Citizenship, Labour Markets and Democratization: Chile and the Modern Sequence (St. Antony's)

Citizenship, Labour Markets and Democratization Chile and the Modern Sequence

Louise Haagh

St Antony’s Series General Editor: Richard Clogg (1999– ), Fellow of St Antony’s College, Oxford Recent titles include: Louise Haagh CITIZENSHIP, LABOUR MARKETS AND DEMOCRATIZATION Chile and the Modern Sequence Renato Colistete LABOUR RELATIONS AND INDUSTRIAL PERFORMANCE IN BRAZIL Greater São Paulo, 1945–1960 Peter Lienhardt (edited by Ahmed Al-Shahi) SHAIKHDOMS OF EASTERN ARABIA John Crabtree and Laurence Whitehead (editors) TOWARDS DEMOCRATIC VIABILITY The Bolivian Experience Steve Tsang (editor) JUDICIAL INDEPENDENCE AND THE RULE OF LAW IN HONG KONG Karen Jochelson THE COLOUR OF DISEASE Syphilis and Racism in South Africa, 1880–1950 Julio Crespo MacLennan SPAIN AND THE PROCESS OF EUROPEAN INTEGRATION, 1957–85 Enrique Cárdenas, José Antonio Ocampo and Rosemary Thorp (editors) AN ECONOMIC HISTORY OF TWENTIETH-CENTURY LATIN AMERICA Volume 1: The Export Age Volume 2: Latin America in the 1930s Volume 3: Industrialization and the State in Latin America Jennifer G. Mathers THE RUSSIAN NUCLEAR SHIELD FROM STALIN TO YELTSIN Marta Dyczok THE GRAND ALLIANCE AND UKRAINIAN REFUGEES Mark Brzezinski THE STRUGGLE FOR CONSTITUTIONALISM IN POLAND Suke Wolton LORD HAILEY, THE COLONIAL OFFICE AND THE POLITICS OF RACE AND EMPIRE IN THE SECOND WORLD WAR The Loss of White Prestige Junko Tomaru THE POSTWAR RAPPROCHEMENT OF MALAYA AND JAPAN, 1945–61 The Roles of Britain and Japan in South-East Asia

Eiichi Motono CONFLICT AND COOPERATION IN SINO-BRITISH BUSINESS, 1860–1911 The Impact of the Pro-British Commercial Network in Shanghai Nikolas K. Gvosdev IMPERIAL POLICIES AND PERSPECTIVES TOWARDS GEORGIA, 1760–1819 Bernardo Kosacoff CORPORATE STRATEGIES UNDER STRUCTURAL ADJUSTMENT IN ARGENTINA Responses by Industrial Firms to a New Set of Uncertainties Ray Takeyh THE ORIGINS OF THE EISENHOWER DOCTRINE The US, Britain and Nasser’s Egypt, 1953–57 Derek Hopwood (editor) ARAB NATION, ARAB NATIONALISM Judith Clifton THE POLITICS OF TELECOMMUNICATIONS IN MEXICO Privatization and State–Labour Relations, 1928–95 Cécile Laborde PLURALIST THOUGHT AND THE STATE IN BRITAIN AND FRANCE, 1900–25 Craig Brandist and Galin Tihanov (editors) MATERIALIZING BAKHTIN C. S. Nicholls THE HISTORY OF ST ANTONY’S COLLEGE, OXFORD, 1950–2000 Anthony Kirk-Greene BRITAIN’S IMPERIAL ADMINISTRATORS, 1858–1966 Laila Parsons THE DRUZE BETWEEN PALESTINE AND ISRAEL, 1947–49

St Antony’s Series Series Standing Order ISBN 0–333–71109–2 (outside North America only) You can receive future titles in this series as they are published by placing a standing order. Please contact your bookseller or, in case of difficulty, write to us at the address below with your name and address, the title of the series and the ISBN quoted above. Customer Services Department, Macmillan Distribution Ltd, Houndmills, Basingstoke, Hampshire RG21 6XS, England

Citizenship, Labour Markets and Democratization Chile and the Modern Sequence Louise Haagh British Academy Research Fellow St Antony’s College Oxford University

in association with St Antony’s College, Oxford

© Louise Haagh 2002 All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No paragraph of this publication may be reproduced, copied or transmitted save with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, 90 Tottenham Court Road, London W1T 4LP. Any person who does any unauthorised act in relation to this publication may be liable to criminal prosecution and civil claims for damages. The author has asserted her right to be identified as the author of this work in accordance with the Copyright, Designs and Patents Act 1988. First published 2002 by PALGRAVE Houndmills, Basingstoke, Hampshire RG21 6XS and 175 Fifth Avenue, New York, N. Y. 10010 Companies and representatives throughout the world PALGRAVE is the new global academic imprint of St. Martin’s Press LLC Scholarly and Reference Division and Palgrave Publishers Ltd (formerly Macmillan Press Ltd). ISBN 0–333–80385–X This book is printed on paper suitable for recycling and made from fully managed and sustained forest sources. A catalogue record for this book is available from the British Library. Library of Congress Cataloging-in-Publication Data Haagh, Louise, 1967– Citizenship, labour markets and democratization : Chile and the modern sequence / Louise Haagh. p. cm. — (St Antony’s series) Includes bibliographical references and index. ISBN 0–333–80385–X 1. Democratization—Chile. 2. Industrial relations—Chile. 3. Chile—Politics and government—1988– I. Title. II. Series. JL2681 .H32 2001 322’.2’0983—dc21 00–069603 10 11

9 10

8 09

7 08

6 07

5 06

4 05

3 04

Printed and bound in Great Britain by Antony Rowe Ltd, Chippenham, Wiltshire

2 03

1 02

In memory of my mother Birte Haagh and for Ato and Alan

This page intentionally left blank

Contents

x

List of Tables and Figures

xv

Acknowledgements List of Abbreviations and Spanish terms Introduction Political and social democratization Social citizenship and development Political transitions and citizenship: the Chilean case

xvii xxiii xxiii xxiv xxvi

Part I Theoretical Issues 1

Citizenship and Democratization Bases of occupational citizenship Occupational citizenship and development Marshall on citizenship reconsidered The development of social citizenship Conclusion

3 3 5 7 14 22

2

Human Resources and Market Reforms Labour market flexibilization: the core controversies Labour freedom and the development of human resources Market reforms and labour institutions in Chile (1973–90) Conclusion

23 23

Part II

26 32 42

Labour Politics and Chile’s Transition

3

Business Leaders and Democracy’s Limits Background The CPC in transition politics Conclusion

47 48 54 60

4

Unions, Parties and the Sacrifice for Democracy Unions before and during Pinochet Politicians and bureaucrats: the labour leaders and their advisers

62 64

vii

71

viii Contents

Accommodation of the opposition parties Conclusion 5

The Transition to Democracy and the Enforcement of Markets Managerial autonomy and pacts of legitimation Non-intervention and transition politics The question of sectoral pact-making The re-reform of the Labour Code: labour issues under Frei (1993–2001) Conclusion

Part III 6

74 75

77 78 79 83 90 93

Social Citizenship and Chile’s Labour Reforms

Democratic Legitimation and the New Labour Code That Never Was Labour flexibility in a comparative perspective The politics and development discourse of labour legislation The legal reform under scrutiny Conclusion

99 104 121

7

Training Policy and Decentralization Chile and the liberal model The training policies of the 1990s The politics of an unemployment insurance regime Conclusion

123 124 128 138 145

8

Labour Relations and Investment in Workers Labour relations and human resources Forms of labour flexibility and democratization in industrial firms Inter-firm flexibility and the organization of work The contexts of workers’ training Conclusion

147 147

Union Strength, History and Effects Local unions and labour flexibility Definitions of union strength Local labour leaders and the flexibility debate

178 179 181 194

9

97 98

152 162 168 173

Contents

The traditional industrial relations institutions Conclusion

ix

199 205

Conclusion Social and political democratization Labour institutions and economic development

209 209 211

Appendix 1:

Tables to Accompany Chapter 1

216

Appendix A:

Notes on Survey Sampling

222

Appendix B:

Statistics to Accompany Chapters 8 and 9

226

Appendix C:

Notes on Regression Tables

231

Appendix D:

Notes on Loglinear Tables

234

Notes

236

Select Bibliography

264

List of Interviews

277

Index

281

List of Tables and Figures

Tables 2.1 Wages in costs per sector, 1976–89, and investment as a percentage of value added in enterprises with over 50 workers 2.2 GDP, labour and capital inputs and sources of growth in output per worker; Chile compared with average of four regions, 1950–89 3.1 Annual budget of the SOFOFA, 1991 (pesos 1992) 4.1 Number of unions in the enterprise, 1992 4.2 How the economic crisis of the 1980s affected the respondents’ firms 4.3 The most important measures taken in the firm to overcome the crisis, in the estimation of the union leader 4.4 Was there a relationship of co-operation between workers and the employer during the 1980s? 4.5 How far did co-operation with employers during the 1980s benefit workers? 4.6 The most important goal for the local union 4.7 The most important goal for the union movement in the current environment 4.8 Emphasis on training as a goal for the union movement crossed with degree of occupational security 6.1 Aggregate figures for unionization and employment 6.2 Percentage of firms where unfounded dismissals tend to be referred to tribunals 6.3 Comparison of Article 155.f of the 1979 Labour Code with Article 3 of Law No. 19.010, as percentage of total separations 6.4 Should there exist (by law) only one union and bargaining group in a firm? 6.5 In which of the following ways did the signing of the last convention take place? 6.6 How was the negotiating group which signed the last convention formed? x

35

35 54 65 66 67 68 68 69 70 71 100 105

106 106 109 110

List of Tables and Figures

6.7 6.8 6.9 6.10 6.11 6.12 6.13 7.1 7.2 7.3 7.4 7.5 7.6 7.7

7.8 8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8 8.9

Comparison of a convention and proposals for a collective contract in a metallurgical firm, 1991 Collective bargaining by type of contract Unions versus bargaining groups in collective bargaining, selected years Bargaining groups (as opposed to unions) by size of bargaining unit, 1998 Mode of agreement on wages by type of bargaining group, 1998 Aggregate strike statistics Strike data by size of bargaining unit, 1990 and 1992 Percentage distribution of participants and spending on different SENCE programmes, selected years Percentage distribution of participation through the TBS, by occupation, selected years Percentage distribution of participants by area, 1988–96 Training programmes carried out by, and in association with, SENCE, 1992 Typology of course content, third bid (licitacíon) of the Youth Training Scheme (YTS), 1992 Representation of dynamic export sectors in the Youth Training Scheme (YTS), 1992 Percentage of courses in the YTS programme compatible with demand from small and medium-sized enterprises, by region Comparison of the proposals for an Unemployment Insurance Regime Elements of the organization of work and labour relations in four textile and clothing firms, 1992 Formalization crossed with precarious flexibility Formalization crossed with task diversity Levels of labour turnover, figures from INE compared with figures from the survey of local labour leaders Labour turnover crossed with level of formalization Labour turnover crossed with labour flexibility and productivity Cross tabulation of training and productivity by levels of turnover Labour turnover: analysis of variance (excluding size of firm) Labour turnover: analysis of variance (parsimonious model)

xi

111 114 115 116 117 119 120 129 130 131 132 133 135

136 139 156 158 160 162 163 164 165 166 167

xii List of Tables and Figures

8.10 Workforce training crossed with basic characteristics of firms 8.11 Pearson correlations: workforce fluctuations crossed with investments and types of investment 8.12 Workforce training crossed with indicators of workers representation 8.13 Logistic regression: determinants of workforce training 8.14 Influence of turnover on relationships between training and workers’ representation in the firm 8.15 Influence of wage system (arbitrary versus formal) on the relationship between turnover and training 9.1 Unionization levels in firms, various indicators 9.2 Odds that union leaders with contracts find that labour relations are bad 9.3 Age of union crossed with indicators of formalization of labour management 9.4 Age of union crossed with labour turnover and training 9.5 Age of union crossed with union leadership and level of workers’ integration 9.6 Odds that training is high (depending on union age and type of bargaining instrument) 9.7 Odds that labour turnover is high (rather than medium or low), depending on union age and type of bargaining instrument 9.8 Opinions on wage variability crossed with those on occupational security 9.9 Odds that a high level of training coincides with the view that incentive wages can be a source of long-term participation in the firm 9.10 Employers’ response to union’s request for dialogue on wage flexibility 9.11 Availability of information for collective bargaining crossed with existence of Workers’ Health and Safety Committee 9.12 Existence of Workers’ Health and Safety Committee (WHSC) and availability of information for collective bargaining in the scenarios of low and high levels of workers’ training 9.13 Existence of Workers’ Health and Safety Committee (WHSC) and availability of information in the scenarios of low and high levels of labour turnover

168 169 170 172 174 175 182 184 186 187 189 190

192 197

198 200

201

202

202

List of Tables and Figures

9.14 Union leaders’ support for making obligatory the response to union petitions for bargaining above the firm-level 9.15 Odds that influence from the federation (advice on bargaining) coincides with union opinions regarding workers’ training

xiii

204

206

Figures 1.1 3.1 3.2 8.1

Three dimensions of the development of citizens’ occupational rights Industrial employers within the organization of Chilean business ‘Influential entrepreneurs’ in Chile on the question of a national dialogue with union organizations Dynamics in local labour relations in Chilean industrial firms

20 50 56 151

Appendix tables A.1.1 Some trade-offs associated with the development of social citizenship under different forms of national governance A.1.2 Values used for box-plot in Figure 1.1 (Chapter 1) A.1 Details related to universe on which sampling is based A.2 Importance of industrial sectors in sample universe, by unionized industrial employment, industrial employment and value added A.3 Survey sample and weighting formula A.4 Levels of workers’ training by sector (according to level of technological development) A.5 Levels of turnover by sector (according to level of technological development) B.1 (To accompany Table 8.7) Significance of the reductions of the original model when progressively controlling for the least significant factors B.2 (To accompany Table 8.12) Significance of the reductions of the original model when progressively controlling for the least significant factors B.3 Logistic regression to explain high levels of workforce training B.4 Logistic regression to explain high levels of workforce training, parsimonious model

216 220 222

223 224 225 225

226

227 227 228

xiv List of Tables and Figures

B.5 (To accompany Table B.6) Significance of the reductions of the original model when progressively controlling for the least significant factors B.6 (To accompany Table 8.7) High technology sectors and low technology sectors, levels of turnover by formalization of labour management B.7 Levels of training by existence of Workers’ Health and Safety Committees, small and medium-sized firms, with up to 199 workers, and large firms with 200+ workers

229

229

230

Acknowledgements A large group of people deserve my gratitude for having made this book possible. Above all I am indebted to Alan Angell, my academic supervisor at St Antony’s College, for his scholarly inspiration and dedication throughout my D.Phil. years and during the tenure of my British Academy Postdoctoral Fellowship at St Antony’s College. I also owe a special debt of gratitude to Samuel Valenzuela for his invaluable advice and support since acting as examiner of the D.Phil. on which the present volume is based. Two institutions of research funding provided financial backing for my research and writing. The Economic and Social Research Council of the United Kingdom funded my D.Phil. and the award of a fellowship from The British Academy gave me the opportunity to revisit this text. During my research period in Chile I was lucky to encounter a truly vast number of people who gave unselfishly of their support and time. Among them are those who subjected themselves to my interviews and inquiries, and others who gave me companionship and insight into the world of Chilean politics and labour relations. In the second group I owe special thanks to my colleagues at the Centro de Estudios Sociales (CES), where I was fortunate in having an academic base throughout my stay. Manuel Barrera, María-Angelica Ibañez, Agustín Quevedo, Humberto Arco and many others whom I encountered there through CES’s engagement with the trade union world, ensured that I soon became familiar with the concerns of the ‘floor’ of Chilean unionism. It was this rich experience that convinced me of the relevance and feasibility of doing a survey of local union leaders, and to investigate further the relationship between social relations and labour management in Chile. In carrying out the practical aspects of this task I relied on the professionalism, speed and experience of the Centro de Estudios de la Realidad Contemporánea (CERC). Luísa Andrade and Ema Lagos in particular, deserve very special thanks for their expertise and good spirits. In helping me to organize the interviews with union leaders in many parts of the extended and difficult Chilean terrain, and later in helping to tabulate the data, they were truly invaluable. Without their help, the survey of union leaders would not have been possible. I am also deeply indebted to Volker Frank, for introducing me to CERC. Among the local union leaders who helped to prepare the detailed case studies of local unions, I particularly xv

xvi Acknowledgements

need to thank Maurício Navarro, Gloria Andrade, Jaime Gonzales, José San Martín, Manuel Osorio, Elba López and Juan Pizarro. Among the many other people who agreed to give their time for interview there are some deserving of a special mention here. Guillermo Campero of the Labour Ministry, Eugenio Díaz of the CUT’s advisory, CIASI, Miguel Soto of the metallurgical workers’ confederation CONSTRAMET, Raúl Requena of the CUT, and Eugenio Heiremans of the insurance holding AChS all spent many long hours answering my questions. Each provided freely unsolicited information, much insight and introductions to further contacts within their sector. Equally, I need to thank the many people who several times turned the task of collecting statistics into enjoyable occasions, especially Jaime Carríl, Head of Studies of the Labour Inspectorate, and Vuscovic Huerta, Director of Economic Statistics at the National Statistics Institute. Both gave me insights into the ‘politics’ of collecting statistics, and both were enthusiastic about pursuing new ideas and methods. On my return from Chile I received help from many people in my statistical work. Among them I especially need to mention Mario Cortina of the Department of Statistics in Oxford, Clive Payne of the Social Studies Faculty, Paul Griffiths of the Statistics Institute at Oxford, and Professor Bell of Stirling University. Gabriel Palma generously read my whole manuscript at an early stage, kindly gave of his time, and provided me with many valuable suggestions throughout. Finally, there were many people who provided me with companionship along the way. Among them Teresa and Paula Hammond deserve my warmest thanks. Elmer Cuba, Nelly and Volker Frank, Manuel Barrera, Adnan Nawaz, Maria-Angelica Ibañez and Edward and Tessa Epstein contributed in special ways to making the research period enjoyable. Gwyneth Doherty-Sneddon, Roberta Middleton and Stephanie Mitchell all gave invaluable moral support later on. During the years of writing I spent most of my time at home with my son Ato. He deserves my greatest thanks, for his company, patience and loving ways. LOUISE HAAGH

List of Abbreviations and Spanish Terms

AA AALC

Aprendizaje Alternado (Apprenticeship Programme) Asociación de Abogados Laboralistas de Chile (Association of Labour Lawyers of Chile AB Asociación Bancaria (Bankers’ Association) Acuerdo Marco Framework agreement – name given to the yearly tripartite national social accords AChS Asociación Chilena de Seguridad (Chilean Accident Compensation Association) AFP Administrador de Fondos de Pensiones (Private Pension Funds) ASIMET Asociación de Industriales Metalúrgicos (Metallurgical Employers’ Association) Caja de Compensación Insurance policy institution Cámara ChilenoAlemana de Comercio e Chilean–German Chamber of Commerce and Industry Industria CED Centro de Estudios del Desarrollo (Centre for Development Studies) CEP Centro de Estudios Públicos (Centre for Public Studies) CERC Centro de Estudios de la Realidad Contemporánea (Centre for the Study of Contemporary Reality) CES Centro de Estudios Sociales (Centre for Social Studies) CIASI Centro de Investigación y Asesoría Sindical (Union Research and Advice Centre) CIEPLAN Corporación de Investigaciónes Económicas para América Latina (Corporation for Economic Research on Latin America) CINTERFOR El Centro Interamericano de Investigación y Documentación sobre Formación Profesional (The Interamerican Centre for Documentation and Research on Professional Training) CGCDPIC Confederacion Gremial del Comerio Detallista y de la Pequeña Indústria de Chile (The Confederation of the Retail Trade and Small Industry of Chile) xvii

xviii List of Abbreviations and Spanish Terms

CMPC CNC CNS CNT CODELCO CONAPAN Concertación Consejo de Diálogo Social CSNTT FENTEMACONSFETEMA

Compañia Manufacturera de Papeles y Cartones (The Paper Company) Cámara Nacional de Comercio (National Chamber of Commerce) Confederación Nacional Sindical (National Confederation of Unions) Central Nacional de Trabajadores (National Labour Central) Corporación Nacional del Cobre (The National Copper Company) Confederación de Trabajadores del Pan (The Confederation of Bakery Workers) Opposition alliance founded in 1987 to contest the plebiscite of 1988 and the national elections of 1989 Council for Social Dialogue Consejo Superior Nacional del Transporte Terrestre (National Council of Over-Land Transport)

Confederación de Trabajadores Electrometalúrgicos (Confederation of Electro-metallurgical Workers CONSTRAMET Confederación de Sindicatos de Trabajadores Metalúrgicos y Metalmecánicos (Mechanical and Metallurgical Workers’ Confederation) CONTEVECH Confederación de Trabajadores del Textil y Vestuario de Chile (Confederation of Textile and Clothing Workers of Chile) CONTEXTIL Confederación de Trabajadores del Textil (Confederation of Textile Workers) CONUPIA Confederación Nacional Unida de la Mediana y Pequeña Industria, Servicios y Artesanado (National Confederation of Small and Medium-sized Industries and Craftsmen) CORCAPLAM Corporación de Capacitación Ocupacional y Desarrollo Laboral de la Industria Metalúrgica (Corporation for Occupational Training and Labour Development of the Metallurgical Industry) CORDEMET Corporación Deportiva y Cultural de ASIMET (The Sports and Cultural Corporation of ASIMET) CORDETECMET Corporación de Desarrollo Tecnológico (Corporation for Technological Development)

List of Abbreviations and Spanish Terms

COREDUCMET CORFO CORMA CPA CPC CPTC CSNTT CTB CTDRH

CTF CUOC CTC CUT Dirección del Trabajo EDF EEC ENADE EIS FCJ FECHIPAN FESIMA

xix

Corporación Educacional de ASIMET (Education Corporation of ASIMET) Corporación de Fomento de la Producción (Chilean Development Corporation) Corporación de la Madera (The Forestry Producers’ Association) Confederación de Productores Agrícolas (Confederation of Agricultural Producers) Confederación de la Producción y el Comercio (Confederation of Production and Commerce) Consejo de la Producción, el Transporte y el Comercio (Council of Production, Transport and Commerce) Consejo Superior Nacional del Transporte Terrestre (National Transport Council) Confederación de Trabajadores Bancarios (Bank Workers’ Federation) Centro Tecnológico para el Desarrollo de Recursos Humanos (Technological Centre for the Development of Human Resources) Confederación de Trabajadores Forestales (Confederation of Forestry Workers) Confederación Unidad Obrero-Campesino (Unitary Confederation of Rural Workers) Confederación de Trabajadores del Cobre (Confederation of Copper Workers) Central Unitaria de Trabajadores (United Workers’ Central) The National Labour Inspectorate El Diario Financiero (The Financial Times) European Economic Community Encuentro Nacional de la Empresa (National Meeting of the Business Sector) Employment Insurance Scheme (in Korea) Formación y Capacitación de Jovenes (Training and Education Programme for Youths) Federación Chilena de Panaderos (National Federation of Bakers) Confederación Sindical de Maipú (Union Confederation of Maipú)

xx List of Abbreviations and Spanish Terms

FONDECYT Foro del Desarrollo Productivo Historia de la ley Højskole

IADB IC ICARE

ILO INACAP

INE INSALCO IPAC ISAPRES ISI KOMA MIDEPLAN MITI MOC NGO NTI

Fondo Nacional de Desarrollo Científico y Tecnológico (The National Science and Technology Fund) Forum for Productive Development

History of the Law (Documents covering the parliamentary proceedings of a legislative project) Literal translation from Danish reads ‘Highschool’. The phenomenon of the Højskole refers to the nationwide network of schools of adult education founded by Grundtig in the 19th century with the purpose of providing working people with the chance to gain a wider spiritual and intellectual stimulus and consciousness of civic values. Inter-American Development Bank Investment coordination Instituto Chileno de Administración Racional de Empresas (Chilean Institute for the Rational Administration of Firms) International Labour Organisation Instituto Profesional y Centro de Formación Técnica – ex Instituto Nacional de Capacitación Profesional (Institute for Professional Training and Centre for Technical Education 1976 – ex-National Institute of Professional Training 1966–76) Instituto Nacional de Estadísticas (National Institute of Statistics) Instituto Superior Alemán de Comercio (Higher German Chamber of Commerce) Indústria Procesadora de Acero S.A. (Steel Processing Industry) Instituciones de Salud Previsional (Institutions of Health Insurance) Import-Substitution Industrialization Korea Manpower Agency Ministerio de Planificación y Cooperación (Ministry of Planning and Co-operation) The Ministry of Trade and Industry of Japan Movimiento Obrero Campesino (The Peasant and Workers’ Movement) Non-governmental Organization Network-thickening initiatives

List of Abbreviations and Spanish Terms

Nuevo Código del Trabajo OECD OTE OTIR Papelera PC PDC PEM POJH PPD PR PREALC

PROTRAC PS PTC REFA RN SASK SENCE SERCOTEC SITC SM SNA SOFOFA SONAMI TD UDI UIR

xxi

New Labour Code Organisation for Economic Cooperation and Development Organismo Técnico de Ejecución (Institutions that offer professional training, and bid for public funds); see also PTC Organismo Técnico Intermedio (Association of firms formed for the purposes of pooling resources for training) See CMPC Partido Comunista (Communist Party) Partido Demócrata Cristiano (Christian Democratic Party) Programa de Empleo Mínimo (the Programme of Minimum Employment) Programa de Ocupación para Jefes de Hogar (The Occupation Programme for Heads of Households) Partido por la Democracia (The Party for Democracy) Partido Radical (Radical Party) Programa Regional del Empleo en América Latina y el Caribe (Regional employment programme for Latin America and the Caribbean – under the World Employment Programme) Protección al Trabajador Cesante (Programme for the Protection of Unemployed Workers) Partido Socialista (Socialist Party) Private Training Corporation Verband für Arbeitsstudien (Corporation for Labour Studies) Renovación Nacional (National Renovation) Suomin Ammattiliihojen Solidaarisuuskeskus (The Trade Union Solidarity Centre of Finland) Servício Nacional de Capacitación y Empleo (National Training and Employment Services) Servicio de Cooperación Técnica (Technical Co-operation Services – Economic Ministry of Chile) Standard International Trade Classification Small and Medium-sized (Firms) Sociedad Nacional de Agricultura (Agriculturalists’ Association) Sociedad de Fomento Fabril (The Industrialists’ Association) Sociedad Nacional de Minería (National Mining Association) Task Diversity Unión Democrática Independiente (Independent Democratic Union) Unemployment Insurance Regime

xxii List of Abbreviations and Spanish Terms

USEC Vicaría WHSC WPP YTS

Unión Social de Empresarios Cristianos Vicariate Comité Paritario (Workers’ Health and Safety Committee) Workers’ Preference Principle ‘Programa de Jovenes’ (Youth Training Scheme)

Introduction Political and social democratization The period of government that reinstated political democracy in Chile (the years 1990–93) had a special significance. It brought to the fore the difficulties faced by many countries in combining the expansion of citizenship with the opening of labour markets during the preceding decade and a half. Before 1973, Chile was characterized by a long tradition of democratic government, a fairly sophisticated state bureaucracy, a highly democratic labour movement, and an extended system of welfare provision. Later, the country witnessed the most thorough modern attempt at expanding markets. Not surprisingly, by the late 1980s concern was being expressed about the need to extend social citizenship once more following the radical tone of the market reforms. In fact, having achieved economic, and later political, liberalization, Chile appeared to be in the best possible position to recover citizenship in the social domain. The ways in which this expectation was posed, and the reasons why it was not fulfilled, carry lessons that are beyond the narrow confines of the Chilean case. This book then is about social citizenship and political democratization in Chile. However, its scope is beyond the Chilean case. The book also assesses the relationship between political and social democratization in general terms. Moreover, it discusses the importance of the sets of institutions that promote what I define as occupational citizenship. In the broadest terms, occupational citizenship refers to the existence of effective means of sustaining and developing a productive existence. As other forms of citizenship it has to be measured in terms of type and degree. The philosophical and social bases behind the concept are hardly new, yet changes in labour markets have added to its complexity and meaning. More than that, the need to assess social citizenship issues is re-emerging, having been submerged by more pressing matters of economic adjustment and state reforms for the last three decades. Such concerns directed the democratization literature through identifiable stages: the first being focused on the process of the breakdown of democracy in Latin America and parts of Europe; the second on problems of the transition to electoral democracy; and the most recent wave surveying the issues of the consolidation of democratic regimes. 1 The xxiii

xxiv Introduction

writings listed remained essentially concerned with political democracy in the minimal sense defined by Dahl. 2 This delimitation of focus was understandable, because achieving political democracy was naturally seen as a precondition for having a debate about any kind of democratization at all. On the other hand, the delimitation was helpful because it allowed us to reassess the question about how the nature of political democracy’s foundation affects democratization in other forms. Still, the democratization literature remained quite general and focused primarily on the national scene. As our attention turns from political democratization as the dependent variable to its relationship with the social sphere, our inquiry has to be deeper and more narrowly focused. For this reason, the chosen field of our empirical study, labour reforms, is a delimited one. At the same time, the approach we use entails moving across different fields of inquiry. The return to citizenship studies implies a recovery of the individual as the focal point of analysis. Therefore it should involve an attempt to connect the different elements that define individual welfare, in this case as conceived in relation to labour markets. To understand citizenship as we shall define it, it is not enough to assess public policy or social pact-making at the national level. Nor will it suffice to quantify skill investments, analyze labour relations in the sociological sense, or investigate labour law. It is the connections, conceptual and empirical, between these domains that need to be drawn.

Social citizenship and development The analytical resources available for this task are not well-developed. Indeed, the understanding of citizenship as productive existence has been absent not only from democratization studies but also from disciplines where one would have thought that this notion would have a natural home, such as in development writing. In this case the reasons for neglect may be found in an understandable emphasis on the immediate provision of services, especially primary education and health. We agree with Marshall, however, that citizenship is about more than a composition of social rights. It is first of all about substantive conceptions of ‘membership of the community’, from which rights and duties then flow. 3 The area of the development literature that has considered membership of communities most directly in terms of productive existence is the analysis of human capabilities and entitlements.4 This line of work continued a long tradition in political philosophy and economics, but at the same time made a powerful break with convention by introducing

Introduction

xxv

the broader concern with membership to the study of developing economies. On the other hand, the capability approach has mainly been concerned with acute problems of poverty and crisis and not with more complex institutional issues, such as those that affect labour markets and more developed institutions of welfare. Therefore there still is no single established tradition that we can draw on to study the development of social citizenship in emergent middle-income economies, or indeed across a more wide-ranging set of cases. In view of this, the first two chapters of this book seek to broaden the philosophical concerns about membership into a workable framework of analysis of occupational citizenship. This must include discussions both of the processes involved in the formation of this form of citizenship relative to other aspects of national development and of more specific institutional issues in public policy, labour markets and labour relations. In Chapter 1 we review Marshall’s work on the substance and the development of citizens’ rights in Europe up to the Second World War, and analyze what we see in some ways as the reversal of the sequence he studied. It is often thought that Marshall’s analysis is relevant mainly to developed societies and to Europe in particular, since social service institutions and industrial relations there are more widely developed. But this is mistaken. Several Latin-American countries had relatively advanced social security institutions by the time of Marshall’s writings, as well as extensive systems of labour regulation, and in the post-war decades they pursued the ideal of unifying and eventually broadening welfare services to cover all. Moreover, the Latin-American states experienced the challenge of developing market institutions in welfare earlier than did most countries in Europe. In that sense they are especially relevant to the study of the reversal of Marshall’s story and to the present-day issues of citizenship raised in the occupational realm. What those peculiar problems entail in the context of contemporary moves to make labour markets more flexible is assessed in general terms in Chapter 2. In particular, we discuss what the greater fluidity of labour markets implies for different theories of the role of labour institutions in relation to individual freedom and the development of human resources. The hypothesis made is that well-established labour market institutions above the level of firms have become more, and not less, important to citizenship and development. The chapter lists some of the specific institutional problems that could be identified in Chile by the early 1990s as a result of the neo-classical approach to market reform.

xxvi Introduction

Political transitions and citizenship: the Chilean case The period immediately after 1990 in Chile was crucial precisely because labour reform became an issue of citizenship at this time. The second and third parts of the book are therefore concerned with change and continuity in public policy and labour relations up to and during this phase. Although we seek to carry the political story through to the middle of the year 2001, the years of the first transition government, during 1990–3, are given special attention. This government can be considered as the founding period of the new democratic regime. It lasted only three years, but during that time crucial decisions and reforms were made which affected social citizenship in a lasting way. Moreover, the transition years in Chile coincided with the end of a prolonged period worldwide of economic downturn followed by de-regulation. By the early 1990s, some countries were still embroiled in deep economic reforms. However, enough had been done for a consensus to have developed around such reforms, and for the question to surface again as to what shape social citizenship should take in the new milieu. What is significant about Chile is that the country had not only started structural change early on, but within the narrow terms of the liberal reforms had also completed such change, possibly with greater success than any other contemporary case. 5 Chile’s early arrival at the juncture where market reform and political democracy appear to enter a fruitful union makes a study of the reality of the country’s achievements a particularly compelling task. Chile had at the time under study a number of institutional capacities, which should it make it possible to take steps to deepen occupational citizenship. These included the presence of macroeconomic stability and good prospects for economic growth. Both facilitated the building of long-term contracts between social and economic actors. Moreover, Chile had an uncorrupt and relatively efficient state bureaucracy and unified national organizations of employers and unions. Respect for the law and institution building among the citizenry (including labour law) was extensive, and the capacity of the state to monitor law enforcement and property rights well established. To this should be added the existence of economic institutional capital – for example, a well-functioning capital market, some institutional framework of social insurance, and relatively developed institutions of education and training. An analysis of individual institutions, however, will not be sufficient. We also need to consider how far existing institutions operate as a network, and how political factors impinge on their function. In short, we

Introduction

xxvii

need to take into account the broader framework of governance – to be discussed in Chapter 1 – which dictates how such capacities as we have highlighted are linked and put to use. This framework includes the developmental orientation of the state, the density of social governance and the shape of political governance. The first encompasses the willingness and ability of government caretakers to set independent development goals and effectively communicate these. The second entails the existence of an entrepreneurial class willing to engage (in an associate form) in the development of long-term strategies of productive development. It also includes labour organizations with a preference-ranking that allows ‘control trade-offs’ to form.6 Meanwhile the third factor covers, in our case, the impact of the shift in national political governance. The challenge of moving to an open political regime takes us back to our initial question concerning the relationship between political and social democratization. Would labour reform initiate the series of ‘precedent-setting political confrontations’ through which, according to Valenzuela, ‘virtuous institutionalisation’ and consolidation of political democracy proceed? 7 If so, it would appear to provide a perfect illustration of how political democratization can be made to dovetail neatly with a parallel shift in the social domain. Alternatively, perhaps the former became consolidated at the expense of the latter through the process of labour reform. Indeed, the relationship between political and social democratization is not straightforward. The second and third parts of this book are concerned with providing the different pieces that one hopes will make for a clearer picture. As we noted, the national framework of governance was a political construction through which the ability to change labour institutions evolved. In the second part of the book we analyze how labour institutions came to attain such a central role in Chile’s transition. Chapter 3 looks at the political organization of employers in relation to labour reform, and Chapter 4 at the orientation of the labour sector and the political parties in opposition to General Pinochet. Chapter 5 surveys the political framing of the labour regime during the crucial years. In the final part of the book we look to more specific issues in Chile’s institutions of labour. It is the areas studied in this part that allow us to determine how far actual steps towards occupational citizenship were taken during Chile’s transition. Our first case study, in Chapter 6, assesses the relationship between labour representation and the flexibility challenge from a legal-institutional angle. It looks at the political process of labour reform and discusses the implications of a weak representative framework for the possibility of establishing long-term

xxviii Introduction

relationships between employers and labour. Chapter 7 then looks at public training policy, defined as the central bridge between social and economic policy in a refined development strategy. In the chapter we identify the interception of institutional and ideological constraints involved in reforming an atomized market regime. The last two chapters of the book concentrate on labour relations inside Chilean firms. This is done through a survey carried out in 1992 of local union presidents in the manufacturing sector. It is argued that the limited nature of the national reforms made relations inside Chilean firms of even greater importance to labour freedom and empowered mobility. By looking, respectively, at styles of labour management and the role of unions, we examine the sources that promoted and impeded the establishment of long-term relationships and investment in workers. The survey reveals the sources of rigidity in Chile’s highly liberal regime. Thereby it questions the notion that deregulation necessarily makes labour market agents more mobile and free. In the conclusion, finally, we are able to connect the different elements of labour reform. In the light of our review of local labour relations, it is possible to consider the impact of the national reforms at a local level. We are able to judge how far the national reforms had expanded the scope for social democratization during the crucial initial phase of political democracy’s reinstatement in Chile. In more general terms, we are able to establish the intricate ways that modern citizenship is shaped by labour markets and labour market reform.

Part I Theoretical Issues

This page intentionally left blank

1 Citizenship and Democratization

Citizenship is an elusive concept. The ideal of social inclusion that it espouses is timeless. However, the substantive meaning of citizenship varies as political and material conditions change. Our understanding of citizenship therefore needs periodic revision. In this chapter we first provide working definitions of different forms of citizenship. We then set the contraction of social citizenship since the 1970s in the context of earlier formative periods. We argue that the loss of the idea of a mainstream divides the post-war phase from later efforts to restore social rights. This leads us to consider issues of methodology in the analysis of the development of citizenship. It is suggested that a stages approach is inadequate. Not one, but several factors independently shape the form that social citizenship takes. Analyzing the interplay between key aspects of economic governance allows us to identify the national framework of occupational citizenship and to assess the Chilean case in a comparative light.

Bases of occupational citizenship Definitions of citizenship Political citizenship In the simplest terms, modern citizenship is synonymous with membership of the nation-state, or the holding of the passport or residence permit of a nation-state.1 With the rise of the modern state, a double process occurred, whereby citizenship in this limited sense came to embody the formal equalization of subjects in various aspects of civil law, while other forms of citizenship became detached. Thus the modern dilemma concerning the form of social citizenship, and particularly the role of the state in supporting it, came to the fore. 2 The historical division of citizenship signals the difficulty with substantive definitions of political democracy that conflate the different elements of citizens’ 3

4 Theoretical Issues

rights. A minimal (separate) classification is called for. Valenzuela’s composition of O’Donnell and Schmitter’s ‘procedural minimum’ and Dahl’s eight conditions provides a good definition. The procedural minimum refers to ‘secret balloting, universal adult suffrage, regular elections, partisan competition, associational recognition and access, and executive accountability’. Meanwhile Dahl’s conditions include freedom of information, and to organize, speak and stand for office.3 Political citizenship is, then, synonymous with the existence of political democracy under this definition. Social citizenship Social citizenship, on the other hand, could refer to other rights which in principle apply to all citizens, and that are, if not legally established, at least de facto protected by the state.4 In our case, we shall be mainly thinking of rights and duties that relate to the individual’s position in labour markets. Since the 1970s, the general debate about social citizenship has been heavily influenced by changes in this domain. Such changes include the greater precariousness of work contracts and the resultant clash with the idiosyncratic nature of investments in human resources. These trends brought to light the problems associated with unregulated labour mobility. Long-term or even intermittent exclusion from labour markets reappeared as a welfare issue in a context where work increasingly bought social security. Precarious contract forms expanded as state protection contracted. Together these conditions explain the importance of providing a long-term context for the development of the individual in the occupational sphere. Occupational citizenship implies the existence of such a context, in effect a form of continuous membership of the labour market. In a more practical sense, occupational citizenship would point towards integrated ways of approaching employment problems. Occupational citizenship The notion of occupational citizenship restores the classical association of social and economic status with citizenship (or full membership) which the liberal carving-out of the political sphere had divorced. This association was based on the idea that social exclusion is not compatible with a meaningful conception of citizenship. The Greeks resolved this tension by excluding those without property or ‘office’ as did the European states of the nineteenth century. In contrast, the modern idea of occupational citizenship extends the means of ‘office’, interpreted liberally, to all.5 The abhorrence of passive assistance as a means of alleviating social exclusion has been broadly shared. It has emerged from conser-

Citizenship and Democratization

5

vative liberal as well as socialist quarters, from the Malthusians in post-revolutionary France to the Fabians (such as the Webbs) in early twentieth-century England.6 The philosophical foundations of the notion of occupational citizenship are best ascertained, however, through the treatment within political philosophy of the concept of individual liberty. At least from Mill onwards, an implicit distinction has been made between human essence as one of consumer of utilities (which Mill saw as too limited) and one as user of human capacities. 7 In modern socialist theories of democracy Mill’s critique is made more concrete. The freedom to pursue individual happiness is seen as ‘an equal effective right of the members [of the polity] to use and develop their human capacities’.8 The notion of negative liberty, or the utilitarian’s absence of intervention into a given personal space, is translated into freedom from the extractive power of others.9 The development of human capacities is now interpreted in a more full sense of positive freedom. According to Market Socialists, the minimal notion of liberty as ‘absence of constraint’ is inadequate because it does not encapsulate the value of autonomy that most people associate with freedom. Individuals must be able to retain the ‘capacity for choice itself’.10 Occupational citizenship has this notion of developmental freedom at its core.

Occupational citizenship and development In theories of development the conception of developmental liberty has generally given way to the utilitarian vision. Social exclusion is associated with wants in the latter area, and not the former. This is evident in the human capital literature, which, as Sen comments, has had a tendency to see human skills not as ends in themselves, as sources of human fulfilment, but as means of achieving growth.11 To this critique however we need to add that human development is more than just an individual end. The problem of individual development also has a social dimension, which captures our human essence as social animals. Archer refers to this as the ‘axiom of sociality’.12 It leads us to comprehend the classical association of citizenship with the capacity to hold public office, what we might equate with developing as an individual by being socially useful.13 Sen himself refers to ‘agency achievement’, the ‘general (usually social) goals (other than immediate personal well-being) which the person has reason to have’. In particular, he highlights the importance of the ‘occurrence of such things [that one values, as] brought about by one’s own efforts’. 14 Sen’s focus on a ‘life . . . of serious options’ and his emphasis on preferences based on ‘values’ and ‘counter-factual choices’15 as an aspect of freedom (separate from the access to a certain level of

6 Theoretical Issues

income)16 hints at the importance of the continuity and substantive nature of social inclusion. To this, however, we need to add the notion of meaningful productive existence as the making of socially recognized contributions, or the freedom to contribute. Hence the notions of developmental freedom and of the freedom to contribute have strong affinities with occupational citizenship. The former notion however, can be seen as temporally finite. In contrast the notion of occupational citizenship lends itself to an analysis of the problems involved in securing the permanence of social inclusion. It is difficult to deny the essential good of contributing and developing through productive existence. The concept of occupational citizenship is, however, concerned with the problem of maintaining this ability over time. Therefore, what currently seems to be lacking is an inquiry into the institutional means of sustaining that function. Both national (established) and incipient forms of social (or occupational) citizenship may be considered relevant. The latter form is pertinent as a social model that aspires to state protection. It therefore assumes a particular importance during transitions. This is a time when various models of citizenship come into contention. Indeed, Chile’s transition coincided with a broader debate about labour rights. For this reason alone our discussion will be both about rights that are in principle national, and about their incipient forms. Here it is useful to remember how other rights, like the provision of basic health care, may, become adopted within the popular understanding of citizenship. If, for instance in the context of western Europe, regular health care was denied to some individuals, it would undoubtedly be judged that this group had been effectively excluded from the citizen body. We could also say that this makes healthcare access a dense form of social citizenship in this case.17 The divergent trajectory of the US and the retrenchment of public services elsewhere since the 1970s, confirms the historical contingency of such a development. On the other hand, the continued high political status of health reform affirms the contagion effect that the example of health care density in the core welfare states has had. Indeed, the example of health signals the expanded relevance of occupational citizenship. The concept has acquired a new meaning as a range of social goods formerly provided by the state have become more closely defined by individuals’ labour market position. In other words, occupational citizenship has become increasingly relevant for national citizenship to be or remain of a ‘high-intensity’ kind. This entails that the means to sustain developmental freedom and the freedom to contribute have to be invented anew.

Citizenship and Democratization

7

Marshall on citizenship reconsidered Reversal of the classical sequence To understand the significance of the contemporary invention of citizenship in this broader sense, it may be useful to revisit earlier defining moments. T. H. Marshall is relevant, not only because he remains unsurpassed as a historical and definitional source on the subject, but also because he wrote his lectures at an important stage in the development of social rights, shortly after the Second World War. This is the phase that culminated in the aim of the welfare state, during the post-war boom, and whose decline and then reform are at the heart of our study. As Marshall saw the eighteenth to the twentieth centuries as the ‘three centuries of citizenship’ (civic, political and, finally, social), the period from the 1970s to the 1990s marked in some ways an equally seminal three decades. 18 They represent in a sense a reversal of Marshall’s story. The movement was not, as Marshall had interpreted the progress he studied, one of ‘strengthening civil rights . . . by creating social rights’, which he regarded the rise of collective bargaining in the late nineteenth century as doing. He viewed the establishment of a ‘secondary industrial citizenship’ (through bargaining in particular sectors) as unsustainable, and so labour rights had to be brought into the political sphere.19 The logical consequence was that the ‘living wage’ had to be accepted as a social right, not by employers, but by Parliament, and extended to all. From welfare to pro-poor policy In contrast the equalization of precarious conditions of the recent past represented a collective move downwards. It was not simply that the fiscal means for providing universal welfare shrank. Perceptions of justice followed reality: in Marshall’s time ‘the diminution of inequality [had] strengthened [demands] for its abolition’.20 Now the realist psychology, ‘whatever is, is right’, 21 which Alfred Marshall ascribed to Adam Smith’s followers, applied to the embrace of a reduced welfare doctrine. The distributive conflict became defined as a struggle between the extreme poor and the average poor and working classes, and in this way citizenship was divided anew. This happened in Chile in at least three ways. In the first, poverty relief and general entitlements to social services came to be seen as competing claims.22 This entailed the abandonment of financing social welfare via taxation in favour of individual accounts insurance. In parallel with this the remaining fiscal resources were diverted towards targeted and demand-based schemes. Second,

8 Theoretical Issues

most remaining universal (state-financed) programmes became decentralized in their administration and finance, and in the process general standards (in education, health and so on) were diluted. Finally, a selective emphasis on certain users in public goods markets became common, as happened in the case of training, as we shall see in Chapter 2, where firms, not workers (including unemployed or casual workers) came to be the central consumers. Whereas previously selective (targeted) programmes had been vehicles to bringing marginal groups within the norm, the norm itself ceased to be the objective. In principle, the approach was shortsighted. In the end, targeting the marginal would be no longer sufficient to eradicate marginality. The mainstream itself would be interspersed with periods of marginality for all individuals, with precariousness as a general norm in labour markets. This phenomenon of precariousness can be identified in various welfare domains that we do not enter. However, we see the labour market as being particularly relevant, as the nexus that connects the different elements of welfare that have come to depend directly on income.23 Occupational citizenship and the mainstream Yet what does the mainstream signify in relation to citizenship? It is important here to emphasize again that citizenship, even when identified in relation to individual rights, is above all a social phenomenon. The passage most often quoted in Marshall is the idea of ‘membership of the community’. 24 But Marshall himself did not only see social citizenship as a contract between individuals and the state in which certain duties were extended for rights. It was in a collective meaning about a ‘way of life’ that came into being when ‘the provided service, not the purchased service, becomes the norm of . . . welfare’.25 It is possible to interpret that the core idea that Marshall and others sought to express at this time was that of a mainstream. While some people may at times be able to afford luxuries, essentially all should contribute to and have a place in the mainstream. One can identify three stages in this idea. The first in modern times is one that we may associate with classical liberal thinkers such as Alfred Marshall, J. S. Mill and other contemporaries, when they emphasized the importance of initial education, not merely as a right, but also as a compulsion. 26 T. H. Marshall saw ‘the growth of public elementary education during the nineteenth century as the first decisive step on the road to the re-establishment of the social rights of citizenship in the twentieth’. 27 In other words, Alfred Marshall and Mill implicitly, and T. H. Marshall explicitly, recognized that skill is at the core of citizenship,

Citizenship and Democratization

9

not simply sequentially (in time), but logically as well. It is at the core because it is the individual’s productive existence that makes him social. This implies that ‘skill’, and indeed that is how we shall understand the term in this book, does not necessarily signify technical or theoretical knowledge of an advanced degree. Rather it encompasses the learning of a recognized trade, or at least the formal acquisition of work experience of a kind that allows the individual to build on it and have it recognized in the wider community. Alfred Marshall or Mill, however, did not reason that far, out of fear that the concept of compulsion in economic life would entail a restriction on individual freedom that they abhorred. T. H. Marshall, on the other hand, recognized the incongruence between social assistance as a citizenship right extended through the political sphere (Parliament and the public administration) and ‘industrial citizenship’ as an institution that extends the social wage only to exclusive members. His answer, as noted, was to unite the two ‘to bring into the sphere of policy-making and public law the concerns of trade unions’ to an extent, and provided that trade unions also recognized their duty to the wider community. At the same time he seemed resigned to the sharp distinction between formal schooling and the labour market, and only lamented fleetingly what he considered to be the lasting mark of early educational achievements (or lack thereof) within later occupational life. On the other hand, he foresaw the essence of a third stage in the idea of a mainstream that we could argue pertains to society today. He spoke of ‘giving as much opportunity as possible for second thoughts about classification’. And he stressed that this would pertain ‘both in the educational system itself, and in after-life’.28 This idea, in modified form, is a core element of occupational citizenship. Its modern form and problems derive from the social mobility and flexibility that is associated with labour markets today. These make the return to the mainstream, second chances, or bridges that sustain individuals between phases in their lives, like the ability to return to studying, finding a new job in later life, and so on, of vital importance. Public policy and the mainstream At the same time, occupational citizenship cannot simply be reduced to social policy. Even so-called ‘pro-poor’ policies, while they inspire sympathy in being focused on immediate relief, do not in themselves contribute to the advancement of citizenship. We shall argue that they do not, unless they are tied to other policies that (i) seek to sustain a mainstream;

10 Theoretical Issues

and (ii) seek to bring the destitute within such a mainstream. Contrariwise, they are essentially forms of charity, in the sense Miller defined. 29 Redistribution need not be thought of in this case simply as a matter of taxation financing a core service (such as health). For example, practices or contributory services that are not strictly state-provided, such as apprenticeships or unemployment insurance, can be linked to ‘provided’ services such as education or income support. But this, of course, requires that qualifications and entitlements are mutually recognized and regulated on a nationwide scale. Citizenship is therefore, almost inadvertently, tied to institutions in the private sector as well. The mainstream and equity It is also important to emphasize that the existence of a mainstream is not synonymous with the goal of economic equality. Rather, it is about continuous access to equality of opportunity, and about an acceptable level of control of individuals over their economic (productive) life. As we have already seen, this particular concept of control, or developmental freedom, was highlighted by key writers of the nineteenth century, and, of course, by T. H. Marshall, although it may be said that none attempted a workable or practical definition. Alfred Marshall’s understanding of citizenship as ‘living the life of a gentleman’ (and presumably woman) could be equated with Marx and Engels’ ideal of human beings as able to engage freely in a variety of activities.30 It also conforms with the later notions of personal development expressed in the writings of Macpherson, Miller and Sen. Marx and Engels, of course, also indicated with this that people should not be slaves to their occupation, although for practical purposes this problem could be avoided by the increase in leisure time and by the notion of second chances already discussed. In fact, Marx abhorred the idea that social justice could be attained by paying all workers an equal wage. 31 Occupational citizenship certainly would not imply equal incomes. It could accommodate quite a considerable difference in accumulated wealth among individuals. However a line would have to be drawn where a certain level of income was necessary to pursue basic development chances. Similarly, there are levels of inequality in access to what we could call social mobility amenities that would be incompatible with the notion of a mainstream. At a minimum, our case studies suggest the need for a certain level of redistribution to link the systems of social insurance and employment continuity, for certain obligations on the private sector to contribute to human development, and for public regulation to guarantee a mainstream in markets of skills.

Citizenship and Democratization

11

Development of a mainstream The mainstream is also a changing phenomenon, and it is difficult to define an ideal state. The key point to observe, both for labour market institutions and for specific areas of public policy (training, unemployment and social assistance) is one of direction; that is, whether the policy or institution in question is forming part of a progressive cycle that is contributing towards constructing the mainstream. It is often said about Latin-American countries such as Chile (Uruguay, Argentina, Brazil, Costa Rica and Cuba after 1959 are other cases in point) that during the post-war period till the market reforms of the 1970s (and Cuba beyond), they aspired to follow the progressive cycle that had inspired the architects of the welfare state in Europe. This cycle pointed towards progressively including more people within the mainstream of society through the vehicle of social policy. In turn social policy itself would become a mainstream service, as excluded sectors would cease to exist. The LatinAmerican countries had focused primarily on the classical services (initial education, health, and pensions), but we have argued that labour markets are a necessary element in a similar strategy for educational ‘after-life’. Accordingly, the challenge to constituting mainstreams today would be to integrate the social services more closely with labour markets, without the latter merely dictating the former as services that sweep up excluded people that labour markets produce. Then the purpose of study should not be to determine whether occupational citizenship ‘exists’, but where and how forms of this kind of membership develop. The central question is how far specific labour institutions or policies detract from or build institutions which contribute towards a mainstream within labour markets. The modern sequence To recapitulate, it is the absence of the idea of a mainstream that sharply divides the current stage of welfare reforms from the post-war phase. The notion of a mainstream is not an accepted doctrine today because it is perceived as clashing with the economic component of civil rights. The Chilean case is symptomatic of this clash because of the emphasis on the economic component since the mid-1970s. On the other hand, the clash is often concealed in the discourse because of Chile’s clear and marked restoration of the political component of civil rights (in 1990), and of political citizenship, both of which we tend to associate historically with a certain ‘superation’ of the crudeness of market rights. We still lack a deeper understanding of this economic sphere of civil rights, and its relationship to social and political rights.

12 Theoretical Issues

It is interesting that Marshall himself, at least in his own definition, did not distinguish overtly between the political and economic components of civil rights. The reason, no doubt, was that even he, to some extent, saw the sequence as being ‘natural’ that introduced economic rights appropriate to capitalism coincidentally with the development of the basic freedoms that underpin political citizenship. These included ‘liberty of the person, freedom of speech, thought and faith’ as well as ‘the right to justice’. What in his definition of civil citizenship we would call the economic components covered ‘the [universal] right to own property and to conclude valid contracts’. 32 For practical purposes, however, Marshall did perceive the difference between the political and economic components in civil rights. He noted for the late nineteenth century how it was political, and not civil, rights that were a danger to capitalism, because civil rights were anchored in principles of liberal economics, principally the freedom to contract. By comparison the universalizing logic of political franchise in the late nineteenth and early twentieth centuries had, economic liberals could argue, an unfortunate connotation. When the move was made to extend the political franchise to the socially disenfranchised in the early twentieth century, there could only be two possible political outcomes. It was inconceivable to remove social assistance in exchange for political citizenship. Hence the logical step was to extend the principle of social assistance to all; that is, as a sine qua non of citizenship. This ‘dangerous’ sequence has been avoided in modern times. Universal political franchise has already been accepted and people have ceased to ponder on the logical implications that this process might have. In other words, the renewed emphasis on the classic economic component in civil rights is not broadly emancipatory at the time of writing. If at all, it is so only to a limited degree, because the historical context has changed. The kind of equalizing movements that first pushed social rights to the core of modern citizenship, industrial citizenship, social assistance, and the social mobilization entailed by the First World War, all prior to universal franchise, have been wholly absent from the development of political citizenship in the contemporary age. The basic extension has already occurred and the re-introduction of the classical conception now has the effect of diluting the substance of economic rights. The acceptance of a new welfare doctrine As a consequence, the last quarter of the twentieth century saw the acceptance of a reduced welfare doctrine. This has to be seen in the context of the particular sequence that re-established the notion of the ‘deserving

Citizenship and Democratization

13

poor’, and of which the Chilean is a suitably representative case. The new supremacy of the economic component in civil rights was aided by economic contraction and crisis. These developments justified shifting the socialization of resources from the reproduction of a mainstream to the sustenance over the line of extreme poverty of a minority. That the resource shift had to occur ceased to be questioned, and academic analysis turned to the most appropriate means of securing support. 33 Emergency measures drifted slowly but steadily into the definition of a new welfare doctrine, advocated by donor institutions and state reformers alike.34 In the process, the role of the social services was redefined. From being a generator of the mainstream, these institutions turned to supporting outsiders. 35 The latter, in short, were the new non-members of economic society. In the case of Chile, the neo-liberal reformers understood that the classical sequence must not be repeated, by which outside notions, notions that compete in principle with the individual right to contract (like industrial citizenship) are brought into the political sphere. They understood that the best way of guarding against such a development would be to ensure that practices such as industrial citizenship do not emerge at all. This would prevent the incongruence between industrial and political citizenship from arising, because logically the latter would then seek to encompass the former. The result would be the fuller concept of citizenship as the existence of a social mainstream we delineated above. Entrepreneurial leaders in particular perceived this astutely, and this was the reason for their adamant refusal to discuss substantial themes with labour, not just in the firm, but at every level. We could say that labour reforms in Chile in the 1990s represented an attempt to stall the replication of the classical sequence, even as they were sold as an exemplary instance of that sequence, as was necessary in order to appease the expectation of social deepening at this crucial time. To complete the picture, let us reinterpret Marshall again: the re-amalgamation of the three strands of citizenship, unified prior to the Industrial Revolution, occurred in a particular sequence. Civil rights in the economic sphere came first. But what is important is that the establishment of collective rights in the civil sphere soon followed. This helped to define the extent to which social rights became inherent in the broadening of citizenship in the political sphere. It was important to the sequence that allowed the ‘three runners’, as Marshall personified the separation of citizenship in the seventeenth century, to ‘come abreast of one another’, presumably, following the tone of his prose, in order soon to become fully amalgamated in a ‘bionic’

14 Theoretical Issues

person of the post-war age. In contrast, the process of the last quarter of the twentieth century was neo-liberal, not simply because civil rights in the economic sphere experienced a revival but because they remained supreme, and because of the new role of economic liberalism in relation to politics. In the political sphere, economic governance was no longer accepted as a political question, but seen as a neutral process. And therefore, as in the case of Chile, political democracy became established without the social base that we associate with it (rightly or wrongly) in the classical form. The ‘three runners’ have been kept on separate tracks.

The development of social citizenship The evolutionary conception of democratization However, in order to understand the Chilean case counterfactually, we need to develop a comparative approach to the development of citizens’ rights. In reinventing citizenship in modern times, reference tends to be made to the axiomatic relationship between political and social democratization. But this evolutionary conception of democratization misleads us. Even if one accepts Marshall’s original sequence, it does not follow that the process was a mechanical one. Nor can general inferences be made about the relationship between political and social democratization on the basis of the sequence history followed in Britain. Reversals and contradictory tendencies often appeared. These include the more substantive welfare approach of the Speedhamland system in an early stage of the Poor Law. The eventual ‘reunion’ of the three runners, for a time, was never a given. This complicates the analysis of social democratization (that is, the deepening of social citizenship rights). But to this we may add the problem of sorting out what key factors influence these processes at particular points in time. Perhaps political democratization (as indeed we shall argue) is only one among these.36 It is also possible that within the same broad category of citizens’ rights (for example, those associated with employment) different types (as opposed to degrees) of social citizenship have developed, depending on the constellation of factors in particular cases. In the recent democratization literature, the evolutionary thesis can be identified in the guise of four supportive assumptions. The first refers to the supposition that political democratization is a process that occurred originally (or normally) ‘from below’. O’Donnell and Schmitter, for example, alerted us to the fact that recent pacts for democracy in Latin America were ‘democracy by undemocratic means’.37 A related assumption refers to the idea that dominant groups become in

Citizenship and Democratization

15

some way less likely and less able to use their power to influence political outcomes when the universal franchise is re-introduced. The idea that a renewed separation between the public and private domains occurs has been advanced with regard to the role of employers in the Chilean transition. Along the same lines, Przeworksi has noted how a ‘typical democratising coalition’ is likely to accommodate the preferences of all social sectors. 38 The idea of political democracy as spearheaded by socially progressive forces is similar.39 However, serious questions have been raised as to whether the theory of ‘progressive forces’ can be sustained even for the ‘classical’ (European) democracies. 40 So-called progressive actors may even complicate the transition to political democracy. They might fear that this process could derogate established privileges, as one interpretation of labour during the Spanish and Portuguese cases has proposed.41 In the case of Chile, it was not the force that favoured democratization the most – labour – but the one that most resisted it – the entrepreneurial sector – that became dominant in shaping the process. But the most important assumption, for our purposes, refers to expectations regarding the social democratizing role, which the first elected government is likely to play. The revisionist analysis of early state-building shows that the relevant reforms sought to integrate the inhabitants of controlled territories as citizens. However, it was only in a few cases, or in a limited sense, that such reforms coincided with political democratization.42 Early social reforms were more consistently associated with the war effort and with military conscription. 43 More to the point, forms of social citizenship that are never questioned today, such as universal primary education, do not have any necessary or direct historical association with political democracy, although at the philosophical level the two may be closely related (as in Marshall and Mill). 44 In many modern states, universal primary education, and public education generally have been vastly developed without the practice of political democracy having ever taken root. Singapore and South Korea in the 1970s and 1980s are good cases to illustrate this point. The influence of political democracy on social citizenship is deceptive in other ways. Degrees of political competition and the longevity of political democracy may ensure political rights. However it does not automatically follow that they promote social rights. The failure effectively to grant universal health care rights in the United States as noted is a good example of this. Moreover, the fact that shifting governments should, in principle, be able to change the nature of social rights does not necessarily translate into an ability to do so. Welfare states are difficult to retrench.

16 Theoretical Issues

Above all, it is important to recognize the influence of other factors (apart from political democracy) on the development of social rights. Scholarship that makes clear distinctions between aspects of democratization is therefore the most useful. Recent work has focused on the conditions leading to the inception of democratic elections, the ‘first’ transition, and then to the consolidation of political democracy in the classical sense, the ‘second’ transition.45 For the Chilean case, Valenzuela has observed how authoritarian rulers were in a uniquely strong position to assert political control over both the first and second transitions. 46 At the same time, the strength of the economic component in civil rights since the 1970s demands that the factor of economic governance is introduced to the study of social democratization. Indeed, the continuance of anti-democratic views in Chile was tied up with notions of non-intervention in the economy. The focus on economic governance allows us to study the intersections between different arenas of labour reform (the firm, public policy, national pact-making and so on), or the network between different services that are important to occupational citizenship.47 Different factors in economic governance Economic governance as a developmental concept can be divided again into two forms. These include the involvement in economic governance assumed by the state (‘from above’) and by interest associations (‘from below’). These, together with the nature of the political regime, constitute the macro-factors that we regard as important in shaping the development of citizens’ occupational rights. Whereas the three variables are related, it is understood that none can simply be reduced to any other in the set; each may independently influence the development of social citizenship. The state’s developmental role Recent work in historical scholarship on the role of the state has reemphasized the importance of looking at the nature of state-society links. 48 Building on Johnson’s notion of the ‘developmental state’, Evans has suggested that the latter are those that have achieved a certain degree of ‘embedded autonomy’, defined as the existence of a balance between an efficient bureaucratic tradition on the one hand, and smooth state–society ties on the other. 49 For Latin America, adding the second condition would seem to make sense, since in most countries of the region the state has been very involved in the economy, without this necessarily translating into an effective developmental tradition. 50

Citizenship and Democratization

17

According to Johnson ‘industrial rationalisation’ (in the case of Japan) meant ‘state policy at the micro-level . . . state intrusion into the detailed operations of individual enterprises with measures intended to improve those operations . . . the attempt by the state to discover what its individual enterprises are already doing to produce the greatest benefits at the least costs, and then, in the interests of the nation as a whole, to cause all the enterprises of an industry to adopt these preferred procedures and techniques’. [Italics added]51 There may however, also be a variety of social functions associated with the degree to which the state is developmental, and which have important indirect effects on citizenship. This is so even though this citizenship has not always developed into statutory rights in the East Asian cases. For example, a series of studies have indicated how the aggressive focus on industrialization by these states may have contributed in some measure (and in a short space of time) to a kind of occupational citizenship. It has done so by improving the effective capacity, through support for industrial dynamism, skill creation and relatively long-term employment practices, of individuals to sustain themselves in labour markets.52 As Garon shows for Japan, even though the state refused to develop a European-style welfare system, it was successful in organizing and sometimes financing forms of ‘community-based’ charity.53 Korea did not have a formal labour code, guaranteeing statutory rights, until 1997, and yet the courts before then effectively protected a high standard of job security in the case of large firms. At the same time, the point here is not to suggest whether social citizenship on the whole was adequate or not in these countries.54 The fact that such policies were not always developed in the context of political democracy (although in Japan they were) or with greater independent space for trade unions (in Sweden, however, they were)55 should not lead us to judge a priori that they were inherently incompatible with forms of political and social democratization. With these considerations in mind, Johnson’s thesis regarding the role of the state in development can be applied. Close to one end there is the ‘regulatory, or market-rational, state [which] concerns itself with the forms and procedures – the rules, if you will – of economic competition, but [which] does not concern itself with substantive matters’.56 Close to the other end, there is ‘the plan-rational state, [in which] the government will give greatest precedence to industrial policy, that is, to a concern with the structure of domestic industry and with promoting the structure that enhances the nation’s international competitiveness’.57 Along this spectrum, developmental states such as Singapore and South

18 Theoretical Issues

Korea are very close to the plan-rational ideal type. At the same time the ‘plan-rational’ state could also be concerned with social development, as noted above, and as Evans suggests for the Indian state of Kerala.58 In this case, the Scandinavian countries would be a good way up on the scale between laissez-faire (or regulatory) to developmental state, because of the role of the state in social development. 59 The role of social organizations in expanding citizenship The potentially dominant role of interest associations in expanding social citizenship has been amply demonstrated in the literature on corporatism and interest mediation.60 In Scandinavian societies, for example, national social agreements between employer associations and unions either preceded, or emerged in parallel with, the development of key social rights guaranteed by the state. 61 It is hardly a coincidence that these societies both have some of the most comprehensive social agreements and some of the most comprehensive state-provided welfare rights. This has shaped a tradition of social citizenship that places great weight (relative to the other ideal types) on individual rights as workers’ rights, as opposed to rights and duties as national subjects (in the developmental state) or as market agents (in the regulatory state).62 The role of interest associations has also been important in Chile. The relationship between the state and industrial elites was coloured, to say the least, by ‘the question’ of industrial relations. 63 In contrast with the welfare states, interest organizations had not found a self-sustaining formula independently of the state. During dictatorship, the retraction of the state was combined with an explicit and detailed rejection of a public law conception of the labour market, which would, in Therborn’s phrasing of the European phenomenon ‘provide [such] unions’ and employers’ organisations with public functions, rights and duties.’ 64 It is debatable how far interest organizations in Chile ever had strong public functions in the European sense, but the clear stress on market neutrality after 1973, and the greater emphasis on individual as opposed to collective representation in the 1980 Labour Code, will justify our grouping of Chile during and after Pinochet within the Anglo-American common law tradition which ‘had great difficulties in dealing with “combinations” (between the state and private actors) in any positive way, geared to the safeguarding of individual freedoms’.65 This will help us to explain how Aylwin’s Chile, after all, remained most closely identified within the group of laissez-faire states.66 Other important differences can be gauged between the historical cases. A distinct trade-off in the laissez-faire model (where political and

Citizenship and Democratization

19

individual rights typically are high, and social interest mediation and industrial development policies typically limited) is that the advantages of free organization and individual rights may in practice be overridden by the greater organizational power that capital has. The same trade-off does not exist in the (corporatist) welfare states. Meanwhile, a central difference between the two other regimes seems to be the role of interest representation. Strong associations in the welfare states appear to facilitate universal coverage and statutory rights. Meanwhile the weakness of social organizations in the developmental states seem to induce the subordination of the individual in favour of economic nationalism. However too far-fetched extrapolations from these differences should be avoided. The comparison of ‘historical’ examples of forms of national governance is meant to highlight the importance of different factors in influencing citizenship over time. The point we particularly wish to emphasize is the delicate historical relationship between different forms of citizenship. The case of the welfare states compared with the other regimes holds important lessons concerning the sequence of social and political reforms. Perhaps Sen’s insistence on the importance of the ‘prior’ level of social development (prior to the onset of rapid economic growth) to the level of ‘participatory growth’67 may be extended to the political sphere. In other words, what particularly matters (to social democratization) is not how early political democratization occurs, but rather the extensiveness of social development, and of associative practices, when it occurs. This point is crucial to the analysis of the Chilean case. Comparing factors in the development of social citizenship Let us now consider tentatively how different countries would be placed in a figure featuring the three dimensions of national governance (see Figure 1.1). The spectrum on the y axis goes from a low to a high level of independent influence of interest associations on occupational rights. The spectrum on the x axis is as taken from Johnson (1982) (p. 17), from laissez-faire to developmental characteristics of the state’s role. The political forms of rule dimension (z axis) is not so easily associated with intensity, but runs through a series of regime types on a much rougher scale according to the degree of political competition for government.68 The most important aspect to consider in the figure is Chile’s changing position (from 11 to 13). Of particular importance is how the country has moved in relation to the three ideal types of national governance that we have discussed. In this respect, the little lines drawn between

20 Theoretical Issues

7 10

13 5 4

15

3 6

1

10

11

8

2 12

Political regime

10

9

5

0 2 0

7 Developmental orientation of the state

12

10 Social governance 17

20 Key: 1. Denmark 2. Sweden 3. Germany 4. France 5. Italy 6. Spain 7. United States 8. Japan 9. South Korea (1960s–) 10. Singapore (1960s–) 11. Chile (–1970) 12. Chile (1973–89) 13. Chile (1990–5). Figure 1.1 Three dimensions of the development of citizens’ occupational rights Note: The period for Spain is 1975. See also Appendix E.

cases indicate where countries tend to cluster around particular governance forms. We see that the welfare states are in a strong position along the y dimension, and the developmental states similarly placed along the x axis. The line between 7 and 13 connects the two regimes that can broadly be called laissez-faire. In order to provide some idea of comparison, it also tentatively features other countries (in intermediary positions). What should be less controversial here than the exact placement of individual countries is the fact that this is generally not radically affected by the policies of any one government. This includes cases of radical upheavals such as in Chile in 1973. After that date the country moved considerably along all three dimensions. 69 Even in such cases it is likely that previous long-term legacies will in some way manifest themselves. In the case of Chile after 1990, placement 13 (although the box does not illustrate this very clearly), the country remained very low on

Citizenship and Democratization

21

the state dimension. In this case the retraction of the state from a developmental orientation and the low degree of social governance explain the continuing supremacy of the economic component in civil rights. The figure also suggests that there may be a tendency for the ‘other’ two variables to behave in certain ways for each type of national governance. For example, welfare states will typically have political democracies that are either system- or party-hegemonic, depending on whether it is a social democratic regime or party that has consistently been in power. On the other hand, the developmental states tend to have limited political competition. Typically this will be in the form of a democratic one-party state (post-war Japan upto the early 1990s), or one-party or authoritarian state. Finally, laissez-faire regimes – those that are weak on the state-indevelopment and social governance dimensions – will often be tied to bipolar party-political arrangements. This is likely because state policies are not embedded to the same extent within consociational agreements of different kinds. The USA, the United Kingdom, and post-Pinochet Chile are cases in point. In between there are the intermediary cases, such as pre-Allende Chile (pre-1970), as well as post-war France and Italy. In these cases interest organizations have not been strong enough to force a social democratic project on the state, and the state not developmental enough to institute a party- or system-hegemonic form of governance. On the other hand, these forces have been strong enough to prevent the development of a ‘pure’ laissez-faire regime. It was mainly this relationship that changed in the interlude of Pinochet’s dictatorship. The state became sufficiently laissez-faire, and unions sufficiently weak, that a progressive deepening of social citizenship rights, as before, had become unlikely. What about the Chilean state during the Pinochet period? The Pinochet regime’s placement in Figure 1.1 indicates that the state was very strong, and heavily intervening. However, this occurred only on the political dimension – that is, in forcing institutional changes. It did not manifest itself in the promotion of either social or industrial development. This helps to place in perspective what may appear as a paradox: how the strength of the state’s role in removing the political institutions and ties with society that would support strategic development projects also created a new constituency in favour of occupational citizenship. In other words, the step seemed not to distant from equality of status (or non-status) to the formulation of new rights and duties in universal terms. Or at least this was so within the remaining pockets of social organization where demands could still be collectively formed.

22 Theoretical Issues

Conclusion This trend – the universalization of informality – takes us back to the early debates regarding the format of universal education in Europe. Universal education was then seen as a way of reinventing citizenship after the dislocations produced by the break-up of separate estates. Malthus, one of the greatest opponents of universal state provision, eventually proposed it as a medium of social discipline. 70 The fact, though, was that the idea of compulsory education in France was as politicized in its time, as Donzelot relates it to us, as the idea of rights of access to training or employment information for Chilean workers became in the 1990s. It is therefore only in a historical light that the suggestion is radical, in principle – that if we accept compulsory schooling, we should have no great difficulty with accepting occupational citizenship as well. The question is why societies would in principle stop at the ages of thirteen or fifteen in terms of actively ensuring that the individual has the means, as well as the duty, to retain a productive existence. What matters, however, as much as the principles, or even the institutional feasibility of reforms, are the political perceptions that proposed changes engender. The purpose of comparing the three types of national governance here has been to suggest how the development of forms of social citizenship should be analyzed as the product of a constellation of factors. It should not be seen as a gradual or natural process of democratization. The Chilean case in 1990–1 is a modern story of the consolidation of a laissez-faire democratic regime. Still, that is only half of the tale, because the country embodied other possibilities: the years of the first democratic government, between 1990 and 1993, were decisive in settling the future role of social organizations and the state in development. The early period of political transition, in particular, allows us to observe the complex interplay of overlapping legacies as each asserted or reinvented itself at a time of political flux.

2 Human Resources and Market Reforms

In this chapter we set our discussion of occupational citizenship in the context of the challenges posed by the market reforms of the 1970s and 1980s. We first look at the general debate about labour flexibility, and then discuss the institutional issues raised by occupational citizenship in two areas where this notion may be said to have important efficiency functions. These are the advancement of labour freedom; and the development of human resources. These areas also demonstrate important ways that internal and external labour markets are linked. We then provide a brief review of the labour market reforms of the Pinochet period, considering local labour institutions first and then reforms to public policy and the role of the state. These reforms may be seen as representing an attempt to perfect a neo-classical model of labour markets where the economic component in civil rights prevails. Finally, we discuss the specific challenges to labour institutions faced in Chile by 1990, and suggest the need for an analytical framework that provides some scope for linking political dimensions of citizenship to the social and institutional problems involved in the development of human resources. The study of occupational citizenship provides one possible framework.

Labour market flexibilization: the core controversies Neo-classical and institutional approaches The flexibility debate today covers a broad gamut of areas. These include labour costs (both aggregate, differential and over time), conditions of employment (type of employment contracts) and the organization of work. Also involved are issues concerned with mobility between jobs, education and training and forms of social security that the state no longer provides.1 The policy controversy in the debate centres on 23

24 Theoretical Issues

the question as to how far labour flexibility should be regulated and supported by frameworks that support individual security. The position of the neo-classical development economics is that flexibility is achieved in the measure that the labour market is deregulated.2 Deregulation makes contractual relationships (at least those that cover labour) as contingent as possible, the view of contract that Williamson refers to as ‘classical contracting’.3 In contrast with this neo-Marxian scholars and critics from the institutional school believe that pure market relationships are at best not achievable in complex modern societies. At worst the attempt to enforce them legally is damaging to other social and economic objectives. These include developmental freedom as well as long-term efficiency goals such as the promotion of human resources.4

Premises of the neo-classical development economics These positions reflect differences in methodology. The neo-classical development position does not deny that the development of human resources is crucial.5 However, it contends that the levels and forms of human capital that are required will emerge spontaneously if the market is allowed to work without hindrance.6 It is also implicitly assumed that the market provides effective mobility of human resources. This position needs to be addressed in terms of both the positive and negative assumptions behind it. The positive assumption refers to what will/does happen in the atomized market regime. In this case the natural sequence claim could be discredited by showing that social institutions played an autonomous part, even within the very extension of markets that had occurred in a country like Chile. In underestimating the power of social (non-market) institutions, the neo-classical perspective fails to explore the connection between internal and external labour markets and institutions. 7 The proposed independent role of social and political factors would also question the very idea that intervention in markets can be neutralized. Managers are never isolated in the decisions they make. The Chilean case allows us to ask how far deregulation itself might complicate later social and industrial development efforts by creating new institutional barriers. The negative assumptions behind the neo-classical argument refer to the likely outcomes of restricting or not maximizing markets. A central rationale is the possibility of market collusion (by unions). This would prevent the operation of the market mechanism in determining the price and (via its signals to managers) the appropriate organization of work.8 In other words, giving too much of a say to unions will hinder

Human Resources and Market Reforms 25

the ability of managers to manage human resources efficiently. There is also the standard moral hazard argument that providing workers with too much security will dampen their incentive to work hard and/or, if unemployed, to seek work. Interestingly, the neo-classical position is not usually associated with the argument that too great a sense of job security might also dampen efforts to find better jobs and progress within markets. This, we assume, is because of the primary focus on the freedom of managers to act in the literature. However, the existence of incentives for employees to change jobs could also be seen as a key feature of a vibrant labour market. In this study we discuss both negative assumptions regarding social organisations in the neo-classical development economics. We do so by looking at the role of unions and the opinions of local labour leaders on issues of work and mobility. Our general hypothesis is that in trying to protect labour freedom in a scenario of universal informalization of work, unions may have a different affinity with development goals than they did in the past. In particular, they may be more likely to push for universal citizenship rights. These include forms of occupational citizenship, starting with the demand for the formalization of work in the firm. In the case of Chile, key factors made the democratizing role of unions directly relevant to the development of human resources. Among them was the unions’ inclination, for a number of historical reasons, to use control gains as a medium to introduce more long-term relationships to Chile’s labour institutions and to formulate workers’ demands in terms of universal citizens’ rights.9 Formalization was also intimately linked to the investment of workers in Chilean firms.10 In other instances, employer organizations played a democratizing role. This, however, would typically depend on the existence of a history of associational behaviour, as in the exceptional case of the metallurgical association in Chile, ASIMET. In respect to the second question, we tend to find that attitudes to mobility are closely related to the perception of opportunities in labour markets. In other words, the reactions of unions and individual workers to collective bargaining and to individual mobility and work are not fixed. They vary according to preference ranking, education and the history of labour relations, and to the existence and credibility of employmentrelated services in labour markets. The key question, then, is how far the opportunities for political opening in Chile were exploited towards the building (whether through incremental or larger steps) of forms of social citizenship which had elements both of historical feasibility and economic efficacy at this time.

26 Theoretical Issues

Labour freedom and the development of human resources Occupational citizenship and the analysis of individual freedom Voice and mobility Finding the institutional means of supporting occupational citizenship both within and between firms poses complex problems. In particular, it questions how far citizenship in the firm can be considered in isolation from broader issues of labour mobility. Let us consider two extremes of labour freedom. At one extreme, workers would need to be in full control of the firm’s decision-making in order to maximize their choice over occupational development in the firm. 11 At the other extreme, workers would need to be presented with the full array of resources that would make it possible for them to be as mobile as other forces in the market, such as capital. 12 Such resources would include, at a minimum, access to unemployment benefit, employment information and subsidized training. The alternative of maximum mobility is not an adequate substitute for attaining control in the firm. Without such control, workers would have to exercise their exit rights and leave their employment every time a conflict arose, and this would be costly.13 In short, relatively long-term employment is the most realistic scenario of work. However in order to be compatible with labour freedom it must involve a combination of control through voice and measures to protect workers’ occupational development when employment fails. Occupational versus firm-based citizenship The position adopted here is that the concept of occupational citizenship, compared, for example, with firm-specific forms of citizenship, is particularly relevant to understanding problems of labour freedom in the modern context, by virtue of its universal scope. To some degree, the exclusive focus on control at the level of the firm has also led the Marxian literature to ignore the relationship between internal and external labour markets. This failure to link external and internal labour markets is therefore shared between the neo-classical development economics and neo-Marxian writing. The selective focus on self-management leaves unresolved the trans-firm freedom of workers. This includes some of the key problems of social equality implicit in the value-domain that underlies most self-management theory. Issues that are not considered include the possibility of large pay differentials between firms, an individual’s effective property right over his/her skills, the effects of dismissal and mobility on human resources and on developmental freedom, and the question of who should bear the cost of investment risk-taking. 14

Human Resources and Market Reforms 27

Therefore, although self-management and occupational citizenship may be compatible, they are quite different. The distinction between them derives, in fact, from their emphasizing different aspects of labour freedom. 15 For example, the self-management theorists appear to narrow exploitation of workers down to the question of alienation from the work process. This is associated, in turn, with the existence of the traditional wage contract. The forms of alienation, on the other hand, have at least two components, variously emphasized: alienation because of lack of shared ownership over the means of production, or alienation because of surrender of authority to an employer. Archer sees resolution of the latter through the former, whereas Bowles and Gintis suggest that the former is not necessary to the latter. 16 Without pretending to resolve this debate here, we may note that the sense of alienation experienced may depend on what it is over which ownership is lacking (output, individual skills, the means of production, capital, land and so on) and on the occupation in question. In a sense, both the concepts of alienation and developmental freedom represent existential claims about the reality of work, which do not need to be seen as exclusive. Here we have chosen to emphasize the importance of ownership (or at least some control) over occupational mobility and skills. How far the retaining of the ability to choose and develop professionally over time (developmental liberty) contributes towards other aspects of individual freedom associated with work is something that is touched on tangentially in our discussion of local labour relations in Chile (in Chapters 8 and 9). Indeed, the concept of occupational citizenship also affords us the possibility of ascertaining the complex relationship between voice and exit. It is a least conceivable, for example, that real powers of exit might buttress the capacity of workers to demand that the firm prioritizes skills investments. In return for various forms of occupational security, workers might trade rights to direct consumption.17 In summary, the self-management perspective tends to imply that the eradication of authority relations is feasible.18 In contrast the focus on occupational citizenship restricts itself to assumptions about the effect on individual freedom when such authority relations at the local level are traversed by professional (that is, in principle, universal) criteria. Ideally, the latter would encompass a situation in which risk (including the investment risks of firms – of any type of governance) is not displaced on to individuals in a way that their personal development is affected (as in the dismissal of the worker without professional associations outside the firm). In the absence of this ideal, smaller steps towards occupational citizenship would include more subtle aspects such as

28 Theoretical Issues

the formalization of work rules in the firm, and/or the extension of protected and tradable qualifications above it. Occupational citizenship and the development of human resources The powers of exit depend, among other things, on the access to education and training. But how to assess the availability of this kind of access? Two problems follow: the unavailability of good data, especially data on training post-schooling, and the tendency in the literature to focus narrowly on quantifiable ‘outcomes’.19 In our case, we are concerned to investigate how far skill investments or outcomes empower individuals in labour markets. Therefore we also need to analyze the importance of institutions and labour relations, not only in promoting or impeding investments in skills, but also in the proper certification of such investments and of work experience. Indicators of initial education are important, but do not in themselves suffice.20 Social institutions and human resources The reasons behind the scant attention to social and political institutions, particularly post-schooling, have to be seen in the context of the origin of human capital theory as a branch of mainstream economics, in the 1950s, 21 and its late revival, in the early 1990s, after the neoliberal interlude. The issue of skill development has presented a methodological challenge to economists. In part this is due to the importance of time-horizon to the development of human resources and the implication that ‘prior’ social factors and institutions, not just markets, influence economic behaviour in this domain. Therefore institutional economists developed the notion of asset specificity. This incorporated time factors and property claims over human resources.22 Human skills are assets that require long-term investment. Mainstream economists such as Marshall, and later Solow, also emphasized the importance of social custom to decisions about savings and investment behaviour.23 Yet the notion that power relations affect the development of human resources remains contentious. 24 Different strands within institutional economics disagree. For example, both property approaches and organization theory are concerned with efficiency implications of the governance of contractual relations, 25 yet they differ in their emphasis on the mechanisms of change. The transaction cost economics (a branch of organization theory) places more faith in the self-selection of efficient contracts as a historical aspect of the development of institutions. This is still the most widely accepted approach.26 Hence the empirical challenge remains to show systematically how social institutions can also

Human Resources and Market Reforms 29

act, independently, both as facilitators and as barriers, to the establishment of efficient contracts. Our research on Chile indicates the importance of self-sustaining elements in governance structures. Internal and external labour markets The importance of time-horizon to skill investments seems to imply a similar logic as that associated with protecting labour freedom. In this area, as we recall, the analytical challenge is how to connect the analysis of firm and supra-firm institutions. For example, property approaches tend to distinguish sharply between occupational regimes, based either in the firm (as in Japan), or at the sectoral or national levels. 27 Yet focusing on only one level is unrealistic. In reality, subtle trade-offs between occupational rights within and outside the firm are likely to form. In cases where occupational citizenship is already high (for example, where the professional domain predominates, as in Germany), the educational system is likely to have a structuring influence on labour relations inside the firm. 28 Conversely, in systems where local enterprise has greater power in structuring labour relations and the labour market, local unions may be important in pushing for the formalization of work from below. 29 Even an atomized regime (such as Chile’s) will be traversed by distinctions between firms, distinctions for which the variegated evolution of unionism is partly responsible. In summary, the key empirical question to establish is how the firm and supra-firm levels of labour relations relate, and what efficiency and inefficiency functions social institutions contain. Our argument is that these questions in the governance literature have to be made historical. In order to understand the political challenge, the challenge of occupational citizenship, we must try to discover the factors that make the relevant institutions change or remain resistant to change. One of the most important actors in the broader institutional environment is the state. The case for a developmental role for the state is particularly strong in the area of human resources, given the importance of long-term investments in this domain. Nevertheless, the standard neo-classical arguments against state intervention have tended to dominate in recent decades. The most common source of inefficiency noted in the literature on training is associated with the lack of integration with firm strategies and capacities of the economy. One such case is inefficient manpower planning.30 Although it may increase labour freedom, at least for some workers, for some time, it may not have a beneficial effect on the development of human resources. An exception is Singapore, where planning was supported by extensive information resources. 31

30 Theoretical Issues

A similar example might be the case of unemployment benefit systems not connected to job relocation. Finally, there is the case of inefficient interventions where unions are segmented sectorally and unable and/or unwilling to establish trade-offs with employers that take the public interest into account. For reasons that will become clear (mainly the extent of deregulation and the attitudes of the labour movement) such problems were not generally relevant in the Chilean case. Human resources and state intervention Instead, a different type of inefficiency prevailed. This was associated with an excessive reliance on the economic component in civil rights, on a common-law philosophy or market neutrality in the framing of public policy. In particular, we shall emphasize this bias (in Chapter 7) as an obstacle to the development of a public regulatory framework with sufficient clout to build linkages between different systems of social inclusion and social assistance.32 This prevented flexibility in the formulation of policy. For example, the market neutral position objects to sectoral policy which would mean that if in a given region, some sectors were thought to be more likely to contribute to sustainable or better-quality jobs than others, it would not be possible to seek to expand opportunities in these fields. Further, the market neutral position advocates a strict separation of service domains, to maximise the freedom to contract. And yet, separate markets in pensions, unemployment insurance, and training, may not effectively link entitlements in these different areas for individuals changing jobs. Establishing institutional links that connect private and public services is an essential part of building the mainstream, but it is not a simple task. We have already seen (in Chapter 1) that a mainstream service is one that entitles all to continuous membership and benefits in particular situations of need in exchange for contributions according to means. Classical examples of mainstreams are a national health service, primary education, and a national pensions system. These are relatively simple as they are supported by general taxation and not vulnerable directly to markets. In wider education and the labour market, the issues become more complex. In theory, economic agents (such as individuals and corporations) would have to act in a manner that would imply responsibility not just to society understood as a system of civil law that protects negative freedom. In addition they would have to act responsibly towards society understood as a mainstream that protects equality. The activities of private and public institutions would need to be compatible with the promotion of positive freedoms of the individuals they affect. The kind of institution–building that supports the mainstream is one which links

Human Resources and Market Reforms 31

the experience gained in one place of work or study to future entitlements. Typical cases in point include nationwide co-ordination of skill certification, employment information and social assistance. The relationship between standards and markets So the mainstream in wider education and labour markets refers to work experience and qualifications that are socially valid (as opposed to merely compulsory). Since it is specific institutional designs that make them valid (such as participation in a recognized course of study or work experience), belonging to the mainstream signifies having access to those designs. The problem of the relationship between standards and markets is partly one of transition. Once certain general standards exist, these can facilitate markets because they make skills (in this case) tradable. However, in the transition, in the process of generating those standards (or markets) there is a fine line between compulsion that is prohibitive of (or unconnected with) economic activity, and incentives that channel economic activity in a desired direction. Therefore, developing continuous education systems that are integrated effectively with labour markets does not depend on state initiative in a simple way, as do primary education or the provision of health care. Where public activity is not well linked to private activity, the mainstream may cease to be an effective tool for occupational citizenship. For example, the famous state-sponsored ‘Højskole’ in Denmark remains essentially a leisure and therapeutic activity. This is vastly different from cases of continuous education in labour markets as in Germany. Here the activities of formal (state-sponsored) education has over time locked participating firms into a dynamic institutional network of training. At the same time, there is no blueprint of what an effective institutional network looks like, what it contains or how it evolves. Arguments that apprenticeship systems have tended traditionally to be focused on industry, or that the development of service economies requires a more ‘market-orientated’ approach can also be qualified. Evidence to the contrary includes participation of service and manual skill occupations both in the German system itself and in the skills-led employment ‘activation’ programmes of Scandinavian countries.33 The general point remains that providing better chances for continuous membership of the employment sphere depends on sophisticated links between the activities of firms and the systems of general education and welfare. Such links make social mobility and recognition effective for the wider majority. The transition government in Chile itself linked the reform of labour institutions to enhanced economic efficiency. Its programme combined

32 Theoretical Issues

a political rationale for labour reform; that is, the need to restore workers’ freedoms and representational rights,34 with an emphasis on how increasing participation (in firms) would also facilitate more efficient work methods and more investment in workers.35 In this way, the programme touched on the domains of both individual freedom and skill development, which, as we have argued, have strong affinities with the notion of occupational citizenship. In the next two sections we examine in general the labour market challenges that this government faced.

Market reforms and labour institutions in Chile (1973–90) The ‘Chilean model’ The market reforms introduced under Pinochet are generally known for their radical nature. By 1980, over 400 state-owned enterprises and banks had been privatized, public-sector employment had been reduced sharply, domestic prices (except wages and the exchange rate) and the capital market had been liberalized, and the exchange rate regime and tariffs (down to 10 per cent by 1977) unified. The overall performance of the economy by the late 1980s is also recognized widely. Chile’s trade balance was in surplus throughout the decade. Fiscal austerity had turned a budget deficit of 6.6 per cent of GDP (in 1974) into a surplus of 5.5 per cent by the mid-1980s. In the five years from 1981 to 1986, exports moved from a mere 17 per cent of GNP to 30 per cent. Unemployment was lowered, to around 5 per cent after the 1980s crisis and it remained at this level till the late 1990s.36 The weakness of technological progress However, the important empirical question for our purposes is how far investments in long-term capabilities was an element of the private sector’s success. Looking at general trends (in the economy) first, there is agreement that Chile’s recovery was not in general based on additions to value added. Levels of investment and savings in the private sector continued to be low. 37 Moreover, technological innovation was not a strong feature of Chile’s recovery. The introduction of new technology in the 1980s was concentrated in large and medium-sized establishments, and it tended to evade the manufacturing industry. Moreover, it was characterized by product- and continuous-process technology, and only rarely with the type of machinery associated with leaps in value added. The simultaneous deregulation of the domestic capital market, opening of the capital account and intensive privatization had provided greater incentives towards financial compared with productive investments.38

Human Resources and Market Reforms 33

The country also continued to face the challenge of deepening its export base. Mining still represented more than half of Chile’s exports towards the end of the 1980s (and 44 per cent by 1994). ‘New’ export sectors were concentrated in agricultural crops with little or no processing involved. 39 Despite its expansive opening, Chile’s participation in world trade, in value terms, declined during the Pinochet years, from 0.44 per cent in 1970 to 0.22 per cent.40 Between the 1960s and the 1980s, industry’s overall share in GDP fell from 25 per cent to 20 per cent, while its participation in employment went down from 18 per cent to 14 per cent. The component of manufactured goods in Chile’s total exports was still only 15 per cent in 1996, compared with 92 per cent for a country such as Korea, and 84 per cent for Singapore. 41 In the labour market, jobs of higher quality failed to materialize in the 1980s, and the subject had by the late 1990s become a national electoral theme. New employment in the 1980s tended towards the creation of precarious jobs in low-productivity sectors. Numbers of males employed in the informal urban sector increased by roughly 50 per cent between 1966 and 1983, a change that remained constant until 1990.42 Their participation in the secondary sector, meanwhile, fell by 25 per cent. Increases in female participation in the labour force also to a large extent meant an influx into commerce and services, where their participation increased from 15 per cent to 40 per cent between 1966 and 1990. The percentage of people employed in agriculture barely changed in the 25-year period under review, but temporary and seasonal employment advanced dramatically, in some estimates involving 70 per cent of the workforce.43 Although poverty decreased in the 1990s, the mainstream was not recovered as income distribution failed to improve.44 The problem of low value added and the role of the state Accordingly, in the recent literature, emphasis has been placed on the state’s withdrawal from a developmental role as a factor in the weakness of Chile’s technological performance. For example, Petrobelli finds that the ‘absence of selective support’ for industrialising efforts . . . may have prevented the up grading into more complex manufacturers’. 45 ‘And the absence of interventions produced a pattern of internationalisation that reflected the static sources of . . . natural resource endowments’. On the other hand, to note that the state withdrew from development planning is not to explain how the private sector did not use the new space to engage in a programme to expand productive capacity and develop human resources. Why did the private sector itself not setout on such a path? The argument of the neo-classical school is that such

34 Theoretical Issues

an initiative will emerge spontaneously when markets are allowed to develop unhindered. The fact that a higher value-added and humanresource-based style of growth did not emerge, or was not promoted by democratic governments, even after more than a decade of strong macroeconomic performance and liberalized capital, goods and labour markets suggests that certain political and economic barriers existed within the liberal model. This interpretation is reinforced by indications that unfulfilled demand for worker training did exist among Chilean firms by the early 1990s. At this time, supra-firm training provision was failing to meet even expressed demand for skilled labour. On average, 43 per cent of manufacturing firms surveyed in 1990 stated that lack of skilled labour was one of the greatest obstacles to the development of their firm. By comparison, 23 per cent cited lack of administrative capacity and 15 per cent lack of technical means.46 At the same time, only 8 per cent of respondents to the same survey believed they had the equipment or personnel to carry out training themselves.47 There were also indications that market institutions responded weakly to the high levels of labour mobility. The low dynamism of Chile’s labour market in terms of absorbing the increase in forced labour mobility and creating new quality jobs became apparent in the wake of the spill-over of the 1997 Asian financial crisis. Chile’s unemployment peaked at 11.5 per cent, higher still than that of Korea. The development problems Chile faced by the late 1980s cannot simply be reduced to passivity on the part of the state. The analysis of labour use in Chile in the 1980s Institutions and human resources in Chile The literature has, however, not placed much emphasis so far on the structure of labour and welfare institutions as a factor in the development of both social citizenship and human resources in Chile. It has been assumed, to some extent rightly, that the labour market was simply ‘free’, and hence firms dictated human resource development, while state institutions dictated welfare and citizenship. For example, Bosworth and Marfán and others have argued that quantitative factors dominated labour use in the 1980s, which makes sense, given the ability of the private sector to reabsorb labour following the crisis of the early 1980s, when open unemployment came close to 30 per cent. The fact that labour costs tended to drop significantly in the crisis years (see Table 2.1) supports the supposition that by the early to mid-1980s the room for ‘easy efficiency’ (the removal of excess labour, the coverage of ‘dead’ time with existing labour and so on) had been filled. 48 Consequently, output growth in

Human Resources and Market Reforms 35 Table 2.1 Wages in costs per sector (VBP – VA + wages/wages)* 1976–89, and investment as a percentage of value added in enterprises with over 50 workers Industry sector

1975 1980 1985 1989 Investment Investment as % of VA** as % of VA***

311–2 321 322 323 324 331 332 341 381 382 383 384

15.5 38.4 31.3 18.6 35.1 41.3 39.2 23.2 35.9 35.4 42.4 51.2

Foodstuffs Textiles Confectionery Leather Footwear Wood Furniture Paper cartons Production of metal Metal machining Electrical Transport equipment

15.9 34.7 29.6 17.4 26.6 29.4 31.4 17.5 23.8 36.4 26.0 49.0

9.1 16.1 18.9 11.2 19.6 15.1 20.2 10.1 18.4 18.9 21.9

11.6 19.2 20.8 10.5 12.9 12.6 21.5 9.5 17.4 24.7 12.2

12.6 9.2 3.4 6.1 3.5 5.6 7.8 9.1 11.6 12.6 6.6

13.2 16.1 9.0 4.9 9.4 17.6 7.7 16.1 10.6 22.5 17.7

Notes: *VBP = Valor Bruto de la Producción (gross value of production); VA = Valor Agregado (value added); ** 1979; *** 1989. Source: Calculated from INE – ‘Tomos Industriales’, various years.

the remainder of the decade was sustained in large measure because of a quantitative extension of labour use. Overall, productivity did not improve at all throughout the 1980s, and Chile compared un-favourably with the rest of the region (see Table 2.2). Productivity growth continued to be very unstable in the first half of the 1990s.49 In general, Bosworth Table 2.2 GDP, labour and capital inputs and sources of growth in output per worker; Chile compared with average of four regions, 1950–89

Labour quantity Physical capital Labour quality Output per worker

1973–80 1980–9 1973–80 1980–9 1973–80 1980–9 1950–73 1973–89

Chile

Latin America

Asia

Korea

OECD

1.88 2.93 2.34 1.88 0.94 0.96 2.87 0.63

2.85 2.58 6.44 3.41 1.05 1.43 3.32 0.17

3.21 1.72 10.79 8.33 2.01 2.61

3.74 1.54 15.50 11.83 1.94 2.85

5.09

5.34

−0.22 0.50 4.81 3.54 0.40 0.63 4.82 2.23

Notes: *Latin America (Argentina, Brazil, Colombia, Mexico, Venezuela); Asia (Korea, Taiwan, Thailand); Iberia (Portugal, Spain); OECD (France, Germany, Japan, Netherlands, United Kingdom, United States). Source: Selected figures from Bosworth and Marfán (1994), pp. 170 and 211–12.

36 Theoretical Issues

and Marfán emphasize the low regard for labour quality in Chile. In particular, they stress the relatively greater emphasis on labour quantity compared to a greater emphasis on physical capital investment in Asia. The data, however, does not in itself provide any clues as to the underlying institutional bases behind the relationships posed. First, it seems obvious that the lack of incentives in Chile towards physical capital investments by the state, as explored by Pietrobelli and others, naturally also affected the propensity to invest in high-level technical skills. But to be satisfied with this as an explanation for the low emphasis on developing human resources generally we would have to accept the neo-classical position discussed above that institutions to develop human skills only emerge after technological breakthroughs as defined from the outside. 50 Already, the incorporation of initial schooling as a factor of technological innovation in new growth theories has undermined the position that human development follows growth ex post. The neoclassical position therefore now rests on a strict distinction between initial schooling as an ex ante factor and the labour market and attendant institutions of learning and labour recycling as post hoc or external factors. Even supposing that this division between initial schooling and the labour market can be maintained, we are back to the question of why liberalized markets and a balanced macro economy were not sufficient to lead the private sector to innovate spontaneously towards higher valueadded activities, given the available schooling rates. We would also have to explain how a large share of unfulfilled demand for training existed in Chile. But there is also the question of whether it is analytically sound to consider enrolment rates (initial schooling) as the only ex ante important facet of labour quality. It is often difficult to establish categorically which came first (investments in physical or human capital) even for East Asian countries such as Korea, where the creation of basic technical schools, from the early 1960s onwards, pre-dated the stage of rapid growth of the 1970s. Even in liberalized labour markets, such as Chile’s, it may be hypothesized that the experience of local unionism may make the labour management of firms diverge. In addition, firms may respond differently in their management of human resources, depending on the external institutions in which they engage – for example, whether a market in recognized training experience exists in their sector. In the case of Chile, school enrolment rates do not in themselves fully account for the hypothesized tendency of firms to rely on labour quantity (the simple extension of labour use) in the 1980s. From Table 2.2, it can also be seen that Korea had high levels of labour quantity

Human Resources and Market Reforms 37

relative to quality in the period when it industrialized heavily, (the 1970s) and clearly relied on the expansion of a skilled labour force. 51 Moreover, Chile has been catching up with Korea – as far as simple enrolment rates are concerned – albeit slowly. Chile fares well in particular in primary education, with enrolment rates higher than Korea’s, though Chile falls behind at the secondary and tertiary levels.52 This suggests that enrolment rates alone are not sufficient to explain how far the private sector relies on skilled labour, and takes part in its development. Finally, good enrolment rates do not in themselves say much about the quality of the education provided, or indeed of its connectedness with private sector institutions. As Angell has discussed with regard to Chile, the key task that democratic governments faced after 1990 in primary education related to bringing under-performing schools up to a certain standard (Angell, forthcoming). As we shall discuss in Chapter 7, the main problem in both primary and secondary education in Chile was its varying quality. This, in turn, can be linked to the structure (of finance and administration) of the system, an aspect that the simple enrolment rates do not show. In summary, although initial education is the proxy most generally used (in statistical studies), it is clear that little is said in this respect about the provision of workers’ training at the point of entry to, and especially once a part of, the labour market. A related problem is clearly the shortage of adequate statistics in this field. Pietrobelli cites vocational education enrolment in the population, where Chile has caught up somewhat relative to how far it was in the mid-1970s behind countries such as Korea. At that time Chile had, according to Pietrobelli, about a tenth of Korea’s population in vocational education, whereas in 1986 it had a half. However, even these statistics can be very deceptive, when not taking into account the design of training systems. In the case of Korea, initial vocational education tends to be orientated towards lower-to-middle-ranking salaried or blue-collar workers, many of whom are trained in state-certified schools.53 In Chile, vocational education statistics of publicly-financed courses (see Chapter 7) can encompass a large proportion of management training. In summary, the weakness of training statistics links with a different shortcoming in the existing literature on the development of human resources, that of institutional and political study. Apart from the question of total investment levels in workers’ skills, which is as yet quite impossible to measure, there are the attendant political and historical questions as to why and how investments in workers will tend to be made or not, and when and how workers themselves are able to enter a bargaining position with respect to the organization of work. For example,

38 Theoretical Issues

how far was workers’ institutional defencelessness against short-term labour contracts a significant factor in explaining the persistence of such contracts in Chile of the 1980s? How far was institutional deregulation important to the persistence of a style of growth based in simple numerical flexibility, in the ability to shift the quantities and cost of labour quickly, rather than in a deepening of labour quality? The analysis of Bosworth and Marfán indicates that institutional changes in labour relations may have been crucial, since the ability of employers to reincorporate labour on their own terms in the 1980s is an implicit factor in their analysis. However, it remains unexplored because of the exclusive focus on initial education and productivity outcomes. In general, the roles which labour relations and labour market institutions play in the development of human resources deserves more analysis. For the purposes of our discussion of occupational citizenship we can distinguish between two areas of labour reform undertaken during the Pinochet years. The first includes reforms to labour relations, with the expansion of employer prerogatives and the second covers deregulation of occupational services and a changed role for the state. The constitution of a liberal industrial relations Core labour institutions, such as collective bargaining, have traditionally been decentralized in Chile. Moreover, most union functions, even prior to 1973, took place at this level. Nevertheless, the Pinochet Government sought to atomize the institutions of labour further. The purpose was to create a perfect market in labour and establish the supremacy of firm managers in decisions about human resources. Prior to 1973, unions in Chile had acquired quite extensive legal rights, beginning with the 1925 Labour Code. Legal measures had been particularly important because of the relative weakness of unions in the local firm structure of industrial relations. 54 Shortly after the coup, collective bargaining was suspended followed by limitation on the matters legally accepted at union meetings, and measures to supplant old union leaders at firm and sectoral levels. The practice of bargaining above the firm level was a political reminder of the Allende period, when negotiations on minimum wages were imposed in a number of sectors. Institutionalization of a new labour market structure began in earnest in 1978. The measures, popularly known as the ‘Labour Plan’, introduced a new fragmented union and collective bargaining system with the 1980 revised Labour Code. 55 A successful attempt was made to weaken the presence of the unions in Chile, in the quantitative sense. The rate of unionization oscillated around 11 per cent in the early 1980s, down from

Human Resources and Market Reforms 39

its historical level of around 30 per cent. By 1991 it had climbed to some 15 per cent, to fall slightly thereafter. 56 Taken together, the labour reforms of the late 1970s onwards can be seen as an attempt to reproduce something close to the economic component in civil rights in the Marshallian sense. This is a kin to the classical liberal system of property rights, what Miller has referred to as the liberal or ‘proprietary’ notion of property relying on ‘unrestricted control’, ‘where a person is seen as a self-determining agent controlling his material environment and recognizing no obligations to others around him’ (Miller, 1990, Chapter 1). Essentially, this is the notion whereby property rights over resources become concentrated to a maximum degree, through legal and other means, in the owners of physical capital. A relevant contrast can be made with other national systems of governance where this legal ‘personification’ of the firm, as Chomsky has called it,57 has been countervailed by the development of other forms of property right, principally of the agents of human capital, namely labour. For example, this process has taken shape in certain Scandinavian or East Asian societies, principally by making individuals the key agents demanding from integrated systems, in one form or another, of social welfare.58 The neglect of the development of the markets and the property rights of labour in Chile had a passive form (as in the structuring of public finance for training), and a proactive form, in terms of maximizing limited and unsophisticated forms of contract in the labour market, what Williamson has referred to as the presentialist or classical form, which ‘favours escape clauses, limited commitments and short-term relationships . . . to cause to be perceived or realised at present’.59 Such contracts curtail the voice of workers, the first element in developmental freedom. Examples of the latter in Chile’s revised 1980 Labour code included: (i)

(ii)

The development of ad hoc and parallel forms of bargaining entities inside firms, to seek to replace the old ‘one union per enterprise’ structure.60 Dilution of the substance of consultation and bargaining with workers. This included the abolishment of negotiation on any matters which infringed the employers’ right to organize work (Article 12), and the limitation of strikes to 60 days’ duration. Employers were allowed freedom to hire and rehire during strike action (Decree Law 2758). Temporary workers, and workers in agriculture, were effectively excluded from bargaining; and profit sharing, a traditional form of participation in the firm, was abolished.

40 Theoretical Issues

(iii)

(iv)

The promotion of precarious contracting. For example, Decree Law 2200 replaced Vols I and II of the 1925 Labour Code; and this and associated measures gave employers the facility to modify work contracts without consultation. Dismissal without the need to state a cause was introduced, and rights to indemnity reduced, especially for part-time and temporary workers. The prohibition of regulation of the working day and/or night as well as the workload. In this case Decree Law 942 (Article 3) of May 1975 made any agreement on the limitation of daily work illegal. Article 345 (following the ILO’s Article 20) prohibiting night work between the hours of 9.00 pm and 4.00 am, was also repealed.

In short, the framework for labour flexibility which was established in the 1970s tended to promote the heavy use of what the OECD refers to as numerical flexibility forms. This includes the emphasis on extracting, and manipulating, physical qualities, including work-time, pay, mechanical skills and so on.61 Following our earlier argument, the problem was not that new flexibility measures were introduced. Rather it was the lack of parallel frameworks to negotiate trade-offs at the firm level (for example, pay restraint for job security) or an external framework to provide greater sources of secure flexibility for workers. Deregulation of labour market services The passive measures with which labour’s access to secure flexibility was curtailed encompassed the establishment of labour market institutions that weakened the links between services on which waged and salaried workers rely. This harmed the prospects for occupational mobility, the other element of developmental freedom. The relevant measures included: (i)

(ii)

A further segmentation of the educational system. In the case of Chile, the Pinochet Government introduced a market-based design, and used financial decentralization to encourage competition between local privately-run and publicly administered schools. This design produced systematic inequalities in educational attainments with the development of ‘waste-basket’ schools within each municipality62 (Chapter 7). In turn the bases for establishing a broad link between initial education and further training were weakened. The promotion of short-term contracting in the training market comprised a second key feature of deregulation. This was done by centring the power to demand within individual firms, in a context where industry and the state play no role in the setting of

Human Resources and Market Reforms 41

(iii)

(iv)

standards. With Decree Law 1446 in 1976, the Pinochet regime created a new national system of professional training, and a year later a public supervisory body was set up, the Servicio Nacional de Capacitación y Empleo (SENCE), essentially to distribute public funds. By 1985, 90 per cent of public funds for training were channelled through the so-called tax rebate scheme, which permitted firms to claim back one per cent of their labour costs from the state for the purpose of providing training for workers. In this way, employers, not workers, were constituted as the training markets’ central consumers. A third feature of deregulation was the leasing-out of training services on short-term contracts to a multitude of private providers. After the coup, the state-owned Instituto Profesional y Centro de Formación Técnica (INACAP), the largest training corporation in the country, was passed on to the national entrepreneurial association, the Confederación de la Producción y el Comercio (CPC). In total, more than seventy centres for worker training were transferred to the private sector, and by the late 1980s some 400 effective corporations existed.63 However, no universal certification system was put into place. Public programmes using the institutes relied on short-term contracts. One of the longest state-provided contracts to private institutes, through the Youth Training Scheme (YTS) in the 1990s, ran for only six months. Another important feature of the atomized labour market was the expansion of individual-account social insurance without links to other forms of state-guaranteed welfare provision. In the 1980s, as was noted in Chapter 1, the state’s welfare functions came to be identified increasingly with marginality, as the distinction between welfare and social insurance was scrapped.64 The two alternative avenues for secure mobility that workers have, skills and social assistance, were split.

There was a liberalizing trend within the training market across the Latin-American continent. For example, tax rebate schemes were also established in Brazil in 1976, and in Argentina in 1981. Yet, in contrast with Chile, where all traits of tripartism in the training market were purged, the authorities in these countries retained the possibility of planning and targeting training provision. In Chile, the short-termism of training finance, together with the centring of ownership over skills in firms, meant that no incentives were provided for developing the common, tradable standards in work experience and skills on which secure mobility for workers depends. Other relevant contrasts to note

42 Theoretical Issues

are the cases of Singapore and Korea, because of their success as export economies in the 1970s and 1980s. During this time Singapore also had a mixture of public and private training institutes (as in Chile) which competed to some extent for public funds. However, ministerial initiative remained important in the setting of national manpower targets. In addition, specialized ministerial bodies benefited from a high degree of co-ordination of labour market information, which the tertiary institutions were compelled to provide. 65 In the case of Korea, vocational training centres were established nationwide as early as 1967 (modelled on Germany) which provided training free of charge to middle-school graduates and high-school drop-outs, as well as high-school graduates at different levels, who would then gain some share in a tradable skill. 66 Even as labour market flexibility was expanded in the 1990s, firms were charged heavier indirect levies (through the employment insurance scheme) to help finance retraining schemes offered to unemployed workers.67 In both countries, then, the state’s strategic co-ordination of training provision was based on an understanding of skill as the ‘engine’ of growth. Within this set-up the demand for skill had, if necessary, to be supported by supra-firm bodies.68

Conclusion It would be difficult to dismiss the observation that a broader political rationale had informed the Pinochet Government’s programme of economic reforms. A key objective of these reforms was to achieve economic opening and macro-economic stability. But in addition a principle of nonintervention was applied. This contained three basic characteristics. These were (i) a rejection of development planning; (ii) no role for interest representation; and (iii) the isolation of the management of individual firms. The poverty of labour market institutions was the issue that linked each of these features, and which compounded the problems of asset specificity associated with investments in human resources. In some ways this shortcoming folds with the political challenges associated with the transition to democracy in Chile (see the Introduction). An expansion of public authority in each of the three areas listed was precisely what the labour movement and transition government had set out to achieve. Nevertheless, whereas there is little shortage of documentation on the general decline of labour institutions on the one hand, and on the performance of the Chilean economy on the other, it is difficult to find approaches which seek to integrate the two. Efficiency in the usage of human resources was assessed in two principal ways by Bosworth and

Human Resources and Market Reforms 43

Marfán, in the rate of labour productivity growth and in capital–labour substitution. However, labour productivity is not always a good indicator of the actual quality of labour use. Often it can conceal a simple extraction function, as it did in Chile in the 1970s.69 Capital–labour substitution may be a better indicator, because at least it suggests the presence of technological up-grading that presumably will require the parallel acquisition of workers’ skills. Nevertheless, this indicator could also conceal a situation in which rigid and hierarchical labour management precludes dynamic learning functions. In other words, we are back to the question as to which form of labour flexibility is being pursued, and the factors that allow this form to persist. This issue is at the core of our inquiry into the development of occupational citizenship. In this chapter and in Chapter 1 we have introduced two levels of analysis. First, we looked at how the national framework of economic governance develops. Second, we discussed the specific design of the institutions of labour. At the time of Chile’s transition to democracy, the shape of labour institutions became a core issue of citizenship and a test-case of how far the national framework of economic governance could and should be modified at this crucial point in the nation’s politics. We shall argue that the structure of free labour markets, together with the political barriers against associational practice were also important factors that actively constrained the development of human resources. Modifying Chile’s market institutions was both politically and institutionally difficult, once economic liberalization had taken shape. In the changeable period of the formation of market institutions, the success of any state initiative will depend on how the private sector is already organized, and how it perceives its private domain relative to that of the public. The period that we study in Chile was one such transition phase. Democratic governments were clearly constrained in terms of what they could do, even in what they proposed to do, by the organization of the private sector. In short, we might say that occupational citizenship (and the idea of the mainstream in educational ‘after-life’) demands a complex form of stateness or public sphere, or a strong connectedness of the social and educational services with market institutions. The problem in a country such as Chile was how to rebuild and reconnect social services following privatization, leasing out and market restructuring, without such connections collapsing into the multiplication of short-term relationships between service agents, individuals and firms. Specifically, our thesis is that for labour markets to be in reality fluid, a much tighter network between social and labour institutions would be required than that which the latest generation of market reformers had sought.

This page intentionally left blank

Part II Labour Politics and Chile’s Transition

This page intentionally left blank

3 Business Leaders and Democracy’s Limits

Entrepreneurial leaders in Chile came to play an authoritative political role during the country’s transition. This role appears to be a paradox at first glance, since business leaders were not active supporters of parliamentary democracy. However, the question as to which are the decisive social forces in pushing through a democratic regime is quite different from the problem of who become decisive in shaping it. In the first place, the specific modalities of the Chilean type of transition, its pacted nature and the opposition to democracy of key elites, 1 makes it possible that weaker actors will refrain from pressing their interests for fear of reversing the process. On the other hand, stronger actors may try to use pact-making to institutionalize the exclusion of large spheres of economic governance from the public domain. These positions describe the situations of union and entrepreneurial leaders in Chile. But in order to understand the eclectic attitude of entrepreneurial leaders in Chile to political democracy, two further factors need to be noted. First, the importance of the unifying ethos of nonintervention which, as a development ideology, had strong international support. And, second, the long experience that entrepreneurial leaders had with political democracy, including the way this had shaped their internal organization and attitudes to labour relations. We disagree that a ‘distancing of economic life in general from the political world’ had taken place by the start of the country’s transition to democracy.2 This view can be seen as representing a tendency in recent scholarship about entrepreneurs in Chile to take the economicscientific reasoning behind the liberal reforms at face value. It is implied that an extended pragmatism exists in the relationship between entrepreneurs and the state, and between entrepreneurs and other forms of public institution that the state protects, including trade 47

48 Labour Politics and Chile’s Transition

unions. Such pragmatism has been ascribed to countries such as Japan or Germany where, according to some authors, the state’s involvement with the private sector was as such not a major historical issue.3 But this does not fit the Chilean case.4 In particular, we wish to highlight the long-standing dissociation of entrepreneurial associations in Chile from social questions. The neo-liberal programme only exacerbated this tendency. Distinguishing between what was tactical politics and what was political ideology within the entrepreneurial sector in Chile may help us to establish this point. Below, we shall first discuss general factors that help in understanding the role of leading employers in the transition. The strength of the authoritarian coalition provided entrepreneurs with a solid platform. However we also need to set out the historical reasons why employers could unite around this coalition, and indeed represent one of its most lucid ideological sponsors. Lastly we look at the practical role of the CPC in the early phase of transition, and at how internal divisions on labour relations in some ways served to aid the Confederation’s cause.5

Background The longevity of the authoritarian coalition The continuance of strong anti-democratic views on economic governance in Chile, even after 1990, is explained partly in the solidity of the authoritarian coalition. This encompassed the military, the parties of the right, entrepreneurial leaders in the CPC, and the Pinochetista press. 6 The two new parties of the right, the Unión Democrática Independiente (UDI) and Renovación Nacional (RN), owed their senate majority to the binominal system designed by Pinochet, and the senators designated by him, some of whom were to play a key role in the negotiation of the labour reform. 7 The entrepreneurial community’s early support of Pinochet meant that its leaders were in a strong position to influence the political right later. The strength of the entrepreneurial community at the time of transition can be ascertained from what the authoritarian coalition was not willing to include as an aspect of pact-making. For instance, the formal pacts between opposition and authoritarian government prior to 1990 did not encompass significant concessions on socio-economic affairs. The same was reportedly true for informal talks at this time. 8 This contrasts with other pacted transitions (such as those in Spain after 1975, and Venezuela after 1958), where business was weak relative to the political class. Chilean entrepreneurs were not only well organized, but also politically motivated to an unusual degree.

Business Leaders and Democracy’s Limits 49

The political insertion of employer organizations The embrace of an individualist ethos, from the mid-1970s onwards, by leading Chilean employers, was influenced by the liberal ideology that came to inspire economic reforms. Yet it is equally important to emphasize the long-standing political legacies that mitigated against the development of a corporatist understanding of employer representation among the country’s largest capital holders. It is a well-known fact that national models of industrial relations are often heavily influenced by developments within manufacturing industry. The Industrialists’ Association, the Sociedad de Fomento Fabril (SOFOFA) was not the oldest employer association in Chile (see Figure 3.1), but it was part of the first crucial wave of employer associations, and arguably it came to play a dominant role within the national federation, the CPC. Two aspects in particular are important to stress in explaining the abortive social history of industrial relations in Chile. The first is the historical separation between big business (including the SOFOFA) and the organization of small and medium-sized entrepreneurs (see Figure 3.1). The earlier inception of the organizations of big business meant that their leaders were in a powerful position to forestall the growth of labour market institutions at the bipartite level. The CPC’s member associations were not established in response to the rise of unions, as were the separate associations catering for smaller firms, a movement that occurred half a century later. The second factor to note is the early integration of the associations of large capitalists as part of the Chilean political class. This occurs before the broad extension of political franchise, a point stressed in Chapter 1 as being particularly important for the development of social citizenship. Moreover, it happens almost half a century before the formal inclusion of unions, in 1925. Entrepreneurs, in short, were used to seeking solutions to social problems through political (and, if necessary, military) supremacy.9 Hence, the coexistence in the early parts of the century, as highlighted by Valenzuela, between political inclusion (political citizenship) and social exclusion (the harsh repression of unions), 10 gives us some clues to later developments: for example, how neither the introduction of a labour code in 1925, nor the existence of successive left-centre governments in the decades that followed, ushered in a comprehensive process of democratization in the field of labour relations, local or national, until the 1960s. In this interpretation, reformist government did not develop on the basis of social compromise, but existed despite of its absence. 11 Employers’ early integration within the political establishment had also set a certain precedent to their later search, in the late 1950s and early 1960s, and again in

50 Labour Politics and Chile’s Transition

Small and medium-sized firms1

Medium-sized and large firms

CPC

CPTC

Membership in 1984 Membership representing almost all large firms in Chile:3 estimated at 40 815 firms, 6-member board composed of the heads of each of the estimated in 1984 at 180 0002 six member organizations. All decisions are taken by consensus.

Asociación de Cámara Chilena de la SNA SONAMI Cámara Central SOFOFA Bancose e de Comercio Construcción (Construction) (Mining) (Commerce) (Industry) (Agriculture) Instituciónes Financieras 1858 1838 1951 1883 1883 (Banks) 1943

Membership in 1984 estimated at 6500 of a potential of 9962 firms Executive Council 58 members

CONUPIA (Industry) 1970 Membership estimated at 15 000 of potential in 1984 of 35 000

Direct members

Indirect members

559 large firms (1992)

6 regional organizations + 23 sectoral associations

It is estimated that these represent close to 90% of industrial value added4

ASIMET (Metal) 1938 247 members in 1989 (of 784 firms in 1989)5

Figure 3.1 Industrial employers within the organization of Chilean business Notes: 1 The dividing line between the two groups may be set at 50 workers, but this is not strictly observed. 2 Of an estimated potential membership of 538 757 (Camper (1984) Sectors of relatively small-scale producers/businesses include (i) Agriculture (CPA, 1973); (ii) services and vendors (CGCDPIC, 1938); and (iii) transport (CSNTT, with origins back to 1918). 3 Campero (op.cit., p. 319) estimated 44 518 potential member firms in the CPC in 1984. According to CPC leaders, membership had not changed significantly since the mid-1980s. (interview with Manuel Feliú Santiago (14 Septemer 1992)). 4 SOFOFA membership list (September 1992) and interview with Jaime Alé Santiago, (8 September 1992), Head of Studies of the SOFOFA. 5 ASIMET (1992a), p. 12, and membership list ASIMET (1992b).

the 1980s, for a hegemonic role within Chilean politics. But if employers were by the 1960s already well established as ‘political animals’, 12 what was the nature of leadership within this group?

Business Leaders and Democracy’s Limits 51

The existence of a politicized leadership elite with a high degree of autonomy must be considered as a key factor behind the CPC’s later role in support of market neutrality, particularly in light of the extreme hardship that affected the majority of Chilean firms during the first reforms. The analysis by Rozas and Marín indicates that economic diversification occurred mainly from within the established economic groups represented in the CPC.13 However, it is particularly important to emphasize the continuity in political leadership. Entrepreneurial leaders interviewed in 1992 broadly agreed that the traditional associations, the SNA, and particularly the SOFOFA, increased their influence in the 1980s in the CPC.14 Certainly, it was the traditional sectoral associations who concentrated the economic power behind the CPC front. Next to the stately mansions of the SOFOFA and the SNA, the CPC was a poor organization, with its rented headquarters in an ordinary office building and a budget representing less than 10 per cent of that of the SOFOFA, according to one informed estimate. Important individuals and areas of economic activity were not directly active in the CPC or represented on the board. These included a number of primary export sectors and the new private social security companies. On the other hand, it has been noted that the low-key existence of unrepresented groups in some ways consolidated the traditional political voices in the CPC, in so far as the CPC was still uncontested as a political medium.15 Other entrepreneurial leaders suggested that group inter-linkage meant that the absence of representation was a mere formality, and that broad agreement existed on the CPC’s position on policy issues. On the other hand, the point hardly needs to be stressed that economic power cannot be distilled into political clout, or vice versa. In other words, the influential political position of the SOFOFA was maintained through a certain political seniority that it had developed within the fraternity of the CPC. Indeed, while CPC decisions had to be weighted between six executive board members (of which SOFOFA was only one), influential entrepreneurs from various camps interviewed in 1992 were almost unanimous in their depiction of the SOFOFA’s power of veto there. This veto was particularly felt on issues of non-intervention. A decision-making style of consensus in the CPC had allowed the SOFOFA to act as a ‘rebel’ within the CPC and ‘a headache to all CPC presidents in the 1980s’.16 Elitism and continuity in leadership inside the SOFOFA itself are also important factors in explaining this organization’s forthright style and dominant role in the 1980s and early 1990s. The association was driven by political leadership, rather than by bureaucratic structure. The

52 Labour Politics and Chile’s Transition

SOFOFA’s key strategist and spokesman, Jaime Alé, highlighted the influence of old family ties and a ‘cultural heritage’ among representatives of key economic groups, through which the Paper Company (CMPC – the Compañia Manufacturera de Papeles y Cartones) had spun an inner circle of political activism and influence within the SOFOFA in past decades. Jorge Alessandri, in addition to his presidency of the country (between 1958 and 1964) and of the SOFOFA, during the previous eleven years, had been president of the CMPC, and CMPC board members were said to have a dominant voice in the SOFOFA. 17 The Paper Company was also said to be one of the five principal donors to the SOFOFA,18 although in the light of the above information it would be difficult to ascribe the influence of its associates to this fact alone. There is no doubt, however, that the largest member firms played a dominant role in the Association’s finance, and that this fact had some bearing on their standing within the group. 19 This may also be deduced from the internal leadership structure (Figure 3.1). Employers’ views on democracy and social organization By the 1990s, leading entrepreneurs’ acceptance of political democracy was possibly more conditional than it had been even in the pre-coup period. In fact, it is fair to say that up to the mid-1960s, Chilean employers had been exceptionally successful in protecting their interests within the democratic regime. What was different about the political engagement of the 1980s and early 1990s was the intensity of purpose to ensure that this time political democracy would really not ensue in a process of social democratization. Hence an important aspect of a new identity was the reassessment of past experience with democratic rule. It came to be widely held, for example, that import substitution industrialization since the 1930s had relegated the entrepreneurial class to an economically dependent and politically defensive role. Entrepreneurs had clearly benefited from protectionism, but this experience tended to be recollected as somewhat shameful. Eugenio Heiremans was a prominent corporatist within the entrepreneurial community of the 1960s, and later became one of the strongest neo-liberal voices from within the SOFOFA. As he put it, ‘The Chilean entrepreneur was lethargic in mentality and instead of worrying about efficiency in his enterprise was raised to seek paternalism from the state’. 20 Leading employers also clung firmly to the view that ‘conflicts in labour relations have always been induced by politicians’.21 Montero herself explores how the behavioural pattern of the ‘new entrepreneur’ in Chile was based on the adoption of a pure market definition of modernity, by which

Business Leaders and Democracy’s Limits 53

productive activity itself and factors such as rational calculation and predictability were regarded with open disdain.22 A similar emphasis was apparent among CPC leaders interviewed in 1992, who made it a point to define themselves against other employers’ leaders in what they saw as their own adhesion to true economic liberty.23 The general political outlook among entrepreneurial leaders had clear implications for their approach to labour and the development of human resources. Corporate initiatives in the area of skill development were never strong in Chile, compared with Brazil, Argentina and, of course, Germany, which in the 1990s became the model policy-makers, hoped to emulate. 24 There were also structural factors behind the lack of attention to skill development among the central entrepreneurial associations in Chile, including, as we shall later observe, the absence of trained personnel and an apparent lack of interest in supra-firm activities within individual firms. Hence the importance of the negative position adopted by leaders in associations like the SOFOFA on issues of formalization of markets in skills. The CPC had at its disposal the institutional means of deepening concerted activities between firms in the area of skill development; indeed, some of the most important and publicly-funded training corporations in Chile had been under the control of the SNA and the SOFOFA since the mid-1970s. Yet these were regarded as independent profit-making bodies, no more. For example, the CPC or SOFOFA were not involved in raising private funds for these bodies. It was thought that no mechanisms of planning or strategic evaluation, no functional ties, should develop between the private training centres and the SOFOFA. The Association emphatically rejected a role as provider of technical advice and assistance to members: ‘This has been a conscious decision. We have always viewed such activities as something remote from us.’25 This was based on the belief that workers’ training had to be guided by the activities of individual firms, and by the demands of these firms. This orientation also helps to explain how the SOFOFA spent a considerable amount of its annual budget on public relations, compared with insignificant sums on specific assistance to members (see Table 3.1). For example, the Association spent nearly 52 per cent of its budget on personnel and almost 40 per cent on various public events and promotional activities in 1991. But the point to note is that the core entrepreneurial leaders in Chile, and particularly those in industry, were concerned with ideological questions of market doctrine, and not with pragmatic aspects of industrial relations. And this had direct implications for the nature of their concerns and activities during Chile’s transition.

54 Labour Politics and Chile’s Transition Table 3.1

Annual budget of the SOFOFA, 1991 (pesos 1992)

INCOME Membership fees Direct members Indirect members

EXPENDITURE

236 895 364 9 986 126

Other Current Income Fairs & expositions Industrial Board

92 583 087 8 663 667

Non-operational Income Fees from earlier activities Interest and adjustments Investments Other income

1 193 184 7 019 236 5 449 991 12 301 738

Total Deficit

374 092 393

Personnel Office needs

198 367 371 31 926 421

Transfers to related organisations Public opinion Fairs and expositions Presence abroad In relation to members

13 666 421 4 385 188 67 877 571 3 836 383 47 011 623

Social programmes

13 933 484

Depreciation Total

4 532 135 385 536 597

11 444 204

Source: ‘Balance general al 31. de Diciembre de 1991’, Sociedad de Fomento Fabril (SOFOFA), F.G.

The CPC in transition politics From 1986 onwards, the yearly entrepreneurial gatherings, the so-called ENADEs, came to represent one of the main political events in the country. The CPC also acted to organize direct political support for Pinochet in the forthcoming plebiscite, with the formation of the ‘Civic Committees’ in August 1987.26 In SOFOFA, this shift is exemplified by the decision of Ernesto Ayala to become directly involved in entrepreneurial politics. Meanwhile, entrepreneurial leaders commonly regard the election of Feliú to the CPC in 1986 as a qualitative shift towards a more forceful public role, against the ‘mediocre’ performance of the preceding decades. 27 According to the CPC’s president, the objective of a series of television commercials was to ‘make a national diffusion of the principles that inspire private enterprise. We wished to implant our opinion and insure that the virtues of the economic model would be placed above the change in political governance’.28 It is important to emphasize the CPC’s strong position within the right, because it was by no means obvious that the entrepreneurial community as a whole should be represented by the purist position on market neutrality. For example, the association of small entrepreneurs, CPTC, took up a more critical stance. This became diluted when the

Business Leaders and Democracy’s Limits 55

CPTC was integrated into the campaign entitled ‘Entrepreneurs for Development’, another campaign organization set up in 1988, by the SOFOFA. Other entrepreneurial initiatives that diverged from the orthodoxy of non-intervention, including the forum for Christian entrepreneurs (USEC), employers from the metallurgical sector, truck owners and small merchants were limited in number, and they did not have media access. On the other hand, they were weakened by their unwillingness to be seen as part of the opposition to Pinochet.29 The CPC also exerted its influence within the right at a time when its parties were still in the stage of formation. The entrepreneurial veto, for example, was a key factor in the selection of former Finance Minister, Hernán Buchi, as the candidate of the right in the presidential elections of 1989. Individuals supporting SOFOFA organized the supply of funds for the UDI, whereas the campaign for Jarpa in RN was allegedly starved of funds. 30 And yet, the relationship between leaders in the CPC and the right political parties was not an automatic one. Indeed, the political leadership role that entrepreneurial leaders assumed was hardly based on trust, but rather on the belief that ‘politicians’ are likely to ‘destroy the economic system’, with the tendency of ‘all politicians’ to ‘fall for economic solutions which are communist’.31 Hence the CPC did not leave the conduct of its diplomacy vis-à-vis the opposition to the parties of the political right. In order to understand the success of the CPC’s offensive, we need to examine the strategic significance of certain differences over tactics among the Confederation’s top leaders. ‘Closed’ and ‘open’ sectors in the CPC An invisible divide, a ‘civil war’ according to Feliú existed in the CPC between those leaders who feared the consequences of any contact at all with national unionism, on the one hand, and those, on the other, who not only saw this association as being inevitable, but who also felt confident enough to shape it. 32 The division of the most important ‘influential entrepreneurs’ around the CPC into an open and a closed sector is represented in Figure 3.2. The situation of the main exponents of a conciliatory position in key leadership posts was an important factor in the eventual adoption of an open policy. The entrepreneurs interviewed for this investigation all pointed to Feliú as the main protagonist. It was his personal initiative to set up a much-contested committee on labour relations within the Confederation, with the purpose of devising an offensive strategy vis-à-vis labour. Raúl García, vice-president of the SNA and responsible for

56 Labour Politics and Chile’s Transition

The ‘Closed’ Sector

The ‘Open’ Sector

Hernán Briones (SOFOFA president 1991–3)

Manuel Feliú (CPC President (1988–91) President of the private bank Banco de Concepción

Ernesto Ayala (CMPC Board President) Eugenio Heiremans (Founder and President of the Asociación Chilena de Seguridad – a private workers’ health insurance company)

Raúl García (Head of the Labour Relations Committee within the CPC during the Aylwin Government)

José Antonio Guzmán (CPC President 1991–) Represents the construction sector (Cámara Chilena de la Construcción) Pedro Lizana (SOFOFA Vice-President (–1993) (SOFOFA President 1993–)

Figure 3.2 ‘Influential entrepreneurs’ in Chile on the question of a national dialogue with union organizations

labour relations under Feliú and later CPC presidents, was a key ally. 33 At a time when the CPC could not ‘commit itself to anything because of the internal opposition we faced’, García was sent by Feliú to meetings in the Vicariate and elsewhere as a sounding board and ‘to maintain a nexus with the labour centrals’ on Feliú’s behalf.34 Díaz, head of the labour central’s main advisory body, the Centro de Investigación y Asesoría Sindical (CIASI), recalls a 15-day trip to Europe in 1987, organized by a Christian Democrat research institute, as his first personal contact with a member of the CPC’s inner circle: ‘Before we just sent tough messages through the press. Now Raúl and I could call each other and have formal meetings arranged.’ 35 Another figure within the core circle of CPC activists whom we need to mention here is Pedro Lizana, because of his position as vice-president of SOFOFA, the circle most vehemently opposed to Feliú’s forthright leadership style. Lizana was initially included through his university contact with the Christian Democrat intellectual Boeninger, the man who initiated alternative discussion forums. Through his participation in a number of seminars Lizana became convinced of the possibility of ‘defending the CPC’s position’ among the labour leaderships. In fact,

Business Leaders and Democracy’s Limits 57

the confidence of the open sector was no more evident than in García and Lizana’s conviction that the national union representatives could comfortably, even preferably, be made parties in the establishment of ‘liberal’ systems of workers’ training and unemployment insurance. 36 Tactics like these were not supported by other weighty sectors in the CPC. One top SOFOFA leader confessed to the inclusion within the SOFOFA Directive of the most ‘backward’ entrepreneurs, an observation shared by both labour and political leaders as well as Feliú himself, who recalled ‘ferocious battles’ arising from the resistance of successive SOFOFA presidents. 37 This sector was able to halt or delay a number of initiatives taken by Feliú and García in the 1980s. By way of example, it was resistance from the SOFOFA and the banking sector that at the last minute prevented Feliú from signing a symbolic accord with labour drawn up by USEC in 1988. Hence, the development of an assertive strategy to deal with the labour challenge was largely because of the leadership style of the CPC’s president. Feliú would ‘manipulate the consensus mechanism’ and ‘take the Confederation into arrangements with labour without necessarily having the support of the Board’. 38 In the case of the SOFOFA, Lizana would sometimes choose to ignore what in fact he knew to be the Board’s position. Reluctant backing was achieved ‘un poco con los hechos consumados’.39 The first formal accords with the Central Unitaria de Trabajadores (CUT – or United Workers’ Central formed in 1988) in late 1989 and early 1990 were arrived at in exactly this fashion. However, the divide that we have identified within the CPC’s inner circle must be characterized as one of strategy rather than substance. The central fears that the cautious sector entertained about the possible consequences of extending recognition to national unionism, and the ‘horrendous’ prospect of being drawn into industry-wide talks and a corporatist agenda,40 were shared by the open sector. García’s ‘greatest fear’ was precisely the negotiation by sector. Lizana noted how ‘in the SOFOFA, we are united in the rejection of conducting any labour relations at the sectoral level’. And, as Feliú himself put it, ‘Relations with labour at the sectoral level ? Never !’ 41 The purpose behind drawing in the labour centrals within a bipartite relationship prior to the transition was precisely to depoliticize the existing institutions of labour. And hence the shared view that ‘specific themes’ and ‘practical accords’ must be kept off the agenda. In García’s words, ‘bipartism must be kept sufficiently limited so that we do not interfere in the affairs of the enterprise. Nothing about working conditions, nothing about wages’. 42 With this rejection flowed the very

58 Labour Politics and Chile’s Transition

themes of economic modernization with which the CPC political class identified, such as flexible working methods and variable wages. The first direct meetings between CPC entrepreneurs and labour leaders took place at a time when economic recovery had not yet manifested itself. Meetings with the Central Nacional de Trabajadores (CNT) in December 1984 and January 1985, as well as a more formal meeting that followed in July 1985, did witness the participation of leaders from both strands within the CPC. 43 But entrepreneurial leaders were clearly uncomfortable with the equiparity that the format and substantive content of these meetings implied.44 At the July meeting, Manuel Bustos, leader of the CNT, linked any impending social pact-making to a tripartite framework.45 The reaction of the CPC’s cautious sector was to cut off further dialogue considering ‘conflictive themes’. No further formal meetings of this nature between the CNT (and its successor, the CUT) and the CPC board were to occur until late 1989, by which time the latter had gained the initiative. 46 From mid-1985 onwards, the Confederation presented its own separate project of social pact-making. Spearheaded by García, a first public meeting was held in August 1985, followed by two highly publicized gatherings in May 1987 and November 1988, under the leadership of Feliú. The smaller event of 1985 paraded ‘social understanding’ in the context of the ‘free-market economy established in Chile’. It allowed no discussion of specific themes. The document elaborated at the Jahuél meeting of May 1987 acknowledged the importance of worker participation in the firm, but omitted any reference to the role of the union. The CPC-sponsored meeting in the town of Ducal the following November characteristically had no set agenda. The outline of industrial relations The CPC’s task was not an easy one. The successive electoral victories of the opposition in 1988 and 1989 presented the Confederation with a problematic arena.47 It was at moments like these that the existence of an open sector proved to be crucial. In the months subsequent to Aylwin’s victory, the CPC entered into three arrangements, the symbolism of which had so frightened its guarded sector: recognition of the CUT; talks on labour reform; and a tripartite pact. 48 Each of these initiatives was based on the capacity for reasoning and the prior contacts that the open sector had formed. The first formal statement of mutual recognition between the CPC and the CUT, signed one week after the presidential elections, as well as the more substantive accord of 31 January 1990, were spearheaded by

Business Leaders and Democracy’s Limits 59

Feliú and García. And it was these two who ‘dragged the rest of the Confederation by the ear’ to the subsequent tripartite accord in March.49 It was the CPC’s open wing, indeed, which rationalized that the inevitable prospect of labour reform might be used to serve the Confederation’s own ends. Hence, ‘influencing the CUT and the Concertación directly’ represented the CPC’s best defence. Not surprisingly, the CUT–CPC commissions that were set up to study the labour legislation in March/ April 1990 staged the leadership of Lizana and García (against the reticence of a suspicious SOFOFA). The law commissions were hardly imposed on the CPC by the new government, because their establishment had formed part of the CPC’s earlier accord with the CUT. Thus Feliú’s observation, ‘The Government boarded the train which we had set in motion in January.’50 The agreement in March to a real rise in the minimum wage, and the parallel public commitment to a review of the Labour Code, both represented a keen sensitivity to the need of a successful transition government to differentiate itself on issues of poverty and workers’ rights. In reality, entrepreneurs never lost the initiative in social pact-making that they had established in the 1980s. The length and controlled nature of the transition process favoured the CPC. The accord with labour of January challenged a CUT which even CPC conservatives were convinced would at this point ‘sign anything to obtain recognition’. 51 Thus the first social pact of the incipient democracy established that national labour relations would concern only ‘general matters’, and that issues pertaining to ‘existing conditions of work’ would be ‘negotiated directly within the firm’. Existing property rights were affirmed, the ‘trickle-down’ effect was accepted as the means of improving social welfare, and the role of the state was reduced to that of a watchdog over ‘the stability and equilibrium of the macro-economy’. 52 In short, the CUT had encountered the extent of the possible at the CPC’s behest, and thus had liberated the Concertación from imposing a politically painful control. Entrepreneurs hence contest the view sometimes advanced by government functionaries that they developed the idea and format of pactmaking. Guzmán notes, ‘With the bipartite meetings with the CUT we confirmed what came before, the world as it had developed during the military regime.’53 Meanwhile, the idea of tripartite meetings were dropped after a few ineffectual attempts. The labour commissions prepared the political ground for a limited legal reform. This engagement not only committed the CPC politically, but also the CUT. As it turned out, entrepreneurs were later able to conduct direct negotiations with the

60 Labour Politics and Chile’s Transition

government on the points that appeared most sensitive to them. In summary, entrepreneurs from the open sector felt confident that with the first social accords of democracy they had averted the threat of national unionism and drawn the outlines of the labour institutions of the immediate future. This was certainly García’s view: ‘We are able to explain the convenience of the model to the labour leaders and their advisers. They too are rational, and hence they will appreciate.’ 54 Subsequent events proved them right.

Conclusion Both Montero and Abramo argue that the organization and political integration of employers in Chile were crucial to democratic stability in the early 1990s. Montero indicates how entrepreneurs had at this time assumed their ‘normal place’ in the national polity, with the redefinition of a demarcation between the ‘economic’ and the ‘political’. 55 Abramo makes reference to Brazil and the successful cases studied by Moore. 56 These arguments are relevant in key respects that we have highlighted in this chapter. For example, one can only speculate on the question as to the potential problems it might have caused for the transition caretakers had there not existed a group of politically astute, or what we have referred to as ‘open’, entrepreneurial leaders within the CPC. Johnson quotes Byron Marshall on the case of Japanese businessmen in the pre-war period. These had felt obliged to ‘denounce[d] the philosophy of economic individualism for its stress on the pursuit of personal gain, and claimed that Japanese businessmen were motivated by patriotic devotion and a willingness to sacrifice for the common good’.57 To Johnson, this quote reveals the more fluid relationship between public authority and the economy in Japan relative to that obtaining in most other countries. Be that as it may, a similar trajectory had been frustrated in Chile and in many ways the ethos of entrepreneurs in Chile had in the 1980s more in common with the North American tradition, where ‘The size, scope and domain of government is the most controversial and conflictual value question in [American] history and politics’, 58 than with the European continental tradition which Chilean politics before 1973 might at first appear to resemble. 59 The Chilean case after 1973, in particular, highlights the possibility of a process ‘in reverse’ from that indicated by Johnson: how the development of anti-state views in the private sector might affect the autonomy of a (new) democratic government to pursue substantial development

Business Leaders and Democracy’s Limits 61

goals. The key entrepreneurial leaders in Chile did not accept democracy as a form of government beyond the minimal regulatory state. This raises questions about their real endorsement of democracy in the sense of accepting the outcomes of the process of democratic governance. Although there were alternative employer traditions present in the realms of (at least) Chilean industry, it is true to say that these were prevented from finding political expression. Indeed, through the 1970s and 1980s, large business in Chile had finally managed to construct a working relationship based on unity between the SNA and the SOFOFA, to match the many underlying areas of coincidence of interest between these associations and their long shared history in Chilean politics. An indication of the unity and political strength of the CPC in the late 1980s was the fact that tactics in the approach to labour was the most important divisive issue within the organization, not issues pertaining to sectoral differences within the dominant class itself. And entrepreneurs no longer relied on the right political parties to express policy differences based on sectoral issues. 60 Instead they represented an independent and united front. The different positions on the labour issue were largely a matter of style, because there was a clear coincidence in the need for retaining a maximum level of atomization in labour affairs. Employers found themselves in a uniquely strong political position to defend their perceived gains within the former regime, and to forestall the replication of the classical (Marshallian) sequence, where social follows political democratization.

4 Unions, Parties and the Sacrifice for Democracy

The literature on transitions to democracy encouraged us to focus on the unions’ capacity to produce conflict. This was implicit in the focus on the unions’ restraint versus mobilization (basically, their likely impact on democratic stability). This in turn precluded a more in-depth look at labour relations and the factors that motivate organized labour. In Chile, for example, we do not see the bell-curved shape of mobilization that O’Donnell and Schmitter associate with the role of labour during democratic transitions.1 The concern with the ‘restraint’ versus ‘mobilization’ of labour leaders also appeared to support the unfruitful assumptions about the limited preferences of labour leaders for wage gain, and their tendency to pursue those at all costs, which characterizes neo-classical writing on unions. In fact, the importance of the unions was primarily their potential role as democratizing agents in the social and economic arenas. There are two basic reasons for emphasizing the unions’ potential role. First, the superficial analysis of outcomes might easily disguise the fact that strong alternative positions from those represented in framework agreements existed. But, second, an understanding of labour’s position will illuminate the importance of conflicts over social democratization to occupational citizenship and the development of human resources. What we need to reiterate here is that the expansion of occupational rights that either accompanies or precedes the development of human resources, is also conceivably a political goal to relevant actors. In times of transition, new rulers must either ‘reinvent’ or ‘reconfirm’ the rules that went before, or they must invent new and credible ones. Indeed, it is the idea that political democracy should ultimately contribute to 62

Unions, Parties and the Sacrifice for Democracy

63

expanding the means of individual liberty to an ever-increasing share of society’s members which makes the demand by workers to reform labour institutions at a time of transition to political democracy both legitimate and difficult to refuse by political elites. It follows from this perspective that the weakness of unions is also relevant, in terms of accounting for ‘lost’ opportunities for development and democratization. The general capacity of Chilean unions, in the early 1990s, to make valid proposals for reforming occupational rights is, then, the focus of this chapter. By ‘valid’ here is meant such proposals that had both universal applicability, at least in principle, as well as an element of strategic potential. 2 As one of the ILO’s chief policy advisers put it, ‘the key to union revival is acceptance of the flexible labour market as an economic reality’. 3 It would be reasonable to expect that the extreme degree of labour flexibilization witnessed in Chile would lead to the retrenchment of unionists into a call for defensive and parochial forms of employment security. If this were the case, then the unions’ potential influence on social democratization (as the expansion of universal citizens’ rights) in the Chilean transition would be limited. Much of this chapter, however, is devoted to an explanation of how a position of this kind was not adopted by the union movement in Chile. We shall stress two factors in particular. The first is the local-based structure and historical legacies of the union movement in Chile. The second is the prolonged and deep nature, indeed the universal impact, of the process of liberal reforms. In this and later chapters we shall focus primarily on the experiences of local leaders within light manufacturing firms. Our assumption is that informality had pervaded this stratum of enterprises just as it had other sectors of the economy (see Chapter 2). Furthermore, as we shall see, the national labour organization (or central, as it was known) was also seeking to define occupational rights within the terms of citizenship. The focus on political reforms that peaked at the time of transition served to strengthen this tendency. Hence the local-based struggles of the unions in the diversity of industrial enterprises towards formalization of employment arrangements were hardly incompatible with the universal definition of rights. In a sense, the debate on flexibility was captured by labour leaders (in what could be seen as a stubbornly optimistic pose partly created by the very arduousness of union office in Chile) as an opportunity to bring the tacit management strategies of employers ‘above the board’.

64 Labour Politics and Chile’s Transition

Unions before and during Pinochet Labour insertion and political parties The Chilean Labour Code of 1925 provided for a system of local bargaining, and relatively weak measures to encourage the economic independence of unions. The isolation of local unions made them highly dependent on the provision of assistance from political parties. In fact, labour protest remained, in many instances, closely linked to the latter’s activities. 4 Therefore, the social progressive projects that the left and centre parties pursued in the pre-coup period provided the Chilean labour movement with a degree of cohesion that it could not otherwise have attained. This included the formation of an extensive social security system with a basis in specific sectors or occupations, and negotiation at state level of aspects such as sectoral minimum wages and firm-level working conditions.5 On the other hand, this link did not necessarily diminish the local leaders’ isolation in the area of labour relations. Local leaders would also make a distinction between ‘their long-term political beliefs’ and their ‘conduct of day-to-day union affairs; or between the way they act[ed] in the political arena and the way they act[ed] in the economic arena’. 6 Angell’s analysis, interpreted liberally, suggests how far structures of worker representation were divorced from the mechanics of economic development in Chile – the small extent to which they were in fact a ‘social democracy’ in the Scandinavian form, where state reformism and class collaboration have gone hand in hand. Severance of links between national and local levels The special problem that labour presents to any authoritarian regime, its strategic economic position and its potential role as a political opposition platform,7 was arguably dealt with in the most effective way possible by the military authorities in Chile: by legalizing a segmented union structure, as examined in Chapter 2. 8 Several factors contributed to labour’s weakness, even as unionization increased in the 1980s. Indeed, the latter never regained historical levels, as we have noted, and in 1991 almost half of existing unions had experienced fewer than six years of existence (there were 9858 unions in 1991, against only 4994 in 1985). In autumn 1992, 46 per cent of respondents to our survey of industrial unions stated that more than one union existed in their firm (see Table 4.1).9 Leaders at the federal level noted how earlier sources of motivation for sectoral leadership, including allegiance or affiliation to political parties, had weakened over the last couple of years.10

Unions, Parties and the Sacrifice for Democracy Table 4.1

65

Number of unions in the enterprise, 1992

Number of unions Not known/no answer 1 union 2 unions 3 unions 4 unions or more Total

Percentages

Frequencies

3.5 52.0 30.5 4.5 9.5

9 127 75 11 23

100.0

245

Source: Union survey, author, 1992.

Employment trends had also changed, so that there was no longer an obvious sectoral base for national unionism. Industry is often the place where forms of industrial relations take hold, and indeed industry still accounted for approximately 24 per cent of organized labour in 1992, against 22 per cent in 1976. In the years after 1979, 76 per cent of strikes were carried out in this sector. Fifty three per cent of the workers involved in collective bargaining came from industry. But new job creation in industry was not impressive, with 7100 new jobs in 1990–2, against 35 900 in the financial sector.11 The union movement was also highly fragmented. In 1992, Chile had over 300 federations and confederations, of which only a third were affiliated to the CUT. The great majority of the federations remained ill-equipped and basically lifeless.12 More often than not, each sector had more than two or three federations, organized around the different party affiliations of their leaders. In most federations, fewer than a third of local member unions paid their dues, and those that did were generally several months in the red. Almost all federations were run by unpaid labour.13 It was estimated that the CUT itself received only between 10–15 per cent of the dues it was owed in 1991. 14 And yet, if we are to comprehend the union response, we also need to emphasize the unintended consequences of the Pinochet reforms, in terms of aiding the long-term possibilities for expanding universal social rights through union work. These included the erosion of previous legal divisions between categories of workers (between unions for blue- and white-collar workers, for example), and even the very informality of work (see Chapter 2). This created a certain element of commonality between the work-related problems experienced by ‘outsiders’ and those (currently) in employment. Traditional sectoral divisions were also weakened by the abandonment of sector-specific social insurance and the equalization of productivity levels between sectors (see Chapter 2).

66 Labour Politics and Chile’s Transition

The position of local leaders on labour relations For the local union leaders, however, the experience of increased informality was mainly confined to conditions within the leaders’ own firms. The tradition of defending workers’ rights through unions must have been in the collective memory of workers; for otherwise it is difficult to understand how thousands of new unions, and federations, emerged while dictatorship was still in place. At the same time, the re-creation of unionism, by old as well as new unions, had to be linked to the immediate problems that extreme labour flexibilization presented within individual firms. It is in this area that we most clearly perceive the subjective impact of atomization and economic crisis, and its broader reverberations on union demands. The impression that workers’ co-operation was necessary to ensure the survival of the firm may not have been factually correct. However in the 1980s, this impression led many union leaders to take an accommodative role upon themselves. Of the local leaders we surveyed in 1992, 86 per cent recounted that severe economic hardship had affected their place of work in the course of the 1980s (Table 4.2). Among the most commonly observed measures taken by management were dismissals and the lowering of wages (see also second mentions – in Table 4.3). This was followed by the introduction of new forms of supervision and machinery. The union leaders responded to what appeared to be a rationalization focused mainly on labour costs, with an essentially adaptive approach. The single most frequently cited response to economic crisis was the ‘freezing of demands’ by the workforce itself (18 per cent first mention), indicating a sense of duty. A similar trend may also be observed in the

Table 4.2

How the economic crisis of the 1980s affected the respondents’ firms Percentages

Not known/no answer It did not have any effect It had a slight effect It had a marked effect, which has not been overcome It had a marked effect, which has basically been overcome The firm was established after 1980 Total Source:

Union survey, author, 1992.

Frequencies

6.1 15.9 11.7 4.0

15 39 29 10

58.3

142

4.0

10

100.0

245

Unions, Parties and the Sacrifice for Democracy

67

Table 4.3 The most important measures taken in the firm to overcome the crisis, in the estimation of the union leader (first and second mentions) 1st mention Percentages

Frequencies

Percentages

Frequencies

7.2 18.5

17 45

26.0 18.5

64 45

14.6

35

19.3

48

12.8

31

1.3

3

11.2

27

5.5

13

10.5

25

9.9

24

18.4

45

5.4

13

2.8

7

12.2

30

1.3

3

0.6

2

2.7

6

1.3

3

100.0

245

100.0

245

Not known/no answer Not applicable (no effect of crisis on the firm) Dismissals and change of personnel Change in work organization Introduction of new technology Change in destination of products (markets) Freezing of wage demands by workers Lowering of wages, or maintenance of low wages, by management More participation by workers in the firm Other Total Source:

2nd mention

Union survey, author, 1992.

finding that while 73 per cent of the presidents concerned said they had engaged in ‘forms of co-operation’ with their employer in the course of the crisis (characteristically, 29 per cent said this had been sporadic; see Table 4.4), only 14 per cent considered that such co-operation had been ultimately beneficial to workers (see Table 4.5). Thirty nine per cent indicated that it had been beneficial only ‘sometimes’. These responses indicate that the later restraint that local leaders exercised ostensibly for the sake of democracy, was partly an inclination consolidated in the 1980s through the day-to-day experience of work. In Chapter 8 we shall further argue that, while party political loyalties continued to be important among union leaders in the early 1990s, such loyalties did not have a significant influence on the local leaders’ views on (local) labour relations.15 It should be noted that union leaders emphasized wage demands. Indeed, when asked which was the most important goal they longed to

68 Labour Politics and Chile’s Transition Table 4.4 Was there a relationship of co-operation between workers and the employer during the 1980s?

Not known/no answer Yes, always Yes, but only sporadic* No, there was no co-operation Not applicable (the firm was established after 1980) Total

Percentages

Frequencies

10.1 44.6 28.6 14.0 2.7

25 109 70 34 7

100.0

245

Notes: * ‘Puntual solamente’. Source: Union survey, author, 1992.

see fulfilled by their union, wage increases above the rate of inflation was a clear first priority: 71 per cent chose this answer (see Table 4.6). However, this should not really surprise us. The emphasis on goals related to economic benefits in fact scored somewhat less in our 1992 survey than it did in a survey carried out in 1962 by Manuel Barrera, where 79 per cent prioritized this.16 If we really wish to ascertain the nuances in the positions of local leaders, and the extent to which wage-related goals might be modulated by other attainments, we need to look at first preferences and at second preferences. A direct comparison between our own and previous findings is not possible, because the organization of work and workers’ say in it tended not to be developed as a theme in earlier surveys. Nevertheless, it is

Table 4.5 How far did co-operation with employers during the 1980s benefit workers? Percentages Not known/no answer It always benefited workers It benefited workers sometimes It never benefited workers There was no co-operation Not applicable (the firm was established after 1980) Total Source: Union survey, author, 1992.

Frequencies

10.2 13.8 39.1 23.1 11.1 2.7

25 34 96 56 27 7

100.0

245

Unions, Parties and the Sacrifice for Democracy Table 4.6

The most important goal for the local union (first and second mentions) First mentions Percentages

Not known/no answer Raising of the real wage More family benefits Better working conditions Levelling of wage levels (formalization) Profit share Improvements in workers’ training and rewards Some participation Improvement in the system of bargaining Total Source:

69

Second mentions

Frequencies

Percentages

Frequencies

2.5 71.4 5.7 1.6 9.8

6 175 14 4 24

10.1 0.3 6.2 3.7 26.6

25 1 15 9 65

1.2 7.4

3 18

26.6 16.0

65 40

0.4 0.0

1 0

2.9 7.6

7 18

100.0

245

100.0

245

Union survey, author, 1992.

noteworthy that as many as 20.5 per cent of respondents to our survey selected themes that related to a democratization of flexibility forms as their first preference. In 1962, 10 per cent had cited non-economic demands in the area of strengthening the union as their first preference. In our survey, 9.9 per cent selected greater equalization between workers’ wages in the firm as their first preference, and another 9.6 per cent had mentioned other themes, including training, participation and better working conditions. The comparison between second preferences is more revealing still. In 1962, 48 per cent of respondents selected increased economic benefits as their second choice, with another 20 per cent in favour of welfarerelated demands, such as housing, help with schooling and holidays. In our survey, 0.3 per cent selected economic benefits as their second preference, although 26.6 per cent wanted a mandatory share in profits, and 6.2 per cent selected family-related claims. Another 26.6 per cent selected equalization of wages, and a further 26.5 per cent prioritized either training, more participation or improvements in the system of bargaining as their second preference, with 3.7 per cent opting for better working conditions. In other words, 56.8 per cent selected as a second preference goals that were in some ways related to a democratization of flexibility forms.

70 Labour Politics and Chile’s Transition

But what about the preference for training? The picture in this area is rather complex. It was found that those leaders who had emphasized the preference for more workers’ training as a demand for the union movement (see Table 4.7) were more likely to report high levels of occupational insecurity in their firms, relative to leaders who had emphasized more conventional preferences (see Table 4.8). And while only a tiny minority of unions had some influence on the selection of workers for training (8.6 per cent said they always did; and 15.3 per cent said they sometimes did), a fairly important number of unions believed that the union or the union federation had been important in instigating the exercise of workers’ training inside their firm. Thirteen per cent said they had been the most important actors. This is probably unlikely to be true, but the fact that union leaders believed it to be the case is significant. This belief was more prevalent in unions with contracts (the stronger legal instrument) and it was particularly prevalent in older unions (established prior to the coup), and in unions that in the 1980s had participated in co-operation with the employers. 17 In short, there were many indications that principles of equal and fair treatment, transparency, and access to self-improvement through training were conceived of as basic democratic rights among union leaders. This also indicates (of which more in Chapters 8 and 9) that the interest of local leaders in flexibility forms originated from an independent and critical stance. In other words, it did not reflect acquiescence to a kind of mutualism sought by employers from unions.

Table 4.7 The most important goal for the union movement in the current environment Percentages Not known/no answer Raise unionization levels Organize mobilizations Work to attain more influence on national labour and economic policies Work more closely with political parties Seek more links with Chilean society at large Seek to improve the dialogue with employers Raise the level of training Improve the social relations inside the firm Total Source:

Union survey, author, 1992.

Frequencies

1.9 39.5 2.4 17.4

5 96 6 43

– 1.5 8.7 21.1 7.5

– 4 21 52 18

100.0

245

Unions, Parties and the Sacrifice for Democracy

71

Table 4.8 Emphasis on training as a goal for the union movement crossed with degree of occupational security (levels of labour turnover)* Turnover rates under 30%

Turnover rates over 30.01%

Totals

Percentage Frequency Percentage Frequency Percentage Frequency Raise unionization levels Raise the level of training

88

(80)

12

(11)

100

(91)

70

(35)

30

(16)

100

(51)

* Defined as variations in the total size of the workforce (see Appendix B). Source: Union survey, author, 1992.

Politicians and bureaucrats: the labour leaders and their advisers The isolation of local unions had a varied effect on the labour leaders who were acting above the firm level. The representational gap had in some cases forced the leaderships of sectoral associations to focus their activities away from party-political issues. This was the case with CONSTRAMET, the metal workers’ union. This organisation was run by a dedicated set of leaders linked to the Communist Party, but had very few member unions in 1992 with party links. CONSTRAMET had to seek new members by dealing with the issues of labour relations that concerned its constituency. An example of this was CONSTRAMET’s attempt to establish direct control trade-offs through consultation with employers for the first time in the transition period (see Chapter 5). But the adoption of a ‘realist’ or incremental approach to labour relations at the national level, and which in many ways reflected the orientation of local leaders, requires a different account. We need to make a distinction between the labour leadership proper, and the labour advisers. The actual contents of an agenda of labour relations in the second half of the 1980s was left almost entirely in the hands of a set of professionals and labour lawyers organized in the Centro de Investigación y Asesoría Sindical (CIASI), an independent NGO that housed the CUT’s key advisers. The latter had experience from the floor of Chilean unionism, but as technocrats they were free from the usual constraints of elected union office (such as providing a radical discourse on legal and wage-related demands). One of the factors behind CIASI’s important programmatic role was the continuation of ideological and party-political differences among the

72 Labour Politics and Chile’s Transition

labour leadership (unelected before 1988, for obvious reasons). Indeed, the alliance between the Christian Democrats (PDC) and the Socialists (PS) which would shape the new CUT of 1988 only emerged victorious after a decade that had seen the existence of four labour centrals. The day-to-day affairs of national labour leaders continued to be closely associated with the party-political world. 18 Not surprisingly, the CUT elections of October 1991 occurred around the presentation of prestructured party lists, written up by the respective party committees of trade union leaders. 19 The position of core union leaders also continued to be strengthened by their capacity to bring foreign money to the CUT, partly through party contacts, and subsequently to control a sizeable part of international aid. In 1992, 75 per cent of the CUT’s budget was still made up of foreign donations. Federal leaders in sectors of relatively new occupation, such as fruit-packing or forestry, felt bitter at the lack of practical assistance from the CUT.20 Disappointment by sectoral leaders at the lack of consultation concerning national pact-making was the major theme at the second CUT Conference, in 1991. On the other hand, the preoccupation among the labour leaders with political issues conferred a high degree of effective autonomy to CIASI in defining the CUT’s approach to employers as well as legal affairs. 21 CIASI intellectuals had followed closely the process of renovation towards the centre. In some respects its expertise was equal to that of the Concertación technocrats, with whom it had fairly close links. 22 It was CIASI’s president Díaz who initiated the idea of national social pactmaking with employers, through his contacts with Raúl García in the SNA, and he had to face a wave of scepticism at a host of union meetings over the course of 1988. Nevertheless, what emerged eventually was a fairly coherent stance. Ultimately, the labour leaders relied on CIASI’s expertise in their attempts to present the CUT as a valid political actor. The ‘new unionism’ of the CUT The weak organizational base of the CUT explains the focus of its leaders on legal affairs.23 Broadly speaking, the CUT had embraced the market model. An illustration of this was the lack of any critical reference to the open market model as such in CUT publications. In fact, the degree of effective unity within the CUT leadership around these parameters was remarkable. The Communists, who tried unsuccessfully to re-include reference to the class struggle and the abrogation of private property in the central’s Declaration of Principles, placed unity first. Its leaders either

Unions, Parties and the Sacrifice for Democracy

73

kept a low profile or participated in pact-making with the Concertación Government alongside the Socialists and the Christian Democrats through the first crucial pacts under Aylwin. But, as noted earlier, it was a small circle of CUT leaders and CIASI advisers who had elaborated the CUT’s strategy for the transition period: the CUT should abandon ideas of mobilization, should not rely exclusively on the state, including state corporativism, and should develop alternative themes to those of wages and job security in the traditional sense of fixity.24 It was suggested that the idea of linking remunerations to productivity would improve labour relations and be beneficial to workers. One concrete suggestion involved the establishment, between the CUT and the CPC, of a national institute to study ‘technical’ ways of linking wages to productivity gains and involving workers more comprehensively in production. This institute could formulate ‘models’ which might be of use for collective bargaining in firms.25 Díaz saw such institutions as an opportunity for the union movement at large, ‘The room for manoeuvre that workers have today is very large, if they use it well. In the developed world they stress “quality”, “productivity”, and say that “this depends on workers’ skills”, on “good working conditions”, “consideration for the human factor in production”, “the labour factor”, “labour relations”, etc. So, it is up to the union movement to take advantage of this.’26 In fact, the CUT did put several suggestions that would increase transfirm security and mobility of workers. These included the establishment of a National Agency of Employment, as well as a tripartite unemployment fund, encompassing criteria of solidarity. Díaz suggested more vigorous training schemes, with INACAP, the country’s largest training corporation coming under public scrutiny, and programmes of industrial re-conversion, among others. In summary, it is clear that the CUT perceived that alternative and universal forms of occupational rights (to pure job security, occupational statism and rigidly occupation-specific wages) represented the most constructive defence of workers’ rights within the new setting. This was no doubt a fairly realistic assessment. 27 And yet, even with this moderate strategy, the CUT was confronted with obvious limits. CIASI retained a weak link with the membership base, particularly the sectors that lacked a strong political proximity to the CUT. In 1992, Díaz had to admit that the plans to invent novel guidelines for local bargaining had failed to come through.28 Instead, the CUT relied on the policy intentions of the Concertación, a unilateral faith that proved to be problematic, and on the CPC’s willingness to accept a discussion of matters that pertained to the sectoral and enterprise

74 Labour Politics and Chile’s Transition

levels. Otherwise, the CUT’s very acceptance of efficiency criteria could turn out to be no more than a ‘sell-out’, or worse still, a sealed commitment to the atomized model. 29

Accommodation of the opposition parties The coherence of the Concertación relied in reality on a certain distancing from labour by both the Socialists and the Christian Democrats, whose ideas in this area had varied greatly in the past. When pact-making between the left and the centre began to take shape in the early 1980s, social pact-making was still viewed as a vehicle to transform the existing development model. By the mid-1980s, however, the stance of the political opposition began to change, as preparations went ahead to engage in the plebiscite for or against eight more years of Pinochet, scheduled for 1988. Entrepreneurial leaders and the right responded with the initiation of an aggressive media campaign that emphasized the importance of maintaining the existing privatizations, fiscal discipline and the labour regime. 30 At the business gathering (ENADE) of 1988, to which Aylwin’s prospective ministers were invited, it was pointed out how ‘substantial changes to collective bargaining and strike provisions’ or the installation of a ‘regime of stability in employment’ would be considered contrary to the established ‘labour regime’. Any such changes would make it ‘impossible’ to ‘generate that stability and confidence’ that the business community demanded the political opposition undertake to secure. 31 After 1988, the attempts at social or political pact-making with entrepreneurs virtually died out. 32 The emphasis of the Concertación was shifted towards the presentation of a prospective ministerial team of a conservative blend.33 The success of the military regime’s economic reforms had a forceful impact on opposition leaders. As one top technocrat put it: ‘Chile has a tremendous reputation, because it is one of the few countries in the South which classifies for the catalogue of triumphant Western values à la “Fukuyama” . . . It was Pinochet who carried through the revolution and who paid all the costs. Today, we reap all the benefits.’34 Intellectual publications by members of the political elite still highlighted the need for ‘an aggressive policy of technological innovation’ with the state as the promoter of ‘a new style of growth’, based on national systems of ‘permanent training’ in addition to “vigorous” technological assistance’ by the state. 35 However a political formula soon took shape. The dichotomy between an intellectual and a political formula can be traced in the Concertación’s programme of government. For example,

Unions, Parties and the Sacrifice for Democracy

75

the stress on the pursuit of ‘vigorous industrialization’,36 which would integrate workers in the ‘tasks of technological innovation and adaptation’37 was more a general aim than something the state could or should actively promote. It was presupposed that the ‘rules of the game’ should be accorded ‘primarily’ between the parties themselves.38 On the other hand, the Concertación did see a role for the state in establishing ‘integral’ labour relations. This would be done through ‘profound changes’ to the Labour Code of 197939 on the basis of which substantive pact-making should extend to the sectoral and national levels. 40 The Concertación promised that both parties would participate in national institutions of training. The alliance also wished to create a new universal unemployment benefit system. At this stage, then, the Concertación evidently aimed at establishing basic forms of occupational citizenship, without calling it such. Once in government, this formula proved difficult to sustain.

Conclusion It was a mixture of factors, including the local-based tradition of union work, and the impact of social and economic reforms, that sensitized the unions in Chile to flexibility issues. These and other political factors were important in explaining how the union response remained an essentially independent one. In particular, it is interesting to highlight how long-term legacies blended with more recent union experiences. In the new scenario after 1973, both local leaders and national advisers built on the pre-coup union tradition, which had always emphasized unity (despite, and in some ways because of, the difficulty of attaining it, caused by the often divisive impact of the political parties) and the universal aspect of rights. The same ideals were implicit in the universal programmes for social transformation that the left parties had entertained. It was also manifest in their attempts to co-operate in achieving them. It is plausible that this tradition was carried within the local unions themselves (if not in individual leaders), in that most of them had had more than seventeen years of existence, 41 and in the national union leaders and advisers, many of whom had started their union work prior to the coup, or who belonged to political parties. As we shall see later (Chapter 9), it was not the union leaders’ political associations which motivated them to become directly involved in the negotiation of trade-offs with employers. But nor would we expect this to be the case, since the authoritarian period only served to widen the gulf that had always existed, according to Angell, between the unions’ daily role, and the principles that inspired their daily existence.

76 Labour Politics and Chile’s Transition

Our intention has been to highlight the general influences that made the unions use their democratic authority, however slight, to focus on rights of a universal kind. Taken together, Chapters 3 and 4 have allowed us to observe two historical features that were shared between the organization of employers and labour in Chile: a strong reliance on national political solutions, and an underlying functional distance from labour relations at the local level. In fact, the failure to connect CIASI’s project with local unionism in a functional way indicated that legal measures to strengthen the local and intermediary level of unionism might have been crucial in order to advance towards forms of occupational citizenship at this time. This is an issue to be discussed in Chapter 6. Suffice to say that the history of firm unionism in Chile proved to be a doubleedged sword in labour’s attempt to define a new role for itself in the 1990s. On the one hand, it meant that union leaders understood and dealt with the issues of labour flexibility at the level of firms. But on the other, it remained difficult to bridge the gap between firm unionism and national leadership, and consequently to retain the support of political parties. In short, if the labour movement achieved only a small degree of political articulation in the transition period, this was due to its weakness and not to the poverty of its ideas.

5 The Transition to Democracy and the Enforcement of Markets

Despite the Concertación’s declared intentions of doing so, it was not easy for a new government to provide an alternative labour policy, given the authoritarian government’s success in establishing a deregulated liberal economy and, at least on the surface, in making it work. During Aylwin’s term in office a subtle political process took place which legitimized a decentralized and weak industrial relations substantially similar to what had existed before. The transition government came to accept the notion of non-intervention in the economy, and this helped to consolidate the dominant role that entrepreneurial leaders had attained in the previous years. In fact, this set-up was mutually convenient because it gave transition government considerable managerial autonomy, as we discuss in section one. One of the costs was a reasoned debate about labour reform. The very subtlety of this process is an indication of the important part that political discourse played. Development ideology was the invisible liquid that oiled a policy of non-intervention in industrial relations (second section). It made it possible, for example, to present what was consolidated from the previous period, through the social pacts concluded during the time under study, as the mark of genuine progress. One of the weaknesses that characterized the labour movement in the transition period was the fact that labour mobilization was dominated by sectors with what might be called fairly ‘traditional’ aims. At the same time, the CUT was barely given an alternative framework through which to channel its more thoughtful concerns. The challenge in labour reform was largely about the level at which different labour issues would be discussed. As we have observed, a decentralized industrial relations was nothing new in Chilean history. On the other hand, it is also true that events during the decades prior to the dictatorship 77

78 Labour Politics and Chile’s Transition

had pointed towards a greater role for labour institutions at the sectoral and national levels. Moreover, new challenges of development had surfaced during the dictatorship that made such a framework relevant. These included a greater level of manufacture in exports, more emphasis on workers’ skills and the negotiation of viable flexibility forms. The breakdown of such a framework is the subject of the last three sections of the chapter, in which we deal with the specific issue of industrial relations reform.

Managerial autonomy and pacts of legitimation Earlier we argued that the literature on transitions to democracy has tended to view pacts in terms of a certain equiparity between elites, with ‘authoritative . . . class associations . . . agreeing on essential goals’. This conception is similar to Olson’s idea of a ‘balance of power or stalemate’ as the origin of a democratic regime. 1 There is a need to differentiate such pacts of ‘stalemate’ from others which serve to work more actively as a ‘legitimation’ of the status quo. Chile’s social pacts between 1990 and 1993 represented the latter kind.2 Our argument is that Chile’s form of pacted transition embodied a strong inducement to policy-makers towards continuity. In effect, the transition caretakers won their spheres of autonomy by accepting constraint. Acceptance of Pinochet’s timetable for transition won the Concertación alliance access to a centralized executive. Later, acceptance of non-intervention won the new government the ‘facilitating confines’ to carry through a successful management of the economy. Indeed, the Aylwin Government’s labour policy was more successful and politically expedient than that of the other LatinAmerican states precisely because it was constricted in nature. Ascent to government did give the Concertación a certain sphere of autonomy vis-à-vis the political right and entrepreneurs, but this was exercised with considerable care. Despite its presidential powers, the Aylwin Government tied itself to a restricted labour policy in the nature of the working relationship that it established with Renovación Nacional (RN) in Congress. In fact, the powers of presidential rule were exercised most firmly vis-à-vis the Concertación’s own.3 In co-operating with the vetoholding right, the Concertación could please the business sector without being seen to do so. The Concertación also knew how to exploit the wariness of entrepreneurs as to their new role and image in the early months of the nascent democratic regime. Hence, the Aylwin Cabinet enticed the CPC to transform its initiative in social pact-making into a political commitment to

The Transition to Democracy and the Enforcement of Markets

79

the new regime.4 In the case of the CUT, the central’s advisers were called to the Ministry to review the first social accord (written up by the government) only 48 hours before their signatures were due.5 Two years later the CUT’s Socialist leadership (represented by Vice-President Martínez) refused to sign the yearly accord, because of its lack of substance, and was called to order by the Socialist Economy Minister himself, according to one source present at the relevant meeting of unionists and government officials of the Socialist Party.6 The aim of the accords was ‘basically political’, according to the Minister of Government, Enríque Correa, a way to distinguish the present from the previous regime. 7 This allowed the CPC to continue to limit the pacts. The first tripartite accord of the transition government bore a close resemblance to the pact that the CPC had organized with the CUT two months before. It reiterated almost literally the earlier reference to a limited role for the state and, in nebulous terms, to the desirability of concentrating labour relations in firms. 8 The format and content of the accord sanctioned the separation between poverty alleviation, on the one hand, and the productive sphere, on the other. The role of the state and public co-ordination (tripartism, for example,) was allowed in the former, but not in the latter. 9 The limited nature of this first accord set a trend in motion. Tripartite accords on substantial themes never managed to materialize in the decade that followed. After 1994 no further formal accords were entered. A Forum for Productive Development (el Foro del Desarrollo Productivo) was fleetingly established, but this and another effort at continuous dialogue, the Council of Social Dialogue (el Consejo de Diálogo Social) of 2000, also failed.

Non-intervention and transition politics The role of development ideology Schull defines ideology as a discourse that is shared by many groups, with different aims and interests. He explains how ‘having to respect certain conventions for meaningful communication . . . [can] entail further commitments for [an] agent’s pattern of action’.10 This in some ways describes how the theme of non-intervention became embedded in the dynamics of Chile’s transition. Our present argument is that what was at one time an instrumental adjustment on the part of the political élite changed in flavour when this élite assumed the office of government. It changed towards a broader ideology of development in the Hirschmanian style. For entrepreneurs, the convention of non-intervention

80 Labour Politics and Chile’s Transition

was an instrument to maintain social power and political influence. To the Concertación it became a medium to obtain the private sector’s concert. Hirschman’s ‘perversity thesis’ refers to the belief that adverse effects will arise whenever mechanisms of ‘natural order’, including the competitive market mechanism, are tampered with.11 We would argue that the Concertación used the perversity thesis as a medium to explain its timid policies in respect to building capabilities in labour market. A top adviser in the Labour Ministry favoured a ‘uniform policy’ and believed that the state had to be extremely cautious in initiating any sector- or industry-specific focus in public policy. It was feared that would amount to ‘protectionism’. Minister of Government, Correa, was emphatic that the state should not provide ‘any special treatment to small firms’, despite his concession that ‘jobs of higher quality’ was an ‘economic necessity’.12 The unwillingness to distinguish firms by size in policy instruments continued to be an obstacle in the design of unemployment insurance right through the 1990s. 13 Correa believed that social policy should be centred on poverty relief and the control of inflation. It could not be a policy of ‘welfare’ or for the ‘labour market’. Such views were reflected in a public critique by a newly appointed technocrat in the Economy Ministry. She spoke out against an alleged ‘taboo’ within the policy establishment which counterpoised ‘horizontal’ with ‘sectoral’ instruments. 14 The first applied ‘to all’, and hence was acceptable, whereas the latter was targeted at specific groups, and hence dismissed. The Government’s leading policy-maker on workers’ training used precisely this dichotomy to explain why training policy was based on ‘neutral’ criteria of geography rather than of sectoral modernization.15 Interestingly, Henríquez believed that the taboos in question were being practised essentially with respect to industry and to policies designed to protect labour. Indeed, selective intervention was customary in agriculture, mining or transport. In some way her observation fits remarkably neatly with our argument that ideas of market neutrality in Chile were bound up with negative perceptions of the country’s experience with state-induced industrialization prior to 1973. The idea that representative groups should participate in market coordination was also rejected in the interest of maintaining market neutrality. ‘Social Democracy’, according to Minister of Government Correa, was impossible in Chile because ‘it entails acknowledgement of sectional interests’. Social or economic projects with reference to ‘any particular group in society’ (‘proyectos estamentarios’) should be rejected on the grounds that such interference was ‘incompatible with democracy’. 16

The Transition to Democracy and the Enforcement of Markets

81

Correa’s views had changed greatly since the mid-1980s, when he worked closely with the CUT in CIASI: ‘Two central beliefs we have abandoned concern the convenience of bargaining at the sectoral level and the idea of tripartism as a means to cope with labour concerns.’ 17 Finally, the Aylwin Government adopted Hirschman’s twin ‘jeopardy thesis’, which relates to the argument that a new proposed policy will put in jeopardy a previously attained goal.18 To Enríque Correa, Chile faced ‘a unique – and perhaps last – historical opportunity to achieve development’, namely to continue the economic project upon which the military regime had embarked: ‘In all the essential aspects the current regime is a good one.’ Correa linked the prospect of changes that might be sensitive to the business community to the possible ‘loss’ of this last chance: ‘If we raise taxes or if we had proposed a more rigid labour reform, economic growth would probably be two or three digits below the level at present.’19 The political weight of the CPC Political campaigning by entrepreneurial leaders played an important permissive role in making the continuity of economic policy appear essential. The Concertación was charged with carrying out a ‘silent contrarevolution’ in its failure to continue the privatization process, 20 and the subject only subsided when the copper company itself was opened to private capital the following year. In fact, employers seemed less than happy to accept basic rules of the democratic game, such as the prerogative of government to propose changes to the country’s political institutions, including the composition of Congress and the electoral system that, it was avowed, would destabilize the Chilean economy. 21 The question is, if entrepreneurs had not abandoned their instinctive view of the state as the enemy, then how were they to co-operate with the state to achieve the common good? As we saw in Chapter 3, the very notion that market regulation, even by private-sector institutions, might be helpful in building capabilities and industrialization was so unacceptable to the CPC’s associations that they specifically avoided any demands for target-specific programmes or aid to their own sector. By 1993, entrepreneurs were displaying a relative cessation of hostilities towards the government. This reflected their recognition that further liberalization of the economy was being carried out. One CPC leader mentioned the lowering of uniform tariffs from 15 per cent to 11 per cent: ‘To me, that was it!’ (Dije, eso ya!).22 Alé, SOFOFA’s key technocrat, noted how new measures meant that Chilean capitalists now operated abroad ‘basically without restrictions’. To him this was the evidence of

82 Labour Politics and Chile’s Transition

a sound economy. In June 1993, Correa conceded to a renewed privatization effort and promised that all future public sector projects would be carried out through private initiative. 23 And in international forums, Raúl García and Cortázar staunchly defended the country’s private pension system against attacks from the ILO itself and from Chilean labour.24 Tripartism, bipartism and the formation of industrial relations A series of labour conflicts underpinned the weakness of the CUT as a peak actor and provided the Aylwin team with a chance to affirm its ‘market neutral’ position. For example, in its conflict with copper workers (the CTC) the state-owned copper corporation (CODELCO) maintained its commitment to the government’s anti-inflationary strategy and to a decentralized management of the dispute. The Communists, who had led the strike, fell to the lowest position among the political parties in subsequent union elections. A few months later, in the summer of 1991, the leaderships of public health and education sector unions, which together represented more than 400 000 workers, were decisively weakened. The transition government used the yet-tobe-revised Pinochet law against strike activity and employed police repression to break up the sectors’ mobilization for institutional changes and higher pay. But the larger question of a future industrial relations had to be settled within the private sector itself. According to the maxim of nonintervention, the central issue became defined in terms of a choice between tripartite and bipartite institutions, that is, between institutions which involved an element of the state, and others that did not. When the terms of this restricted definition of what is ‘public’ was accepted, it became quite easy for the employer sector to sabotage any attempt at institution-building at all. Government policy was also an important factor in forming a limited bipartism, both in form and in content. Leaders of the CUT and the CPC had met a few times in the months following the 1990 accord. 25 However, the process of labour legislation had made entrepreneurs wary, and in November the CPC council elected José Antonio Guzmán as president, to replace the outspoken style of Feliú. In the accord of 1991, and in a bipartite media event arranged later that year by the CPC, the CUT agenda was ignored altogether. This included further discussion on a national employment information system and an unemployment benefit scheme. When the CUT used its weapon-of-last-resort, a direct appeal to the father-figure of President Aylwin, 26 it was met with a

The Transition to Democracy and the Enforcement of Markets

83

cautious silence, which allowed the SOFOFA and the CPC to reiterate their rejection of any form of labour relations outside the firm.27 Labour Minister Cortázar followed with a public distinction between national accords on general themes (such as the minimum wage) and problems pertaining to labour relations in the firm which had to be resolved at that level. Aylwin backed his Labour Minister. 28 The National Social Accord of 1992 did stipulate the establishment of a tripartite conference, but the contents of talks were cleansed of the concrete labour market themes proposed by the CUT. The CUT leaders rejected the first accord that the new body produced, a propaganda film in favour of the labour legislation, on the grounds that its ethos was individualist and anti-union. The CUT was further weakened as a result. The affair concerning the propaganda film exposed the tension between CIASI staff, who negotiated the media deal, and CUT leaders, who resented the independence and ‘technocratic’ attitude of the advisers upon which they depended. The CUT ‘froze’ relations with government temporarily, but its top leaders never managed to organize the promised general strike. The CPC, on the other hand, emerged strengthened from this experience. It seized the chance to reject any further talks with the CUT, and President Aylwin responded by restating that the state had no role to play in labour relations. He also kept the government’s promise to the right by refusing to reconsider aspects of the new labour reform. 29 The politics of constraint was a mutually reinforcing game played between Concertación and business leaders. Entrepreneurs did not want to sit in on formal bodies, but the Concertación was characterized in private as the most ‘open’ and the ‘most receptive’ government ‘we have known’. 30 The CPC’s key leaders had a direct link to every important minister and this was used several times a week.31 In the view of many, Minister Cortázar was ‘the most successful Cabinet Minister’.32

The question of sectoral pact-making At the heart of the debate about the structure of industrial relations was the issue of sectoral pact-making. According to García, the CPC’s progressive man, ‘our biggest fear are negotiations by sector’.33 Images of history tied in neatly with neo-conservative assumptions about the distortionary effects of labour monopoly and of ‘picking winners’ at the sectoral level. When CUT leaders spoke of sectoral pacts or tripartite commissions, these became associated with the experience of the Allende

84 Labour Politics and Chile’s Transition

period, when government and labour leaders allied against entrepreneurs to dictate wage scales by sector. These impressions had little relationship with the proposals put forward by the CUT at the time of transition. These proposals entailed the establishment of talks at which any decisions would be made by consent. Even if labour leaders in some sectors, such as metals, had sufficient power to use sectoral talks as a platform to mobilize for unilateral gains, the fact was that labour leaders were not inclined towards this type of strategy. The CUT spoke of discussing ‘minimum working conditions’ in areas of precarious employment, such as temporary work in the new export sectors. Other initiatives entailed the building of institutions to help relocate workers, and measures to promote workers’ training, productivity and industrial exports. 34 However, ideas of sectoral pactmaking like these were often confused with the legal issue of collective bargaining above the firm level. This did not help in establishing a more technical discourse. In fact, the CUT’s project (and the original intention of the Concertación), had been to make it obligatory for more than one employer to respond to union petitions. Once in government, however, Labour Ministry officials changed their position in favour of a voluntary system.35 The argument of senior officials and ministers finally evolved to reflect agreement with the entrepreneurial thesis that sectoral accords and minimum wage agreements would undermine the whole export model. Experiences of sectoral talks in the Chilean transition The national accords did not usher in the establishment of sectoral talks ‘downwards’, because it was only in sectors where employers were accustomed to sectoral talks already that such steps were taken. The successful and unsuccessful cases cited below illustrate this point. The metallurgical sector accords The association of employers of the metallurgical sector, the Asociación de Industriales Metalúrgicos (ASIMET), and two smaller labour federations initiated a pattern of yearly minimum wage accords in May 1990. By 1992 the dominant labour federation, the Confederación de Sindicatos de Trabajadores Metalúrgicos y Metalmecánicos (CONSTRAMET), led by Communists, finally agreed to join. The accords then comprised more than 75 000 workers, and represented beyond doubt the most comprehensive experience of its kind during the Chilean transition. A relatively high level of skill in the labour force has historically favoured the bargaining position of metallurgical workers. The sector’s

The Transition to Democracy and the Enforcement of Markets

85

skill composition is, however, fairly diverse (and many low-skilled or unskilled workers start work in the sector). Hence the continued high density of union membership (at least 60 per cent, and in some estimates close to 80 per cent, in the early 1990s) was perhaps a more important bargaining factor. Moreover, a high level of commitment and technical knowledge, especially in CONSTRAMET, characterized union work at the federal level. 36 It was on the initiative of CONSTRAMET that the 1992 accord was furnished with new agreements to discuss union participation in the design and implementation of training programmes and to include courses in the education of unions.37 The accord asked all ASIMET members to give more information to unions for the purposes of an informed process of bargaining. On the initiative of CONSTRAMET, the parties agreed to study an unemployment insurance scheme and they proposed that similar institutions be established at the national level.38 Organizational history is also important in explaining the willingness in ASIMET to enter into comprehensive accords on labour matters, given the otherwise heterogeneous composition of the metallurgical sector, comprising activities from basic iron- and steel-works through to the production of spare parts, equipment and machinery for transport, and for domestic or scientific use, and occupations ranging from welding and joining to industrial mechanic. ASIMET had been involved with problems of the productive sphere for decades, principally in the area of providing legal and technical assistance to members. It held an independent position within the entrepreneurial setting, and it diverged from the CPC in its critique of the Pinochet regime’s treatment of industry. Finally, entrepreneurs in the metallurgical sector had a long legacy of involvement with workers’ welfare, even if this was paternalistic in nature. And this involvement had been made concrete and tied to enterprise modernization through the initiation of institutions of training.39 ASIMET’s prior engagement with labour on concrete issues meant that its attitude both to unionism per se, and to the question of sectoral accords was based not on ideology, but on pragmatic judgement. The key to the success of the 1992 accord, according to ASIMET’s president at the time, was the inclusion of the Communist-led confederation: CONSTRAMET . . . organizes people of the left, Communists . . . very conflictive people. It organizes the great majority of unions in the metallurgical sector. It was an organization that had always given us a lot of problems in firms. It produced tremendous conflicts. During

86 Labour Politics and Chile’s Transition

my period I tried to get closer to them. I met with them on countless occasions . . . They began to understand what we [ASIMET] were about. Our preoccupation with social issues, particularly in the area of housing, health and recreation . . . So as it turns out we have had a lot of success. At the moment we are attending some of their sessions with their member unions. Last week we held the annual dinner of ASIMET, and they participated; their leaders, and our leaders and lawyers. 40 Paredes acknowledged that the accord was a conflictive theme within the entrepreneurial sector as a whole, but he did not identify problems within ASIMET itself. 41 Of the three labour organizations, Paredes regarded CONSTRAMET as the best organized. Participation of the Communists added credibility and reliability to the accord: ‘We haven’t had any problems. In our case, the fact that we have signed a framework agreement has meant that practically speaking we have not had a single strike in the firms affiliated to ASIMET. The entrepreneurs have understood.’42 The metallurgical accords also represented the single most important attempt at establishing a professional domain, as discussed in Chapter 2, at the time under study. The accords could be seen as the culmination of a process that had been initiated by the employer federation, ASIMET, but which required the incorporation of all relevant interest associations in order to maximize its legitimacy and universal scope. Certain features of the process of social pact-making, and the initiatives in skill development that preceded it, contribute to justifying the characterization of ASIMET’s project in skill development as an opening towards occupational citizenship at the sectoral level. First, there is the sheer quality and comprehensive character of the envisaged programmes of worker training. These were not simply modelled on an abstract picture of the German dual model of education. Asimet had already benefited from extensive financial and practical help from the German Chamber of Commerce in Chile, the Instituto Superior Alemán de Comercio (INSALCO), and from the Chilean–German Chamber of Commerce, the Cámara Chileno-Alemana de Comercio e Industria, as well as from a convention signed with the German federal state of Baden-Württemberg. The convention with Baden-Württemberg implied the posting in ASIMET’s Chilean–German Industrial School of technical experts from Germany. Contrary to previous experiences of dual education in Chile, which had been concentrated around white-collar occupations through INSALCO, the new programme developed with ASIMET concentrated on the quali-

The Transition to Democracy and the Enforcement of Markets

87

fication of industrial mechanics.43 The courses envisaged would entail two years’ training in the Chilean–German Industrial School, and a further two years combining formal education with a contractually agreed course of apprenticeship in a firm associated with ASIMET. After two years in employment, the young person would qualify for further training at the level of higher-level technician (técnico a nivel superior) with the possibility of qualifying for a university-level degree within the area of choice. The second important feature of ASIMET’s programme is connected with the first. This concerns the programme’s possible effect on the formalization of skill markets in the metallurgical sector. Specifically, the associative form of the initiative, with the involvement of ASIMET and the workers’ federations, increased the likelihood that the certification of the skills obtained would have a universal validity, and hence would contribute to the developmental freedom of workers. It would also increase the likelihood of future investments by individual employers in skills that had a trans-firm validity. The initiatives taken by ASIMET’s leaders to set up an employment information bank (bolsa de trabajo), and the commitments made by ASIMET and the workers’ federations, with the 1992 accord, to invigorate this bank as well as to establish an unemployment benefit system, were likely to have a similarly beneficial impact on the occupational citizenship of the sector’s workers. Talks in the banking sector A more latent form of social pact-making was experienced in the banking sector. The resourcefulness of the confederation that organized bank workers, the Confederación de Trabajadores Bancarios (CTB), is exemplified in the organization of a discount service for electrical household goods for its members and, in 1988, the establishment of a thriving pension fund, ‘AFP Futuro’, with 7500 affiliates.44 However, there was an important difference between this case and that of metals. The employers’ organization, the Asociación Bancaria (AB) had only agreed to talks with the labour sector in 1991 in an attempt to influence the CTB’s plans to establish a workers’ health fund. Employers were worried about competition with services they had previously offered. The AB declined to incorporate other themes within an agenda for talks. This included a discussion of modest proposals for forms of employment stability.45 The AB’s rejection effectively curtailed broader sectoral agreements in the sector of banking. The position of transition government The Aylwin Government had defined its position regarding the sectoral issue by abstaining from labour relations beyond the national social

88 Labour Politics and Chile’s Transition

accords (Acuerdos Marco). But its resolve to enforce a labour framework deemed functional to the export model was tested most sharply in the case of exceptional sectors. These included temporary workers and workers organized in the so-called inter-firm unions. Reforms in the ports A top-down reform, in 1988, of the labour system in Chilean ports had allowed contracting companies to hire any person on a daily basis, without the provision of contracts. The Aylwin Government supported these changes, on the grounds that ‘a return to a registrar of port workers, giving these a monopoly of their own labour, would threaten the whole export model’ in Chile. However it was clear that a democratic government had to initiate some modifications.46 Port workers had strong union organizations,47 and so the Aylwin Government arranged for a tripartite accord to be signed, in February 1991. When labour leaders opposed the more detailed legislative reform a year later, they were ignored by the government and by employers, who withstood strike action by relocating their section of permanent staff. 48 In the end, the Labour Ministry’s reform legitimized and consolidated the existing port system. Offers of training and safety courses were made, but the practices of contracting companies were hardly checked by provisions that the companies pay their licence to port authorities and maintain an office in each harbour of operation. The new reform permitted the signing of a ‘provision of work’, and for work stability through a restricted, if adjustable, registrar. However, the decision of whether to use a ‘provision of work’ remained at the discretion of the employer, including the length of time it should run, and it gave workers no formal right to bargain or strike. Temporary work in new export sectors The government paid less attention to areas of temporary work that had a shorter history of worker organization, such as the fruit and forestry sectors. For example, the organization that represented forestry workers (the CTF) had little clout, with a density of fewer than 20 per cent of the sector’s workforce, comprising 8803 members and sixty-nine unions as of April 1992. Payment of union dues was basically non-existent. The president and two members of the executive had been sacked from the sector’s largest firm, Arauco, for their union activities, and after that had worked at Arauco through various contractors’ services. Written invitations from the CTF to the employer association, the Corporación de la Madera (CORMA), to discuss items related to formalization and basic working conditions, had passed without a response. 49

The Transition to Democracy and the Enforcement of Markets

89

As early as September 1992, Labour Minister Cortázar said privately that he was determined not to change the law so as to make bargaining obligatory for temporary workers, as it was in other sectors of labour. He also emphatically rejected the idea of ‘tripartite councils’ at the sectoral level in the areas of fruit and forestry.50 Cortázar had made the point in his writing that productivity wages require negotiation,51 yet aides rejected the ‘humanist’ position, the idea that there is a ‘right’ price for a piece of work: ‘We have to have wages that can fluctuate . . . in these sectors of export. Or else the model will die.’52 The case of inter-firm unions in the bakery sector In sectors such as leatherwork and bakery, a decade-long tradition existed of conducting bargaining above the firm level. This had continued through the 1980s, when such practices were not strictly legal. With the Aylwin reforms, such negotiations were allowed, but the new regulations proved to be stifling to the continuation, in practice, of sectoral bargaining.53 Indeed, the new provisions had made each individual employer who wished to negotiate in concert with others register formally through a bureaucratic procedure which involved a written testimony (jointly with each union) before a magistrate (ministro de fe). The affiliation of an establishment to the employer association, the Federación Chilena de Panaderos (FECHIPAN) was not regarded in the eyes of the law as evidence of delegation of bargaining rights. In addition, the new Legal Code presented special obstacles to the organization of the inter-firm unions. The Confederación de Trabajadores del Pan (CONAPAN) comprised unions which spanned large geographical areas (with up to a hundred affiliated establishments). Amassing magistrates for each member unit for the purposes of continuing joint bargaining presented an impossible task. Stipulations that the inter-firm unions were to represent only a limited geographical area set the existing centralized membership services in disarray. The stipulation that only workers with permanent contracts could enter into bargaining (Article 81) also proved troublesome. CONAPAN presented these legal problems to the new labour cabinet repeatedly during 1992, but continued to complain about rejection, and finally extended its appeal to a sympathetic but powerless parliament.54 However, the position of the Ministry was most telling in the areas where a prior consensus existed between sectoral leaders. These included the regulation of bread production and the convenience of facilitating sectoral pacts. The presidents of CONAPAN and FECHIPAN jointly approached the Labour Minister on more than one occasion to encourage him to initiate a revision of the new

90 Labour Politics and Chile’s Transition

stipulations. The reason given at the Ministry for its cautious stance was fear of the precedent that the case of the bread sector would set.55

The re-reform of the Labour Code: labour issues under Frei (1993–2001) The Government of Eduardo Frei sent a labour reform to Congress in January 1995, which sought to modify certain aspects of the initial reforms of 1991–2 (to be discussed in Chapter 6). For nearly five years, however, as the very idea to legislate yet again on the subject was opposed by the right, the Concertación government refrained from turning the matter into a major political issue.56 The peaceful relations it sustained with the entrepreneurial sector were only upset in the second half of 1999. Frei had decided to give the labour reforms first priority in reaction to the attempt by the Right candidate for the forth-coming presidential election, Joaquin Lavin, to mount a social agenda. 57 In the last months of 1999, the entrepreneurial sector showed once again its determination and ability to halt further significant reforms in this field. As soon as the decision to give legislative urgency (a faculty the constitution in Chile gives to the President) to the labour reforms, on 15 November 1999, the CPC and the SOFOFA mounted a concerted campaign, and the Christian Democratic Party split on the issue. Senior Christian Democrats who were the key actors in negotiating with entrepreneurial leaders in the early years (1990–3), such as then Finance Minister, Alejandro Foxley, openly criticized the Frei Government’s legal project. He proposed, with the right, to make further amendments, and was joined by an important group of Christian Democrats. The public disagreements that ensued with other Concertación parliamentarians, principally from the union movement, and the Socialist Party and the PPD, emboldened the right to reject the labour reform bill in the Senate outright on December 1st.58 The previous day, entrepreneurs had made clear that they wished the matter postponed to the ensuing government. They emphasized that their favoured negotiators in the Concertación remained Alejandro Foxley, Edgardo Boeninger and Rene Cortázar, respectively the Ministers of Finance, the Presidency, and Labour under Aylwin. When the new Concertación Government of Lagos, installed in early 2000, retabled the labour reform issues, the new project was substantially weakened. The bill that the Chilean Senate finally agreed to look at, in Spring 2001, had abandoned the 1999 proposal to legalise voluntary collective bargaining between more than one firm and sets of

The Transition to Democracy and the Enforcement of Markets

91

unions. In light of the entrepreneurial sector’s principled defence of liberty, this was ironic. There is no doubt then that the entrepreneurial sector successfully vetoed significant changes to the Labour Code throughout the 1990s and beyond. In two key respects the political processes of 1999 and early 2000 reflect on the lasting significance of the accords, implicit and explicit, made during the period 1991–3. It did so, first, in the language the business sector (successfully) used to oppose reforms, and, second, in the significance attached to the atomized labour regime. Again, entrepreneurial leaders wove together the demand for an almost unconditional respect for negative liberty in the private sector with the viability of the democratic regime. In the week leading up to the vote in the Senate on December 1st they raised a media campaign which sought to combine legal expertise, technical authority and threats of economic and political reversal to bring pressure to bear on the political process. The SOFOFA paid a prestigious constitutional lawyer of the Catholic University, José Luis Egaña, to prepare a report on the ‘legality’ and ‘institutionality’ (a veiled wording for a threat to the democratic regime) of the labour reforms. The right organized the signing of an open letter by a series (47 in total) of prominent economists (many of them officials during the Pinochet regime), warning of the effects on the economy of the reforms. This was supported by statements from a collection of experts on Wall Street which helped to whip up expectations in the foreign investment community, whether true or false, that Chile would be a country not worth investing in if faced with further labour reforms. The President of the CPC, Walter Riesco, threatened that the ‘democracy of the accords’ had come to an end.59 The key points entrepreneurs were worried about echoed the debate of 1991–3. These were the prospect of collective bargaining above the firm level and by temporary workers, and the idea that workers could be replaced during strike action. We have already noted the Aylwin Government’s support of the entrepreneurial position on the first two issues of collective bargaining (the strike issue will be discussed in Chapter 6). The Frei Government (although it only insisted on this position towards the very end) wanted to make collective bargaining above the firm easier but, as we have seen, was opposed by key senators of its own party, who had held key positions during the period of Aylwin. The aforementioned legal report commissioned by the SOFOFA established (not surprisingly) that the labour reform bill was ‘substantially unconstitutional’ on at least six counts. These included the legal impossibility of unifying two economic entities for the purposes of

92 Labour Politics and Chile’s Transition

bargaining, the legal establishment of collective bargaining as a voluntary act, and by firms; the rights of entrepreneurial confidentiality and secrecy (non-disclosure of information during bargaining); and the pursuit of free economic activity (which would be impeded by ‘prohibitions’ such as those proposed on hiring during legal strike action). In effect, the SOFOFA was defending the supremacy of the economic component in civil rights (see Chapter 1) of the 1991 Legal Code. The document published by the SOFOFA stated the legal avenues for declaring the government’s labour reform unconstitutional. The SOFOFA again played an important role in the first meetings of the year 2000 with the new Lagos Government. Whereas it maintained an independent stance within the CPC (this time publicly) it was the academic institute most closely linked to the SOFOFA, the CEP, that hosted the first prolonged formal talks between ‘influential’ entrepreneurs and Lagos and his key ministers.60 The SOFOFA did not cease in its efforts to dilute the proposals for labour reform. It referred to the reform as an ‘attack on entrepreneurial capacity’ in Chile and, in May 2001, the CPC stated a set of ‘conditions’ prior to entering any talks on labour reform including the postponement of reforms to the private health system – another sensitive issue for the CPC – and a limit of 90 days for settling the contents of labour reform. The Lagos Government was forced to bow to these pressures. The CUT had no comparable pressure tools, showing how quickly and completely the momentum of the early transition phase had been lost. As in the early 1990s, entrepreneurs’ concerted press campaign was dominated by simple cost-cutting concerns, for example staunch opposition to the cutting of the legal length of a continuous stretch of work from 14 to 12 days. This shows how dependent entrepreneurial strategies in Chile were on simple labour hours and cost, a point we pick up on in Chapters 8 and 9. Moreover it suggests how even minor modifications to labour regulation, and therefore largely irrational fears, were allowed to dominate a major civil rights debate. Entrepreneurs also took the position that inter-firm bargaining might be allowed if restricted to certain themes such as health, hygiene and training, and as long as small and medium-sized firms (concentrating 80 per cent of employment in Chile) would be exempted. As emerged in the Lagos Government’s abandonment of the issue of supra-firm bargaining, entrepreneurs were more than successful. In Chapter 7 we shall discuss the likelihood that employers will engage in concerted initiatives in the area of training. Of particular interest to note, in respect to labour reform, is that the Concertación did have (for the first time) a majority in the Senate, as former president Frei became a senator for life. By this time,

The Transition to Democracy and the Enforcement of Markets

93

however, sufficient time had passed since the dictatorship that the Concertación no longer needed the right to act as a scapegoat for the alliance’s inability to pass substantial labour reforms. Concertación senators closely involved in the negotiations of the first labour reform, such as Alejandro Foxley, maintained the position adopted during the crisis of December 1999, against inter-firm bargaining.61 In key respects, the political pacts that had been entered in the first crucial period of Aylwin’s government remained on the books.

Conclusion The sources of the Concertación governments’ adoption of ‘market neutrality’ were many. But above all we have argued that the Concertación chose to respond to the fears, rational or symbolic, of the entrepreneurial sector, and that there were compelling reasons for it to do so. An important factor was the power that policy-makers believed entrepreneurial leaders had in affecting the economy, if these leaders were to initiate a campaign against the government.62 Another factor was the understandable temptation to co-operate with the right in securing a successful management of the economy and of the transition. Our discussion of sectoral accords supports these conclusions. First, the accords that worked did so because employers in the sectors concerned had a pragmatic attitude to labour. Second, the position of the transition government reflected the need to accommodate to the fears of the CPC. And yet, despite the government’s good reasons, the effects of its ‘market neutral’ position were hardly positive. As we have tried to show, the Aylwin Government’s decision to abstain from social pactmaking, and to discourage its bipartite form, was synonymous with allowing employers to dismantle institution-building in general. This meant that a chance was missed to consolidate the more technical approach that many labour leaders and some entrepreneurs had developed, on issues such as worker training, availability of employment information, and security between jobs. The accords of the metallurgical sector indicated that the ‘technical’ argument against such agreements was unsupported. The accords had no real effect on wage setting. They were flexible enough not to compromise the survival chances of the individual firm, despite the heterogeneity of this sector. And finally, the case of the metal sector showed that even the most dedicated Communist sectors of the union movement were inclined to focus on precisely the sorts of labour issues that tended to favour the development aims that the transition government had set out for itself.

94 Labour Politics and Chile’s Transition

Above all, the metallurgical-sector accords demonstrated that strong unions at the sectoral level were not necessarily a hindrance to economic development. Indeed, as the president of ASIMET had himself indicated, the very existence of a powerful union had made a form of long-term regulation both a desirable and a feasible goal. The experience suggests that similarly specific accords at the national level might have influenced the relationship between the state and the private sector in a more general sense, by making the private sector more accountable to public policies and the interests of a larger community. In this respect, the significance of the Aylwin Government’s position went beyond the confines of its own term in office. This leads us to restate the point that it was not so much the fact that the private sector was large that mattered. What mattered was the attitude that its leaders had towards working with democratic institutions, and what tools it developed to that effect. Thus the question has to be raised as to how far and in what sense the newly elected government had an ‘embedded autonomy’ vis-à-vis the private sector, in Evans’ sense of being able to establish original goals and enticing the fluid co-operation of the private sector in the pursuit of them.63 State autonomy in Chile during the early 1990s was hardly of this kind, limited as it was by the confined spaces for democratic authority that the preceding regime had left behind. It was not so much a question of what a capable and independent set of policy-makers could ‘offer the private sector’, to paraphrase Evans, but of what the private sector could offer the nascent political élite in terms of making governability possible – and on what terms.

Part III Social Citizenship and Chile’s Labour Reforms

This page intentionally left blank

6 Democratic Legitimation and the New Labour Code That Never Was

Social citizenship is a state that is composed of a number of factors. Yet constitutional rights carry a special political weight. Therefore, the case of legal reform epitomized, more than any other area of labour reform, the symbolism and the popular expectation raised by political democracy’s inception in Chile. Above all, it called to mind the classical Marshallian sequence, when political democracy was a seminal breakthrough with substantial legal significance in several spheres. The expansion of republican freedoms was (at least, according to the authoritative analyses) a vehicle for the development of the liberal economy; and, hence, the expansion of social freedoms became an inexorable part of the process. 1 We have insisted that the modern sequence is different. In the Chilean case, legal political rights, and the economic component in civil rights, became substantially strengthened. However social rights were not. The openness of the modern market economy entails that fewer social rights are possible to modify by states without strong private sector support. Chile therefore did not follow the classical sequence. Entrepreneurs in Chile were concerned, primarily, to cement the privileges, as they were perceived, of the past, and the shape of workers’ rights was the issue that was foremost on their agenda. It was not, as in Laski’s interpretation of the French case, a sub-issue, a trivial byproduct of other concerns. This area of difference has important implications for how we understand the questions of change and sequences of change during Chile’s transition. In this chapter we discuss how far the prospects for establishing occupational citizenship were affected by the political legacies and conceptions of labour’s place in production which meshed to dominate the legal debate at this crucial time. There are differing views about how far the organizations of labour could be strengthened by encouraging centralization with legal measures in Chile. 97

98 Social Citizenship and Chile’s Labour Reforms

Certainly, this was a political step almost impossible to take in the early 1990s, as we saw in Chapter 5. This aside, our discussion so far would seem to suggest that strengthening local unions was an important and basic objective. It was important given the tradition of firm-based bargaining and the need for sectoral and national unions to develop a strong local base (Chapter 4). The empowerment of the local union’s role would also provide one element of pressure towards a definition of labour flexibility that encompassed arrangements for the development of human resources. Hence there was both a political and a technical rationale for focusing the bulk of a new reform at the lowest level. Accordingly, our comparative discussion (in the first section) is intended to review the role of the local union in contexts where labour flexibility has been modulated to encompass the protection of human resources. In the second section we look at the nature of the political debate about labour reform. In particular, we seek to pinpoint the clear political rationale behind the pressures to maintain a framework that facilitated a short-term style of management. Again, we observe the tacit acceptance by the transition government of the ‘confines’ of its period, an acceptance that served it in the task of management. Finally, we examine the details of the Labour Code that was revised during 1990–1, with an emphasis on dismissal, representation, contractual flexibility and collective bargaining, and assess its significance in the light of present day legal debates.

Labour flexibility in a comparative perspective The deregulation school of theorists has dominated the discourse on labour flexibility since the 1980s. It is contended that the more legal interference from unions is minimized, the greater the chances must be of achieving a ‘true’ market wage. It is on this basis, for example, that Japan is sometimes placed on the US side of the equation vis-à-vis a centralized European variety. The validity of this narrow focus on centralization can be questioned. For example we may point to similarities within the ostensibly ‘opposed’ labour systems of Germany and Japan, 2 and specifically the long-term definition of workers’ rights (the element of occupational citizenship) that distinguish both. In both countries, organizational cohesion and continuity of local unions has been important. The competition for workers’ loyalties – including multiunionism – did not enter the picture.3 On the other hand, unions remained connected with day-to-day work affairs, and this underpinned their authority when bargaining on the organization of work (and not just wages).4 These features can be taken to indicate the minimal kinds

Democratic Legitimation and the New Labour Code 99

of legal measures required for establishing a form of ‘in-house’ protection of human resources in a case like that of Chile, where the role of the union was already central to the definition of contracts. Firm-level unions in Chile were weak. Workers covered by collective contracts did not go over 10 per cent in number. The national unionization rate was very low, roughly 12.5 per cent in 1990, compared with historical levels of about 30 per cent to 40 per cent prior to 1973. Most union members (roughly 70 per cent) belonged to firm-level unions (in the private sector the rate was much higher), and on average unions were small (see Table 6.1). To this should be added the diffusion of representation, within both sectors and firms.5 Turnover rates were also high.6 In short, workers were unable to take advantage of the falling levels of unemployment that characterized the period between the mid-1980s and the mid-1990s to achieve control gains ‘from below’ (Table 6.1). But what were the attitudes of decision-makers in Chile regarding legal reform?

The politics and development discourse of labour legislation General issues in labour reform The Aylwin Government’s need for political legitimacy was perhaps the central reason why labour reform became a key issue of the transition. But the resort to legal regulation by both labour and employers’ leaders, and the latter’s marked politicization of it, was also an indication that the social actors had failed to compromise on a sustainable regime of flexibility at their own behest. The centrality of provisions for dismissal among the concerns of both parties was an example of a shared defensive position. 7 The CPC was vehemently opposed to what was a fairly moderate set of proposals by the CUT. These entailed the specification of a cause for dismissal by either (a) the ‘economic development’ of the firm; or (b) market conditions that would make current operations unviable. The Confederation opposed any statement of cause, and proposed instead that the ceiling on individual dismissals should be set at no less than 25 per cent per year. The CPC also opposed any proposals that would enhance the role of labour representation. It remained vehemently opposed to provisions that would restrict the right to collective bargaining to unions (excluding bargaining groups – see below), and it resisted any broadening of the right to strike. In addition, the CPC rejected more positive inducements to labour participation in company life. For example, it resisted the broadening of matters that could be discussed legally in collective bargaining; it rejected the idea that agricultural workers should be able

100

Table 6.1

Year

1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999

Aggregate figures for unionization and employment (a)

(b)

(c)

(d)

(e)

(f)

(g)

Total union members (firm-level) (000s)

Average size of union (000s)

Unionization rate (%)

Total labour force (000s)

Unemployed (000s)

Rate of unemployment (%)

270,9 296,9 311,6 349,7 417,5 455,9 473,9 460,2 448,4 417,8 426,8 401,6 390,5 359,4

72 72 69 71 69 71 67 60 54 50 48 45 43 39.4

4312.0 4425.3 4656.3 4805.3 4888.6 4983.9 5199.8 5459.0 5553.8 5538.2 5600.7 5683.8 5851.5 5883.4

449.2 424.0 370.8 341.9 363.1 353.2 322.4 349.7 431.1 363.8 302.0 303.6 419.2 529.1

10.4 9.6 8.0 7.1 7.4 7.1 6.2 6.4 7.8 6.6 5.4 5.3 7.2 8.9

i

ii

iii*

Employed workers covered (%)**

9.0 9.5 9.6 10.6 12.4 14.1 13.9 12.5 11.9 11.5 11.7 10.9 10.5 9.9

6.3 7.7 6.7 7.3 8.5 9.2 9.1 8.4 8.1 7.5 7.6 7.1 6.7 6.1

10.0 10.6 10.4 11.4 13.4 15.1 14.9 13.4 12.9 12.3 12.4 11.5 11.3 15.3

6.6 7.1 1.0 7.7 8.9 9.5 9.9 9.5 9.6 8.7 8.2 7.7 7.4 4.3†

Notes: * 1999 – first six months; (i) all unionized as percentage of total labour force; (ii) unionized in firm-level unions as a percentage of total labour force; (iii) all unionised as a percentage of the employed. ** Covered by collective contract. † October 1999. Source: Dirección del Trabajo; ILO (1998).

Democratic Legitimation and the New Labour Code 101

to negotiate through permanent organizations, or at all; and it refused measures to broaden the collection of union dues as well as the lifting of the ban on unionization up to one year after the formation of a company. Most of all, entrepreneurs resisted any notion of establishing forums for discussion above the firm level, even in areas where firm bargaining was already proscribed or difficult, and despite the CUT’s position that all decisions by such commissions would be made by consent. One area where the concern with what might be termed ‘petty’ costcutting was particularly evident was export agriculture. Employers persistently resisted proposals such as an increase in annual holidays from 18 to 21 days (Article 67); the calculation of overtime work as Sunday payment; and the calculation of Sunday and holiday pay on the basis of average income over a period of time (Article 44). Employers did not want to ‘guarantee’, but only to ‘proporcionar’ (proportion) the life and health of temporary workers (Article 171). 8 Long arguments were sustained over the distance between lodgings and the workplace, which would determine an employer’s obligation to supply transport, as well as the obligation to provide proper hygienic conditions in which to prepare food, and the idea that nurseries should be provided for the use of female temporary workers. 9 The CUT also proposed to adjust a section of the Pinochet Labour Code, Article 12, which gave employers the facility to modify the contents of the work contract. It is interesting to note that the CUT did not want to repeal this legislation, but merely to allow workers some participation in modifications of work tasks. The CUT proposed that a revised Article 12 should include the following additions: (i) the possibility that the union take part in such alterations at a general level; (ii) the obligation that the union be consulted in cases of individual changes, including the permanent insertion of overtime work; (iii) that, in such cases, it would be the most representative union that was consulted; and (iv) that abuse of Article 12 be sanctioned as a ‘disloyal practice’ in law. The CPC wanted to maintain Article 12 as it stood, and opposed all these suggestions. The role of the CPC The position of entrepreneurs appeared to be far more open to change at the time of the bipartite labour commissions of spring 1990, set up after the accord between the CPC and the CUT earlier that year. A ministerial memo suggested that the following themes were at the time subject to consent in Commission 2 on collective bargaining: that the union be the sole medium for bargaining; that indefinite strikes be installed, and that a prohibition be placed on hiring during legal strikes. In addition it

102 Social Citizenship and Chile’s Labour Reforms

was proposed that tripartite commissions and minimum wages (tarifados) be established in areas where collective bargaining was not permitted.10 A plausible reason for this flexibility might be the dominance, in the early stages, of lower-ranking employer representatives, who were less predisposed towards an ideological discourse. Many of these were technical staff or employers with experience in dealing with labour in quasicorporatist settings.11 However, the CPC’s core leaders eventually became very much involved in the process of labour reform. The political profile of CPC leaders was accentuated through a variety of channels, including the media and the research institutes of the Right, as well as through the activities of constitutional senators close to the former dictator, in particular William Thayer.12 The Right drew strength from the fact that a similar coincidence of opinion did not, and could not, exist between the CUT and the Concertación. Finally, prominent CPC leaders from the open sector took it upon themselves to represent their sector directly at Congress.13 They were, even in the opinion of Socialist representatives, far better equipped and more timely in the presentation of their lobbying than were labour leaders.14 The ‘facilitating confines’ of the Labour Ministry The contours of a non-interventionist approach also became progressively apparent in the government’s evolving approach to labour reform, and in the Cabinet’s working relationship with the veto-holding opposition in Congress. Evidence that a political process of rationalization was taking place can also be deduced from the way in which the Aylwin Government single-handedly scaled down important areas of its legislative programme before actual negotiations with RN took place. This occurred partly in response to consultation with CPC leaders. The Concertación had previously entertained the idea of making the response to union petitions for bargaining above the level of the firm obligatory, even when any agreement was subject to the consent of both parties, and negotiations could be abandoned. 15 By the late summer of 1990, at least, this idea had been rejected in favour of a purely voluntary scheme. According to the SOFOFA’s leading technocrat, this theme ‘was discussed and we found an accord’.16 Other areas that were dropped prior to the presentation of a law project included the following: the ending of discretionary dismissal; allowing temporary workers to negotiate through permanent unions and repeal of the ban on matters of negotiation. Also dropped was the elimination of the right to hire workers during strikes. The labour Cabinet pinned itself down to a restricted reform through the nature of the political relationship it established with RN. According

Democratic Legitimation and the New Labour Code 103

to a leading RN negotiator of labour reform, real negotiations took place outside Congress in the form of several personal meetings with the Labour Minister, and sometimes with union leaders. In exchange for its co-operation, the Concertación promised not to re-reform the labour code.17 A most evident sign of the Concertación’s covert consensus policy was the decision to dispatch the key law projects to the Senate first, and not to the House of Deputies, where the government held the majority. In this way a more limited project was guaranteed, and RN was protected from public political fuss. Constitutional senator William Thayer did not lament the government’s two-track policy: ‘They wished to shout out loud that a new Labour Code established profound changes to the existing code, and blame us, the designated senators, for certain defeats.’18 UDI, the right-wing party with which most entrepreneurs chose to identify, had not participated in the labour reform. Nevertheless, by the time the key bills had been passed, and to some extent tested, the evaluation of employers was decidedly favourable. According to Heiremans, who represented the most intransigent sector within the CPC, ‘The area in which the government consulted us most was on labour, especially the labour legislation, which resulted in a prudent reform.’19 ‘After all our discussion, the labour reform came out moderate in the end’, was the appraisal of the CPC’s president, Guzmán. But the evaluation of the constitutional senator, William Thayer, is perhaps the most important for us to mention, because all sides tended to agree that he was indeed the most qualified individual in parliament on the subject. He intimated that the Aylwin Government and the opposition had a tacit agreement not to politicize the small extent of the new reforms. The Government had a spectacular task. It had to give the impression that much was reformed, whilst in reality reforming little . . . In order to create a ‘new labour code’, they transformed the old into new separate laws . . . something that[also] . . . happened during Chile’s independence. They wished then for Chile to be a monarchy and not a republic, so that it would not resemble Spain . . . They declared that they had taken their inspiration in the French Revolution and the English monarchy, all as long as there was no hint of Spain . . . But 80 or 90 per cent of the Spanish legislation continued on the books: I know the texts . . . but it could not be said. This is what is happening here. The old laws continue on the books, but it cannot be said that they do. This is the issue. The old laws continue in some 80 to 90 per cent. And in my view the changes have been for the better.20 The emperor has kept his clothes.

104 Social Citizenship and Chile’s Labour Reforms

The legal reform under scrutiny Stability in employment: Law No. 19.010 on the termination of contracts The significance of legal restrictions on dismissal depends on the extensiveness of other mechanisms that protect workers’ occupational rights. In the case of Japan, for example, a fairly liberal text (Articles 627 and 628 of the legal Code) was countervailed by extensive employment security schemes, as we observed. In Germany, measures to protect and reemploy workers were quite extensive; and unions in practice could demand that a dismissed worker remain in his/her job, at least until the final hearing of any appeal. 21 The Chilean labour market, of course, did not have workable reemployment institutions. Nor was an exchange between job security and flexibility at the firm level a common phenomenon. Indeed, job security was fairly slight. In 1991, the rate of dismissal (as a percentage of the workforce) was very high in some sectors, such as industry (16.6 per cent) and services (25.8 per cent). 22 The legislative solutions, (basically three) appeared to represent a compromise deal.23 The ceiling on the number of years of work used to calculate the level of indemnity was raised from 5 to 11 (but not eliminated, as the government wanted). The fine for unjust dismissal was raised, albeit not as much as the government wanted (25 per cent versus 20 per cent). Most importantly, discretionary dismissal (Art. 155f) was replaced by a new set of rules, which implied the statement of a cause.24 In reality, how far did these changes distinguish the new legislation from the Pinochet Code? The clause relating to dismissal that replaced Article 155f, ‘Article 3’ – or, popularly, ‘the necessity of the firm’, appeared so broad as to be virtually empty: ‘an employer can terminate a contract of work using as a cause the necessities of the firm, establishment or service; such as those deriving from rationalisation, modernisation, falls in productivity, changes in market conditions or in the economy, which make it necessary to dismiss one or more workers.’ Further, the clause added as a permissible reason ‘inadequate technical or work performance’ (by the worker).25 The law on dismissals of 1966 (Law No. 16.455) also had a clause that related to the ‘necessities of the firm’ (Article 2, No. 10), but this made reference ‘only . . . to necessities which have to do with the technical functioning of the firm . . . not to any event which may provoke a lowering of profits or an economic regression of firms’.26 The association of labour lawyers in Chile, the Asociación de Abogados Laboralistas (AALC) proposed an alternative wording which, they said, was more in line with Chilean tradition. It read, ‘Those

Democratic Legitimation and the New Labour Code 105

shall be considered necessities which derive from economic, technological, or structural causes, or which relate to rationalisation or modernisation, as long as there is a direct relationship between the phenomenon in question and the dismissal.’ 27 The real, more than merely formal, significance of the ‘statement of cause’ hinged on the efficacy of the appeal system. But this remained unattractive to workers, who could ill-afford to postpone the collection of their indemnity – effectively their unemployment insurance – for the eight months or so that an appeal could be expected to last. In our 1992 survey of local labour leaders, 79 per cent of respondents felt that some of the dismissals that occurred in their firm could be challenged in court. However of these, two-thirds had refrained because of the length of time, and the costs involved in the process (see Table 6.2). As a percentage of the workforce, the number of dismissals during the first year of the new legislation was greater than in any of the previous five, at 8.5 per cent versus an average of 7.5 per cent. 28 The increase stems partly from the use of other clauses of dismissal apart from Article 3 (particularly Article 2) between 1990 and 1991 (see Table 6.3). This removed the last seeming advantage of the new legislation: the raising of the ceiling of years that formed the basis for calculating the size of indemnities from 5 to 11 years. Under Article 2, which constituted 37 per cent of dismissals in 1991, the worker collected no indemnity at all. Lawyers acting on behalf of employers in related cases were also inclined to believe that the new dismissal regime made little practical difference.29 Representation and competence of the collective contract In our comparative section we emphasized two key characteristics of qualitative flexibility at the level of firms: (i) the organizational cohesion of unions; and (ii) the linkage between the union and consultation.

Table 6.2 tribunals

Percentage of firms where unfounded dismissals tend to be referred to

Dismissals have not been referred Dismissals have been referred No such dismissals in the last year Missing case Total Source:

Union survey, author, 1992.

Percentage

Frequency

52.8 25.9 21.3 –

159 78 64 1

100.0

302

106 Social Citizenship and Chile’s Labour Reforms Table 6.3 Comparison of Article 155.f of the 1979 Labour Code with Article 3 of Law No. 19.010, as percentage of total separations Years

Total separations

Art. 155.f./Art. 3

Percentage of total

1990 1991

348 693 372 520

76 674* 56 640

21.9 15.2

Source: Dirección del Trabajo.

How far were such characteristics reflected in the debate on labour reform in Chile; and in the final outcome; and why? Multi-unionism Concertación spokesmen had made it a point from an early stage to promote competition between unions.30 A single-union system was not commanded by law in either Japan or Germany, we recall, but it operated effectively in both.31 In Japan, only 4 per cent of firms had more than one union. 32 Even the ILO regarded the single union as desirable for stable industrial relations. 33 In Chile itself, it is questionable how far the existence of more than one union translated into real competition. This argument is more likely to apply in contexts of union strength and where national bureaucratic structures play a dominant role. Not surprisingly, 82 per cent of union presidents in our survey preferred a single-union system by law (see Table 6.4). We also found no evidence to suppose that unions in double-union firms were any younger (and hence more dynamic) than unions in single-union firms. In other words, double unions reflected at best a fragmented type of representation, and at worst a source of divisionism that employers could exploit.

Table 6.4 Should there exist (by law) only one union and bargaining group in a firm? Answer

Percentage

Yes No It dosen’t matter Not known/no answer Total

Frequency

82.8 13.6 1.0 2.6

250 41 3 8

100.0

302

Source: Union survey, author, 1992.

Democratic Legitimation and the New Labour Code 107

Coverage The coverage of union agreements was one area in which the government did seek to expand the basis for union representation. In France and Spain, unions could make up for their weak membership base through the practice of extending collective agreements to workers outside the union. 34 We noted earlier the role of the German courts. A similar process was very difficult in the Chilean case. In fact, the Concertación’s proposal was relatively modest, in that it aimed at making the extension of collective provisions automatic only in the case of newly-registered workers. Not surprisingly, this project was rejected in Congress. The government’s negotiated proposal, to make union dues obligatory for non-unionized workers only when these had provisions in the collective contract extended to them, was passed.35 Another telling weakness in this reform was the placement of the collection of union dues under the authority of the employer. In our survey of local labour leaders, 40 per cent of respondents said that 75 per cent were not being paid in their firms.36 Finally no obligation had been placed on employers to provide information on the wage structure of the firm. As a consequence unions were forced to consult non-unionized workers on an individual basis regarding their level of benefits and pay. 37 Workers in precarious employment Certain groups of workers remained outside the central framework of the legal Code. For example, no provisions ensured the right of the vast group of temporary workers that existed in Chile to bargain collectively. Maritime workers were allowed to sign a ‘provision for a place of work’. This could be individual, however, and did not involve a bargaining relationship (see Chapter 5). 38 A similar situation pertained in the area of export agriculture, where the national grouping of agricultural unions, the Comisión Nacional Campesina, had petitioned for the establishment of a ‘communal union’, with the right to strike.39 The government’s project divided unions into four sub-groups, and ignored the communal union.40 Permissible subjects The restriction on permissible issues of collective bargaining in the Pinochet Labour Code was significant. It was hardly realistic to expect that a process of flexible democratization could begin, or that union activities could coincide with day-to-day consultation, as long as this provision remained on the books. It is noteworthy that Japanese law

108 Social Citizenship and Chile’s Labour Reforms

contained no ban or specification of matters for negotiation.41 Nor (in reality) did German law.42 The Aylwin Government’s project represented at least a modest attempt to countervail the all-inclusiveness of the legal proscription (it did not propose to eliminate it) by legitimizing agreements that were related to ‘information systems, communication, proposals, consultation, and other similar areas’.43 Congress rejected this addition, however, and the final formulation adopted literally the most important overall proscription of the 1980 Code.44 The final ban contained any matters that ‘may restrict or limit the employer’s exclusive right to organize, lead and administer the firm’.45 There was another clause in the Chilean Labour Code that placed an obligation on employers to provide the ‘information necessary’ to explain the ‘economic circumstances’ of the firm. But if matters to be discussed were already limited, how did this provision affect the discretion of management on the question of disclosure? In the survey of local unions cited earlier, 65 per cent of respondents said that the information made available to them was not sufficient.46 Unions versus bargaining groups: a case of complementarity? But the most significant drawback in the outcome of labour reform was perhaps the fact that bargaining groups were allowed to remain, because their presence affected the competence of the local unions directly. Bargaining groups were introduced as a feature of the Pinochet labour reform of 1979–80 (as we noted in Chapter 2), ostensibly to provide the forum for a less conflictive form of labour relations than traditional unionism had been. In reality, however, their definition was such as to constitute an alternative form of labour integration to unionism itself. Not surprisingly, the Aylwin Government felt compelled to propose the repeal of bargaining groups in its legislative project, despite the strong opposition it faced from the CPC.47 Ultimately, the annulment of bargaining groups failed to pass in Congress, and two years after the negotiations, Minister Cortázar had adopted a more optimistic perspective on the question of the bargaining groups. This was based on the notion that bargaining groups could somehow act as a form of initial support for unionization. 48 But was it realistic to view bargaining groups as a source of support to unions? Could they be an extended arm of collective bargaining somewhat similar to the ‘works councils’ in Germany, which had a legal right to information about the firm,49 or the practice of consultation in Japan? Bargaining groups had no more access to information, and they had no more rights to discuss matters that violated the discretion of employers,

Democratic Legitimation and the New Labour Code 109

than had the unions. Bargaining groups were not demarcated from unions in Chile. They were set up to compete with them.50 By definition, a bargaining group was any number of workers which, on the initiative of the workers, assembled for the purposes of bargaining, and which, upon the signing of a contract, dissolved. 51 It was a form of worker representation with no permanence, and no coherent, long-term purpose or role within the firm. But that was not all. As in the case of temporary workers, bargaining groups would normally sign conventions, although this form of contract could also (and was occasionally) used by unions. A collective contract was associated with a procedural framework for negotiation, including the obligation of employers to respond to union petitions, eventually with a ‘last offer’, the strike vote in response to this; and, most importantly, the right to strike. On the other hand, bargaining groups signing conventions could not strike, and no rules or procedures were stipulated to help guarantee that a real negotiation take place. 52 In Chile, in 1990, 25 per cent of establishments negotiated with more than one group of workers (see Table 6.7, col. A).53 This situation needed to be critically addressed by the legislation, and not encouraged by it.54 There were also indications that two essential preconditions for the Chilean convention to function as an act of bargaining often failed to apply. The first such condition would be that an actual dialogue should take place. The survey of industrial unions cited earlier showed in this respect that 38 per cent of unions (44 out of 117) which had negotiated conventions rather than contracts had chosen not to present an alternative project to that composed by the employer (Table 6.5). Of the remaining 62 percent, 12 percent had a convention imposed on them. Another, and perhaps a more essential, prerequisite for bargaining groups to function (and one which was stipulated in the law), was that Table 6.5 In which of the following ways did the signing of the last convention take place? Answer Only the employer presented a project The union also presented a project Sub-total Non-valid cases* Total

Percentage

Frequency

62.4 37.6 100.0 –

73 44 117 185

–

302

Note: * Non-valid cases include unions which did not sign a convention. Source: Union survey, author, 1992.

110 Social Citizenship and Chile’s Labour Reforms Table 6.6 formed?

How was the negotiating group which signed the last convention

Answer Voluntarily, without initiative from the employer By initiative of the employer Sub-total Non-valid cases* Total

Percentage

Frequency

72.7 27.3 100.0 –

80 30 110 192

–

302

Note: * Non-valid cases include unions that have not signed convention. Source: Union survey, author, 1992.

their formation should be ‘spontaneous’ or ‘voluntary’. Our survey suggested that, in this respect, in 1 in 4 valid cases (or 30 out of 110), the negotiating group that existed in parallel with the union had been formed by the employer (See Table 6.6). A single case of a metallurgical firm illustrates the way in which conventions could become an instrument for undermining the authority of the union. Table 6.7 compares the contents of a convention signed in the metallurgical firm, Sports Wagon, in November 1991 with the contents of the union and employer proposals for a collective contract that ensued a few weeks later. Table 6.7 clearly shows that the employer offered far better terms to those workers who had been willing to sign a convention. It also indicates the clear contrast between the nature of reward and human development emphasized by the employer and by the union, respectively. Whereas the former highlighted rewards for individual workers, the latter highlighted human development for workers on equal terms, including access to study time, and professional training. The union also highlighted the formalization and equalization of the wage system as well as broadly-based production incentives. On the other hand, the fact that all of these items were rejected testifies to the importance in the labour reform of 1991 of the maintenance of the employers’ rights to exclude all such themes from collective review. 55 It is a common observation that, in comparison with other situations of democratic expansion in Latin America and elsewhere (Spain and South Korea, for instance), the rise in the level of union activism and industrial relations conflict in Chile was remarkably low. 56 Levels of unionization increased by a third in the first two years of the Aylwin Government (from 11.5 per cent in 1989 to 15.7 per cent in 1991), yet both the levels of unionization and the number of (legal) strikes were to fall notably in

Table 6.7

Comparison of a convention and proposals for a collective contract in a metallurgical firm, 1991*

Items

Convention signed outside the union (1.11.1991)

Union proposal for a contract (17.12.1991)

Employer’s first response to union (27.12.1991)

Employer’s last offer to union (21.1.1992)

Duration Minimum wage

4 years 50 000 pesos p.m.**

2 years 99 000 pesos p.m.

4 years 40 000 pesos p.m.

Initial wage rise, and subsequent adjustments Incentives

With inflation, adjusted 3 times p.a.

30% real rise, adjusted 4 times p.a.

With inflation, adjusted once p.a. Rejected

As before

Rejected

As before

Rejected

As before

With inflation, adjusted once a year 10 045 pesos 7 757 pesos

As before

–

–

Seniority pay

–

Subsequent wage rises

With inflation, adjusted 3 times p.a.

Marriage bonus Birth bonus

20 000 pesos 12 000 pes. p. child

Up to 120 000 pesos 120 000 pes. p.child

15 000 pesos 9 000 pesos

111

Formalization of wages

An incentive will be introduced per taxi or lorry produced if all standards observed The workers’ base wage will be adjusted to the next scale after a fixed period Single bonus: After 3 yrs 7 000 pesos After 4 yrs 10 000 pesos After 5 yrs 13 000 pesos With inflation, adjusted 4 times p.a.

45 000 pesos p.m. As before

(contd.)

Items

Training

Studies

112

Table 6.7

Convention signed outside the union (1.11.1991) Allowance will be made as long as it is paid for by the state through the tax rebate scheme –

Reward for best comrade (compañero) Reward for best worker Reward for punctuality Clothes

Holiday for two (not to exceed 120 000 pesos) Ditto (not exceeding 200 000 pesos) A party

Protective mask Union office

– –

Extension of benefits

–

–

Union proposal for a contract (17.12.1991) 30 workers p.a. to receive training either at the expense of the firm or the state (tax rebate scheme) Permission to attend courses (not paid for) –

Employer’s first response to union (27.12.1991) Rejected, but will allow hours off for courses related to the metallurgical sector Rejected. Refer to training rubric –

Employer’s last offer to union (21.1.1992) As before

As before –

–

–

–

–

–

–

3 pr trousers p.a. 2 pr safety shoes 3 cotton sweaters Yes incl. noise guard Cancelling of fee for renting union office To new union members

2 pr overalls & 2 pr safety shoes accepted – Rejected

Overalls changed to slacks – As before

Rejected

As before

Notes: * The table contains only some of the items discussed; ** In 1992, 100 dollars corresponded roughly to 36 000 pesos. Source: Copies of the relevant convention and proposals for a contract; Sport Wagon (1991a), (1991b), (1991c), (1992).

Democratic Legitimation and the New Labour Code 113

the following years (see Tables 6.1 and 6.12). Moreover, there were other signs that the qualitative strength of both representation and of strikes had weakened in notable ways. The percentage of the workforce covered by a collective contract (around 10 per cent) had barely changed since 1980, and, as we saw, the trend was to become negative after 1992 (see Table 6.1). Firms with more than one bargaining instrument continued to represent a large share of the total. The share of workers involved through bargaining groups more than doubled between 1987 and 1998, both in relative and in absolute terms. The same occurs for the number of workers involved in bargaining through a convention as opposed to a contract (see Table 6.8). But the position of the union was also weakened in a number of ways during this time. Its average size fell from 100 in 1981 to 71 ten years later, and finally to 43 members in 1998 (see Table 6.1). The decline in union strength was relative to bargaining groups. Workers involved in bargaining through unions fell from 90 per cent in 1987 to 75 per cent in 1998, and the size of each unit by almost a third (from 156 to 109 members). The representation of bargaining groups, meanwhile, almost tripled: from 10 per cent to 26 per cent from the late 1980s to 1998 (see Table 6.9). It is clear that bargaining groups are a feature of bargaining situations where workers are weak. This can be seen inasmuch as they are far more prevalent relative to unions where bargaining units are small. In 1998 they constituted 75 per cent of instruments in units with 7 or fewer workers, compared with 24 per cent in units with 100 to 249 members (see Table 6.10). Finally, a real wage adjustment is more likely to be achieved and bargaining less likely to fall through where it is a union and not a bargaining group that negotiates (see Table 6.11). Still, as far as wage gains are concerned, what matters in the context of our discussion is not so much the level achieved, but the trade-offs involved. We saw this clearly in the case of the union in Sport Wagon above. The subject of bargaining was absent from both the Frei Government (1999) and the Lagos Government’s (2001) legislative proposals for labour reform. The right to strike and act collectively The strike can be an important political weapon, even in systems that are characterized by compromise and consensus. For example, strikes have been modest in Japan in terms of person-days lost, but short and preventive strikes or related job actions have been important during some periods. The ‘right to act collectively’ was stipulated in the country’s constitution, and was used to keep employers to the job-security regime with support from the courts.57

114

Table 6.8

Collective bargaining by type of contract

Year

A* No.

1987 1988 1989 1990 1991 1992 1998

na na na 1 926 2 137 2 212 na

Instruments %

24.6 18.3 17.6

Workers involved

Contract

%

Convention

%

Contracts

%

Convention

%

1 332 889 1 513 1 570 1 839 821 1 343

79.1 64.0 64.8 65.4 72.2 64.1 63.4

352 506 821 829 709 459 775

20.9 36.0 35.2 34.6 27.8 35.9 36.6

154 777 78 956 140 489 137 642 186 337 80 691 134 877

84.2 61.4 63.4 74.7 80.5 75.0 67.0

28 971 49 557 81 150 46 741 45 085 26 841 66 341

15.8 38.6 36.6 25.3 19.5 25.0 33.0

Notes: * A = Firms with more than one bargaining instrument simultaneously. Source: Dirección del Trabajo.

Table 6.9

Unions versus bargaining groups in collective bargaining, selected years

Year

Percentage of workers involved through

Conventions* as a percentage of bargaining instruments

Unions

Bargaining groups

Unions

Bargaining groups

Total

Unions

Bargaining groups

90 82 79** 74

10 18 21** 26

13 13 14 18

34 53 69 71

24 33 42 37

156 125 110 109

31 43 44 69

1987 1991 1992 1998

Average size of bargaining unit

Total 109 90 84 95

Notes: * As opposed to a contract; ** First two semesters. Source: Dirección del Trabajo.

115

116

Table 6.10

Bargaining groups (as opposed to unions) by size of bargaining unit, 1998 (percentages)

Size of bargaining unit

1–7 workers

8–24 workers

25–49 workers

50–99 workers

Workers involved* Total no. involved** Instruments*

73.3 363 75.3

47.8 8 200 49.9

32.0 21 264 32.2

28.4 33 993 27.7

24.4 44 313 25.4

77

525

599

491

291

Total number**

Notes: * Percentages; ** Bargaining groups and unions. Source: Dirección del Trabajo.

100–249 workers

250–499 workers 23.2 25 966 24.4 78

500– workers 21.8 67 119 24.6 57

Total number – 201 218 – 2118

Table 6.11

Mode of agreement on wages by type of bargaining group, 1998

Mode of agreement on wages

Percentage of instruments Unions

A percentage depending on inflation A nominal percentage Other mode No agreement is reached Total involved

67.5 65.7 57.3 53.7 1 362

Bargaining groups 32.5 34.3 42.7 46.3 756

Percentage of workers involved Total 100 100 100 100 2 118

Unions 77.1 77.9 61.0 59.3 148 986

Bargaining groups 22.9 22.1 39.0 40.7 52 232

Total 100 100 100 100 201 218

Source: Dirección del Trabajo.

117

118 Social Citizenship and Chile’s Labour Reforms

The strike issue was certainly a major concern to the CPC in Chile. It objected to proposals in the government’s project to put a stop to the legal 60 day strike limit,58 and to ban the right to hire workers during strike action.59 But it was difficult for the right to dismiss indefinite strikes, since this had come to be seen as a feature of the democratic regime. 60 The change implemented was cleverly counterbalanced by other provisions, however, and it is possible to argue that it left labour worse off in many ways than it had been under the Pinochet code. The automatic dismissal of workers after 60 days of strike action was abolished, but it remained something of a cosmetic change, because strikes rarely lasted that long. Other legislative changes were far more significant. The date after which individual workers could resume work, for example, was reduced from 30 to 15 days,61 and a provision was introduced by which a strike would be automatically called off after 50 per cent of the workers had dropped out of the protest.62 According to sectoral leaders, this made the sustaining of strike action almost impossible.63 In the years 1991–3 legal strikes became more frequent, but after this initial transition period, and after the new legislation, they began to decline (see Table 6.12, col. a). The average duration of strikes was notably below the 1980s average and under the crucial 15-day limit, particularly for smaller unions and groups (Table 6.12, col. g; Table 6.13, col. d). The rate of participation in strikes in Chile is remarkably low. It remained so even in the early transition years where legal strike action was at its most fervent. Moreover, the number of strikes carried out was far lower in small bargaining units, and illegal strikes occurred mainly in larger firms (see Table 6.13, cols. b and c). A final setback in the new Labour Code was the maintenance of the right of employers to replace workers from the first day of strike action. The new provision that employers must offer at least an adjustment for inflation hardly diverged from the established norm. In the area of temporary work, employers objected to bargaining, basically because of their opposition to the right to strike. 64 Their opposition to bargaining above the firm level was based in their belief that the magnitude of consultation with labour was equal to the magnitude of potential conflict. In Chapter 5 we saw that the temporary workers’ capacity to strike remained curtailed in the Lagos Government’s early reform proposals. The lifting of the ban on hiring during strikes was no longer supported by the alliance’s own senators. When, in Spring 2001, the Senate finally voted to legislate again in the area of labour reform, the proposal to eliminate hiring during strikes was completely abandoned. The only amendment proposed was a slight increase in costs to employers of hiring replacements during this time.

Table 6.12

Aggregate strike statistics (a)

Year

No. of strikes

Legal 28 68 57 31 36 38 42 41 81 72 101 176 219 247 224 196 187 183 184 121

(c)

Workers involved – public & private (000s)*

Illegal

Public

112 106 132 263 245 282 268 316 262 288 289 207

10.7 22.5 14.9 6.9 3.6 3.6 8.5 3.9 9.9 5.6 17.9 25.0 46.2 27.0 59.7 16.2 24.7 25.8 19.5 12.6

Notes: * From 1987 including illegal strikes. Source: Dirección del Trabajo; ILO, 1998.

Workers involved – private sector (000s)*

Average size – public & private*

Public

Private

30.0 29.3 92.9 82.4 359.9 113.7 108.2 113.3 423.6 361.4 318.5 357.0

(d)

– – – – – – 3.8 29.7 27.7 68.7 38.7 33.1 56.5 34.1 35.1 228.3 34.2 – –

381 331 267 223 99 95 203 96 122 78 177 142 210 126 266 83 132 141 106 104

(e)

(f)

(g)

Average size – Private only*

Persondays lost in strikes

Average duration (days)

– – – – – – 98 268 262 541 150 143 200 132 119 932 – –

213.4 472.7 314.3 131.3 46.5 67.6 69.0 69.0 144.9 180.5 441.8 374.4 1 219.5 503.3 545.5 482.7 903.7 3 062.0 479.0 1 752.9

20 21 21 19 13 12 21 15 14 14 16 15 12 12 12 13 12 10 10 –

Private

268 276 704 313 1 469 403 404 358 1 617 1 254 1 102 1 725

119

1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998

(b)

120

Table 6.13

Strike data by size of bargaining unit, 1990 and 1992

Size of bargaining unit (persons involved)

(a)

(b)

(c)

(d)

(e)

Person-days lost in strikes

No. of strikes approved

Percentage carried out (including illegal strikes)

Average duration

Rate of participation (%)

1990 8–24 24–49 50–99 100–249 250–499 500– Total*/Average** Source: Dirección del Trabajo.

1991

1990

1992

1990

1992

1990

1992

1990

1992

9 702 37 582 28 533 60 541 38 073 70 761

2 949 27 573 51 674 83 213 62 555 484 237

84 136 73 56 17 18

28 66 61 45 9 4

38 47 48 43 53 67

57 50 56 51 67 0

19 17 11 16 13 6

13 13 11 13 20 9

38 48 53 54 60 62

55 35 65 60 62 72

245 192*

712 201*

384*

213*

46**

54**

15**

15**

57**

61**

Democratic Legitimation and the New Labour Code 121

Conclusion In our discussion of legal reform we did not pose the question as to whether the new Labour Code had allowed economic democracy to be established or even whether it allowed for supra-firm aspects of occupational citizenship. Instead, the criteria we used were minimal, and had to do with the role of the local unions. Our more specific argument was that the unions’ roles in day-to-day aspects of work were as essential in building control through voice as was their competence at the hour of bargaining. Only through day-to-day involvement would the union risk making concessions in the interests of long-term development, and only in this way would such information and precedence as would be needed in a relationship of this nature be at hand. It is striking how badly the new Labour Code performed, even when measured against these modest criteria. In the main, it performed badly because it was in so many ways simply a remodelled version of previous legacies, above all that of 1980. The chief heritage with which the reformers had to contend was perhaps that of the adversarial principle. The new Labour Code did not cease the promotion of a conflictive industrial relations. This can be ascertained by the way in which the role of the union was understood. First, the union tended to be set up as the sole outlet for formal consultation. Given the politicized tradition of union-management relations in Chile, this amounted to the equation of all forms of permanent consultation with politicization and with union strife in the minds of employers. Second, the mechanism envisaged as an alternative location of labour relations to the union, bargaining groups, was not a permanent or even a regular form of consultation. It was merely a tool for ad hoc bargaining set up to compete with the unions. Thus it in fact invigorated, in one sense, the conflictive nature of union presence. In addition, the legislation did not help to move labour relations forward by allowing new themes to be introduced, either with the union or through alternative routes. Instead, it perpetuated the tradition of focusing solely on short-term wage issues, despite the important steps that had been taken by unions away from this tradition in the preceding years. Finally, weaknesses in the dismissal regime remained. It is plainly plausible to arrange measures of reemployment and reeducation above the firm that will allow for a number of dismissals in the firm each year. Yet, the level of job insecurity may be felt to be so high, or so arbitrary, that labour will fail to see the sense in compromise. Little was done with the new Legal Code to reduce dismissals, or to make the regime operate in a less arbitrary way.

122 Social Citizenship and Chile’s Labour Reforms

The role of the transition government also merits a conclusive note. It could be argued that a ‘historic opportunity’ really was missed on its own account. It took nearly a decade before employers, and the right, could be enticed to seriously reconsider the subject of labour reform. By this time, some of the most salient issues of local labour relations had been muted, such as all those to do with the quality of bargaining, the continuity of worker representation and its cohesion, and specifically the question of the bargaining groups. On the other hand, the way in which the Labour Ministry team of the early 1990s managed to steer a course involving all relevant actors in the legislative process, can only be described as impressive. Even the later retabling of labour reforms as a revision of the 1991 Code was politically apt. Labour reform was indeed a master act of legitimation: to employers, of the status quo; to labour, of political inclusion; and to citizens, of democratization. Which of these audiences had more real cause to celebrate is a different question.

7 Training Policy and Decentralization

Labour law reform was the key source of public debate about social rights during Chile’s transition. However it was not the only arena through which citizenship could be deepened at this critical juncture. Social policy offered a parallel route. It was certainly the major channel through which the expansion of citizenship had been sought in the post-war period, not only in Europe but also in countries such as Chile, which had strived to establish the social mainstream in at least education and health. The potential for social policy to play a leading role in Chile was in some ways even improved during the authoritarian period, by the removal of sectoral distinctions in fields such as social insurance and the organization of labour. This both enhanced the prospects for social policy to enter the field of occupational rights, and the need for it to do so. The basic problem of occupational citizenship, as we have posed it, is related to ensuring continuity in productive existence. This might entail continuity in a particular job, or through the labour market broadly conceived, and through work or training experiences that are connected. If applied to a policy context, the key problem to examine is the nature of the incentives that policy instruments and available institutions provide to optimise such links. During the Pinochet period, a few non-selective schemes to promote training were established by the state, as was noted (in Chapter 2). The cornerstone of the new system had been to decentralize decision-making regarding the allocation of resources from the national level to individual employers. Employers would purchase services directly from the new group of private providers. The new institutions were presented as market neutral. In reality however, they were highly selective, because workers, whether employed or unemployed were not considered to be legitimate consumers of 123

124 Social Citizenship and Chile’s Labour Reforms

training. The non-interventionist perspective that inspired the Pinochet reforms therefore raises the question of whether markets, in this case individual employers, can really make neutral decisions in the area of training, and if not, what outcomes market decisions favour. In the sections below we evaluate the available evidence in favour of public co-ordination in the training market, and we introduce the case of market failure in the Chilean context. It is in the institutional and political dimensions of these market failures, that our evaluation of the Aylwin programmes should be seen.

Chile and the liberal model The case for a training policy Abundant natural resources and capital scarcity are among the economic rationales typically cited in favour of giving the state a strategic role in skill development in emergent economies. 1 Since Chile had both an open capital market and privatized training institutions, explanations for the continuing low level of human capital would have to centre on factors that are less commonly cited. These might include the lack of institutions to co-ordinate skill investments and to relocate labour in the private sector itself. Such factors suggest the need for a more engaged public policy. Trends in Chile’s labour market, specifically the process of reformalization in the 1980s, made an active training policy feasible. Following an official unemployment average of 19.6 per cent for the period 1980–4, with a further 11 per cent of workers on public employment programmes, by 1990–4 unemployment had been brought down to below 5 per cent. The destruction of the previous (pre-1973) sectorally segmented system of social provision had also allowed the most marginal and poorest sectors to enter the orbit of social benefits for the first time.2 These conditions were highly propitious. However, the nature of the reformalization – its concentration on precarious (though formal sector) wage jobs – increased the need for an active policy.3 Over this period public employment fell dramatically (from 21 per cent to 11 per cent between 1970 and 1986), while manufacturing collapsed and traditional and non-traditional exports and service sectors increased in importance. By 1994, tertiary unemployment represented no less than 64.4 per cent of urban employment, followed by secondary employment at 27 per cent. Labour reincorporation covered groups in weak bargaining positions and without training access, such as women, the young, and a continuing stream of rural–urban migation into Santiago. 4 The labour programmes of the period did not provide an adequate

Training Policy and Decentralization

125

response to this challenge. PEM and POJH were in reality no more than temporary work projects, covering 9.2 per cent and 1.8 per cent of Chile’s workforce in 1982. Another programme involving the subsidizing of labour-intensive projects only covered some 25 000 people at its height, whereas a project employing unemployed professionals included at most 250 people a year. A national training programme for youth comprised only some 11 000 workers during its single year (1986–7) of operation.5 Hence, public policy in the 1980s did not exploit other trends inadvertently favourable to the advancement of occupational citizenship. One trend was the empowerment of the state’s fiscal capacity to influence the incentive structures of firms. 6 Another was the generalization of employment conditions among workers (albeit precarious). Training instead became essentially a private sector affair. How effective were the private institutions in organizing skill formation in Chile? Market failures: problems of demand and supply In Chapter 2 we noted how a large share of SM firms felt that they required worker training (43 per cent), but that only a small percentage (8 per cent) felt able to carry it out. To this we should add that more than 95 per cent of such firms rejected the management courses on offer on the Chilean market as incompatible with their situation. This indicates that the failure of demand interacted with the nature of the structure of supply. Part of the problem was that the PTCs tended to cater exclusively to larger firms. For example, In the early 1990s, Chile’s largest grouping of private PTCs, Achecyd, directed over 90 per cent of courses under its aegis towards enterprises with over 200 workers.7 A more general problem, and perhaps a more serious one, was the tendency for supply to focus on management training. In 1990, for example, 60 per cent of participants on training courses through listed PTCs were either executives, professionals, supervisors or administrators (the latter with 45 per cent). Of the 40 per cent of participants who were workers, 32 per cent were already skilled. The figures for management versus workforce had changed further, to 65 per cent versus 35 per cent, by 1992. 8 In the case of Achecyd, one of the largest intermediary PTCs, 80 per cent of supply throughout the 1990s was anchored in courses of administration, English and computing.9 In Chile, courses directed at middle-level management represented 34 per cent of the total, compared with an average of 2 per cent for South America. Skill upgrading was lower in Chile than anywhere else (45 per cent of courses compared with an

126 Social Citizenship and Chile’s Labour Reforms

average of 72 per cent for the continent). Initial vocational education was also quite small.10

Education, training and the apprenticeship question Segmentation of education Moreover, the PTC system was not designed to mediate the key social problems that were associated with the Chilean labour market. The rate of youth unemployment over average unemployment was fairly high in Chile (2.3) in 1990 compared with the rest of Latin America (1.99) and with the developed world (1.89). 11 The particularly high vulnerability of youth in Chile was closely related to the lack of institutional links between schooling and further training, and the latter and work. By 1995, following extensive privatization, 39 per cent of primary school students and 43 per cent of students in secondary education were in the private sector. Although private schools on average had higher standards than public ones, the private sector had no uniform regime. Fifty-nine per cent of young people (20–24 years) completed secondary education, but desertion rates were commonly as high as 50 per cent in municipal schools. Only 34 per cent of those in the lowest income quintile completed. Socio-economic profile and ties to the job market were particularly tenuous for students from the technical branch, the so-called ‘circuito de los pobres’ (up from 35 per cent of total enrolment in 1992 to 40 per cent in 1997).12 At primary level, municipal schools performed worse than privately-run schools, mainly because the latter scooped up the best students. At secondary level, the trend would reverse, as under-performing students would have by then been weeded out. Only the best students remained. For those deselected, however, there was no mainstream educational system, whether in the private sector (like apprenticeship) or in the public sector, in which to continue. Moreover, in the secondary system, differentiation persisted. Helgø found that private fee-paying school students did substantially better in the labour market, and that the significant difference lay primarily in the greater resources enjoyed by those schools, not in the fact that these were privately run.13 The point to note is that early educational segmentation made it difficult to introduce the tradable (mainstream) standards needed for a fluid labour market. Vulnerability would increase further in times of crisis. During the latter years of the 1990s, when Chile was hit by the effect of the East Asian crisis, unemployment nationally peaked at 11.5 per cent, and in mid-2001 remained in two-digit figures,

Training Policy and Decentralization

127

reaching 13.8 per cent in Santiago. Youth unemployment (the age group between 15 and 24 years of age) remained at over 20 per cent.14 Apprenticeship and the PTC system Very uneven conditions of entry of young people on to the labour market generally helps to sustain the social stratification of firms. Thereby it tends to increase the difficulty of eventually introducing the occupational streams and standards that might facilitate investment in, and rotation of, skilled labour. The PTCs in Chile placed a very low emphasis on retraining courses. Unemployed workers, or workers trying to enter the labour market, were not sufficiently well organised to ‘demand’ them. But this meant that the PTCs failed to contribute to efficient labour rotation, despite the high level of turnover practised by Chilean firms. 15 Only 17.3 per cent of Chilean entrepreneurs in the SERCOTEC survey mentioned earlier were aware of the rules that governed apprentice contracts; 74.4 per cent had no knowledge of them at all.16 The existing structure of supply was, again, an important factor. Achecyd, for example, had no apprentice courses on its curriculum of training activities. The head of the CPC’s Education Department (Comisión de Educación), Carlos Schlessinger, conceded that the seventy technical schools under the administration of the CPC lacked both teaching expertise and the standard of equipment that would justify linking the courses offered there directly to work experience in actual firms. It was also thought that the tradable value of the skills gained there was weak, since workers usually had to undertake further training after finding employment.17 Furthermore, it was believed that the state subsidies that financed these establishments were insufficient, particularly since no financial commitments were forthcoming from the private sector towards strengthening the schools. But, again, the structure of supply interacted with that of demand. Indeed, it was thought that individual firms lacked the technical staff, as well as the willingness, to ‘open up’ to a dual education programme. 18 Hence, the CPC did not support the initiation of dual education schemes. This was consistent with the principle against concerted action, which might intervene in the internal affairs of individual firms. Political constraints on training policy The Chilean labour movement, through the CUT, had from an early stage in the transition advocated tripartite involvement in the training market (see Chapter 4). However, the entrepreneurial associations had remained against such reforms. They saw the demand-based

128 Social Citizenship and Chile’s Labour Reforms

system of training as one of the greatest successes and a pillar of the market reforms. CPC leaders emphasized that investment in human capital should emerge exclusively through ‘private initiative’ and in the context of further deregulation.19 The Concertación, on coming to power, established a consultative body under the Labour Ministry, but it had no policy-making powers. A significant force against reform was the weakness of the state apparatus, not to mention, of course, the enormity of the presence that the private sector itself had achieved. Chile’s SENCE, the public supervisory body, according to CINTERFOR, ‘never developed a training policy’, and had, in 1990, ‘the continent’s lowest level of planning’. In 1990, Chile was ostensibly the only Latin-American country in which the authorities were unable to carry out an adequate labour market survey, primarily because of a lack of funds. Furthermore, SENCE had to rely on newspapers and the municipal ‘oficinas de colocación’ to run the faltering grant scheme. The most important reason was a lack of co-operation from the private sector on information exchange.20 In short there was a number of indications that the public authorities had little choice after 1990 but to adapt their programmes to the PTC system.

The training policies of the 1990s The ambitious aims of the Aylwin Cabinet in the area of training were exemplified by the creation of a new administrative entity to run the ministerial team’s own initiative, the Programa de Jovenes, or Youth Training Scheme (henceforth YTS). This involved an initial increase in spending, in 1991, of 74 per cent (down to 19 per cent in 1992). In practice, however, the ministerial approach sanctioned the trading of roles that had been established over the previous fifteen years, when the public sector turned from policy planning to playing a key role as financier of the PTC system. The tax rebate scheme Of all the training initiatives, the tax rebate scheme (see Chapter 2) remained the most important. According to this scheme, any firm could claim back 1 per cent of its annual wage bill to carry out training. In 1980, this programme concentrated 66 per cent of SENCE resources, with the other 34 per cent going to a grant scheme which targeted workers directly (see Table 7.1). By 1985, the firm-based tax rebate scheme commanded 91 per cent of resources. Despite this, usage of this option was slack. By 1992, only 54 per cent of existing firms were included in SENCE’s registers, down from 58 per cent in 1988, and 70 per cent in 1985.21 The number of

Training Policy and Decentralization

129

Table 7.1 Percentage distribution of participants and spending on different SENCE programmes, selected years (percentages) Years

1977 1980 1985 1990 1991 1993 1996

Tax deduction programme

Apprentice scheme

Grant programme

Youth Training Scheme

1

2

1

2

1

2

1

2

41 65 86 97 91 91.1 93.7

40 65 91 98 73.8 65.4 77.3

– – – 0.2 0.3 0.1 0.1

– – – 0.0 0.3 0.3 0.3

59 35 14 2.4 2.2 1.3 1.4*

60 35 9 0.0 3.6 3.8 2.3**

– – – – 6.5 7.5 3.8

– – – – 22.3 30.5 18.8

Total

100.0 100.0 100.0 100.0 100.0 100.0 100.0

Notes: 1 = Participants; 2 = Spending; * The figure would have been 2.4% if we had included the 1995 programme aiming at training managers of SM firms; and ** 3.6% in the following box. Source: Elaborated on the basis of SENCE data, Labour Ministry, Chile; Bulletin of Statistics No. 4, 1992.

firms that activated the tax rebate option was even smaller, averaging one in four. 22 But the key question is, how far did the tax rebate scheme remedy some of the general failures of the PTC system noted above? The bias towards larger firms and management ranks The tax break scheme continued to be heavily weighted in favour of larger firms. In 1991, for example, only 4 per cent of SENCE’s resources was spent in establishments with fewer than 50 workers, despite the fact that 63 per cent of Chile’s workforce was concentrated within this group. Table 7.2 lists the occupational breakdown of SENCE participants through the tax rebate scheme. It shows that less than two-fifths of SENCE participants came from the ordinary workforce in the early 1990s, down from nearly 50 per cent ten years before. The remaining 62–63 per cent pertained to management – executives, administrators and supervisors, of which the latter represented the smallest group. Occupational distribution barely changed under the Aylwin and ensuing Frei Governments, and when it did it was in favour of management training. The figures for professional targeting were almost identical to those that characterized the PTC system overall.23 But even the occupational breakdown does not present us with a clear enough picture of where the real weight of SENCE’s resources tended to fall. If we use course content as our indicator, we find that around 94 per cent of courses can be classified under subjects that have to do with

130 Social Citizenship and Chile’s Labour Reforms Table 7.2 Percentage distribution of participation through the TBS, by occupation, selected years Year

Management

Workforce

Executives & Adminis- Supervisors Skilled Semi-skilled Unskilled Total Professionals trators workers workers workers 1977 Totals 1985 Totals 1990 Totals 1992 Totals 1994 Totals 1996 Totals

14 21 21 20 19 18

18 52 25 63 26 61 28 62 31 63 34 64

20

27

17

28

14

30

14

30

13

30

12

29

16 48 6 37 7 39 6 38 5 37 5 36

5 3 2 2 2 2

100 100 100 100 100 100 100 100 100 100 100 100

Source: Before 1992, handwritten sheets from SENCE archives; after 1992, Boletín Estadístico 1994–1996.

management (including computing and English), services or sales (see Table 7.3). In 1992, only 2 per cent of courses sponsored by SENCE through the tax rebate scheme were centred on activities that related to industry. This increased to 3.6 per cent in 1997, when including the food industry, but falling again to 1.7 per cent in 1999. The 1997 reform A reform of SENCE of 1997 introduced a provision for firms to introduce bipartite committees to discuss the assignation of training resources. However, the remit of the new committees was weak. The committees were advisory only and were not effectively supervised by the state. As with the Health and Safety Committees of the 1980s (comités paritorios), which in our survey were found to exist in only a quarter of firms, the bipartite committees rarely functioned. In 1999 they were found to exist in only one out of 10 firms. Moreover their effect seemed at best neutral. Only 9.6 per cent of training through the TBS was carried out in firms with such committees. In summary, the tax rebate scheme presents us with a concrete example of how public policy came to follow the private sector trend. This need not have occurred. It seemed possible to introduce relatively

Training Policy and Decentralization Table 7.3 Area

Percentage distribution of participants by area, 1988–96 1988

1990

1992 1994* 1996* 1997**

Administration 38.4 38.2 40.2 31.6 36.9 (Administration of (20.8) (20.8) (22.9) – – Personnel and Marketing) (Administration of – – – – – Production) Services and Commerce 8.2 5.3 6.3 11.2 15.3 (Personal services) (3.3) (5.4) (4.3) – – (Financial) – – – – – (Health) – – – – – Basic and applied 43.3 44.3 43.2 16.8 18.1 sciences (Computing) (12.4) (14.2) (18.7) (15.0) (16.5) (English) (14.9) (8.5) (5.2) (4.1) (5.2) (Mechanics – general, – – – – – auto, diesel) (Welding) – – – – – Industry 3.0 2.5 2.2 2.4 2.5 (Automobile industry) – – – – – (Food industry) – – – – – (Textiles) – – – – – Other 15.3 15 14.4 18.9 5.5 Total

131

100.0

100.0 100.0 100.0

1999

37.2 (25.8)

30.7 –

(1.9)

–

15.3 20.2 (4.1) (11.6) (2.3) (6.8) (1.3) (1.8) 39.4 8.3 (16.8) (2.8) (1.8) (0.7) 3.6 (0.4) (1.8) (0.2) –

14.8 6.7 1.7 – 1.7 – – – 15.9**

100.0 100.0

Notes: * The greater detail up to 1992 reflects the handwritten (and more detailed) presentation of statistics at that time. ** The only large post was 4.9% for education and training. Source: Before 1992, handwritten sheets from SENCE archives; after 1992, Boletín Estadístico 1994–1996, and Liquidaciones de Capacitación Visadas (2001).

modest changes to the tax rebate scheme in order to correct some of the flaws noted above. For example, earmarking of a certain percentage of resources could have targeted SM firms, and/or specific occupations and/or sectors in connection with regional policy. In Mexico, similar schemes at this time favoured small and medium-sized firms, 24 a feature also found to be successful in Singapore. 25 At the very least, the resources could have been aimed at the training of workers. 26 In fact, the non-selective bias of the tax rebate scheme (that is, the universal extension of a 1 per cent pay-back from taxes) had serious flaws. Indeed, for most small firms, which were often squeezed financially, the sum was too little to initiate training at all. For large firms, who had the necessary capital, if not perhaps the will, it was of little significance. However, policy-makers followed almost literally the liberal ideal that any intervention from the state had to be universal to all, or ‘non-selective’.

132 Social Citizenship and Chile’s Labour Reforms Table 7.4 1992

Training programmes carried out by, and in association with, SENCE,

Name of programme

1 2 3 4 5 6 7 8

Number of participants

Tax rebate Grant scheme Apprentices*** YTS (Youth) FCJ (Marginal youth)* Coal reconversion Womens’ training** Dual education

Percentage participating

259 783 2 133 655 21 460 2 400 369 734 340

90.24 0.74 0.23 7.46 0.83 0.13 0.25 0.12

287 874

100.00

Total

Cost (millions of 1992 pesos) 11 866 216 59 5 111

Notes: * This programme (through FOSIS) covered approximately 12 000 young people. ** In 1992–3, 1467 women completed training courses through SERNAM. *** The Apprenticeship Programme offered up to 60% reimbursement of salaries (maximum being the minimum salary per month for the apprentice to a maximum of one year). Source: SENCE, Boletín Estadítico Numero 4, Enero-Diciembre, 1992.

A decision had been made on principle, not to target SM firms. 27 And the Director of SENCE, despite his overall commitment to industrialization (see above), dismissed any idea of future plans for policies that would involve specificity by sector, occupation or industry.28 Other training programmes A number of new training programmes in the 1990s subsidized individuals directly, as opposed to firms, and hence had the potential to strengthen workers’ property rights, by giving them, at least in an immediate sense, control over acquired skills. In fact, and despite their undoubted immediate relevance, the programmes remained extremely small, and isolated not only from each other and other poverty programmes, but also from any mainstream system of education and training. Moreover, they lacked credibility in the business sector, which was very reluctant to participate in the employment relocation offices that existed at municipal level. Among the alternative programmes, the YTS was undoubtedly of the greatest importance. The Youth Training Scheme The YTS had a very high propaganda value for the new government. It aimed to cover no fewer than 100 000 young people in three years, and

Training Policy and Decentralization

133

Table 7.5 Typology of course content, third bid (licitación) of the Youth Training Scheme (YTS), 1992 Course type

Percentages

Sample of specific courses

1

24.5

Metal workers (7%); textiles (5%); electricity (5%); computing (4%); industrial processing of food exports (0.4%); other (3%)

2(a)

21.7

Basic skills within ceramics, electricity, shoe-making, upholstery (5.9%); maintenance and assistance (3.4%); machine operation (2.5%); basic secretarial (2.1%); food preparation (1.3%); bakery assistant (0.9%); basic mining and agriculture (1%); other (4.6%)

34.3

2(b)

12.6

Basic cultivation of fish and other food products (4.2%); basic food processing (3.8%); fruit sorting and packing (2.1%); forestry (1.3%); other (1.2%)

3

41.2

Office helper (6.1%); security services (5.7%); basic sales technique (3.6%); basic food presentation (3.6%); bar and hotel services (2.5%); telephone operators and supermarket cashiers (2.3%); basic personal services (1.9%); basic services and various (15.5%)

Notes: 1 Areas approximate to a typical technical education content. 2(a) Areas requiring some development of skill and/or (b) related to dynamic export sectors. 3 Areas involving very low skill development and/or unrelated to dynamic export sectors. Source: Elaborated on the basis of unpublished SENCE statistics, Labour Ministry, Chile.

it was costly. Of the new money that was injected into SENCE in 1990, 67 per cent went into this programme. The high unit costs of the new programme made it reasonable to expect that the quality of education would be high.29 Positive evaluations of the YTS tend to focus on three features of the programme. These are its interactive linkage with the private sector, its targeting of youth, and its usage of the dual education model in the design of the training scheme. The dual education model implies that theoretical training interacts with practice in an actual firm.30 But was the link to the private sector as smooth as suggested?

134 Social Citizenship and Chile’s Labour Reforms

Low skill content Table 7.5 presents us with a typical breakdown of course contents within the YTS. The breakdown is based on two criteria. These are the level of technical skill content, and associations with Chile’s dynamic export sectors.31 The first box in the table contains courses that qualify at least under the first criteria. It shows that only 25 per cent of courses within a typical YTS auction reflected activities that are normally associated with the technical content of a dual education scheme. A further 21.7 per cent featured activities involving a sufficient level of skill to warrant a 5-month training programme, but about which it would be difficult to say that they exhibited strong links with productive change. The remaining 53.8 per cent featured activities with a very low skill content. A mere 12.6 per cent of these might be associated with dynamic export sectors such as fish and food processing, fruit packing and forestry. Typical courses within the remaining 41.2 per cent included private security, bar-tending and cashiers’ services. In short, only 25 per cent of courses appeared to offer a realistic opening to an occupational stream. The courses were in any case too short to compare with basic technical education or a standard apprentice scheme. We could also assess the YTS merely on the question of how far it had any relationship to the quest for productive transformation within the new export areas. For this purpose, Table 7.6 is focused on the nature of the activities learned in the new export areas (14.6 per cent of the total). It distinguishes between functions associated with higher technical competence and higher aggregate input activities and others that are connected with basic and precarious work activity. Here we find that a mere 0.8 per cent of activities were associated with a high skill content, while 3.6 per cent represented tasks in the middle range, including the processing, smoking and preservation of fruit and fish products. Finally, the remaining 10.2 per cent accounted for the simplest and, from a social perspective, the most vulnerable, chores, such as packing and forestry. The final alternative way to evaluate the YTS is to look simply at the evidence of existing demands from firms. Table 7.7 compares the areas of demand for skilled or semi-skilled workers by region which were detected in the SERCOTEC survey noted above, within the YTS courses undertaken (as part of the YTS third auction) in the same regions. The table shows that a great number of the regions cited featured no correlation at all. The overall average was 6.6 per cent. 32 Public policy and the YTS In summary, the involvement of the private sector within the YTS had the potential to ensure competition, maximum efficiency and

Training Policy and Decentralization

135

Table 7.6 Representation of dynamic export sectors* in the Youth Training Scheme (YTS), 1992 Type of course

Percentages

Sample of specific courses

1

0.8

Quality control; operation and management of irrigation equipment; crop management; management of chemical products in agriculture

2

3.6

Food industry processing and preserving; fish-meal production; smoking, processing of salmon

3

10.2

Cultivation, plantation assistant, handling of fruit-trees, packing, forestery; forest plantation, packing, and other

4

14.6

Total

Notes: * New dynamic export sectors are defined here as forestry, fishing, fruit and other foodstuffs; 1 Advanced skills in production or management; 2 Middle-level skill; 3 Simple skills; 4 Percentage total on courses (527 – with 11 017 participants) in the third bid of the YTS. Source: Elaborated on the basis of SENCE statistics, Labour Ministry, Chile.

accountability. However, if SENCE did not set any criteria as to the technical content, occupational focus or sectoral linkage of the courses on offer, it was hardly encouraging the PTCs to upgrade their infrastructure and to develop their links to firms. Even among the more sophisticated PTCs, such as INACAP, there was a tendency towards seemingly casual skills. Nineteen of INACAP’s thirty-five courses on the YTS’s Third Auction were centred in the third group of Table 7.5, including cookery, floor services in hotel, cake decoration, and ‘sales technique’.33 Furthermore, if SENCE’s demand for bidders outweighed the supply, then SENCE was not promoting competition in any significant way. The director of SENCE confessed that at least 20 per cent of available resources remained unused after each SENCE auction.34 This could be an indication that the PTCs found the participation in the YTS less lucrative than other activities, or it could be that there was not the sufficient infrastructure available in Chile to cater for worker training on the YTS scale. The link to firms, through what resembled an apprenticeship period – and which seemed to justify referring to it as a ‘dual’ scheme, like that in Germany, was in fact one of the scheme’s main flaws. None of the twenty-four employers interviewed in a ministerial survey five years after the programme’s start had had any input into, or indeed had any

136 Social Citizenship and Chile’s Labour Reforms Table 7.7 Percentage of courses in the YTS programme compatible with demand from small and medium-sized enterprises, by region Region

1 2 3 4 5 6 7 8 9 10 11 12 13† Average

Areas in demand*

Percentage coincidence in YTS**

Carpentry, textiles, metals Carpentry, metals Metal (welding) Food conservation (oysters), confectionery, furniture making, welding, metals Confectionery, textiles Textiles, wood preparation, welding Furniture making, welding Designers (clothes/furniture), furniture-making, welding, textiles, food conservation, printing Furniture making, whetting of shutters and similar Making of shutters, welding, lacquering, varnishing, machine operation Pottery, furniture making Furniture making, welding Confectionery, textiles, shoe-making, furniture making

12.2 0.0 0.0 6.5 7.4 8.3 0.0 22.7 4.0 0.0 0.0 0.0 25.0 6.6

Sources: * SERCOTEC survey, complemented by local government information; ** Course list by area of third bid of the YTS (involving 545 courses and over 11 000 participants; † Metropolitan region.

opinion, on the educational content of the practice period. Seventy three per cent were at a loss as to its technical purpose. Some 93 per cent were not even aware of the standard module that was set up to accompany their assignment. Only 20 per cent had had any information about supervision of the programme by SENCE itself.35 This signals the failure to use the YTS as an instrument for introducing a set of widely applicable and universal apprenticeship standards, although the opportunity to do that was one of the few real advantages of the exercise, as a result of the broad inclusiveness of the scheme. As it was, the extensive coverage of the scheme remained a serious drawback, because the resources had to be so thinly spread. Moreover, individual PTCs were isolated from one another within the atomized system, and had difficulty in identifying the needs of firms for trained personnel. This may help to explain the finding in one SENCE survey that the PTCs were repeatedly over-producing particular courses in relation to regional need. 36 In order to give an incentive to infrastructural development, SENCE would have had, at

Training Policy and Decentralization

137

the very least, to offer longer-term contracts, and to establish a broaderbased system for the certification of skills. The aprendizaje alternado programme This fact, along with the absence of a functioning network of long-term contracting in the business sector itself, may help to explain how apprenticeships through public subsidies had failed to develop. The relevant programme (Aprendizaje Alternado (AA) begun in 1988, and since modified a number of times) still counted for no more than 0.1 per cent of SENCE participants in 1999 (see Table 7.1). Schemes of this nature had a higher unit cost than the YTS, and for them to be sustainable as well as inclusive would have required a greater input from the private sector, as well as longer contracts than the preponderant two-month contracts offered by SENCE. The job-scoring rate of participants on the apprenticeship scheme was, in fact, remarkably good. Seventy five per cent gained not simply jobs, but an actual work contract, compared with 56 per cent in the control group (and with rates of 67 per cent in the programme group and 61 per cent in the control group, respectively, on the YTS). Despite the YTS being designed specifically for marginal youth, there was greater participation of young people from the three lowest income brackets in AA – 34 per cent compared with 25 per cent on the YTS. The formal value of the work contracts subsequently gained on the AA was also better. Ninety one per cent of participants on the apprenticeship scheme had a form of social security affiliation, compared with 70 per cent on the YTS. Nevertheless, it would indeed be odd if the AA programme escaped the underlying institutional weaknesses so manifest in the larger youth training scheme. The weakness of institutional links in the former is seen in the fact that as many as 42 per cent of participants found an apprenticeship in a firm insufficiently related to the theoretical part of their course. Both the YTS and the AA compare unfavourably with public (KOMA) training schemes in Korea, where 90 per cent found jobs after training. To this should be added the greater value of training certificates there, because of the publicly regulated and generalized character of the training provided. 38 SENCE, together with the Education Ministry, had taken some initiative to remedy faults in the Chilean system. For example, in 1994 it started a programme to educate workers appointed by companies as a form of foreman (maestro-guía) to take charge of apprenticeship in the firm. On the whole, however, such efforts have failed to characterize training provision on a significant scale. The AA represented only 132 of a total of 6576 courses for youth alone, financed by SENCE in the years 1994–5.

138 Social Citizenship and Chile’s Labour Reforms

Finally, the YTS, like the FCJ (targeting socially marginal youth; see Table 7.1), were not designed to qualify the participants for other mainstream educational options. This could be a source of frustration for the participants, who would not find the certificate from courses designed for young people with social-psychological problems easy to trade. Nevertheless, gaining a regular occupation was expressly what a majority of the participants had wanted. Fifty nine per cent of them had entered the programme with the wish to ‘learn a trade’. The next-largest category, ‘to have better possibilities of getting waged work’ received only 28.6 per cent of responses. The failure to target youth unemployment was borne out by the steady dilution of youth programmes, down to a combined total of 1.4 of SENCE participants by 1999. The Labour Ministry had described the objective of the Youth Programme in terms of socialization, as a ‘chance to become familiar with the discipline of the work place . . . The Programme is not based on the presumption that the young people will actually find work.’39 Following our earlier discussion, this seems to be a truthful and fair description of the Youth Programme. However, it also reveals that the programme, like the other training schemes that have been mentioned, could hardly be described as an exercise in skill formation, in institution-building or in the advancement of workers’ property rights.

The politics of an unemployment insurance regime An unemployment benefit system for Chile was finally put through parliament in April 2001 after a decade of attempts by Concertación governments and the CUT to promote such a scheme. The length and difficulty of this process illustrates the extent of employer resistance to institutions that would connect the training system to efforts to make labour mobility more empowering for workers and to establish tradable skills. The progressive dilution of the project was informed by employers’ cost-cutting concerns and their insistence that the administration of the system would be strictly separate and privately run. The political process around the reforms clearly demonstrates the path dependence initiated by the limited nature of framework agreements for labour at the point of the new democracy’s founding. Outlines of an unemployment insurance regime (UIR) began to be drafted by both employer organizations and labour in the late 1980s. However a government proposal was not put to parliament until 1993. This project (later reformed into PROTRAC) was delayed in the Senate for over three years and dropped in early 2000. After this a yet more

Training Policy and Decentralization Table 7.8

139

Comparison of the proposals for an Unemployment Insurance Regime

Type scheme

Years of service at the time of dismissal 1

2

3

5

7

9

11

Rate of unemployment benefit received at the time of dismissal* 1 2 3(i) 3(ii) 4 (a) (b) (c) (d) 5****

1 1.8 0.82 0.5 2.8 1.0 0.0 0.2 1.6 0.6

2 2.8 1.64 1 3.8 2.0 0.0 0.8 1.1 1.4

3 3.8 2.45 1.5 4.8 3.0 0.0 1.3 0.5 2.2

5 5.8 4.09 2.5 5.9 3.0 1.0 1.4 0.4 3.7

7 7.8 5.73 3.5 7.0 3.0 2.2 1.6 0.2 4.6

9 9.8 7.36 4.5 8.4 3.0 3.5 1.8 0 5.7

11 11.8 9.00 5.5 10.1 3.0 5.0 2.1 0 6.8

Notes: 1 Existing indemnity scheme. 2 Ditto, plus existing unemployment benefit. 3(i)** 1997 proposed scheme. If dismissed by ‘necessities of firm’ clause. Employer contributions only. (Total would be like Item 1); (ii) 1997 scheme. Dismissal under other articles. Employer contributions only. 4 1993 proposed scheme, total according to Labour Ministry calculations. Components of unemployment benefit: (a) Indemnity for years of service (up to three years maximum). One wage per year. (b) Any-event sub-account (financed by employer over 3–10 years). (c) Firing sub-account (financed by employer for first three years and by worker over 3–10 years). (d) Credit (co-financed by workers in first three years and state contributions). 5 1993 proposed scheme calculated on the basis of employer and state contributions only***, total including: (i) Deduction of credit – to be repaid by worker to fund upon finding new employment; (ii) Deduction of worker contribution (first three years = 50%; last 7 years = 32.7%). * Measured in number of monthly wages. ** This scheme operated as an indemnity of dismissal, financed exclusively by the employer. *** Low-income workers only. **** Excluding contributions made by workers (as under the old scheme). Source: Labour Ministry, Chile.

diluted version was revived and passed a year later. Below we review the different proposals. Evaluating the Chilean schemes The 1993 proposal The first government project of mid-1993 was drafted to accommodate employer preferences for an individual account insurance system operated

140 Social Citizenship and Chile’s Labour Reforms

by existing institutions in the private sector, like the Pension Funds (AFP). It entitled its recipients to three months’ worth of indemnities. Moreover, it contained a new ‘annual account’, a dismissal sub-account that would offer immediate contributions and benefits, and a credit option for low-income workers. The scheme would be financed basically on a tripartite basis.40 The benefits that the system would offer were very limited. For example, the enlargement of the workers’ freedom that the proposed scheme would provide was substantially modest. As can be seen from Table 7.8, row 4(b), protection when leaving a job only operated after the fourth year of employment, and remained very small compared to the benefit received when dismissed under the existing regime of indemnity payments.41 Hence, despite the fall in employer contributions that the new scheme implied, from 100 per cent to 58 per cent, the former retained discretion in the ‘movement of people’. Still, the 1993 proposed scheme did contemplate more substantial earlier benefits compared with the later formula. It was abandoned in part because a loan scheme was opposed by unions, and greater co-ordination with a fiscal alternative, as well as too wide changes to the operation of the AFPs, opposed by business groups. 42 This illustrates one of the most paradoxical aspects of the Chilean situation. On the one hand, the very strength and transparency of the social insurance institutions could make the future inclusion of some excluded sectors possible, such as workers on fixed-term contracts and domestic workers. On the other hand, the resistance to information-sharing and co-ordination with representative bodies in the public or social sectors would be likely to impede such a move. This shows again that it was not an absence of institutional capital (training institutions), which lay behind the poverty of investment coordination in the area of human resources.43 Rather the main obstacle was again a collective action problem, in this case the absence of institutional links. The 1997 proposal Because of the many obstacles encountered with the 1993 proposal, a new scheme from 1997, the so-called PROTRAC, was made simpler, but this time objections (mainly from employers) centred on the short-term increase in costs. According to the 1997 proposal, compulsory membership would consist of contributions towards individual accounts of 0.8 per cent and 3.6 per cent, respectively, of an individual’s salary by the employee and employer per month for eleven years. The plan was that benefits equivalent to around 50 per cent of the monthly salary could be paid out for one month per year of service (up to a maximum of five months). This arrangement would come into effect after one year and

Training Policy and Decentralization

141

regardless of the reason for leaving the firm. If the employee was dismissed under the articles claiming ‘necessities of the firm’, he or she would be entitled to the full indemnity operating under the existing indemnity system. In this case the difference would be paid by the employer, while the state would bail out individuals through a minimal income assistance programme based on a descending percentage of the minimum wage.44 There were many financial improvements in the new scheme. Table 7.8 reveals that none of the government proposed schemes, either the 1993 or the 1997 proposals, would represent a net increase in costs for employers. In fact, clearly the financial structure of the new UIR would lower transaction costs. The new system would help to spread costs, which under the indemnity system had to be paid in full at the time of dismissal. 45 This should make it attractive to firms, and therefore compatible with an overall trend of formalization of work and a further inclusiveness of the scheme. Flexibility would be further increased through the lowering of the costs of dismissal under the ‘necessities of the firm’ clause. In respect to the latter, the UIR represented an improvement in important ways, by levelling out the means of effective movement, at least in principle, to cover workers voluntarily leaving employment, or who were sacked with clauses imputing fault.46 These positive points notwithstanding, it is clear that the extent of the flexibility gain to workers would depend on how far the system’s coverage and links to other institutions, would maintain occupational continuity overall. With respect to coverage even for workers in the best possible category, namely those in stable and formal employment, it would seem that the new system was not a significant improvement on the old. It would still take five years of employment to earn the coverage of five months’ unemployment, which is deemed to be an adequate term. The assumption of a five-year average job span, with a five-month relocation period, was based on a scenario of continued employment growth, and would remain inadequate for those who left their employment after less than five years. It is hard to see, based on these calculations, how the scheme would be able to cope with an unemployment crisis of the magnitude that hit Chile in the wake of the Asian crisis of 1997. At this time unemployment peaked at close to 12 per cent and fluctuated between 9 per cent and 11 per cent for over three years. The legislative proposal did envisage a preferential access for adherents to the existing grant scheme, but we should remember how slight its resources were, and how weak the certification of skills in Chile remained. To this should be added the tenuous nature of the links between municipal labour relocation efforts and firms. It would remain uncertain how

142 Social Citizenship and Chile’s Labour Reforms

far the hundreds of thousands of users of the new UIR could be accommodated within SENCE, when the institution still budgeted over 90 per cent of its total funds to the tax rebate scheme. In short, whereas the UIR created a basic institutional framework for administering the property rights of labour, the training system continued to situate the property rights over skills in firms. This mismatch limited the extent of the institutional linkage that the system would be able to foster. The employment insurance of 2001 Similar problems characterized the scheme that was finally legislated in 2001 after the new Lagos Concertatíon Government, installed in early 2000, proposed a replacement for PROTRAC. In this scheme earlier access to funds was possible but the period of potential coverage (of five months) remained inadequate in relation to the average period of unemployment and, more importantly, the funds for insurance payments had become reduced yet again. Employers’ contributions were cut this time from the 3.6 per cent of wages envisaged in PROTRAC to a mere 1.6 per cent (workers’ contributions being cut from 0.8 per cent to 0.6 per cent). This lowering of employers’ contributions was what had made the passage of the project possible in the end. Moreover, the separate and exclusively private administration of the insurance scheme would continue to mitigate against future adjustments to the scheme. Such adjustments might include the establishment of a better functional link to institutions of labour. It needs to be noted that the AFPs already had an unfavourable record of non-payment (by employers) of contributions from workers. The new legislative proposal, while it raised workers’ entitlement to take action against such offences, did not specify the role of the state or specific agencies in speeding up possible claims.47 Rectifying such basic regulatory flaws represented a basic step to improving institutional links. In the end the provision of state finance for low-income workers was just the lowest common denominator in terms of what could be agreed between labour and business sectors. It represented an increase in the previous benefit in the short term (extended only to five months). However it hardly compensated for the abandonment of earlier plans (implicit in the 1993 proposals) to finance a common fund with resources for training purposes. Nor did it address the CUT’s earlier proposal to make sure that job relocation and training were given a regional and/or sectoral focus. Such plans had initially led the CUT to oppose handing over the running of the UBR to the AFPs.48 Indeed indications from entrepreneurial leaders from an early stage were that a strong and vocal political resistance would be mounted to

Training Policy and Decentralization

143

any joint administration, and to any increase in costs to either employers or to the state.49 Given the government’s general emphasis on consensus politics, it is inconceivable that it would have wanted to promote a system that faced strong opposition from business groups. In summary the discussions during 2000 and 2001 suggested how far, over the years, successive Concertación governments had been forced to climb down visà-vis the business sector in their proposals for unemployment insurance. The key discussions continued to be about the level of the contributions to be paid by employers. Whereas PROTRAC had contemplated a total of 4.4 per cent, the new proposals centred around a reduction to somewhere between 3.5 per cent and 4 per cent, of which now employees would have to pay two-thirds, and not one-third, as first envisaged. In the end, employers’ contributions were cut by more than half. The idea of handing the administration of the insurance to a private firm was by then taken for granted, and the CUT’s earlier strong resistance to this idea was muted. Taken as a whole, the political process of unemployment insurance reform of the 1990s demonstrates how a rather narrow concern over levels of contributions could be made to override potentially far more significant improvements in the system of labour mobility, a system in which the cornerstone could have been unemployment insurance. The system of registration of wage earners that such a scheme entails makes it possible to link a variety of services through it. The discussions of 2000 and 2001, however no longer included how the system could be linked to training, or to other forms of social assistance and employment creation. In all, the decision to lease out administration did not bode well for linking the insurance to active welfare schemes in the future. Employment insurance in Korea Korea presents an interesting contrast. This country, where the unemployment crisis of the late 1990s hit particularly hard, was better prepared institutionally. It was not simply that an employment insurance scheme (EIS) had been introduced in 1995, but that the structure of the scheme was designed to provide an integral response to employment issues. Its focus was different from the outset: ‘While “unemployment insurance” is an ex-post facto and a remedial measure that pays unemployment benefits to jobless persons, “employment insurance” is a preventive and positive measure to prevent unemployment in advance through employment stabilization and the development of workers’ ability.’50 Contribution levels in Korea are similar to those in Chile, but because the state in Korea assumes all administration costs, the actual size of available

144 Social Citizenship and Chile’s Labour Reforms

funds is greater. In Korea’s employment insurance, employers pay 1.5 percent, and workers 0.5 per cent. Employers pay 0.3 per cent towards a scheme designed to prevent job loss (employment stabilization). This provides subsidies for specific measures by firms to keep workers employed (through job relocation, job-share schemes, temporary leave and so on). They then pay 0.7 per cent (smaller firms pay as little as 0.1 per cent) of salaries towards a joint fund for vocational training, for which, again, individual firms can apply. Finally, they pay 0.5 per cent (as do the workers themselves) towards unemployment insurance for workers. The state supports the system by assuming all administration costs and actively administering the different schemes. The regional employment stability centres, for example, administer the unemployment insurance, whereas employment promotion subsidies are centrally run. Second, the state infuses the system with money at times of extra demand for EIS resources. For example, the Labour Ministry provided roughly a third of the resources spent on training via this scheme in 1998 (5.3 trillion won).51 It is still early days to judge the Korean system, but evaluating it purely from an administrative point of view, it seems that the choice to leave administration in state hands made it feasible to use ‘the best possible level’ (local or central) to organize different tasks. It was clearly possible to co-ordinate efforts to cope with different aspects of the employment crisis. In fact, it is very doubtful that a scheme not run by the state would have been able to muster that level of co-ordination, as well as the trialand-error approach to help overcome the crisis that authorities in Korea employed. 52 State administration in Korea, and indeed the flexible approach to the state’s role in the economy, also allowed the authorities there to use a differential rate scheme to assist smaller firms. The scheme operated during 1999 with up to five different levels of contributions depending on size. This stands in contrast to Chile, where the concerns of PYMEs continued to come up against the doctrine of market neutrality. In this case the notion of market neutrality restrained differential treatment of firms (such as by size), as was seen with the tax rebate scheme. In contrast, employers in Korea were to some extent compelled to co-operate with the state. In response to the crisis affecting the country, they agreed to increase their share of contributions, it seems without a major politial fuss. With effect from January 1999, the employers’ contributions alone were raised as part of an agreement with the private sector to help combat the problem of unemployment. Employers, then, were to finance 2.1 per cent, and workers (as before) 0.5 per cent. In summary,

Training Policy and Decentralization

145

the lowering of unemployment in Korea was far more speedy compared with the Chilean case (where the economic crisis itself was on a much smaller scale) which suggests that public administration of social interventions (in the case of Korea) had not lowered the labour market’s capacity to adjust. Moreover, employment insurance in Korea had been launched on a sound organizational basis, although still inadequate and in its infancy. In contrast with its counterpart in Chile, it was not bogged down in its development by simple cost-cutting concerns.

Conclusion There were many conditions that diminished the technical problems involved in applying a comprehensive training and employment policy in Chile. These included the existence of a basic computing system for employment at the municipal level, a large number of training institutions, and the linkage between these and the entrepreneurial groups. And yet there was also a pervasive legacy of ideological polarization – as well as institutional inertia present in Chile. This had led to a perceived dichotomy between the state as a structure separate from the productive sphere, and a private sector seen as merely an ensemble of unconnected firms. The establishment of the YTS was an indication that democracy had brought with it a greater determination to use public policy in the training market, and to focus training on workers. Nevertheless, policy-makers stated clearly that they had no intentions of establishing an occupational or a sectoral perspective within the YTS. In the case of the tax rebate scheme, it could be argued that its ‘universal and uniform’ bias had reproduced one of the most troublesome precepts of the Beveridge model (that all should benefit from the same share of state subsidy, regardless of income or situation), the very model that the neo-liberals had been so keen to suppress. It is ironic that this feature of uniformity in the classical welfare model should combine so smoothly, in the TBS for example, with the purist market argument against selectivity. The truth was, of course, that the TBS was selective in a way that ought to concern policy-makers. The selection however was located in the PTC system, and not in accountable institutions. It is also significant that the Aylwin Government chose to cut the grant scheme even further, from 5 per cent to 2.7 per cent by 1992. The grant scheme was the only programme that had involved a role for the public bureaucracy in gathering information about training. It was also the only scheme which had taken its point of departure as identifying the demands and needs of workers as well as

146 Social Citizenship and Chile’s Labour Reforms

firms and deprived communities and regions. It was the only available scheme, in short, that had the potential to perform social policy as well as industrial policy functions. The case of a UIR has been analyzed separately in this chapter because it represented a direct advance towards the development of occupational citizenship. It did so by departing from the position of the individual as he or she faced the situation of exclusion from the employment order. In fact, the case of the UIR also represented a bold advance by abandoning ideas of universal and uniform coverage. In that sense it could be described as an apparent attempt to be consistent with the liberal ideal of focusing on the individual. And yet in doing so it also exposed the inherent flaws in this ideal when it is understood in a rigid sense, that is, in its extreme form. In other words, it was seen to be admissible from the perspective of policy-makers and the entrepreneurial community to individualize the contributions and benefits of a UIR. Yet it was not acceptable to combine decentralization of this kind with the establishment of national institutions which could co-ordinate a broad array of services from which the individual would gain. Hence the proposals for a UIR that circulated during the Aylwin and Frei Governments could not exploit the true advantages of the individual-centred approach. In summary, it is beyond dispute that the labour market programmes of the 1990s contained a series of positive features, such as those we have cited. Nevertheless, there were also some rather negative aspects to the schemes. In the case of the YTS there was a decisive move away from the construction of actual qualifications; and in the case of the UIR, what appeared to be a permanent leasing of the system to the private sector. This begs the question: why not the private sector? Our answer has been, not the private sector alone, when this amounts to a critical loss of institutional links. Indeed, it is not only necessary when one evaluates the nature of a new institution to think about what the buried alternatives are, but also, and more importantly, to take stock of the possibilities for future change that the new regime can be expected to foster. With a wholesale leasing out, the power of public policy in Chile was permanently curtailed. In the light of this it is an open question whether the labour market programmes of the Aylwin government, like the other labour reforms, overcame the legacies of the immediate past.

8 Labour Relations and Investment in Workers

The national reforms represented the political opportunity implicit in Chile’s transition. At this level larger steps towards the building of citizenship could be taken. What we have yet to investigate is the nature of the scattered attempts towards crude forms of occupational citizenship inside Chilean firms. It is in the light of the challenges that persisted there that we can hope to comprehend, in a fuller sense, the implications of what was and was not reformed in Chile’s institutions of labour. We have observed that where occupational citizenship is low outside the firm (as it had remained in Chile), long-term association with the firm, and some control over work aspects, may become obvious elements of social security. By implication, they become the immediate focus of social struggles. Our assumption is that this very fact made the state of local labour relations, the existence of long-term versus short-term relationships with workers, a central development issue in the Chilean context. Our present analysis will assess the central role of social relations and of unions in shaping investments in workers. It will also look at the variegated nature of the factors that motivate workers’ leaders. In this way we will pose the relevance to development studies of labour relations. Labour relations should be conceptualized as a key factor in the development of human resources. Then, the importance of social relations to investments in workers (in long-term relationships and in training) is the general concern of this chapter. Following on from this Chapter 8 will then assess the democratizing role of the unions.1

Labour relations and human resources In Chapter 2 we saw how the neo-classical development economics (or neo-liberal theory) assumes that deregulation ‘unties’ firms from external constraints.2 Yet, in our discussion so far we have implied that certain 147

148 Social Citizenship and Chile’s Labour Reforms

patterns characterized Chilean labour relations across individual firms. At the crux of the problem lies the failure, in neo-liberal theory, to distinguish between the effects of different institutions. There is an explanation for negative economic outcomes when the economy is regulated (too much regulation), but there is not one for negative economic outcomes when the market has been made to work to a maximum. Hence the neo-liberal perspective restricts us to a small set of market indicators, at best, in explaining the management of human resources; for example, in accounting for how it was that many Chilean firms had very high levels of turnover, even after the crisis years of the early 1980s had passed. In the case of training, the assumption behind the market neutral perspective is that firms will invest in training if they ‘need’ it (see Chapter 6).3 What spurs this ‘spontaneous’ need is not considered to be a relevant question. By contrast, in the governance literature discussed in Chapter 2, the area of choice is perceived as a larger, greyer one, in which it matters considerably what the costs or gains would be from losing or maintaining a labourer. Whether there is a cost in terms of ‘lost trust’ might depend on how far the forms of flexibility that are used involve a negotiation with workers. Whether there is an economic cost might depend on legislation. Whether there is something to be gained by keeping a worker would vary in accordance with how much investment in training has previously been made, and so on. Internal and external labour markets In the case of the OECD, there are already several studies which suggest that deregulation does not improve labour market efficiency. It does not necessarily lower unemployment or create new skills.4 At the same time, it is clear that ‘bidding down wages’ can be made to work more easily in the context of low skills and low industrialization. 5 It is in this way that neo-liberal theory with respect to the labour market appears to be supported in these primitive contexts. Still, we endeavour to suppose, even for the Chilean case, that the context of deregulation itself (including the Labour Law and the absence of formalized skill markets) constituted a macro-institutional context that encouraged firms to adjust via the most expendable component, their human resources. It hardly matters how far the origins of short-term forms of labour management in Chile were mainly cyclical (an effect of the recession of the early 1980s) or structural (following economic liberalization and the removal of legal constraints on labour use). The two are in any case difficult to separate, historically and empirically, in Chile.6 Rather, the important observation for our purposes seems to be the following: if

Labour Relations and Investment in Workers 149

associations could still be found – a decade after the key economic and legal reforms and the recession – between the ‘economic’ aspects of labour management (for example, turnover and training), on the one hand, and workers’ representation on the other, then it may be possible to speak of a situation of ‘lock-in’. This ‘lock-in’ worked independently of market forces to some extent. 7 Lock-in in this case could refer to the persistent weakness within the individual firm of social forces that might act against a short-term management of labour. It might include the low formalization of work rules and the lack of worker representation. In Chile factors like these were sustained by external conditions like the labour legislation and the absence of formalized markets in skills. Then distinctions between firms in the atomized system could well be especially marked, because of the absence of trans-firm criteria other than permissive rules. As we have suggested such distinctions were likely to be not simply a reflection of ‘technological and budgetary constraints’, to use Solow’s phrase. 8 Other intervening variables, would also include the social history of the firm. Social history and macro-institutions Two sources in the governance literature in particular are important to our discussion. These can be divided into macro-institutional and behavioural approaches. The first links the nature of workers’ occupational status to investments in workers.9 The second source could be called the ‘social theory of the labour market school’. This highlights the role of values and motivation in shaping labour relations.10 Such ideas had challenged key behavioural assumptions within the marginal utility analysis behind neo-classical theory. For example, they questioned the assumption that managers are perfectly endowed to select the appropriate organization of human resources, presumably on the basis of reading market signals. They also challenged the idea of limited preferences of workers. 11 In the literature, however, the two approaches are delinked. It is not the same to focus generally on the impact of values and motivation in the formation of labour market demands, as does Solow, and then to stress the supremacy of macro-institutional factors in orientating such behaviour.12 How can we accommodate the institutional perspective with the more subtle type of social factor, including the subtle impact of work rules and of unions over time? How far does social history matter? How does social history interact with the macro-incentives? Seeking to answer this question is, in effect, to isolate the social element in institutional change, the element which remains as the main contention between sociological and economic approaches to institutional change in the macro

150 Social Citizenship and Chile’s Labour Reforms

literature discussed in Chapter 2.13 Figure 8.1 provides a general overview of macro- and firm-specific factors. It indicates how social factors in the firm influenced the management of human resources in the general context of deregulation (a.i) and (a.ii), extending our analysis in Chapter 4. In Chapter 9 we shall expand on the role of the local unions (b.ii) within this framework. We should note that the figure highlights the framework presumed to exist ‘prior’ to exchange. Hence it does not account for all factors that motivate labour relations, especially not for short-term factors. However the assumption is that the relationships listed constitute a filter, a set of possibilities for alternative responses, to ‘external’ or market influences. In the case of employers, (a.ii) constitutes the set of macro-institutional incentives as far as labour management is concerned. It includes the margin of what is ‘feasible’ in terms of the usage of hiring-and-firing flexibility (legislation on labour contracts and lay-offs), for example. The assumption made implicitly in Figure 8.1, is that (b.ii) may attenuate how far this margin is used. We should note that (a.ii) also includes national (or supra-firm) incentives for investment in worker training (production system), which we know were very low in the Chilean case. Again, the implicit assumption is that (b.ii) may intensify the degree to which training is nevertheless emphasized in the firm.14 In short, the figure supposes that even where atomization (and hence market incentives towards short-term management) is very strong (as in Chile) ‘softer’ local factors, may still affect the type (the time-horizon) of flexibility pursued by firms. In fact, the centrality of the firm in this system means that they are particularly likely to do so. We noted earlier that the influence of social factors on employers (b.ii) and (b.i) is in fact rarely mentioned in the literature on union effects. We are thinking, especially, of the influence of non-union representation (but forms of representation, which tended to be strengthened by union presence), and of qualitative indicators of union presence, to be explored in Chapter 9. In both these cases the term ‘institutional memory’ of the firm is applicable to describe the long-term formation of social work rules and the relationship between the establishment of such rules and the propensity to invest in workers.15 The question is whether managerial acts with respect to labour are ever socially neutral. If they are not, then there must be at least more than one possible response to markets by the firm. With respect to labour flexibility, it follows that the central issue is not whether wage and mobility flexibility, the two ‘conventional’ forms, will occur. What matters is the level at which they occur, the readiness with which they are preferred to other options, and the degree to which social trade-offs

Labour Relations and Investment in Workers 151

Workers’ preferences (the day-to-day role of local union leadership)

General sources of (a.i) influence (broadly shared preferences) Experiences (b.i) specific to the firm

struggle for democracy neo-liberal ideology and deregulation employer behaviour/ frameworks of security

Institutional and social principles for the management of human resources in the firm (more or less protective of human resources)

(a.ii) production system/legislation

Figure 8.1

(b.ii) history of unionism/social relations in the firm

Dynamics in local labour relations in Chilean industrial firms

affect their coincidence with long-term objectives. Then, our claim that social trade-offs can both ‘prevent’ and ‘shelter’, a priori, a flexibility style that is protective of human resources, can be best defended in the context where decisions on human resources have been atomized to the utmost. Essentially, it needs to be demonstrated that both employers and workers may still be motivated, in their ‘economic’ decisions, by social norms. Finally, we should note that since our universe is confined to unionized firms, we cannot control for union presence. In Chapter 9, for example, we proceed by distinguishing between weaker and stronger unions. Meanwhile, for the following discussion, the assumption will have to be made that the labour relations described represent a ‘best possible’ scenario, given that workers must have less authority in the firms where there was no union presence at all. The survey of union leaders The main source for our analysis of local labour relations is a survey of 245 local labour leaders carried out in autumn 1992. Extensive consultation with local leaders revealed that a very comprehensive picture could be drawn of the organisation of work. This even included issues such as fluctuations in the size of the labour force over the previous year, how many workers (excluding management) had received formal training, whether a formal system of remuneration was present, and so on. In accounting for the role of local union leaders as informants about the context of work, we cannot forget the observations that

152 Social Citizenship and Chile’s Labour Reforms

Angell first made about the peculiar positions of local leaders in Chile as full-time workers in their firms (with the exception of only a small number of sectoral and national figures).16 This situation had not changed in seventeen years of dictatorial rule. Unions in sectors in or ‘almost in’ industry (in the case of the incipient export industries) were chosen as the most relevant universe, given our focus on the development of human resources and the involvement of workers in deciding on flexibility options. The sample was obtained via random selection, as explained in Appendix A.17 It is reasonable to suppose that differences between sectors (for example, a textile firm and another firm producing dried fruit) would matter; for example, that certain technological arrangements make pressures for negotiation with workers more evident. However, it is far from obvious how far such factors were important in distinguishing firms in our sample, particularly as all the data presented was checked for sector effects.18 Sophisticated technology tended to characterize only certain elements of production, even in well-stocked firms.19 Furthermore, we found that it was quite feasible for employers to introduce continuous-process technology, and advanced computer-controlled machinery, while continuing arbitrary wage and employment routines.20 In fact, a certain amount of technological heterogeneity was to be preferred. If important associations could still be found across the firms, our hypothesis regarding the importance of power relations (and external factors such as weak legislation) would only be strengthened. We have included four case studies to add some qualitative and historical evidence to the statistical work.

Forms of labour flexibility and democratization in industrial firms Local union leaders in our survey demonstrated a widespread recognition of new management forms introduced during the 1980s. These included a greater range of tasks (54 per cent), the existence of systems to measure unit output (71 per cent), and incentive wages (61 per cent), of which about a-third (20 per cent) were pure piece-rates. On the other hand, 16 per cent of respondents reported that more than 15 per cent of the workforce had received training in the previous two years; 21 per cent cited figures between 10 per cent and 20 per cent, and another 29 per cent cited figures below 10 per cent. Finally, 34 per cent of respondents reported no training at all in the previous two years. 21 The existence, according to these figures, of only a small number of high-training companies could be a very general indication that the new systems of

Labour Relations and Investment in Workers 153

labour management did not necessarily involve the long-term development of human resources. But in order to explore this idea we need to have some criteria by which to assess the arrangements for labour flexibility that were used in the firms under study. Definitions of labour flexibility and Chilean labour relations We have already observed how the OECD’s distinction between ‘functional’ and ‘numerical’ forms ignores authority relations. Therefore it cannot explain how both forms may be associated with long-term relationships and skill development in some contexts and not in others.22 We propose to begin by looking instead at a more manageable and elementary aspect of workers’ control (and which, as we have noted, constitutes a basic element in the construction of occupational citizenship). This is the degree of formalization of labour management. Descriptions of wage flexibility tend to illustrate the importance of formalization. For example, in a tailoring firm (Cierres Wings), new machines had been introduced three months prior to the interview, and this had resulted in an immediate increase in the production of zips from 300 to 5000 per day per worker. The particular function in question was to separate the zips from one another, which had previously been done by hand. However, no change in the wage had followed. In a textile firm, the productivity increases had not been so obvious, but there was still resentment about the way in which workers had been excluded from decision-making and from sharing in the gains from increased production: Before, it took three minutes to do a sock. Now it takes one minute . . . after new technology arrived, when we still did not know how to operate the new machines, our tariff (piece-rate) was lowered. But our real wage dropped each time adjustments were made, the first time by 7 per cent. So up to a point we were deceived. Furthermore, management’s control over tariffs had facilitated the maintenance of informal wage setting, which was regarded as unfair: In the machine section we work in groups, and the payment is according to the volume of production. But wage setting is individual. There are cases among newer personnel where one man is paid more than another for performing the same function. There has always been tension among the workers. Our aspiration is that there should be a bit more justice in this field.

154 Social Citizenship and Chile’s Labour Reforms

Finally, local union leaders saw the achievement of formalized qualifications, that is a form of training that they could use elsewhere in the labour market, as crucial to widening the individual freedom of workers. To the question of what the term ‘modernization’ in the firm meant to him, one leader answered: For me it ought to mean raising our standards. In the sense of taking advantage of the new technology that arrives, and making provision for workers’ training, so that we will have skills (conocimientos) which will enable us to make out in any other place as well, which is taking advantage of the new technologies. In summary, it appears that a democratization of flexibility, its formalization, and eventually negotiation with workers, represented an immediate, everyday concern to local leaders in Chile. Some ways to assess formalization in Chilean firms Aggregate responses to our survey indicate that formalization in the firms studied was generally very low. For example, only 23 per cent of firms combined the existence of professional management (in which the owner was not involved on a day-to-day basis) with the following features of formalization: (i) the extension of formal contracts to more than 80 per cent of workers; and (ii) the existence of a general wage structure which was known to the union (see Table 8.1). A formal wage structure could be found in only 37 per cent of the companies studied. It may be useful at this point to introduce the cases of four specific firms, whose leaders were among those interviewed in depth prior to the actual survey. The cases were selected because of their varying levels of formalization. They are ranked in Table 8.1, starting with the firm where formalization was greatest, Firm A (San Joaquin).23 In this firm, workers classified by a similar broad category (‘operarios’) received the same wage, despite the fact that six different functions were involved, and were only distinguished by formal seniority arrangements (mechanics, cleaners and helpers). Administrative staff were waged at a separate level. In Firm B there were five different base wages in the broad area of production, as well as seniority. The experience of the other firms was entirely different. For example, in the two largest sections of Firm C, employing approximately 150 workers, the union leaders knew of at least 18 different wages (11 in one, and 7 in the other). Hourly rates were set individually. For example, for doing the same job, dyeing thread, there was at least four different wage levels, rang-

Labour Relations and Investment in Workers 155

ing from 150 to 220 pesos per hour. The circumstances of Firm D were described with the quotes cited above (p. 177). Approximate wage levels, and the internal wage differentials, also varied between firms B, C and D.24 But what can the formalization of pay tell us about other aspects of the work process? In comparing the four unions in question there seemed to be no straightforward relationship between the level of technological innovation in the firms, on the one hand, and the degree to which the wage system was arbitrary, on the other. Nor did there seem to be a clear relationship between the latter and the nature of the production system (continuous process or auto-mechanical). Care also needs to be exercised in using simple ‘group’ features of the wage system as a reflection of the commitment to human resources in firms. For instance the arbitrary piece-rate system in Firm D was based partly on team-work. There are still important distinctions to be made about how far such systems implied a sense of stability. This tended to depend on how far reward was calculated, or even negotiated, in a collective way. This emerges from the nature of the incentive system in Firm A, which had come about through negotiation. In this case the system was based on overall performance over a longer period, and did not form part of the base wage. In short, differences in labour relations, rather than differences in products or production processes, seemed at first glance to provide a more convincing explanation for the different levels of formality and the organization of the wage system in the four firms. Formalization and labour flexibility The extension of the more precarious forms of flexibility (the OECD’s ‘numerical’ forms) can be illustrated by ranking firms according to how many such forms they used. In doing so we found that of all the union leaders interviewed, only 38 per cent cited a yearly wage variability of under 25 per cent (their own wage).25 Another 37 per cent, on the other hand, experienced not only a greater than 25 per cent variability, but they also reported a level of labour turnover in the previous year that exceeded 30 per cent, and had experienced at least once the situation that the employer dismissed workers in order to rehire them on different terms.26 Furthermore, we found that there was also an association between low formalization, on the one hand, and the indicators of precarious flexibility, on the other. For example, as Table 8.1 shows, only 16 per cent of the firms that had a formal wage system as well as formal management, housed all precarious flexibility forms. This contrasts with 53 per cent of the firms where formalization was low.

156

Table 8.1

Elements of the organization of work and labour relations in four textile and clothing firms, 1992

Elements of the organization of work Production and training Nature of the production system

Hilandería San Joaquín (Firm A)

Cierrez Wings (Firm B)

Pichara (Firm C)

Textil Matex (Firm D)

Continuous process

Combination continuous process and auto-mechanical

Mainly continuous process

Combination continuous process and auto-mechanical

Estimated level of technological innovation in the last two years

Regular investments in new technology

Regular investments in new technology

No recent Investments

Regular investments in new technology

Training

None

None

None

None

Wage systems A formal wage system can be identified

Yes

Yes

No

No

A scale of four wage levels according to function

Apporox. five different wage levels corresponding to seniority

Multiple and arbitrary wage-setting (11 in one section with the same function), with attempts to formalize

Wage setting is totally individualized

Detail

Less than 10 per cent

Less than 10 per cent

More than 25 per cent

More than 25 per cent

Wage type (stability)

Fixed monthly wage

Fixed daily wage

Fixed hourly pay

Strict piece rates

Form of incentive wages

A collective production bonus spread over three months’ production

No added incentive

For some a weekly incentive (but operated arbitrarily)

No added incentive

Wage dispersion (the percentage difference between the lowest and the highest among ordinary workers)

Forms of representation Type of collective agreement

Convention

Contract

Contract

Convention

Workers’ say in training

Union took the initiative and gained some commitment from employer

None

None

None

Age of union

New (post-1988)

Old (pre-1973)

New (post-1979)

New (post-1988)

Source: Interviews with union presidents, March 1992.

157

158 Social Citizenship and Chile’s Labour Reforms Table 8.2

Formalization crossed with precarious flexibility

Chi Sq.2

Formalisation1 – from high to low, (i)–(iii)

Total

(iii) (ii) (i) 30% 100% 47% 23% 55 cases 113 cases 70 cases 238 cases −2306** Precariousness

1. None of the below (2 or 3) 2. Respondent had more than 5 per cent variation in his/her warnings last year 3. As under 2. and dismissals have increased and/or there has been rehirings with new type of contract

16%

39%

54%

31%

23%

19%

53%

38%

27%

= 100% = 100%

= 100%

Notes: 1. (i) Firms which have none of the characteristics below; (ii) Firms in which more than 80% of workers have formal contracts and (possibly) only one of the two characteristics cited under (iii); (iii) More than 80% of workers have formal contracts, a formal wage system exists and management is identified as separate from owner, or shared between owner and professionals. See also Haagh, 1997, Appendix F for detailed definitions of variables for all subsequent tables in this chapter. 2. * indicates significant (below 0.05); ** indicates highly significant (below 0.01). Source: Union survey, author, 1992.

We shall now examine the four textile firms again, to illustrate the association between formalization and two specific flexibility forms. Formalization and incentive wages A distinction may be made between incentive wages as a positive inducement and incentive wages as a negative stimuli. The first is where the workers’ sense of creativity and commitment to the firm is stimulated. The second is where workers’ output is optimized only in the quantitative and immediate sense, by making the workers’ basic income related to effort. The usage of piece-rates, at least in the Chilean case, may correspond rather well to the latter form. The intermediary form might be the case where the incentive is still a decentralized measure based on short-term outputs, but where it is merely added to the base wage. Finally, there is the type that is based on the performance of the firm as a whole

Labour Relations and Investment in Workers 159

over a period of time.27 In theory, it could be argued that only an entrepreneur who is willing to invest in the workforce in the broadest sense would grant forms of reward (such as the last mentioned above), which carries the risk of increasing the bonds of solidarity between workers (as well as the level of association between management and the workforce). In fact, it is interesting that it was only in Hilandería San Joaquín (Firm A), the firm where formalization was greatest, where the union leaders proposed their own incentive system, a collective ‘production bonus’, to be tied to monthly targets. The employer’s reaction is also noteworthy. Although positive, on the whole, he had not wanted the incentive formula to become permanent. Hence the final settlement (in collective bargaining) had established only a three-month ‘incentive period’, during which the workers would have to ‘earn’ a 10 per cent real increase in wages, by achieving a determined level of output. On the whole, though, the relatively open stance of the employer in Firm A was commensurate with the higher level of formalization, and conceivably of vested interests (or trust) in that firm, compared with the others. As the president of the union put it, ‘The employer [who took over the firm in 1989] comes from a family in which the firms have always had good relations with unions. I have a lot of confidence in him, without trusting him 100 per cent.’ The case of Firm C provides a useful contrast. The incentive system here was not as negative as the piece-rate system, because it operated in the form of an ‘extra’ bonus if a determined number of cones were rolled per week (approximately 2500). The problem was the system’s apparently arbitrary nature, an aspect that ties in with the informal character, overall, of the wage system in that firm. The union president said: In the spinning section where I work an incentive is paid. Where the trouble starts is where the employer does not keep to his share. For example, the worker may be producing 70 kilos a day. The employer observes that the worker is working towards achieving the final incentive [the actual pay]. So he stops the production at the back, by failing to leave the raw materials, and the worker cannot do as much work in the last days of the week. This impression in turn affected the leaders’ perception of the employer. The secretary of the union said ‘You see, he is not an entrepreneur, he is only a merchant. He is not a true industrialist, like other employers, like Machasa, that wants to produce. He stops the production,

160 Social Citizenship and Chile’s Labour Reforms

sells what he has and starts again from scratch.’ President of the union: ‘We say this because an entrepreneur would take on a manager and an administrator. Without education he can neither administrate nor lead a firm.’ Finally, Firm D demonstrated perhaps an extreme case of social division in the firm (and of the second principle of incentive wages noted above). In this case, workers could receive a certain percentage of the firm’s monthly production only if they agreed to work on Sundays. As the union president explained, for most of the 1980s working on Sundays had been a chore that workers could ill-afford to refuse. Supervisors had an interest in pushing workers to work on Sundays because their bonus was tied to periodic production targets (whereas workers received no bonus). Union president: ‘it is in their interest that we go on Sundays. That is their job, to egg us on, demand that we are there on Sundays. Up till the year 1989 we were obliged to go, even if we couldn’t or didn’t feel like it. There was a lot of pressure. Threats of dismissal and harassing us, psychologically the pressure was there.’ Formalization and task diversity But we should also consider one of the ‘functional’ flexibility forms, at least in passing. As we recall, the OECD assumed that this type of flexibility was associated with an emphasis on workers’ involvement, but we found that task diversity was in fact associated with low formalization in the Chilean firms under study (see Table 8.3). In Firms B and C, where this practice was recognized by union leaders, it tended to be associated with close supervision of output and time measurement systems. The descriptive term preferred was that of

Table 8.3

Formalization crossed with task diversity

Chi Sq.1

−1339*

Formalization – from high to low, (i)–(iii)

Existence of task Yes diversity No

(i) 23% 55 cases

(ii) 48% 113 cases

48% 52% = 100%

50% 50% = 100%

Notes: * indicates significant (below 0.05). Source: Union survey, author, 1992.

Total

(iii) 30% 100% 70 cases 238 cases 65% 35% = 100%

Labour Relations and Investment in Workers 161

‘work load’.28 The workers were frequently asked to sign contracts that stipulated work load, sometimes minding up to four machines at a time. The Union President said: the problem is that as a worker I cannot sign for four machines, since I may not be able to attend to them. If I sign this contract [which includes a specified workload to be attained], then I can be accused of breach of contract [if this workload is not attained] . . . So right there [the employer] is putting a cause for dismissal in the contract . . . But the workers allow themselves to be pressured by the employer, in fear of losing their jobs. Still, we need to note that one cannot simply infer from the low level of influence that the union leaders had in Firm D, or indeed from the pressures received from management in relation to aspects of the incentive system, that the union leaders rejected the use of variable wages. The union leader of Firm D, the firm where formalization was lowest, preferred the current system of piece wages to receiving a fixed wage: I like the tariff, but I would like it to be fair. Pay according to production, that is fine, but there should be formal rules, so that disruptions that are no fault of ours should be taken into account. Like when the electricity was cut during the years when there were protests. We did not receive anything for the hours that the machines stopped. 29 The union leader’s support for piece-rates appears, on the one hand, to be almost a paradox, given the high levels of supervision that a piece system entailed. On the other hand, given the level of distrust that evidently existed between labour and management in that firm, it is understandable that workers like himself might prefer a strict system of accounting, even though it might lead him to exhaust himself. As he put it, ‘If we had a basic wage then there probably would be even more pressure. Now we work with a certain measure of freedom.’ In summary, the example above illustrates the tensions felt by local unionists in Chile in their quest for a share of developmental freedom: how, for example, they saw bargaining on flexibility as a source of greater independence. On the other hand, the extent to which Chilean employers regarded flexibility at work as their prerogative was reflected in the hesitance, even of their more forward members (Firm A), to introduce

162 Social Citizenship and Chile’s Labour Reforms

incentive wages as a negotiated strategy which might improve the longterm security both of workers and the firm.

Inter-firm flexibility and the organization of work One of the key areas in the debate on flexibility is the practice of labour turnover. This had been made highly accessible by the Pinochet legislation (Chapter 6). The key question with regards to labour turnover is how far this could be associated with the relations of authority in the firm. The involvement of workers, for example, would be an indication that dismissals were indeed used as a ‘last resort’. On the basis of firm level statistics from the National Statistics Institute in Chile (which match nearly all cases in our survey), we have calculated roughly comparable indicators of variations in the total size of the workforce. 30 Table 8.4 compares the figures obtained from INE with those of our own survey (bearing in mind that we are comparing different years). Although the INE figures also do vary between the two periods cited, the variation between these and our own survey is greater. There are at least two reasons for this. This is partly because the INE figures are likely to underestimate the extent of variation in the size of the labour force.31 Second, INE figures may count only workers with formal contracts, whereas our figures reflect the number of (waged) workers in the firm. The INE figures are presented here to validate the generally high levels of variation in the size of the labour force found in the survey.32 Turnover and workers’ involvement The cross-tabulation in Table 8.5 indicates that among the firms that had a very high level of turnover of labour, only very few had a high

Table 8.4 Levels of labour turnover, figures from INE compared with figures from the survey of local labour leaders Selected INF categories

> 15% 0.1–15% 0.1 < Source:

Turnover 1986–88 (INE)

Turnover 1988–9 (INE)

Number of cases

%

Number of cases

%

75 58 67

38 29 34

52 96 50

26 49 25

Turnover, union survey

Turnover 1991–2 (approx.)

< 30% 10–30% > 10%

INE, Santiago, Chile and author’s survey of labour leaders, 1992.

Number of cases

%

107 72 53

46 31 23

Labour Relations and Investment in Workers 163 Table 8.5

Labour turnover crossed with level of formalization

Chi Sq.1

Turnover – from high to low, (i)– (iii) selected cells from crosstabulations1 (ii) 31% 72 cases

(iii) 23% 53 cases

18%

50%

62%

52%

27%

21%

30%

23%

17%

= 100%

= 100%

= 100%

(i)** 46% 107 cases −2529** 1 (2) and formal wages & (a) contracts & (b) formalisation 2 Owner is not or only partially involved in management and (a) or (b) 3 None of the above

Total

100% 232 cases

Notes: 1. * indicates significant (below 0.05); ** indicates highly significant (below 0.01). Source: Union survey, author, 1992.

level of formalization: 18 per cent compared with 62 per cent for those with low turnover. This is a first indication that there was an association between the liberal use of turnover and social aspects of labour management. In our qualitative cases, we found that it was only in the firm where the level of formalization and involvement had been comparatively great, in Firm A, that a broader exchange of flexibility had been achieved. In this case, the union president recounted how a period of low sales, the holiday period of January–February of the previous year, had led to an understanding that wages could remain unaltered (the automatic adjustment according to inflation would not take place). In addition, a few more holidays (without pay) could be allowed.33 Turnover was in the middle range in Firm D and in the high range in Firm C (see Table 8.4). But the contrast to note is Firm C, because of the overtly contestatory context that surrounded the usage of dismissals. According to the testimonies, the employer in Firm C had in recent years dismissed up to a third of the labour force in a link to the holiday season. It was suggested that these dismissals were made easier by the fact that there were habitually twenty to thirty workers (from an average of around 150) without permanent contracts. Furthermore, since the

164 Social Citizenship and Chile’s Labour Reforms

new labour legislation had begun to operate, the employer had used dismissal clauses that imputed fault to the worker (Article 2).34 The cases of dismissals in the textile firms add further credence to our earlier suggestion (see Chapter 6) that the dismissal regime in Chile was so liberal that it was feasible for employers to use dismissal as a disciplinary device. At the same time, they show how a greater level of authority by workers (Firm A) would increase the likelihood of a negotiated flexibility style. Turnover and labour flexibility Table 8.6 shows that the forms of flexibility that were least democratic also tended to have a significant association with levels of high turnover. In the case of task diversity (TD), 60 per cent of the firms with high turnover used TD in some form, whereas only 39 per cent of the firms with low turnover did so. Furthermore, it was more than twice as likely to

Table 8.6

Labour turnover crossed with labour flexibility and productivity

Chi Sq.1

Turnover1 – from high to low, (i)–(iii) (ii) 31% 72 cases

(iii) 23% 53 cases

46% 54%

40% 60%

22% 78%

= 100%

= 100%

= 100%

60% 40%

56% 44%

39% 61%

= 100%

= 100%

= 100%

20%

18%

40%

24%

39%

37%

56%

43%

23%

= 100%

= 100%

= 100%

(i)** 46% 107 cases +2552**

Measurement of individual output

+1665* Task diversity

Productivity: Introduction −2479** of last new technology meant respondents output per unit of time

Yes No Yes No Improved 0–39% Improved 40–80% Improved < 80%

Total

100% 232 cases

1. * indicates significant (below 0.05); ** indicates highly significant (below 0.01). Source: Union survey, author, 1992.

Labour Relations and Investment in Workers 165 Table 8.7

Cross tabulation of training and productivity by levels of turnover High levels of turnover* (58 cases)

Low training** Recent productivity increases*** Low High 72% 28% (34 cases)

High training Recent productivity increases Low High 54% 46% (23 cases)

Low and middle levels of turnover (41 cases) Low training Recent productivity increases Low High 55% 45% (22 cases)

High training Recent productivity increases Low High 32% 68% (19 cases)

Notes: * As defined in Chapter 8. ** Low training corresponds to > 9.99% of the workforce having experienced some form of training in the previous two years, and high training < 10% (see Appendix F, training b). *** Productivity defined as: increase in production per unit of time (case of respondent) with the introduction of last new technology: (i) Low = None – 40%; (ii) High = <41%. Source: Union survey, author, 1992.

find that individual output was measured in the firms with high turnover (54 per cent) compared with the low turnover group (22 per cent).35 There is also the important question of the presumed effects of a pressured environment (the ‘negative’ forms of inducements – in this case, the threat of job loss) on workers’ productivity, and which appears to be supported in Table 8.5, col. (iii).36 However, Table 8.7 presents a slightly revised scenario. There are, admittedly, relatively few cases listed in the table (cases with unreliable data were discounted). However, this indicates, at least, that the presence of workers’ training also seemed to have an inducing effect, but not in conjunction with high turnover. Training tended to have a positive association with productivity in the firms with low turnover.37 Social factors and turnover In recent revisions of the neo-classical paradigm, it has been proposed that the relevance of social factors might attenuate sharply the conventional assumptions that attitudes to work and the use of labour should be affected principally or solely by standard descriptive, as well as ‘marketdetermined’, factors of the firm’s development, such as ‘technological and budgetary constraints’.38 Our review has lent some support to this proposition. But it is clear that our argument would be slightly stronger if we could control for factors that provide an indication of general features, such as the firm’s ‘technological and budgetary constraints’. This is what we seek to do with the analysis of variance below (see Tables

166 Social Citizenship and Chile’s Labour Reforms Table 8.8

Labour turnover: analysis of variance* (excluding size of firm)

Variable to be explained (dependent): level of labour turnover* Predictor variables (independent)

Basic factors Low ratio of workers per machine High ratio of workers per machine High technology sectors Low technology sectors** Social factors (i) Labour management Use of task diversity Low level of skilled workers in the firm (ii) Authority More than 1 bargaining group in the firm Non-existence of formal wage system Residual = 420.977 DF = 8 Explained: (215 cases) 11.8% missing (29 cases)

Sum of squares

F

Sig. of F

0.041 8.035 0.002 12.123

0.020 3.945 0.001 5.953

0.887 0.048 0.977 0.016

8.476 20.552

4.162 10.091

0.043 0.002

11.428 22.136

5.611 10.869

0.019 0.001

146.215

8.975

0.000

Notes: *Analysis of variance was used rather than an ordinary regression, since the response variable was continuous and the predictor variables all categorical. See Appendix C, which also includes a definition of all variables. **As defined in Appendix A. Source: Union survey, author, 1992.

8.8 and 8.9). In this we include the size of the firm, the approximate level of recent machine investments relative to workers, and the firm’s placement according to sector.39 Size of firm was discarded, as this factor proved to be insignificant.40 In Table 8.8, where this factor is excluded, we find that firms where no new machines, or very few relative to workers, have been introduced tended to have higher levels of turnover. Placement in low-technology sectors is also an important indicator, as we would expect. However, the important finding in this table is that the factors relating to social aspects of labour management are at least as important. This is so even when we control for the influence of the first set of factors. 41 In the second analysis of variance table (Table 8.9), where the factor for approximate levels of machine investment is excluded (as the least significant factor), the significance of the social factors, as well as the levels of skill in the firm, become even clearer. At the same time, it appears that placement in low-technology sectors cannot be excluded completely. Table

Labour Relations and Investment in Workers 167 Table 8.9

Labour turnover: analysis of variance (parsimonious model)

Variable to be explained (dependent): level of labour turnover* Predictor variables (independent)

Basic factors High ratio of workers per machine Low technology sectors Social factors (i) Labour management Use of task diversity Low level of skilled workers in the firm (ii) Authority More than 1 bargaining group in the firm Non-existence of formal wage system Residual = 421.172 DF = 6 Explained: (216 cases) 11.8% missing (25 cases)

Sum of squares

F

Sig. of F

8.361 17.756

4.145 8.802

0.043 0.003

8.459 20.982

4.193 10.401

0.042 0.001

11.460 22.216

5.581 11.013

0.018 0.001

146.172

12.077

0.000

Notes: *As Table 8.8. Source: Union survey, author, 1992.

B.2 in Appendix B indicates that the overall significance of the model would be reduced significantly if this factor were to be removed. On the other hand, the ‘cycle’ whereby factors such as low formalization of labour management seem to reinforce high levels of turnover, and vice versa, is evident in both high- and low-technology sectors (although it was most intense in the latter). 42 Placement in a low-technology sector makes the association between turnover and the precariousness of labour management generally, more likely, or more pronounced. The significance of the social factors, however, is still striking. It is not a surprising finding, given our earlier discussion, that a lower proportion of skilled workers in the labour force would tend to predict that turnover would be higher, or that the practice of task diversity should be important in explaining the tendency to use dismissals to control the workforce. The two remaining factors, those that allude directly to aspects of authority, are also so significant as to make this explanation seem at least plausible. The number of negotiating groups was included here as an overt indicator of workers’ authority. Needless to say, the existence of a wage system that was barely known, and the existence of competition to represent workers, offered two key indicators that workers had little authority to counteract the usage of dismissals as a first resort.

168 Social Citizenship and Chile’s Labour Reforms

The contexts of workers’ training The time-horizon of labour management and investment in workers We have already disputed that actual demand should be sufficient to produce a desirable level of workers’ training in Chile (see Chapter 6). But perhaps for this reason, it is particularly interesting to inquire further about the sources that motivated employers to train their workers, and particularly about the importance of the form that labour flexibility had taken in many semi-industrial and industrial firms. We found that basic or traditional capability factors, such as the size of the firm and the level of mechanization, were relatively important in explaining the level of training (see Table 8.10).

Table 8.10

Workforce training crossed with basic characteristics of firms

Chi Sq.2

Level of workers’ training, from low to high, (i)–(iii)1

Total

(iii) (ii) (i) 37% 100% 29% 34% 83 cases 70 cases 91 cases 244 cases 0.2232** Size of the firm 1

Fewer than 49 workers 2 50–99 workers 3 100–199 workers 4 200+ workers

0.1763* Number of new machines introduced in the previous two years 0.0720

22%

12%

8%

28% 29%

19% 33%

20% 33%

21%

36%

40%

None or fewer 52% than 2 2 Between 3 and 26% 9 machines 3 More than 22% 10 machines

22%

27%

45%

43%

33%

30%

33%

42%

29%

21%

22%

18%

46%

36%

52%

1

Ratio of workers 1 Fewer than 10 workers per machine 2 40–100 workers 3 100 + workers

Notes: 1. (i) 0–0.99% of workers; (ii) 1–9.99% of workers; (iii) <10%. 2. * indicates significant (below 0.05); ** indicates highly significant (below 0.01). Source: Union survey, author, 1992.

Labour Relations and Investment in Workers 169 Table 8.11 Pearson correlations: workforce fluctuations crossed with investments and types of investment Level of workforce No. of fluctuation1988–9* firms Level of workforce fluctuation 1986–8 (high to low) Investment as a percentage of value added 1988** Investment as a percentage of value added 1989 (low to high) Investment in machinery 1988*** Investment in machinery 1989 (low to high)

0.43942 −0.33372 −0.17451

200 179 163

0.1257 0.19051

140 141

Notes: 1. Significance level 0.05; 2. Significance level 0.01. * The minimum number of workers as a percentage of maximum number in the same firm in 1988 and 1989 – lower figures representing higher turnover. ** Investment includes investments in land, buildings and machines as is calculated as a share in value added. *** Defined as investment is machines as percentage of total investment (land, buildings and machines). Source: Firm level data, INE, Santiago, Chile.

Labour turnover and training On the other hand, when the ratio of workers to machines was taken into account (col. (iii)), then mechanization was not so strong as an explanatory factor. The question of time-horizon needs to be brought in. For this purpose, we shall first make use of the INE data on labour turnover that was mentioned earlier. The first three figures in Table 8.11 are chi squares of labour turnover (as defined above), with levels of total new investment in 1988 of the firms in our sample. Surprisingly, perhaps, the two middle figures show that the higher the labour turnover, the higher the investment, as a share in value added tended to be. Labour turnover appeared to be a force of economic dynamism. The last two correlations, however, suggest a different scenario. This time, labour turnover is correlated with the level of machine investment in total fixed investment (other investment being in land and buildings). The assumption could be made that investment in machines was a form of long-term investment. And, indeed, now low levels of turnover correlate positively with high investment. Our survey is not capable of corroborating these provisional findings. However we also found that lower levels of turnover were more compatible with investment in human resources (see below). We shall begin by pursuing our more specific argument that labour relations had a

170 Social Citizenship and Chile’s Labour Reforms Table 8.12

Workforce training crossed with indicators of workers representation

Chi Sq.1

Level of workers’ training from low to high, (i)–(iii)2

Total

(iii) (i) (ii) 37% 100% 29% 34% 83 cases 70 cases 91 cases 244 cases (ii) and formal wages (a) and contracts (b). Owner is not, or is only partially, involved in management. None of the above.

14%

24%

31%

40%

50%

51%

46%

26%

18%

(ii) and (iii) and some participation in organisation of wage incentives and/ or work tasks. (iii) and existence of WHSC (see below) Some profitsharing. None of the above.

17%

35%

44%

22%

28%

24%

41%

14%

13%

20%

23%

19%

Functions Exists only in name Does not exist

20% 29% 51%

26% 53% 21%

53% 30% 17%

1.

−2549** Formalization 2.

3. 1.

Extent of non-union 2. −2449** worker representation 3. 4. −3488** Workers’ Health and Safety Committee (WHSC) (comité paritario)

Notes: 1. * indicates significant (below 0.05); ** indicates highly significant (below .01); 2. As Table 9.9. Source: Union survey, author, 1992.

perceptible relationship with the time-scale chosen by firms. In fact, the first cross-tabulation in Table 8.12 indicates that a high level of training was much more likely where labour management was formalized and where the wage system was known to workers.

Labour Relations and Investment in Workers 171

Participation and training Similar findings emerge from the cross-tabulations in cols (ii) and (iii). These are concerned with more explicit forms of worker representation (though still non-union). We found that, overall, only 19 per cent of companies had all the features of participation considered – that is, a form of profit-sharing, in addition to the existence of a Workers’ Health and Safety Committee (WHSC), and some participation in the organization of the wage incentive system and/or in work tasks. Nevertheless, this combination featured in 44 per cent of the firms that provided a high level of workforce training, whereas it existed in only 17 per cent of those that did not train the workforce. Existence of a Workers’ Health and Safety Committee seemed to be a particularly good indicator. Only 17 per cent of the high-training companies did not have one, whereas 51 per cent of the companies without workforce training did not have one. However, these observations are at the very least controversial, because the dominant market rational approach does not consider the relevance of labour relations in the firm, as we have noted. Again, by using a regression, we can get a better idea of how important the factors that relate to authority were, compared to the factors relating to the firm’s market position in a more neutral sense, such as the size of the firm, the approximate ratio of new machines to number of workers, and location by sector (depending on technology levels). The significance of the factors that described the organizational and relational aspects of work deserves a special comment. The first such factor was the level of turnover, and the second aspect was the practice of measuring individual output. This variable was included because not only was it associated with low formalization but also with the usage of individually-based incentive wages which, as we saw, tended to represent an exclusive flexibility form in Chile. The regression further adds to the supposition that such flexibility forms would not tend to coincide with, or to produce, an emphasis on human resources. Finally, we included a variable that described an aspect of workers’ authority directly. The functioning of the WHSC was found to be extremely significant. At the same time, some aspects of the physical space of production were also significant to training levels. Again, we found that the ratio of new machines was not an important factor. 43 Sectoral location was also not significant. On the other hand, the regression in Table 8.13 (without this factor), seems to confirm what was said earlier with respect to the size of the firm. For example, it was found that the enterprises with more than 200 workers were far more likely to train the workforce,

172 Social Citizenship and Chile’s Labour Reforms Table 8.13

Logistic regression: determinants of workforce training

Variable to be explained (dependent): training of more than 5% of workforce in the previous two years.1 Explanatory factors (i)

(ii)

Basic factors Sectors with a higher level of technology [dumhil 1] Sectors with a lower level of technology [dumhil 3] Big firms (rather than medium or small) [big 1] Small firms (rather than medium or big) [small 1] Labour management Existence of WHSC [whsc Low levels of turnover [r] Not measurement of individual piece-rate output [meas]

Constant

B

S.E.

Wald

Sig.

−0.5880

0.4180

1.9787

0.1595

−0.6302

0.4327

2.1209

0.1453

−0.8733

0.3961

4.8610

0.0275

0.7844

0.4199

3.4897

0.0618

−1.6511 0.2861 1.1688

0.3581 0.1120 0.3902

21.2564 6.5292 8.9717

0.0000 0.0106 0.0027

0.5908

0.4410

1.7952

0.1803

Classification table = explained 62.35% for value 0 of training; and 83.65% for value 1 of training = overall 74.82% Notes: See Appendix C for regression model and definitions of all variables. Source: Union survey, author, 1992.

compared with the two lower categories. However, cross-tabulations in Table B.7 in Appendix B indicate, interestingly, that the association between one of the social factors considered (the WHSC) and levels of training was very strong across large, medium-sized and small firms. We see again an indication that the ‘cycles’ between social relations in the firm and labour management were not entirely dependent on the physical space of production (in this case, size), and yet how the latter seemed to make such associations more likely. In the case of the four textiles firms, only in one, Firm A, which had the highest level of formalization, had there been any training for the mechanical staff. In Firm A, in 1991 the union had also made worker training a subject of bargaining with the employer, and while no changes had so far taken place, the employer had promised to consider the matter. In summary, there seem to be at least some grounds for supposing that the choice of training the workforce on the part of employers was to some extent based on the choice of labour management in a broader sense.

Labour Relations and Investment in Workers 173

That is, whether workers were involved in decisions on the shop floor, and indeed, whether or not workers were seen as a source of long-term capital or not. Internal and external factors and training The last issue we can explore, briefly, is how far some of the positive relationships we have discovered – that is, between skill development and participation, on the one hand, and between skill development and job stability, on the other, could survive within the atomized framework of flexibility in Chile. Did external constraints affect them? In Table 8.14 we have chosen to focus on the strong association we found earlier between training and representation in the firm. In fact, the table shows clearly that the tendency for low training and the absence of a WHSC to be present became far greater as the levels of labour turnover increased. The odds were almost 8 to 1 that this coincidence would be found where turnover was high compared to the context where turnover was low. In Table 8.15 we assess the question of occupational security in an even more direct way. In theory, the firm-based aspect of occupational citizenship would be optimized in contexts of high training and low turnover. How far was this relationship affected by the tendency towards low formalization in Chile? We have assumed that it was partly the relationship to authority that led to very different outcomes from the various flexibility forms. Hence it is to be expected that the association between high training and low turnover would be more likely in the firms where the system of remuneration was the subject of shared knowledge in the firm. Table 8.15 supports this supposition, with odds of more than 4 to 1. Again, this illustrates how far external and internal labour markets in Chile were linked in subtle ways. Social history mattered, but in turn it is likely that its positive effects were helped or hindered by certain external factors, including legal or other supra-firm inducements towards greater formalization inside Chilean firms.

Conclusion One of the central justifications behind the labour market reforms in Chile of the mid-1970s onwards was the assumption that, if external constraints were removed, the country’s firms would automatically produce a form of labour management that was optimal for economic development. The possibility was ignored that an abundance of freedom can also be a source of constraint – that is, that it can gear decision-making in such a way as to

174

Table 8.14

Influence of turnover on relationships between training and workers’ representation in the firm

Odds that low training coincides with the non-existence of a Workers’ Health and Safety Committee Type of firm

Odds**

Observed frequencies (%) High training

Low training

Yes

WHSC No Yes

No

Significance of logit models and two-way relationships

Sig.

Total

Firms with high turnover

7.7

15

15

8

61

100 (107) cases

Logit: (1) the relationship between training and representation in the firm is affected by turnover

0.03

Firms with medium turnover

4.4

25

22

11

42

100 (73) cases

Two-way odds: – high training by turnover high 0.4 medium 0.9 low 0.8

0.06

Firms with low turnover

1.0

10

34

14

42

100 (53) cases

– high training with representation

0.00 3.6

Notes: ** The odds are extracted from the observed frequencies as follows: (Low training and No/Low training and Yes)/(High training and No/High training and Yes). For further details about odds tables, see Appendix D. Source: Union survey, author, 1992.

Table 8.15

Influence of wage system (arbitrary versus formal) on the relationship between turnover and training

Odds that low turnover coincide with training (of more than 5% of the workforce in the previous two years) Type of firm

Odds**

Observed frequencies (%) High training

High

Low training

Turnover Low High

Significance of logit models and two-way odds

Sig.

Total

Low

Firms where wage system is arbitrary

0.6

13

6

46

36

100% (100 cases)

Logit: The relationship between turnover and training will vary depending on whether the wage system is formal

0.000

Firms where wage system is formal

2.9

16

42

22

20

100% (129 cases)

Two-way odds: – Absence of a formal wage system with low training – Low level of turnover with formal wage system – High turnover with low training

0.000 5.9 0.003 2.3 0.005 2.2

Notes: ** The odds are extracted from the observed frequencies as follows: (High training and Low turnover/High training and High turnover)/(Low training and Low turnover/Low training and High turnover). Source: Union survey, author, 1992.

175

176 Social Citizenship and Chile’s Labour Reforms

make a short-term perspective the most rational management form. In the case of Chile, the likelihood that a short-term perspective would be taken was further enhanced by the speedy liberalization of other areas of the economy – that is, of financial markets and of trade – and by the fact that this liberalization had not been accompanied by measures to cushion local producers, or to help them develop their capabilities in the long term. More importantly, we found the central premise behind the Pinochet labour reforms to be false. We found that labour management was not simply a swift response to external forces, even in the Chilean case, where these had been allowed the maximum possible space. Some of the strongest associations we found were between simple aspects of representation (for example, the WHSC, and the level of formalization) on the one hand, and the time-horizon of labour management on the other. These associations speak of a ‘lock-in’ effect that, to some extent, also preceded the pervasion of market forces. The presence of ‘lock-in’ effects, of course, questions the thesis of a natural process of selection of the level of human capital required for economic development, not least because of the many social components from which such capital is derived. We found some evidence to support the neo-liberal model of productivity growth in the firm (based on workers’ exertion and fear of job loss), or at least that this model could be made to operate in the primitive context of work. Still, it did not follow from this that the positive sources of workers’ performance (the experience of training, in this case) may not also operate in this primitive context. Furthermore, our survey indicated that high turnover had a negative effect on skill development, which may represent a more sustainable source of productivity growth. Hence, we have at least indicated that, in the methods of achieving productivity outcomes, there may be an element of choice. In addition, we could not rule out the fact that labour relations might be important in explaining the use of high turnover by firms, even when we controlled for certain regular factors. In other words, the hypothesis could not be discounted that deregulation itself can engender a pattern of high turnover, and hence that it can sustain a culture in which turnover becomes the preferred option. It has been beyond the scope of this chapter to assess all the elements that led it to be the case that a key stratum of employers in Chile placed so little emphasis on human resources. Equally, it is difficult to establish precisely the origins of the forms of labour integration which, in some firms, would seem to be mitigating the incentives towards short-term forms of management that the Pinochet reforms had provided (we shall explore this point in Chapter 9). However, we have been able to indicate that labour relations played at least a key role in guiding labour management in the context of the external constraints it faced.

Labour Relations and Investment in Workers 177

Our discussion has suggested how central elements of occupational citizenship was an implicit feature of day-to-day labour relations. The lack of foresight and access to personal planning and preparation remained a continuing preoccupation for workers. The very absence of alternative forms of occupational continuity, the very arbitrariness of labour management, and the isolation of workers in individual firms made the search for something akin to developmental freedom an ingrained part of the emergent understanding of rights. In fact, we found that the search for an element of control over daily aspects of work, such as the wage system, was so pronounced among workers’ leaders, and the attainment of such control so elusive, that some of them even preferred the short-termism of piece-rates simply to escape the prevailing lack of accounting. It was partly this position of the workers’ leaders that made it plausible, as we suggested at the beginning, that the local unions might perform a democratizing role in developing incipient forms of occupational citizenship inside Chilean firms. Finally, the concentration of these struggles within the individual firm reminds us of the important aspects of the labour reforms that ‘never were’ of the early 1990s. Chile’s Labour Code, even after 1991, barely provided a forum in which workers’ representatives could agree to forms of flexibility other than the use of dismissals. Nor could they promote the types of incentive wages that motivated workers (see Chapters 4 and 6). For example, we might recall that the restrictions on matters that could be brought up for collective bargaining had remained so tight that even the issue of workers’ training could legally be excluded as a theme for debate. The Workers’ Health and Safety Committees were restricted to dealing with matters of health and safety, and in any case did not operate in most of the firms. The employers’ right to modify the contents of the work contract was also not amended. This faculty had been crucial to the employers’ ability to introduce multiple task practices of the kind we described. In a sense, an unintended consequence that followed upon the expansion of individual freedom had been the creation of a new collective constraint, since there was subsequently no easy way to make individual firms emphasize human resources. Local entrepreneurs were hardly likely to generate new practices that could form the basis of a legal shift. Indeed, the fact that labour relations were so powerful a factor in explaining labour management, and that many employers saw minimal forms of inclusion as a form of risk, indicate that there was also a cultural element present in many firms which acted against investment in human resources. At the very least, it seems plausible to suggest that the atomized enterprise system had found the means to sustain itself.

9 Union Strength, History and Effects

The thoroughness of the economic reforms of the 1970s and 1980s diminished union presence worldwide. It also raised new issues in the analysis of unions. The alternative faced was posed as one between classical unionism and classical markets. Even in mainstream academic analysis there has been a tendency to emphasize the importance of territorially based movements compared with labour unions.1 In short, it remains controversial, to say the least, how far workplace unions did play or can play a role in the new scenario. And it remains to be seen how far this role is appropriate, at least in principle, to the universal definition and application of rights. In this chapter we analyze the factors that motivated the operation of industrial unions in Chile. We also assess their incipient role in building forms of occupational citizenship. Privatesector unions were remarkably quiet during Chile’s transition, above and beyond what would be expected to accompany a period of political opening. 2 This quietness might suggest that unions were not capable of exploiting the new environment. Furthermore, a unionization rate of 15 per cent was not impressive. It was even within the lower range of an average estimate of unionization rates in periods following authoritarian rule.3 But the question remains whether it is only important to analyze unions when they promote open conflict in the classical sense, or when they are imposing in number. We shall use an approach that diverges from the mainstream tendency to focus on the indices of peak conflict and deduce from these the effects that unions have on economic and political order. The low level of open conflict in the private sector during the period under study leads us to think about other ways in which the power of workers’ representatives is applied. In addition, as noted earlier, there were certain basic characteristics of the enterprise unions in Chile that had remained important since Angell first defined 178

Union Strength, History and Effects 179

them in the early 1970s and which suggest that the local unions were not simply co-opted by employers (as ‘company’ unions), the industrial relations system or the state to behave in a certain way.4 Because of this aspect of independence (and accountability) there was a reasonably good basis for studying how the day-to-day context was shaping the unions and was in turn shaped by them. In essence, it was still sufficiently difficult in the early 1990s to sustain a local union in Chile, and sufficiently risky and unprofitable for local leaders to engage in union activity, for it to be probable that unions were created for the basic purpose of defending the interests of workers at the firm level.5 We shall continue to use the survey of local union presidents and the qualitative interviews from pilot surveys. However it is important to note that, in order to assess the influence of the unions, we have to proceed by distinguishing between different types of union, based on the unions’ levels and forms of strength, given the fact that there are no non-union firms in our sample. On the other hand, this perspective evidently makes it possible for us to explore the problem of what union strength is, as well as the costs some unions may have incurred for the influence they had attained during the authoritarian period. The chapter is divided into four sections, of which the first is a general discussion of union analysis. We argue that neo-liberal and mainstream analysis of the role of unions as a simple matter of numerical strength rests on two assumptions. The first is that unions are the sources of conflict, and the second, that workers’ motives are fixed. In the remaining three sections we question the validity of these assumptions in the Chilean case, by posing three questions: Was the impact of unions on labour management necessarily destructive, and how can union strength be defined? What were the sources that influenced labour’s opinions on flexibility and the likelihood of flexibility bargaining? And, did the traditional institutions (such as sectoral associations) play other roles apart from providing a vehicle for union mobilization?

Local unions and labour flexibility As Franzosi has observed, it is noteworthy that, from the time when labour statistics were first gathered (in some countries as crime data collected by the police) to the present day, the literature on strikes has been dominated by indicators that place their root cause implicitly in the activities of unions.6 The nature of this focus may carry the unfortunate inference that it is in the first instance simple characteristics of the unions that underpin and shape industrial relationships, such as

180 Social Citizenship and Chile’s Labour Reforms

the unions’ sizes and levels of centralization. The nature of labour management is absent from the picture. We shall suppose that the definition of union strength must also focus on qualitative dimensions, and that it must be historically sensitive. As was noted in Chapter 6, much of the recent literature on union effects is based on the traditional indicators of union strength. In general, the tendency to focus on peak labour conflicts7 arguably experienced a certain revival as neo-classical approaches came to wield a dominant influence on development studies in the 1970s and 1980s. 8 The business cycle explanations put forward then may be highly relevant to explaining wage and employment patterns in some contexts, such as Chile’s economic crisis of the early 1980s and its aftermath.9 On the other hand, we need to note that extreme levels of unemployment accompanied the 1983 protests in Chile, and these demonstrations were political protests rather than strikes. Relative deprivation is hardly more plausible as a cause, because not only were the leaders of the protest movement among the most well-off workers in Chile, and among the best organized, but their associations and motives were highly political.10 Perspectives that focus on the capacity of unions to mobilize resources and on their political links seem better equipped to interpret the labour movement’s actions during this period.11 But nor can they explain the low-key existence of Chilean industrial unions at the time of transition. On the other hand, whereas institutionalization theory abandoned the exclusive focus on peak (and periodic) conflicts, it provided little scope for analysis beyond the observation that piecemeal integration of workers through democratic institutions can increase industrial peace. When does institutionalization occur? What are the roles of different unions, and different forms of collective contract, of norms, and of power?12 The governance approach can provide a more nuanced picture of union development and action. In the sociological stream of governance literature, for instance, institutionalization is no longer seen as a phenomenon occurring discreetly as a function of economic development.13 This provides for greater sensitivity to historical variation. Indeed, both the shape and costs of acquiring rights in Chile must correspond in someway to the peculiar methods of organizing the relationships of power between labour and employers there. For example, in Chapter 8 we argued that the decentralized form and informal character of labour relations in Chile meant that, in most instances, labour representatives could only influence labour management in discrete ways. A relevant question to ask, therefore, is whether union presence had ameliorated, indirectly, the incentives towards short-termism that the liberal reforms

Union Strength, History and Effects 181

of the 1970s and 1980s had wrought? The importance of the local union is also underlined by the fact that issues of occupational security and skill development were delegated to the level of the firm in Chile. For example, had local unions been stronger, or/and tied to a broader professional domain, it is unlikely that local employers would have been able to use 62 per cent of the state credits that were intended principally for workers for management training (see Chapter 7). In this chapter we further explore some ways that the ‘harder’ institutional factors (such as the role of unions and contracts) and the ‘softer’ social factors mentioned in writings on preferences, such as time factors and motivation, interact. With respect to the behaviour of workers’ leaders, our reasoning is slightly different. We refer again to Figure 8.1 on page 151. It is assumed that factors under (a.i) have tended to produce what we might refer to as a broad inclination towards control or security trade-offs among local labour leaders (as our data will show). On the other hand (b.i) refers more directly to the actual behaviour of the local union. As we shall see, despite a strong general disposition towards social trade-offs, other variables influenced strongly how far unions were able to channel the preference in a constructive way. Chief among them were factors that affected the developmental freedom of workers, for example the inclusion through representation and the time-horizon of labour management.

Definitions of union strength Union strength by density and legal instrument Our survey of industrial unions revealed that the traditional indicators of union strength were barely significant in explaining basic union characteristics. For example, unionization levels did not increase significantly with the size of the firm, and they made little difference to the type of collective bargaining instrument (see Table 9.1). Variables that were associated with the quality and specific history of unionism in Chile had stronger explanatory power. And only this type of variable displayed associations with labour management. We selected two such indicators that seemed of particularly interest, given the nature and recent evolution of Chile’s industrial relations. One of these was the difference between the use of contracts and conventions as the instruments of collective bargaining. The controversy about legal instruments in Chile has resided partly in the fact that there are few legal guarantees associated with conventions and, if this instrument is chosen, workers have no right to strike. In addition, the use of conventions came to be associated with the ad hoc forums for bargaining

182 Social Citizenship and Chile’s Labour Reforms Table 9.1

Unionization levels in firms, various indicators

Chi sq.

Unionization levels (i) > 60% (136 cases)

+1282*

+2302**

+1328*

Respondent is union leader for the first time

Yes No

Size of firm

> 49 50–99 99–199 < 200

The union has a

Contract Convention

1

(ii) < 60% (93 cases)***

50% 50%

63% 37%

100%

= 100%

17% 24% 28% 31%

9% 19% 37% 35%

100%

= 100%

63% 37% 100%

76% 24% = 100%

Notes: * Significant below 0.05 level; ** Significant below 0.01 level; *** As in Chapter 8, see Haagh, 1997, Appendix F for detailed definitions of variables in this and all subsequent tables. 1. In this case the differences became more marked when unionization levels above and below 40% are compared. Source: Union survey, author, 1992.

that were set up in 1979 to compete with the unions – the ‘bargaining groups’. This is because these bargaining groups were not allowed to sign contracts (see Chapter 6). The transition government of 1990 had initially attempted to repeal the use of bargaining groups, but later the view was expressed, by President Aylwin’s Labour Ministry team, that the formation of bargaining groups might serve as a ‘forerunner’ to real unions in the aftermath of dictatorship. And, of course, the 1979 reformers had originally conceived of bargaining groups as an alternative to adversarial labour relations. In the light of this debate, two relevant questions seem to be the following: could conventions protect workers to the same extent as contracts, despite their weak legal powers; and, if contracts were associated with a more adversarial labour relations, in what sense was this the case? On the first point, there seems to be some support in our data for the assumption that unions with contracts were stronger organizations in the firm. Not only were density levels slightly higher in unions with contracts (see Table 9.1), but contracts also seemed to offer greater formal protection than did conventions, at least in some fields. For example, in 52 per cent of the unions with conventions, it was estimated that more than 20 per

Union Strength, History and Effects 183

cent of the workers did not have formal individual contracts. For unions with (collective) contracts, the figure was 37 per cent. There were also fewer cases of ‘unlawful’ dismissals reported in unions with contracts (74 per cent in unions with conventions against 55 per cent in unions with contracts).14 Second, union leaders with contracts tended to be more critical, on the whole, of the employer. For example, only 25 per cent of those with conventions believed that the employer ‘tries to eliminate and/or prevent the growth of the union’, compared with 41 per cent for the unions with contracts.15 And yet the tabulation in Table 9.2 suggests that the more negative view of employers among unions with contracts only occurred in the firms where a great number of workers had no individual contracts – that is, where formalization was very low. Not surprisingly, this indicates that unions with contracts were more sensitive to the problem of informality as a form of control. However, an equally valid interpretation is that they were rational in their evaluation of management. In summary, our discussion in this section could be seen as a general note of caution that union strength (in the legal sense) cannot necessarily be associated with a more conflictive labour relations per se, even during the period of transition when there was more scope for unions to have their voice heard. The importance of union continuity But type of legal instrument is only one element in understanding union strength. We found that older unions in Chile were more likely to have consolidated an ‘embedded’ presence in the firm (see Chapter 6), and that they needed the protective framework of contracts less than did other unions. That union age should matter is far from obvious in Chile, because of the severe repression of unions after 1973, and hence the tempting conclusion that the old unions had been neutralized effectively. On the other hand, our findings highlight an aspect we discussed earlier, namely the importance of subtle forms of influence as an aspect of union strength in Chile during the period under review. In order to explore the importance of union history, we decided to distinguish between four periods, as the cross-tabulations below will show. These periods were: pre-Allende; between the Allende Government and the fall of democracy; between the year of the coup and the year of the plebiscite for democracy (1988); and, finally, the transition period (between 1988 and 1992 – the year of the survey). According to our general hypothesis, the longer the union had existed, the greater the level of workers’ authority (in relation to labour management) was likely to be. However, the most important distinction

184

Table 9.2

Odds that union leaders with contracts find that labour relations are bad

Types of firm

Odds**

Observed frequencies (%) Contracts Conventions Labour relations Good

Bad

Good

Bad

Significance of logit models

Sig.

Total

Firms where more than 80% of workers have individual contracts and in which a formal wage system could be identified

0.4

32

33

10

25

100% (177) cases

Firms where only one or none of these conditions pertain***

2.8

13

57

11

18

100% (61) cases = (231)

Where % of individual contracts does effect the relationship between legal instrument and labour relations. 0.06

Notes: * Significance: below 0.05 is considered to be significant, and below 0.01 to be highly significant. ** The odds are extracted from the observed frequencies as follows: Contracts and Bad/Contracts and Good/Conventions and Bad/Conventions and Good. Source: Union survey, author, 1992.

Union Strength, History and Effects 185

from our perspective was that between the unions established prior to the coup (both pre- and during the Allende Government) and those established after the Pinochet code. Aside from the differences in context, the time-gap was also considerable (hardly any unions were formed between 1974 and 1979).16 Furthermore, it could be presumed here that unions that had managed to survive after 1973, and during the period in which collective bargaining and many other union activities were banned, must possess, in one way or another, a significant presence in the firm. The hypothesis is that the firms that had grown accustomed to local unions when these had relatively greater powers (single-unionism, for example) and greater political clout, might have retained a form of institutional memory, even despite the fact that unions were officially dormant between 1974 and 1979. Union age, formalization and time-horizon The hypothesis that union presence had remained significant, even when unions had been stripped of their legal powers, suggests that unions might have had a bearing, first of all, on the day-to-day aspects of labour management. The cross-tabulations in Table 9.3 indicate that the older the unions were, the smaller was the wage scale and the greater the likelihood that there was a formal wage system in the firm.17 In the case of the existence of a WHSC, there was a clear dividing line between the unions that were formed before the coup and those formed in the Pinochet period. But, following our analysis in Chapter 8, where we saw that the level of formality was strongly linked to the time-horizon of labour management, we might expect that the unions’ age would also influence the firms’ levels of turnover and worker training. And indeed, in the case of both variables, there was some difference between the pre-coup and post-coup groups of unions (see Table 9.4). Furthermore, training was particularly likely to be low and turnover was particularly likely to be high in the firms with the most recently-formed unions (post-1988), compared with the others. In addition, in the case of union age (as with size of firm) we found that its removal from the regression when seeking to explain the levels of training in the firm would imply a significant reduction in general significance (see Appendix B, Tables B.3–B.5). These tentative indicators, then, seem to corroborate in some way the hypothesis that the presence of unions prior to liberalization and deregulation may at least have contributed to the amelioration of a tendency towards short-term forms of labour management. On the other hand, the long-term presence of the union also seems to have created a form of institutional memory that affected the union

186

Table 9.3

Age of union crossed with indicators of formalization of labour management Date of union establishment

Chi sq. > 1969 22% (48 cases)

+1908*

+1677*

+1548*

Number of different (known) wages per 100 workers**

0–3 4–10 > 11

Firm has formal wage system

Yes No

Existence of WHSC**

Functions Exists only in name Does not exist

1970–1973 34% (75 cases)

1974–1988 23% (51 cases)

Total 1989–1992 21% (47 cases)

68% 22% 0%

40% 48% 13%

38% 54% 8%

33% 41% 27%

= 100%

= 100%

= 100%

= 100%

63% 38%

48% 52%

44% 56%

37% 63%

= 100%

= 100%

= 100%

= 100%

39% 34% 27%

40% 38% 22%

24% 41% 35%

25% 34% 41%

= 100%

= 100%

= 100%

= 100%

Notes: * Significant below 0.05 level; ** Workers’ Health and Safety Committee (comité paritario). Source: Union survey, author, 1992.

100% (221 cases)

Table 9.4

Age of union crossed with labour turnover and training1 Date of union establishment

Chi sq. > 1969 22% (48 cases) −2033**

−1801**

Percentage of ordinary workers who have experienced training in previous two years

Level of labour turnover

None or > 0.99% 1–9.99% 10% and greater

More than 30% Between 15 and 30% Between 0 and 10%

1970–1973 34% (75 cases)

1974–1988 23% (51 cases)

Total 1989–1992 21% (47 cases)

27% 35% 38%

26% 30% 44%

36% 39% 25%

50% 17% 33%

= 100%

= 100%

= 100%

= 100%

48% 28% 24%

36% 34% 30%

50% 38% 12%

67% 22% 11%

= 100%

= 100%

= 100%

= 100%

100% (221 cases)

Notes: 1. As defined in Chapter 8; ** Significance level below 0.01. Source: Union survey, author, 1992.

187

188 Social Citizenship and Chile’s Labour Reforms

leaders’ perception of labour relations. The importance of institutional (as opposed to individual) memory is highlighted by the fact that the very old unions did not seem to have more cases of experienced leaders. Exceptions were the unions formed after 1988, in which 75 per cent acted for the first time (Table 9.5). The last two columns in Table 9.5 show that the sense that the employer wanted his/her workforce for the long term became gradually stronger the older the union. These figures can be interpreted in two ways. On the one hand, a literal interpretation must have some validity, given the fact that labour management did tend to be more inclusive of the workforce in the older unions. On the other hand, we cannot discount the possibility that leaders of established unions had been induced to accept current practice as acceptable, during a period of authoritarian rule where consultation with unions could also be used as a way to extend a form of benevolent authority over the unions. In other words, the flow of authority was not unidirectional and served to constrain the behaviour of both parties concerned. Union strength and union history We can now return to the question of the importance of the bargaining instrument depending on the age of the union. The idea that union age affected the usage of legal powers needs some further clarification. Conventions were also used in the pre-coup period, but it may be possible that conventions served a different function for the unions that were established when having a single union18 was the norm (in the pre-coup period), and continued to do so because of the influence of institutional memory, which we have already established to be an important factor. In short, we may have reasons to suspect that the importance of legal protection (contract) in Chile might have been highly contingent on the degree to which relationships of mutual constraint had already been formed between the union and management in the firm. It was dependent on the degree to which the time-horizon of labour management had already been affected by the presence of unions. Our survey data seemed to provide some support for this thesis. In Table 9.6 we can see that training was far more likely in the firms with older unions, irrespective of legal instrument. However, there was also a strong difference between conventions in new and old unions. In fact, the odds were more than 4 to 1 that training would take place in the firms where it was an old union that had signed a convention, compared with those where a new union had done so.19 We also found that labour turnover was far more likely to be high in new unions with conventions than in older unions with the same legal

Table 9.5

Age of union crossed with union leadership and level of workers’ integration Date of union establishment

Chi sq. > 1969 22% (48 cases) −2504*

First time as union president

Yes No

1.

+2147**

Level of worker Integration.

2.

3.

(2) and incentive wages enhances sense of commitment to the firm and employer seen to regard workforce in long-term perspective. Some training (more than 1% of workers) and some say in the usage of incentive wages. None of the above.

1970–1973 34% (75 cases)

1974–1988 23% (51 cases)

Total 1988–1992 21% (47 cases)

45% 55%

48% 52%

53% 47%

82% 18%

= 100%

= 100%

= 100%

= 100%

32%

17%

18%

9%

41%

57%

46%

41%

27%

26%

36%

50%

= 100%

= 100%

= 100%

= 100%

100% (221 cases)

Notes: * Significance level below 0.05; ** Significance level below 0.01. Source: Union survey, author, 1992.

189

190

Table 9.6

Odds that training is high (depending on union age and type of bargaining instrument)

Type of firm

Odds**

Observed frequencies (%) Training

Unions established before 1973 and using contracts Conventions Unions established after 1973 and using contracts Conventions

Significance of logit models

Sig.*

Total

High

Low

0.86

47

53

0.94

48

52

0.42

30

70

0.17

14

86

100% (86) cases 100% (33) cases 100% (64) cases 100% (28) cases

Only legal instrument has an effect on training Only age of the union has an effect on training

0.25

Both have an effect, which varies with each level of the other variable

0.21

0.001

= (211) cases Notes: * Significance: below 0.05 is considered to be significant; and below 0.01 to be highly significant. ** The odds were extracted on the basis of the observed frequencies as follows: low training (> 9.99% of workforce)/high training (10% of workforce and <). Source: Union survey, author, 1992.

Union Strength, History and Effects 191

instrument, (see Table 9.7). The figures make it difficult to rule out the hypothesis stated above that conventions in firms with older unions served rather like informal an mechanisms to sustain a more inclusive, or at least a longer-term, relationship with the workforce. On the other hand, it seems plausible that, in the newer unions, conventions had been introduced in association with a highly precarious form of labour management – that is, where turnover was very high. We have already discussed the possibility (in Chapter 5) that unions (and presumably this was particularly the case for new unions) sometimes had little choice but to accept the employers’ preference for a convention, which provided workers with few legal guarantees, and which did not allow them to strike. Regarding the four case studies introduced in Chapter 8 the two out of the four that had signed conventions happened to be less than three years old in 1992. This would lead us to expect that both should have very little authority to influence the forms of management. And yet, in one of them, San Joaquín, a relatively long-term commitment to the workforce had been made at the bargaining forum (see Chapter 7), where the employer had accepted a three-month bonus scheme based on overall performance of the firm. Relative to other Chilean firms in our sample, the level of dismissals was low and formalization of the wage system was high. San Joaquín was also the only firm among the four where worker training was discussed (and had been raised by the union – see Chapter 8). The point to note about this case is that it was possibly the arrival of a new employer with long-standing experience of unions in his other firms that explained the extension of these commitments to the union. This was the account the union leaders gave when comparing the firms’ labour management and relationship with the union before and after the firm had changed hands. The case of San Joaquín is interesting here, then, primarily because of the way in which institutional memory was believed to have affected the employer’s understanding of his relations with workers. The union president defined his choice of a convention in relatively positive terms: We signed a convention, not a contract. There is more conversation . . . We had consulted with older workers, and there were lots of meetings and exchanges, but no legal steps. We also wanted to see first how much we could negotiate without entering into the scenario of a contract, which implies the possibility of striking. Then we probably would have had more problems [with the employer].

192 Table 9.7 Odds that labour turnover is high (rather than medium or low), depending on union age and type of bargaining instrument Type of firm

Odds

Observed frequencies (%) Labour turnover

Unions established before 1973 and using contracts Conventions Unions established after 1973 and using contracts Conventions

High

Medium + low

0.7

48

52

0.3

21

79

1.0

51

49

2.8

72

28

Significance of logit models

Sig.*

Total

100% (82) cases 100% (33) cases 100% (63) cases 100% (25) cases = (203) cases

Only legal instrument has an effect on turnover Only age of the union has an effect on turnover

Both have an effect which varies with each level of the other variable

0.48 0.01

0.00

Notes: * Significance: below 0.05 is considered to be significant; and below 0.01 to be highly significant. ** The odds were extracted on the basis of the observed frequencies as follows: high labour turnover (30.01% of workforce and 30.00% of the workforce). Source: Union survey, author, 1992.

Union Strength, History and Effects 193

By contrast, the other union with a convention had clearly been constrained to sign a convention because of the origins of the union in the experience of bargaining through ‘bargaining groups’ during the 1980s. The advent of democracy had been the key element in provoking the formation of the union itself, in 1991, through the initiative of the representatives of the previous ‘groups’. Still, the ad hoc format of bargaining from previous years did not change during the latest round of bargaining. This included rules stating that the employer would set the level of wage adjustment (adjustment with inflation) and that the convention should contain no other agreement. It was also in this firm that complaints were made that the union was not allowed a day-to-day presence in the firm (see Chapter 7), and it was the only firm among the four where a core group of workers did not belong to the union. We should also compare the two unions that had contracts, because of the different contexts in which this instrument had originally been used. In the case of the old union in Cierres Wings (formed in 1969) a contract had been chosen as part of the return to bargaining practices in 1979. This came about after a period in which the union was dormant (paralizado), but during which time certain past practices had been observed, including the customary adjustment of wages linked to inflation (between 1973 and 1979 this was not compulsory). In the other firm, Pichara, a contract had been chosen for a very different reason. Essentially, the leaders of the new union (formed in the early 1980s) were conscious that they needed the powers of this instrument for their protection.20 On the one hand, this meant that the dynamism in labour relations was essentially adversarial, as emerged from the high level of disciplinary dismissals that seemed to follow from the determination of both sides to ‘show their strength’ (medir fuerza) during bargaining rounds. On the other hand, the Pichara union had also had some success since 1991 in using collective bargaining to have new themes accepted, including demands for a formalization of wage scales. Finally, the possibility needs to be mentioned that there had been some loss of autonomy to the unions that had attained their influence on labour management largely because of the employer’s familiarity with unions. For example, although the old union with a contract (Cierres Wings) apparently had a good understanding with the employer (see Chapter 8), and despite the extension of a range of traditional benefits, 21 there were evidently important elements of inertia in the relationship. This was the only firm where incentive wages had not been brought in, and while union leaders would like a bonus scheme, they had not dared to propose one. On the other hand, in the case of

194 Social Citizenship and Chile’s Labour Reforms

San Joaquín, a convention had been signed in the hope that the employer’s prior familiarity with unions would provide them with sufficient protection. Hence, there was clearly an element of coercion present, because the union leaders still felt that they might cause offence by proposing the use of a contract.

Local labour leaders and the flexibility debate The principled support for wage flexibility In Chapter 4 we saw that Chilean labour leaders, by 1992, were highly sensitized to flexibility issues, and from the outset this contradicts the neo-liberal assumption about the limited preferences of workers. However, it might be interesting to examine the strength as well as the roots of the commitment of labour leaders to those aspects of flexibility that involved an immediate sacrifice – such as wage restraint or variable wages. The overwhelming level of support found among local leaders for wage flexibility in principle ranged from national to local issues and compares with the strength of the commitment to democracy and to democratic forms of decision-making in society generally, which the local leaders also expressed. A sense of duty, for example, was implied in the support for the national minimum wage accord of 1991, by which adjustment levels became linked to productivity increases and future inflation (bearing in mind that inflation was experiencing a downward trend). Only 13 per cent of the union leaders interviewed thought that this accord was neither fair nor necessary. Meanwhile 84 per cent of the union leaders interviewed supported the accord, even though, of this number, a large majority (57 per cent overall) thought that it was unjust. 22 These leaders had ticked the response, ‘the accord is unjust, but necessary’. The possibility that union leaders linked their support for the 1991 formula to their wider sense of responsibility in consolidating the democratic regime is also reflected in the high level of support (79 per cent overall) that workers gave for to idea that the moderation of workers was necessary in order to consolidate democracy; 14 per cent thought such moderation was unnecessary, and 7 per cent had no view on this. 23 Another question inquired whether respondents were concerned with the productivity of their own firm even if this might mean accepting lower wages at times. Seventy five per cent agreed, and only 18 per cent felt union leaders should not ‘mix the interests of the employer with those of the workers’. Only 7 per cent confined their concern to ‘crisis situations’. At the same time, local leaders did expect a form of exchange.

Union Strength, History and Effects 195

For example, while 78.5 per cent of respondents to another question preferred the response that wages ought to vary with the firm’s productivity rate, most of them (84 per cent) thought that a fair exchange for such variability would be some participation in the organization of work. Sixteen per cent felt that wages should be variable without participation by workers, and a final 7 per cent, thought that wages should not vary with firm productivity at all. Explanations of labour’s opinion on wage flexibility How much did the links to the political parties matter? The overwhelming level of support found among local union leaders for wage flexibility in principle suggests that we are concerned with a phenomenon that cut broadly across a range of characteristics that we found tended to distinguish the unions. Nevertheless, it is also possible that certain factors served to reinforce (or alternatively to dampen) the principled commitment to democratic compromise, and the sense that the workers played an important role in this field. First, we need to assess the suggestion that the concern with flexibility among union leaders had emerged as a result of the weakening of the unions, including the weakening of their links with the political parties. In fact, we found that there was no direct evidence in our survey data to support such an argument. For example, choice of legal instrument, affiliation to a federation, age of the union, unionization levels or partypolitical ties and preferences had no significant impact on any of the evaluations made before. 24 On the other hand, it is not very clear how far the local union leaders had become significantly less politicized since the 1960s, in the sense of stating a preference for a political party. It is significant that one survey in the early 1960s suggested that 66 per cent of local leaders interviewed had specified their political position by identifing with a political party, compared with 60 per cent in our survey – not a great difference.25 Evidence in favour of the hypothesis that it was a weakening of functional ties between local and national unions that had moderated the former, also seems thin. Indeed, there is already some doubt as to how much party affiliation mattered to the nuts and bolts of local labour relations in the 1960s. For example, in Barrera’s 1969 survey of 230 local union presidents, only 1 per cent said their long-term objective was to raise the political awareness of workers.26 And Angell noted, with regard to labour relations in the 1960s, how, ‘Chilean trade union leaders seem able to make a distinction between their long-term political beliefs and

196 Social Citizenship and Chile’s Labour Reforms

their actual conduct of the day-to-day union affairs.’27 Our own survey suggested that day-to-day labour relations continued in the early 1990s to be basically independent of the political affiliation of unions. Even where the functional link between local unions and political parties was most tangible – that is, in the federations at the mid-level – there was broad agreement, and regret, among sectoral leaders that local unionists had little interest in political parties.28 In summary, if we could find no direct association between the support for political parties and the acceptance of wage flexibility, and if the relationship between the local unions and the parties had remained distant, then it is likely that it was a different type of historical learning that had influenced the unions. An experience similar to (though not dependent on) that of the political parties was of having to struggle to regain the democratic regime, a concern which we have seen to be very prominent in the mindset of labour leaders. In fact, the continuing isolation of the local leader in Chile in the task of defining labour relations no doubt sharpened the coincidence in the firm of the struggle for rights to economic and other forms of security as well as for democratic rights. Local unionists could not afford to sacrifice one for the sake of the other. Indeed, this situation makes it plausible that the experience of labour management would have an effect on the opinions of local leaders. But how did this influence manifest itself? The experience of labour management The argument of neo-liberal theorists regarding the simple economicrational motives of workers could be supported to some extent if it could be shown that it was principally the local union leaders’ experience of hardship that had caused them to accept a discussion of wage flexibility in Chile. It certainly cannot be dismissed that there was a general impact of this nature because we have seen that labour leaders had been in some ways constrained to seek accommodation with employers following the crises years of the early 1980s (see Chapter 4). However, we may need to conceptualize the existence of a threshold of tolerance, above which a certain level of crisis might have made co-operation with employers and wage flexibility seem a necessary and feasible step, but below which such ideas were rejected as unrealizable. The difference between one level and another was perhaps not so much the extent of economic hardship, but the degree to which management allowed for solutions that included workers. Our data appears to support this idea, because it was in the firms where short-term forms of management were practised (such as a high level of group dismissals),

Union Strength, History and Effects 197 Table 9.8 security

Opinions on wage variability crossed with those on occupational

Chi sq.

Opinion on wage variability With union say 77% (176 cases)

Less +2340** Group than 20% dismissals have affected More than 20%

+1662*

Workers have an influence on selection for manpower training

Yes No

Total

Without union say Against 6% 100% 17% (38 cases) (14 cases) (228 cases)

65%

38%

32%

35%

62%

68%

= 100% (89 cases)

= 100% (15 cases)

= 100%

26% 74%

28% 72%

8% 92%

= 100%

= 100%

= 100%

Notes: * Significance level below 0.05; ** Significance level below 0.01. Source: Union survey, author, 1992.

or where it was less likely that workers would have say on aspects of work (manpower training), that labour leaders tended to express doubts about wage flexibility (see Table 9.8). But the evidence that the experience of labour management affected the position of unionists on flexibility issues is even stronger where union leaders’ reflections on experiences of wage flexibility in their own firm is concerned. There was a strong association between the experience of training in the firm and the union leader’s belief that incentive wages could be a long-term form of association with the firm. At the same time, this relationship was not affected by the party association of the union leaders (as Table 9.9 shows), or by their preferences for parties of the centre or left. On the other hand, it is not surprising that a scant knowledge of the wage system applied would bring a sense of disillusionment to a negotiation of wage flexibility. The odds were 3.7:1 that low formalization29 would coincide with a pessimistic view of incentive wages. This association was not affected by either political association or the local leaders’

Type of firm

Odds*

Observed frequencies

Expected odds

Level of training** Low High

Significance of two-way associations Odds

Sig.

Optimistic view of incentive wages***

Firms where the union leader is active within or identifies with a particular political party Firms in which the union leader does not identify with a political party

2.4

3.0

Yes

No

Yes

No

Total

27

50

28

21

100% (107 cases)

17

48

18

17

High training with positive view on incentive wages*

0.008 0.00 2.6

100% (99 cases)

Identifies with a party with high training 1.1

0.65

= 206 cases

Identifies with a party with positive view on incentive wages 1.4

0.70

Notes: * The odds are calculated as follows: Low training and Pessimistic view of incentive wages/Low training and Positive view of incentive wages/ High training and Pessimistic view/High training and Positive view. ** Training defined as low when (> 9.99% of workforce) and as high when (10% of workforce and <). *** Optimistic view as defined in table 9.5, where optimistic view indicates a positive response to the question. ‘Do incentive wages represent a way for workers to participate in the firm in the long-term?’. Source: Union survey, author, 1992.

198

Table 9.9 Odds that a high level of training coincides with the view that incentive wages can be a source of long-term participation in the firm

Union Strength, History and Effects 199

views on national development strategy, which in Chile at the time was the subject of a great deal of contention. The proclivity of employers to allow an actual dialogue with the union (27 per cent of valid cases) 30 was also motivated by the overall principle with which they managed the workforce. We recall that employers in Chile had full discretion on whether or not to discuss such issues with unions.31 In addition, various indicators of the unions’ immediate strength (such as unionization levels and a stronger profile during collective bargaining) seemed to have a positive impact on the likelihood that employers would be open towards a discussion of variable wages (see Table 9.10).32 In one sense, then, it could be said that stronger unions were functional because they forced a necessary debate to take place. In summary, our discussion so far makes it difficult to sustain the simple supposition that it is the influence of political parties that acts to radicalize Chilean unions, even during a period of political transition where, presumably, the importance of political loyalties might increase. In fact, the sources that motivated unionists and employers might easily be the reverse of what we are led to expect from standard neo-liberal theory. The position of unionists on flexibility in their own firm seemed to depend to a large extent on the degree to which employers had invested in long-term relationships. A degree of hardship might have motivated workers to co-operate with employers in the 1980s, but their acceptance of flexibility in exchange did not appear to be reinforced by factors such as fear (for example, fear of job loss) or exclusion (through ad hoc bargaining forms). On the other hand, employers appeared to invest in workers when they were already accustomed to unions – not when the unions did not exist. But what about the traditional industrial relations institutions, such as collective bargaining and sectoral associations? These represented the apex, as it were, of a pyramid of conflict. Is it possible that their operation was also influenced by the time factors and day-to-day relationships that moved the pyramid’s lower tier?

The traditional industrial relations institutions Labour management and collective bargaining There is an alternative to the perspective on bargaining that sees the levels of wage adjustment as the principal mark of achievement. It could be argued that bargaining provides an opportunity to strengthen the sense of fairness used to build a long-term context of work, and it may be supposed that the prospect of formal bargaining itself may be an

200

Table 9.10 Employers’ response to union’s request for dialogue on wage flexibility Employers’ stance

Chi sq.

Rejecting

Responsive

−0.1948**

The union has had a say in the organization of the wage incentives

Yes, in collective bargaining Yes, outside collective bargaining No say at all

33% 16% 51% (142 cases)

50% 20% 30% (54 cases)

−0.1517

Level of unionization

Less than 40% More than 40%

37% 63% (145 cases)

54% 46% (54 cases)

19% 18%

35% 18%

29%

25%

34%

22%

(146 cases)

(54 cases)

1. 2. −0.1781*

Level of union authority

3.

4.

None of the options below. At least 40% of workers in the firm are unionised. (2) and union leader does not think that workers are too weak to sustain a strike action. (2) and (3) and union participated in the negotiation of flexibility in the 1980s.

Notes: * Significance level below 0.05; ** Significance level below 0.01. Source: Union survey, author, 1992.

Union Strength, History and Effects 201

important source of authority for workers which informs the day-to-day relations with management (see Chapter 6). It is interesting that, of the four cases studied in detail, it was in the firm where the system of wages was most transparent that variable wages in exchange for job stability had been agreed to, and where bargaining had explicitly involved the theme of incentive wages. Indeed, since the 1979 Labour Code had given employers full discretion on disclosure as well as the right to veto bargaining themes (except wage adjustment), it was probably the case that availability of information represented a good indicator of the general principles that informed the representation of workers in the firms. 33 In fact, we found that the strongest associations with availability of information were to be found in the aspects of labour management that described how far a longer-term association with the firm was promoted among employees (see Table 9.11). Thus, of those that said that they ‘nearly always’ had enough information, 60 per cent had a WHSC. Where there was ‘never’ enough information, only 19 per cent had a WHSC.34 This is an important finding, because it suggests that the quality of collective bargaining in Chile was indeed powerfully influenced by non-union forms of worker involvement (although the length or legacy of union presence affected these indirectly). 35 Finally, we can see from Tables 8.12 and 8.13 that a union’s embeddedness in a firm was closely associated with the time-horizon of labour management overall. The association between a WHSC and information for bargaining was particularly negative where, respectively, training was low and labour turnover high. Table 9.11 Availability of information for collective bargaining crossed with existence of Workers’ Health and Safety Committee Chi sq.

**.3240 Existence of Yes No Workers’ Health and Safety Committee

Never Always Sometimes Very rarely 31% 40% 6% 23% (52 cases) (14 cases) (91 cases) (72 cases)

25% 60% 40%

7% 37% 63%

38% 33% 67%

Notes: * Significance level below 0.05; ** Significance level below 0.01. Source: Union survey, author, 1992.

30% 19% 81%

Total 100% (229 cases)

202 Social Citizenship and Chile’s Labour Reforms Table 9.12 Existence of Workers’ Health and Safety Committee (WHSC) and availability of information for collective bargaining in the scenarios of low and high levels of workers’ training Existence of WHSC Yes No Availability of information

Low training High training

Yes

No

Yes

No

7% 29%

14% 27%

14% 11%

65% 33%

Total number of cases Source:

(138 cases) (86 cases) (224 cases)

Union survey, author, 1992.

In summary, the discussion in this section has reaffirmed the importance of issues in the area of developmental freedom to the formation of contracts inside industrial firms. At the same time it has underlined a point we made earlier, namely that the innovative aspects of the labour reforms of the 1970s and 1980s – for example, the support for the nonunion forms of involvement such as the WHSC – would tend to make little sense if the unions’ independence was not strengthened in other ways – for example, by granting access to information about the firm. In fact, this aspect of powerlessness was a strong motivating factor behind the local unions’ link with organizations at the sectoral level.

Table 9.13 Existence of Workers’ Health and Safety Committee (WHSC) and availability of information in the scenarios of low and high levels of labour turnover Existence of WHSC Yes No Availability of information

High turnover Low turnover

Yes

No

Yes

No

11% 19%

14% 12%

18% 30%

57% 39%

Total number of cases Source:

Union survey, author, 1992.

(139 cases) (87 cases) (226 cases)

Union Strength, History and Effects 203

The sources motivating association with federations Political parties When neo-liberals in Chile speak about the destructive impact on the unions of the political parties, it is particularly the prospect of sectoral bargaining that they have in mind (see Chapters 5 and 6).36 But the assumption that it was the association with political parties that inspired the local unions’ attachment to federations, in 1992, has to be qualified, at the very least. It is true that the last years of military rule and the transition period saw a great upswing in the affiliation of local unions to federations: 37 per cent of unions in our survey had affiliated after 1987, and only around a third (another 37 per cent) were not affiliated at all. However, only 1 per cent of leaders interviewed cited political affiliation of the federation as the reason behind their choice; 14 per cent said they had received help from a particular federation; and a further 14 per cent that their particular association was the ‘most practical’. Finally, 11 per cent had sought membership of ‘the strongest’ federation, while 9 per cent acknowledged that their choice was ‘the only one that we knew’. Labour management and long-term investment If it was not association with, or sympathies for, political parties that inspired the links with federations, what motivated the local leaders’ views on intervention by federations? Overall, we found that 59 per cent of the local leaders were supportive of obligatory negotiation by sector, whereas 41 per cent expressed either an outright rejection, or some hesitancy (19 per cent). Hence, local leaders certainly seemed able to identify with the terms and issues of the national debate. And yet the results of our survey also suggest that experiences within the firm were the most significant motivating factors behind the labour leaders’ stance. For example, the odds were 3.4:1 that union leaders who believed they did not have sufficient information for local bargaining would support obligatory negotiation by sector. But it may be more important that among the facets of labour relations inside the firms that influenced the perceived need for sectoral bargaining, it was not those that had an immediate political relevance, such as the degree of information, that were the most powerful. It tended to be the extent of long-term involvement that particularly mattered for the perception of union leaders. This is clear when we look at the combined influence of the levels of turnover and training on their support for obligatory supra-firm bargaining. Table 9.14 shows that in the group of unions where the most long-term combination obtained, low turnover

204 Social Citizenship and Chile’s Labour Reforms Table 9.14 Union leaders’ support for making obligatory the response to union petitions for bargaining above the firm-level Number of cases

Unions with a high level of training* for the workforce and (a) a low level of turnover (b) a high level of turnover** Unions with a low level of training for the workforce and (a) a low level of turnover (b) a high level of turnover Total

Support for obligatory bargaining by sector Yes

No

55 29

54% 62%

46% 38%

58 63

79% 98%

21% 2%

205

* Training defined as low (> 9.99% of workforce), and as high (10% of workforce and <). ** Turnover defined as low when (0–30%), and as high when (30.01% <). Source: Union survey, author, 1992.

and high training, 46 per cent did not support this bargaining mode. On the other hand, where the most short-term combination obtained, only 2 per cent did not. These associations are really rather significant. They indicate at least that the time-horizon of labour management influenced the cultural context of work in a forceful way, and that workers’ demands for more inflexible forms of contract were strongly contingent on the level and forms of occupational citizenship that had already been offered inside Chilean firms. Finally, there is the question of whether the federation in fact had an influence within the local firms, and what its direction was. For example, did it make a dialogue about flexibility inside the firm less likely? The impact of federations on labour management We found that although 86 per cent of affiliated unions participated in the federation’s meetings, only 8 per cent answered affirmatively when it came to help in solving problems in the area of working conditions. And yet it was possible to find at least some agreement between the various unions. It is particularly relevant to investigate the subject of worker training, because it is one of the few areas within which there had, in fact, been attempts to set up a policy between organizations of employers and labour leaders at the sectoral level at the time of transition. For example, we could compare the influence of federations inside the firm

Union Strength, History and Effects 205

in two groups of manufacturing activities where an equal amount of training was taking place, but where it was only within one (that of metallurgy) that there had been an accord at the sectoral level on worker training. The other cluster of sectors compared grouped activities connected with the new export areas, such as the preparation of wood and of food products such as fruit, fish and vegetables. We found that it was far more likely in the metallurgical sector that there would be a coincidence between the federation’s involvement in local bargaining (advice) and the situation where the local union would have a say on training. In an odds analysis, the difference gave odds of 6.4:1 (see Table 9.15). A possible explanation could be that there was a greater level of receptiveness on the part of employers in the area of metallurgy as they had seen that a dialogue on training with the unions had worked at the sectoral level (Chapter 5). This suggests the influence of professional associations need not be conceived of in the traditional sense, where federations dictate minute details of labour management and wage levels that might not be compatible with the formation of long-term strategies as well as short-term adjustment within individual firms.

Conclusion We can draw three interrelated conclusions from this chapter: about what recent Chilean labour history reveals about union analysis; about the implication of the Chilean case for neo-liberal theory on the universal role of unions; and finally about the usefulness of studying social institutions during such periods where fundamental changes in the regimes of national governance take place. The development of a ‘quiet’ role for unions in Chile with continued relevance should not be associated with the constraints imposed on unions by the 1979 labour reforms. The possibility of bringing out this role in our analysis has instead entailed a revision of the way in which unions are usually studied. We had to allow for the subtler operation of unions and for the possibility that this was itself shaped by the broader environment, a factor that neo-liberal theory does not concede and that the mainstream of union studies often fails to consider. Given the nature of Chile’s industrial relations, the strategies of employers were inevitably a key feature of this environment. It is this revision of method that has allowed us to conclude that in Chile it was primarily the established unions that played a functional role in terms of extending the time-horizon of labour management, and hence had ensured that labour flexibility became less exploitative. We can refer to their role as a deepening of

206

Table 9.15

Odds that influence from the federation (advice on bargaining) coincides with union opinions regarding workers’ training

Type of firms

Odds*

Obs. frequencies (%) (no. of cases) Influence from union federation*** Yes

Expected odds Significance of probit models Sig.

No Union have a say on training

Yes

No

Yes

No

Total

Unions in areas of new export industry** and no experience of sectoral pact-making

1.0

10

40

10

40

100% (60) cases

Unions in an area of established industry (metallugical sector) plus experience of sectoral pact-making on training

6.4

28

25

7

40

100% (57) cases

The relationship between outside influence and union’s say on training will vary depending on sector

7.3–3

0.06

Notes: * The odds were calculated on the basis of observed frequencies as follows: No advice from federation and No say on training/No advice from federation and A say on training/Advice from federation and No say on training/Advice from federation and A say on training. ** Wood, fish products, fruit and vegetable products. *** Influence from the union federation was defined as providing advice in connection with specific rounds of collective bargaining in the individual firms. Source: Union survey, author, 1992.

Union Strength, History and Effects 207

citizenship in this case precisely because they had done so without losing sight of the universal aspect of rights, at least within the confines of their operation (the individual firm), a capacity which was itself a product of historical features of the organization of unions in Chile. The fact that they had done so at the expense of their own autonomy must also be considered as a consequence of the historical aspects of industrial relations, which dictatorship had served to deepen, including the scope for authoritrianism in the firm, given the personalized form of labour relations. The strongest indication we found that the long-term (preceding) building of voice in the firm had a positive effect on developmental freedom – that is, on the incipient expansion of occupational citizenship – was the influence of union age. In fact, we found that the influence of age superseded that of legal protection, a factor that is much closer to the type of quantitative statistic usually employed to predict union effects. The point to note here is that, despite the apparent similarities with other cases where the market has been sought as a means of disciplining adversarial labour relations (for example, in Britain), the firm tradition in Chile prior to 1970 was in fact so totally distinct as to render comparison illusive. Scholars from different perspectives emphasize the British tradition of adversarial relations,37 including the existence of a conspiracy against reform throughout the twentieth century as all major actors agreed that the forces of capitalism had to be allowed to work without constraints. In pre-coup Chile there was also an escape from social compromise of a kind, but in this case escape was possible partly because the key forces, including employers, adhered to the legal-democratic institutions (variously parliament, the bureaucracy and labour legislation) as the natural avenues for the resolution of conflict. Hence, while social compromise was not autonomous, as in the Scandinavian case, all agents up to the early 1970s accepted as a fact of life legally enforceable interventions in the operations of economic exchange. Employers might not have been happy about the fact, but they were still used to working with single-unionism in Chile, recognized in law since 1925, and with enforceable collective contracts, two crucial aspects of labour law that never became established in Britain. 38 Even Pinochet’s new Labour Code sanctioned the use of legally binding collective contracts. It could be said, then, that there were at least two competing traditions present in Chile’s unionized firms by the early 1990s. One had come into its own during the 1970s and 1980s and impressed upon employers that labour was the most expendable component in the firm, while the other was the paternalistic tradition, with its elements of both authoritarianism

208 Social Citizenship and Chile’s Labour Reforms

and reform. The survival of the latter, despite the former, suggests that, perhaps, the neo-liberals ‘got away with it’ in the Chilean case not because of the thoroughness of the reforms, but in spite of them, and indeed because of the long tradition of firm-level bargaining there. There could hardly be a greater affirmation that the labour market works as a social institution than the Chilean case. In fact, these features of continuity may help to explain how there was never a rank and file rebellion during Chile’s transition, despite the harsh conditions of dictatorship that should have made this a highly plausible case.39 But this raises a final question of how far the return of democracy ameliorated the new liberal tendency in labour management and expanded the reformist one. In one sense, the data presented here supports Valenzuela’s general argument, because in our analysis it was the unions that had experienced the most direct consequences of political ‘closure’ (for example, the new unions, and particularly those with conventions) and/or of exclusive forms of management (‘social harshness’), which were the most likely to develop the basic undercurrents of union radicalism.40 Their leaders had also been more likely than any other group of leaders during the democratic transition to face rejection by employers when trying to establish a dialogue on the shape of flexibility in their firm. As we saw in Chapter 6, bargaining groups continued to increase in importance during the 1990s at the local level, but in the national legal debate (the retabling of the labour reform in 2000), their existence had virtually ceased to be questioned. The transition caretakers had revealed an awareness of the constraints upon them, by insisting that it was the role of the local union to influence the firms’ decisions on labour management and human resources. And yet our analysis has suggested that there was a clear incongruence between this expectation and the degree to which the political élite had in fact been able to provide the unions with the means to perform this role. On the other hand, the transitional élite had been able to rely on the strong sense of duty that local leaders felt towards the democratic regime, by appealing to the myth that the local unions had a special task to perform. This myth, towards which all actors contributed, was highly successful as an element of social control. The question remained of how far it had also been successful in reducing insecurities and promoting the investment in workers inside Chilean firms.

Conclusion

Social and political democratization The relationship between social and political democratization is complex. As the analyses of O’Donnell and Valenzuela have shown, consolidating political democracy is a delicate process in its own right. 1 In this study, however, we have been concerned to show how the stakes involved during the inception of political democracy may also affect the prospects for social democratization. In particular, we have highlighted in the Chilean case how it was quite feasible to progress rapidly through what Valenzuela calls ‘precedent-setting political confrontations’ 2 – that is, confrontations that contribute to a process of consolidation, and at the same time set clear limits to future efforts at social democratization. In particular, we have argued that the relationship between democracy and development in Chile has been coloured by the force of the neo-liberal movement of the 1980s, and the overriding concern with the separation between public and private spheres. Hence, the case of occupational citizenship presents a challenge in the Chilean context largely because of the contradiction that appears to exist between the nature of human development, which requires time and deliberate efforts, on the one hand, and the generation of short-term relationships and withdrawal of planning in public life, which was a legacy of the neo-liberal phase. In order to emphasize the feasibility of labour reform, we selected labour issues that are relevant to the contemporary context. This context has been shaped by the struggle towards the flexibilization of labour. Hence, we have not been concerned with the classical issues of ownership of physical capital or directly productive assets, or of income distribution for immediate consumption. Instead, we have focused mainly on questions of the ownership of human assets, and questions of the distribution of opportunities for personal development (or developmental freedom) over time. The rationale for doing so was our reasoning that these (and perhaps more so than the classical) issues of property and redistribution are relevant to the extension of labour freedom, to 209

210 Social Citizenship and Chile’s Labour Reforms

citizenship and to the development of human resources in the present scenario. We have used the term occupational citizenship to describe the universal nature of problems like these. We highlighted two sets of factors as influential in the development of occupational citizenship. The first was the existence of an economic climate and institutional infrastructure suitable for the development and policing of long-term occupational rights. The second was the approach to economic governance by the key actors, particularly how far this was conducive to the universal application of rights, and which was orientated towards longer-term goals. We have shown how the first set of conditions were, in a very large measure, uniquely fulfilled in Chile in the early 1990s. Indeed, we emphasized how problems of application might have been overcome if the second set of political conditions noted above had been fulfilled to a similar degree as the first. Elements of inertia or lock-in constraining institutional change in labour markets were intervoven with political factors. Hence, had it not been for our account of the unity and diligence of entrepreneurial leaders in reproducing the minimalist values of public authority first promoted in the 1970s with such vigour, it would be difficult to comprehend the real limits of the nascent democracy as the country emerged from authoritarian rule. The transitional Cabinet did take important and bold initiatives. But the citations in this book suggest that the political élite was concerned to respect the tolerance level of the private sector, not because of ‘neutral’ economic effects of intervention, but because of their political effects within an ideologically-driven business community and the negative international signals a dissatisfied business class would send out. The important exception to this rule was the case of the metallurgical sector. A range of activities developed in almost parallel fashion. They included pooling investments for worker training, establishing secure qualifications, respecting new professions and involving individual firms in apprenticeship duties. They also covered the sharing of employment information, the active integration of all the sector’s union federations, and the establishment of forms of unemployment and social insurance. As individual initiatives these were all separate. However, in their causal structure they were linked. Each initiative made another appear feasible. This goes to show how the establishment of forms of occupational citizenship is a complex and, above all, a dynamic process; a process that involved strong elements of cumulative causation. The labour reforms of the early 1990s in Chile had more than a passing significance. They represented the political confirmation of the kinds of practices in the social and economic spheres (that is, the spheres ‘out-

Conclusion

211

side’ and ‘below’ national government and public policy) that would be compatible with a polity worthy of the name ‘democratic’.3 The political opposition claimed their legitimate right to oppose any further revision of Chile’s institutions of labour. How far this was taken seriously became evident during the ensuing Frei Government when the right confidently blocked legal revisions put in place during 1995. By 1999 it had become abundantly clear that the business sector in Chile had had a significant impact on the policies presented by the political élite, independently of any military threat of return to authoritarian rule, and independently of the right. Policy-makers may have little choice in the wake of authoritarian regimes in terms of where they start the democratization process, but where and how they start may make a difference. In the Chilean case, consolidation of political institutions and economic stability were, perhaps inevitably, the overriding concerns. Slowing down the democratization of economic and human development was one of the costs.

Labour institutions and economic development Because of the wave of transitions to democracy in Latin America during the 1980s, the concern of scholars has naturally focused on political democratization. The relationship with economic development problems is more difficult to deduce. Our task has not been an easy one, because it is more complicated to show the political and social links between forms of public authority and economic development in the liberal polity. In neo-classical theory, such links are not supposed to exist, or they are very weak. Moreover, they are not readily observable, as in the case of social democratic or developmental states. In Chapter 2 we saw that even the most outstanding recent works focusing on institutions in the area of human resources, have in fact not given much credit to such institutions as autonomous factors. In his 1993 volume, for example, Schultz dismissed the approach (which he referred to as ‘institutional economics’), in which ‘institutions are treated as an exogenous variable in the sense that they are altered by political acts. 4 From his perspective it is primarily when technological changes ‘appear’ (in this case fertilizer) that a demand for new institutions arises. But we have argued that the ‘political’ decision to enforce radical market institutions may have affected the direction of economic development in Chile. There was the case of the private training market, which provided certain incentives favouring management training. More importantly, perhaps, both economic and institutional incentives were provided for firms to use labour as an expendable element. That institutional

212 Social Citizenship and Chile’s Labour Reforms

incentives were important independently of the ‘economic cycle’ seemed to be corroborated by the fact that rates of turnover and worker training inside Chilean firms were strongly related to social institutional factors. For example, they clearly interacted with the legacy of unionism and labour relations. Therefore, it would be too simplistic to say that merely lowering the unemployment rate in Chile would be sufficient to increase the bargaining power of workers. If entrepreneurs were already relying on unskilled labour to a large extent, or if skilled labourers continued to be isolated by the lack of certified markets in skills, then the effects of tighter labour markets would be reduced. If the use of precarious contractual forms had become part and parcel of waged employment, then a strong social buffer would have been erected which, in effect, protected employers from the market effects of the ‘labour market’. On the other hand, the question would have to be asked whether we can safely predict labour’s ‘economic’ behaviour if pockets of industrial workers were to attain a significantly higher level of bargaining power because of the ‘labour market’ effect. It is not inconceivable that a stronger force (than the labour market) in establishing shades of labour monopolies in Chile was simply past organizationbuilding and institutional memory. The case of the bakery sector is particularly illustrative, as it concentrates a stratum of relatively low-skill labour. In the 1980s it was not labour market pressures that made entrepreneurs participate in forms of regulation with labour leaders at the sectoral level, but the fact that past practice had made it clear to both parties that the benefits of regulation outweighed the costs. Then, just as Franzosi asks for management statistics that can supplement the understanding of strikes, so the study of human resources, the degree to which they are valued and developed, begs for a new type of data. Perhaps it is because of their absence that, when we talk about national labour institutions, we tend to worry about whether or not they promote economic disruption – and not, for example, about how far they may help to promote responsible economic behaviour. The idea of association between social institutions and the organization of work presents a methodological challenge to the new liberal perspectives on labour markets. The case of Chile raises the stakes in the debate, because managers had virtually a free hand (free from distortions imposed by unions and other forms of regulation of labour management). In this context, why would the level of dismissals and pay variability over time be associated strongly with the degree to which labour management was formalized? Why would turnover be associated with the existence or not of a formal wage system? And why would training be linked to levels

Conclusion

213

of formalization and to the extent of participation and representation of workers? A likely answer is that these associations existed because the firm is a social entity, composed of relationships. These relationships in turn depend on assessments about past conduct and expectations of the future, not just on static signals coming from the ‘outside’. On the other hand, it has been presumed until very recently that the presence and particular role of unions can be isolated as factors that intervene from the outside. 5 This raises difficult methodological issues. A problem with all approaches that assume causality in labour relations is that it is inherently difficult to meet the condition of appropriate time-order.6 Do particular social institutions come about only because of the nature of employer strategies, excluding the possibility that the relationship also works the other way round? In turn, are employer strategies always determined by ‘market’ (neutral) signals?7 We have suggested that the answer must be negative to both of these questions. The strength of the associations (between formality and the time-horizon of labour management and between different dimensions of time-horizon) in our data, is supportive of this supposition. In the case of the association between training and turnover, for example, it is plausible both that a low level of prior training would make high turnover less costly, and that the existence of a regime, or expectation, of high turnover would make investment in training a risky and hence unlikely investment (partly because, as argued above, of the absence of formalized markets in skills). In this case, the relationship is likely to be self-sustaining, rather than a one-way causal one. In another phrase we have used, it could be seen as a ‘cycle’, either virtuous or vicious. 8 The possibility also should not be discarded that there are ‘internal’ causes for firms to choose particular forms of legal intervention. For example, we saw that the association between contracts and labour management varied strongly depending on union age. Again, legal rules mean different things for different governance structures. Firms not only adopt, but actively appropriate, the legal formula that suits them at the time when bargaining relationships begin to take root. This suggests that the focus on governance structures over time still cannot be dismissed, even when discussing formal bargaining tools. There were some areas of the analysis, however, where it made relatively more sense to speak of a particular time order between two variables and/or factors. It is likely that these factors could act independently to increase the intensity of the virtuous or vicious cycles. Let us consider sector and size first. 9 In Chapter 8 we saw that the association between time-horizon (turnover or training) and the social factors in

214 Social Citizenship and Chile’s Labour Reforms

question was statistically significant in both small and big firms, and in low technology sectors versus all other sectors. At the same time, however, the associations in question were consistently more intense depending on size (in the case of training) and sector (in the case of turnover). Let us consider the case of union age. The study of union presence over time is at least a possible alternative way of analyzing possible union ‘effects’. For example, might the presence of unions over time (coupled with their insertions in very different industrial relations contexts) have made a difference to the kinds of social relationships, or the cycles (vicious or virtuous), that became dominant in the firms? First, we know that with the exception of seven cases (less than 3 per cent of the sample), the firms in which unions were interviewed in autumn 1992 had been established prior to 1973. Moreover, only four unions were formed between 1973 and 1980 (after which year most activities were banned). This means that we have a long period over which to assess the differences between firms that had had a union established prior to dictatorship and those in which unions were only established sometime after the new Labour Code came into effect in 1980. In the light of this, the argument was hardly convincing that the nature of labour management in firms with older unions was more likely to be, for example, long-term, because the firms themselves were older and had therefore been able to develop over a much longer period of time, or that, conceivably, they were strong organizations because they had managed to survive through the 1970s and the years of crisis. A possible explanation here might be that firms that did not have unions before 1973 had excluded them overwhelmingly because they were authoritarian from the outset. This, however, makes it difficult to explain why they should tolerate a union being set up after 1979 when it was much easier for employers to prevent this from happening than it had been prior to 1973. The possibility that the toleration of unions prior to the dictatorship could have been because they were already economically more successful also cannot be discounted. However, it is unlikely that such differences from over two decades prior to the interview would still be the explanation for differences in labour management in the early 1990s. Following the economic success explanation as a reason for accepting unions, we would have to assume that it applies even more after dictatorship when, again, employers had much greater chance of avoiding unions. If this economic argument was correct, then the newunion firms, more so than the old-union firms, should have had good labour management, but we saw that the reverse was the case, both in the economic sphere (turnover and training) and in terms of the resolution

Conclusion

215

of conflicts. Hence the new unions were not consistently set up as a consequence of good labour relations emerging as a product of economic success. It is more likely that their influence was weaker because they were incipient and emerged in an unfavourable legal and political context. Suffice it to say that we hope we have illuminated the sturdiness and capacity for survival of the civil society of labour in Chile, both employers’ social memory and the institutional memory of unions, during a period when it has commonly been supposed that unions were effectively repressed. The analysis of unions has emphasized the revival of national union leadership during the second half of the 1970s and, of course, the 1980s.10 In respect to the local unions, a similar revival has understandably been associated with the aftermath of the 1979 Labour Code. But the unions that were created after this period were in many ways weaker than were those that had been forced to survive the most repressive period of Chile’s dictatorship. At least we are reminded of the force of latent relationships (the importance of which may reappear later) and the aspect of time in the study of both union effects and human resources. Finally, we have learnt that the maintenance of a latent influence of Chilean unions also had its costs. Indeed, it appears that the freedom gained for workers through the long-term presence of local unions was only partial – that is, it had also retained the unions within a certain informal relationship of authority with the employer; one, perhaps, that only a fuller (that is, a trans-firm) form of occupational citizenship could radically transform. In particular, the experience of dictatorship served to accentuate the paternalistic–authoritarian aspects of participation, which was in all likelihood a legacy with long historical roots in Chile’s industrial history.11 On the other hand, local unions had not ceased to be essentially independent (see Chapter 8), partly because of the very constraints exerted on them by management.12 In his study of latent social democracy in late-nineteenth-century Chile, Valenzuela cites the critically moderate tone of union and social democratic discourse – the idea that unions would also be a ‘factor of order’ in the enterprise, and not merely a source of disruption.13 It is interesting how far this tradition survived both as a problem (because of the absence of real social compromise that fed it), and as a renewable source of constructive pressure on labour management. Whether the unions would be able to expand this pressure to change the shape of labour’s professional rights and the nature of skill development in Chile remains open to question. But what seems probable is that the transition to electoral democracy in Chile had not only served to reopen this task, but also to consolidate many of the new obstacles developed in the liberal decades.

Appendix 1: Tables to Accompany Chapter 1 Table A.1.1 Some trade-offs associated with the development of social citizenship under different forms of national governance. Static and evolutionary advantages and disadvantages – broken down according to whether they are broadly positive or negative for the development of social citizenship Forms of National Governance@

Characteristics of Social Citizenship 216

a. evolutionary

Traditional Western Democracies

Developmental State^

b. static

Shared characteristics

Singapore, Rep. of Korea Positive

a. * Late development pushes state to prioritise industrialisation and associated employment goods.

* Representative groups that form more likely to be democratic in longstanding political democracies

Welfare States^^ (continental European)

Laissez Faire Regimes

Scandinavia, Germany

United States, Britain (post-1979), Chile (post-1990)

* Representative groups more likely to be able to develop unified and socially cohesive programmes which are required to turn social goods into social citizenship.

* Same as under political democracy generally.

* Political stability favours larger social investment schemes and institution-building.

* The resulting citizenship * Social citizenship rights will usually be rights more likely broadly accepted by to have proper social groups (including constitutional protection. employers).

* Business sector often made subordinate to state (national) goals. b. * Income disparaties in firm and in society at large relatively low.

* The individual enjoys statutory rights.

* Training pushed by state and companies.

* The individual has formal access to appeal and/or protest about eligibility and so on.

* Unemployment low: employment security or industrial dynamism. * Savings systems: long-term household savings, social investment funds. * State promotes long-term employment. From productivity wage guide-lines to help for part-time employees.

* Easier for representative groups to apply political pressure openly for the deepening or change of citizenship rights, by protest and other means.

* Rights to unemployment insurance on individual basis (not tied to family situation, for example). * Rights to free child care for mothers, protecting occupational continuity for women.

*–”–

Etc.

217

(contd.)

218

Table A.1.1

Forms of National Governance@

Characteristics of Social Citizenship a. evolutionary

Traditional Western Democracies

Developmental State^

b. static

Shared characteristics

Singapore, Rep. of Korea Negative

a. * It may be difficult for social organisations to develop autonomous and democratic internal structures.

* Political parties may become subsumed within the state.

Welfare States^^ (continental European)

Laissez Faire Regimes

Scandinavia, Germany

United States, Britain (post-1979), Chile (post-1990)

* Social groups may * Compared with become too close to developmental state parties, and monopolise (though not always): weak ability to subordinate social representation business to national goals, within their segments.^^^^ particularly to accept national social agreements.^^^

* Turn to the type of political democracy.

* Repetitive policy reversals/swings due to absence of hegemonic policy regime [note 1]. Makes it difficult to develop and consolidate social rights, particularly of a universal or near-on universal character (which is required for it to be said that occupational citizenship exists).

b. * Limits on or lack of political citizenship.

* Turn to type of political democracy.

* Long-term state assistance and exclusion from work may affect large minority.^^^^

* Substance of stateengineered or de facto state-protected rights (i.e. citizenship rights) will be very limited at any one time.

* Employment and social goods often contingent on suppression of individual rights. * Social goods promoted by the state, typically in association with private sector practices, often not as formally inclusive, and protected within system of statutory rights and appeal, as under democratic forms of citizenship.

219

@ Intermediary cases: France, Italy, Britain (pre-1979), Chile (pre-1970). ^ The special case of Japan (not considered) shows that the developmental state can co-exist with a long-standing practice of political democracy, although as a type of political democracy it should be bracketed within the one-party type for most of the postwar period, along with, for example, Sweden. ^^ In addition to the Scandinavian countries, Germany could be included here on the grounds of the extensive coverage of social agreements similar to those characterising the social democratic regime in Scandinavia as well as the influence of such agreements on public policy and regulation of the labour market. See Maurice, Sellier and Silvestre (1986). Streeck (1989). ^^^ This is less likely to happen in the social democratic states because of the generally strong influence of unions and employer-union social pact-making. But it is relevant for particular periods in the ‘intermediary’ cases, such as Britain pre-Thatcher as well as France and Italy. ^^^^ See literature defending corporatism cited on p. 19 for a qualification of these problems.

220 Appendix 1 Table A.1.2 Number in Figure 1.1

1 2 3 4 5 6 7 8 9 10 11 12 13

Values used for box-plot in Figure 1.1 (Chapter 1) Countries (all periods ending 1995)1

Denmark (post-war) Sweden (post-war) Germany (post-war) France (post-war) Italy (post-war) Spain (1975 onwards) USA (post-war) Japan (post-war) South Korea (1961–1980s) Singapore (1970 on) Chile (post-war – 1970) Chile (1973–1989) Chile (1990 onwards)

Values on the three dimensions in box-plot x

y

z

Role of the state in economic governance*

Role of interest associations in economic governance**

Nature of political regime***

8 10 9 9 8 10 4 15 17 18 11 3 4

17 18 15 12 12 14 3 8 12 62 9 2 6

12 11 13 16 18 12 15 11 10 9 17 2 16

Notes: * As explained in Chapter 1, this refers to levels (from low to high) of the state’s strategic role in industrial development, following the analysis of Johnson (1982, 1987) and Evans (1995), specifically adapting Johnson (1985), p. 19. ** This refers to the degree of involvement (from low to high) of autonomous employer associations and labour unions in economic governance, including in the shaping of public policy and of labour market institutions. *** As explained in Chapter 1, this categorization is not perfectly associated with ‘degrees’, but runs though a series of regime types roughly associated with levels of political competition for office. All values above 10 are democratic regimes according to the definition in Chapter 1. The regime types draw on, among others, Sartori (1976), Linz (1978), Valenzuela (1988, 1992), Mainwaring et al. (1992). It should be noted that the classification should not be seen as being too rigid, and is open to reinterpretation. With this caution in mind, we can list the following regime types and their corresponding values: Values 16–20: Polarised pluralist (such as Chile pre-1973 (17) and post-war Italy (18)); Values 14–15: Bipolar configurations (such as USA (15), and France and Chile post-1990 (16), with France and Chile accorded higher values because of the greater extent of party competition; Values 10–13: Other forms of democratic regime, including 12–13 = ‘system hegemonic’ (for example, persistence of social democratic coalition governments (12), or centrist governments (13)); Value 11 = persistence of one party in power through fully competitive elections or ‘one-party hegemonic’ democratic regimes (such as Sweden or Japan); and Value 10 = for cases that during the period considered have been in a stage of transition between authoritarian and democratic regimes (such as Korea); Values 1–9: various forms of non-democratic government (according to our definition), from mildly exclusionary, such as Singapore (Value 9), because of the absence of real competition for national office, through to harshly authoritarian regimes, such as Pinochet’s Chile (2).

Tables for Chapter 1 221 1. In each country, there was, of course, significant change over the course of the period which is missed in this type of summary. The figure represents the broad tendency or direction in which a country was moving during the period. 2. Singapore is lower than Japan and Korea here, because, while tripartism is far more institutionalized and important in Singapore, there is still a question mark over the autonomy of interest associations compared with the others.

Appendix A: Notes on Survey Sampling In the practical implementation of the Union survey I was able to rely on an institute experienced in producing surveys for academic analysis,1 and to train a group of interviewers in the use of the questionnaire, who would head teams in different regions. The survey was confined to plant-level unions (henceforth, firm unions), excluding inter-firm, transitory and independent unions.2 Firm unions represented 65.6% of Chilean unions. Attention was focused on firm unions in key sectors of Chilean industry (selected at random in selected sectors), to be able to say something more profound and statistically relevant about them. Economic importance was the basic criteria for selection, with size of employment and/or the place of a given activity within dynamic export growth areas being the two key indicators. These would also be the areas where the development of human resources could be seen to be especially important. Firm unions dominate the sectors studied, with 94% of industrial unions being of this type. Choosing a limited selection of sectors made it possible to ensure that unions in each represented at least 13% of the total number of unions in that sector. For all runs with the 236 cases together, the representation of each sector is weighted, so that all are fixed at 16.38%, corresponding to the overall representativeness of the sample. 3 The sectors chosen, and those excluded, are listed in table A.1. See Haagh (1997), pp. 343–5 for a more extended discussion of the sample.

Table A.1

Details related to universe on which sampling is based

Union classifications

No. of unions

Sectors by classification of SITC

No. of Chilean unions (official figures, end 1991)

9858

–

No. of Chilean unions (Labour Ministry records, as at 26 June 1992)*

8490

–

No. of firm unions**

5569

–

No. of industrial unions

2045

–

No. of industrial firm unions***

1922

–

Core universe Selected sectors of industrial firm unions

– 1447

Fruit, vegetable and fishing industries (3113, 3114, 3115) Bakery industry (3117) 222

Notes on Survey Sampling

223

Textiles and clothing industries (321, 322) Wood, furniture, paper and carton industries (331, 332, 341) Chemicals and related industries (35) Metal industries (38) Sectors of industrial firm unions not in sample

475

Meat, milk products, mill products, sugar, drink and tobacco, animal foods (3111, 3112, 3116, 3118, 312, 313) Leather and footwear (323, 324) Printing (342) Glass and cement; iron and steel (basic) (36. 37) Various (39)

Notes: * All subsequent figures are based on the list of firm unions printed on the 26 June 1992. ** These represented 65.6% of all unions at the end of 1991, a figure quite consistent with previous years. Our sample has 22 unions which are not of the firm type: the 22 bakery unions are inter-firm unions. Collective bargaining practices differ for the bakery unions, which negotiate jointly, but on the day-to-day aspects of labour relations considered in this book, the unions’ function is quite similar to that of other unions. *** 94% of industrial unions were firm unions. Source: Labour Directorate, Chile.

Table A.2 Importance of industrial sectors in sample universe, by unionized industrial employment, industrial employment and value added Sector of industry

31

Percentage of total unionized industrial employment

Percentage of total industrial value added*

Percentage of total industrial employment*

Food, drink and tobacco 3113 Fruit and vegetable processing 3114 Fish processing 3117 Bakery

2.5

21 1

33 1

6.9 4.6

2 6

2 22

Textiles, clothing and leather 321 Textiles 322 Clothing

9.9 7.1

14 7 4

20 9 7

33

Wood

9.2

5

10

34

Paper and printing 341 Paper

2.8

7 2

6 1.2

12.8

18

10

2

3

32

35

Chemicals, plastics, petrol and coal

36

Non-metallic minerals

224 Appendix A Table A.2

(contd.)

Sector of industry

37

Percentage of total industrial value added*

Percentage of total unionized industrial employment

Basic metal industry

38

Metal industry

39

Other industry

17.5

3 Total industry Percentage representation in sample universe

73.3

Percentage of total industrial employment*

17

1

15

16

1

1

100 61

100 79.2

Notes: * The column on the left represents the total of the sector concerned. When it is not broken down by data in a column to the right, the sector as a whole is represented in the sample universe. When it is broken down, only the left-hand column is represented in the sample. Sources: Chilean Labour Inspectorate; unpublished list of Chilean unions, 26 June 1992.

Table A.3

Survey sample and weighting formula

Industry sector

Weighting formula: Total universe (1447)/No. cases (237) = 16.38% (the percentage each sector should represent when weighting is applied) No. of Percentage No. of cases when adjusted cases in rep. of to represent sample universe 16.38% of universe

Processing of fruit, vegetables, fish Paper Wood and forestry* Furniture Textiles Clothing manufacture Chemicals Metal Bakery

Percentage Weighting change that used weight implies

24

13

30

25

1.25

11 32 12 32 28 25 51 22

21 13 32 17 21 14 15 25

9 40 6 31 22 30 55 15

18 25 50 3 27 20 8 32

0.82 1.25 0.5 0.97 0.73 1.2 1.08 0.68

Notes: * Weighting formula: total universe (1447)/no. of survey cases (237) = average representation (16.38%). It should be noted that large weight adjustments were only made downwards and in the cases of sectors where a very large number of unions were represented.

Notes on Survey Sampling

225

Table A.4 Levels of workers’ training by sector (according to level of technological development)* Level of technological 0–0.99% of 1–9.99% of 10% and more development* workforce workforce of the workforce High** Middle Low

25.6 47.8 31.8

24.4 31.3 31.8

50.0 20.9 36.4

Totals

100% (90 cases) 100% (67 cases) 100% (88 cases)

Notes: * For simplicity, all the sectors in the survey were divided into three groups, according to an approximate estimate of their level of technological development. The grouping attaches weight to both ‘hard’ factors, such as types of machine, and ‘soft’ factors, such as process of elaboration and organisation of work. ** High: paper products, metals and chemical manufacture; Middle: textiles, furniture and clothing, and basic chemicals; Low: fruit, forestry, wood and fishing industries, copper, and bread production. Source: Survey of Union Presidents (1992).

Table A.5 Levels of turnover by sector (according to level of technological development)* Level of technological development* High** Middle Low

High turnover (%)

Mid-level turnover (%)

Low turnover (%)

Total

40.7 40.0 56.1

31.4 38.5 24.4

27.9 21.5 19.5

100% (86 cases) 100% (65 cases) 100% (82 cases)

Notes: * and ** as Table A.4. Source: Survey of Union Presidents (1992).

Appendix B: Statistics to Accompany Chapters 8 and 9

Table B.1 (To accompany Table 8.7) Significance of the reductions of the original model when progressively controlling for the least significant factors Original and reduced models* Model

1

2

Original

Minus: machilow, dumhil1 3 Minus: machilow, dumhil1, big1 small1 4 Minus: machilow, machihig big1, small1, dumhil1** 5 Minus: machilow, machihig, big1, small, dumhil1, divtarea

Variables remaining

Sum of squares (explained)

Degrees of freedom

Reduction

Sig.

Dumskill, dumform, mt1union, divtarea, machihig, machilow hilo1, hilo3, small1, big1

147.348

10

–

–

147.227

8

0.121

0.99

146.172

6

1.176

0.98

138.992

5

8.356

0.14

126.935

4

20.413

0.000

Notes: * As long as the reduction remains statistically insignificant (above 0.5), the variable(s) in question can be deduced without significantly altering the explanatory power of the model. ** Hence, in this case, Model 4 is the parsimonious model, because taking out further variables would significantly reduce the explanatory power of the model.

226

Statistics for Chapters 8 and 9 227 Table B.2 (To accompany Table 8.12) Significance of the reductions of the original model when progressively controlling for the least significant factors Original and reduced models

Size and significance of reduction

Model

Variables remaining

1

Original

Meas, whsc, r, dumhil1, dumhil3, small1, big1

2 3 4

Minus: dumhil1 Minus: dumhil3 Minus: dumhil1 and dumhil3 Minus: dumhil1, small1 Minus: dumhil1, dumhil3, small1** Minus: dumhil1, dumhil3, small,big1

5 6

7

Notes:

Sum of squares (explained)

Degrees of freedom

Reduction Sig.

70.958

7

68.963 68.795 68.242

6 6 5

1.995 2.163 2.716

0.92 0.91 0.76

65.617

5

5.341

0.33

65.041

4

5.917

0.20

50.240

3

20.718

0.00

–

–

* See text under Table B.1, ** In this table, Model 6 is the parsimonious model.

Table B.3

Logistic regression to explain high levels of workforce training

Variable to be explained (dependent): training of more than 5% of workforce in previous two years.1 Explanatory factors Basic factors Sectors with a higher level of technology Sectors with a lower level of technology Large firms (rather than medium or small) Small firms (rather than medium or large) Union presence Union established after 1979 (rather than before 1973) Labour management Existence of WHSC

B

S.E.

[dumhil1]

−0.6093

0.4240

2.0649

0.1507

[dumhil3]

−0.6314

0.4389

2.0694

0.1503

[big1]

−0.8118

0.4030

4.0568

0.0440

[small1]

0.8779

0.4276

4.2144

0.0401

[duman3]

0.9895

0.4399

5.0590

0.0245

−1.6551

0.3642

20.6560

0.0000

[whsc]

Wald.

Sig.

228 Appendix B Table B.3

(contd.)

Variable to be explained (dependent): training of more than 5% of workforce in previous two years.1 Explanatory factors Low levels of turnover No measurement of individual piece-rate output

[r] [meas]

Constant

B

S.E.

Wald.

Sig.

−0.2580 1.1948

0.1127 0.3939

5.2412 9.2002

0.0221 0.0024

0.4077

0.4543

0.8054

0.3695

Classification table = explained 62.73% for value 0 of training and 82.6% for value 1 of training = overall 74.39% Notes: See Appendix E for regression model and definition of all variables. 1. A single dummy was used here for simplicity, distinguishing between two levels of training, low and high. Standing (1992) also uses variables with just two values in his analysis of training in Malaysia, but distinguishes between presence or absence of training activities. Source: Standing (1992), p. 342.

Table B.4 Logistic regression to explain high levels of workforce training, parsimonious model Variable to be explained (dependent): training of more than 5% of workforce in previous two years Explanatory factors Basic factors Large firms (rather than medium or small) Union presence Union established after 1979 (rather than before 1973) Labour management Existence of WHSC Low levels of turnover Not measurement of individual piece-rate output Constant

B

[big1]

[duman3]

[whsc] [r] [meas]

S.E.

Wald.

Sig.

−1.2238 0.3441

12.6506

0.0004

0.8983 0.4283

4.3999

0.0359

−1.6688 0.3457 −0.2717 0.1054 1.2710 0.3862

23.2987 6.6463 10.8332

0.0000 0.0099 0.0010

0.3597 0.3360

1.1465

0.2843

Classification table = explained 61.39% for value 0 of training and 81.47% for value 1 of training = overall 73.17% Notes:

See Appendix E for regression model and definition of all variables.

Statistics for Chapters 8 and 9 229 Table B.5 (To accompany Table B.6) Significance of the reductions of the original model when progressively controlling for the least significant factors Original and reduced models

Size and significance of reduction

Model

Variables remaining

1

Original

Meas, whsc, r, dumhil1, dumhil3, small1, big1, Duman3

2 3 4

Minus: dumhil1 Minus: dumhil3 Minus: dumhil1 and dumhil3 Minus: dumhil1 dumhil3, small1*** Minus: dumhil1, dumhil3, small1, duman3

5

6

Notes:

Sum of squares Degrees of Reduction (explained) freedom

Sig.

76.376

8

–

–

74.291 74.265 73.640

7 7 6

2.085 2.111 2.736

0.96 0.96 0.83

69.722

5

6.654

0.25

65.041

4

11.335

0.03

* See text under Table B.1. *** In this table, Model 5 is the parsimonious model.

Table B.6 (To accompany Table 8.7) High technology sectors,* levels of turnover by formalization of labour management Variables correlated

Levels of turnover Row % and total no. ( )

Pearson Chi Square: 0.00746

High

Medium

Low

Row total

47 19

30 49

23 32

100 (115) 100 (37)

Dumform (formal wage system)

No Yes

Low technology sectors**, levels of turnover by formalization of labour management Variables correlated

Levels of turnover Row % and total no. ( )

Pearson Chi Square: 0.00023

High

Medium

Low

Row total

68 20

14 55

18 25

100 (62) 100 (20)

Dumform (formal wage system)

No Yes

Notes: * Representing the two first categories listed in Table A.4; ** Representing the last category listed in Table A.4.

230 Appendix B Tables B.7 Levels of training by existence of Workers’ Health and Safety Committees, small and medium-sized firms, with up to 199 workers Variables correlated

Levels of turnover Row % and total no. ( )

Pearson Chi Square: 0.00000 Workers’ Health and Safety Committee

No Yes

High

Low

Row total

19 56

81 44

100 (103) 100 (52)

Levels of training by existence of Workers’ Health and Safety Committees, large firms, with 200+ workers Variables crossed

Levels of training Row % and total no. ( )

Pearson Chi Square: 0.00088 Workers’ Health and Safety Committee

No Yes

High

Low

Row total

45 85

55 15

100 (51) 100 (26)

Appendix C: Notes on Regression Tables Labour turnover and workers’ training A different method is used in each regression: for the regression of turnover, a general factorial analysis of variance is used, since the dependent variable is continuous (and while it is bounded at zero, it has only 8 values at zero and it is not bounded upwards and has a normal distribution), and all independent variables are categorical. To display the t statistics, a linear regression is also used. In the case of training, a logit regression is used, given the fact that the dependent variable takes only the values 0 and 1. All the independent variables are entered in the regression as dummies of either a simple category (0 or 1) or two categories, such as the level of mechanization or size of the firm (the latter whereby ‘big1’ has values 1 = large firms and 0 = all other firms; and ‘small1’ has 1 = small firms and 0 = all other firms). The variables are summarized below.

Variables for analysis of variance for turnover Since the dependent variable (labour turnover as defined below) is a continuous variable, and the independent variables are all categorical variables a factor model of variance was used, whereby: R (turnover) = size of firm + ratio of workers to machine + sector + level of skilled workers in the firm + level of formalization of labour management + practices of task diversity or not + single or multi-union firms. A linear regression was also included in Appendix B, to display the t statistics, whereby: R (turnover) = (beta1 ∗ big1) + (beta2 ∗ small1) + (beta3 ∗ machihig) + etc.

Dependent variable Turnover: continuous variable (R) defined as: (y − x)/y (from low to high levels of turnover), where x = the minimum number of workers that existed at any given time during the previous two years. y = the maximum number of workers that existed at any given movement during the previous two years.

Independent variables (i) Basic firm indicators Ratio of workers to machines: dummy (machilow/machihigh): machilow: 1 = low ratio of workers/machines (between 1 and 40 workers (34%). 0 = other. 231

232 Appendix C machihig: 1 = high ratio of workers/machines (more than 500 workers) (18%). 0 = other. Size of firm: dummy (small1/big1) – small1: 1 = small size (between 1 and 99 workers) (36%); 0 = other. big1: 1 = large size (more than 200 workers) (33%); 0 = other. Sector: dummy (dumhil1/dumhil3) − dumhil1: 1 = high technology sectors (as defined in Appendix A); 0 = other (medium and low). dumhil3: 1 = Low technology sectors (as defined in Appendix A); 0 = other (medium and high).

(ii) Labour management indicators Level of skilled workers in the firm: dummy (dumskill): 1 = high level of skilled workers in the firm (more than 30%) (46%); 0 = low level of skilled workers in the firm (less than 30%) (54%). Practice of moving workers between tasks: dummy (divtarea): 1 = Yes (54%); 0 = No (46%). Level of formalization: dummy (dumform): 1 = high ( .. .) (23%); 0 = no (low. ..) (77%). Level of workers’ bargaining authority: dummy (mt 1 union): 1 = more than one bargaining unit in the firm (union or non-union)* (49%); 0 = only one bargaining unit in the firm (51%). * The non-union forms of bargaining (the ‘negotiating groups’) are discussed in Chapter 8.

Variables for regression on workers’ training Since the dependent variable (workers’ training as defined below) is a categorical variable, probit was used. Logit regression model for regression on training: Logit (p) = log P(x)/1 − p(x) = B[0] + B[1] × [2] + . . . B[p] × [p].

Dependent variable Training: dummy (cap 1z): 0 = high training (more than 5% of the ordinary workforce in the previous two years) (62%); 1 = low training (none or less than 4.99% in the previous two years) (38%) (245 cases).** ** Defined as percentage of workers in the firm who received training in the previous two years.

Notes on Regression Tables

233

Independent variables (i) Basic firm indicators Ratio of workers to machines: dummy (machilow/machihigh) – as above. Size of firm: dummy (small1/big1) – as above.

(ii) Labour management indictors Workers’ Health and Safety Committee: dummy (whsc): 1 = a WHSC operates in the firm (33%); 0 = a WHSC does not operate in the firm (67%). Labour turnover: (R) continuous variable – defined as above (from low to high levels of turnover). Measurement of output system by individual/small group: dummy (meas): 1 = no measurement of individual output (29%); 0 = measurement of individual output (71%).

Appendix D: Notes on Loglinear Tables Each loglinear table is presented as an ordinary cross-tabulation, with the cells of the observed frequencies. In addition, and for simplicity, the relevant odds are extracted as described below. In the right-hand box, the significance of the crosstabulation is interpreted as a logit model, in which the factors in the heading of the tables are taken as the dependent variables, and the factors in the first lefthand column are taken as the independent variables. Finally, the odds and the significance of underlying two-way relationships between the variables is cited. The purpose of these odds is to demonstrate how far any two variables in the analysis are highly correlated, and if they are, whether that correlation becomes the most important underlying association, and thus the association which most strongly ‘explains’ the odds of the cross-tabulation.

Odds The usage of odds is the simplest way to represent the significance of the associations considered. They are extracted by presenting a cross-tabulation, and then subtracting its results in the following way: Training Low

High

90 (/) 85

30 (/) 75

1.059

0.4

= 2.65

The odds in this case are 2.65/1 (2.65 to one) that small firm size will be associated with low training of workforce rather than high.

Significance of two-way odds We present the odds of each two-way association. We also present the level of significance of that two-way association within the three- or four-way model (depending on the number of variables used). Traditionally, 5% is considered as the accepted level of significance; below that, significance increases. However, levels of significance close to 5% (say 6% or 7%) are also worth considering. When none of the two-way associations are at 5% significance or below, then the independent model (three- or four-way) may be considered on its own.

Calculation of significance of two-way odds This is done via a X table. Depending on the degrees of freedom used in each revised two-way model, the Ratio Chi Square will have to be reduced by a certain 234

Notes on Loglinear Tables 235 amount to correspond to the 5% significance level of P (in the last column of the table titled ‘P’). If the degrees of freedom used are 2, for example, the Ratio Chi Square will have to be reduced by 5.991. In the tables, only the Ratio Chi Square and Degress of Freedom (DF) of the independent model are listed, and the calculated percentage significance level of the two-way associations or revised models.

Logit By using logit, one or two factors are taken as a response variable. Multinominal logit models are used to model the relationship between one or more dependent categorical variables and a set of (or one) independent categorical variables. See further, M. Norusis, SPSS Advanced Statistics 6.1, Chicago: SPSS, 1994, p. 209.

Notes

Introduction 1. Linz (1978), O’Donnell (1986), Valenzuela (1989), Mainwaring et al. (1992), Linz and Stepan (1996). 2. Dahl (1971), pp. 3–4. 3. Marshall (1950), p. 28. 4. Sen (1991, 1992, 1994). 5. Chile had reduced trade barriers to 15%, privatized close to 500 state enterprises, liberalized capital and labour markets, and had enjoyed five years of successive growth, and declining rates of inflation and unemployment. For good reviews, see Bosworth and Marfán (1994), Pietrobelli (1998). 6. See Archer (1995), pp. 103–14. In this case, it would be important to establish whether workers’ leaders at various levels favoured the possibility of trading immediate consumption rights for more long-term occupational rights. 7. Valenzuela (1992), p.71.

1

Citizenship and democratization

1. As in the dictionary definition, Scruton (1982), p. 63. Bottomore refers to this as ‘formal citizenship’. Marshall and Bottomore (1992), p. 65. 2. Mills’ modification of the utilitarian doctrine, as Lindsay noted (1940, pp. xvi– xvii, xxi), including the acceptance of state intervention to protect liberty (and his sympathies for trade unions) was an early exposition of this new problem. 3. Dahl (1971), p. 3, O’Donnell and Schmitter (1986), p. 8, Valenzuela (1992), pp. 60–1. In this definition, Sartori’s ‘representative and electoral democracies’ (1987, part 1, pp. 102–11) are subsumed. See Sartori (ibid., pp. 8–14), for a discussion of the need to sub-define different forms of democracy. 4. I will stick to the term ‘citizenship’ (as opposed to ‘human’) right (despite Bottomore’s suggestion 1992, p. 85) to the contrary), as nation-states remain the core agents and enforcers of law, certainly with regard to the labour market. There is the caveat that even legally working residents are not always conferred the full rights of formal citizens, but that is a problem beyond our scope here. 5. Scruton (1982, p. 64), sees citizenship (at least since Aristotle) as implying the ability to ‘hold office’. In most European states, receiving assistance meant losing the right to vote. For Denmark, this was the case until 1907, when unemployment benefit was separated from the question of political citizenship (although it was not until 1961 that all limitations were abolished constitutionally). In Japan, the Military Assistance Law of 1937 exempted soldiers in receipt of public assistance from losing their right to vote, but the provision still remained for all other recipients; Garon (1997), p. 58. 6. On the Malthusians, see Donzelot (1977), pp. 73–4; and on the Webbs, Garon (1997), p. 44. 236

Notes 237 7. Mill (1910) for example, ch. 3, ‘On Individuality’, p. 125. 8. Macpherson (1973), p. 51 and pp. 40–1. The free access to the pursuit of these is understood as ‘developmental liberty’ (pp. 97, 119) (henceforth ‘developmental freedom’). 9. Macpherson (1973), pp. 97, 118. According to this principle, intervention that is beneficial to some without harming others (except their extractive abilities) would be justified. 10. Archer (1995), pp. 16–17, 19–22. 11. Sen (1992, pp. 110–11), (1997, abstract). 12. Archer (1995), pp. 8, 23–6. 13. See also Taylor on ‘developed freedom’ as an understanding of self that is ‘always created and sustained by the common expression and recognition [it] receive[s] in the life of the society’, Taylor (1985), p. 209. A related notion of ‘republican freedom’, freedom as being active members of the public space is found in the Kantian tradition, and in Arendt, as eloquently discussed in Beiner (1984). 14. Sen (1992), p. 57. 15. Sen (1992, pp. 41, 66), (1991, p. 17). 16. Sen (1992), pp. 110–11. 17. And hence we might say that the more national citizenship encompasses such dense forms of social citizenship, the greater does its ‘intensity’, in O’Donnell’s (1992) terminology, become. 18. Marshall’s understanding of political and social citizenship was essentially the same as ours. Although he emphasized the whole range ‘from a modicum of economic welfare and security to the right to share to the full in the social heritage’ and ‘live the life of a civilised being’, in practice he referred to institutions linked to the labour market, such as the educational system and the social services, T. H. Marshall (1950), p. 12. In the ‘civil element’ he referred to basic rights ‘necessary for individual freedom’ upheld (principally) through the courts of justice. 19. T. H. Marshall (1950), pp. 43–4. 20. Ibid., p. 47. 21. A. Marshall (1966), p. 180. 22. Nelson (1992) is representative of texts that counterpose the two poor constituencies. 23. For a volume that discusses other social policy areas where the mainstream has been diluted as under (i) and (ii) above, see Haagh and Helgø, forthcoming. 24. All who possess the status are equal with respect to the rights and duties with which the status is endowed. 25. T. H. Marshall (1950), p. 58. 26. As enforced by the state as against (if necessary) the wishes of families. Mill, ‘On Liberty’ (1982 [1859]), p. 176; A. Marshall [1873] in Pigou, cited in T. H. Marshall (1950, p. 27), and A. Marshall (1890), pp. 180, 596. 27. T. H. Marshall (1950), p. 27. 28. Ibid., p. 67. 29. Miller (1990), pp. 98–9. 30. Marx and Engles (1975) (The German Ideology, vol. 5). 31. Marx and Engels (1968) (Critique of the Gotha Programme), pp. 234, 306. 32. T. H. Marshall (1950), pp. 10–11.

238 Notes 33. 34. 35. 36.

37. 38. 39.

40. 41. 42.

43.

44.

45. 46.

47.

48. 49.

As represented well in Nelson (1992). A good summary is Tendler (1999). See Haagh (2001b). Some of the most important advantages of political democracy (to individual freedom) lie in the realm of statutory rights (state guarantees, independent complaints procedures and judiciary, and so on), and are undoubtedly necessary for making social citizenship under any form complete. O’Donnell and Schmitter (1986), pp. 38, 46. Przeworski (1986), pp. 59, 62; Olson (1993), p. 573. Even where the bourgeoisie is the progressive force (Dahrendorf, 1959; Barrington-Moore, 1966), we are used to thinking that deepening will follow. Rueschemeyer et al. (1992, pp. 297–8) emphasize the role of labour in bringing about political democracy. Skocpol (1979, p. 141) explained the timing and manner of early political reforms by international events and state capability. J. S. Valenzuela (1988), pp. 16, 24. See, for example, Skocpol (1979), pp. 82–4 on the early centralization of the Imperial Russian State in the absence of political citizenship, and pp. 106–8. Chile was another case where state-centralization arguably preceded democratic reforms; Zeitlin (1984). Skocpol (1979, p. 85) writes about Russia, how ‘juridical equality for the peasantry was a prerequisite for the establishment of a modern army of “citizen” conscripts’ in the 1860s. But rural emancipation also immediately preceded conscription in France (in 1793) and Prussia (1807–14), although it was only in the former case that this process could be associated with the installation of political democracy. When Germany decreed universal health insurance in 1883, the first country to do so, it was hardly a full-fledged democratic state. General schooling was introduced in Denmark as early as 1814, and the country’s illiteracy rates were as low as 10–15% by mid-century, compared with 40–45% in the developing political democracies of England and France at that time; Senghaas (1985), p. 101. However, Denmark did not introduce parliamentary government till 1901, with universal suffrage following in 1915. Mainwaring et al. (1992), O’Donnell (1992), p. 18. See also Valenzuela (1992), pp. 60–1. Because Chile combined transition through reform (or ‘pacts’) with the existence of firmly anti-democratic authoritarian rulers; Valenzuela, 1992, pp. 77–8. We argue that in specific instances ‘reserved domains’ such as that Valenzuela highlights (for example, the presence of constitutional senators and credible threats from entrepreneurs of destabilization, among others) served variously to consolidate democracy and to delay social democratization, sometimes simultaneously. Macro-perspectives are considered in this chapter, and micro-perspectives in Chapter 2. For a discussion of economic governance structures, see Bowles and Gintis (1996), and Williamson (1985, 1995, 1996), from p. 325. Polanyi (1957 [1944]); Hirschmann (1958, 1963); Gerschenkron (1966); Johnson (1982, 1987); Senghaas (1985); Evans (1994, 1995). Evans (1995, p. 12) sees autonomy as being characterized by meritocratic and self-contained bureaucracies, and embeddedness as a smooth relation-

Notes 239

50. 51.

52. 53. 54.

55. 56. 57. 58.

59.

60. 61. 62. 63.

64. 65. 66. 67. 68.

69.

70.

ship with the private sector. He identifies Korea, Taiwan and Singapore as versions of the developmental state (Brazil and India being intermediary cases). See also Johnson’s analysis (1982) of Japan. Williamson (1993), p. 144. Johnson (1982) (1995); Evans (1995). For Japan, Johnson mentions a long list, from the retirement of state officials into industry to the low influence of shareholders, and government control over financial institutions; Johnson (1982). For a good collection skill, see Ogawa et al. (1993), and Haagh (1998), ch. 1, for more detail. Garon (1997). Table A.1.1 in Haagh (1998) summarizes positive or negative ways that the welfare, developmental and laissez-faire states have contributed to the development of occupational citizenship, as well as some possible trade-offs. The terms laissez-faire and regulatory state are used interchangeably. For such policies in Sweden, see de Vylder (1996); and Boyer and Durand (1997), pp. 49–54. As described for the USA, in Table A.1.1 in Appendix 1. Johnson (1982), p. 19. Evans (1995, pp. 235–40). In this case, the state’s role was chiefly in promoting social (as opposed to industrial) development, but in a way that is described as being radical (compared with the small European welfare states). See also Sen (1994), pp. 12–13. Late development was not a factor to spur an aggressive industrial developmental role; although the state was also industrially developmental in some periods, particularly in Sweden; see de Vylder (1996). For a good summary, see Scholten (1987). On the role of associative practices and the de-concentration of ownership in the ‘Scandinavian development scenario’, see Senghaas (1985), pp. 89–93. In the latter case, what Therborn (1992) referred to as a ‘common-law’ (as opposed to a ‘public law’) conception of rights. Hence, Moulian (1985, p. 86) spoke of the element of party-induced ‘ideological supply’ to social conflict in the 1960s. See also Angell (1972) pp. 76, 192 and (1988). Therborn (1992), p. 30. Ibid. Parenthesis added. See Haagh (1998), Table A.1.1, for a detailed comparison of laissez-faire, welfarist and developmental regimes. Basically, the level of poverty and social inclusion that accompanies growth. Sen (1994), pp. 26–8. The actual value given to each country along the three dimensions are set out in Table A.1.2 in Appendix 1. All countries that are above 10 (out of 20) on this dimension are political democracies. Chile moved to an extremely low position on all three: extremely centralized authoritarian rule, repression and dismantling of interest organizations, and withdrawal by the state in the orchestration of economic and social development. Donzelot (1977), pp. 73–4.

240 Notes

2

Human resources and market reforms

1. Piore (1986), OECD (1986a, pp. 8–9) Boyer (1988, pp. 223–7), Blank and Freeman (1993), Archer (1995, pp. 174–6) and Boyer and Durand (1997). 2. Hayek (1980), Krueger (1983, pp. 25–7, 186–7) and Madsen-Pirrie (1995). 3. Williamson (1985), p. 69. 4. Boyer (1988), pp. 240–6, Pagano (1991), Archer (1995), Bowles and Gintis (1996a, 1996b) and Boyer and Duran (1997). For the institutional critique, see Williamson (1985) and Eggertson (1990). 5. See World Bank (1993, pp. 39–40) on the importance of skill formation as ‘the supply-side of entrepreneurial activity’, and World Bank (1997, p. 52) on the ‘large externalities to society’ and ‘high returns’ from primary level education; and Krueger (1983), p. 27. 6. Again, this is in evidence in the World Bank’s (1997, p. 52) almost exclusive focus on primary education in mentioning what it regards as acceptable ‘interventionist’ forms of skill formation. 7. For a theoretical discussion of internal and external labour markets, see Grimshaw and Rubery (1998). 8. Supposedly making wages equal to the economic value of output (Hayek, 1980, p. 56) and making labour move freely according to firms’ needs (ibid., p. 18). See also Krueger (1983), pp. 186–7, and Madsen-Pirrie (1995). 9. In this case, control gains refer not only to classical issues such as ownership over production assets, but also to more subtle forms of control over the work process, including the attainment of formal work rules. 10. As Maurice et al. (1986) have observed for the French case, national tripartite agreements have tended to be renewed in France only when struggles in individual firms have made the conditions of work between establishments converge to a degree which makes national rules both necessary and feasible. 11. This is the ideal Putterman (1990) refers to as ‘self-management’; Archer (1995) as ‘economic democracy’; and Bowles and Gintis (1996a) as ‘the democratic firm’. All these authors tend to assume that the market will still govern relations between firms. 12. Mergeability, range of applicability and ease of transformation are three distinct advantages that capital has over labour. For the individual, the attainment of skills is the best way to approximate these qualities. Organisation is another; Archer (1995), p. 94. 13. Because of the costs involved in continuously renegotiating individual labour contracts. See Williamson (1985, pp. 49–53) and Williamson et al. (1986, pp. 142–3). See also discussion in Archer (1995, pp. 49–52) and Pagano (1985, ch. 8) on the costs of litigation and arbitration. 14. Putterman (1990, pp. 164, 166–7) discusses inequality and dismissal. 15. For example, see Putterman (1990), pp. 157–9. 16. Archer (1995); Bowles and Gintis (1996a). 17. Hirschman (1970), pp. 82–6. 18. Putterman (1990). 19. In this regard, we assume that the overall long-term relationship between the development of human skills and economic growth (for example, Barro, 1991) has been established. 20. Human Development Report (1996), pp. 135, 144, 146; Angell (1996, pp. 11–13, 16–17) gives detailed data.

Notes 241 21. Solow (1957). 22. Williamson (1985), Williamson et al. (1986). See also Piore (1986, p. 162) on the New York garment industry, in which case flexibility was possible because implicit (trans-firm) forms of social insurance ensured maintenance of the resource (specifically the human resource) base. We could refer to this and the German cases as formalized skill markets. 23. Marshall (1966) pp. 180, 202–3. 24. For example, Schultz (1993, p. 244) would sharply disagree with the ‘Radical and institutional economists’ that ‘property rights and power relations shape technology and the organisation of labour’. Pagano and Rowthorn (1985), p. 121, nt 4. 25. For the first, see Putterman (1986), Pagano (1991, 1995), and Bowles and Gintis (1996a, 1996b); and the second, in Williamson (1985, 1996) and Williamson et al. (1986). 26. For example, Williamson (1985) does not believe that it has been convincingly shown that power relations can affect the shape of contractual forms. 27. In a German-style model, workers have a property right in an industryspecific occupation and individual employers readily invest in certified sector-specific skills. In a nationwide Scandinavian-type model, national pact-making ensures that workers own the right to ‘a job of quality close to the social norm’, Pagano (1991), p. 332. 28. Maurice et al. (1986). 29. In this sense, Pagano’s analysis may have underestimated the potential role of unions in firm-based regimes, including Japan (see Chapter 5) and Chile. 30. World Bank (1993), pp. 136–9. 31. Selvaratnam (1991). 32. This is where the absence in the World Bank’s (1997, pp. 4–5, 26, 56) advocacy of the private provision of social insurance of a discussion of linkages to other systems of social provision (assistance, skill development and so on) stands out. 33. On the latter, see Kongshøj-Madsen (1999). 34. Programme of government of the ‘Coalition (Concertación) of Parties for Democracy’ (1989), p. 25. (Parentheses inserted). The Concertación alliance, led by the Christian Democrat and Socialist Parties, was formed in 1988 between sixteen parties to contest the plebiscite that year for or against eight more years of Pinochet. Below, the Chilean name Concertación will be used for this coalition. The terms Aylwin and Concertacíon Government will be used interchangeably. 35. Ibid., p. 26. 36. For general reviews, see Scott (1995) and Bosworth and Marfán (1994). 37. Bosworth and Marfán (1994), p. 204. Savings in Indonesia, Japan, Malaysia, Singapore and Korea in 1993 were 31%, 30%, 38%, 41% and 35%, respectively; compared with Chile’s 24%, and in which pension fund savings represent a major share; World Bank (1995), pp. 78–9. 38. The differential with respect to international interest rates was more than 20% in the early 1980s. See further, Mizala (1992, pp. 168–70), and Bosworth and Marfán (1994, p. 185), for analyses of the decrease in productive investments. 39. Echeñíque (1990a), p. 154. Still, in 1994, the ten principal export products, after copper and other mining products, included, 5.2% for cellulose, 5%

242 Notes

40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58.

59. 60.

61. 62. 63. 64.

65. 66. 67. 68.

for fresh fish (unprocessed); 4% for fishmeal; 3.1% for fresh grapes; and 1.5% for uncut wooden planks; Ffrench-Davies and Sáez (1995), p. 88. Ominami and Madrid (1989), pp. 25, 36. World Bank (1999), World Development Report, p. 228. Infante and Klein (1992), p. 6. Ibid., p. 11. For good detail, see ILO (1998), pp. 65–6. Pietrobelli (1998), p. 117 SERCOTEC (1990), p. 99. The survey sampled 602 firms, of which textiles, wood and metallurgical firms accounted for over 90%. Ibid., p. 85. Bosworth and Marfán (1994). IADB (Inter-American Development Bank) (1994), Appendix table 2.a.1. Schultz (1993), p. 244. Devised from Bosworth and Marfán (1994), p. 180. Pietrobelli (1998), p. 109. Park (1994). On the structure of the labour movement and the system of industrial relations in the pre-1973 period, see Angell (1972), chs 3 and 4. See Barrera and Valenzuela (1986) for a good review of the revised 1980 Labour Code. By the end of the decade, roughly 10% of the labour force were covered by collective contracts. On the personification of the firm, see Eggertson, (1990, p. 7); and Chomsky in Burchill (1998), p. 19. For example, in Scandinavia, unemployment insurance is tied directly to other social assistance and training at national level, whereas in East Asia, national training systems and ‘copying’ have tended to set an occupational welfare standard between firms. Macneil (1978), pp. 863, nt. 25; 890), cited in Williamson (1985), p. 69. Although many large enterprises did have several unions prior to 1973, principally because of structural divisions in the firm and the separation of white- and blue-collar workers into separate unions. For details from this section, see Chapter 6). OECD (1992), p. 97. Helgø, C. (forthcoming). Of the 732 training corporations registered in 1988, at least 300 were not operational; CINTERFOR (1990), pp. viii, 52. Up to 1970, state-sponsored education included around 90% of the population. This figure had fallen to roughly 60% by 1990; IADB (1994), pp. 9, 11. In 1992, the public health care system still sustained over 80% of health care, but it managed only a third of the health resources available to the private insurance corporations. Selvaratnam (1991), p. 2. Park (1994), pp. 156–9. Haagh (forthcoming). Selvaratnam (1991), p. 1. Sharma and Luh Luh (1994) review, among other items, the regulation of labour supply, and the promotion of variable and participatory wages in return for stability and flexibility in employment. See

Notes 243 also Leggett (1993, pp. 237–8) for the state’s support of industry-wide unions and a continued role of tripartism, despite ‘Japanization’ in the 1980s. 69. I gratefully acknowledge the advice of Samuel Valenzuela regarding the usage of labour productivity indicators.

3

Business leaders and democracy’s limits

1. Valenzuela (1992), p. 77. 2. Montero (1993), p. 65. 3. See Johnson (1995, p. 43) on the ‘sociological’ conception of the market in Japan, and Hutton (1995, p. 287) remarked on ‘republican’ attitudes in Germany. 4. Campero (1984; 1991), Muñoz (1988), Montero (1990, 1993). 5. In the following text, Confederation and CPC are used interchangeably. 6. Kaufman (1986, pp. 89–107), by contrast, assumes that disintegration of the ‘exclusionary coalition’ is a feature of political liberalisation. 7. The right had moved from having no organizational vehicle at all in the mid1980s to the attainment of roughly 33% of the popular vote by the end of the decade. According to William Thayer, a close supporter of Pinochet who was appointed constitutional senator after 1990, the UDI and RN were each ‘given’ a senator. Thayer himself was RN’s. Interview with William Thayer, Minister of Labour under the Frei Administration (1964–70), Santiago, 4 September 1992. 8. Various interviews. The exception was a formal accord on the reform of the Central Bank. 9. A characteristic feature of this integration was the absence, compared to other countries, of significant economic or political cleavages between the agrarian and the expanding capitalist sector (Zeitlin and Radcliff, 1988), at least after 1891 (Zeitlin, 1984, pp. xi, 4–8). The SNA and the SOFOFA continued to display important differences over policy matters right up to the 1960s, as Carriere (1981) discovered, but Zeitlin and Radcliff’s analysis (1988, pp. 171–82) indicates that dominant families continued to have economic interests in both sectors. 10. Valenzuela (1996). 11. Northern Europe is the obvious contrast. In Denmark, for example, the great labour market accord, the so-called September Agreement, of 1899 preceded the major social democratic reforms. 12. According to Zeitlin and Radcliff (1988, p. 246), of the 229 executives studied in Chile’s top 37 non-financial corporations, 64 were personally holding some form of political office. Unfortunately, the figures do not reveal association by federation and type of office. See also Silva (1998). 13. Marín and Rozas (1988). Rodríguez and Venegas (1989) have surveyed the dominant influence of traditional land-owning families (those represented in the SNA) in the bulk of modern export agriculture. Some of SOFOFA’s most powerful members were based in sectors of subsequent export success, such as the export of paper pulp (the Matte group), whereas others were able to diversify into other profitable activities, such as the fishing industry. In fact, the SOFOFA saw a significant growth in direct members, from 300 in the early 1980s to just over 500 in 1992. Among them were some of Chile’s largest companies, who prior to 1987 had no association, including Entel, Chiléctra, Endesa and the Compañia de Teléfonos.

244 Notes 14. The information for this chapter is largely derived from interviews with twelve top leaders of the CPC, the SNA and the SOFOFA during 1992. For exact dates and details of all interviews henceforth consult list of interviews at the end of this volume. These include the current and penultimate presidents of the CPC (Manuel Feliú, and José-Antonio Guzman); the then head of labour relations in the CPC and SNA vice-president (Raúl García); SOFOFA vice-president, and later president (Pedro Lizana); and a top influential figure in the CPC, and former president of the SOFOFA (Eugenio Heiremans), among others. The author has decided not to include names for all references; for full detail, see Haagh (1998). 15. More moderate voices included Angelini and Lúksic. This information was provided by Heiremans (1992) and the national labour leadership’s top adviser, Eugenio Díaz, who expressed the hope that those would, in future, participate more in entrepreneurial politics. Various Interviews with Eugenio Díaz (1992). 16. Interviews with José Zaballa (1992), president of the moderate forum of Christian entrepreneurs, the Unión Social de Empleadores Cristianos (USEC), and Faúd Garib, SOFOFA councillor (1992). Zaballa refers to this as ‘the great taboo’ in the Chilean entrepreneurial community. 17. The Papelera’s board president, Ernesto Ayala, who held SOFOFA’s highest elected office for six years in the 1980s, was described by prominent entrepreneurs as the ‘power behind the throne’ during the succeeding presidencies of both Fernando Aguero and Hernán Briones (1991–3). Eugenio Heiremans, meanwhile, first president of the SOFOFA in 1952, worked very closely with Alessandri and was known as a key source of political influence within the SOFOFA and the entrepreneurial community as a whole in the transition period, without the need for elected office. 18. The other main contributors include the large forestry and cellulose plants and the country’s main fishing companies. 19. According to the SOFOFA’s annual accounts for 1991 (see Table 3.1), the contributions from the 559 largest firms listed as direct members (this figure dates from 1992), a membership form associated with size and prestige, accounted for 61% of the organization’s internal finance, against 3% from the majority of members represented through the sectoral or regional associations. There was not a set membership fee. 20. Interview with Eugenio Heiremans (1992c). 21. Interview with Raúl García (1992). Heiremans stresses the ‘many efforts made’ by entrepreneurial leaders from mid-1972 onwards ‘to push the armed forces to intervene’. The army was repeatedly described as ‘hens’; interview with Eugenio Heiremans (1992c). 22. Montero (1990) pp. 108–9. 23. And there was the alleged obsession sustained by SOFOFA’s president (Hernán Briones) with the use of the phrase ‘to prohibit’ in political discourse, for fear that a precedent might be set for future state intervention. Petitions from the textile and metal federations during 1991 and 1992 for state help had reportedly aroused this internal debate. 24. See CINTERFOR (1990) vol. I, part 1.1 for a comparative discussion of early initiatives in Latin America), and World Bank (1993, pp. 270–1). 25. Interview with top official of the SOFOFA.

Notes 245 26. 27. 28. 29. 30. 31.

32. 33.

34.

35.

36. 37.

38. 39. 40. 41. 42. 43. 44.

45.

See Campero (1991), pp. 152–7 for detail regarding other events. Interview with Eugenio Heiremans (1992). Interview with José-Antonio Guzmán (1992), president of the CPC. Zaballa, head of USEC, explains (in interview, 1992) how entrepreneurs of such alternative persuasion were marginalized from gatherings organized by the CPC. Heiremans, Ayala and Briones, three of the most powerful figures within the SOFOFA, were regular suppliers of funds to the right. Such views, expressed in 1992 by top SOFOFA leaders, had no doubt been confirmed by RN’s decision to initiate negotiations on both tax and labour reform in accord with the party’s decision in January, according to RN Senator Miguel Otero (in interview, 1992) to pursue a selective policy of ‘political accord’ with the Aylwin Government. Interview with Manuel Feliú (1992). García was viewed as ‘one of the most enthusiastic participants from 1984 onwards’, despite the ‘continued scepticism’ he met with from other SNA leaders. Interview, Manuel Feliú (1992). The Vicariate was a semi-autonomous entity set up within the Catholic church, ostensibly to help provide for the poor and needy. But effectively it provided a shelter during the Pinochet years for opposition and human rights groups. Interview with Eugenio Díaz (1992c). Guillermo Campero, top adviser to Aylwin’s Labour Minister, similarly point to the key role of Raúl García as a mediator to the ‘hard’ sectors in the CPC. Interviews with Guillermo Campero (29 January, 4 February, 26 March, 8 May, 13 August 1992). Interviews, Pedro Lizana (1992), and Raúl García (1992). A leading entrepreneur recalls how Ayala only agreed to meet a group of pro-Pinochet labour leaders in the late 1980s when persuaded to do so by Heiremans, a figure who was himself vehemently opposed to Feliú’s leadership and accords with the CUT. As he says bitterly about its leader, ‘Bustos was elected by the parties and legitimized by us’; Interview, Heiremans (1992a). Interviews with CPC leaders. For detail, see Haagh (1998). ‘More or less as a fáit accomplí’. Interview, Eugenio Heiremans (1992). Quotes from interviews already cited. Interview, Raúl García (1992). All the entrepreneurs interviewed rejected this notion. Interviews, Eugenio Díaz (1992d), and Raúl García (1992); CNT (1985). The history of meetings and labour relations recorded in this and ensuing chapters is based on a vast collection of documents, which for lack of space has not been footnoted in this text, nor listed in the bibliography. Please refer to Haagh (1998) and the asterisked note in the Bibliography. CPC (1985), p. 1. Manuel Bustos, from the textile workers’ confederation Confederación de Trabajadores del Textil y Vestuario de Chile (CONTEVECH), had by this time assumed the key leadership position within the opposition labour movement. He was also the foremost Christian Democrat union leader in Chile, but the broad base of his authority stemmed partly from the fact that he had championed the co-operation of centrist labour leaders with the left from the late 1970s onwards.

246 Notes 46. At this time, entrepreneurs from the open sector were careful not to commit themselves. García, Lizana and Garib, among others, participated in seminars organised by CED–CIASI–USEC in 1987, but did not sign the statement of intent. 47. All the leaders of the ‘open wing’ interviewed recalled their concern with the low public opinion rating of the entrepreneurial sector, and the need to influence a government programme that still appeared ambiguous. According to Campero, in the Aylwin Government’s later attempts to involve entrepreneurs, ‘the public opinion argument always worked the best’; Interview, Guillermo Campero (1992). 48. According to Guzmán, ‘Our greatest fear was labour reform’. Alé comments, ‘I am not against voluntary accords. It is the law about which I am worried’; Interviews, José Antonio Guzmán (1992) and Jaime Alé (1992b). 49. Interview, Manuel Feliú (1992). 50. Ibid. 51. Interview, Heiremans (1992). 52. Acuerdo Marco – Chile, una Oportunidad Histórica, 27 April 1990. 53. Interview, Guzmán (1992). 54. Interview, García (1992). 55. Montero (1993), pp. 37–9, 65. 56. Abramo (1992), p. 30. 57. Johnson (1995), p. 108. 58. Ibid., p. 106. 59. Principally because of the country’s party system. 60. Prior to the mid-1960s, the Liberal Party had been concerned mainly with issues pertaining to industry, whereas the National Party expressed the policy interests of the SNA.

4

Unions, parties and the sacrifice for democracy

1. O’Donnell and Schmitter (1986), pp. 27 and 48–56. 2. By this is implied a reasonable expectation that control trade-offs of an incremental nature could be established with employers. For an extended discussion of strategic feasibility, see Archer (1995), ch. 3. 3. Standing (1990), p. 458. 4. Indeed, Angell’s original work established this point for the Chilean case. Angell (1972), pp. 1, 62, 128–30. And, as he notes (p. 7), the few strong national federations that existed did so in spite of, and not because of, the Labour Code and its provisions. 5. Angell (1972), p. 54. 6. Angell (1972), p. 78. 7. Barrera and Valenzuela (1986), p. 230. 8. Detail is contained in Barrera and Valenzuela (1986), pp. 235–6. 9. Although multi-unionism was also sometimes a legacy of the pre-1973 period. A more in-depth analysis of this survey follows in Chapter 8. Details of sample selection for the survey may be seen in Appendix A. In the following List of Interviews at the end of this volume all tables with the title ‘Union survey’ will refer to this source. 10. Various interviews with sectoral leaders, as listed in the Bibliography. 11. And 155 600 for overall job creation, National Statistics Institute, Chile.

Notes 247 12. Of the Confederations cited in this chapter, only the bank workers’ federation (Confederación de Trabajadores Bancarios (CTB)) and the dominant metallurgical-sector workers’ federation (Confederación de Sindicatos de Trabajadores Metalúrgicos (CONSTRAMET)) could afford to run a photocopier. 13. CONSTRAMET had an unusually high number of staff. With 20110 members, it employed six people on a nationwide basis in 1992. The textile workers’ federation (Confederación de Trabajadores del Textil y Vestuario (CONTEVECH)), with 6758 members, had two paid staff. The main federation of forestry workers (Confederación de Trabajadores Forestales (CTF)), with 4235 affiliates, had no paid staff. 14. Interview with Mario Alburquerque, CUT adviser (1992). 15. Although it is clear that many union leaders, both old and new, saw the antecedence of ‘politicization’ in a negative light, 56% of respondents believed the experience of the previous twenty years signalled the need for a more moderate strategy. On the other hand, 90% of respondents believed that these experiences indicated the need to establish more autonomy from political parties. 16. Barrera (1969), p. 55. 17. For further detail on the figures cited here, see Haagh (1998), pp. 103–5. 18. The PDC and the PS each had a labour leader in the senate and the house of deputies. We should note, though, that in the Chilean case, the achievement of an independent source of union support and standing was a prerequisite for party support. By comparative standards, the Chilean labour movement remained highly democratic. 19. With the October 1991 vote, the PDC gained 20 seats on the National Council, the PS 15, and the PC 9. The composition of the CUT executive, meanwhile, reflected the power-sharing agreement between the Christian Democrats and Socialists: the PDC 5, PS 5, PC (Communist Party) 3, PR (Radical Party) 1, MOC (independent workers’ movement) 1. 20. Interviews with José Abello Jara (1992), president of the forestry sector confederation CTF, and Juan Corvalán (1992), leader of the mixed agricultural sector confederation, Confederación Unidad Obrero-Campesino (CUOC). 21. CIASI, of 1975, was led by an ex-union leader and lawyer, Eugenio Díaz, and it had advised the national union leaders since before the inception of the labour centrals. 22. CIASI’s budget was completely separate from that of the CUT, and with its annual size of nearly US$200000 was close to seven times larger than the modest US$28000 nominally commanded by the CUT itself. The bulk of CIASI finance came from abroad: US$58000 from the Finnish SASK; US$30000 from the German Ebert Foundation; around US$80000 from the EEC, and US$18000 from the Belgiam NCOS (all figures are approximate) interview with Eugenio Díaz (1992e). 23. The Chilean CUT in particular wanted to see an end to the ‘bargaining groups’ which had authority outside the union. 24. Personal notes, meetings spring and summer 1989 and interviews, Díaz (1992). 25. Díaz (1993). Díaz does not mention Germany, but the comparison with the German association for labour studies REFA (Verband für Arbeitsstudien), an independent agency with some 45000 members (including unions and firms) which specializes in research and education in the area of remunerative techniques, is inescapable. It is clear from the analysis of REFA by Maurice

248 Notes

26. 27. 28. 29.

30. 31. 32. 33.

34. 35.

36. 37. 38. 39. 40.

41.

5

et al. (1986, pp. 84, 130–1) that variable wages within firms could not have been legitimate to the same extent in Germany if it had not been for the participation of unionists at all levels of REFA, and for the technical basis of the role of REFA-educated union and management advisers. But a REFA-style agency clearly represented a public law conception of the labour market, and therein lay the radical nature of Díaz’s proposal. Arguably, the success of such an agency in Chile would also depend on the development of professional domains, or of trans-firm forms of occupational security for workers (for example, broadly recognized and transferable qualifications, or apprenticeship-based professions), which would make it realistic to separate the problem of variable wages from the question of organizational hierarchy in the firm. Interviews, Díaz (1992). CUT (1989), pp. 11, 18–19. See Haagh (1998), ch. 3. Interviews, Díaz (1992a). Díaz’ belief that the law commissions of spring 1990 represented the dawn of substantive industrial relations illustrates the CUT’s dependence on that exchange. Interview with Eugenio Díaz (1992a). Supposedly shared views on a public policy of training, unemployment benefit, and a national employment information agency later turned out to be illusory. See, for example, Revista Hoy, 7 May 1990. ICARE (1989), pp. 69–70. Interview, Campero (1992b); Castíllo (1991), annex 1. René Cortázar was elected Aylwin’s Minister of Labour only after an internal conflict between the sector of labour lawyers, represented by Francisco Tapia, on the one hand, and party economists on the other, interviews with Eugenio Díaz (1992a) and Guillermo Campero (1992e). Cited in Castíllo (1991), pp. 85–6. Ominami and Madrid (1990), pp. 120–44. Ominami was the Concertación’s choice for Economy Minister. Bitar (1993), then President of the Party for Democracy (PPD), the 1987 off-shoot of the Socialist Party. Concertación (1989), p. 15. Ibid., p. 18. Ibid., p. 25. Ibid., p. 25. These will be analyzed more closely in Chapter 6. This is despite early key concessions made to entrepreneurs. The Concertación at this stage had already backed down from an original intent to make the negotiation of union petitions above the firm level obligatory. It had also conceded to very flexible rules for dismissal, the other main law package. Interview with Guillermo Campero (1992a). And only 11 of the new unions in our survey were not either affiliated to a federation or had leaders with prior experience of union work. See further, Haagh (1998), p. 306.

The transition to democracy and the enforcement of markets

1. O’Donnell and Schmitter (1986), p. 46, and Olson (1993), p. 573. 2. Social pacts in the Aylwin Government refer to the formal yearly minimum wage agreements, as well as the wider ‘politics of social accord’, ostensibly aimed at drawing in social actors in aspects of policy-making.

Notes 249 3. Socialist and Christian Democrat parliamentarians agreed that they were receiving strict voting instructions from the government on labour and other issues. Interviews with Raúl Requena, (1992b); Di Giogio (1992). 4. As the SNA’s vice-president put it, the CPC was placed in a ‘political checkmate’; interview, Raúl García (1992). 5. Interview with Eugenio Díaz (1992). 6. Interview with CUT treasurer, Fanor Verdugo, (1992). 7. Interview with Minister of Government, Enríque Correa, 4 September 1992. 8. Acuerdo Marco – Chile, 27 April 1990. See also Epstein (1992) and Frank (1995). 9. The most concrete aspect of the deal was an agreement to raise the minimum wage by no less than 19.6% (received by an estimated 3.3% of the population in 1993), and to revise existing transfer benefits. Increases settled down in the next few years, with 3.1% in 1991 and 2.4% in 1992. A similar trend occured with social spending in GDP: ILO (1998). 10. Schull (1992), p. 732. 11. Hirschman (1993), pp. 292–3. 12. Interview, Enríque Correa (1992). Despite the fact that small firms comprised over two-thirds of the workforce considered poor in Chile. 13. See, for example, La Tercera, 16 March 2000. 14. Henríquez (1992), p. 2. Lysette Henríquez was general manager (gerente general) of SERCOTEC, the Technical Co-operation Service under the Economy Ministry. 15. Interview with Mario Cerda (1992). 16. Interview with Enríque Correa (1992). Correa defined this situation as one where, ‘decisions are made which affect all as a consequence of their membership of a particular class . . . That is not compatible with democracy’ (‘class’ here could also be translated as ‘group’). 17. Ibid. 18. Hirschman (1993), pp. 296–7. 19. Interview, Enríque Correa (1992). 20. See in particular, interview with Hernán Briones, El Mercurio 16 and 17 November 1991. 21. See, for example, El Mercurio 9, 12 and 14 June 1992; 23 and 27 July 1992; and El Diario Financiero (henceforth El Diario), 10 and 12 June 1992. 22. Interview with Faúd Garib (1992). 23. El Diario, 16 June 1993. 24. ILO, Memoria (1992). 25. Meetings over the summer 1990, for example, were scheduled to discuss ‘CUT themes’, such as private-sector training grants, unemployment benefit insurance and early retirement for workers involved in physically taxing work. See, for example, El Diario, 21 July 1990. 26. Letter to President Aylwin, 15 January 1992, p. 1. See also, Letter to José Antonio Guzmán, 27 March 1992. 27. SOFOFA’s response is contained in a letter to Manuel Bustos, 28 January 1992, p. 2. For details of correspondence at this time, see Haagh (1998), pp. 132–4. 28. Letter to Manuel Bustos, 9 March 1992, pp. 9–10. 29. El Diario, 10 November 1992. 30. This view was expressed by several entrepreneurs interviewed, including Lizana (1992), Feliú (1992), García (1992) and Heiremans (1992).

250 Notes 31. 32. 33. 34. 35. 36. 37. 38. 39.

40. 41.

42. 43.

44.

45.

46. 47. 48. 49. 50. 51. 52.

Interview with José Antonio Guzmán, president of the CPC (1992). Interview with Miguel Otero, vice-president of RN (1992). Interview with Raúl García (1992). Letter to President Aylwin, 15 January 1992, p. 3. Interview with Guillermo Campero (1992). See also, El Diario, 10 November 1992. CONSTRAMET had only six regular staff members in 1992, but had managed to compile a database on the firm of each affiliated union. Carta de Intenciónes ASIMET–CONSTRAMET, April 1992. Acuerdo Marco entre ASIMET, Consfetema – Fentema y CONSTRAMET, May 1992, pp. 3–7. ASIMET had created three corporations, a corporation for sport and culture, Corporación Deportiva y Cultural de ASIMET (CORDEMET); one for education, Corporación Educacional de ASIMET (COREDUCMET); and one for worker training, the Corporación de Capacitación Ocupacional y Desarrollo Laboral de la Industria Metalúrgica (CORCAPLAM), of 1979. CORCAPLAM trained over 5000 workers in 1991, and had 283 member firms. ASIMET (1991), p. 47. The example of one firm, the Indústria Procesadora de Asero, S.A. (IPAC) is typical. The firm joined CORCAPLAM in 1988, and since then it had not only begun to use the yearly tax rebate available to firms for worker training, but had also invested an additional 32% of its own profits. With 145 workers in 1992, it had offered training to 302 individuals since 1988, with an average of 24 courses a year purchased through CORCAPLAM: ASIMET (1992c), pp. 8–9. Interview with Jaime Paredes (1992). The minimum wage set by the metal accords in 1992 was 51000 pesos against a sectoral medium wage estimated at 108000 pesos. The national minimum wage was 38600 pesos in that year. Interview with Jaime Paredes (1992). The German government would contribute 3 million Deutsche Marks over three years from 1993, whereas the Ministry of Education would contribute some 195000000 pesos over a similar period. Only 50% (or 15 out of 31 unions) had failed to pay the union dues of the previous month (September 1991), and only 29% were four or more months behind: CTB (1991). This compares with 66% and 48%, respectively, for CONSTRAMET (in December 1991). The CTB’s proposals involved the establishment of an option for workers aged over 50 (women) or 55 (men) to trade lower pay and lower-skill tasks for job stability. Interviews with Raúl Requena (1992), and Carlos Garay (1992), secretary and president of the CTB. Information on the government’s position and reasoning is taken mainly from an interview with Guillermo Campero (1992). Unlike other sectors of temporary work, such as construction, with a density of about 9%, and whose pleas for sectoral talks were ignored. El Mercurio, 20 February 1990. See also El Diario 26 and 29 November 1992. Interview with Abello Jara et al. (1992). Interview with René Cortázar (1992). Cortázar (1993), pp. 33–4. Interview with top ministerial aide (1992).

Notes 251 53. Interview with Guillermo Cortéz, president of CONAPAN and general secretary of the CUT (1992). A document of September 1992 listed the other areas in which employers and labour leaders were in agreement, including regulation of night work and maximum workload, facilitation of bargaining at the sectoral level, and, the establishment of a tripartite commission for the bakery sector to discuss those and other themes. See further, Haagh (1998). 54. 10 Concertación deputies asked Cortázar to include the CONAPAN’s proposals in a law project. Letter to René Cortázar, 30 April 1992. 55. CONAPAN–FECHIPAN, joint letter to René Cortázar, Minister of Labour, 12 February 1992, and interview with Guillermo Cortéz (1992). 56. After it became clear in 1995 that the Concertación would not get sufficient legislative support to revise the original labour reforms (from the right, who still dominated the upper house), the governing alliance agreed to a ‘flexible’ version of its original proposal, as negotiated between a constitutional senator, William Thayer, and the then Labour Minister, Arrate, sent to the Senate on 30 September 1997. After an initial positive reception in the Finance Committee of the Senate, the full house rejected the idea of legislating at all on the matter of labour reform, with the votes of the right and most constitutional senators. 57. For a comprehensive discussion of the politics of the elections of December 1999, see Angell (2000). 58. There had been concerns on the right that opposition to the labour reforms in Congress would reverberate negatively on their candidate, Lavín, who in November had felt sufficiently worried about his social image to note that ‘some changes in labour law were required’. 59. See La Tercera, 25 November 1999. 60. CEP (Centro de Estudios Públicos), a think tank, is run by Eliodoro Matte, possibly the most influential figure in the SOFOFA. 61. See, for example, La Tercera, 19 March 2000. 62. And since entrepreneurial leaders had in so many ways stated that their support for democracy was contingent upon non-intervention, there was a very thin line between the policy-makers’ fear of economic disruption and their fear of an authoritarian reversal. 63. Evans (1995).

6

Democratic legitimation and the new Labour Code that never was

1. As Laski (1936, pp. 146–8) reminds us, the working class was not favoured very much in a direct way by the French Revolution, and the subsequent Civil Code. The expansion of civil rights that accrued to working people, in the principles established then of public assistance, general education, and, in the Civil Code, regulations of contract of service and of labour conditions and disputes (p. 148); not to mention the securing of peasant property and the abolition of seigniorial rights, which occurred despite and not because of any concern for working people’s rights. They accrued, precisely, as a sort of byproduct of the attempt by another class to secure for itself a new source of political power. This is the type of opportunity, later further expanded by the struggles of the working class to establish progressive taxation, the ‘social

252 Notes

2.

3. 4.

5. 6.

7.

8. 9.

10. 11.

service state’ and other goods (Laski, 1936, p. 156; Rueschemeyer et al. (1992, pp. 87–91), which we tend to associate with political democracy’s inception. Studies which have set out to prove the universal validity of the association between union strength and economic rigidity have found positive results when applied to the USA and UK (Blanchflower and Freeman, 1992; Hirsh, 1992; Addison and Chilton, 1993), but their overall conclusions have been contradicted by studies on other OECD countries, including Japan (Blanchflower and Freeman, 1992). For Japan, see Morishima (1991). On what is effectively a single union system, and on the role of the courts in upholding single-union agreements, see Hanami (1983), pp. 162–6. On co-determination and local unions in Germany, see Weiss et al. (1989), and on day-to-day consultation in Japan, Matsuda (1983), p. 200; Koike (1983), pp. 50–1. In a 1992 union survey of 240 manufacturing firms (to be discussed below) it was found that only 29% of unions had density levels above 60%. For example, according to firm-level data obtained from Chile’s National Statistics Institute (INE) of a sample of 240 manufacturing firms, the average rate of turnover in 1988–9 was 19.4% (reflecting the difference between the lowest and the highest recorded numbers of workers in the two years). If the supposition is made that dismissals were distributed evenly across the workforce, then this would correspond to roughly 9 jobs in the course of an average 45-year working life. The average Japanese worker would experience roughly 5.5 jobs (3.7 jobs for large firms, and around 5.7 for all sizes below that), according to Hashimoto and Raisian (1985, p. 726). However, since the calculation of the INE data for Chile was done by measuring the difference between the number of workers on the day of the (yearly) INE survey in 1988 and 1989, the rate of turnover noted above for Chile is almost certainly underestimated. In a survey of local union leaders in the same set of manufacturing firms (in 1992), the average level of turnover was found to be no less than 31%, which would correspond to roughly to 14 changes of job during an individual’s working life. The dismissal regime figured as the first or second most prominent theme in legal documents and in press statements. For the (similar) position of the CPC and the intellectual right, see speeches contained in ICARE (1990) and CPC (1990c) and Pérez Mackenna (1990). The discussion of positions taken by the CPC and the CUT are based on a vast collection of documents, which cannot be noted here for lack of space. Many observations below emanate from an internal memo on the (by then) separate negotiations sustained by the Labour Ministry with the CPC and the CUT, entitled ‘Memorandum: Principales controversias respecto de las modificaciones a la legislación laboral: Proyecto: negociación colectiva’, July 1990. For point-by-point referencing, see Haagh (1998), ch. 5. El Diario, 6 April 1993. El Diario, 28 September 1993; and interview with Vidaurre (1992), chief adviser to Senator Rolando Calderón. The last item was rejected in the Senate. Ministry of Labour (1990), pp. 5–7; and CIASI (1990c), pp. 1–2. For the members of Commission No. 2, see Haagh (1998), p. 158. This composition may be contrasted with Commission No. 3, which dealt with more

Notes 253

12. 13. 14. 15.

16. 17.

18. 19. 20.

21. 22.

23. 24. 25. 26. 27.

28. 29.

general issues ‘Materias Económicos-Sociales’. This included Jaime Alé and Pedro Lizana, leading opinion formers in the SOFOFA. For example, as president of the technical commission on labour affairs of National Renovation; interview with William Thayer (1992). Raúl García, head of the CPC Labour Commission said he did all lobbying in person. Interview with Vidaurre (1992). This policy was seriously considered during the Concertación’s elaboration of a government programme in 1988–9, and was later a key point of discussion with the CPC; interview with Guillermo Campero (1992a). Interview, Jaime Alé (1992). Interview, Angel Fantuzzi (20 and 28 August 1992), entrepreneurial leader from the metal sector and RN deputy. In Fantuzzi’s opinion, the government showed a surprising openness to RN’s proposals; interviews, Guillermo Campero (1992) and William Thayer (1992). Interview, William Thayer (1992). Interview, Eugenio Heiremans (1992a). Interview with William Thayer (1992). Emphasis added. For the full original quote in Spanish, see Haagh (1998), p. 365. Thayer was not only a labour lawyer by training, but he was also Labour Minister in the Frei Government, was consulted in the imposition of a new Labour Code during Pinochet (of whom he was a personal friend since 1938, when Thayer spent some time in the army), and has written a number of specialized books on the subject. Weiss, Simitis and Rydzy (1989), p. 86. These figures may be at least double, since Article 3 has been used, as an overall average, in only 28% of cases of job termination in 1991, against 37% for Article 2, and 35% for Article 1. The Government’s proposals are contained in Senado de Chile (Chilean Senate), Sesión 17 (1990), p. 1181. As reflected in Nuevo Código del Trabajo (1991), Law No. 19.010, Articles 5, 1, 2 and 3, respectively. Ibid., Article 3. Corte Suprema Recurso de Queja, sentencia de 16 de Marzo de 1971, cited in AALC (1990), p. 8. AALC (1990), p. 11. Ironically, this ‘radical’ proposal (compared with Article 3 on discretionary dismissal eventually adopted in Chile) appears rather similar to the controversial and ostensibly ‘liberal’ nature of new laws on dismissal passed in Korea in early 1997, which specified ‘the conditions and procedures for employment adjustment or dismissal due to compelling managerial reasons which have been ruled over by judicial precedents of the supreme court’, and the stipulations of which read as follows: ‘(1) There should be urgent managerial conditions such as continuing aggravation of management, structural adjustment and technological innovations for improvement of productivity and change in business; and (2) Employers are required to make prior efforts to avoid dismissal of workers and conduct sincere consultation with trade unions or workers’ representatives’, Ministry of Labor, Republic of Korea (1997), p. 5. From 76674 dismissals to 56640, Labour Directorate, Chile (1992b). Interview with Miguel Otero (1992b).

254 Notes 30. 31. 32. 33. 34.

35. 36. 37.

38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50.

51.

52. 53. 54.

As in Foxley’s speech at the ENADE 1988, ICARE (1989), p. 67. For Japan, see Hanami (1983), pp. 162–6. Matsuda (1983), p. 191. ILO (1990), pp. 111–12. France and Spain had, in 1992, a unionization rate of 12 per cent and 16 per cent, respectively, but through the practice of extended coverage, 80 per cent and 82 per cent of the workforce in these countries were covered by union-negotiated collective contracts, ILO (1992a), p. 58. Although the amount to be paid was lowered to 75%, Nuevo Código . . . (1991), Article 122 of Law No. 19.069, p. 98. See further Haagh (1998), p. 339. According to leaders of the Bank Workers’ Federation, for example, this meant failure to collect the 75 per cent from more than 2000 workers in the Banco de Chile’s central branch; that is, from nearly 50 per cent of its employees; interviews, Raúl Requena (1992) and Carlos Garay (1992). Ministry of the Presidency, Mensaje No. 408-321 (1991), pp. 8–9. La Nación, 22 July 1990. These were: the firm union, the inter-firm union, the independent workers’ union and the temporary workers’ union. Matsuda (1983), p. 188; Hanami and Blanpain (1989), p. 12. Hanami and Blanpain (1989), pp. 12, 16. The government’s law project is contained in Senado de Chile, Historia de la Ley No. 19.069, Sesión No.10, Article 82, p. 661. Ley No. 18.620, Libro IV, Article 291(3), Nuevo Código . . . (1991), p. 52. Ibid. (1991), Article 82, inc. 2., p. 92. See further, Haagh (1998), p. 308. The government proposed to make unions the ‘only subjects allowed to enter . . . collective contracts’, Senado de Chile, Historia de la Ley, No. 19.069, p. 635. Interview with René Cortázar (1992). Hanami and Blanpain (1989), pp. 5, 206. Like the code it sought to reform (Articles 280 and 301, 302, Libro IV, Titulo I of Law No. 18.620), the 1991 text stipulated that the negotiations an employer undertook with unions and bargaining groups, respectively, had to be simultaneous (with exceptions being subject to the employers’ discretion); that is, within 30 days of each other, Nuevo Código . . . (1991), Articles 94–96, pp. 93–4; and for the previous legislation, pp. 50–3. The legal formulation is, ‘a group of workers which is formed for the sole purpose of negotiating’. The rules governing the negotiating commissions of unions and negotiating groups were the same. Nuevo Código del Trabajo (1991), Article 102, inc. 3, a–d, p. 95. These conditions are clearly stated in Articles 90 and 127 of the new Code, Nuevo Código . . . (1991), pp. 93, 99. Calculated from the Labour Directorate, Chile. This evaluation fits with the ILO’s Convention 94, which promotes consultation on matters not dealt with in other machinery. ILO (1990), p. 51. It also chimes in with the ILO’s ruling of 1991 against the so-called ‘Solidarist’ organizations of labour promoted by employers in Costa Rica, in which it was stated clearly that direct settlement between workers’ representatives and management did not promote collective bargaining, ILO (1992), pp. 12–13.

Notes 255 55. Between the time of signing of the convention and the employers’ response to the union’s petition for a contract, 20 unionized workers were dismissed. Another 40 workers wished to move from the convention they had signed to form part of the collective contract, but were legally prevented from doing so. 56. For example, Valenzuela and Frank (1992, p. 10). 57. Shirai (1983a), pp. 137–8, 164; (1983b), p. 216. 58. ICARE (1989), p. 57. 59. Ministry of Labour, (1990). Interestingly, Commission No. 1 of the Spring Commissions agreed tentatively to both changes. 60. Interview with Senator for RN, Miguel Otero (1992b). In his book on the labour reforms, constitutional senator William Thayer also indicates as much. Thayer (1991), p. 164. 61. The 1979 Code stipulates 30 days (Libro IV, Titulo VI, Article 348, inc.1); compared with Article 157 (b), inc. 3 of 1991, which allows reinstatement after 15 days; Nuevo Código . . . (1991), pp. 62, 105. 62. Article 159, inc. 2, ibid., p. 105. 63. Interviews with leaders of federations of textile workers (CONTEVECH, CONTEXTIL) and metal workers (CONSTRAMET). 64. Ministry of Labour, Chile (1990).

7

Training policy and decentralization

1. World Bank (1980), p. 104; (1987), pp. 21, 208; Haddad et al. (1990), p. 6; OECD (1992), pp. 254, 268. 2. Graham (1992). 3. Bosworth and Marfán (1994), p. 170. Our own survey suggested that in 41% of the industrial establishments, at least 20% of the (waged) workforce had no employment contracts. 4. The share of female urban employment between 1966 to 1994 rose from 24.6% to 33.7%, whereas urban employment in total employment (in line with urbanization generally) rose from 74.9% to 87.1% over the same period; Díaz (1997), pp. 164, 167. 5. Graham (1994). 6. The increase in fiscal regulatory capacity occurred with a reform of the fiscal system in 1983; see Palma (1997). 7. SERCOTEC (1991), p. 18. It is estimated that around 70 per cent of employment is in firms below that size. 8. Elaborated from SENCE data, the Labour Ministry. 9. ‘Metas de Productividad Impulsarán la Capacitación’, El Diario, 21 September 1992. 10. CINTERFOR (1990), vol. I, p. III. 9. 11. Developed country figures are from 1989. Overall unemployment in Chile was 8.8 per cent in 1986, but that of 15–19-year-olds, 18.4 per cent. By 1990, the ratio was 5.7 per cent to 15.9 per cent. Meanwhile, between 1986 and 1995, the proportion of job-seekers among young unemployed increased from 65.5 per cent to 83.3 per cent for 20–24-year-olds; ILO (1998), pp. 189, 194. 12. Labarca (1996), p. 69, and ILO (1998), pp. 233–5. 13. Helgø, forthcoming. 14. National Statistics Institute (INE), Chile, 1999.

256 Notes 15. Within Latin America, certification of skills was found to be particularly wanting among the Chilean PTCs; CINTERFOR (1990), pp. VII, 24–5. 16. SERCOTEC, Holz (1991), p. 38. 17. Hence, the schools administered by the employer organisations themselves were not much use to firms in terms of providing them with professional criteria according to which they might classify skills, and, by association, also not much use in terms of enhancing labour security and mobility by providing professions with a trans-firm, tradable value. We should also note how this unwillingness to associate the social hierarchy within the firm with supra-firm professional norms would naturally complicate the establishment of supra-firm bodies involved in helping to modernize the flexibility of work, including institutions along the lines of REFA (see Chapter 4). 18. ‘Educación Técnico-Profesional: Un Atraso en la Modernización?’, El Mercurio, 2 December 1991. There seemed to be some justification for this view, since even ASIMET had encountered a variety of problems in the initial phase of the development of its dual programme, particularly in terms of a lack of trained personnel in the sector’s firms who could undertake apprenticeship responsibilities, as well as a reluctance on the part of the permanent staff to pass on their skills to what they saw as younger competitors. But the commitment of ASIMET’s leaders to the concept of concerted skill development meant that they took the initiative to set up a programme with German co-operation to establish a new qualification, the ‘trained foreman’ (‘maestro calificado’). These foremen, having thus gained a more secure professional status, had less incentive to oppose the transmission of their skills. 19. ICARE (1990), pp. 154–5; (1991), p. 32. 20. CINTERFOR (1990), vol. I, p. I. 27–28, vol. II, p. 38, p. V. 50. 21. Figures in this section are from The Labour Ministry, SENCE. 22. Ibid. In all, it was estimated that only 13% of eligible firm used the tax break option in 1992. 23. For example, in the years 1990 and 1992, the distribution of participation on registered PTCs between management and the workforce was 59.9% and 65.3% for the former, and 40.1% and 34.7% for the latter. Based on SENCE data, Labour Ministry, Chile. 24. CINTERFOR (1990), pp. III.23, V .28–30, VIII.46–49. 25. World Bank (1991), p. 38. 26. For example, a tax break scheme run in Singapore in the 1980s earmarked resources for worker training, with the grants paid out post hoc; World Bank (1991), p. 38. 27. Interview with Enríque Correa (Chapter 5, p. 00). 28. Interview with Mario Cerda, director of SENCE (1992). 29. The programme covered only 7.7% of the participants in SENCE, but it represented 28% of its resources. 30. El Diario, 5 November 1992. The dual system was ostensibly modelled on the German technical schools. 31. The table represents the total of courses (527) of one YTS auction. Each course featured 15–25 participants with a total of 11017; SENCE. 32. And hence less than the success rate of 34.5% of the municipal relocation offices mentiond above (35.9% for 1992).

Notes 257 33. 34. 35. 36.

37.

38.

39. 40. 41. 42. 43. 44. 45.

46. 47.

48. 49.

50. 51. 52.

Third licitation of the YTS, Labour Ministry, SENCE (1992b). Interview with Mario Cerda, director of SENCE (1992). Third licitation of the YTS, SENCE (Santiago, 1992). This was found in a SENCE study carried out in September 1994, ‘Caraterizacion de empresas e instituciones que han otorgado practicas . . . ’ Labour Ministry, December 1997, pp. 107–9. We should note that the Ministry of Education’s attempt to scrutinize the activities of the seventy technical schools under the CPC’s control in 1992, by putting a question-mark over the state’s renewal of the leases of these establishments to the CPC, was met with overt threats from the entrepreneurial community of authoritarian reversals (El Mercurio, 26 November 1991). From a ministerial inquiry it was found that roughly 80% of the establishments had no links with firms, despite a clear stipulation in the leases to that effect. This uncertainty lasted only a few days, however, and then the leases were renewed. See Park (1994), p. 159. Interestingly, the traditional state-certified training schools, such as KOMA in Korea, continued to sustain job placement rates of over 80% compared with the private training centres that mushroomed after Korea’s unemployment crisis of 1997 onwards, which sustained rates of around 20%; Haagh (2001a). These figure are taken from KOMA and Labour Ministry archives in 1999. Interview with high-placed official in the Labour Ministry (26/3/1992). The Labour Ministry, published in El Diario, 3 May 1993. Compare row 4, where workers’ contributions have been deducted, with row 2, the old system, in Table 7.8. The rows following row 4 in Table 7.8 illustrate the different ranges of accounts. Which the World Bank (Development Report – 1997, p. 73) sees as the key impediment. Proyecto de Ley Sobre Sistema de Protección al Trabajador Cesante (PROTRAC), 7 July 1997. In 1992, the unemployed person received roughly 27.5% of the minimum wage for the first three months, reducing to 18.3% for the next three, and 14% for the last six. The usage of which had increased after the 1990–1 reforms; see Chapter 6. Chile had, in 1992, roughly 250000 cases of arrears on workers’ AFP contributions by employers, 70000 of which were held up in the courts. Approximately 1 million more were ‘phantom’ registrars, with no record of payment. El Diario Financiero, 7 and 22 September 1993. According to the President of the AFP Habitat (linked to the construction sector), the AFPs should be permitted to run the administration of the system ‘from day one’; El Diario Financiero, 9 September 1993. Ministry of Labor, Korea (1999), p. 7. Interview with Gee Tae Song, Director-General, Human Resources Development Bureau, Ministry of Labor (August 1999). For example, according to Keum Jaeho of the Korea Labour Institute (the government think-tank on labour matters) the structure of financing and subsidies to firms was changed ‘at least ten times’ during the second half of 1998 and the first half of 1999; interview, July 1999.

258 Notes

8

Labour relations and investment in workers

1. The survey that forms the basis for our analysis in this chapter was carried out among 245 local union leaders in a mixed group of manufacturing firms across Chile in autumn 1992. For more detail, see Appendix A. 2. A position which, as we noted, rests on the assumption that managers are detached socially and respond perfectly to markets, as Marshall (1966 [1890], p. 180 and appendix A, p. 621)) also admitted. 3. A toned-down version of this approach can be found in the World Development Report, World Bank (1995), pp. 36–40. 4. See Schultze (1985), Blanchflower et al. (1987) on Europe; and Bewlay (1993), on the USA. 5. Solow also admits that extreme circumstances, including deep recessions, can have this effect. Solow (1990), pp. 37, 39. Presumably this could also make undercutting more acceptable among employers, and hence affect social norms. 6. Union repression was immediate after the coup, but liberalization of the economy preceded the establishment of a new Labour Code, and this in turn preceded recession and mass unemployment followed by ‘labour-intensive’ recovery (see Chapter 2). 7. The concepts of ‘lock-in’ and cumulative causation are explored in Brian Arthur (1989) and Hirschman (1993). 8. Solow (1990), p. 23. 9. According to Pagano and Rowthorn, ‘property rights and power relations shape technology and the organisation of labour’, Pagano and Rowthorn (1996), p. 121, nt 4. See further, Bowles and Boyer (1990), Marglin and Schor (1990), Pagano (1991), Bowles and Gintis (1996) and Grimshaw and Rubery (1998). 10. Community standards (Solow, 1990) or trust (Lorenz, 1992) and variable preferences (Akerlof, 1982, 1986) are emphasized. For workers’ preferences, see Pagano (1985), pp. 169–73 and Aoki (1990), p. 282, Dahl (1985), pp. 113–15, and OECD (1986b) p. 10. 11. Solow (1990), pp. 6, 23, 39. 12. Boyer and Durand (1997), for example, explore the effects of the ‘model of production’ in orientating the choice of labour flexibility; and Maurice et al. (1986) focus on the structuring role of the education system. 13. Where the sociological approaches are summarized in Note 9 above, and the economic approach represented in Williamson (1985). 14. As we observed earlier, see especially williamson (1985), pp. 207–14. if national incentives are weak, the importance of firm-level incentives will increase. 15. Institutional memory implies the use of authority (over time), as distinct from power in Arendt’s (1970, pp. 44–5) definition, which refers to numerical qualities and direct coercion. Unions, while they are a numerical form of strength, may serve to build the authority of workers over time, as expressed indirectly through formalization or non-union forms of representation. 16. Angell (1972), pp. 1–3. 17. The president of the union was targeted for interview; see further, Haagh (1998), pp. 202–3. 18. And levels of training did not vary significantly by sector, see Appendix A, Table A.5.

Notes 259 19. Hence the usual distinction that is made between auto-mechanical or ‘Taylorist’ assembly line production and automated production (Franzosi, 1995, pp. 79–80) is not always strict in Chile. 20. This is consistent with Díaz’s suggestions that so-called formal (waged) employment in Chile had become characterized by high levels of contractual informality in the 1980s, Díaz (1997). 21. There are no good national indicators on worker training inside Chilean firms (or indeed most LDC firms). Díaz (1993), pp. 43, 46. Chile’s INE (the National Statistics Institute), which has an impressive (classified) bank of over 168 economic indicators for every firm, does not include information on investment in training; see further, Haagh (1998), p. 203. 22. OECD (1992), p. 97. 23. All four firms were medium-sized and located in Santiago. Three were textiles firms and one clothing manufacturer. 24. In Firm A, variability according to seniority was identified (80000 to 85000 pesos per month), whereas wages in Firm B varied between 52080 and 57000 pesos per month. In Firm C, the difference was 36000 to 69000 pesos per month, and in Firm D 36000 to 56000 pesos per month; 36000 pesos was roughly equivalent to US$100 at 1992 values. 25. This was defined as the lowest and highest point of the union leaders’ own monthly wage in the previous year. 26. A proxy for the level of labour turnover was defined by calculating the difference between the maximum level the workforce had reached in the previous two years and the minimum level during the same period (see further, Appendixes B and C). The disadvantage to this method is that labour leaders might not provide an entirely accurate figure; however, there is no reason to suppose that the inaccuracy would be so great as to influence the statistical analysis in any significant way. In fact, since turnover is a controversial issue, union leaders tended to be clear about the minimum and maximum levels relating to the workforce in recent years. There is also the advantage that workers without contracts would be considered, and rather than recording the number of workers on a particular day in the year (as the National Statistics Institute does), the actual (although approximate) maximum and minimum levels would be recorded. 27. Buchele and Christiansen (1992, p. 85) use the presence of bonus payments as an indicator of more co-operative labour relations. 28. Teresita Selamé notes the important distinction between multi-task work (‘multivalencia’), and multi-skills (‘polivalencia’), Selamé (1986), pp. 3, 6. 29. In Chapter 9 views on variable wages are discussed in more detail. 30. Because of a change in classification after 1989, this was the last year for which figures were available. INE figures correspond to 210 of our firms in the years 1986–8, and 198 firms in the years 1988–9. 31. As noted on p. F134. 32. The INE sample of firms had an average rate of turnover of 19.4% in 1988, using the calculation mentioned earlier p. F134. 33. The case of Firm A perhaps highlights a point which Marshall has made, that ‘the popular notion of fairness’ (such as higher formalization of the wage system (whereby ‘every man . . . ought to be paid for his work at the usual rate for his trade and neighbourhood’) is a sort of prerequisite for

260 Notes

34. 35. 36.

37. 38. 39.

40. 41.

42.

43.

9

the acceptance of flexibility on the part of workers (when the doctrine of fairness in pay ‘is modified by the admission that changes of circumstance may require changes of wages in one direction or another’), Marshall (1925), p. 217, quoted in Solow (1990), p.17. The faults cited included low productivity, failure to obey the supervisor, the breaking of equipment, and working too slowly. The measurement of individual output also showed clear positive correlations with low formalization. This presumption is made in the ‘conflict model’, on which the neo-liberal perspective also relies. For a review on the controversy on this point, see Buchele and Christiansen (1992), p. 78. For Malaysia, Standing also found that firms with a high proportion of casual labour were less likely to train the workforce; Standing (1992), p. 344. Solow (1990), p. 23. The level of machinery investment is a tentative indicator based on the number of machines introduced relative to firm size. In respect to sector we separated those represented in the survey into three according to approximate technology levels. Cross-tabulations between these sectors on the one hand, and turnover and training on the other demonstrated a moderate relationship; see Appendix A, Tables A.4 and A.5 in Appendix A. See Haagh (1998), p. 319. The easiest way to assess the significance of the different variables is to look at the ‘significance of F’ column. It is common to suppose that values below 0.05 are significant, whereas values below 0.01 are highly significant. Cross-tabulations Table B.6 in Appendix B show that the association between turnover and one of the social factors considered (level of formalization of labour management) was highly significant in both low and high technology sectors. See Haagh (1998), p. 319.

Union strength, history and effects

1. For example, for Oxhorn (1995), organized labour in Latin America of the 1990s ‘is increasingly reduced to the role of an even more self-interested actor that is incapable of regarding the interests of non-members and pursuing public goods’. 2. Valenzuela (1988), pp. 8–10. See also Valenzuela and Frank (1992) for an analysis of the importance of strike levels during Chile’s early transition years. 3. Valenzuela (1988), p. 13. 4. Angell (1972), p. 2. 5. As before the dictatorship (Angell 1972, pp. 61–3), only an infinitely small number of union leaders, nearly all in national office, received any form of payment. Furthermore, where expenses were paid by union members, the latter tended to be rather jealous at how the money was spent. Even the leaders of sectoral associations tended to spend most of their time as regular workers in their firms, and nearly all the local leaders had to do their union work outside working hours. Although union leaders were protected from dismissal during their term, they lived with the fear of dismissal that could follow when the period of protection ended. The new Labour Code of 1979

Notes 261

6. 7.

8.

9. 10.

11. 12. 13. 14.

15.

16.

17.

18.

19.

20.

21.

increased the financial weakness of unions by terminating the mandatory share of unions in company profits. Franzosi (1995), p. 15. We should note that by ‘peak’ we are not referring to the levels of conflict (national strike versus a strike within the firm), but to the ‘overtness’ of conflicts, such as the strike itself. The assumptions of classical strike models discussed by Franzosi, such as Ashenfelter and Johnson (1969), are mirrored in the more politically inspired analysis of neo-liberal theory, such as Hayek (1988). Both assume a limited preference on the part of workers (and their leaders – a distinction is not made), based on the assumption that work is a ‘disutility’. Bosworth and Marfán (1994), chs 1 and 7. For relative deprivation theories, see, for example, Graham and Gurr (1969); see Barrera and Valenzuela (1986) for an analysis of the protests in 1983. A recent volume that stresses the importance of national organizational capacity and political coalitions is Baglioni and Crouch (1990). See Gallie (1983, pp. 14, 157) for a discussion of institutionalization theory. Aglietta (1979), Maurice et al. (1986), Bowles and Boyer (1990), Marglin and Schor (1990), Putterman (1990), Bowles and Gintis (1996). Although in both cases figures were high. This reflects the continuing high level of distrust and controversy associated with the dismissal regime in Chile. But this also indicates that leaders with conventions were hardly motivated by the negative views of their employers when they reported high levels of dismissal and low formality. Only seven firms in our sample were established after 1980. As to the union presidents, 15% had worked in the firm for less than five years (of which 8 had been there for four years), and 54% had worked in the firm for more than ten years. In evaluating wage scales we included only a cluster of sectors that had a roughly similar wage dispersion overall (fruit, wood, chemicals, textiles and metals). Although there was a distinction between ‘workers’ and ‘employees’ in the period prior to 1976, the principle of the closed shop was still in force with regard to the enterprise unions that grouped the first category, Angell (1972), p. 6. The hypothesis could be made that the firm’s market position, in particular its level of technological development, would affect the levels of training or turnover. Tables A.4 and A.5 in Appendix A indicate in this respect that levels of technology (by sector groupings) did not have a strong impact on either variable (nor was it associated with union age or bargaining instrument, as shown in Haagh, 1998, pp. 351–352). The president of this union was not associated with a particular political party, but he had been actively involved in the textile federation CONTEVECH since the early 1980s, and before joining Pichara he had been union leader in another firm. These benefits resembled almost point by point those Angell (1972, p. 77) recorded for the 1960s, including housing, medical attention, leisure-time

262 Notes

22. 23.

24.

25. 26. 27. 28.

29. 30. 31.

32.

33.

34.

35. 36. 37. 38.

39. 40.

entertainment, etc. Then, the employer could also use his ability (and the unions’ inability) to extend such benefits, as a source of authority over the union. See also Figure 6.1 in Chapter 6. Only 3% of respondents had no views on this issue. In his survey of local labour leaders Frank (1995, p. 360) also found that local leaders distinguished between ‘possible’ and ‘not possible’ demands, and that 57% of leaders believed that it was necessary to moderate their demands for the sake of the stability of political democracy (ibid., p. 365). The chi squares between the acceptance of wage flexibility in principle and these variables were, in the order cited: 0.0500, 0.0503, 0.0170, 0.0276, 0.0751, bearing in mind that chi squares below 0.15 are not normally significant. Barrera et al. (1964), p. 417; 19% were classified as apolitical. Barrera (1969), p. 64. Angell (1972), p. 78. Sectoral leaders in CONSTRAMET (Communist) and CONTEVECH (Christian Democrat) confessed that they deliberately toned down their political discourse at general meetings, for fear of scaring off local leaders various interviews. See further Haagh (1998), Appendix F. Since this was also phrased as an open question, a rich and detailed selection of responses were collated. For example, in the group of employers which had rejected a discussion with the union, only 40% had practised a formal wage system, whereas in the group of ‘open’ employers, 60% had done so. High levels of turnover and low levels of training also had a negative association with the employers’ tendency to reject discussion with the union. The odds were 4.8:1 that new unions which had signed conventions would face rejection by the employer. On the other hand, the odds were only 1.7:1 that new unions with contracts would do so. For example, the deal on incentive wages agreed to in San Joaquín was dependent on the goodwill of the employer, who could have invoked the ban on discussing matters that infringed his right to lead, organize and administer the enterprise. The overwhelming majority of respondents to the survey (68%) said that they rarely or never received sufficient information for bargaining; See Haagh (1998), p. 286. This coincides with our earlier finding that contracts (as opposed to conventions) promoted ‘embeddedness’ of unions in the firm. We recall that tripartite commissions at the sectoral level came to wield direct power over wages in many sectors during the Allende period. Hutton (1995a, 1995b), Madsen-Pirrie (1995). In pre-1973 Chile there were a number of large firms with several unions. However, this often emerged as a consequence of the legal division between blue- and white-collar unions, or through to functional divisions within the firm. As Valenzuela (1988, pp. 15, 22) set out. Valenzuela (1988), pp. 2–3. Although, this author evidently has no responsibility for our own, admittedly rather stretched, interpretation of his theory.

Notes 263

Conclusion 1. 2. 3. 4. 5.

6. 7.

8.

9. 10. 11. 12.

13.

O’Donnell (1992), Valenzuela (1992), Linz and Stepan (1996). Valenzuela (1992), p. 71. As Sartori’s (1987a, p. 9) social democracy as a ‘style of society’. Schultz (1993), p. 244. Krueger (1983), Hayek (1980), Madsen-Pirrie (1995), Addison and Chilton (1993), Hirsh (1992); see further, Introduction and Chapter 5 of this volume. Agresti and Finlay (1997), p. 357. As we saw in Chapter 7, mainstream economists such as Marshall (1966 [1890]), in his time, and Solow (1990) in recent years, have been doubtful on this point. For a sociological analysis, see Granovetter (1985). Virtuous being, for example, between high training and low turnover, or between high formality and high training, or between a long-term timehorizon and the possibility of negotiating control trade-offs, among others. In Appendix B, Tables B.6 and B.7. Barrera and Valenzuela (1986). Angell (1972); Barrera (1968). Conceivably, this was a circumstance that allowed local (as well as national) unions to engage in sometimes highly informal relations with employers, without this participation collapsing into a mere mutualism. Valenzuela (1996).

Appendix A: Notes on survey sampling 1. I thank Volker Frank for introducing me to the staff of the Centro de Estudios de la Realidad Contemporánea (CERC). 2. With the exception of the bakery sector; see note to table A.A.i. 3. See further, Haagh (1997), p. 344.

Select Bibliography Note: This bibliography includes only sources to which references are made directly. The original bibliographies for the D.Phil. thesis that formed the basis of this book had extensive sections covering labour relations documents, legal texts and statistical references, letters and notes. For lack of space these have not been included here. The original full references can be found in Haagh (1997), bibliographies i–v. Details of interviews can be found in List of Interviews at the end of this volume. Legal texts and labour relations documents are only quoted where absolutely necessary, and with a few exceptions are listed directly in footnotes. AALC (1990) Corte Suprema Recurso de Queja, sentencia de 16 de Marzo de 1971, Informe de la Corte del Trabajo de Santiago. Recurso: Club de deportes Colo Colo; in Asociación de Abogados Laboralistas de Chile, ‘Comentarios al Proyecto de Ley Referido a las Normas Sobre Terminación del Contrato de Trabajo’, Santiago, July. Abramo, L. (1992) ‘El Empresariado en Tanto Actor Social: Análisis de la Literatura Reciente’, Instituto Latino Americano de Estudios Transnacionales, Santiago, Chile. Addison, J. T. and Chilton, J. B. (1993) ‘Can We Identify Union Productivity Effects?’, Industrial Relations, 32: 2, pp. 124–32, Winter. Aglietta, M. (1979) A Theory of Capitalist Regulation: The US Experience, London: Verso. Agresti, Alan and Finlay, Barbara (1997) Statistical Methods for the Social Sciences, 3rd edn, New Jersey: Prentice Hall. Akerlof, G. A. (1982) ‘Labor contracts as partial gift exchange’, Quarterly Journal of Economics, 96, pp. 543–69. Akerlof, G. A. (1984) ‘Gift Exchange and Efficiency Wages: Four Views’, American Economic Review, 74, pp. 79–83. Akerlof, G. A. and Yellen, J. A. (1986) Efficiency-Wage Models of the Labour Market, Cambridge: Cambridge University Press. Angell, A. (1972) Politics and the Labour Movement in Chile, Oxford University Press. Angell, A. (1988) ‘Some Problems in the Interpretation of Recent Chilean History’, Bulletin of Latin American Research, 7: 1, pp. 91–108. Angell, A. (2000) ‘The Chilean Elections of 1999’, Bulletin of Latin American Research, June. Angell, A. (forthcoming) ‘The Politics of Education Reform in Chile: the Programa 900 Escuelas and the Mece-Básica’, in L. Haagh and C. Helgø, Social Policy Reform and Market Governance in Latin America, London: Palgrave, St Antony’s. Aoki, M. (1990) ‘A New Paradigm of Work Organisation and Co-ordination? Lessons from the Japanese Experience’, in S. A. Marglin and J. B. Schor (eds), The Golden Age of Capitalism, Oxford: Clarendon Press. Archer, R. (1995) Economic Democracy – The Politics of Feasible Socialism, Oxford: Clarendon Press. Arendt, H. (1970) On Violence, London: Allen Lane. 264

Select Bibliography

265

Ashenfelter, O. and Johnson, G. E. (1969) ‘Bargaining Theory, Trade Unions and Industrial Strike Activity’, American Economic Review, 59: 40, pp. 35–49. ASIMET (1991) Memoria, Santiago: ASIMET. ASIMET (1992a) Guía de la industria metalúrgica-metalmécanica, Santiago, Chile: ASIMET, Publicaciones Lo Castillo. ASIMET (1992b) Membership list, Santiago, Chile: ASIMET. ASIMET (1992c) Boletín, Santiago, Chile: ASIMET, January–February. Association of Labour Lawyers in Chile (1990) ‘Comentarios al Proyecto de Ley Referido a las Normas Sobre Terminación del Contrato de Trabajo’, Santiago, Chile, July. Baglioni, G. and Crouch, C. (eds) (1990) European Industrial Relations: The Challenge of Flexibility, London/Newbury Park/New Delhi: Sage. Barrera, M. (1969) El Sindicato Industrial: Anhelos, Metodos de Lucha, Relaciones con la Empresa, Universidad de Chile, Edicion Publicaciones INSORA, No. 48, Santiago, Chile. Barrera, M., Landsberger, H. and Toro, A. (1964) ‘The Chilean Labor Union Leader – A Preliminary Report on His Background and Attitudes’, Industrial and Labor Relations Review, 17: 3, April. Barrera, M. and Valenzuela, J. S. (1986) ‘The Development of Labor Movement Opposition to the Military Regime’, in A. Valenzuela and J. S. Valenzuela (eds), Military Rule in Chile. Dictatorship and Oppositions, Baltimore and London: Johns Hopkins University Press. Barrington-Moore, Jr, B. (1966) Social Origins of Dictatorship and Democracy – Lord and Peasant in the Modern World, Harmondsworth: Penguin Books. Barro, R. J. (1991) ‘Economic Growth in a Cross-section of Countries’, NBER Working Paper No. 3120, Cambridge, Mass.: National Bureau of Economic Research. Beiner, R. (1984) ‘Action, Natality and Citizenship: Hannah Arendt’s Concept of Freedom’ in Conceptions of Liberty in Political Philosophy, Pelczynski, Z. and Gray, J. (eds) The Athlone Press, London. Bewlay, T. (1993) ‘Research Findings Presented to the Happiness Conference’, LSE, November. Bitar, S. (1993) ’Chile: Hacia una Inserción Internacional (III)’, El Diario Financiero, 2 March, p. 2. Blanchflower, D. G. and Freeman, R. B. (1992) ‘Unionism in the United States and Other Advanced OECD Countries’, Industrial Relations, 31: 1, pp. 56–79, Winter. Blanchflower, D. G., Oswald, A. J. and Garrett, M. D. (1987) ‘Insider power in Wage Determination’, Mimeo: London School of Economics, Centre for Labour Economics. Blank, R. M. and Freeman, R. B. (1993) ‘Evaluating the Connection between Social Protection and Economic Flexibility’, National Bureau of Economic Research, Working Paper No. 4338, Cambridge, Mass., April. Bosworth, B. P and Marfán, M. (1994) ‘Saving, Investment and Economic Growth’, in B. P. Bosworth, R. Dornbusch and R. Laban (eds), The Chilean Economy – Policy Lessons and Challenges, Washington DC: The Brookings Institution. Bottomore, T. and Marshall, T. H. (1992) Citizenship and Social Class. Bowles, S. and Boyer, R. (1990) ‘A Wage-led Employment Regime: Income Distribution, Labour Discipline, and Aggregate Demand in Welfare Capitalism’, in S. A. Marglin and J. B. Schor (eds), The Golden Age of Capitalism, Oxford: Clarendon Press. Bowles, S. and Gintis, H. (1996a), ‘Is the Demand for Workplace Democracy Redundant in a Liberal Economy?’, in U. Pagano and R. Rowthorn (eds),

266 Select Bibliography Democracy and Efficiency in the Enterprise, USA and Canada: Routledge Studies in Business Organisation and Networks, Routledge. Bowles, S. and Gintis, H. (1996b) ‘Efficient Redistribution: New Rules for Markets, States and Communities’, Politics and Society, 24: 4, December, pp. 307–42. Boyer, R. (ed.) (1988) The Search for Labour Market Flexibility – The European Economies in Transition, Oxford: Clarendon Press. Boyer, R. and Durand, J.-P. (1997) After Fordism, London: Macmillan Business. Brian Arthur, W. (1989) ‘Competing Technologies, Increasing Returns, and LockIn by Historical Events’, Economic Journal, 99, March, pp. 116–31. Buchele, R. and Christiansen, J. (1992) ‘Industrial Relations and Productivity Growth: A Comparative Perspective’, International Contributions to Labour Studies, No. 2, pp. 77–97. Burchill, S. (1998) ‘Human Nature, Freedom and Political Community: An Interview with Naom Chomsky’, Citizenship Studies, 2: 1, pp. 5–22, February. Campero, G. (1984) Los Gremios Empresariales en el Período 1970–1983: Comportamiento Socio-Político y Orientaciones Ideológicas, Santiago: ILET. Campero, G. (1991) ‘Entrepreneurs under the Military Regime’, in P. W. Drake and I. Jaksic (eds), The Struggle for Democracy in Chile – 1982–1990, Lincoln, Neb. and London: University of Nebraska Press. Carriere, J. (1981) Landowners and Politics in Chile – A Study of the SNA 1932–1970, Amsterdam: Centrum voor Studie en documentatie van Latijn Amerika. Castíllo, F. (1991) La Fuerza del Diálogo, Santiago: Ced, November. CIASI (1990a), ‘Terminación del Contrato de Trabajo’ – Notes from Spring 1990 Labour Commissions (Number 1), April. CIASI (1990b) ‘Comisión No. 2’ – Notes from Spring Labour Commissions, April. CIASI (1990c) ‘Marco de Referencia para el Diálogo entre la CUT y la CPC’, Santiago, 31 January, 1990, Reprinted in Serie Documentos, CIASI, No. 5, June. CINTERFOR (ILO) (1990) La formación profesional en el umbral de los 90, vol. I, Montevideo: ILO. CNT (1985) Minuta Reunión con Dirigentes Empresariales, 30th July. Concertación, ‘Programa de Gobierno – Concertación de Partidos por la Democracia’, la Concertación, The Aylwin Government’s Programme of Government, 1989. Cortázar, R. (1993) Política Laboral en el Chile Democrático – Avances y Desafíos en los Noventa, Santiago: Ediciones Dolmen. CUT – Pamphlet (1989) ‘Propuesta para la transición a la democracia’, Santiago: CUT, April. CUT – Pamphlet (1991) ‘Propuestas para el debate’, Primer Congreso Nacional Ordinario – unidad para reforzar la democracia, CUT, 28–30, Santiago: CUT, October. Dahl, R. (1971) Polyarchy: Participation and Opposition, New Haven, Conn.: Yale University Press. Dahl, R. (1985) A Preface to Economic Democracy, Cambridge: Polity Press. Dahrendorf, R. (1959) Class and Class Conflict in Industrial Society, Palo Alto, Calif.: Stanford University Press. de Vylder, S. (1997) ‘Is the Chilean Economy finally becoming Coherent ?’, Paper presented at the Annual Conference of the Society for Latin American Studies, 4–6 April 1997 at the University of St. Andrews. Díaz, A. (1997) ‘Chile: Neoliberal Policy, Socioeconomic Reorganization, and Urban Labor Market’, in R. Tardanico and R. M. Larín (eds), Global Restructuring, Employment, and Social Inequality in Urban Latin America, Miami: North-South Center Press.

Select Bibliography

267

Donzelot, J. (1977) The Policing of Families – Welfare Versus the State, London: Hutchinson. Echeñique, J. (1990a) ‘El Sistema Alimentario. Políticas de Desarrollo para Chile’, in Chile: Inversión para el Crecimiento Equitativo, Santiago, Chile: Prealc, ILO. Echeñíque, J. (1990b) ‘Las Dos Caras de la Agricultura y las Políticas Posibles’, in Proposiciones – Chile, Sociedad y Transición, No. 18, Chile, pp. 145–58. Eggertson, T. (1990) Economic Behaviour and Institutions, Cambridge: Cambridge University Press. Epstein, E. C. (1992) ‘Labor and Political Stability in the New Chilean Democracy: Three Illusions’, Paper presented at the International Congress of the Latin American Studies Association, Los Angeles, 24–27 September. Epstein, E. C. (1997) ‘Chilean Labor and Labour Rights under the Free Trade Area of the Americas’, Paper presented at the International Congress of the Latin American Studies Association, Guadalajara, Mexico. April 17–19. Esping-Andersen, G. (1985) Politics Against Markets: The Social Democratic Road to Power, Princeton: Princeton University Press. Evans, P. B. (1985) ‘Transnational Linkages and the Economic Role of the State: An Analysis of the Developing and Industrialised Nations in the Post-World War II. Period’, in P. B. Evans, D. Rueschemeyer and T. Skocpol (eds), Bringing the State Back In, Cambridge: Cambridge University Press. Evans, P. B. (1987) ‘Class, State and Dependence in East Asia: Some Lessons for Latin Americanists’, in F. C. Deyo (ed.), The Political Economy of the New Asian Industrialism, Ithaca, N.Y.: Cornell University Press. Evans, P. B. (1994) ‘Predatory, Developmental, and Other Apparatuses: A Comparative Political Economy Perspective on the Third World State’, in A. Douglas Kincaid and A. Portes (eds), Comparative National Development – Society and Economy in the New Global Order, Chapel Hill and London: The University of North Carolina Press. Evans, P. B. (1995) Embedded Autonomy – States and Industrial Transformation, XXX. Ffrench-Davies, R. and Sáez, R. (1995) ‘Comercio y Desarrollo Industrial en Chile’, Colección Estudios CIEPLAN, 41 (Santiago: CIEPLAN). Frank, V. K. (1995) ‘Plant Level Leaders, the Union Movement, and the Return to Democracy in Chile’, Unpublished Ph.D. dissertation, Department of Sociology, University of Notre Dame, Indiana, August. Franzosi, R. (1995) The Puzzle of Strikes – Class and State Strategies in Postwar Italy, Cambridge University Press. Freeman, R. B. and Medoff, J. (1984) What Do Unions Do?, New York: Basic Books. Gallie, D. (1983) Social Inequality and Class Radicalism in France and Britain, Cambridge University Press. Garon, Sheldon (1997) Molding Japanese Minds – The State in Everyday Life, Princeton, NJ: Princeton University Press. Gerschenkron, A. (1966a) Bread and Democracy in Germany, New York: Howard Fertig. Gerschenkron, A. (1996b) Economic Backwardness in Historical Perspective – A Book of Essays, Cambridge, Mass: The Belknap Press of Harvard University Press. Graham, C. (1991) From Emergency Employment to Social Investment: Alleviating Poverty in Chile, Washington DC: Brookings Institution Occasional Papers, Brookings Institution.

268 Select Bibliography Graham, C. (1994) Safety Nets, Politics and the Poor – Transitions to Market Economies, Washington DC: Brookings Institution. Graham, H. D. and Gurr, T. R. (1969) (eds) Violence in America; Historical and Comparative Perspectives, New York: FA Praeger. Granovetter, M. (1985) ‘Economic Action and Social Structure: The Problem of Embeddedness’, American Journal of Sociology, 91: 3, pp. 481–510, November. Grimshaw, D. and Rubery, J. (1998) ‘Integrating the Internal and External Labour Markets’, Cambridge Journal of Economics, 22, pp. 199–220. Gurr, T. R. (1969) ‘A Comparative Study of Civil Strike’, in H. D. Graham and T. R. Gurr (eds), Violence in America; Historical and Comparative Perspectives, New York: Praeger. Haagh, L. (1998) ‘Redemocratisation, Labour relations and the Development of Human Resources in Chile (1990–1993), D.Phil. Thesis, University of Oxford. Haagh, L. (1999) ‘Training Policy and the Property Rights of Labour in Chile (1990–1997): Social Citizenship in the Atomised Market Regime’, Journal of Latin American Studies, 31, pp. 429–72. Haagh, L. (2000a) ‘Social Policy and De-centralisations – The Liberal Model in Chile Compared’, Working Paper LAC-02-2000, The Latin American Centre, University of Oxford. Haagh, L. (2000b) ‘The Politics of Employment Policy in Korea’, Paper presented to the Brazil–Korea Conference at the Institute of Latin American Studies, University of London, 9–11 December 2000. Haagh, L. (2001) ‘The Challenge of Labour Reform in Korea: A review of Contrasting Approaches to Market Enhancement’, in Labor Market Reforms and Financial Crisis in Korea, The World Bank. Haagh, L. (2002a) ‘The Emperor’s New Clothes – Labour Reform and Social Democratisation in Chile’, Studies in International Comparative Development, University of Berkeley, 37: 1, May 2002. Haagh, L. (2001b) ‘Social Policy and Themes in Market Governance in Latin America’, in L. Haagh and C. Helgø, Social Policy Reform and Market Governance in Latin America. Haagh. L. (forthcoming) (2001c) ‘Human Resources and De-centralisation in Chile’, in L. Haagh and C. Helgø, Social Policy Reform and Market Governance in Latin America. Haagh, L. and Helgø, C. (forthcoming) (2001a) Social Policy Reform and Market Governance in Latin America. (London: Palgrave, St Antony’s) Haddad, W. D., Carnoy, M., Rinaldi, R. and Regel, O. (1990) ‘Education and Development – Evidence for New Priorities’, World Bank Discussion Papers No. 95. Hanami, T. A. (1983) ‘The Function of the Law in Japanese Industrial Relations’, in T. Shirai (ed.), Contemporary Industrial Relations in Japan, Madison, Wisc.: University of Wisconsin Press. Hanami, T. and Blanpain, R. (1987) ‘Introductory Remarks and a Comparative Overview’, in Industrial Conflict Resolution in Market Societies. Hashimoto, M. and Raisian, J. (1985) ‘Employment Tenure and Earnings Profiles in Japan and the United States’, The American Economic Review, September, pp. 721–35. Hayek, F. A. von (1960) The Constitution of Liberty, London: Routledge and Kegan Paul.

Select Bibliography

269

Hayek, F. A. von (1984) 1980s Unemployment and the Unions – The Distortion of Relative Prices by Monopoly in the Labour Market, Hobart Papers No. 87, London: The Institute of Economic Affairs. Hayek, F. A. von (1986) The Road to Serfdom, London: Ark Paperbacks. Hayek, F. A. von (1988) The Fateful Conceit: The Errors of Socialism, edited by W. W. Bartley, London: Routledge. Helgø, C. (forthcoming) ‘Market Oriented Education Reforms and Social Inequalities Among the Young Population in Chile’, in L. Haagh and C. Helgø, Social Policy Reform and Market Governance in Latin America. Henríquez, L. (1992) ‘Oh! Esos falsos dilemas y la modernización de las pymes’, El Diario Financiero, 11 December, p. 2. Hirschman, A. O. (1958) The Strategy of Economic Development, New Haven, Conn.: Yale University Press. Hirschman, A. O. (1963) Journeys Towards Progress – A Study of Economic PolicyMaking in Latin America, New York: Twentieth Century Fund. Hirschman, A. O. (1970) Exit, Voice and Loyalty, Cambridge University Press. Hirschman, A. O. (1993) ‘The Rhetoric of Reaction – Two Years Later’, Government and Opposition, 28: 3, pp. 292–314. Hirsh, B. T. (1992) ‘Firm Investment Behaviour and Collective Bargaining Strategy’, Industrial Relations, 31: 1, pp. 95–121, Winter. Hutton, W. (1995a) The State We’re In, Reading: Vintage, Cox and Wyman. Hutton, W. (1995b) ‘The German Model’, Paper presented at the Conference on Different Models of Capitalism, London School of Economics, 15 November. ICARE, CPC (1989–91) Annual publications of the national business gathering, Encuentro Nacional de la Empresa, ENADE ILO (1990) International Labour Standards – A Workers’ Education Manual, Geneva: ILO. ILO (1992) Booklet Memoria, Conference of American States, June. ILO (1998) Chile – Crecimineto, empleo y el desafío de la justicia social, Geneva: International Labour Organisation. ILO (Various years) World Labour Report. Infante, R. and Klein, E. (1992) ‘Chile – Transformaciones del Mercado Laboral y sus Efectos Sociales: 1965–1990’, PREALC – World Employment Programme, ILO, Working paper No. 368, October. Inter-American Development Bank (1994) Modernizar con Todos: Hacia la Integración de lo Social con lo Económico en Chile, Informe de la Comisión Piloto del Programa Reforma Social del Banco Inter-Americano del Desarrollo, Grupo de Agenda Social, Washington: IADB, January. Johnson, C. (1982) MITI and the Japanese Miracle: The Growth of Industrial Policy, 1925–1975, Stanford, Calif.: Stanford University Press. Johnson, C. (1987) ‘Political Institutions and Economic Performance: The Government–Business Relationship in Japan, South Korea and Taiwan’, in F. C. Deyo, The Political Economy of the New Asian Industrialism, Ithaca, NY and London: Cornell University Press. Johnson, C. (1995) Japan: Who Governs? The Rise of the Developmental State, New York: W.W. Norton. Kaufman, R. R. (1986) ‘Liberalisation and Democratisation in South America: Perspectives from the 1970s’ in Venezuela’, in G. O’Donnell, P. Schmitter and

270 Select Bibliography L. Whitehead (eds), Transitions from Authoritarian Rule – Comparative Perpectives, Baltimore and London: The Johns Hopkins University Press. Koike, K. (1983) ‘Internal Labour Markets: Workers in Large Firms’, in T. Shirai (ed.), Contemporary Industrial Relations in Japan, Madison, Wisc.: University of Wisconsin Press. Kongshøj-Madsen, P. (1999) ‘Denmark: Flexibility, Security and Labour Market Success’, Employment and Training Papers 53, ILO. Korpi, W. (1983) The Democratic Class Struggle, London: Routledge. Koshiro, K. (1983) ‘Development of Collective Bargaining in Postwar Japan’, in T. Shirai, (ed.), Contemporary Industrial Relations in Japan, Madison, Wisc.: University of Wisconsin Press. Krueger, A. O. (1983) Trade and Employment in Developing Countries, Vol. 3: Synthesis and Conclusions, Chicago University Press. Labarca, O. (1996) ‘Education in Basic Skills and Training for Productive Work’, CEPAL Review, 59. Labour Directorate of Chile (1992a) Unpublished list of unions from Direcció de Trabajo, 26. June. Labour Directorate of Chile (1992b) Internal Memo, ‘Finiquítos y Avisos de Terminación de Contratos de Trabajo’ and data-base data print-outs. Lagos, R. A. (1994) ’Qué se Entiende por Flexibilidad del Mercado de Trabajo?’, Revista de la CEPAL, 54, pp. 81–95, December. Laski, H. (1936) The Rise of European Liberalism: An Essay in Interpretation (London: George Allen & Unwin). Leggett, C. (1993) ‘Corporatist Trade Unionism in Singapore’, in S. Frenkel (ed.), Organised Labor in the Asia-Pacific Region, Ithaca, NY: ILR Press. Lindsay, A. D. (1940) ‘Introduction’ to J. S. Mill compilation, Utilitarianism, Liberty and Representative Government, Everymans Library No. 482, London: J. M. Dent & Sons. Linz, J. (1978) The Breakdown of Democratic Regimes: Crisis, Breakdown and Re-equilibrium, Baltimore, Md: Johns Hopkins University Press. Linz, J. and Stepan, A. (1996) Problems of Democratic Transition and Consolidation, Baltimore, Md: Johns Hopkins University Press. Lorenz, E. H. (1992) ‘Trust and the Flexible Firm: International Comparisons’, Industrial Relations, 321: 3, pp. 455–73, Fall. Macneil, I. R. (1978) ‘Contracts: Adjustments of long-term Economic Relations under Classical, Neo-Classical, and Relational Contract Law’, in Northwestern University Law Review, No. 72, pp. 854–906. Macneil, I. R. (1980) The New Social Contract: An Inquiry into modern Contractual Relations, New Haven: Yale University Press. Macpherson, C. B. (1973) Democratic Theory, Oxford: Clarendon Press. Mainwaring, S., O’Donnell G. and Valenzuela, S. (eds) (1992) Issues in Democratic Consolidation, Notre Dame, Ind.: University of Notre Dame Press. Marfán, M. and Bosworth, B. P. (1994) ‘Saving Investment and Economic Growth’, in B. P. Bosworth, R. Dornbusch and P. Labal (eds), The Chilean Economy – Policy Lessons and Challenges, Washington DC: The Brookings Institution. Marglin, S. A. and Schor, J. B. (eds) (1990) The Golden Age of Capitalism, Oxford: Clarendon Press. Marín, G. and Rozas, P. (1988) Los Grupos Transnacionales y la Crísis, el Caso Chileno, Buenos Aires: Nueva América.

Select Bibliography

271

Marshall, A. (1966) Principles of Economics – An Introductory Volume (8th edn – first published 1890), London/Melbourne/Toronto: Macmillan. Marshall, T. H. (1949) Citizenship and Social Class, Cambridge University Press. Marshall, T. H. and Bottomore, T. (1992) Citizenship and Social Class, London: Pluto. Marx, K. and Engels, F. (1968) Selected Works, London: Lawrence and Wishart. Marx, K. and Engels, F. (1975) Collected Works, London: Lawrence and Wishart. Matsuda, Y. (1983) ‘Conflict Resolution in Japanese Industrial Relations’, in T. Shirai (ed.), Contemporary Industrial Relations in Japan, Madison, Wisc.: University of Wisconsin Press. Maurice, M., Sellier, F. and Silvestre, J.-J. (1986) The Social Foundations of Industrial Power – A Comparison of France and Germany, Cambridge, Mass. and London: MIT Press. Mill, J. S. (1848) Principles of Political Economy, vol.11. Mill, J. S. (1910) Utilitarianism; Liberty; and, Representative Government, London, New York: Dent, Dutton. Mill, J. S. (1982[1859]) On Liberty, The Penguin English Library, Richard Clay (The Chaucer Press) Bungay, Suffolk. Miller, D. (1990) Market, State and Community: Theoretical Foundations of Market Socialism, Oxford: Clarendon Press. Ministry of Labour, Chile (1990) Memo, ‘Memorandum: Principales Controversias respecto de las Modificaciones a la Legislación Laboral. Proyecto: Negociación Colectiva’, Ministry of Labour, Santiago, July. Ministry of Labour, Chile (1997) Legislative project for PROTRAC (Sistema de Protección al Trabajador Cesante), 7 July. Ministry of Labor, Korea (1999) Statement on Employment Insurance in Korea. Mizala, A. (1992) ‘Las reformas económicas de los años setenta y la indústria manufacturera Chilena’, Colección Estudios Cieplan, Número Especial, 35, pp. 153–99, September. Montero, C. (1990) ‘La Evolución del Empresariado Chileno: Surge un Nuevo Actor?’, in Colección Estudios Cieplan, No. 30, pp. 91–122, December. Montero, C. (1993) ‘El actor empresarial en transición’, Colección Estudios Cieplan, 37, pp. 37–68, June. Morishima, M. (1991) ‘Information Sharing and Firm Performance in Japan’, Industrial Relations, 30: 1, pp. 37–61, Winter. Muñoz, O. (1988) ‘El Estado y los Empresarios: Experiencias Comparadas y sus Implicancias para Chile’, Colección Estudios Cieplan, No. 25, pp. 5–54, December. Nelson, J. M. (1992) ‘Poverty, Equity, and the Politics of Adjustment’, in The Politics of Economic Adjustment, Princeton, N.J: Princeton University Press. Nitta, M. (1989) ‘Conflict Resolution in the Steel Industry’, in T. Hanami and R. Blanpain (eds), Industrial Conflict Resolution in Market Societies, Deventer: Kluwer. O’Donnell, G. (1986) ‘Introduction to the Latin American Cases’, in G. O’Donnell, P. C. Schmitter, and L. Whitehead (eds), Transitions from Authoritarian Rule – Latin America, Baltimore, Md: Johns Hopkins University Press. O’Donnell, G. (1992) ‘Transitions, Continuities and Paradoxes’, in S. Mainwaring, G. O’Donnell and S. Valenzuela (eds), Issues in Democratic Consolidation, Notre Dame, Ind.: University of Notre Dame Press. O’Donnell, G. and Schmitter, P. (1986) The Transition to Democracy: Tentative Conclusions about Uncertain Democracies, (Part IV), G. O’Donnell, P. Schmitter and L. Whitehead (eds), Baltimore, Md: Johns Hopkins University Press.

272 Select Bibliography OECD (1986a) Labour Market Flexibility – Report by a High-Level group of Experts to the Secretary-General, Paris. OECD (1986b) Flexibility in the Labour Market – The Current Debate – A Technical Report, Paris: OECD. OECD (1992) Technology and the Economy – The Key Relationships, Paris: OECD. Ogawa, N., Jones, G. W. and Williamson, J. G. (eds) (1993) Human Resources in Development along the Asia-Pacific Rim, Oxford: Oxford University Press. Olson, M. (1993) ‘Dictatorship, Democracy, and Development’, American Political Science Review, 87: 3, pp. 557–76, September. Ominami, C. and Madrid, R. (1989) La inserción de Chile en los mercados internacionales, Santiago, Chile: Prospel. Ominami, C. and Madrid, R. (1990) ‘Chile: Elementos para la Evaluación del Desarrollo Exportador’, in Proposiciones, No. 18, pp. 120–44, Ediciones SUR, Santiago, Chile, January. Oxhorn, P. (1995) ‘Neo-Liberal Economic Reform and Disarticulation of Civil Society in Latin America’, Paper Presented at the Conference, ’Alternativas de Izquierda al Neoliberalismo’, Havana, Cuba, 11–13 February. Pagano, U. (1985) Work and Welfare in Economic Theory, Oxford/New York: Basil Blackwell. Pagano, U. (1991) ‘Property Rights, Asset Specificity, and the Division of Labour under Alternative Capitalist Relations’, Cambridge Journal of Economics, 15, pp. 315–42. Pagano, U. and Rowthorn, R. (1996) Democracy and Efficiency in the Enterprise, Routledge Studies in Business Organisation and Networks, USA/Canada: Routledge/Helsinki: UNO World Institute for Development Economics Research. Palma, G. (1997) ‘When Workers Save Nothing and Capitalists Consume Everything’ – Was Kaldor Right about why Latin America had such a poor savings performance?, in P. Arestis, G. Palma and M. Sawyer (eds), Capital Controversy, Post-Keynesian Economics and the History of Economics Essays in Honour of Geoff Harcourt, vol. I, London: Routledge. Park, Y.-B. (1991) ‘Union/Nonunion Wage Differentials in the Korean Manufacturing Sector’, International Economic Journal, 5: 4, pp. 79–91, Winter. Park, Y.-B. (1994) ‘State Regulation, The Labour Market and Economic Development: The Republic of Korea’, in G. Rodgers (ed.), Workers’ Institutions and Economic Growth in Asia, Geneva: International Institute for Labour Studies. Pérez Mackenna, F. (1990) ‘Mantención del Despido sin Necesidad de tener que Probar alguna Causal Específica’, in Puntos de Referencia, CEP, Number 46, March. Pietrobelli, C. (1998) Industry, Competitiveness and Technological Capabilities in Chile: A New Tiger from Latin America?, Basingstoke: Macmillan – now Palgrave. Piore, Michael J. (1986) ‘Perspectives on Labor Market Flexibility’, Industrial Relations, 25: 2, Spring, pp. 146–66. Pirrie, M. (1995) ‘The Anglo-Saxon Model’, Paper presented at the Conference on Different Models of Capitalism, London School of Economics, 15 November. Polanyi, K. (1957 [1944]) The Great Transformation, Boston, Mass.: Beacon Press. Przeworski, A. (1986) ‘Some Problems in the Study of Transitions to Democracy’, in G. O’Donnell, P. Schmitter, and L. Whitehead, (eds), Transitions from Authoritarian Rule – Comparative Perspectives, Baltimore and London: The Johns Hopkins University Press.

Select Bibliography

273

Putterman, L. (1986) The Economic Nature of the Firm – A Reader, Cambridge: Cambridge University Press. Putterman, L. (1990) Division of Labor and Welfare – An Introduction to Economic Systems, Oxford University Press. Rodríguez, D. and Venegas, S. (1989) ‘Los empresarios frutículas Chilenos: Origenes, estilo de vida, opiniones’, GEA Documents, No. 14, Santiago, Chile: GEA. Rueschemeyer, D., Stephens, E. H. and Stephens, J. D. (1992) Capitalist Development and Democracy, Cambridge: Polity Press. Sartori, G. (1976) Parties and Party Systems: A Framework for Analysis, New York: Cambridge University Press. Sartori, G. (1987b) The Theory of democracy Revisited – Part Two: The Classical Issues, New Jersey: Chatham House Publishers. Scholten, I., ‘Introduction’, in Political Stability and Neo-Corporatism: Corporatist Integration and Societal Cleaveages in Western Europe, XXX Schull, J. (1992) ‘What is Ideology? Theoretical Problems and Lessons from Soviet-Type Societies’, Political Studies, XL, pp. 728–41, 732. Schultz, T. W. (1993) The Economics of Being Poor, Oxford/Cambridge, Mass.: Basil Blackwell. Schultze, C. L. (1985) ‘The Macroeconomic Costs of Microeconomic Efficiency’, American Economic Review, pp. 1–35, March. Scott, C. D. (1995) ‘The Distributive Impact of the New Economic Model in Chile’, Revised version of paper presented to the Conference on ‘The New Economic Model in Latin America and its Impact on Income Distribution and Poverty’, London, 11–12 May. Scruton, R. (1982) A Dictionary of Political Thought, London: Pan Books Macmillan. Selamé, T., ‘Evolución Tecnológica: Su Impacto sobre las Ocupaciones y el Perfil Educational de la Fuerza de Trabajo’, Materiales de Discusión, No. 1, Centro de Estudios Sociales, Santiago, Chile, December 1986. Selvaratnam, V. (1991) ‘Innovations in Higher Education – Singapore at the Competitive Edge’, World Bank Technical Paper, Asia Technical Department Series. Sen, A. (1991) Markets and Freedoms, Development Economics Research Programme, Suntory–Toyota International Centre for Economics and Related Disciplines, London School of Economics, DEP No. 31, September. Sen, A. (1992) Inequality Reexamined, Oxford: Clarendon Press. Sen, A. (1994) Beyond Liberalisation: Social Opportunity and Human Capability, London School of Economics, DEP No. 58, November. Sen, A. (1997) What’s the Point of a Development Strategy?, Development Economics Research Programme, Suntory–Toyota International Centre for Economics and Related Disciplines, London School of Economics, DERP No. 3, April. SENCE (1992a) ‘Ranking de Cursos Ejecutados’, SENCE statistics, The Labour Ministry, Chile. SENCE (1992b) ‘Programa de Capacitación Laboral de Jovenes – Resultado del Tercer Llamado a Licitación Pública del Subprograma Capacitación y Experiencia Laboral en Empresas’ SENCE, Ministerio del Trabajo, Chile, Supplement to La Nación 7 July 1992, pp. 25–8. SENCE (1992c) Unpublished statistics, various, The Labour Ministry, Labour Ministry, Chile.

274 Select Bibliography SENCE (1992d) ‘Los Jovenes – Prioridad del Gobierno de Chile. Programa de Capacitación Laboral de Jovenes’, 29 Reunión de la Comisión Técnica Cinterfor-OIT, Rio de Janeiro, August. SENCE (1992e) Subprograma Capacitación y Experiencia Laboral en Empresas. 4. Llamado – August. SENCE (1993a) The Labour Ministry, Chile, Apartado Estadístico No. 1, April. SENCE (1993b) The Labour Ministry, Chile, Apartado Estadístico No. 2, April. SENCE (1997) Boletín Estadístico 1994–1996, Servicio Nacional de Capacitación y Empleo, SENCE, Santiago, Chile. Senghaas, Dieter (1985) The European Experience: A Historical Critique of Developmental Theory, trans. from German by K. H. Kimmig, Leamington Spa/Berg. SERCOTEC (Servicio de Co-operación Técnica – Economy Ministry, Chile) (1990) Antecedentes sobre la Capacitación Empresarial y Laboral en la Pequeña y Mediana Indústria y Artesanado, Document No. 9, Santiago, March. Sharma, B. and Luh Luh, L. (1994) ‘Labour Market Flexibilities as HRM Strategies: The Example of Singapore’, Journal of Asian Business, 10: 1. Shirai, T. (1983a) ‘A Theory of Enterprise Unionism’, in T. Shirai (ed.), Contemporary Industrial Relations in Japan, Madison: University of Wisconsin Press. Shirai, T. (1983b) ‘A Supplement: Characteristics of Japanese Managements and Their Personnel Policies’, in T. Shirai, (ed.), Contemporary Industrial Relations in Japan, Madison: University of Wisconsin Press. Silva, P. (1998), The Politics of Expertise in Latin America, New York: St. Martin’s Press. Skocpol, T. (1979) States and Social Revolutions, Cambridge/New York: Cambridge University Press. Skocpol, T. (1985) ‘Bringing the State Back In: Strategies of Analysis in Current Research’, in Peter B. Evans, D. Rueschemeyer and T. Skocpol (eds), Bringing the State Back In, Cambridge University Press. Solow, R. (1957) ‘Technical Change and the Aggregate Production Function’, Review of Economics and Statistics, 39. Solow, R. (1990) The Labour Market as a Social Institution, Cambridge, Mass./ Oxford: Basil Blackwell. Sport Wagon (1991a) Convenio Colectivo, between the company Sport Wagon and workers, 7 November. Sport Wagon (1991b) Proyecto de Contrato Colectivo de Trabajo, Sindicato de Trabajadores de la Empresa “Sport Wagon”, 17 December. Sport Wagon (1991c) Letter to the President, Secretary and Treasurer of the union, from Adolfo Nudman Lerner, Owner of Sports Wagon, 27 December. Sport Wagon (1992) Letter to the President, Secretary and Treasurer of the union, from Adolfo Nudman Lerner, Owner of Sports Wagon, 21 January. Standing, G. (1990) ‘Labour Flexibility and Insecurity: Towards an Alternative Strategy’, in R. Brunetta and C. Dell’Aringa (eds), Labour Relations and Economic Performance, London: Macmillan – now Palgrave. Standing, G. (xxxx) ‘Do Unions Impede or Accelerate Structural Adjustment? Industrial versus Company Unions in an Industrialising Labour Market’, in Cambridge Journal of Economics, No. 16, pp. 327–354. Streeck, W. (1989) ‘Skills and the Limits of Neo-Liberalism: The Enterprise of the Future as a Place of learning’, Work, Employment and Society, 3: 1, March, pp. 105–18.

Select Bibliography

275

Taylor, C. (1985) Philosophy and the Human Sciences – Philosophical Papers 2, Cambridge University Press. Tendler, J. (1999) ‘The Rise of Social Funds: What Are They a Model Of?’, draft MS, Cambridge Mass.: MIT Press, January. Thayer, Arteaga, W. (1991) Sindicatos y Negociación Colectiva – Análisis de la Nueva Legislación Ley No. 19.069., Colección Manuales Jurídicos, Santiago de Chile. Therborn, G. (1990) ‘Lessons from Corporatist Theorisations’, in J. Pekkarinen, M. Pojhola and B. Rowthorn (eds), Social Corporatism – A Superior Economic System?, Oxford University Press. United Nations (1996) Human Development Report 1996 – Growth for Human Development, UNDP, Oxford: Oxford University Press. Valenzuela, A. (1978), The Breakdown of Democratic Regimes – Chile, Baltimore and London: Johns Hopkins University Press. Valenzuela, A. (1989) ‘Chile: Origins, Consolidation and Breakdown of a Democratic Regime’, in L. Diamond and Juan J. Linz (eds), Democracy in Developing Countries – Latin America, Col., Lynne Rienner. Valenzuela, J. S. (1988) ‘Labour Movements in Transitions to Democracy: A Framework for Analysis’, Kellogg Institute for International Studies, Working Paper No. 104, June. Valenzuela, J. S. (1992) ‘Democratic Consolidation in Post-Transitional Settings: Notions, Process, and Facilitating Conditions’, in S. Mainwaring, G. O’Donnell and S. Valenzuela (eds), Issues in Democratic Consolidation, Notre Dame, Ind.: University of Notre Dame Press. Valenzuela, J. S. (1996) ‘Why No Social Democracy in Chile: 1880–1940’, Talk given as part of the Latin American Seminar Series, St Antony’s College, Oxford, 15 November. Valenzuela, J. S. and Frank V. (1992) ‘The Labor Movement and the Return to Democratic Government in Chile’, Latin American Labor News, 5, pp. 9–11. Weiss, M., Simitis, S. and Rydzy, W. (1989) ‘The Settlement of Labour Disputes in the Federal Republic of Germany’, in T. Hanami and R. Blanpain (eds), Industrial Conflict Resolution in Market Societies, Deventer: Kluwer. Williamson, J. G. (1993) ‘Human Capital Deepening, Inequality, and Demographic Events along the Asia-Pacific Rim’, in N. Ogawa, G. W. Jones and J. G. Williamson (eds), Human Resources in Development along the Asia-Pacific Rim, Singapore: Oxford University Press. Williamson, O. (1985) The Economic Institutions of Capitalism, New York: The Free Press. Williamson, O. E. (1996) Industrial Organisation, Aldershot: Elgar. Williamson, O. E. (ed.) (1995) Organisation Theory – From Chester Bernard to the Present and Beyond, Oxford: Oxford University Press. Williamson, O., Wachter, M. and Harris, J. (1986) ‘Understanding the Employment Relation: The Analysis of Idiosyncratic Exchange’, in L. Putterman (ed.), The Economic Nature of the Firm, Cambridge University Press. World Bank (1980) ‘Indonesia – Employment and Income Distribution in Indonesia’, World Bank Country Study, Washington DC: World Bank. World Bank (1987) ‘Korea – Managing the Industrial Transition’, vol. II, World Bank Country Study, Washington DC: World Bank. World Bank (1991) ‘Vocational and Technical Education and Training’, World Bank Policy Paper, Washington DC: World Bank.

276 Select Bibliography World Bank (1993) Skills for Productivity – Vocational Education and Training in Developing Countries, J. Middleton, A. Ziderman and A. van Adams, Oxford: Oxford University Press World Bank. World Bank (1992, 1995, 1997, 1999) World Development Report. Zeitlin, M. (1984) The Civil Wars in Chile (or the Bourgeois Revolutions That Never Were), Princeton, NJ: Princeton University Press. Zeitlin, M. and Radcliff, R. E. (1988) Landlords and Capitalists: The Dominant Class of Chile, Princeton, NJ: Princeton University Press.

List of Interviews

Labour leaders Aguírre, Sergio, General Secretary of the Maritime Workers’ Union (CONGEMAR), and third Vice-President of the CUT (leading Communist Party figure in the CUT). (a) 19 May 1992. (b) 23 May 1992. Ahumada, Manuel, President of the National Catering Union (affiliated with the Communist Party), Santiago, 14 May 1992. Arco, Humberto, labour leader from the metallurgical sector and labour advisor at the Centre for Social Studies (CES), Santiago, 24 January 1992. Barilari, Enríque, President of the ‘Coordinadora’ of municipal health workers, Santiago, 15 July 1992. Bustos, Manuel, President of the CUT (affiliated with the Christian Democratic Party), Santiago, 15 September 1992. Carrizo, Leonor, member of the executive council of the public sector Health Workers’ Union (FENATS), Federación Nacional de Trabajadores del Salud, (affiliated with the Socialist Party), Santiago, 16 July 1992. Coñoman, Patricia, President of the Union of Textile Workers (CONTEXTIL), (affiliated with the Communist Party), Santiago, 20 May 1992. Cortéz, Guillermo, President of the Bakery Workers’ Federation (CONAPAN), and General Secretary of the CUT (affiliated with the Socialist Party), Santiago, 26 August 1992. Corvalan, Juán, representative of the Fruit Workers’ Association (CUOC), Santiago, 5 June 1992. Garay, Carlos, President of the Confederation of Bank Workers (CTB), (affiliated with the Christian Democratic Party), Santiago, (a) 26 May 1992. (b) 16 July 1992. Garrido, Carlos, President of the union of the CMPC company, Santiago, 7 May 1992. Jara, José Abello, President of the Confederation of Forestry Workers (CTF), and two executive committee members, Juán Miranda and Jorge Gonzales, Santiago, 4 April 1992. Matus, Luís, President of the Supervisor’s Union of the CMPC, Santiago, 20 May 1992. 277

278 List of Interviews Requena, Raúl, Secretary of the Association of Bank Employees (CTB), head of the CUT’s Commission for the Economy, Employment and Productivity, and member of the CUT Executive Council (affiliated with the Socialist Party), Santiago. (a) 17 July 1992. (b) 24 July 1992. (c) 13 August 1992. (d) 27 August 1992. (e) 4 September 1992. Rojas, Sergio, President of the Public Workers’ Central (CEPCH), Santiago, 1 April 1992. Soto, Miguel, President of the Metal Workers’ Confederation (CONSTRAMET), (affiliated with Communist Party), Santiago. (a) 20 January 1992. (b) 27 January 1992. (c) 31 January 1992. Members of the executive of CONSTRAMET: Humberto Arco, Santiago, 24 January 1992: Capelo, Alburquenque, Chavez, Arco, and Moraga, Santiago, 3 February 1992. Members of the executive of the Confederation of Textile and Clothing Workers’ in Chile (CONTEVECH), including Oscar Cáceres, Santiago, 11 March 1992; and, Margarita Carrízo, Octavio Díaz, Raúl Sepúlveda and German Cortéz, Santiago, 18 March 1992. In-depth interviews with local leaders March–April 1992, including Jaime Gonzales, Maurício Navarro, Manuel Osorio, Elba López, José San Martin, Gloria Andrade, José Abello Jara, Oscar Cáceres, Patrícia Coñioman, Margarita Carrízo, and Raúl Sepulveda, among others.

Labour advisors Alburquerque, Mario, member of CIASI, Santiago, 14 August 1992. Arrieta, Adolfo, CIASI economist, Santiago, 7 April, 1992. Bonifáz, Rodolfo, CIASI economist, Santiago, 15 April 1992. Díaz, Eugenio, Director of CIASI, CUT advisory team, Santiago, (a) 6 January 1992. (b) 13 January 1992. (c) 15 January 1992. (d) 3 March 1992. (e) 23 September 1992. Ester-Féres, María, CIASI’s key advisor on legal matters to the CUT, Santiago, 3 April 1992.

List of Interviews 279

Entrepreneurial leaders Alé, Jaime, Director of Studies, SOFOFA, Santiago, (a) 8 April 1992. (b) 4 September 1992. Feliú, Manuel, President of the CPC (1986–1990), Santiago, 14 September 1992. Fantuzzi, Angel, deputy for Renovación Nacional (RN) and entrepreneurial leader from the metal sector, (a) 20 August 1992, Valparaíso. (b) 28 August 1992, Santiago. García, Raúl, Vice-President of the SNA and head of the CPC’s commission for labour relations, Santiago, 26 May 1992. Garib, Faud, textile entrepreneur and head of the textile tripartite commission during the Allende Government (1970–1973), and, in 1992, member of the SOFOFA General Council, Santiago, 30 July 1992. Guzmán, José-Antonio, President of the CPC, Santiago, 1 April 1992. Heiremans, Eugenio, head of the insurance holding AchS, and prominent member of the SOFOFA and the CPC, Santiago. (a) 15 June 1992. (b) 21 July 1992. (c) 29 July 1992. Paredes, Jaime, President of the Metal Association (ASIMET), Santiago, 28 July 1992. Lizana, Pedro, Vice-President of the SOFOFA, Santiago, 8 September 1992. Zaballa, José, head of the Social Union of Christian Employers (USEC), Santiago, 15 June 1992.

Members of the political right Otero, Miguel, Senator and Vice-President of Renovación Nacional, (a) 20 August 1992, Valparaíso. (b) 21 August 1992, Santiago. Thayer, William, Constitutional Senator (1990–) and Minister of Labour under the Frei Government (1964–1970), Santiago, 19 August 1992.

Government officials and party leaders Campero, Guillermo, Head of Cabinet, the Labour Ministry, Santiago, (a) 4 February 1992. (b) 26 March 1992. (c) 8 May 1992. (d) 13 August 1992. (e) 29 January 1992.

280 List of Interviews Cerda, Mario, Director of the Labour Ministry’s training programmes, and of SENCE, Santiago, 24 August 1992. Correa, Enríque, Minister of Government, Santiago, 4 September 1992. Cortázar, René, Labour Minister under the Aylwin Government (1990–1993), Santiago, 25 September 1992. Di Giorgio, José Ruiz, labour leader from the petrol sector, Senator for the PDC and formerly Vice-President of the PDC and the labour central CNT, Valparaíso, 19 August 1992. Escalona, Camilo, member of the Chamber of Deputies for the Socialist Party, Valparaíso, 20 August 1992. Schnake, Erich, President of the PPD, Santiago, 3 September 1992. Vidaure, Advisor to Rolando Calderon, member of the Senate for the Socialist Party, 19 August 1992.

Index Abramo, L., 60 Addison, J. T. and Chilton, J. B., 98, 213 Agresti, A. and Finlay, B., 213 Alburquerque, M., 65 Alé, Jaime, 102 Alienation, 26–7 Angell, A., 18, 28, 38, 64, 90, 150, 179, 195–6, 215 Apprenticeship: in Chile, 31, 127; and private training corporation (PTC) markets, 127; and institutional networks, 31; compared with Germany and Singapore, 31, 141; and dual education model, 132, 135–6; and youth training schemes, 133–5 Arendt, H., 150 Argentina: and training, 41; and public policy, 10 Arthur, Brian, 149 Ashenfelter, O. and Johnson, G. E., 180 ASIMET, 25, 28, 127; and pact-making with labour, 84–7 Asset specificity, 28, 42 Ayala, E., 55, 57 Baglioni, G. and Crouch, C., 180 Bakery sector, 89–90, 212 Banking sector: and pact-making in Chile’s transition, 87–8 Bargaining groups, 108–13, 188–94, 208; see also Collective bargaining Bargaining instrument: and contracts versus conventions, in law, 108–13; in firms, 181–3, 188–94, 208 Barrera, M., 38, 64, 68, 180, 195, 215 Barrington-Moore, B., 14 Barro, R. J., 27 Bewlay, T., 148 Bipartism, and industrial relations in Chile, 82; and non-intervention in Chile’s transition, 93–4

Blanchflower, D. G., 148 Blanchflower, D. G. and Freeman, R. B., 98 Blank, R. M., 21 Boeninger, E., 56, 90 Bottomore, T., 3 Bowles, S., 16, 24, 26–8, 180 Boyer, R., 17, 21, 24, 180 Boyer, R. and Durand, J. P., 149 Brazil, 60; and training, 41; and public policy; 10 Briones, H., 53, 55, 81 Buchele, R. and Christiansen, J., 139 Burchill, S., 38 Business leaders in Chile, 13; and democracy, 47–8, 60–1; and authoritarinism, 48; and conceptions of development, 49; and history of, 49–50; organisation of, 50; closed and open sectors in, 55–7; and views on social organisation, 52–3; and budget, 54; and transition politics, 54–5; and views on democracy, 52–3; and views on social organisation, 52–3; history of, 49–50; and relation to politics, 60–1 Business sector in Chile relations between small and large firms, 49; and the rise of unions, 49; and concerted action on human resources, 53; compared with other countries, 53; and view on market neutrality, 52–5; and the media, 54; and relationship with the state historically, 48–50; relationship with the state during transition, 60–1; views on relationship with the state, 52–3; corporatist business leaders in metallurgy, 84–7; and threats if labour code is reformed, 90–3 Bustos, M., 57, 83 281

282 Index Calderón, R., 101 Campero, G., 48, 54, 56, 58, 74, 75, 84, 88, 102 Capabilities, 31, 80; see also Human resources Carrière, J., 49 Castillo, F., 74 Cerda, M., 80, 131 Certification of skills, and youth schemes, 135–6; and education, 126–7; and private training corporations, 127; in Korea and Singapore, 41; in Korea, 37 Chile, market model, 32–4; privatisation, 32; deregulation, 32; labour market, 32–4; economic liberalisation, 32; and labour mobility, 34; and development planning, 33; and institutional capacities, xxv; and education, 36, 40; and industrial relations, 38–41; and training policy, 123–38; and labour law, 97–118; and political parties, 74–5; and local unions, 176–97 Chomsky, N., 38 Christian Democratic Party, 31, 58, 72 CIASI, 72, 56; technocrats in, 71; programmatic role for the CUT, 71–2 Citizenship, definitions, 3–4; see also Social citizenship, Political citizenship, Occupational citizenship Civil rights, 11; political and economic components, 11, 251, see also citizenship Classical sequence, in T. H. Marshall, 6–8, 97; and labour law, 97–8; and the modern sequence, 11–13; see also Political democratisation, Social democratisation, Transition analysis CNT, 58 Collective bargaining in Chile, and precarious employment, 107; and permissible subjects, 107–8; contents by law, 107–8; coverage compared with France and Spain,

107; and day-to-day labour relations in Chile, 153–62; and usage of legal instrument, 181–3; and union age, 188–94; and sectoral federations, 204–5; and bargaining mode, 155–7; and incentive wages, 158–62, Communist Party, and metal unions, 71; in the CUT, 72–3; and social pact-making in metallurgical sector, 84–7, 93 CONAPAN, 89–90, 212 Concertación, 31; and view on social pact-making in the 1980s, 74; and accommodation to business in late 1980s, 74–5; and relations with the CUT, 73–4, 79; and programme of government, 75; and autonomy of government by accepting constraints, 78; and political pact-making in early 1990s, 78–9; and re-reform of the labour code, 90–3; and accommodation to business over labour law reforms, 102–4 CONSTRAMET, 65; and the Communist Party, 71; and local leaders, 71; and pact-making in Chile’s transition, 84–7; and metallurgical sector accords and local unions, 204–5 CONTEVECH, 65 Contracts and contractual relations, 28; neo-classical position, 34; time-horizon, 28; efficient contracts, 28; ‘presentialist’ contracting, 39; short-term contracting, 40; conventions and contracts and bargaining groups in Chile, 108–13, 181–3 Correa, Enríque, 131, and Chilean Government’s view of pact-making, 79–81 Cortázar, R., 88–90, 108; and limiting sectoral pact-making in Chile, 83, 89, 90 Cortéz, Guillermo, 89 Costa Rica, 10, 110, 254–5

Index CPC, and authoritarian coalition, 48; history and composition, 49–50; leadership in and market neutrality, 51; representation in, 51; decision-making style, 51, 57–9; in relation to the SOFOFA, 52; and views on labour relations and development, 52–3; and views on training, 53; and transition politics, 54–5; closed and open sectors in, 55–6; and conciliation with labour in the 1980s and early 1990s, 56, 58–9; position on labour relations, 57; and the political opposition, 55–8; and social pact-making 55–8; and non-intervention, 81; and influence on Concertación, 81–2; and opposition to specific labour law changes, 99, 101–2; role in limiting labour law reforms, 102–3; in stalling labour reform, 90–4 CTB, 64 CTF, 65 Cuba, 10 CUT, 72, 88; and local unionism, 73, 76; and meetings with entrepreneurs in 1980s, 57–9; organisation of, 65; and labour advisors, 71–2; establishment of 71–2; and views on labour relations, 73, and strategy during transition, 73–4; and federations, 65; and finance, 65; proposals for pact-making with the CPC, 73; and relations with the Concertación, 73–4; and proposals for labour law reforms, 99–101, 173 Dahl, R., xxiv, 4 Dahrendorf, R., 14 De Vylder, S., 17 Decentralization, and education, 126; of industrial relations in Chile, 38, 78; and welfare, 7–8, 12–13; and training 40–1; and

283

social pact-making 83–90; and training policy, 79, 123–5; see also Deregulation Democratisation literature, xxiii–xxv; and evolutionary thesis 14–16; and social forces, 47; and business leaders, 47–8; and role of unions, 62–3, 75–6; and transition government, 78; and labour law reforms, 99–101, 121–2; and social policy, 123, 144–5; and local labour relations, 147–8, 173–4; and local unions, 178–180, 205–6 Denmark, education, 31 Deregulation, and labour market flexibility, 147–9; and new institutional barriers, 24; and transition government, 77–8; and Aylwin Government’s accommodation to, 79–81; and business leaders’ demand for, 54–6, 81–3; and social pact-making, 83–90; see also Internal and external labour markets, Decentralization Deserving poor, 12; see also Pro-poor policy Development ideology, 42–3; and Chile’s transition, 72–5, 78–81 Development of human resources, xxiv, 27–8, 34–8, 53–4; see also Human resources Development planning, and training policy in Chile, 33, 127–8; view of, by business leaders, 52–5; view of, by transition government, 79–81; see also Developmental state, Deregulation Developmental state, 16, 21, 33, 216–21 Di Giorgio, J. R., 78 Díaz, Alvaro, 120 Díaz, Eugenio, 56, 58, 73–4 Dismissal legislation in Chile, 104–5 Dual education, 127, 132–8; see also Apprenticeship Durand, J. P., 17, 21, 24

284 Index Echeñíque, J., 32 Economic governance, 16–21, 42; and Chile 19–20; see also Governance literature Education in Chile, primary, 126; secondary, 126 Education, as compulsion 8; importance to citizenship; and welfare 9; and labour market 9, 37; in Chile 36–40; primary, 126; secondary, 126; in Korea 37; and decentralization, 126 Eggertson, T., 24, 38 Embeddedness, of local unions, 183–5 Employment security, 104–5; negotiations in firms of, 192–7; local unions’ view of 66–70; national union leaders’ view of, 72–4; see also Turnover Employment trends, and unionisation, 65 Engels, F., 10 Epstein, E. C., 79 Equality, of opportunity 10; and wage equality, 10; and social mobility amenities, 10 Evans, P. B., 16–17, 94 Exit and voice, 27 Fantuzzi, A., 103 FECHIPAN, 89–90 Feliú, M., and relations with labour in the 1980s, 54–5; leadership style in the CPC, 56–8; and pact-making, 59–60 Ffrench-Davies, R., 32 Formalization, inside Chilean firms, 25; and unions’ role, 29, 149, 154; and labour flexibility, 155–8; and incentive wages, 158–62; and task diversity, 160–2; and turnover, 163–4 France, 25, 251; and civil code, 92; and unionisation, 107; collective agreements in, 107 Frank, V. K., 79, 110, 263 Franzosi, R., 179 Freedom, developmental freedom (developmental liberty, used

interchangeably), 5–6, 25–7, 177, 209–10; freedom to contribute, 6; individual freedom, xxiv, 25–7; excess of, 31, 173–7; positive and negative freedom, 30; labour freedom 26, 209; abundance of, 173–4 Freeman, R. B., 21 Gallie, D., 180 García, R., 52, 56–7, 83; and relations with labour in 1980s, 55, 58–60 Garib, F., 58, 81 Germany, 47, 99; and labour law, 104, 107–8; and professional domains, 29, 31 Gerschenkron, A., 16 Gintis, H., 16, 24, 26–8, 80 Governance literature, 147–51; and approaches to unions, 180–1, 213 Graham, H. D. and Gurr, T. R., 180 Graham, Carol, 124–5 Grimshaw, D. and Rubery, J., 24 Growth, sources of in East Asia and Latin America, 34–5; in Chile, 32; see also Investment Guzmán, José-Antonio, 54, 58, 59, 82–3 Haddad, W. D., 124 Hanami, T. and Blanpain, R., 108 Hanami, T., 98 Hayek, F. A. von, 24, 180, 213 Heiremans, L. E., 52, 54–5, 57, 59, 83, 103 Helgø, C., 40, 126 Henríquez, L., 80 Hirschman, A. O., 16, 26–7, 49 Hirsh, B. T., 98, 213 Human capacities, 5; see also Human resources Human resources, and labour market 23–4; and labour freedom, 25–7; and occupational citizenship, 27–31; flexibility, statistics on 34–5, 212–13; and firm flexibility, 150–1; and labour relations in firms, 147–51; and governance literature, 149; and neo-classical position, 148; and deregulation, 148–9; and social history and

Index macro-perspectives, 149–50; development of, xxiv, 27–8 Hutton, L. W., 207 IADB, 34 ILO, 33, 82 INACAP, 40 Incentive wages, in Chilean firms, 158–60 Indemnity payments, 104–5 Indonesia, 32 Industrial citizenship, 7, 9 Industrial policy, bias against in Chile, 80 Industrial relations, and trade-offs, 29; in 1980s Chile, 38–41; the gap between the local and national levels, 76; centralization and decentralization, 78; and democracy as viewed by the Concertación, 80; and social trade-off, 150; and sectoral federation, 199–205 INE, 99 Infante, R. and Klein, E., 33 Informality, universalization of, 21; see also Formalization, Precariousness Institutional links, 30; and unemployment benefit insurance, 141 Institutional memory, 150; of local unions, 183–8, 212–5, 258 Institutional network, 31 Institutionalisation theory, 180 Institution-building, social policy and democracy, 146 Internal and external labour markets, 29, 36, 147–51; and training, 173; and lock-in effects, 175–6, 205–8 Investment, in Chile, 34–5; in technology and value added, 32; types of, 34–5; types of in Chile and labour turnover, 168–9 Japan, 16, 28, 32, 98; and labour law, 104 Johnson, C., 16–17, 19, 47–60 Kaufman, R., 48 Kongshøj-Madsen, P., 32

285

Korea, 16–17, 32; labour law in; employment insurance in, manufacture in exports, 32; education in, 36–7; and training corporations, 41 Krueger, A. O., 213 Labarca, O., 125 Labour analysis, Chile 1980s, 34; see also Union analysis Labour code of 1925, 38–9, 49, 64 Labour code of 1980, 30–9, 201–5; and dismissals, and multi-unionism, 106; and coverage of collective bargaining, 107; and workers in precarious employment, 107 Labour Code of 1990–1991, and dismissals, 99, 104–5; multi-unionism, 106; coverage of bargaining, 107; precarious employment and bargaining, 107; content of bargaining, 101, 107–8; unions versus bargaining groups, 108–13; the right to strike, 113–21; and the legacy of the adversarial principle, 121–2; and democratic legitimation, 122 Labour code of Aylwin, discussion of re-reform of (1993–2001), 90–3; and pressure from business leaders (1993–2001), 90–3 Labour federations, and local labour relations, 199–205; and local bargaining, 203–5 Labour flexibility, forms of in Chile, 152–4; numerical and functional, 153; wage flexibility, 153; inter-firm flexibility, 162–7; and turnover, 164–7; time-horizon of, 149–51 Labour management, and social factors, 162–7 Labour market flexibility, neo-classical and governance approaches, 147–51; in Chilean firms, analysis of, 23–5, 27–31; analysis of, in Chile, 34–8; implementation of, during Pinochet, 39–41; local unions’

286 Index Labour market flexibility – continued view of 66–70; national union leaders’ view of, 72–4; and local labour relations, 145–65; and labour law, 97–117 Labour markets, 28; and welfare, xxiii; and schooling, 36; internal and external labour markets, 29, 36; theories of labour flexibility, 23–5; flexibility of, xxiv; and social factors, 211–5 Labour ministry, and social pact-making, 82–4, 87–90 Labour mobility, 26–7; see also Labour flexibility, Occupations Labour plan, 38–9; see also Labour code of 1980 Labour policy, in Chile 1980s, 34, 124–5 Labour, local leaders, labour mobility, 26; labour market flexibility, 24; and control trade-offs, xxvi Laissez-faire, 1; and regimes, 18, 20; see also Deregulation, Decentralization, Neo-classical position Laski, H., 97 Leggett, C., 42 Liberal Party, 60 Linz, J., xxiii Linz, J. and Stepan, A., 209 Lizana, P., 56, 57, 83, 102; and the open sector in the SOFOFA, 56 Local union leaders in Chile, and position on labour relations, 66–7; and the restraint for democracy, 67; and co-operation with employers in 1980, 68; and flexibility in firms, 69; and views on training, 70–1; and the sensation of duty, 66–7; and key goals 67–9; and federations, 70; and views on wage flexibility, 194–9; and relations with federations, 203–4; and influence of political parties, 195–6 Local unions in Chile, and investments in workers, 147; history of leadership of, 151–2; and

definition of union strength, 181–3; and contracts versus conventions, 188–94; and impact of federations, 203–5; and time-horizon of labour management, 183–94, 205–8; and transition analysis, 207–8 Lock-in, 149, 258; and labour management, 176; and transition, 210 Macpherson, C. B., 5 Madrid, R., 32, 74 Mainstream, 7–11; and social policy, 123–4; and education and training, 126–7; and youth training schemes, 134–8; and unemployment insurance, 141–3 Mainwaring, S., xxiii, 15 Malaysia, 32 Malthus, Thomas, 22 Managerial autonomy, and Chile’s transition government, 77 Manufacture exports, 32 Marglin, S. A. and Schor, J. B., 180 Marín, G. and Rozas, P., 51 Market failure, and training policy in Chile, 125 Market neutrality, 29; and threats by business sector if labour code is reformed, 90–4; and transition, 210; see also Neo-classical position Market socialists, 5 Markets, and standards, 30; and social factors, 211–15, in training and use of public funds, 128–36; see also Neo-classical position, Internal and external labour markets Marshall, Alfred, 8, 147, 163 Marshall, T. H., xxiii–xxiv, 6–11, 13–15, 28; and membership, xxiii; and developed economies, xxiii; and the mainstream, and the classical sequence, 6–8, 97; and the modern sequence, 11–13 Marx, K., 10, 19 Matsuda, Y., 99, 106, 108 Maurice, M. et al., 25, 29, 73, 194, 255

Index Metallurgical sector, 84–6; and pact-making in Chile’s transition, 84–7, 210–11 Mexico, and tax incentives for training, 130 Mill, J. S., 4–5, 8, 15 Miller, D., 3 Mizala, A., 32 Mobilization theory, 180 Montero, C., 47–8, 52–3, 60 Morishima, M., 98 Multi-unionism, in Chile, 106 Muñoz, O., 48 Nelson, J. M., 7, 12 Neo-classical development economics (inter-changeable with neo-liberal theory), 23–4; see also Neo-classical position Neo-classical position, and skills, 34; and labour markets, 34–5; and labour law, 98; and the management of human resources, 147–9; and social factors, 165–6; and models of productivity growth, 176; and analysis of unions, 180–1 Non-intervention, a convention in Chile’s transition, 77, 79–81; as viewed by the Concertación, 74–5; in industrial relations, 38–41; as viewed by business leaders, 52–4; and business leaders’ role in, 56–8; as convention in Chile’s transition, 82–90 O’Donnell, G., xxiii, 4, 6, 14–15, 62, 78, 209 O’Donnell, G. and Schmitter, P. C., 4 Occupational citizenship, xxiii–xxv, definition, 4–6; and individual freedom, 5–6, 25–7; and institutions, xxvi–xxvii, 25; and social democratisation, xxv–xviii; and transition, 6, 21–2, 147; and long-term association with firms, 147; and local unions, 206–7; and social and political

287

democratisation, 209–11; and governance, 3, 16–21; and the mainstream, 8–11; and the new welfare doctrine, 12–13; and human resources, 23, 27–31; and human resources in Chile, 37–8; and property rights of labour and deregulation in Chile, 38–42; and Chile’s transition, 42–3; and unions in Chile, 63, 73, 76; and government policy, 80; and the metallurgical sector accords, 86–7; and political process of labour law, 102–3; and labour law, 121–2; and training policy, 123–4; and apprenticeship and occupational streams, 126–7; and tax break schemes, 131; and youth training schemes, 135–8; and unemployment benefits, 141–5; and local labour relations, 147–50, 176–7; and local unions, 178–9, 207–8; and cycles in labour relations, 166–72; and time-horizon of labour management, 188–90, 199; and union leaders’ support for democracy, 194–6 Occupations, 133–4, 140; and idea of office and citizenship, 4; and membership, 4; and developmental freedom, 4–6; and citizenship, 8–9; and productive existence, 6; and labour policy in 1980s, 32–4; gaining one on youth scheme, 136–7; and use of tax break programmes, 128–30; and PTC system, 126–8; and certification, 37; Korea, Singapore and Chile, 41 Ogawa, W., 17 Olson, M., 78 Ominami, C., 32, 74 Organisation theory, 28 Otero, M., 53, 83, 105, 118 Ownership, 27 Oxhorn, P., 178

288 Index Pact-making, and training policy, 127–8; by CUT and CPC in early 1990s, 82–4; role of Concertación, 82–3; see also Social pact-making, Sectoral pact-making Pagano, U., 24, 28–9 Paredes, J., 85–6 Park, Y. B., 37, 41, 137 PDC, alliance with PS, 72; in the CUT, 73 Piore, M. J., 23, 28 Pirrie, Madsen, 24, 207, 213 Polanyi, K., 16 Political citizenship, 4; see also Political democratization Political democracy, definitions of, xxii; see also Political democratization Political democratisation, and relationship social democratisation, xxii, xxvi, 209–11; and evolutionary thesis 14–16; and social forces, 47; and business leaders, 47–8; and role of unions, 62–3, 75–6; and transition government, 78; and labour law reforms, 99–101, 121–2; and social policy, 123, 144–5; and local labour relations, 147–8; 173–4; and local unions, 178–80, 205–6, 209–11; and economic development problems, 211–2 Political pact-making, and Chile’s transition, 78; and the Concertación’s political use of, 78–81, 102–4; see also Social pact-making, Pact-making Political right, role in securing peaceful transition, 93; political parties of, 48–9; in authoritarian coalition, 48–9 Ports, labour reform in, 88 Precariousness, in Chilean firms, 155–8; and incentive wages, 158–60; and task diversity, 160–2; and labour turnover, 164, 166–7; and time-horizon, 170–2; and union leaders, 184–6; and workers’ preferences, 196–7;

compared, 121; and strikes, 118; and bargaining instrument, 108–13; and subjects of bargaining, 107; and information, 108; versus the mainstream, 8; and certification, 37 Private training corporations, Chile compared with Singapore, Brazil, Argentina and Korea, 40–1; and youth schemes and vocational training in Chile, 136–8; and apprenticeship, 124–31, 134–6 Property rights, approaches to, 28; and human resources, 28; and unions, 29; as individual control over skills and productive life, 8, 10, 26–31; and unemployment benefit insurance in Chile, 141–2; and training in Chile, 131–2; and youth programmes, 138 Pro-poor policy, 7, 9; see also Deserving poor PS, alliance with PDC, 72; in the CUT, 73 Public policy, and institutional links, 10 Putterman, L., 26–8, 180 Rationalisation of firms in 1980s, and labour costs, 66; union response, 66 REFA, 72–3 Requena, R., 78, 87, 107 Rights, universal definition of and unions, 63; see also Property rights, Civil rights, Occupations, Occupational citizenship RN, in authoritarian coalition, 48 Rueschemeyer, D., 14 Scandinavia, 17, 31 Schmitter, P., 62 Schull, J., 79 Schultz, T. W., 28, 34, 211 Schultze, C. L., 148 Scott, C. D., 32 Scruton, R., 3 Sectoral labour federations, 199–205; see also Sectoral pact-making

Index Sectoral pact-making, as theme in Chile’s transition, 83–4; experiences of, in Chile’s transition, 84–90 Self-management, 26–7 Selvaratnam, V., 30, 41–2 Sen, A., xiii; and capability approach, xxiii–xxiv; and human capital literature, 5; and freedom, 5; and social development, 19 SENCE, 40; and training policy, 127–8, 131–2 Senghaas, D., 15, 18 Sharma, B. and Luh Luh, L., 42 Silva, P., 50 Singapore, 16, 17, 32; and tax incentives for training, 130; and training, 41 Skills, and statistics and measurement, 26, 28; and certification, 30; markets in, 10; and institutions, 28, 34; and sources of investment in, 147–9 Skocpol, T., 14–15 SNA, 53, 55, 60–1, 79 Social citizenship, definition of, 4, 12, 13–16; see also Social democratization Social democratization, 13–19; and universal franchise, 12; and other forms of democratization, xxii–xxvi, 209–11; and evolutionary thesis 14–16; and social forces, 47; and business leaders, 47–8; and role of unions, 62–3, 75–6; and transition government, 78; and labour law reforms, 99–101, 121–2; and social policy, 123, 144–5; and local labour relations, 147–8, 173–4; and local unions, 178–80, 205–6 Social exclusion, 5; see also Precariousness, Informalization Social factors, and labour management, 162–7; see also Union analysis Social governance, 17–20 Social history and macro-perspectives, 149–50; and macro-institutions, 149

289

Social institutions, and markets, 211–15 Social insurance in Chile, 41, 138, 140, 257; and pension funds (AFP), 142–3 Social insurance, 7, 30; see also Unemployment insurance Social pact-making, in the 1980s, 55–8; and the early transition, 58–60; in Chilean ports, 88; in the metallurgical sector, 84–7; in sectors of temporary work, 88–9, in the bakery sector, 89–90; in banking, 87 Social policy, and Latin America, xxiv; and occupational citizenship, 9; and the mainstream 10–11; and Chile’s transition, 123–4; training and industrial policy, 145 Social security, and Latin America, xxiv Socialist Party, 31, 72 SOFOFA, historical insertion, 49, 61; hegemonic role within business community, 51, 61; in CPC decision-making, élitism in, 51–2; views on democracy, 52; and training corporations, 53; and budget, 54; structure of, 51–2; and views on training, 53; leadership role in the CPC, 54–5; and ‘closed’ sectors in the CPC, 57; and views on labour relations, 57, 59; and threats if labour code is reformed, 90–3; and initiative in pact-making, 58–9, 83, 92 Solow, R., 149 Spain, and unionisation, 107; and collective agreements in, 107 Standing, G., 63, 163 State, 29; developmental state, 16, 19–20, 29; welfare state 17–18, 20; and intervention, 29; efficient and inefficient interventions, and regulation, 29–30; and different kinds of national governance, 216–21 State autonomy, ‘embedded autonomy’, 16, 94; of the Concertación during the transition, 78

290 Index State in Chile, and relationship with the business sector historically, 48–50; relationship with the business sector during transition, 60–1; business sector’s view on, 52–3 Stepan, A., xxiii Strikes, in Chile, 113–21; in Japan, 113; and the democratic regime, 118 Survey, of unions, universe of, 151–2; make up of, 222–5; local manufacturing unions in, 151–2 Taiwan, 16 Targeting, 7 Tax rebate scheme, 40 Tax rebate scheme, in Chile, 128–31; in Mexico and Singapore, 130 Taylor, C., 237 TBS, 146–7 Tendler, J., 12 Textiles, 118 Thayer, Arteaga, W., 48, 90, 102; role in and views on labour reform, 103–4, 118 Training in firms, and time-horizon of labour management, 168–70; and union leaders 70–4; and turnover, 169; and firm size and mechanization, 168; and formalization, 170; and bipartite committees, 129–30; and business leaders 53–4, 85–7 Training policy, and Chile’s transition, 123–5; case for, in Chile, 124–5; and market failures, 124; and decentralization, 127; and weakness of the Chilean state, 127–8; tax rebate scheme, 128–31; Youth Training Scheme, 132–6; and other youth schemes, 136–8; and democracy, 145 Transaction costs, and unemployment benefit insurance, 140–1 Transition theory xxii–xxiii, 13–16, 47–8, 209; and employers, 60–1; and unions, 62–3; and local unions, 178–9, and measurements

of union strength, 179–80; and the analysis of strikes, 178–9; and evolutionary thesis, 14–16; and social forces, 47; and business leaders, 47–8; and role of unions, 62–3, 75–6; and transition government, 78; and labour law reforms, 99–101, 121–2; and social policy, 123, 144–5; and local labour relations, 147–8; 173–4; and local unions, 205–8. Turnover, and the organisation of work, 162; and formalization, 162–3; and social factors, 165–7 UDI, in authoritarian coalition, 48, 55 Unemployment insurance, 29; in Chile, 80, 138–43, 145–6; in Korea, 143–4 Unemployment, in Chile, 32, 34, 124 Union age, 183–4; and union history, 183–5; and union effects, 214–15; and contracts, 188–94; and time-horizon of labour management, 188–94; and training, 188–91 Union analysis, and transitions, 62; and embeddedness, 183–5; and institutional memory, 183–8; and union history, 183–5; and definitions of union strength, 151; and social factors, 171–2; and formalization, 154–62; and neo-liberal theory, 180–1, 205–8; see also Neo-classical position Union dues, 107 Union movement in Chile, structure of, 63–4; historical insertion of, 64; links to political parties, 64, 75–6; and tradition of unity, 75 Unions in Chile, and democratisation, 62–3; and occupational rights, 63; and positions on employment security, 63; numbers of, 64; density and coverage, 65, 99–100; see also Local unions Uruguay, 10 Utilitarianism, 5

Index

291

Valenzuela, J. S., xxvi–xxvii, 4, 15, 38, 42, 47, 49, 64, 110, 178, 180, 208–9, 215 Virtuous and vicious cycles, in labour relations, 213–4 Vocational training, Korea and Singapore, 41; Chile, 40–1, 125–38

Welfare, definition of, 8–9; see also Mainstream, social policy Williamson, O., 16, 24, 28, 39, 149–50 Workers’ preferences, 149–51 World Bank, 24, 29, 31–2, 53

Weiss, M., 99, 104 Welfare state, 6–7, 9, 17–18, 20, 216–21

Zeitlin, M. and Radcliffe, R. E., 49–50 Zeitlin, M., 49

Youth Training Scheme, 41, 132–6, 146–7