevailed in Portugal and Greece. Why did state power become so centralized in Southern Europe? The prestige of the Napoleonic model, so powerful throughout continental Europe in the early nineteenth century, undoubtedly contributed to this outcome. But the Germanic states, for example, were on the whole able to resist it. More important were internal factors: the fear that the pieces which had just been put together to form Italy would fly off in separate directions, if not tightly bound to it; the identification of the regionalist cause with both leftist and rightist extremism in the Spain of the 1840s; the need to bring under control the strong regional forces that had developed during Greece’s long struggle for independence against the Ottoman Empire; the rivalry between Oporto and Lisbon in Portugal. Once established, the centralization of state power was firmly consolidated during the several, often lengthy, periods of ‘emergency’ or ‘exceptional’ rule that the southern nations experienced in the nineteenth century, before going one step further to the open and usually long-lasting dictatorships of the twentieth century. The frequency of such periods of emergency or dictatorial government was in itself a second salient feature of the heaviness of the traditional Southern European state. Throughout the period under examination and far more often than in Western Europe, the liberal heritage was allowed to lapse in the region, and, whether covertly or overtly, authoritarian governments assumed power. These were most frequent in the Iberian Peninsula, which, aside from the Salazar/Caetano (1928–74) and Franco (1939–75) authoritarian regimes, experienced the Pais (1917–18) and Primo de Rivera (1923–30) dictatorships, as well as several periods of semiauthoritarian rule under purportedly parliamentary governments (in Portugal, 1842–6, 1849–51, 1906–8, and 1914–15;

Introduction

13

in Spain, 1840–3, 1844–5, 1852–4, 1866–8, and 1874–6). Greece endured three open dictatorships (Pangalos 1925–6, Metaxas 1936–41, Papadopoulos/Ioannidis 1967–74), and constitutional rule was little more than a fac¸ade in 1915–17, 1922–4, and 1935–6. Italy experienced one outright dictatorship (Mussolini, 1922–43). But from 1860 to 1866, constitutional guarantees were suspended in the entire Southern half of the Peninsula, and between 1898 and 1900 the country narrowly avoided a reversion to a kind of royal absolutism. Numerous as they are, the above listings do not fully convey the degree to which extraconstitutional government characterized Southern Europe during this period. At several other moments, the state exceeded its prerogatives without going quite as far as in the instances cited. Use of its coercive powers as the need arose against those who threatened to upset the status quo was a third salient feature of state heaviness identified with the traditional Southern European state and linked closely to its centralist tradition. Recourse to coercion formed part of state behavior in both dictatorial moments or in normal times, when the state was nominally bound by the rule of law. This long tradition of severe repression was established in the first half of the nineteenth century, during the especially bitter struggle between absolutism and constitutionalism that characterized the three Latin nations. The chief victims at that time were middle class liberals and democrats, who filled the jails and flooded England and France as refugees. In the latter half of the century, the repression would come to be directed especially against working class protests. To be sure, this latter occurred all over Europe during that time. But in the South its circle of victims was broader, and included many peasants, since agrarian protests continued to be powerful there until well into the twentieth century. The institution whose name became almost synonymous with this tradition of repression was Spain’s Guardia Civil, a national police corps organized in 1844 as part of the military and run by army officers. Lauded by the established classes, it was so widely hated among intellectuals and the masses for the brutal almost mechanical quality of its violence that its abolition became a popular demand of every successful revolutionary movement. But even under the most progressive governments, Madrid tended to be reluctant to weaken its coercive powers, and when competing police forces were created, they soon became almost as unrestrained in their tactics as the Guardia Civil. The Italian gendarmerie never acquired the same nefarious reputation, especially among foreign observers, but despite its sprightly uniform and manners, it resembled the Guardia Civil in its actions as well as in its structure. In Greece, the national police service was not as clearly articulated or autonomous, and seems to have had a higher proportion

14

Introduction

of short-term personnel in its ranks than the other two corps, but this did not make it any the less reluctant to shoot into crowds to disperse them, or to apply unrelieved, grinding pressure on a daily basis against local dissidents. The lack of respect for civil liberties was also prevalent in Portugal, where police brutality was common not only under the Salazar regime but even during the Republic of 1910–26, that country’s supposedly most democratic and idealistic period prior to 1975. Whenever police action proved insufficient, the scale of repression tended to be escalated in all four countries; martial law of varying degrees of severity was proclaimed, and army units were sent in to ‘restore order’. By relying so heavily on physical force, and by using it so consistently in favor of the economically advantaged against the poor, the Southern European state failed miserably in what should have been its mission of incorporating the underprivileged into society. This had serious consequences, among them the fact that it helped prevent labor in Southern Europe from evolving in the same directions it would follow in Western Europe and the United States. Not until the 1950s in Italy, the 1970s in Greece and Spain, and the 1980s in Portugal would moderate trade unionism become the rule in Southern Europe. Prior to those times, anarchism and anarcho-syndicalism had waxed strong in the three Latin nations, especially Spain, where, for much of the period between 1870 and 1936, they clearly constituted the dominant working class forces. In Greece, though a specifically anarchist ideology never took hold, a kind of ‘revolutionary plebiscitarianism’ typified labor conflicts, as workers desperately sought to break out of the cage into which the repressive policies of the state had forced them. There was also an interesting linkage between social repression and regional activism that should not pass unnoticed. Whereas regional autonomists normally demanded greater liberties and the reduction of central state power, they changed their stance whenever social protest seriously threatened regional authority. This pattern was firmly established in Catalonia, above all among members of the conservative Lliga. In 1907–9, 1917–19, 1923, as well as during the Civil War, many Catalanists found the presence of strong state power comforting and welcomed it, although their relations with Madrid were characterized by great pendular swings throughout the first four decades of the twentieth century. But while the Catalanists never allowed social fears permanently to alter their regionalist aspirations, the established classes in Southern Italy in effect abandoned regionalism as a goal in return for constant state protection against social upheavals. The process (immortalized in Giuseppe di Lampedusa’s The Leopard) got underway during Garibaldi’s invasion of Sicily in 1860, and was fully consolidated during the Giolittian era of 1903–14, although the police

Introduction

15

continued to shoot into crowds of protesters in Southern Italy with some frequency, even though repression had eased considerably in the northern and central parts of the country. Indeed, many aspects of this unwritten understanding between Rome and local southern elites continued to characterize the post-World War II Christian Democratic era of political dominance and in some ways perhaps remains in effect even today. A conspicuous feature of the heaviness of the Southern European state in the twentieth century has been its interventionist role in the economy. Italy set the pace, with several hundred firms (some very large), in which the state was a major shareholder, organized under vast state-run holding companies, above all Istituto per la Ricostruzione Industriale (IRI), created in 1933, and the giant Ente Nazionale Idrocarburi (ENI) oil and gas conglomerate, founded in 1953. This was in addition to the Italian state’s adherence to the continental European model of exercising control over all the basic means of transportation and communication (railways, motorways, airlines, telephones, telegraphs, postal services, television), and adding in 1962 control of the electrical industry as well. Finally, several state agencies were charged with promoting economic development, the most notable being the Cassa per il Mezzogiorno, created in 1950, whose aim was to address the perennial problems of underdevelopment plaguing Southern Italy. A major portion of the Spanish economy also became either a government monopoly or was run by mixed enterprises under parastatal control. Primo de Rivera initiated the trend in the 1920s, with the nationalization of the petroleum industry and with his cuencas hidraulicas projects (which in some ways anticipated the Tennessee Valley Authority) in Spain’s major river basins. State intervention was carried much further by the Franco regime, which proclaimed economic autarky as its goal during its first two decades in power, and which, in 1941 followed the Italian fascist model by creating Instituto Nacional de Industria (INI), a Spanish counterpart of IRI. In Greece, too, the state presence in the economy not only became very strong during the 1960s and 1970s, but as in Italy and Spain, was remarkable for its wide dispersal throughout the industrial, financial, and service sectors. Ironically, it was Salazar’s Portugal, which had long trumpeted its dedication to corporatism, that had the smallest (though still significant) public sector in the 1960s, a sector which would balloon enormously for a decade or so, after the 1974 Revolution added to it a wide range of nationalized enterprises. Although nothing like this strong parastatal sector existed prior to the 1930s, the economic presence of the state was not negligible even then at least in Italy. Alexander Gerschenkron uses late nineteenth-

16

Introduction

century Italy as a prime example of how in backward economies the state can act as a surrogate for private enterprise, by assuming responsibility for the task of catching up. State intervention mostly took the form of tariff protection for selected industries, preferential treatment of local industries in military procurement, and other such indirect measures. The state also undertook the completion of the railway system in the 1870s and 1880s, after the private companies which had built the prime lines lost interest, and played a key role in the establishment of Italy’s first major steelworks. To be sure, military considerations provided some of the motivation for both these latter initiatives, but the desire to promote economic development was also critical. A related dimension of the interventionist nature of the traditional Southern European state was the widespread adoption of practices which substantially impeded the functioning of market mechanisms. These practices reflected the role of ideas in economic policy (Hall 1989), which in the case of Southern Europe were interventionist and corporatist. Some of these interventionist practices were protectionist efforts to defend the interests of economic elites working through their political and administrative allies in the state. Others were manifestations of corporatism, whose origins could be traced back to the nineteenth century, but which, as just noted, came to full fruition under the authoritarian regimes of the interwar period. This was particularly important with regard to the overweening regulation of the labor market by corporatist arrangements, in which the state and employers sought to constrain workers, restricting or preventing their right to engage in collective bargaining, regulating the terms of employment, and, over the long term, suppressing wage demands. A final factor which supports the conception of Southern societies as being ‘state heavy’ is the role the state played historically in providing jobs, especially for those precarious sectors of the middle classes who were educated but nonpropertied. State bureaucracies overstaffed with people whose major concern was to hang onto their jobs became characteristic fairly early. In Spain, for example, Raymond Carr (1966) makes empleomania, the obsessive pursuit of public employment, his central explanatory concept for political life in the 1840s through the 1860s, and Benito Pe´rez Galdo´s’s novel Miau provides a moving portrait of a party loyalist who lost his ministerial job in the closing decades of the nineteenth century. For Greece, Kolettis in the 1840s first opened the public trough to ‘his Greeks’ so they could ‘eat’, setting in motion a snowballing of the number of state employees throughout both the nineteenth and twentieth centuries. In Italy, the flocking of Southern Italian intelligentsia to governmental posts in Rome began

Introduction

17

with the coming to power of the Sinistra in 1876, and after about 1900 became notorious for its proportions. To be sure, political patronage has existed in all countries, not least in the United States, especially before its fierce spoils system began to be weakened by civil service legislation. But the phenomenon seems to have been more widespread and entrenched in the South than in other regions of Europe during the nineteenth century. More importantly, it did not tend to diminish with time, but either retained or actually increased its strength during the twentieth century. The unusually large parastatal economy mentioned above was one of the reasons. Those who controlled the state could offer employment not only in one of the standard governmental agencies, but also in the parastatal entities that sprang up in all four countries. Another factor which seems to have contributed to this phenomenon was the sharp alternation in power between rival political forces (between dictatorial undemocratic regimes, and between rival governing parties within democratic regimes), which was a prime feature of Southern European political life during much of our period. With each change in regime and, especially in twentieth-century Greece, election within democratic regimes, many of the state or parastatal employees who had gotten their jobs during the previous government in one way or another managed to hang onto them, while a whole new layer of state employees chosen from among the new government’s partisans was added to them.

The ‘Weakness’ of South European States If, for the reasons just discussed, Southern European states can be considered to have been heavy, in what ways can they also be thought of as having also been ‘weak’ states? Let us begin our examination of the attributes of the Southern European state that point to its weak nature, by recalling that centralization, however extreme, did not imply strength. The legal status of local and regional government might have been severely emasculated, but the local elites nevertheless retained considerable power, since mutual need bound them to the central state authorities and offered them both privilege and protection. The basis of this symbiosis was that, especially in the liberal parliamentary regimes which governed Southern Europe for most of our period, the central authorities depended on the acquiescence or active support of the local elites, not only on election time but also in the normal course of events. Civil governors in Spain could, for example, unseat municipal councils at will, but they exercised that power sparingly (except when dealing with ‘subversive’ bodies, where extrasystemic forces had

18

Introduction

somehow gained control), since, if one set of local elites was alienated, that made civil governors all the more dependent on the other factions, rendering their position potentially weaker. A de facto collaboration between national and local powerbrokers provided the underpinning for Southern European liberalism, especially during the latter part of the nineteenth century, when the socioeconomic structures were still overwhelmingly rural. The most important privilege the local elites received in return for their acquiescence to the status quo was probably that of not being taxed in proportion to their wealth. This occurred in two ways, through the influence of these elites on what sorts of national taxes would be imposed, and via the direct control they exercised over what local taxes would be levied and who would pay them. This led to an unjust distribution of taxes of both kinds, with indirect taxes which fell equally on all, regardless of income, predominating at both the national and local levels, and with direct local taxes sometimes actually being tilted in favor of the wealthy. But since for the local elites the best kind of tax was no tax at all, their power helped cause the chronic inadequacy of both state and municipal tax revenues, although the inefficiency of the state bureaucratic machinery and its inability to carry out delicate or complex operations also contributed to the shortfall. Without sufficient revenues from the well off to carry out its tasks, the state was forced into almost constant deficit spending, which created a vicious cycle, whereby the shortage of funds needed to meet the rising debt service charges resulted in periodic crises that were settled through moratoria and restructuring of debt payments, only to be followed in a few years later by a new crisis. The catastrophic Portuguese fiscal crisis of 1890–2 is perhaps the most dramatic illustration of this syndrome, since revolutions, wars, and crop failures did not contribute to it, as tended to be true with the crises of other countries. The Greek crisis of the mid-1890s triggered by the collapse of the main export crop, the currant, was also severe and its adverse consequences long term. Spain staggered from fiscal crisis to fiscal crisis from the 1820s to the early 1900s, without succumbing to any of them to the degree that Portugal and Greece did in the cases just mentioned. The same was true of Italy, although in the late 1860s and 1870s the threat of bankruptcy caused the conservative government in power to try to balance the budget by introducing severe new taxes which fell mostly on the peasantry, a mistake which provoked widespread revolts and was never again repeated. If excessive state use of its instruments of coercion for repressive purposes caused Southern European societies to appear state heavy, the state’s inability to control those instruments was a sign of its

Introduction

19

weakness. Far more frequently than elsewhere in either Western or Eastern Europe, the military became a law unto itself, repeatedly taking matters into its hands, either by seizing complete control of the state, or by forcing it to do its will in lesser ways. The Iberian Peninsula provides the most frequent examples of both kinds of military intervention into politics, in part because of the impetus established during the fierce civil wars that ravaged Spain and Portugal during the early and mid nineteenth century, in part in reaction to the consciously more radical regimes that won power in the two countries during the early twentieth century. Greece followed closely. The insubordination of the military there originated during the war of independence of 1821–7, and the subsequent period of 1829–33, when semianarchic conditions prevailed for much of the time and there were no clear lines of authority. It manifested itself anew in the 1840s and 1860s, when the ineptness and unpopularity of Greece’s Bavarian-born first king, Othon I, offered ample opportunities for conspiracy. But it required the deep internal divisions and the disasters of twentiethcentury Greek history to cause praetorianism to become almost systemic. As to Italy, it managed to escape the more extreme manifestations of this political curse, after tottering on the edge prior to 1860. But if the military never openly seized power, lesser forms of military insubordination were not uncommon, above all during the final years of liberal Italy, when army and police defiance of civil authority and their de facto collaboration with the nationalist extremists were crucial to the rise of fascism. The tardiness and inadequacy with which the state provided basic social services is another indication of the weakness of the traditional Southern European state. Well into the mid twentieth century, social insurance programs were underfunded and underdeveloped, while social assistance or welfare programs also exhibited characteristics that departed markedly from the more universalistic programs that, over time, gradually became the rule in advanced industrialized societies. Foremost among these were the enormous differences in access to these welfare benefits between full-time employees and a growing class of precariously employed or unemployed individuals. Corporatist labor practices deepened this divide and, over the long term, made it virtually unbridgeable by new entrants into the labor market. Access to benefits was extremely uneven, and welfare programs were highly fragmented. In addition to worsening problems of economic inequality, these structural defects distributed these scarce state resources inefficiently. One reason why resources were scarce is that the taxation systems employed by Southern European states were highly regressive, susceptible to tax evasion, and incapable of generating an

20

Introduction

adequate volume of revenues to support state activities at levels comparable with those of more mature democracies. This pattern was to effectively persist until (re)democratization, opened the way for the improvement of the state’s tax-collecting capacity and the implantation of modern welfare state structures in all four countries. The slow and faulty growth of public educational systems is another example of the weakness of the traditional state in Southern Europe. In the three Latin nations, illiteracy rates prior to the 1960s remained considerably higher than they should have been, given their relative levels of urbanization and economic development, as well as their much longer histories as independent nations, when compared to Eastern Europe. Jurisdictional conflict with the Catholic Church contributed to the paralysis, as endless struggles occurred over the degree to which education should meet religious rather than secular needs, and how much the church should contribute to the costs of the enterprise. Equally important, all three states refused to use national revenues to finance primary education, delegating this burden to the municipal governments, where the local elites in charge usually starved school budgets to avoid tax increases. The situation was somewhat better in Greece, since the Orthodox Church, much less politically organized than the Catholic, was not as great a complicating factor. But on the other hand, the linguistic split unique to Greece, between an elite purist language patterned after classical Greek and the popular demotic tongue created a social chasm that mere primary school education could not bridge. The incompetence of the administrative apparatus was another dimension of weakness characterizing the traditional Southern European state. As Sotiropoulos points out in his contribution to this volume, this particular phenomenon stemmed in part from the prominent role political patronage played in staffing the bureaucracy, as well as from the inability to develop the type of administrative elite that made it possible for France to maintain effective government despite the continuous reshuffling of cabinets that characterized the Third and Fourth Republics. In fact, Sotiropoulos produces an ideal type outlining the salient features of traditional Southern European public administration and highlights the various ways in which its operation contributed decisively to the impaired functionality of the traditional Southern European state. Among the distinguishing characteristics of Southern European state bureaucracies were pervasive clientelism; the uneven development of administrative institutions, with some departments overstaffed while others lack sufficient employees to perform their tasks at an acceptable level; an excessive and rigid legalism; the lack of a stable, institutionalized, expert administrative

Introduction

21

elite (except in Spain and, to a lesser extent, Portugal); and widespread perceptions of administrative inefficiency and corruption at the lower level. Incompetence was evident in many aspects of state activities, but above all in that ultimate test of state power, the ability to wage war. The military record of the Southern European nations in the nineteenth and twentieth centuries, both on the battlefield and in the organization of the war effort at home, was a sorry one for the most part. Spain was incapable of bringing any war to a speedy conclusion, whether it was the failed efforts of 1814–24 to restore its control over Latin America, the Carlist wars of the 1830s and 1870s, the Cuban wars of the 1860s and 1890s, the Moroccan war of the 1920s, or the Civil War of 1936–9. Something similar can be said for Portugal’s civil wars of the 1830s and 1840s, its participation in World War I, and its African wars of the 1960s. United Italy went from defeat to defeat, starting with its 1866 war against Austria and continuing to World War II, the only exceptions being the Italo-Turkish war of 1911–12 and the Ethiopian war of 1935–6, when the unrestrained use of Italy’s enormous technological superiority enabled Mussolini to overwhelm his victims. Even nominal victories, as in World War I, were really defeats, since they were due mostly to the efforts of Italy’s allies and Italians themselves played secondary roles in achieving them. As to Greece, despite disastrous defeats in its wars with the Ottoman Empire in 1897 and with emerging Turkey in 1922, as well as a miserable performance in the Civil War of 1946–9, its record is on the whole more balanced, since there were periods of effective military action between 1912 and 1920, and especially in the Metaxas dictatorship’s preparation for, and conduct of, the 1940–1 war against Italy. The only other instance of a high degree of organizational ability in the war record of any Southern European state is the Spanish Republican government’s ultimately vain effort to create new military forces in 1937 and 1938, and to replace those wrecked by the nationalist insurrection and by social revolution. A further factor weakening the Southern European state was its lack of an ideology capable of causing people to sacrifice for it in times of crisis. Only Greece is a partial exception, since irredentism provided a unifying cause until 1922 of such intensity that nationalism remained strong even after catastrophic defeat in the Greek–Turkish war of 1919–22 which had extinguished all hope of territorial expansion. In Italy, neither colonialism nor irredentism were strong enough to serve as the basis for unifying and mobilizing myth, as indicated by the quick abandonment of the first Ethiopian venture after the Adowa defeat in 1896, and by the exceptionally high desertion rates that plagued the

22

Introduction

Italian army in World War I. Fascist propaganda tried to fill this ideological vacuum by aggressively inculcating a wide variety of political myths, but the regime’s successes were both brief and superficial, as its unimpressive record in World War II shows. For a few years in the 1880s, Portugal seemed on the verge of finding a myth capable of rallying its people in the dream of establishing an imperial belt across the width of Africa, but the British ultimatum of 1890 forced it to abandon this cause, with the net result that the Portuguese monarchy was further weakened because of the abject humiliation it had endured. In Spain, the availability of galvanizing myths appealed only to specific sectors of society (social revolution for the anarchists, Carlism for very devout Catholics, etc.) and so tended to divide the country rather than unite it. Nationalism, so strong elsewhere in Europe and the nineteenth and twentieth-century’s, remained remarkably underdeveloped in Spain. Nevertheless, Spain provided a good example of the power of nationalist ideology in the rise of Catalanism. While the Spanish state remained weak and disoriented, in part because it lacked a cause that could mobilize significant portions of the population, the Catalans united as a people behind the autonomist myths they created in the nineteenth and early twentieth centuries and produced a powerful seminationalistic movement of their own. Merged with, and underpinning, its inability to elicit intense loyalties was the fragile legitimacy of the traditional Southern European state, which tended to keep its people from giving it the benefit of the doubt even in everyday matters. What were the sources of this precarious legitimacy? One was the weakness of royal authority, a crucial element of state validation during the nineteenth century. This authority was badly undermined in the Iberian Peninsula by the exile or imprisonment of the Portuguese and Spanish monarchs during the Napoleonic wars, by the family conflicts that divided the Braganzas and the Bourbons during the 1830s, and by the personal deficiencies of the heads of both dynasties during much of the nineteenth century. The monarchy also came in question in Greece, though less fundamentally so until World War I, when the struggle for power between Eleftherios Venizelos and King Constantine resulted in the National Schism, which would polarize Greek politics for more than two decades. The long rearguard battle of the Catholic Church against its loss of power in the secular sphere also seriously undermined the legitimacy of many of the governments (whether politically progressive or merely conservative but independent-minded), which held state power in the Latin countries. This was especially so in Italy, where the Papacy boycotted the new state for a full half-century after its creation, and in Spain, where integrist and Carlist movements were strong.

Introduction

23

An additional factor contributing to the weak and fragile legitimacy of the traditional Southern European state was the precocious introduction of liberalism and of political institutions associated with it in the four countries. It made its appearance in each society at slightly different times and under different circumstances, but had become embedded in all of them by the 1840s, long before the social and economic evolution of the Southern European nations had reached the point it achieved in Britain, France, and in the other early European converts to civil liberties and parliamentary rule. A wide gap thus opened between political theory and political realities, reminiscent of de Tocqueville’s important distinction between the pays le´gal and the pays re´el. The first major effect of the constant violation in practice of political liberties that were accepted in principle was fierce conflict between progressive liberals who sought to actually implement more of the liberal ideal, and conservative liberals who feared chaos were this to happen. This is what lay behind much of the civic strife of the 1830–70s, especially on the Iberian Peninsula. By the 1870s, the idealists had lost most of their fervor, so that the two branches of liberalism could compromise their differences and agree to alternate in power, setting in motion a process that came to be known as the turno pacı´fico in Spain and rotativismo in Portugal. Politics became more peaceful for a few decades, but this only reflected the deep structural corruption of the political process that had taken place, as the governments in power shamelessly manipulated elections so as to produce the desired results. Such tawdry practices would undoubtedly have been overlooked had the liberal state been effective in sponsoring economic development. But its attempts to do so either were or seemed to be failures, with the result that the agony of modernization in Greece, Italy, Portugal, and Spain was especially severe, as the high degree of social conflict endured during the nineteenth and twentieth centuries indicates. The process was very lengthy and was all the more tormenting because progress kept being made toward the goal, but the goal kept receding. A labor of Sisyphus in short, which was not completed in Italy until the 1960s—the first stage at which the economy of that nation can be said to have laid down solid foundations for the future. In Spain, a sense that the economy had finally approximated advanced European levels did not become general until the 1980s, while a similar consciousness began to emerge in the early 1990s in Portugal, as well as in Greece (which in the 1980s squandered the opportunities deriving from entry into the European Community). A final factor weakening the legitimacy of the traditional states in Southern Europe was their inability to live up to the great expectations

24

Introduction

created by their radiant past histories. This was true for the Iberian nations, which had the memory of their golden ages to haunt them. It was even more valid for Greece and Italy, the two entirely new state constructions, many of whose citizens shared the illusion that once the Turks and the Austrians had been expelled, the glories of the past would quickly be recovered, only to be bitterly disappointed by the unpleasant realities which immediately confronted them. As the contrasting experience of Germany after 1871 indicates, it was not the newness of the states itself which caused the problem. Rather, the legitimacy of the traditional Southern European state came into question because of disillusionment with the meagerness of its accomplishments in every field—political, economic, and cultural. For these and other reasons, the state in Southern Europe never fit into any of the models of state–society relations that arose in Western Europe. It could not approximate the state’s ascendancy over society which characterized imperial Germany, nor the kind of grudging but nevertheless deep respect that the French have long displayed toward their state, nor the comfortable assurance with which the English tended to ignore their state, knowing that in moments of crisis, like the Napoleonic or the World Wars, everyone would rally round it, giving it great strength. A vicious cycle operated in Greece, Italy, Portugal, and Spain. Being poorly articulated, impoverished societies, they could not produce strong states, and since stable, competent, rational states were lacking, the societies, in turn, could only progress slowly. Since southern Europeans were historically and culturally close to the dominant nations of Europe, they fervently wanted to follow in the paths set by Britain, France, and Germany and had many fewer mental reservations than those held by non-European societies. Since they were not colonies but independent states, they were at least nominally responsible for determining their own destiny. The goal of full Europeanization could not be rejected, except briefly and occasionally by a few intellectuals, but the means to achieve it could not be found, and there was nobody outside themselves to hold responsible for this failure. Therein lies the source of the frustration and despair so characteristic of Southern European history in the nineteenth and much of the twentieth centuries. Even though a great deal was accomplished, as our four nations managed to keep the gap that separated them from the leaders from growing still wider (and so unknowingly positioned themselves to be able to benefit from the great socioeconomic breakthroughs of the post–World War II era), their distance from those who had once been their equals or inferiors could not but torment them.

Introduction

25

Authoritarianism and its Legacies While the dictatorships of the mid twentieth century exhibited characteristics that went well beyond the dysfunctionalities of the traditional Southern European state outlined above, several of the traits or tendencies of the nineteenth and early twentieth centuries persisted and reached their most extreme degrees of exceptionalism under these authoritarian regimes. This is particularly true of Greece, Portugal, and Spain, whose departure from the democratic norms that were fully entrenched elsewhere in Europe were strikingly anachronistic by the early 1970s. State repression of political dissidence in all forms of partisan organization, except those dominated by the regime under Franco, Salazar/Caetano, and the colonels was, at various points, much more excessive than had been typical of earlier periods. In Spain, this difference between the milder forms of repression in the past and the harshness of the dictadura (dictatorship) of Franco and the more benign rule of Primo de Rivera (1923–30) was captured by coining the term dictablanda (blanda meaning ‘soft,’ while dura is the Spanish word for ‘hard’). For the governments which came to power after the transition to democracy, coping with the legacies of this repressiveness was a major policy consideration of paramount importance, which made possible the effective restraining of the coercive impulses of the police and ensured that judiciaries would strictly adhere to the rule of law and to the norms concerning the defense of civil and political liberties associated with democratic polities. In addition to the repression of political organization and expressions of dissent, these regimes vigorously suppressed trade unions and coerced workers into corporatist, vertical ‘syndicates’. While the elimination of collective bargaining and the right to strike facilitated the exploitation of labor and held down wages, over the long run employers would find themselves adversely affected by the imposition of rigid controls of labor relations that impeded the flexible reassignment of workers to new tasks, and, in effect, prevented employers from dismissing incompetent or redundant workers. These rigidities of the labor market undermined the productivity and competitiveness of Southern European firms, and this legacy of the past would assume great salience for subsequent democratic governments in the context of their entry (in the 1980s) into the European integration process. Similarly, the large parastate firms and holding companies that were established or expanded in accord with corporatist economic philosophies under the authoritarian regimes in Italy and Spain, in particular, were ill-suited for the open competition which integration into the

26

Introduction

European Community would require. Their inefficiencies and lack of competitiveness were insulated from international market pressures by heavy-handed state economic intervention and, in Portugal and especially Spain, by the autarky and isolation of their economies under Franco and Salazar/Caetano. In Portugal, the property seizures and waves of nationalization that took place during the Revolution of 1974 further contributed to a bloated and inefficient industrial sector. At the same time that these manifestations of the heaviness of the Southern European state were exacerbated under the authoritarian regimes of the mid twentieth century, so, too, was the weakness of the Greek, Portuguese, and Spanish states during this period. Levels of public spending on social services, in particular, lagged far behind those found in other Western European societies. Perhaps the most extreme example was that of Spain in the mid-1960s, which spent less of its national income on education (1.5 percent) than any other country in Western Europe. (Spain’s lagging education expenditures were followed, in inverse rank order, by Portugal’s 1.9 percent and Greece’s 2.3 percent; the next closest Western European country was West Germany, with 4.2 percent [Ministerio de Educacio´n y Ciencia 1969: 174].) While lagging economic development certainly contributed to the inadequacy of this important sector of social spending, the fact that Spain was tied with Angola in 122nd place out of 131 countries in one study (Taylor and Hudson 1972: 32), and its placement in the lowest quartile (alongside such countries as Upper Volta, India, Togo, and Paraguay) in terms of enrollment in higher education at about that same time (Wilensky 1975, Table 16) clearly suggests that the level of socioeconomic modernization cannot provide a sufficient explanation of this manifestation of state weakness. And even though Franco had been a career military officer and became Head of State while also serving as Generalı´simo of the Army, Spain in 1973 was also spending less on its military (1.5 percent of GDP) than any other country in Europe—East or West (Gunther 1980: 49). In contrast, it should be noted that Portugal’s 4.5 percent of GDP devoted to military spending was the highest among West European countries (undoubtedly as a reflection of its involvement in colonial wars at that time in Angola and Mozambique), and Greece’s 4.1 percent also placed it near the top among Western European countries (reflecting its tense relations with its Communist neighbors to the north and with Turkey to the east). An important factor contributing to low levels of social spending in all three countries was clearly the regressive and evasion-prone taxation system upon which the state depended. The yield of tax revenues as a percentage of GDP placed Spain (20 percent) in last place among Organization for Economic Cooperation and Development (OECD)

Introduction

27

countries in 1971, with Portugal’s 21.8 placing it third from last, and Greece’s 24.5 percent ranking it 17th among the OECD’s 22 member countries at that time (OECD 1972). While administrative inefficiency and delayed socioeconomic development may partly account for these inadequate flows of revenues to the state, it is more likely that cronyism and the close association between the elites of these authoritarian regimes and the upper social strata in each country contributed substantially to these aberrant features of Southern European public policies. This, in turn, suggests that the demise of those dictatorships and the opening up of these political systems to democratic contestation by the full array of social groups might have paved the way toward substantial changes in key public policies and the administrative structures and practices of the state in these three Southern European countries. The fact that Italy had democratized decades earlier will help to test this proposition in the case studies of statestructural subsystems and key public policies that will follow in this book.

The Evolution of States in Southern Europe Since the 1970s To what degree has the contradictory heritage of the past been overcome in recent decades? The central concern of this series has been to trace the overarching transformative process which, in the course of the past four decades or so, has made possible the emergence of the New Southern Europe. Under the combined impact of socioeconomic development, democratization, and Europeanization, the trinity of long-term processes constituting the major analytical instruments of this project, Greece, Italy, Portugal, and Spain have, each in its own way, succeeded in shedding their old, ‘exceptionalist’ identity as constitutive parts of a region regarded as the ‘backwater of Europe’, and to assume the characteristics of modernity, in the process significantly approximating the general attributes associated with advanced (post)industrial societies and democracies in Western and Northern Europe. Change has been rapid, indeed assuming leapfrogging qualities in certain areas, such as politics, and slower in others. With regard to the state, which constitutes the focus of this volume, the overall impact of long-term change has been such as to gradually alter most aspects of the ideal type derived from its traditional organization and behavior. The following chapters in this book describe and analyze changes in several public policy domains and state structures. These include a detailed study of the judiciaries of Greece, Italy, Portugal, and Spain during the transition to democracy and in democratic politics today, as

28

Introduction

well as a broad comparative analysis of the state bureaucracies and efforts at administrative reform over the past three decades. The public policy case studies range from quantitative, time-series analyses of aggregate levels of social welfare spending and qualitative examinations of the evolution of specific welfare programs, to studies of taxation and environmental policies. An analysis of supranational integration and its impact on the dynamics and structure of subnational governance (especially in Spain) is also presented in this volume. To be sure, this sample of institutions and policy sectors does not capture the full range of potentially relevant policy areas or state subsystems. The scope and variety of state activities in advanced postindustrial societies, such as those currently existing in Southern Europe, are simply so enormously widespread as to defy efforts to capture all of their significant features in a single volume. Rather, they represent a careful selection of state subsystems and policy areas where change and continuity can best be observed and analytical insights based on these can most profitably be derived. Taken together, they point to the great strides forward taken by the state in our four countries that have rendered it more stable, more effective, more just, more legitimate, more citizen-friendly, and less distant than at any other time in the past two centuries. Put otherwise, they have rendered the state more functional. At the same time, they also serve as reminders of new challenges lying ahead and of the need to respond to inevitable demands for further improvement and reforms. At the very least, these are likely to include increased citizen demands for improved state performance in the areas of public administration, social services, taxation, and education, further strengthening of mechanisms and structures designed to enlarge the scope of state accountability vis-a`vis the citizen; ability to accommodate waves of incoming migration in Spain, Italy, and Greece from the northern shores of Africa or from Southeastern Europe; the change of traditional social norms and the rise of multicultural society, with the incorporation of variable, postmodern lifestyles and different ethnic and religious communities in South European societies; the emergence of new social and political identities (regional, national, and European); and enhanced capacity to profit from opportunities arising from growing integration in the unfolding project of the EU and from more intensive and extensive participation in the world of advanced postindustrial societies and of mature democracies. A primary objective of this series of books, however, is to attempt to explain both the systematic transformation of the Southern European State and the significant differences among countries and across sectors of state activity. For this purpose, it is perhaps useful to fully

Introduction

29

elaborate the theoretical underpinnings of a number of hypotheses that will be developed by the contributing authors in their respective studies. We return to these hypotheses in a synthetic overview of these transformative processes in the concluding chapter of this book.

Toward Explanation: Modernization, Democratization, and Europeanization In our efforts to explain processes of social and political change in Southern Europe, we have repeatedly encountered evidence of the importance of three clusters of explanatory factors. Simply stated, these are modernization, democratization, and Europeanization. The first of these refers to socioeconomic modernization—that cluster of changes in economic relations, social structure, and culture that many studies undertaken in the 1950s and 1960s found to have a major impact on political behavior at the individual level and, over time, at the level of the political system. With regard to some of the public policy processes and outputs that we analyze in this volume, numerous quantitative studies have found that a country’s level of socioeconomic development can have a substantial impact on its ability to adopt and implement policies central to the definition of the socialwelfare state. While quantitative studies analyzing cross-national variations in social welfare policy outputs among advanced industrialized countries only (such as Castles and Mckinlay 1978) have repeatedly found that such operational indicators of modernization as per capita income have little or no explanatory impact, those analyzing policy outputs among the full array of the world’s countries from the poorest to the richest (beginning with Cutright 1965 and Wilensky 1975) have found quite the opposite, suggesting that the achievement of a certain ‘threshold’ level of modernization is a prerequisite for the development of a social-welfare safety net for all citizens. Since three of the four Southern European countries lagged far behind the rest of Western Europe in terms of socioeconomic development until quite recently, and did not initiate their most rapid periods of modernization until the 1960s or later, this ‘modernization threshold’ hypothesis is relevant to an explanation of lagging social-policy development at the beginning of the period under study in this volume. In the first two volumes in the series, we found powerful evidence that some of the socioeconomic and cultural variables embedded within this cluster of social transformations have potent and pervasive impacts on processes of political change, but we also concluded that by themselves these factors are insufficient to account for the often

30

Introduction

complex patterns of change that we have observed. Accordingly, rather than rely upon some simple form of socioeconomic reductionism, we have explored the impact of these modernization variables as components of multivariate processes that sometimes involve complex interactions with other sets of explanatory factors. Among these are the process of democratization (which is heavily influenced by the behavior of strategically placed political elites), and the long-term impacts of democratic government and partisan institutions on other facets of society. In short, in the context of Southern Europe, democratization emerges as a second powerful cluster of factors that can substantially affect the fundamental nature of politics, public policy, and the state itself. Finally, in the particular case of Southern Europe in the final decades of the twentieth century, various structural and policy changes relating to European integration assumed a powerful role with regard to the transformation of certain state subsystems and public policy domains. Let us examine in somewhat greater detail how these three processes of change—socioeconomic modernization, democratization, and European integration—have affected fundamental patterns of political behavior that we have analyzed in earlier volumes in this series. Building on these earlier findings, let us then formulate a number of key questions relating these three processes to the public policies and state subsystems that are the principal focus of this book. In Southern Europe as elsewhere, these three large-scale processes are mediated by political and administrative institutions, which are examined in detail in the individual chapters of this volume. Socioeconomic modernization has often been shown to have a powerful impact on the relationship between elites and masses in a political system. In its classic formulation, modernization is seen as leading to the conversion of ‘subjects’ into protocitizens, insofar as increases in literacy, urbanization, and exposure to mass communications and transportation networks give rise to the demand for a more active participatory role in the political system, and provide citizens with many of the tools (cognitive and other) necessary to play an active and efficacious role in politics. Following the substantial economic development that unfolded in Greece and Spain in the 1960s, for example, docile acceptance of authoritarian rule was no longer tenable. This led both to occasional popular mobilizations in support of regime change, but also to widespread expectations that democratization was ‘the only game in town’. While Portugal’s economy lagged somewhat behind those of Spain and Greece, its level of socioeconomic modernization had certainly surpassed the minimal levels typically regarded as prerequisites for independent and self-sustaining mass participation. Indeed, the level of participation by ordinary citizens in the

Introduction

31

revolutionary process of political change in Portugal substantially exceeded the mobilization of Greeks and Spaniards in their democratic transition processes. In short, socioeconomic modernization contributed both to the capacity of citizens to participate independently in politics (as is required in any definition of democracy), and to widespread expectations that democracy would be a more viable regime option that it had been in earlier periods of nineteenth- and twentieth-century history in these countries. Socioeconomic modernization also contributed to the transformation of the dynamics of partisan politics, to the emergence of the new campaign politics, and to fundamental changes in the organizational characteristics, programmatic orientations, and mobilization strategies of political parties after the mid-1970s in Southern Europe, as in other modern democratic systems. Insofar as socioeconomic development creates a sizable and relatively affluent middle class, and insofar as political elites believe that the largest block of voters is located near the moderate center of the political spectrum, their votemaximizing strategies will lead them to adopt moderate policies, images, and campaign styles. In short, the emergence of modern economic and social structures can contribute to the development of stabilizing centripetal drives within the party system, in sharp contrast with the polarizing dynamics of premodern systems. This has provided a powerful incentive to parties of both left and right to converge toward the center of the political spectrum, if they wish to be successful in increasing their respective shares of the vote. Finally, socioeconomic modernization has gone hand-in-hand with technological advances that make it possible for party elites and candidates to communicate directly with voters, rather than mediate their campaign messages through face-to-face linkages based upon sizable mass-based party organizations. Particularly important in this respect has been the emergence of television, which has become the principal medium of political communication in virtually all contemporary democracies. This makes it possible for elites to conduct modern election campaigns in the absence of sizable partisan secondary associations. Again, we can conclude that socioeconomic modernization may facilitate moderation of party ideologies and programmatic stands, as well as to transform party organizations toward more modern forms that can effectively participate in the new campaign politics. However, we have also seen plentiful evidence that socioeconomic reductionism would be incapable of providing an adequate explanation of the complex political trajectories that we examined in these preceding volumes. This was particularly true in Volume 1, where we treated democratic consolidation as the ultimate dependent variable. In

32

Introduction

contrast with the common economic argument that prosperity can contribute to regime support, we found no such relationship with regard to support for either the former authoritarian regimes or for the newly emerging democratic systems. Spain and Greece, in particular, had undergone far-reaching processes of socioeconomic development during the 1960s and early 1970s, and yet this contributed nothing to the overall level of popular support for the regimes of General Franco or the colonels. Conversely, the establishment of new democratic systems in these countries coincided with a protracted period of economic stagnation and occasional economic crises, and yet this did not impede their consolidation or the development of regime legitimacy. The timing of democratic consolidation processes among these four countries, moreover, was also inconsistent with predictions that could be derived from a reductionist economic argument: Italy was certainly the most affluent and modern of these four countries, and yet its process of democratic consolidation was the most protracted. Instead, while not denying that the socioeconomic context within which democratization processes unfold has a significant impact on their chances for success, we identified a number of explicitly political factors, such as institutional legacies, historical paths along which public policies have evolved in the past, and strategies of collective actors, that better account for the patterns of change that we observed in these four countries. These included the nature of the authoritarian regime (especially, in the cases of Greece and Portugal, the existence of ‘nonhierarchical’ military regimes), the pace and fundamental characteristics of the transition itself (most clearly distinguishing the abrupt revolutionary transformation of Portuguese politics from the more incremental process of change in Spain), the partisan institutions that emerged in the early stages of the transition, and, most importantly, the strategies, calculations, and behavior of political elites. Similarly, in Volume 2, we saw mixed evidence of the impact of socioeconomic modernization on partisan politics. To be sure, the incentives implicit in a modern, affluent social structure contributed to the centripetal drives that we observed in three of these four party systems (i.e. excluding Italy to some extent), while the emergence of television as the dominant medium of political communication certainly contributed to the emergence of the new campaign politics in all four countries. At the same time, not all parties or party systems have followed trajectories that strictly accord with this ‘modernization’ logic. Most importantly, socioeconomic modernization and secularization had substantially transformed Italian society by the 1970s, and yet political polarization (particularly along the lines of the historic religious cleavage), and ‘the old campaign politics’ persisted unchanged

Introduction

33

until the early 1990s. Similarly, while the long-term trend in Spanish politics has been toward moderation and a gradual disappearance of the religious and class cleavages as principal determinants of electoral behavior, this has not unfolded in a direct or unilinear manner. Indeed, a notable polarization of Spanish politics occurred in 1982 and its effects remained evident throughout the entire decade. Finally, not all parties behaved in accord with the centripetal logic implicit in these modern social structures, as the Portuguese Communist Party’s stubborn adherence to orthodox Marxism-Leninism illustrated. We concluded, instead, that a number of purely political variables have a substantial impact of their own on the dynamics of partisan politics. One of these was democratic consolidation, which has profoundly influenced the strategies, calculations, and behavior of political elites, and, in turn, mass political behavior. In contrast with unconsolidated, unstable democracies, the context of a consolidated democratic regime establishes a centripetal and self-reinforcing dynamic. Acceptance of a single, unambiguous set of ‘rules of the game’ and of the legitimacy of the established institutions of democracy can channel conflicting demands through structures specifically designed to regulate, if not resolve, conflict. The absence of consensus that these are indeed the only appropriate venues for the expression of conflict can, as was apparent during prior democratic or semidemocratic interludes in these four countries, encourage the articulation of demands (e.g. in the streets or through armed clashes) in a manner not conducive to the dialog and mutual concessions implicit in democracy. This often leads to polarization, and may result in instability, destabilization, deconsolidation, and potential breakdown. (It is noteworthy that the partial exception of Spain’s Basque country provides further evidence in support of this contention: in this region, the legitimacy of the existing Spanish democratic regime is not firmly established, and, accordingly, politics within this region has followed a polarizing, unstable, and occasionally violent trajectory.) Widespread agreement on the rules of the game—including explicit guarantees of civil and political liberties—also lessen suspicions among contending elites and gives them confidence that even if they lose one round, they will still be around to fight again in the future. Taken together, these characteristics of consolidated democracies lower the stakes inherent in a given conflict, avoid dynamic processes that can lead to polarization, and improve the chances that a compromise solution will emerge. They also set in place a centripetal dynamic in which elites are encouraged to moderate their rhetoric and behavior, because, in a stable, consolidated democracy, that is the most likely way to secure victory. Finally, civility breeds civility: cordial and mutually respectful interactions among elites can be self-perpetuating

34

Introduction

over time. In short, democratic consolidation establishes contexts conducive to political moderation. It also provides powerful incentives to parties that may have initially maintained antisystem or semiloyal stands to fully embrace the existing democratic regime, since failure to do so will alienate most voters and can consign such parties to the opposition benches indefinitely. In short, democratic consolidation encourages both political moderation and the unequivocal acceptance of the existing democratic regime on the part of all parties wishing to maximize votes and to eventually come to office through the electoral process. Another set of political variables that have long-term impact on the nature of partisan politics involves the institutionalization of electoral rules and specific types of political parties. Electoral laws have a direct and immediate impact on the representation of social interests in parliament, and, in addition, establish powerful incentives to political elites to adapt their strategies and electoral behavior over the long term. The clearest and most dramatic evidence of this proposition can be seen in the case of Italy. Political parties were institutionalized in the 1940s, and as a result many of the salient features of politics in that era were perpetuated over the ensuing decades, even though the social or political circumstances that had given birth to those particular features had long since disappeared. The outbreak of the Cold War, and the related institutionalization of the religious cleavage into the very structure of the dominant political party of the so-called first Republic, established a durable cleavage in Italian politics that, among other things, led to permanent government under DC-dominated coalitions, as well as to the permanent exclusion from power of the second largest partisan block led by the Communist Party of Italy. It also institutionalized both the old campaign politics until the early 1990s (despite the fact that television was available to virtually all voters), as well as clientelism, which permeated all the governing parties (despite the fact that its origins could be traced to premodern structures and practices of the undeveloped south of Italy prior to socioeconomic modernization). Finally, it ‘froze’ into patterns of electoral behavior a religious cleavage until the 1990s, even though growing secularization should have substantially weakened this factor as a determinant of the vote. It was only when these partisan institutions disintegrated in the early 1990s that these distinguishing features of partisan politics of the first Republic abruptly disappeared. In short, the consolidation of democracy and the institutionalization of electoral laws, parties, and other core institutions of democratic systems exert their own independent influence on various dimensions of political behavior.

Introduction

35

A new and highly significant causal factor emerged from the empirical studies contained in Volume 3: the process of economic and political integration of Western European countries within the EU. Europeanization affected in several ways the various sectors of the Southern European economies analyzed in that volume. First, significant (and long overdue) structural reforms were often introduced in an effort to prepare these economies for the challenges they would face within the broader European market. A central aspect of these structural reforms concerned the abolition or, at the very least, sharp reduction of traditional, corporatist, protectionist mechanisms and arrangements, which long constituted such a salient feature of Southern European economies and societies, and have over time produced profound rigidities in these countries’ labor markets. Second, some policy options were effectively imposed on the Southern European states through complex negotiations which took place in Brussels and had a profound impact on a number of sectors of these countries’ economies and the social strata associated with them. These changes included, among others, the massive inflow of EU funds, the introduction of the value added tax (VAT), the abolition of tariff barriers, and the elimination of certain subsidies to state-controlled enterprises (parastatal organizations). In these instances, member states had little alternative but to adopt these policy reforms and did so with little variation. This notwithstanding, not all responses were identical. Indeed, discernible differences in the individual national responses to the challenges associated with entry to the European Community can be identified in each case. Greece, preoccupied with security and political considerations relating to the need to further consolidate the country’s democratic regime, at a time when the memories of the colonels’ authoritarian regime were still fresh and the confrontation with Turkey over Cyprus still very much alive, opted for a strategy of ‘enter first, negotiate later’, which placed a premium on expeditiousness rather than on detailed and protracted negotiations concerning reforms envisaged by the European Community. Not faced with similar constraints, Spain and Portugal, on the other hand, though eager to secure admission in order to ‘reclaim their place in Europe’, were able to insist more on resolving areas of particular interest to them prior to entry. The outcome of all this, as Volume 3 meticulously documents, was that, as a result both of entry into the European Community and of the fact that the Community is simultaneously an institutional power external but also internal to the member states, the logic of Europeanization became a powerful parameter directly affecting the behavior of the state in Southern Europe, and left its clear imprint on a wide range of state subsystems and policy domains.

36

Introduction

To what extent would this process of Europeanization affect the structures and policy outputs of the Southern European states beginning in the mid-1980s, by which time all four had become members of the EU? To what extent would these same fundamental processes affect the institutional structures, behavioral patterns, and policy outputs of the Southern European states? Would socioeconomic modernization create pressures for change and adaptation in the direction of the more mature democracies of the industrialized world? To what extent would democratization establish channels for the articulation of demands for change? And would the transformation of Southern European states or public policies be characterized by leapfrogging? Would the nature of change be unilinear and/or uniform or would it differ from one policy area to another, reflecting the particular properties of each and revealing variation in the behavior of state subsystems associated with specific policy options? On a broader level, would the causal processes linking public policy to our three transformative processes conform to the unidirectional logics described above (in which policy is treated exclusively as a dependent variable), or would the evidence reverse the direction of causality or suggest reciprocal influences, allowing policy options to serve as independent variables in particular circumstances? Would the impact of particular policy choices be felt equally across the whole range of state subsystems and upon the democratic regime as a whole or would they have differential impacts on specific subsectors of state and regime? Finally, what are the theoretical implications of such a potential finding, and how transferable are these insights to the general literature concerning the comparative study of the state and of public policy? These are some of the broader questions arising out of the interplay between and among socioeconomic development, democratization, and Europeanization (which have been mediated by historical legacies, old and new institutions, and political ideologies) and relating to the role of the state in the New Southern Europe. The ensuing chapters attempt to address these and to provide more detailed answers to them.

The Contents of This Book In the following chapter, Francis Castles systematically analyzes the substantial impact of democratization and socioeconomic development on the evolution of social security expenditures in Southern Europe over three decades 1960–90 characterized by rapid social, economic, and political changes. At the beginning of this period, Greece, Spain, and Portugal lagged far behind other Western European countries with

Introduction

37

regard to the provision of basic social services, but by 1990, they had approximated the levels of spending on social programs typical of more mature democracies in advanced postindustrial societies. But the data presented and analyzed in this chapter also reveal significant variations among these three countries that underwent democratization during this period. In so doing, Castles confirms two central insights arising from this five-volume series: first, that exceptionalist interpretations of Southern Europe are no longer credible or convincing, and, second, that the four countries of the region do not constitute a distinct regional type of state, society, or economy, but rather differ among themselves as much as other Western and Northern European democracies. His analysis of these data also enables him to generate separate measures of the impacts of socioeconomic modernization and democratization processes. While generally confirming Castles’ findings concerning the extent to which social policies in Southern Europe have approximated the Western European norm, Marisol Garcı´a and Neovi Karakatsanis forcefully demonstrate the absence of any leapfrogging in state behavior with respect to this policy area. On the contrary, the authors’ conclusions point to significant lags with regard to several aspects of social policy which still separate Southern European states from their counterparts to the north. More specifically, they point out in Chapter 3 that, democratization notwithstanding, social citizenship has not been sufficiently developed in the region. Their search for an explanation of this phenomenon leads to a detailed examination of the Southern European welfare regimes and to the labor market structures associated with, and underpinning, them. Their findings highlight the significance of legacies of the past, both authoritarian and preauthoritarian, as factors contributing to the emergence and consolidation of fragmented, corporatist, and familistic welfare regimes, characterized by entitlement fragmentation, labor market segmentation, clientelistic practices, and familistic influences. In the course of their analysis, the authors raise important questions concerning how our three transformative processes have affected the development of social welfare in Southern Europe, which are addressed in the concluding chapter of this volume. The chapters by Pedro Magalha˜es, Carlo Guarnieri, and Yorgos Kaminis, dealing with judicial institutions and behavior in Greece, Italy, Portugal, and Spain, and by Dimitri Sotiropoulos, on the public administrations in these four countries, seek to determine the extent to which the policy areas associated with these sectors of the state evolved substantially over the past few decades or, conversely, lagged behind patterns prevailing in Western and Northern Europe. If this is the case,

38

Introduction

what accounts for this outcome? Put otherwise, what was the impact on these sectors of the three transformative processes that constitute the central analytical tools informing the essays in this volume? With the partial exception of Spain, Sotiropoulos’s analysis of state bureaucracies in Chapter 5 reveals that, despite perceptible improvements as a result of democratization and Europeanization, these public administrations continue to subscribe to an administrative culture that is not service oriented or consistent with the precepts of new public management or the idea that public administration has a legal and moral duty to serve the citizen efficiently, politely, and promptly. Magalha˜es, Guarnieri, and Kaminis also conclude in Chapter 4 that the judiciaries in the New Southern Europe have adjusted and adapted to the democratic regime without undergoing profound transformation. Both chapters seem to identify a certain degree of institutional autonomy in these state subsystems vis-a`-vis the predecessor authoritarian regime. To what extent, if any, can this autonomy help to explain the lag observed in the adjustment of these state subsystems to the logic and imperatives of the new democratic regime, and, more generally, of our three transformative processes? Finally, how has the advent of democratic politics affected these two state subsystems which, for a variety of reasons, enjoyed a significant degree of insulation under authoritarian rule in each of these countries? Iva´n Llamazares and Gary Marks focus on the impact of one of our three transformative processes, Europeanization, on the relationship between national states, subnational actors operating within their respective territories, and supranational structures linked to the emerging European polity. Their major concern is to explore the extent to which European integration and regional decentralization have opened up new structures of political opportunity and have made new institutional alternatives available to regional actors, be they parties or governments. Implicitly but centrally, the chapter’s concerns also touch upon another of our transformative processes, since decentralization was a direct outcome of democratization in Spain and, more indirectly, of democratization and Europeanization in Italy, and, to a lesser extent, in Greece and Portugal. A number of important questions emerge from this analysis: What is the impact of such an empowerment of regional entities on the central state in Southern Europe? Is there a discernible trend across all four countries toward a uniform reinforcement of regional actors and a corresponding decline in the salience of the central state in the politics of each country? Or, conversely, do we find in this area of state activity evidence of a significant degree of cross-national variation which would further undermine the argument for an ostensible convergence toward a ‘distinctly Southern European

Introduction

39

pattern’ concerning the evolution of the state over the past few decades? What lessons and insights concerning the relationship of regional actors with the central state and its structures has regional empowerment brought to the fore? Put otherwise, has the gradual immersion and socialization of our four countries in a culture of political contestation that is more geared to bargaining, compromise, and the positive sum logic underlying both than to confrontation and zero sum conceptualizations of conflict management had any discernible impact on the political behavior of subnational actors operating within the political culture gradually being shaped by the emerging European polity? Last but not least, does regional empowerment within the context of a broader entity privilege positive sum approaches to conflict management, and how is it likely to affect the ‘separatist logic’ traditionally identified with many regional entities in Southern Europe? In addressing such questions, Llamazares and Marks shed light on yet another dimension of the evolving nature of the state in the New Southern Europe which constitutes the central concern of this volume. The chapter by Geoffrey Pridham and Jose´ Magone sets out to examine the extent to which each of our three transformative processes has affected the emergence, development, and implementation of environmental policy in Southern Europe. Chief among these are the following: (1) Did regime change and democratization affect environmental policies in Greece, Portugal, and Spain? If so, is democratization linked to policy changes relating to the environment in a direct, causal manner, or more indirectly, as, for example, by opening up opportunities for contestation on issues related to environmental policy? And how do these policies in the three newer democracies compare with those of Italy, which was democratized three decades earlier? (2) To what extent has socioeconomic change contributed to the emergence of postmaterialist values and greater sensitivity concerning environmental challenges, and how has this cultural transformation affected political behavior and the policy stands of parties and governments? (3) Has Europeanization played a role in the development of environmental politics in the region leading, over time, to a closer approximation of patterns found in the more socioeconomically developed societies and more mature democracies in Western and Northern Europe? More generally, has change in this policy domain been gradual and evolutionary or has it been marked by sudden spurts and acceleration associated with the phenomenon of leapfrogging so salient in the realm of political change in the region? Finally, in his examination of the impact of democratization on the fiscal regimes of Greece, Portugal, and Spain, Stavros Thomadakis finds strong correlations between changes in the tax systems of these

40

Introduction

countries and all three of our transformative processes. His analysis, however, identifies both the different ways in which each of the three countries responded to stimuli generated by these processes, as well as a two-step process of adaptation, whereby reform in the economic realm explicitly and deliberately followed similar but prior changes in the political arena. As Thomadakis puts it, the gradual process of alignment of these countries with the logic and imperatives of the European Community/EU necessitated that they go through two distinct moments of consolidation: in this respect, he distinguishes between what he calls ‘fiscal consolidation’ from the consolidation of democracy, per se. This chapter places great emphasis on Europeanization and raises broader questions concerning the direction of causality linking Europeanization to these other policy areas. Indeed, it suggests that the causal arrows can point in either direction, or may be better conceived as reciprocal and mutually reinforcing: policy changes in some areas may be mandated by the EU, but compliance with those mandates may subsequently open the door to broader and/or deeper integration with the European Community. But whatever the cause of these tax system changes, it is important to examine the extent to which fiscal consolidation has helped to transform the Southern European state, especially insofar as a larger and more stable flow of revenues may have helped to overcome the traditional shortcomings of the weak state in Southern Europe. In the concluding chapter, we reflect on the empirical findings of these sectoral analyses in an effort to refine our initial explanatory hypotheses and theoretical perspectives. We reexamine the specific predictions of each of these three clusters of explanatory factors— modernization, democratization, and Europeanization—and determine the extent to which they can adequately account for the varying patterns emerging from these case studies. Even though a comparative study based upon just four cases cannot sufficiently ‘confirm’ or ‘disprove’ any of these relevant hypotheses, our analysis will benefit from the fact that the timing of each of these macrolevel changes varied from case to case, thus enabling us to sidestep the usual problem of colinearity that often bedevils cross-national comparative studies of this kind. Specifically, Greece, Italy, Portugal, and Spain underwent the most decisive stages of socioeconomic modernization, democratization, and Europeanization at different times. This temporal distance enables us to separately measure the impact of each transformation on the policy or state-structural sector in question. In the end, we find that each of these three clusters of explanatory factors has contributed to these structural and policy transformations. Nonetheless, we also argue that, even when all three are taken into

Introduction

41

consideration, they cannot provide a sufficient explanation of some significant aspects of the patterns of change that we observe in the following chapters. Accordingly, the concluding chapter of this volume will augment the three macrosystemic transformative processes discussed above with hypotheses growing out of the case studies that provide considerable additional explanatory power. These involve the extent of autonomy of the affected sector from the previous authoritarian regime, and the extent to which defenders of policies or structures inherited from the ‘old order’ possess institutional or other power resources vis-a`-vis reformers. In short, we argue that key political actors, historical legacies, lingering political ideologies, past policy choices, and institutions can exert influence on policy processes and outputs in a manner that is somewhat autonomous from the powerful forces of modernization, democratization, and Europeanization.

................ 2 ................

The Welfare State and Democracy: On the Development of Social Security in Southern Europe, 1960–90 Francis G. Castles

Viewing the world in terms of the research output of studies in the field of quantitative political economy between the late 1970s and the present, it is almost as if the New Southern Europe did not exist. This very important strand of political economy research does possess its own Mediterranean littoral, but it is only a truncated one, consisting, in the vast majority of studies, exclusively of France and Italy. The great debate of recent decades on the determinants of welfare state development in western societies has been—and continues to be—framed with virtually no reference to the experience of Greece, Portugal, and Spain. Prior to the 1980s, such a neglect was, at least, intellectually defensible. Now, after several decades of social security growth in Southern Europe, it is progressively less so. In the late 1980s, in household transfer terms, each of these nations had a claim that they were bigger welfare states than the United States, Britain, Canada, Switzerland, or Australia. In 1989, Greece spent a larger proportion of national income on social security than did Germany, the country which pioneered the concept of social insurance just over a century earlier. The reasons for the initial neglect of Southern Europe in the comparative literature are not difficult to comprehend. Much of the thrust of the comparative endeavor—whether the comparative economics of the OECD Secretariat or the comparative political economy of academia— was premised on the validity of the ‘method of similarity’ (Przeworski and Teune 1970), that the most useful comparisons were those which sought to locate the determinants of diversity within a universe of discourse tightly defined in terms of basic similarities. But, in the 1960s and 1970s, the Southern European countries were only gradually

The Welfare State and Democracy

43

emerging from a syndrome of underdevelopment which seemed to distinguish them radically from the rest of the advanced world. From a welfare viewpoint, this syndrome had at least four components. First, these countries’ social policy outputs were extremely low— in 1960, according to OECD data sources (OECD 1992a), Spain and Portugal spent a smaller proportion of national income on social security transfers than any other OECD nations apart from Turkey, and Greek spending only exceeded that of these countries plus Japan and Finland. Second, there were some conspicuous gaps in the national accounts data for these countries in the years immediately after the foundation of the OECD in 1960 and doubts persisted rather longer about the statistical adequacy of some of the data collected by state bureaucracies, which themselves could be characterized as underdeveloped in various respects. Third, Greece and Portugal were, and remain, the poorest countries of Western Europe, with Spain only marginally better off and just exceeding Irish levels in the period since 1960 (Summers and Heston 1991). Finally, the Southern European countries were also politically distinct from the advanced nations in having experienced greater or shorter periods of nondemocratic rule in the postwar era. Almost certainly, this was, ultimately, the critical factor in the exclusion of Southern Europe from systematic comparative analysis, particularly for the academic comparativists, whose major focus of interest was on establishing the impact on policy outcomes of democratic rule by parties of varying political complexion (Castles 1982). For these writers, Southern Europe was strictly a ‘no-go’ area, for how could one establish the impact of parties in countries where, until quite recently, there had been no democratic party competition? Obviously, today, the syndrome of underdevelopment in the sense outlined above no longer characterizes Southern Europe. As noted already, and as demonstrated in far greater detail subsequently, these countries have massively extended their social security provision, with a profile of growth, which has, over time, made their performance approximate to the OECD norm. Bureaucratic performance is also far less in doubt than it once was, not least as a consequence of the standardization of government reporting and accounting procedures that has come with membership of the European Community. While these countries have remained poor compared with the rest of the OECD, postwar economic growth rates have been high compared with all other OECD countries except Japan, so that the Southern European countries have improved their economic status relative to the remainder of the advanced world (OECD 1992a, Table 3.1). Above all, the democratic regimes newly instituted in the mid-1970s have

44

The Welfare State and Democracy

endured, becoming more institutionalized with each successfully negotiated change of government. And yet, the neglect of Southern Europe in academic comparative analysis persists. While single country monographs appear with increasing frequency, and while there have been important attempts to capture the features defining Southern European particularity (see Esping-Andersen 1993; Leibfried 1993; Castles and Ferrera 1996; Ferrera 1996; MIRE 1997), the vast majority of studies in comparative political economy using quantitative methods to locate the determinants of policy outcomes still perpetuate a lacuna which is increasingly difficult to justify.1 Given frequent laments on the methodological problem of ‘too many theories, too few cases’ (see Przeworski 1987), the tendency of such studies to ignore the New Southern Europe, when no longer legitimately excluded by an emphasis on the method of similarity, seems quite extraordinary. It seems equally odd to witness the emergence of a monograph literature on the social policy experience of the Southern European countries, which often examines national experience in terms of categories, problematics, and models generated by a comparative literature that altogether neglects systematic analysis of these very countries. In this chapter, I seek to make at least a tentative start in bringing the Southern European welfare state experience within the ambit of comparative social policy research. In the context of this volume, it seems appropriate to do this explicitly rather than implicitly; not by simply including these countries in an analysis of some particular aspect of social policy, but rather by asking whether the inclusion of these countries in our broad-gauge analysis of the development of social security modifies the conclusions of that analysis and whether it tells us anything interesting about the New Southern Europe. Obviously, within the ambit of a single chapter, it is impossible to analyze all aspects of social policy. A focus on social security and welfare transfers is chosen because this constitutes the largest aggregate of welfare state spending in most advanced nations. Readers are, however, warned that there is considerable evidence that the correlates of different components of welfare state expenditure differ quite markedly, and that any conclusions offered here would have to be supplemented by substantial further analysis before they could be generalized to the welfare state as a whole. In particular, readers should note that the evidence presented here offers only the most fragmentary support for the view that welfare expenditure is a function of the ideological conflict of left- and right-wing parties. This does not necessarily mean that ideological influences on Southern European welfare development were negligible. In other areas than social secur-

The Welfare State and Democracy

45

ity—for instance, the development of health services and the expansion of educational opportunity—they may have been considerable. The inquiry proceeds in a number of stages. First, I provide some basic descriptive statistics comparing various aspects of these nations’ overall aggregate social security expenditure development with that of the other OECD nations in the period 1960–90. This time-period is selected for detailed analysis in this chapter because it simultaneously captures the era in which the Southern European nations joined the European mainstream both in terms of levels of socioeconomic modernization and in terms of democratic consolidation. It should further be noted that there are many other ways of characterizing expenditure development than in aggregate terms (see discussion in EspingAndersen 1990), but that the data required for other approaches are more difficult to obtain—especially for the countries of Southern Europe. Nevertheless, it should be stressed that the conclusions arrived at here apply only to aggregate social security spending and to no other dimensions of welfare state development. Second, with a view to gaining a perspective on the social policy impact of the authoritarian regimes which characterized these countries in the 1960s and the democratic transitions which took place from the mid-1970s, I compare the course of change and development of social security transfers in Greece, Portugal, and Spain with the experience of the other OECD countries. Third, I reexamine a number of the key propositions of the comparative literature—that social security transfers expenditure is ultimately a derivative of socioeconomic maturity, that social security manifests a convergent trajectory of development driven by socioeconomic forces, and that partisan control of government and governmental policy responses strongly influence at least some kinds of welfare policy outcomes—and discuss the extent to which the inclusion of the Southern European countries in the analysis modifies conclusions based on the more restricted OECD data-set. At this stage, I also look at the experience of each country over time with a view to locating the role of economic, social, and political factors in each country’s social security development. Fourth, I relate the Southern European experience of social security development to the concepts of ‘welfare regimes’ (Esping-Andersen 1990) and ‘families of nations’ (Castles 1993) with a view to establishing the extent to which these nations share a distinctive cultural identity either as a grouping set apart from other advanced nations or in common with some sub-set of such nations. It is my hope that, irrespective of the extent to which this review of the data on social security level and development from a Mediterranean perspective contributes anything novel to comparative public policy research, it will serve as a valuable prologue to further research and a worthwhile

46

The Welfare State and Democracy

overview of the forces influencing the historical development of social security in Southern Europe. There are four additional points that should be noted before embarking on the analysis. First, the methodology adopted in the chapter is necessarily statistical, utilizing a combination of simple descriptive statistics and regression analysis. With a host of causal variables to review, three countries to analyze for a period of three decades, and comparisons to make between these countries and the rest of the OECD, no other strategy is possible. Readers should not be put off by the statistics, which do no more than assert that certain features of welfare development are associated with certain economic, social, and political factors. Sometimes, the numbers and figures tell us stories that would not be readily apparent from other methodologies. Just as often they warn us of the ambiguity of stories often taken for granted and of the urgent need for the kind of detailed historical exegesis to be found in the other chapters of this volume. The second point to be noted is that the careful reader will find some minor discrepancies between the data used in this study to describe cross-national differences in social security transfers in 1960 and in 1990 and the data used to describe time-series change in social security and welfare transfers between the early 1960s and the late 1980s. The former data come from OECD Historical Statistics (OECD 1992a) and are shown in Figures 2.1a, 2.1b, and 2.1c in the next section. The latter data come from a closely related source: the OECD Government Household Transfers Database, as reported by Varley (1986) and with further data supplied to this author by the OECD Secretariat, and are shown in Figures 2.2a, 2.2b, and 2.2c. On the whole, there is little reason to believe that these discrepancies are conducive to any serious risk of misinterpretation, with the major difference between the two being the exclusion of some relatively minor items in the Government Household Database in an attempt to increase comparability. Although, in consequence, the expenditure figures reported in the time-series data are generally somewhat lower than those reported for the cross-sectional data, cross-national, and cross-time variances in the two data-sets appear quite consistent. I have, indeed, taken advantage of this fact to extend the Portuguese time series in the Government Household Transfer Database by means of interpolating expenditure increases reported in OECD Historical Statistics for the years 1986–9. A third way in which this chapter differs from the others in this volume is that Italy is excluded. The reasons lie in both the subject matter and the method of this research. With regard to social security development and its immediate correlates, Italy does not share at least two of the crucial features of the syndrome of underdevelopment typical of

The Welfare State and Democracy

47

the countries of the New Southern Europe. At the beginning of the period, in 1960, Italy was quite the reverse of a social security laggard, and the trajectory of social security thereafter was, therefore, wholly different from that of the other Mediterranean countries, approximating the pattern typical of the OECD nations generally. The other difference was the absence of authoritarian government in Italy in the period under survey here. It is, of course, possible that these two deviations from the syndrome of underdevelopment sketched above are connected, and much of the analysis of this chapter is relevant to an understanding of the links between democratization and policy outcomes, but, certainly, normal statistical methodologies do not permit comparisons to be made of trajectories of development separated by several decades. All we can really say is that, in terms of a comparison of social security expenditure development in the first decades of the postwar era, Italy was not a part of what is now seen as the New Southern Europe, but was rather a part of contemporary mainstream European social security development. Ferrera has noted that, already, by ‘the outset of the Second World War, the Italian welfare state existed as a fairly well articulated institutional system,’ amounting to ‘some 15–17 of the state budget,’ but that it ‘remained for two decades (after the war) a fragmented arena of marginal adjustments, additive expansions and clientelistic exchanges’ (Ferrera 1986: 389–90). Those who concentrate on the Southern European character of the Italian development concentrate on these latter features. The data presented in Figures 2.1 and 2.2 below confirm the former proposition—that, in expenditure terms at least, Italy was a fully developed social security state by 1960, and followed a pattern of subsequent expenditure growth quite different from that elsewhere in Southern Europe. Finally, it should also be noted that the analysis presented in this chapter is restricted to the period between 1960 and 1990. This is because my primary objective is to measure the extent to which cross-national differences and patterns of policy change are the products of the ‘modernization’ and ‘democratization’ processes that are a central theoretical focus of this book and, indeed, the entire series on the New Southern Europe. At the beginning of this period, Greece, Portugal, and Spain were (along with Ireland) the poorest and least developed countries in Western Europe. By the end of the 1980s, all three of them had undergone impressive processes of modernization. In the early 1970s they were the only three nondemocratic countries in Western Europe. By the end of the 1980s their new democratic regimes had been fully consolidated. In order to systematically explore the impact of these two transformative processes, we focus our analytical attention on those decades when these two processes were unfolding most rapidly.

48

The Welfare State and Democracy

No Longer the Rearguard Figure 2.1 charts the changing cross-national distribution of the OECD nations in respect of social security transfers as a percentage of GDP in the period 1960–90. Figure 2.1a makes clear the extent to which the countries of the Iberian Peninsula constituted a social policy backwater in 1960. In that year, of the OECD countries, only Spain and Portugal spent less than 50 percent of the OECD average of 7 percent of GDP. Greece was also close to the bottom of the distribution, but stood out as a laggard to a far lesser degree than its Mediterranean cousins, being one of an appreciable grouping within 2 percentage points of the OECD average. Historically and culturally defined groupings of countries, of which we shall say far more later, are quite apparent in Figure 2.1a. Apart from the Southern rearguard, it is possible to distinguish other geographical or cultural consanguinities. On the whole, the Englishspeaking countries cluster somewhat below the OECD mean and the Scandinavian countries just above it. Outstandingly, there is a central Western European vanguard grouping containing not only social insurance pioneers like Austria and Germany, but also the Mediterranean nations of Italy and France. Here certainly, there is no indication of a Mediterranean commonality, but rather a clear distinction between a social security rich Mediterranean core and a social security poor Mediterranean periphery. As Figure 2.1b shows, this distinction was not to last. In 1990, the rearguard was no longer Iberian or indeed Mediterranean. Rather it consisted of a miscellaneous grouping—Australia, Finland, Japan, and the United States—that had all been low in the distribution three decades earlier and were now at the bottom of the OECD leaguetable. Iceland, the one OECD country that actually experienced a relative decline in transfers over the period, made up the rear. Of the Mediterranean countries, Portugal had the lowest expenditure, but, even so, with 12.1 percent of GDP devoted to social security in 1989, it came not far behind the OECD average of 13.3 percent. By the late 1980s, Spain and Greece exceeded the OECD average and both were marginally within the top-half of the distribution. Moreover, these countries had now caught up with Italy, a former Mediterranean transfer expenditure leader and, to the extent that a distinctive Mediterranean grouping is distinguishable, it is one located at the center rather than the bottom of the distribution. This grouping had leap frogged all the English-speaking countries and lay just behind the Scandinavian cluster and the still quite distinct central Western European vanguard.

UK Australia Belgium Luxembourg Norway United States Switzerland

Japan

Greece

UK

Australia

Spain

Ireland

Switzerland

Switzerland

Ireland

UK

Germany

Iceland

Greece

Denmark

Italy

Norway

Sweden

Finland

Canada Sweden

Denmark

Norway

Netherlands

Denmark

Italy*

Ireland

Sweden

Netherlands

Greece

Austria

Belgium

France

Luxembourg

Belgium

Germany

Portugal

Luxembourg

Austria

Spain

Netherlands

France

49

Japan Finland

The Welfare State and Democracy

Italy

(a)

Finland

France

16

United States

Austria

14

United States

12

Japan

Canada

10

Portugal

Canada

8

Germany

6

Portugal

4

Spain*

Australia

2

0

(b) 30

25

20

15

5

0

10

(c) 0.2 0.18 0.16 0.14 0.12 0.1 0.08 0.06 0.04 0.02 0

Figure 2.1. Social security transfers as percentage of GDP in (a) 1960, (b) 1990, and (c) 1960–90 growth rate

Iceland

Iceland

50

The Welfare State and Democracy

An important hypothesis in the comparative literature suggests that socioeconomic transformation leads to a process of welfare catch-up, and data for the annual average percentage growth of social security expenditure (see Figure 2.1c) show that, in general, those were the countries with the lowest spending in 1960 which experienced the most rapid growth thereafter. Spain and Portugal at the bottom of the distribution in 1960 expanded their social security effort more than any of the other OECD countries and all three countries of the New Southern Europe were in the vanguard of growth measured in this way. Although, with the focus on growth, distinct country groupings are less obviously apparent, one can discern a tendency for relatively rapid growth by the Scandinavian countries and a tendency for the central Western European social security leaders of the 1960s to be in the bottom half of the distribution in respect of change. Both of these tendencies are compatible with the notion of social policy convergence which will be examined more explicitly in a later section.

Dictatorship, Democracy, and Social Security The data reported in Figure 2.1 provide only a very summary picture of social security development in the three countries of the Mediterranean periphery. To begin the task of isolating the forces pushing that development requires a more detailed examination of the pattern of spending in each of these countries on a year-by-year basis. At this level of detail, it starts to become possible to enter into, at least, a preliminary discussion of questions which are clearly of vital importance in understanding these countries’ trajectories of social expenditure growth: namely, whether authoritarian rule and the transition to democracy have made a difference to social policy outcomes. The assumption that democracy is more propitious to welfare state development than authoritarianism cannot be taken for granted. Indeed, empirical research into the emergence of welfare states in Western Europe suggests that ‘the propensity to introduce social insurance schemes was much higher in the constitutional-dualistic monarchies (Austria, Denmark, Germany, Sweden) than in the parliamentary democracies’ (Flora and Alber 1981: 72). One potent reason for the early adoption of such schemes in the former nations was a need on the part of nonelected governments to defuse the potential for emergent working class mobilization. Moreover, an explanation based on the imperative of maintaining popular support may also account for the maintenance of existing social security schemes under the dictatorial regimes of the 1920s and 1930s, with the earliest postwar data avail-

The Welfare State and Democracy

51

able (see Flora 1987) indicating that Austria and Germany remained at the forefront of European social insurance expenditure in the immediate postwar period. Ferrera also notes that the Fascist period in Italy ‘witnessed a great increase in welfare programs, especially in the field of sickness and family policy’ (Ferrera 1986: 389). On the basis of this evidence, it is arguable that there is nothing inherently contradictory in the notion of authoritarian regimes fashioning and maintaining a strong social insurance state. Whether such an argument is applicable to the Iberian authoritarian regimes or the Greek junta is much more disputable. On the one hand, there is, at least some evidence of social security growth. Gunther, in a pioneering study of Spanish public policy under Franco, points to the seeming ‘anomaly’ of increased social service spending in a nation in which the regime’s primary supporters had an interest in ‘a small public sector’ (Gunther 1980: 48–51). On the other hand, and the logic underpinning the very notion of a Spanish anomaly, the Southern European regimes were, far more than those of central Western Europe, conservative rather than mobilizational in ideology. Their rationale was to resist the winds of modernizing change of which enhanced public expenditure and taxation were seen as conspicuous elements. On that reading, one might expect the pretransition regimes to constrain social security growth and the process of transition to involve an upsurge of social security expenditure. Such an upsurge might be considered the more likely, since newly installed democratic regimes desperately needed to maintain a delicate balancing act between political extremes, and social rights extended to the working class interests had a real potential to serve as a counter to a constitutional garantismo (Di Palma 1986) accorded to the nostalgic right.

The Impact of Dictatorship There are in fact two quite distinct kinds of questions we may ask about the impact of dictatorship: did authoritarian rule reduce or increase the level of spending compared with some past reference point, or did it impede or enhance expenditure compared with the trend of future development. The substantive statistical analysis of individual countries in this study focuses exclusively on the latter, but, here, quite briefly, and in a nonstatistical manner, we discuss evidence concerning the former. One reason for focusing on a comparison with later stages of welfare development is that it is only in the Greek case that we have reasonably complete evidence permitting an assessment of the impact of

52

The Welfare State and Democracy

authoritarian rule. In Greece, because the Colonels’ rule fell entirely within the period for which we have data, it is possible to come to somewhat more solidly based conclusions as to whether the nondemocratic regime was associated with an increase or decrease in social expenditure. In fact, this evidence suggests that change was quite strongly retarded by the advent of the regime. According to the OECD Government Household Transfers Database on which we rely for our time-series data, Greece was the single country in the OECD to experience a decline in social security and welfare transfers to households as a percentage of GDP (from 5.9 to 5.7 percent) in the period 1965–73. However, as is apparent from Figure 2.2a, the trajectory over this period was not uniform, with an increase in expenditure until 1968 and a more or less continuous decline thereafter. Such a pattern is consistent with budgeted increases from the preauthoritarian regime being reversed by the Colonels’ regime after several years of lag. Other analysis of data emanating from the same source (OECD 1985: 22) shows that Greece over the period 1960–75 manifested the lowest income elasticity of social expenditure of any of the OECD countries.

(a)

1962

1964

1966

1968

1970

1972

1974

1976

1978

1980

1982

1984

1986

1988

1990

1962

1964

1966

1968

1970

1972

1974

1976

1978

1980

1982

1984

1986

1988

1990

16 14 12 10 8 6 4 2 0

(b) 16 14 12 10 8 6 4 2 0

The Welfare State and Democracy

53

1987 1986

1989

1985 1984

1983

1981

1979

1977

1975

1973

1971

1990

1988

1982

1980

1978

1976

1974

1972

1970

1968

1966

1964

16 14 12 10 8 6 4 2 0 1962

(d)

1969

1967

(c) 16 14 12 10 8 6 4 2 0

1990

1989

1988

1987

1974

1968

1960

(e) 16 14 12 10 8 6 4 2 0

Figure 2.2. Southern European social security transfers as percentage of GDP in (a) Greece 1962–90, (b) Portugal 1962–90, (c) Spain 1967–90, (d) Italy 1962–90, (e) OECD Average 1960, 1968, 1974, 1987–90

Overall, the implication is strong that the Colonels’ rule was associated with a major constraint on social policy expenditure in a nation which, until the mid-1960s, was appreciably farther along the path of social security development than the other countries of the Mediterranean

54

The Welfare State and Democracy

periphery. This does not, of course, mean that the decline in expenditure was directly caused by the imposition of nondemocratic rule, and we shall have more to say on this topic in the context of our final multivariate statistical model of Greek social security development. The available expenditure data on Portugal and Spain provide evidence that is at best inferential and fragmentary. Since we dispose only of data relating to these regimes’ later years, we cannot directly assess the extent of any decline from preauthoritarian times. We do know that Spain had initiated social insurance schemes dealing with pensions and unemployment in 1919 (Guille´n 1997), and may surmise that Spain’s very low level of reported government expenditure in 1960 implied a cut back since the time of the Republic. In Portugal, all schemes existing in 1960 had been instituted under Salazar’s regime and one can at best argue that the 1960 level of expenditure hardly demonstrates that the regime was seriously expansionary in its early decades. However, in the period from 1960 to the fall of the authoritarian regimes in the mid-1970s, there were some indications of change. In Portugal, as shown in Figure 2.2b, this came after Salazar left office, with some upturn in the very gradual growth pattern of expenditure in the two years preceding the fall of the regime in 1974. However, in Spain, growth after 1960 was, according to OECD Historical Statistics, far more rapid, with an increase of 2.3 percentage points in spending as a percentage of GDP between 1960 and 1967. Growth thereafter is shown in Figure 2.2c, with a further increase of 2.7 percentage points up to Franco’s death in 1975. The reasons why authoritarian Spain was different from the other Southern European nations in embracing change in the late 1960s and early 1970s requires some explanation. One real possibility is that much of the earlier growth of expenditure was more apparent than real. The 1963 Basic Law of Social Security consolidated previously fragmented corporatist schemes which, from 1967 (the date at which the time series in Figure 2.2c begins) for the first time counted as public expenditure in National Accounts (see Esping-Andersen 1993). However, Spanish social security growth after 1967 seems real enough and it is certainly possible to locate historical factors inhibiting change in Greece and Portugal which had no obvious counterparts in Spain during this period. In Greece, the inhibiting factor was, of course, the rise of the Colonels, with power falling into the hands of a military clique obsessed by the pursuit of democratic enemies and the worsening crisis in Cyprus. In Portugal, it may be attributed to long drawn out colonial wars which led to a substantial diversion of resources to the defense budget. Military appropriations in Portugal averaged around 7

The Welfare State and Democracy

55

percent of GDP throughout the 1960s and early 1970s, much the highest percentage of any country in Western Europe (SIPRI various dates). On the other hand, and looking ahead to our discussion of the hypothesis that the development of social expenditure is largely determined by socioeconomic forces, a view which would also make the distinction between dictatorship and democracy irrelevant, the explanation of the difference between these countries in the latter years of authoritarian rule cannot be one based on the social policy impact of economic development alone, for, in the decade of the 1960s, economic growth rates in all three Southern countries were virtually identical and much higher than those prevailing in the rest of Western Europe (OECD 1992a, Table 3.2).

The Impact of Democratic Transition Turning now to the impact of democratic transition on social expenditure change, we are confronted with the obvious difficulty of defining the length of the transition period. Here, we suggest three possibilities, all of which have some appeal. First, we may argue that transition may be best conceived of as consisting of a brief period immediately following the fall of an authoritarian conservative regime. The underlying assumption, here, would be that this period is likely to be one in which there would be an immediate surge of pent-up popular demands in the social security arena. Taking a two-year period, and seeking to capture the possibility of immediate impact by allowing only a few months of lag-time, would imply that transition effects occurred in both Greece and Portugal in 1975–6 and in Spain in 1977–8. Second, we might suggest that transition effects are likely to be felt over a rather longer period which allows resources to be shifted between different programs in the state budget. Assuming four years to be an appropriate time for such an impact, and allowing a lag of somewhat over a year for demands to be crystallized as budget priorities, suggests transition impacts in Greece and Portugal in 1976–9 and in Spain from 1978–81. Lastly, given that policy change takes time and may have to surmount many hurdles, it might be argued that transitional effects might be felt over a very long period indeed. In effect, this amounts to a catch-up argument, suggesting that in the period from the overthrow of the authoritarian regime until the present we might expect a convergence in performance between each of the Southern European countries and OECD social outcomes in general. Looking at Figures 2.2a, 2.2b, and 2.2c in terms of immediate transition effects, a rather mixed picture emerges. There is no sign of

56

The Welfare State and Democracy

immediate social security upsurge in Greece, but quite substantial effects in both Spain and Portugal, with a two-year growth of 2.3 and 2.5 percentage points of GDP respectively. In the Portuguese case, this may be regarded as a particularly dramatic growth spurt, coming off a low base and representing a 50 percent growth of expenditure in a very short period of time. Certain caveats are, however, appropriate in discussing the Portuguese case: first, that the governments under whose auspices it occurred were not democratic and second, that the observed changes may have been due to confounding effects, most notably an unemployment level which more than tripled in these two years (from 1.7 to 6.3 percent of the labor force [OECD 1992a]). In other words, it may well be that Portugal’s leap forward in social expenditure in these years was not necessarily an immediate consequence of the outpouring of pent-up popular demands, but might as plausibly be attributed to the leftist ideology of the military regime which presided over the transition or to repercussions of the first Oil Shock. Examining the data in terms of a transition period articulated to pick up rather longer term budget impacts, the story for Greece remains unaltered: the latter part of the 1970s was, at best, a period of modest recovery for the Greek welfare state and Greek expenditure levels were not to show any real growth until the early 1980s. However, taking this somewhat longer perspective, the initial upsurge of Spanish spending in 1977–8 is now seen as merely the beginning of a much more dramatic welfare transformation. In the period 1978–81, Spanish social security transfers grew from 8.2 to 12.6 percent of GDP, an increase of approximately 50 percent, much the most rapid period of growth shown in Figure 2.2c. Finally, the Portuguese picture of a postauthoritarian social security explosion is substantially modified in the longer view, with Figure 2.2b showing a very nearly steady state development between 1976 and the end of the decade. Finally, we may ask whether there was a process of convergence taking place between the fall of the authoritarian conservative regimes and the present. Certainly, between the dates at which the respective juntas fell and the latest date on each country for which we have data, growth in all these countries was markedly faster than in the OECD in general. Greek social security grew by 8.3 percentage points, Portuguese by 5.0 points, Spanish by 5.4 points, and the OECD average (1974–90) by only 4.4 points. Moreover, measured in growth rate terms, which take account of each country’s initial level of social security spending, the three Southern nations were all amongst the four fastest growing OECD countries, with Greek expenditure and Portuguese spending doubling measured in GDP terms and Spanish expenditure increasing by around 80 percent. For Greece and Portugal that marked an enormous speedup in the

The Welfare State and Democracy

57

process of welfare transformation. For Spain, change in the rate of social security development was less marked, given the not insubstantial growth in the latter years of the Franco period.

Measuring Democratic Transition Having looked at the impact of dictatorship and democratic transition on social security in a broadly descriptive fashion, I now embark on a process of analysis which will culminate in the presentation of multivariate statistical models which attempt to account for the observed patterns of social security development in Greece, Portugal, and Spain over a period from the late 1960s to the late 1980s, with the starting and finishing dates for the various countries differing depending on the precise years for which complete data on all relevant variables is available for each. The need for such models is strongly underlined by the constant reference in the descriptive account above to confounding factors, which might possibly modify or radically change interpretations derived from a simple examination of the trends manifested in Figures 2.2a, 2.2b, and 2.2c. The adumbration of these models will proceed in tandem with the articulation of models for OECD social security in both 1960 and 1990. The data coverage here is for a group of twenty-one nations, excluding only New Zealand and Turkey (on the grounds of missing data) and Japan (because of a wish to explore family of nations differences only pertaining to nations of European cultural descent). It is my hope that this combination of cross-time and cross-national analysis will provide a reasonably detailed picture of the processes of social security development in the New Southern Europe and of how those processes related to more general trends in the Western democracies. It might be thought that the evidence of cross-national comparison was sufficient to establish the determinants of social security development both in the countries of Southern Europe and in the OECD as a whole, rendering a detailed analysis of the cross-time experience of the individual nations superfluous. However, that is not so, since the factors pushing change over time within countries may not be the same as those accounting for variation among countries.2 It is necessary to examine the cross-time experience of the Southern European nations in some detail in order to establish how their social security effort shifted in relation to the general OECD pattern, since there is no inherent reason why the factors impelling that shift should have been generally manifested across the OECD or, indeed, why such forces should have been the same in each individual nation. Hence, this

58

The Welfare State and Democracy

study shifts backward and forward between a cross-time and a crossnational perspective in the hope of locating the forces shaping the trajectories of development by which countries of Southern Europe ceased to be welfare laggards.

Cross-time Variation Here, I begin by examining the correlations between each of our measures of democratic transition and social security expenditure as a percentage of GDP in Greece, Portugal, and Spain for each year of periods beginning as early as possible in the 1960s and culminating as nearly as possible in 1990. Later sections of the chapter will introduce a range of additional factors, with the overall purpose being to provide as convincing account as possible of the cross-national and cross-time variation illustrated in Figures 2.1 and 2.2. At this stage of the analysis, the correlations presented are only bivariate in character and significant findings are to be read only as indicative of associations between variables and not as demonstrating proven causal relationships. In Tables 2.1a, 2.1b, and 2.1c, UPSURGE refers to a two-year transition effect, BUDGET IMPACT to a longer term four-year transition effect mediated through state budgets, and THRESHOLD to the years after the fall of each respective authoritarian conservative regime. It should be noted that THRESHOLD merely measures the difference between the authoritarian conservative regime and the post-transitional period and that a positive correlation, indicating high social security levels in the

Table 2.1

Social security development and democratic transitions

Variable

Correlation

T-statistics

(a) Greece: 1962–89 THRESHOLD UPSURGE BUDGET IMPACT

0.37 0.20 0.19

2.1 1.0 1.0

0.93 0.09 0.05

13.3 0.5 0.2

0.91 0.07 0.21

10.6 0.3 1.0

(b) Portugal: 1962–89

(c) Spain: 1967–90

The Welfare State and Democracy

59

posttransitional period, can as readily be interpreted in terms of the negative impact of the prior dictatorial regime. Each transition measure is a dummy variable and lags are as indicated in the discussion of the previous section. The final column of each table reports t-statistics, a measure of the significance of the relationships. T-statistics of 2.0 or more indicate degrees of association which are statistically significant. These results, which are a necessarily preliminary to our later analysis, are not surprising in most respects. UPSURGE and BUDGET IMPACT effects would certainly be expected to be experienced in terms of a major change from the former level of spending, and, as we have already noted, such changes can be discerned in both the Portuguese (UPSURGE) and Spanish (BUDGET IMPACT) cases (see Figures 2.2b and 2.2c). However, they are rather unlikely to be experienced as a level of development higher than in subsequent years. The insignificant findings for these variables in all cases merely confirm this. The important question for the later multivariate stage of the research is whether these transition effects become more significant in the context of multivariate models which seek to control for the factors leading to higher expenditures throughout the 1980s. On the other hand, THRESHOLD is highly significant in every case, showing an association between high levels of spending and the posttransitional period. That, of course, is what would be expected in light of our knowledge of increased expenditures over time in each country, and it leaves for subsequent analysis the question of whether the pattern of development in each country is attributable to the turn to democracy or to other socioeconomic and political factors with which that transformation was covariant. Perhaps, the only cause for even minor surprise is the relative weakness of the THRESHOLD finding for Greece, but that is, of course, compatible with the implication of our earlier analysis: that because Greece was initially an appreciably more advanced welfare state than the Iberian nations, and despite the possibly retarding effect of the Colonels’ regime, there was less of a difference between expenditure levels at the beginning and the end of the period than in the other nations.

Cross-national Variation Given the apparent strength of the THRESHOLD effect in each of these countries, it is also of interest to assess the impact of democracy on the cross-national variation manifested by the OECD nations. In other words, was one of the reasons Portugal and Spain spent less than other Western nations at the beginning of the period that they were dictatorships and the other nations were not? One simple way of assessing this

60

The Welfare State and Democracy

proposition in statistical terms would be to use a dummy variable for dictatorship, but such a procedure is extremely crude. For 1960, it would involve the attribution of any difference between the Iberian nations and the rest of the sample to the fact of authoritarian rule and, in 1990, it would simply mean that there was no variation to measure. Rather than a simple dummy variable, I use a measure of democratization which, not merely distinguishes dictatorship from democracy, but further distinguishes between different degrees of democratization. The measure used is Vanhanen’s (1984) index of democratization, which is a multiplicative measure of the percentage of the population voting in elections and the share of the vote cast for all parties bar the largest in any given parliamentary election. The logic linking social security expenditure to democratization in this sense is the argument that the greater the degree of democratic participation and of party competition, the greater will be the pressure on politicians of all parties to concede welfare demands (see Castles 1990). In Table 2.2a, I report the bivariate correlation between OECD social security expenditure as a percentage of GDP in 1960 and the democratization index. In Table 2.2b, the correlation between the same variables is reported for 1990. In addition to correlations and t-statistics for the entire sample of twenty-one OECD nations which is the focus of analysis here, I report in parentheses correlations and significance levels for a sample excluding the three nations of the New Southern Europe, thereby permitting an assessment of the degree to which the results derived from the broader sample reflect the peculiarities of these latter nations. In 1960, the larger sample manifests a very strong relationship between democratization and expenditure, implying that the greater was the degree of democratic participation and party competition, the higher was the level of social expenditure. Moreover, the only somewhat weaker result for the sample excluding the countries of Southern Europe shows that, at this time, the positive effect on expenditure was one which applied generally across the OECD nations, deriving almost

Table 2.2 OECD transfers and democratization Variable

Correlation

T-statistics

(a) 1960 (21 cases with the countries of New Southern Europe; 18 without) Democratization

0.74 (0.57)

4.8 (2.8)

(b) 1990 (21 cases with the countries of New Southern Europe; 18 without) 0.31 (0.28)

1.4 (1.2)

The Welfare State and Democracy

61

as much of its explanatory force from the low expenditure of democratic laggards, like the US and Switzerland,3 as from the more egregious authoritarianism of Spain and Portugal. However, by 1990, the bivariate impact of democratization is no longer significant in either sample. One possible interpretation of these findings, compatible with an observed tendency for democratization to have increased in virtually all the nations of the sample over this period of three decades, is that the relationship with social expenditure depends on there being a structural shift in the degree of responsiveness to political demands. Such an argument would suggest that, where there is considerable cross-national variation in the extent of democratization, differences between nations might well have a major impact on expenditure, but that, once all or most nations experience a structural shift to greater responsiveness, the impact would be much diminished or even disappear. It should be mentioned that such an interpretation does not require that we establish the existence of a cross-time democratic transition effect in any or all of the Southern European countries. If the structural shift involved a polity moving to a situation in which popular demands could mediate other influences—the socioeconomic, political, and policy forces to be discussed in subsequent sections— democratization could not appropriately be seen as a sufficient condition of welfare expansion, although, in respect of certain influences, it might possibly be a necessary one. Because it is the purpose of this chapter to focus on the distinguishing characteristics of policy development in the countries of the New Southern Europe, I have discussed issues concerning the social policy impact of dictatorship and democracy at far greater length than is usual in the comparative public policy literature. In remaining sections, I turn to issues which have been far more thoroughly rehearsed in the literature in order to see to what extent they can shed light on the social policy experience of these countries over the three decades 1960–90.

On the Socioeconomic Determinants of Social Security This section explores one of the most influential theories in the field of social policy determination: that expenditure is a function of a country’s level of socioeconomic development. This hypothesis, which harks back to pioneering research by Cutright (1965) and Wilensky (1975), has been formulated in a variety of ways. Obviously, a measure of economic modernity is crucial and that most generally used is GDP per capita. The percentage of the aged in the population has also featured in the literature as the immediate channel by which socioeconomic modernity

62

The Welfare State and Democracy

intrudes into the social policy arena, both as a demographic category in need and as a political agency of change (see Pampel and Williamson 1989). Finally, there has been some consideration of the impact of institutional inertia, with the years since a program was established serving as a rough measure of the extent of incremental push for program expansion (see Wilensky 1975). Since, in 1960, the countries of New Southern Europe were very economically underdeveloped compared with the rest of the OECD and manifested relatively youthful demographic profiles, and since social programs were, in general, introduced relatively late in these countries, there are prima facie reasons to believe that such factors may have been implicated in these countries’ weak social security performance as shown in Figure 2.1a. Moreover, since the rate of socioeconomic development has been more rapid on the Mediterranean periphery than in the OECD in general, these factors also appear as possible candidates for explaining the rapid pace of social security development in the New Southern Europe as demonstrated in Figure 2.1c.

Cross-national Variation In Table 2.3a, I report the bivariate correlations between OECD social security expenditure as a percentage of GDP in 1960 and real GDP (data calculated from Summers and Heston 1991 and measured in thousands of US dollars per capita), the proportion of the population 65 years of age or older (OECD 1992d) and the years elapsed since the passing of initial legislation to establish an age pension system (US Department of Health, Education and Welfare 1982 and later years). In Table 2.3b, correlations for precisely the same variables are reported for 1990. Table 2.3 OECD transfers and socioeconomic variables Variable

Correlation

T-statistics

(a) 1960 (21 cases with the countries of New Southern Europe; 18 without) Real GDP pc (000s) 65þ Program Experience

0.18 ( .32) 0.61 (0.50) 0.46 (0.40)

0.8 (1.4) 3.4 (2.3) 2.2 (1.7)

(b) 1990 (21 cases with the countries of New Southern Europe; 18 without) .09 ( .25) 0.54 (0.53) 0.17 (0.14)

0.4 (1.0) 2.8 (2.5) 0.7 (0.6)

The Welfare State and Democracy

63

The results in Table 2.3a show a moderately strong socioeconomic impact on social security expenditure levels in 1960. Both the age structure of the population and program experience are significantly associated with outcomes and these results are not seriously influenced by the exclusion of the three Southern nations. The story regarding per capita real GDP is very different, with the relationship positive but well below the level of statistical significance for the sample including all three Mediterranean countries. With these countries excluded, the relationship actually becomes negative, although still insignificant. With the exception of the age structure variable, the 1990 OECD comparison reported in Table 2.3b suggests a weakening of socioeconomic influence over time. In the bivariate analysis, there remains no sign that a country’s level of economic resources impacts positively on cross-national variance in expenditure and, with or without the three Mediterranean nations included in the sample, the relationship is negative. A quick glance back to the countries toward the bottom of the distribution in Figure 2.1b seems to confirm that national wealth is, of itself, insufficient to account for high levels of transfers expenditure. Program experience has also declined very markedly as an influence, a finding perhaps more consonant with a notion of program maturation, and hence ultimate expenditure convergence, than of the incremental development originally postulated by Wilensky (1975). At this very preliminary stage in the analysis, age structure looks as if it might have a crucial role, since, unlike both economic level and democratization, it remains a significant positive correlate of expenditure irrespective of time and sample.

The Convergence Thesis The socioeconomic hypothesis of social security development was in its initial articulation closely tied to the view that a process of social policy convergence was taking place driven by the forces of economic development or, as Wilensky put it (1975: 27): ‘economic growth makes countries with contrasting cultural and political traditions more alike in their strategy for constructing the floor below which no one sinks’. In order to assess this argument more fully we need not merely to look at levels of development, but also at the changes taking place over time. Did the changes described in Figure 2.1c constitute a process of convergence and were they driven by socioeconomic forces? Table 2.4 seeks to answer these questions by examining the correlations between annual average growth in social security expenditures as a percentage of GDP and the level of expenditure in 1960, the change in real GDP per capita

64

The Welfare State and Democracy

and change in the age structure. The association with the 1960 level tells us whether there is an a priori case for convergence, that is, whether the countries that started out with the lowest levels of expenditure grew most thereafter. The reported associations with real GDP change and age structure change tell us something about the extent to which that process may have been driven by socioeconomic factors. Since virtually all countries already had the full range of social security schemes in place by 1960, we do not analyze the relationship between growth in social security and change in program experience. The choice of measuring change in growth rate rather than first difference terms biases the data in favor of the convergence thesis, since countries with low initial spending have their subsequent change magnified in growth rate terms. If the convergence thesis cannot pass this test, it cannot pass any test. As previously, we report in parentheses findings based on a sample excluding the countries of the New Southern Europe. That is important, since a number of writers have suggested that there was no convergence tendency amongst the richer advanced countries constituting the normal OECD sample for comparative research (Castles 1982; O’Connor 1988). In this view, apparent signs of convergence would be purely an artifact of including initially relatively underdeveloped nations, such as Greece, Portugal, and Spain, in the sample. The results in the first column of Table 2.4 suggest that there was a moderately strong pattern of convergence exhibited between 1960–90 for the OECD sample as a whole, and that it was, seemingly, driven more by GDP growth than by changes in the age structure of the population. However, the much diminished and statistically insignificant result for the relationships between expenditure growth and initial level of spending and between expenditure and GDP growth in the smaller sample make it clear that this convergent tendency is substantially a function of the inclusion in the sample of countries of the New Southern Europe. This latter finding supports the view that extreme caution is necessary in identifying forces conducive to a general process of convergence in welfare outcomes in the advanced industrial nations. More sensibly, perhaps, research should be focused on the factors Table 2.4 Growth of OECD transfers and socioeconomic variables, 1960–90 (21 cases with the countries of New Southern Europe; 18 without) Variable

Correlation

1960 Level Change GDP Change 64þ

.60 ( .29) 0.51 (0.24) 0.33 (0.23)

T-statistics 3.2 (1.2) 2.6 (1.0) 1.5 (0.9)

The Welfare State and Democracy

65

peculiar to the individual nations of Southern Europe or to the group as a whole which may have been conducive to lessening the welfare gap between them and the rest of the OECD.

Cross-time Variation The processes of democratic transition already discussed may constitute one such factor. Another might well be a trajectory of socioeconomic development peculiar to the nations of the New Southern Europe. Again, it is worth emphasizing that the factors accounting for change time within countries may be quite different from those accounting for variation amongst countries. For instance, while changes in real GDP across OECD countries may not have had a major impact during the period 1960–90, they may well have had such an influence in the New Southern Europe, the difference being attributable, at least, in part, to the relative stage of social policy maturation of the various nations, an interpretation more than somewhat congruent with the strong relationship between the initial level of expenditure of OECD nations in 1960 and their subsequent growth rate observed in Table 2.4. In Tables 2.5a, 2.5b, and 2.5c, I report bivariate correlations between social security expenditure as a percentage of GDP in Greece, Portugal, and Spain for periods beginning in the 1960s culminating in the late 1980s and real GDP per capita (Summers and Heston 1991) and the proportion of the population of 65 years or over. Both of the independent variables are lagged by one year to facilitate a causal interpretation of the models subsequently derived from this analysis. It should be

Table 2.5 Transfers and socioeconomic variables Variable (a) Greece: 1962–89 Real GDP 65þ

Correlation

T-statistics

0.72 0.71

5.3 5.0

0.94 0.97

14.6 18.7

0.77 0.93

5.7 11.5

(b) Portugal: 1962–89

(c) Spain: 1967–90

66

The Welfare State and Democracy

noted that age structure data is missing for Portugal for the period 1976–9. In all these tables, a very close coincidence is apparent between socioeconomic development, whether measured by the log of real GDP or the age structure of the population, and social security development. Some caveats are, however, appropriate concerning premature conclusions as to which of these factors is most important and whether they are the only, or even the main, determinants of policy outcomes in this field. First, real GDP and age structure are themselves closely associated, leaving it unclear which has the most significant impact in each case. Second, both the findings for Greece and that concerning real GDP for Spain leave an appreciable degree of variance statistically unexplained.4 A possibility strongly hinted at in the comparative public policy literature is that, in order to account fully for outcomes, we may need to supplement socioeconomic with political and policy variables. In an earlier section, I have already established a prima facie case that one political variable, a structural shift in responsiveness to political demands attendant on the transition from dictatorship to democracy, is closely associated with the pattern of development under examination here. Now, however, it is appropriate to assess the effect of other political variables, especially the impact of party and ideology. On the policy front, it seems appropriate to examine the policy response to unemployment, which, on a priori evidence one would expect to exhibit a similar trend of development to real GDP, age structure, and social security expenditure in this period. Conclusions concerning the determinants of cross-time variance in the countries of the New Southern Europe, therefore, have to wait on the analysis provided by later sections which introduce further possible determinants of expenditure change and offer a more complex statistical analysis aimed at distinguishing the separate impacts of the whole range of factors discussed here.

On Politics and Policy Responses From the late 1970s onward, the notion of a process of socioeconomic determination of social policy outcomes was strongly challenged by the view that politics matters (see Stephens 1979; Castles 1982; Hicks and Swank 1992). In addition, more technical research emanating from the OECD Secretariat insistently pointed to the fact that the extent of social expenditures was likely to be related to the changing size of the clientele for existing schemes of provision (OECD 1985). The first

The Welfare State and Democracy

67

strand of thought argued that the differential support and incumbency of parties of Left, Right, and Center, with the emphasis varying depending on the writer, might be expected to impact on the extent of social expenditure. The second pointed out that the size of welfare clienteles changed over time. That was also, of course, central to the demographic strand of the socioeconomic hypothesis, but with the economic slowdown and cultural transformation of the 1970s and 1980s, it was apparent that new or, at least, substantially increased welfare clienteles were emerging, most particularly the unemployed and single parent families. Space restrictions prevent me from exploring the full range of hypotheses concerning the impact of parties. In the subsection on cross-national differences, I have chosen to explore whether the Right-incumbency variable which I have shown to be important in OECD comparisons excluding the countries of the New Southern Europe (see Castles 1982) remains relevant in comparisons including these countries. In the following subsection on the developmental tendency of social security in the three nations of Southern Europe, I change tack and look at the impact of Socialist government. That is partly because measuring the impact of the Right would involve problems of distinguishing statistically between the impact of Rightist rule under authoritarian and democratic auspices in what is a small sample with very limited degrees of freedom. It also follows from the more pragmatic consideration that domestic commentary in these nations is more likely to be interested in whether Socialist rule has made a difference. The notion that a positive policy response to unemployment, particularly in the later period when unemployment had risen massively in some OECD countries but not in others, might help to explain cross-national social security variation rests on the argument that relief rolls increase more or less automatically and that the state will be progressively forced to take greater steps to care for the needs of a much expanded group in dire need. It should be noted, however, that there may be countervailing effects, since unemployment also may lead to the use of welfare mechanisms, such as early retirement schemes and disability pensions, to retire workers from the active labor force (see Esping-Andersen 1990; Von Rhein-Kress 1993). As previously, I assess the impact of unemployment and of partisan control of government in relation to variation in levels of expenditure across the OECD in 1960 and 1990 and expenditure development in each of the countries of the New Southern Europe between timepoints in the 1960s and the late 1980s.

68

The Welfare State and Democracy Cross-national Variation

Results for the comparison of levels of expenditure in OECD countries are to be found in Tables 2.6a and 2.6b. Right incumbency is measured as the number of years a prime minister on the Right was in office in the ten years preceding the 1960 and 1990 expenditure comparisons (definitions of parties of the Right as in Castles 1982). Unemployment data are from OECD Historical Statistics, Table 2.15. The data are unlagged for both time points and there are three missing cases— Iceland, Luxembourg and Switzerland—for 1960. The findings concerning unemployment tell a clear story. At neither timepoint, and irrespective of the inclusion of the Southern countries, do unemployment levels make a significant contribution to crossnational social security variation. The conclusion we should draw from this is not that unemployment is unrelated to welfare expenditure. Rather, it seems probable that different countries cope with unemployment in such diverse fashions (Castles and Mumford 1992) that the nature of the policy response is not obvious in data of the kind analyzed here. The whole point of early retirement and disability schemes as remedies for a slack labor market is that they involve spending the welfare dollar to reduce unemployment rather than to compensate for it. The story concerning Right incumbency is rather different. In 1960, there is no apparent association. By 1990, however, a dramatic change seems to have occurred, with the rule of a strong and united party of the Right being moderately and negatively associated with outcomes, irrespective of sample. This confirms earlier findings (Castles 1982: 62–3), that demonstrate a strengthening inverse relationship between a variety of measures of Rightist influence and of welfare outcomes after 1960.

Table 2.6 OECD transfers and political and policy variables Variable

Correlation

T-statistics

(a) 1960 (for Rightist incumbency, 21 cases with the countries of Southern Europe, and 18 without; for unemployment, 18 and 15 cases, respectively) Right Incumbency Unemployment

.15 (0.13) .19 ( .23)

0.6 (0.5) 0.8 (0.9)

(b) 1990 (21 cases with the countries of the New Southern Europe; 18 without) .51 ( .51) 0.16 (0.20)

2.6 (2.4) 0.7 (0.8)

The Welfare State and Democracy

69

Cross-time Variation Tables 2.7a, 2.7b, and 2.7c seek to explore the impact of partisanship and unemployment growth on policy development in Greece, Portugal, and Spain. In this case, we use a measure of Left incumbency, which simply denotes whether a socialist party was in office in a given year. In the Portuguese case, the years of the left military transitional government are not counted as years of left incumbency, since otherwise the democratic transition effects of the UPSURGE variable would be conflated with the policy impact of the Soares governments. The left incumbency variable is lagged by only one year, a compromise between budget realities and the fact that these parties’ supporters expected them to effect sweeping social change as rapidly as possible (Maravall 1992). Unemployment is unlagged on the assumption that the expansion and contraction of unemployment insurance clienteles and the use of early retirement and disability schemes to cope with unemployment follows automatically on the growth and decline of unemployment. The magnitudes of the correlations relating to unemployment are strong in each case. While the plausibility of a link between rising unemployment and an increasing trend of social security development is undoubted, it is true that, as in the case of the THRESHOLD effects of democratic transition, real GDP and age structure, there was, in all these countries, a clustering of high values of unemployment toward the end of the period when social security expenditure was at its peak. With regard to partisanship, the most obvious point to note is the major difference between Greece and Spain on the one hand and Portugal on the other, with the strong (and in Greece’s case, extraordinarily strong) correlations between left incumbency in the former countries contrasting with the negligible relationship in the latter. Of course, this, too, Table 2.7 Transfers, party politics, and policy Variable (a) Greece: 1962–89 Left Incumbency Unemployment

Correlation

T-statistics

0.94 0.85

13.6 7.6

0.35 0.84

1.9 7.1

0.70 0.95

4.6 14.1

(b) Portugal: 1962–89

(c) Spain: 1967–88

70

The Welfare State and Democracy

could be a question of timing and not inherently indicative of a superior welfare performance by the Greek and Spanish left. In these latter cases, left governments were in office for much of the 1980s, whereas, in Portugal, the first Soares administration came rather earlier and well before expenditure peaked. It is precisely because all the factors I have adduced to account for social security expenditure in these countries tend to cluster with strong to very strong correlations that, in the concluding section of the chapter, I seek to elaborate multivariate models for each of the countries. Statistically, it would clearly have been more expeditious to move to such models without the prior elaboration and discussion of bivariate associations. But, in the context of what is an exploratory overview of social security development in Southern Europe, it seemed desirable to demonstrate the diverse character of the relationships between variables in each of the different countries. These, it is hoped, may serve as a check against the assertion of possibly spurious accounts of social policy development in individual country studies, since the sort of coincidence and clustering of variables noted in this and previous sections is precisely the sort of situation in which rival monocausal hypotheses can be advanced as telling the whole truth and nothing but the truth. For analysis to proceed beyond mere plausibility, it has to be demonstrated that simple explanations are insufficient if only because there are just too many simple explanations which are inherently plausible.

Families of Nations and Worlds of Welfare Before proceeding to the concluding statistical analysis, there is one further set of issues to be discussed which may contribute to our understanding of the cross-national variation in social security expenditure manifested by the OECD nations in both 1960 and 1990. In my initial descriptive account, I pointed to various groupings of nations which appeared to be characterized by rather similar welfare outcomes (see Figures 2.1a and 2.1b). Apart from an assumed Southern Mediterranean grouping, I noted commonalities which characterized the English-speaking and Scandinavian nations. In addition, I suggested that virtually all the countries toward the top of the OECD social security distribution in 1960 were core Western European nations and that, despite the growth of expenditure in Scandinavia during the period, this basically remained the case in 1990. The issues I wish to explore here are, first, whether this core grouping of welfare leaders has more in common than geographic contiguity

The Welfare State and Democracy

71

and, second, whether the countries of the Southern European grouping, so radically different in welfare performance from the European core in 1960, could be on a trajectory of development that might over the very long-term make their welfare performance rather more akin to this leadership group. One key to a possible commonality of the core nations is to be found in Esping-Andersen’s work on social policy development, which distinguishes diverse liberal, socialist, and conservative ‘worlds of welfare’ on the basis of the characteristics of the policy instruments through which they deliver social policy outcomes (Esping-Andersen 1990). In this account, the liberal, socialist, and conservative worlds of welfare correspond almost precisely with the English-speaking, Scandinavian, and core Western European nations, with the five most conservative nations being Austria, Belgium, France, Germany, and Italy, all of them in the vanguard of welfare performance in 1960. Research by Castles and Mitchell (1992), premised on the notion that ‘families of nations’ with shared historical and cultural attributes are characterized by similar policy outcomes not only in respect of social policy outcomes but also across a wide range of policy arenas (see Castles 1993), identifies four groupings rather than three, with an additional ‘radical’ family of nations comprising a subset of the English-speaking nations, including the United Kingdom, Australia, and New Zealand. However, with this modification, the research strongly confirms Esping-Andersen’s basic classification and identifies a conservative world of welfare or family of nations very nearly identical to that located in Esping-Andersen’s study. For Esping-Andersen, the defining characteristics of the conservative world are the corporatism and etatism manifest not only in these countries’ schemes of social policy provision, but also in their culture and social structure more generally. In respect of the welfare state, corporatism betokens a high degree of status segregation in the organization of social policy, operationally defined in terms of the substantial number of occupationally distinct social insurance schemes. Etatism denotes a strong bias in the welfare system in favor of state employees, which is demonstrated by the generosity of welfare schemes directed to this class of employee. Esping-Andersen notes links between these features of conservative welfare systems and the teaching of the Roman Catholic Church in the Encyclicals Rerum Novarum (1891) and Quadragesimo Anno (1931), setting out the Church’s opposition to both capitalist and socialist beliefs and advancing the principle of subsidiarity as the basis for state intervention in the field of welfare. Later work by one of Esping-Andersen’s students (van Kersbergen 1991, 1995) articulates in greater detail why we might expect to

72

The Welfare State and Democracy

encounter a distinctive Catholic syndrome of welfare provision or, rather, a world which he identifies as ‘social capitalism’. First, subsidiarity directly implies a corporatist approach, since the state is not enjoined to treat all equally, but to work through existing social groups. The issue is not one of social citizenship, but of treating existing social groups according to their social worth. Second, subsidiarity involves a reluctance to hand over power to the state, and hence a preference for what is a further highly distinctive feature of both Esping-Andersen’s and Castles and Mitchell’s conservative groupings; namely, the funding of social insurance schemes through employer and employee contributions rather than direct taxation.5 Third, a doctrine of natural inequality reinforces the notion of status, but is counterposed by the idea of the just wage, which leads to the belief that ‘benefits for adult male employees . . . ought to be characterized by a capacity to replace the family income at the level of the present status’ (van Kersbergen 1992: 21). In this last feature of Roman Catholic doctrine, it is possible to locate a potent force conducive to the high levels of social spending manifested in the core conservative nations. Whether defined in terms of specific religious doctrines or as broad characterizing features of culture and social structure, this notion that welfare performance systematically varies as between different families of nations merits empirical examination. The focus of our research here argues against too great a specificity of attribution of the determinants of difference. We want to locate which world of welfare or family of nations Southern Europe inhabits. Roman Catholicism is too specific because it excludes Greek Orthodoxy and specific features of modern Catholic social doctrine might even be argued to exclude both Spain and Portugal, given that these were the backwaters of Catholic Christendom into which such doctrines found it most difficult to make progress until recently. On the other hand, there is much in the picture of a conservative world of welfare which is redolent of the characteristics of social provision in Southern Europe. Fragmented insurance schemes catering for different social groupings on the basis of status and a strong reliance on insurance contributions are characteristic features of each of these nations’ welfare systems (see OECD 1992b), and this is just as true of Greece (see Kremalis and Yfantopoulos 1992) as of the Iberian Roman Catholic nations (see Moreno and Sarasa 1991; Mozzicafreddo 1992). Our hypothesis, then, is that the countries of the New Southern Europe may share with the core countries of Western Europe institutional arrangements of social policy conducive to a particular trajectory of social security expenditure development.

The Welfare State and Democracy

73

If, at the start of the period under examination here, the Southern European countries of the Mediterranean periphery strongly resembled the European core in all but expenditure levels, what this family of nations argument requires is the identification of a factor or factors initially retarding expenditure growth in Southern Europe and promoting it thereafter. At least three factors discussed at previous stages in this analysis immediately suggest themselves. Levels of socioeconomic modernization were substantially lower in the countries of the Southern periphery than in the rest of the OECD, the Iberian nations plus Greece differed from the other OECD nations in being ruled by authoritarian conservative regimes until the mid-1970s, and rates of unemployment were relatively low prior to the mid-1970s. On this argument, rapid rates of socioeconomic development, the transition to political democracy or the rise in unemployment after the mid1970s could, separately or in combination, have unleashed an inherent potential for welfare development of a kind consonant with the observed convergence finding of Table 2.4 above. This is the guiding hypothesis for our multivariate analysis of OECD expenditures in the next section, but before we embark on that analysis, it is first necessary to investigate how well the families-of-nations hypothesis accounts for cross-national variation in expenditure levels leaving these other factors aside. In order to do this, it is necessary that we give more operational precision to the idea of a core Western and Southern European family of nations. That is difficult. A sophisticated methodology would demand that we specify the determinants of the institutional features of the welfare state these countries have in common and relate these to expenditure outcomes. Various previous studies which have not included the countries of Southern Europe have examined factors that are clearly relevant to our theme. Both Wilensky (1981) and van Kersbergen (1991) have shown that the strength of Christian Democratic political parties is associated with welfare outcomes. Esping-Andersen (1990) has developed a measure of welfare conservatism resting on the corporatism and etatism of welfare arrangements. However, the former, clearly, does not fit because of our wish to incorporate the Greek case, and the latter does not allow us to assess the degree of conservatism of any of the Southern European cases. The solution we adopt here is necessarily quite rough and ready. We locate (by use of a dummy variable) what we shall designate as the Conservative family of nations consisting of Austria, Belgium, France, Germany, Greece, Ireland, Italy, Luxembourg, the Netherlands, Portugal, and Spain. The countries in the grouping include all the nations Esping-Andersen labels as ‘strong’ conservative (Austria, Belgium,

74

The Welfare State and Democracy

France, Germany, and Italy), two of the countries he labels as ‘medium’ (Ireland and the Netherlands) and none of the countries he labels as ‘weak’ conservative. Since Luxembourg has a welfare state akin to that of Belgium, and the Southern European countries manifest corporatist and etatist features characteristic of the conservative type, this designation seems to offer an appropriate basis of differentiation between these countries’ welfare systems and those of the Anglo-American and Scandinavian families of nations. Obviously, were this analysis to exclude Greece, this grouping could also just as easily be described as a Catholic family of nations (see Castles 1994). It includes all the countries of Western Europe with very high nominal levels of Roman Catholic adherence (see Barrett 1982), in which, presumably, the Catholic hierarchy has had the means of influencing elite and lay opinion on the proper institutions of social welfare. In addition, it also includes those countries in which Roman faith has been rather less dominant (Germany and the Netherlands), but in which Christian Democratic parties have, nevertheless, dominated politics in the postwar era. Our broad focus and preferred designation of a Conservative family of nations does not require us to show the basis of the institutional similarities between Greece and the otherwise exclusively Catholic membership of this group. Two speculative points, however, may be worth some brief mention. First, although the Conservative family of nations cannot properly be called Catholic because of the inclusion of Greece, the divide drawn is more or less exactly that between the countries most and least touched by the Protestant Reformation. In that sense, Greece is properly part of the Europe untouched by Protestantism and its corrosive effects on earlier cultural traditions such as corporatism and etatism. The Reformation was the shaping force of a political liberalism which fell on rather stony ground throughout much of Southern Europe, including Italy, until well into the twentieth century. Second, it may be, in Portugal and Spain, as well as in Greece, that the continuance of such cultural traditions in areas other than social policy made these countries ultimately quite receptive to the diffusion of welfare state ideas that resonated positively with (because informed by) similar traditions in other countries within the same family of nations. In other words, while Greece was Orthodox and not Catholic, and while Portugal and Spain were relatively untouched by Catholic social doctrine until long after the core Western European nations, they may well have had a natural proclivity to look to nations in which such doctrine was strong for appropriate design features for the new institutions which were required when, however, late by international standards, they came to tackle the issues involved in developing a modern social security system.6

The Welfare State and Democracy Table 2.8 Variable

75

OECD transfers and family of nations Correlation

T-statistics

(a) 1960 (21 cases with the countries of the New Southern Europe; 18 without) Conservative 0.36 (0.77) 1.7 (4.8) (b) 1990 (21 cases with the countries of the New Southern Europe; 18 without) 0.53 (0.64) 2.7 (3.3)

Tables 2.8a and 2.8b explore the social expenditure impact of membership of the Conservative family of nations as defined above in both 1960 and 1990. These findings, while far too crude to speak to notions of the kind we have just been discussing, are certainly broadly supportive of a family-of-nations hypothesis. Taking the figures in parentheses first, it is apparent that, both in 1960 and in 1990, the hypothesis appears to account for a substantial part of the variance of the OECD sample excluding the countries of the New Southern Europe. Moreover, for each period, the strength of these relationships is as strong or stronger than any earlier reported for democratization, socioeconomic, political, or policy variables for this sample of countries in the earlier tables analyzing cross-nation comparisons. In 1960, however, just as would be expected from their lowly position in the OECD welfare league table, the introduction of the three Southern European nations into the comparison vastly reduces the association, which is now appreciably below the level of statistical significance. However, by 1990 there can no longer be any question of a family of nations influence that was OECD-wide. In this respect, Table 2.8b seems to show that the OECD was now of one piece. If the bivariate results for 1960 seem to offer only the most tentative support for the notion that cultural differences common to groups of nations could be a major influence on welfare outcomes, by 1990 the existence of a clear social policy divide between different families of nations had apparently become a quite dominant feature of the OECD’s social security expenditure landscape. The question that we need to ask is whether this represented a dramatic change in the factors driving social security performance in the countries of the New Southern Europe or whether a trajectory of development preordained by their cultural inheritance had been unleashed by the forces of socioeconomic and political modernity.

Putting it All Together This chapter has been constructed a little like one of those Gargantuan Russian novels that constantly redirect the reader’s attention to and

76

The Welfare State and Democracy

from among different groups of characters. The reason for structuring the chapter in such a manner, constantly shifting focus between the individual countries of the New Southern Europe and the broader OECD comparison, has been an effort to locate which of a wide range of factors may have contributed to the process by which the countries of the Mediterranean periphery became gradually more aligned to the general experience of social security development among the advanced democratic nations. Russian novels finally bring their characters together to work out their fates, and that is what I do here by bringing together the various factors discussed in previous sections by means of multivariate analysis, and by offering some final thoughts as to how the development of social security in the Mediterranean countries is related to the general pattern of OECD outcomes. Some general warnings about the status of the findings is required, especially for the statistically uninitiated. The number of cases— between eighteen and twenty-four—in the equations reported here is quite low and insufficient to establish any but the strongest relationships beyond reasonable doubt. In the cross-national comparisons, potential problems of multicollinearity are actually reduced by the introduction of the Southern European nations into the sample, but in the time-series regressions, as stressed many times above, there is a strong clustering of variables. The point of the analysis here is to unpack that clustering, but the small number of cases leaves some room for uncertainty. That is particularly true because the number of cases has made it necessary to present ‘best fit’ rather than ‘fully specified’ models. In the time-series models, for instance, we have included either a real GDP or an age structure variable, but never both, since, generally, to do so would mean reducing apparently strong relationships with either variable separately to negligible ones for each together. Under these circumstances, some modesty in interpretation is appropriate, with the conclusion in the instance cited being not that the trend of real GDP or of demographic growth is the definitive cause of social security development, but rather that a case has been established for a strong impact of socioeconomic processes of transformation. In order to allow the individual country findings to contribute to the overall cross-country analysis, the time series are presented first.

Cross-time Variation Tables 2.9a, 2.9b, and 2.9c report the findings of multivariate regressions seeking to account for social security variation in Greece, Portugal, and Spain for the period starting in the latter part of the 1960s and

The Welfare State and Democracy

77

Table 2.9 Transfers expenditures Variable (a) Greece: 1967–89 (23 cases) Intercept T-1 THRESHOLD 65þ Unemployment Left Incumbency Adj. R2

Coefficient

5.19 .73 .35 .48 .48 1.35

T-statistics

5.0 1.3 3.1 4.4 1.8

.98

DW ¼ 2.23

1.21 .64 .86 1.00 .17 .57 .98

5.6 3.0 2.9 2.3 1.7 DW ¼ 2.04

2.74 .49 1.00 .58 .15 .87 .99

2.8 4.5 2.4 2.3 1.9 DW ¼ 1.89

(b) Portugal: 1966–89 (24 cases)

UPSURGE Real GDP pc (000s)

(c) Spain: 1968–90 (23 cases)

BUDGET IMPACT

culminating at or around 1990. The figures reported in the first column are unstandardized regression coefficients. The figures in the second column are again t-statistics, which may, as before, be interpreted as significant when having values over 2.0. As is commonplace in timeseries research of this kind, a term is included for the lagged dependent variable, which simultaneously reduces the risk of first-order autocorrelation and permits the inference of association between the other independent variables and change in the dependent variable. Durban Watson statistics (DW) offer a standard test for the presence of firstorder autocorrelation, with a score around 2.0 being quite satisfactory. It should be noted that all three time series are rather shorter than the full run of expenditure data reported in the majority of the earlier bivariate time-series regressions. The reason is that we only have reasonably adequate unemployment data for these countries from the late 1960s onward.

78

The Welfare State and Democracy

The model reported in Table 2.9a accounts for virtually the entire cross-time variance and shows that, apart from the marked impact of the lagged dependent variable, two factors were influential in driving Greek social security development in the period. A one percentagepoint higher level of unemployment and, quite fortuitously, a one percentage-point increase in the size of the aged population had an identical effect: namely, an increase in social security expenditure of 0.48 percent of GDP. Since the aged population increased somewhat more (4.3 percentage points) than the unemployed (1.8 percentage points) over this period, the substantive effects of the socioeconomic variable considerably outweigh those of the policy response variable. Left incumbency, on the other hand, had no significant effect, with the negative coefficient standing in some contrast to findings which have suggested that, in the course of PASOK’s first term in office, the party made and implemented a major manifesto commitment in the area of social service provision (Kalogeropoulou 1989). Perhaps, the one surprise, given our earlier analysis of the retarding impact on social security development of junta rule, is the apparent absence of a THRESHOLD effect. In terms of the model presented in Table 2.9a, a weakness of the earlier bivariate account lies in a failure to take into account the trend of unemployment in the period. In fact, unemployment followed a downward trajectory from the late 1960s to the mid-1970s, accounting for an estimated decline in expenditure over the period of around 1.5 percentage points. In other words, the model suggests a reason why Greek social security expenditure might have been static or declining irrespective of the actions and ideology of the regime. It remains possible, of course, that were we in a position to include data for the earlier part of the 1960s, prior to the Colonels’ takeover of power, some negative impact on expenditure would be more apparent. The Portuguese model reported in Table 2.9b also offers a very comprehensive account of social security development. For Portugal, there is no choice but to use real GDP as an indicator of socioeconomic development, since data on age structure is incomplete. The GDP effect is significant and of substantive importance, suggesting a contribution of around 3.5 percentage points to social security expenditure over the period as a whole. As in the Greek case, unemployment is a positive determinant of the rise in expenditure, but of lesser importance than the socioeconomic effect, accounting for only around 1 percentage point of expenditure between the beginning of the period and the highpoint of Portuguese unemployment in 1986. In the Portuguese model, there is a clear UPSURGE transition effect, with the left transitional military regime markedly increasing expenditure in the revolutionary two

The Welfare State and Democracy

79

years prior to the full establishment of democracy. There is, however, absolutely no support for a left partisan effect, with the administrations under Soares contributing virtually nothing to social security expenditure development. As I have suggested earlier, it is an open question whether the massive impetus to change under the military transitional government reflected the rise of democratic forces or the impact of leftist ideology. Certainly, it left only minimal room for further social security expansion in the following years, irrespective of which party may have been in office. An explanation of Spanish social security development no less comprehensive than in the other countries of Southern Europe is similarly driven by socioeconomic and policy variables. A one percentage-point increase in the aged population contributed more than 0.5 of a percentage point to social security growth and a one percentage-point increase in unemployment contributed 0.15 of a percentage point to spending. Given that Spain’s aged population grew by 5 percent and unemployment by 13 percent, age has a somewhat higher substantive impact (2.9 points against 1.95). As in all the other countries, the regression provides no evidence that Left incumbency had any positive effect on spending. In fact, the result is almost significant but negative, offering a degree of support for Esping-Andersen’s counterintuitive finding that ‘counter to traditional theory, socialist cabinets (in Spain) depress welfare share’ (Esping-Andersen 1993). In the Spanish case, the government that presided over the really significant expansion of the welfare state was the Sua´rez administration, although, here, we interpret that effect not as a partisan centrist one, but rather as evidence for the existence of a very substantial democratic transition impact mediated through the state budget. The absence of a significant positive effect for the THRESHOLD variable in any of these cross-time equations is one of the most interesting findings of the cross-time analysis. It suggests that the convergence of the Southern European countries with the OECD norm cannot have been a function of the removal of the dictatorship as such, but rather, insofar as democratization affected the process at all, it must be interpreted as an unleashing forces pent-up during the period of the regime’s incumbency. Again the idea of a structural shift seems appropriate. In a given country, the extent of democratization does not seem to determine outcomes in a linear fashion. Rather, once a country has made such a shift, other factors come into play through the mediating influence of popular demands. In terms of this analysis, the possible candidates for convergence inducing factors unleashed by the shift to democracy are socioeconomic transformation and unemployment.

80

The Welfare State and Democracy

Arguably, the most crucial feature of these findings is their consistency across countries. In all three of the nations of the New Southern Europe, social security expenditure change was driven by a combination of socioeconomic development and unemployment. In all three countries, the substantive impact of socioeconomic variables somewhat outweighed the policy response to unemployment. Partisan rivalry appears to have had little effect in any of the countries and there is nothing in these findings to suggest a coherent and potent party political cleavage determining policy change in the decade and a half following the fall of the Southern European authoritarian regimes. Perhaps, over such a period and concerning such a topic as social security, it is a little unrealistic to expect that this would be the case. As Maravall (1992) has pointed out, the Southern European Socialists in the 1980s were, to a still greater degree than their Northern colleagues, between a rock and a hard place in seeking to combine radical reform with the economic disciplines imposed by a stalled international economy and high domestic levels of unemployment. Only in respect of the democratic transition process were there any appreciable differences between the countries. In the Iberian Peninsula, transition effects of somewhat different kinds led to temporary advances in welfare expansion fuelled by newfound opportunities to express popular demands through the political and budget processes. That the Portuguese transition effect was expressed in an immediate upsurge in expenditures, while the Spanish effect percolated more gradually through the budgetary process, clearly reflects the political character of the processes by which the respective nations shrugged off the authoritarian legacy of the past: in Portugal, through popular revolution and, in Spain, through a more gradual evolution to liberal democratic legitimacy. In Greece, no transition effect of any kind is discernible, presumably because any impact of the relatively shortlived period of authoritarian rule on expenditure development was insufficient to depress Greek performance markedly below that appropriate for her level of socioeconomic development.

Cross-national Variation I turn now to conclusions concerning the factors which impinge on cross-national variation and to reflections on what they have to tell us about the trajectory of Southern European social security development in comparative perspective. Tables 2.10a and 2.10b report the findings of multivariate regressions seeking to account for social security variation in the OECD in 1960 and 1990. As previously, I report in

The Welfare State and Democracy Table 2.10

81

OECD transfers, 1960 and 1990

Variable

Coefficient

T-statistics

(a) 1960 (21 cases with the countries of the New Southern Europe; 18 without) Intercept Democratization Real GDP pc (000s) 65þ Right Incumbency Conservative Adj. R2

6.26 ( 5.45) .17 (.16) .43 (.40) .53 (.48) .06 (.07) 3.12 (3.27) .77

5.0 2.0 2.2 0.7 3.4 .71

(3.7) (1.6) (1.7) (0.8) (3.1)

(b) 1990 (21 cases with the countries of the New Southern Europe; 18 without) 12.52 ( 11.68) .05 (.04) 0.5 (0.4) .56 (.51) 2.2 (1.5) 1.31 (1.35) 3.5 (2.7) .46 ( .48) 2.2 (1.9) 6.67 (6.52) 3.8 (3.0) .67 .64

parentheses findings for a smaller sample of OECD nations excluding the nations of Southern Europe. Since they have no discernible effect in either period, and take up valuable degrees of freedom in what are already very small samples, I include neither program experience nor unemployment in these models of OECD social security outcomes. In the case of unemployment that immediately seems to suggest a prima facie conflict with the single country time-series results, in all of which unemployment proved to be a significant positive influence on the expansion of social security. In fact, as noted earlier, there is no necessary correspondence between time series and cross-sectional effects. That rising social expenditure is likely to flow from increasing unemployment in most of the countries in which it occurs (including all three countries of the New Southern Europe) does not preclude some countries from using the mechanisms of the social security welfare state to avoid lower labor force participation in the first place, with the consequence that no general crosssectional relationship can be observed. The four significant results in Table 2.10a relate to the positive impact of the two socioeconomic variables—real GDP, the age structure of the population—in the larger sample and of membership of the Conservative family of nations and democratization irrespective of sample. That socioeconomic factors only appear relevant when the

82

The Welfare State and Democracy

New Southern Europe is included in the analysis goes some way to confirming that, at this time, and, at least in respect of factors related to economic development, this area marched in tune to a different drummer from the rest of the OECD (cf. Castles 1982). On the other hand, the consistent finding of a strong positive link between membership of the Conservative family of nations and high welfare expenditure underscores that the divide between the South and the rest was not absolute, and that there were affinities even with those nations that headed the OECD social security league-table. Turning now to political factors, it is apparent that the substantive impact of democratization is very considerable in both samples, and, in the larger of them, it makes a difference of somewhat more than 6 percentage points of GDP between the least and the most democratic nations. On this basis, we may conclude that, in the earlier period, the single most important factor accounting for variation in social security effort in the OECD lay in the very substantial differences in the degree to which different regimes permitted the political articulation of popular demands. Rather surprisingly, the findings for the smaller sample confirm the result from the bivariate analysis of Table 2.2a that the extent of democratization was also a factor distinguishing between the welfare performance of the more advanced nations. On the other hand, there is no sign that, in this period, party politics was of any great importance and the degree of association between Right incumbency and outcomes is negligible in both samples. In the early 1960s, it would certainly appear that politics mattered, but what made the real difference was not so much the partisan complexion of government as the extent of popular participation. Differences in the extent of democratization constitute key variable in understanding the differences between the welfare performance of the Mediterranean countries in 1960. Harking back to my introductory remarks concerning the reasons for excluding Italy from the New Southern Europe, at least in the context of the development of social security policy, it will be apparent that the great expenditure divide between Italy and the Iberian nations at this time was almost entirely a function of this difference. Many questions are appropriate about the true character of Italian democratic practice both in the 1950s and subsequently, but it can hardly be denied that electoral turnout and votes for nongovernmental parties were both high, allowing, according to the measure of democratization used here, a strong articulation of political demands for welfare provision. Remembering, too, that the time point of this cross-section is 1960, before the Colonels came to power in Greece, it is also possible to attribute Greece’s greater welfare development than the remainder of the New Southern Europe to the fact that its degree of democratization,

The Welfare State and Democracy

83

although not as high as that of much of Western Europe, was far higher than in Portugal and Spain. At the beginning of this period, Italy had made the structural shift required for the development of a modern welfare state. In contrast, the countries of the New Southern Europe had either not yet made that transition or, as in the Greek case, its permanence remained in some doubt. The role of democratization is far from being the only point of interest in this analysis. Whereas earlier, in Table 2.3a, the evidence for a positive bivariate relationship between real GDP per capita and outcomes is negligible and even negative, in the larger sample here it is both positive and significant. Moreover, the same phenomenon of a stronger association in the multivariate than in the bivariate analysis is also apparent in respect of the Conservative variable. It is this finding which constitutes the decisive evidence for an interpretation based on worlds of welfare or families of nations, for, where previously, in the larger sample (see Table 2.8a), there was only a tenuous and indicative association between the Conservative variable and outcomes, in the multivariate analysis, it is not only strong, but virtually as strong as for the smaller sample. In effect, the commonality of this family of nations in 1960 is masked by their great divergence in socioeconomic levels and diverse degrees of democratization. Certainly, on this reading, the nations of Southern Europe cannot be considered inherently different from the core countries of Western Europe in their potential for welfare development at the beginning of the period under review, but rather may be seen as having been prevented from achieving that potential by a syndrome of political and economic underdevelopment inherited from the past. Even though economic development features in this explanation, there is a real irony in this interpretation for the protagonists of the socioeconomic convergence hypothesis, for the only reason that the impact of real GDP was not clearly apparent in the earlier bivariate analysis was that all of these countries remained the captives of historical and cultural forces shaping their social security systems. Apart from democratization, itself a political rather than a socioeconomic variable, what mattered most for social security outcomes in 1960 was membership of distinctive families of nations in the context of which had been elaborated quite diverse worlds of welfare capitalism. Indeed, the very fact that program experience, which is a variable that captures some of this historical and cultural difference, drops out in face of a families-of-nations explanation is simply because these variables are locating something of the same reality. If, in 1960, the social security expenditure levels of the Southern European welfare states reflected the legacy of the past, that is no less true of the advanced nations as a whole.

84

The Welfare State and Democracy

The story which emerges from Table 2.10b has some important similarities to that of thirty years earlier. Real GDP and age structure both still make a difference in the larger sample and the difference between Conservative and non-Conservative nations has become still greater, being around 6.5 percentage points of social security spending in both samples. There are also three very important differences. Democratization no longer has any discernible effect in either sample, Right incumbency is now significantly negatively related to expenditure in the larger sample and almost so in the smaller one, and age structure now emerges as significant in the smaller as well as the larger sample. The substantive impact of age structure is very considerable indeed, with the difference between the two countries (Iceland and Sweden) at the extreme ends of the age structure distribution making an estimated difference of around 9 percentage points of social security expenditure between them. The overall conclusion that emerges is one of modern social security systems which continue to reflect the cultural and institutional imperatives of their historical design, while simultaneously finding means to respond to the demographics of an aging population. Diverse levels of democratization, it would seem, are no longer important because the OECD as a whole had made a crucial structural shift to greater political responsiveness to popular demands for welfare. By 1990, all these nations combined a degree of democratic participation and party competitiveness sufficient to allow the expression of popular demands for welfare and so remaining differences in the degree of democratization no longer mattered. On the political side, what now did make a difference, at least to a limited extent, were ideological divisions expressed through partisan control of government. Still more important were socioeconomic differences and, most particularly, differences relating to the age structure of the population, itself a most potent source of political demands for welfare. The modern state is unable to shrug off its responsibility for this category of social dependence and cannot easily disguise the character of its intervention in the way that varied responses to labor market downturn disguise the state’s responsibility for unemployment. But, while demographics matter, thirty years of social security development demonstrate no tendency for socioeconomic forces in general to have superseded cultural factors. On the contrary, in 1990, families of nations remain, as much or more than they were in 1960, the vital parameters of cross-national social security variation and the key to understanding the pattern of OECD outcomes. Moreover, in this story, the Mediterranean periphery no longer stands out as all that different from the general OECD experience. Dropping real GDP from the large sample 1990 equation no longer radically reduces the estimated impact of membership of the Conser-

The Welfare State and Democracy

85

vative family of nations, even though it does diminish it somewhat. By 1990, the Southern European nations, although still substantially less developed in socioeconomic terms than the majority of OECD nations, had made very substantial strides in expanding their levels of social spending. To the extent that economic growth and demographic changes continue to transform their societies, they seem to have every prospect of continuing on that path. Indeed, and this is the only possible conclusion warranted by the families of nations hypothesis so clearly vindicated in this analysis, these nations have every prospect of eventually moving beyond the OECD norm, which, in simple average expenditure terms, all but Portugal have already attained. In fact, it is only when seen as members of the Conservative family of nations that expenditure levels in these countries can any longer be seen as particularly low. Without any judgment being entered into as to whether it means they are endowed with better or more caring welfare states, the fact is that members of the Conservative family of nations spend far more on social security than nations in other families of nations. In this perspective, the countries of Southern Europe must be seen as quite typical members of the Conservative family of nations. They spend somewhat less than others in this grouping because, until relatively recently, they were poor and had relatively youthful populations, because their decades under authoritarian rule had repressed popular demands for these social policies, and because the development of massive nationwide programs of this kind implies a significant time lag before they can catch up with their European counterparts. Readers are again reminded, as they were at the beginning of this chapter, that social security is not the be-all and end-all of the welfare state and that aggregate expenditure patterns are not the only way to measure social policy outputs. Nevertheless, the conclusion is an important one and means that these nations had, by the early 1990s, overcome their socioeconomic and political inheritance to become a part of the European social policy mainstream.

Notes 1. As far as I am aware, the only studies including Southern European countries in routine policy outcomes analysis are my own 1998 volume Comparative Public Policy: Patterns of Post-war Transformation and Lijphart’s (1999) Patterns of Democracy: Government Forms and Performance in 36 Countries. 2. I owe my formulation here to an identical point made by Glyn and Rowthorn (1988) concerning cross-time and cross-national variation in unemployment.

86

The Welfare State and Democracy

3. In the United States, the virtual exclusion of black voters in the South until the 1960s and generally very low levels of voter registration until the present, and in Switzerland, the failure to include women in the franchise until 1970, meant levels of democratization well under half those typical of the core states of Western Europe. 4. The strength of the finding for Greek real GDP contrasts with the finding of Kremalis and Yfantopoulos (1992: 85) of a linear bivariate relationship between GDP and social expenditure for the period 1962–84 of 0.99. The discrepancy in the strength of the findings appears to be a consequence of that study’s measurement of social expenditure in monetary terms rather than as a share of GDP. 5. Although I do not explore the notion further in this chapter, there are very strong reasons for believing that this aspect of the tax system is a key mechanism for translating Catholic social policy ideas into policy outcomes. Social security contributions are inherently productive of higher levels of expenditure, since (a) they confer rights to benefit and (b) are less conducive to taxpayer backlash than other forms of financing. The association between the measure of membership of the Conservative family of nations employed below and the social security contribution share of total taxation is quite strong—more than 0.76 in the mid-1960s (the earliest date for which tax data are available) and 0.69 in 1990. Moreover, cross-national research on types of tax systems (Peters 1991: 60–4) has identified four distinctive systems, of which two are exclusively composed of countries which may be classified as part of the Conservative family of nations. It is also worth noting that the relationship between the social security contribution share of total taxation and social security expenditure outcomes has been confirmed in a pooled time-series analysis of the standard 18-nation OECD sample (Castles 1990). For further analysis of this topic, see Castles (1998). 6. Clearly, too, for the Iberian nations, such patterns of policy diffusion would have been much facilitated by the fact of the linguistic affinity of the romance languages and, for the New Southern Europe generally, the use of French and Italian as second languages throughout the Mediterranean area.

................ 3 ................

Social Policy, Democracy, and Citizenship in Southern Europe Marisol Garcı´a and Neovi Karakatsanis1

In his classic 1950 work, Citizenship and Social Class and Other Essays, T. H. Marshall set forth the compelling argument that all citizens are of equal worth, and thus universally entitled to a minimum standard of living. Linking civil citizenship with social welfare, Marshall maintained that the universal provision of education, health, social security, and welfare benefits leads to the development of full citizenship—first, social citizenship and, through it, political citizenship. As civil rights protect the individual from arbitrary or discriminatory treatment, and as political rights assure that power is not confined to the hands of a few, social rights offset the vagaries of the free market and correct its inherent inequalities. When states respond to the basic needs of disadvantaged citizens by providing them with social welfare, and when they do so universally without stigmatizing those in need, they enable the less fortunate to acquire the minimum competence required so that they too may participate—fully and with dignity—in their political communities as equals. Esping-Andersen (1990) has maintained that, in this conception, the welfare state does not only intervene in an attempt to correct social inequalities, but becomes an active force in structuring social relations. As labor is ‘decommodified’ through redistributive social policies, becoming more autonomous from employers and market forces, it becomes more politically active and the balance of class power is fundamentally altered. Accordingly, politics—indeed democracy—and social citizenship are critically linked. Utilizing the concept of social citizenship as an ideal-type construct—an endpoint on a continuum of social welfare policies— this chapter examines the development of social welfare provision in

88

Social Policy, Democracy, and Citizenship

postauthoritarian Southern Europe. By examining the cases of Spain, Portugal, Greece, and Italy, we attempt to discern to what extent politics in these four countries affected the development of their social welfare states. As would be expected, we argue that politics have clearly had a decisive impact on the development of social welfare in Southern Europe. As we illustrate, and as Frank Castles has demonstrated in his contribution to this volume, democratic governments in Southern Europe exhibited a markedly increased commitment to the expansion of crucial welfare programs when compared to their authoritarian predecessors. Aggregate levels of spending on social security in Southern Europe have come to approximate West European levels and, in some cases, even to exceed them. Nevertheless, we emphasize that social citizenship has not been achieved. Despite the overall increase in aggregate-level expenditures and other important welfare reforms legislated by Southern European democratic governments, the social welfare regimes of the South do not fit with the more generous, universalistic models of welfare provision present in much of the western industrialized world. In a word, the ideal-type of social citizenship continues to be absent. This chapter surveys the many similarities of the Italian, Portuguese, Spanish, and Greek social welfare states despite a number of important differences: Spain’s much higher rate of unemployment; Portugal’s high female participation rates, its universal family allowances, and its benefit for young unemployed persons living alone; Italy’s longer democratic experience, as well as its and Greece’s very generous old-age pensions (Ferrera 1996: 17–37; Flaquer 2000: 26) are but a few examples. These differences notwithstanding, this chapter focuses primarily on the many similarities among the Southern European social welfare states. Rather than being characterized by the principles of universalism, Southern European welfare regimes remain fragmented, corporatist, and familistic. Citizens enjoy widely varying degrees of entitlements in pension, unemployment, and other benefits depending on their position in the labor market, their gender, age, and place of residence. Furthermore, labor market rigidities have propelled the emergence of a dualistic labor force in the south in which certain individuals—those enjoying full-time careers in the formal labor market—enjoy extremely privileged positions and extensive social welfare protection, while others—the unemployed and those working in the informal economy and in very short-term contracts—are unprotected and lack a social wage altogether. This fragmentation in welfare provision is reinforced by gender and age. Southern European women and youth find it particularly difficult to accumulate social rights because of their unfavorable positions in the labor force. Many women and young

Social Policy, Democracy, and Citizenship

89

people continue to work in the informal economy or as unpaid labor in family-owned businesses and farms, thus having no formal access to the Southern European welfare states. In the absence of universal coverage, women, in particular (and, increasingly, migrant women), continue to play an essential caring role in social welfare provision— a role unparalleled by their counterparts elsewhere in Europe (Karakatsanis and Swarts 2003, 2004). Moreover, traditional patron–client relations, with roots in agricultural society, have been transplanted into the urban centers and continue to enjoy a high level of popular acceptance even today, particularly in Greece and Southern Italy. Such clientelistic practices have been incorporated—institutionally and functionally—into the modern social welfare state, further limiting (in some parts of Southern Europe) its political accountability and universalism. In short, based on such peculiar characteristics—entitlement fragmentation, labor market segmentation, clientelist practices and familistic tendencies—the Southern European welfare state is unquestionably distinct and peculiar. We maintain that to truly understand the peculiarities of the Southern European social welfare regime, one must identify the factors underlying its distinctiveness. We stress the inheritances of the past. To reiterate one of the arguments of the introductory chapter of this volume, social-policy reform has been constrained by existing institutional arrangements, which have made change incremental and pathdependent. Thus, a central argument of this chapter is that while posttransition politics and ideology have greatly expanded the scope of the Southern European welfare state, historical, political, and cultural traditions have limited the extent of change in social welfare provision. We argue that the democratic governments of Italy, Spain, Portugal, and Greece have to varying degrees inherited from their predecessor governments an uneven, employment-driven corporatist model of welfare that continues to produce a highly uneven pattern of social insurance coverage. Rigidities in the labor force, also derived from the corporatist past, persist and continue to reinforce a number of ‘traditional’ cultural norms and practices, further fragmenting social welfare coverage. Thus, a number of ‘traditional’ cultural and political characteristics endure even today, giving the Southern European welfare state a ‘flavor’ of its own and making it quite distinct from its Northern European counterparts. In this regard, our study considers the relationship between the state, the labor market, and the family, considered the three cornerstones of the welfare state (Esping-Andersen 1999). We argue that similar public policy patterns in Southern Europe do not result simply from cultural affinity nor from a common socioeconomic and historical experience but from a dynamic mixing of

90

Social Policy, Democracy, and Citizenship

both. In short, societal values are quite often stable and distinctive, and therefore capable of shaping social structure but, on the other hand, they also undergo constant transformation and change. Thus, it is important to note that we do not see culture as static or as a permanent and primordial attribute of society. Indeed, we argue that while ‘traditional’ cultural norms continue to persist in the south, allowing the family to play a crucial function as social welfare provider, society is also experiencing profound modernization and change, altering cultural norms and traditional expectations. Indeed, the question is not whether the Southern European family will cease to be a social welfare provider, relinquishing that function to the state or some other entity, but when and how it will do so. It is this continually shifting interaction of historical, cultural, and political forces that we seek to describe in this chapter.

Worlds of Welfare The notion that distinct worlds of welfare exist, each exhibiting unique characteristics and approximating to a greater or lesser degree the ideal-typical social citizenship construct, is not new. In 1990, EspingAndersen identified three welfare regimes which have historically developed in the western world, each exhibiting a different degree of labor decommodification. According to Esping-Andersen, the ‘social democratic’ regime of welfare, also known as the Scandinavian model, emerged in countries where citizens were historically more accustomed to political participation due to a tradition of direct democracy at both the local and national levels. In these societies, the universalization of social services was effectively achieved through class conflict as workers’ interests became politically incorporated into national parliaments. Thus, state institutions played a central role in the provision of social assistance programs from the beginning, and relations between national and local institutions and the individual beneficiary were clearly defined and uniform. This social democratic regime of welfare has become the model of social citizenship—a system which permits citizens the greatest degree of independence from market forces in constructing individual life chances through universal welfare services and a relatively high degree of labor decommodification.2 Continental Europe, including Germany, France, Austria, Belgium, and Italy (with some variation), is characterized by a second, more conservative regime of welfare, one strongly influenced by Catholicism. Esping-Andersen maintained that the emergence of this ‘corporatist’ welfare regime resulted either from the fact that the governments of

Social Policy, Democracy, and Citizenship

91

Continental Europe were historically less interested in social redistribution, or from a political balance of forces in these societies that was traditionally more inclined toward privileged groups. Distinguished by contributory systems in which social entitlements are dependent on employment, this regime is characterized by a considerable degree of differentiation in welfare programs and benefits, depending on workers’ occupational status and previous earnings. Those outside the labor force, particularly women who remain in the home as mothers and general family caregivers, are excluded from social security entitlements. Thus, in addition to the greater reliance on family support for welfare assistance, this welfare regime tends to have a greater degree of institutional decentralization as well as an increased involvement of the private social sector in social assistance programs. Esping-Andersen’s third model of welfare—the ‘liberal’ regime—is characteristic of Anglo-Saxon societies (the United States, Britain, Canada, Australia, and New Zealand). A main feature of this regime is that the state takes relatively little responsibility for the welfare of its citizens, allowing welfare provision to be bought and sold privately in the free market. The state’s formal role is quite limited, providing residual coverage to those who lack the basic means to purchase their own in the private market. This has a number of negative repercussions: the level of labor decommodification is minimal; those who receive welfare support from the state are often stigmatized as nondeservers; most taxpayers see little justification for their contributions to the welfare state since they do not accrue any direct social welfare transfers; and class differences tend to be reinforced as the gap between rich and poor is widened. Based on a number of other variables, alternative typologies also emerged in the sociology and political science literature. For example, B. Guy Peters (1991) constructed a typology based on the percentage of total tax revenue for all levels of government derived from eleven types of taxes. Four tax clusters emerged from his analysis: (1) an AngloAmerican group, coinciding with Esping-Andersen’s liberal regime; (2) a broad-based taxation model, overlapping with the continental model, but including Spain; (3) a cluster incorporating the Scandinavian countries; and (4) a ‘Latin’ cluster comprised of Italy, Greece, Ireland, and Portugal. According to Peters, both the Latin and broad-based systems are characterized by strong corporatist traditions. However, the Latin model is distinct as it is more heavily based on indirect taxation, exhibits significant difficulty in tax collection, and is characterized by a higher degree of centralization. Other scholars have also identified the existence of a distinct Southern European model of welfare. Some contend that Italy, Greece,

92

Social Policy, Democracy, and Citizenship

Portugal, and Spain are characterized by a rudimentary type of welfare state. They argue that while the constitutions of these countries formally commit them to building ‘modern welfare states’, and while certain social security programs serve as basic income measures, comprehensive implementation is nonetheless lacking. They emphasize the distinctiveness of the Southern European family as a provider of social welfare and argue that rudimentary levels of provision set Southern Europe apart from other types of welfare regime.3 Still others have made a cultural argument, claiming that Southern European culture— a culture highly influenced by the social doctrine of the Catholic Church in Italy, Spain and Portugal—has contributed to the emergence of a distinct brand of welfare heavily reliant on notions of solidarity (Castles 1994: 19–39; van Kersbergen 1995). This chapter is clearly in the latter camp, identifying Southern Europe as a region with a model of welfare clearly distinct from those described by Esping-Andersen. Following Ferrera (1996), who first showed the distinctiveness of Southern Europe in terms of social policy, we argue that Southern European welfare states do not constitute a residual case in the typology of welfare regimes. With a number of important distinguishing characteristics, discussed below, we consider there to be an identifiable, distinctive Southern European type of social welfare, one conditioned by a unique set of historical, political, and cultural circumstances.

The Southern European Model: A Distinct Welfare Type Having argued for the existence of a distinct Southern European type of welfare provision, we now turn to a discussion of the chief distinguishing characteristics of the Southern European model. Specifically, we identify four key characteristics which shape social welfare in Southern Europe: severe labor market rigidities and segmentation; an extraordinary level of fragmentation in welfare provision structured largely along occupational lines and employment status; the persistence of patron-clientelism not only in politics, but in the structure of social services; and the prevalence of familism, or family welfare provision as a way of compensating for the widespread inadequacies of the Southern European welfare systems. While differences exist within Southern Europe, we regard these differences as variations within the same basic model. As Flaquer (2000) argues, strong family and kinship patterns, a weak welfare state, and a highly segregated labor market reinforce each other, shaping the Southern European model into one where the ‘welfare family is the welfare state’ (Bahle and Pfenning 2000: 4).

Social Policy, Democracy, and Citizenship

93

Labor Market Rigidities and Segmentation The Southern European model has been characterized by significant rigidities in labor markets coupled with a high degree of segmentation in welfare provision, privileging employed workers in the formal economy and marginalizing and neglecting the unemployed or those in informal and part-time, contract work. As we will illustrate later in this chapter, the origins of these pathologies lie in previous experiences with authoritarian rule as well as in earlier periods, including in some respects the nineteenth century. In sum, all four Southern European countries share a tradition of limited democracy, when workers were appeased, and their loyalties bought through a system of rigid labor laws which protected and benefited workers, primarily male heads of household, but in exchange, denied them the right to strike. A rigid labor market had the further, and long-lasting, consequence of allowing governments to maintain artificially high levels of male employment, and thereby avoid high unemployment insurance costs. However, while democratization made the strike legal, the rigidities inherited from these corporatist labor regimes were not reformed during these countries’ transitions to democracy. Instead, when democratization was finally achieved in Greece, Italy, Portugal and Spain, a mass of labor market rigidities was bequeathed virtually intact from preceding governments to the democratic era. These institutionalized rigidities restricted dismissals by private and public employers, prevented unilateral reductions in working hours by management, and awarded generous severance payments to dismissed workers. However, in the effort to facilitate peaceful transitions to democracy and to assure democratic consolidation by maintaining the support of workers, all four democratic governments left virtually untouched this traditional labor relations structure. Even after the labor market reforms of the 1980s and 1990s, Southern European labor markets continue to work to the benefit of privileged ‘core’ workers who enjoy full-time employment, long-term contracts, high wages, generous social security, and other fringe benefits and virtually complete protection from dismissal. A few examples of the generous protection enjoyed by the employed will suffice. In Greece,4 terminations cannot become effective until notice is given to the labor ministry, whose authorization is difficult to acquire and routinely denied. Salaried employees may also receive as much as twenty-four months of severance pay plus a proportion of annual legal bonuses if dismissed (Price Waterhouse 1992a: 76–7). In Spain, too, dismissed workers receive generous severance payments— until 1997, 45 days of salary for each year of service, up to a maximum

94

Social Policy, Democracy, and Citizenship

of 42 months wages (Estatuto de los Trabajadores 2004). Equally restrictive termination policies exist in Portugal where, although the concept of ‘layoffs’ was introduced in 1983, it was not effectively put to use. Instead, employment can only be terminated by mutual consent, when a contract of employment expires, when there is ‘just cause,’ or when the employee voluntarily initiates termination. Collective dismissals require the approval of the Portuguese Ministry of Labor, while ‘just cause’ dismissals require consultation with the workers’ council or trade union representatives and labor tribunals—bodies heavily weighted in favor of workers (Price Waterhouse 1992b: 85–6). By maintaining this rigid protection for privileged workers—usually male heads of household—democratic governments in all four countries failed to address the thorny problems associated with such rigidities which, on the one hand, were quite popular with workers but which, on the other, served to make firms less efficient and competitive as hiring and firing decisions were largely outside their control. As a result of the exorbitant costs associated with such generous severance payments, unprofitable firms have found it difficult to terminate employees at times of economic downturn, thereby discouraging job creation that would particularly benefit ‘outsiders’ at times of economic growth. Spain is a case in point. For many years, Spanish firms declared bankruptcy rather than pay the huge indemnities required by law for dismissals. In so doing, the number of unemployed workers increased by an even greater margin. Moreover, extremely costly severance and retirement contributions made by employers have made them highly reluctant to create new jobs and to formally employ ‘outsiders,’ particularly youth and women, except on a part-time, contractual basis (which does not require extensive social protection) (see Esping-Andersen 1995: 18–19). As a result of introducing labor laws aiming at promoting a more flexible labor market, Spain in the 1990s registered an alarming proportion of temporary employment, mainly taken by the youth and women. Despite more recent reforms that have begun to partially dismantle some of the rigidities in Southern European labor markets,5 the obvious fact that trade unions are composed of and dominated by employed—and therefore privileged— workers has tended to exacerbate the existing rigidities and to reinforce the ‘insider–outsider’ distinction. In fact, when labor-market liberalization has occurred in Southern Europe, it has tended to take place in a somewhat peculiar fashion. In Spain, increased flexibility diminished the job security of the newly employed (through the use of fixed-term contracts) without diminishing the job security of those with permanent full-time contract jobs (Jimeno and Toharia 1994: 109). Since workers, who have an interest in maintaining the rigidities of

Social Policy, Democracy, and Citizenship

95

the labor market are also the ones who wield the most political power within the unions, trade unions have been extremely reluctant to accept any government policies that would hurt the permanent workforce (Estivill and de la Hoz 1990; Rhodes 1997; Richards and Garcı´a Polavieja 1997).6 Moreover, since outsiders are denied trade union membership, they are effectively denied the ability to voice their demands through this critical political resource. As Andrew Richards and Javier Garcı´a Polavieja (1997) argue, Spanish unions have steadily lost contact with, and helped promote the alienation of, a growing proportion of the . . . workforce. The unemployed, the precariously employed, women and the youth are gradually being left out of the predominantly male, blue-collar culture of a trade union movement increasingly confined to the core permanent workforce (also see Rhodes 1997: 115–16; and Estivill and de la Hoz 1990: 297–8). In short, the Southern European labor force is segmented and dualistic, pitting this permanent group of privileged insiders who dominate jobs and labor unions against an equally permanent group of excluded, voiceless outsiders. These same labor markets and the welfare states based on them systematically work to the detriment of ‘non-core’ workers, the unemployed, women and youth. One particular characteristic of Southern European labor markets is the low rate of female participation, coupled with high levels of female unemployment (see Table 3.1). In Spain and Greece, the gap between male and female unemployment rates has actually grown in the 1990s, with male heads of household being the least likely to be unemployed, especially over the long-term (Flaquer 2000: 22; also see Cousins 1999). Portugal is exceptional in this regard, with a female participation rate actually higher than the EU average. This is due to the structure of the Portuguese labor market, where labor costs are lower. Women enter the formal labor market where they often receive very low wages because additional money to the household income is very important in order to meet consumption needs and because formal employment assures them some unemployment benefits in case of dismissal (Moller and Hespanha 2002: 137–56). Moreover, extended family networks provide caring support for children and they allow young mothers to work outside the household. In contrast, in the other three countries, selfemployment and unpaid employment in small family businesses play a much larger role and help ‘hide’ female employment (Carlos and Maratou-Alipranti 2000: 41–6). This relegates many women (and youth) to unpaid, unprotected, and ‘uninsured’ positions in the workforce. Reconciling family and work in Southern Europe is also made difficult by a second labor market characteristic: the paucity of part-time work. Only 8.4, 11.6, 12.7, and 16.6 percent of women in Greece, Portugal,

96

Social Policy, Democracy, and Citizenship

Table 3.1 Unemployment rates in Southern Europe, 1999

Italy Spain Greece Portugal

All males

All females

Male youth*

Female youth*

8.9 11.2 6.6 4.0

15.7 22.9 15.9 6.2

28.6 21.7 21.4 7.5

38.3 37.3 39.3 11.1

* Under age 25. Source: OECD (2001b).

Italy, and Spain, respectively, engage in part-time work, compared to the EU average of thirty-one percent (Eurostat data, quoted by Flaquer 2000: 19). Thus, a particularly adverse consequence of severe labor market rigidities is the creation and maintenance of underprivileged workers in the informal economy. With a few new permanent jobs being created and with a little hope of attaining such protected positions, many in Southern Europe have turned to employment in the burgeoning ‘black’ economy. The informal sector is estimated to be about twenty percent of GDP in Italy, thirty percent of GDP in Greece, and about twenty to twenty-five percent of GDP in Spain and Portugal.7 However, since welfare in Southern Europe is largely occupation-based, such workers are denied access to the wide range of social rights accruing to protected workers. Remaining excluded from formal employment with no right to a social wage,8 their employment trajectories tend to be highly unstable, as firms continually terminate their employment in a vicious cycle of hiring, firing, and rehiring—all to avoid turning temporary positions into permanent ones. In sum, within the dualistic labor force that has emerged in Southern Europe, outsiders have little hope of entering the highly protected world of formal employment. Thus, contrary to the universalism inherent in the notion of social citizenship, access to welfare provision in Southern Europe is fundamentally determined by one’s employment status—whether one enjoys a highly protected formal position, works at an unprotected informal job, or is unemployed with virtually no protection. Coupled with extreme labor market rigidities, these systems produce and perpetuate extreme dualism in worker rights and social welfare access.9 While a minority of highly protected workers enjoy statutory job protection and social insurance, ‘the rest’ of society is relegated to informal or shortterm employment with little or no access to welfare benefits, including unemployment compensation (with the partial exception of Portugal). Finally, it is also important to note that the institutionalization of

Social Policy, Democracy, and Citizenship

97

highly rigid labor markets as well as government policies that help keep unviable firms in operation (largely to maintain the employment status of male heads of household!) at extremely high budgetary costs have allowed Southern European governments to spend less on active redistributive policies. Traditionally, the male head of household’s generous benefits and permanent employment status meant that the state could be less involved in explicit social welfare provision, leaving the family—and its women, in particular—as the main social welfare provider. It is important to note that cultural norms have reinforced these practices as Southern Europeans have largely approved of the employment of core employees (primarily heads of household) and the dependence of youth and women on their families, fathers, and husbands. Specifically, Southern Europeans have favored job stability for male heads of household, whom society has expected to be the main (but not the only) provider of the family. One indication of the importance of family status is the criteria used in the 1980s in Greece to make hiring decisions in the country’s most coveted sector—the public sector. Number of children, level of family income, and other indicators of family status were to be weighed more heavily in the hiring decision than professional qualifications and work experience (OECD 1991a: 88)! As Karakatsanis (2000: 252) has argued, ‘this informal stopgap measure of the state is itself a culturally supported practice that ultimately reinforces the reciprocal relations inherent in familism’ and permits the state to take a less proactive stance in income redistribution throughout Southern Europe.

Fragmentation in Provision A second, closely related characteristic of the Southern European social welfare regime is fragmentation in the provision of social benefits, even to protected workers. At the heart of this fragmentation is a system of occupation-related funds that finances social insurance and is supported primarily by employer and employee contributions. Compounding the inequalities in provision between permanently excluded outsiders and privileged, employed insiders, these occupation-related funds provide widely varying levels of coverage to employed Southern Europeans, privileging certain occupations over others. The highest degree of such fragmentation can be found in Italy and Greece where a complex system of highly differentiated, occupationally specific funds exists—with over 120 funds in Italy and over 236 funds in Greece.10 Spain is characterized by a medium degree of fragmentation, with different funds for private, public, agricultural, self-employed, and

98

Social Policy, Democracy, and Citizenship

other workers, and Portugal, the least fragmented, still has a social security system that distinguishes between private and various categories of public employees (Ferrera 1996: 17–37; Guille´n 1997; Trifiletti 1998). As a result of this fragmentation, even ‘insiders’ receive widely varying benefits. Some workers—especially public employees, white collar workers and private wage earners in medium and large enterprises— are overprotected, while others—irregular employees in small businesses, the traditional service sector, and agriculture—are drastically underprotected (Ferrera 1996: 17–37, 1997: 13–24). Moreover, even pension benefits which, as Castles (this volume) illustrates, are quite high relative to other West European countries, vary from person to person, depending on a number of employment-related factors. In Spain, for example, pensioners who have contributed through wages can easily receive eighty percent of the basic salary. However, those who have not contributed—such as elderly women—or who have not fulfilled their contributions receive very low pensions—so low, in fact, that today around eight percent of the total number of pensioners— those on social assistance pensions—receive only one half of the minimum wage. Whereas the lowest pension in the social security system based on contributions equals the minimum wage (Herce 2004: 52–63; INE 2005). These disparities among various employment-related social insurance funds are paralleled by significant geographic fragmentation as well. Such geographic imbalances reinforce the fragmentation inherent in the Southern European welfare regime. In Italy, Spain and Portugal, where minimum income programs were introduced (although not fully implemented) in the 1990s,11 regional and local authorities have been given authority over these programs. As a result, programs tend to exacerbate, rather than reduce, fragmentation as certain cities and regions offer more generous benefits than others.12 For example, whereas means-tested benefits are available in the Basque country, in other regions these are either limited or barely existent. This territorial dimension is also quite apparent with regard to health care in Spain where, thanks to devolution, institutional arrangements differ from region to region (Fargion 1996: 152). Such regional disparities are particularly prominent in Greece and Italy.13 In Greece, the country’s two largest cities—Athens and Thessaloniki—have a predominant position in terms of health care provision, education, employment opportunities, and cultural development. For instance, even though statutory provisions provide that health care should be available throughout the country, it is often difficult to secure adequate attention in rural areas. This can be partly explained by unequal spending on

Social Policy, Democracy, and Citizenship

99

health, regional economic development imbalances, delays in decentralization, and shortages of trained medical staff willing to work in rural communities (Symeonidou 1996). It is reinforced by the fact that the country’s largest and best hospitals and clinics are located in Athens and Thessaloniki. Thus, while, in theory, residents of rural Greece have health insurance coverage similar to those in large cities, in practice, they must travel to Greece’s larger cities to receive highquality care. Many settle for the lower-quality care found in the periphery due to long waiting periods and high travel costs. Italian regional discrepancies also deserve special mention. In Italy, over twenty years after the 1948 Constitution entrusted the regions with legislative power in the field of public assistance, the devolution process finally began in 1970 with the establishment of regions. Two years later, the regions were given responsibility for certain functions, but no guidelines were provided as to how services were to be performed and benefits distributed. This has led to and worsened regional fragmentation with northern and central regions providing care superior to that of southern regions (Fargion 1996: 139–41). Moreover, the rapid industrial development of the North in the 1960s and 1970s promoted South to North migration and exacerbated, rather than diminished, unequal provision. As the number of wage earners in the North increased by 800,000 between 1966 and 1970, while those in the South decreased by 200,000, regional disequilibria were greatly reinforced. Moreover, comparatively high social welfare expenditures in the South have done little to successfully develop that region, which is still plagued by inadequate and uneven provision. Instead, public spending in the South has worked to greatly reinforce the extensive networks of patron-client relations (Putnam 1993; Ferrerra 1996: 17–37). In sum, Italy is highly dualistic and fragmented on geographic lines, and the south remains economically dependent and is a prime example of the Southern European model of welfare provision, while the north and center have successfully made the leap to the more modern, northern model. Thus, in Italy as in all of Southern Europe, universality and equity in social provision is lacking. Where Southern Europeans reside and what type of work they perform has much to do with the quality of their social entitlements.

The Resilience of Clientelism: Italy and Greece A third important characteristic that distinguishes the allocation of social welfare benefits in Southern Europe from that found in the North is the pervasiveness of clientelist practices within the social

100

Social Policy, Democracy, and Citizenship

welfare state. Particularly common in Greece and Italy, clientelism has further fragmented and limited the universalism of social coverage in these two countries by using public resources to manage consensus.14 Typically, welfare manipulation of this type has taken the form of a direct exchange of some kind of favor—such as unemployment benefits, income assistance, disability and old-age pensions, or employment in the public sector—in return for electoral support. The origins of such practices lie in the period before full democratization. In the absence of truly participatory, representative workers’ associations, clientelism became a method by which the ‘democratic’ states of Italy and Greece earned legitimacy as well as the complacency and support of key societal sectors. As a result, a system of clientelist relations profoundly affected the development, implementation, and character of both the Italian and Greek social welfare states. In Southern Italy, for example, the reciprocal clientelist relations that had developed during the Italian unification process had, by the end of the nineteenth century, also been cultivated between the state and its existing civic associations. During fascism, clientelist practices especially flourished and became explicitly linked with the provision of social welfare. After World War II and the demise of fascism, clientelist practices were inherited by Italian democracy. The Christian Democratic Party (DC), in particular, learned how to manipulate and use this system of quid pro quo relations in its own electoral favor (van Kersbergen 1991). As Sassoon (1997: 35–6) argues: The Italian state intervened massively, but without a plan. Public intervention was conceived as an instrument of social control . . . as a way of entrenching the electoral supremacy of the DC and as a way of providing private enterprise with favorable conditions for growth. In the formation of this new political power block the use of the state was essential.

The postwar development of the Italian welfare state reflects these clientelist tendencies. In order to obtain increased electoral support from the Italian middle class, the Christian Democrats further fragmented the occupational differentiation of the social security system, thereby contributing to its increased particularism. Thus, it is interesting to note that, while labor policies, social insurance, and unemployment benefits all emerged at this time, it was through patron-client relationships (operated through the enti publicci15) that allocations were made. By offering social services, particularly disability pensions and rural unemployment insurance to their supporters, the heads of the Enti could build up local constituencies of support. In this way, the old fascist strategy of securing social consensus through

Social Policy, Democracy, and Citizenship

101

social policy—without breaking the strength of Catholic action in the social assistance sector—was reinforced (Ascoli 1984; Pacci 1984). To date, partisan grass roots institutions, linked to the political parties, continue to regulate individual access to welfare (Ferrera 1984). As in Italy, clientelistic practices have had a long presence in Greece and have likewise been an obstacle to the universalization of social welfare provision. There, clientelism has its roots in Ottoman rule when Turkish overlords ruled through local Greek notables (cotsambasides), serving as intermediaries between the Ottoman state and Greek peasants. After Greek independence in 1830, the particularistic/personalistic organizational forms persisted, albeit in a modified way. Oligarchic families (tzakia) relied on patronage and clientelism to control the voting process and safeguard their representation in parliament. Using representative government in a clientelist fashion, they were able to protect their political privileges (Mouzelis 1978: 143; Kourvetaris and Dobratz 1987: 40). Such practices also existed in Portugal and Spain, albeit to a significantly lesser degree. While clientelism had less distributive power in Portugal than in Italy, it operated nonetheless—mainly in rural areas through personal contacts and favors. In Spain, however, Franco created and protected new welfare institutions but failed to establish clientelist practices through the state’s agencies. For this reason, it would be misleading to overemphasize the role of clientelism in Spain. Even though the franquist elite used its power and influence to occupy key decision-making posts in public sector enterprises (Instituto Nacional de Industria), banks and the more prominent private enterprises (Linz 1988), the use of clientelism as a mechanism to redistribute income and social services was limited. Unlike the Italian postwar state whose political system was based on party competition, the Spanish dictatorship did not need to buy votes, in the absence of national elections. Instead, the exchange of small favors was most widespread at the local level, mainly involving issues of land speculation with the purpose of economic benefit rather than social welfare. Such clientelistic trade-offs remain intact in Southern Europe and are a particularly obvious and egregious violation of the principle of social citizenship. By emphasizing the conditional nature of welfare provision—an exchange of benefits for political support—clientelism seriously undermines the principle of universality on which social citizenship is based. In Italy, the clientelistic use of the welfare state by governing parties for electoral purposes (particularly the DC party in the underdeveloped South) has been a notable feature of the social welfare state. In exchange for electoral support, governing parties have regularly provided citizens with individual benefits, particularly

102

Social Policy, Democracy, and Citizenship

highly protected public sector jobs which offer generous social benefits. Moreover, even the number of disability pensions—another way in which votes are exchanged for social welfare provision—are indicative of the clientelistic manipulation of the welfare state. Moreover, even the number and type of pensions—another way in which votes are exchanged for social welfare provision—are indicative of the clientelistic manipulation of the welfare state. Italian pension expenditure is the highest in the EU, both as a percentage of GDP (15.1 percent in 1999 compared to an EU15 average of 12.7 percent) and as a percentage of total social spending (62 percent of total social expenditure in 1999 compared to an EU15 average of 47.9 percent [Abramovici 2002: 2]). This is particularly true with respect to disability pensions—a favorite instrument of politicized social policy. The last forty years have seen a tremendous growth in the number of such pensions—for instance, from 1960 to 1972, the number of such pensions virtually tripled and actually surpassed the number of old-age pensions (Ferrera 1986: 446–55; Ferrera 1996: 17–37). While a tightening of eligibility requirements and enforcement reduced the number of disability pensions somewhat in the early 1990s, by 1994 there were still 7.2 million disability pensions in Italy, a country of about 56 million people (OECD 1997: 94). Moreover, regional variations in eligibility rules reflect the use of social policy as a clientelistic tool. In 1994, 25 percent of all pensions in Northern Italy were disability pensions; at the same time, disability pensions accounted for 46 percent of pensions in the South (OECD 1997: 93). Thus, while successive governments implemented a number of important reforms to the pension systems, extending entitlements, and thereby making welfare provision somewhat more inclusive and universal, they simultaneously extended the reach of patronage, consolidating their own clientelistic practices. In Greece, too, the reintroduction of electoral politics in the mid1970s actually promoted clientelism as it intensified the competition for support between parties. As one observer argued: The parties’ traditional reliance on the purchase of support by state patronage was increased by the appearance of genuine competition between them, and by the new power of voters. It was largely for this reason that state expenditure on welfare and jobs in the public sector—both of which had risen gradually in the twenty five years up to 1974—increased steeply in the following fifteen years. (Close 1993: 228)

Specifically, with the establishment of mass party organizations in Greece, traditional clientelist control by local notables gave way to a new form of clientelism in which the party leadership played the definitive role in allocating state spoils. This distinct form of clientelism

Social Policy, Democracy, and Citizenship

103

consisted of the systematic penetration of the state apparatus by party members and the distribution of political favors through it. To accommodate the party faithful and provide them with much-coveted public sector employment, the number of existing positions was increased and new public sector departments created—all in an attempt to maintain the party’s electoral base. To create as many positions as possible, the state encouraged early retirement (with generous pensions), freeing up positions for new hires. Data confirm the widespread use of such methods in Greece, as general government employment grew at three times the rate of private sector employment, increasing at a rate of about 30 percent in the 1980s. The Panhellenic Socialist Movement (PASOK) was especially prone to such ‘bureaucratic clientelism’, and used it extensively when making public appointments and granting other political favors.16 Thus, based on the modern party machine, clientelism and personalism are deeply entrenched in Greece. Another way in which clientelism has permeated social welfare provision in Greece is through the manipulation of political business cycles by both Nea Demokratia and PASOK, who increased social welfare payments in the run-up to elections. While there are many reasons for the increase in state subsidies—including an increase of the aged as a percent of the total population, a monetary increase in the average pension, and a cost of living adjustment for pensions17—it is important to note that electoral concerns, explicitly couched in clientelistic terms, also greatly influenced the level of government expenditures. In short, the important point about both the Greek and Italian welfare states is that, unlike in most other countries, where the relationship between voting and welfare is impersonal, in these countries there is a one-forone exchange of votes for social welfare favors. Once again, the universalism of the social welfare state is greatly undermined in the Southern European model.

Familism In the absence of a formal, well-developed and universal welfare state, informal, family-based strategies of welfare provision have developed and become widespread in all four countries. As Esping-Andersen (1999: 51) has said about Southern European public policy, ‘familialism easily goes hand-in-hand with a very passive and undeveloped’ social welfare state. The lack of universal welfare coverage in Southern Europe reinforces the family’s role as the main provider in society. More specifically, as Naldini (2000: 71–2) illustrates for Italy and Spain, welfare policy throughout Southern Europe continues to be

104

Social Policy, Democracy, and Citizenship

‘gender-specific, promoting motherhood and aimed at reinforcing the traditional distribution of authority and power within the family— between genders and between generations—as well as strengthening solidarity between kin members’ (see, also Nash 1991; Saraceno 1991; Naldini 2003). So, for example, when family allowances were introduced in Italy and Spain in the 1930s, they quickly became pronatalist instruments ‘aimed at increasing population, supporting the malebreadwinner family and implementing a ‘‘family wage’’ ’ (Naldini 2000: 72). They were paid to husbands/fathers and, for women to be eligible, they had to be widows, single mothers or married to men who were unable to work. Given such historical patterns of development, Southern Europeans and, especially, women have learned to provide each other with welfare support and assistance in an effort to offset the imbalances of formal provision (see Guerrero and Naldini 1996; Martin 1996; Trifeletti 1999; Simoni and Trifeletti 2004). In this way, individuals, rather than the state, have taken the lead in increasing the lifechances of disadvantaged family members. Several explanatory variables—economic need, the rigid and gendered labor market, the underdeveloped welfare state and traditional family values—can account for this provision of family welfare. First, peasant culture—where access to inheritance was explicitly linked to the obligation, rather than the choice, to care for fellow kin— has not fully disappeared, but has been reproduced in the urban context.18 Despite socioeconomic modernization and concomitant attitudinal modification, women (mothers, daughters, and daughters-in-law) still continue to construct their self-identities largely around family responsibilities, continuing to serve, as is often expected, as the backbone of family care.19 This has been facilitated by persistently high levels of female unemployment as well as relatively low levels of female labor force participation (eight to ten percentage points below the European average) in Greece, Spain and Italy (Flaquer 2000: 22). Specifically, Southern European cultural norms and practices ‘teach’ family members to assist each other at times of need rather than rely on the state (on Spain and Italy, see Guerraro and Naldini 1996). As Karakatsanis (2000: 248) points out about Greece: In this cultural milieu, where pre-marital cohabitation and single-person living arrangements are neither widespread nor positively sanctioned by society, unemployment benefits for first-time job seekers are nonexistent and are largely viewed as unnecessary both by society and government alike. Instead, generous income assistance is provided by the extended family and most firsttime job seekers are happy to rely on such financial support for relatively long periods of time rather than accept undesirable jobs that they do not consider up to their high standards.

Social Policy, Democracy, and Citizenship

105

As a result of persistent cultural norms, the significant socioeconomic modernization experienced in all four countries has not significantly weakened family networks or kinship relations. On the contrary, demographic and family structures retain their distinctive pattern, marking them out as more ‘traditional’ than those of Northern Europe. Despite the fact that fertility rates have significantly declined in Southern Europe, the number of divorces, cohabitating couples and births out of wedlock continue to remain low (Guerrero and Naldini 1996: 42–66). Seen from this perspective, the extremely high levels of youth unemployment, the structural rigidities of the labor market, and the lack of social provision for the young have simply reinforced traditional cultural norms supporting family solidarity and dependence. In this regard, Southern European families serve at least two important and complementary functions. On the one hand, the family serves as employer, providing employment in family owned businesses and farms or helps job-seekers find work through personal and family connections. On the other hand, it also functions as a provider of social assistance when family members and other loved-ones find themselves in need. As employer, many families run small family-owned businesses that often employ several family members. As a result, in the early 1990s an extremely high proportion of young Southern Europeans were employed by the family—69 percent in Greece, 65 percent in Italy, 58 percent in Portugal, and 61 in Spain (Jurado 1995).20 It is also interesting to note that a substantial number of young people in the south gain employment through family connections. In Greece, for instance, personal influence and connections, meson, often serve as the most reliable route to gainful employment. Such connections are so important that, according to one survey of Greek workers, less than ten percent of those surveyed indicated that the government-run job placement office (OAED) had actually helped them find work (OECD 1996: 83). Instead, Greek workers readily acknowledge their reliance on informal networks of personal assistance to find employment. Thus, the Southern European family serves not only as employer, providing both formal and informal work for unemployed family members in family-owned businesses, but also provides informal job-placement assistance, helping find employment for job-seekers through personal connections and friends. Perhaps the most important way the family serves as a redistributive agent, however, is by pooling resources to support family members in need. Take, for example, the case of child care and care for the elderly. Southern European welfare states have been extraordinarily reticent to provide public child care services (especially for children under three years of age) to working parents (Flaquer 2000: 19), fewer than five percent of whom are covered in

106

Social Policy, Democracy, and Citizenship

Italy, and less than three percent in Greece. Instead, a great deal of societal pressure is placed on working mothers and nonworking, usually maternal, grandmothers to assure that childcare will be provided by older female family members (Esping-Andersen 1994: 16), who are among the few trusted in these societies to provide high-quality, loving care to children. Proof of this widespread attitude is offered by an Italian study, which confirmed that for every young working woman surveyed in Turin there was at least one corresponding mother or mother-in-law who played an active role in childcare provision (Ruspini 2000: 234–5). A second example of such attitudes is Greece, where there are remarkably few childcare facilities,21 and where more than half of working parents rely on grandmothers for the care of preschool children. Greek parents are suspicious of public childcare centers (Sims-Schouten 2000). As one Athenian working father commented: ‘My mother-in-law is taking care of my child. I will never use child care facilities for my children, I don’t know the people who work there, so why would I trust them?’ (quoted in Sims-Schouten 2000: 281). Indeed, such attitudes persist throughout Southern Europe, where most people believe that family-provided services are superior to those provided by the state (Flaquer 2000: 26). Thus, public demand for state provision of such services is relatively low, and the family continues to be the main provider. These arrangements are made possible by the fact that the elderly continue to live with their children (37 percent in Spain, in 1985, and 39 percent in Italy, in 1990—OECD 1994) or in very close proximity to them. Moreover, as Flaquer (2000: 20) argues, such ‘intergenerational dependency takes place both at the beginning and at the end of the life cycle’. Since the number of people over seventy-five who live alone in Southern Europe (with the exception of Italy) is below the European mean, this practice makes possible the delivery of services requiring daily interaction, including both the provision of childcare by grandmothers, and care of the elderly by the next generation of women, rather than by nursing homes or other assisted living programs.22 Thus, facilitated by the still-widespread practice of cross-generational cohabitation and collaboration, the Southern European family provides important social welfare functions for children, the elderly and other family members who are unemployed, aged, disabled, or in need of assistance. Portuguese and Greek practices illustrate familism particularly well.23 In neither of these historically rural and traditional societies did the ‘breadwinner’ model predominant in the advanced industrial countries to the North, in which the male head of household generated the family’s single source of income, take root. Instead, a family wage

Social Policy, Democracy, and Citizenship

107

emerged, that is, a domestic collective income to which all family members contributed and from which all potentially benefited. In the face of unprecedented levels of unemployment, the practical need to generate this domestic collective income has been reinforced and become even more widespread. Typically, this involves pooling the income of a highly protected head of household with those of other family members, some of whom may hold full- or part-time jobs in the formal economy, while others are engaged in contract or informal work in the underground economy. In addition to the wage of the male head of household, women contribute income from formal, informal and/or paid irregular work (baby-sitting, cleaning, catering, secretarial work done from the home, casual service jobs, retail tourism, industrial home working, etc.), while young people contribute by engaging in door-to-door sales, express delivery, distribution of advertising materials, summer jobs, etc. A third important source of income to the family wage are government subsidies (especially to farmers), old-age and disability pensions as well as interest and other types of income from property ownership, rentals, and savings. Finally, the family wage benefits from the unpaid ‘assistance’ of family members who work informally as unpaid family help (Hadjimichalis and Vaiou 1990a: 94–5). Such practices are particularly common among workers in the traditional agricultural sector of Greece. In addition to working on small, rather unprofitable, family-owned farms, they tend also to seek full- or part-time employment (‘pluriactivity’) elsewhere, usually in the informal economy (Hadjimichalis and Vaiou 1990b). It is important to note, however, that this practice is also widespread in towns and cities, especially among individuals who own small family-owned enterprises that rely on the often unpaid ‘assistance’ of wives, children, and other family members. Such income pooling is believed to benefit the entire family, reducing its dependence on any single source of income, lowering the threshold of the minimum necessary and acceptable wage, creating a larger collective income than otherwise possible and providing a surplus income that is used to assist those family members in need. Through these kinds of arrangements, the quality of life of individual family members, particularly the young, is greatly enhanced (Karakatsanis 2000). The pattern is similar in Portugal, where a survey revealed that only twelve percent of the long-term unemployed relied on state benefits as their main source of financial support, while over fifty percent of respondents were supported by their families (Mendes and Castro Rego 1992; Pereirinha 1996: 210; quoted in Bermeo 2000: 274). This arrangement also allows families to receive income from an occupation-related

108

Social Policy, Democracy, and Citizenship

fund linked to formal employment of a family member (usually the father). As long as one family member maintains this link with the formal labor market and acquires the privileges and social benefits associated with ‘insider’ status, a family safety net emerges and compensates for the lack of universal social benefits (Gough 1996: 14). As Trifiletti (1999) correctly argues, while the state does not guarantee a family wage, it does allow families to pursue strategies which ensure that at least one of its members—usually the male head of household— has a good, stable job with access to social entitlements. Within this context, although the provision of welfare services in Southern Europe remains low and nonuniversal, income transfers to families can be regarded as relatively generous. Formal state welfare policies have been formulated to complement, rather than replace, the family provision of services. A particularly revealing example of such ‘intrafamiliar redistribution’ can be found in a recent study comparing Spanish and French youth, in which Teresa Jurado Guerrero (1999) observes that young Spaniards remain in their parents’ homes for longer periods of time than their French counterparts. This is true despite the fact that young people in both countries have to enter a labor market with high unemployment and short-term employment. Guerrero maintains that young Spaniards, like other Southern European youth but unlike their French counterparts, connect a final departure from their parents’ homes not with job-seeking, but with marriage. The figures presented in Table 3.2 are striking, and fully reveal the extent to which the family’s role as social welfare provider is reemerging in the face of economic hardship in an otherwise modern and affluent society. Specifically, between 1989 (a time when the Spanish economy was relatively healthy but when unemployment was high) and 1997 (at the end

Table 3.2 Young people residing at home in Spain 1989

1997

Ages 19–24 Males Females All

88.3 76.8 82.6

90.1 80.2 85.1

Ages 24–29 Males Females All

46.0 32.2 39.3

59.7 43.4 51.5

Source: CIS Surveys, July 1989 (1,813) and 1997 (2,262).

Social Policy, Democracy, and Citizenship

109

of a minor recession), the percent of young people between the ages of 19–29 (both male and female) living in the homes of their parents increased. Guerrero claims that, whereas French youth appear to establish their independence with the support of active employment policies, extensive public housing supply, housing allowances, comparatively high wage levels and the frequent support of parents, in Spain, the difficulty in finding work is not mitigated by public housing or extensive and generous public allowances. On the contrary, young Spaniards tend to receive subsistence-level wages and are forced to rely on longer-term parental assistance. Based on such factors, some have argued that the institution of the family is experiencing a ‘renaissance’ (Kumar 1997). The absence of comprehensive state provision regarding housing and family-services has forcefully encouraged the resilience of familialism in Southern Europe (see, e.g., Allen et al. 2004). It is further interesting to note that, in Spain, this phenomenon is reinforced by legal provisions (la obligacio´n de alimentos) requiring family members to financially support each other. Parents are obliged to take care of their minor and, at times, mature children. Likewise, children are liable to care of parents when the need arises. And siblings have an obligation to provide ‘restricted support’ (alimentos restringidos) to each other in the case of temporary and exceptional need. Failure to fulfill these obligations may lead to criminal prosecution (Alberdi 1995; Flaquer 1998; Oinonen 2000: 148). Nancy Bermeo (2000: 274) has argued that as a result, ‘the Spanish family. . . ensures that extreme poverty and social exclusion are very limited despite the record levels of unemployment’ (also see Laparra and Aguilar 1996: 108). Similar phenomena can be discerned in Greece and Italy, where the percent of men and women aged 25–29 living with their parents increased by thirteen percent and fifteen percent, respectively, between 1986–96. While this increase is also indicative of the later ages at which young people marry in the south, Ferna´ndez Cordo´n (1997) reminds us that it also demonstrates the extent to which the situation of young people has worsened as a result of high youth unemployment, labor market rigidities, and the lack of social provision for the young (Flaquer 2000: 20; also see Aassve et al. 2002). In sum, while the Southern European family provides social assistance to the unemployed, it relieves the state of the need to expend increased resources on such benefits (Karakatsanis 2000: 247). To take the argument one step further, continuing reliance on such strong kinship solidarity has allowed some to assume that public welfare provision is redundant, perpetuating the underdevelopment of the welfare state in Southern Europe (Bahle and Pfenning 2000: 2).

110

Social Policy, Democracy, and Citizenship

Finally, it is important to emphasize that economic need is not the only explanation or cause of the Southern European family model of welfare. As previously mentioned, a strong cultural component—the high value these societies place on family solidarity—continues to reinforce the practice, teaching family members to assist each other at times of need rather than rely on the state. This is reflected in the case of single-person living arrangements. Whereas the 1995 European average for 25–29 year-olds living alone was twelve percent for men and ten percent for women, the average in the south was well below the mean, with Spain and Portugal being very much below the European average: in Portugal, two and three percent of single men and women respectively, and in Spain only one percent of either sex live alone (Flaquer 2000: 19). Young people who remain in their parents’ homes for extended periods of time (often until they marry) do so not only because of economic need but also because of cultural expectations and norms.24 In other words, in addition to constraints imposed by the labor and housing markets, and the paucity of state welfare services, there is a cultural expectation that young people remain within the parents’ household until marriage. Thus, public policy and culture are linked in a reinforcing manner: the absence of universalistic welfare provision requires young people who have not secured full-time employment to depend on their families for support, while cultural expectations regarding cross-generational family cohabitation have undermined pressures for the development of universalistic welfare policies. More specific manifestations of this interplay between cultural values and social policy development (or the lack thereof) can be seen in three sets of case studies. The first focused on single mothers in Turin: This study found not only that the majority had never contacted welfare authorities (social services, unemployment offices, regional social insurance offices), but also that most single mothers had negative opinions of such offices, generally perceiving them as opponents rather than helpers (Morner 2000: 258). Single mothers’ economic dependence on the family was not perceived as a social problem, while dependence on public welfare was viewed as ‘bad’ (Saraceno 1994). As Ruspini (2000: 237) maintains, ‘the hidden assumption is that the family, through the unpaid work of women, is the ‘‘natural’’ main provider of welfare’. The second specific example is that of first-time job-seekers in Greece, where premarital cohabitation and single-person living are neither widely practiced nor widely condoned by society. Most firsttime job-seekers are content to rely on generous parental support for relatively long periods of time rather than accept jobs that are not to

Social Policy, Democracy, and Citizenship

111

their liking. Many remain in their parents’ homes and dependent on their families until a relatively advanced age while awaiting highly prized positions (particularly in the public sector, which two-thirds of Greeks have expressed a desire to obtain—OECD 1990a: 24). In effect, this is tantamount to a preference for unemployment and dependence on the family over independence and low-pay private sector jobs. Moreover, Greek parents tend to reinforce this dependent relationship by providing generous support for children and expecting little in return. Indeed, parents who opt out of such arrangements are stigmatized in accord with widespread cultural values. A third example is the dowry system in Greece (which was abolished by legislation enacted in 1983). While regional variations existed, traditionally, the dowry system was one aspect of arranged marriages, in which the bride’s family was obligated to provide its future son-in-law with a negotiated amount of cash, property, livestock and other goods in exchange for his willingness to marry their daughter.25 The exact size of the dowry varied, depending on the bride’s socioeconomic status, and came from years of family savings, the earnings of the girl herself, the sale of family property, immigrant remittances, and even borrowing (Kourvetaris and Dobratz 1987: 157). While dowries have been formally abolished, Greek parents continue to provide monetary gifts, homes, household furniture, appliances, linens, land and/or rental properties to them at the time of marriage (Symeonidou 1996: 81). This ‘marriage subsidy’—often in the form of rent-free living—goes a long way toward offsetting the consequences of low pay, underemployment and unemployment for many young couples and is yet another way in which the Greek family continues to serve a crucial, albeit informal redistributive function. It should be noted that this emphasis on rent-free living, or home ownership, is not unique to Greece but exists throughout the south and has been seen by many as a form of insurance against social risk (Trifiletti 1998; Flaquer 2000). Specifically, Southern European societies have the highest proportions of owner-occupied housing (Castles and Ferrera 1996: 163–84), some 20 percentage points above the European average (with Portugal closer to the mean). Flaquer (2000: 24) maintains that this insistence on home-ownership facilitated by government incentives (special subsidies and tax benefits in Spain and Greece, and reduced mortgage rates in Italy) can be interpreted as a form of ‘family self-protection’ in the face of very limited and unequal social welfare coverage. It is important to note, however, that while societies which rely on such family solidarity may, on the one hand, be ‘preventing a great number of people from falling into exclusion’, they may also, on the other hand, be ‘leaving a small part

112

Social Policy, Democracy, and Citizenship

of the population in a completely unprotected situation’. As Laparra and Aguilar (1996: 91) point out, ‘There is no way of granting a right to family solidarity’. Finally, as stated in the introduction to this chapter, it is important to emphasize that we do not view tradition and culture as static; they are continually changing and in flux. We recognize that in Southern Europe declining marriage rates, later ages at which young people marry, rising divorce rates26 and single-parenthood,27 as well as attitudinal change concerning premarital sexual relations, gender roles and family values, are bringing about profound cultural transformation. The attitudes of young people are being altered as more of them begin to seek greater levels of independence from family and kin. Divorce rates, for example, have increased markedly, from fewer than two out of one hundred marriages in Portugal and five out of one hundred in Greece in the 1960s to almost nineteen percent in both countries by 1995 (Carlos and Maratou-Alipranti 2000: 38). Women, long the backbone of ‘welfare-provision’ are also challenging their traditional and dependent roles within the family as they attain higher levels of education and formal work outside the home. Thus, the rates of female participation in the labor markets in Southern Europe remain below the EU average of 59 percent (with the exception of Portugal), they have nonetheless increased dramatically over the past several decades (see Table 3.3). Also, the higher the level of women’s educational attainment, the higher their participation rate. Since younger women in Southern Europe are generally bettereducated than their older counterparts, and since increased levels of educational attainment are positively related to labor force participation, female participation in the formal labor force is likely to substantially increase in the near future, bringing inevitable consequences to the family and its traditional welfare function. In fact, two independent Italian surveys on family and fertility already indicate that increasing numbers of married women and mothers wish to remain employed after childbirth, and that a growing demand for public services among more educated women is emerging. In the late 1990s, studies imply that Southern European attitudes about fatherhood and the proper role of fathers as caregivers are beginning to change (Bimbi and La Mendola 1999; Sabbadini 1999; quoted in Sundstro¨m 2000: 207). Despite the fact that practices continue to lag behind their northern European counterparts (Carlos and Maratou-Alipranti 2000: 45), it is expected that over the coming decades Southern Europeans and their children are likely to turn increasingly to the state, rather than the family, for social assistance. As Carlos and Maratou-Alipranti (2000: 34)

Social Policy, Democracy, and Citizenship

113

Table 3.3 Female labor force participation rates in Southern Europe

Italy Spain Greece Portugal EU-15 Avg.

1989

1999

43.8 40.2 43.5 59.1 54.2

46.0 48.9 49.0 66.8 59.2

Source: OECD (2001b).

maintain, changing gender roles within the family—particularly with respect to higher levels of education, labor-market participation, and financial independence of women—will give rise to new demands for public social services, and pose new challenges to the current Southern European model of welfare. In sum, although modernizing social trends have come slowly, divorce rates remain low, and cohabitation and single-person living arrangements are still relatively rare in Southern Europe, liberalizing attitudes have nonetheless taken root. As Martin (1996: 28) maintains, attitudinal modernization is likely to accelerate from one generation to the next. As more and more Southern Europeans adopt modern attitudes, as greater numbers of women decide to exit the home and to enter formal employment, and as increased numbers of young people leave their parents’ homes to live on their own or with roommates, the number of women available to engage in domestic caregiver functions is likely to decline. Thus, cultural and social transformations will continue to have an undeniable effect on the family provision of welfare (Guerrero and Naldini 1996: 62). As Haris Symeonidou (1996: 84) rightly asserts about Greece, the ‘family cannot be regarded as a welfare state substitute in the long term’.28 We maintain that this holds true for the other countries of Southern Europe as well. Thus, we conclude that four key characteristics—labor-market rigidities, fragmentation in the delivery of social welfare benefits, widespread patron-client relations, and familistic social welfare patterns—comprise a unique model common to the four Southern European countries. To understand how that model came to be, one must examine the emergence of social welfare in Southern Europe. As we shall see, along with democratization, there has been significant movement toward social citizenship, but the traditional underlying patterns of welfare provision remained largely the same.

114

Social Policy, Democracy, and Citizenship

The Historical Emergence of the Southern European Welfare State Welfare Development Before Democratization: The Impact of Corporatism and Authoritarianism To understand the emergence of such fragmented, familistic, and informal patterns of welfare provision, we maintain that it is crucial to understand the historical, political, and social legacies of Southern Europe. Specifically, it is important to emphasize that the Southern European social welfare regime was profoundly affected by corporatism as well as by decades of authoritarianism in Portugal, Spain, and Italy and restricted democracy in Greece. Unlike the inclusive corporatist model found in Germany, France, Austria, and Belgium, Southern European corporatism was limited and ‘imperfect’. In contrast to workers in the North, especially those in countries characterized by strong social democratic traditions, Southern European workers played a limited, circumscribed role in the design and implementation of social welfare policies. Thus, while their Northern neighbors achieved a greater degree of social welfare universalization through class conflict, Southern workers remained outside the public policy arena and did not participate in the formation of social welfare programs. In the absence of working-class participation, public policies were introduced, instead, through government fiat, with governments taking the lead in formulating, designing, managing, implementing, though not always funding, social policy. This was true both of authoritarian Spain and Portugal in the 1940s and 1950s, but also of ‘democratic’ Italy and Greece. In all four countries, trade union strength was consciously limited and workers were prevented from influencing legislation through their political parties and elected representatives. In short, Southern European workers’ organizations lacked the strength even to participate in a limited manner in the crafting of social welfare legislation. In the absence of powerful workers’ movements and labor unions taking the lead in the development of social welfare, the Italian model was initially influenced by the Catholic Church and its emphasis on charity and, later, by Italian fascism (see Feltrin 1991). Up to the end of the nineteenth century, religious (opere pie) and municipal charities—not the state—provided relief to the poor. In 1890 a large number of charitable organizations were brought under state control and placed under the supervision of provincial authorities. However, no clear provisions were made to control the scope and content of their

Social Policy, Democracy, and Citizenship

115

activities. Thus, despite their legal status as public entities, the charitable organizations operated as private bodies (Fargion 1996: 138). Fascism later transformed the Italian welfare state into a corporative authoritarian system, relying on a high degree of state direction. While the fascists dissolved the traditional self-help societies, they replaced them with the Enti Nazionali Autarchici di Assistenza e Providenza (Autarchic Organizations of Social Assistance), state-controlled entities that placed welfare distribution into the hands of the regime’s own political party. The enti endured long after the demise of fascism (see Ferrera 1993), and important reforms leading to increased workers’ rights, better working conditions, and decentralized bargaining were postponed. As Paci (1984) argues, the authoritarian and interventionist role played by the Italian state was paralleled by the inability of civil society to organize and by the colonization of trade unions by political parties which effectively excluded them from direct interest representation in the economic and political decision-making process. Workers were apprehensive about confronting government with their demands, and even in the more dynamic sectors of civil society, particularly those in Northern Italy, citizens remained largely outside the policy arena. The state, facilitated by a weak liberal ruling class and by fragmented workers’ organizations, seized the opportunity to control both the design and management of social services. Thus, when the state introduced compulsory insurance for disability and old age in 1919, these provisions were not in line with the aspirations of workers themselves. In short, Italian corporatism served to exclude rather than incorporate the interests and demands of citizens into the formulation of social welfare policies (Salvati 1982: 173–4). In Portugal, the Estado Novo was heavily influenced by Catholic corporatism as well as by the example of Italian fascism. The Salazar regime privileged professional organizations and corporations willing to subordinate themselves to the tutelage of the state. Government representatives had the right to be present in all corporation meetings and even to veto decisions taken by the organizations’ members. Public policy was designed by the state representatives and, in the early period (1928–40), annual budgets were drafted by Salazar himself (Schmitter 1975). Even the ‘right to public assistance’ was abolished in the 1933 Constitution and replaced by a corporatist commitment to ‘promote and support institutions of solidarity, welfare, co-operation and mutualities’ (Guibentif 1996: 222). This system of state-sponsored corporatist tutelage effectively eliminated worker influence over public policy. Within the associations workers were impotent (Caetano 1935: 37, quoted in Schmitter 1975),

116

Social Policy, Democracy, and Citizenship

and it was virtually impossible for them to organize outside the statesponsored corporations. This, in addition to the fact that the right to strike was illegal, made independent workers’ movements extremely weak and kept their membership low. As late as the 1980s, union membership was only around six percent of the total population. In addition to limiting the ability of workers to organize and denying them the right to strike, corporatism in Portugal also played an important role in the fragmentation of social welfare provision. One example were the Caixas de Aposentac¸a˜o (Lodging Boxes) which had been introduced as early as 1875. These caixas were responsible for collecting contributions from employers, workers, and the state and redistributing them in the form of social security benefits to workers. Different professional sectors belonged to different caixas, however, and therefore received different levels of coverage. With the emergence of the Estado Novo, new caixas were created, many of them without any source of public support. They, along with other organizations (such as the mutual help societies and the Catholic Misericordias), created a complex and fragmented map of social protection, leaving some geographical areas and persons unprotected while heavily overprotecting others. Only in the 1940s were efforts undertaken to create a more comprehensive network of caixas around the country and to extend coverage of the existing caixas to new categories of beneficiaries. However, the funds were extremely heterogeneous and varied—there were national and regional level funds, company funds, district funds, and family benefit funds. The range of benefits varied from fund to fund, and each fund had different rules. Thus, the caixas reforms of the 1940s did little to provide common, let alone universal, standards of welfare protection for all Portuguese citizens (Guibentif 1985: 945–59). As in Spain, state support for welfare in Portugal was minimal prior to its transition in 1974. While several compulsory insurance programs were created in 1919 to cover work-related accidents, disability, old age, and illness, these programs were financed through employee and employer contributions and administered by the confessional organization, Misericordias, the nonconfessional mutualities and the caixas. In 1931–2, legislation made the mutualities, not the state, responsible for providing members with health insurance, old-age, disability, unemployment, maternity, and survival pensions and, in 1935, the Estatuto do Trabalho Nacional (National Statute for Work) explicitly stated that ‘social providence’ should be financed by employers’ and workers’ institutions (Maia 1985). Unlike the other three Southern European countries which, by the 1960s, were experiencing rapid economic growth, the Portuguese economy lagged considerably behind. Despite this, however, reforms aimed

Social Policy, Democracy, and Citizenship

117

at achieving nationwide coverage of social benefits were introduced and a general regime of health care was established in 1963 which, in theory, was to cover all workers. In 1963, social assistance institutions ( previdencia social) were also reformed in such a way that the caixas were homogeneously organized on a regional basis, coming to constitute a federation. Despite such reforms, however, universalization was not achieved. Social welfare continued to be provided along highly fragmented, unequal, and largely employment-driven lines. In short, welfare-state development in authoritarian Portugal fits the standard Southern European model: fragmented, uneven, and corporatist. In Spain, state corporatism was only half-heartedly institutionalized under Franco, yet there were clear corporatist tendencies in the provision of social welfare nonetheless (Linz 1988). As the Salazar regime in Portugal had done, franquism destroyed parliamentary democracy and, with it, autonomous working-class organizations. Repressive labor policies were introduced, democratic unions were outlawed and collective bargaining abolished. During the regime’s early years, when the fascist Falange was most influential in labor policy, social corporatism emerged between employers and workers as well as between regional and local organized interests. Later, when the Fuero del Trabajo was passed in 1938, membership in industry-wide, corporatist vertical unions (sindicatos) became compulsory. Composed of workers, management, and government representatives, these sindicatos remained officially active until the end of the franquist dictatorship in 1976. Thus, as in Italy and Portugal, corporatism in Spain greatly weakened labor. In effect, it destroyed the ‘resistance from the working class and traditional left-wing unions and parties to the creation and functioning of this corporatist labor-repressive institutional structure’ (Maravall 1978: 22). It was not until the 1950s, when economic liberalization began, that industrial relations began to change, giving management greater scope to iron out informal agreements with workers, especially in the larger industries. Later, from about the early 1960s, workers themselves began to acquire increased strength, leading to intensified industrial conflict throughout the 1960s and 1970s. What is particularly relevant, however, is that welfare provision in Spain was not historically confined to the vertical unions, or corporations. Instead, the franquist Ministry of Labor created new structures that served to perpetuate the regime’s rhetoric that it was directing its caring function toward ‘all Spaniards’. It was unsuccessful in doing so, however. Its attempt to create a truly ‘organic democracy’ within labor relations failed in the face of post–civil war animosities which deeply divided Spanish society. Moreover, Franco’s efforts to establish clientelist opportunities through state agencies were largely unsuccessful as welfare

118

Social Policy, Democracy, and Citizenship

provision was not confined to the regime’s sindicatos. Mirroring the unwillingness of the Southern European corporatist state to directly fund social welfare, the franquist regime studiously sought to limit its welfare obligations. While compulsory insurance programs covering oldage (1919), maternity (1926), and labor accidents (1932) had been incrementally established, as in Italy and Portugal they were financed through either ‘mutualities’ or the Catholic church. Even in the 1940s, when Spaniards saw the development of health insurance (covering some 30 percent of the population), family support programs, and old-age and disability insurance, the provision of such benefits continued to be highly particularistic. While the state subsidized some of these program costs, it was the private sector which was largely responsible for covering most contingencies. Even the rapid economic development of the 1950s and 1960s did not encourage a revision of the regime’s social policies. Instead, Franco and his supporters remained most interested in conserving the state’s resources (Gunther 1980). Even though unemployment insurance was formally established in 1961 and while the 1963 Basic Law of Social Security aimed, in theory, to conform to the principle of income maintenance, in practice, the system remained fragmented and seriously underfinanced.29 Ironically, Spain’s democratization in the 1970s actually served to dampen down workers’ demands for improved social services. Whereas early protests had centered on wages and working conditions, by the time of Franco’s death and the transition to democracy, protestors were increasingly demanding democratization and political liberalization (Maravall 1978: 19–43). These larger political issues had come to predominate, as issues of welfare provision were deemphasized and relegated to the background. It would not be until after the establishment of democracy that unions would seriously address social welfare issues. By that time, however, the welfare structures of the state had yet again been engineered by state representatives in the absence of workers (Garcı´a 1994). In contrast with the other Southern European countries, Greece experienced much shorter periods of authoritarian rule. Almost from the time it gained its independence in 1830, the country experienced a Western parliamentary form of government. However, because its democracy was oligarchic and restrictive, Greece actually closely resembles the pattern of welfare evident in the rest of Southern Europe. Mouzelis (1986) describes the period between 1830 and 1909 as one by which formal liberal, pluralist representation existed but where the majority of the Greek lower classes were excluded from active politics by the use of political manipulation and a system of cooptation and incorporation. Moreover, Greece never really developed a sizable, politically-active

Social Policy, Democracy, and Citizenship

119

urban population despite rapid urbanization and the growth of its middle class. This is partly explained by the job opportunities that migrants from rural areas found in the urban centers, which during the early nineteenth century helped keep unemployment low. In the absence of active trade union organizations, many Greek workers (like other Southern Europeans) belonged to mutual aid societies which provided them with welfare assistance, albeit in fragmented and particularistic form. Thus, even when this period of oligarchic rule was finally brought to an end in 1909, Greek politics continued to be characterized by the absence of autonomous trade unions and mass working-class parties. Greek workers, like their Southern European counterparts, played a small and unimportant role in active politics. Even when trade unions finally appeared in Greece around 1909, their leadership lacked any form of ideological and organizational cohesiveness. As a result, they exercised little influence over social legislation, leaving such initiatives to the liberal government of Eleftherios Venizelos. However, the emergence of Venizelos as a champion of working-class interests was, in many ways, a negative phenomenon for workers. As Mouzelis (1986: 60) argues, while Venizelos granted certain rights and benefits to the working class—including the implementation of laws covering working hours, women, child labor and industrial accidents—he included legal provisions that placed the entire trade union movement under state tutelage, thereby effectively eliminating its autonomy. At the same time, his ‘mildly reforming measures helped to neutralize the development of strong socialist and agrarian movements’ (Clogg 1992: 77). In short, while de jure corporatism did not exist in Greece as in the other Southern European countries, a system of de facto incorporation functioned within the existing rules of parliamentary democracy. And while secondary associations were effectively free for most of Greek history, they remained weak and vulnerable to state manipulation. The General Confederation of Greek Labor (GSEE), for example, created in 1918 by the Venizelos government, was a prime example of state paternalism actively sabotaging the development of autonomous secondary associations. As with the implementation of laws protecting workers, the initial drive for the creation of the GSEE came neither from the trade unions nor from the socialist intelligentsia, but from the liberal Venizelos government which wished to set up a unified trade-union movement under government control (Mouzelis 1986: 60–75). As Theodore Katsanevas (1984) argues, the ‘paternalistic unionism’ developed by Venizelos not only regulated unions but also acted on the belief that the solutions to labor problems ought to come from above. And as Schmitter (1995) and others have demonstrated, forms of state corporatism

120

Social Policy, Democracy, and Citizenship

continued throughout the twentieth century (also see Sotiriopoulos 1995). As a result, unions were effectively controlled by their integration into a three-tiered edifice. While it was difficult for the state to control the lowest level labor organizations, successive Greek governments made certain that these first level organizations were not represented in the second and third tiers, and that demonstrably progovernment unions were over represented at these higher levels. Moreover, ‘untrustworthy’ individuals were denied union membership, radical first tier unions were not permitted to join second tier organizations, and bargaining rights were given only to the third tier, especially the progovernment GSEE. In short, state control of the labor movement was enhanced by government policies that intentionally encouraged the creation of a large number of small, weak unions and channeled compulsory union dues through the state-run organization, Ergatiki Estia, or Workers’ Hearth. Under this system of de facto corporatism, the Greek state firmly controlled unions, manipulating their government-imposed three tiered structure and depriving unions of any real bargaining role. Thus, as in the other three countries, workers in Greece were ‘incorporated’ into politics but at an extremely high cost: the loss of voice and influence. As elsewhere in Southern Europe, the development of social welfare provision in Greece was incremental, initiated by governing elites, rather than by labor, and usually a response to ‘critical’ historical events (the War of Independence, World War I, and the Asia Minor Catastrophe). Early provisions were largely funded by one of many self-help societies. For example, health care, the oldest of social welfare provisions in Greece, was introduced in the early years of the newly independent state as the aftermath of the War of Independence induced the government to respond to the medical needs of individuals who could not provide for themselves (Philalithis 1986: 145; Coudroglou 1985: 145). By the mid-1800s mutual aid organizations of a number of different professional groups also provided rudimentary social services to their members. By 1861, funds provided health care coverage to disabled veteran seamen and offered a variety of pensions, including those for retired sailors, persons injured on the job, and widows and orphans of sailors. A number of other funds were also created, including those of the merchant navy, miners, the military, and civil servants (Kremalis and Yfantopoulos 1992; Katrougalos 1996: 39–60). The creation of specific funds was not coincidental. As Philalithis (1986: 146) maintains, funds emerged to protect those ‘sectors of the community either with economic importance or with influence on

Social Policy, Democracy, and Citizenship

121

decision making’. World War I saw the emergence of rudimentary health care, financed totally by employer contributions. And the Asia Minor Catastrophe of 1922—in which 1.5 million refugees sought refuge in Greece—spurred the creation of funds for a variety of workers, including tobacco workers, typographers, health providers, bakers, lawyers, newspaper vendors, and clerical workers. Despite this, funds emerged slowly and coverage remained narrow. By 1932–3, seventyseven different insurance funds provided coverage for only 6.7 percent of the population (Yfantopoulos 1990: 221). As in the other Southern European countries, these programs were primarily funded from employee and employer contributions, with the state’s involvement remaining minimal. In 1937, in an attempt to win support from the working class, the dictatorship of General Metaxas created the Social Insurance Foundation (IKA)—the fund for private sector workers.30 George Katrougalos argues that with the creation of IKA, which provided generalized but not universal coverage, the foundations of the Greek welfare state were laid (Katrougalos 1996: 47). However, important adjustments were made in favor of certain politically powerful groups, including employers and occupational lobbies (Tsalikis 1967). Metaxas also introduced compulsory arbitration in labor disputes, collective bargaining, a minimum wage for all workers, and a moratorium on the debts of peasant farmers (Demertzis 1979: 111–12). In addition, secondary insurance programs—designed to work in conjunction with the existing primary insurance funds—were created in the 1930s and provided additional coverage for bakers, metal workers, cement workers, and fertilizer workers. In short, under the Metaxas regime, social welfare provision was expanded in an effort to appease and incorporate workers—but at a relatively low cost to the state as funding came primarily from employers and workers themselves (Petmezidou-Tsoulouvi 1992: 124–8). The post–World War II period was colored by the tragic events of the Greek Civil War (1946–9), an upheaval that dramatically influenced the provision of social welfare throughout the 1950s and 1960s, and brought the limited earlier expansion to a virtual halt. No sooner had the nationalist forces emerged victorious from that war than they established a restrictive parliamentary democracy whereby the losers were excluded from access to the state and its resources via a complicated system of legal and illegal restrictions. Laws were passed which outlawed the ‘subversive’ activities of left-wing trade unions, while certificates of ‘civic mindedness’ were necessary for employment in the public sector and, for a while, to enroll in an institution of higher education. In addition to exclusionary forms of state control, incorporative methods were also used to maintain the anticommunist state.

122

Social Policy, Democracy, and Citizenship

Patronage employment in the state bureaucracy, as well as the extension of public goods and services, were based on ideological criteria, and the ‘non-nationally minded’ were systematically excluded.31 Postwar efforts to expand social welfare quickly stagnated (Philalithis 1986). As the dominant conservative forces attempted to maintain their economic and political predominance, the medical care and health insurance sectors were unable to develop in the direction of universalism. Only those measures providing a minimum of medical services to the public and those reflecting the economic and political power of privileged, ‘nationally minded’ groups effectively operated. The left and center-left remained excluded and marginalized. Largely as a result, well over half of the population (composed primarily of agricultural workers) had meager insurance coverage at best during much of the postwar period. Moreover, administrative posts in the state bureaucracy were not created in order to satisfy real needs, but to reward the ‘nationally minded’ for their regime loyalty. George Katrougalos (1996: 48) argues that ‘this irrational development of the state’s structures resulted in an expansion of government expenditure, which almost doubled in twenty years without being accompanied by an increase in social expenditure as in most other Western countries’ (also see Petmesidou 1991: 31). Rapid socioeconomic modernization in the 1950s and 1960s did, however, begin to loosen the political controls of the post-civil war period. While urbanization weakened the right’s clientelistic networks in the countryside, progovernment forces increasingly lost control over the trade union movement, strike activities began to intensify, and, in 1958, the left-wing United Democratic Left (EDA) emerged as the official opposition party in parliament. In social policy as well, limited but important steps were taken. In 1961, the Organization for Agricultural Insurance (OGA) was created by the conservative government of Constantine Karamanlis to provide health, old-age, and death benefits for all agricultural workers. The scheme was quite unusual in that it was financed by the Greek state. However, it quickly atrophied as a result of insufficient economic resources (Katrougalos 1996: 51). Nonetheless, the creation of OGA was significant in that, with its inauguration, approximately 95 percent of the population were covered with some form of social insurance, despite the persistence of wide variations in the quality of insurance coverage from fund to fund, with OGA providing among the least generous benefits. In 1963, George Papandreou’s Center Union party displaced the Conservative ERE in government. Paralleling contemporaneous events in Italy, an ensuing popular mobilization forced Papandreou to loosen wage controls and to increase spending on social welfare and

Social Policy, Democracy, and Citizenship

123

education. However, the breakdown of Greece’s repressive post-civil war state and the emergence of dynamic and urbanized sectors of civil society with a greater voice in politics, had paradoxical effects. Rather than leading to increased democracy and government accountability, these developments triggered a reaction by Greece’s alarmed military which, in 1967, established military rule. For the next seven years Greece was ruled by the colonels’ junta (Petras 1992: 100–1). The authoritarian regime’s initial economic policy was characterized by strict control of the labor market, the prohibition of strikes, imposition of wage and price controls, and cuts in social welfare expenditures. Expenditures for education, health, and social services declined, while those for defense and public order almost doubled (Kafiris 1975: 40). As the regime wore on, however, social unrest began to grow as workers became restless, demanding economic reform and the restoration of democracy. As General Metaxas had done previously, the colonels attempted to ‘buy’ complacency by increasing wages and adopting expansionist monetary and fiscal policies (Papadopoulos 1991: 200–4). Thus, the regime can best be characterized as pragmatic, initially adopting restrictive economic policies but later giving way in order to bolster public support. It was only with the establishment of full democracy in Greece that significant social welfare change would take place.

The Social Welfare Impact of Democratization in Southern Europe As has been shown, universalistic, state-funded social policies were virtually lacking in Southern Europe prior to democratization. Because the Southern European social welfare state had developed incrementally over a long period of time, with deep roots in authoritarianism and restrictive democracy, it remained underdeveloped and highly particularistic. Universalism and social citizenship were nowhere to be found. In the absence of democracy in Southern Europe, welfare services, to the extent they were provided, were granted to important and privileged societal groups as a way of securing their loyalty and complacency. In this peculiar fashion, the state supported limited expansions of welfare prior to democratization but did so arbitrarily and without accountability. However, as Frank Castles persuasively argued in the previous chapter, this situation was dramatically altered by democratization. In all four countries, the emergence of full democracy produced dramatic effects, leading to a significant expansion and strengthening of the social welfare state.

124

Social Policy, Democracy, and Citizenship

In Italy, the most important moves toward social welfare expansion began quite early, reflecting that country’s earlier democratization process. In fact, as early as the 1960s, Italians were beginning to see the expansion of major programs, especially in health care. Highlighting the importance of democratic confrontation (largely absent in the other three countries) was the pressure exerted by trade unions and social movements. In the face of widespread social and political pressure, Italian political parties were given little choice but to address the concerns and demands of workers. As a result, welfare issues were incorporated into the parties’ electoral campaigns and public policy programs. Thus, while the size and scope of the welfare state lagged considerably in Greece, Portugal, and Spain, it underwent significant expansion and growth in democratic and developing Italy. Clearly, dramatic economic development took place in Italy in this period, permitting the expansion of social welfare, but we believe that the effects of democracy were paramount. In fact, Italy was not unique among the Southern European countries in experiencing dramatic economic growth in the 1960s. However, only in fully democratic Italy did the demands of trade unions and urban social movements actually produce effective social welfare reform. In authoritarian Spain, Portugal, and Greece, workers’ organizations remained weak and their demands ineffective. The connection between political democracy, union strength, and social welfare provision in Italy is clear. After experiencing relative difficulties in the 1950s, Italian trade unions were, by the 1970s, relatively stronger than their Southern European counterparts. By 1976, 51.2 percent of the Italian workforce was unionized, an increase of nearly eighteen percent since 1968 and a figure comparable to that of the UK (52 percent), and in excess of that of France (25 percent) and Germany (35 percent) (Coi 1979: 205). Despite the fact that union membership was uneven, however, with higher membership rates in the industrial regions of the North and the Center, Ferrera (1993) contends that trade union strength was nonetheless closely related to the concomitant expansion of welfare. A few illustrative examples attest to this. In 1952, the Italian pension system, which had been limited to industrial workers, was fully reorganized to incorporate new occupational groups, including small farmers (1957), fishermen (1958), small traders and artisans (1959). Health insurance was also reformed, with coverage being extended to retired public employees (1953) and later to all pensioners (1955). Under the center–left coalition of 1963–5, Italians were presented with a national social security system as well as with substantial pension reform, including the establishment of a social pension (fondo sociale) partly funded by the

Social Policy, Democracy, and Citizenship

125

state (Ferrera 1986). In the 1970s, pensions were further expanded (especially old-age pensions, which reached 11 percent of GDP by 1980) and health care increasingly universalized. Thus, a relatively modern social welfare system (albeit a hierarchical and uneven one, largely fragmented along occupational lines) was established, thanks in large part to democratic structures, and the consensus reached among political parties. From 1976 to 1979, the National Solidarity coalition between Catholics and the Italian Left furthered this development and brought with it some limited efforts to universalize welfare provision. Policies aimed at creating compulsory social insurance and overcoming the existing variation in entitlements across social and occupational categories were passed. Particularly important were reforms aimed at providing unemployment benefits to several occupational groups and to the young. Then, in 1978, the Workers’ Statute equalized the cost of living adjustment for all citizens, reformed education and public housing, unified the social security agencies into a national system, and established the National Health Service as well as the unique Cassa Integrazione Guadagni (Income-Support Fund).32 Despite such important efforts, however, social welfare provision in Italy retained the classic Southern European characteristics of particularism and fragmented corporatism. In contrast to the Italian case, the Spanish, Portuguese, and Greek states did not greatly expand social welfare provision during the 1950s and 1960s. In the absence of full representative democracy, their governments saw little need to accede to demands for improved social programs. Thus, it was only after their transitions to democracy that workers began to participate more effectively in the design and implementation of welfare as social pressure began to improve the scope and quality of welfare services. But again, in all three cases, this occurred in a limited, nonuniversalistic way. In Spain, the democratization of the 1970s ushered in a new era of social welfare reform. While franquist welfare structures actually outlived authoritarianism, considerable reform took place during and after Spain’s democratization. The social security system was restructured and expanded in 1978, unifying over twenty particularistic ‘special regimes’ into four major institutions: the National Institute of Public Health (INSALUD), the National Institute of Social Security (INSS) for income-maintenance policies, the National Institute of Social Services (INSERSO), and the National Institute of Employment (INEM). Moreover, the Real Decreto de Creacio´n de las Entidades Gestoras (Decree on the Creation of Management Entities) transformed the old social security structures into six management entities. Each of these reforms represented

126

Social Policy, Democracy, and Citizenship

efforts to rationalize the organization of social services and to broaden their coverage. Laparra and Aguilar (1996: 97) maintain that this was partially a consequence of the ‘obvious need for building the legitimacy of the new democratic governments’. However, social security continued to be based on workers’ and employers’ contributions, and the traditional system of unequal benefits remained intact. Pensions remained closely linked to years of service, the amount of one’s contributions, and occupational status. It is important to note that, despite such important reforms, politics was given priority over economics during Spain’s transition to democracy as the government attempted to secure social harmony to ensure the success of democratization. It was thus only with the socialist Partido Socialista Obrero Espan˜o (PSOE) accession to power in 1982 that social welfare provision became a more explicit government objective. Based on the 1978 Constitution which stated that social security should reach all Spaniards, the independent system of social assistance that had been developed in the 1960s was partially abandoned. Today a series of means-tested benefits exists aimed at ensuring all citizens with a minimum monthly income. However, the benefits are not well integrated, frequently overlap, and are incoherently designed and managed (Aguilar, Laparra, and Gaviria 1996). As a result, different categories of people continue to receive widely varying levels of benefits. The Portuguese case was broadly similar. Despite incremental reform under authoritarianism, major transformations in the social welfare state did not take place until the April 1974 Revolution, when the military began to dismantle corporatist organizations and to create an integrated system of social security. The democratic constitution of 1976 also broke with corporative principles, placing responsibility for the organization and coordination of social security on the state. It stipulated, for example, that health care was a social right of all citizens and an official state responsibility. Moreover, given the fact that social security was henceforth conceptualized as a public service to be provided by the state, both the Caixas and the previdencia social (social assistance) systems were replaced. Whereas only 50 percent of the population was covered by the public health system in the 1970s, by the end of the decade coverage was nearly universal. Despite these positive reforms, however, Portugal followed a Bismarckian model. To this day, a high proportion of benefits are still derived from social contributions rather than the state. As a result, a high degree of occupational and regional fragmentation persists (Santos 1990). When the transition to democracy occurred in Greece after seven years of authoritarianism, its social welfare state grew more gradually than did those of Spain and Portugal. As Frank Castles shows in this

Social Policy, Democracy, and Citizenship

127

volume this was partly because the Greek social welfare state was already relatively more developed than that of Portugal and Spain. In addition, the Greek transition process was heavily elite driven, dominated by the conservative Nea Demokratia, and basically nonconsensual: workers did not engage in negotiations with the state or employers. Nevertheless, a number of important symbolic gestures were made by the conservatives in government, indicating, if nothing else, a changed attitude toward the legitimacy of the social welfare state. Specifically, while Constantine Karamanlis, the prime minister who led Greece’s transition to democracy, imitated Adolfo Sua´rez in emphasizing the political, rather than the social, aspects of democratization, and despite the fact that the social welfare state was not at the core of his party’s political program, the government nevertheless presented itself as a modern liberal democratic party committed to the standards of ‘social democracy, political freedom and social justice’ (Nea Demokratia 1974; also see Thomadakis 1994). More importantly, the party maintained that ‘political freedom [could] only be guaranteed if there [was] a fair sharing of national income’ (Nea Demokratia 1974). Twenty-two constitutional articles on individual and social rights were included in the 1975 Constitution (Articles 425), including the important Article 21, which codified the notion that ‘The State shall care for the health of citizens and shall adopt special measures for the protection of youth, old age, disability and for the relief of the needy’. Moreover, the right to work for all citizens and of equal remuneration for both men and women were recognized in Article 22. The progressivity of these positions reflects a public recognition that the state was now perceived as playing a legitimate role in the provision of social welfare. A number of concrete policy reforms were also implemented. Between 1974 and 1977, modest tax reform transferred 45 billion drachmas from the upper to the lower classes, and farmers’ pensions were increased by 150 percent (Loulis 1988: 191, 66). Collective bargaining (which had been suspended under the dictatorship) was restored, and price controls were abolished. During this period real wages made considerable gains while working hours were reduced (Spourdalakis 1988: 133). A number of nationalizations also took place, as Karamanlis nationalized both Olympic Airways and a majority stake in the Stratis Andreadis banking empire. The state also took over urban transportation, water supply, and sewage companies. This was such a departure from the past conservative public policy that one Athenian daily asserted in the early period of transition that the government’s ‘basic aim’ was ‘social and economic democracy’ (quoted in Katsoudas 1991: 5). The second-term Nea Demokratia government even took the first steps toward reforming health care provision in Greece, introducing a bill

128

Social Policy, Democracy, and Citizenship

calling for the establishment of a National Health Council to plan health services, the reorganization of hospitals, and the creation of Health Regions and Health Centers to provide primary health care in rural areas. This bill was rejected however by both the far-right and left—by the right because it allegedly failed to safeguard the privileges of the medical establishment, and by the left because it did not establish a single unified, public health service (Philalithis 1986: 150–1).

The Impact of Ideology: Socialism in Spain, Portugal, and Greece As we have argued thus far, in all four postauthoritarian countries, both public pronouncements as well as concrete policies appear to support the notion that ‘democracy matters’ in the configuration of social welfare programs. After democratization, constitutional articles acknowledged the proper role of the state in providing public assistance, political parties made pronouncements indicating their embrace of the social welfare state, and governments (even those of the right) took concrete steps toward expanding the scope and coverage of social welfare. However, the notion that ideology ‘matters’ must also be stressed here. The accession to power of socialist governments in Spain, Portugal, and Greece in the 1980s signaled a decisive shift toward generous, increasingly universal provision. Even as Italy in the 1980s began to rationalize its social policies, with a strong emphasis on the need to cut social spending, the socialist administrations of the other three Southern European countries went in the opposite direction. As left-wing governments in all three countries felt the need to respond to voter demands in the field of social welfare, the 1980s saw dramatic growth in the provision of social welfare benefits despite the constraining influence of economic downturn.33 In Spain, the socialist PSOE came to power in 1982, emphasizing that it would put the constitutionally guaranteed rights to social welfare into practice. To this end, several new welfare policies were introduced and important institutional restructuring took place. The effective participation of trade union representatives in the social security institutions (which had been guaranteed in the National Employment Agreement of 1981) was finally put into effect. Furthermore, the number of pensioners grew dramatically during the decade, with disability pensions amounting to one-fourth of all pensions. The paralyzing general strike of December 1988 led to the indexation of pensions to the cost of living and increased the number of unemployed who were covered from 28 to 40 percent of the total. Perhaps the PSOE’s

Social Policy, Democracy, and Citizenship

129

greatest contribution to social welfare was to make health care coverage much more universal. In 1986, the PSOE established a National Health System of decentralized, free, and public health care, operating almost entirely on the basis of tax revenue (95 percent in 1996) that has been operating since about 1989.34 In contrast with Spain, where high unemployment was the greatest concern of government in the 1980s, in Portugal, the government benefited from full employment, thanks in large part to the maintenance of low salaries, a relatively large proportion of people still involved in agriculture, as well as the expansion of the industrial and service sectors. In 1981, however, efforts were made to make the pension system appear more universal by creating the National Pensions Center which was to take the place of the caixa. Accordingly, the number of pensions grew by approximately 40 percent during the 1980s. In 1982, the health system was also fully decentralized to the regions and was further reformed in 1984 (Mozzicafreddo 1992). However, as in the other Southern European countries, social security continues to be based even today on social contributions. Thus, while in the mid-1980s efforts were made to expand the state’s transfer payments to include social services not covered by contributions, 91.2 percent of social security funding is still derived from employer and employee contributions. Overall, benefits have been gained by employed workers rather than citizens as a whole (Guibentif 1996). Nonworkers continue to be excluded. In Greece, the 1980s brought to power that country’s first socialist government, led by the PASOK, and with it promises of greater universalization. Its strong electoral appeal had roots in popular demands for social justice, effective reductions in inflation and unemployment, and improved social services (Thomadakis and Seremetis 1992: 204). On coming to office, the party’s social and economic policies were designed to benefit the middle and lower strata of Greek society: PASOK legislated the establishment of public health and social security systems and granted substantial increases in wages and salaries while indexing pensions and salaries. In addition, the party passed the Democratization of Trade Unions Act which provided better union protection, introduced proportional representation in union elections, reformed the method by which union subscriptions were collected, and made strikes easier to call. PASOK also proposed to ‘socialize’ large sectors of the economy, claiming that this would make corporations more responsive to the needs of consumers and their employees. Continuing in this reformist spirit, in 1983, it proposed a plan for a National Health System—one that would equalize benefits to all citizens, irrespective of their economic, social, and occupational status. Passed into law in the fall of 1983, this attempt at universalization was not

130

Social Policy, Democracy, and Citizenship

fully implemented. A substantial reform of the Greek National Health system was attempted by the Simitis (PASOK) government in 1997 and again during its second term in office (2000–4). The financing of the system was streamlined, while public hospitals obtained new management, hired from the private sector. Overall, this reform met with resistance from the unions of public-hospital doctors and nurses, and was not fully implemented. More successful was the decentralization of health services, as the national health system was divided into seventeen regional health systems. Changes to the social security system were made however. Health, old-age, and disability insurance was provided to individuals and their dependents with no prior work history or stable employment—individuals who, in the past, would have had to provide for themselves, rely on the family unit or depend on private charity at the local level. These people were now covered by OGA, albeit the least generous of all social insurance funds.35 In short, we see the impact of ideology again. With the accession to power of Greece’s first socialist government, there was, for the first time, the explicit acknowledgment that universal social service provision was a government target. While the achievement of full social citizenship is still unrealized in Greece and, with the possible exception of health care provision, throughout Southern Europe, we clearly see the importance first of democratization and second of partisan ideology in moving the Southern European social welfare state toward more universalistic models of welfare provision. In the 1990s, however, the impact of ideology (both socialist and conservative) on the social welfare state has weakened as Southern European governments have had to deal with economic problems, including in some cases unprecedented levels of unemployment as well as structural reforms made necessary by EMU criteria. Much of this has moved southern governments, both conservative and socialist, toward ‘rationalization’ of the welfare state, including a heavier reliance on means-testing of entitlements. Thus, in Spain, for example, family allowances, which were universal (albeit inadequately low until 1990), are today increased but means-tested (Flaquer 2000: 18). Increased financial pressure on the Greek social welfare state led PASOK to rationalize benefits there as well. Family benefits are means-tested, public childcare centers give priority to low-income families, and family cash transfers are increasingly being replaced with care services partly provided by the state and partly by the voluntary sector (Karantinos, Ioannou, and Cavounidis 1992; Carlos 2000: 59). In short, even socialist governments in Southern Europe that traditionally appeared committed to universal social citizenship rights have in more recent times abandoned both rhetoric and practice.

Social Policy, Democracy, and Citizenship

131

Conclusion This study has contextualized the development of the Southern European social welfare state within the specific cultural, political, and economic milieus of the four Southern European countries, highlighting the traditional political and cultural vestiges that have structured social programs in this part of the world. First, our analysis illustrated that even though democratization was instrumental in extending the coverage of Southern European welfare, many of the welfare structures found in Southern Europe today actually preceded democratization and were constructed without the political participation of civil society. Political parties and secondary associations also remained organizationally weak in Southern Europe, failing to develop strong, healthy links to civil society. Unlike in Northern Europe, where labor movements and other secondary associations articulated workers’ interests and demands, in the south, they played a very circumscribed role, both in the design and in the implementation of welfare policies. Thus, even though the social welfare states of all four countries developed incrementally over a long period of time and therefore had roots that predated democratization, we showed how, under long periods of authoritarianism and restricted democracy, the Southern European social welfare state remained underdeveloped and highly particularistic. Workers and other groups in civil society lacked the full capacity to organize politically so as to exercise pressure on the state, either forcing or coaxing it to fulfill its protective social welfare function. Welfare services were granted as a way to maintain consensus and assure the loyalty and complacency of workers to the state. In this peculiar fashion, the state supported welfare prior to democratization but did so without any sort of democratic accountability. As a result, social welfare reforms were generally initiated by unaccountable governing elites, who often failed to respond to the real needs of the populace. Thus, based on this historical trajectory, social welfare in Southern Europe remains fragmented and poorly developed even today. We have also shown that, in the absence of well-developed participatory, representative workers’ associations, clientelism became the method by which Southern European governments maintained consensus and the support of key societal sectors. Not surprisingly, one mechanism by which such consensus was maintained and support purchased was through the social welfare state and its benefits. As we have seen the particularistic nature of the Southern European welfare state is maintained even today through the persistence of patron-client relations. Another distinguishing characteristic of the

132

Social Policy, Democracy, and Citizenship

development of the Southern European welfare state has been the familistic and gendered provision of social welfare. As a result of poor social welfare development, Southern Europeans who are already culturally inclined to value strong family ties and large extended family networks, are compelled to construct family strategies of welfare provision, leaving citizens without families to rely on religious institutions (in the Catholic countries) and mutual-assistance societies. This too has left its imprint on the Southern European welfare state. Individuals in need of protection often remain reliant on family rather than state provision of welfare. Thus, in Southern Europe, to the extent that an institution has historically improved the life-chances of individuals, making workers more autonomous and independent from their employers and the vagaries of the market, that institution has been the family, rather than the state. Over the last several decades, however, a rapid process of political, economic, and social modernization, accelerated by membership in the EU, has pushed these societies toward convergence with Northern Europe—demographically, economically, and culturally, including an approximation of their social welfare states to those of the North. Political change has proven to be a critical factor. During the 1970s, the three Southern European countries with authoritarian modes of governance (Greece, Portugal, and Spain) experienced a transition to parliamentary democracy. One consequence of this transition, and the popular political participation it allowed, has been the weakening of political polarization and the rise of social democratic parties as structuring political agents. Complementing this shift has been the modernization and moderation of conservative parties, which have moved toward the center of the political spectrum, thereby winning elections and entering government. A crucial element of such change in Southern Europe has been the inclusion of welfare issues onto the political agenda. Political parties, both conservative and socialist, committed themselves (at least rhetorically) to implementing social welfare reform and expansion. Thus, one of the main arguments of this chapter has been that, first democracy, then partisan ideology have mattered considerably in the development of the Southern European social welfare state. However, despite the fact that democratic institutions and practices have been established in these countries, and Southern European social policies have approximated in some ways those of Northern Europe, certain characteristics of the Southern European welfare state indicate that while democratic political structures allow for the incorporation of citizen demands, traditional particularistic practices do not change easily or quickly. Even under full democracy, the organization and membership of citizens in labor movements and secondary

Social Policy, Democracy, and Citizenship

133

associations has not been fully successful. While playing a greater and more active role, labor has still found it difficult and, often not entirely in its interests, to alter the rigid institutional structures which have traditionally shaped the welfare state in this part of the world. As a result, labor-market rigidities, the clientelist nature of the welfare state, the fragmentation of benefits, as well as the role of the family as a provider of social services remain. Features that benefit ‘privileged’ core workers—and through extension their families—continue to persist, while reforms likely to benefit ‘outsiders’ are ignored. Finally, we must emphasize that structural transformations necessitated by unprecedented levels of unemployment in Southern Europe and the need to bring southern economies in line with the rest of Europe in the face of EMU convergence, have pushed Southern European governments, both conservative and socialist, to enter an era of economic discipline in the late 1980s and 1990s. To this end, an important cornerstone of such government policy has been economic restructuring. In Greece, firms are increasingly being forced to rely on their own efforts and efficiency gains rather than on generous government subsidies and fiscal exemptions (OECD 1992c: 34, 82). Restructuring in the manufacturing sector led industry to shed some 21 percent of its employees between 1981 and 1995 (OECD 1996: 56). Furthermore, the last decade has seen successive Greek governments implement a wide-ranging privatization program. From 1998 to 2001, for instance, the state partially or fully privatized over 30 formerly-state-owned corporations worth 11.8 billion Euros. This restructuring agenda has been far-reaching, including such icons of the Greek public sector as Olympic Airlines, Hellenic Petroleum (EP), the Hellenic Telecommunications Organization (OTE) and the Public Power Corporation (DEI) (Wright and Pagoulatos 2001: 259–62, 269; OECD 2002: 118–21). Therefore, restructuring efforts continue to have widespread consequences for the labor force: many workers have already lost their jobs and an increasing number— including ‘core’ workers in the formal economy—are in danger as further job losses are likely. In Spain, too, labor-market reforms were seen as vital by the PSOE and, more recently, by the conservative Popular Party (PP) as the way to improve the country’s competitiveness within an expanded EU. As a result, Prime Minister Jose´ Marı´a Aznar pushed for reforms that would reduce the cost of dismissals and allow for greater flexibility of hours for part-time contracts. Based on the government’s new regulations, the amount of compensation paid to dismissed workers will go from forty-five to thirty-three days for each year worked, and the maximum allowance is reduced from forty-two to twenty-four months of pay, and is due to be reduced further, to twelve months. Moreover, employers have been given greater freedom in setting

134

Social Policy, Democracy, and Citizenship

part-time employees’ shifts and hours, with greater labor flexibility in companies, greater geographical mobility of the workforce, and an increased willingness to change job description and variable hours of work and wages (Short 2001). In sum, Southern European social welfare reveals itself to be a product of a wide variety of competing influences: traditional practices from predemocratic periods; clientelist and corporatist legacies; the impact of democratization and the ‘opening up’ of politics; as well as the notentirely-successful attempts by social democratic parties to universalize social welfare provision. As a result, social citizenship has yet to be achieved in Southern Europe. Even today, we see that access of women and youth to social benefits continues to be problematic because of the close connection between these benefits and employment in the formal, paid economy. Furthermore, as this chapter has argued, compared to their counterparts in other European countries, Southern European women are not only less involved in paid work but are also occupied by unpaid care-work. This serves not only the family, but also indirectly the state, allowing it to take a back seat in the provision of social welfare. Thus, a circularity has been created between a lack of universal coverage, on the one hand, and the absence of demand for such coverage, on the other. As Flaquer (2000: 26) argues: As in a sort of self-fulfilling prophecy, the outcome . . . is that difficulties faced by individual families are not tackled via public mobilization but through people’s private strategies. This creates a kind of negative feedback situation in which the system is reproduced and even reinforced. In Southern European countries, family solidarity both explains and is the result of an undeveloped family policy.

Finally, it is important to note that placing the responsibility of social welfare provision on the family is largely possible because both social welfare provision and the existing labor market in Southern Europe are ‘gendered’. As we have illustrated in this chapter, Southern European women do not generally have equal access to employment in the formal labor market, and consequently lack access to social welfare benefits, while being the main providers of welfare within the family. Because gender relations throughout the region are institutionalized and reflected in the three institutions discussed in this chapter—the welfare state, the labor market and the family–women’s social rights— indeed their very citizenship, has lagged in Southern Europe (see Sundstro¨m 2000: 202). Despite moves to modernity, traditional cultural, political, and institutional influences persist in the south. Thus, the safest conclusion to reach is that the Southern European social welfare state continues to be characterized by features that are uniquely and decisively its own. But, perhaps, not for long.

Social Policy, Democracy, and Citizenship

135

Notes 1. We would like to thank P. Nikiforos Diamandouros, Richard Gunther, Gianfranco Pasquino, and Jonathan Swarts for reading and commenting on earlier drafts of this chapter and especially for their assistance and advice during the editing process of this project. Marisol Garcı´a is very grateful to Manuel de Lerena, Pierre Guibentif, and Manuel Villaverde Cabral for giving some of their time and knowledge in our research on Portugal, and to the Social Science Research Council for providing a travel grant to support this field work. All errors and omissions are, of course, our own. 2. This model has been an ideal-type for the other European regimes as long as the Fordist system of production and accumulation has been predominant. However, at present, even in Scandinavian countries targeted social policies are being introduced. 3. Leibfried 1992; Saraceno 1994; Castles 1995: 291–313; Ferrera, 1996; Gough 1996: 1–23; Rhodes 1996: 1–22; Moreno 1996: 1997; Trifiletti 1999. 4. See, for example, Hadjmichalis and Vaiou 1990a: 1–24; Hadjmichalis and Vaiou 1990b: 79–106; Chletsos 1993; Kostaki and Ioakeimoglou 1997: 185–200; and Ioakeimoglou, Kaminioti, Kostaki, and Skopelitou 1998. 5. See, for example, the OECD Economic Surveys for Greece (1998), Italy (1999a), and Portugal (1999b). 6. In 1988, the unions forced the government to retreat from its plan to increase youth employment by reducing employers’ social security contributions and fixing payment at the level of the statutory minimum wage. In May and October 1992, general strikes were organized to prevent the reduction in unemployment benefits aimed at increasing the incentives to search for employment. Finally, in January 1994, a general strike took place in protest at what the unions called an ‘unprecedented deregulation of industrial relations’ (Richards and Garcı´a Polavieja 1997; Rhodes 1997). 7. Sources: For Italy, Blades 1982: 34–6; for Greece, Kanellopoulos 1992; for Spain, The Economist 1987: 22; and for Portugal, OECD 1986a: 79. 8. It is important to note, however, that many in Southern Europe who hold jobs in the informal economy do so as a form of secondary employment (moonlighting). In cases where workers also hold permanent jobs in the formal economy, they receive social welfare protection from their formal employment status. On moonlighting, see Mingione 2001. 9. For a discussion of the Southern European dualistic welfare regime, see Ferrera 1996: 17–37; 1997: 13–24; Moreno 1997; Pringle 1998; and Trifiletti 1998. 10. The Greek figure is for 1995. See, also, Papadopoulos 1997; Matsaganis 2000; and Symeonidou 1997: 337–62. 11. In contrast, Greece has no guaranteed minimum income. See, Symeonidou 1996: 81; and Saraceno 2002. For a Southern European comparison of minimum income policies, see Matsaganis et al. 2003. 12. On local schemes with discretionary levels of provision, see Symeonidou 1997; Trifiletti 1998; and Flaquer 2000.

136

Social Policy, Democracy, and Citizenship

13. This is despite the fact that Greece has not undergone effective devolution as has Italy and Spain. 14. As we see below, in both Portugal and Spain such practices have been less widespread and, perhaps as a result, very few studies of clientelism there have emerged. One possible explanation for this lack of clientelism in the Iberian Peninsula may be the absence of political party competition for extended periods of time during the twentieth century. We argue below that authoritarianism probably made clientelism unnecessary there as elites were not compelled to institutionalize such practices. 15. The Enti publicci of the welfare state have their origins in the Fascist period. These institutions comprise the Intituto Nazionale della Previdenza Sociale for old age, invalidity, unemployment, family allowances; Instituto Nazionale per le Assicurazione contro gli Infortuni sul Lavoro for invalidity, and the ´ ssicurazione contro le Malattie for sickness. Instituto Nazionale per l A 16. Lyrintzis 1984; Spourdalakis 1988; Koulouglou 1987. 17. Expenditures on pensions (as a percent of GDP) are the only category of social expenditures for which Greece exceeds the OECD average. On this, see Provopoulos and Tinios 1993. 18. On ‘compulsory altruism,’ see Land and Rose 1985: 74. 19. For further analysis on the relation between family care and the role of women, see Comas 1994. 20. The EU12 average is 39 percent; Gough 1996: 14. 21. In 1992, there were fourteen state nurseries (for 8 month to 2.5 year olds) and thirty-five nursery sections within kindergartens (for those between two and one-half and five years of age) in the entire country. (Source: Karantinos, Ioannou, and Cavounidis 1992: 46, quoted in Sims-Shouten 2000: 278.) 22. On Greece, see Pantelidou-Malouta 1988; Stratigiki and Vaiou 1994: 1221–34. 23. For Italy and Spain, see Guerrero and Naldini 1996: 42–66. 24. On Italy, see Mingione and Pugliese 1995: 118–44; and Menniti, Rossella, and Laura 1997. On Spain, see Guerraro 1999; Guerraro and Naldini 1996: 42–66. 25. There were, however, variations on this general pattern from location to location. 26. It should be noted that divorce rates are not the best indicator of break-up in Spain and Italy, where a significant number of separated couples do not have access to divorce. 27. In 1995, live births outside marriage were 3 percent in Greece, 8 percent in Italy, 11 percent in Spain, and 19 percent in Portugal, compared to the EU average of 23 percent (Eurostat data, Quoted in Flaquer 1998: 29). 28. In a forthcoming publication, Karakatsanis and Swarts argue that, in the absence of a universal social welfare state and as increasing numbers of women leave the domestic sphere for professional employment, a significant demand for migrant women to perform social welfare functions has been created in Greece.

Social Policy, Democracy, and Citizenship

137

29. The underfunding of social insurance programs is obvious from the fact that contributions to the systems were calculated only on the basis of basic wages, not on the real level of workers’ incomes. 30. A single Social Insurance Agency for all workers in the private sector had been adopted into law in 1934 by the liberal Venizelos government. The law failed to be implemented as the Venizelos government collapsed. The conservative government which succeeded Venizelos adopted a law in 1934 creating IKA, however it was not inaugurated until 1937. 31. Vergopoulos 1978; Mouzelis 1986; Tsoukalas 1987; Maloutas and Oikonomou 1988. 32. This was a state fund which aimed at preventing redundant workers from being dismissed. Assuming that their status was temporary, workers remained at home but continued to receive a reduced wage from the Social Security agency (see Mingione 1998: 17). 33. According to Eurostat public spending on social protection increased in the four countries at a higher rate than the EU average. Social protection by inhabitant grew in the 1980s: 3.3 times in Greece; 2.6 times in Italy; 2.5 times in Spain and Portugal. The average for EU was 2.2 times higher than the previous decade. 34. In 1982, about 30 percent of all hospital beds were managed by the Catholic Church. By 1989, the Health System was practically universalized and financed mostly through income taxes (70 percent). The seven Autonomous Communities which have health protection transfers are Catalonia, Valencia, Andalusia, Basque Country, Galicia, Navarra, and the Canary Islands. Most of these regions are located in the periphery of the Peninsula and have a long tradition of cultural distinctiveness. In the other ten Autonomous Communities, the INSALUD (National Institute of Public Health) was put in charge of public health care (see Rico 1996: 128–30). 35. While this was an important reform—indeed, the only one of its kind in Greece—it should be noted that it did not substantially increase the number of new beneficiaries as approximately 95 percent of the population was already covered.

................ 4 ................

Democratic Consolidation, Judicial Reform, and the Judicialization of Politics in Southern Europe Pedro C. Magalha˜es, Carlo Guarnieri, and Yorgos Kaminis1

Political scientists in general and students of democratization in particular have been increasingly interested in the study of institutions. This interest has coincided with an overall shift in democratization studies, from a focus on the social and structural prerequisites of democracy to a focus on concrete political actors and their choices in transitional contexts. Among the kinds of choices that have received the attention of scholars are those involving core political institutions, like electoral systems or the rules regulating executive-legislative relations. As the editors of this volume imply in the introductory chapter, large-scale processes—such as modernization, democratization, and Europeanization—and past institutional arrangements—such as structures and legacies inherited from past political regimes, may constrain (but do not determine) future policy choices. A number of works produced in the last decade have persuasively argued that those choices make a difference in the prospects for democratic consolidation and stability.2 However, the causes and consequences of the creation of other types of political institutions have been widely neglected, even in regions like Southern Europe, whose processes of democratization have been exhaustively studied. One of those neglected institutions is the judiciary. Depicting the judiciary as being a ‘political institution’ may still raise a few skeptical eyebrows. The reluctance to view courts and judicial decisions as being at least partially political in nature has prevailed until recently in comparative studies, and helps to explain the neglect

Democratic Consolidation and Judicial Reform

139

of the issue of judicial reform in most empirical and theoretical accounts of democratic transition and consolidation. However, there are several good reasons to believe that a full understanding of how countries develop democratic institutions requires greater attention to the process of judicial reform. First, it has been persuasively argued that one of the indispensable features of consolidated democracies is the existence of a functioning state apparatus, usable by democratic governments and organized according to rational-legal norms (Linz and Stepan 1996: 10–11). In other words, a necessary (although not sufficient) requirement of a consolidated democracy is the limitation of political power through stable, coherent, and enforceable legal rules,that is, the ‘rule by law’.3 This requirement clearly involves the judiciary. One of the tasks of democratic consolidation consists in creating the institutions and fostering the norms that, among other things, allow courts and judges to adjudicate cases according to generally recognized procedures, using stable legal codes and/or precedents, and reach decisions in an impartial, efficient, and nonarbi trary manner. For students of democratization, this raises the immediate questions of the extent to which judges that served non-democratic regimes are able to become part of this ‘usable and functioning’ democratic state apparatus, and of how political and instiutional changes may affect that ability. Second, and in spite of the controversies found in democratic theory in this respect, even the staunchest critics of judicial power are ready to concede that some form of judicial review is required in order to assure the protection of the primary political rights that allow for the free formulation of preferences in democracies.4 Furthermore, majority rule must arguably be counterbalanced by equipping democracies with self-binding mechanisms that protect them from future majority ‘passions’ or from the unwillingness to stick to prior policy commitments and institutional arrangements (Elster 1993: 189–93; Majone 1996). In other words, democracies require the legal culture and institutions of constitutionalism and the ‘rule of law’. This, in turn, implies the submission of ordinary legislation to a higher law beyond the reach of simple majorities, the attribution of constitutional interpretation and adjudication to an independent judiciary, and the support of these principles and institutions by a strong legal culture in civil society (Linz and Stepan 1996: 10). Finally, modern polities in general (and the Southern European cases in particular) have undergone changes that do not allow us to further disregard the political relevance of courts. Decades of research of the United States judiciary have established not only how political and ideological factors influence judicial decision-making, but also how

140

Democratic Consolidation and Judicial Reform

those decisions have an impact on the social allocation of resources and the functioning of other political institutions. In contrast, the still prevalent treatment of the continental European civil law judiciaries depicts them as mostly bureaucratic organizations, composed by politically irrelevant judges whose role can be summarized in Montesquieu’s words: to be the ‘mouth that speaks the words of the law’. As it happens, this depiction is inconsistent with the facts. On the one hand, the merely bureaucratic role of judges in civil law systems has often been more an element of the judges’ own cultural and professional selfpresentation than of political reality. This has led to the unwillingness to recognize their political role in the past even in its most blatant manifestations, as in the well-documented case of the Weimar Republic (Kirchheimer 1961; Ehrmann 1987: 75–92). On the other hand, the political significance of judiciaries in consolidated democracies appears to be on the rise. An important body of literature has emerged since the late 1980s dealing with the so-called ‘judicialization of politics’, a phenomenon broadly defined as ‘the expansion of the province of the courts and judges at the expenses of the politicians and/or the administrators’ (Vallinder 1995: 13). That expansion that has occurred both through the impact of judicial review on the policy-making process5 and through what has already been called the ‘criminalization of political responsibility’, of which the ‘Clean Hands’ operation in Italy stands as the major example (Guarnieri 1997). This chapter has three different but interrelated objectives. The first is to look at the recent past of the Southern European democracies, analyzing the process of judicial reform as it unfolded during these countries’ democratic transitions and accounting for the political factors behind the adoption of particular judicial institutions. In line with the approach presented in the introduction to this volume, we hypothesize that not only historical legacies, but also general strategies and specific choices of political elites influence the institutional development of the justice system. We argue that the outcomes of judicial reform in Italy, Greece, Portugal, and Spain were the result of two major factors: the legacies of the nondemocratic past in Southern Europe and the self-interested strategies of political actors involved in processes of institutional design. Our second objective is to estimate the extent to which new judicial institutions have contributed to democratic consolidation in Southern Europe. Two aspects of that contribution are discussed. On the one hand, the role of constitutional courts in the protection of political rights and, as a source of important cultural changes inside the judiciary, diffusing principles of constitutionalism and fostering the adoption of modern legal doctrines to which judges socialized under authoritarian regimes were not particularly sensitive

Democratic Consolidation and Judicial Reform

141

at the time. On the other hand, we analyze the role of judicial review as an ‘arbitral’ mechanism, and its contribution to the completion of transitional pacts that had been left open in what concerned the most complex and divisive issues facing the new Iberian democracies. Finally, our third objective is to assess the contemporary role of these countries’ judiciaries and its peculiarities, discussing the wave of judicial activism in Italy in the early 1990s, its effects on the political system, and the extent to which the Greek, Portuguese, and Spanish cases have replicated that experience.

The Politics of Judicial Reform in Southern Europe The extant literature on democratization and institutional change tends to be divided in two major perspectives.6 The first highlights the causal role of the social, cultural, and institutional legacies from the preceding nondemocratic regimes—legacies that are believed to constitute impeding or facilitating factors for the adoption and stability of particular institutional arrangements.7 In this view, the legacies of the nondemocratic past (visible in civil and political society, the economy, the rule of law, and the state apparatus) influence both the complexity of the tasks that new e´lites have to undertake to build democratic polities and the likelihood of their success. The second perspective emphasizes the autonomy of the new political actors, their incentives and interests, and the intentional design and crafting of new democratic institutions. In this view, institutional change is conceived as the result of bargaining between self-interested political actors, interested in choosing the rules of the game that will benefit them the most.8 How do these alternative explanations fare against the evidence from the processes of democratic judicial reform in Southern Europe? Should they be conceived as alternative and exclusive explanations, or rather as complementary approaches to judicial reform? These are the theoretical questions addressed in this first section.

Authoritarian Legacies and Judicial Reform The so-called civil law (or Roman–Germanic law) tradition results from a number of historical experiences, legacies or influences shared by most continental European nations: the revival of the Roman-Justinian code in Medieval and Renaissance Europe by legal scholars; the liberal revolutions and the transfer of sovereignty from the ‘King’ to the ‘Nation’ (represented in parliament); and the Napoleonic code and its

142

Democratic Consolidation and Judicial Reform

spread throughout continental Europe. There are relevant differences between nations and families of nations within the so-called civil law tradition, and it is even debatable whether the actual functioning of their legal systems ever met the civil law ideal-type (Shapiro 1981: 126–56). However, that ideal-type remains influential in the minds of legal scholars and political and judicial actors, and some of its features have undeniable empirical relevance for the distinction between civil law and common law countries. First, civil law systems share the notion that written law and preexisting legal rules predominate over custom, the judicial discovery or creation of law or even judicial precedent. Second, the interpretation of those rules is to be aided by doctrine produced by legal scholars and professors. The combination of the will of the parliament—the sole producer of legal rules—and the ‘science of law’ produced in universities becomes then the main source of the law, and judges are left no other role but to mechanically apply rules and doctrines to particular cases (Gilissen 1979). This basic model had important consequences in terms of the organization of civil law judiciaries, turning them into something much closer to a state bureaucracy than to an actual third branch of power. These features are historically well exemplified by the Southern European judiciaries. Applicants for the judiciary were—and still are— selected on the basis of their general knowledge of several branches of the law, as tested by written and oral exams and guaranteed by a university degree in law. Professional training and experience were acquired inside the judicial organization, starting from the bottom of a pyramid-like hierarchy, since the candidates for the judiciary were encouraged to enter the competition soon after graduation. The organizational units of the judiciary, as well as its organizational roles, were ordered according to the principle of hierarchy. Promotions to higher positions were granted by hierarchical superiors enjoying great discretion, according to traditional criteria as seniority and merit, with the government playing an important role in the appointments to the highest positions. Finally, the approach to work performance and role assignment was—and still is—of a ‘generalistic’ type, where members of the judiciary were supposed to be able to perform indifferently all organizational roles formally associated with their rank (Merryman 1969; Di Federico 1976). Considering these institutional features of civil law judiciaries, it should not come as a surprise that, once established, the twentieth century Southern European authoritarian regimes did very little to change them. In Italy, due to the rather wide powers traditionally enjoyed by the executive toward the judiciary, the Fascist regime

Democratic Consolidation and Judicial Reform

143

made remarkably small efforts to change the latter’s institutional setting. The process of recruitment remained basically bureaucratic, while the powers of the top ranks of the judicial hierarchy and of the Minister of Justice were moderately increased. In Greece, the colonels’ regime also maintained promotions, appointments, transfers, and disciplinary actions as responsibility of politically appointed ‘judicial councils’, as had been the case before the dictatorship. While the Greek Constitution was altered to entrust the selection of the presidents and vice-presidents of the higher courts to the executive, here as well the regime limited itself simply to constitutionalize the predictatorship legislation. However, institutionally speaking, the placement of directly or indirectly politically appointed ‘judicial councils’ on the top of the organizational structure of the judiciary and the absorption of most of their powers by the government did not represent dramatic breaks with past liberal regimes. Political control of the judiciary was backed by a system of internal evaluation of judicial performance controlled by the higher ranks, as well as by a system of promotions according to ‘merit’ that was aimed at assuring the basic compliance of judges to governmental policies (Magalha˜es 1995: 58). Beyond this point—the maintenance of a traditional civil law structure—any further attempts at increasing the political penetration of the Southern European judiciaries were moderate and, in any case, virtually unnecessary. Given the general conservatism of the great majority of Italian magistrates, only 17 out of more than 4,000 were purged in the transition from the Liberal to the Fascist regime (Saraceno 1993). As the Justice Minister Rocco underlined in 1925, ‘the judiciary must stay away from any kind of politics . . . and this is what in the great majority of cases happens’ (Modona 1973: 142). Only in the late 1930s, following the German example, did the Fascist regime increase its pressure on the judiciary. However, perhaps because of lack of time, it achieved no substantial results, and the ‘Fascistization’ of the Italian judiciary, as well as of all public administration, remained a rather superficial phenomenon.9 Likewise, the successive political purges that had taken place from 1935 onward in Greece had already resulted in the removal of the most free and radical spirits from the judiciary, making it hardly surprising that, at the time of the coup—April 21, 1967—the vast majority of Greek judges held extremely conservative views, similar to those expressed by the new authoritarian leaders (Pikramenos 1990: 64–7; Alivizatos 1994: 71). The only major incident of political-judicial confrontation occurred in 1968, when a constitutional act lifted the lifetime tenure of the judiciary, to facilitate the firing of thirty judges. The highest administrative Court, the Council of State, annulled the dismissals in June 1969, but the reaction of the

144

Democratic Consolidation and Judicial Reform

colonels was immediate: the President of the Council was forced to resign and replaced, together with one of the vice-presidents and eight other members. From then on until the regime’s collapse in July 23, 1974, the judiciary was not a cause of concern to the colonels. In Portugal, the resilient conservatism of the judiciary during the First Republic allowed Salazar’s regime a light hand in terms of purges, and a greater investment in terms of institutional rearrangements that would ensure judicial compliance to political orientations. The exception to this overall pattern is perhaps Spain, where the ideological composition of the judiciary under the Second Republic reflected the political polarization of society. During and especially after the civil war, Franco and the Falange included many judges (as civil servants in general) as targets of the extensive political purges (Andre´s Iban˜ez 1986). Institutionalized forms of political control, genuine convergence of political views—already existing or assured by purges—and the typical passivity and positivism of civil law judges contributed to the coziness of the relationship between these authoritarian regimes and their courts and judges. However, this coziness was also possible because Southern European dictators were careful enough to spare the regular judiciary the most ominous tasks of political repression and control. In Italy, most politically significant judicial cases under the Fascist regime were devolved to a special court—the Tribunale Speciale per la Difesa dello Stato, staffed mainly by military and party militia’s officers—or, in a less formal way, to the police forces, whose powers were increased. Therefore, although in a context of strong limitation of civil and political rights, political interference in the ordinary matters dealt with by the judiciary was rare, since all politically significant cases were assigned to separate structures that were fully loyal to the regime. Even in this case, the actual involvement of the Tribunale Speciale in the regime’s repressive activities must not be overstated: between 1926 and 1940 the Tribunale sentenced ‘only’ nine persons to death (Neppi Modona 1973: 103). In the first stage of the Portuguese Estado Novo, political crimes were dealt with summarily by military courts. Although jurisdiction was transferred in 1945 to special criminal courts, the Tribunais Plena´rios, trials took place in a setting that only slightly resembled the rules and procedures followed by regular courts, under the strict control of PIDE–DGS, the infamous secret political police. In Spain, the end of the World War II also marks an evolution toward greater legalism, correcting some of the more repressive and quasi-totalitarian features Franco’s regime had previously exhibited. That evolution culminated in the curtailment of the wide jurisdiction of military courts in 1963, and in the substitution of the

Democratic Consolidation and Judicial Reform

145

Tribunal de Represio´n de Masonerı´a y Comunismo [Tribunal of Repression of Masonry and Communism] by a politically controlled Tribunal de Orden Pu´blico (Iban˜ez 1986). Finally, in Greece, from 1967 to 1971, political crimes under the martial law were assigned for trial by military courts. During this period, the courts tried 3,364 individuals, imposing very harsh penalties on more than 2,000. In 1971, jurisdiction was indeed transferred to regular criminal courts, but that did not mean that the regime had decided to judicialize or even reduce political repression. While the number of political trials was significantly reduced, with courts showing much greater clemency than in the previous period, the colonels resorted to the more informal and extrajudicial methods of dealing with political dissidents shared by the remaining authoritarian regimes: unlawful arrest, imprisonment, deportation, and torture (Koundouros 1978: 29; Alivizatos 1979: 507). Interestingly enough, Greece, Portugal, and Spain had been forerunners in the adoption of systems of judicial review of legislation in Europe.10 However, judicial review was doomed to failure in all three countries. In Greece, the colonel’s dictatorship instituted a Constitutional Court with the power of settling conflicts of constitutional interpretation between the highest courts and, additionally, the power to dissolve political parties and to deprive citizens of the totality of their constitutional rights. Clearly, the inspiration was the West German Federal Constitutional Court, but the similarities stop here. On the one hand, the members of the court would be appointed by and subordinate to the executive. On the other hand, most Greek courts recognized the right of the colonels’ dictatorship to adopt executive ordered constitutional amendments, something that withdrew any normative power that the Constitution might have independently of the executive’s will. In Portugal and Spain, judicial review was inoperative even before the installation of Salazar and Franco’s regimes. Educated in a positivistic legal culture and inserted in a civil law organizational setting that curtailed judicial autonomy and discretion, the Iberian judge was hardly equipped to put in practice the ideas of contemporary constitutionalism, which after all have taken firm roots in Europe only after World War II. Therefore, the consequence of authoritarianism for the practice of judicial review in these countries was simply the complete neutralization of something that had never actually worked in any successful way. Franco simply eliminated judicial review from the set of powers enjoyed by the Spanish judiciary. The Portuguese 1933 Constitution kept it, but under severe institutional and practical limitations. While courts could theoretically review the content of legislation, they shared that power with an obviously inefficient rubber-stamp parliament. More importantly, every judge was quite aware that in

146

Democratic Consolidation and Judicial Reform

order to maintain his11 career expectations, he should avoid the ultimate extravagance of denying constitutional validity to legislation. This is, after all, a good example of how the constraints to judicial independence operated in the Southern European authoritarian regimes. The fact that many magistrates who served under these regimes were later able to swear upon the absence of explicit, and everyday governmental interference in the work of courts should not come as a surprise. In fact, that interference was hardly necessary. All that judges were entrusted with was the application of the legislation enacted by authoritarian political bodies and upholding the doctrines of politically controlled higher courts. On the one hand, the congruence between the preferences of judges and authoritarian leaders and their responsiveness to political authorities were assured by the mechanisms of judicial selection and appointment, as well as by the organizational and institutional settings in which judges were integrated. On the other, the most politicized activities that judges were asked to perform were carefully detached from the main structure of the judiciary. Therefore, the role of the Southern European judiciaries during the authoritarian regimes fits to a good extent the typology elaborated by Toharia with reference to Franco’s Spain (Toharia 1975). Unlike what happens in totalitarian regimes—where the judiciary tends to be strictly integrated, organizationally and ideologically, into the regime—in an authoritarian regime the judiciary tends to enjoy a moderate ‘independence’ in its day to day activity. This moderate independence is compensated by a low political significance, given that judicial jurisdiction is usually limited to the private sphere of social life, and judicial review is abolished or severely restricted. All politically salient cases are dealt with directly by the police or by special (and strictly controlled) courts. Finally, the incentives related to the judicial career, together with the mechanisms of professional socialization, tend to induce conformist behavior inside the judicial corps, discouraging any truly independent stance. As we shall see, as the democratic transitions in our countries unfolded, the democratic political e´lites in Southern Europe did not exactly see eye to eye regarding the specific paths that judicial reform should take. However, in each of these countries, a broad consensus was possible around a few major issues, and judicial reform in Southern Europe ended up exhibiting three basic common features. The first was the transfer of all or most supervisory powers over the judiciary from the executive to revitalized ‘judicial councils’. These were now conceived as self-governing bodies of the judiciary that would absorb those executive powers that had been previously used to neutralize judicial independence: recruitment, appointment, disciplinary sanc-

Democratic Consolidation and Judicial Reform

147

tions, promotions, and transfers. In Italy, the 1948 Constitution envisaged the creation of the Consiglio Superiore della Magistratura, and so did the Greek (1975), Portuguese (1976), and Spanish (1978) constitutions, creating, respectively, the Anotato Dikastiko Symvoulio (ADS), the Conselho Superior de Magistratura (CSM), and the Consejo General del Poder Judicial (CGPJ). All four constitutions also included articles stating the independence of the judiciary from elected branches and assuring judicial life tenure, as well as banning special tribunals and unifying the previously fragmented jurisdictional systems, a measure specifically addressed at avoiding the repetition of the past experiences of ‘special’ tribunals created to deal with ‘political crimes’. Second, purges and punishment of judicial personnel for involvement in repressive activities in the past were very limited in all four countries. In Italy, this was true not only of judges but also of most of the civil servants and other state officials. To be sure, a strong initial effort was made to prosecute officials of the past regime, especially those belonging to the Nazi-sponsored Repubblica Sociale Italiana. The effort led to an enormous amount of legislation: in seventeen months, between May 1944 and October 1945, the provisional government enacted eleven decrees to such effect (Battaglia 1962: 74). However, the sanctions against Fascist leaders were applied with declining energy, the expurgation of the civil service was slow and later reversed, and almost all judges were altogether left out of the process as objects of prosecution. In Spain, the pacted and consensual nature of the transition process precluded any sort of prosecution of former officials of the authoritarian regime, in exchange for the legalization of opposition parties and the overall acceptance of democratic reform by the most staunchly conservative sectors of the regime (Maravall and Santamarı´a 1986). To some extent, and despite the different nature of the country’s democratic transition, this was also true in Greece, as Constantine Karamanlis, the Prime Minister during the transition, was able to balance the contentious legalization of the Communist Party with limited purges of the Colonels’ collaborators from the state apparatus. In the end, very few judges were penalized and even fewer were dismissed, also a result of the fact that many of the Junta’s collaborators among the judiciary, who were mainly members of supreme courts and of an advanced age, had already retired when the authoritarian regime collapsed (Alivizatos and Diamandouros 1997: 39–40). The complex and tumultuous transition to democracy in Portugal, however, did pose the threat of widespread purges of the civil service and the judiciary. During the revolutionary years of 1974 and 1975, Portuguese judges faced not only the invasion of their normal jurisdiction by the extraordinary military and executive powers, but

148

Democratic Consolidation and Judicial Reform

also the far more threatening process of open defiance of legal order and judicial authority by spontaneous popular movements (Costa Pinto 1998). These developments were accompanied by a generalized radical left-wing discourse that depicted the judiciary as a ‘bourgeois’ institution, and by decrees providing for the dismissal and prosecution of officials involved in the authoritarian apparatus of political repression, including the judges. However, the initial intents of prosecuting the judges of the Plena´rio were rapidly replaced by more pragmatic considerations, especially when revolutionary discourse and practice gave way to the language of moderation and reconciliation since 1976. In the end, very few judges were officially or legally dismissed from office, and most of those that had been suspended were later reinstated (Magalha˜es 1995: 66; Costa Pinto 2001: 75–6). Finally, the third commonalty among our cases is that the important changes introduced in the Southern European judiciaries’ institutional independence and jurisdiction were not matched by comparable changes in their bureaucratic organization. In the years immediately following democratic transitions in Southern Europe, the judiciaries’ hierarchical structure in these countries was basically maintained, preserving the low level of judges’ internal independence and individual autonomy. Recruitment remained limited to young and basically inexperienced law graduates, whose career advancement remained dependent on systems of evaluation combining merit and seniority, with the former ascertained by the judicial councils and their inspectors. The possibility of ‘lateral’ recruitment of experienced jurists into the corps, which would have reduced the judiciaries’ insulation from society, did become formally possible. However, the high-ranking judges in control of the judicial councils did not conceal their hostility to this procedure, and used their powers to neutralize it de facto. As we shall see in section three, a few aspects of this basic structure of civil law judiciaries have undergone changes in later periods. However, what we find until the 1960s in Italy and the early 1980s in Portugal, Greece and Spain is that the ‘hierarchical-bureaucratic articulation’ of the Southern European judiciaries was kept virtually intact, while at the same time its institutional insulation from political bodies and authorities was increased (Andre´s Iban˜ez and Pe´rez Marin˜o 1991). Therefore, the processes and outcomes of judicial reform in Southern Europe were, to a great extent, remarkably similar. First, there was a high degree of continuity from the authoritarian to the democratic regimes in organization and personnel. Second, the few but crucial innovations introduced in terms of the expansion of judicial jurisdiction and independence, which adapted the Southern European democracies to the procedural criteria of the rule of law, were not seriously or

Democratic Consolidation and Judicial Reform

149

actively challenged by any major political group. Both outcomes were obtained without causing fundamental political conflicts on the issue of judicial reform, dramatic calls for the purging and prosecution of judges who had served the authoritarian regimes, or any kind of open defiance of the new democratic regime and its priorities by the judges themselves. What is striking about these similarities is the fact that they have occurred in rather different cases of democratic transition. Continuity and consensus in judicial reform do not defy greatly what would be expected in the case of Spain, the prototypical example of a regimeled transition, where guarantees to groups integrated in the authoritarian state apparatus were successfully negotiated and assured. However, Italy, Greece, or Portugal—distinct but unmistakable cases of regime collapse—would suggest rather different developments. If the relatively unproblematic nature of judicial reform in Southern Europe still does not sound particularly surprising, a look at recent events in Eastern Europe might help changing the reader’s mind. In Bulgaria, for example, much of the time between 1994 and 1996 was spent in acute conflicts between the Socialist (former Communist) parliamentary majority and the Constitutional and Supreme courts, as well as with attempts by the government to curtail judicial independence, arguably making judicial reform one of the most contentious issues in the short life of Bulgarian democracy (Melone 1996; Howard 2001: 96–7). In Poland and Hungary, judicial reform appears to have occurred in a somewhat smoother fashion, but not without the shadow of disruptive conflicts between majorities and oppositions on the issue of judicial appointments, or serious skepticism not only about the technical expertise of judges but also (and more fundamentally) about their loyalty to democratic rule (Oltay 1992; Sabbat-Swidlicka 1992). This contrast between the success and smoothness of judicial reform in the Southern European cases and the difficulties faced in Eastern Europe confirms the importance of the ‘legacies of the past’ for successful judicial reform. As Linz and Stepan note, in comparison with the East European cases, Southern European countries have benefited from several features of their previous authoritarian civil law judiciaries: the already existing principles of Western law and a developed legal scholarship; the highly formalized, civil-service style of judicial recruitment; the technical, ‘executory’ way judicial functions were traditionally defined; and the lower politicization of state officials (including the judges), as well as their lower involvement in repressive functions (Linz and Stepan 1996: 248–9). This lower politicization meant basically two things. First, that the traditional professional criteria in the recruitment and socialization of judicial personnel were at least allowed to coexist with political ones. Second, and most importantly,

150

Democratic Consolidation and Judicial Reform

that the judiciary as a whole was largely spared the need or the duty to perform any sensitive political roles, allowing for the resilience of mostly technical and professional standards in decision-making. Thus, the transition to democracy did not require costly purges or general overhauls of the judicial system, or the need to create anew both the procedures and the legal culture of the rule of law. The issue of judicial reform in the Southern European democracies was as important as the political role of the judiciary had been in the former authoritarian regimes. In other words, it was basically a secondary and unproblematic one, at least compared with the far more difficult tasks of overcoming the mutual distrust of Italian Communists and Christian Democrats in an emerging cold war context, accommodating military and civilian authority in Portugal, salving the wounds of the Civil War in Greece, or making the military, the Communists, and the Basque and Catalan autonomists agree on a set of democratic rules in Spain. This stands in stark contrast with the situation that later would arise in the East European transitions. Here, ‘legal nihilism’ had prevailed as the dominant view of the role of law in society, implying its use as an active instrument for the attainment of the political and revolutionary goals of the Communist parties. This element of legal culture had very concrete institutional and behavioral consequences. First, the direct and unmediated role of party authorities in the appointment and promotion of judicial personnel, for which being regarded as loyal Communists was a sine qua non condition. Second, the role of the State Prokuratura in closely monitoring judicial behavior. And third, phenomena such as ‘telephone law’, depicting the fact that judges were directly instructed by party officials on the desirable outcomes of cases.12 This legacy meant that in the more recent democratic transitions in Eastern Europe reformers faced more complex tasks regarding the adaptation of judicial systems to political democracy, given the higher politicization of judges, the inadequacy of their professional and legal qualifications, the general public distrust of legal institutions, and the destruction of the standards of impartiality in legal decision-making that some of those countries had inherited from their pre-Communist past. In contrast, for the Southern European regimes, those ‘continuities’ were seen as less problematic in terms of its adequacy to political democracy, at least in the light of more pressing reforms and of the equally pressing problems of scarcity of judicial expertise and the need for some continuity in the administration of justice. In theoretical terms similar to the conceptual framework of this volume, past institutional choices and ongoing political conjunctures have not ‘locked-in’ the choices of political actors, but have certainly

Democratic Consolidation and Judicial Reform

151

limited the range of their options. In other words, the authoritarian legacy seems to have decreased the stakes and the political centrality of the issue of judicial reform and to have simplified its agenda, allowing unproblematic continuities in organization and personnel.

Political Interests and Institutional Choices In spite of explaining the surprising similarities among our different cases of democratic transition, as well as distinguishing between them and the more contentious and problematic processes of judicial reform in Eastern Europe, the ‘legacies of the past’ approach fails to account for several concrete institutional outcomes of judicial reform. A ‘new institutions’ approach to judicial reform suggests that the institutional design of judiciaries is the result of struggles between political actors over the benefits and advantages brought about by different institutional arrangements. Political actors are interested in the shape of judicial institutions for two reasons. First, those institutions affect the congruence of the judiciary with the interests of the dominant political actors, because they determine the extent to which majorities can affect the composition of courts, allowing or preventing the appointment of judges whose preferences are known to be close to theirs. Second, the rules regarding the management of judicial careers also influence the responsiveness of judiciaries to political actors. They do so to the extent that they may place in the hands of current majorities the control of a system of punishments and rewards associated with judicial careers (promotion, assignment, recall), which can be used to condition judicial behavior, to some extent independently of the actual composition of courts and the policy preferences of judges. Thus, a ‘new institutions’ approach to judicial reform assumes that dominant political actors involved in processes of constitutional design attempt to create judicial institutions that foster the congruence of the judiciary with their interests and its responsiveness to their priorities, under the constraints posed by their present power and the uncertainty about who will control the elected branches in the future (Magalha˜es 1999; Ginsburg 2003). In fact, in the constitutional and legislative debates on judicial reform in the Southern European democratic transitions, the generalized commitment to the principles of expansion of judicial independence and jurisdiction was accompanied by important disagreements on relevant institutional details. In Italy, Portugal, and Spain, the parties to the left of the political spectrum were generally less favorable to the creation of institutions that would insulate the judiciary from political

152

Democratic Consolidation and Judicial Reform

pressures and increase its jurisdiction. The Italian Socialists and Communists, although rather more concerned with assuring the future control of a parliamentary majority, also expressed their reservations regarding the introduction of judicial review and the depolicitization of judicial appointments and supervision desired by the right (Rigano 1982: 38 ff). Similarly, in Portuguese constitutional debates, the advocacy of the principle of judicial independence coexisted awkwardly with attacks on the elitism and conservatism of the so-called ‘judicial establishment’, voiced especially, again, by the Socialist and the Communist parties. And although the generally leftist political environment in Portugal induced almost all parties to infuse some of these themes into their political discourse, in the Spanish case the partisan cleavage was more clear cut. While the Spanish Communists and Socialists were less than enthusiastic about strengthening the ‘political unaccountability’ of the judiciary, the conservative Unio´n de Centro Democra´tico (UCD) curiously became one of the staunchest defenders of the positions previously held by the progressive association of Spanish judges (Justicia Democra´tica), especially those concerning the monopoly of judges in the management of their own careers, to be obtained through the institutionalization of a judicial council composed by judges elected by their peers (Gerpe Landı´n 1991: 145–50). It was precisely around the issue of the judicial councils’ composition that partisan divisions were most clearly expressed. For example, in the Portuguese Constituent Assembly, while Socialists and Communists argued for a council whose majority would be designated by the elected branches, the center-right Partido Popular Democra´tico (PPD/ PSD) and the conservative Centro Democra´tico Social (CDS) preferred that judges elect the council’s members from among their own ranks. In Spain, the controversy between the UCD and the leftist opposition concerned a similar issue, namely, the proportion of council members to be designated among professional judges and among other kinds of jurists, these latter predictably with more clear political and even partisan connections (Gerpe Landı´n 1991: 145–50). A possible explanation for these positions would lie in the traditional ideological cleavages in Europe in what concerns the role of the judiciary. In fact, the European left has always been strongly influenced by the Jacobin tradition, which emphasizes the role of a powerful legislative assembly and distrusts any check on institutions representing popular sovereignty. Marxism further undermined the commitment to notions of judicial independence, portraying judicial institutions as unimportant ‘superstructures’ or bourgeois apparatuses. However, the Greek case illustrates how ideological traditions only go so far in explaining institutional preferences. Here, the positions of leftist and

Democratic Consolidation and Judicial Reform

153

rightist parties vis-a`-vis judicial institutions were the exact mirror image of their Southern European counterparts. During the constitutional debates in 1975, both the PASOK (Socialist) and the KKE (Communist) argued that the presidents and vice-presidents of the three supreme courts13 should be elected by their fellow members, while the moderate parties of the left proposed that promotion should be based merely on seniority. However, the conservatives in New Democracy (ND) favored the political appointment of the presidents and vicepresidents of the higher courts by the cabinet, and their control of no less than 73 percent of seats in the constituent assembly made the constitutionalization of those institutional preferences a very easy matter. Therefore, if—as the comparison between the Greek and the remaining cases illustrates—doctrinal and ideological traditions do not explain partisan positions in relation to judicial reform, then what does? New Democracy’s support for a greater politicization of judicial appointments cannot be understood without considering the fact that the approval of the Greek constitution would not be followed by elections, and therefore that the constitutional provisions concerning the judiciary would be functioning under a ND government with an overwhelmingly strong parliamentary support. Conversely, in Italy, the right-wing and centrist parties’ fears of what appeared to be an almost certain leftist victory made them more supportive of the strengthening of judicial independence and, in general, of any kind of check on legislative power (like judicial review). This insulation was all the more convenient given the perception that a judiciary untouched by the transition (and therefore, staunchly conservative as it had always been in the past) would be ideologically favorable to them. A similar perception about the Iberian judiciaries seems to have shaped the positions of political parties in Portugal and Spain. However, in these cases, the absence of hegemonic parties such as the Greek ND and the high level of uncertainty about the results of the first free elections after the approval of the constitution led to rather vague dispositions regarding the composition of the councils and left details to later legislative development, given the partisan disagreements on the issue and in the light of more fundamental bargains that had to be struck. As elections took place and the balance of powers between political parties was updated in Italy, Portugal, and Spain, partisan strategies also changed. After the 1948 elections in which they gained an absolute majority in parliament (with 52 percent of the seats), the Italian Christian Democrats showed the remarkable flexibility of their previous commitment to principle of judicial independence, actually freezing the implementation of the judicial council for as long as eleven years.14

154

Democratic Consolidation and Judicial Reform

That strategy was only made unsustainable by DC’s electoral setback of 1953, which weakened its grip on power. Under the new political circumstances, the center-right had to give in to increasing pressures for implementation of the Consiglio Superiore della Magistratura, although control of the latter by the upper echelons and the conservative factions of the judiciary still made it rather palatable to the Christian Democrats.15 In Spain, the UCD’s victory in the 1979 elections was short of an absolute majority (with 48 percent of seats in the Lower House), something that prevented a freezing of judicial reform similar to that which had occurred in Italy. However, it did not prevent the conservative majority from pursuing its interests in a different fashion. The 1980 Ley Orga´nica del Consejo Superior del Poder Judicial specified that the twelve career judges that would hold a majority in the CGPJ would be elected by their peers, and that the Consejo would assume particularly wide powers, including the nomination of two members of the constitutional court, oversight of judicial behavior and disciplinary actions, and all the decisions concerning promotions and transfers (Ferna´ndezMiranda Campoamor 1993: 55). The UCD’s active support of this arrangement, apparently reflecting its abandonment of efforts to exert institutional control over the judiciary, can be more clearly understood in the light of its connections with Spanish judicial factions. After the progressive Justicia Democra´tica (JD) had integrated a national and much wider Asociacio´n Professional de la Magistratura (APM) in 1977, the institutional insulation of the Spanish judiciary became acceptable to the executive as long as the dominant conservative faction inside the APM also controlled the judicial council.16 The postconstitutional evolution of judicial institutions in Portugal also brought a decisive change in the direction of a greater insulation of the judiciary. However, unlike what occurred in the Italian or Spanish cases, this process was not supported merely by the conservative players in the party system, but rather by a broader coalition that included the center-left Socialists and the center-right Social Democrats. The reasons for this peculiarity have to be found in the specificity of the Portuguese transition. That transition was, in a sense, a ‘double transition’, made not only against right-wing authoritarianism but also, at a second stage, against left-wing revolutionary radicalism. In this context, the potential cleavages between Socialists and Social-Democrats were frequently overridden by a deeper and more fundamental cleavage—between the extreme-left (including the Communists) and the remaining actors in the party system. This partially explains why, while in control of the executive, neither the Socialists nor the SocialDemocrats (who have basically alternated in government since 1976) found any obstacles to reinforcement of the institutional insulation of

Democratic Consolidation and Judicial Reform

155

the judiciary, despite occasional political rhetoric to the contrary (Magalha˜es 1999: 72–6). This ultimately paid off, since awarding the supervision of the judiciary to its overwhelmingly majoritarian conservative sector resulted in the neutralization of the ‘revolutionary’ remnants inside the judiciary. This was something that the CSM did willingly throughout the late 1970s, either by punishing magistrates whose behavior was permeated by nonlegalistic standards or by defining highly positivistic criteria of jurisprudential correctness that were weighted heavily in the classification and promotion of judges (Magalha˜es 1999: 83; Hespanha 1999: 21). The institutional outcomes of judicial reform in Southern Europe are described in Table 4.1. In Greece, the president and the executive retained their ability to appoint the higher ranks of the career judiciary, a power that ultimately facilitated their ability to influence the composition and leadership of the Greek judicial councils. Conversely, the Italian, Portuguese, and Spanish judges were allowed a rather strong insulation from current political majorities. That insulation resulted from the neutralization of all appointment powers previously held by political authorities and from the transfer of all supervisory powers to judicial councils where political appointees were in the minority— a highly original solution among other civil law countries. Italy took originality even further, by extending those guarantees to the public prosecutor. In most democratic countries, the public prosecutor’s office is strictly separated from the judiciary, and is made politically accountable either through direct political appointments or by its hierarchical dependence on the executive branch. This remained the case in Greece, Portugal, and Spain, but not in Italy. Here, judicial and prosecutorial careers were completely integrated and prosecutors came to enjoy the same guarantees of independence attributed to the judges. In conclusion, we can now get a glimpse of how a ‘legacies of the past’ explanation of institutional change in the judiciary fits together with another that focuses on political self-interest and ‘new institutions’. As we have seen, the authoritarian legacy of Southern Europe consisted of institutions and practices that reduced the political stakes around the issue of judicial reform and the tensions that would otherwise arise between personnel and organizational continuity, on the one hand, and the responsiveness of the state apparatus (including the judiciary) to the new democratic order, on the other. However, these legacies did not eliminate the ability of actors to pursue benefits through particular institutional choices. The fact that the Italian, Portuguese, and Spanish judiciaries ended up acquiring a high level of institutional independence from political authorities while the Greek did not can be seen neither as the result of different normative commitments with

Table 4.1 The institutional outcomes of judicial reform in the Southern European democratic transitions: rules of appointment and supervision of the career judiciary Politicization of judicial appointments

Powers of judicial councils

Composition of judicial councils

Italy (1948–59)

Very low: No appointment to judicial or prosecutorial positions by political bodies

High: Consiglio Superiore della Magistratura holds all transfer, promotion, and disciplinary powers over judges and prosecutors (Constitution, Article 105)

Majority of magistrates elected by their peers (Constitution, Article 104)

Greece (1975)

Medium: President of the Republic appoints chairs and vice-chairs of the supreme courts, under motion of the executive (Constitution, Article 90)

Medium: Transfers and promotions by presidential decree with concurring opinion of Anotato Dikastiko Symvouli. Disciplinary action against judges by separate disciplinary council (Constitution, Articles 90 and 91)

Headed by politically appointed members and composed of judges chosen by lot. Disciplinary council with majority of politically appointed members (Constitution, Articles 90 and 91)

Portugal (1976–7)

Low: No appointments to judicial positions by political bodies

High: Conselho Superior da Magistratura holds all transfer, promotion, and disciplinary powers over judges (Constitution, Article 223)

Majority of judges elected by their peers (Constitution, Article 223, and relevant legislation)

Spain (1978–80)

Low: King appoints chair of Supreme Court, under motion of CGPJ (Constitution, Article 123)

High: Consejo General del Poder Judicial holds all transfer, promotion, and disciplinary powers (Constitution, Article 122)

Majority of judges elected by their peers (Constitution, Article 122, and relevant legislation)

Democratic Consolidation and Judicial Reform

157

the principles of the rule of law and by law, nor as a result of traditional doctrinal positions on the issue of the judiciary. The Italian Christian Democrats in 1959, the Spanish UCD in 1980, and the Portuguese Socialists in 1976 (as well as the PSD/CDS center-right coalition in the early 1980s) actively fostered the institutional insulation of their countries’ judiciaries. First, they did so in the knowledge that complementing the conservative political leanings of most judges was also a professional and organizational culture that, following the civil law tradition, was characterized by the clear preservation of the boundaries between judicial and political spheres. Governing parties did nothing to change this view of the judicial role. In fact, they did quite the contrary, promoting organizational changes that marginalized the more progressive and activist factions and allowing fundamental continuities in personnel and bureaucratic settings. In this way, they preserved a traditional definition of the judicial role, allowing at least in the short-run for the preservation of the ‘dejudicialization of politics’ and ‘depoliticization of justice’ that had characterized these political systems in the past. Second, conservative and centrist parties fostered judicial independence not only because the control of those judiciaries by their overwhelmingly conservative sectors was potentially favorable to them, but also because they lacked a clearly hegemonic position in the party system. In political systems where no single party holds an absolute majority (and particularly in transitional contexts of heightened political uncertainty), it is less likely that governing parties will seek self aggrandizement by weakening constitutional guarantees in general and judicial independence in particular. This can be attributed not only to the difficulty in obtaining support for such measures (a result of insufficient bargaining power in the relevant decision-making arenas where institutional design takes place), but also to the perceived higher likelihood that current minorities will become majorities in the future and be able to use those weakened guarantees of judicial independence against them. Thus, awarding independence to conservative judiciaries untouched by the transitions provided the best strategy for these governing parties, allowing some degree of ideological congruence and a low degree of responsiveness to future political opponents. As the Greek and Italian examples also illustrate, whenever governing parties enjoyed a strong political position and electoral uncertainty was low, this logic was suspended. In the majoritarian settings of Greek and Italian democratic consolidation, ND and the Christian Democrats used their grip on power to generate and/or preserve judicial institutions that allowed for a low level of judicial independence.

158

Democratic Consolidation and Judicial Reform

Constitutional Justice and Democratic Consolidation in Southern Europe The story of judicial reform in Southern Europe will remain incomplete until we have focused on the birth and role of constitutional courts. Italy, Portugal, and Spain adopted the so-called ‘Kelsenian’ model of centralized judicial review, named after the Austrian legal scholar Hans Kelsen. In this model, judicial review is entrusted to special politically appointed courts, in contrast with the ‘diffuse’ US model, in which all courts are entitled to review the constitutionality of legislation. Another novelty in relation to the US model lies in the power of abstract review of legislation, exerted when a predefined set of public authorities (for instance, the prime minister, the president, the ombudsman, and others) refers specific pieces of legislation to the court for analysis of their conformity with the Constitution before [a priori] or after [a posteriori] they are promulgated by the Head of State and take effect. This contrasts with concrete review, in which the declaration of unconstitutionality can be only made in the context of a specific judicial dispute (Tate 1992). The introduction of systems of centralized judicial review in these countries is a particularly striking phenomenon, considering the rupture with sixty years of legal tradition in Portugal, the previous failure of the Tribunal de Garantı´as Constitucionales in the Spanish Republic II, and the complete lack of a tradition of judicial review in Italy. The general postwar cultural and political climate favorable to a politics of human rights, constitutionalism, and the expansion of judicial review may have favored these developments. However, to understand the successful institutionalization of constitutional courts in Italy, Portugal, and Spain, and the reason why Greece remained as the outlier among the remaining Southern European cases in terms of the introduction of the ‘Kelsenian’ model of centralized judicial review, we need to focus on two distinct roles these courts came to perform. As ‘transitional devices’, constitutional courts fostered the responsiveness of lower courts—where judges appointed by previous authoritarian regimes prevailed—to the interests of new democratic e´lites, as well as to diffuse legal doctrines more adapted to the protection of political rights. As ‘arbitral’ mechanisms, constitutional courts addressed dilemmas created by highly divisive and politically costly issues that the elected branches and e´lites were unable or unwilling to solve during the democratic transition, and which were postponed and delegated to institutions that enjoyed some degree of insulation from the electoral process.

Democratic Consolidation and Judicial Reform

159

Constitutional Courts as ‘Transitional Devices’ In Italy, the concept of constitutional review of legislation was almost completely novel. Before 1948, the Constitution of Italy was the Statuto Albertino (thus named because of Carlo Alberto, King of Sardinia, who had granted it), a ‘flexible’ text whose amendment did not require any special supermajoritarian procedure. Therefore, no court could strike down a law on grounds of its unconstitutionality, since the Statuto itself was considered to be nothing but a law; such an attempt would have been regarded as a violation of the principle of separation between legislative and judicial powers. In the constituent assembly debates of 1947, the defense of centralized judicial review derived from a generalized perception that judges who had been selected, promoted, and trained under Fascism were not likely to interpret the Constitution in the more democratic and progressive spirit of its framers (Di Federico and Guarnieri 1988; Cappelletti 1989). However, the institutionalization of the Italian constitutional court was not immune from the same kind of political calculations that permeated the design of other judicial institutions. Predictably, while the parties of the right defended judicial review, the left depicted it as a threat to parliamentary sovereignty and argued for a sort of ‘political review’, which means constitutional review of legislation by the parliament itself (Volcansek 1992: 93). In the end, however, the approval of provisions dealing with the constitutional court was relatively consensual. Mutual concessions resulted in the creation of the Corte Costituzionale, a constitutional court with powers that included the arbitration of conflicts of jurisdiction between state organs, as well as concrete and abstract judicial review of legislation under referral of political authorities or judges. In a pattern that replicated the protracted institutionalization of the CSM, the Italian Corte Costituzionale would only start functioning in 1956. After the DC’s electoral victory in 1948 and the exclusion of the left from power, the Christian Democrats succeeded in postponing the institutionalization of the court, and were only forced to accept it after the 1953 elections increased the bargaining power of the left, which had by that time discovered the court’s potential role in checking the centrist majority. The provisions of the Spanish constitution that concern judicial review of legislation were also the subject of a rather broad consensus among the four major parties, reinforced by the secretive nature of the work of the subcommittee that produced the first constitutional draft. As in the Italian case, the major dissension came later, during the discussion of the Organic Law of the Constitutional Court in 1979,

160

Democratic Consolidation and Judicial Reform

particularly concerning the UCD’s imposition of abstract a priori review of organic laws and statutes of autonomous regions. However, and except for this particular provision, later withdrawn in 1985, the Tribunal Constitucional Espan˜ol closely resembled its German counterpart. It was entrusted with concrete review of legislation (in response to queries from ordinary judges when dealing with cases where the constitutionality of applicable legislation was in question), abstract review (in response to referrals from a wide number of political and public authorities), amparo (enabling ordinary citizens to appeal judicial and administrative acts that allegedly violated fundamental rights), and, finally, responsibility for settling conflicts of competencies between the central government and the regional autonomı´as. In the Portuguese Constitution of 1976, the limitations that the authoritarian regime previously imposed on the ability of the ordinary judiciary to exert concrete judicial review of statutes were fully lifted. However, the establishment of a system of judicial review proper was also protracted, because of wider political constraints related to the role of the military in the transition. The mode of review adopted in the 1976 constituent assembly was predetermined by the pacts established between the parties and the military stipulating that the tutelary powers of the latter had to be accommodated, at least temporarily, also in the system of constitutional control of laws (Antunes 1984; and Arau´jo 1995). Therefore, an anomalous element was left in the Constitution, only to be removed in 1982. This element was the Conselho da Revoluc¸a˜o (CR), a body composed by the top hierarchy of the armed forces that, among other powers, retained the power of ‘judicial’ (or rather, in this case, ‘political’) review of legislation. However, the Constitution also defined the role of the Comissa˜o Constitucional (CC), an advisory body to the CR in matters of abstract judicial review, and with its own jurisdiction as the last court of appeals from judicial review decisions made by ordinary courts, composed by members appointed by methods similar to the ones used for regular constitutional courts. In this way, this highly convoluted system combined centralized and diffuse, political and judicial, and concrete and abstract review. Greece constitutes the exception among our cases in the sense that no constitutional court charged with judicial review of legislation was created after the transition. Instead, democratization brought a return to the predictatorship system where all courts and judges were entitled to review the constitutionality of legislation in concrete cases—in other words, to undertake concrete judicial review (Spiliotopoulos 1983: 477ff). However, there are special and rare circumstances in which one Greek court can conduct centralized judicial review. The Special

Democratic Consolidation and Judicial Reform

161

Highest Court (SHC) was created by the 1975 Constitution (Article 100) in order to, among other things,17 settle controversies among the highest courts in the land (the Council of State, the Areios Pagos, and the Court of Auditors) concerning the unconstitutionality of statutes, and thereby to increase legal certainty by unifying case law. The framers of the 1975 Constitution specifically used the term ‘Special Highest Court’ in order to avoid the identification of the SHC with the Constitutional Court which had been foreseen (but never functioned) in the colonels’ constitution. The SHC is composed by members of the three Supreme Courts and is enlarged to include law professors when the Court reviews the constitutionality of statutory law provisions or settles conflicts between courts and administrative authorities. Table 4.2 summarizes the functioning of judicial review in each of these countries. In an article discussing the role of judiciaries in South America, Owen Fiss describes constitutional courts as ‘transitional devices’ in new democracies, ‘a way of constraining the influence of judges who were appointed by or served dictatorial regimes’ (Fiss 1993: 72). As Samuel Valenzuela argues, democratic consolidation may be impeded by unaccountable political actors who impose reserve domains of policymaking and ‘have privileged access to crucial elements of state power to make credible their threat of destabilization’ (Valenzuela 1992: 65). The judiciary may be one such actor, particularly when judges who served authoritarian regimes remain unscathed by the transition and simultaneously see their insulation and power greatly increased, such as in the cases of Italy, Portugal, and Spain. In the previous section, we have seen how that increase in insulation and jurisdiction came about, and why the dilemma it created was comparatively less dramatic in Southern Europe than in posttotalitarian new democracies. However, the dilemma persisted. In Italy, for example, ordinary courts were quick to reinstate civil servants who had been purged during the transition, but rather slow in enforcing the large panoply of constitutional political and civil rights, whose effective exercise by citizens required the abolition of most of the Fascist legislation.18 For example, despite strong criticism by opposition parties and a substantial academic doctrine, the Italian Court of Cassation always managed to find creative ways to refuse to declare the bulk of Fascist criminal law unconstitutional, leaving untouched the blatantly excessive powers of police and prosecution (Battaglia 1962). Likewise, ordinary courts and career judges in both Portugal and Spain did not turn out to be particularly sensitive to political rights and constitutionalism, preferring in the Portuguese case to behave as a legal channel for the grievances of the old political and administrative

Table 4.2 Judicial review of legislation in Southern Europe Abstract review of legislation Italy

Greece

Corte Costituzionale —

Referral in abstract a priori review

A posteriori







Referral in abstract a posteriori review

Concrete review

Government, Regions, Judges in ordinary Autonomous courts file issues of Provinces constitutionality —

All courts

Appeals for constitutional redress —



Portugal (1976–82)

Conselho da Revoluc¸a˜o (Comissa˜o Constitucional as advisory body)

A priori and a posteriori

President, Council of the Revolution, Ministers of the Republic

All courts, CC as President, Speaker, final instance of Prime Minister, Ombudsman, Attorney appeal General, Assemblies of Auton. Regions



Portugal (1982)

Tribunal Constitucional

A priori and a posteriori

President, Ministers of the Republic, Prime Minister, 1/5 of MPs

All courts, TC as President, Speaker final instance of Prime Minister, Ombudsman, Attorney appeal General, 1/10 of MPs, Ministers, Assemblies & Execs. of Regions



Spain

Tribunal Constitucional

A posteriori



Prime Minister, Ombudsman, 50 members of Congress or Senate, Execs. and Assemblies of Auton. Regions

Judges in ordinary Citizens courts file issues of appeal directly constitutionality with TC

Democratic Consolidation and Judicial Reform

163

e´lites affected by the transition. This was done not only (in the Portuguese case) through their role in reintegrating purged civil servants, but also, for instance, through a generally lenient attitude in both Portugal and Spain toward extreme-right newspapers and public figures charged with abuse of freedom of the press. Tellingly, their jurisprudence on freedom of expression placed otherwise a priority on the right to honor and privacy at the expense of freedom of the press, as journalists in the now uncensored media became a preferential target for law suits and even criminal charges (Martı´n Pallı´n 1982: 74). What allowed constitutional courts to become effective ‘transitional devices’ and help solve the dilemma created by the strict insulation of untouched judiciaries? Two basic features of these systems of centralized review were particularly helpful: the rules regarding their composition and their monitoring powers over lower courts. First, unlike the career judiciaries organized according to the system of corporate self-government we described in the previous section, constitutional courts were responsive to new democratic political actors. That responsiveness was assured by the partial or complete political-partisan nature of the justices’ appointment (see Table 4.3), given the quotas assigned to different branches of government and the supermajority rules required in the different legislatures. Parties have typically negotiated the attribution of seats to ‘their’ justices, often former politicians, law professors, and career judges whose party credentials are well known. In the Italian and Portuguese (after 1982) cases, informal arrangements have even produced courts whose composition allows for an equilibrium between majority and opposition appointees, as occurs in the German case. In Spain, the lottizzazione was initially limited to the two major parties,19 but was later expanded to the appointment of justices supported by the Basque nationalist PNV, the Catalan nationalist Converge`ncia i Unio´ (CiU), and even the Communists of Izquierda Unida. In any case, the requirement of supermajorities for appointment of justices has often converted these institutions into the last bulwarks of consensual politics even in majoritarian political contexts, thereby assuring their responsiveness to electoral majorities and coalitions broader than the ones that control parliament and executive. This responsiveness is encouraged by the absence of institutional guarantees of judicial independence enjoyed, for instance, by the American Supreme Court justices (e.g. life tenure) or even by the ordinary judges of Southern European judiciaries. Second, constitutional courts were entrusted with powers through which that broad political responsiveness was transmitted even to the staunchly insulated Italian, Portuguese, and Spanish career judiciaries. This was obtained by changing the architecture of the judicial

Table 4.3

The composition of constitutional courts in Southern Europe Number of justices

Italy Corte Costituzionale

15

Length of term

Appointing authorities

Eligibility requirements

9 years, no reelection

5 by President of the Republic, 5 by Parliament (2/3 majority), 5 by highest ordinary and administrative courts

Law professors, lawyers with more than 15 years prof. experience, highest judges

Portugal (1976–82) Comissa˜o Constitucional

9

4 years

1 by Supreme Court, 3 by CSM, 1 by President, 1 by Parliament, 2 by CR

4 justices appointed by judicial bodies from career judiciary

Portugal (1982–98) Tribunal Constitucional

13

6 years

10 by Parliament (2/3 majority), 3 by the first 10 appointed

6 from career judiciary, all must be lawyers

Spain Tribunal Constitucional

12

9 years, no reelection, 1/3 replaced every 3 years

4 by Congress (3/5 majority), 4 by Senate (3/5 majority), 2 by government, 2 by CGPJ

All must be lawyers with more than 15 years of professional experience

Democratic Consolidation and Judicial Reform

165

systems, granting its top constitutional courts de facto authority over the ordinary courts from which they were supposedly detached. In Italy, the very first case heard by the Corte resulted from an appeal for concrete review of a Fascist security law argued to be contradictory with freedom of expression (Volcansek 1992: 95). By declaring the law unconstitutional, the Court indicated not only its willingness to assure strict protection of political rights but also to abrogate laws enacted prior to the promulgation of the constitution, a power that it proceeded to use against the conservative and sometimes undemocratic leanings of the career judiciary. For example, the Court of Cassation’s treatment of the reformed Code of Criminal Procedure produced an interpretation less favorable to the defendant, which triggered a long conflict with the Constitutional Court. By the second half of the 1960s, the jurisprudence of the latter had finally prevailed in practice. In this and other cases, the Constitutional Court was instrumental in adapting the Italian legal system to the new democratic constitution. In the Italian parliament, whose political inertia had prevented it from assuming such a role, this activity does not seem to have been negatively received.20 Monitoring of lower courts by constitutional courts was even more exhaustive in the Iberian democracies. This is due to the fact that, in Italy, the lower court judges themselves have to refer issues of constitutionality to the Corte Costituzionale in order for the latter to acquire concrete review jurisdiction, something that judges were initially quite reluctant to do (De Franciscis and Zannini 1992: 72). Judges in Spanish lower courts also made initially little use of the similar cuestio´n de inconstitucionalidad, as could be predicted from their generally positivistic and passive legal culture. However, the Spanish framers devised a far more efficient strategy in order to assure that the Tribunal Constitutional would endure less information and monitoring problems in their control of lower courts. The mechanism of amparo entitles citizens in general to seek reparation from the infringement of rights committed by any public authorities. These appeals have been extremely frequent in Spain (on average, since 1980, over 2,000 per year), and are, in a quantitative sense, the main activity of the Tribunal. How this mechanism can possibly have contributed to the control of career judges is not immediately evident. However, when we look at the object of appeals for constitutional redress, we realize that all of them are directed against judicial decisions made by ordinary courts. Legal scholars and observers of the Spanish case have not ignored this role of amparo in Spain. Rubio Llorente (1993) remarks that the constitutional court’s exhaustive control of ordinary courts’ decisions has converted it into a ‘super cassation’, ‘usurping’ the competencies of the

166

Democratic Consolidation and Judicial Reform

Spanish Tribunal Supremo. Bon, on the other hand, asks himself in the same vein if ‘it would not be more logical if the constitutional justice judged more the law and less the ordinary judge’ (Bon 1993: 59). But whatever the normative implications of the fact may be, the reality is that this form of control seems to have been always, one way or the other, in the minds of the leading political e´lites during the transition.21 Through the use of amparo, a diverse array of citizens and interest groups provide the constitutional court with valuable information on the lower courts jurisprudence, allowing it to choose among a large number of judicial decisions that it may wish to select and then reverse. Lower court judges are not unaware of this, and the incentives to deviate from higher jurisprudence are low, since response is sure and perhaps even swifter than what could be expected if a case had to be propelled upward by the usual and cumbersome avenues of normal judicial appeal. In Portugal, concrete review is directly connected with the ordinary judiciary’s own powers of judicial review. Those powers were articulated in such a way that when any ordinary court at any of the three tiers of the judicial system (Supremo Tribunal, appellate, and trial courts) is faced with constitutionality issues, its decision is bound to find its way to the Constitutional Court.22 The result of this framework is a system where the Supremo Tribunal, which previously reigned supreme over the judicial system, ended up by becoming a ‘lower’ court in relation first to the CC, and later to the Constitutional Court, at least regarding the judicial review of legislation. From a quantitative point of view, the main activity of the Portuguese Tribunal Constitucional (as well as of its predecessor, the CC) has been the concrete review of legislation: from 1976 to 1982, the Comissa˜o issued 481 decisions on concrete review, while the Tribunal Constitucional proceeded with this exhaustive control, issuing more than 7,600 decisions in concrete review from 1983 to 1998. To be sure, decisions that have confirmed previous rulings by lower courts have exceeded overrulings by a ratio of 2:1 (Rocha Marques 1993). However, the crucial aspect of concrete review by the Constitutional Court lies in the constraints imposed on ordinary courts’ ability to deviate from its doctrine or even to engage in judicial review in the first place. The result of this conversion of constitutional courts into ‘super cassations’ was twofold. First, like in the Italian case, Iberian constitutional courts have adopted antiformalist and anti-positivistic stances in the application of civil and political rights, operating major changes in jurisprudence and law in the field of equality, criminal procedure, and judicial protection of defendants. Although empirical data are unavailable, most observers recognize in both courts a scrupulous but

Democratic Consolidation and Judicial Reform

167

also widely creative approach to civil and political liberties, with the adoption of broad conceptions of protected rights (da Fonseca 1985; Rubio Llorente 1991). Second, the relationship between the Italian, Portuguese, and Spanish constitutional and lower courts has been, at times, filled with hostility. In Italy, the contribution of the constitutional court to the erosion of the traditional influence of the Court of Cassation explains some of the conflicts between the two in the early 1960s. In Portugal and Spain, controversy about, respectively, centralized concrete review and amparo, is recurrent. Career judges have periodically expressed their concerns with what they call ‘the politicization of constitutional justice’ and the decline of supreme courts, and have even proposed the creation of constitutional sections inside each Supremo Tribunal. However, conflicts have been defused by the complete unwillingness of any political party to acknowledge those claims, the changes in the prevalent attitudes in the judiciary, and the sheer institutional resources at the disposal of constitutional courts. Jurisprudential conflicts have always ended with the supremacy of the constitutional jurisdictions and, in the long run, a more collaborative relationship between the specialized constitutional jurisdiction, and the ordinary courts has actually emerged in the three countries. In short, the institutionalization of centralized judicial review by politically appointed constitutional courts was the ‘transitional device’ introduced in the architecture of Italian, Portuguese, and Spanish judicial systems that allowed for the coexistence of strict insulation of career judiciaries and continuities in organization and personnel. The Greek exception regarding the creation of centralized judicial review is particularly interesting to the extent that it confirms the arguments about why the combination of a career judiciary socialized and recruited under an authoritarian regime, on the one hand, with the institutional insulation of that judiciary, and on the other, creates pressures for the installation of some kind of device for the political monitoring of judges. In the Greek transition, the very notion of centralized judicial review was closely associated with the poor imitation of a ‘constitutional court’ that existed under the colonel’s dictatorship; there was a prevalent notion in political discourse that centralized and abstract constitutional review of legislation was associated with dictatorial regimes, or at best, to parliamentary systems with manifest authoritarian aspects. However, this instinctive rejection of centralized judicial review was also made possible by two aspects in which the Greek case is distinct from the remaining cases. First, the short duration of the Colonels’ regime meant that the majority of the Greek judges in office in 1974 had entered the judiciary before the coup of

168

Democratic Consolidation and Judicial Reform

April 21, 1967. Therefore, contrary to their Southern European counterparts, they had, as a rule, dispensed justice under the previous democratic regime and were much less mistrusted by the citizenry and the political e´lites. Second, the more politicized system of judicial appointment imposed by Nea Demokratia in the transition, and the political responsiveness of the judiciary it assured, eliminated the dilemma faced by the remaining Southern European new democracies, and helps to explain why a ‘transitional device’ such as centralized judicial review was not introduced.

Constitutional Courts as ‘Arbitral’ Institutions: Two Case Studies Diffusing the ‘culture of constitutionality’ from above throughout the career judiciary, and making it politically responsive was not the only positive contribution of constitutional courts to democratic consolidation. Judicial review of legislation also seems to have functioned as an ‘arbitral’ mechanism, insulated from the will of narrow political majorities and able to fill in the gaps of incomplete constitutional contracts (Lijphart, Rogowski, and Weaver 1993: 306; Guarnieri and Pederzoli 1996: 153–4). The following two case studies illustrate how this important role was performed in Portugal and Spain regarding the two most important political obstacles to democratic consolidation in both countries: the reserve powers of the military Council of the Revolution (CR) in Portugal, and the allocation of powers in a system of multilevel government in Spain.

Portugal: The Comissa˜o Constitucional and the Council of the Revolution The major anomaly of the post-1976 Portuguese regime from the point of view of its conformity to the criteria of procedural democracy was the CR, composed of 19 members of the military.23 The transitional pacts signed between the political parties and the military attributed to the CR important roles, ‘constitutionalizing’ the military’s tutelary powers in Portuguese democracy at least until 1982. The competencies of the CR included not only autonomous legislative powers in military matters, but also the power of abstract constitutional review of legislation (see Table 4.2). As several authors have pointed out, these de jure powers prevent us classifying the Portuguese regime as a consolidated

Democratic Consolidation and Judicial Reform

169

political democracy until 1982, when the Council was finally eliminated from the Constitution and a true constitutional court was created (Linz and Stepan 1996: 126). However, a more detailed observation of the activities of the CR tell us a somewhat more subtle and complex story—the story of the emergence of a genuine and vigorous form of judicial review that slowly undermined the institutional role of the military in the regime. The most outstanding power of the Council was in fact the abstract a priori review of legislation, through which it was integrated in practice as a supplementary chamber in the regular legislative process. As we can see in Table 4.2, both the military presidency and the CR itself were able to refer legislation for the latter’s abstract constitutional review. However, this power seems to have been used with surprising self-restraint. From a total of 3,843 legislative bills and executive decrees24 sent to the President between 1977 and 1982, the CR analyzed only 1.6 percent in a priori review. A breakdown of parliamentary bills and executive decrees that were reviewed by the CR during this period, as well as the number struck down by the CR as unconstitutional, can be seen in Table 4.4. If the activism of the CR in a priori review does not seem entirely negligible, neither does it seem particularly outlandish, at least when compared with other systems of centralized judicial review where the abstract a priori mode is available.25 If we take in account the de jure powers available to the CR, we realize that it would have been easy for this institution to assume complete veto power over all parliamentary and governmental legislation due to the Council’s ability to refer legislation to itself and then repeal it. However, of all the statutes picked by the CR itself for constitutional scrutiny, only about half were repealed, and in the case of legislative bills referred by the president this percentage dropped to 30 percent. Part of the explanation for this relative self-restraint is rooted in the system of checks and balances defined by the constitution and Table 4.4 Abstract a priori review in Portugal (1977–82) Referred to the CR as Pct. of all bills submitted for promulgation

Pct. struck down by CR (pct. of all submitted for promulgation)

Parliamentary bills

7.3% (29)

2.0% (8)

Executive decrees

1.0% (35)

0.6% (20)

170

Democratic Consolidation and Judicial Reform

the constraints it entailed, and, in particular, the role of the CC. The Comissa˜o was initially devised as a technical advisory body to the CR. The Council was required to consult with the Comissa˜o in all rulings concerning abstract review of legislation, but the opinions of the latter had no binding character (except with respect to the review of decisions by the lower courts). Nevertheless, in practice, the Comissa˜o acquired a much more important role. Its composition was determined by a method that was rather similar to those used by other European constitutional courts, combining political and judicial appointments. Its members, particularly those who were recruited from outside the career judiciary, were among the most prestigious lawyers and legal scholars in the country, at least among those not connected with the previous authoritarian regime. Its already discussed role on concrete review gave the Comissa˜o importance and prestige, and its opinions on abstract review cases were public and, in both form and content, no different from typical judicial rulings of European constitutional courts, with their complex and exhaustive legal argumentation, use of comparative precedent and jurisprudential coherence. The CR could indeed contradict those opinions in its final ruling, but that was a decision not to be taken lightly. Its juridically unprepared members could do nothing else but to summarily confirm or disconfirm the CC’s opinion, and when the latter was the case, had then to be able to withstand the public costs associated with engaging in what could only be perceived as overtly political intervention in judicial matters. In the end, few rulings by the Conselho da Revoluc¸a˜o were contrary to the CC’s opinions: in 213 abstract review decisions, in only thirteen (6 percent) of those cases did the Council reverse the opinion of the Comissa˜o. Moreover, a close analysis of those thirteen cases also shows how the conflicts between the CC and the CR in terms of constitutional interpretation were carefully circumscribed to two types of legislation. First, military issues, where the rather anomalous power of the CR to decide on the constitutionality of its own decrees was most evident. Second, the social/economic policies (especially regarding social rights and the definition of economic sectors to be open to private enterprise of the centerright PSD/CDS governments of 1979–82), which were clearly antithetical to the leftist economic clauses of the Constitution, and which embroiled the government in political conflict with President Eanes. But not even here does the Council’s reactions seem particularly abnormal: first, because judicial activism vis-a`-vis nonincremental legislation seems to be a common feature in most systems of judicial review in Europe; and, second, because in eight of the thirteen cases where the Council decided against the Committee’s opinion, the latter had been obtained by a narrow majority of its members, something that ran contrary to the

Democratic Consolidation and Judicial Reform

171

general trend of wide consensus in that consultative body’s decisions.26 This indicated that ‘in those cases, the CR’s decision [to repeal the CC’s advisory opinion] could be seen as supported by reasonably sound juridical-constitutional grounds’ (Antunes 1984: 324). Therefore, the contribution of the CC to the consolidation of Portuguese democracy went beyond the monitoring of the ordinary judiciary and the protection of political rights. The Comissa˜o was also instrumental in mitigating the deviations from procedural democracy introduced by the institutionalization of military power through the CR: first, in a general sense, by influencing the final rulings of the Council, whose members were not unaware of the inadequacy and abnormality of their judicial powers; second, and even more strikingly, the CC was successful in ‘inviting’ the CR to abdicate from some of powers, by issuing several opinions where it declared the Council’s decrees incompatible with the Constitution. In what may come as a surprise to those who have stressed the tutelary role of the Conselho da Revoluc¸a˜o in those years, the CR actually confirmed most of these opinions in final rulings, accepting therefore the judicial limitation of its own powers. Thus, the CC played a decisive role in mitigating one of the most undemocratic aspects of the 1976 Constitution—the reserve powers of the CR—and ended up prefiguring the Constitutional Court that emerged in 1982, with significant continuity of institutional rules and jurisprudence.

Spain: Judicial Review and Multilevel Government It has been argued that the very rationale for the introduction of judicial review in democratic regimes is the role of courts as arbiters in systems of multilevel government. Simple correlation seems to support this assessment: as Lijphart points out, judicial review and ‘rigid’ written constitutions tend to go together with federalism and decentralized government in modern democracies (Cappelletti 1989; Johnson 1993: 34–6). However, we are still far from understanding what exactly is the role performed by courts in the relations between central governments and federal subunits. Stone and Shapiro, for example, suggest that: ‘[The constitutional court’s] potential usefulness to the central government in holding the member states in line is a political resource of such magnitude that it can be employed to extract from the central government obedience to its decisions, even on questions unrelated to federalism, and on decisions on federalism that run against the central government . . . Each of the member states of a federal system may typically be sufficiently concerned about the potential misconduct of its fellow members that each will welcome federal judicial discipline directed against the others and even accept that discipline

172

Democratic Consolidation and Judicial Reform

against itself as a small price to pay for insuring the discipline of the others. (Stone and Shapiro 1994: 408)

This hypothesis invites further development. One of its possible interpretations would be that, in a federal system, the role of courts is to ‘keep the member states in line’, furthering the interests of central government and the majorities that control it. However, there is another interesting interpretation. In some cases, the authority of courts may be used against both central governments and federal subunits, and still be welcomed by all participants if and when it is perceived that the benefits thus brought in terms of the surveillance of other subunits and/or the maintenance of the overall federal system outweigh short-term losses. In fact, ‘keeping member states in line’—an undeniably useful political resource for majorities in the central government—could hardly have worked in democratic Spain. The main obstacle to Spanish democratic consolidation undoubtedly was and remains the ‘stateness’ problem (Linz, Stepan, and Gunther 1995: 89). During the Franquist regime, the historical claims to political autonomy produced by some of Spain’s diverse cultural and linguistic subunits were not only dismissed, but also severely repressed by all available means. Part of the Herculean task that political reformers had to face in the transition was to come up with an institutional framework that assured the unity of the state, and at the same time accommodated some of the long-repressed claims of nationalist political forces, which particularly in the cases of the Basque Country and Catalonia had not only deep historical roots and institutional precedents, but were also being expressed in some cases by overtly violent methods and secessionist objectives.27 It should not come as a surprise that the question of the Comunidades Auto´nomas’ degree of political autonomy was the most deeply divisive in the pacts that paved the way to democracy as well as in the constitution-making process (Cotarelo 1989: 320). As a result of that process, instead of searching for an impossible maximalist agreement which would detail the division of competencies and powers between the regions and the central government, the framers opted for a minimal consensus that would not harm the remaining goals of the transitional process (de Esteba´n 1989: 304–5). That outcome was a system that hardly fits in the basic legal and theoretical categories of federalist, unitary, regional, or decentralized state28. However, the vagueness and ambiguity of its constitutional definition satisfied the partially contradictory views of conservative and progressive parties, allowed for the recognition of the so-called ‘historic communities’, and did not close entirely the prospects of expanding political autonomy for the nationalist forces in Euskadi and Catalunya.

Democratic Consolidation and Judicial Reform

173

That vagueness did not preclude the establishment of certain fixed characteristics of the Spanish system of multilevel government. First, Spain was divided into comunidades that would enjoy a sort of ‘autonomy a la` carte’, awarding autonomy status to the regions that desired it, to the extent they desired it (within the basic limits of the Constitution), and through two basic transitional processes, one faster and more extensive than the other.29 Second, the Constitution created a wide variety of legislative instruments for mutual delegation and redefinition of competencies between central government and comunidades, suggesting that the allocation of powers would be subjected to a constant bargaining process, contingent upon political conditions and the evolution of the democratic transition itself. One final rule in this constitutional agreement was that the Tribunal Constitucional would serve as this game’s umpire. The executive was allowed the possibility to refer legislation produced by the regional governments to the Tribunal Constitucional, and the executives and legislatures of the Comunidades Auto´nomas were similarly allowed access to the court to dispute the constitutionality of central government legislation interfering with their competencies. Finally, a further supplemental mechanism of conflictos de competencia (constitutional controversies between the central government and the regions on administrative acts) was created. However, it remained an open question as to whether the court could be trusted to enforce the basic and minimal rules of the game that were implicit in the constitutional pacts. The first and crucial test would come in 1982. The approval of the Basque, Catalan, and Galician estatutos created a chain reaction by which all the other autonomous regions— where the historical, political, and cultural claims to self-rule had been weaker or simply non existent—took advantage of the precedents set by the three ‘historic communities’ and explored constitutional opportunities for autonomy that, apparently, some of the framers did not have in mind. In fact, the attempted coup of 1981 has been interpreted as a counterreaction to this process from the sectors that most feared its consequences to national unity (Gunther, Montero, and Botella 2004). The political response of the center was led by leader of UCD and Prime Minister, Leopoldo Calvo Sotelo, who concocted a political pact with PSOE to rationalize and discipline the ‘autonomic system’ defined by the 1978 Constitution. The product of that pact was the ‘Law for the Harmonization of the Autonomic Process’, known by the acronym LOAPA. Basque and Catalan nationalists, excluded from these negotiations and perceiving LOAPA as a major breach in the transition pacts, reacted quite negatively to this initiative, and the danger of deconsolidation of Spanish democratic institutions loomed large (Gunther 1991:

174

Democratic Consolidation and Judicial Reform

251–2). The constitutional court was then called into play. In August 1982, the Tribunal received five requests for a priori review of LOAPA from the executives and parliaments of Catalonia and the Basque Country, as well from a group of 50 MPs. Following a full year of deliberations, the court reached its crucial and historical decision. Not only did the court declare fourteen of the LOAPA’s specific provisions unconstitutional, but went so far as to explicitly reject the very notions that central/regional government relations could be regulated by legislative means and that the powers of the federal subunits could actually be ‘harmonized’ (Ruling 76/1983). The political meaning of this decision was twofold. First, it was perceived as a boost to claims for the devolution of power from the central government to the Comunidades Auto´nomas. And second, it meant that the distribution of powers between the regions and the central government was to remain open-ended, heterogeneous, and subject to political negotiation and case-by-case constitutional interpretation by the Tribunal Constitucional itself. Indeed, throughout the 1980s, the court was flooded with requests to scrutinize the constitutionality of legislation and administrative acts that allegedly violated the division of competencies between the central government and the Comunidades. On the one hand, conflictos positivos de competencia became the second major source (after amparos) of the court’s caseload, averaging 32 per year from 1980 to 1997.30 On the other hand, and unlike what occurs in most constitutional review systems, the power to refer legislation to the court for abstract review has not been primarily used by the opposition to challenge the policies of the governmental majority. Instead, it has been almost exclusively used by the Prime Minister against legislation passed by the Comunidades or, conversely, by the latter against central legislation (especially by the Basque and Catalan governments) (Bon 1993: 59). One of the results of this intense judicialization of multilevel conflicts has been, according to Gunther, Montero, and Botella’s estimation (2004: 322), that something like 10 percent of all autonomic legislation has been challenged before the Tribunal Constitucional. After the initial boost to decentralization resulting from the LOAPA decision, the court’s jurisprudence has defied a clear classification as being for or against decentralization. Legal analysts have been somewhat vague in this respect,31 and some have even described that jurisprudence as neither centralizing nor decentralizing, but instead as unclear, casuistic and fragmentary (Pe´rez Royo 1986; Moderne 1993). However, the court’s unwillingness to define a rigid and dogmatic rationale behind the central-regional allocation of powers, together with its ability to convert political conflicts into somewhat arcane

Democratic Consolidation and Judicial Reform

175

legal controversies, may have been precisely the reason why it has maintained compliance and contributed to defuse this most divisive issue in Spanish democratization. Although observers have been unanimous concerning the centrality of the court in the construction of the Spanish Estado de Autonomı´as, the 1990s have brought an apparent decline in that role, as fewer and fewer issues concerning central-regional conflicts have been brought before the court.32 Part of the explanation for that decline lies in the mere passage of time, as the court progressed in building a jurisprudence that reduced the number of potential new conflicts and made the outcome of old ones predictable (Alberti, Bon and Moderne 1995: 320; Gunther, Montero and Botella 2004, Ch. 6: 33). In addition, the autonomic pact signed between PSOE and PP in February 1992 for the greater harmonization of the powers of regional governments may also have contributed, at least temporarily, to a reduction of central-regional conflicts. A more troubling reason for the reduction in Constitutional Court appeals since 1990 is that Basque regional government authorities have simply ceased to refer legislation or administrative acts and decrees to the Court. Although this could be interpreted as merely reflecting a preference for the political and bilateral settling of conflicts with the central government, it could also represent a challenge to the Court’s legitimacy, among Basque nationalists in particular. In fact, statements by Catalan and Basque political leaders indicate that the nationalists perceive (or want their constituencies to perceive) the Court’s jurisprudence as extremely centralist, and want to have an institutionalized say in its composition (Aja 1999: 135). On the other hand, the decline in judicial conflicts may only be temporary. In a system characterized by constant pressures for competition between regions with asymmetric powers and insufficient institutional mechanisms for cooperation,33 the importance of the Tribunal in the management and arbitration of the Estado de las Autonomı´as is not likely to disappear anytime soon, as the new rise of conflicts since 1997 seems to indicate.34

The Judicialization of Politics in Southern Europe In the previous sections, we described the extent to which Southern European judiciaries were transformed by democratic transitions, how institutional mechanisms were put into place to assure the successful integration of judiciaries in the new democracies, and how judicial institutions themselves contributed to democratic consolidation. This final

176

Democratic Consolidation and Judicial Reform

section addresses some aspects of the role of Southern European courts, judges, and prosecutors following consolidation of these democratic regimes. In these countries, the passive civil law judiciaries of authoritarianism were partially replaced by aggressive judges and prosecutors, deeply involved in the investigation and trial of cases of political corruption and abuses of power. In all four countries, these scandals and their investigation contributed to the discrediting of political e´lites, major shifts in public opinion and important electoral changes. Indeed, in Italy, these effects assumed the form of a political earthquake. David Nelken depicted it as a peculiar kind of revolution, one where ‘it was the renewed insistence on legality which itself had revolutionary consequences’ (Nelken 1996: 191). Others have described it as a second transition, this time not from authoritarianism to democracy but rather from one democratic regime to a different democratic regime, the Italian ‘second republic’ (Mershon and Pasquino 1995: 42). This last section attempts to explain these shifts, focusing especially on the reasons behind the particularly intense form of judicialization of politics that took place in Italy. We conclude that the judicialization of politics in Italy was the result of both institutional conditions and political incentives including, on the one hand, a particularly powerful and institutionally insulated prosecutorial branch, and, on the other hand, a politically factionalized judiciary, facing a party system plunged into stagnation and crisis. In terms of the theoretical approach shared by several authors in this volume, in Italy historical legacies of institutional development provided political actors with a limited range of choices. In spite of apparently similar manifestations of judicial activism in the remaining Southern European democracies, the peculiar conjunction of factors that caused the ‘Clean Hands’ wave of judicial activism does not seem to have been present in Greece, Portugal, or Spain.

Behind ‘Clean Hands’: Institutional Opportunities and Political Incentives The events surrounding the famous ‘Clean Hands’ investigations are well known and have been described elsewhere (Nelken 1996; Guarnieri 1997; Giglioli 1999: 6381–94). Starting in 1992, the prosecutors of Milan [soon baptized Tangentopoli (Kickback City)] brought to light a widespread and deeply rooted web of corrupt transactions involving politicians, bureaucrats, and business people. Soon, those investigations started to hit politicians of the governing coalition, as well

Democratic Consolidation and Judicial Reform

177

as managers of public and private companies. The Amato government, a coalition formed in July 1992 by four parties of the traditional establishment in order to deal with the difficult economic and financial situation (including a substantial devaluation of the Lira) survived less than a year, but long enough to see seven of its ministers resign, after being notified that they had been placed under investigation. On December 15, the former Prime Minister and still head of the Socialist Party, Bettino Craxi, was notified of the fact that he was also under investigation. In early 1993, the prosecutor general of Milan summarized the results of the operation mani pulite: 380 people were under investigation, among them fifteen members of the Parliament. By the end of the year, that figure had risen to 1,135, partially due to the Enimont scandal, which involved virtually all major members of the political class who governed the country in the 1980s, under the accusation of having solicited and received a so-called ‘mother of all bribes’ of about 100 billion lira (Canosa 1996: 205). In such a climate, the traditional hold of ‘party government’ on the Italian political system began to crumble. In April 1993, the former governor of the Bank of Italy, Carlo Azeglio Ciampi, was asked to lead a government openly sponsored by President Oscar Luigi Scalfaro. Three members of the Democratic Party of the Left (PDS, the heirs of the Italian Communist Party) took part in the coalition for a few hours but later withdrew, as a protest against the Parliament’s decision to grant immunity to Bettino Craxi. The local elections which took place in the fall of 1993 in some important cities seemed to open the way to the PDS, which until then had been only marginally affected by ‘Clean Hands’. But the national elections held on March 1994—with the newly enacted quasi-majoritarian electoral system— had a quite different result. The center-right alliance, Forza Italia, forged just before the elections by the television magnate Silvio Berlusconi, triumphed in the polls. However, attempts at establishing peaceful relations with the judiciary were pointless.35 In an ironic twist of fate, Prime Minister Berlusconi was himself notified of being under investigation while attending a UN conference on the fight against crime in November 1994. The internal weaknesses of the heterogeneous coalition led by Berlusconi, along with his tarnished public image, had as a consequence the fall of the government in December 1994. Once again the President of the Republic played a crucial role in facing yet another difficult political situation. A government of ‘experts’ was appointed under the leadership of Lamberto Dini, a former executive of the Bank of Italy, with the external but decisive support of the Democratic Party of the Left. Later that year, for the first time in the postwar

178

Democratic Consolidation and Judicial Reform

period, a nonconfidence vote in parliament supported by center-left political forces forced the resignation of a minister—a former magistrate in charge of the key portfolio of justice—because of his decision to send inspectors to verify prosecutors’ activities in Milan and Palermo. Finally, in the aftermath of ‘Clean Hands’, the elections held on April 1996 brought the success of a heterogeneous center-left alliance headed by the PDS, a victory that seemed to be welcomed by the judiciary. The magnitude of the political change to which mani pulite contributed cannot be understood just by looking at cabinet instability and the number of investigated politicians. In fact, many actors who had played a central role in the First Republic were forced to leave the political scene, at least for a while, thus accelerating the turnover of political personnel initiated with the 1992 elections. The party system underwent dramatic changes. The parties hit the hardest were those that had traditionally composed the governing coalitions for almost half a century. Some of them, as the Social Democrats, simply collapsed. Others, like the Liberals and the Republicans, suffered severe electoral setbacks and lost much of their previous importance. The Christian Democrats, who had dominated Italian coalition politics over the previous half a century, split several times and were ultimately converted into a new, purged, and weaker Partito Popolare Italiano. The Socialists officially disappeared, came back under a new label, but received just 2 percent of the vote. Overall, the sudden burst of judicial activism, together with major shifts in electoral behavior, and economic and financial crises, contributed to the breakdown of the old political equilibrium. The weakening of the executive and the legislative branches, with many of their members under investigation, was only partially balanced by the growing role of the President of the Republic (Cotta and Isernia 1996: 15). Most important was the shift of public support and confidence from the political branches, largely discredited, to a ‘new’ institution, the judiciary, that was perceived to be far more responsive to societal demands than the political branches, thanks also to the support of the media (Giglioli 1999). How can we explain the behavior of judges and prosecutors and their role in this transformation? The judicial institutions whose emergence we described in the first section of this chapter were, in all likelihood, a necessary condition for mani pulite. The political protagonismo of the Italian judiciary cannot be understood without considering the institutional opportunities provided by a high level of external independence and the great powers with which magistrates in general and public prosecutors in particular were endowed. Very much unlike what occurs in most democratic polities—including the other Southern European cases—the means available to Italian excutives to influence the action

Democratic Consolidation and Judicial Reform

179

of public prosecutors are extremely limited, considering the unification of prosecutorial and judicial careers and their subjection to a largely politically insulated judicial council. Therefore, it is not surprising that, especially in the last twenty years, public prosecutors have been able to act in a way that is increasingly autonomous from the interests of governing coalitions. The growth of prosecutorial independence has also been facilitated by the fact that the Italian criminal process is strictly governed by the constitutionally prescribed principle of compulsory prosecution (Article 112). On the one hand, this principle has been interpreted in such a way as to imply a ‘judicialization’ of the role of the public prosecutor and to justify the extension of judicial guarantees of independence also to prosecuting magistrates.36 On the other hand, by requiring prosecutors to start an investigation and request a judicial decision every time there is some evidence that a crime has been committed, the principle of compulsory prosecution also assigned magistrates a task impossible to perform. In practice, when the prosecutor finds some evidence of a crime, she or he must open a file. Given the practical impossibility of concluding all criminal investigations initiated in this manner, those investigations often lag behind until the statute of limitations has to be applied. But most importantly, the dismantling of substantial controls on prosecutorial activity has also made the prosecutors virtually unaccountable for the choices they inevitably have to make concerning which investigations they actually pursue. The weak accountability of public prosecutors is coupled with their significant and increasing powers. Especially since the 1970s, the fight against terrorism and organized crime has steadily led to an expanding role of magistrates—prosecutors and investigating judges—in the criminal process.37 The trend was not reversed by the 1989 reform of the code of criminal procedure. Indeed, subsequent legislation and the Constitutional Court’s rulings further reinforced this trend, to a large extent entrusting the prosecutor with the power of instructing the criminal process.38 Moreover, the wide powers enjoyed by the prosecution and the organizational connection between the prosecutor and the judge, both belonging to the same corps, tend to strike an uneven balance between the two conflicting parties, openly disadvantaging the defendant. These trends toward the growth of external independence and powers of the Italian magistrates provided institutional opportunities for judicial protagonismo. The political incentives derived from a third trend: the erosion of the power of the higher ranks of the judicial hierarchy and the dismantling of the judicial-political arrangement that had been established in the past between them and the Christian

180

Democratic Consolidation and Judicial Reform

Democrats. That was the same arrangement which, as we have previously seen, had frozen the implementation of the judicial council until 1958, allowed the DC’s final acquiescence to judicial self-government, and prevented any substantial change in the internal organization of the judiciary until much later. The beginning of the end of that arrangement can be traced back to the emergence of the so-called correnti, organized judicial factions formed in the aftermath of the expansion of the judiciary’s size as well as of its increasing social and political diversification brought about by democracy. That increased pluralism would soon result in ties to a variety of political parties, including those in the opposition. Parties and correnti found mutual advantages in the establishment of such ties. For their part, political parties had strong incentives to carefully cultivate relationships with Italian magistrates. Obviously, this was particularly true for the Communist Party since it could introduce yet another check on the power of the Christian Democrats, given the PCI’s exclusion from participation in the governing coalitions that dominated most of the Italian First Republic. During the 1950s and 1960s, meanwhile, many magistrates sought allies among the opposition political parties in their efforts to reform (actually, to abolish) the traditional career system and procedures for internal evaluation of work performance and promotions (Freddi 1978; Guarnieri 1992). These efforts were overtly opposed by the higher ranks of the judiciary, and were further frustrated by the government’s unresponsiveness to their demands. The opposition parties, with their growing influence in parliamentary politics (Di Palma 1977; Cotta 1991), were willing and increasingly able to assist the magistrates in implementing these reforms. As a result, between 1963 and 1973, the Italian parliament passed a series of laws that allowed for a step-by-step dismantling of the traditional system of promotions. Those changes, in turn, reinforced both the centrality of correnti in judicial life and the politicization of magistrates, by making the ties between a candidate for promotion and a specific corrente or party highly relevant for judicial appointments (Di Federico 1990: 279–97; Rebuffa 1993: 91–8). Therefore, all the terms of the past alliance between the higher ranks of the judiciary and the dominant parties of Italian centrismo had been put in question. The dominance of the conservative higher ranks was undermined by the dismantling of the judicial hierarchy, which produced the additional consequences of further increasing and diversifying the magistrates’ links to political parties, public opinion, and interest groups. On the other hand, the DC’s dominance in Italian governing coalition politics had itself been diminished, with its electoral support dropping to a postwar low of 33 percent in 1983. Both

Democratic Consolidation and Judicial Reform

181

conservative judges and the DC were now confronted with new and powerful competitors—the progressive and activist judges allied to the Socialists and Communists in parliament. The political and ideological congruence between the interests of the dominant actors in the judiciary and the political system had been replaced by a set of opportunistic and mutually beneficial exchanges between multiple judicial factions and political parties. Those exchanges resulted, for instance, in parliamentary generosity in setting judicial salaries (now on average double those of higher civil servants), the satisfaction of most demands for increased judicial independence and powers on the part of the magistrates, and the systematic failure of most judicial investigations into political corruption, which had been going on Italian politics since the inception of the democratic regime but which intensified after the mid-1970s. However, these exchanges between multiple parties and correnti proved to be far less stable than the previous alliance between DC and the judicial higher ranks. This became clear in the 1980s, when the Socialist party assumed an increasingly relevant role as the center-left actor in Italian coalition politics, bringing to it what its leader Bettino Craxi called decisionismo. In practice, decisionismo meant two things: First, an effort to end consociational policymaking practices, or, at least, to limit them to the governing coalition partners, excluding the Communists not only from participation in the executive but also in the broader legislative coalitions and exchanges of which the PCI had been a part in the late 1970s (Pasquino 1996: 148). Second, a more personalistic and confrontational political style (although it was seldom more than just a ‘style’) that placed an emphasis on governability, in opposition to the past policy immobilism that had characterized Italian politics (Hine 1993: 202–4). Inevitably, the judges were ultimately targeted as objects of this confrontation. In 1987, Craxi opened the hostilities between politicians and magistrates by trying, with limited success, to reduce their powers. In this context, he used a popular referendum to abolish the traditional civil immunity of magistrates, in what was interpreted as a negative judgment on the overall performance of the judiciary itself. The second episode occurred throughout 1990 and 1991, when President Cossiga and justice minister Martelli engaged in a variety of different conflicts with the judiciary, motivated by a variety of different cases but always with the systematic intent of restricting the powers of the CSM (Balboni 1992). The third episode of political-judicial confrontation was mani pulite, but it ended up being much more than an ‘episode’. Previously, politicians could exert some influence over the judiciary primarily through

182

Democratic Consolidation and Judicial Reform

the lay members in the Higher Council or by developing the sorts of mutually beneficial exchanges described above. By the early 1990s, however, there was little left to exchange. Having to some extent reduced the role of the Communists in legislative political bargains, the Italian governing coalition was now trying to neutralize the most progressive elements of the judiciary, while the latter were eager for an opportunity to denounce the partitocrazia in which they had a decreasing stake. The opportunity came when that governing coalition was reduced to shambles in the April 1992 elections ‘earthquake’. In those elections, the DC’s share of the vote fell to a postwar low of 29.7 percent, the decade-long increase in the Socialist vote was halted, and a new political force, the Northern League (Lega Nord) emerged with considerable support. Exploiting widespread dissatisfaction with tax increases and the worsening performance of the state administration, and demanding in an unusually aggressive way a federal reorganization of the state aimed at granting autonomy to the northern regions of the country, the Lega Nord became the fourth largest party in parliament, with close to 9 percent of the vote. The combination of these factors plunged the traditional governing coalition into crisis, further aggravated by the resignation of the President of the Republic Francesco Cossiga. A seven-week stalemate followed, where the quadripartito, left with the slimmest of majorities and plagued by internal divisions, was only able to elect a President after fifteen rounds of votes. A governmental coalition was ultimately formed, but it was now weak, uncertain, and totally discredited. This crisis-induced political vacuum was filled by the judiciary. The judges’ activism in the prosecution of corrupt politicians could no longer be checked by the ruling parties, and that activism ultimately led to their near disappearance and to a radical restructuring of the Italian political system.

The Remaining Southern European Cases: The Limits of Judicial Activism In the late 1980s and throughout the 1990s, Portugal, Spain, and Greece also provided examples of the kind of ‘criminalization of political accountability’ that characterized Italy. Starting in 1988, the Socialist government of Spain became a target for media and judicial investigations concerning its alleged ‘dirty war’ against Euskadi Ta Askatasuna (ETA). This slow and cumbersome process ultimately led to the accusation (and ultimately, to the conviction in 1998) of Jose´ Barrionuevo—former Interior minister—for sponsoring the activities

Democratic Consolidation and Judicial Reform

183

of the Grupos Antiterroristas de Liberacio´n, allegedly responsible for the killing of more than twenty ETA members. The GAL case was just the beginning of an incredible succession of cases that involved the highest ranking figures of the party and government (including Prime Minister Felipe Gonza´lez), and played a decisive role in PSOE’s electoral defeat in 1996 (Heywood 1995a; Jime´nez 1999: 80–99). Portugal, usually more low-key even in its political scandals, did not escape this trend. The Social Democratic party, which controlled the executive from 1985 to 1995, was deeply affected by media allegations and judicial investigations of financial corruption, tax evasion by party members, misuse of EU funds and illegal campaign funding. These and other scandals were seized upon by opposition parties and the Socialist President Ma´rio Soares as evidence of the alleged clientelistic practices and lack of accountability of the PSD’s majority government. The image of the party and its leader, Anibal Cavaco Silva, were tarnished, and these scandals contributed to PSD defeats in both the 1995 parliamentary and 1996 presidential elections. Finally, the 1988 ‘Koskotas affair’ in Greece created a political crisis that resulted in the trial of Prime Minister Andreas Papandreou, government ministers and highranking cadres of the Socialist party. Georgios Koskotas, a young entrepreneur who already had a criminal record in the United States, rose from a simple bank employee in late 1970s to become one of the country’s most important economic figures, controlling banks, newspapers and mass-circulation magazines, as well as the most important Greek football team. Koskotas’ rapid rise, combined with the close links he had developed with leading circles within PASOK, upset the equilibrium which had developed, following the fall of the dictatorship, in the shady zone where the mass media are entwined with certain enterprises that undertake major state procurements. His reversal of fortune began when the other major media groups united against him, leading to continual revelations about his illegal activities and forcing the public prosecutor’s office (which until then had demonstrated an inexplicable inactivity) to initiate proceedings. Ultimately, two government ministers were condemned to prison sentences, while Papandreou was acquitted by a small margin. Several scholars have advanced explanations for this peculiarly ‘Southern’ or ‘Latin’ explosion of political scandals and corruption: the exceedingly fast pace of social change after democratization, creating economic and political parvenu e´lites who were (too) eager to take advantage of previously closed opportunities; weakly organized and rooted political parties, with incentives to bypass legality in order to finance their (increasingly costly) competition for power; and changes in ownership, organization and professional standards in the media,

184

Democratic Consolidation and Judicial Reform

just to name a few.39 But why did Italy emerge a singular case among our Southern European countries in terms of the breadth in scope of judicial investigations and their enormous political impact? If our analysis of the particular constellation of institutional opportunities and institutional incentives that led to the ‘Clean Hands’ earthquake is correct, then we have to determine whether that constellation has been also present in the remaining countries. In spite of partial and occasional similarities, the answer is negative. First and foremost, there is no equivalent to the political crisis and weakening of executive power surrounding the April 1992 elections in Italy. In Portugal, the governmental instability that prevailed from 1976 until 1987 period has been replaced by very high levels of cabinet durability: the Social-Democrats secured two consecutive absolute majorities, and governed without interruption until 1995; this was followed by stable Socialist governments supported by near-majority in the parliaments elected in 1995 and 1999. In Spain, the year 1982 marks a profound party-system change that involved the disappearance of UCD and a shift to a markedly majoritarian political system: majority governments were formed by the PSOE between 1982 and 1993, and by the PP between 2000 and 2004; and even the single-party minority governments formed by the Socialists between 1993 and 1996, and the PP between 1996 and 2000 are substantially above the West European average in terms of cabinet durability. Only Greece comes close to have exhibited a similar political crisis, brought about by the ‘Koskotas affair’, resulting in short-lived cabinets between February 1989 and November 1990. However, Papandreou ultimately returned to the political scene in 1993, obtaining a new absolute majority. His death in 1996 and the ensuing leadership crisis within PASOK was also successfully overcome, as Costas Simitis became Prime Minister with the support of a new majority in parliament. In what concerns the judiciary’s organization and institutional insulation, similarities with Italy are in most cases elusive and superficial. The complete lack of ideologically based judicial factions in Greece, for example, is accompanied by a low level of judicial independence, which has favored the political control of the top echelons of the judiciary by the executive. It is true that since the 1990s the Greek judiciary has gradually acquired some autonomy. For instance, the chairpersons of first-instance and second-instance courts are elected from among the judges of the corresponding courts by all the judges of each court. By contrast, the chairpersons of the highest-instance courts (e.g. the Areios Pagos) are selected by the government. However, from a comparative perspective, the insulation of the Greek judiciary from other political institutions is still limited. In this respect, Greece has been the

Democratic Consolidation and Judicial Reform

185

exact mirror image of the Italian case, initially to the benefit of the politically hegemonic ND, but later to the benefit of the Socialists. The major attack on judicial independence occurred in the winter of 1988, when the Papandreou government tried to neutralize the judicial investigation of the economic scandals. Justice minister Agamemnon Koutsogiorgas—who one year later would become the main accused in the trial of the Koskotas Scandal—submitted a draft law to Parliament with the ultimate aim of avoiding, or at least delaying, the convocation of the Athens Appeals Court, which had the responsibility for initiating criminal proceedings concerning financial scandals. The Papandreou government’s draft law entrusted the convocation of the plenary session to the principal of the court, who would be chosen by the Supreme Judicial Council. Given the progovernmental composition of the Council, it was clear that the principal of the Athens Appeal Court would do everything possible to avoid judicial investigation of the scandals. In an unheard-of-event in Greek judicial history, the entirety of professional associations of those working in the judicial system (judges, lawyers, and court clerks) rallied against the draft law, but PASOK’s governing majority passed the law by adopting emergency procedures in parliament. In other less overt threats to judicial independence, the successive governments of Andreas Papandreou from 1981 to 1989 did not hesitate to ignore the seniority criterion for promotions, in order to appoint individuals with well-known prosocialist views to the leadership of the judiciary, a practice that was followed by subsequent governments—both ND and PASOK—in the 1990s. The result has been that every change in the leadership of the judiciary has triggered intense wrangling among the political forces, with the ultimate result of reducing the prestige of the leaders of the judiciary in the eyes of public opinion. An intermediate situation in terms of judicial independence is found in Spain. The shift to a majoritarian political system in 1983 brought about important institutional changes in the judiciary, designed by the PSOE to reduce the influence of the conservative sectors within the judiciary and, ultimately, reduce the political insulation of a powerful and assertive Consejo General.40 The Socialists accompanied this move with changes in the retirement age for judges, blatantly directed at purging the judiciary of its most conservative elements appointed during the earlier years of the Franco regime. The long-term consequences of this far-reaching institutional reform are difficult to evaluate. The choice between a politically appointed CGPJ and one elected by the judges continues to generate discussions about the proper balance between the values of judicial independence and democratic legitimacy of the judiciary.41 However, those theoretical debates hide

186

Democratic Consolidation and Judicial Reform

deeper political undercurrents. The parliamentarization of the CGPJ has strengthened the ties between political parties and organized ideological factions inside the Spanish judiciary, as the former acquired a decisive say in the Council’s composition, and have appointed members who tend to behave as disciplined partisan representatives (Toharia 2000: 109). This, in turn, has had two contradictory consequences. On the one hand, judges investigating alleged cases of corruption and abuse of powers by members of PSOE received extensive support from PP and the media connected to the opposition (most prominently, the newspapers ABC and El Mundo, who placed these investigations at the top of their news agenda and appear to have benefited from extensive and systematic breaches of the secrecy of judicial proceedings. However, on the other hand, the politicization and factionalization of the judiciary (like in Italy) is accompanied by the lack of strong guarantees of independence (unlike in Italy). Throughout the 1990s, the Socialists were accused of controlling a majority of politically subservient members of the CGPJ, and the Consejo’s internal investigations of the behavior of politically inconvenient judges have not helped to change that perception—a perception that the PP and its allies in the media proceeded to use in their efforts to further discredit the PSOE government.42 Besides, the Spanish case remains similar to the Greek with regard to the low level of independence of the public prosecution, which is governed by an executive-appointed Attorney General under strict bureaucratic, centralized, and hierarchical principles. On the whole, the mix of favorable political incentives and unfavorable institutional opportunities for judicial activism has also caused a mixed record in comparison with the Italian case. Public attention was directed toward judicial investigations of politicians, and the reputations of Felipe Gonza´lez and his government were greatly damaged, arguably leading to the PSOE’s defeat in 1996. In actual fact, few of the politicians investigated were ever sentenced, as judicial procedures were in themselves incredibly protracted, and damaging verdicts were often reversed or mitigated by higher courts.43 From a purely institutional point of view, Portugal is probably the case that shares the most similarities with Italy. Changes in the direction of greater majoritarianism in the political system that occurred in the mid-1980s were not followed by a reduction of judicial insulation vis-a`-vis the executive. Unlike the PSOE, the PSD failed in its successive attempts to ‘parliamentarize’ the CSM, given the opposition of the Socialists to constitutional amendments and the vigilance of the Constitutional Court. At the same time, however, even though the prosecution remains organizationally separated from the judiciary, the independence and powers of the Portuguese Ministe´rio Pu´blico

Democratic Consolidation and Judicial Reform

187

have increased substantially during the 1980s, representing a substantial departure from the model (that Portugal once shared with Spain or Greece) of a prosecution integrated into the administrative apparatus. The extraordinary corporate activism of prosecutors and the changes in criminal procedure—that entrusted prosecutors with the responsibility for criminal investigation—brought about increasing similarities between the institutional setting and status of the judiciary and the prosecution, as in the Italian case. Moreover, control of the prosecution by political majorities is precluded by the need for a joint consensus between the President and the executive over appointment and dismissal of the Attorney-General, together with the fact that the presidency and the cabinet were controlled by different parties until 1996. These similarities should not, however, obscure the deep and important differences between the two cases. Unlike in the Italian case, Portuguese judges have exhibited an almost complete a lack of factionalism and politicization of its personnel, who have persisted in interpreting judicial independence as a condition for corporate and bureaucratic self-rule and judicial passivity. Successive governments have adequately rewarded this tendency by answering favorably to most of the ‘corporatist’ demands of the judges.44 Instead, it is in the Ministe´rio Pu´blico that we find greater affinities with the legal and political culture of the Italian judiciary. An ‘expansive’ and ‘progressive’ understanding of their role in the administration of justice has been defended since 1974 by the young magistrates who today fill the top echelons of the prosecution. This, together with the eight-year period of majoritarian cavaquismo, helped make them allies of the Socialist and Communist opposition (inadvertent or not), and turned AttorneyGeneral Cunha Rodrigues into one of the most prominent public figures in the country. However, the opportunities and incentives for prosecutorial activism have undergone important changes. On the one hand, starting with the 1995 election, the minority and coalition governments that followed caused a shift toward more consensual and bipartisan practices of governance, and thus to a decline in public and political support for judicial or prosecutorial activism. On the other, such support has also declined as a result of both the ‘corporatist wars’ in which judges, prosecutors, and police officers have been involved since 1995, and of the complete incapacity of the Ministe´rio Pu´blico to effectively bring to justice politicians and officials investigated in the past. At the time of this writing, it is still unclear whether the Socialist Party’s absolute majority in the 2005 elections will result in any renewed institutional opportunities for activism on the part of the Portuguese public prosecution. Therefore, the similarities among the Southern European cases do not seem to be sufficient for the replication of the wave of judicial activism that characterized Italy in the early 1990s. Greece, Portugal,

188

Democratic Consolidation and Judicial Reform

and Spain have not remained immune to the general trend toward an increasing significance of law and courts shared by most contemporary democracies. Political and social conditions throughout the 1980s, particularly the long tenures enjoyed by single-party majorities, have favored the emergence of opposition support for investigations of corruption and abuses of power. Judges and prosecutors have begun to reevaluate their public role, with the growing media attention leading to the emergence of ‘judicial entrepreneurs’ like Baltazar Garzo´n, Cunha Rodrigues and others, while the Italian case provided a case that magistrates elsewhere have found worthy of emulation. However, the activism of investigating judges and prosecutors, in Greece, Portugal, or Spain has lacked the simultaneous sense of ‘crusade’ and the effectiveness Italian judges have achieved. The lack of institutional opportunities (strict judicial and prosecutorial independence) and/or political incentives (politicization of magistrates and political stagnation and crisis) appear to account for this fundamental difference.

Conclusion At the beginning of this chapter, we proposed to answer several questions about democratic judicial reform and the judicialization of politics in Southern Europe. What explains that, in contrast with the East European cases, judicial reform was achieved in the absence of purges of judicial personnel, deep and divisive political conflicts, or the endangerment of the legitimacy of judicial institutions? And, in spite of these similarities, what explains the differences among these countries in terms of the institutional design of their judiciaries? Starting with the answer to the second of these questions, what we find is that Italy, Portugal, and Spain break decisively with their civil law tradition, at least with regard to the high level of independence and institutional insulation awarded to their democratic judiciaries. On the one hand, although the maintenance or increase of executive powers vis-a`-vis the judiciary would have been the most advantageous outcome for those parties victorious in the first free elections, that solution was made impossible by the requirement of supermajorities in the constituent assemblies, together with the obvious opposition of minorities to such an arrangement. On the other, continuity with the civil law model may not only have been impossible but also contrary to the best interests of those political parties. First, because in contexts of heightened political uncertainty, as is characteristic of all democratic transitions, enshrining the executive’s control over the judiciary in constitutional rules means that future governments under different

Democratic Consolidation and Judicial Reform

189

political parties will be handed an extremely powerful resource that they can easily turn against the creators of those rules. And second, because for the Italian Christian Democrats, the Spanish UCD, and the Portuguese centrist parties, insulating the judiciary had an additional advantage, since the dominant conservative groups in the judiciary were, to a great extent, close to their fundamental political interests and preferences. The exceptions to this break with civil law traditions were, respectively, Greece and, in the 1948–53 period, Italy. In Greece, ND dominated a constituent assembly with no immediate elections in sight, while the Italian Christian Democrats emerged from the 1948 elections with a surprising victory, establishing themselves as the hegemonic party of posttransitional Italy. This led, in both cases, to a freezing of judicial institutions in accord with the previous civil law model of executive control. In Italy, judicial independence was finally implemented as soon as DC lost that hegemonic position, and the country became one of the prototypical examples of a consensual political system (in Arend Lijphart’s sense of that term). In Greece, however, the succession of ND and PASOK majority governments created few incentives for executives to abdicate from their direct or indirect control over judicial appointments and supervision. The story of judicial reform in Southern Europe cannot, however, be entirely explained in terms of political self-interest. Elsewhere, the continuity in office of judges who had served the authoritarian regimes was perceived as an attempt to preserve reserve domains of power in the hands of authoritarian elites or specific political actors, and therefore as serious threats to democracy (Melone 1996; Magalha˜es 1999). This was not the case in Italy, Portugal, or Spain. This was a result of the legacy of lower politicization, the shared professional and technical legal standards inherited from the authoritarian regime, and the limited pluralism existing inside the judiciary. It was also a product of the consensual establishment by political actors of institutions that, while responsive to majorities broader than those supporting the government, were able to monitor and control the decisions of mistrusted ordinary courts and judges. Constitutional courts were those ‘transitional devices’, and their contribution to the protection of political rights, the legitimacy of legal institutions, and the consolidation of democracy could hardly have been achieved if judicial review had been entrusted to the ordinary judiciary.45 Again, Greece and Italy (for a short period) were the exceptions. In the Italian case, as we have seen, it is clear that delay in the institutionalization of the Constitutional Court led to a less than satisfactory protection of political and civil rights in the period between 1948 and 1956. It is true, however,

190

Democratic Consolidation and Judicial Reform

that the absence of centralized judicial review in Greece seems to have been less problematic. The short length of the colonels’ regime (seven years) meant that it did not have time to exert a deep influence over the ethos or perceptions of the Greek judges. And, as we saw before, the institutional insulation of the career judiciary in Greece has not taken place to the same extent as in the other three countries. Thus, the need to create a constitutional court that would be appointed by Parliament and guarantee the political responsiveness of the judicial power has not been so intensely felt. In Greece, the system of appointment of the leadership of the judiciary guaranteed such responsiveness. However, the question remains whether that ‘responsiveness’ has not been excessive and too directed at the political party that happens to be in control of the executive. Can we speak of clear similarities between the outcomes of judicial reform in Southern Europe? To some extent, and discounting the resilient Greek exceptionalism, the answer is yes. The traditional civil law judge enters the judiciary through competitive public examinations immediately after university graduation, and remains in that career throughout her or his entire working life. However, perhaps because the initial examination is not believed to be sufficiently reliable, civil law judges remain subject to various forms of control that limit her or his independence, including continuous evaluations from higher ranking colleagues or, in certain cases, direct monitoring by the Minister of Justice. In Italy, Portugal and Spain, the distinguishing features of traditional civil law judiciaries have all but disappeared. The institutionalization of judicial councils triggered a multifaceted process of change that has deeply affected the relationships between courts and politics, as well as the internal organization of the judicial corps. First, it brought an unprecedented increase of external independence vis-a`-vis the elected branches, which in turn created opportunities for a more daring intervention by judges and prosecutors in the political process. Second, entrusting the task of promoting and appointing judges to collegial bodies such as the judicial councils—where normally all judicial ranks are represented—collided with the traditional hierarchical principle, according to which only higher-ranking judges are able to evaluate their lower-ranking colleagues. Thus, as the internal independence and individual autonomy of judges grew significantly, judicial hierarchy began to erode. However, we would probably have to discount much more than Greek exceptionalism in order to speak of a distinctive ‘Southern European’ type of judiciary. There are important differences among our cases, as the preceding discussions of judicial protagonismo and political scandals made clear. Moreover, some of these developments cannot be said

Democratic Consolidation and Judicial Reform

191

to be unique to Southern Europe. Even France, the prototype of ‘jacobin’ parliamentary sovereignty and civil law legalism, has recently begun to exhibit several of the ‘distinctive’ features we have identified here as breaks with the civil law tradition: the reduction of the powers of the Minister of Justice, the increased independence of both judges and prosecutors, and their growing intervention in the prosecution of prominent politicians and businesspeople.46 Several issues regarding the judicialization of politics in Europe in general and Southern Europe in particular are likely to dominate future research in this area. The preliminary question is whether the type of judicialization that has prevailed in Southern Europe, centered on the criminalization of political responsibility, will continue to develop, perhaps spreading its influence to other civil law nations (as appears to be happening in France or Belgium) or not (as in Germany). The second issue concerns a different manifestation of the judicialization of politics, that is, judicial policymaking and the intervention of courts in the legislative process. In this respect, attention to judicial review of legislation and the role of constitutional courts is required. This phenomenon remains a largely unexplored area of research in these countries from a political science perspective, unlike what has already happened, for instance, with the French and German cases.47 The political significance of the Italian constitutional court has already been underlined and to some extent analyzed.48 However, the policymaking role of such obviously important and powerful constitutional courts such as the Portuguese or the Spanish remains virtually unexplored. Both manifestations of the judicialization of politics raise important normative and empirical questions concerning the quality of democracy. Can these developments be defined as ‘favorable’ to democratic rule? While modern democratic theory accepts or even prescribes limitations to the political power of electoral majorities, the proper extent of these limitations is not clearly determined. The judicialization of politics also raises questions regarding the reactions of the political class. How will politicians take into account the growing power of the judiciary? Will Southern European political actors try to control judicial appointments more closely—following the American model—in lieu of the century-old tradition of technical, bureaucratic recruitment? Or will politicians continue to rely instead on the more or less informal practice of bargaining with the representatives of the judiciary (and, in particular, with judicial associations) in their efforts to influence judicial decisions? Will this practice foster phenomena already found in Italy and Spain, where the judiciary is divided along more or less open political lines? On the other hand, we have also provided evidence that

192

Democratic Consolidation and Judicial Reform

the extent of judicialization of politics may have been overestimated, and that its growth is very much dependent on specific political contexts, particularly with regard to the adoption of majoritarian or consensual norms and practices. Finally, will the availability of other, more efficient, channels of political participation substantially reduce the political role of courts in these democracies? Further study of these and related issues concerning the role of the courts in democratic polities will help to fill significant gaps both in our knowledge of contemporary political systems and in democratic theory.

Notes 1. We would like to thank the editors of this volume for their insightful comments to earlier drafts of this chapter and for their guidance during the long process that led to this final version. We would also like to acknowledge the comments and suggestions received of the participants at the Istanbul SSRC conference (June 1994), the initial contribution of Anto´nio Vitorino, involved in the earlier stages of this research, and the useful comments and suggestions received from several people who read the entirety or parts of earlier versions of this chapter: Paloma Aguilar, Miguel Lobo Antunes, Anto´nio Arau´jo, Gregory Caldeira, Alison Rimsky, Sara Schiavoni, and Joa˜o Ramos e Sousa. During the period of research that led this chapter, we received generous support from the Centro de Estudios Avanzados en Ciencias Sociales of the Juan March Foundation, the Department of Organization and Political System of the University of Bologna and the staff of the Portuguese Constitutional Court. Finally, we are deeply indebted to all the judges, prosecutors, and jurists in each country that gave up some of their precious time to answer our insistent queries. Their anonymity is preserved, but without their contribution this chapter would not have been possible. Pedro Magalha˜es did most of the work for this chapter as a Ph.D. candidate at the Department of Political Science of the Ohio State University, where he was a fellow of the Calouste Gulbenkian Foundation and the Fulbright Commission in Lisbon. 2. See, among other works, Linz and Valenzuela 1994; Lijphart and Waisman 1996; Heper, Kazancigil, and Rockman 1997. 3. A conceptual distinction between ‘rule of law’ and ‘rule by law’, where the latter is associated to the notion of Rechtsstaat, can be found in Reitz 1997. 4. For a brief and excellent discussion of the issues raised by the role of judicial review in democratic theory and the problem of judicial ‘quasiguardianship’, see Dahl 1989: 187–91. 5. See, for example, Kommers 1976; Landfried 1988; Stone 1992, 2000; Shapiro and Sweet 1994; Shapiro and Sweet 2002. 6. The contrast between ‘legacies of the past’ and ‘new institutions’ approaches is most clearly advanced in Crawford and Lijphart 1995.

Democratic Consolidation and Judicial Reform

193

7. See Linz, Stepan, and Gunther 1995: 81–3; and Linz and Stepan 1996: 55–65. 8. Although much of the so-called ‘transitions’ literature emphasizes rationality and interest maximization of individual political actors, this approach is most clearly subscribed in works such as Di Palma 1990 or Przeworski 1991. Empirical applications of this perspective to institutional choices in new democracies can be found in Geddes 1995; Frye 1997; Negretto 1999; Benoit and Schiemann 2001. Theoretical discussions of this perspective on institutional design can be found in Moe 1990; Knight 1992, 1995. 9. Knight 1995. In the late 1930s, magistrates were compelled to enroll in the Fascist party. However, Justice Minister Grandi wanted them all inside the Rome’s organization, he said, to protect them from the pressures of local party bosses, and therefore, safeguard their impartiality. 10. In Greece and Portugal, judicial review had functioned very much in the ‘concrete all-courts’ American mode since, respectively, the late nineteenth and early twentieth centuries. As for Spain, judicial review appeared as part of the first stage of diffusion of the so-called ‘Kelsenian’ model, with the creation of a specialized court, the Tribunal de Garantı´as Constitucionales [Tribunal of Constitutional Guarantees] that enjoyed the monopoly of constitutional jurisdiction and abstract review of laws during the Spanish Second Republic. See Spiliotopoulos 1983; Costa 1989; and Rousseau 1992. 11. At the time, and until democratization, access to the judiciary was restricted to males. 12. Sajo´ and Losonci 1993; Heydebrand 1995; Reitz 1997. 13. The Supreme Civil and Criminal Court (Areos Pagos), the Supreme Administrative Court (Council of State—Symvoulion tis Epikratias), and the Court of Auditors (Elengtiko Synedrio). 14. Until the end of the 1950s, the Italian judiciary was governed mainly by the higher-ranking magistrates, acting very often with full agreement of the Ministry of Justice, thanks also to the fact that magistrates were in charge of all important positions in that ministry. This collaboration is well exemplified by the case of a high magistrate (president of a panel of the Court of Cassation) elected to Parliament in 1948 on the DC ticket, who resigned in 1952 in order to be appointed—by the government—President of the Court until his retirement two months later, but elected again to Parliament at the 1953 elections and shortly after appointed Minister of Justice. See Di Federico 1976; and Di Federico and Guarnieri 1988. 15. Under the circumstances, the institutionalization of the Consiglio was no great defeat for the left either, that increasingly perceived the potential of a judiciary protected from explicit governmental control as soon as ties between progressive and activist sectors of the judiciary and the leftist parties emerged, particularly between the politically excluded Communists and Magistratura Democratica, one of the correnti born in the early 1960s. See Pappalardo 1987. 16. To make sure that would be the case, the small print in the legislation that created the CGPJ reinforced the authority of that conservative faction and

194

17.

18.

19. 20. 21.

22.

23. 24. 25.

Democratic Consolidation and Judicial Reform

of the higher ranks of the judiciary, precluding the constitution of regional associations and defining a rather high minimum percentage of membership for the formation of national ones, with the obvious purpose of limiting the influence of Justicia Democra´tica. The introduction of a majoritarian system for the election of CGPJ members, the strict incompatibility between judicial and political functions, and the hierarchical criteria in the formation of the CGPJ, with a system of quotas that overrepresented the higher ranks, can also be seen in the same light. See Andre´s Iban˜ez 1986. Its competencies includes the review of: (a) elections to the national and European parliaments, (b) the results of referenda, and (c) incompatibilities of office leading to forfeiture of their seats by Members of Parliament. The Court also settles: (a) conflicts of competence between courts and administrative authorities, and (b) controversies related to the designation of rules of international law, as generally acknowledged. The Court of Cassation, equivalent to the Portuguese or Spanish supreme courts or the Greek Areos Pagos, is the highest level of appeal in the judicial system. The rationale for leaving authoritarian legislation in force was introduced by the Court in 1948, when it distinguished between ‘programmatic’ and ‘preceptive’ constitutional norms, and among the latter between those requiring additional legislation and those that could be immediately enforced. Only the latter were deemed binding for the judged. UCD and the Socialists of PSOE until 1983, and PSOE and AP/PP (Alianza Popular, later renamed Partido Popular, a conservative party) thereafter. See Tranfaglia 1975: 189; Zagrebelsky 1988: 481; and Hine 1993: 162–5. In fact, other mechanisms of hierarchical control had been proposed in the constituent assembly, like a disposition strongly defended by PSOE’s Gregorio Peces-Barba that would formally allow the court to explicitly review the jurisprudence of the lower courts in Spain. This arrangement was abandoned in the final stage of constitutional drafting. See Rubio Llorente 1993. When one of the parties in the trial unsuccessfully raises the question of the constitutionality of a statute, it becomes possible to make a direct appeal to the constitutional court. However, that possibility becomes a certainty when any of the ordinary courts decides not to apply a major regulatory instrument (law or decree) on grounds of its unconstitutionality when deciding upon a concrete case. The reason is simply that, in these cases, appeal to the constitutional court becomes compulsory for the public prosecutor in that particular judicial district. The President (General Ramalho Eanes), the Chiefs of Staff of the Armed Forces, Navy, Army, and Air Force, and fourteen other officers. In Portugal, the executive enjoys legislative powers in some issue areas, conditioned to parliamentary ratification, if requested. The French Conseil Constitutionnel, for instance, has repealed approximately 10 percent of all parliamentary bills sent for promulgation from 1981 to 1992. See Stone 1992.

Democratic Consolidation and Judicial Reform

195

26. In a priori review decisions made by the CC between 1976 and 1982, 63 percent were unanimous and 26 percent were supported by two-third majorities. In a posteriori review the percentages were roughly the same (65 percent and 21 percent, respectively). 27. The most striking example of this is ETA, a terrorist militia formed during the Franquist regime that quickly shifted its targets from authoritarian officials to the military, police forces, and democratic politicians during the transition, and continues to be the thorn on the side of Spanish democracy. 28. Gunther, Montero, and Botella (forthcoming). For a different view favoring the federal nature of the Spanish system see, for example, Linz 1997. 29. The option for a generalized though heterogeneous system of autonomic competencies was in itself a compromise, since although the ‘stateness’ problem was mainly a Catalan and Basque one (and to a certain extent, Galician also), a system accessible to all provided for some degree of rationalization and control that rendered it more acceptable to less decentralization-oriented political forces. See Moderne 1987: 61–84. 30. Sources: Campo 1998: 202; and statistics from 1997 produced by the Gabinete Te´cnico del Presidente de Tribunal Constitucional. 31. Pedro Cruz Villalo´n (who became the Court’s president in 1999) describes the first years as ‘moderately centralist’ (Cruz Villalo´n 1991), while Juan Jose´ Soloza´bal Echavarrı´a (1991) points to the Court’s ‘obsession’ with ‘the essential unity of the Spanish legal order’. 32. Following the acute judicialization of conflicts between central and regional governments in the 1980s, the last decade has witnessed a marked decline in that kind of litigation. In 1997, for example, the number of requests to solve conflicts of competencies between the central and the regional governments had dropped to ten, very far from the eighty-two and ninety-five requests received by the court, respectively, in 1985 and 1986. Litigation in abstract review initiated by the central government or the comunidades against each other’s legislation has also decreased, never to reach again the incredible figures of 1985 (forty-eight requests). 33. Colomer (1999: 48–52) discusses that absence and the inadequacy of the Spanish Senate to perform such role. 34. After receiving only twelve requests for abstract review related to centralregional conflicts in 1996, the court received thirty-nine of such requests in 1997, one of the highest figures ever and a complete reversal of the previous trend of decline in litigation. 35. Antonio Di Pietro, the most famous Milan prosecutor, was offered the control of the Ministry of Interior by Berlusconi (a key portfolio, since it is in charge of police and internal security). 36. Also Portugal and Spain have in their Codes of criminal procedure the same principle, but its implementation seems to be less cogent. See Van den Wyngaert 1993. 37. A sign of this growing power has been the influence magistrates have been able to exert on police forces, which today tend often to act only under direction by the prosecution. See Di Federico 1995: 233–42.

196

Democratic Consolidation and Judicial Reform

38. See Fabri 1994: 211–16; Ferrarese 1994: 653–66; Chiavario 1995. 39. See, for instance, Della Porta 1992; Della Porta and Meny 1995; Pujas and Rhodes 1999: 41–63. 40. In 1985, the Socialists approved the so-called ‘Bandre´s amendment’, that specified that all of the Council’s members, including the twelve judges, were now to be designated by a three-fifth majority of Congress. What had been a body dominated by judges connected to the conservative sectors of APM became in 1985 the result of a pact between the Socialists and its Basque and Catalan allies in parliament on the one hand, and Alianza Popular on the other, with the former keeping the lion’s share of appointments. 41. For a recent analysis by one of the foremost scholars on the judiciary in Spain, see Toharia 2000. 42. See Tassara 1987: 157; Heywood 1992: 735. For two recent and (predictably) different positions on an alleged politically motivated action against a prominent judge connected to PP, see ‘Editorial: Una Canallada’, El Mundo, October 16, 1999; and Pe´rez Royo 1999. 43. For a recent chronicle of ongoing developments on many of those judicial cases, some of them concerning events that occurred as far back as the early 1980s, see de La Cuadra 2000: 208–9. 44. For example, between 1989 and 1997, the judiciary was the only body of public servants whose wages have grown at a rate well above inflation. 45. For a similar argument favorable to centralized constitutional jurisdictions in East European cases, see Teitel 1994. For a defense of an American-style system of judicial review for the new East European democracies, see Utter and Lundsgaard 1994. 46. See Garapon 1996; Garapon and Salas 1996; and Salas 1998, esp. 77. In recent years two acting ministers—Carignon and Longuet—have resigned after being placed under judicial investigation. 47. See footnote 9. 48. See Zagrebelsky 1988; Cheli 1996; De Mucci 1996; Volcansek 2000.

................ 5 ................

Old Problems and New Challenges: The Enduring and Changing Functions of Southern European State Bureaucracies1 Dimitri A. Sotiropoulos

At the end of the twentieth century, southern European democracies had both to face old problems relating to the long overdue effort to build accountable, transparent, and meritocratic bureaucracies, and to meet new challenges stemming from a transformed international order. The latter included the end of the Cold War era, globalization, pressures to retrench state intervention in many policy areas, introduction of new technologies and management methods, and the requirements of accession to the EU. Some of these challenges were familiar to all West European bureaucracies (Wright 1994: 104–7). The modern bureaucracies of Greece, Italy, Portugal, and Spain have developed along the lines of the Napoleonic state tradition, but they have also differed from this tradition. Other Western and Northern European bureaucracies had become modern and, with the exception of the interwar period, accustomed to function within democratic regimes. In the final decades of the twentieth century, they embarked on a process of adaptation to the diffusion of new technologies and the intensification of international economic competition. Southern European bureaucracies did not have the time to proceed in the same manner, that is, by meeting the aforementioned successive challenges in a stepwise fashion. Instead, the bureaucracies of Greece, Portugal, and Spain had to confront all of these tasks simultaneously—organizational modernization, democratization, and adaptation to new technologies and methods of management, to European integration, and to new international environment. Because of

198

Southern European Bureaucracies

their delayed economic development and, especially, democratization and adhesion to the EU, the state mechanisms of Greece, Portugal, and Spain had no alternative but to confront all of these challenges at the same time. Italy was and still is much different. From the point of view of this chapter, it may be placed at the crossroads between Western and Southern Europe. This is so since Italy made its transition to democracy at the end of World War II along with other West European countries, whereas Greece, Portugal, and Spain followed several decades later. Italy followed the postwar economic boom of most West European countries more closely than other Southern European economies. This achievement placed it among the most advanced economies of the world in the 1980s and the 1990s, while Greece and Portugal (and, less so, Spain), lagged behind other EU economies. The larger Italian public sector had become important for the economy much earlier than the corresponding sectors of Greece, Portugal, or Spain. And Italy preceded Spain by initiating its administrative decentralization in the early 1970s, through the creation of twenty regional authorities (whose performance has varied considerably according to Putnam 1993). This decentralization into numerous provincial and communal authorities clearly distinguished the evolution of the Italian state from the timid attempts of the Greek and the Portuguese states to decentralize. Nonetheless, in the twentieth century the Italian state and political system in general shared enough common administrative and political traits with their Greek, Portuguese, and Spanish counterparts to merit inclusion alongside them in the Southern European group. One similarity was its extensive period of authoritarian rule. The Italian central public administration, like its Southern European counterparts, also resisted several attempts at reform since the end of the War (Cassesse 1993a: 326). Political clientelism in Italy was similar to patterns of state–citizen relations witnessed in Greece, Portugal, and Spain. Finally, the Italian state functioned for a long time as a ‘social shock absorber’ (Spence 2000: 135), employing people who could not or would not find a job in the labor market. This was a function all too familiar in Greece, Portugal, and Spain under different political regimes. We would like to argue that, even though Italy is in many respects a very advanced West European economy and polity, its state bureaucracy has remained Southern European. In order to discuss differences and similarities among the four Southern European bureaucracies and their recent evolution, this chapter addresses the following questions: Is there a distinct traditional type of Southern European state bureaucracy and how did it evolve over time

Southern European Bureaucracies

199

during the last quarter of the twentieth century? More specifically, what were the effects of democratization on the state bureaucracies of Greece, Italy, Portugal, and Spain, and how have state bureaucracies changed after consolidation was completed? This chapter does not address either the topics of privatization of the public sector or Europeanization of public administration or administrative decentralization, although, when necessary, these aspects are included in the analysis. The focus of the chapter is on the central public administrations of Southern Europe from the period of transition to and consolidation of democracy through the late 1990s. We use the conventional dates as benchmarks for the consolidation of these regimes: the early to mid-1980s for Greece, Spain, and Portugal (Linz, Stepan, and Gunther 1995: 96, 107, 113), while we regard the case of Italy as being weakly consolidated by the late 1950s (Morlino 1995: 376), with full consolidation being achieved in the late 1970s, when the effective exclusion of the PCI from the Italian political system ended.

The Historical and Sociopolitical Context of Southern European Bureaucracies Over the past two centuries, the formal structure of the state bureaucracies of Greece, Italy, Portugal, and Spain was not substantially different from that of their North and West European counterparts. In the nineteenth century the structure of the four bureaucracies was influenced by the master plan of Napoleonic administration. As is well known, this system involved centralization of authority, unity of command, adherence to the liberal principle of equality before the law, and recruitment of administrative personnel on the basis of merit, usually through a competitive entrance examination (Beltra´n 1988: 256). Over time, and particularly in the final quarter of the twentieth century, the above structure was formally modified in the direction of granting greater administrative autonomy to the periphery of these countries, although to a much greater degree in Spain and Italy than in Greece and Portugal. The trend toward decentralization was coupled in all four countries with changes in the size and the outlook of the wider public sector, resulting from fiscal constraints, the international prevalence of ideas favoring a more limited state intervention in the economy, and the requirements of integration into the EU. This externally induced convergence of public administrations was to some degree common to all EU member states. Overall, by the end of the twentieth century, along with the consolidation of democracy in Southern Europe, domestic and external influences have made

200

Southern European Bureaucracies

Southern European states more similar to the rest of Western Europe (Gunther and Diamandouros 2001: 398). While this trend holds for most features of the political systems of the Southern European societies, it seemed less marked in their administrative subsystems. These seemed to have changed less than other subsystems, such as, the economy or the party system. Some of the traditional traits of the Southern European state bureaucracies, such as their latent function as employers of excess labor, may indeed have faded away, probably owing to the streamlining of the macroeconomic policy of EU member states. However, other traits, such as their inefficiency or, in some cases, the absence of an administrative elite, seem to endure. Until the end of the twentieth century, similarities between Southern European bureaucracies and their northern counterparts were formal rather than substantive: they shared formal Napoleonic structures, but these did not extend to certain subterranean, substantive politico-administrative traits that were typical of Southern Europe and familiar to South Europeans. Public opinion data indicate that contact with the public administration leads to widespread feelings of frustration and dissatisfaction among citizens in all four countries. Poll data collected in 1985 in Portugal, for example, show that Portuguese citizens believed that central public services had worsened since 1974 (Graham 2001: 217–19). Southern Europeans have often felt more like subjects than as citizens endowed with constitutionally guaranteed rights. To some extent, such beliefs may be attributed to extensive periods of authoritarian rule. The extent and variety of ‘bureaucratic pathologies’ in Southern Europe have extended well into the democratic era. While discrimination, illegal administrative decisions, red tape, corruption, inertia, and delays in the delivery of state-administered goods and services are not unique to Southern Europe, such symptoms of dysfunctionality have been more pronounced in Greece, Italy, Portugal, and Spain than elsewhere in contemporary Western Europe.

A Southern European Model of Bureaucracy? Seven common characteristics, first visible in the nineteenth century, have distinguished the Southern European bureaucracies under both authoritarian and democratic forms of government from those found elsewhere in Western Europe (Sotiropoulos 2004). (1) Political clientelism ‘at the top’: The first trait is the one that most observers immediately associate with Southern European states:

Southern European Bureaucracies

201

extensive and enduring political clientelism. By clientelism, we mean ‘a particular mode of social and especially political organization, whose typical structural element and characteristic building block is the patron-client dyad’ (Mavrogordatos 1983: 5). During the last quarter of the twentieth century, the old person-to-person clientelism in Southern Europe was probably replaced by party-led clientelism as far as appointments at the top echelons of the bureaucracy are concerned. Clientelism in promotions and transfers to high-ranking civil service posts has assumed a more organized form and is engineered by party organizations. (2) Political clientelism ‘from below’: This second trait involves the relation between political parties and society. This relation, which is mediated by party mechanisms, facilitates the selective recruitment of individuals at entry-level jobs in the bureaucracy. Through the intermediation of parties, the public sector used to fulfill a particular latent social function in Southern Europe. This was the function of alleviating social pressures from below—from unemployed, unemployable, or professionally insecure social categories (including graduates of political science and humanities faculties, high-school graduates without university education, internal migrants, etc.). In this sense, the public administration functioned as a social shock absorber by offering such individuals job opportunities in the public sector, particularly during periods of rising unemployment or just before the conduct of general elections (the ‘political-electoral cycle’). (3) The uneven character of the public sector: The third trait, possibly associated with the social shock absorber function discussed above, is that Southern European bureaucracies were characterized by an uneven distribution of personnel and other resources. Southern European bureaucracies were overstaffed in some quarters, owing to successive waves of recruitment by alternating political regimes and parties, while at the same time they were understaffed in other administrative domains. The Italian postal service and the Greek state airlines, for example, have an excessive number of employees, while regional public services have remained understaffed. (4) Legal rigidity and excessive legalism: The legal principles regulating all administrative actions are applied in Southern European states in an excessive and fragmented manner which has led to an overproduction of laws and decrees. This is the trait of legalism which is probably associated with compartmentalization of regulations and formalism—that is, with a longer than usual distance between what the law stipulates and what really happens. In Southern Europe, this distance between legal provision and eventual outcome does not simply reflect the problems of implementation faced by every modern state.

202

Southern European Bureaucracies

In that area of Europe, the distance is to a certain extent the result of the superimposition of Western political and administrative institutions on premodern, agrarian societies during the nineteenth century. It may also be the result of the informal workings of all sorts of interests, related to social class conflicts and local cultures. (5) Lack of an institutionalized administrative elite: In Southern Europe, particularly in Greece and Italy, there is no such thing as a typical European administrative elite (Cassesse 1993b: 336;Sotiropoulos 1993: 47, 53). This trait may not apply to Portugal and, especially, Spain where higher civil servants enjoy some power and prestige, and are often members of the social elite of their respective societies. But it certainly is the case with Greece and Italy. Attempts to construct an administrative elite—as with the dirigenza in Italy, and efforts in the early 1980s to found elite national schools of public administration in both Greece and Italy—have failed. As a consequence, there is no equivalent to the British or the French higher officialdom, with its considerable political influence and social status. (6) Perceived administrative inefficiency: The sixth trait is the inefficiency and sluggishness of Southern European bureaucracies. However, given that all large, complex organizations today suffer from some inefficiencies, it is difficult to assess the extent of Southern European bureaucratic inefficiency in comparative perspective. Attitudinal data from the early 1980s show that bureaucratic efficiency was perceived to be much higher in the United Kingdom, France, and Germany than in Spain, Italy, Greece, and Portugal, in that order (Christopoulos 1998: 4). Indeed, public hospitals in Greece and the south of Italy have acquired such a bad reputation that local residents sometimes prefer to be hospitalized in other countries or regions, or in private hospitals. The performance of the Italian postal service also remains unpredictable, while parts of the Greek railway network are still grossly underdeveloped. Throughout Southern Europe, including Spain and Portugal, inefficiency is still widely regarded as a typical pattern on which there has been little improvement over time.2 (7) Widespread political corruption in the lower ranks of the administration: Finally, corruption at the lower echelons of the bureaucracy is abundant in Southern Europe, although it is by no means of East European or African proportions. Corruption may take two forms: one is linked with white-collar crime and the promotion of big-business interests; the second, ‘petty’ corruption, is related to strategies of the lower classes and their encounters with low-ranking bureaucrats (De Sousa Santos 1986), such as is manifested in the disproportionate number of disability pensions issued in Greece and Italy by state-run medical committees.

Southern European Bureaucracies

203

The last two characteristics reflect common perceptions about South European bureaucracies. The extent to which they are real is debatable. Taken together, the above seven characteristics may constitute a model of Southern European bureaucracy, which is useful at least as a sketch of the bureaucratic contexts within which democratization processes have unfolded in Greece, Italy, Portugal, and Spain.

Southern European Bureaucracies During Democratic Transitions and Consolidation A democracy is consolidated ‘when all politically significant groups regard its key political institutions as the only legitimate framework for political contestation, and adhere to the democratic rules of the game’ (Gunther, Puhle, and Diamandouros 1995: 7). During the transition to and consolidation of democracy, the state bureaucracy has three options: it may play a positive, negative, or a neutral role in these democratization processes. Some bureaucrats whose careers and interests had been associated with the deposed regime may choose, out of loyalty to the nation or out of pure opportunism, to work with the representatives of the new democratic regime. Other bureaucrats, who stand to lose a lot (e.g. power, status, even opportunities for personal enrichment) because of regime change, may opt for resistance to incoming democratic elites, particularly during the first tenuous and uncertain phase of transition when the tide has not yet turned decisively against authoritarianism. Alternatively, bureaucrats may assume a neutral stance, based on a ‘wait-and-see’ attitude, stressing the importance of their technocratic or managerial skills useful to any regime, or their ability to facilitate the continuity of the state during unstable periods of regime change.3 Political elites who lead the process of democratization face a dilemma with regard to bureaucracies inherited from the authoritarian period: should they ‘cleanse’ the ranks of bureaucracy of possible antidemocratic elements, or should they try to co-opt them into the new regime? If they choose the first option, they may eliminate a possible source of threat to the emerging democratic regime and may proceed more quickly with administrative reform. However, they would simultaneously lose any acquired organizational experience and technocratic skills which the dismissed bureaucrats would offer to the new democratic governments. They may create new lines of cleavage and conflict at the very heart of the state, and add a new source of instability to the democratization process which, by its very nature, is already fraught with uncertainty. If democratic political elites choose the

204

Southern European Bureaucracies

second option, the continuing presence of prodictatorial elements in the bureaucracy would constitute an obstacle to administrative reforms threatening the interests of the entrenched bureaucratic groups, and might pose a challenge to the governability or even legitimacy of the new regime. In the four Southern European cases, the responses of bureaucracy to regime change and the strategies of incoming democratic elites toward incumbent bureaucrats have varied, depending on the type of regime transition and the particular political context. In historical institutionalist terms, bureaucratic structures, norms, and patterns of behavior, which were dominant in the authoritarian past, limited the scale of administrative reform during the period of democratic transition and consolidation. Greece: The Greek transition resembled those of Italy and Spain insofar as democratic elites chose to co-opt rather than to purge proauthoritarian elements in the state bureaucracy. The authoritarian regime of 1967–74 did not have sufficient time to institutionalize itself (Diamandouros 1986). Thus, after the fall of the colonels’ regime, there was a need for changes in administrative personnel, rather than in the structures of the state bureaucracy. New Democracy, the conservative party founded by Constantine Karamanlis which controlled absolute majorities in the Greek Parliament between 1974 and 1981, took steps to eradicate projunta officials from the upper levels of public administration. It was rather successful in this task. In some quarters of the state, such as the judiciary and the universities, many, but not all, projunta elements were led to early retirement. Apart from that, no other major reform was attempted in the 1970s. Karamanlis, who served as prime minister from the very beginning of the democratic transition in 1974 until his elevation to the Presidency of the Republic in 1980, was preoccupied less with administrative reform itself than with the strengthening of the peak of the executive branch of government. Toward that objective, the Greek Parliament passed a constitution in 1975 which accorded important powers to the President. The liberal constitution insisted upon the political neutrality of civil servants, and subjected the public administration to the will of the elected government. The constitution stressed the complete separation of politics from administration. The Karamanlis government expanded the public sector. It nationalized ailing enterprises (e.g. the Commercial Bank) and even bought Olympic Airways from Aristotle Onassis. In the late 1970s, a new ministry, responsible for the environment and town planning, was founded. In the second half of the 1970s, Karamanlis passed a law reorganizing the government and another law which codified all legis-

Southern European Bureaucracies

205

lation concerning the status of civil servants. Most relevant regulations had been included in the original Civil Service Statute (or Code) voted in the immediate postwar era (in 1951). Under ND rule, patronage practices were continued as recruitment to the public sector was on the basis of formal competitive entrance examinations conducted separately by each ministry or public agency. In practice, this allowed considerable leeway to conservative ministers, parliamentary deputies, and managers of public agencies to influence the selection process. On the other hand, the conservative government passed a law which severely limited the right to strike in the public sector. Other widely disputed practices of ND involved political patronage in transfers and promotions in the public sector as well as sporadic police surveillance of voters of the Left. According to the government’s critics, the regime change of 1974 did not evolve into an adequate reform of the postwar Greek state. In contrast to Spain, where civil servants were represented in the political elite that effected the transition to democracy, the political role of civil servants in Greece was minimal in the immediate aftermath of the authoritarian period. This was reflected in the representation of civil servants among parliamentary deputies: in the 1974 parliament, just 3 percent of deputies were civil servants, rising slightly to 4.6 percent in the 1977 legislature (Metaxas 1981: 46; cf. Drettakis 1991: 62–5). The recruitment of ministers from among former civil servants was also negligible. As in the case of Italy, the political participation of civil servants was commensurate to their relatively weak position in the political system. In the postauthoritarian period, the confederation of Greek civil servants (ADEDY) was more submissive to the governing elites, as its leadership was controlled first by the ND party and later by PASOK, after it took power in 1981, through the intervention of the progovernment judiciary in the civil service unions. The composition of the top board of ADEDY was closely monitored by the government itself, sometimes through appeals to the courts which appointed a union leadership composed of progovernment trade unionists. Such practices were introduced by ND, and were imitated by PASOK governments between 1981 and 1989. In short, the period 1974–81 in Greece was generally devoid of major administrative reforms, which reflects the different priorities of Karamanlis, the architect of the Greek transition to a majoritarian, unicameral, unitary, centralist, parliamentary regime (Bruneau et al. 2001: 56–8). Italy: While Italy’s transition to democracy unfolded between 1943 and 1948, democratic consolidation was not achieved until about thirty years later. Italian politics during the intervening decades was

206

Southern European Bureaucracies

characterized by ideological polarization between the Right and the Left, low political trust among the main political parties, violent mobilizations of the working class (such as during the ‘Hot Autumn’ of 1969), and bombings and political assassinations by terrorist groups of both Left and Right. Democratic consolidation was largely achieved with the de facto ‘Historic Compromise’ between the DC establishment and the PCI in the late 1970s (Gunther, Puhle, and Diamandouros 1995: 24–5). Under Mussolini, most of the top civil servants had been conservative and elitist, if not profascist. Between the fall of the authoritarian regime and the decisive victory of the DC party in the general elections of 1948, many civil servants shifted their political allegiance to this newly hegemonic political party, while others remained quietly profascist. Generally, promotions in the civil service hierarchy were facilitated for those who chose to join DC and its union, CISL. The lack of reform was probably due to a contradiction in the relationships linking politics to administration: the constitution of 1948 maintained the prewar organization of the state apparatus, reflecting a liberal political philosophy according to which public administration was (ideally) immune to politics. The constitution did not rejuvenate the public administration, as it did with other political institutions (Cassese 1993b: 337; Franchini 1993: 428; Della Cananea 1997: 194), while it insisted that administrators maintain strict impartiality. Simultaneously, it accorded ministers full political responsibility for the activities of their ministries. The purpose of this arrangement was to prevent the bureaucracy from becoming an organ of the parliamentary majority and the government (Battini 1998: 205). This contradiction was not unique to Italy, since it can be found in the constitutions of other modern democracies. What was particular about the Italian case was the complete separation of the political class from the higher echelons of the civil service and the simultaneous reluctance of the civil service to exercise its responsibilities. We could argue that the state in postwar Italy evolved in four directions, most of which were a continuation of pre-World War II trends. First, the Italian state continued its prewar pattern of increasing tax revenues, and government expenditure, as well as the number of ministries and parastate corporations (Di Palma 1979: 158; Cassese 1993a: 326; Lewansky 2000: 216). Second, the trend of growing public employment, and particularly of employing people coming from the South of Italy in disproportionate numbers, also continued. Third, political parties assumed a greater role in state management and subjected the administration to ‘clientelistic party interference’ (Della Cananea 1997: 195). The particular forms which party influence took have been referred to as partitocrazia—the penetration of every political

Southern European Bureaucracies

207

and administrative sphere by parties—and lottizzazione, i.e. the division of functions and spoils among parties (Spence 2000: 128). Finally, since 1970 a slow transfer of power toward the regions has taken place (Cassese 1993b: 337), although until recently regional governments still had limited jurisdictions. The Italian higher civil service remained a politically secure body, consisting mostly of law school graduates who were linked to the DC and were able to thwart several attempts at administrative reform. In the postwar period, several commissions suggested plans for reform, but all relevant initiatives were absorbed and neutralized by the bureaucracy. Inertia was the result of power struggles among factions of the dominant party (the DC) and the lack of political will on the part of that party to reorganize the state apparatus. It was also a product of unionization of the civil servants who safeguarded their limited occupational privileges (e.g. tenure, salary increases on the basis of seniority). Such privileges may have been relatively minimal, but they were still important given the general poverty and unemployment which characterized the south, from which most civil servants were recruited. In short, with regard to the structure and personnel of the Italian central public administration, continuity rather than change characterized the period before and after the transition to democracy in Italy. Change was more marked in the wider public sector, however, which experienced a remarkable enlargement in the 1950s. New mechanisms of state intervention were established, such as the ENI and the Ministry for State Investments (1956). Their emergence gave impetus to the patron–client ties between the conservative and the centrist parties, governing in coalition under the aegis of DC, and large sections of Italian society (Morlino 1995: 354). However, the mentality and skills of the employees of the wider public sector were different from those of the employees of the state administration per se. Generally civil service personnel were more ‘generalist’ and more traditional in their outlook than equivalent personnel of the wider public sector. Like their Greek counterparts, few upper-level Italian civil servants acquired governmental offices. Party elites controlled all avenues of access to a career in politics. However, when it came to personnel policy, the Italian civil servants used to exercise considerable influence on the decision-making process. Similarly, in postauthoritarian Greece, particularly after 1981, it was not higher civil servants but party-led trade unions of civil servants which exercised influence in matters of personnel policy. In Italy, beginning in the 1960s the politicized trade unions—i.e. the leftist CGIL, the centrist UIL, and the Catholic CISL—were involved in personnel decisions to such an extent that one may speak of ‘comanagement’ (cogestione). Representatives of

208

Southern European Bureaucracies

unions were systematically included in the relevant committees (Lewansky 2000). The trade union of top civil servants (DIRSTAT) became a bargaining body with close ties to DC and substantial veto powers. This involved the domination of the procedures of recruitment by networks of patronage and the management of other personnel matters (e.g. promotions, transfers). The end result was usually a combination of political favoritism (serving the clientele of DC) and trade union interests. Throughout the period between the fall of fascism and the consolidation of Italian democracy in the late 1970s, there was little change in the civil service which seemed to have acquired certain quasi-permanent characteristics. These included inefficiency, the delivery of low-quality services, and an inability, if not outright reluctance, to implement new policies. Such symptoms were the result of deeper problems with the organization of the Italian public administration rooted in its legal framework and procedures (Della Cananea 1995: 9). The statutory framework of the civil service, first passed in 1908, changed under fascism in 1923 and modified in 1957, was very rigid. This rigidity strengthened the role of administrative courts, rather than the government or the parliament, in determining the organization and functioning of public administration. Resorting to the courts, which was reminiscent of a similar trend in the case of Greece, did not result in a better administration. The functions and organization of administrative units were not clearly differentiated, the procedures were long and complex, and the financial rules and controls were inadequate. The overall result was inefficiency, continuous increases in public spending, and the growth of public debt (Ginsborg 1990: 422–3; Cassese 1993a: 317). Other important problems were related to the recruitment, career mobility, and organizational culture prevailing among civil servants. To a much greater extent than in other West European bureaucracies, recruitment to the Italian civil service was based largely on family ties to politicians and top civil servants (Aberbach, Putnam, and Rockman 1981: 74, Table 3–8). Recruitment was also characterized by a pronounced ‘southernization’: in the postwar period, civil servants came disproportionately from the south (Cassese 1984: 40–1; Cassese 1993a: 319; Partridge 1995: 713; Lewanski 1999: 105, Lewanski 2000: 218). Civil servants were hired without competitive entrance examinations into ‘temporary’ positions (ostensibly on the grounds of urgent need to perform irregular tasks) which later became permanent through a process called ‘titularization’. In 1966 temporary employees made up approximately 75 percent of all civil servants (Cassese 1993a: 325). Once recruited in the civil service, horizontal occupational mobility was very low, while vertical mobility—promotion based on seniority—was

Southern European Bureaucracies

209

semiautomatic. Similarly, beginning in 1956 salary increases have also been tied to seniority, and were not based on merit (Lewanski 1999: 104). The organizational culture prevailing in the civil service was parochial and formalistic, and there was little adaptation to changes in response to evolving social circumstances. The Italian higher civil service in particular has been described by Sabino Cassese as ‘an ossified world’ (Cassese 1999: 55, 59) made stagnant by ‘a high degree of legalism and low level of e´tatisme’, manifested in the absence of the kind of administrative elite found in France (Cassese 1993a: 318). The state was paternalistic toward society, but it did not guide social and economic development in the way, for instance, the French state supervised the postwar development of France. Indeed, the mentality and style of work of Italian civil servants remained preindustrial, in the sense that they reflected a concern for job security rather than responsibility, higher earnings, and efficiency (Lewansky 2000: 218–19). The end result was described by some observers as ‘sclerosis’ (Lewanski 1999: 101) or ‘rigidity’ of the postwar Italian public administration (Cassese 1984: 58–9). The general weakness of the postwar Italian administration was in stark contrast to the periodic robustness of the Italian industry and service economy (Lewanski 1999: 98, 108). There were efforts to battle such problematic aspects in the 1950s, the 1960s, and the 1970s. In 1972, for example, a presidential decree provided for the creation of a superior and better-paid administrative category of directors within the civil service (la dirigenza) selected on a competitive basis (Cassese 1984: 49–52; D’Orta 1990: 443–4; Battini 1998: 208–9). Appointment to these positions would be assigned to a special body, independent of the ministries. Unfortunately, none of these provisions were implemented, except insofar as they provided for an increase in the salaries of directors (Battini 1998: 209), and there were no concerted efforts to reform the state until the mid-1990s. Civil servants practically refused to assume more responsibilities, while the political class did not diminish its involvement in major decisionmaking. As an observer of the Italian administration put it, ‘managers took the money, but preferred their safe careers and willingly left their responsibilities to the ministers’ (Lewanski 1999: 120). Why was there no change? One interpretation of this inertia contends that it was the result of a piecemeal series of ad hoc reforms that did not constitute a coherent whole. It was yet another manifestation of the Italian tendency to succumb to sectoral and particularistic interests (Franchini 1993: 429, 432; Della Cananea 1997: 199). A second view is a more political explanation of the nonimplementation of these reforms: Cassese speaks of ‘a pact founded on mutual renunciation’, a ‘tacit agreement’, on the basis of which the political class kept its power,

210

Southern European Bureaucracies

while the higher civil service retained its security (Cassese 1984: 54–5; Lewansky 2000: 213). Accordingly, bureaucrats would not interfere with the formulation of public policy, unless it touched upon issues of administrative personnel and remuneration. In effect, Italian bureaucrats exchanged power for security and freedom to manage their own conditions of employment (Cassese 1984: 46, 57). In addition, bureaucrats actively participated in the formation of patronage ties between the state and particular interest groups (Lewanski 1999: 98, 100). A third view claims that, although society bore the costs of the bureaucracy’s inefficiency, there was a wide consensus that contributed to the impasse regarding administrative reform. The wider society gained advantages through participating in clientelistic relationships, which were translated into patronage appointments in the public sector and into tax evasion (Lewansky 2000: 212–13, 237). Although this sounds like a functionalist and probably teleological argument, it nevertheless addresses realistically the question how the Italian bureaucracy externalized the costs of its inefficiency. To sum up, in postwar Italy, no major purge of the Italian bureaucracy was undertaken. The majority of higher civil servants survived the transition from fascism to democracy, and the organizational structure of the bureaucracy was not altered. Administrative reform was a low priority in the Italian postwar political scene. Until the completion of democratic consolidation, sometime in the 1970s, the political struggle between the Right and its allies (DC and its coalition partners), on the one hand, and the Left, on the other, monopolized the interest of political elites. Reforms were resumed much later, in the 1990s, in the wake of the crisis of 1992–4, which in certain respects resembled a regime change. Portugal: The collapse of the ‘Estado Novo’ in April 1974 triggered a far-reaching radical change in the institutional structure, functions, and personnel of the state administration in the first postauthoritarian period. In part, this was a reaction against the extent to which Antoˆnio Salazar relied upon the state administration as the core of his authoritarian regime. Salazar, following a career as a professor of finance and as finance minister, had undertaken his saneamento (cleansing) of the bureaucracy as part of his efforts to establish his dictatorial regime. The basic nature of his regime, coupled with a rather high turnover of ministers, meant that upper-level civil servants would function as the principal policymakers in this unusual political system (Opello 1985: 162), and the state apparatus, without ever challenging Salazar’s supremacy, played a pivotal role throughout the long life of the Estado Novo. For almost half a century, public bureaucrats asserted the authority of the

Southern European Bureaucracies

211

Portuguese state over competing societal interests and kept any possible challenges to the regime at bay. As Graham has noted, ‘As the regime became institutionalized, the making and implementation of public policy in the New State shifted to the bureaucratic arena, where it could be regulated under strictly hierarchical arrangements’ (Graham 1986: 2–3). These administrative structures therefore became prime targets of the revolution that followed the collapse of the Estado Novo in 1974. Under pressure from the gonc¸alvistas—the followers of the procommunist Captain Vasco Goncalves who became prime minister in several postdictatorial provisional governments—most directors general and managers of public enterprises abandoned their posts in 1974–5. The dismantling of institutions, such as the Ministry of the Corporations (which controlled the Portuguese labor), the internal security service, and the Overseas Ministry (which oversaw colonial territories), was also deemed necessary (Graham 2001: 215). In 1974–5 the communists (the PCP) controlled the mass media (Gallagher 1983: 210, 217). During the same period the gonc¸alvistas had taken control of a few ministries (Graham 1993: 97), and dismantled the Presidency ‘as a supra-ministry overseeing and regulating the administrative system as a whole’ (Graham 1983: 227; Opello 1985: 161–2). For a relatively brief period of time, left-wing forces managed to control the former Ministry of Corporations (before its abolition), the Ministry of Education, and parts of the military. Purges were not as dramatic in the Ministry of Foreign Affairs or in the Ministry of Justice. On the whole, the Left did not totally dominate either the public administration or the military, but was influential. One obvious consequence of this intervention in the bureaucracy was a fragmentation of authority and jurisdictions. This was accompanied by ‘the politicization of public organizations, the difficulty of coordinating government policy at the top, and the problems created by the lack of organizational coherence within the mainline ministries’ (Graham 1983: 240). The overall level of public employment also rose substantially. In the immediate postauthoritarian period, over 140,000 new employees were hired by the central government and about 20,000 by local governments. However, these figures included 35,000 to 45,000 civil servants who had served in colonies of Portugal before 1974 (Graham 1986: 6; Corte-Real 2000: 4). A parallel trend occurred in the wider public sector, which grew very rapidly during the transition to democracy. This was facilitated by provisions of the constitution of 1976 and legislation passed in 1977 which were conducive to state expansion. Under pressure from the MFA, the parties of the Left, trade unions, and

212

Southern European Bureaucracies

grassroots movements, most banking and insurance companies, and major industries fell to the hands of the state (Morlino 1995: 357). The mass media were also politically controlled from above. While these particular characteristics of the Portuguese transition contributed to the growth of public employment, such an expansion is also a predictable manifestation of our model of Southern European state bureaucracy. As noted earlier, its social shock absorber function is conducive to extensive patronage-based appointments in an effort to give employment to surplus labor. Finally, the transformation of the Portuguese political environment in the mid-1970s brought about changes in recruitment and promotion patterns as well as in the attitudes of Portuguese civil servants. Walter Opello claims that Portuguese directors general in the postauthoritarian period were less elitist than most of their West European counterparts, but they had also become rather tolerant of conflict. This was probably due to changes in the composition of the civil service because of the post-1974 purge (Opello 1985: 172). According to Graham, ‘New personnel have entered who are identified with political perspectives and outlooks impossible under the old regime; lower-level bureaucrats have found new opportunities to move ahead and to voice their opinions; competent professionals have survived, and probably improved their status and income; and old-regime bureaucrats find themselves ill at ease and constantly threatened by the new environment’ (Graham 1983: 243). These altered attitudes and behavioral norms were also the product of legislation enacted from the mid-1970s through the mid-1980s (Corte-Real 2000: 4). In 1974, a new directorate general in charge of civil service reform was created; an office of Ombudsman was founded in 1975; and in 1979, a School of Public Administration (Instituto Nacional da Administraca˜o—INA) was created, after the model of the French ENA, to promote professional training. In that same year there was a first attempt to reform the status of higher civil servants and the bureaucratic career ladder (Orvalho and Nabais 1990: 478; Corte-Real 2000: 4). To sum up, in Portugal, in contrast with the Greek, Italian, and Spanish cases, there was a far-reaching purge of supporters of the fallen authoritarian regime. This was a consequence of the dictatorship’s deep roots in the state administration, as well as the more abrupt and less consensual nature of its transition. Accordingly, the removal of civil servants who had supported the dictator was deemed necessary, and democratization of the state mechanism was deemed to be a high priority. Struggles over political control of the state administration were particularly intense during the turbulent revolutionary

Southern European Bureaucracies

213

phase of the transition. Although some civil service laws (e.g. on recruitment and careers) were revised in the late 1970s, most of the legislation implementing administrative reform was deferred until the 1980s. Spain: The Spanish transition to democracy is considered to be a model of elite settlement. The consolidation of democracy took only a few years and was completed by 1982 (Gunther, Puhle, and Diamandouros 1995: 21). In contrast with the Portuguese case, during the Spanish transition to democracy, no major purge of franquist civil servants was effected, and no drastic organizational restructuring took place (except, obviously, for the dismantling of the Movimiento Nacional). The civil service was neither a protagonist of regime transition nor an obstacle to fundamental change in the political system (Crespo 2001: 80, 85). In the civil service, a culture of modesty and consensus, which characterized the Spanish transition as a whole, allowed the hardliners among the top state officials to gradually ‘fade away’ to the backstage of the Spanish political scene (Beltra´n 1988: 260). A total of 30,000 franquist officials were integrated into the postauthoritarian civil service (Alba 1995: 391–2). While the discreteness of these changes is similar to the Italian and the Greek cases, Spain was unique insofar as it has implemented a major decentralization of the public administration beginning in 1980, eradicating one of the cornerstones of the fallen authoritarian regime—the extreme centralization of state authority. The state administration also became more differentiated in terms of organizational complexity, comprised of bureaucratic structures at the national, regional, and local government levels, as well as numerous public agencies and state-owned enterprises (Beltran 1988: 261). In the meantime, strong civil service trade unions emerged to give voice to widespread demoralization among civil servants which resulted from inadequate incomes and the incoherent manner in which employees had been integrated into the civil service under the Franco regime (Alba 1995: 392; Heywood 1995b: 130). The democratic constitution that was ratified in the December 1978 referendum provided for the political neutrality of the civil servants, the submission of the public administration to the elected government, the principle of merit in the recruitment of administrative personnel and, most importantly, the principle of administrative decentralization. In the first few years after Franco’s death, there were few changes in the organization of the central administration (Gunther 1980: 288–9). One innovation introduced in 1977 was the creation of a new upperlevel administrative rank—that of Secretary of State—which was roughly the equivalent of junior minister in the British administrative hierarchy, and was located just above the rank of Subsecretario

214

Southern European Bureaucracies

[Undersecretary]. Each Secretary of State supervised a number of ´ lvarez Rico and Gonza´lez-Haba Guisado 1992: directors general (A 86–7; Baena del Alca´zar 1993: 226–7). Secretaries of State were soon to play an important role in the formulation of policy. There were several of them in the powerful Ministry of Finance but far fewer in other ministries (Heywood 1999: 112). In 1979, the Minister of the Presidency was given the task of initiating more substantial administrative reforms, although the reform initiatives of the UCD governments were prevented from becoming law by that party’s sudden demise in 1982. For Alba (1995: 393), the lack of significant administrative reform during 1975–82 was the price to be paid for the political consensus which characterized the Spanish transition. Although several top Spanish civil servants participated in politics ‘from above’, that is, in parliament and the Cabinet, and ‘from below’, through the labor struggles of their unions, the Spanish civil service was not ravaged by partyled factionalism, as was the case of the Greek civil service. The internal divisions of the Spanish civil service were less of a political and more of a social and organizational nature. Divisions mirrored the lines among the cuerpos (corps) of civil servants (Baena del Alca´zar 1993: 463–4)— organized bodies of specialists, somewhat reminiscent of the French grand corps. Internally, however, Spanish cuerpos experienced much less of the political strife and factionalism that characterized other Southern European countries. Indeed, the existence of cuerpos allowed Spanish civil servants to preserve their autonomy from political parties to a greater extent than their Greek, Italian, and Portuguese counterparts. Insofar as they were politicized, the activities of cuerpos were oriented toward securing governmental posts as an extension of careers in the civil service. They also took the form of expert intervention in policymaking. To sum up, after the fall of the Franco regime, democratic governments encountered little, if any, bureaucratic obstruction upon coming to power (Beltra´n 1990: 334). The Spanish transition to democracy was probably the most consensual of all Southern European transitions, although Spanish civil servants did mobilize to obtain better salaries. Franquist bureaucrats were either gradually integrated into the postauthoritarian bureaucracy or were quietly marginalized. The first priority of the post-1975 governments was to rebuild democratic political institutions, leaving administrative reform for the postconsolidation period. However, the new socialist government encountered strong resistance from the elite corps when it attempted to divest them of their powers through a series of ambitious reform initiatives after the PSOE came to power in 1982.

Southern European Bureaucracies

215

Administrative Reform and Democratic Persistence Democratic persistence is ‘a phase in the democratization process subsequent to consolidation, in which political arrangements, crafted during the transitions and institutionalized during consolidation, acquire further roots and contribute to the deepening of democracy and to the improvement of its quality, or, conversely, to its erosion and possible deconsolidation or re-equilibration’ (Diamandouros and Gunther 2001: xii). This postconsolidation phase began in Italy in the late 1970s, in Greece and Spain roughly in the early 1980s, and in Portugal a little later, around the mid- or late-1980s, and coincided with pressures to adapt to European integration. How did bureaucracies evolve in this period, and which administrative reforms were introduced in these four countries? In all four countries, reform legislation was enacted in parliament, and some hesitant steps were taken to implement it. Among other things, this legislation affected the decentralization of authority of the central government, the organizational structure of the Cabinet, and the career patterns of civil servants. In parallel with the official effort to improve on the performance of public administration, most Southern European governments also employed informal, clientelistic practices to intervene in the civil service and influence its size and composition. Later, they switched their attention to better management methods and improved relations between citizens and the public administration. Greece: Between 1981 and 1989, PASOK, the socialist/populist party of Andreas Papandreou, ruled on its own with a comfortable majority of parliamentary seats. In contrast to the previous period of conservative government (1974–81), many administrative laws were passed. PASOK had an ambitious administrative reform program which aimed both at the further democratization of the public sector, the government structure and the public administration, as well as their use for electoral purposes. To an extent, some transformation was necessary in a state apparatus that traditionally had served authoritarian elites (1967–74) or conservative charismatic leaders (Karamanlis for a time before the colonels’ regime and again in 1974–81). In 1982, for instance, PASOK demilitarized one of the then only two television channels (YENED). That same year, PASOK abolished the highest civil service posts—the posts of director general and deputy director general—pensioning off the incumbents and creating new staff positions. These new, political appointee positions were inserted between the ranks of minister and director, and were filled en masse by PASOK party cadres and sympathizers. At the same time, PASOK’s own trade

216

Southern European Bureaucracies

union in the public sector (PASKDY) acquired extended influence over personnel matters. Further changes in 1985 and 1986 modified the structure and pay scales of the various ranks in the administration, making the structure even more top-heavy than it had been previously, with the highest ranks (grade ‘A’) becoming overstaffed. In practice, few civil servants had tasks corresponding to their high level, as the management of the state administration was assumed by the fifty-plus PASOK ministers and their entourage of appointees. Creating top-heavy administrative structures was not unique to PASOK: in 1975 the ND government had also created a new high-ranking post, inserting the new rank of ‘deputy director general’ between those of director general and section chief (as noted above, this inserted post was abolished in 1982). In the mid-1980s, PASOK tried to create a professional school—a semiautonomous ‘center’ (EKDD)—for the preservice and in-service training of civil servants. The school, which was conceived after the French ENA, was founded in 1983 and started functioning two years later. Neither civil service unions nor socialist ministers themselves supported this endeavor to renew and train the civil service personnel. The school soon met with the resistance of the judiciary and the civil servants of major ministries, such as the Ministry of Foreign Affairs, which always operated their own recruitment systems. Neither the new school nor the introduction of a ‘points system’ for the recruitment of new lower- and middle-rank civil servants (based on standardized evaluation of one’s social and academic qualifications) culminated in a satisfactory level of professionalization. The primary effect of the points system (which was applied only once, in 1984) was to favor the recruitment of socially disadvantaged candidates into the civil service. The system functioned adversely for young, highly educated candidates. The latter systematically obtained fewer points than candidates with fewer educational credentials but with more visible income needs. In 1984 and 1986, PASOK also passed legislation to decentralize the government. Largely in response to pressure from the European Community relating to the distribution of regional investment funds (such as through the Integrated Mediterranean Programs), thirteen regional authorities were created. These authorities were understaffed, controlled by the Ministry of National Economy, and did not possess enough human or other resources of their own. Their managers were appointed by the government. Prefectures (which in Greece constitute a tier between that of the region and the municipality) did not obtain any substantive autonomy or resources either. PASOK’s decentralization legislation was not fully implemented until the mid-1990s. Many other pieces of administrative legislation suffered the same fate in the 1980s.

Southern European Bureaucracies

217

A great organizational fluidity was characteristic of state structures after 1981, particularly with regard to the creation of new governing structures (committees, secretariats, and agencies), which functioned in parallel with the traditional ministries. In part, this reflected PASOK’s oscillation among three contradictory goals: professionalization of the civil service, the use of the state mechanism for electoral purposes, and the exercise of social welfare and labor policies through the exploitation of state resources. The latter was accomplished through the hiring of new, often unnecessary administrative personnel on the basis of financial need or political favors, and through the periodic transformation of fixed-term labor contracts into tenured posts in the civil service (the equivalent of titularization in the Italian public administration). However, the adoption of populist measures, such as the flattening of the administrative career pyramid or the recruitment of socially disadvantaged candidates to public jobs, dampened efficiency without increasing responsiveness. After the fall of PASOK from power in 1989, and following an interlude of short-lived coalition governments in 1989–90, ND won a decisive victory in the elections of 1990. As with PASOK, Nea Demokratia was beset with conflict between two groups of party cadres and government ministers: the modernizers, seeking to trim the public sector and introduce modern methods of management in the administration without completely foregoing the political benefits of patronage practices, and the traditionalists, who favored continuation of heavy monitoring of the public administration by the governing party in order to reap the electoral benefits of clientelistic appointments. At the same time, incumbency enabled ND’s civil service trade union (DAKDY) to acquire large influence. In 1990, the government of ND reinstituted the administrative rank of director general which had been abolished by PASOK in 1982. In 1991 it also reintroduced the process of competitive entrance examinations for the recruitment of new personnel to the civil service. This time, examinations were not conducted on a ministry by ministry basis (as under previous ND Cabinets of the mid-1970s) but were organized by the government in a centralized fashion. In addition, the conservative government formed new expert planning and policy units in each ministry. The ND ministers were supposed to consult with these units, but in practice they hired their own advisers and appointed them in the cabinets ministeriels, the staff units which had been formed by PASOK. Generally speaking, ND did not refrain from the familiar practices of patronage employment exercised by Greek governing parties (Sotiropoulos 1996: 134, 137). For instance, hundreds of new temporary employees were hired by the public Public Power Corporation (DEI).

218

Southern European Bureaucracies

The socialists returned to power in 1993, but this time they attempted to implement modernizing reforms that differed substantially from past ‘populist’ practices. PASOK passed a law in 1994 creating an independent authority (ASEP) in charge of administering competitive entrance examinations. For the rest of the decade of the 1990s, this law became a symbol of the standardization and transparency of recruitment to the public sector. Strict provisions of the law about hiring new permanent personnel did not, however, prevent ministers and managers of public corporations to continue the practice of hiring temporary employees who would later acquire permanent status. Second, the new modernizing ‘philosophy’ was reflected in the creation of a new grade scale which reversed the ‘flatter’ and ostensibly more egalitarian structure created by PASOK. A third innovation occurred in 1994, when there was a breakthrough in the process of decentralization: for the first time in Greek history, there were local elections for prefects (the equivalent of the French prefets) and for prefecture councils. An additional wave of modernizing reforms was launched in the late 1990s by PASOK Prime Minister Costas Simitis. Among the more significant innovations was the creation in 1997 of the new office of the Greek Ombudsman and the Data Protection Authority. By 1999 all the fragmented pieces of administrative legislation were collected and codified in a new Statute of the Civil Service. In the same year, a law providing for collective negotiations between civil service unions and the government was passed. Finally, measures to facilitate contacts between citizens and public service agencies were announced, including electronic tax declarations, applications for certain certificates by telephone, and ‘one-stop shops’ dealing with a variety of administrative services. Other legislative measures, such as those dealing with the improvement of skills of the civil servants or the introduction of ‘e-government’, however, were not substantially implemented. To sum up, the administrative policy of PASOK in the first period of democratic persistence had very mixed results. It involved repeated interventions in the administrative system, and oscillated among the contradictory goals noted above. ND stayed in power for a short period (1990–3) and did not have time to have a substantial impact on the public administration. Its policies were also a mixture of modernizing and traditional elements. The resumption of power by PASOK in 1993, coincided with fiscal pressures on the state and with a realization that Greece might be left out of the European integration process. The country had to adapt quickly to the requirements of this integration. Realization that the country had to adapt quickly to the preconditions of this integration reinforced the momentum toward modernizing reforms, at least at the level of policy formulation. A more transparent

Southern European Bureaucracies

219

and standardized mode of recruitment to the public sector and the foundation of independent administrative authorities were fruits of the administrative modernization in Greece in the 1990s. Italy: Several waves of thwarted reforms took place in Italy following the consolidation of its democracy in the late 1970s. The multiplication and diversification of administrative structures that occurred until the early 1990s reflected the lack of an overall design for reform. In 1984, a new Ministry of the Civil Service was created to supervise the civil service system and to oversee the functioning and expenses of the Italian public administration as a whole. In 1989, another ministry, responsible for universities and scientific and technological research, was created. Several public institutes were rearranged. Also in the 1980s, some reforms dealt with the organization of the Council of Ministers, regulatory policies, and the 1972 decree which had created the dirigenza (D’Orta 1990: 443–4; Franchini 1993: 431). In the 1980s, two other attempts at reform stood out. The first refers to the internal organization of the civil service, while the second to labor relations in the public sector. More concretely, in the early 1980s, the Giannini reforms organized the civil service in a novel way, somewhat reminiscent of the French corps or Spain’s cuerpos. Under the previous system, civil servants were placed in four different careers linked respectively to auxiliary, executive, conceptual, and directorial tasks. Through the reform, the first three careers were suppressed, while the dirigenza was kept intact. Except for the directors, all other civil servants were then classified in categories with a professional profile based on education and type of activity. Nine of these more general categories were formed, each characterized by a functional qualification (qualifiche funzionali).4 There was no hierarchy among qualification categories, but civil servants advanced within their own category. This reform was gradually modified by sectorial interests (Lewansky 2000: 223) through a torrent of new legislation—a tendency also familiar from the Greek case, which again reminds us of legalism and rigidity, a main aspect of our traditional model of South European bureaucracy. The modification and eventual failure of the reform was due to the passage of special laws which allowed certain categories of civil servants to slip into qualification categories different from those in which they had originally been placed; it was also partly attributable to the multiplication of qualification categories (D’Orta 1990: 446). This gradual evisceration of the law was facilitated by the second major reform which, in 1984, introduced negotiations between the government and the employees of the public sector concerning compensation and the organization of work. Matters pertaining to such areas

220

Southern European Bureaucracies

as the organization of administrative structures and recruitment to the public sector, in contrast, were to be regulated unilaterally by the government. This partition of legal competencies was not respected; special laws, for example, granted certain categories of employees additional compensation on top of the increases that had been agreed through negotiations (D’Orta 1990: 446). In the second half of the 1980s, part-time employment and temporary work on a project basis were introduced in the Italian public sector. This innovation was intended to increase flexibility, productivity, and efficiency, and entailed, among other things, a transformation of top bureaucrats into managers (D’Orta 1990: 448–9). These reforms took more definite shape in the 1990s, beginning in 1990 with a reform of local government, a law regulating administrative procedures, and legislation delimiting the right to strike in the public sector (D’Orta 1990: 449). This wave of reform continued in the 1992–4 ‘Cassese reforms’, and again in 1998 with the enactment of the ‘Bassanini laws’. The reforms of the 1990s differed from earlier ones in several ways. To start with, the reforms of the 1990s were put forward by technocratic governments which were relatively independent from the parties that had ruled Italy since the end of World War II. Second, they were launched in the wake of the crisis which shook the Italian political system in the early 1990s. The events that triggered this crisis, known collectively as Tangentopoli, had affected not only the political class but also the civil service. As a consequence, reformers could draw on a wider consensus on the need for changing the politico-administrative system. Third, the earlier ideological polarization between Right and Left (rooted in historic tensions between the two Italian subcultures) had abated following the collaborative relationship between the PCI and DC in the late 1970s, and, especially, the end of the Cold War. The partisan landscape was further transformed by the emergence a new political party, the Lega Nord, which gave voice to the dissatisfaction of Northern Italians with the public administration. Fourth, these political developments combined with the economic strains created by the fiscal condition of the Italian state and the macroeconomic requirements of European monetary integration, to produce an environment conducive to reform. The renewed concern about the condition of the Italian state was evident in the fact that seven referendum initiatives in April 1993 concerned the administration (Della Cananea 1997: 201). Finally, reforms were preceded by extensive studies of the condition of the Italian public administration (Franchini 1993: 433) and were more systematic and more comprehensive than the reforms of the 1970s and the 1980s (Della Cananea 1997: 195–6; Battini 1998: 209–13; Lewanski

Southern European Bureaucracies

221

1999: 120–1). In short, there was a move from ‘administrative reform’ to ‘reform of the administration’ as a whole (Franchini 1993: 430). The 1992–4 reform not only affected again the higher civil service, but also other aspects of the administration (Battini 1998: 211; Lewanski 1999: 127). The thrust of the reforms was the empowerment of the senior civil servants, who were granted wider management jurisdictions and corresponding responsibilities. Ministers and their staff would continue to establish policy guidelines, but would not interfere with the management of public administration (Battini 1998: 209–10). In 1994, the massive realignment of the party system was accompanied by a change of government in which the technocratic cabinets of Amato and Ciampi were succeeded by the government of Silvio Berlusconi. The leader of the new party, Forza Italia, did not have much interest in administrative reform. His successor as prime minister, Lamberto Dini, also had other priorities, many of which concerned pressing financial matters. As a result of the low priority given to administrative reform, as well as of the short time in office of reform governments, the reforms of 1992–4 were not implemented, with a few exceptions in the area of relations between citizens and administration. Also impeding the implementation of reforms were the fragmentation of administrative structures, which made wholesale change difficult, and the resistance of vested interests. These included categories of public employees who would be required to adapt to unwelcome changes in their employment status, as well as interest groups outside the administration which had forged links with the traditional administration. Administrative reform was given a new impetus by the government of Romano Prodi, who served as prime minister between the Spring of 1996 and the Autumn of 1998. The new reforms strengthened some of the aspects of the 1992–4 reforms but also moved in new directions (Batttini 1998: 213–16; Lewanski 1999: 109–10, 114, 117). Two laws were passed in 1997 and one in 1998 which aimed at decentralization, ‘privatization’ of labor relations in the public sector, a clearer distinction between politics and administration, and the elimination of red tape (Lewanski 1999: 110). Employment conditions in the civil service were to be determined by bilateral agreements between the government and civil service unions (this time including directors general). There was also greater emphasis on delivering better services to citizens, including by keeping offices open in the afternoon, devoting additional resources to the training of civil servants, providing for the evaluation of the management of public services, and establishing ‘citizens’ charters’. Some of these innovations paralleled similar trends

222

Southern European Bureaucracies

in Portugal and Greece, and to some extent reflected New Public Management ideas disseminated by international organizations. In terms of administrative structure, the reforms of 1998 led to the merger of two ministries and, more importantly, the ‘federalization’ of Italy, which attempted to transfer human and financial resources, as well as policy jurisdictions to the regioni (Lewanski 1999: 114). This decentralization was given considerable political impetus by the substantial political power secured by the Lega Nord. Finally, several large public administration units were given the freedom to restructure their internal organization. The most important difference between the reform of 1992–4 and that of 1998 was that the latter reinforced the power of the ministers by making the highest-ranking civil servants of each ministry dependent on the minister. This reflected a return to the political class of powers that had been devolved to the administrative elite in 1993. After 1998, senior civil servants were to serve in their posts on the basis of a renewable contract and to be removed with every change of government. They could also be removed on the grounds of substandard performance. At the same time, the reform of 1998 gave ministers a freer hand in organizing their own staff (Battini 1998: 215). In short, the reform introduced a new kind of relations between bureaucrats and politicians. In short, the Italian reforms of the 1990s went beyond those of earlier periods by dealing not only with personnel matters, but also almost all other aspects of a modern administration—regarding structure, procedures, oversight, and resources. These more ambitious reforms coincided with the substantial transformation of the Italian political system as a whole in the 1990s, and with international trends altering the role of the state in society. In contrast to its interventionist role in the postwar period, the state was now to assume a regulatory role, enabling society to develop on its own, rather than to closely monitor and intervene in many sectors of social and economic activity. In keeping with fashionable currents of thought circulating through the international environment in the 1990s, this altered relationship was portrayed as encouraging initiatives by individuals and social organizations, and as transforming the state from an obstacle to a partner in bottom-up, private sector initiatives. Nonetheless, problems of implementation persisted. While, on the one hand, the privatization of public employment conditions has progressed considerably, implementation of citizens’ charters has not (Spence 2000: 146). Portugal: The first decade following the 1974 Revolution was characterized by fluidity of administrative structures in Portugal and an indecisiveness as to the correct course that administrative reform

Southern European Bureaucracies

223

should take. This meandering course of administrative change paralleled the uncertainty and sometimes turbulent developments in the postauthoritarian evolution of the Portuguese political system. Fluidity and indecisiveness was evident in the overall size of the public sector, in the changing course of major public policies (e.g. from initial tolerance of private enterprise to nationalizations and back to privatizations), and in the form of state agencies founded to bring about administrative reform. The rapid increase in public employment in the first two years following the collapse of the Salazar/Caetano regime was one manifestation of the wave of nationalizations that swept the Portuguese private sector in the course of the Revolution. In sharp contrast, beginning in the 1980s, and particularly after the rise of PSD to power in 1985 and the constitutional reform of 1989, fiscal reform and reprivatization of business enterprises shaped a new environment conducive to administrative modernization. These modernization efforts proceeded along two tracks: debureaucratization, and improvement of the quality of service delivery and relations with ordinary citizens. The Portuguese administration had been characterized by legalism, formalism, and rigidity, all of which fit within our ‘ideal type’ of traditional Southern European bureaucracy. A reform of the career system began in 1985, initially oriented toward reversing the previous decade’s deterioration of public employees’ salaries. This was followed by the establishment of a new salary structure in October 1989 whose aims were to standardize the fragmented and unequal remuneration of public employees, to introduce objective criteria for any remaining differences in compensation, and to make salaries competitive with those in the private sector (Orvalho and Nabais 1990: 480–1; OECD-PUMA 1992: 243). The range of salaries was substantially broadened, and increases in compensation were to be based on seniority, and to be decoupled from promotion to a higher rank, for which additional specific qualifications were required. This severing of the link between promotion and compensation was similar to reforms effected by PASOK in Greece and by the PSOE in Spain in the mid1980s. Between 1982 and 1986, the Ministry for Administrative Reform was replaced by the Secretariat of State for Public Administration, which, in turn, gave way to a smaller and looser structure, the Secretariat for Administrative Modernization. A director general was put in charge of the new Secretariat, and in 1989 the post of the Secretary of State for Administrative Modernization was established (see Corte-Real 2000: 9). Overall, a change of ‘administrative philosophy’ took place in the mid-1980s. Previously, administrative reform efforts took the form of

224

Southern European Bureaucracies

grandiose plans for wholesale reform formulated by the central government which were supposed to be dutifully implemented in a uniform fashion by all relevant administrative units. Following the rise to power of Cavaco Silva’s PSD, more pragmatic and flexible reform policies were enacted, and tailored to suit the particular circumstances of individual administrative units. Major emphasis was placed on ‘debureaucratization’—an attempt to decrease the number of printed forms to be submitted by citizens requesting government services, and the number of procedures to be followed in contacts between businessmen and the administration. In 1987, a special ‘Enterprise–Administration Committee’ was set up to facilitate business–bureaucracy relations. On balance, however, despite some symbolic gestures, such as proclaiming ‘debureaucratization day’ (which passed without any real effect), the simplification of procedures had not progressed far enough by the end of the 1980s (Fernandes 1990: 8). Many new measures were neglected by administrative units, and cumbersome bureaucratic procedures were retained even when there were explicit initiatives to streamline them. On the whole, the Portuguese public administration reacted very slowly to reform initiatives. The economic recession of the early 1990s further complicated these reform efforts, and culminated in a ‘need to improve services within a context of scarce resources and budgetary constraints’ (Corte-Real 2000: 6). This more complex environment led to contradictory trends in public administration. Approximately 30,000 new public sector employees were hired between 1996 and 2001 (Pu´blico, September 20, 2001), most of them in the education and health sectors. While this may reflect an expansion and enhancement of the welfare services offered to citizens, it also was a product of continuing clientelism. Between 1991 and 1994, modernizing efforts included the drafting of ministry-by-ministry plans for reform, a new Code of Administrative Procedures, the construction of databases and public exhibitions to familiarize the general public with the public administration, and a program of vocational training for civil servants. Following the coming to power of the Socialist Party (PS) under Antonio Guterres, emphasis was again given to relations with citizens and businesses. Between 1996 and 1999, new initiatives included a ‘Citizens-Administration Forum’, a one-stop shop for businesses and a ‘citizens’ shop’, the aim of the latter two being to consolidate government service agencies in one location (Corte-Real 2000: 16). In order to check the growth of public employment, a policy of hiring only one civil servant for every four who were pensioned off was announced in 2001. Such a measure had been attempted also in Greece in the early 1990s, but to no avail since the clientelist hiring of temporary employees did not abate.

Southern European Bureaucracies

225

In the end, administrative reform efforts in Portugal have fallen short of achieving intended objectives, some notable improvements notwithstanding. On the one hand, there have been signs of progress in some areas since the mid-1980s: new channels of citizen participation and communication were opened, numerous reform initiatives were introduced, and some change in the administrative culture was apparent. On the other hand, procedures have remained slow, cumbersome, and complex (Corte-Real 2000: 25), a proposal to decentralize the administration was rejected by a nationwide referendum in November 1998, and a multitude of new public organizations and agencies were established on the basis of EU funds, often parallel to previously established central administration bodies. This same strategy of creating parallel administrative mechanisms in order to cope with the requirements of European integration was also followed in Greece in the 1990s, perhaps as a product of concerns over the low level of skills and inefficiency of the existing state administration. An unanticipated consequence of establishing these parallel administrative bodies, however, may have been the lessening of pressures for reform of the central state administration (Graham 2001: 220). In general, it appears that at some point Southern European governments gave up on wholesale reform of the state bureaucracy, shifting, instead to an ‘agencification’ of the state (which could be found in many Western countries in the last two decades of the twentieth century). The Portuguese government, for example, has often preferred to create new administrative units (such as smaller and more flexible public agencies) rather than to reform existing ones (such as traditional, large ministries). Spain: The coming to power of a PSOE government coincided, by and large, with the consolidation of Spain’s democratic regime, except for the partial regional exception of the Basque Country. Given this increased confidence in the stability of the regime, administrative (and economic) reform and political decentralization could be vigorously pursued. The most dramatic consequence of these processes of change was the transformation of one of the most highly centralized states in Europe into a ‘quasi-federal’ state, with substantial legislative and executive powers transferred to the seventeen autonomous regions [Comunidades Auto´nomas], each with its own government, parliament, and administrative mechanism with wide policy jurisdiction. By the late 1990s, the seventeen ‘Autonomous Communities’ employed approximately one-third of all public employees in Spain. The socialist government’s commitment to administrative reform was manifested in several laws, two of which implemented the transfer of authority to the autonomous regions—one intended to simplify administrative processes, and the other regulating conditions of work and

226

Southern European Bureaucracies

labor union elections in the public sector. One symbolic gesture signaled this change of the climate: the traditional small windows [ventanillas] through which citizens used to interact with civil servants in public buildings were brought down. From a structural perspective, two other legislative initiatives were of the greatest significance. The Ley de Incompatibilidades enacted in 1983 prohibited public employees from holding more than one post in the public sector and restricted their access to private sector jobs. The second and most important was the 1984 Ley de Medidas [reform measures act], which, among other things, reduced the number of cuerpos [corps of civil servants], introduced a new 30-level grade scale, modified pay scales, standardized and modernized recruiting and in-service training practices, made the transfer of civil servants from one sector of the public administration to another more flexible, lowered the retirement age from 70 to 65 years, and introduced new technology and modern management practices (see Baena del Alca´zar 1993: 456). Prior to these reforms, the particular cuerpo to which a civil servant belonged determined the salary level and the posts for which he/she was qualified throughout his/her career. Under the Ley de Medidas, however, it was the administrative level one occupied, rather than the cuerpo to which one belonged, that determined salary levels and job qualifications. This new system required the careful construction of detailed organizational charts, as well as specific and accurate descriptions of civil service posts. In addition to determining salaries on the basis of the posts held, ‘step’ increases were given for every three years of service (the trienio), and ‘job performance’ (productividad) increases were also introduced (Sficas 1989: 50; Molina 1999: 45; Parrado 2000: 255–6, 261). The PSOE government’s sweeping reforms were motivated in part by suspicions concerning a bureaucracy inherited from previous authoritarian and center-right democratic governments (much as PASOK was moved to reform a Greek bureaucracy that had served ‘the colonels’ regime and previous right-wing governments). But these reforms were also stimulated by a desire to reduce the power of the grandes cuerpos (such as the Abogados del Estado, the Inspectores de Hacienda, and the Letrados del Consejo del Estado), which were often more interested in advancing their own particularistic interests (especially regarding the hiring and promotion of their own members and their de facto ability to veto certain policy initiatives) than in administrative efficiency (Beltra´n 1990: 346). Indeed, under the Franco regime, certain ministries had fallen under the control of specific cuerpos, and ‘turf ’ battles between rival cuerpos had undermined performance and contributed to the internal fragmentation of the state administration.

Southern European Bureaucracies

227

The 1984 reform attacked this de facto colonization of specific bureaucratic posts by cuerpos, making it possible to freely transfer civil servants from one position to another, and to fill open positions through competitions [concursos] regardless of the individual’s cuerpo affiliation. However, while these reforms may have greatly reduced the control over recruitment and promotion by cuerpos, some skeptical observers contend that it replaced one particularistic appointment system with another, establishing a new ‘spoils system’ that systematically favored the appointment of career civil servants who were members or supporters of the governing party to the upper-level ‘political positions’ in the bureaucracy (Beltra´n 1990: 347). By 1996, 80 percent of such posts were held by civil servants who were considered to be supporters of PSOE (Quermonne and Jime´nez Asensio, cited in Molina 1999: 51). The prevalence of civil servants among political appointees was still evident in the 1990s even though, in comparison to the Franco era and to the 1980s, the percentage of outsiders (i.e. individuals who were not career civil servants) appointed to political posts had increased sharply (Parrado 1999: 17). On that count, the Spanish case was diametrically different from the case in Greece under PASOK, in which few civil servants occupied political posts. Other persisting problems following enactment of the 1984 reform involved the inequality of remuneration and the continuing influence of the elite corps, despite the initial attempts of PSOE governments to curb it. The strongest resistance to reform was mounted by powerful elite cuerpos in the Interior and Finance ministries. These traditionally elite corps had attracted well-educated members of the urban upper classes, whose interpersonal contacts served as the basis of a closed network capable of resisting external political pressures (Molina 1999: 43–4). Their positions were further strengthened by the heavy reliance of modern governments on experts who, in the case of Spain, can be found among its civil servants rather than solely among party cadres (as in Greece) or technocrats from universities or other prestigious state institutions (as in Italy). The PSOE’s reforms also included provisions for vocational training, particularly under the auspices of the Institute of Public Administration (INAP), which was reorganized in 1987. In-service training programs were established within each of the major ministries of the central government, as in the regional governments of Valencia, Catalonia, and the Basque Country. And to assist the government in the formulation and implementation of administrative policies, and to oversee relations among central, regional, and local levels of government, a Ministerio Para las Administraciones Pu´blicas [Ministry of Public Administrations] was founded in 1986 (parallel to a similar

228

Southern European Bureaucracies

organizational reform in Portugal). The new ministry was dominated by the cuerpo of general administrators (Molina 1999: 48). Other noteworthy developments in the late 1980s and early 1990s included the establishment of collective bargaining between the Spanish state and civil service labor unions, particularly concerning work´ lvarez and Gonza´lez-Haba 1992: 311–12). In 1992, the ing conditions (A first Minister of Public Administrations, Joaquı´n Almunia, announced a new modernization program intended to improve services to the general public, to transform tradition-bound civil servants into autonomous ‘public managers’, and to introduce modern methods of management and budgeting (Heywood 1995b: 132–3). These reform proposals were strikingly similar to the Portuguese and Italian reforms of 1992–4. They were accompanied by efforts to make public organizations less rigid by subjecting them to private sector laws regarding hiring and contracting procedures. Under the terms of a law passed in 1991, thirty agencies, including the National Airports and the Post Agency, fell under these provisions (Parrado 2000: 263). These efforts appear to have been at least partially successful. Throughout the first half of the 1990s, the hiring of new personnel was sharply constrained, at least in part as a result of the fiscal crisis of that period. Between 1990 and 1997, only 8,000 new employees joined the central administration (Parrado 2000: 256, 258, based on Ministry of Public Administrations data). This was in stark contrast to the massive hiring of the 1980s and to the great expansion of the bureaucracies of the seventeen comunidades. The coming to power of the PP government of Jose´ Marı´a Aznar in 1996 was not followed by any dramatic changes except, obviously, for a complete turnover of top administrators as a reflection of the PP’s drive to establish political control over the bureaucracy. Despite the PP’s election promise to decrease the number of top civil servants, by 1998 the top levels of the administrative grade scale had been greatly inflated. This was the net product of patronage appointments by the PP, coupled with the fact that tenured civil servants who had occupied senior posts remained in the service and legally kept their highranking status, even after they had been dismissed from their posts. Another minor change was a law passed in 1996 that eliminated restraints imposed by Socialist government legislation limiting horizontal occupational mobility of civil servants wishing to work in the private sector. Perhaps the most important change in the 1990s was the 1997 law that expanded the career possibilities of civil servants. Subsequently, all top posts in Spanish ministries, except for the staff of ministerial cabinets [gabinetes], were to be filled by civil servants. This continued the trend (called funcionarizacio´n by Molina [1999]) toward

Southern European Bureaucracies

229

consolidation of the powerful position held by senior bureaucrats in the core of the executive branch of the political system.

Conclusion While the public administrations of Greece, Italy, Portugal, and Spain used to differ from their Northern counterparts along the lines captured by our model of Southern European bureaucracy, and perhaps still do to some extent, our detailed examination of the structures, norms, and practices of these bodies has revealed substantial differences among the four Southern European cases. Just to cite one source of significant cross-national variation, high levels of government turnover (commonly regarded as helping to insulate the civil service from political control by elected officials) did characterize Portugal until 1987 and Italy throughout the postwar period, but Greece and Spain have experienced very high levels of cabinet durability since the founding of their respective democratic regimes in the mid-1970s (see Bruneau et al. 2001: 26–8). Another example of significant cross-national variation involves the level of professionalism and corporatist organization of the civil service. Spain has a tradition of skillful and cohesive bodies of civil servants (cuerpos), and the level of professionalism among Spanish civil servants can only rarely be found in Greece or Italy. Conversely, the Greek and Italian cases share a strong tradition of ties between civil servants and political parties, at least since the end of World War II. In this and several other ways, the Greek and Italian bureaucracies appear to share more common features than any other combination of state administrations within the region, and these similarities became even stronger when we turned our attention to their respective experiences with various attempts to implement administrative reforms. Nonetheless, there remain important differences between the Greek and Italian cases. While Greek party governments of the second half of the twentieth century used to actively intervene in the day-to-day operations of the Greek civil service, the Italian coalition governments of the postwar period reached an arrangement with the civil servants. Except for the recruitment of new personnel, which in Italy has habitually fallen victim to quotas and exchanges among the ruling political parties under the lotizzazione system, other aspects of the functioning of the state administration were left to civil servants themselves. This trend was the result of the ‘art of arrangement’, as Cassese has called it (1993b: 335, 339). In the case of Portugal, the intervention of political parties seems to have been more extensive in the immediate

230

Southern European Bureaucracies

postauthoritarian period and again after 1987. In Spain, this kind of partisan intervention became more notable after 1982. There were also differences of administrative development within each of the four cases under study. Between 1977 and 1983, for example, Spain became a quasifederal state through its division in seventeen Communidades Auto´nomas, each with its own structure of public administration. Italy has also been divided into twenty regions since 1970, and these regional governments have been governed by different political parties. Accordingly, the administrative system of some regions of Spain or Italy may be quite different from those to be found in the rest of the country. The Basque Country has a very high level of autonomy, complete with most of those self-government powers typical of a modern state, and the regional administrations of Catalonia and Galicia also have very substantial self-government rights. In Italy, the quality of public services obtained in some areas of the South (e.g. Naples, Calabria, Sicily) is much lower than that of the equivalent services in the North of Italy. In addition to these subnational geographical variations, there are important sectoral differences in administrative performance. In Italy, for example, the treasury, the diplomatic service, the Council of the State, and the Audit Office all function quite well (Meny 1993: 452–3). So, too, do the Council of the State, the Bank of Greece, and the Ombudsman in Greece. These cross-national and subnational variations notwithstanding, there are significant similarities among these four cases that have continued well into the stage of ‘democratic persistence.’ One of these commonalities is the existence of politicized informal networks within the state bureaucracies, which differ somewhat from the kinds of informal ties found in most other large modern organizations. To a much greater extent than is found elsewhere in Western Europe, the nexus of politics and administration is woven around political fractions, with national variations corresponding to the political party system of each country. Patronage or spoils systems have been unofficially constructed by political parties. This has been facilitated by the fact that Southern European democracies are party democracies par excellence, partly as a result of the dominant role played by political parties in the process of democratic consolidation. Patronage appointments have established factions within the state administration. While factionalism is not unique to Southern Europe (e.g. it is quite pronounced in the Belgian bureaucracy), the bases of Southern European factionalism within the state administration are less of a social, professional, ethnic, or linguistic nature, and much more overtly partisan in character. Another shared characteristic is that in Southern Europe recruitment and access to the state bureaucracy was and, to a certain extent,

Southern European Bureaucracies

231

still is permeated by political party considerations and priorities. This partisan particularism undermines the professionalism and quality of services, and has led citizens in these four countries to view their respective bureaucracies as omnipresent and distant. Negative attitudes toward these state administrations have been further undermined by their past service to nondemocratic regimes—a trait which they share with Eastern European bureaucracies following the fall of Communism. In Eastern Europe and, to a lesser extent, in Southern Europe the state seems to combine an awkward aloofness, often insensitive to wider economic, technological, and cultural changes, with a superimposing presence (Sotiropoulos 2002: 63–4). The latter was especially true at times of authoritarianism when deviant political behavior was closely monitored by state security agents. To reform such administrative practices was one of the immediate tasks of the transition to and consolidation of democracy. In Southern Europe, bureaucrats did not resist democratic transitions, for reasons that vary from one country to another. In Spain, bureaucrats were largely insulated from the control of Falangist cadres, and by the mid-1970s, representatives of the formerly Falangist faction had disappeared from the Council of Ministers. Reinforcing the bureaucracy’s willingness to acquiesce to a change of regime (and, in several prominent instances, to play leading roles in that process) were bitter memories of the Spanish Civil War which led to the exercise of caution during the delicate stages of the transition. The traditional professionalism of the Spanish civil service corps also helped, placing in key government positions competent administrators who were committed to the continued smooth functioning of the state. The case of Portugal was diametrically opposite to that of Spain. Bureaucrats had been closely identified with the regime of Salazar and Caetano, and the abrupt collapse of the authoritarian regime did not allow for the protracted evolution that had transpired in neighboring Spain. Rather than the remarkable continuity of personnel that characterized the Spanish bureaucracy following the death of Franco, purges of bureaucrats were initiated by the left-wing governments of the revolutionary period, supported by the mobilization of left-wing workers, farmers, and students. This extreme discontinuity contributed to the political instability that for a while made the Portuguese transition unpredictable. The Greek and the Italian cases were somewhat in between the above two extremes. In Greece, there were some limited purges since the military junta did not have the time to institutionalize itself in the state administration. In Italy, profascist officials were allowed to fade away. The cleansing of bureaucratic ranks was not a top priority, since the main preoccupation of governing elites was how to contain the Italian Left.

232

Southern European Bureaucracies

One way or another, Southern European bureaucracies quickly learned how to adapt to democratic rule. By the time democracy became consolidated in Greece, Italy, Portugal, and Spain, their bureaucracies were fully supportive of the new democratic regimes. Still, administrative reform was necessary in order to make bureaucracies more accountable and responsive, as well as sensitive to the goals of economic modernization and European integration. Despite the announcement of numerous reform initiatives, however, new policies were often aborted due to bureaucratic resistance or to abrupt changes in reform plans after a new government came to power. The basic precepts of ‘New Public Management,’ which were popular with other West European bureaucracies (Clark 2000: 25) have failed to take root. This is not to say that no significant change has taken place, or that the Southern European model described at the beginning of this chapter has remained intact. Changes away from this model can be observed along three notable evolutionary paths. The first of these was toward decentralization which, as noted above, has been most extensive in Spain since the beginning of the 1980s. In Italy, decentralization progressed slowly following its initiation in the 1970s, but developed much more rapidly by the end of the twentieth century. While the Portuguese mainland has been governed by a centralized bureaucracy, the archipelagos of the Azores and Madeira acquired substantial autonomy. The relatively modest decentralization of Greece beginning in the mid-1980s was a response to pressure from the EU, and was undertaken as a means of receiving and administering funds from the Integrated Mediterranean Funds. The country was divided into thirteen regions with very thin regional administrative structures, although in the 1990s additional resources and authority were transferred from the central administration to the regions and prefectures. The second direction of the evolution of the state in Southern Europe in the final years of the twentieth century followed the worldwide trend of state retrenchment, mostly through privatization of parts of the public sector (Wright and Pagoulatos 2001: 234–6). Privatization was wholeheartedly adopted in theory by successive Southern European governments, but in practice was only implemented to a substantial degree in the 1990s. In Greece, privatization began in earnest only after the 1996 rise to power of Costas Simitis, Papandreou’s successor. In Italy, there was a pressure to increase the efficiency of the public administration and to eliminate public monopolies, particularly in the second half of the 1990s (Lewanski 1999: 112). In Portugal, substantial waves of privatizations occurred in 1989–90 and 1995–6. In Spain, privatizations picked up after 1993. By the second half of the 1990s,

Southern European Bureaucracies

233

the size of Southern European public sector had been considerably decreased, and these trends have still not come to an end. The third direction of change was related to the effects of European integration on the organization of Southern European bureaucracies. These were not limited to the push for decentralization, but also involved increased administrative efficiency. The EU exerted its own pressures on Southern European bureaucracies, but also served as a source of inspiration for domestic pressures toward administrative reform: the modernizing political elites of Southern Europe used the prospect of integration into the EU as a political weapon to press for change in their states and societies. Paradoxically, this led to a multiplication of administrative structures of various types: first, in the form of new units (secretariats, divisions, or sections) responsible for relations with the European Community (and later the EU) inside ministries; second, in the form of task forces entrusted with responsibilities relevant to European integration; and third, in the form of new public agencies created to sidestep the rigid hierarchies and cumbersome procedures of the existing central public administration. Finally, these changes were accompanied by the emergence of new ideas about the state. The state was now seen as a catalyst of change, enabling individuals and collective actors to accomplish their own, self-assembled projects (economic, educational, or otherwise), rather than guiding them closely from above. These changes parallel trends that can be observed in other European states, suggesting that there may be some convergence in the basic characteristics of bureaucracies in both Northern and Southern Europe. Domestic developments—such as the replacement of the postwar party system in Italy in 1992–4—and, more importantly, international changes—such as the advance of European integration, the end of the Cold War, the intensification of global competition among national economies, and the spread of new technologies—have created a new political, economic, and technological environment for public administrations, including those of Southern Europe. This new environment is quite different from the historical context of regime transition and consolidation with which we began this chapter. How Southern European bureaucracies will adapt to the new environment remains an open question.

Notes 1. Some ideas contained in this chapter were inspired by the late Vincent Wright. The author would like to thank Gianfranco Pasquino for helpful comments on Italy; also Carlos Gaspar, B. Filipe Nunes, and Maria Asensio

234

Southern European Bureaucracies

for bibliographic advice and information on Portugal; Alejandro Quiroga for comments on Spain; the two Greek and the two Spanish ministers who granted him interviews; and the two editors of this volume for their encouragement and editorial work on various versions of this chapter. 2. See Alba 1995: 391; Heywood 1995b: 138–40; Molina 1999: 50, 52; CorteReal 2000: 24; Graham 2001: 218. 3. I wish to thank Gianfranco Pasquino for making this point. 4. D’Orta 1990: 443–4; Franchini 1993: 429–30; Furlong 1994: 87; Lewanski 1999: 117; Lewanski 2000: 223.

................ 6 ................

Multilevel Governance and the Transformation of Regional Mobilization and Identity in Southern Europe, with Particular Attention to Catalonia and the Basque Country Iva´n Llamazares and Gary Marks

The territorial organization of decision-making in Europe has been transformed in the last twenty years under the influence of European integration and subnational decentralization. With the creation, enlargement, and deepening of the EU, decision-making competencies in a range of policy areas, including regional policy, agricultural policy, the internal market, and monetary policy have been reallocated from the national to the European level. At the same time, central state executives in several European countries, including Spain and, to a lesser extent, Greece, Italy, and Portugal, have shifted a variety of competencies to regional, local, or urban governments. This chapter examines the consequences of these developments for regional governments and regional political parties. The first set of questions we ask has to do with how the changing territorial organization of decision-making has influenced regional political activity. How have European integration and regional decentralization shaped the structure of political opportunity for regional parties and regional governments? To what extent have they been drawn directly to the European level, and how can one explain the resulting pattern of representation? The second set of questions we ask has to do with the goals of regional actors. To what extent has their participation in an emerging European polity influenced the ambitions of regionalists for political autonomy and the conceptions of institutional alternatives available to them?

236

Multilevel Governance

The countries of Southern Europe—Greece, Italy, Portugal, and Spain—are an interesting focus for such questions because they share membership of the EU and many sociocultural features, yet have contrasting systems of territorial relations. Variation in the territorial relations of these four countries is just about as great as that within the EU as a whole. At one extreme, Spain is one of the most ethno-territorially diverse countries in Western Europe, while Portugal and Greece are among the most homogenous. Catalonia and the Basque Country have a combination of strongly entrenched regional governments and electorally successful nationalist parties; Northern Italy has a weaker regional party and much weaker subnational competencies; while Portugal and Greece are unitary states with largely national parties. An examination of territorial politics in Southern Europe is therefore a study in contrasts. This is the first theme of this chapter. Our answers to the questions posed above emphasize the very different starting points of subnational actors both across and within the countries of Southern Europe, and we find few signs of convergence in the territorial structuring of authority among these countries. Our second and more contentious theme is that the role of national states is being transformed in some subtle—and not so subtle—ways. In regions where the reallocation of competencies and the mobilization of regional parties has progressed furthest, particularly in Catalonia and the Basque Country, we find that subnational governments are intimately involved at the European level, that the national state has lost appeal as an institutional form for realizing political autonomy, and that there has been a significant shift toward multiple territorial identities. In short, the central state is in decline in Southern Europe, both as a normative model for regional autonomy and as an actor representing domestic territorial interests in the EU.

From National Decision-Making to Multilevel Governance The reallocation of decision-making competencies away from territorial states that has taken place over the past three decades in Southern Europe is an historically anomalous development that reverses a centuries-long process. The overall direction of power redistribution and the locus of power creation in the process of state-building from the thirteenth to the twentieth centuries was toward the center. While the absolute degree of centralization varied across Western European states, all became more centralized as monarchs and state builders struggled to monopolize legitimate authority, to create secular

Multilevel Governance

237

hierarchical systems of justice, and to deepen and widen taxation. This process was greatly intensified with the rise of nationalism across Europe and the identification of centralized extraction, provision of welfare, and control of various sectors of the economy and society with the interest of the nation as a whole.1 State-building followed divergent paths in Southern European countries.2 Portugal and Spain developed state-like institutions at a very early historical stage, whereas Greece and Italy did not become independent and unified countries until the nineteenth century. In the Iberian Peninsula after the secession of Portugal from the Spanish crown in the seventeenth century, central states evolved gradually within stable territorial boundaries. In Spain and Portugal, political unification preceded the development of modern nationalism. However, these two countries differed in that the Spanish crown encompassed a much more heterogeneous collection of cultures and political institutions than the Portuguese crown. In Spain, encroachment by the center had to surmount serious resistance on the part of peripheral movements. The revolt of the Catalans in 1640, the War of Succession at the beginning of the eighteenth century, and the continuous political conflicts and civil wars of the nineteenth century were decisive chapters in a drawn-out struggle leading to a modern centralized state. The Greek and Italian experiences were very different. In both countries nationalism played a critical role in the expulsion of foreign powers and the emergence of centralized political institutions. However, there are also important contrasts between the Greek and Italian paths. In Greece, absorption of territories previously dominated by foreign states continued until well into the twentieth century. In Italy, a national state was constructed from discrete political units by the end of the nineteenth century. Although their historical paths were distinctive, all Southern European states were developing highly centralized political systems by the beginning of this century. This trend was reinforced by the centralizing tendencies of authoritarian regimes suffered by each of these countries in the twentieth century. The contention that guides this chapter is that this development has been reversed over the past two decades. Of course, as with the previous process of central state-building, we find large and persistent variations across countries. But there are strong grounds for believing that the overall direction of development in Western Europe is toward the devolution of competencies both up the European level and down to various levels of subnational government. We describe the resulting pattern of policymaking as one of multilevel governance in which

238

Multilevel Governance

governments at different territorial levels share, rather than monopolize, public decision-making over wide areas of policy. The past several decades have seen the development of highly variable and nonhierarchical patterns of interaction among diverse actors in dispersed territorial arenas.3 As we noted above, policymaking is highly heterogeneous across policy areas. In some policy areas, such as monetary policy and agricultural subsidies, major decisions are made exclusively at the European level; in others, such as fiscal policy and most areas of social policy, they remain largely at the national level. In many areas of service provision, subnational governments play the major role; while in others, such as cohesion (structural) policy and the environment, governments at several levels participate in decision-making. This policy patchwork is overlaid by wide territorial variation, for there are sharp differences in the extent to which regions are mobilized politically, either by subnational governments or by regional parties. In Greece, Portugal, and Spain territorial relations have been shaped by the democratic transition that took place in each country during the 1970s. In Portugal, and particularly in Spain, decentralization has gone hand in hand with democratization. Decentralization has been promoted by the negative experience of centralism under former authoritarian regimes, by regional movements (most notably in Spain, and, to a lesser degree, in Portugal’s Atlantic islands during the 1970s,4 and by the need to strengthen democratic consolidation by integrating local and regional elites within the new political system. As we detail below in our discussion of ethno-territorial movements in Catalonia and the Basque Country, regionalization has been most ambitious in Spain. However, steps in this direction have been taken in all Southern European countries. In Portugal, the Constitution of 1976 called for the creation of the autonomous regions of the Azores and Madeira, each of them having its own executive and an elected assembly with significant competencies.5 The Constitution of 1976 also mandated administrative regions and regional assemblies in the rest of Portugal, though this was not put into practice until 1991.6 Regional Coordinating Committees were set up in Portugal in 1977 to design and implement regional development plans, a function that has been strengthened as a result of the EU’s cohesion policies (Opello 1992). However, in 1998, a referendum to create elected regional councils was defeated. Decentralization in Greece has been the most modest. Despite administrative reforms in 1982 and 1986 the regional councils [nomarchia] and administrative regions [perifereies] remain of marginal

Multilevel Governance 7

239

importance. The most important decision makers at the regional level are prefects who are appointed by central government. Council assemblies at the level of the prefecture (nomarchia) were created in 1982, but until the elections of October 1994, they were not directly elected, but selected by the governing party from among local officials, and their decisions can be annulled by prefects. However, against the background of the extreme centralism of the past, particularly under the military dictatorship of 1967 to 1974, the reforms of the 1980s can be considered a noteworthy innovation. The most important development in Greece was the passage of a new law in 1997, which reorganized the more than 6,000 municipalities and communes into 1,033 larger municipalities and enlarged the competences and jurisdiction of municipal authorities. Italy has not experienced a recent democratic transition nor has it had a legacy of strong ethno-territorial movements. However, it shares with other Southern European countries the experience of authoritarian rule and a corresponding reaction against centralized governance once it ended. Pressures for decentralization have been a continuing feature of the Italian polity. The regionalization process started with the 1948 Constitution, which distinguished between ‘special’ and ‘ordinary’ regions (see Leonardi 1992). Special regions (Trentino-Alto Adige, Val d’Aosta, Sardinia, Sicily, and Friuli-Venezia Giulia) were created to deal with ethnic, territorial, and linguistic conflicts, and they were given their own institutions in the 1950s (see Putnam 1993). The establishment of regional governments in ‘ordinary’ regions did not take place until 1970, largely because key Christian Democratic interests opposed the policy and because fundamental institutional change was impeded by the polarization of Italian politics. In 1978, the administrative resources of regional governments and the share of public spending devoted to them increased significantly (Leonardi 1992: 231). Despite such steps, Italian regions remain weak. They lack significant decisional competencies, have limited administrative capability, and receive nearly all their resources from central government.8 This gap between expectation and reality has helped to trigger regional mobilization, most notably in the Northern Leagues. Regional demands have also been voiced by the Conference of Regional Presidents which was established in 1983. In the early 1990s, the Conference of Regional Presidents called for regional financial autonomy, clear-cut areas of competency, regional representation at the national level and the introduction of subsidiarity as a constitutional principle.9 The same dissatisfaction has led Italian regions to cultivate transnational links with other regions and demand direct representation in the EU (Leonardi 1992; Grote 1996).

240

Multilevel Governance

Regional Mobilization in the EU A key feature of multilevel governance is that subnational governments are no longer constrained to dyadic interactions with central state actors, but connect instead with a variety of actors in different arenas. Until the post-World War II era the logic of conflict for regions that attempted to resist the process of state-building was bipolar, pitting recalcitrant regions against nation states intent on consolidating their grip on their respective territories. Even when ethno-territorial movements within the same state shared the goal of defending themselves against state encroachment, they were rarely able to coordinate their activity.10 The defining feature of such movements was their particularism, the defense of exclusive cultures and institutions, and as a result they found it difficult to establish durable cross-regional alliances. Regions that straddled national borders (e.g. the Basque Country) developed along different lines in each country and regionalists were drawn into political conflict with their respective national state. National states not only structured political conflict within their territories but served as a barrier to alliances between subnational actors located in different countries and between subnational actors and foreign states. In the EU, relations among territorially defined political actors are characterized by multipolarity in which actors participate in subnational, national, and supranational political arenas simultaneously.11 These arenas connect subnational actors with subnational actors in other states, with supranational actors, with their respective national governments, and even, in some cases, with foreign governments. The result is an intricate web of relationships. State executives no longer mediate relations between subnational actors and extranational actors. Supranational, national, and subnational arenas are more accurately described as interpenetrative than as simply interconnected (Jones and Keating 1995). The boundaries between these arenas are diffuse rather than nested. Instead of being nested exclusively within their respective national arenas, subnational actors are now engaged in a variety of transnational relationships in diverse supranational, national, and subnational arenas. The shift from bipolarity to multipolarity has transformed coalitional possibilities. Whereas under bipolarity, regional actors were confronted by centralizing states on a whole range of issues, in the emerging European polity they are embedded in a pluralistic setting in which they may ally with different actors at different levels on different issues.12

Multilevel Governance

241

The issues that divide mobilized ethno-territorial actors from state actors are no longer mutually reinforcing or predictable. Regional actors face a fluid situation characterized by cross-cutting rather than mutually reinforcing cleavages. For this reason, we hypothesize in the following section that multipolarity has a profound influence on the political orientation of regional actors, favoring pragmatic bargaining and demands for regional autonomy rather than separate statehood.13 But multipolarity by no means tames regional demands. In fact, it does the reverse. Insofar as regional actors have new sources of political influence and the means to build ethno-territorial constituencies, they may continuously upgrade their demands for an increased share of competencies and resources. Multilevel governance has undermined violent nationalism, but it has also intensified regional demands, and prompted a series of experiments with regional autonomy short of state creation. The situation we are describing is unprecedented in modern Europe, but it is worth emphasizing that we are not forecasting the demise of the territorial state. European integration and decentralization have pulled some important competencies away from national states, but they have by no means negated the immense resources still concentrated in the hands of central state executives. State executives remain the most influential political actors in the EU by virtue of their representation in the Council of Ministers and the European Council. The difference now is that national arenas are just one among a variety of channels that are available for subnational actors. The political reality of state executive power within the multilevel polity described above creates an ambiguous, double-edged, situation for regional actors, particularly those entrenched within their respective states. On the one hand, they are concerned that state executives may preempt their competencies by making laws at the European level over which they have little control. State executives have been able to project their power into the supranational arena by virtue of their dominance of treaty making in the EU.14 Entrenched domestic interests, including particularly subnational governments, are fearful of being outflanked by state executives as decision-making is shifted to the European level. The result is an ongoing struggle among contending institutions at the supranational, national, and subnational levels concerning decision-making, voting rules, and the role of subnational governments at the European level. On the other hand, regional actors see enhanced opportunities for influence in many areas of policymaking. While national governments dominate the process of institutional creation, they do not dominate many areas of policymaking (see Marks, Hooghe, and Blank 1996:

242

Multilevel Governance

343–78). The capacity to create new institutions does not imply the capacity to control how those institutions actually work. Recent years have seen the creation of new policies, especially in the area of cohesion policy, that are not exclusively determined by national governments and which bear directly on subnational actors and institutions (Marks 1992). European integration has, in a nutshell, created powerful supranational institutions and has provided regional actors with new avenues for influencing decision-making. Regional actors have mobilized in the European arena through diverse channels. Local and regional governments from several countries have set up autonomous offices directly in Brussels; subnational governments have established numerous formal and informal networks; in regions designated for EU cohesion funding, subnational officials help design and implement economic development plans alongside national and Commission officials; and subnational governments are represented in highly visible, though primarily symbolic, assemblies—most notably the Committee of the Regions established in the Maastricht Treaty. Subnational governments in Southern Europe, and particularly from Spain, have taken advantage of these channels. They are represented in the Assembly of European Regions, the Conference of Peripheral Maritime Regions, the ‘Four Motors for Europe’, Eurocities, the Association of Regions of Traditional Industry, C-6 (a network of six French and Spanish Mediterranean cities), the European Association of Border Regions, the Union of Capital Regions, associations covering the Alps and the Pyrenees, the Association of Frontier Regions, and the Coalfields Communities Campaign. Catalonia has been particularly active in creating subnational networks. Catalans have chaired the Assembly of European Regions and the Council of Municipalities and Regions, and Pasqual Maragall, President of the Catalan Government, has served as President of the Committee of the Regions (Mun˜oz 1993). But, and this is a distinctive feature of multilevel governance, regions do not engage in these activities equally. There is no congruence, or even convergence, in the political role of regions in the EU. Instead, there are enormous disparities in the level of organization, financial resources, political autonomy, and party-political mobilization of regions across Southern Europe. The creation of new avenues for regional mobilization brings into focus persistent differences in the organizational and political capacities of regional actors. At the one extreme, Spanish Comunidades Auto´nomas (along with German La¨nder and Belgian regions) are well funded, strongly institutionalized, entrenched within their respective states, and active in the European arena. At the other extreme, regional governments are much

Multilevel Governance

243

weaker in Greece or Portugal. If one wishes to find subnational political actors in the latter countries, one has to look to town and city mayors who, even if they are aware of opportunities in the European political arena, lack the resources to mobilize there. There is, therefore, little sign of the kind of territorial convergence implied in the notion of a ‘Europe of the Regions’. In perhaps no other area of political institutionalization in the European polity is there greater diversity than in the territorial politics of the member states. Multilevel governance seems to be leading not to uniformity but to continued diversity as contrasting regional actors are brought together within an overarching polity.

The Transformation of Ethno-Territorial Movements What is the effect of multilevel governance on territorial conflict? In this section we examine this question by looking at two regions in particular: the Basque Country and Catalonia. These are regions with strongly entrenched ethno-territorial movements that go back well before European integration was underway and where, as a consequence, the effects of European integration and multilevel governance may be analyzed in historical perspective.15 The Basque Country and Catalonia have the strongest and most durable ethno-territorial movements in Southern Europe, and are two of the four most influential movements (along with the Flemish and Scottish) in Western Europe. In both the Basque Country and Catalonia, ethno-territorial movements challenge their respective central states directly in demanding new competencies. For these reasons they are interesting cases in which to examine the dynamics of multilevel governance. The consolidation of democracy in Spain went hand in hand with farreaching decentralization. The reasons for this go beyond the pressures mounted by Basque and Catalan nationalists. Those pressures were effective due to the vulnerability of the democratizing political elite during the transition process and the widespread rejection of the Francoist model of traditional Spanish nationalism involving the suppression of peripheral cultures.16 Regionalism in Spain has been a dynamic process. Concessions to Basques and Catalans precipitated demands for decentralization on the part of other, less distinct, regions that feared they would lose resources and influence if they did not mobilize on their own account. Central policymakers believed that they could tame Basque and Catalan pressures by creating federal institutions for the country as a whole. In one sense, those pressures have been tamed, but in another sense they have been intensified. Violent nationalism has been decisively

244

Multilevel Governance

weakened where it once had widespread appeal, in the Basque Country. Participation in a system of multilevel governance and the multipolar character of political coalitions in this new setting have made political–territorial conflicts less explosive and more mundane. The logic is essentially the same as that underlying the moderation of revolutionism under liberal democracy: channels for authentic participation and a share of control reduce the intensity of political opposition. However, at the same time, these new channels and regional powers have ratcheted up demands for self-government and provided ethnoterritorial parties more political leverage to attain their goals. Moderation of means—operating within the political system—does not imply timidity of goals. In both the Basque Country and Catalonia, the demands articulated by nationalist movements go beyond federalism. The constitutional and financial reforms that have been undertaken intensify, rather than dampen, the political struggle about how to structure a multilevel polity. This underlying similarity exists despite sharp contrasts in the character and context of ethno-territorial nationalism. Whereas the claims of Catalan nationalists concern the territory of the autonomous community of Catalonia (a former principality within the crown of Arago´n), the territories claimed by Basque nationalists are divided into two states (France and Spain) and into two autonomous communities within Spain (the Basque Country and Navarra). Basque nationalists have been far more extreme in their tactics and goals than Catalans. Independent statehood has long been a staple demand of the Basques; Catalans have been readier to explore independence short of statehood. Finally, and perhaps most importantly, there is a strong terrorist organization in the Basque territories (ETA) that has no parallel in Catalonia.

Basque Nationalism: Bargaining and Confrontation Under the Threat of Terrorism The Basque nationalist movement has long rejected the Spanish state and demanded full independence.17 After Francoism, radical nationalists were willing to use violence in pursuit of their goals. The more moderate Partido Nacionalista Vasco (Basque Nationalist Party [PNV]), which is the largest party in the Basque Country, at first refused to legitimate the Spanish democratic transition. The PNV campaigned, with some success, for abstention in the 1978 constitutional referendum. Although the PNV was initially reluctant to endorse the Spanish democratic path, once democratic regional institutions were set up,

Multilevel Governance

245

the party was gradually integrated in a multilevel polity. With the exception of the militant Euskadi Ta Askatasuna (ETA) and its political wing, Herri Batasuna (HB), Basque nationalists opted for enhanced self-government within Spain rather than the creation of an entirely separate state. The coalition between the PNV and the Socialist Party (PSE–PSOE) in the Basque regional government was perhaps the most telling example of the changing nature of alliance and opposition in the new institutional framework.18 This coalition governed from 1986 to 1998, except for a brief period in 1991 when a nationalist coalition, composed of the PNV and two nationalist parties, Eusko Alkartasuna (EA) and Euskadiko Ezkerra (EE), prevailed. Despite the potential for conflict between the PNV and PSE on the national question, the coalition proved remarkably resilient—in sharp contrast to the nationalist coalition which lasted less than ten months. Multilevel governance is conducive to complex alliances and crosscutting cleavages rather than a single line of combat. The coalition between the PNV and PSE–PSOE operated at the regional and local level, but not in the national political arena. After the 1993 General Election, the PSOE turned to the Catalan CiU rather than to the PNV for support. But this did not prevent the PNV from moderating its attacks on the Socialist government. Whereas the Conservative Partido Popular (PP) and the left-wing Izquierda Unida repeatedly called for the resignation of the Socialist prime minister, Felipe Gonza´lez, the PNV supported Gonza´lez and his Socialist government. Although the Socialists resisted Basque demands for more political autonomy, the PNV feared (wrongly as it turned out) that a conservative government would centralize authority. After the victory of the PP in the 1996 general elections, the PNV supported Jose´ Maria Aznar’s government on budgetary and other issues in return for concessions on certain nationalist demands. However, this did not prevent the PNV from continuing regional, provincial, and local coalitions with the PSE in the Basque Country. Between 1979 and 1998 the PNV gradually dropped its demand for full independence and national sovereignty. Jose´ Antonio Ardanza, President of the Basque government from 1985 until 1998 and a leader of the PNV, argued that multilevel governance made traditional Basque demands for independence and statehood less pressing: Nations—and I am referring both to those nations that identify with a state and to those without a state—will no longer consider the state as their only model of self-fulfillment. The maintenance or achievement of sovereignty and statehood will no longer serve as a priority and obsessive objective . . . .

246

Multilevel Governance

We think that in the new Europe we want to build, it is possible and convenient to concentrate our efforts on national construction free from statist obsessions. As far as we are concerned, we are ready to untie the knot that has indissolubly linked nation and state.19

Regionalism has weakened violent nationalism. Both ETA, the terrorist organization, and HB, the radical national party, lost support from the mid-1980s as regional self-government was introduced. Reformist nationalist parties that participated in government were drawn into closer collaboration against violence.20 The Pactos de Ajuria Enea of 1988 united all Basque political parties except HB in support of continued regionalism and against violence. The Basque government and its police force were controlled by a Basque nationalist/socialist coalition and they, rather than the Spanish (or French) police, were in the front line against nationalist violence. It was one thing for Basque terrorists to attack Spanish police, but it was quite another thing for them to confront a Basque police force that was widely regarded as legitimate in the region. The police campaign against ETA has been increasingly successful. The successive imprisonment of ETA leaders created disarray within the organization. At the same time, antiviolence movements have gained considerable support in the Basque Country. The murder of several members of the PP, especially of the city counselor of the Basque city of Ermua, Miguel Angel Blanco, in 1997, after he was kidnapped and held hostage, led to a wave of civic revulsion that surprised ETA and HB and induced them to alter their strategy. Paradoxically, the weakening of ETA and HB paved the way for the creation of a broad front calling for national independence. In September 1998, the PNV, EA, and HB, alongside other nationalist groups, concluded the Lizarra agreement which declared that the Basque conflict could only be resolved on the basis of self-determination for the citizens of Euskal Herria. Those living in the Basque territories should have ‘the last word’.21 Neither the national frontier between Spain and France, nor the preferences of those living outside the Basque territories should weigh in finding a solution to the Basque question. The Lizarra declaration also called for the immediate creation of new institutions to meet ‘the aspirations of sovereignty’ of Basque citizens.22 By signing the Lizarra agreement the PNV moved toward a radical assertion of sovereignty and self-determination, though it did not demand a separate state. A few days later ETA called for a truce, the result, according to ETA, of a secret compact between ETA and the PNV signed in August 1998. According to ETA, the PNV and EA agreed ‘to sever their ties with

Multilevel Governance

247

Spanish parties and create a unique and sovereign institution that would include all Basque territories’ (i.e. in France and Spain).23 These agreements among Basque nationalists led to the establishment, in February 1999, of an assembly of nationalist representatives in local institutions (udalbiltza).24 They also led to the formation of a coalition among the PNV, EA, and EH in the Basque parliament, even though EH did not enter the Basque government and continued to contest its legitimacy on the grounds that it reinforces the division between the Basque territories. The transformation of the PNV was paralleled by a realignment of the Basque party system. After the Lizarra agreement, HB reorganized itself as Euskal Herritarrok (EH) in an attempt to broaden its appeal to disillusioned militants and nationalists who had publicly condemned violence. The reorganized party increased its share of the vote in subsequent regional (1998) and provincial, local, and European elections (1999).25 At the same time, the PNV and EA formed an electoral coalition to stem the loss of votes to EH on one hand and the Socialists and PP on the other. The PSE and especially the PP sought the votes of moderate nationalists who had formerly supported the PNV, and indeed both parties gained support in regional and local elections at the expense of the PNV and EA. The net result was a sharp reversal for the PNV. It lost control over several local and municipal governments, including the Basque capital (Vitoria-Gasteiz) and the provincial government of Alava, which shifted to the PP.26 Although the PNV went so far as to declare the Basque statute null and void, demand radical restructuring of the Basque institutional framework, and sign a parliamentary agreement with EH, ETA suspended the truce in November 1999.27 However, this did not change the position adopted by the PNV leadership. In December 1999, a few days after ETA suspended the truce, the PNV reaffirmed its will to move toward a common political–institutional framework for all the Basques. Nevertheless, due to the persistence of violence, the internal divisions of the nationalist forces, and the weakness of nationalist support in certain Basque territories, the nationalist front failed to unleash a process leading to self-determination, and ended up collapsing among increasing reproaches between democratic and violent nationalists. The folding of the nationalist front preceded the polarized regional elections of March 2001, in which the coalition formed by PNV and EA benefited from the sharp decline of radical separatists and managed to improve its position in the Basque parliament.28 After these elections, the PNV and EA constituted a new coalition government under

248

Multilevel Governance

the leadership of Juan Jose´ Ibarretxe.29 However, these were also the regional elections in which electoral support for all nationalist parties dipped to its lowest level since the statute of autonomy of 1979.30 In 2004, the Basque parliament passed a proposal to drastically reform the Basque statute of autonomy, the so-called Ibarretxe Plan. According to this project, the Basque Country would enter a status of ‘free association’ with Spain. This proposal was rejected by the Spanish parliament with the votes of the Socialist and Popular parties. Following this rejection by the Spanish parliament, Juan Jose´ Ibarretxe called new elections aimed at strengthening the institutional and bargaining position of both the PNV and the Basque government. However, in the 2005 Basque elections the PNV experienced a serious setback. Although the PNV–EA electoral coalition remained the most voted for political force, it lost more than one hundred and fifty thousand votes (falling from 42.7 to 38.6 percent of the vote) and four seats in Parliament. By contrast, the Socialist Party, which was now adopting a less confrontational profile and proposing a negotiation on a renewed statute of autonomy compatible with the Spanish Constitution, experienced a strong rise in both votes (from 17.9 to 22.6 percent) and seats (from thirteen to eighteen). The radical nationalist left, now represented by the Communist Party of the Basque Territories (EHAKPCTV) attained also better results than in the 2001 elections (12.5 percent of the votes in 2005 vs. 10.1 of Batasuna in 2001) and seats (nine in 2005 vs. seven in 2001). The electoral advance of the radical separatist left took place after a long period without ETA killings and was paralleled by declarations made by the radical left underscoring that the times of confrontation should be replaced by politics and negotiation among all political forces in the Basque Country. The PNV setback makes it very difficult for this party to reach stable majorities in parliament and to advance unilaterally toward the implementation of the Ibarretxe Plan. Recent developments have shown both the failure of nationalist sectors to give rise to a new political framework and the resilience of nationalist orientations among the population of the Basque autonomous community. Neither nationalists nor non-nationalists can unilaterally impose a stable institutional framework, and all political groups, and first and foremost supporters of radical nationalism, will have to eventually accommodate in a very plural political setting. This development may be ultimately facilitated by the increasing weakness of ETA, and by the awareness, on the part of the separatist left, that a strategy based on institutional and political participation must displace the politics of violence and confrontation.

Multilevel Governance

249

Catalan Nationalism: Pragmatism and Assertiveness In contrast to Basque nationalism, Catalan nationalism has historically been oriented toward greater political autonomy within the Spanish state rather than the creation of an entirely separate state.31 This facilitated the integration of Catalan demands into the Spanish Constitution of 1978 and the Statute of Autonomy of the following year. Converge`ncia i Unio´ (CiU), a moderate nationalist center-right coalition, became the major political party in the Catalan regional parliament in 1980, and its leader, Jordi Pujol, has presided over successive Catalan governments since that time.32 The CiU has combined pragmatism and assertiveness. From the beginning of the Spanish transition Catalan nationalists pressed their demands in ways that were compatible with the Spanish Constitution. In fact, the part of the Constitution that specifies the position of regions resulted from a compromise with Catalan nationalists and, in particular, their speaker in the Spanish Parliament, Miquel Roca. Nevertheless, there has been continual conflict between the government and Catalan nationalists. After the failed military coup of 1981, PSOE felt obliged to reassure Spanish nationalists by setting limits on regionalism. The same years saw an escalation of Catalan demands for greater political autonomy. Like the PNV, the CiU has been willing to enter governing coalitions with national and other parties. But only recently has this been necessary for the CiU at the regional level. From 1984 to 1996, the CiU had absolute majorities in the Catalan Parliament, and from 1996 to 1999 the party could rely on the support of several smaller parties without having to enter a formal coalition. However, in the regional elections of 1999, a left-wing coalition led by the former mayor of Barcelona, the Socialist Pascual Maragall, received more votes than the CiU. The CiU remained the largest party in the Catalan Parliament due to the overrepresentation of rural areas, but it was compelled to gain the support of the PP and/or the center-left nationalist Esquerra Republicana de Catalunya (ERC). In November 1999, Pujol was reelected to the Presidency of the Catalan Government with the support of PP deputies, while ERC deputies abstained. Relations between Spanish and Catalan parties have reflected the party identity and parliamentary strength of the Spanish government. From 1982 to 1993, the PSOE had absolute majorities in the Spanish parliament and the result was a standoff between two entrenched governments (Morata 1995). By contrast, from 1993 until 1996 the PSOE depended on the parliamentary support of the CiU, which in

250

Multilevel Governance

exchange demanded a higher level of financial transfers from the central state on the grounds that there is a large gap between the tax revenue extracted from Catalonia and central state expenditure there.33 Under the new system, regional governments received 15 percent of the direct taxes that the Spanish government collects in their respective territories (see El Paı´s, February 6, 1994). The alliance between CiU and the PSOE became more apparent—and more precarious—in the wake of the corruption scandals implicating several wellknown incumbents, including the former governor of the Spanish central bank and the chief of the Civil Guard. Despite growing discontent among the electorate and deep political crisis in the government, the CiU supported Felipe Gonza´lez’s refusal to resign. This placed the CiU in a difficult political position insofar as it ran the risk that some of its voters might switch to the PP. The CiU supported the Gonza´lez government until the end of 1995, when the CiU failed to vote for the budget thus prompting new general elections for 1996, one year before the end of the legislature. After Aznar’s limited victory in the 1996 elections, the CiU maintained a similar relationship with the PP. After the elections, Catalan nationalists helped to elect Aznar as president of the Spanish government, and they supported the government budget from 1996 to 2000. This close cooperation followed years of tensions between the CiU and the PP.34 After the 1996 agreement between these parties, the percentage of direct taxes that reverted directly to regional governments rose to 30 percent. In addition, after 1996 regions could themselves introduce exemptions or targeted reductions of certain taxes.35 In the general elections of the year 2000, the PP attained the majority in the Spanish parliament. After these elections, no stable parliamentary alliance between CiU and the PP was formed. However, further steps toward the fiscal autonomy of regional governments were adopted. In 2001, a new fiscal agreement between the Spanish government and fifteen autonomous communities (including the Catalan) was reached. By virtue of this agreement, autonomous communities directly receive 33 percent of the national income taxes and 35 percent of the valueadded taxes that are collected in their respective territories. In addition, regional governments will now have full legislative control over their shares of income taxes. The CiU’s electoral success has generated opportunities for participation in governing coalitions, and this has given rise to internal strains within the party. Before the 1993 general election, the CiU was riven by the Socialist party’s offer of a coalition. After an intense debate on the issue at the October 1992 congress of Converge`ncia Democra`tica de Catalunya (a constituent of CiU), the General

Multilevel Governance

251

Secretary of CDC, Miquel Roca, resigned on the grounds that he favored closer cooperation with the Socialists.36 The issue arose again after the election when the Socialist party repeated its offer. The CiU still refused in principle, but agreed to cooperate on a case-by-case basis. This strategy was controversial, both within the moderate wing of Converge`ncia and particularly within its Christian democratic wing, Unio´ Democra`tica de Catalunya. However, the prestige of the President of the Catalan government, Jordi Pujol helped to neutralize dissent and establish a stable pattern of parliamentary association without participation in the Spanish government. The ongoing conflict between the CiU and the Spanish government over competencies and the CiU’s adeptness at exploiting existing political channels created the perception that a CiU-led regional government is the most effective way to get more for Catalans. This helps to explain the frequency of split ticket voting in Catalonia, with the Socialist party receiving the majority share of votes in national elections and the CiU most often receiving the majority at the regional level (see Montero and Font 1991: 7–34; Riba 2000). Rather than satisfying demands for political autonomy, Catalan selfrule, unmediated by overarching political parties, seems to have deepened ethnic identity, reinforced the center-periphery cleavage, and ratcheted up political ambitions. In 1978, the CiU bargained for a Spanish ‘state of autonomies’, an arrangement that fell far short of federalism. Twenty years later, the goal of a Spanish federation has been taken up by many CiU leaders. In 1994, a document published by the Catalan government proposes that all organs of the Spanish state in the region should become subordinate to the Catalan government which would assume nearly all the competencies of the Spanish executive within Catalonia. The Catalan President would antecede the Spanish Prime Minister as the main representative of the Spanish State in Catalonia. In the proposed framework, the Catalan President would have a direct relationship to the Spanish King, who would continue to play a symbolic, moderating role.37 This proposal which points in the direction of a loose Spanish confederation, is consistent with statements by Jordi Pujol, President of Catalonia, stressing that the situation prior to 1714, in which the monarchy was the chief link among the different parts of Spain, is more suitable for Catalonia today than the current system.38 In July 1998, in the Declaration of Barcelona, the CiU joined the PNV and the Bloque Nacionalista Galego in a common demand for the recognition of the national status of Catalonia, Euskadi and Galicia, and for a deepening of the devolution process. This declaration was followed by other steps, such as the Agreement of Santiago de Compostela, which demanded a new

252

Multilevel Governance

interpretation of the Spanish Constitution in order to fit the multinational character of Spain.39 The Catalan elections of 2003 and the Spanish general elections of 2004 involved crucial changes in the relative importance of Catalan nationalist parties. They also created a new scenario for the advance of demands of autonomy. After the 2003 Catalan elections, and even though the CiU attained more seats than any other party, a coalition government including the PSC, the nationalist ERC, and the left-wing Iniciativa per Catalunya-Els Verds was formed. This new government was presided over by the former major of Barcelona, the Socialist Pascual Maragall, and included also a strong institutional presence of the ideologically independentist ERC. The 2004 Spanish general elections further increased the political importance of the ERC. Not only did this party experience a sharp rise in these elections, but also became a key and necessary parliamentary ally for the Socialist party. These changes in the Catalan and Spanish political landscape are determining important institutional changes. These changes include the development of a new statute of autonomy for Catalonia and the initiation of a process of constitutional reform in Spain.

Toward Multiple Identities in the Basque Country and Catalonia Might multilevel governance affect territorial identities as well as the intensity of territorial conflict? In this section we examine the extent to which individual identities have been transformed in the context of multilevel governance. Our guiding hypothesis is straightforward: individual identities to a territorial community are strengthened by experience of legitimate authority exercised by actors representing that community. Our presumption is that political identities are somewhat flexible over time, and that they are shaped by the political activities in which citizens are engaged—as participants in the body politic, as voters, as members of social movements, interest groups, and associations that are mobilized in the territorial arena, and as subjects of authoritative decisions.40 To the extent that competencies are diffused across different territorial levels within a system of multilevel governance, we hypothesize the emergence of multiple identities, that is coexisting identities to several territorial communities. Historical variation in identity is almost certainly linked to political conflict. Exclusive national identities emerged with the development of strong states that were intent on gaining the total allegiance of those living in their territories. In the experience of Western Europe, war was

Multilevel Governance

253

the crucible in which local, regional, and national identities were melded into strong national identities, pitting one nation against another. Multiple identities, by contrast, arguably moderate conflict among communities to the extent that they share an overarching loyalty. The logic of multiple identity appears to parallel that of crosscutting cleavages which diminish the intensity of conflict because they join members of otherwise contending groups (Lipset 1981/1959). There is some evidence for the existence of such multiple identities among Europeans. Surveys conducted by Eurobarometer in November 1991 and May/June 1995 find on average that individuals report as strong an attachment to their region as to their country. Only in Denmark, Finland, Ireland, and the United Kingdom is attachment to country stronger than attachment to region, while in Greece, Italy, and Portugal the two were evenly matched. The surveys include the usual battery of social background questions, and a study commissioned by the EU reports that strength of regional attachment is positively associated with youth, education, and positive attitudes to the EU (Ravet 1992: 26). However, data on change in individual orientations over a number of years is difficult to come by. The questions on territorial attachment that were posed in Eurobarometer surveys in 1991 and 1995 were not asked before, nor have they been repeated since. In searching for comparable data over time, one faces a rather sharp trade-off between breadth of territorial coverage and comparability over time. Fortunately, survey items included in a number of DATA S.A. surveys in the 1970s and 1980s were repeated in surveys carried out by the Centro de Estudios Sociolo´gicos in 1996. This gives us the possibility of examining changes over a period of critical political change in two regions that are of particular interest to us (see the footnote on Table 6.1 for details). The first survey was conducted in 1979, the year in which the statutes of autonomy for Catalonia and the Basque Country were passed. The second survey was conducted roughly fifteen years after the establishment of regional governments in the Basque Country and Catalonia. Finally, the third survey was conducted during Aznar’s second term as prime minister, in a period marked by a PP absolute majority in the Spanish parliament and serious polarization on the Basque political scene. Tables 6.1 and 6.2 show the overall frequencies of territorial identities in the Basque Country and Catalonia. These tables reveal that from 1979 to 1996 the percentage of people having balanced multiple identities, that is, who claim to be both Spanish and Basque or both Spanish and Catalan, increased by 10.2 percent in the Basque Country and 3.6 percent in Catalonia.41 From 1996 to 2002, this percentage increased by 0.7 percent in Catalonia, but

254

Multilevel Governance

Table 6.1 Identities* in the Basque Country (Comunidad Auto´noma Vasca) Self-identity

1979

1996

2002

Only Basque More Basque than Spanish Both Basque and Spanish More Spanish than Basque Only Spanish Don’t know/no answer N¼

38% 12 26 6 14 4 1,011

21% 30 36 4 5 4 276

25% 19 34 3 5 15 576

*Respondents in the 1979 survey were asked to choose among the following: ‘Espan˜ol,’ ‘ma´s espan˜ol que vasco,’ ‘tanto vasco como espan˜ol,’ ‘ma´s vasco que espan˜ol,’ and ‘vasco.’ Response options were worded and ordered somewhat differently in the 1996 CIS survey: ‘Me siento u´nicamente espan˜ol,’ ‘me siento ma´s espan˜ol que vasco,’ ‘me siento tan espan˜ol como vasco,’ ‘me siento ma´s vasco que espan˜ol,’ or ‘me siento u´nicamente vasco’. Sources: For 1979, Linz 1986: 43 (based on data from DATA survey for Richard Gunther, Giacomo Sani and Goldie Shabad, as described in Gunther, et al. 1986: 425–45); for 1996, CIS survey (2,228); and for 2002, CIS survey (2,455).

Table 6.2 Identities* in Catalonia Self-identity

1979

1996

2002

Only Catalan More Catalan than Spanish Both Catalan and Spanish More Spanish than Catalan Only Spanish Don’t know/no answer N¼

13% 11 33 6 28 9 1,232

11% 26 37 12 13 2 782

16.2% 24 37 8 12 2 1,682

*Response options in the two surveys were identical to those described in Table 6.1, except that ‘catala´n’ replaced ‘vasco.’ Sources: For 1979, Linz 1986: 43 (based upon data from DATA survey for Richard Gunther, Giacomo Sani, and Goldie Shabad, as described in Gunther et al. 1986: 425–45); for 1996, CIS survey (2,228); for 2002, CIS survey (2,455).

increased by 2.5 percent in the Basque Country. Table 6.3, which combines responses of all those who have multiple identities, whether balanced or one-sided, shows dramatic changes from 1979 to 1996. In this period, in the Basque Country multiple identities rose by 25.9 percent, and in Catalonia by 23.8 percent. By contrast, from 1996 to 2002 multiple identities dropped 14.1 percent in the Basque Country and 4.5 percent in Catalonia. However, Basque figures for 2002 must be

Multilevel Governance

255

Table 6.3 Percentages of respondents in the Basque Country and Catalonia with multiple identities* Self-identity

1979

1996

2002

Basque Country Catalonia

44% 50

70% 74

56% 69

*These percentages are the sums of those selecting the second, third, and fourth response options in Tables 6.1 and 6.2.

taken with special caution due to the dramatic increase in the percentage of DK/NA answers (from 4 to 15.1 percent). It did not take into account DK/NA answers, and in the Basque Country multiple identities would have dropped from 72.8 percent in 1996 to 65.7 percent in 2002. In both regions, the proportion of the population who regard themselves as exclusively Spanish has declined sharply and is now small: 5.4 percent in the Basque Country and 12.9 percent in Catalonia. Likewise, in Catalonia, the proportion who are exclusively Catalan is just 11 percent. In the Basque Country it is higher, but has fallen from 38 percent in 1979 to 20.7 percent in the 1996. A large majority of Catalan and, in more recent years, Basque citizens, view their peripheral and Spanish identities as coexisting rather than mutually exclusive. Without overinterpreting these data, we can say that they are consistent with the general argument we have developed concerning multilevel governance and our hypotheses concerning multiple identities. We do not merely find that exclusive Spanish identity has declined in the Basque Country and Catalonia, but we find as well that multiple identities have increased in both regions after 1979. However, whereas from 1979 to 1996 there were dramatic increases in nested identities, from 1996 to 2002 multiple identities have declined, particularly in the Basque Country. The 1998 shift of the PNV to a policy aiming at Basque sovereignty is one of the most crucial factors affecting these developments. In addition to this, the fact that Aznar’s government adopted a response of uncompromising resistance against these upgraded nationalist demands, presenting itself as the main bulwark against the risks of secession and emphasizing the links between those demands and ETA, may have contributed to the decrease of nested identities. Even if we assume that the decisions made by political leaders were conditioned by their electoral opportunities (the competitive advantages of adopting staunch positions regarding the Basque crisis, both in the Basque Country and outside of it) and institutional capabilities (the absolute majority attained by the PP in the Spanish

256

Multilevel Governance

Parliament in 2000), our data show that the strategies adopted by political elites can have direct effects on the evolution of territorial and cultural identities.42

Conclusion The emergence of multilevel governance in Southern Europe has created a new context for peripheral nationalism as well as new ways for regional governments to press and express their interests. The emerging European polity offers them expanded opportunities for influence and new potential coalition partners in diverse political arenas. In short, peripheral nationalists are being drawn into a variety of multipolar relationships that reward bargaining more than confrontation. On the other hand, the creation of a multilevel polity has by no means tamed demands for regional empowerment. The sheer political weight of several regional movements has increased over the past three decades in Italy and especially in Spain, and regionalization is on the agenda even in highly centralized countries, such as Greece and Portugal. Over the past decade, the channels available to regional actors have multiplied beyond recognition. The shift of competencies to the European arena has exerted a powerful magnetic force on subnational actors, particularly those in Spain which are entrenched within their national political arena. Alongside the Committee of the Regions, established under the Maastricht Treaty, and encompassing subnational actors across Southern Europe, are less publicized, but more potent, channels of regional representation: regional offices, transnational regional coalitions, and associations of every kind. Regional actors may now mobilize in a trans-European polity, with diverse associational and coalitional possibilities, alongside their respective national polities. The cases of the Basque Country and Catalonia suggest that nationalist movements can escape from the view that the national state is the dominant, indeed the only, political form in which national aspirations can be realized. The emergence of multilevel governance encompassing states and regions within an overarching polity poses a competing vision in which regions and supranational actors are empowered at the expense of central states. Scenarios of multilevel governance do not challenge the defining feature of states as institutions monopolizing legitimate coercion in a given territory. But this minimal notion of state sovereignty no longer captures the realities of power in Europe, and it is less and less useful for distinguishing among alternative modes of governance. As both the former Basque regional president, Jose´

Multilevel Governance

257

Antonio Ardanza, and the former leader of the Northern League, Gianfranco Miglio, have pointed out, regionalization and European integration are eroding sovereignty and paving the way for a new kind of polity in which the material, legal, and symbolic character of the state will be transformed (Ardanza 1993; Miglio 1993). European integration poses an alternative to national states as the basis for ordering political life in contemporary Europe. To the extent that regions find a way of gaining meaningful autonomy in the European multilevel polity short of statehood, so the national state is less likely to be perceived as the culmination of ethno-territorial ambitions. Even from the perspective of nationalist parties which have (at least rhetorically) separatist leanings, the new European polity creates a framework in which the national state loses its exclusive appeal, both symbolically and as a strategic goal. Along these lines, Jordi Pujol, the former President of Catalonia, has argued that Catalonia’s problems will not be solved through secession, but rather by restructuring the Spanish political system in a way that reflects its multinational character (El Paı´s, February 9, 1994). The weakening of state control has led some peripheral nationalists to argue that traditional demands for independent statehood are becoming irrelevant. In Jose´ Antonio Ardanza’s words, ‘we are ready to untie the knot that linked nation and state’ (Ardanza 1993: 38). This is not to suggest that peripheral nationalist movements are withering away. Separatism has become weaker, but territorial movements in general are at least as assertive as they were in the past. Territorial identities are, if anything, becoming stronger, not weaker, in the process of European integration. Instead of subsuming national identities under a common European identity, the continued existence of strongly entrenched state executives in the Council of Ministers may actually enhance the sense of divergent national interests. This, we have argued, sharpens subnational identities, as distinct ethno-territorial groups countermobilize to press their particular demands at the European level through a variety of formal and informal channels. Regional empowerment is potentially self-reinforcing. Competencies not only provide regional governments with the capacity to make policy, but they give them the opportunity to deepen their own bases of support. Political patronage, cultural, and linguistic policies are as important for regional governments as they were for state builders in shaping identities. As Levi and Hechter indicated, regional empowerment is immensely helpful for the electoral growth of ethno-territorial groups, which can now allocate scarce resources to foster ethnic mobilization (Levi and Hechter 1985). Regional empowerment has led the Catalan, Basque, and even Galician governments to ratchet up their demands. The proposals of

258

Multilevel Governance

the former Galician regional President (a Conservative politician who was a minister in the Franco regime) and of the former Catalan President for a thoroughgoing federal transformation of the Spanish state indicate that subnational demands have grown wider since the Constitutional reform of 1978. While we diagnose a common direction of development across Southern Europe, there are few grounds for supposing that these countries will converge. In the first place, their starting points are vastly different. The powers of subnational governments and the degree of regional mobilization vary from country to country about as much as any political variable one can think of. Second, the creation of a common set of political opportunities in the EU has had very different effects depending on initial dispositions. Entrenched subnational governments are best positioned to exploit the changing structure of political opportunity. New channels for the mobilization of regional governments at the European level have actually widened the gap between strongly articulated Spanish regional governments and weak subnational units in Greece and Portugal which are virtually silent. Heterogeneity of territorial relations across Southern Europe is therefore not a passing phase, but a durable feature of the emerging Euro-polity.

Notes 1. For a more detailed discussion of state-building and European integration, see Marks 1997. 2. For a detailed analysis of the similarities in the economic, social, and political histories of Southern European countries, see Giner 1986. 3. On the characteristics, types, and implications of multilevel governance, see Hooghe and Marks 2003. 4. On the conflicts leading to the special political situation of the Azores and Madeira, Gallagher 1979. See also Guilherme Reis Leite 1992. 5. These transfers enabled the autonomous regions to pass legislation in matters of their special interest, to regulate national laws, and to initiate legislation in the national parliament. See Pereira 1995. 6. However, not all the members of the regional assemblies would be elected by direct suffrage. Some would be elected by the municipal assemblies of each region. See Opello 1992: 162–86. 7. See Papageorgiou and Verney 1992: 139–61; Verney 1994; Featherstone and Yannopoulos 1995; Ioakimidis 1996. 8. For instance, whereas the central administration employs more than 2,000,000 civil servants, only about 80,000 civil servants work for the regional administration (see Desideri 1995). Although the regions control between 20 and 25 percent of public spending, more than 90 percent of

Multilevel Governance

9.

10.

11.

12. 13. 14. 15.

16.

17.

18. 19.

259

that amount is directly transferred from the central government. In addition, most of the regional income has to be directed toward areas in which regional governments lack autonomous competencies (see Hine 1993: 268–9). On the characteristics of the taxation system of the special regions, see Desideri 1995. These demands were expressed in the documents ‘Per un nuovo stato regionale’ (1991) and in the ‘Carta delle Regioni d’Italia.’ See Leonardi 1992. Even though some peripheral movements attempted to establish links with national political projects, as some wings of Catalan nationalism did in the first third of this century, the centralizing trends of national states promoted bipolar conflicts between peripheral movements and their respective states. We use the term ethno-territorial movement to distinguish ethnic movements that are territorially based. Hence, it enables us to exclude a movement that has a distinctly ethnic character, but whose constituency lacks a connection with a defined territory, a category which would include many religious and almost all immigrant groups. We should note that the concept of multipolarity here is different from the concept as it is used by neorealists, in that we encompass not only nation states, but a variety of institutions having heterogeneous constituencies and competencies across different levels of government often in non-exclusive territories. On the fluid and complex character of the relationships among territorial actors within the new European polity, see Hooghe 1995. On pragmatic bargaining among subnational, national, and supranational institutions in the Belgian case, see Hooghe 1995. See Moravcsik 1993: 473–524; and Marks, Salk, Ray, and Nielsen 1996: 164–92. In this respect the Basque Country and Catalonia are unlike Northern Italy, where party-political mobilization, in the form of the Northern League, is a recent development. On the causes and characteristics of the process of regionalization in Spain see Pe´rez-Dı´az 1993: 194–204. An analysis of the relationships between the party systems of Catalonia and the Basque Country, and the Spanish democratic transition can be found in Gunther, Sani, and Shabad 1986. On the characteristics of Basque nationalism and the diffusion of separatist orientations in the movement, see Linz 1986; Dı´ez Medrano 1995; and Juaristi 1997. A study that emphasizes the rhetorical character of Basque separatism and traces its historical origins is Juaristi 1994: 127–30. After the absorption of a part of a former left-wing nationalist organization, the current name of the PSOE-PSE is PSE (Partido Socialista de Euskadi). ‘Las naciones—y me refiero tanto a las que se identifican con un Estado propio como a las que no disponen de e´l—dejara´n de ver en el Estado su u´nico modelo de realizacio´n. El mantenimiento o la consecucio´n de su soberanı´a y estatalidad dejara´ de ser un objetivo prioritario y obsesivo. . . . Pensamos que, en la nueva Europa que queremos construir, es posible y

260

20.

21.

22. 23. 24.

25.

26.

27.

Multilevel Governance

conveniente concentrar nuestros esfuerzos en una construccio´n nacional descargada de obsesiones estatalistas. Por nuestra parte, estamos dispuestos a desatar el nudo que unı´a indisolublemente la Nacio´n al Estado’ (Ardanza 1993: 37–8). For the independentist orientations of the President of the PNV, Xabier Arzalluz, see El Paı´s, April 3 and April 7, 1994. Interestingly enough, some other declarations by Xabier Arzalluz point in a direction similar to that of Ardanza. See Unzueta 1994: 166. In the Spanish general elections, the percentage of vote for HB in the Basque autonomous community dropped from 17.78 percent (1986) to 16.98 percent (1989), 14.83 percent (1993), and 12.47 percent (1996). In the autonomous elections, HB attained 17.47 percent in 1986, 18.33 percent in 1990, and 16.29 percent in 1994. Data excerpted from the official website of the Basque government. See http://www. euskadi.net. With the exception of the Basque branch of Izquierda Unida, Izquierda Unida-Ezker Batua (IU-EB), all the political forces that joined the Lizarra agreement had nationalist orientations. IU-EB experienced a dramatic setback in the regional and local elections that followed the Lizarra agreement. In the Basque parliament, IU-EB lost four out of its six seats in the 1998 elections. Excerpt from the Declaracio´n de Lizarra, September 12, 1998, see http:// www.iu-ebberdreak.com/docu/lizarra.html See http://www.gara.net/99/eta/gehi2.2k.html Euskal Herriko Udal Ordezkarien Biltzarra. This structure allowed nationalists to surpass the borders that separate all the Basque territories. In addition, this assembly aimed at strengthening the position of nationalists by giving the same weight to small villages (where nationalists are strong, particularly in Guipu´zcoa and Vizcaya) and large capitals (where nationalists are weaker than non-nationalists). In the 1998 autonomous elections EH received 17.91 percent of the vote in the Basque autonomous community, 1.62 percent more than in the 1994 autonomous elections. After the elections of 1999, only one of the four capitals (Bilbao) of the Spanish territories claimed by Basque nationalists remained in nationalist hands. At the time of writing, the mayors of Vitoria-Gasteiz and Pamplona are PP and the mayor of San Sebastian is a socialist. In the general election of March 2000, the PP received a greater share of the vote than the PNV in every Basque provincial capital city and in most major towns. In the provincial capital of Vitoria, the PP received 40.9 percent of the vote, whereas the PNV received just 16.8 percent. El Paı´s, November 28, 1999. The main reason ETA alleged to suspend the truce related to the slowness and lack of determination of the PNV and EA in the fulfillment of their compromises. According to ETA, a new agreement was reached in August 1999 among ETA, the PNV, and EA. By virtue of this agreement, Basque nationalist forces would work to call for a national Basque parliament elected by citizens of all Basque territories. That parliament should elect the Basque president and lead the process toward a

Multilevel Governance

28.

29. 30. 31. 32.

33.

34.

35.

36. 37.

38.

39.

261

sovereign Basque state. However, the PNV argued that neither the PNV nor EA actually did sign that agreement because they considered it unfeasible. See El Paı´s, November 28, 1999. EH dropped from fourteen to seven parliamentarians, and from 17.9 to 10.1 percent of the vote. The coalition formed by PNV and EA attained thirtythree seats, one more than the seats won by PP and PSE. In October 2001 IU-EB joined the PNV-EA government, assuring it thirtyfive of the seventy-five seats of the Basque parliament. Nationalist parties received 57.8 percent of the vote in the 2001 regional elections. On the emergence of Catalan nationalism, see Sole´ Tura 1970. The competing Catalan nationalist party, Esquerra Republicana de Catalunya (ERC), has played a secondary role, receiving less than 10 percent of the vote in regional elections. The ERC has campaigned for separatism, though it has not supported violence. In this critical respect, the Catalan political scene is very different from that of the Basque Country. According to CiU, 28 percent of the money the Spanish state extracts from Catalonia as taxes never return to this autonomous community. See http:// www.convergencia.org/pactefiscal/index.html This tension was aggravated by the PP attacks on the linguistic policies of the Catalan government and on the support that the CiU gave to the Gonza´lez government. The agreements of the PP and nationalist parties involved many other decentralizing reforms. These reforms included the transference of competencies in employment policies and infrastructures, as well as regional access to the Spanish policy-making process in European issues. See Cambio 16, no. 1090, October 1992. See also Antich 1994. A version of this document can be found in El Paı´s, February 11, 1994. Although Jordi Pujol refused to dismiss the document, he acknowledged that its goals were probably too ambitious. Josep Antoni Dura´n Lleida, leader of Unio´ Democra`tica de Catalunya (the party that forms part, with Converge`ncia Democra`tica de Catalunya, of CiU) holds a similar project: a confederation of Catalonia with the Basque Country, Galicia, and the rest of Spain. That view explicitly excludes independence (see Cambio 16, no. 1098, December 1992: 31). In the 1999 Catalan elections the CiU decided not to run on the basis of the Declaration of Barcelona, and, so far, it has not engaged in a single confrontation with the Spanish government in order to bring about fundamental restructuring of the Spanish state. Although the declaration of Barcelona was never suspended by the parties that signed it, it lost public relevance after 1999 due to both the different tracks followed by Basque and Catalan nationalists, and the dependence of the CiU government on PP support in the Catalan parliament during the final years of the Pujol governments. In December 2004, a new declaration was issued by the BNG, CiU, and PNV calling for the establishment of a multinational state. The political impact of this declaration was clearly limited by the

262

Multilevel Governance

fact that the Catalan ERC, now a key player in the Catalan and Spanish political arenas, did not sign it. 40. One may also expect that the exercise of authority that is regarded as illegitimate may intensify identities in an alternative territorial community. 41. Other surveys conducted in 1990 show very similar figures for balanced nested identities. See Ferrando, Lo´pez-Aranguren, and Beltra´n 1994: 16. 42. It must be kept in mind that changes in the strategy of the Basque Nationalist Party were not triggered by growing public demands for self-determination nor, as the CIS 1996 data clearly show, by the development of exclusively Basque identities.

................ 7 ................

The Environment, Socioeconomic Transformation, and Political Change in the New Southern Europe Geoffrey Pridham and Jose´ Magone

The countries of Southern Europe are not usually regarded as among the most environmentally advanced in Western Europe. Images of air pollution in Athens and coastal degradation in other parts of the northern Mediterranean combined with the negative impacts of rapid economic growth in these countries encourage the view that Southern Europe is the least ecologically modern part of the EU. This dubious reputation has been underlined by reference to a ‘Mediterranean syndrome,’ whereby any kind of public good is likely to encounter fundamental obstacles. These include a civic culture sanctioning noncooperative behavior, legislative processes that are fragmented and reactive, and administrative traditions that make policy enforcement difficult or haphazard (La Spina and Sciortino 1993). In the 1990s, there was even concern over whether environmental policy implementation in Southern Europe was markedly different from the rest of Europe (Pridham and Cini 1994), so the impression conveyed was hardly that of a new Southern Europe in the making. Nonetheless, more recent studies have revealed that environmental consciousness in Southern Europe has increased substantially, and in some policy areas these countries have done far better than their northern counterparts. The Europeanization processes induced by the European Commission may have been the major catalyst for this convergence (Boerzel 2001, 2003). The negative reputation of the 1990s was justified by some objective criteria if Southern Europe is compared with Northern European countries like Germany, the Netherlands and those in Scandinavia, where environmental awareness is more advanced and public behavior more

264

The Environment and Political Change

environmentally motivated. But such cross-regional contrasts between South and North in Europe risk downplaying important changes that have recently occurred in these Southern countries in the environmental field. Furthermore, crude North/South comparisons ignore some significant differences between the countries in the Southern European region in this respect. While Greece and Italy may still be laggards in implementing EU environmental directives, Portugal and Spain have moved up to a middle-ranking position among EU countries (Boerzel 2001). In our view, a more fruitful approach here is diachronic withinregion rather than synchronic cross-national comparison, despite the value of the latter. This is because the diachronic approach has more potential in this case for capturing the dynamics of change and its complexity. It will also allow us to draw out useful comparisons and differences among the four countries we are examining—Greece, Italy, Portugal, and Spain. This preference is dictated by the need to accommodate basic dimensions of change that have occurred in Southern Europe over the past generation. First, socioeconomic modernization has been considerable in these countries, its consequences being both political and environmental. They were political because modernization fostered new attitudes and greater openness to outside influences in the international environment; and environmental since economic development brought degradation in its train. That in turn stimulated environmental awareness, which also owed something to international influences. Second, political change in these countries involved the replacement of authoritarian by democratic regimes over three decades ago, offering the prospect of new directions in public policy. For the environment, the kind of basic opening-up that comes with democratization is crucial both in stimulating awareness (where public information is an important precondition) and in requiring public authorities to respond. Thus, we can see different sets of interactions taking the environment as the center of concern. This scenario is somewhat modified by national variation in the timing of socioeconomic modernization and, in Italy’s case only, in regime change. That difference, however, allows some richness of comparison and helps to distinguish between the dynamics of these two forces for change. It is our contention that these countries have, to varying degrees, arrived in the world of ecological modernization although this is a late and recent development. But there is no real sign that, in doing so, they have been able to leapfrog developmentally over the Northern countries through learning essential lessons from the latters’ experience of industrialization and urbanization and avoiding their worst mistakes.

The Environment and Political Change

265

Delayed development has, by and large, been followed by delayed environmentalism. The intention is, therefore, to explore this evolution of the environment as a political concern, and to ask how much this is primarily a direct consequence of socioeconomic transformation and its effects or whether political change, including institutional level as well as policy change, has been decisive in its emergence. Explaining that emergence obviously depends in part on how the environment is conceived as a political issue area in its own right. That is the case when this policy sector is recognized as autonomous and enjoys special governmental structures to support it. This test is significant since environmental policy has historically derived from previous disparate policy concerns. This policy emergence in this area may be stimulated in different ways. It can occur as a result of environmental crisis which both lends a visibility to environmental concerns, and this can open the way, perhaps under public pressure, for a rationalization of institutional structures to favor a more decisive pursuit of environmental quality, for example, the Seveso crisis in Italy of 1976, and the cross-national consequences of the Chernobyl crisis of 1986. Or, the environmental sector may emerge in a more gradual, less dramatic, manner through growing social and political pressures prompted by mounting environmental degradation. This may well happen when there is a perceived threat to economic interests, such as tourism. At the same time, it is difficult to argue that the recognition of the environment as an autonomous policy sector involves a major change in policy priority. Although an environmental policy priority has at times been evident in Southern Europe, it has hardly been continuous or even consistent. To some extent like all policy concerns, interest has fluctuated over time according to circumstance, government party commitment and, often, electoral cycles. Furthermore, while environmental policy is inherently complex and often highly technical in content, it has in some respects a strong potential for political interest due to a link with public health matters. So, much depends on the kind of environmental issue in question. There are also divided strategies that may be employed. An emphasis on the environment per se traditionally suggests a shift in priority away from conflicting concerns (notably, economic); whereas, ecological modernization directs attention, instead, to the marriage of economic and environmental interests. The evolution of environmental policy is, generally, not merely a response to growing environmental degradation or simply the outcome of particular party or ministerial commitments, although those are usually crucial factors. Measuring the dynamics of a new policy area also requires some attention to constraints as well as stimuli. The

266

The Environment and Political Change

former may include problems of political will and opposition, as well as the difficulties that may arise in creating viable bureaucratic space in a public administration that is well established if not always efficiently run. In more abstract terms, institutional legacies leftover from past regimes, combined with bureaucratic inertia, often constrain the dynamics of new policies. Constraints may also derive from the stage of economic development and the extent to which priorities and interests here act as a check on environmental initiatives. In parallel, other forces may help to counter or modify such constraints. In the eyes of decision-makers, the activity of environmental groups, and particularly the degree of public environmental awareness, will be important variables. International pressures may or may not have an impact, though usually a secondary one. The EU is, nevertheless, a special case of external actor, both in general because its member states have to adopt legislation even if they have not been particularly active in promoting it; but also as, on the environment, Brussels came to adopt a more active line from the mid-1980s, this being highlighted in the Single European Act and the Maastricht Treaty. There are, furthermore, broad international tendencies concerning the environment which may influence policy thinking and policy approaches. While retaining many cross-sectoral implications, the environment came to be redefined as a concern in a more holistic manner in the 1970s. And then, as Albert Weale has shown, ‘the persistence and intensification of old pollution problems and the growth of new issues provided the occasion for a new politics of pollution to emerge in the 1980s’ (Weale 1992: 28). This process of ecological modernization has involved a new scale of problems, new patterns of interaction within relevant policy networks and, ultimately, new intellectual and ideological conceptions of policy issues. It almost goes without saying, too, that environmental policy became more ‘internationalized’ in this same period, where the enhanced powers of the EU in the environmental field are arguably the most significant change. In short, we are looking at a changing context, both domestic and international, to this issue area that must reflect on the way it has emerged. Thus, there are different sets of interactions focusing on the environment, involving four major dimensions: political, including regime, change; economic development but also social change; cultural patterns and public awareness; and, international tendencies and pressures. These dimensions appear throughout this analysis although to varying degrees, according to the actual problem under discussion. For instance, international impacts may change somewhat between different successive time contexts. Moreover, while attention naturally

The Environment and Political Change

267

focuses in the early part on regime change, later in the chapter interest turns to how the new democracies have functioned. Environmental degradation is in places treated as a separate factor, but it is one that derives directly from socioeconomic change, with obvious effects on public awareness. In exploring these interactions it is possible to define in particular the relative importance of socioeconomic and political determinants in the development of environmental concerns. In doing so, we adopt a diachronic approach while taking note of comparisons with other parts of Europe where relevant to our discussion. By turning first to socioeconomic transformation we can estimate how much this formed a principal factor in this development. This dimension will subsequently appear when detailing interactions with relevant institutional structures and the political process. The latter then becomes our main point of discussion, all the more as it includes looking at regime change and its effects; while the international dimension receives attention at different points in our discussion, given its importance for both socioeconomic change and policy progress and constraints.

Socioeconomic Transformation and the Environment in Southern Europe There are many characteristics common to the countries in the northern Mediterranean which mark them off from other parts of Western Europe. In particular, the environmental situations of the four Southern European countries that we are examining differ substantially from those of Northern Europe. With their preponderance of unspoiled, undeveloped natural areas, bordering a semi-enclosed sea, possessing rare flora and fauna and suffering from poor water supply, the environmental situation and problems of the South are undoubtedly specific to that region. A warm climate and physical properties of the region mean, for example, that marine degradation has notably affected these countries, though clearly Portugal with its Atlantic seaboard differs in this respect. The Mediterranean has been vulnerable to environmental damage because of its relatively stable seawater, low volume of water in relation to growing coastal population densities and weak tidal regimes (Luciani 1984: 61–4). The total population surrounding the Mediterranean increased rapidly in the postwar generation from about 212 million in 1950 to 356 million by 1985 although low birth rates since the 1980s have slowed down this growth. Around the coast are 537 cities with populations of 10,000 or more, 70 percent of them in EU

268

The Environment and Political Change

member states. Compounding the natural growth of population has been the continued rise of mass tourism, with the Mediterranean having become the goal of a third of the world’s entire tourist market (in 1984 some 52 million foreign tourists visited the coast not counting almost as many local tourists).1 Tourism is highly intensive, being seasonally as well as spatially concentrated, and it produces or accentuates environmental problems—water and air pollution, water shortages, traffic congestion, and the destruction of traditional landscapes (Williams and Shaw 1991). This is in addition to other forms of environmental damage in the Mediterranean, such as industrial and chemical waste, oil pollution, and domestic sewage. Over time, these mounting problems have created some conflicting pressures for public policy in the region—between the urgency of environmental control, which is costly, and the compelling need for economic development. Environmental degradation in these countries is first and foremost an outcome of economic modernization in general, of late, precipitous, and unplanned industrialization and urbanization, and in particular of such developments as the growth of mass tourism in these Mediterranean countries. The overall tendency from the beginning was to subordinate environmental concerns to the dictates of economic growth. Clearly, climate has contributed markedly to this situation in some environmental respects (notably, atmospheric), but as an exacerbating rather than causal factor. All the countries of Southern Europe recorded very high growth rates at different stages in the postwar period. There have been substantial changes in the structure of employment, with a relative decline in agriculture and a shift to industry and the tertiary sector. The expansion of the manufacturing sector has been an outstanding feature of Southern European development (Williams 1984: 8–9). This phenomenon has been commonly viewed as a case of ‘industrialization without development,’ occurring in the absence of adequate infrastructure to accommodate the social effects, above all those arising from urbanization and considerable internal migration from rural areas. The failure to provide housing was perhaps the most visible example of uncoordinated socioeconomic change (Hudson and Lewis 1985: 13). Furthermore, the persisting exploitation of workers in large informal sectors of the economy discourages the integration of a large section of the population into the social system, and the resulting culture of social exclusion and marginality prevents the emergence of environmental awareness (see Task Force on Human Resources 1991 and 1993). The lack of planning combined with the inevitable trend toward a consumer-oriented society has run strongly counter to any easy integration of environmental concerns into public policy and to promoting

The Environment and Political Change

269

ecological awareness. Nevertheless, there has been a gradual tendency for environmental concern to grow in the Southern countries, especially from the 1980s up to the present time. This has occurred somewhat more slowly and later than in some (but not all) Northern countries in Europe, but is still significant. The European Commission Task Force report of 1989 (European Commission 1989) noted that both Italy and Greece had a high rating among EU member states for seeing environmental problems as urgent—a pattern that became even more pronounced over the following few years (as we see in Table 7.1). Certainly, compared with a decade earlier, public pressure had emerged concerning the environment and had begun to impinge on policymaking circles in these Southern countries.2 As one policy adviser in Athens remarked in 1992: ‘the sensitivity of the public has in general increased tremendously; and, now, under pressure of this sensitivity of the public, more or less, every type of government tries to conform with new legislation regarding the protection either of the environment or of health’.3 And, this environmental awareness has continued to grow noticeably in Greece.4 The same has taken place in the other Southern countries, although there are differences of pace and to some extent timing in this development. While this increasing pressure on policymakers is real, some qualifications are in order. Environmental awareness has certainly grown, but environmentalism as such—pointing to some changes in lifestyle— has not developed fast so far. This is due to the lack of the kind of civic culture that relates collective interests with the particular and advances the former. For instance, one study of air pollution in Athens stressed the problem of developing environmental values in the face of the individualism that was a distinctive trait of Greek society and ‘excludes the real development of a collective consciousness’ (Vlassopoulou 1991: 167). The same difficulty was met out in the provinces. An official of the municipality of Rethimno in Crete remarked in 1994: ‘everyone is talking about the environment, and everyone is interested in the environment, but when they have to take action for the protection of the environment many people turn their backs and they go’.5 Similarly, in Spain, most people tend to regard industry and not individuals as the main cause of pollution, creating a pressure for the imposition of fines but not for the alteration of individual conduct (La Spina and Sciortino 1993: 232). Environmental awareness is particularly characterized by an abstractness which does not easily translate into concrete support for policies and their social implications. This gap was not exclusive to Southern countries in Europe, although evidence usually suggests it is greater there. For example, an ISSP survey of 1993 showed that

270

The Environment and Political Change

support for environmental taxes in the abstract was much stronger in Europe than for specific taxes on petrol consumption, but differences between the two levels of support were smallest in Germany and greatest in Italy (Witherspoon 1996: 54). This survey also concluded that there were generally gradations of public concern, ranging from general support for environmental protection through support for environmental policies to personal willingness to make sacrifices for the environment (Witherspoon 1996: 53). Education will obviously in the long run lead to a qualitative change enhancing environmental awareness and the greater diffusion of postmaterialist values. The present transformation of Western societies from a class-based social stratification to a knowledge-based one should further promote the importance of education in Southern Europe. The move from a quantitative approach toward education, focused on creating modern education structures, to a qualitative one is still underway. The introduction of new technologies in primary and secondary schools, on the one hand, and the persistence of cultural, political, and structural illiteracy, on the other, produce conflicts which remain unresolved. In this respect, one can speak of a fragmentation in the educational condition of Southern Europe which inhibits the growth of environmentalism. Alongside these various common problems, there is significant crossnational variation among the Southern European countries. Looking at the four countries in question, there have been differences both in the timing of socioeconomic change and in the incidence of particular environmental problems. Italy’s period of great economic growth was earlier in the postwar period, beginning in the 1950s, while that in Spain and Greece occurred from the 1960s and in Portugal somewhat later. There are also differences of specific environmental concerns. Spain has had a special interest in desertification and soil erosion as well as water supply, while Italy and Greece, for instance, have been rather more concerned about the state of coastal water quality. In Italy, industrial pollution and the nuclear power issue have also been prominent. From this it appears that, despite cross-national variation, there is no clear North/South divide, except for such issues as petrol prices. Cleaner air in cities is, for instance, something that concerns the public more or less equally across the EU. Nevertheless, such cross-national differences have affected the degree and focus of public perception of environmental problems, as well as the extent of environmentalism in each country, as is apparent in the tables which follow. The data presented in Tables 7.1 and 7.2 do, however, reveal some differences concerning the extent or intensity of environmental feeling in general between Southern and Northern Europe, as well as within

The Environment and Political Change

271

each region during the 1980s and 1990s. Table 7.1 provides clear evidence of a substantial increase in concern over the environment both within Southern Europe and in most EU countries between 1988 and 1992. By the end of that period, the percentages of Eurobarometer survey respondents regarding the environment and fighting pollution as urgent problems in Italy and, especially, Greece significantly exceeded the EU average, and the level of concern with these environmental problems in Spain was similar to those found in Northern European countries. Portugal, however, anchored the other end of the continuum, with only Ireland having a lower percentage of respondents regarding environmental problems as urgent. In 2004, the vast majority of respondents in Greece, Portugal, and Spain regarded environmental policy as important as economic and social policies (with 95 percent of Greeks sharing that opinion). Italians, in contrast, were below the EU average in this respect. Table 7.1 Percentages of respondents regarding the environment and fighting pollution as an urgent problem (1988–92) and importance of environment in relation to economic and social policies (2004)

Italy Spain Greece Portugal Belgium Denmark West Germany/Ger. after 1990 France Ireland Luxembourg Netherlands UK Finland Austria Sweden Slovenia Poland Cyprus Estonia Lithuania Hungary Czech Republic Latvia EU 12/EU-25

1988

1992

2004

85 74 82 64 73 82 88 59 63 84 65 67

91 82 97 73 85 87 88 80 70 83 84 82

74

85

79 87 95 85 80 84 81 87 81 83 75 87 84 74 86 94 92 91 91 91 91 89 85 85

Source: Eurobarometer, 37,1992; Special Eurobarometer 217:31.

272

The Environment and Political Change

However, examined from the broader perspective of ‘postmaterialist values’ (of which, concern over the environment is a significant component), we can see that as late as the mid-1980s Southern European culture lagged somewhat behind Northern Europe in terms of basic orientations generally regarded as conducive to support for environmental protection policies. All the countries of the North, except for Ireland, were above the EU average for postmaterialist values, and all four Southern European countries were below average. By the end of the century, however, a substantial narrowing of the gap between Northern and Southern European countries had taken place, especially in Italy. Only in Portugal did materialist values remain predominant, and postmaterialist values increased only slightly. Given the importance of cross-national differences among the Southern European countries, their processes of economic development, environmental degradation, and public perception of environmental issues in each country will therefore now be examined in more detail. Italy embarked on economic modernization a decade or two before the other three countries. As a result, problems of environmental damage and with them environmental concern developed rather earlier than elsewhere in the region. The primary forms of degradation have been industrial contamination, even outside the main industrial concentrations (a fact which reflects Italy’s diffuse development), the decline in coastal water quality, and urban traffic pollution—not to

Table 7.2 Distribution of materialist and postmaterialist values in Southern Europe and other EU countries, 1986–7 and 2000 Percent with materialist values

Italy Greece Spain Portugal Belgium Denmark Germany Netherlands France Ireland UK EU-12 Average

Percent with post-materialist values

1986–7

2000

1986–7

2000

35 46 42 52 38 21 17 17 33 37 22 30

14 19 25 36 23 9 28 12 28 20 NA NA

11 8 12 6 15 18 24 25 14 9 15 15

28 17 17 10 20 16 17 22 18 14 NA NA

Sources: Inglehart 1990: 93; and Inglehart et al. 2004: 383–4.

The Environment and Political Change

273

mention negative effects on the country’s architectural heritage. Materialist consumer concerns have persisted in Italy and the familiar attachment to the automobile testifies to this phenomenon. On its central role in Italian life, it has been noted that ‘the car is a consumer good surrounded by a mythical halo, as an imaginary collective that has a consistent part in influencing the level of demand’ (Lega per l’Ambiente 1990: 30). As Table 7.1 suggests, the ratio of postmaterialists relative to materialists (11:35) is certainly behind that for the EU average, but ahead of other Southern European countries. There are, furthermore, some signs in the Italian case of greater environmental awareness than elsewhere in Southern Europe. Traditionally, the persistence of consumerist sentiment has acted as a check on any qualitative shift to postmaterial values. The media in Italy has been somewhat ambivalent toward environmental issues, with a traditional partiality for the sensationalist. However, since the mid-1980s, there has been more regular press coverage of environmental issues and a new tendency to inform the public on such matters as urban smog, and drinking and coastal water. But television has tended to be much more neglectful of environmental issues than the quality press. The 1992 Ministry report on the state of the environment in Italy reported a survey on RAI during the period 1986–90 which showed that the environment accounted for only 1.7 percent of news items (Ministero dell’Ambiente 1992: 26). By the 1990s, however, a shift in attitudes to more acute concern has become noticeable, although it is regionally variable, being stronger in the north than in the south (Alexander 1991: 105; Financial Times, April 24, 1989). Thus, in this crucial decade, public pressure on decision-makers gradually intensified setting a new trend. In Greece, economic development and consequent pollution have been more concentrated, specifically in the Athens area and a few other cities, following rapid growth in the 1970s. During the period 1960–80 Greece’s growth rate was one of the highest in the world and second only to Japan among OECD countries (OECD 1983: 9–11). The most calamitous problems are oil tanker pollution in the Saronic Gulf and the famous pollution ‘cloud’ in Athens. Since then, there has been progress in cleaning up the Saronic Gulf; but waste management problems have worsened. As the OECD report of 1983 noted, ‘public concern about environmental problems is fairly recent but an increasingly important force’, and was already becoming a central issue in the major cities (OECD 1983: 12). In Greece environmental awareness came rather late relative to the rest of Europe, but when it did it was largely ‘a spontaneous reaction under the pressure of a rapidly deteriorating environment and

274

The Environment and Political Change

the initiatives of many individuals working at various levels within and outside the public administration’ (DocTer 1987: 127). But there is a seeming ambivalence in the attitude of Greeks to the protection of the environment, as the very low proportion of postmaterialists (Table 7.2) and a high level of membership in environmental and antinuclear movements indicate. This is probably due to the fact that the rapid increase in concern over the environment (15 percent) over the years 1988–92, for instance, was not matched by a corresponding growth in environmental commitment on the part of the population as a whole. This obviously relates to the difference between the abstract and the concrete discussed above. There is thus less general public pressure on government, although there is some pattern of localized pressure notably in the Athens case. In Spain, the main forms of damage have been soil erosion, forest fires, and water pollution. Soil erosion has been linked, for instance, to the abandonment of agricultural areas through heavy internal migration with the creation of large urban and industrial centers. As to water pollution, much of this is due to the fact that 66 percent of Spanish industry has been located in coastal zones and, as a consequence, 80 percent industrial emissions are to the sea (Estevan 1991: 186). These changes began to occur from the mid-1960s, with the Franco regime’s adoption of a growth strategy and of economic liberalization together with the start of mass tourism. This provoked, by the early 1970s, a new concern for the effects of environmental damage on local territory and in particular on landscape preservation which has become a pattern since (Morris 1992: 23). But this must be contrasted with the political ambivalence of many Spaniards. Official publications emphasize that Spanish society is ‘greener’ than the rest of the Western world. Indeed, in 1987, 62 percent of all Spaniards thought that the level of environmental protection should be raised, even at the risk of jeopardizing economic growth. But they appear reluctant to translate these views into action; indeed, a Demoscopia survey in the late 1980s identified a growing awareness of environmental issues, but at the same time a reluctance to accept more taxes on cars and petrol (Ministerio de Obras Pu´blicas y Urbanismo 1987; El Paı´s Internacional, October 8, 1990). This tallies with evidence suggesting that materialist values are still far more predominant than postmaterialist ones, as we saw in Table 7.2. In Portugal, the quality of water resources is seen as the top priority of environmental policy.6 The degree of environmental degradation has reached a point where it has become less tolerated by the public, with ´ vora, where seventeen such scandals as in the district hospital of E patients with renal insufficiency died because the water had a high

The Environment and Political Change

275

level of aluminum, alerting opinion to the state of environment quality (The Expresso, June 5, 1993). At the end of the 1990s, about 38 percent of water resources were polluted, with the highest levels of pollution in the most populated regions of Portugal. This is reinforced by a weak network of water supply, which in 2000 covered only 90 percent of the population (as compared with the 94 percent EU average). Only 70 percent of the population is supplied by a sewage system (vs. 85 percent in the rest of the EU); and only 50 percent is supplied by water treatment stations. In the early 1990s, the sea coast was also highly affected by nontreated residuals from private households, industry, harbor activity, residuals from agriculture, and intensive aquaculture. In 2000, one could say that such problems were reduced to specific areas and that the vast majority of sea coast areas had improved the quality of water by the introduction of environmental measures. The quality of air in Portugal is, in comparison, with the other countries of the EU quite good, although there are critical zones where air pollution is highly concentrated in the main industrial areas. In 1990, about 75 percent of all dangerous residuals were ‘eliminated’ with old-fashioned types of treatment, leading to enormous risks of soil and water contamination. In 2000, one could say that a substantial number of old-fashioned dump sites and treatment methods were closed year on year, and the government gained more control over this worrying issue (Ministe´rio das Cidades, Ordenamento do Territo´rio e Ambiente 2002). The problems of soil erosion and forest fires contribute to the picture of serious multifaceted environmental degradation in Portugal.7 The relative economic backwardness of Portugal has resulted in the most consistently materialist and unconcerned population in the EU regarding environmental matters. Portugal has the highest number of materialists in Southern Europe (Table 7.2), the second lowest percentage of population interested in joining or already a member of an environmental/antinuclear movement, and the next to lowest percentage worried about the state of the environment, although this is on the increase (Table 7.1). Perhaps one reason for this is that as a small country Portugal is not a serious polluter by European standards (The Independent, December 6, 1993). Nevertheless, there is an evident potential for political pressure from the public when environmental matters have a direct bearing on public health. In conclusion, our focus in this section on cross-national comparison has generally confirmed that Northern European countries are in several respects more environmentally advanced than Southern ones. But it has also highlighted important differences within both regions, while suggesting that differences between them should in no way be presented as stark contrasts. With further environmentalist progress in

276

The Environment and Political Change

the Southern countries, such differences from the North might begin to dissolve, all the more given cross-national variation within the two parts of Europe. But, meanwhile, the former show many characteristics of modernizing poorer countries: they have acquired more of the potential of richer countries to pollute without yet possessing the technological and material means for relieving pressure on the environment. From our diachronic approach, it seems clear that across the four countries major economic change produced various social pressures that have tended to have detrimental effects on the environment. There emerged a strong propensity for consumerist attitudes typical of fast developing societies, with a new prestige accorded material acquisitions. This occurred first in Italy, with the other countries following when their economic development accelerated. The persistence and worsening of environmental damage subsequently created a growing concern for environmental protection in the Southern European countries, qualified by a traditional reluctance over personal sacrifice to this end. Widespread emergence of postmaterialism comparable to that seen in Northern European countries has so far been strongest in northern Italy and certain specific areas in the other three countries. Overall, however, the materialist/postmaterialist gap between Northern and Southern European countries had narrowed considerably by the beginning of the twenty-first century. Growing environmental degradation and public responses to it have thus provided a fairly consistent but also increasing pressure on decision-makers to adopt policies to protect the environment. But the traditional priority accorded economic growth in these countries has remained a countervailing influence. At this point, our diachronic approach requires us to turn to the political dimension, since the relationship between environmental degradation and policy response is clearly not a straightforward one. Intervening variables, especially political ones, are important to consider. The political and administrative institutions of South European states may fulfill the function of such intervening variables. Large-scale processes of political and social transformation, such as modernization and Europeanization, produce a variety of effects as they are diffused through the different institutional channels found in the four South European societies under study. Furthermore, insofar as historical legacies are a factor, it is necessary to consider the effects of political system change from authoritarianism to liberal democracy. That also highlights another difference between Southern and Northern European countries, since authoritarian regimes persisted in most of the former until the mid-1970s. Regime change has a bearing on the question of timing when exploring the relationship between socioeconomic modernization and political

The Environment and Political Change

277

change. To varying degrees, there was a rough coincidence in the South between these two developments, but there are also noteworthy differences. While in Italy democratization occurred (in the latter half of the 1940s) before economic modernization (1950s and 1960s), in Spain, and to a lesser extent Greece, economic modernization began shortly before democratization occurred from the mid-1970s. In Portugal, however, economic modernization started to develop slowly from the 1960s, shortly before the onset of democratic transition, but the most substantial period of economic growth took place a decade later. This cross-national differentiation allows for subtle handling of diachronic patterns of change and its various dimensions.

Regime Change and the Environment: Does Democracy Make a Difference? It is commonly believed that democratic regimes can boast a better environmental performance than authoritarian regimes. Basic features of liberal democratic systems like the relative openness of decision-making, alternative mechanisms for political participation, and competition as well as public information all tend to facilitate the identification of environmental concerns. Above all, democratic mechanisms for exposing environmental problems—notably, the media and pressure groups—and the accountability of democratic institutions much facilitate the emergence of environmental issues as a matter of public debate. Mobilization on these issues lends a legitimacy to the environmental cause and is likely to create policy pressures (Lafferty and Meadowcroft 1996). Does this, however, suggest that in the absence of democracy, environmental problems will inevitably result? Such comments as ‘pollution flourishes where there is no democracy’ certainly seem to ring true given the revelations of pronounced environmental degradation following the collapse of communist rule in Eastern Europe and the Soviet Union. According to one report in 1990, ‘forty years of centrally-planned over-investment in heavy industry, and of total disregard for environmental costs or economic efficiency, have left a horrifying toll of cancers and respiratory diseases, dying forests, poisoned rivers and lakes and polluting factories’ (The Times, June 18, 1990). However, the relationship between environmentalism and democracy is more complex than may appear at first sight, since it is impossible to view it outside the context of economic development and time period. Under communist rule, a certain dominant policy thinking favored rapid industrial development geared to ambitious production

278

The Environment and Political Change

goals. These overrode environmental considerations, for policy was driven by the belief that nature’s resources were to be exploited for the interests of the people, who were promised a better standard of material living (Elsom 1987). Systemic secrecy, the lack of environmental accountability, and limited environmental awareness undoubtedly served to aggravate the state of pollution under communist rule. A distinction should then be drawn between the regime as such and its policy strategy, since the latter reflects among other things the stage of economic development of a country in question. It is thus hypothetically possible that an authoritarian regime might, in different economic circumstances, take a less intransigent line toward the environment. As it was, the process of democratization certainly had a liberating effect for environmentalism in Eastern Europe, just as in several of the countries in that region environmental movements had provided a central focus for opposition to communist rule (Waller 1989). It also appears from Eastern European experience that the link between environmentalism and democratic transition was largely situational. While environmental concerns were fairly prominent in the initial shift to democracy, this was followed within a year or so by a sharp decline in the salience of the environment as a policy concern, above all because of the overriding attention to economic transformation and its disrupting effects (Waller and Millard 1992). In order, therefore, to assess the impact of regime change on environmental policy in Southern Europe we examine the policy trends under the authoritarian regimes there, and then follow them through time to the present in this and the next section as a way of determining institutional legacies. In this way, it becomes possible to distinguish links between the political, the economic, the cultural, and the environmental. For, in essence, we are concerned here with how far politics matters. As a whole, the right-wing dictatorships in Southern Europe had pronounced economic concerns but scant regard for the environment. In this respect, there is some similarity with Eastern Europe under communist rule, except this environmental neglect was less ideologically driven and less rigidly imposed by a systematic policy strategy. Despite this, environmental legislation of a kind was decreed, though largely as a result of international pressure. The event which had a decisive impact even on these authoritarian governments was the United Nations conference held in Stockholm in 1972. This established among other things the UN’s Environmental Program, which effectively became the formal start of ‘green diplomacy’—the idea of aiming at framework conventions for environmental cooperation. Thus, in the case of Greece, Timagenis and Pavlopoulos see the ‘birth of real environmental law, as in most other countries’, being placed in

The Environment and Political Change

279

the mid-1970s, notably after the 1972 conference (Timagenis and Pavlopoulos 1982: xxxi). As for Spain, the Franco government played an active part at Stockholm emphasizing the importance of natural conservation (Estevan 1991: xix); and references in the Spanish press to environmental themes became more frequent following this event (Ministerio de Obras Pu´blicas 1989a: 295). The reasons for this response were mixed, but an awareness of regime legitimacy problems was certainly present in these last years of the authoritarian regimes. Spanish economic policy had already shifted from an autarchic to an internationalist line to favor growth, a policy change influenced by the desire for closer commercial relations with the then European Economic Community (EEC). It was not, however, so much the latter’s own early interest in environmental affairs, at this time a low priority for it, as the general change in international thinking which prompted the Franco regime’s willingness to show an interest. Spain was certainly not alone in responding to this international environmental agenda. Several international conventions were signed by all three authoritarian regimes, including the World Heritage Convention (Greece, 1971) and the convention for the prevention of marine contamination by spillage from ships (Spain, 1973). The Stockholm Conference also had an important impact on the environmental policy of Caetano’s government in Portugal. A policy concern with the environment was adopted, particularly as the new government (which replaced Salazar’s) wished to create a more positive image of the authoritarian regime at home and abroad. This environmental policy interest formed part of Caetano’s slogan ‘Evolution in Continuity’, an attempt by the regime to reform itself in order to prevent real democratization. Whilst these regimes were responding to the international agenda, they were hardly environmentally motivated as such. Their policy shift was indeed very limited, legislation being a piecemeal response for economic or diplomatic reasons. It was usually technical in the sense of focusing on special environmental problems in a rather disjointed manner. Nonetheless, it was not so untypical of the period, when policy responses in Western Europe relied on more traditional administrative regulation (Weale 1992: 16, 22). For example, the Franco regime created several national parks and passed laws on woodland sites, forest fires, and the regulation of tourism. A new departure was signaled by the 1972 law setting up an interministerial committee (CIMA) to start coordinating environmental policy (DocTer 1987: 341). At the same time, the authoritarian regime in Portugal created a first structure for environmental questions in the National Commission for the Environment (Comissa˜o

280

The Environment and Political Change

Nacional do Ambiente—CNA), as part of the National Junta for Scientific and Technological Investigation (JNICT), in 1971. Later, when more autonomous, the CNA became dedicated to environmental education and studies on the quality of the Portuguese environment. It brought together experts concerned with the environment who were to play an important role in promoting environmental issues in the first phase of Portuguese democracy. The CNA had a very technical and interministerial character with the aim of integrating environmental concerns with other policy areas—for those times quite a forwardlooking idea (Comissa˜o Nacional do Ambiente 1971–4). Such a ‘technical’ approach to environmental problems was evident in the policies of all three authoritarian governments. In Greece too, the military junta passed laws on national parks, the protection of game, and the control of shipping. One may even refer back to Fascist Italy and its laws on such matters as the country’s artistic and archeological heritage and on urban waste (DocTer 1987: 786–7). But in those far-off days such matters were hardly conceived of as components of a potential environmental policy, as was becoming the case from the mid-1970s. This change and its limitations may be illustrated by reference to Greece. The official most concerned with drafting the first effort at environmental planning in the early 1970s commented on the situation then. Acknowledging the influence of UN pressure, as the junta had to respond to the demand for a report on the state of the environment, she noted how difficult it was finding expertise and information: We had a hard time looking around in the universities and in various places to pick up those who could have something to say about the environment. It was really a very first approach to this issue . . . this global approach, it was the first time we had to look around and find data and information, establish a methodology and so on.8

There was no evident hostility from the junta toward information collection for policy advice and even its publication. The chairman of the then Committee on Bathing Beach Pollution, established before the dictatorship, in 1962, admitted they had feared the junta would ban the reports required under strict new legislation of 1965, but in fact this did not happen. He also admitted the junta did not take a great interest in the environment.9 Its priorities were elsewhere, for during this period Greece was undergoing an intensive phase of industrialization. In Athens, building speculators were allowed a free run contributing much to the city becoming an ugly urban sprawl. To summarize, the authoritarian regimes in Southern Europe did not leave any dramatic or considerable environmental legacies comparable

The Environment and Political Change

281

to the disgraceful conditions in which communist rule ended in Eastern Europe. On the other hand, their policies of growth were already showing signs of environmental damage—at least, enough to start mobilizing new ecology groups—and these were to worsen with time. While these regimes had in fact taken the first albeit modest steps on the path of environmental legislation, this was a response less to democratic pressure than to the wave of international attention to the environment in the early 1970s. The right-wing dictatorships in the South were diplomatically and politically isolated at that time, but their countries were becoming economically and culturally more open to European concerns and ideas. Thus, there appears to be no absolute distinction between types of political system and the extent of pollution problems. Environmental degradation and poor policy response have not been confined to communist systems nor indeed to authoritarian regimes in general. Late industrializers—whether democratic or not—are those countries most likely to disregard any imperative to protect and preserve the environment. Political openness is certainly a crucial intervening variable between policy response and economic activity, but some democracies are among the world’s worst polluters. However, political openness accompanied by growing environmental awareness at the mass level in the context of mounting environmental problems and international tendencies favoring action is the most probable combination for producing a determined public policy response favorable to the environment. This leads us to consider the question of time period and the relationship between democratization and the growth of environmental concerns. Looking at the three periods of democratization in European politics—the postwar cases, Southern Europe in the 1970s, and Eastern Europe in the 1990s—there is a notable difference in this respect. Environmentalism has been much more prominent in the third period than the first two. This is not merely because of the particular degradation in Eastern Europe, for international tendencies have meanwhile become much more sensitive to environmental concerns. While the Eastern European transitions occurred, therefore, in a rather different setting from the earlier regime changes, those in Southern Europe happened shortly after an earlier wave of international environmental interest—in the late 1960s to early 1970s—that was short-lived and gave way to the predominant concern with recession by the mid-1970s. On the other hand, the postwar transitions—including Italy—took place well before the environment was conceived as a policy area as such, and before environmentalism appeared. It seems unlikely that environmentalism can be a dominant or continuous priority issue in the politics of transition to democracy, if only

282

The Environment and Political Change

because system-building concerns are predominant. Insofar as uncertainty and even instability usually accompanies early democratization, then placing a special emphasis on environmental concerns might appear a luxury, at least for the time being. At the same time, the emergence of civil society is likely to present the best opportunity for environmentalism to flourish, although by itself this may not be sufficient. Indeed, there was a noticeable change for the advent of political pluralism in Southern Europe that led among other things to mobilization by environmental groups, as will be seen below. This was the first sign that, whereas under authoritarian rule environmental policy formulation had been firmly top-down, it could under democracy be subject to bottom-up pressures. The gradual state of environmental degradation in the Mediterranean, described above, was already getting serious, but—with our countries—it worsened more definitely after rather than before or during regime change. It is clear that at the time of the three Southern European transitions in the 1970s, environmental policy was at an early stage, although other European countries had only recently begun to take the issue seriously too. This casts a somewhat cautious note about the prospects for change in policy approaches in relation to the relative timing of socioeconomic and political change. Was there simply a continuation of environmental degradation and a similarity of policy response (or indeed neglect) following dictatorial rule? In other words, did past environmental policy decisions or policy inertia lock in democratic elites or did the shift to democratization stimulate more serious interest in a policy strategy with the arrival of new democratic elites? How much did environmental management structures change and improve, and if so how much was this the result of domestic democratic pressures? How far did the emergence of civil society after its suppression activate public attitudes on this question? These questions also have longer-term implications for the nature of democratic politics in these new democracies. Our starting point is to consider the various national constitutions as a normative expression of the new democratic procedures and as a reference point for later policy. All four countries in fact included some reference to the environment, even Italy’s postwar constitution. Italy’s Constitution of 1948 stated briefly in Article 9 that the Republic ‘safeguards the natural beauties and the historical and artistic wealth of Italy.’ This bland requirement showed a traditional preoccupation with the country’s strong cultural heritage rather than the environment as such. The national constitutions of the 1970s, on the other hand, reflected the different concerns of a generation later. The Greek Constitution of 1975 provided for environmental protection in general and the protection of workers, public health, urban development, and

The Environment and Political Change

283

physical planning in particular. Article 24 announced that ‘the protection of the natural and cultural environment constitutes a duty of the state,’ which is required to ‘adopt special preventive and repressive measures for the preservation of the environment. . . . ’ The Spanish Constitution of 1978 was less detailed on this question, but it still reflected the mood of the time, colored by international concern over the environment. Article 45 stated: ‘The public authorities shall safeguard a rational use of all the natural resources with a view to protecting and improving the quality of life and preserving and restoring the environment, by relying on essential public cooperation.’ Article 148 provided that the regions would be responsible for environmental protection management. Various articles of the 1976 Portuguese Constitution, revised in 1989, similarly dealt with the environment, although with a more programmatic edge. Article 9e sees the defense of nature, of the environment, and the conservation of the natural resources as a basic task of the state connected with a ‘correct planning of the territory’. Article 66 requires the state to promote territorial planning, taking into account a balanced socioeconomic development and a concern for the biology of landscapes. Following the European Year of Environment, two laws on the environment were adopted by the Portuguese Parliament. The Basic Law of the Environment (No. 11/87) and the Law on Associations for the Defense of the Environment (No. 10/87) regulate Portuguese environmental policy and the mechanisms of decision-making. Article 2 of the Basic Law defines the objective of environmental policy as ‘to optimize and guarantee the continuity of the use of natural resources, quantitatively and qualitatively, its basic principle being that of sustainable development’. These formal commitments did amount to a significant reorientation in public policy thinking, with the principled recognition that the environmental issue is one of importance to government and citizens alike. In looking at its real effects, we turn first to state structures and how much these changed with democratization. Was change simply evolutionary, continuing, or reinforcing the previous pattern of environmental administration, or did it introduce a radical new departure? To what extent was change path dependent? Did it facilitate a more effective and perhaps more strategic approach to environmental problems? Italy is comparatively an awkward case in this respect, as environmental policy began to emerge in the 1970s with the trend elsewhere in Western Europe; although environmental management properly speaking did not really develop until the 1980s. Some preliminary structures were eventually replaced by a new Ministry of the Environment in 1986. This suggested, however, that Italy was responding as

284

The Environment and Political Change

much as these other countries to the same international pressures for action on the environment. In Greece’s case, the state structure has remained highly centralized. Following the 1975 Constitution, several laws and decrees extended and expanded the responsibilities and functions of environmental administration, with the National Council for Physical Planning and Protection of the Environment established in 1976 and the creation of a new Ministry for Physical Planning, Housing and the Environment in 1980 (OECD 1983: 13). In 1985, this Ministry was merged with the Ministry of Public Works. Spain has differed structurally since, as the Constitution dictated, responsibility for the environment was to a large degree decentralized after Franco. The central government retained exclusive responsibility for promulgating basic laws on the protection of the environment, while allowing the regions the right to introduce supplementary regulations. Madrid was also granted certain special tasks relating to forestry and scientific research as well as international relations over the environment. The CIMA set up under the Franco regime was abolished but, as a coordinating body, was not replaced. Environmental administration was assigned in the first instance to the Ministry of Public Works, but, as in Italy and to some extent Greece, several other ministries retained or acquired special functions in this area. It was only much later (1996) that a separate Ministry of the Environment was in fact established. The regions received miscellaneous environmental responsibilities, including health services, managing water resources, and town and country planning (DocTer 1987: 341–2). However, regional environmental administration did not really start before the mid-1980s. The same pattern of administrative reform rather than radical change is evident in Portugal’s case. By the mid-1980s, the state structure for the environment had undergone several modifications. This was due initially to government instability after the 1974 Revolution; but then to priority given to policies promoting economic growth. The CNA, although transferred to the responsibility of different ministries during and after the Revolution, survived until the end of the 1970s. In 1975 it was incorporated as a consultative and coordinating office into the State Secretariat of the Environment (Comissa˜o Nacional do Ambiente 1975). Between 1979 and 1985, the latter formed part of the Ministry for Life Quality. An important step was the creation of the State Secretariat on the Environment and Natural Resources (Secretaria de Estado do Ambiente e Recursos Naturais [SEARN]) within the powerful Ministry for the Planning and the Administration of the Territory. But in 1990, following the Basic Law of the Environment, the SEARN was transformed into a Ministry of the Environment

The Environment and Political Change

285

and Natural Resources (Ministe´rio do Ambiente e dos Recursos Naturais [MARN]). However, its competences are very few, due to the diffusion of environmental competences among different ministries. There is not a single issue on which the Ministry of Environment may act without involving several other ministries and official institutions (see Joanaz de Melo and Pimenta 1992: 17–19).10 As far as environmental policy in South European countries during democratic transition and consolidation is concerned, we see institutional continuity combined with some institutional reform rather than any radical break with previous institutional arrangements. Thus, it could be said that policymaking was neither locked in owing to past policy choices nor subjected to a radical break from the past. These countries henceforth developed different models of environmental management ranging from the centralized to the decentralized, and from partial concentration in one new or reformed ministry to the location of competences in an already established senior ministry. This was not out of line with environmental administration patterns elsewhere in Western Europe, and reflected different national responses to the cross-sectoral aspects of environmental policy. In creating the formal basis for a more active environmental policy, these new democracies also favored procedures for public participation and information. In Spain’s case, however, this was regulated indirectly through different laws—including one passed under Franco in 1958! Democratization certainly allowed the free expression of environmental concerns, which in turn seems to have encouraged the development of environmental movements. Environmental groups started becoming more active in Greece from the mid-1970s (DocTer 1987: 129). A number of particular local issues surfaced under the junta and were taken up by ecological groups, such as the opening of factories at Delfi and Methana and the removal of olive groves for building a factory at Megara. In the last case, a link was in fact made between this local protest and the Polytechnic movement in Athens which proved decisive in its action against the regime late in 1973.11 There has been a mushrooming of environmental groups since, invariably along very sectoral or specialist lines. In Spain, according to one of the early figures of the environmental movement, environmental protest was not well organized under the authoritarian regime in Spain and involved small numbers although it drew some inspiration from roots in the anarchist movement.12 In Catalonia, environmental groups tended to focus on corruption and speculation under the dictatorship rather than on environmental concerns, although contacts were made with international groups.13 As one environmentalist figure in Spain at this time recalled, in 1975

286

The Environment and Political Change

there were only about five or six such associations and then more than a hundred the following year. He also noted the marked expansion of different environmental concerns in press coverage compared with the Franco period, when the only public information was on a few restricted matters like forest fires.14 By the end of the 1980s, the number of environmental groups in Spain had grown to some 700. In Portugal, a few journalists like Afonso Cautela had tried to cover up environmental problems during the authoritarian regime (Cautela 1977). There is evidence of only one important organization, among the 147 in existence, which tried then to alert people to environmental problems early on, namely the League for the Protection of Nature (Liga para a Protec¸a˜o da Natureza [LPN]), but it was very academic in character and concentrated on exposing local issues (Pimenta and Joanaz de Melo 1993: 147–8). On the whole, early environmental action in Portugal involved a small number of committed individuals, who had much greater environmental awareness than the public in general. Only after the Revolution of 1974 did several movements begin to appear. In 1978 there was an important movement against the nuclear power station in Ferrel, involving several associations. Then in 1982, there was protest against the National Energy Plan, which wanted to introduce nuclear energy in Portugal (Lemos 1988: 285–7). In 1981 the Group for Studies on Territory and Environmental Planning (Grupo de Estudos para o Planeamento do Territo´rio e do Ambiente [GEOTA]) was founded. The most prominent founder of this Group was Carlos Pimenta, who was active in mobilizing public opinion for the protection of the environment. At the moment, the GEOTA and LPN are among the three most important associations for the protection of the environment. The QUERCUS (National Association for the Defence of Nature) is recognized as an association which covers all environmental issues and promotes environmental improvement at local and regional level. It tends to use sensationalist actions only as a last resort.15 Most of the 147 associations now registered at the Institute of the Environment (Instituto do Ambiente [IA]) receive financial support from the government through IA (Instituto do Ambiente 2004). They are present in all parts of Portugal, although most are concentrated in the western, more industrialized regions.16 Overall, it is estimated that there are 120,000 environmental association members (about 0.1 percent of the population [da Silva 2002: 162]). The mass media has probably been more active than environmental groups are inclined to be. The new private television channels have tended to adopt a critical line toward the government’s environmental policy, while television gives much coverage to the views of environmentalist associations. In the 1990s, many environmental associations cooper-

The Environment and Political Change

287

ated with other European umbrella associations in order to prevent environmental damage resulting from the construction of a second bridge between Lisbon and the southern part of the country. This was the first time that Portuguese environmental NGOs used the multilevel governance structure of the EU in an attempt to halt this project. This effort was unsuccessful, but it reflected the growing self-confidence of the environmental movement in Portugal (Bukowski 2004). The arrival of democracy in these countries clearly favored environmentalism with its constitutional recognition, improved and more elaborate policy structures and the development virtually ab initio of environmental groups. At the same time, it is necessary not to overrate the extent to which environmental concerns flourished with the transition to democracy. In Spain’s case, the resurgence of civil society that came after Franco’s death was powerful, but it did little to help substantially environmentalist concerns at this time. There were other priorities to do with the process of systemic reform or standard ideological issues. The harsh truth was that international concern with the environment—which had been influential in prodding even authoritarian rulers to confront the issue—had decreased in the face of recession by the time the Southern European transitions actually took place. Combined with the dominance of transition problems, this meant the environment was squeezed out of policy priorities until international and especially EU action on the environment brought it back onto the agenda from the mid-1980s. Meanwhile, problems of environmental degradation continued to worsen.

Environment and Democratic Politics in the New Southern Europe In the view of some green political theory, there should be a normative relationship between environmentalism and democracy. That line of thinking has implications for the nature if not quality of democracy; and hence, for the type of democratic system that emerges from regime change. In other words, while there may be cross-national variation in the specific forms and outcomes of democratic consolidation, so environmentalism may play a role of greater or lesser significance or even no role at all in new democracies. As Paehlke has argued, ‘there is a strong link between environmentalism and enhanced democratic openness and participation’ in the sense that the former challenges established bureaucracies and political and economic interests to respond more effectively (Paehlke

288

The Environment and Political Change

1990: 40). Taking this argument further, Torgerson and Paehlke raise ‘the issue of whether a centralized hierarchy—particularly a bureaucratic organization—could ever be sufficient for, or appropriate to, the task of effective environmental management. For the advent of environmental problems tends to challenge, rather than reinforce, the conventional presuppositions of administration, particularly those favoring centralized hierarchy and a closed process of decision-making’ (Torgerson and Paehlke 1990: 12). There is thus considered to be an ‘ideal’ form of environmental government which is democratic, open, flexible, and decentralized. In this way, environmentalism became seen as conducive to ‘the quality of democracy’ during an early phase of international concern over pollution in the late 1960s and early 1970s, a perspective which continues today. Any such development may well contribute to the process of democratic consolidation by helping to promote viable political pluralism. Democratic consolidation involves firming up norms and structures of a new regime but also broadening its legitimation. Factors promoting the ‘quality of democracy,’ such as environmentalism, may increase support for a new democracy as well as influence the course of democratic politics following consolidation. At the same time, environmentalism may also have disrupting effects on politics given its challenge to conventional norms of economic growth. And new democracies, with their limited resilience, cannot always accommodate too much early basic conflict. Our attention here will, therefore, now be directed to the environmental issue and how it reflects the type of the new political systems in Southern Europe once they were inaugurated. And, we are equally concerned with the functioning of those systems with respect to the environment. But positive environmental performance is not a result merely of the existence of environmental institutions and formal constitutional provisions. More relevant is how these institutions are influenced by general democratic conditions in the context of which they operate (Janicke 1996). Here, international influences may also impinge, but last and by no means least is the state of the economy and the possibilities this creates. Hence, the question of timing of both the political and socioeconomic dimensions reappears now that the special tasks of regime transition have been achieved. Undoubtedly, the possibilities for interaction between them were significantly altered with regime change, for democratization created different and more diffuse mechanisms for policy influence and generally a more open setting in which conflicting demands could be made. This meant that certain aspects of the socioeconomic dimension—specifically, those emanating from society—were

The Environment and Political Change

289

given greater room for direct, also no doubt arbitrary, impact in the political arena. As we noted in the introduction, environmental policy is more complex than simply a response to growing pollution and environmental degradation. Therefore, it is necessary to evaluate key influences on the policy process over time in these new Southern European democracies. These are grouped and defined as follows: The functional: The responsiveness of a new democratic system to environmental problems (assuming they are salient enough) as evidence of system effectiveness. This involves looking at system structures and bureaucratic mechanisms, as affecting the prospects for policy change; The political: This focuses on policy priorities and patterns, with particular reference to the economy, but also how environmental issues are handled, and the role of political actors and linkages, for example, how much environmental groups reflect on the evolution of the ‘movement’ factor in public opinion or general associational activity in political life; The cultural: The relationship between the environment and political– cultural change in newly established democracies, as expressed through public concern. Does this promote democratic values and therefore contribute to system legitimacy?; The international: The relationship between environmental policy and the international dimension; and whether this represents a separate and special influence or has wider bearings on the policy process. These four themes guide the discussion which follows and are examined in turn.

The Functional While the Southern European countries are not particularly unique in their structures of environmental management, they have nevertheless faced special difficulties of administrative competence. They have been rather weak in effective coordination at both horizontal (interministerial) and vertical (center–periphery) levels, the latter being most problematical in devolved or quasifederal systems like Italy and Spain. But it was especially a problem in Athens where national-level environmental management was highly fragmented among a wide range of different ministries (Weale et al. 1996). Ministerial rivalry and bureaucratic conservatism have continued to carry much weight, making it difficult for a new policy area like the environment to acquire

290

The Environment and Political Change

suitable bureaucratic space. This was also true of Italy, for the creation of the Ministry of the Environment in 1986 met with considerable resistance from well-established ministries to transferring to it some of their environment functions. As a result, the new ministry has remained small and has enjoyed only limited power among government structures in Rome. Old practices die hard in public administration even with democratization. This is evident in a lack of openness in bureaucratic habits, such as in Greece and for a time in Spain; and this tends to foster inefficiency. In Italy, too, administrative and procedural problems have been conspicuous. As Bianchi has explained, policy effectiveness has been marred by ‘the many technical and bureaucratic hindrances of the Italian legal and administrative system, especially in the field of environmental protection, where many overlaps of interests and procedures occur’ (Bianchi 1992: 88). Such administrative inefficiency presents less of a problem in Spain, however. But there are also infrastructural deficiencies. Basic problems arise from the lack of qualified human resources, equipment, and financial means to implement a consistent environmental policy. These countries have been rather deficient in the kind of expertise and facilities that make for firmly grounded policy, such as the regularity and reliability of environmental information, efficient monitoring mechanisms, and the incorporation of research (Pridham 1994: 88–90). Modern environmental management and administration requires some radical restructuring or at least introducing effective interministerial coordination. Of the cases in question radical restructuring occurred in Spain, but that was because it formed part of a general drive toward granting substantial regional autonomy. Moreover, Spain has passed various reforms in recent years aimed at improving center– periphery coordination (Pridham 1996: 68–9). In Portugal, however, the division of powers between the central and local governments is still not settled, with implications for management of the environment. Regional coordinating commissions lack autonomy vis-a`-vis the powerful central government, while partisan changes in control of the national government undermine stability in environmental management (Magone 2004a). In Italy, the national legislative process has become more efficient under a law of 1989 aimed at speeding up the incorporation of EU legislation. However, ponderous bureaucratic procedures have not really been reformed despite the introduction of regional government in the early 1970s. Any discussion of the changing functions of the state in recently established democracies should include subnational structures of government. In these countries, this is not simply a matter of regional and

The Environment and Political Change

291

local functions but often also of localistic cultures that may influence policy choice and practice and thus underpin democratic politics. For instance, the statist tradition of policymaking in Spain has tended to encourage distance between rulers and the public; but the new regions together with local authorities have provided a means for countering that problem. The fact that these local councils were closer to the public, both physically but possibly in outlook, was emphasized by a party functionary responsible for environmental matters: Spain is a country little enamored with institutions. Let’s say society is perfect and the institutions are bad. There is a history of little respect towards the institutions, so the messages that come from the institutions are seen as remote. The paradox, then, is that environmental messages that are sent out by the administration, such as laws on water with a strong environmental content, all the legislation of the EU—the citizen sees this as a bit distant and does not internalize it. Now, what he does internalize—and that has had a certain force in the process of democratic consolidation—is what is most directly related to him. In this sense, the town halls play a very important role in environmental management because the results are visible. For example, the collection of rubbish and waste, the matter of water in Madrid, the canalization of rivers etc. The central administration only has a competence of a general kind, as a factor in the quality of life.17

Such patterns of local democracy no doubt contributed to the rooting of these systems even at a time when, as in the mid-1990s, the national government ran into discredit for corruption and environmental scandals (e.g. the Don˜ana National Park) were in the news. At the national levels, the Southern European democracies have generally responded institutionally to the new environmental agenda, but in a gradual and limited way constrained by traditional bureaucratic practices.

The Political Some studies of environmental policy success give a low priority to such factors as the party composition of governments (Janicke 1996: 73–5). This would seem to be supported insofar as economic constraints but also well-established vested interests may leave limited scope for ideological considerations; but this question merits closer examination. In these Southern countries, which are relatively poor compared with the rest of Western Europe and place a high priority on economic modernization, portfolios like public works have usually reflected a predominant concern. That was highlighted in the Greek and Spanish cases where the environment was combined with public works in ministerial structures and clearly came second in priority. However,

292

The Environment and Political Change

individual ministers with a strong commitment to the environment have at times taken policy initiatives; and their tenure in office has left a mark especially if not short lived. At the same time, this has been the reason why these have not led to policy consistency under their successors. Vested economic interests have often reasserted their influence over government policy thinking, as was shown in PASOK’s case in the mid-1980s, following its initial stress on environmental matters when Tritsis was minister (1981–4). The same could be said of Ruffolo’s relatively long and at times proactive tenure of the environment portfolio in Italy (1988–92). Looking more broadly at policy priorities and performance, the superior economic imperative has however tended to persist over time. It has tended to stifle sustained departures in environmental policy. As Ruffolo complained at the end of the 1980s, ‘environmentalist policy is conceived still, to a large degree, as something external, peripheral and sectoral with respect to the production and consumption processes’ (Ministero dell’Ambiente 1989: 37). Similarly, an overriding concern in Spain with high unemployment has presented a major barrier to innovative environmental policy. As admitted in official sources, there is a real problem there of breaking the deeply rooted traditional view that the environment and employment generation are mutually antagonistic (Ministerio de Obras Pu´blicas 1989b: 220). Spanish government leaders have often taken a dismissive line toward the environment; and, if they have explicitly or implicitly linked policy priorities with system consolidation, the economy has been uppermost in their minds. And, the everpresent consideration of environmental costs has remained strong in the minds of policymakers, acting as a disincentive to implementing EU regulations (Pridham 1996: 66, 69). The environment has, therefore, as a whole tended to remain of limited concern to policy-makers, notwithstanding the profusion of environmental measures on the statute books of countries like Italy and Greece. Ad hoc approaches have been much more common than strategically inspired policies, although the latter have developed more in the past few years under EU influence, especially concerning the EU priority to sustainable development. Ruffolo certainly demonstrated a capacity for pushing through new legislation and for mobilizing opinion in its favor. The effort of the first Cavaco Silva government in Portugal to promote environmental concerns is also worth mentioning. The crisis affecting Portuguese industry and agriculture was seen as a chance by Cavaco Silva to relaunch a more qualitative development strategy integrating environmental concerns (Magone 1997). In addition, some regional governments in these countries (notably, EmiliaRomagna in Italy) have sought to integrate environmental concerns

The Environment and Political Change

293

into broader policy planning, just as a limited but increasing number of local authorities have shown initiative in the environmental field (Pridham 1994: 88). This trend represented a significant change in Southern environmental policies, notwithstanding its limitations. For, in the past, the predominant policy style in these countries was traditionally reactive. Environmental events often forced attention to these concerns on governments. The increasing frequency of pollution episodes in Athens usually led to a variety of short-term ‘extraordinary’ measures such as complete bans on private cars on days with critical meteorological conditions. Italian leaders in particular were notorious for regarding environmental policy as largely a response to emergencies. Of these there have been plenty, whether oil tanker accidents or problems of toxic waste. Some events like the Seveso crisis over industrial pollution in 1976 and, of course, Chernobyl in 1986 had a dramatic but also longer-lasting effect on policy concern in Italy. At the same time, a more consistent political pressure from the parties has not been present save in Italy, essentially because of the established presence of the Greens but also because of greater public sensitivity to environmental concerns. The main political parties in these countries have rarely placed much emphasis on environmental policy. In the first national elections in Spain in 1977 the failure of the environmental groups to organize themselves and cooperate weakened the impact of environmentalism on the political scene (La Calle Domı´nguez et al. 1991: 8–9); and this was typical of later developments. And in Greece the parties did not highlight environmental concerns, except for PASOK and the Athens nefos in the 1981 election and momentarily during some later national and local elections. Election programs have invariably highlighted environmental matters when public or international concern has coincided with election periods. Environmental matters have not as a rule been controversial or divisive between the established parties in these countries. This is perhaps not surprising in countries where postmaterialist values are still not yet pronounced. In Spain, it has been said, ‘politicians are convinced green measures do not win votes’ (The Economist, April 27, 1991). The absence of party differences may also be due simply to a lack of intensive interest in such matters. In Portugal, environmental issues are not usually controversial among the parties, both main laws on the environment being adopted without any debate in 1987. In Italy the situation only began to shift in the 1980s with the arrival of the Greens in the Parliament from 1987. Such an event is always a sure pointer to a likely response from established parties, which in Italy have usually been allied to economic interests. However, the disruption

294

The Environment and Political Change

in Italian politics in the 1990s, producing among other things a rather different party system, has interfered with this development. In Greece, on the other hand, the Greens have continued to be very weak. The political impact of environmental groups has turned out to be limited in the sense of access to decision-makers though with some cross-national variation. In Spain, for instance, there persisted for long a statist, closed, and highly bureaucratic policy style that did not welcome institutionalized channels for consultation with interests involved (Aguilar 1991: 1, 7); but that began to change in the 1990s once European policies began to take effect. In Portugal, the environmental associations have gained in reputation throughout the 1990s and have become an important ally of the department of the environment. These are mainly dedicated to environmental education at the local level, and in this are supported by the central government. In 1991, they increased their organizational capacity by founding the umbrella organization Portuguese Confederation of Environmental Associations (Confederac¸a˜o Portuguesa de Associaco˜es de Ambiente—CPADA), which comprises 110 associations from all over the country. They meet annually and exchange views on how to enhance their performance (CPADA 2004). In Greece, environmental groups are quite numerous, but their lack of organized party–political links has inhibited their input into the policymaking process (Pridham, Verney, and Konstadakopoulos 1995: 261). There is, thus, much evidence to support the view that environmental groups in Southern Europe have tended to be rather less aggressive than their counterparts in Northern Europe (The European, March 26, 1992). This has, however, not meant they have remained passive at the societal level, for some of these organizations in Italy and also Greece have been prominent in pushing and publicizing their environmental demands, in conflict with the government of the day. The larger ones have now begun to benefit from their international links, in particular from support in Brussels for their new role as a channel for information on environmental policy deficiencies in member countries. The public airing of the rate of implementation of environmental directives such as those on air pollution, for example, has had a certain galvanizing effect, this being an environmental problem that directly affects public health.

The Cultural We have so far emphasized the predominant policy concern with economic modernization, this allowing limited space to environmental matters. But what about the national publics which may only in part

The Environment and Political Change

295

be influenced by the judgements of politicians? The data presented in Table 7.1 reflect a growth in environmentalist attitudes, but this has not been clearly or consistently manifested in changes in behavior over environmental problems. Certainly, a flowering of consumerism in the past few decades has held environmentalism in check, while a traditional emphasis on particularistic over collective interests (such as environmental quality) suggests that environmentalism was unlikely to cut deep. For example, introducing environmental impact assessment (EIA) in Italy encountered problems not only of bureaucratic inefficiency but also resistance elsewhere in the country. Originally a procedure adopted from the United States, EIA was often seen in Italy as a phenomenon foreign to Italian legal culture (Pridham 1996: 61–2). In Greece, traditional suspicion toward the state in general appears to have acted against ready compliance with environmental regulations as indeed with those in other policy fields (Pridham 1996: 66–7). All the same, there are other indications that some change has been occurring but only very gradually and essentially from the mid-1990s onwards. The activities of environmental groups and greater media attention are undoubtedly having a diffuse influence on public opinion. At the same time, there is an evident gap between national policymakers and the public on this and indeed other issues. This may have been somewhat reduced by more detailed and seemingly more reliable official information on national environmental situations not least under pressure from the EU. This is much less so in Greece than the other Southern countries. Quite often local interests have influenced action, including with respect to tourism. While economic concerns have still predominated, a new link with interests like the tourist industry has opened up the possibility for sectoral appreciation of growing environmental expectations in the international and domestic market. One such case was the island of Mallorca which introduced new quality regulations and controls on traffic for strengthening the appeal to upmarket tourists (Pridham 1994: 88). In the city of Athens—from the time Tritsis became mayor in 1990—there have been more determined efforts to deal with the pollution cloud, not least because of growing public pressure in the city, while events like planning for the Olympics there had some effect too. If there has developed a growing public or ‘cultural’ sensitivity to the environment, it has often included a strong territorial focus. Catalonia, a Spanish region known for a sense of territorial identity, has a large urban population, high level of education, and awareness of nature, which combined tend to make conservation an important issue, for example, the debates that surfaced over the despoliation of the Costa

296

The Environment and Political Change

Brava by mass tourism (Morris 1992: 25). Such a strong localistic focus has long roots in these countries. Local press coverage is often more intense than the national, especially in Spain and Italy. Localism in Southern Europe was, for instance, marked in the campaigns conducted for the European Year of the Environment which mobilized local pride in the early 1990s (Pridham 1994: 96). In Portugal at the time of the Rio international conference of 1992, the government organized a series of public debates on its environmental proposals. At the same time, it declared the environment as a public policy priority for the next decade.18 According to a Gallup survey carried out in 1992 the Portuguese see the environment as among the most serious problems of their society after hunger/housing, crime/violence, health, and living costs (Schmidt 1992a, 1992b). In 1998, a further governmentsponsored study conducted by the prestigious Institute of Labor and Entrepreneurial Sciences found out that environmental issues are among the two most serious problems for 13.8 percent of the population. Overall the environment ranked in fifth place in terms of issues that concerned (Silva 2002: 164). Public policy has begun to respond in different ways to changing attitudes at the public level, notably with regard to environmental information. Environmental education in schools has grown, together with—as in Italy—vocational training in environmental matters (Ministero dell’Ambiente 1989). It is not absolutely clear whether these developments reflect deeper changes in political culture, but they point toward a more definite trend of ecological modernization in the future even if in a diffuse way as shown up to now.

The International The transnational nature of many environmental problems—accentuated by the close proximity of so many different European countries and their often dense populations—means it is hardly possible to discuss the link between environmentalism and democracy without reference to a variety of external influences. As we have seen, international influences may either be diffuse, emanating from the external context in different and sometimes not very tangible ways; or they have a direct and more concrete impact arising from common membership of international organizations. It is the latter we now discuss. Given common environmental problems facing these Mediterranean countries, there is a certain logic to their international cooperation. They participate in several organizations and are signatories to a number of conventions on the Mediterranean Sea from the Barcelona

The Environment and Political Change

297

Convention of 1976 (concerning marine pollution) onwards. Important has been the Mediterranean Action Plan (MAP), resulting from this convention, and administered by the UN Environmental Program. Other actions have proliferated under the MAP, including priority action programs to promote environmental management practices, and so encourage more permanent change in policy approaches with less emphasis on crisis measures (see Pridham, Cini, and Porter 1995). These international involvements influence policy-thinking and thus help to condition the outlook of decision-makers. But, it has been the EU that has had the most incisive effect on the environmental policies of these Southern European countries. The EU has acquired increasing weight in the environmental field in forcing national governments to respond to its policy initiatives; and, as a consequence one may speak of Europeanization effects also in the Southern European member states (Eder and Kousis 2001b: 401). EU membership has made a substantial difference in the evolution of these countries’ environmental policies. It has advanced these both by contributing the bulk of legislation, but also by urging new policy approaches, with the emphasis in the Fifth and Sixth Environmental Action Program on sustainable development. Even for Italy, which had passed some rudimentary laws in the decade before the mid-1970s, Brussels has been crucial as a determinant of policy. As one study concluded, ‘what distinguishes Italian legislation from others’ legislation, above all in Northern and Central Europe, is the absolute prevalence of rules which owe their inspiration to the EU rather than the national level’ (Capria 1991: 1). Greece was only really beginning to develop an environmental policy exactly when joining the EC in 1981; and, so, there it is possible to speak of the EU having some formative effect on that country’s approach to environmental matters. In Spain’s case, this dependence on Brussels has perhaps been the most marked of these Southern European countries. A high official in the Spanish Ministry of Public Works (environmental policy department) noted that his country had accepted all past EU legislation on the environment as one price for membership. He referred also to the benefits this promised of underscoring democratization: The only thing we did without considering any economic costs was to enter the Common Market. As to our entry, we approved en bloc all the Community directives on the environment without any time or other reservation. Portugal made reservations, but not us; we approved everything. And we said to ourselves that we would not study the economic cost: it was necessary to approve it and then we will see. And that caused us much effort. . . . Now it is a matter of realizing the economic cost of measures. At times, it is very difficult; at others, easy or relatively easy. Afterwards, there are benefits also.

298

The Environment and Political Change

Q: Why this decision to accept all the legislation? A: Because politically it was important to enter the Common Market. All Spain wanted to enter the Common Market, and our only possibility of stability and political change was the Common Market.19

This adoption of the whole acquis communautaire of EU legislation during the two years after entry caused some administrative blockage for a time. This was hardly surprising since, as admitted above, the decision was taken for high political reasons without attention to the consequences. Involvement in European environmental policy has had various beneficial effects, both legislative and other. It has forced some rationalization of environmental management, through ministerial reorganization. For example, as noted in a cross-national study of the impact of EU air pollution legislation, ‘it is clear that the directives are bringing about major changes to air pollution control practices in Greece, Spain, Italy and Portugal; the changes amount effectively to a comprehensive re-ordering of existing policies, procedures and administrative arrangements’ (Bennett 1991). EU membership has also nudged the Spanish administration, for instance, to open up to consultation with the private sector (Aguilar 1993: 8–9; Aguilar 2001: 258–60) and broaden the sectoral conferences with the autonomous regions. A strong Europeanization of Spanish environmental policy has unfolded (Boerzel 2003). In the case of Portugal, the public has been involved in policy discussion, particularly over the EIA of big local projects, following a belated application of the EU directive on this matter.20 Overall, Brussels has acted as a consistent pressure for firmer policy commitment to environmental matters. The stick of legislation has, however, combined with the carrot of special EU resources for environmental improvement, such as the Cohesion Fund which included funds for environmental infrastructure. Furthermore, since Iberian entry in 1986, the EU has begun to focus specifically on environmental problemsolving in the Mediterranean. There have followed a series of special projects, including the Mediterranean Strategy and Action Plan (MEDSPA) and the Environmental Regional Program (ENVIREG) (Pridham, Cini, and Porter 1995: 53). In this way, the EU has come to recognize that there are certain ‘Southern’ socioeconomic characteristics which require special treatment. This has not prevented European legislation on the environment highlighting tensions in the South arising from EU membership. These countries have felt under increasing pressure from Brussels in recent years over keeping pace with stricter environmental standards, and are particularly concerned that such measures as a carbon tax

The Environment and Political Change

299

would harm their economic development. Spain was at the forefront of opposing this proposal, although some Northern states (such as the UK) and European industry have not regarded it favorably. Perceptions have also been colored by the general feeling, occasionally resentment, that much EU legislation in this area is ‘Northern’ in inspiration, which is true to a large extent (Redclift 2001). The Southern European member states had a poor reputation for implementing environmental legislation in the 1980s and mid-1990s, but they have made significant progress since then. This is justified in certain respects, although it does get exaggerated. Generally speaking, there have been problems in carrying out European legislation in most member states. The question is the degree to which this has happened. Problems have arisen over incorporating European legislation; however, there has been a marked improvement in Italy and Greece, as shown in Table 7.3. Most of all, it has been practical implementation which has illustrated the inadequacies of Southern member states. The reasons have varied cross-nationally, but they relate very much to the functional, political, and cultural factors discussed above (Pridham 1996). Implementation of EU directives has been difficult for all member states, as the data in Table 7.4 reveal. Italy and Greece are among the laggards in this regard. It is noteworthy, however, that Spain and Portugal are near the middle of the range of EU member states. Table 7.3 Percentage of environmental directives translated into national law, by country and year

Italy Greece Portugal Spain Belgium Denmark Germany France Ireland Luxembourg Netherlands UK Austria Finland Sweden

1990

1991

1992

1994

1995

1997

63 79 95 92 86 99 92 92 87 89 97 91

59 76 94 93 81 98 92 89 84 86 95 85

83 92 89 91 94 99 92 96 90 92 97 93

56 85 82 86 85 100 91 94 97 93 98 82

85 88 87 90 83 98 94 95 95 92 98 93

97 97 97 99 87 100 94 96 98 98 99 96 97 96 97

Sources: European Commission 1993, 1995a, 1995b, and 1998.

Table 7.4 Noncommunication cases, nonconformity cases, bad application of horizontal cases,* 2000–1 and 2003 (in percentages) 2000–1

Italy Greece Portugal Spain Belgium Denmark Germany France Ireland Luxembourg Netherlands UK Austria Finland Sweden Total absolute numbers

2003

Noncommunication

Nonconformity

Bad application horizontal

Noncommunication

Nonconformity

Bad application horizontal

6.3 8.7 5.5 8.7 6.3 4.8 8.7 9.5 6.3 6.3 4 15.1 3.2 4 2.4 126

9.3 1.2 3.5 2.3 8.1 3.5 8.1 8.1 5.8 1.2 4.7 7 22.1 8.1 7 86

7 9.3 9.3 8.1 11.2 3.5 7 10.5 5.6 5.6 3.5 8.1 5.6 2.3 5.6 89

8 10.2 3.4 8 9.1 1.2 10.2 12.5 5.7 5.7 8 5.7 4.5 4.5 3.4 88

13.6 3.4 2.5 8.5 5.9 3.4 6.8 14.4 6.8 5.9 6.8 7.6 6.8 4.2 3.4 118

10.5 13.7 5.3 8.4 5.3 2.1 3.2 10.5 16.8 6.3 5.3 7.4 3.2 1.1 1.1 95

Source: European Commission 2002: 117–19, and 2003: 38–40. *Explanation of terms: ‘Non-Communication’ means that the member states fail to adopt the measures (national laws, regulations and administrative provisions) to transpose directives and to communicate them to the Commission within the prescribed time limit; ‘Non-Conformity’ means that the member states’ transposition measures do not conform with requirements of directive; and ‘Horizontal Bad Application’ means that the member states fail to implement certain derived or secondary obligations contained in the Community acts, such as setting out plans, classifying sites and designating areas, adopting programmes, monitoring data, reporting etc.

The Environment and Political Change

301

Thus, comparisons with the rest of Western Europe should be made in a differentiated manner. Viewed cross-nationally, Southern Europe is rather behind many Northern European countries, not to mention the United States, in policy designed to protect the environment. But it is important not to underrate differences of state tradition, economic development, strength of civil society and of cultural values between North and South, and not to ignore differences in the longevity of democratic government. And, last but not least, differences between the Southern European countries themselves should not be forgotten for they are significant (Eder and Kousis 2001b: 396), just as differences between North and South should not be exaggerated.

Conclusion If there is any grand lesson from this study, it is that the environment as an issue area has a distinct autonomy of its own insofar as it is a direct consequence of secular change. It owes much also to international trends, which may even have an effect on relatively isolated authoritarian governments. In fact, their very isolation and vulnerability may make them more sensitive to international trends. In Europe, transnational policy pressures have been institutionalized through the EU, which itself tends to be driven on many environmental issues by the ecologically modern Northern member states, such as Denmark and Germany. The twin influences of secular change and common international trends are highlighted by the case of Italy, whose political regime change had occurred a generation earlier than in the other countries. The context of Italy’s postwar democratization from the later 1940s was entirely different. We have also seen that there can be no absolute distinction between authoritarian regimes and liberal democracies, for the timing of economic development and international pressures counts. Nevertheless, democratization does open up new possibilities for policy response to environmental concerns, given the right secular and international conditions—as the Iberian and Greek cases show. Such possibilities may be slow in developing not least because the very preoccupations of regime change provide serious constraints on new departures for a given period, although this cannot exclude the impact of new environmental concerns and of international influences in the future. The gradual democratization of these countries has contributed to enhancing environmentalism through political openness, allowing the development of environmental associations and the diffusion of environmental information and education. At the same time, the new

302

The Environment and Political Change

democracies reorganized and extended state administration on the environment including the introduction of local structures, which in some of these countries have developed into an important area of policy action. To put it in terms of modern institutionalist theory, democratization had multiple institutional effects both at the level of central government and local institutions. However, it is difficult to argue that environmentalism has in turn contributed equally in these countries to democratization, although its relevance to the development of political pluralism is obvious. This is because the environment has not been a dominant concern; and in the country where it has advanced most (Italy), democracy had already been established several decades before. The prior attention to economic problems and growth in the three new democracies of the South has tended to raise conflicts with environmental interests. At the same time, international pressures and social change have provided counterinfluences to this tendency. The superiority of economic concerns, understandable in these countries of Europe’s semi-periphery with their desire to stimulate growth and overcome unemployment, had a more obvious and perhaps more direct link with political regime consolidation. However, the rising challenge to these conventional beliefs from environmentalist and international changes and demands did not apparently distract from the achievement of democratic consolidation. Conceivably, timing is relevant here since this change began to occur once consolidation was well underway. Italy is, as usual, difficult to incorporate, since her democratic consolidation—commonly seen as completed belatedly by the 1970s—occurred before the impact of environmentalism. In short, our study demonstrates that political democratization is a vital if not central ingredient in the growth of environmentalism, but not exclusive of our parallel changes at the socioeconomic and international levels. Of the four Southern European countries, Italy has been the most advanced in growth of environmental awareness, and the activity of ‘green’ organizations has, if anything, had a stimulating effect there on patterns of democratic participation. The other countries are, historically, relatively younger democracies. Here, the notion of ecological modernization is not a matter of widespread conviction, but there are limited and separate signs, including at the local level, as to a change in this respect especially from the mid-1990s. These may provide a pointer to political patterns of the future. Significantly, the link between environmental concern and economic interest has now started to manifest itself across the Southern European region, especially with respect to degradation and its effects on the important tourist industry in these countries (Pridham 2001). Such recognition of the problem

The Environment and Political Change

303

enhances the desire to rectify it and, in turn, that may be transformed into policy. Thus, in judging the significance of the environmental issue area, much depends on the frame of reference. If that is cross-nationally comparative, then the countries of Southern Europe are still easily found wanting in their response to the ‘new politics of pollution,’ although as shown this should not be written up large as a North vs. South dichotomy. There are, furthermore, differences between the Southern European cases worth noting, in not only their degree of environmental awareness but also in their environmental administration, ability to carry out EU directives, and specific environmental problems. If, on the other hand, the comparison is diachronic, looking at each national case, then there has indeed been significant change although gradual and so far limited. Overall, therefore, focusing on the relationship between the environment, socioeconomic transformation, and political change has been instructive in assessing the complexities that lie behind the emergence of a relatively new policy area in Southern Europe and in reflecting on the types of democratic systems now established there. Last, but not least, since 1985 all four Southern European countries have been under considerable pressure to adjust to the Europeanization process, and this has had substantial impact on environmental policies. The importance to Southern Europeans of forming part of this community of democratic nations, engaged in a process of collectively improving the quality of life of their populations, has had a positive impact on policies adopted and implemented, and may have even contributed to the convergence of the values of Southern European populations with the rest of Europe in terms of postmaterialism (Magone 2004).

Notes 1. The Economist, December 21, 1991. Also see World Bank 1990. 2. Vlassopoulou 1991: 31; and interviews with Giovanni Zapponi, Istituto Superiore di Sanita`, Ministry of Health, Rome (May 19, 1992), and with Iannis Papadakis, Ministry of Health, Athens (December 1992). 3. Interview with Iannis Papadakis. 4. Interview with Georgia Valaoras, head of WWF office, Athens. 5. Interview with Vangelis Archontakis, Municipality of Rethimno, Crete, October 1994. 6. Speech of the Minister of the Environment on the Day of the Environment, broadcast on RTP Channel 1 television, June 5, 1993.

304

The Environment and Political Change

7. Pimenta and Joanaz de Melo 1993: 77–8; also see Borrego et al. 1991. 8. Interview with Dimitra Katohianou, KEPE, Athens (November 1992). 9. Interview with Iannis Papadakis, Ministry of Health, Athens (December 1992). 10. Beginning in the second half of the 1990s, environmental matters have been assigned to the Ministry of the Planning of the Territory. 11. Interviews with Thanos Veremis, University of Athens, and Leonidas Louloudis, Agricultural University of Athens (November 1992). 12. Interview with Joaquı´n Araujo, journalist, Madrid (October 1992). 13. Interview with Jordi Bigas, editor, Integral, Barcelona (October 1992). 14. Interview with Joaquı´n Araujo, journalist, Madrid (October 1992). 15. Interviews with Conceic¸a˜o Martins, GEOTA, Lisbon (April 1993), and with activists of QUERCUS (April 1993). 16. Interview with Fernando Rebeiro, National Institute of the Environment, Lisbon (June 1993). 17. Interview with Angel del Castillo, PSOE, Madrid (October 1992). 18. Ana´lise Tema´tica 1993; Sı´ntese Estrate´gica 1993; and interview with Pedro Liberato, Vice-President of INAMB (June 1993). 19. Interview with Joaquı´n Jos Vicent, Ministerio de Obras Pu´blicas, Madrid (October 1992). 20. Nogueira and Pinho 1988; Nogueira 1993; and interview with Pedro Liberato (June 1993).

................ 8 ................

Democratic and Economic Consolidation in Southern Europe Stavros B. Thomadakis

Over the last third of the twentieth century the landscape of Mediterranean Europe has changed radically. Previously, the region of Mediterranean Europe embraced the three most traditional societies and backward economies of Western Europe: Spain, Greece, and Portugal. These three societies and economies have undergone significant transformation. Their transformation is evident to any interested observer but is not a simple one, nor is it to be explained by a single argument or a solitary factor. It has been a process that involves politics, economics, international events, and global changes. The central question which informs this essay is about the role of politics and whether ‘politics matters’ in a process of economic and structural change. More specifically, the impact of democratization on fiscal policy and structure informs the central question. This question has at least two significant facets. The first inquires about the role of politics, as opposed to long-acting socioeconomic factors, in shaping paths of development and structural change. The second focuses on the influence of party politics and chosen policies upon processes of change. In this essay an attempt is made to capture both aspects of the role of politics in a specific historical period. In terms of the theoretical approach of the introductory chapter of this volume, we could say that both questions address aspects of large-scale transformations, such as modernization, democratization, and Europeanization, as well as aspects of past policy choices and institutional legacies. We can generally think of a sequence of questions that underlie the approach of the paper. Is the South European transformation the outcome of predetermination by earlier long-term factors, or is it shaped by political action? Does the choice of policies have a lasting

306

Democratic and Economic Consolidation

effect? Is transformation shaped by domestic or international conditions? Does the balance of factors change as circumstances change and as the democratic regime matures? Democratization is a central and common historical point in the process which the three societies and economies have undergone in the middle 1970s. With the hindsight of three decades, it can be argued that transformation without democratization would have been impossible. Even under the fairest economic conditions, transformation always involves choices for the allocation of scarce resources, strategies that eliminate or modify existing privileges, conflicts of interests and conflicts of visions. It therefore requires mechanisms that can reproduce and maintain a twin condition: legitimacy and stability of the sociopolitical arrangement. Democracies in Southern Europe have managed to produce this twin condition over the last quarter of the twentieth century. They have maintained their legitimacy and have gained both political and economic stability. Focusing on the last third of the twentieth century requires that we have a basic understanding of starting conditions. The three South European countries shared two important features that were significant to their subsequent economic paths. The first common feature is important because it serves to illuminate existing systems: The state as an economic entity was quite small in all three countries. Total government outlays hardly exceeded a fifth of GDP of the three economies in the middle 1960s, as compared with a European average of over onethird of GDP at the same time. As systems of collection and distribution of resources, therefore, the southern states were small and weak, compared to their other European neighbors and future partners. The small economic size of the state in southern societies seems to contrast to the notion of the strong authoritarian state that many consider a standard historical feature of the three South European countries. This contrast requires comment. An authoritarian state may exercise economic intervention without being itself economically large, that is, without directly collecting or spending monies. It can allocate resources by fiat, fix prices by decree, and can license some activities and forbid others. Indeed, in all three countries the state wielded an extensive web of administrative powers for the regulation of economic activity. This made compatible the coexistence of authoritarianism with economically small public sectors. These forms of intervention however became harder to maintain in a context of rapid growth and new economic opportunities. Here it is relevant to recall the second common feature of the three economies: all three were star performers in

Democratic and Economic Consolidation

307

terms of economic growth in the 1960s. For over a decade the three economies exhibited average growth rates near or exceeding 7 percent annually, with Europe at large growing at an average rate of about 4.5 percent annually. Thus, there was a powerful tendency for convergence but initial levels were of course still very disparate, and actual convergence was distant even at those differential growth rates. Nevertheless, high economic growth generated changes and new pressures within the southern societies. These had to be absorbed by existing systems of allocation of resources, benefits, and investments, and the ability of existing systems was therefore put to a severe test. The accumulation of private capital, the pressure of market choices, the opening of the economy to international competition were factors that inevitably pushed for the elimination of administrative interventions. In this process, these three southern states were presented with a new historical choice: on one hand, they could allow the private sector and the forces of the market to fill new spaces and needs created by growth, but maintain the public sector within its original small economic limits; on the other, they could opt for an economically active public sector which would base its activist role on the growth of its perimeter, through an expanding role in the collection and allocation of resources, rather than through antiquated regulation. The three South European societies followed the second route. It is argued in this chapter that democratization and European integration were powerful factors that determined the result. The comparative approach is central to this essay and inevitably acquires a double dimension: a comparison of the three countries with each other, and a comparison of the three countries as a group with the ‘European average’. Two points must be borne in mind about these comparisons. The first is that comparisons of the three countries with each other reveal important differences as well as common elements that always inform particular histories. There is no particular predilection to emphasize similarities beyond the obvious initial backwardness of the three societies. In fact, differences among the three are important because they reveal both the role of national politics and the limits and challenges of convergence to a common model. The second point is that the ‘European average’ is not a very clear standard of comparison. In this chapter the European average is represented by EU means and this clearly includes countries with very different social models, politics, and economic structures over the last quarter century. They represent a meaningful standard only insofar as they are all democratic and developed economies, belonging to an economic

308

Democratic and Economic Consolidation

alliance. We therefore assume that convergence of the southern democracies to the ‘European average’ reveals an adjustment that makes them compatible and economically viable within the open space defined by the economies of their partners. This does not necessarily endorse the ‘European average’ as a meaningful or successful socioeconomic model. In this paper, the focus is placed on fiscal conditions and structures. These represent a specific but central part of the transformation that has absorbed the direct impact of both political and economic processes. The significance of the fiscal sphere for interpretation and analysis begets a special weight because of the importance that attaches to the public sector. Indeed, one of the major changes which is apparent over the last thirty-five years in the three South European countries is in the size, the role, and the structure of the public sector. The fiscal sphere condenses the interplay between economic forces and political choices. Its study can therefore be very revealing about the effectiveness of policies in pushing structural change. Even more to the point, the fiscal sphere is the epicenter of the social mechanism that can produce the twin condition of democratic support and stability. The success or failure of fiscal management can regulate the maintenance or elimination of the twin condition. The process of transformation that the three South European societies underwent in the last thirty-five years of the twentieth century passed through distinct phases. The phase of democratization formed a transitional period that involved primarily the role of domestic factors and actors in each country. In terms of the approach employed in the introductory chapter of this volume, we could say that domestic institutions, leftover after the demise of past political regimes, were influential in the aforementioned transformation. Such institutions, and related norms and legacies, constrained fiscal policymaking. The phase of consolidation of democracy and of the European orientation of the three societies was characterized by the emergence of the importance of international factors, actors, and constraints. Finally, the phase of European convergence, which determined the definite path of the three southern democracies toward participation in the European Monetary Union brought fully to the forefront the forces and institutional pressures of the EU on its three least developed members. The shift of importance from domestic to international factors as we pass from each phase to the other is not only an important point from a methodological respect. It is also a substantial change in the makeup of political and economic forces that determined social outputs in the erstwhile backward and relatively isolated countries of Southern Europe.

Democratic and Economic Consolidation

309

The Fiscal Legacy of Democratization in Southern Europe: Public Spending The examination of the causes and the consequences of democratization in the three South European states can be subsumed under four questions. The first is whether the collapse of the dictatorships was the outcome of long-term forces arising from prior socioeconomic development or short-term forces related to economic crisis. Theory and evidence from other case studies suggest the possibility of both types of effect. The second is whether democratization had an immediate and visible fiscal effect. Democratization is almost synonymous with a demand for political rights. It is reasonable to hypothesize that it also engenders expectations for the provision of economic and social goods. This can lead to considerable expansion of fiscal expenditure, and must be tested in the case of South European democratization. The third question is whether the fiscal-expansionary impact of democratization, if it proves significant, is materially affected by national politics, the character of political parties managing democratization, and the ideological makeup of the forces leading the transition to democracy. The fourth question is whether fiscal expenditure expansion also mobilizes policies for revenue expansion and how are these policies dependent on national political conditions. These questions are analyzed in this section. In the years 1974–6 the three dictatorships of Southern Europe were successively eliminated. Portugal, Greece, and Spain became young democracies and began, in that order, their process of transition toward stable democratic life on a road full of challenges and obstacles. The challenges were not only political. They were also economic, and formidable at that. It is remembered that 1973–4 was the period of the first oil shock that plunged the world economy into a serious crisis and created economic uncertainty of large proportions. Its first impact was on price inflation that jumped from 8.6 percent in 1973 to 13.3 percent annually in 1974 in the European Community as a whole. The three South European economies were much harder hit by this inflationary spiral: Portuguese inflation peaked at 29.2 percent in 1974, Greek inflation peaked at 26.9 percent in the same year, whereas in Spain the shock was milder, but nevertheless serious, with 15.7 percent in 1974. Continued inflation would be a serious problem for the young democracies. The appearance of economic crisis shortly before the collapse of the three dictatorships furnishes a likely factor, but by no means an exclusive one, which heightened the pressures and hastened the establishment of democracy. It has been argued persuasively that as dictatorships seek to

310

Democratic and Economic Consolidation

draw legitimacy from good economic performance, they also become vulnerable to economic crisis (Maravall 1997: 6). The three South European dictatorships could lay claim on good economic performance, since up to 1973 their economies were the fastest growing in Europe. In any event, the very argument that economic crisis hastened the demise of dictatorships also serves to underline the difficulties that young South European democracies had to face, coming to grips simultaneously with a legacy of political and economic crises that were intertwined. It is important to gain a macroscopic comparative view of economic development prior to and during the period of democratic transition of the three South European economies. In Table 8.1 are shown GDP growth rates, inflation rates, growth rates of real investment, and growth rates of government consumption spending, over three subintervals of the period 1960–79. The three South European economies are compared to the benchmark of corresponding averages for the EEC. In the interval 1960–68 we can gain the basic characterization of the decade. The three South European economies were model performers. Their economic growth rates were much higher than those of EEC

Table 8.1 Economic performance, 1960–79

GDP annual growth rates Greece Spain Portugal EEC average Annual growth rates of fixed investment Greece Spain Portugal EEC average

1960–8

1968–73

7.3% 7.5 6.6 4.6

8.2% 6.8 7.4 4.9

1973–9

3.7% 2.5 3.1 2.4

9.5 12.5 6.0 5.9

10.6 7.8 10.7 4.9

0.2 1.1 0.2 0.3

Consumer price inflation Greece Spain Portugal EEC average

1.9 6.6 3.2 3.6

5.6 7.1 8.4 6.1

16.1 18.3 23.7 11.1

Annual growth of government consumption Greece Spain Portugal EEC average

6.2 4.1 9.4 3.9

6.2 5.3 6.6 3.8

7.5 5.4 9.2 3.0

Source: OECD (1986b).

Democratic and Economic Consolidation

311

countries. Inflation rates were lower in Greece and Portugal, compared to the EEC. Spain however was a clear exception on the inflation front, with an average annual rate of 6.6 percent, which for the standards of that period was quite high. The growth of real investment was also very high in Greece and Spain and slightly over the European average for Portugal. The growth of government consumption, that is, the purchase of goods and services by government, was generally higher in the three southern economies than in the EEC, but this was not inconsistent with the much higher general growth rates that these economies enjoyed. The period 1968–73 has been separated out in the Table in order to furnish a more immediate, but again sufficiently representative, picture of economic conditions prior to democratization. The strong performance of the southern economies continued and the only significant change is that inflationary pressures built up everywhere during this period. The acceleration of consumer price changes was appreciable in Greece and Portugal, but small in Spain where the inflation rate was already higher in the previous interval. Of course, inflation of consumer prices also accelerated in the EEC. Nevertheless, it can be surmised that a more than doubling of the inflation rate in the two poorest countries of Europe, Greece and Portugal, had a negative political impact on the respective dictatorships.1 Gathering inflationary pressures notwithstanding, the broad picture that emerges is that until the time of energy crisis, the three South European economies had been growing substantially faster than their European neighbors for over a decade and a half, and had been experiencing conditions of macroeconomic stability at the same time. This long period of economic boom undoubtedly fostered incomes and economic well-being in the European south. Equally undoubtedly, it brought to the fore new pressures and expectations for the fairer distribution of the benefits of growth, and for change that would accommodate new activities and new initiatives. The contribution of economic progress to the ‘obsolescence’ of the long-standing dictatorships in Spain and Portugal can be analyzed at many levels. A meaningful approach focuses on the machinery of policymaking and policy outputs under a dictatorial regime.2 Yet, the long economic boom cannot be taken as a sufficient condition for democratization in Southern Europe. Three arguments support this. First, on a theoretical level the long boom may have in fact increased support for the existing regimes and presented them with new margins for political and economic reforms which would offer palliatives but not harm the fundamental character of the regime. Second, contrary to the Iberian countries, in Greece the direction was reversed. A military

312

Democratic and Economic Consolidation

dictatorship was established in 1967, precisely in the midst of the economic boom. It is nonsense to ascribe the later collapse of the same dictatorship to the same economic boom. Third, as history always takes its revenge on experimental social scientists, the observations from the 1970s are not clean. Economic boom ended relatively abruptly in the middle 1970s, not by a measured economic slowdown but by a sharp crisis. The experimentalist is placed in a quandary: did South European democratization result from long boom or from sharp short crisis? The question itself is misplaced, of course. Economic conditions cannot sufficiently explain change, but they can offer powerful insight into the choices, the challenges, the demands, and the constraints which shaped policies and policy outputs. In the case of South European democratization, considerable insight can be gained by looking at the last column of Table 8.1, and at how sharply conditions changed in the 1970s, as compared to the 1960s. The period of the two energy crises is captured fully in the observations of the interval 1973–9, and this was clearly a time of sharp economic problems for the three South European economies. Average growth rates halved, inflation rates almost tripled, real new investment evaporated. These features paralleled developments in the EEC, except that there the changes were much milder. The fragility of the weaker economies of the European south became obvious, at precisely their time of crisis and general economic malaise. From a more macroscopic viewpoint one might also argue that as the economies of South Europe had overtaken their northern neighbors during the boom, so they also overtook them, in the negative direction, during the sharp recession following the energy crisis. This formalism however does not capture the important specificity of democratization that intervened in 1974–6. The societies of the south had to put together new political institutions, to engage in the reconstruction of democratic politics, while at the same time they had to come to grips with their worse economic crisis in almost two decades. The economic effects observed after 1973 in the three economies of the south embrace both the impact of the international crisis and the features of the national collective path to democracy. Democratization is almost synonymous with a universal demand for political rights. Its primary content is therefore the provision of ‘political goods.’ Yet, the liberation of expectations and of pent-up demands for wages, economic well-being, social protection, and economic opportunity always appear simultaneously, as they are anchored in repressed needs from the undemocratic past. Thus, it is normal to expect that democratization will heighten pressures for the provision of ‘economic’ as well as ‘political’ goods. If, in addition, an economic

Democratic and Economic Consolidation

313

crisis is present, the need for protection from the effects of the crisis can become a heated political issue in economic policy choices. On the part of those who shape policy, the need to gain legitimacy is heightened during a period when political forces seek to put together a new democratic order that can work, while engaging in competition with each other at the same time. The basic hypothesis suggested by this argument is that the primary fiscal response to democratization will be an increase of public expenditure, since this is the most readily available instrument with which democratic governments can quickly accommodate demands for protection and build support for their exercise of economic policy. A careful look at the last panel of Table 8.1 gives a first indication of the verification of the hypothesis in the case of the three southern economies. In the lowest panel are shown the growth rates of government real consumption expenditure.3 Observing these growth rates, we note that in all three countries they accelerate in 1973–9 as compared to 1968–73. This happens despite the fact that the overall rate of economic growth in these countries is reduced by more than half at the same time. It also diverges sharply from the European trend, where the rate of growth of government consumption declines appreciably from an average of 3.8 percent in 1968–73 to an average of 3 percent in 1973–9. Furthermore, it is also notable that there are large differences among the three southern countries. The rate of growth of government consumption jumps to a high level in Portugal (9.2 percent), to an intermediate level in Greece (7.5 percent), and to a level only slightly higher than in the previous period in Spain (5.4 percent). This significant variation between the three is repeated in other evidence, and is interpreted more fully below as the outcome of widely divergent political paths followed in the three countries during the course of democratization. Consistent with the logic of path dependency, different policy choices made earlier in the transition period eventually constrained the range of policy options which were available later in the consolidation process. Portugal’s transition was managed by and large by leftist forces, whereas Greek and Spanish transitions were managed by moderate-conservative governments. National political differences played an important role in the pattern and structure of fiscal impacts of democratization, as we observe in a later part.4 To look more closely at public expenditure during the years of democratic transition we can consult the data in Table 8.2, where are shown current disbursements of government as a percent of GDP, and their main components (consumption, social security, other transfers) over the interval 1973–9. The data of Table 8.2 offer a more detailed view than the broad averages of Table 8.1.

314

Democratic and Economic Consolidation

Table 8.2 Government expenditures as percentage of GDP, 1973–9 1973

1974

1975

1976

1977

1978

1979

Greece Consumption Social security Other Total

11.5 6.8 2.8 21.1

13.8 7.1 4.1 25.0

15.2 7.4 4.1 26.7

15.1 7.7 4.6 27.4

16.0 8.5 4.5 29.0

15.9 9.2 4.8 29.9

16.3 8.8 4.6 29.7

Spain Consumption Social security Other Total

8.6 9.6 1.5 19.7

8.8 9.5 1.5 19.8

9.2 10.3 1.7 21.2

9.8 11.1 1.9 22.8

0.0 11.7 2.0 23.7

10.4 13.2 2.6 26.2

10.9 14.5 2.3 27.7

Portugal Consumption Social security Other Total

12.8 4.6 2.1 19.5

14.1 5.3 3.3 22.7

15.0 8.6 3.6 27.2

13.7 11.6 5.6 30.9

14.0 10.8 6.8 30.8

13.9 10.1 7.8 31.8

13.9 9.5 7.8 31.2

EEC average Consumption Social security Other Total

15.7 13.6 5.1 34.4

16.6 14.3 5.4 36.3

17.5 16.3 6.0 39.8

17.3 16.6 5.9 39.8

17.2 16.9 6.1 40.2

17.4 17.3 6.5 41.2

17.4 17.3 6.6 41.3

Source: OECD (1986b) Tables 6.2, 6.3, and 6.4. Note: The computation of ‘Other’ is made by subtracting consumption expenditures and social security transfers from total current government disbursements.

Within seven years of democratization in South Europe, the current expenditures of southern governments gained an order of magnitude of 10 percentage points of GDP, moving from the region of 20 to the region of 30 percent. This constituted a huge increase of government spending and its weight in the overall economy. Thus, there is to be no doubt that the young South European democracies chose to wield the tool of fiscal expenditure generously. The general economic crisis made this all the more necessary, as it apparently did in all EEC countries. The expansion of government spending was a general phenomenon, but the EEC averages grew more slowly, and by less than South European rates of spending. Nevertheless, we must also observe that South European governments started from very low levels of spending compared to the rest of Europe. Thus, even after the intense increases in the 1970s, government spending levels as a percent of GDP remained roughly 10 percentage points lower in South Europe than in the EEC. Nevertheless, with the mediation of economic crisis, and the pressures of

Democratic and Economic Consolidation

315

democratic politics there was a measure of convergence between the south and the EEC in the 1970s. The broad similarity in overall spending growth among the three southern states recedes when we look at greater detail of time pattern and structure. In Greece and Portugal, most of the growth of the share of public expenditure takes place by 1977. In Spain, on the contrary, growth is rapid after 1976. Domestic politics explains these differences. In Greece, an election was held in 1977, and the conservatives who had ruled since 1974 were seeking to consolidate their own prospects, in part by boosting public expenditures, and were successful.5 In Portugal, the political upheavals, the sharp movements from Right to Left, a wave of nationalizations, salary and pension increases, and other attempts at rapid redistribution measures explain the huge increase of current public expenditure as a percent of GDP from 19.5 in 1973 to 30.9 in 1976. It is notable that in the remainder of the period, the only component of Portuguese expenditure that grows is ‘other transfers’ that by and large consists of interest payments on debts that were run up in prior years. This item was higher in Portugal than in any of the other two countries, and was a direct result of the earlier explosion of public spending. In Spain, the growth of public expenditure as a share of GDP is concentrated in later years, notably after 1976. In fact, until 1976 Spanish public expenditure grows much less than the EEC average. It is probable that, as a result of the gradualist type of Spanish transition, the pressures for expenditure growth that had appeared much earlier in the other two countries, became operational in Spain only after the general election of 1977. It was soon after the election that the UCD government succeeded in establishing the Moncloa Pacts in the autumn of 1977, that committed the major political parties to a macroeconomic stabilization plan and to a number of reforms in the area of social protection.6 The growth of public expenditure after 1977 was in large part due to social security outlays, as can be seen from the data of Table 8.2. This was the outcome of both policies, as embedded in the Moncloa pacts, and circumstance. Unemployment rose continually in Spain throughout this period, and it would be expected that social security spending would experience substantial pressure. This is also evident in EEC averages where the rise of social security spending over the interval 1973–9 takes up more than half the total increase in public spending in that period. It is impossible to ascribe the increase of social security spending specifically to political causes. It is nevertheless notable that both in Portugal and Spain the combination of democratization and economic crisis appeared to force something like a break in a barrier: social security transfers acquired the role of driving factor for

316

Democratic and Economic Consolidation

the augmentation of public expenditure. Again, Greece is at variance in this instance. The primary category of expenditure increase in Greece was direct government consumption spending. This reflected both the facts that unemployment did not rise as sharply in Greece at that time, and the implicit policy choice to maintain employment levels by maintaining public sector hiring as a tool against unemployment.7 The data and analysis so far enable the formulation of a central argument. National politics and institutional constraints shaped the paths and patterns of change in public expenditure of the young southern democracies, but the substantial overall growth is still the overriding common characteristic. The combination of economic crisis and democratization furnished a powerful impetus for an expansion of the spending role of government. Change in the southern democracies was rapid. Within six years the share of public spending in GDP grew by 50 percent, whereas in the EEC it only grew by 20 percent. As a major feature of the period of democratization, the very large growth of public spending created an obvious fiscal requirement that could put to the test the stability of the economies and the viability of the fiscal system: growing expenditures had to be financed. Debt finance is an immediately available response which usually presents few political problems but can lead to major economic problems. Debt becomes a cause of erosion of fiscal confidence as it accumulates. Tax finance, on the other hand, is politically more difficult to impose but economically a more appropriate response that avoids the destabilization caused by deficits and the accumulation of debt. Policies to increase taxes are politically as difficult as policies to increase expenditures are easy. At times of economic crisis when economic relief is high on the political agenda, the collection of additional taxes is a challenge for governments. In the period of democratization in the European south, the additional pressures of legitimization, acceptance of democratic processes, and political competition created obvious tendencies for policymakers to put off reforms and policies that would bring about a radical increase in taxation to match the evolving radical increase in public expenditure. As a general proposition, this must be moderated by two considerations. The first is that some tax increases normally follow increases in expenditures: Public expenditures create new taxable incomes both by pushing existing incomes onto higher tax brackets and by drawing resources from the unofficial (and untaxed) to the official economy. The second is that governments can overcome political resistance to increased taxation if they manage to present persuasively the fairness, the ‘just character’ of the taxes. Measures for the elimination of tax evasion, policies of increased progressivity of taxes, campaigns against tax fraud furnish classic cases

Democratic and Economic Consolidation

317

where tax reform can become a vehicle for gaining rather than losing public support for economic policy on account of taxes. In the particular case of the period of democratization of the three South European economies, the response of tax revenues to the needs created by the great increase in expenditure is therefore an empirical question. It depended on the extent to which governments of the period were able to overcome the circumstances and political constraints of the times. Tax receipts and social security contributions in four signal years are shown in Table 8.3. They are computed as percentages of GDP.8 An immediate and general observation is that between 1970 and 1980, the interval in which we would expect to capture the tax effects of transitions to democracy, taxation as a share of GDP grew very little in all three southern economies, with Spain remaining by far the lowest. From 10.6 percent of GDP in 1970, Spanish tax revenue amounted to 11.6 percent in 1980. This is also observed for the European average, where tax share grew very little. It would appear justified to conclude that in the midst of economic crises, additions to tax burden were not politically feasible or desirable anywhere, all the more in the young democracies of Southern Europe. The picture changes considerably if we look at social security contributions. We saw already, when studying expenditures, that social security outlays absorbed a large portion of the increase in total public expenditure in the late Table 8.3 Tax and social security receipts as percentage of GDP 1965

1970

1975

1980

Greece Tax receipts Social security Total

12.4 5.7 18.2

14.6 6.3 20.9

14.8 6.2 21.0

16.1 7.9 24.0

Spain Tax receipts Social security Total

10.5 4.2 14.7

10.6 6.3 16.9

10.3 9.3 19.5

11.8 11.1 22.9

Portugal Tax receipts Social security Total

12.4 3.5 15.8

15.1 4.7 19.8

13.9 7.3 21.2

17.3 7.3 24.6

EEC average Tax receipts Social security Total

21.5 6.3 27.8

23.9 7.3 31.2

24.8 9.3 34.1

25.6 10.1 35.8

Source: OECD (2000), Tables 3, 4, and 14.

318

Democratic and Economic Consolidation

1970s. Here, we observe that social security contributions also underwent considerable increases from 1970 to 1980. Spain is the front running case with contributions rising from 6.3 percent of GDP in 1970 to 11.1 percent in 1980. This was a very large increase. Increases in Portugal and Greece were respectively lower than in Spain, but in both cases social security contributions expanded more than taxes. If we compare the three southern economies to the EEC average we note that in the EEC also social security contributions grew faster than tax receipts over the relevant interval, going from 7.3 percent of GDP in 1970 to 10.1 percent in 1980. From a quantitative point of view, the rapid growth of social security receipts, especially in Spain and Portugal, led to visible convergence of social security receipts between the southern democracies and the EEC. This is an important observation and can draw both economic and political interpretation. The large expansion of social security contributions was a rapid response to the increase in the corresponding expenditures, which were clearly driven by the looming economic crisis of the 1970s. Since in all three southern countries, as in most of Europe, social security is funded separately from the general state budget and operates as an autonomous system, its receipts are earmarked and can be directly tied to the finance of expenditure. This transparency of use and the flowthrough character of the link between receipts and expenditures within an autonomous system, make it easier to calculate, enforce, and justify politically increases in social security contributions. Indeed, the three southern democracies made this adjustment as the most significant step toward the rectification of fiscal deficits that were created by large increases in social spending. Both in terms of fiscal behavior and in terms of the size of the intervention in the area of social security, the three southern economies exhibited convergence upon the EEC average. The international crisis of the 1970s must be clearly credited with this convergence.9 The three young democracies of Southern Europe successfully survived the challenges of transition from dictatorship in the midst of a serious world economic crisis in the 1970s. Their states were traditionally small and relatively weak economic entities. Taxing and public spending as shares of GDP were the lowest in Europe. Their economies had undergone spectacular economic growth in the sixties, leading their societies to increased well-being (but also giving rise to new demands), and leaving their states with budget surpluses and monetary stability. The economic crisis of the 1970s may have contributed to the demise of the dictatorships, but its most prominent characteristic was that its effects mingled with the political requirements for democratic legitimacy, and produced specific and large changes in the fiscal

Democratic and Economic Consolidation

319

order of the three southern economies. The economic role and size of the state grew rapidly. Public spending was enlarged with social expenditures leading the way, as would be expected in times of crisis and as trends in all of Europe also showed. Public receipts were much slower to increase, thus implying that the young democracies had to forego their previous surpluses and finance expenditures through borrowing. This would then present them with a major challenge in the ensuing period when they undertook to consolidate their democracies by ensuring, inter alia but importantly, the stability and growth potential of their economies. The politics of transition in each country played a large role in the structure and the time path of responses to the challenges of crisis. The left-wing forces that dominated early in the Portuguese transition brought about rapid institutional changes in the economic role of the state that then translated to fiscal results. On the other hand, the moderate-conservative governments which managed the transitions in Greece and Spain did not subscribe to radical institutional shifts but did follow a policy of avoiding major economic conflicts in order to maintain and strengthen the legitimacy of the democratic order. Again these policies translated to specific fiscal results. The growth in the economic role of the states in Southern Europe was clearly fostered by the sense of democratic empowerment which dictated interventions in the economy at a time of a serious international and national crisis.

Consolidation of Democracy and the Economy The democratization of the South European societies during the economic crisis of the 1970s left their states open to two major challenges. One was a broad agenda of economic reform that related to opening up the economy, wiping away protectionist barriers, deregulation of markets, and elimination of subsidies to failing firms. The second was the economic viability of the state itself. This required a resolution of the fiscal dilemma which early and rapid growth of public expenditure had posed. Accumulating deficits are not sustainable. They affect the stability of the economy and endanger its growth. They must be cured either by a rolling back of expenditure or by an increase in receipts. Both options require tough policies. The importance of these challenges for the young South European democracies was not confined to the sphere of economic policy. It had broad political significance. After the turmoil of political transitions in the 1970s, the ability of economies to perform well and to fulfill the expectations which had accompanied democratization became more and more a

320

Democratic and Economic Consolidation

question about whether democracy could fulfill its promise. Furthermore, after the crisis of oil prices, politicians all over Europe and the world were making a central issue of the need to craft new policies that would guide the developed economies to a new paradigm of economic stability and development. Thus, issues of policy and economic structure were central to political agendas everywhere, all the more so in the young democracies of the European south. The quest for democratic consolidation inevitably overlapped with the need for economic consolidation and the crafting of institutions and policies that would produce new economic growth. This did not leave South European policymakers indifferent. Their aspirations for European Community membership and international linkages were important factors in shaping their own responses to crisis and adjustment. The transition to democracy in Southern Europe was linked from its early stages to a strategy of European integration. The isolation of dictatorships from Europe and from the important economic and political trends that were evolving in the European Community was one of the major failings that the new democracies had to address. Indeed, important political forces in all three countries explicitly placed European integration high on their agenda. These included all moderate and conservative parties, as well as socialist parties in Spain and Portugal. In fact, it was the socialist parties of these two countries that negotiated as governing parties the accession of Spain and Portugal to the European Community in 1985. The Greek accession was prepared by governing moderate-conservatives under Constantine Karamanlis. PASOK, the Greek socialist party voiced early objections to membership in the EEC, but fully accepted it when it was obliged to manage, as governing party, the early steps of Greek membership in 1981. The majority of the electorate in each of the three South European countries was constantly in favor of European integration, although on occasion strong dissenting voices were raised. The commitment to European integration had deep political and economic implications. On one hand, it signaled a permanent reinforcement of the democratic order in Southern Europe, by linking national to European institutions through the growth of political and economic interdependence. On the other hand, it also signified that the Southern European countries would have to move quickly to open up their economies and to abide by European Community norms, standards, and common policies. The imposition of fiscal discipline, better control of deficits, and the elimination of inflation were among the major issues that would weigh on the ability and the success of the economies of the south to integrate with those of their European neighbors. The legitimacy of policies that would lead to the correction of macroeconomic

Democratic and Economic Consolidation

321

imbalances would be reinforced by the strategy of European integration, since economic reform would become part and parcel of each country’s European agenda. The link between European integration and the correction of domestic economic imbalances was not reinforced by domestic political debates alone. It was also directly invoked in negotiations between the European Community and each southern country, either in accession talks or, much later, in the preparations for participation in the European Monetary Union. The effectiveness of this external pressure varied among countries however, if one is to judge from postnegotiation behavior. Although this is a point that will be taken up later in this section, there is no doubt that the strategy of European integration was a source of internal and external pressures for pushing through economic and fiscal reforms in each of the three countries of Southern Europe. The overriding need for economic stability and fiscal consolidation became a pressing political priority in the European Community. In the years after the second oil shock, the Community went through one of the worst periods of its history in terms of economic conditions. Stagnation and inflation settled on the European economy in the early 1980s. The real GDP growth rate in Community countries declined from an annual average of 2.4 percent in the period 1973–9 to 0.9 percent in 1980–4. Inflation hovered around an average of 10 percent at the same time, and real investment showed an average annual decline of 0.5 percent in 1980–4—a rare exception in the Community’s history. The protracted crisis that combined inflation and economic stagnation gave rise to the quest for strong policies that would move the European economy forward. The strategy of deepening integration and formation of a true ‘internal market’ started taking shape in that period (see Tsoukalis 1997, Chapter 3, especially 33–40). Part and parcel of that strategy was the need to impose and maintain macroeconomic stability in each participating economy, and this, of course, imposed a serious constraint on the economies negotiating entry in the 1980s. Spanish and Portuguese policymakers, who were actively seeking EEC accession in the early 1980s, could not ignore this constraint, nor could they operate outside the emerging economic policy paradigm of the internal market and the stabilization of exchange rates in the European Monetary System. An important historical aspect of democratic consolidation is that in the 1980s, there were important political shifts in all three countries. Socialist parties came to power in Greece and Spain, where moderateconservative parties had previously managed the transition, and these socialist parties would govern throughout the decade. In Portugal where the transition was dominated by extra-parliamentary forces

322

Democratic and Economic Consolidation

and strong leftist movements, parliamentary politics and parties had started to regain center stage already since 1979. The coalition of socialists and centrists of Bloco Central led by Mario Soares that came to power in 1983 would be succeeded in 1985 by the conservative PSD government of Cavaco Silva that would remain in power through the end of the decade. The alternation in power of major political parties and the formation of stable governments by parties that were in the opposition early in the transition process was practical proof that democratic consolidation had been achieved. The three South European societies were maturing as solid democracies that could accommodate the alternation of parties in power. The serious economic crisis of the early 1980s, that pushed for a change in European policies and for a new policy paradigm for the deepening of economic integration in Europe, affected the stance of the EEC in negotiations with the Iberian countries for accession to the Community. Greece had already become a member in January 1981. Since the new policy paradigm in Europe had not yet emerged as clearly as in the early 1980s, it can be surmised that Greek entry was much more affected by political considerations in the context of democratic transition than the accession of the Iberian countries would be. Besides the effects of explicit European policy, however, we must also repeat that the economic crisis had a direct impact, since a worsening European economy also implied a worsening of conditions for the southern economies. In fact, in the period 1980–84, the southern economies’ growth rates were the lowest of the postwar period, and their inflation rates were the highest.10 All three countries were therefore hard pressed to undertake programs of economic stabilization in the 1980s. In Spain, the new socialist government that was elected in 1982 promptly implemented a stabilization plan that included severe macroeconomic and monetary restraints. In Portugal, the Bloco Central under the leadership of Soares also put in place a stabilization program in 1983. In Greece, the socialist government undertook stabilization only after its reelection in 1985, and a severe program of demand restraint was to remain in effect for two years.11 The important point is that, in every case, a shift toward very unpopular economic policies was undertaken with a fresh electoral mandate and by a political party that rose to power as an alternative to the one that had managed the transition to democracy. That an unpopular policy was adopted implied that economic reform was becoming politically feasible without endangering democracy. The fact that erstwhile opposition parties adopted economic reform when they came to power implied that a broader tradeoff was achieved by the young southern democracies. To experience a smooth alternation in

Democratic and Economic Consolidation

323

power was a major sign of the strength of democratic systems, and therefore, the achievement of a significant ‘political good.’ Stabilization policies are unpopular because they always imply the restriction of economic goods since they are based on curbs of demand and incomes. Despite their unpopularity they became politically feasible for the new governments in the 1980s, because it was becoming clear to electorates that economies were facing an impasse, and also because these governments were no longer under serious threat of a relapse to dictatorial conditions. Basic indices of economic performance of the three South European economies and the EEC for the period 1980–2004 are shown in Table 8.4. If we seek a general characterization of the period 1980–2004, we notice that it shows an alternation of four phases of high and low economic performance: the early 1980s, 1990s, and 2000s were periods of slowdown or recession, while the late 1980s and 1990s were periods of growth. In this general alternation of economic phases, it is visible

Table 8.4 Indices of economic performance, 1980–99 1980–4 GDP annual growth rates Greece 0.8(%) Spain 1.4 Portugal 1.3 EEC average 0.9

1986–9

2.4(%) 4.7 5.8 3.4

Annual growth rates of fixed investment Greece 4.5 1.2 Spain 1.4 12.9 Portugal 1.5 12.0 EEC average 0.5 6.2 Consumer price inflation Greece 21.9 Spain 13.3 Portugal 22.8 EEC average 9.8

15.5 7.4 13.6 4.2

Annual growth of government consumption Greece 3.0 2.8 Spain 3.8 6.7 Portugal 3.7 6.6 EEC average 1.5 2.0

1989–4

1995–9

2000–4

0.8(%) 1.6 2.1 1.5

2.9(%) 3.5 3.4 2.4

4.1(%) 2.9 1.1 1.7

0.4 0.8 2.7 0.3

7.9 6.8 7.2 4.0

7.9 3.4 1.5 0.9

16.2 5.9 9.6 4.2

6.4 3.1 3.8 2.2

3.3 3.2 3.3 2.2

0.5 3.7 4.0 2.0

2.2 2.5 2.3 1.2

n.a. n.a n.a. n.a.

Sources: For 1980–4, OECD (1986b), Tables 3.1, 4.3, 8.11, and 4.7; for 1986–9, OECD (2001a), Tables 3.1, 4.3, 7.11, and 4.6; for 2000–4, OECD (2004), Tables 2, 5, and 18.

324

Democratic and Economic Consolidation

that the South European economies move in step with the EEC—a reflection of the increasing harmonization of their economic movements with the outside world. Theirs could not be considered closed economies in a true sense any longer. Looking more specifically at GDP growth rates and the rates of change in fixed investment (the top two panels of Table 8.4) we note that the early 1980s were a period of dismal performance for the three southern economies, but this was matched by poor EEC-wide performance. Furthermore, Spain and Portugal had instituted stabilization programs during that period, and the restriction on domestic demand that was due to these programs was taking its toll on growth. It is also noted that of the three southern economies, Greece exhibited the worst growth and investment record in the early 1980s, despite the fact that it was the only economy that was engaging at the time in expansionary economic policies. Its stabilization program was instituted in 1985, and it affected its performance in the second half of the decade. In this overall characterization of the early 1980s, we must also note that this was a period of high inflation for all of Europe, but that inflationary conditions were extreme for Greece and Portugal, where annual price level change exceeded 20 percent, but not so extreme in Spain. On the other hand, in this period Spain experienced a big rise in unemployment.12 The second half of the 1980s was a period of recovery and economic development. The EEC economies experienced significant growth, a substantial recovery of investment and an equally decisive decline in inflation. At this time, which also coincided with the first postaccession period for the two Iberian countries into the EEC, these two economies exhibited a spectacular recovery and high growth of both GDP and investment. In fact, the Spanish and Portuguese growth rates exceeded considerably the European average and their high investment rates were reminiscent of the boom conditions of the 1960s. One could persuasively argue that the two economies reaped early the benefits of both their internal stabilization efforts and their entry into the EEC. Growth always creates a positive environment for additional economic reform and structural change. It can be surmised that the interval of Iberian growth in the 1980s offered further impetus to economic change and made the necessary policies much more feasible politically than they would have been otherwise. A case in point is tax reform, at which we shall look more closely later. At the general performance level, we also note that consumer price inflation declined significantly in the three economies. In fact, looking at the long-term picture through the 1990s and the early 2000s, at the third panel of Table 8.4, it is clear that a process of convergence of the southern economies to the low inflation levels of the EEC began in the second half of the

Democratic and Economic Consolidation

325

1980s and would continue uninterrupted through the end of the period. This has been of course the outcome of strong policies that led to, and finally achieved, monetary union and the single European currency. These policies were to be adopted by all three Southern European countries, which were full-fledged members of the Eurozone at the close of the last century. The comparison of performance between the three South European economies shows a remarkable similarity between Spain and Portugal, and a significant differentiation of Greece. In fact, a careful review of the first three panels of Table 8.4 reveals that in terms of GDP growth, growth in investment, and consumer price inflation Greece fared much worse than both the other two countries and the EEC through the period 1980–94. Only in the last subperiods of 1995–2004 does Greek performance improve appreciably. The Greek exception was in part due to circumstance and in part due to policy differences. In the 2000s Greece fared well due to a big investment push motivated by forthcoming Olympic Games, clearly a special factor. The similarity of Spain and Portugal had surely been caused in part by the coincidence of the timing of their stabilization programs and of their accession to the EEC. The difference of Greece had multiple causes that extended in time. In effect, Greek stabilization was greatly delayed and would eventually be successful only in the second half of the 1990s. The stabilization of the 1980s had failed to impose macroeconomic balance and usher in a period of growth in Greece, as had happened with the Iberian economies. The economic policy of PASOK when it took power in 1981 was not constrained by an ongoing negotiation or an impending accession into the EEC, since that was already an accomplished fact. Furthermore, a political platform of broad social change that would touch the unprivileged and disenfranchised strata of society was the foundation of PASOK’s electoral appeal and its meteoric rise. Early policies were redistributive and expansionary, and rapidly led to significant increases in public spending and deficits. Besides boosting domestic demand however, PASOK also instituted policies for the reform of the ‘supply side’ of the economy, in the form of promoting large integrated industrial projects, infrastructures, and sectoral policies.13 These policies did not prove successful mainly because domestic demand expansion did not feed into sufficient domestic investment, but rather soon led to an external payments crises and exchange rate instability.14 These conditions undermined the ability of government policy to intervene effectively on the ‘supply side’ of the economy. More specifically, monetary uncertainty and the risk of devaluation undermined the investment climate, and growing fiscal deficits undermined the

326

Democratic and Economic Consolidation

potential for public investment initiatives. The contrast to PSOE’s policies followed in Spain at the same time is an interesting one. The Spanish socialists also engaged in ‘supply-side’ policies extensively. However, they combined these with severely restrictive monetary policies and macroeconomic stabilization (see Boix 1998: 107–13). In part, this combination was necessary because of their impending EEC accession. Strict monetary policies were also conceived and implemented in conjunction with the eruption of a major banking crisis in Spain that led to a necessary restructuring of the country’s financial system. In contrast with Greece, Spanish policies managed to impose monetary stability and restrain public deficits. These conditions promoted private and public investments, and these in turn became the most important vehicle for successful ‘supply-side’ reform. Spanish policies clearly benefited from favorable timing since they were already in place and working by the time general economic conditions turned favorable in Europe in the mid-1980s. Thus, they were rewarded by rapid growth in the second half of the 1980s, whereas Greek policies were not. After the election of 1985, when the Greek socialists were again returned to power, the government of Andreas Papandreou instituted a formal stabilization policy under the management of Constantine Simitis who took the post of minister of national economy. The policy was initially successful but was abandoned in 1987. Sensing erosion in political support, the government aborted stabilization too soon, changed ministers of national economy, and refashioned once again an expansionary policy until the next election of 1989. The early abandonment of stabilization essentially meant that the economy and society had to absorb the costs and the pains of restrictive demand measures, without having reached the point of reaping the benefits of lower inflation and higher investment and growth rates. Thus, the Greek economy was destined to finish the decade of the 1980s with high inflation, high deficits, and a public debt looming higher than those of the other two countries of the European south (see Thomadakis and Seremetis 1992: 209–23). Nevertheless, the failure of macroeconomic stabilization did not cancel other efforts for structural reform, and, among other policies, a major liberalization of the financial system was initiated at the end of the tenure of the socialist government (see Hellenic Banks Association 1987). This would prove to have long-term effects on the economy in the ensuing period. A serious economic slowdown in the first years of the 1990s, followed by substantial and sustained recovery in the remaining years of the last decade of the twentieth century, shows a general pattern of convergence in economic performance of the southern countries vis-a`-vis

Democratic and Economic Consolidation

327

the EEC. GDP growth rates, rates of change in investment and inflation rates attained comparable levels and changed in the same direction overtime. Greece followed this pattern with a delay, but it is evident in the data for the period 1995–2004, that the Greek economy finally succeeded in achieving stabilization and macroeconomic adjustment, while at the same time expanding with an above average growth rate. Thus, on the whole, the three young democracies of the European South achieved a remarkable convergence to the EU average, in terms of economic movement and performance. Let us now turn our attention to whether this convergence also extended to basic fiscal structures.

The Comparative Evolution of Fiscal Structures The economic phases of the last two decades of the twentieth century found the three Southern European economies on a path of convergence with the other members of the European Community in terms of overall economic performance. Our observations from the earlier period of transition to democracy had shown that fiscal expenditure had been growing at a rapid pace, outstripping public revenues. It is now important to examine how the imbalance between fiscal expenditure and revenue was corrected, so that in the end all three economies were able to obtain membership in the European Monetary Union and to take part in the common currency project. In terms of broad fiscal data, the most meaningful summary statement can be obtained from OECD data, which consistently record, over long periods of time, current disbursements and current receipts of government.15 These are shown as percent of GDP in Table 8.5, for the three South European economies and the EEC average respectively. The main observations from Table 8.5 are the following. First, there was continued growth in the percent of GDP taken up by government disbursements. In the case of Spain and Greece the growth of GDP percentages was quite dramatic over the period covered, whereas in Portugal growth was moderate and more in line with the milder trend shown for EEC countries as a whole. Given the lower starting point of South European economies, these differential developments imply that Greece and Spain converged to the average European level of expenditure faster than Portugal. Second, a similar pattern appeared in the development of current receipts of governments, as a percentage of GDP of the respective countries. Receipts grew in step with disbursements, with more dramatic changes in the cases of Greece and Spain. It should be noted that the relative equilibration of receipts and disbursements occurred in Greece later than in the other countries; this is

328

Democratic and Economic Consolidation

Table 8.5 Fiscal structures (percentages of GDP) 1980–4

1986–9

1989–94

1995–9

Total government outlays Greece Spain Portugal EEC average

36.4 36.8 41.0 47.7

44.9 41.4 40.2 49.5

47.7 45.8 43.3 50.0

52.2 45.3 42.5 49.0

Current disbursements of government Greece Spain Portugal EEC average

36.1 32.5 36.5 43.8

42.0 36.1 36.5 45.2

42.5 40.5 39.7 49.1

49.3 41.3 38.7 45.7

Social security transfers Greece Spain Portugal EEC average

12.1 15.5 11.1 17.2

15.1 15.5 10.5 15.8

14.6 17.3 12.6 17.4

15.4 17.2 11.7 17.2

Current receipts of government Greece Spain Portugal EEC average

32.1 31.8 35.0 41.9

34.5 36.2 35.0 43.4

34.7 39.7 38.1 43.5

46.9 38.7 37.6 44.1

Sources: For 1980–9, OECD (1991b), Tables 6.5, 6.4, 6.3, and 6.6; for 1990–9, OECD (2001a), Tables 6.5, 6.4, 6.3, and 6.6.

consistent with our earlier explanation about the delay in Greek adjustment and successful stabilization. In fact, during the period of the economic downturn in the early 1990s, the two other South European economies had managed to establish basic balance, according to the estimates of Table 8.5. The difference between current disbursements and current receipts in Spain and Portugal did not exceed 2 percentage points of GDP in 1990–4, whereas in Greece, the same difference appears to have amounted to almost 8 percentage points of GDP. This is indicative of the fact that even by the mid-1990s Greece had not achieved fiscal balance and macroeconomic adjustment. These would be attained only in the later part of the 1990s. The measures of Table 8.5 are indicative of trends, but not very accurate descriptors of fiscal structures. Current receipts of government include not only taxes and social security contributions but also items such as receipts from sale of assets or state interests to private parties, revenues from profits and other nontax items. A focus on tax structures proper is now in order since this reflects more permanent

Democratic and Economic Consolidation

329

change, as well as a most important aspect of economic policy. Tax policy has, as we have argued before, both economic and political implications. In the context of the economic consolidation of the three economies of the European south, tax policy was an area of great priority and political sensitivity. In all three economies, taxes had to be increased over time in order to maintain earlier increases in public expenditure which, in many cases, were necessary in the new democratic order that was constructed in the three societies. In all three societies, the legacy of authoritarian regimes made tax imposition and tax collection a more difficult and sensitive issue than it would normally be in the institutional context of northern European economies. Finally, in all three economies there was a common step in the imposition and enforcement of the VAT, as a result of their accession to the European Community. This important tax has played an important role in the convergence of tax structures of the southern three to the Table 8.6

General composition of receipts (percentages of GDP), 1980–2002 1980

1985

1990

1995

2000

2001

2002

Greece Total Tax revenues Social security Workers Employers

24.0 16.1 7.9 3.5 3.5

28.6 18.5 10.2 4.5 4.3

29.4 20.5 8.9 4.4 4.4

32.4 21.9 10.5 4.3 4.8

38.2 26.4 11.8 4.6 5.5

36.6 24.9 11.7 4.8 5.5

35.0 24.1 11.8 4.6 5.6

Spain Total Tax revenues Social security Workers Employers

22.9 11.8 11.1 2.5 8.7

27.9 16.2 11.4 2.0 8.6

33.0 21.3 11.7 1.9 8.4

32.8 21.0 11.9 1.9 8.2

35.2 22.9 12.3 2.0 8.6

35.0 22.5 12.5 2.1 8.6

35.6 23.0 12.6 2.0 8.9

Portugal Total Tax revenues Social security Workers Employers

24.6 17.3 7.3 2.8 4.4

27.1 20.1 7.0 2.7 4.0

29.6 21.6 8.0 2.9 4.8

33.6 23.5 10.1 3.3 6.3

36.4 25.5 10.9 3.4 7.0

35.6 24.6 11.0 3.5 7.0

33.9 24.7 9.2 n.a. n.a.

EEC 15 Total Tax revenues Social security Workers Employers

35.8 25.6 10.3 3.0 6.5

38.6 27.7 11.1 3.5 6.6

39.2 28.3 11.1 3.6 6.6

40.3 28.5 11.8 4.1 6.6

41.8 30.3 11.5 3.8 6.7

41.2 29.7 11.6 3.7 6.8

40.6 29.2 11.4 3.7 6.8

Source: OECD (2000), Tables 3, 4, 14, 17, and 18; OECD (2004), Tables 3, 4, 14, 17, and 18.

330

Democratic and Economic Consolidation

common European standard. Table 8.6 presents tax receipts and social security contributions as a percentage of GDP for selected years, starting from 1980. The data in Table 8.6 reveal two noteworthy features. The first is that the three South European economies have made very significant progress in terms of taxation over the interval 1980–2002. From less than a quarter of GDP, their total tax receipts rose to a third of GDP over that period. This was no mean feat. Such a magnitude of growth testifies to the successful broadening of the tax base in those societies. The second feature is that the three converged among themselves and also converged toward the EEC average, although still remaining below it by about 5 percentage points of GDP. In 1980, the difference between the southern economies and the EEC average had exceeded 11 percentage points of GDP. Thus, it is clear without deeper analysis that politics and policies in the south were successful in achieving a very remarkable improvement and convergence of fiscal receipts to the European norm. More importantly, this success signified that the political system was capable of responding to the challenge of increased public expenditure in a fashion that maintained stability in the macroeconomic situation of the southern economies. The broadening of the tax base was accompanied in many cases, in Spain and Greece notably, by public campaigns against tax fraud and evasion. These contributed to the improvement of tax compliance, which is a basic underpinning of the viability of any fiscal system. The breakdown of total receipts into taxes proper and social security contributions indicates certain variations among countries. These variations are not very significant except in one respect: social security receipts appear to have been structurally lowest in Portugal and highest in Spain, with Greece furnishing the middle case. The difference between the two extreme cases amounts to about 2.5 percentage points of GDP. This may reflect history as well as politics. Spain has experienced over the interval discussed here the highest unemployment rates and this has made it politically expedient and necessary to lean heavily on social security contributions.16 Furthermore, practically throughout the observed period, the governing political power in Spain were the socialists who had since the early 1980s engaged in active reforms of the social security system. Thus, Spain’s social security receipts appear to have converged exactly to the European average by the mid-1990s, overtaking in the 2000s, whereas those of Greece and Portugal came very close or remained somewhat below that average. On the obverse side of the same observation, Portugal’s tax receipts proper (excluding social security contributions) were a somewhat higher percentage of GDP than in the other two countries, implying that more policy

Democratic and Economic Consolidation

331

emphasis was placed on taxation than social security funds in that country. An additional point of divergence among the three countries relates to the allocation of social security tax burdens between employees and employers. In this respect the differences are so significant that they can be considered spectacular. Spain, at one extreme, has placed most of the burden of social security taxes on employers. Consistently through the period shown in Table 8.6, employers were carrying more than four times the social security tax burden than employees. Greece at the other extreme has consistently imposed a higher burden on employees than on employers. This changed only in the years 2000 but the difference with the other two remained. In other words, this is contrary to what is observed not only in Spain and Portugal, but also in the EEC average, where in 1998 the burden on employers appeared to be double that imposed on employees. The Greek difference is notable because it probably testifies to the power of employers who have apparently been able to overcome what we would consider normal allocative policies of social security burdens. More generally, the allocation of social security tax burdens makes up an important element of progressivity in a country’s tax system. In most cases the allocation places more of a burden on employers, and this constitutes a prima facie progressive nature of the system. In the case of Greece, the unorthodox allocation in favor of employers constitutes a significant element of regressivity in the tax system, that has been consistently present in the country’s social security regime. Although it is not possible to offer an accurate measure of regressivity arising from this feature, it is clear that, compared to the other South European economies and to the European average, the Greek system is less progressive. We can now turn to more precise comparative evidence about the composition of tax receipts proper. A three-way categorization of taxes is helpful. One broad category is taxes on income, which includes both personal and corporate income and constitutes what we normally consider direct taxation.17 A second broad category includes taxes on goods and services, which constitutes what we normally consider indirect taxation. A third category includes property taxes. Table 8.7 presents percentage composition of tax receipts, excluding social security taxes, over the interval 1980–2002 for the three southern economies and the European average.18 The estimates of Table 8.7 are quite revealing as they show both change over time and significant divergence among the three Southern European economies, as well as between them and the European average. Let us first focus on observed changes that take place over time. The composition of tax receipts in the EU as a whole has been

332 Table 8.7 security)

Democratic and Economic Consolidation Tax composition (percentage of tax revenues excluding social 1980

1985

1990

1995

1998

2000–2

Spain Taxes on income Goods and services Taxes on property

50.6 40.3 8.9

44.6 48.9 5.9

47.4 43.9 8.5

45.8 44.8 8.6

43.5 45.4 9.3

44.2 45.6 10.2

Portugal Taxes on income Goods and services Taxes on property

27.9 63.7 2.0

34.7 57.8 2.6

35.3 60.7 3.7

35.9 61.6 3.4

38.7 55.4 3.9

39.4 55.7 4.9

Greece Taxes on income Goods and services Taxes on property

28.9 61.4 6.9

27.2 66.3 4.2

28.5 63.8 6.6

32.9 61.0 4.9

33.5* 60.0* 5.6*

37.6 54.4 8.0

EEC 15 Taxes on income Goods and services Taxes on property

47.6 43.7 5.9

47.8 44.3 5.5

47.8 43.9 5.9

43.4 39.8 6.1

48.7 41.8 6.5

50.2 42.9 7.0

* 1997 data for Greece. Source: OECD (2000), Tables 9, 23, and 25; OECD (2004), Tables 9, 23, and 25.

remarkably stable over the 1980s, 1990s, and early 2000s. About half of total receipts have come from direct taxes, about 44 percent from indirect taxes and the rest from property taxes. This very stable European structure has persisted through several phases of the economic cycle and this makes its stability all the more remarkable, testifying to broader institutional stability. In the three economies of the European south, in contrast, tax composition has been changing significantly. The change was not linear in most cases, and this is a significant indication of the sensitivity of change to shifts in the economic cycle, but probably also in the political cycle. Furthermore, 1985 is probably the most critical year in the institutional sense, since it is in or near that time that all three South European economies moved to implement the VAT. As a result, post-1985 trends are the important elements to observe here, since a major tax harmonization took place in the south of Europe. The fact that tax composition changed in the European south reflected both the change in the form of indirect taxation via the imposition of the VAT, and the effort that was clearly underway in each of these countries to increase tax receipts as a whole and achieve fiscal balance. The direction of change in the composition of taxes is contradictory in the three southern economies. In Spain, the share of direct taxes as a

Democratic and Economic Consolidation

333

percentage of total tax receipts declined in favor of indirect and property taxes, especially in the 1990s to increase somewhat in the early 2000s. In Portugal and Greece, by way of comparison, there was an increasing reliance on direct taxation during this same period. But the most notable feature of the data shown in Table 8.7 is that Spain was far ahead of the other two countries in terms of how closely its tax structure resembled the European average, with respect to its heavier reliance on direct taxes in particular. If we consider the degree of reliance on direct taxation as a preliminary indicator of progressivity in a country’s tax system, it is clear that Spain appears the most progressive and Greece the most regressive of the three Southern European states, with Portugal occupying the middle position. Greece’s delay in increasing the role of direct taxes, despite the ‘catch-up’ that is apparent in the late 1990s and early 2000s, was due mainly to the later overall adjustment of the Greek economy that was discussed above. More specifically, it was due to the fact that Greece did not partake in the growth phase of the late 1980s as fully and as successfully as the two other South European economies. It was during that phase that Spain, in particular, managed to bring about significant changes both in legal and in enforcement structures of their respective tax systems, especially the broadening of the tax base and the fight against tax fraud practices (see Boix 1998: 113–15). The overall picture that emerges from the period of economic consolidation of the three southern economies is a very revealing one. Tax receipts have increased and have corrected the structural imbalance that marred the fiscal condition of the three economies during and after their democratic transition. The increase of tax revenues achieved spectacular results and brought the three economies into close convergence among themselves and to average European structures. They thus were able to present appropriate macroeconomic alignment with the criteria required for EMU participation. The uniformity and convergence of fiscal conditions that are apparent in the aggregate do not extend to a more detailed level, however. Both the structure of social security revenue and the composition of tax revenues proper exhibit very significant divergences among the southern economies and between them and the European average. Spain appears to have attained the most progressive system of tax receipts, closely resembling the European norm. On the other hand Portugal and Greece appear to lag behind, with Greece implementing a more regressive system both in taxation proper and in social security contributions. The contrast between Spain and Greece is all the more notable since both countries have had socialist governments in power during most of the interval observed and discussed here.

334

Democratic and Economic Consolidation

Whereas overall adjustment and aggregate convergence have been necessary for European participation, the lack of uniformity in more particular aspects of the tax and social security systems of the three southern economies testifies to the truth of the assertion that ‘politics matter,’ both in terms of the general impact of democratic political processes, and in terms of party ideologies and strategies. The three economies of the south of Europe followed particular paths on their way to aggregate adjustment, and each faced specific constraints, both political and economic. There was also a different process of learning that was unwinding over that period, as policymakers had to respond to complex conditions involving both European and internal strictures and requirements. For example, the behavior of Greek socialists, after the defeat of their conservative opponents in 1993, was clearly influenced by the earlier failure of their unfinished stabilization effort of the mid-1980s. The behavior of Spanish socialists, in contrast, was affected by the fact that the economic boom of the mid-1980s improved greatly the degrees of freedom of their economic policy, and enabled them to continue with economic modernization, despite the dissolution of their implicit alliance with unionist labor that had supported their earlier policy efforts.

Concluding Observations Over the last quarter of the twentieth century, the three states of Southern Europe—Greece, Portugal, and Spain—achieved a major and complex historical feat. They established and maintained stable democracies, entered the EU and later joined the monetary zone of the Euro, achieved macroeconomic adjustment and enforced fiscal equilibrium, and shared the opportunities and the benefits of European growth. These achievements furnish an important historical example of successful integration of less developed economies into a system of more developed ones. The most significant feature of this integration has been the strong tendency for convergence between the less and the more developed parts of the integrating system. As a historical example this is notable because it counters many opposite cases of integration where divergence rather than convergence of the integrating parts has been noted. Furthermore, the success of European integration in the case of the three southern economies has undoubtedly been an inspiration for the policy for enlargement of the EU, which by the spring of 2004 extended to almost all of Eastern Europe. If integration with convergence is repeated in the broader field of this new and more ambitious enlargement, Europe will truly lay claim to a historic achievement of development through extensive integration, the

Democratic and Economic Consolidation

335

successive opening of markets, and the partial subjugation of national to supranational economic policy. In fact, this is precisely the vision that makes supranational economic policy in Europe all the more important and burdens it with complex duties. Democratization has provided a strong and fundamental impetus in the achievements of the three southern economies and societies. The drive to deliver economic well-being and wide participation in the benefits of growth was a strong corollary of the establishment of democratic order. At the same time, the drive to join Europe, and to converge to its norms, became an integral part of the vision of central political players that brought about the transition and the consolidation of democracy in the three southern countries. The two elements of the democratic vision—social benefits from growth and European convergence—worked their way through domestic politics so as to produce a uniform result in the three southern states: an increase in social spending combined with an increase in tax receipts that were more progressively structured. In this fashion fiscal balance and macroeconomic equilibrium were achieved and provided the basis for admission of the southern economies into the eurozone. It certainly must be emphasized that the success of this dual set of aspirations was predicated on two conditions. On the one hand, the significant periods of economic growth for Europe as a whole during the late 1980s and the 1990s amplified the degrees of freedom for economic reform and helped to combine prudent macroeconomic management with structural change in the southern countries. On the other, change in the south passed through an inevitable growth of the economic role and power of the state. The successful democratization-cum-integration could most probably not occur under any other conditions. The southern states were traditionally autocratic but economically weak, and this was a part of the dictatorial legacy that had to be reversed in both its dimensions: the state had to be transformed to a democratic but economically strong entity. And so it was. Part and parcel of European integration of course has been the subsequent transmission of some economic policymaking powers to supranational entities (European Central Bank, European Commission). Yet, state functions as a whole increased in the south, changing at the same time their character as compared to the dictatorial past. Convergence does not mean uniformity. In the south of Europe each country followed its very particular path both in terms of economic change and in terms of domestic politics. The initial conditions right after democratization were very different in the three countries. The terms of political competition and cooperation within their nascent democracies were also quite different during both the transition and

336

Democratic and Economic Consolidation

the consolidation phases of their democratic orders. Although they all achieved similar results, the underlying economic structures were not the same either at the start or at the end of the process. Tax structures and structures of social spending are a case in point. Domestic politics differed greatly in terms of content, speed and even orientation of reforms, although in all cases the overall goals were similar. Reform is a continuous challenge. The differences in economic structures and the ever renewable variation of domestic politics promise to fuel a differentiated path of reform for the three southern democracies of Europe in the future as well.

Notes 1. The comparison of average inflation rates over long intervals masks in this case a sudden shift in consumer prices that took place in 1973, and went on in following years. We come back to this point below. Taking this into consideration, we cannot make any stronger inferences about the effects of inflation. 2. See Gunther 1980 and 1996 for informed analyses of Spanish policymaking both before and after the Franco years. 3. This is the net contribution of government to aggregate demand, includes salaries and purchases of goods and services, but excludes transfers (social security pensions, interest). 4. The Spanish state administration maintained a degree of continuity during the democratic transition, in contrast to the chaotic condition experienced in neighboring Portugal. In Greece also no wholesale purge of the public administration was undertaken. See Gunther 1996: 169–70. 5. In the election of 1977 Nea Demokratia, the conservative party, led by Constantine Karamanlis obtained an absolute parliamentary majority with 42.9 percent of the vote. PASOK, the socialist party, led by Papandreou made an impressive electoral appearance with 25.3 percent of the vote and took the place of main opposition. 6. The pacts of Moncloa are an example of ‘concerted’ political agreement among major players for the exercise of agreed macroeconomic and collective bargaining restraint. Besides its obvious significance for economic stability, this type of consensus building was an important building block in democratic legitimacy and a radical departure from corporatist practices of the Francoist past, as Gunther points out (1996: 172–3). There is no instance of this type of policy consensus in the other two South European countries, during the period of democratization. 7. The moderate-conservative governments that managed successfully the democratic transition in Greece did not engage in serious economic reform but sought to avoid social conflicts that could undermine political commitment to democracy. This feature is shared by the policies of moderateconservative post-Franco governments in Spain. The obvious priority in

Democratic and Economic Consolidation

8.

9.

10. 11.

12. 13.

14. 15.

16. 17.

18.

337

both countries was to build a viable democracy and worry later about a viable economy. Also see Maravall 1997: 78–9. The Table switches to observation of particular ‘signal years’ because of data limitations in the OECD bases available. Expenditure and tax data are not uniformly reported. No obvious distortions have been detected from this change however. See Castles, in this volume, for an extensive comparative analysis of social spending. His findings include at least partial evidence that democratization furnished an important impetus to social spending. A more extensive discussion of economic performance follows in Table 4. For a discussion of political conditions and electoral influences on the choices of economic policy in the three countries at that time see Maravall 1997, Chapter 3. As Maravall notes official unemployment figures in Spain rose from 8.6 to 16.2 percent between 1979 and 1982. See Maravall 1997: 85. A change of leadership in the Ministry of National Economy in mid-1982 altered the pure demand policies of the early months. It brought the team of Gerasimos Arsenis and Constantine Vaitsos as minister and deputy minister, respectively, to the helm of economic policy. They instituted a mix of moderate demand and ambitious supply side policies. Vaitsos, a prominent developmental economist, headed the effort to model and implement sectoral policies. The drachma was devalued in 1983 and again in 1985. The terminology ‘disbursements’ and ‘receipts’ is based on OECD nomenclature in historical statistics. These categories differ somewhat from ‘expenditure’ and ‘revenue’, as the latter refer as a rule to more narrow measures. For example, receipts include nontax items. As trends are the important elements in the Table, we follow OECD nomenclature here. Also see the commentary on this by Castles in this volume. The OECD data do not allow a full decomposition of income taxation between personal and corporate income on all countries and for the entire period. Thus, the aggregate category of total income taxation is maintained in the Table. Some comments on the decomposition in the more recent periods for which there is full data are in the text that follows. The average shown in the Table represents the average of the 15 member states of the EU.

................ 9 ................

Conclusion Richard Gunther and P. Nikiforos Diamandouros

In this book, the fourth volume in the series on ‘The New Southern Europe,’ we have analyzed the basic characteristics of the administrative and judicial structures of the state in Greece, Italy, Portugal, and Spain, as well as the distinguishing characteristics of public policy outputs with regard to taxation, overall levels of spending, environmental protection and, in particular, social insurance and welfare policies. In the introduction and several subsequent chapters, we identified certain distinguishing characteristics of the state in Southern Europe prior to the mid twentieth century, which seemed to constitute a Southern European syndrome that was in several important respects distinct from state characteristics in the rest of Europe. Among the traditional features of the Southern European state (at least as these existed in the late nineteenth and early twentieth centuries) were: its centralization; its distance from and unresponsiveness to ordinary citizens; its frequent use of police and administrative coercion to ‘maintain order’ (which became much more pronounced during periods of authoritarian rule and even under the semidemocratic regime of Greece between the end of its Civil War [1946–9] and the colonels’ seizure of power in 1967); its ‘social-safety-valve’ function, providing employment (largely along clientelistic and/or patronage lines) to a disproportionately large share of the labor force, especially those in the precarious sectors of the middle-class who were educated but unpropertied; and the relatively low level of professionalism and efficiency of three of these countries’ bureaucracies (with Spain’s system of cuerpos representing a partial exception to this pattern) due, in part, to recruitment based on particularistic criteria (linked to clientelism or other versions of patronage) rather than on universalistic, merit-based criteria. Added to these characteristics under the authoritarian regimes of the early-to-mid twentieth century was the development in

Conclusion

339

Italy, Portugal, and Spain of ‘state corporatism,’ whose most salient features included very substantial parastate enterprises and a tightly regulated labor sector which subordinated workers within state-controlled vertical syndicates. While those corporatist patterns of labor relations denied workers the right to strike and engage in effective collective bargaining (whose net effect was to depress workers’ salaries), institutionalized rigidities in the labor market were also disadvantageous to employers, who lacked the ability to freely dismiss workers or even to reassign them in a flexible manner. And this, in turn, contributed to economic inefficiencies which could persist only within largely autarkic economies that were relatively closed to international trade and competition. Finally, domination of government and the state by individuals drawn disproportionately from the upper socioeconomic strata in Greece, Portugal, and Spain introduced a strongly conservative bias into the policy-making process, the net effects of which were low overall levels of taxation, a heavy reliance on ineffective and regressive taxation systems, and extraordinarily low levels of provision of basic services that were widespread throughout other European countries by the mid to late twentieth century. Particularly underdeveloped, in this regard, were social insurance and social welfare programs. One objective of this volume is to determine if substantial changes in the Southern European state have occurred since the reestablishment of democratic regimes in each of these four countries. As we saw in the first two volumes of the series (Gunther, Diamandouros, and Puhle 1995; Diamandouros and Gunther 2001), socioeconomic modernization and democratization culminated in the complete transformation of politics in all four of these countries. Political change has been so sweeping that no significant vestiges of either the predecessor authoritarian regimes or of more distant political traditions of the type that had polarized and destabilized previous democratic and semidemocratic regimes survived into the democratic period. In Volume 3 (Gibson 2001a), dealing with economic structures and interactions, we observed less far-reaching but still considerable change, as many economic sectors were forced to liberalize and become more efficient in their efforts to adapt to the various demands of European integration. The case studies presented in the previous chapters of this book present us with a very different picture from those that emerged in the earlier volumes of the series. Our survey of various dimensions of the Southern European state revealed mixed patterns of change. On the one hand, we have seen that the overall volume of state activity (measured by expenditures and tax revenues as a percentage of GDP), and of social spending in particular, have increased very substantially in Greece, Portugal, and Spain to such an extent that they now

340

Conclusion

approximate the spending and taxation levels of other European countries (including Italy), which had previously developed advanced social-welfare states. In Spain, we also saw that the highly centralized state structure initially established in the late nineteenth century and made more extreme under the Spanish nationalist government of General Franco has given way to a profound process of political decentralization that has culminated in a quasifederal system granting considerable self-government authority and resources to new regional government institutions. On the other, we have also seen that relatively little change has taken place within the state administrations of these countries, and efforts to reform these bureaucracies have been protracted and incomplete. Similarly, there has been little change in the judicial systems of these countries, although this is less the product of resistance to pressures for change than of the greater adaptability of these systems to different types of political regimes. We have also seen that, even though the social welfare programs of Southern European countries have been more adequately funded than in the authoritarian past, they remain fragmented, highly particularistic, and incapable of providing universal coverage to all individuals in need. And while increased concern over the condition of the environment has led to the adoption and implementation of environmental protection policies, the development of public attitudes and policies regarding the environment lagged well behind those in Northern European countries until the late 1990s. Our efforts to explain patterns of change regarding politics and economics have relied heavily on hypotheses derived from three fundamental processes that have unfolded in Southern Europe over the past several decades—socioeconomic modernization, democratization, and Europeanization. In this book’s introductory chapter, we summarized our conclusions concerning the impact of these processes on politics and economics in the region. In that chapter we also employed themes derived from institutionalist theories and, more specifically, certain concepts derived from the historical analysis of South European institutions, such as authoritarian legacies and the large size and weak capacities of South European states. (We will return to some of these themes below, particularly in our discussion of the democratization effect.) A logical starting point for our efforts to account for changes in the Southern European state, therefore, would be to generate a series of hypotheses growing out of these same three processes, and apply them to the specific state subsystems, relevant patterns of social interaction and public policy outputs that we have explored in this volume. In doing so, we shall see that some aspects of state structure and some specific policy outputs have been substantially affected

Conclusion

341

by one or another of these processes, while others have not. We shall conclude that modernization, democratization, and Europeanization provide only partial explanations for several aspects of these processes of change. In some instances, change can be explained only by modifying the initially straightforward hypotheses related to these processes, and in others, we must introduce new hypotheses involving entirely different kinds of causal factors and processes. Before we turn our attention to these more complex processes of causality, it is perhaps useful to set forth a series of simple hypotheses growing directly out of each of these three clusters of alternative explanatory factors. Evidence in support of one set of causal factors or another should be seen in the extent to which specific changes in the state or public policy conform to these predictions. Perhaps more importantly, the pace and timing of state-structural or policy changes in conjunction with the times when modernization, democratization, and Europeanization occurred in each country should also provide evidence in support of one set of causal factors or another. What would we predict should be the impact of socioeconomic modernization on these state subsystems and public policy outputs? First, we would anticipate that, as economies modernize, there should be an incremental expansion in the provision of social insurance and social welfare programs, as well as in the overall magnitude of government expenditure in taxation. This prediction is derived from numerous studies in the field of comparative public policy that have found a strong correlation between various measures of socioeconomic modernization (particularly as measured by levels of affluence and by the age structure of a given country) and the amount spent on such social programs. While this relationship is not linear (it is quite weak when only advanced postindustrial societies are included in the study), a consistent finding is that a certain minimal level of socioeconomic development and affluence must be achieved before the establishment of an advanced social welfare state is a realistic policy alternative. Insofar as modernization in Southern Europe occurred several decades after economic development had unfolded in other parts of Western Europe, one explanation of the small size of the state and the inadequacy of its social programs is simply that these are reflections of lagging socioeconomic modernization in Southern Europe. If this modernization process were the sole determinant of these public policy outputs, we would expect to find incremental expansion of these policy outputs occurring at roughly the same time as the increase in a country’s level of affluence. And as social programs are progressively developed, so, too, should levels of taxation increase, since revenues would have to be raised in order to pay for these expanded services. Accordingly, the overall magnitude of the

342

Conclusion

state—as measured either by the scope of its programmatic outputs or the share of national income consumed by government taxation—would expand in close conjunction with economic growth. Other consequences of socioeconomic modernization would be mediated through more fundamental alterations in social structure and/or culture. In accord with existing bodies of theory relating to the emergence of ‘postmaterialist’ values, we would expect to find that societal concern over the state of the environment would increase gradually, following the attainment of an advanced state of economic development and widespread affluence. And insofar as one might expect that environmentalist public policies would be adopted by governments in response to demands articulated by the general public, we would expect that these policies would follow such a change in mass culture. The adoption of advanced environmental protection policies, therefore, should lag behind the most rapid stages of economic development, and should expand incrementally, in accord with the gradual growth of postmaterialist values. Finally, modernization should culminate in changes in social structures and patterns of behavior that are indirectly relevant to these public policies. Among the core tenets of modernization theory, for example, are trends toward the displacement of extended family networks by nuclear families, as well as the increasing predominance of universalism over particularism, both with regard to citizen access to beneficial social programs and state agencies, and with regard to the recruitment of individuals to staff the state administration. In terms of pace and timing, we would expect the impact of modernization to be gradual, with incremental expansion of government programs occurring in close conjunction with the most rapid periods of economic development. Accordingly, we would expect that Italian social programs would have expanded most rapidly in the 1950s and 1960s, when Italian economic development was most rapid; and that social program expansion in Spain, Greece, and Portugal would occur in conjunction with the rapid economic growth of the 1960s and early 1970s. The timing implicit in a ‘democratization story’ would be quite different. Since such an interpretation is based upon the notion that fundamental characteristics of the predecessor authoritarian regime (such as its domination by individuals drawn disproportionately from the upper socioeconomic strata) would lead directly to the adoption of certain government policies (such as low and/or regressive taxation), or that its repression of political activity would prevent ordinary citizens from articulating demands for beneficial social policies (such as expanded social insurance and social welfare programs), we would

Conclusion

343

predict that the disappearance of that political regime would be quickly followed by changes in public policy. Accordingly, we would expect that the adoption of policies characteristic of advanced social-welfare states would take place only after the removal of authoritarian constraints and the election of more broadly representative governments. From an analytical standpoint, it is fortuitous that in Southern Europe democratization and the most rapid periods of economic growth in these countries occurred at different times: Democracy was established in Italy in the late 1940s, while that country’s most rapid period of economic growth occurred in the late 1950s and 1960s. Thus, political constraints had been removed prior to the time that modernization began to generate the government revenues necessary to establish an advanced social welfare state. Indeed, at one point, during the decades surveyed in this book, Italy’s GDP per capita exceeded that of Great Britain. In Portugal and Spain, in sharp contrast, authoritarian regimes remained in power until the mid-1970s. Spanish economic growth was most rapid between 1960 and 1973, but stagnated over the following decade, picking up pace again in the late 1980s. Portugal’s economic growth was also impressive between 1960 and 1974, but abruptly halted in response to the worldwide economic recession of the mid-1970s, as well as by the chaos of the revolution of 1974–5. Like Spain, Portugal experienced another spurt of economic growth in the late 1980s. Meanwhile, Greece underwent two different changes of regime: from the semidemocratic system that existed from the end of the civil war until the colonels’ seizure of power in 1967, and then from that short-lived authoritarian regime to a fully democratic system in 1974. Its most rapid period of economic growth occurred in the 1960s, but unlike Spain and Portugal the stagnation of the Greek economy that began in the mid-1970s continued for a full two decades (Gibson 2001a, 11–12). The timing of ‘Europeanization’ effects would be quite different. Italy was a founding member of the EEC in 1958, and was therefore subjected to various pressures from Brussels throughout all of the period examined in this book. Greece joined in 1981, and Portugal and Spain in 1986. There are several ways in which membership in the European Community forced substantial policy changes and structural reforms upon new members during the 1980s: tariff barriers had to be abolished; a VAT had to be adopted; various subsidies to business firms (especially parastate enterprises) had to be eliminated; some state-structural reforms had to be adopted in order to administer EC/ EU regional funds; and in those specific sectors within which the authority of Brussels is absolute (such as with regard to international trade, the Common Agricultural Policy, etc.), member-state autonomy

344

Conclusion

had to give way to ‘harmonization’ with European policy requirements. In the 1990s, the EU commitment to monetary union further required all member states to meet certain ‘convergence criteria’, including such important matters as specific limits on the magnitude of budget deficits and monetary inflation. Accordingly, ‘Europeanization’ effects were most pronounced in Greece, Portugal, and Spain in the 1980s and 1990s. Since this time frame is decidedly different from those relating to both democratization (1940s in Italy, mid-1970s in the other three countries) and modernization (1950s in Italy, 1960s in Greece and Portugal, and 1980s in Portugal), these differences of timing facilitate our efforts to differentiate among the causal impacts of these three sets of causal factors.

Modernization Clear empirical evidence in support of a socioeconomic modernization interpretation was presented in the Pridham/Magone chapter on environmental policy. The late modernization of these four economies (at least in comparison with other Western and Northern European countries) helps to account for the late emergence of popular concerns over the environment, as well as the late adoption and implementation of environmental policies in the south than elsewhere in Western Europe. As Inglehart and others have argued that the development of postmaterialist values—among which concern over the environment is prominent—requires the prior attainment of an advanced state of socioeconomic modernization at least a decade prior to the emergence of a mobilized cohort of defenders of the environment. This is because, in accord with postmaterialist theory, such values only emerge among those who had never suffered from significant economic deprivation. Since Greece, Portugal, and Spain experienced the first substantial waves of rapid economic development in the 1960s, with Italy undergoing its most profound transformation in the 1950s, a simple ‘modernization’ argument would predict that such attitudes would not emerge until the 1980s, and that environmentalism would not become a significant political force until the following decade or later. This is, indeed, the pattern that we observed. Until the late 1990s, the development of an environmental consciousness in Southern Europe lagged significantly behind similar processes in other countries that had entered into an advanced state of late modernity much earlier. Similarly, some support for a modernization interpretation is presented in Castles’ analysis of variations (both cross-national and time series) in levels of social security spending, as well as in Thomadakis’

Conclusion

345

study of the fiscal policies of these countries. In general, both argue that one reason for the fiscal policy and social-insurance policy lags visible in Greece, Portugal, and Spain is that their most extensive processes of socioeconomic growth occurred decades after similar transformations in other West European countries. Castles further demonstrates that intra-regional variation roughly accords with the timing of economic development. Italy, which first experienced economic modernization, had by far the best developed social security system in the region at the beginning of the period under examination, while Spain and Portugal were far behind other West European countries with regard to the development of such programs. Also consistent with a ‘modernization’ interpretation, the age-structure of the populations of these countries emerged from Castles’ time series analysis as a variable of some importance, particularly in Portugal. Both Castles and Thomadakis, however, provide empirical evidence to support the hypothesis that ‘democratization’ effects were much more powerful in their impact on the development of these policies. One intriguing modification of the relationship between modernization and political structures, processes, and outputs that has emerged from some of these studies is that the ‘direction of causality’ linking the two was in certain instances the reverse of that posited in classic modernization theory. In the latter, socioeconomic and related cultural changes occur which have important political implications; in short, social changes precede changes in politics, and these altered socioeconomic conditions introduce pressures of various kinds that contribute to political change. Several chapters included in this volume, however, treat modernization as a policy objective. In order to become modern and efficient, structural or policy changes are introduced in an effort to overcome traditional deficiencies, and the direction of the causal arrow is thus reversed, with policy change preceding modernization. Since many of these kinds of pressures have grown out of membership in the EU, we will defer a full exploration of them until the following section on ‘Europeanization’.

Democratization We also find mixed evidence with regard to the impact of democratization on these state subsystems and public policy outputs. The most dramatic and empirically well-grounded evidence of democratization effects was presented by Castles in his time series analysis of social security programs. While one factor contributing to the underdevelopment of these programs in Spain and Portugal was lagging

346

Conclusion

socioeconomic modernization of those two countries, it is also clear that their authoritarian regimes had repressed popular demands for the development of advanced social programs in those countries. Castles presents powerful and persuasive evidence that the opening up of the political arena made possible substantial change in this policy domain. Moreover, differences among the four countries with regard to the timing and extent of the development of their social security programs are clearly reflective of the timing and basic character of the transition to democracy in each country. Italy was democratized in the 1940s, and therefore this obstacle to the development of advanced social welfare programs was absent. Thus, the rapid socioeconomic development of Italy in the 1950s and early 1960s was accompanied by the establishment and expansion of an elaborate social security system. Indeed, by the time Castles’ time series analysis begins, Italy had assumed a position near the center of the range of those ‘Catholic’ countries within which social policies tend to be most well-developed. In sharp contrast, Spain and Portugal remained under the control of right-wing, authoritarian (and, it should be noted, Catholic) regimes until the mid-1970s, by which time their economies had undergone considerable change, particularly in Spain. Nonetheless, despite the widespread tendency for socioeconomic development and greater affluence to be followed by the emergence and expansion of social policies, they remained far behind the West European average. In both cases, the disappearance of the constraints imposed by these authoritarian regimes was followed by substantial change in the social policies. Even more strikingly, the particular pattern or model of social policy change reflects very accurately the nature and pace of the political transition experienced in each country. Portugal was democratized through a revolutionary process of change that entailed the massive mobilization of large sectors of the working class. Accordingly, the increase in social spending was abrupt (in accord with Castles’ ‘upsurge’ model), and was roughly coterminous with the democratization process itself. Spain, in contrast, democratized through a more gradual, evolutionary process, in which nearly two years elapsed between the death of the authoritarian dictator and the coming to power of the first democratically elected government. Moreover, the processes of political change in Spain were characterized by a high level of administrative continuity with regard to both conformity with established bureaucratic procedures (most importantly, continued adherence to the budgetary process of the Spanish state), as well as the personnel occupying key administrative positions. Accordingly, several years elapsed before the full effects of democratization on these policy outputs could be clearly seen. Indeed, the first ‘democratic budget’ to be formulated following the death of Francisco Franco (in November 1975) was

Conclusion

347

that of fiscal year 1979. Accordingly, the expansion of social spending in Spain was somewhat delayed and incremental. In the case of Greece, during the democratic period between the end of the civil war and the colonels’ seizure of power in 1967 the social security system was more extensively developed than it was in authoritarian Portugal and Spain. Thus, despite the cuts in social programs enacted under the colonels’ regime, there was a less substantial increase in program expansion in the first few years following the restoration of democracy. Subsequently, however, spending on these social programs grew at an impressive rate. Thomadakis’ broader-gauged examination of public policy outputs of these states (as reflected in their respective fiscal policies) comes to a very similar conclusion. Like Castles, he acknowledges that one reason for the smaller size of the public sector in Southern Europe may derive from their lagging socioeconomic development. But his principal conclusion is that the increase in overall levels of spending and taxation were greatly affected by the dismantling of the former authoritarian regimes of Greece, Portugal, and Spain. Democratization in the mid1970s was closely followed by sudden and substantial increases in government expenditure. Taxes, however, were not increased immediately. Thomadakis argues that this is at least partly a reflection of the fact that the first democratic governments in Greece and Spain were relatively conservative, and that this postponement of tax increases fits with their basic programmatic orientations. (It should also be noted that this policy outcome was entirely consistent with the conscious decision of the first generation of democratic political elites, particularly in Spain, to put ‘politics first’, in the interest of securing the consolidation of new and potentially vulnerable regimes, by postponing potentially painful and socially disruptive economic restructuring initiatives until after democratic consolidation had been achieved.) This created substantial budget deficits in each of these new democracies in the first few years following democratization, which posed a serious policy challenge over the long term: the resulting ‘structural deficit’ would eventually have to be addressed, and a process of ‘fiscal consolidation’ would have to be completed following the consolidation of democracy in these countries. A third example of the impact of democratization is that of the profound decentralization of the Spanish state. Having militarily defeated the Basques and Catalans in the civil war, Franco and his victorious Spanish Nationalist forces repressed all manifestations of Basque and Catalan nationalism, and established a rigidly centralized state. Thus, for Basques and Catalans, the process of democratization went hand in hand with demands for the restoration of their regional

348

Conclusion

autonomy. Approval of autonomy statutes for those two regions, in turn, triggered a ‘demonstration effect’, in which many other regions of Spain (which had no traditions of regional self-government) demanded their own regional governments. The net effect was the creation of seventeen Autonomous Communities, encompassing all of the territories of Spain, and the transformation of what had been one of the most rigorously centralized states in Europe into a quasifederal system composed of units having about as much autonomy from the central government as American states.

Institutional Interrelationships and Resources While substantial ‘democratization effects’ could be observed with regard to taxation and social security policies, as well as political decentralization in Spain, such politically induced changes were strikingly absent from other state subsystems and policy domains. Before discussing these in detail, let us set forth two hypotheses that help to account for these differing patterns of change. Both of these involve institutional interrelationships and resources. The first involves the extent to which a state subsystem or public policy was intrinsically linked to the former authoritarian regime. Insofar as it was, then the process of democratization would be accompanied by demands for sweeping change in those policies or state subsystems. This was true of the rigid recentralization of the state under the Franco regime, the establishment and maintenance of an inadequate and regressive system of taxation consistent with the economic interests of the upper socioeconomic strata who were linked to the regime and were overrepresented within its governing and administrative elite, and the repression of or inattention to popular demands for the kinds of progressive social policies that had been adopted by other western industrialized societies. Conversely, public policies or state subsystems that maintained some autonomy from the authoritarian regime or were not perceived as closely linked to it would not be subjected to the same political pressures for change in the course of the democratization process. We shall refer to this as the ‘subsystem autonomy’ hypothesis. An important methodological implication of this hypothesis is the need to clearly distinguish between ‘state’ and ‘regime’: the two are conceptually distinct, and it cannot be assumed prima facie that they have common interests. Indeed, this may be a factor that systematically distinguishes authoritarian from totalitarian regimes. In the latter, the regime’s desire to carry out a radical transformation of society usually requires the complete penetration of the bureaucracy

Conclusion

349

by the regime’s single party, placing civil servants under the direct supervision of political commissars. Thus, a change of political regime almost inevitably requires the uprooting of thousands of political commissars from the interstices of the state administration. By definition (Linz 1975), authoritarian regimes do not require such a close interpenetration between state and regime. Democratization, therefore, does not inherently require a far-reaching purge and restructuring of the bureaucracy, although, as we shall see below, this may vary from case to case. A second ‘institutional’ variable pertains to the organizational or political resources possessed by specific groups within or outside the state structure who have benefited from policies and practices identified with the traditional, preauthoritarian polity or from those instituted or perpetuated under the authoritarian regime. Several chapters in this book have dealt with widespread reforms that might make the state administration more efficient, or with public policies that might have been better adapted to altered social or international circumstances. Many of these efforts have been frustrated, and delayed, or seriously compromised, despite widespread calls for change. One major reason why this has occurred relates to the extent of institutionalization, or to the magnitude of organizational or political resources at the disposal of well entrenched special-interest groups benefiting from the status quo. Accordingly, we can see the role of democratization not just from the perspective of allowing previously marginalized social groups to articulate demands for reform, but also as establishing a framework within which particularistic interest groups may advance their own self-interests and prevail over the preferences of the majority. This perspective on the causal role played by democratic politics is more realistic than facile notions portraying democratization processes as simply entailing the triumph of the interests of ‘the people’ over an entrenched privileged elite. This is because, first, it is unreasonable to assume that the people have one common set of interests vis-a`-vis the formerly dominant economic elites. Various social groups, even within a single social class, have specific, divergent and sometimes incompatible interests. Second, democracy per se is substantively neutral in terms of its policy impact (although, to be sure, the specific manner in which democratic regimes are structured may implicitly or explicitly privilege one set of interests over others). The establishment of a democratic regime simply opens up the policy process to contestation among the wider array of social groups, and the outcome of those struggles is not predetermined by the mere existence of democracy. These two hypotheses account reasonably well for the mixed patterns of change and sometimes frustrated efforts at reform that

350

Conclusion

we have seen in the preceding chapters. The analysis of the Southern European bureaucracies by Sotiropoulos, for example, reveals that democratization processes did not, except in Portugal, entail substantial change of the state administration over the short term. This is so for several reasons. First, except in Portugal, these authoritarian regimes did not deeply penetrate into the interstices of the civil state administration for the purpose of resocializing its officials or directly subordinating them to control by the regime’s political commissars. Instead, the authoritarian regimes by-and-large left these state subsystems alone: patterns of recruitment, training and promotion remained intact from the preauthoritarian era, and the state bureaucracy retained a certain organizational autonomy from the regime, especially at the middle and lower ranks of the civil service. The most clear-cut examples of this were the colonels’ regime in Greece (which disdained and largely ignored the state bureaucracy, except for its purges of judges and other top officials) and Spain, where, by the final decade of the Franco regime the Falange (renamed Movimiento Nacional in 1958) had ceased to play a significant role in elite recruitment with regard to most ministries, and the state administration had shifted to the recruitment of apolitical technocrats to occupy the commanding heights of the bureaucracy. Only Portugal departed markedly from this pattern, leading to two minor waves of purges, first, of rightwing officials under the revolutionary junta, then of extreme left-wing individuals following the 1975 countercoup by the more moderate General Ramahlo Eanes. The partial exception of Portugal can be explained by the fact that both Salazar and Caetano (professors of Public Finance, and Public Law and Administration, respectively) were professionally interested in and more involved with the state administration than were their Greek, Italian, or Spanish counterparts, and also used segments of the bureaucracy to repress civil and political liberties. Second, the state administrations of Greece, Italy, and Spain did not oppose or impede the democratization process in any significant way, and thus did not have to be reformed or replaced as an integral part of regime transition. Despite their capacity to make trouble for the newly emerging democratic elites (particularly important in the case of the military), they chose to acquiesce to the political instructions issued by their democratically elected superiors. Indeed, many high-ranking Spanish bureaucrats eventually played key roles in dismantling the Franco regime, and a survey of political attitudes conducted in the early 1980s found that high-ranking bureaucrats were more strongly supportive of democracy than the Spanish public as a whole. In Portugal, on the other hand, where, for reasons just stated, the regime had

Conclusion

351

penetrated the bureaucracy much more deeply, the purging of the bureaucratic apparatus in the postauthoritarian period was much more pronounced. Third, key sectors of the state administration possessed institutional resources that enabled them to resist pressures for reform. In the case of Greece and Italy, where partisan and clientelistic patronage appointments gave civil servants ties to powerful governing elites, reform efforts ran headlong into the particularistic interests of the governing parties. In Spain, the cuerpos parlayed the expert technical training they imparted to their members into an extraordinary ability to control appointments to key bureaucratic positions. While their greater expertise led Sotiropoulos to characterize Spanish bureaucrats as having a higher level of professionalism than their Greek, Italian, and Portuguese counterparts, these institutionalized cuerpo linkages also undermined political control over the State Administration by democratically accountable governments, particularly with regard to certain spheres of public policy: Sofı´a Pe´rez (1997), for example, argues that a remarkably durable group of Finance ministry officials who were closely linked to large banks were able to successfully defend their particularistic interests (e.g. regarding regulation of the banking sector and monetary policy) under both authoritarian and democratic regimes, and under both UCD and Socialist governments. In the Magalha˜es, Guarnieri, and Kaminis study of the Southern European judiciaries, democratization also failed to bring about pressures for radical change. Again, the nature of the relationship between this state subsystem and the previous authoritarian regimes provides a compelling reason why it did not. Unlike the Communist Eastern European regimes (which had purged judicial institutions and appointed judges and court officials who would collaborate with and advance the repressive policies of the Party), when the Southern European authoritarian regimes came into being they largely left the preexisting judicial structures intact. This was for two reasons. First, for the purposes of political repression they created special court systems, completely outside the traditional judicial hierarchy. The mainstream judiciary was thus left to deal with cases that were not directly linked to the authoritarian character of the regime. Second, the civil law traditions that were so strongly embedded in the Southern European countries were rooted in a legal positivism that regarded the state as the sole source of law and rights, and cast judiciaries in the role of executors of laws rather than interpreters of legal principles, as is customary in the common law tradition (see Alivizatos and Diamandouros 1997). In that sense, judges in the civil law tradition are regarded as civil servants who do not question the validity or meaning

352

Conclusion

of the law. Thus excluded from adjudicating the more politicized cases, and limiting their activities to the application rather than interpretation of the law, these traditional judiciaries were relatively insulated from the politics of the authoritarian regime, and were allowed to evolve in accord with their long-standing internal norms, procedures, and recruitment patterns. Since mainstream judicial institutions had not been politicized by the authoritarian regimes, moreover, a purge or restructuring of these judiciaries was not an essential component of the democratization process. With regard to adjudication of civil and criminal law, all that was necessary over the short-term was abolition of the special extrajudicial institutions created for the purpose of meeting out political repression. Over the longer term, however, demands were articulated for reforms of these highly traditional judicial systems, toward the objective of making them more efficient and publicly accountable. (To some extent, it should be noted that these demands may have been stimulated by socioeconomic modernization and democratization processes, both of which have generally led to an increase in the number of lawsuits filed by citizens.) In the cases of Spain and Portugal, democratization had a much more immediate impact on judiciaries in another way: as part of the process of institutionalizing new democratic regimes, it was regarded as necessary to establish constitutional courts and invest them with powers of abstract judicial review. In Portugal, this took the form of the Council of the Revolution—a nondemocratic vestige of the military junta that had handed over power to civilians under the terms of the 1976 Constitution. This body, made up of high-ranking military officers, had the authority to declare legislation unconstitutional, thereby overruling the decisions of duly elected governments. Since this represented the retention of ‘reserve powers’ by the outgoing military elite, its existence technically violated our definition of democracy, and it could have posed a serious threat to consolidation of the new regime. Instead, it wisely chose to intervene in contentious political matters sparingly, and generally deferred to an ‘advisory’ Constitutional Committee (composed of professional jurists). This latter body eventually assumed responsibility for abstract review of legislation following abolition of the CR in 1982. But while elimination of the CR did represent an essential component of the transition to democracy, the paradoxical outcome of this unusual phase of the democratization process in Portugal was to establish the precedent of generally nonpartisan abstract judicial review. In Spain, a special Constitutional Court was established by the Constitution of 1978 and charged with the broad interpretation of constitutional principles, the deepening of the rule of law, and the resolution of disputes among various actors and institutions of the

Conclusion

353

new democratic systems. As in the case of the special courts set up under the authoritarian regimes, establishment of this Constitutional tribunal (appointment to which was explicitly representative of the major political parties) marginalized the mainstream judiciary from involvement with contentious political issues, some of which were tantamount to ‘completing the transition to democracy’. This was particularly important, since a very conservative judiciary had been left behind by the Franco regime. Effectively insulated from the more politically sensitive issues inherent in the transition to democracy (in particular, regulating relations between the central