Regulating for Decent Work: New Directions in Labour Market Regulation (Advances in Labour Studies)

Regulating for Decent Work

ADVANCES IN LABOUR STUDIES Advances in Labour Studies is a wide-ranging series of research titles from the International Labour Office (ILO), offering in-depth analysis of labour issues from a global perspective. The series has an interdisciplinary flavour that reflects the unique nature of labour studies, where economics, law, social policy and labour relations combine. Bringing together work from researchers from around the world, the series contributes new and challenging research and ideas that aim both to stimulate debate and inform policy.

Forthcoming in the series SHAPING GLOBAL INDUSTRIAL RELATIONS: THE IMPACT OF INTERNATIONAL FRAMEWORK AGREEMENTS (edited by Konstantinos Papadakis)

Regulating for Decent Work New Directions in Labour Market Regulation Edited by

Sangheon Lee and

Deirdre McCann

© International Labour Organization 2011 The designations employed in ILO publications, which are in conformity with United Nations practice, and the presentation of material therein do not imply the expression of any opinion whatsoever on the part of the International Labour Office concerning the legal status of any country, area or territory or of its authorities, or concerning the delimitation of its frontiers. The responsibility for opinions expressed in signed articles, studies and other contributions rests solely with their authors, and publication does not constitute an endorsement by the International Labour Office of the opinions expressed in them. Reference to names of firms and commercial products and processes does not imply their endorsement by the International Labour Office, and any failure to mention a particular firm, commercial product or process is not a sign of disapproval. All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No portion 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, Saffron House, 6–10 Kirby Street, London EC1N 8TS. Any person who does any unauthorized act in relation to this publication may be liable to criminal prosecution and civil claims for damages. The authors have asserted their rights to be identified as the authors of this work in accordance with the Copyright, Designs and Patents Act 1988. First published 2011 by PALGRAVE MACMILLAN and INTERNATIONAL LABOUR OFFICE Palgrave Macmillan in the UK is an imprint of Macmillan Publishers Limited, registered in England, company number 785998, of Houndmills, Basingstoke, Hampshire RG21 6XS. Palgrave Macmillan in the US is a division of St Martin’s Press LLC, 175 Fifth Avenue, New York, NY 10010. Palgrave Macmillan is the global academic imprint of the above companies and has companies and representatives throughout the world. Palgrave® and Macmillan® are registered trademarks in the United States, the United Kingdom, Europe and other countries. ISBN: 978–0–230–30217–4 ILO ISBN: 978–92–2–124559–9 This book is printed on paper suitable for recycling and made from fully managed and sustained forest sources. Logging, pulping and manufacturing processes are expected to conform to the environmental regulations of the country of origin. A catalogue record for this book is available from the British Library. Library of Congress Cataloging-in-Publication Data Regulating for decent work : new directions in labour market regulation / edited by Sangheon Lee and Deirdre McCann. p. cm. Includes index. ISBN 978–0–230–30217–4 (hardback) 1. Manpower policy. 2. Labor market. 3. Labor laws and legislation. I. Lee, Sang-Heon II. McCann, Deirdre M. HD5713.R44 2011 331.2—dc22 2011012061 10 20

9 19

8 18

7 17

6 16

5 15

4 14

3 13

2 12

Printed and bound in Great Britain by CPI Antony Rowe, Chippenham and Eastbourne

1 11

Contents List of Figures

vii

List of Tables

ix

Notes on Contributors

xi

Preface

xv

1 New Directions in Labour Regulation Research Sangheon Lee and Deirdre McCann

1

Part I: Theories and Perspectives 2 The Evidence-Based Case for Labour Regulation Simon Deakin 3 Precarious Employment and the Problem of SER-Centrism in Regulating for Decent Work Leah F. Vosko 4 The Growth of Extended ‘Entry Tournaments’ and the Decline of Institutionalized Occupational Labour Markets in Britain David Marsden

31

58

91

Part II: Recent Country Experiences 5 Laws or Luck? Understanding Rising Formality in Brazil in the 2000s Janine Berg

123

6 Labour Market Regulations and the Welfare of Indonesian Workers Iyanatul Islam

151

7 The Enactment of Three New Labour Laws in China: Unintended Consequences and the Emergence of ‘New’ Actors in Employment Relations Fang Lee Cooke

180

Part III: Old Challenges, New Techniques 8 Short Time Compensation as an Employment Stabilization Policy Robert M. LaJeunesse v

209

vi

Contents

9 Gender and the Minimum Wage Jill Rubery and Damian Grimshaw 10 Decent Work for Domestic Workers: An Achievable Goal or Wishful Thinking? Manuela Tomei

226

255

Part IV: Elements of Effective Regulation 11 The Impact of Labour Regulations: Measuring the Effectiveness of Legal Norms in a Developing Country Sangheon Lee and Deirdre McCann

291

12 Governing Regulatory Discretion: Innovation and Accountability in Two Models of Labour Inspection Work Roberto Rocha C. Pires

313

13 A Diagnostic Methodology for Regulating Decent Work Diane F. Frey

339

Index

365

List of Figures 1.1

Net benefits of labour regulations: an illustration

17

2.1

Labour regulation in five countries, 1970–2005

41

4.1

Career tracks for an unstructured occupational entry for a given cohort

96

Growth of weekly earnings 1975–2001 at constant 2001 prices for selected percentiles and occupations

106

5.1

Real minimum wage in Brazil, 1990–2010

127

5.2

Share of formal employment in Brazil (ILO definition), 1992–2008 (per cent)

128

5.3

Real effective exchange rate in Brazil, 1990–2008 (2005 = 100)

131

5.4

Real GDP growth in Brazil, 1992–2008

132

5.5

Number of workers registered as a result of labour inspection in Brazil

138

5.6

Registered domestic workers in Brazil, 1992–2008

140

6.1

Indonesia: aggregate unemployment rate, 2000–2009

163

6.2

The changing share of informal employment in Indonesia, 2001–2009

163

Indonesia: global competitiveness index (1 = lowest rank; higher rank shown as declining fraction), 2000–2007

165

4.2

6.3 6.4

Indonesia: labour market regulations index, 2000–2005

166

6.5

Indonesia: economic freedom index

167

6.6

Labour problems in Indonesia

168

6.7

Unit labour costs in Indonesia (1993–2006: base year = 1993)

169

Export share of unskilled labour-intensive manufacturing in Indonesia

173

Major labour market and employment-related laws in China and key elements

185

Inter-relations between institutional actors

190

6.8 7.1 7.2

vii

viii List of Figures

8.1

Cyclical use of part-time unemployment compensation in the US and for the six most active states

213

Number of workers who are working part-time due to slack work or lack of full-time jobs in the US

215

8.3

Part-time work status by choice in the US

215

9.1

Low pay incidence of women relative to men, full-timers, 2000–2006

231

Average gender pay ratio by low-skill occupation in selected EU member states, 2002

233

8.2

9.2

11.1 Can awareness make a difference? Working time, minimum wages and maternity leave

304

List of Tables 4.1 Changing employment status within selected occupations 1975–2000: self-employed as percentage of total employment in occupation

104

4.2 Summary of growth in real weekly and hourly earnings at selected percentiles 1975–2001

108

4.3 Summary of logit coefficients: factors predicting the probability of an employee’s pay being in the bottom or top 20 per cent of weekly earnings

111

5.1 Annual employment growth by job category

129

6.1 Indices pertaining to employing workers in Indonesia, 2003–2007

165

6.2 Minimum wages and employment: summary of findings

169

9.1 Gender pay gaps at the lowest decile (D1) (full-timers only) by type of wage system, 18 OECD countries, 2005

230

9.2 Low-pay incidence of women relative to men (full-timers only) by type of wage system, 2000

232

9.3 Share of unemployed in receipt of benefits and the gender gap

236

9.4 Advantages and disadvantages of minimum wage policies for promoting gender pay equity

246

A9.1 Country data on collective bargaining coverage, relative level of statutory minimum wage and gender pay gap at the lowest pay decile

253

A9.2 Low pay incidence by gender

254

10.1 Main features of the ‘cheque’ systems in Belgium, France and the Canton of Geneva (Switzerland)

266

11.1 Survey questions about workers’ awareness

301

11.2 Workers’ awareness (percentage of employees: excluding tourism sector: N = 666)

302

ix

x

List of Tables

11.3 Employment contracts without teeth? Percentage of written contract or agreements which provide details on wages, working time and maternity leave (N = 553)

304

11.4 Can awareness make a difference? The results of logistic regressions

305

11.5 Awareness and job dissatisfaction: working hours – regression coefficients

307

11.6 How would workers react? (percentage of respondents)

307

12.1 Total FGTS collection by labour inspection in Brazil, 1996–2005

322

12.2 Comparing the outcomes of FGTS by labour inspectors in the Pernambuco State Office, 2007

323

13.1 An institutional approach to compliance

352

Notes on Contributors Janine Berg is Senior Economist with the International Labour Office, currently stationed in Brasilia, Brazil. She is the author of Miracle for Whom? Chilean Workers under Free Trade and Meeting the Employment Challenge: Argentina, Brazil and Mexico in the Global Economy (2005), as well as co-editor and contributor to In Defence of Labour Market Institutions: Cultivating Justice in the Developing World (2008, with D. Kucera). Her research focuses on the economic effects of labour laws as well as policies for generating jobs and improving employment. She received her PhD in economics from the New School for Social Research. Fang Lee Cooke is Professor of HRM and Chinese Studies at the Department of Management, Monash University, Australia. Previously, she was a professor at Manchester Business School, University of Manchester, UK. Her research interests are in the area of employment relations, gender studies, strategic HRM, knowledge management and innovation, outsourcing, Chinese outward FDI and employment of Chinese migrants. She is the author of HRM, Work and Employment in China (2005), Competition, Strategy and Management in China (2008) and Human Resource Management in China: New Trends and Practices (2011). Simon Deakin is Professor of Law at the University of Cambridge, specializing in labour law, private law, company law, law and economics, and EU law. He is a programme director at the Cambridge Centre for Business Research, and an associate Faculty member of the Judge Business School. His books include Tort Law (5th edn 2003, with B. Markesinis and A. Johnston), Labour Law (5th edn 2009, with G.S. Morris) and The Law of the Labour Market: Industrialization, Employment, and Legal Evolution (2005, with F. Wilkinson). He is a member of the editorial boards of the British Journal of Industrial Relations, the Industrial Law Journal (UK), and the Cambridge Journal of Economics. He was elected to a Fellowship of the British Academy in 2005. Diane F. Frey is a PhD candidate at the London School of Economics, where she was a British Journal of Industrial Relations Teaching Fellow (2007–09). Her research examines worker rights in the context of globalization, and she has published articles on this topic in Advances in Industrial and Labor Relations, the Hastings Race and Poverty Law Journal xi

xii Notes on Contributors

and the American University International Law Review. Additionally, Diane has been a labour organizer for over twenty years. She currently works with the American Federation of Teachers and previously worked for locals of the Service Employees International Union and the National Education Association. Damian Grimshaw is Professor of Employment Studies and Director of the European Work and Employment Research Centre (EWERC) at the Manchester Business School, University of Manchester. His current research interests include co-production and human resource management, low-wage work in Europe, recruitment and retention in the UK care sector and women’s job growth in Europe. Publications related to his present chapter include Fragmenting Work: Blurring Organizational Boundaries and Disordering Hierarchies (2005, with M. Marchington, J. Rubery and H. Willmott), and Managing People in Networked Organizations (2009, with M. Marchington, J. Rubery, M. Carroll and S. Pass). Iyanatul Islam is currently Acting Chief of the Country Employment Policy Unit in the Employment Policy Department of the International Labour Office, Geneva, Switzerland, and Adjunct Professor of International Business, School of International Business and Asian Studies, Griffith University, Australia. His latest publications include two co-authored books on the Indonesian labour market and a coedited volume From the Great Recession to Labour Market Recovery (2011, with S. Verick). Robert M. LaJeunesse obtained his PhD in Economics from Colorado State University, after two years of service in the US Peace Corps. He then spent time working with the US Treasury and the AFL-CIO. For nearly a decade, he worked in academia at SUNY-New Paltz and the University of Newcastle, Australia. His scholarly publications are primarily in the field of labour economics, including a recently released book on the socioeconomic and ecological virtues of work time regulation. He presently holds the position of Senior Economist at the US Equal Employment Opportunity Commission. Sangheon Lee is the Research and Policy Coordinator for the Conditions of Work and Employment Programme of the International Labour Office. He specializes in analysing and monitoring changes in wages and working conditions with a view to developing improved policy responses. Dr Lee has written extensively on working conditions issues. His recent publications include Globalization, Flexibilization and Working Conditions in Asia and the Pacific (2008, co-editor with F. Eyraud), Working

Notes on Contributors xiii

Time Around the World (2007, with D. McCann and J. Messenger). He is also one of the main authors of the ILO’s Global Wage Report 2010/11 (2010) and a coordinator of the Regulating for Decent Work Network. He holds a PhD in economics from Cambridge University. David Marsden is Professor of Industrial Relations at the London School of Economics, UK, and a Research Associate of the Centre for Economic Performance. He did his doctoral work at the Laboratoire d’Économie et de Sociologie du Travail (CNRS), Aix-en-Provence, where he developed his interest in the comparative study of labour markets and labour institutions, and how these affect the internal workings of organizations. Deirdre McCann is a Senior Lecturer in the School of Law at the University of Manchester. Her research is in the field of labour law and policy and has a particular focus on the regulation of working conditions, precarious work, and the influence of flexibility discourses on labour market regulation. Her publications have focused on the regulation of working time, non-standard work and work/family reconciliation. They include Regulating Flexible Work (2008). She was previously a Senior Researcher at the International Labour Office in Geneva and is a coordinator of the Regulating for Decent Work Network. Manuela Tomei is the Director of the Conditions of Work and Employment Programme at the International Labour Office. She is the lead coordinator of the preparatory work towards the adoption of international labour standards on decent work for domestic work by the International Labour Conference in June 2011. A sociologist by training, she is main author of the two Global Reports on discrimination at work produced by the ILO in 2003 and 2007 in the framework of the follow-up to the 1998 ILO Declaration on Fundamental Principles and Rights at Work. Roberto Rocha C. Pires is a researcher at the Department of State, Institutions, and Democracy Studies at the Institute for Applied Economic Research (IPEA), Brasilia, Brazil. He received his PhD in Public Policy from the Massachusetts Institute of Technology. His research focuses mainly on local participatory governance, and on the management of labour regulation and inspection and its effects on social and economic development in Brazil. He has published articles in international journals and edited volumes on the topic. Jill Rubery is Professor of Comparative Employment Systems, Manchester Business School, University of Manchester. She has published

xiv Notes on Contributors

widely on comparative employment systems including the future of the European social models, most recently European Employment Models in Flux (2009). Her research has been funded by various bodies including the European Commission, the ILO, the World Bank and the OECD. Leah F. Vosko is Professor and Canada Research Chair in the Political Economy of Gender and Work at York University in Toronto, Canada. She conducts research on labour and social policy, placing an emphasis on gender, citizenship, and labour market insecurity in comparative perspective. Professor Vosko’s latest book is Managing the Margins: Gender, Citizenship and the International Regulation of Precarious Employment (2010).

Preface Despite the shock of the global financial crisis and its origins in regulatory failure, deregulatory rhetoric has proved remarkably durable. Indeed, the recession has come to be characterized by the dismantling of regulatory frameworks in the desire to create jobs. This volume is a response to that deregulatory project. It aims to contribute to ongoing efforts to build new conceptual frameworks and modes of analysis to investigate the role, value and limitations of labour market regulation. To this end, it brings together an international group of researchers to consider, from a range of disciplinary traditions and perspectives, the future of the field. It is hoped that the volume will enrich and advance the academic and policy debates on post-crisis labour regulation. The book is a testament to the work of its contributors. We are immensely grateful to them for their contributions and enthusiasm for this project. We believe the book showcases the best of contemporary research on labour market regulation and hope that it conveys the importance of this vigorous and exciting field of study. The contributions have their origins in the inaugural conference of the Regulating for Decent Work (RDW) Network, which was held at the International Labour Office in Geneva from 8–10 July 2009. The Network emerged from a widespread unease about the expanding grasp of the deregulatory project, a desire to fully integrate regulation into efforts to advance ‘decent work’, and an intuition that a space should be carved where research on the design and effectiveness of regulatory frameworks could be sheltered from the need to adopt a defensive stance. The Network members are researchers from universities, research centres, government agencies and NGOs across the world and represent a range of disciplinary and theoretical perspectives. We are privileged to benefit from the insights and engagement of this vigorous intellectual community. We acknowledge the support of our friends and colleagues on the RDW Network Organizing Committee: Sandrine Cazes, Colin Fenwick, John Howe, Jill Murray and Anne Posthuma. We are immensely grateful to the International Labour Office for funding and hosting the RDW Conference and for its assistance in the publication of this volume. Particular thanks are owed to Assane Diop for his enthusiasm for the Network and his commitment to the ILO’s role in encouraging and engaging with research communities. He also xv

xvi Preface

very kindly agreed to give the opening address to the Conference. We owe an immense debt of gratitude to Manuela Tomei for her deft political skills, enduring support and invaluable advice, Gerry Rodgers for his guidance and encouragement, and Duncan Campbell and Sandrine Cazes for support both financial and moral. Carola Nolte bore most responsibility for organizing the RDW Conference. She performed this demanding role with patience and good grace and we greatly appreciate her for her outstanding work. Thanks are due to all other individuals who contributed to the success of the Conference, not least to the participants but also to the ILO colleagues who chaired sessions and others who contributed in various ways, including Claire Piper, Coralie Thompson, Susie Choi and Seungeun Lee. We again marvel at the energy and commitment of Charlotte Beauchamp in the ILO Publications Department and her skill in preparing the manuscript for publication while dealing with editors and contributors across institutions, countries and time zones. We thank the staff at Palgrave Macmillan for supporting this first stage of what we expect to be a fruitful collaboration. We are grateful also to Seung-Suk Ryu, who provided invaluable research assistance during the preparation of the manuscript. On behalf of the contributors, we extend sincere thanks to our two anonymous reviewers for their incisive and constructive comments and appreciation for the aims of the project. Finally, we extend sincere thanks to our families, friends and colleagues for their invaluable support.

1 New Directions in Labour Regulation Research Sangheon Lee and Deirdre McCann*

This volume is an international and interdisciplinary response to the most influential account of the role and significance of labour market regulation, namely that derived from orthodox economic theory. It also responds to the most prominent alternative to the orthodox narrative: the contention that the central objective of labour and development policies, including in their regulatory dimensions, should be to realize and sustain ‘decent work’. In recent years, these theoretical and normative approaches have galvanized research on labour market regulation. This volume highlights certain of the more significant and novel developments reflected in these literatures. The aim is to feature innovative ideas and approaches, new subjects and debates, and theoretical perspectives and methodologies that characterize contemporary research on labour market regulation. In doing so, it is hoped that this book will enrich and advance the academic and policy debates on post-crisis labour regulation. This chapter introduces the preoccupations that animate the Regulating for Decent Work (RDW) project and the essays in this book and attempts to draw out their implications for future research agendas on labour market regulation. To this end, the chapter first outlines the recent evolution of the deregulatory narrative then reviews the response of the Regulating for Decent Work Network. After introducing subsequent chapters, it draws on them to identify issues of central relevance to the design of theoretical, conceptual and methodological frameworks through which research on labour market regulation can be advanced. * This chapter is dedicated to the memory of Jonas Agell (1957–2007), whose work inspired the Regulating for Decent Work Network. His plans to be involved were sadly defeated by his failing health. The authors are grateful to Damian Grimshaw, Leah Vosko and the two anonymous reviewers for their comments. 1

2

Regulating for Decent Work

Evolutions in the deregulatory narrative It has long been recognized that the simplistic dichotomy between ‘regulation’ and ‘deregulation’ characteristic of the neoclassical economic tradition and Washington consensus policy agendas is both highly influential and also threatens the distributive objectives of labour market regulation. Theorists of labour regulation of various disciplinary backgrounds have resisted this account, in which labour regulations appear simply to distort the market and generate inefficiencies (see, for example, Berg and Kucera 2008; Manning 2003; Boeri and van Ours 2008). Yet the deregulatory project is not static. The key twist in its evolution over the last decade has been a heightened recourse to empirical methodologies that quantify and compare labour regulations (see further Deakin in this volume). This trend towards quantification has been most pronounced, and most controversial, in the use by the World Bank’s Doing Business project of indicators designed by Botero et al. (2004), which purport to evaluate and compare national regulations shaping ‘the business environment. This project embraces, through the Bank’s Employing Workers Index (EWI), a number of the central features of labour law regimes, including employment protections (dismissal and redundancy), working hours, rest and leave, minimum wages and regulations on fixed-term contracts (see World Bank 2009a and for the underlying methodology, Botero et al. 2004). The Doing Business indicators have been disseminated widely through a series of annual reports (most recently World Bank 2010) and other methods such as the annual selection of ‘top ten reformers’ (see, for example, World Bank 2010). They have also been used as the basis for policy guidance by other branches of the Bank, including in decision making on the allocation of resources by its International Development Association (IDA). These quantification efforts have improved methods of tracking and comparing legal standards and prompted a vigorous literature, in particular on the effects of ‘legal origins’ on regulatory design and economic outcomes (Aherling and Deakin 2007; Deakin et al. 2007; Mitchell et al. 2010). They have also been instrumental in consolidating deregulatory rhetoric, through the pairing of indices that implicitly favour deregulated labour markets with assertions on the association of ‘light’ regulation with economic growth that are empirically difficult to sustain (see further Berg and Cazes 2009; Lee and McCann 2008; Deakin in this volume). Other mutations in this evolution are less often the subject of examination in scholarly and research treatments of the quantitative turn in labour market policy design. Yet the trend towards the construction of

Sangheon Lee and Deirdre McCann

3

legal indicators has encompassed at least two significant refinements of the deregulatory narrative that expand its influence while potentially undermining its efficacy: a shift from the preoccupation with minimum wage and employment protection laws to encompass broader elements of the labour law corpus; and a geographical expansion beyond the advanced industrialized economies to the developing world (Lee and McCann 2008). Also less frequently observed is that this discourse has become more nuanced in response to the growing evidence that the processes of economic globalization are associated with growing inequality (ILO 2008). The preoccupation with the impact of regulations on economic performance remains, but is accompanied by a heightened emphasis on the assertion that labour markets liberated of regulatory interventions also advance equality, in particular by aiding the entry into the labour market of disadvantaged groups such as women and young, uneducated and rural workers (for example, Heckman and Pagés 2004; Basu and Maertens 2007). Finally, it is worth recognizing, as a distinct element of the development of the deregulatory narrative, its persistence. The quantification project, when pursued through unsophisticated methodologies, has generated intense criticism. Most notably, substantial criticism has been directed at the EWI, on its design, its embrace of the social objectives of regulatory interventions, and, most forcefully, its deployment as a satisfactory guide to legal policy (see Lee et al. 2008 for a review). These criticisms have been of such intensity that following a critical report by the World Bank’s Internal Evaluation Unit (IEG 2008), the Bank prohibited the use of the Index in policy advice and assigned it to an ongoing process of reform (see World Bank 2010). The EWI has been relegated to an annex in the most recent Doing Business report and excluded from the overall assessment of countries’ business environments (World Bank 2010). In light of this curbing of the deregulatory project, it may have been assumed that the disruption caused by the global financial crisis, and its origins in regulatory failure, would spur a serious reconsideration of the role of labour market regulation. Yet it has become clear that labour market interventions have not garnered a level of support to match the acceptance of more rigorous regulation of financial markets, now relatively influential at least at the level of rhetoric. In contrast, the response with regard to the labour market realm has more often been a call to persevere with deregulatory reform. In the (albeit strikingly tentative) words of Doing Business 2010, ‘Setting long-term goals and keeping a steady course of reform might help economies recover from shocks, including the current global financial and economic crisis’ (World Bank 2009a, p. 9).

4

Regulating for Decent Work

Regulating for Decent Work The Network on Regulating for Decent Work was established with the goal of nurturing efforts to fashion a coherent response to the deregulatory agenda in its contemporary guise.1 The Network’s objectives are therefore in part reactive: to encourage and disseminate research that investigates this discourse, whether by interrogating its theoretical underpinnings or methodologies or by clarifying and enunciating the potential benefits of regulatory interventions. RDW’s more ambitious project is to prevent research agendas on labour market regulation from being shaped primarily in response to the deregulatory narrative. To this end, it encourages research into dimensions of the field that are being drowned out by the reigning narrative. In particular, conceptual models are favoured that reject a stark regulation/deregulation dichotomy in favour of richer understandings of regulatory frameworks, including those that support detailed empirical work to assess labour law’s de facto operation. To this end, Network participants have, as is evident from the contributions to this volume, continued to be inspired by the notion of ‘decent work’ that was elaborated at the end of the last century as the guiding objective of the International Labour Organization (ILO 1999). The ILO’s Decent Work Agenda has inspired research efforts that elaborate on the policy and regulatory demands of decent work, whether as an overarching concept or in its specific dimensions, and which have incorporated an exploration of the role of legal measures (see, for example, Owens 2002; Boulin et al. 2006). Yet this project is not complete. Participants in the RDW Network are building on this work by further investigating the role of regulatory measures at multiple levels in realizing decent work, albeit in the recognition that this objective may not demand the same solutions in different contexts. The Network is both international and interdisciplinary. It therefore recognizes that the challenges of regulating the labour markets of the contemporary global economy are unlikely to be addressed adequately from a narrower platform. Network participants are based in universities and research institutes in all regions and research the processes of labour regulation in countries across the world. RDW recognizes, however, the urgent need to strengthen the available knowledge of regulatory measures in low-income settings. In this regard, the concern is to avoid too ready an assumption that theoretical models and research themes designed for industrialized countries will inevitably be relevant to the developing world, while also identifying the contexts in which such an assumption might be valid (see further Benjamin 2007).

Sangheon Lee and Deirdre McCann

5

The Network was also established with the goal of fostering the exchange of ideas and experience among researchers from a range of disciplinary and theoretical traditions. The intention in this regard is to underscore and build on the linkages between disciplines in which similar preoccupations are being pursued and similar challenges encountered. Such interdisciplinary engagement, it is hoped, will serve to clarify the themes that animate different literatures, highlight intersecting preoccupations and insights from diverse theoretical traditions, allow researchers to draw on conceptual, theoretical and methodological advances in other fields to inform and enrich their own, and encourage interdisciplinary research. This approach is particularly apt for a field in which doctrinal analyses of labour laws and the interpretive practices of judicial bodies, knowledge of the law’s influence in practice and assessments of its economic repercussions are all required. These points are returned to below.

Chapter overview It is hoped that this first volume of contributions drawn from the work of its participants reflects the founding objectives of the Network. The book is international in scope. Compiled from the work of contributors based in research institutions around the world, it examines the regulatory frameworks of countries as diverse as Brazil, China, France, Indonesia, Tanzania and the United States (US). It also showcases research from a range of disciplinary perspectives, highlighting the vibrancy of scholarship on regulation across fields that include economics, law, industrial relations, sociology and political science. The first part of the book identifies a set of contemporary challenges for labour market regulation that are indicative of key contemporary trends in legal evolution and in the research literatures. Part I (Theories and Perspectives) identifies three central themes that, explicitly or implicitly, underpin most of the contributions in the remainder of the volume: the role of empirical research in assessing and supporting labour market interventions (Deakin, Chapter 2); the historical limitations in the coverage of regulatory frameworks and the recent trend towards extending their reach to excluded constituencies (Vosko, Chapter 3); and the decline of traditional institutional mechanisms and consequent search for regulatory techniques to replace them (Marsden, Chapter 4). Simon Deakin’s analysis of the evidence-based case for labour regulation (Chapter 2) highlights the development of new data sources that are capable of measuring cross-national difference in legal regimes and the resulting proliferation of indicators. Deakin notes that these developments

6

Regulating for Decent Work

have challenged conventional understandings of the economic impact of labour regulations, by suggesting that they are not as negative as hitherto assumed, while also revealing the complexity of this relationship. He concludes that the effects of labour law cannot be predicted through models of universal application, but require more sophisticated analyses of national, regional and industry conditions and complementary institutions, as well as more reliable data on laws and their operation in practice. Deakin cautions, however, that theory still remains more significant in shaping policy-making than empirical findings. In Chapter 3, Leah Vosko examines the disjuncture between labour force participation patterns and regulatory frameworks, focusing on a range of precarious forms of employment. Building on her earlier work, which has highlighted the regulatory exclusion of these work-forms due to the influence of the standard employment relationship (SER), Vosko examines the trend over the last decade towards crafting legal measures designed explicitly to protect workers in precarious jobs. Drawing on international labour standards and laws and policies from Australia, Canada, the European Union (EU) and the US, Vosko reveals these frameworks still to be grounded in the SER, and consequently unlikely to improve working arrangements that deviate sharply from its contours. Proposing instead a model of ‘global labour market membership’, Vosko concludes that strategies of re-regulation must dispense with the notion of a singular employment model and be attentive to gender, citizenship and age, and outlines a range of innovative measures to respond to this need. David Marsden (Chapter 4) singles out for consideration a conventional regulatory institution in decline: the internal and occupational labour markets that until recently dominated in many industrialized countries. Marsden takes the UK as his focus and explores labour markets in which competition for entry has become prolonged over a considerable period of time, which he characterizes as extended ‘entry tournaments’. He suggests that such tournaments are of relevance to the analysis of labour market regulation, in that they appear to have developed in line with the decline in internal labour markets. Marsden questions whether tournament-style transitions are desirable for society as a whole and how far they are compatible with promoting the goal of decent work. He concludes by suggesting certain strategies of legal regulation that could potentially take the place of the declining institutions. The subsequent chapters elaborate on these themes. In particular, they take up the implication of the three foundational chapters that the proper evaluation of the effects of regulatory measures requires sophisticated analyses of national, regional and industry contexts. Part II (Recent Country

Sangheon Lee and Deirdre McCann

7

Experiences) pursues recent trends in labour market regulation through case studies of low-income countries. Janine Berg (Chapter 5) examines the trend towards increasing formalization in Brazil. She concludes that this experience confirms the significant role of government policies in ensuring that workers share in the gains of economic growth. Berg further concludes that the labour market flexibility debate has been exaggerated in the case of Brazil, where the data do not support the claim that labour regulations cause growing informality. Iyanatul Islam (Chapter 6) considers the influence of labour market regulations on the welfare of workers by examining the Indonesian version of the deregulatory narrative. He contends that this thesis is erroneous in the case of Indonesia, on both analytical and empirical grounds. The challenge for policy, he suggests, is to strike the right balance between growth and employment creation on the one hand, and worker protection on the other. The contribution of Fang Lee Cooke (Chapter 7) examines the recent departure from an efficiency-driven development policy in China to one that pursues social justice, social harmony and environmental protection. Her particular focus is on the adoption of a set of three major employment laws and the role of new policy actors that has emerged in this process. Cooke concludes that these new actors have introduced novel dynamics and are playing a significant role in shaping legislative outcomes. Part III (Old Challenges, New Techniques) brings together contributions that, in different ways, examine the role of novel or neglected techniques in addressing long-standing regulatory challenges. Robert LaJeunesse (Chapter 8) interrogates neoclassical theory on its own terms, identifying the predicament of the US in the wake of the crisis as a failure of the ‘flexibility’ model to achieve one of its central goals, of enabling swift job creation. From this premise, he returns to the solution offered by past crises, of adopting work-sharing policies, and examines the role of worksharing in insulating labour markets and their revival during the crisis. Rubery and Grimshaw’s contribution (Chapter 9) also addresses one of the central contentions of orthodox economic theory, in this case that minimum wages distort the functioning of markets. The authors draw on cross-national empirical evidence to suggest instead that minimum wages can counter such distortionary effects and smooth out some of the imbalances caused by the sex segmentation of labour supply. They argue for a renewal of minimum wage analyses that take into account women’s relative employment position. Manuela Tomei (Chapter 10) draws on Vosko’s analysis of precarious work by situating the ongoing progress towards an ILO standard on domestic workers as part of the broader trend towards minimum standards for atypical and highly feminized workforces.

8

Regulating for Decent Work

Through a comparative analysis of regulatory schemes in Belgium, France and Switzerland, Tomei concludes by suggesting approaches and strategies to ensure that the notion of decent work encompasses domestic workers. Part IV (Elements of Effective Regulation) addresses one of the key subjects in contemporary labour market regulation research, the effectiveness of regulatory frameworks. It takes as its particular focus the notion of effective regulation in developing countries and analyses the constraints that characterize these settings, without adhering to the prevalent assumptions about regulatory failure. Lee and McCann (Chapter 11) note the unsophisticated understandings of legal measures in the research on the economic impacts of regulation and call for legal effectiveness to be integrated into the analysis. They draw on new survey data from Tanzania to explore workers’ knowledge of legal standards and the actions they take in response to the breach of legal norms. The authors conclude that there is significant potential for improving working conditions by increasing worker awareness. In Chapter 12, Roberto Pires responds to the widespread characterization of labour inspection as ineffective and lacking in innovation. Noting that the literature has identified innovative practices in various settings, he examines how such initiatives might emerge, drawing on the case study of labour inspection in Brazil. Pires concludes that management models and the organization of street-level regulatory work are crucial and calls for a deeper understanding of these elements of enforcement regimes. Finally, Diane Frey (Chapter 13) presents a methodology geared towards generating insights into the interventions necessary to achieve decent work. She applies institutions theory from the international law tradition to an analysis of working time regulation. Frey presents a complex depiction of the operation of legal norms in Honduras, acknowledging that formal regulation alone cannot account for the observance of legal standards, which also depends upon informal social norms, social conventions and the institutional influence of other dimensions of social and economic life.

Towards new theoretical and conceptual frameworks: the benefits and indeterminacy of labour market regulations It is hoped that the contributions to this volume will be drawn on to advance the broader project of designing theoretical and conceptual frameworks that can guide future research efforts and policy. To this end, the remainder of this chapter identifies and explores certain of the starting points for this project. Drawing on the volume contributions,

Sangheon Lee and Deirdre McCann

9

it elaborates two issues of central relevance to this project: the potential benefits of labour regulation and the complexity of regulatory frameworks. This section is intended both to contribute to the development of a full theoretical synthesis’ (Deakin, Chapter 2 in this volume) and to frame the analyses pursued in subsequent chapters. The potential benefits of labour regulations The first important step in devising the more balanced approach to labour regulation called for by the RDW project is to recognize the potential benefits of regulation and to reflect them in the design and evaluation of labour market policies. This effort is missing from much of the economic literature, which, as elaborated above, has tended to be consumed by the potential detriments of regulation. As a preliminary observation, it is apparent that a central feature of a more balanced theory of regulation is a recognition that the predicted impacts of regulations are heavily dependent on the labour market model assumed in the analysis. It is well known that under the neoclassical model’s assumption of a perfectly competitive market, it is impossible to expect any positive impact from labour regulations. In this model, labour regulations simply distort the market and create economic inefficiencies, in the form of either higher unemployment or slower economic growth. Adjustments to some of the key assumptions, however, such as the introduction of imperfect competition and information asymmetry, alter predictions in favour of labour regulation (see Dickens 1984; Boeri and van Ours 2008 for a recent summary). The employment-enhancing effects of minimum wage regulations in the monopsony model are a case in point (Manning 2003). The contrasting predictions that emanate from competing theoretical models have often been interpreted as suggesting that the outcomes of labour regulations are in essence a matter of empirical findings. However, theoretical models guide empirical research strategies, which typically involve choices over the variables to be considered. In other words, they determine the list of explanatory variables and the relations between them, and hence the ways in which statistical relationships are interpreted. This means that unexpected empirical findings under the neoclassical model tend to be taken to indicate the need for improvement of the statistical models and data, rather than as raising questions about the model itself. Therefore, alternative theoretical models need to be further developed to guide more balanced empirical research and policy discussions. In this respect, the current models of imperfect labour markets appear still to be narrowly focused on certain aspects of

10

Regulating for Decent Work

the labour market, thus failing to capture other important dimensions of the kind discussed below (for example, Kaufman 2007). In this sense, even empirical studies within non-mainstream models may underestimate the potential benefits of labour regulations. A related problem is the tendency to portray the macro-level outcomes of labour regulations in terms of individual enterprise responses. In other words, the macro-outcomes are reduced simply to the sum of these individual responses. This view does not recognize the welldocumented phenomenon that individually rational responses can produce inferior outcomes at the aggregate national level. For instance, during the current economic crisis, many observers have been concerned about the risk that competitive wage reductions (even to below the level of the minimum wage) deployed as a rational enterprise strategy to secure comparative advantages would create shortfalls in consumption demands, in turn making economic recovery more difficult (ILO 2008). The recognition of this fallacy of composition, and of the need to improve collective action among firms, is an important element of an improved understanding of labour regulations, and permits the elaboration of potential benefits. With these observations in mind, it is possible to identify at least three distinct modes in which labour regulations create benefits. Labour regulations as second-best instruments of improving workers’ welfare: ‘risk sharing’ versus ‘rent sharing’ The characterization of labour regulations as the outcome of rent-seeking activities, particularly by ‘insiders’ at the expense of ‘outsiders’, that dominates the economic literature obscures an at least equally important dimension of labour regulations, as the ‘second-best instruments of risk sharing’ (Agell 2002). In a market economy, workers are exposed to various types of risks, notably employment and income risks, which are inherent in the market system itself; yet markets tend to provide only incomplete insurance against these risks. It is difficult to imagine, for instance, that the private insurance market would offer complete insurance to workers for income risks (Bertola 2009). It is known that in the absence of comprehensive insurance, the resulting levels of labour inputs and productivity are suboptimal (see Pissarides 2001, 2010). Further, the risks workers encounter in the labour market tend to be underestimated by individual firms. For instance, job characteristics such as security and workload affect not only job-holders and firms, but also society as a whole, which is not typically taken into account in firm decision making. It follows that, in addition to the limitations of private

Sangheon Lee and Deirdre McCann 11

insurance against labour income risks, firms tend to underestimate the real cost of income or job loss, and thereby to offer suboptimal levels of job quality, including wages and job security. In these circumstances, as Cahuc and Zylberberg (2006) have observed with respect to employment protection, ‘the value of a job for the collectivity – its social value – does not coincide with its private value’ (p. 89, emphasis in the original), and ‘the state must then intervene in order to realign the interest of the firm with that of the collectivity’ (p. 91).2 This kind of approach is compellingly pursued by LaJeunesse (Chapter 8), whose reformulation of the regulatory field of working time is based on a recognition of the role of hours reduction in risk sharing, in that it is premised on a sharing of the costs of unemployment, rather than on permitting these costs to be borne by a relatively small proportion of workers (p. 12). This emphasis on labour regulations as ‘risk sharing’ (rather than ‘rent sharing’) instruments have important implications for two issues that have galvanized the research and policy debates in recent years: globalization and ‘employability’. First, if globalization is to be understood as rendering an economy more open to the rest of the world, and thereby heightening the interconnectedness of national economies, it would also render those economies vulnerable to external shocks and, in so doing, increase the overall risk and uncertainty in the labour market (ILO 2004; Scheve and Slaughter 2004; cf. Bourguignon and Goh 2004). In this case, contrary to the conventional wisdom, globalization is likely to increase workers’ need for public interventions that protect against these increased risks. It should therefore come as no surprise that, among advanced economies, the level of economic openness measured by trade share of GDP is positively correlated with the intensity of job protection and the generosity of unemployment benefits (Agell 1999). If this logic can be extended further, it can be argued that the integration of a national economy into the global economy will be facilitated, rather than discouraged, by the introduction of more effective worker protection mechanisms. For the sake of clarity, we do not argue for a ‘more is better’ approach to labour protection. As will be discussed later (see Figure 1.1), caution should also be directed at the risk of ‘over-regulation’. The point is that economic developments of increasing complexity and sophistication (such as globalization) require corresponding mechanisms that evolve to address new and increasing risks, as is illustrated by the recent policy experiences in Brazil (see Berg in this volume). The second observation is in response to the popular argument that, faced with rapid technological and economic changes, workers should no longer expect more secure forms of employment, and instead accept

12

Regulating for Decent Work

mobility and therefore improve their ‘employability’. In this view, the ‘flexible’ or ‘mobile’ worker is portrayed as the ideal-type of the future generation of workers, since he is able to ensure firms’ adaptability. In effect, the demand is that workers should be encouraged to take more risks in the labour market. However, given that such a change would involve higher risks for individual workers, the drive towards flexible work may not be successful unless the higher risks are taken care of in alternative ways. In other words, as Schmid (2006) has pointed out, ‘[N]ew securities are a precondition for encouraging riskier decisions such as moving from dependent employment to self-employment or even embarking on the venture of changing occupation’ (p. 28). In fact, one can argue that this approach, which may be called ‘riskneutralizing’ flexibility was successfully implemented at the enterprise level in Japan, where functional flexibility (or a high level of adaptability from workers to accommodate substantial flexibility in their job content) was enhanced by the provision of life-long employment security. This understanding of labour regulations as instruments for risk sharing suggests that economic studies have often put too much weight on the disincentive or ‘moral hazard’ effects of labour regulations (see, for example, Dufwenberg and Lundholm 2001). Legal regulation of unemployment insurance is a case in point. It is commonly argued that generous unemployment benefits reduce work incentives and thus contribute to an increased level of unemployment. However, this possibility should be viewed in relation to the ‘liquidity effects’ of unemployment benefits (Boeri et al. 2008). When such schemes exist, workers are more likely to be able to address liquidity constraints during the period of unemployment. The resulting improved stability of household consumption not only improves workers’ well-being but also contributes to macroeconomic stability. These welfare-enhancing effects, which have attracted renewed interest during the current economic crisis, can be expected to be particularly strong for poor households that are vulnerable to external income shocks. If these effects are not taken into account, the negative effects of unemployment insurance, which rest solely on the notion of moral hazard, will inevitably be overstated. Labour regulations as ‘beneficial constraints’ The discussion so far has concentrated on the role of labour regulations in improving workers’ welfare in relation to labour market risks. While such improvements may have beneficial impacts on employers, such as through productivity increases, it is also significant that labour regulations are often developed directly to address employers’ problems, which

Sangheon Lee and Deirdre McCann 13

are especially acute in the context of market failure. One prominent aspect of labour regulations in this respect is their role in addressing coordination failures among employers. Labour regulations need to be understood as ‘devices for coordinating the expectations of actors under conditions of uncertainty’ (Deakin in this volume, p. 35). For instance, we have already seen that the provision of better wages and working conditions harbours potential productivity gains, but the question remains of why it should not be initiated by individual employers rather than through regulation. This is related to the collective nature of such gains. One of the classic examples is working time regulation. It is well known that excessive hours are harmful to both workers’ well-being and productivity, at least in the medium and long term. Many employers are also aware of this and, in the latter part of the nineteenth century, some introduced factory-level initiatives to cut hours. The problem, as is shown by the UK history of industrialization, is that individual initiatives to reduce hours to a reasonable level may not be effective when other firms do not follow such initiatives in order to make additional short-term gains from excessive hours, such as labour cost advantages. Therefore, employers may find themselves in a ‘prisoners’ dilemma’ which requires coordinated actions to reach socially optimal outcomes (see Lee and McCann forthcoming). While different regulatory arrangements can be developed depending on the way different institutional elements are articulated,3 it can be said that the presence of regulations would help in directing the labour market away from the trap of long hours. Similarly, it has been argued that the labour market impacts of minimum wages, job protection and unemployment benefits need to be considered, including their positive impacts on developments in relation to workers’ levels of training and skills, thereby upgrading the employment structure towards good-quality jobs with high wages (see Bertola 2009 for a review). As with the effects of employment benefits on workers’ search behaviours, the unregulated labour market tends to result in an inefficient equilibrium in which low-wage jobs are ‘over-created’. In this situation, employers who can afford more high-wage jobs may have little incentive to create them. The introduction of minimum wages, for instance, can encourage employers to offer good jobs and improve the composition of jobs in order to achieve productivity gains (see, for example, Acemoglu 2001; Mortensen and Pissarides 1999). As Agell (2004, p. 259) puts it (in rather ironic fashion), ‘[I]t is important to note that it is the very fact that the minimum wage creates unemployment among unskilled workers (i.e., the standard objection to minimum wages) that leads to an upgrading of

14

Regulating for Decent Work

the human capital stock.’ In this sense, labour regulations can have ‘push effects’. These observations support the view that, as Deakin emphasizes in this volume, labour regulations are essentially ‘endogenous’ solutions to coordination problems. The beneficial impacts that labour regulations offer to employers as a whole through influencing their behaviours are extended further by Streeck (1997), who has drawn on the Schumpeterian concept of entrepreneurship to highlight the dynamic roles of labour institutions in general. He argues that the essence of capitalist entrepreneurship, and hence economic development, is to convert ‘constraints into opportunities’ through innovation. As noted earlier, when labour regulations are intended to change incentive structures for employers, the responses of employers would be either resistance (or non-observance) or to adjust their strategies (and interests) in line with the new regulatory framework. If the latter strategy pays off, labour regulations can be characterized as ‘beneficial constraints’. In a more general framework, Streek (2004) suggests that ‘the same long-term upgrading of the economic and organizational capabilities of firms that is caused by competitive product markets may also and in basically the same way be caused by social and political regulation’ (p. 432). Combining the two aspects of labour regulations, as risk-sharing tools and beneficial constraints, one can expect that they are associated with productivity improvements. Indeed, some of the empirical studies which explicitly recognize these potential benefits confirm a positive correlation between labour regulations and productivity (for example, Storm and Naastepad 2007). However, as emphasized earlier, it should be remembered that the productivity-enhancing role of labour regulations depends upon their design and implementation as well as the overall economic environment. In this sense, the actual impact of labour regulations is ‘implementation-specific’ (Bassanini and Duval 2006; Boeri et al. 2008; Bertola 2009). Labour regulations as conduits for social values While the recognition that there are potential economic benefits to be derived from regulating labour markets is fundamental to identifying new avenues for research and policy, it is nonetheless insufficient. A further critical element in realigning the narrow economic frameworks dominated by cost–benefit analyses is to acknowledge the contribution of labour market regulation to a broader set of values. The failure of the neoclassical economic tradition to reflect the social objectives served by labour laws has always been a central critique of this literature.

Sangheon Lee and Deirdre McCann 15

The recent turn to regulatory indicators has merely reinforced this tendency, in that they are designed to recognize only economic outcomes (Berg and Cazes 2009; Lee and McCann 2008). Thus, the complex relationship between labour regulations and social values deserves substantially more research attention. That is to suggest, for example, that even if such regulations prove to have certain negative impacts on employment or other quantitative variables, these detriments should be evaluated alongside their contribution to social goals. Academic traditions beyond the economic debates indicate how to integrate a broader set of objectives into theoretical perspectives on labour market regulation. The notion of labour regulations as a repository of social values has always, for example, been prominent in the scholarly literature in the labour law tradition. Indeed, this literature is increasingly turning to the elaboration and analysis of labour protections as a species of ‘human rights’, or otherwise fundamental in nature, perhaps in the search for a narrative as compelling as the neoclassical model (see, for example, ILO 2004; Alston 2005; Fredman 2008; Fenwick and Novitz 2010). Other literatures across a range of disciplines suggest the notion of compatibility between social and economic goals, often by drawing on Sen’s capability approach and its central insight into the interlinked nature of social and economic development (Browne et al. 2005; Lee and McCann 2006; Berg and Kucera 2008). The contemporary debates on the impacts of various regulations are confirming the need to recognize the social values that can be pursued through labour regulations. This is the case, for example, in the rather paradoxical debates on the influence of minimum wage regulations. As part of the revived quest for equity in the deregulatory project, mentioned above, minimum wage sceptics have argued forcefully that these legal measures damage the interests of women by reducing their employment opportunities. Setting aside the empirical validity of this assertion, as Rubery and Grimshaw argue in this volume, it fails to recognize the role of minimum wage laws in combating gender-based pay discrimination. Indeed, the authors contend that the contribution of minimum wage laws to improving women’s wages should be recognized as an objective in its own right, since women tend to be overrepresented in minimum-wage jobs and therefore to benefit more than male workers from minimum wage hikes. The contributions by Vosko and Tomei advance similar contentions: both highlight the importance of legal regulation for a range of precious working arrangements, which tend to include the most vulnerable groups of female workers.

16

Regulating for Decent Work

Moreover, there is a growing body of evidence that workers’ own commitment to notions of equity or fairness may translate into productivity outcomes. The productivity effects of these normative values can be understood if we take into account the incompleteness of the employment contract: the impossibility of specifying in detail a priori the level and quality of workers’ efforts to be required by the employer. Recent experimental studies have shown workers to adjust their efforts in response to their evaluations of the employer’s commitment to fairness (see Fehr and Gächter 2000; Fehr et al. 2009). This observation may explain why employers are often reluctant to seek short-term gains at the expense of their employees, preferring instead to be perceived as ‘generous’. As an example, standard economic theory predicts wage cuts in an economic downturn, due to both underbidding by workers and undercutting by employers. Yet this strategy is not as common as might be expected, apparently due in large part to the concern that such actions would be perceived to be unfair, and labour productivity suffer as a result (Bewley 1999). ‘Underregulation’ bias It has been argued in this section that the potential benefits of labour regulations, both economic and social, should be more explicitly and extensively considered in the analysis than is common in the literature. To conclude this discussion it can be suggested that the failure to take account of these benefits creates what might be termed an ‘underregulation bias’. The above discussion can be summarized in graphic form in line with Figure 1.1, which provides an illustration of the net benefits of labour regulations. The curves in Figure 1.1 demonstrate that the net benefits of labour regulations depend upon their level, and that this relationship can be portrayed as an inverse U shape. This confirms that societies can suffer equally from regulations that are either ‘too high’ or ‘too low’. The minimalist view, which proposes underregulation, is illustrated in the curve with the optimal degree of regulation Da’. Yet, in light of the aforementioned potential benefits which are both economic and social, this assumed level tends to fall short of the actual optimal level of Da*. It should also be noted that labour regulations are both endogenous and implementation-dependent (see Deakin in this volume). As a result, there is no reason to assume that the benefits curve would be the same across countries. In fact, it is more reasonable to expect that the curve would vary, as illustrated by the curve for country B in Figure 1.1,

Sangheon Lee and Deirdre McCann 17

Net benefits of labour regulations

Net benefit curve for country A

Minimalist view for country A

Da'

Da*

Net benefit curve for country B

Db*

Degree of labour regulations

Figure 1.1 Net benefits of labour regulations: an illustration Source: Adapted from Wright (2004, Figure 2).

and confirming that great care should be taken in the cross-country comparative analysis of labour regulations. The figure also shows the inherent risks in the existing indicators of labour regulations, which assume the existence of a universally applicable threshold for ‘good’ labour regulations. These issues are returned to in the remainder of this chapter. The indeterminacy of labour market regulation To complement this awareness of the potential benefits of labour market regulation, it is imperative also to appreciate the complexity and indeterminacy of the regulatory frameworks that harbour them, an insight that applies both to research agendas and policy-sphere decision making on regulatory design and implementation. The notion of regulatory indeterminacy 4 embraces efforts to escape the regulation/deregulation dichotomy that is offered by the neoclassical economic tradition. At the most basic level, it implies an awareness, now widespread, that ‘deregulation’ in the neoclassical orthodoxy does not imply an absence or attenuation of regulation, but rather the retention and reinforcement of institutional frameworks that support market ordering (for example, Standing 1999). On a deeper level, it implies that the abstract and static depiction of legal measures in conventional economic theory should be replaced by models that capture the intricacies of regulatory design and implementation. It has become increasingly apparent that neoclassical analyses carry a distinct narrative of the functioning of regulatory frameworks, which erases the complexities and indeterminacies of their

18

Regulating for Decent Work

design and operation. This model of labour market regulation has been honed and exposed by the quantification project. In the indices, labour laws are assumed to be of precisely-defined coverage and comprehensively adhered to: ‘“perfect” legal ordering to go with the idea of a perfectly competitive market’ (Deakin in this volume, p. 35). Labour regulation indices The need to inject complexity into research design is therefore of particular relevance to the construction of legal indicators. In efforts to improve the accuracy of the existing indicators, a number of elements can be suggested as being worthy of increased attention. First, the level of regulatory detail captured by most of the existing indicators is inadequate. They rarely take account of exceptions or derogations from legal standards, permissible ‘trade-offs’ between related protections or penalties for non-compliance (Lee and McCann 2008). Instead, the indices tend to grasp only the elements of regulatory measures that can be characterized as their ‘primary standards’: the level of the minimum wage, for example, or the daily or weekly working hours limit.5 Contrary to the assumptions that underlie current measurement work, however, regulatory details are of considerable importance in assessing the influence of regulations, and overlooking them risks producing ill-judged policy conclusions. As an illustration, as alluded to above, a number of indices that measure the ‘generosity’ of social benefits have been used to support the policy conclusion that generous benefits are a disincentive to engage in paid employment, captured in notions such as ‘moral hazard’. Yet limited attention has been paid to the regulatory details that govern eligibility conditions for such benefits and the monitoring mechanisms and sanctions that form part of these regimes. The indices have also been criticized for their deficient grasp of the regulatory subject. This point is illustrated by the unselfconscious adoption of the ‘standard worker’ as the model for the EWI. The Index assesses the application of legal measures to a male full-time worker employed by the same firm for 20 years who has a ‘non-working’ wife and two children, is of the country’s majority race and religion and is not a union member. Similarly, the model of the employer is a manufacturing company that employs 250 workers (Botero et al. 2004; see Berg and Cazes 2009). This attachment to the standard models of both worker and employer persists in an era when other strands of scholarship on labour market regulation are vigorously exposing the limitations of these models (for example, Vosko 2000; Freedland 2003; Bosch 2004; Fudge and Owens 2006; McCann 2008). It precludes comparison of the

Sangheon Lee and Deirdre McCann 19

legal treatment of the range of ‘non-standard’ forms of work of the kind examined by Vosko and Tomei in this volume, and therefore fails to capture the nature of the labour market engagement of large numbers of women, minorities and migrant workers. The indices also suggest a model for regulatory frameworks that may only apply to a very small percentage of the labour force of developing countries, whose economies tend to be dominated by small firms. Another dimension of regulatory indeterminacy that eludes these indices derives not from the form of regulatory instruments, but from their implementation. This limitation derives from their incapacity to measure ‘legal effectiveness’ or the extent to which the prescriptions of regulatory measures are reflected in the practices of working life (on the notion of legal effectiveness, see further Lee and McCann 2008). The available indicators have been recognized as failing to take account of the implementation of the labour market regulations that they purport to measure (Aherling and Deakin 2007; Berg and Cazes 2007; Lee and McCann 2008). This limitation could be considered substantial, although possible to remedy, were the indicators not drawn on to ground policy advice on regulatory frameworks. Where, as in the case of the EWI, they form the basis of such guidance, the failure to account for divergence in the de jure and de facto reach of legal measures becomes a fundamental flaw in indicator design. The consequence of these limitations is that, despite laudable attempts to overcome the statistical challenges of measuring and comparing legal interventions, the reliability of most of the available indicators as guidelines for the design of regulatory frameworks remains highly questionable (see Lee et al. 2008 for a review). The statistical relationship between the indicators and key employment variables is characterized as the impact of labour regulations. Yet, as we have seen, the existing indicators suffer from sizable ‘noise’, or uncertainty (Hoyland et al. 2009). There are therefore considerable risks in using them to generate country comparisons and rankings. Not least, and particularly with respect to the highly visible EWI, they may encourage ‘rank-seeking behaviour’, as governments endeavour to improve their ranking in the Index through reforms that have at best an uncertain relationship to the actual economic and regulatory contexts of their countries (Høyland et al. 2009; for country examples, see Bakvis 2009). Research and policy The recognition of both the formal and de facto indeterminacy of labour market regulation can advance research and policy directions

20

Regulating for Decent Work

beyond indicator-based research. In this regard, the notion of regulatory indeterminacy can be unpacked to reveal a number of distinct dimensions. The indeterminacy of regulatory outcomes, first, is closely tied to unpredictability in the response of labour market actors to regulatory reforms (Mitchell and Arup 2006). The role of the judiciary and other interpretive actors has long been recognized. Judges and other decision makers often interpret legislative measures inconsistently with the objectives asserted in the policy literature from which they emerged. Indeed, highlighting such divergence is a staple analytical approach of labour law scholarship (see Anderman 2000). Yet the responses of other actors are also highly influential. Employer responses are known to be critical in determining the influence of regulations. In particular, neither predictions of employer behaviour drawn from narrowly-focused economic analyses, nor the scale of political resistance to a given reform, appear to be useful guides to the evaluation of such measures (Streeck 2004). To illustrate, a study by Edwards et al. (2004) on the impacts on small enterprises of the politically controversial UK National Minimum Wage reported substantial variations in enterprise response and cautioned against exaggerated rhetoric about the impact of such legislation. Similarly, the progressive introduction of a 40-hour work week in the Republic of Korea has been relatively smooth, despite initial predictions of widespread non-compliance (Lee 2003). Also, while wage regulations at the national level (a common focus of empirical studies) are generally predicted to generate rigid wage adjustments, actual wage flexibility can nonetheless be high due to rapid firm-level wage adjustments (see Cardoso 2006 on the Portuguese case). Cooke’s essay in this book explores the actions taken by employers in China to evade the regulatory restrictions introduced by recent laws on labour contracts, employment promotion and dispute resolution. These include forming alliances with other actors such as local government departments and employment agencies. In addition, Cooke’s research reveals that the implementation of this legislation has also been influenced by the emergence of new labour market actors. That bodies such as employers’ associations, employment agencies, human resource management departments, human resources (HR) consultancy firms and non-governmental organizations would be prompted to engage more actively in employment relations was an unintended consequence of the labour law reform process. Cooke’s analysis suggests that an understanding of the role of such ‘new’ actors can be expected to be of critical importance in evaluating the impacts of labour regulations, especially in transition economies such as China.

Sangheon Lee and Deirdre McCann 21

At the conceptual level, such empirical findings have recently prompted reflection on whether legal measures can even be assumed to embody a regulatory ‘purpose’, in the sense of an objective that is clear-cut, readily identifiable and pursued in a linear manner (Mitchell and Arup 2006; Murray forthcoming). One suggestion is to reverse the conventional sequencing of regulatory design and assessment: to permit the findings of empirical sociolegal research to determine the purpose of their regulatory subject (Mitchell and Arup 2006). This recognition of indeterminacy in regulatory purpose might be hoped to prompt further investigation of the circumstances in which implementation will be likely to sustain the protective potential of labour regulations. It may also suggest that certain regulatory strategies harbour a capacity for regeneration in novel social and economic contexts. Examples are offered in this collection, in Lajeunesse’s analysis of working time regulation in its long-submerged role of averting unemployment and Grimshaw and Rubery’s reconfiguration of minimum wage regulation as a gender equality strategy. More authentic understandings of the operation of labour regulations also recognize the potential for their divergent application across the range of regulatory subjects. As precarious work and rising informality prise regulatory frameworks from their traditional subjects and reinforce existing regulatory lacunae, the tasks of discerning how regulatory frameworks govern different labour market constituencies is pressing. As noted above, investigations of the legal treatment of non-standard forms of labour market engagement are already relatively advanced. A particularly promising avenue for future work would be to investigate the extent to which the regulatory treatment of workers can be attributed to regulatory frameworks’ embrace of diverse forms of employing entity (Collins 2006; Davies and Freedland 2006). Further, labour regulations do not operate in a vacuum but generally in conjunction with related regulations or labour market institutions. As a result, institutional complementarities or compatibilities are significant in determining the outcomes of labour regulations. It is now widely recognized, for instance, that both unemployment insurance schemes and employment protection laws are devices to address workers’ income risks, and thus that they can substitute for each other to some extent (Bertola 2009). Indeed, these complementarities underlie the Danish model of flexicurity. Similarly, the performance of minimum wage policies is related to the extent and structure of collective bargaining. Weak bargaining fuels the need for strong minimum wage legislation; yet minimum wages policies, if implemented in an incoherent manner, risk

22

Regulating for Decent Work

‘crowding out’ collective bargaining (Aghion et al. 2008; ILO 2008). This point is illustrated by Berg’s essay on the formalization of the Brazilian labour market, which she attributes to the interaction of a range of policy and regulatory tools, from macroeconomic and educational policies to improved labour inspection and awareness-raising. Finally, research that explores the operation of regulatory instruments must resist a set of influential, yet simplistic, assumptions about the ways in which these measures operate, particularly in low-income settings. This point has become more pressing as labour market regulation is increasingly conceptualized from an international perspective (see, for example, Conaghan et al. 2002; Berg and Kucera 2008). The limited availability of data on developing countries is a profound constraint in understanding the operation of regulatory measures in these settings; yet the problem is not solely one of data availability. The problem of legal effectiveness, for example, unfolds in part as an assumed irrelevance of state legal norms to substantial segments of developing country labour markets, and, in particular, to firms and working relationships that are classified as forming part of the ‘informal economy’ (see, for example, World Bank 2010). Such assumptions, however, are based on simplistic characterizations of labour regulations that do not attend to the complexities of their functioning in low-income settings. Although there is a dearth of empirical studies on the actual effects of labour laws in developing countries (see Boeri et al. 2008), contributions to this volume offer indications of a range of factors that should be taken into account. Both Berg and Islam, for example, reveal that the available data do not support the claim that labour regulations advance informality in either Brazil or Indonesia. Indeed, they conclude that the experience of these countries confirms the significant role of regulatory intervention in promoting formalization. Lee and McCann challenge a related assumption about the relevance of statutory standards: that workers in developing countries are generally unaware of their entitlements under protective legal measures. Their research on awareness in Tanzania suggests instead a developed legal culture in which awareness of norms is relatively high and workers are prepared to take action to assert their legal rights. Frey also rejects the understanding of the nature of regulatory norms reflected in the orthodox economic literature by developing a more complex notion of regulation, as ‘holistic institutions, comprised of rules, norms and actual behaviours’ (p. 339) (see also in this vein Arthurs 1996). A further guise in which the assumed irrelevance of formal legal norms appears in the mainstream literature is the widespread dismissal

Sangheon Lee and Deirdre McCann 23

of conventional enforcement institutions. This account expects labour inspectorates in developing countries to be ineffective, whether as the result of lack of funding, limited capacity or corruption (see generally Fenwick et al. 2007). A literature is evolving, however, that questions such generalizations by highlighting countries in which enforcement mechanisms operate relatively effectively (for example, Piore and Shrank 2008; Pires 2008). In this volume, Pires addresses the characterization of labour inspectorates as inescapably legalistic and bureaucratic. He concludes that, under certain conditions, structures and strategies that manage discretion can create incentives for labour inspectors to adopt innovative practices. It is our hope that the chapters in this collection will encourage further research in this direction as part of broader efforts to investigate the operation of labour market regulation.

Notes 1. For details on the Network, see http://rdw.law.unimelb.edu.au/. 2. Of course, this is not to imply that any form of government intervention can be justified; this market failure is a necessary, but not sufficient, condition for such interventions. 3. It is this dynamic that permits the identification of different types of ‘working time regimes’ (see Anxo and O’Reilly 2000; Lee et al. 2007). 4. Deakin uses a different, although related, notion of ‘regulatory indeterminancy’ to connote the uncertain economic effects of legal rules (Deakin in this volume, p. 36). 5. For a context-sensitive deployment of the notion of ‘primary standards,’ see Lee and McCann (2008).

References Acemoglu, D. 2001. ‘Good jobs versus bad jobs’, Journal of Labour Economics, vol. 19, no. 1, pp. 1–21. Agell, J. 1999. ‘On the benefits from rigid labour markets: norms, market failures, and social insurance’, Economic Journal, vol. 109, pp. F143–F164. Agell, J. 2002. ‘On the determinants of labour market institutions: rent seeking vs. social insurance’, German Economic Review, vol. 3, no. 2, pp. 107–35. Agell, J. 2004. ‘Efficiency and equality in the labour market’, CESifo Economic Studies, vol. 50, no. 2, pp. 255–78. Aghion, P.; Algan, Y.; Cahuc, P. 2008. ‘Can policy interact with culture? Minimum wage and the quality of labor relations’, NBER Working Paper no. 14327 (Cambridge, MA). Aherling, B.; Deakin, S. 2007. ‘Labour regulation, corporate governance and legal origin: a case of institutional complementarity?’, Law & Society Review, vol. 41, pp. 865–908.

24

Regulating for Decent Work

Alston, P. 2005. Labour Rights as Human Rights (Oxford: Oxford University Press). Anderman, S. 2000 ‘The interpretation of protective employment statutes and contracts of employment’, Industrial Law Journal, vol. 29, no. 3, pp. 223–42. Anxo, D.; O’Reilly, J. ‘Working time regimes and transitions in comparative perspective’, in O’Reilly, J.; Cebrián, I.; Lallement, M. (eds) Working Time Changes: Social Integration through Transitional Labour Markets (Cheltenham: Edward Elgar). Arthurs, H. 1996. ‘Labour law without the state’, University of Toronto Law Journal, vol. 46, pp. 1–45. Arup, C.; Gahan, P.; Howe, J.; Johnstone, R.; Mitchell, R.; O’Donnell, A. (eds) 2006. Labour Law and Labour Market Regulation (Sydney: The Federation Press). Bakvis, P. 2009. ‘The World Bank’s Doing Business report: a last fling for the Washington consensus?’, Transfer: European Review of Labour and Research, vol. 15, nos 3–4, pp. 419–38. Bassanini, A.; Duval, R. 2006. ‘Employment patterns in OECD countries: reassessing the role of policies and institutions’, OECD Social, Employment and Migration Working Papers no. 35 (Paris: OECD). Basu, K.; Maertens, A. 2007. The Pattern and Causes of Economic Growth in India, Center for Analytical Economics (CAE) Working Paper 07–08 (Ithaca, NY: Cornell University). Benjamin, P. 2007. ‘Beyond the boundaries: prospects for expanding labour market regulation in South Africa’, in Davidov, G.; Langille, B. (eds) Boundaries and Frontiers of Labour Law (Oxford: Hart), pp. 181–284. Berg, J.; Cazes, S. 2009. ‘Policymaking gone awry: the labor market regulations of the Doing Business indicators’, Comparative Labor Law and Policy Journal, vol. 29, no. 4, pp. 349–82. Berg, J.; Kucera, D. 2008. ‘Labour institutions in the developing world: historical and theoretical perspectives’, in Berg, J.; Kucera, D. (eds) In Defence of Labour Market Institutions: Cultivating Justice in the Developing World (Geneva and London: ILO and Palgrave), pp. 1–31. Bertola, G. 2009. ‘Labour market regulation: motives, measures, effects’, ILO Conditions of Work and Employment Series no. 21 (Geneva: ILO). Bewley, S. 1999. Why Wages Don’t Fall During a Recession (Cambridge, MA: Harvard University Press). Boeri, T.; Helppie, B.; Macis, M. 2008. ‘Labour regulations in developing countries: a review of the evidence and directions for future research’, World Bank SP Discussion paper no. 0833 (Washington, DC: World Bank). Boeri, T.; van Ours, J. 2008. The Economics of Imperfect Labor Markets (Princeton, NJ: Princeton University Press). Bosch, G. 2004. ‘Towards a new standard employment relationship in western europe’, British Journal of Industrial Relations, vol. 42, no. 4, pp. 617–36. Botero, J.; Djankov, S.; La Porta, R.; Lopez-de-Silanes, F.; Schleifer, A. 2004. ‘The regulation of labour’, The Quarterly Journal of Economics, vol. 119, no. 4, pp. 1339–82. Boulin, J.Y.; Lallement, M.; Messenger, J.C.; Michon, F. 2006. Decent Working Time: New Trends, New Issues (Geneva: ILO). Bourguignon, F.; Goh, C. 2004. ‘Trade and labor market vulnerability in Indonesia, Republic of Korea, and Thailand’, in Krumm, K.; Kharas, H. (eds) East Asia Integrates: A Trade Policy Agenda for Shared Growth (Washington, DC: World Bank and Oxford University Press).

Sangheon Lee and Deirdre McCann 25 Browne, J.; Deakin, S.; Wilkinson, F. 2005. ‘Capabilities, social rights, and European integration’, in Salais, R.; Villeneuve, R. (eds) Towards a European Politics of Capabilities (Cambridge: Cambridge University Press). Cahuc, P.; Zylberberg, A. 2006. The Natural Survival of Work: Job Creation and Job Destruction in a Growing Economy (Cambridge, MA: MIT Press). Cardoso, A.R. 2006. ‘Wage mobility: do institutions make a difference?’, Labour Economics, vol. 13, no. 3, pp. 387–404. Collins, H. 2006. ‘Multi-segmented workforces, comparative fairness, and the capital boundary obstacle’, in Davidov, G.; Langille, B. (eds) Goals and Means in the Regulation of Work (Oxford and Portland, OR: Hart), pp. 317–34. Conaghan, J.; Fischl, M.; Klare, K. (eds). 2002. Labour Law in an Era of Globalization: Transformative Practices and Possibilities (Oxford: Oxford University Press). Davies, P. and Freedland, M. 2006. ‘The complexities of the employing enterprise’, in Davidov, G. and Langille, B. (eds) Goals and Means in the Regulation of Work (Oxford and Portland, OR: Hart), pp. 273–94. Deakin, S.; Lele, P.; Siems, M. 2007. ‘The evolution of labour law: calibrating and comparing regulatory regimes’, International Labour Review, vol. 146, pp. 133–62. Dickens, W. 1984. ‘Occupational safety and health regulation and economic theory’, in Darity, W. (ed.) Labor Economics: Modern Views (Boston: KluwerNijhoff Publishing), pp. 133–73. Dufwenberg, M.; Lundholm, M. 2001. ‘Social norms and moral hazard’, The Economic Journal, vol. 111, no. 473, pp. 506–25. Edwards, P.; Monder, R.; Black, J. 2004. ‘Why does employment legislation not damage small firms?’, Journal of Law and Society, vol. 31, no. 2, pp. 245–65. Fehr, E.; Gächter, S. 2000. ‘Fairness and retaliation: the economics of reciprocity’, Journal of Economic Perspectives, vol. 14, no. 3, pp. 159–81. Fehr, E.; Goette, L.; Zehnder, C. 2009. ‘A behavioral account of labor market: the role of fairness concerns’, Annual Review of Economics, pp. 355–84. Fenwick, C.; Howe, J.; Marshall, S.; Landau, I. 2007. ‘Labour and Labour-related Laws in Micro and Small Enterprises: Innovative Regulatory Approaches’, ILO SEED Working Paper no. 81 (Geneva: ILO). Fenwick, C.; Novitz, T. (eds) 2010. Human Rights at Work: Perspectives on Law and Regulation (Oxford: Hart). Fredman, S. 2008. Human Rights Transformed (Oxford: Oxford University Press). Freedland, M. 2003. The Personal Employment Contract (Oxford: Oxford University Press). Fudge, J.; Owens, R. (eds). 2006. Precarious Work, Women, and the New Economy: The Challenge to Legal Norms (Oxford and Portland, OR: Hart). Heckman, J.J.; Pagés, C. (eds) 2004. Law and Employment: Lessons from Latin America and the Caribbean, National Bureau of Economic Research Conference Report (Chicago, IL: University of Chicago Press). Høyland, B.; Moene, K.; Willumsen, F. 2009. ‘The tyranny of international index ranking’, unpublished paper, University of Oslo. Independent Evaluation Group (IEG). 2008. Doing Business: An Independent Evaluation (Washington, DC: World Bank). International Labour Office (ILO) 1999. Decent Work (Geneva: ILO). ILO. 2004. Economic Security for a Better World (Geneva: ILO). ILO. 2008. Global Wage Report 2008/09: Minimum Wages and Collective Bargaining – Towards Policy Coherence (Geneva: ILO).

26

Regulating for Decent Work

Kaufman, B. 2007. ‘The impossibility of a perfectly competitive labour market’, Cambridge Journal of Economics, vol. 31, pp. 775–87. Lee, S. 2003. ‘Political economy of working time in Korea: tensions in the reduction of working hours,’ paper presented at the 15th Annual Meeting on Socio-Economics, Aix-en-Province, June. Lee, S.; McCann, D.; Messenger, J.C. 2007. Working Time Around the World: Trends in Working Hours, Laws and Policies in a Global Comparative Perspective (Abingdon and Geneva: Routledge and ILO). Lee, S.; McCann, D. 2006. ‘Working time capability: towards realizing individual choice’ in Boulin et al. Lee, S.; McCann, D. 2008. ‘Measuring labour market institutions: conceptual and methodological questions on “working-hour rigidity”’, in Berg, J.; Kucera, D. (eds) In Defence of Labour Market Institutions: Cultivating Justice in the Developing World (Geneva and London: ILO and Palgrave). Lee, S.; McCann, D.; Torm, N. 2008. ‘The World Bank’s “employing workers” index: findings and critiques – a review of recent evidence’, International Labour Review, vol. 147, no. 4, pp. 416–32. Lee, S.; McCann, D. forthcoming. ‘Negotiating working time in fragmented labour markets’, in Hayter, S. (ed.) The Role of Collective Bargaining in the Global Economy (Cheltenham: Edward Elgar). McCann, D. 2008. Regulating Flexible Work (Oxford: Oxford University Press). Manning, A. 2003. Monopsony in Motion: Imperfect Competition in Labor Markets (Princeton, NJ: Princeton University Press). Mitchell, R.; Arup, C. 2006 ‘Labour law and labour market regulation’ in Arup et al. Mitchell, R.; Gahan, P.G.; Stewart, A.; Cooney, S.; Marshall, S.D. 2010 ‘The evolution of labour law in Australia: Measuring the change’, Australian Journal of Labour Law, vol. 23, no. 1, pp. 61–93. Mortensen, D.; Pissarides, C. 1999. ‘Unemployment responses to “skill-biased” shocks: the role of labour market policy’, Economic Journal, vol. 109, pp. 242–65. Murray, J. forthcoming. ‘Continuity and reform: time, money and family life in the structural remodelling of Australian labour law’, International Labour Review. Owens, R. 2002. ‘Decent work for the contingent workforce in the new economy’, Australian Journal of Labour Law, vol. 15, no. 3, pp. 209–34. Piore, M.; Schrank, A. 2008. ‘Towards managed flexibility: the revival of labour inspection in the Latin world’, International Labour Review, vol. 147, no. 1, pp. 1–23. Pires, R. 2008. ‘Promoting sustainable compliance: styles of labour inspection and compliance outcomes in Brazil’, International Labour Review, vol. 147, no. 1, pp. 199–229. Pissarides, C. 2001. ‘Employment protection’, Labour Economics, vol. 8, pp. 131–59. Pissarides, C. 2010. ‘Why do firms offer “employment protection”?’, Economica, vol. 77, no. 301, pp. 613–36. Scheve, K.; Slaughter, M. 2004. ‘Economic insecurity and the globalization of production’, American Journal of Political Science, vol. 48, no. 4, pp. 662–74. Schmid, G. 2006. ‘Social risk management through transitional labour markets’, Socio-Economic Review, vol. 4, pp. 1–33.

Sangheon Lee and Deirdre McCann 27 Standing, G. 1999. Global Labour Flexibility: Seeking Distributive Justice (London and New York: Macmillan Press/St Martin’s Press). Storm, S.; Naastepad, C. 2007. ‘Why labour market regulation may pay off: worker motivation, co-ordination and productivity growth’, ILO Economic and Labour Market Paper 2007/4 (Geneva: ILO). Streeck, W. 1997. ‘Beneficial constraints: on the economic limits of rational voluntarism’, in Hollingsworth, J.R.; Boyer, R. (eds) Contemporary Capitalism: The Embeddedness of Institutions (Cambridge: Cambridge University Press). Streeck, W. 2004. ‘Educating capitalists: a rejoinder to Wright and Tsakalotos’, Socio-Economic Review, vol. 2, pp. 425–38. Vosko, L. (ed.). 2000. Precarious Employment: Understanding Labour Market Insecurity in Canada (Montreal/Kingston: McGill-Queen’s University Press). World Bank. 2009a. Doing Business 2010 (Washington, DC: World Bank). World Bank. 2009b. Revisions to the EWI Indicator. Available at http://www. doingbusiness.org/documents/EWI_revisions.pdf. Wright, E. 2004. ‘Beneficial constraints: beneficial for whom?’, Socio-Economic Review, vol. 2, pp. 407–14.

This page intentionally left blank

Part I Theories and Perspectives

This page intentionally left blank

2 The Evidence-Based Case for Labour Regulation Simon Deakin

Introduction There have been empirical studies of the effects of labour and employment laws since the inception of modern social legislation. However, until relatively recently, little attention was paid to the role of the legal system as a causal variable, with the potential to shape social and economic outcomes in its own right. Certain regulatory initiatives, such as minimum wages and employment protection laws, were studied intensively with a view to assessing their impacts on output variables such as unemployment and productivity, but without close regard to their legal form. For the purposes of economic modelling, laws were generally assumed to be self-executing, and issues concerning the incompleteness and imperfect operation of legal rules were left unexplored. As a result, notwithstanding the existence of a large body of literature employing increasingly complex statistical methods, issues of concern to legal researchers and policy makers were only marginally addressed. Over the course of the past decade the picture has changed radically. The role of legal systems in shaping the nature of regulation and, as a result, economic outcomes, has been placed centre stage by the highly influential legal origins hypothesis (La Porta et al. 2008), which has applications to labour law as well as to corporate and insolvency law where it was first developed. Important and influential as it is, the legal origins hypothesis is only one of a number of emerging paradigms in the social sciences with implications for the empirical study of labour law. The varieties of capitalism approach in comparative political economy has opened up a number of insights, stressing, in contrast to the legal origins literature, the endogeneity of legal and related institutions to wider political and economic structures (Hall and Soskice 2001). Despite 31

32

Regulating for Decent Work

their differences, these two approaches share an emphasis on comparative methods, and have prompted the development of new data sources to measure cross-national differences in legal regimes. There has been a proliferation of indicators which attempt to capture the extent of variation across countries in the contents of labour law rules. The growing availability of this kind of data has stimulated interest in statistical methods capable of isolating the impact of legal change on economic development and vice versa, particularly techniques for time series and dynamic panel data analysis. By virtue of these developments, conventional understandings concerning the nature of the law–economy relation are being reassessed, together with some previously accepted wisdom on the supposedly negative economic impacts of labour law regulation. If the advice given to policy makers has become, as a result, less clear-cut than it was, the policy-formation process should in principle benefit from an improved understanding of the role of national and local contexts in shaping the emergence and application of labour law rules. How far it will in fact do so raises questions concerning the feasibility, in practice, of ‘evidence-based policy’. This chapter will review the recent developments to which we have just referred. It begins with an account of theoretical perspectives before moving on to look at the nature of new data sources and techniques for analysing them and the state of the art in the empirical literature on the economic effects of labour law rules of different kinds, with the focus on wage regulation and employment protection. Issues arising from the use of empirical research to guide policy will then be addressed in a concluding section. The focus on this largely quantitative and comparative literature is not intended to imply that other approaches, such as those based on qualitative case study work or experimental studies, are not also valid or important ways of exploring the operation of labour laws; for reasons of space it is not possible to do more than provide a view of one part of what is a huge field. The selection of material here is intended to highlight recent theoretical and methodological innovations of particular interest to lawyers and policy makers and to relate them to the current state of the art on the empirical effects of labour legislation.

Theoretical perspectives on the economic and social impacts of labour laws and their relationship to empirical research The empirical literature on the impacts of labour laws has been shaped by theoretical perspectives, particularly those drawn from neoclassical

Simon Deakin 33

labour economics where the conventional understanding has been that labour law rules operate as an exogenous intervention in, or interference with, the operation of market forces. This idea is vividly captured by Stigler’s analysis of the minimum wage (Stigler 1946), which set the tone for much of the economic discussion of labour law regulation from the late 1940s onwards, and which directly inspired the economic analysis of labour law later carried out by Posner and other members of the Chicago School of law and economics (see Posner 1984). In the neoclassical model, wages and employment are set by the interaction of supply and demand for labour. The market operates as an implicit regulator of decisions to trade, so that firms which underpay risk losing their workers to competitors just as workers who overbid for wages risk exclusion from employment as firms substitute capital for labour or cease to trade. The market also favours equality, in the sense of equal pay for work of equal value; the spontaneous movement of the market to equilibrium ensures that a single price or wage is set for labour of comparable productivity. Where inequalities or imperfections are observed, they are ascribed to non- or pre-market factors, such as differences in individuals’ endowments or preferences for work or leisure, or to employer’s ‘tastes’ for discrimination (Becker 1957). Labour laws themselves are seen as an external source of imperfections; they originate in decisions made in the political sphere and reflect rent-seeking, or distributional demands, by collective groups. Rent-seeking, as it is aimed at redistribution rather than value creation, imposes a deadweight loss on the economy and so constitutes a source of inefficiencies. Further inefficiencies arise from distortions in the operation of the market which are induced by legal interferences with bargaining. Wage regulation (such as minimum wage or pay equity legislation, or the taking of wages out of competition through collective bargaining) depresses demand for labour. It thereby results in the exclusion from the market of those it is designed to help, that is, the low paid. Thus minimum wage laws, for example, are predicted to have particularly adverse effects for the young, workers without formal skills or qualifications, members of ethnic minority groups, and individuals seeking to return to the labour market after a spell of unemployment (Minford 1985). From this point of view, labour legislation is not just inefficient, but discriminatory and unjust in its effects. For a long period after the publication of Stigler’s 1946 paper and the ensuing debate, the neoclassical model proved impervious to empirical challenge. This was in part because of the elegance and simplicity of the underlying theory, and the lack of consensus over alternatives; most

34

Regulating for Decent Work

economists took the view that ‘it takes a theory to beat a theory while facts only pierce the theorist’s hide’ (Samuelson 1951, cited in Kaufman 2007, p. 8). In the early 1990s a challenge to the orthodox view emerged in the form of empirical studies apparently refuting the claim that minimum wage legislation caused higher unemployment. Card and Krueger (1995) demonstrated that US states which raised their minimum wages experienced increases in both earnings and employment levels for younger workers, while Machin and Manning and their colleagues demonstrated that the abolition of minimum wage legislation in the United Kingdom had led to falling employment in the affected sectors (Dickens et al. 1993; Machin and Manning 1994). These studies had a significant impact for a number of reasons. The empirical work was regarded as convincing in part because of the use of what were then relatively novel panel data techniques (including difference-in-differences analysis) to take advantage of the ‘natural experiment’ provided by variations in state-level provision (in the United States) and sectoral regulation (in Britain). We return to this point below in our discussion of the empirical literature on wage regulation (see section below on minimum wages). It also helped that, around this time, theory had developed to the point where a more systematic account could be given of labour market imperfections. The concepts of asymmetric information and ‘dynamic’ monopsony (Manning 2003) were used to explain how imperfections could arise from the spontaneous operation of labour markets, without the need to invoke external interferences such as those derived from trade union activity or labour legislation. Meanwhile, transaction cost economics (Williamson et al. 1975) and behavioural economics (Fehr and Falk 1999) were challenging the neoclassical assumption of perfect rationality on the part of labour market actors. The theoretical reappraisal represented by this work remains, however, somewhat incomplete. Rather than being seen as undermining the competitive model of the labour market, it has been understood as qualifying it in the narrow sense of showing that, in certain specific contexts, the conditions for perfect competition may not obtain. How far this is the case has come to be viewed as an empirical matter, and because more recent studies have found evidence of negative employment effects from the raising of minimum wage rates (at least in the US context; see section below on minimum wages), the consensus position is that monopsony is not an empirically significant phenomenon, ‘thus giving the neoclassical side and the competitive model the upper hand in the theoretical part of the battle’ (Kaufman 2007, p. 11).

Simon Deakin 35

The standard economic analysis of labour law rules is based not just on the idea of a self-equilibrating market which corrects itself in response to temporary dislocations; it also embodies, albeit less explicitly, a theory of the operation of legal rules. There are two linked aspects to this. The first is the idea, referred to already, that legal rules are exogenous to the market relations and so operate as an external imposition upon them. The second is the presumption that legal rules are complete in the sense of being certain in the scope of their application and selfexecuting in their implementation: ‘perfect’ legal ordering to go with the idea of a perfectly competitive market. A neo-institutional account of the operation of legal rules in market settings which sees them as ‘endogenous’ solutions to coordination problems (see Deakin and Sarkar 2008) offers an alternative perspective which has implications for empirical analysis. In this approach, legal rules are understood as devices for coordinating the expectations of actors under conditions of uncertainty. Laws are not simply imposed in a top-down fashion but, just as often, crystallize conventions which first emerge at the level of exchange relations before being formalized in contractual agreements and, at a further level, legal texts. As responses to market failures of various kinds, they originate in particular economic and political settings. Labour law rules may have redistributive aims, while at the same time giving expression to notions of fair treatment which serve as ‘focal points’ in bargaining between the parties (Hyde 2006). The employment contract is a complex legal ‘institution’ in this sense. It represents a standard form for labour transactions, based on a trade-off between the powers of coordination reserved to the employer and the employee’s access to mechanisms for income smoothing and insurance which help to offset risks associated with wage-dependence. The core institutions of labour law systems, including collective bargaining, minimum wage laws and employment protection legislation, express this fundamental trade-off in different ways, which reflect the solutions arrived at in particular national and regional contexts and which have been more widely diffused over time. To understand labour law in this way is to see it not as an external force imposed upon an otherwise ‘unregulated’ market, but as endogenous to market processes and political structures. ‘Endogenous laws’ offer solutions to coordination problems which are incomplete in their specification and imperfect in their application. They are neither self-executing, nor capable of being made operational by enforcement alone. Their successful implementation depends upon the presence of understandings and beliefs among market actors beyond the scope of

36

Regulating for Decent Work

the legal system. Just as the form and content of a legal rule is to a large extent the result of a process of evolution which is context-specific and path-dependent, so the operation of legal rules depends upon contextual factors which vary across time and space. This implies an empirical research agenda which acknowledges the likely indeterminacy of legal rules in terms of their economic effects. Because their origins are context-specific, their effects will not be constant across firms, industries or national systems. Their implementation will depend upon factors beyond the law at the level of social norms and self-regulatory practices, or upon the particular point in the economic cycle at which they are introduced. They may have both positive and negative effects in terms of their impact on efficiency, which may offset each other. To the extent that all or any of these things are the case, it becomes possible to predict the effects of labour law changes only if a good deal is known about the contexts in which they are applied. This implies an empirical strategy based on awareness of the importance of cross-national and regional diversity, and of the need to control for it when modelling the impacts of legal change. In an extreme form, the hypothesis of labour law’s indeterminacy is in danger of tipping over into a claim about its triviality – namely, that by virtue of its endogeneity to economic and political structures and the macroeconomic cycle, the legal system is simply a cipher or conduit for the operation of other forces. By contrast, the legal origins hypothesis holds that the content of laws, and, at a further remove, their economic impact, are shaped by the nature of the ‘legal infrastructure’ or ‘regulatory style’ which predominates in a given country. A few parent systems aside, most countries have inherited their ‘legal origin’ in the sense just described by virtue of an external event such as conquest or colonization, and only rarely through the conscious adoption of another country’s legal institutions. Legal origin can thereby be understood as an exogenous, long-run causal variable, explaining the different economic trajectories of national systems, and the persistence of cross-national diversity (La Porta et al. 2008). Common-law origin systems are said to have a bias in favour of market-creating rules and a laissez-faire approach to economic regulation, in contrast to an emphasis on government ownership and regulation, and on redistribution, in civil-law origin systems. The legal origins hypothesis is open to objection on points of detail, not least its oversimplified account of the common law/civil law divide, but its identification of the legal system as source of path-dependence and institutional lock-in in the development of market economies is

Simon Deakin 37

a significant theoretical contribution which has generated a large and growing empirical literature. In its strong form, it predicts that inherited differences in regulatory style feed through into economic performance outcomes, with common law systems experiencing faster growth by virtue of their greater efficiency. This version of the hypothesis is overly dependent on a few stylized (and arguably misspecified) facts concerning legal systems (for discussion, see Deakin 2009). It is also only weakly supported empirically, given that evidence for common law systems having faster growth rates than civil law ones has proved hard to come by (see La Porta et al. 2008, pp. 301–2). A weak form of the hypothesis may be more defensible: this is one in it is accepted that legal infrastructure could have a long-run influence on economic development, but which sees legal institutions as coevolving with economic and political structures rather than as exogenously determining them (Ahlering and Deakin 2007). A coevolutionary perspective is compatible with the varieties of capitalism approach. This posits a bifurcation of systems between ‘liberal market’ and ‘coordinated market’ economics (Hall and Soskice 2001), a distinction which more or less precisely matches that between the common law and civil law legal families, but which does not view the legal system as the key explanatory variable in the way that legal origin theory does. Instead, emphasis is placed on the role of political structures and the composition of interest group coalitions in shaping economic institutions over the long run. Thus a link has been claimed between proportional representation voting systems, which are said to favour coalition-building and consensus politics, and the emergence of redistributive labour law regimes, solidaristic welfare states and activist industrial policies in coordinated market systems. Majority-voting systems, by contrast, are said to favour the emergence of ruling coalitions which pursue a conjunction of liberal capital markets and flexible labour markets (Iversen and Soskice 2007). Putting to one side, for present purposes, the fine detail of these accounts, their broader significance lies in placing political-economy factors at the centre of the analysis, a view which does not rule out a role for the legal system in influencing outcomes, but which sees the operation of legal rules as embedded in, and contingent upon, the wider political and economic environment. This implies an empirical research agenda which focuses on possible complementarities between features of legal systems and those of economic and political institutions. For example, labour law rules of the kind which are characteristically found in coordinated market systems, such as stringent employment

38

Regulating for Decent Work

protection laws and codetermination-based regimes for employee representation, may be complementary to certain features of the corporate governance environment such as concentrated share ownership and bank-led monitoring. Conversely, flexible labour markets and liquid capital markets, supported by shareholder rights, may be complementary features of liberal market systems (Ahlering and Deakin 2007). In order to assess the economic impact of particular types of labour law rules in a way which is sensitive to these possibilities, data on a range of legal and institutional features are needed, as are techniques capable of isolating the influence of these different variables of interest.

Data sources on legal and institutional variables and techniques for analysing them The emergence of new datasets, quantifying the content of legal rules in the labour law field as well as in related areas of economic regulation including company and insolvency law, has occurred alongside growing interest in the legal system as a variable of interest and in the comparative analysis of institutions and outcomes. However, the status of these ‘synthetic indices’ as reliable data sources remains open to doubt, and their use in statistical analysis raises some problematic methodological issues. The most comprehensive of the new datasets, in terms of the countries and subject matter covered, is the labour index constructed by Botero et al. (2004) in one of a series of papers operationalizing the legal origins hypothesis. This index codes for the labour laws of over 80 developed and developing countries, and contains over 100 indicators which between them cover the areas of ‘employment law’, ‘collective labour relations law’ and ‘social security law’. For each indicator, a score is given to indicate the strength of worker protection in the country concerned, with higher scores (on the whole these are normalized on a zero to 1 scale) reflecting more protection. An algorithm shows, in each case, how the scores are worked out; in some cases, assumptions are made about the operation of the law on the basis of its formal content, in others cardinal variables (such as amounts of redundancy pay or the maximum permitted duration of working time) are used to arrive at the final score. The sources for the coding are simply described as ‘the laws of each country’ with reference also being made, in general terms, to relevant secondary sources; there is no attempt to provide specific legal sources for individual scores. The index reports the laws of the countries concerned at a loosely defined point in the late 1990s. The methods

Simon Deakin 39

used by Botero et al. are broadly similar to those employed in the first legal origin studies, which focused on shareholder and creditor rights (La Porta et al. 1998), but the labour index is more extensive, in terms of its coverage, than its predecessors. The methodology used in the legal origins literature forms the basis for the Doing Business reports of the World Bank. For this purpose the World Bank has developed an ‘employing workers index’ which consists of three sub-indices: a ‘rigidity of employment index’ which among other things covers hiring, working time and other terms and conditions of employment; an index of non-wage labour costs; and an index of firing costs. These sub-indices, in their turn, are broken down into a series of individual indicators. The scores are on a scale of zero to 100 with a higher value indicating ‘more rigid’ regulation. The sources for the codings include surveys of regulations carried out by local lawyers and officials, and the dataset is in the form of a time series updated annually, going back to 2004 (World Bank, various years). The main alternative to the Botero et al./World Bank index is the OECD’s index of the strictness of employment protection legislation (‘EPL’). This covers only part of the labour law field, and is confined to charting the law in 28 OECD member states, but goes back further than the World Bank studies, being based in part on an index first developed in the early 1990s by Grubb and Wells (1993). Four main data-gathering exercises have taken place, referring respectively to the late 1980s, the late 1990s, 2003 and 2008. The index, in its current form, consists of three components: rules affecting the dismissal of workers with regular (that is, indefinite or indeterminate) employment contracts; rules governing fixed-term and temporary agency work; and collective dismissal procedures. The scores are expressed on a scale of zero to 6, with 6 representing maximum ‘strictness’. An overall strictness indicator for each country is arrived at by combining the three sub-indices, with the collective dismissals indicator weighted at 40 per cent of the other two to reflect the extent to which it consists of rules which operate in a supplementary way to those of the other two (OECD 2004; Venn 2009). The two datasets tell a broadly similar story on the extent of crossnational variation in the content of labour laws. Botero et al. (2004) find that labour regulation is highly correlated with legal origin, with countries of French-law origin having the highest scores (indicating more protection for workers) and those of English-law origin the lowest (indicating less protection). The OECD index records Anglo-Saxon countries as having the lowest scores, with those in east Asia, northern Europe and the Nordic systems in the middle, and the highest scores

40

Regulating for Decent Work

being recorded by southern European countries. The main source of variation is the law governing fixed-term and temporary agency work. The OECD index indicates relatively little change in the content of the law in the period covered, the most noticeable trend being a limited degree of convergence brought about by deregulation since the early 1990s. How accurate is this picture of the state of the laws of different countries? It is doubtful that the World Bank datasets measure actual costs to firms of different legal regimes, as they purport to do. The values reported in the indices reflect estimates of the costs of regulation as they apply not to the average enterprise, but to larger firms employing workers on standard (indeterminate) employment contracts; thus they take as their focus the most protective or ‘rigid’ case (Lee and McCann 2007). Since it is assumed that the firm strictly abides by the relevant law but does not go beyond the level of protection mandated in the relevant legislation, not only is the issue of non-enforcement overlooked, but no account is taken of the extent to which firms would observe the standards set out in the law in any event (which ‘standard’ firms would in practice be the most likely to do). The authors of the OECD index more openly accept that it contains limitations which are ‘inherent to most synthetic indices’ including ‘problems of subjectivity, the difficult of attributing scores on the basis of legal provisions that may be applied differently in practice, and the choice of the weighting scheme used to calculate the summary indicator form from the various sub-components’ (OECD 2004, p. 99). They attempt to deal with the problem of the gap between the form of the law and its operation in practice by incorporating information on court practice and industry-level self-regulation through collective bargaining, while accepting that reliable data on some of these issues are not available for all countries (OECD 2004, p. 66). The labour regulation index developed by the Cambridge Centre for Business Research (CBR) provides an alternative approach to legal coding which attempts to address some of these methodological issues (Deakin, Lele and Siems 2007; Armour et al. 2009b). Information from collective agreements and other self-regulatory mechanisms which operate as the functional equivalents to formal laws in some systems is included, and algorithms are developed which allow for the codings to reflect the extent to which labour laws take the form of default rules, applying unless the parties agree otherwise, as opposed to being either completely mandatory or non-applicable. Explanations for codings, providing the primary legal source in each case, are provided (this is not the case with the World Bank index and is only partially achieved in

Simon Deakin 41

the OECD one). The CBR index is in the form of an annual time series, dating back to the early 1970s. Five countries are currently contained in it (France, Germany, India, the United Kingdom and the United States) and it covers the whole labour law field, which is broken down into five sub-indices (form of the employment contract, working time, dismissal, employee representation and industrial action). The picture of the law provided by the CBR index is not dissimilar to that given by the analysis provided in Botero et al. (2004), in that civil law systems are seen to have a significantly higher degree of regulation than common law ones (see Figure 2.1). However, the case of India is an exception; its labour laws are broadly comparable to those of Germany rather than to its ‘parent’ system, the United Kingdom. The time series dimension of the CBR index is also revealing; it suggests that there has been considerable change over time in the content of labour laws, particularly in relation to the treatment of flexible forms of work and to working time controls. It would seem that shifts in the political and macroeconomic environment are capable of bringing about significant changes to the substance of labour law, casting doubt on the idea of stable cross-country differences derived from legal origin, and emphasizing the endogeneity of the law to local political and economic contexts (Deakin, Lele and Siems 2007; Deakin and Sarkar 2008). The main purpose of developing indices of this kind is to carry out statistical analysis aimed at isolating the economic effects of labour laws; here, again, a number of methodological issues arise. Botero et al. (2004) carried out cross-sectional, bivariate analyses which found a number of negative impacts of labour regulation; in particular, higher scores on

France

35

Germany

India

UK

USA

30 25 20 15 10 5 0 1970

1975

1980

1985

1990

1995

Figure 2.1 Labour regulation in five countries, 1970–2005 Source: Deakin, Lele and Siems (2007).

2000

2005

42

Regulating for Decent Work

the labour index were correlated with lower male employment, higher youth unemployment, and a larger informal sector. As Pozen (2006) has pointed out, however, Botero et al.’s labour index is a weak overall predictor of economic outcomes, because of the small size of the identified effects, and a low level of statistical significance in some of the cases. In the case of the OECD index, cross-sectional analyses carried out on the data collected in the 2003 exercise failed to show the negative impacts posited by theory, although a clearer association was found between EPL strictness and flows into and out of unemployment when a time series analysis was undertaken (OECD 2004). The OECD’s 2004 Employment Outlook accepted that bivariate, crosssectional analyses could only provide limited information and should not be used to guide policy. The most important reason for this is the difficulty in drawing inferences about causation in studies of this kind. Evidence of a statistically significant correlation between, for example, a legal or institutional variable, on the one hand, and an economic outcome variable, on the other, need not necessarily indicate causal influence running from the former to the latter. The independent variable may be endogenous in the sense of being correlated to the error term in the regression equation. Where that is the case, a false result may be obtained. For example, a study which shows that more stringent employment protection legislation is correlated with higher levels of unemployment may not be evidence of the economic impact of that particular legal institution, but evidence instead of a change in the macroeconomic environment (higher unemployment) triggering a certain policy response (stricter controls over dismissals). There are a number of techniques available for getting round this problem of ‘reverse’ or ‘simultaneous’ causation. One is to identify an instrumental variable that is correlated with the endogenous regressor, but is not correlated with the error term. For example, in the early legal origins literature, the common law or civil law origins of different countries’ legal systems were used as instruments for the substance of legal rules on shareholder, creditor and worker rights. The assumption here was that legal origin was probably linked to the content of legal rules in the countries, but, conversely, could not itself have been caused by the economic outcome variables that were being studied (and so could not be correlated with the error term). This was because the ‘origin’ of a country’s legal system, and hence its regulatory style, was assumed to be the result of an external event such as colonization or conquest, rather than being internally generated by that country’s process of economic development. By demonstrating a statistical relationship between legal

Simon Deakin 43

origin and the different outcome variables, it was possible to show that the direction of causation ran from law to the economy rather than vice versa, offering support for their theory. However, there are problems with the instrumental variable approach. Not only is a degree of subjective judgement involved in the choice of an appropriate instrument; a variable will not be a good instrument if it could have influenced outcomes through channels other than that of the independent variable which is the main focus of interest. In later papers in the legal origin series, La Porta et al. came to the conclusion that legal origin might be influencing the economy through a number of routes, including interpretive practices and approaches to enforcement, in addition to that of the content of legal rules. On this basis they have now proposed that legal origin be seen as a causal or exogenous variable in its own right, rather than as an instrument for the content of the law (La Porta et al. 2008, p. 298), an approach which creates fresh difficulties since it rests on the strong and arguably untenable assumption that regulatory styles, once fixed, are not susceptible to influences from the economic environment (see Deakin 2009). If longitudinal data are available the impact of legal change on a given jurisdiction or system can be analysed using a differencein-differences approach. This means that there has to be a control group whose experience can be compared to that of the system which is the focus of analysis, in the manner of a natural experiment. Put simply, this approach compares the differences in the means of the before-andafter differences in the two systems. Card and Krueger (1995) used this method in their study of the increase in the minimum wage in New Jersey. As a control on the experiment they used Pennsylvania, a neighbouring state in which the minimum wage had not been increased. Their analysis showed that, after taking into account the effects of the national recession which affected both states, the implementation of the higher minimum rate in New Jersey was associated with an increase in full-time employment in the fast food restaurant sector there, by comparison with Pennsylvania. The result was theoretically plausible because of assumed monopsony effects. In other words, before the minimum wage was raised, employers had used their superior bargaining power and access to information to depress wages below the competitive rate; the increase in wages triggered by the New Jersey law had increased the supply of labour into fast food employment without depressing underlying demand. The difference-in-differences approach can be used to study the shortrun impact of legal changes where a control can be found. If there is

44

Regulating for Decent Work

no control, as is the case if a legal change has a widely felt impact across otherwise similar jurisdictions, its use may not be appropriate. Another problem is that spurious correlations may arise by virtue of autocorrelation, particularly when longer periods are studied (Bertrand et al. 2004). More generally, in longer time series, data may be ‘nonstationary’. What this means is that the time series does not follow a regular path, but is prone to irregular deviations without returning to the previous trend. Where one or both of the two time series is nonstationary in this sense, their error terms are liable to autocorrelate, again producing spurious correlations. This (intuitively) remarkable result has long been understood to be a statistical property of certain time series, but its implications for econometric analysis were not taken on board until much more recently. Techniques for addressing the issue of stationarity through the identification of ‘cointegrated’ time series – that is to say, time series which are, individually, non-stationary, but are linked by a common, stationary trend – began to be developed in the 1980s (Engle and Granger 1987) and these were later combined with methods designed to make it possible to draw causal inferences from correlations in time series analysis. These mostly involve variants of so-called ‘Granger causality’ techniques, which, in their basic form, involve regressing current values of the dependent (or effect) variable against past values of itself and of the independent (or causal) variable. If the addition of the past values of the independent variable makes a different to the result, causation is generally assumed. More advanced cointegration-based techniques for time series analysis are thought to be appropriate only where there is a very long time series; conventionally, at least 25 years of an annual time series are required. Single-country time series can be analysed in this way to indicate causation, or data from a number of countries can be pooled for the purposes of identifying a trend across the group as a whole (‘panel data cointegration’). For example, the legal datasets constructed by the CBR project, which cover 36 years of data, have been shown to be non-stationary in the sense just described. Deakin and Sarkar (2008) use a method known as the autoregressive distributed-lag (‘ARDL’) approach to address the problem of non-stationarity and then carry out a trend analysis of correlations between labour law regulation, employment growth and productivity growth, after controlling for GDP growth, in the four developed countries in their sample. They find few of the negative effects of regulation identified by Botero et al. (2004) in their crosssectional study, and some positive effects (see further the discussion of employment protection legislation, below).

Simon Deakin 45

In principle, a time series provides a more secure basis than a crosssection for testing claims about the effects of law. It enables short-run impacts to be distinguished from longer-term trends and it makes it possible to say something about the vector or direction of causation. However, a limitation inherent in this type of analysis is the unavailability of time series data for many of the variables of interest. This is somewhat less of a problem in the labour law field, given that reliable national-level time series exist for data on employment, unemployment, inequality and GDP growth at national level, than it is for the company law area, in respect of which time series data on one of the key variables of interest, corporate ownership structure, are limited (see Armour et al. 2009a).

Key issues in recent empirical work: minimum wages and employment protection We are now in a position to review the current state of the art on the empirical effects of different kinds of labour and employment laws. Minimum wages Until recently, it was a widely held view among the economics profession that minimum wage legislation causes increases in unemployment, above all among younger workers. This view was, however, based largely on analysis of the US situation, which is by no means typical of the kind of minimum wage arrangements which are seen either in most industrialized countries or in developing ones. The United States has a low-level federal minimum wage, dating back to the 1930s, which is supplemented by minima which are not much higher, in relative terms, in some states. The legislation creating the federal minimum did not provide for automatic increases in the basic rate, and ever since the level at which the minimum operates has been a matter of political contention. For long periods in the 1980s, for example, there was no increase in the rate. In part because of the difficulty in getting political agreement on the issue of uprating the minimum wage, the real value of the federal minimum has fallen over time, and only a small proportion of the workforce, by international standards, is employed on or just above the legally mandated rate, just 2 per cent in 2009. In France, by way of comparison, the statutory minimum covers 16 per cent of the workforce. French legislation ensures that the basic rate is increased annually at least in line with prices and with half the increase in the purchasing power of the average wage. This mechanism for automatic

46

Regulating for Decent Work

increases reflects a long-standing cross-party consensus on the importance of minimum wage legislation and has helped, in turn, to maintain the political legitimacy and practical relevance of this type of labour market intervention (Deakin and Wilkinson 2009). There are other unusual features of the US minimum wage. In part because it set at a low rate, it primarily affects casual or youth workers and those employed in very low-paying industries. The economic literature on the minimum wage has focused largely on these cases, which are driven by the US context. Data availability has also shaped the development of the literature. The US Current Population Survey provides a long time series of teenage unemployment data. It was analysis of this source which provided the foundation for a widely cited paper showing that minimum wage increases led to a reduction in employment (Brown et al. 1982). As noted above, the analysis of Card and Krueger (1995) involved a methodological innovation in using difference-in-differences analysis to estimate the impact of minimum wage increases in selected US states. They were able to take advantage of the natural experiment provided by the rises in the minimum rate that took place in certain states while those of neighbouring jurisdictions left their minima untouched. As we have seen, the increase in the New Jersey minimum wage, when compared to the experience of Pennsylvania, was associated not with a fall in employment, but rather with a rise – a phenomenon that could be explained using the logic of employer monopsony. A further study, of California, again found that teenage wages and employment rose together following a legally mandated rise in the minimum wage. This study was particularly compelling because it found a high level of employer compliance with the legal rate. Survey evidence showed that employers were not, on the whole, responding to the wage rise by laying workers off or cutting non-wage benefits, again suggesting the presence of monopsony power. There was a hugely critical response to the Card–Krueger studies. They were criticized for using incomplete and unreliable survey data, not using appropriate control groups, and assessing the effects of wage increases over insufficiently long periods (Kennan 1995; Neumark and Wascher, 1995). Subsequent panel data studies, over longer periods, have found disemployment effects. A recent overview concludes that in around two-thirds of the post-Card–Krueger literature, there is a relatively consistent (although not always statistically significant) indication of negative employment effects of minimum wages’ (Neumark and Wascher, 2006, p. 121). This has led even some prominent

Simon Deakin 47

supporters of the minimum wage within the economics profession to question whether it can be justified on economic efficiency (as opposed to equity) grounds (Kaufman 2009). However, evidence from other countries suggests that the matter is very far from closed. The British case offers an instructive example. Like the United States, for most of the twentieth century Britain had either no general minimum wage or one set at a very low rate. Between 1909 and 1986 legally binding minimum rates were set for very low-paying sectors, while under wartime conditions and after 1945 the minimum rates set out in industry-level collective agreements more generally had de facto and, to some degree, de jure binding effect. In the early 1980s the system of industry-wide collective bargaining disintegrated and from 1986 changes were made to the statutory systems which first of all saw its scope considerably confined before almost complete abolition in 1993 (only agriculture survived as a sector with legally mandated minimum rates). Around this time a prominent economic critic of labour law ‘confidently postulated’ that abolition ‘would serve to expand employment [and] offer competitive wages for the socially disadvantaged’ (Minford 1985, p. 122). This view was contradicted by the econometric studies of the reforms carried out in the early 1990s. These found that the reduction in the scope of the wage-setting powers of the statutory wages councils from 1986 onwards had been associated with a fall in employment, supporting the hypothesis of employer monopsony (Dickens et al. 1993; Machin and Manning 1994). This research helped pave the way for the reintroduction of the minimum wage from 1998 onwards in a form which gave a prominent role to economic analysis in determining the minimum rate. The National Minimum Wage (NMW) which came into force in the late 1990s set, for the first time in UK history, a basic rate which was to be applied to all categories of workers and all industries. The legislation provided that the rate should be set by the government on the advice of the Low Pay Commission (LPC), a body consisting of employer and union representatives together with a neutral third group, some of whom were appointed on the basis of their economic expertise. The LPC was also given the task of commissioning economic research on the impact of the statutory rate. Its first recommendation, for the introduction of a basic hourly rate equivalent at that time to 45 per cent of median earnings, was accepted by the government and came into force in April 1999. Around 11 per cent of the labour force was covered, with the cleaning, catering and security service sectors being the most

48

Regulating for Decent Work

affected, and women workers benefiting proportionately more than men. The new minimum was not high by international standards and there was no provision in the legislation for it to be increased in line with prices or inflation, but the LPC went on to make a series of annual recommendations which led to above-inflation pay rises, thereby closing the gap between the minimum rate and wages set by employers or in the remaining pockets of collective bargaining. Recently, one of the LPC commissioners has reviewed the substantial body of empirical evidence on the workings of the NMW (Metcalf 2007). The legal minimum had had a discernible effect in increasing the earnings of the lowest paid, particularly after 2002. Between 1993 and 1999, a period in which there were no minima in place except in the agricultural sector, the earnings of the lowest quartile had fallen considerably by reference to those of other groups. This trend was reversed after 1999. Overall wage inequality began to fall for the first time in twenty years. The NMW also contributed to a narrowing of the gap between male and female earnings. Despite this clear impact on earnings, the consensus of the empirical research was that the NMW had not had a negative effect on employment. The analyses commissioned by the LPC looked specifically at the sectors where the minimum wage could have been expected to have most impact, that is, the low-paying sectors of retail, hospitality, social care, cleaning, agriculture, security services, clothing and footwear, hairdressing, and textiles. With the exception of textiles, a sector that was in long-term decline, there was no evidence of disemployment effects. Nor was there evidence of falling employment among those groups of workers most affected by the NMW (women, the young, and ethnic minorities). Metcalf’s study attributes the lack of a negative impact to employer monopsony prior to the introduction of the law: survey evidence reported that firms set wages within a range that was only loosely determined by supply and demand. The profitability of firms in those sectors most affected by the statutory minimum declined, again suggesting monopsony and indicating that the legislation had a significant redistributive impact. Metcalf (2007, p. 53) concluded that ‘the LPC, via its evidence-based approach… has raised the real and relative wage of low-paid workers without adverse employment consequences’. The experience of the British case is significant given the wide range of workers and sectors affected by the reintroduction of the statutory minimum. In this respect it has more in common with other systems in Europe and in the developing world than the US minimum wage, with its limited coverage. Most continental European systems rely on

Simon Deakin 49

a combination of collective agreements and a statutory floor of rights to govern wage determination; here, statutory regulation and collective bargaining tend to act as complements. In many developing systems, statutory minimum wages have a high coverage rate because of the limited role played by collective bargaining. A growing body of research is looking at the role played by industrial and political conditions in shaping the nature of minimum wage laws. The forms of statutory minimum wages, as with other forms of labour regulation, can be seen to be endogenous to national contexts (Koçer and Visser 2009). This suggests that the effects of minimum wage legislation need to be modelled in a way which is more sensitive to historical and institutional forces shaping the path of economic development in particular countries. Employment protection legislation A prolonged debate on the merits of employment protection legislation (EPL) has been driven by significant cross-national differences in the degree of regulation provided by this type of labour law. The United States largely adheres to the common law rule of employment at will, under which the employer may dismiss the employee without good cause and with little or no notice. There has been some judicial erosion of this rule since the 1970s, with courts in some states carving out limited exceptions for ‘bad faith’ terminations and dismissals in breach of implied commitments made by employers. However, the basic rule remains intact in the overwhelming majority of state jurisdictions. In Western Europe, by contrast, unfair dismissal legislation imposes substantive constraints on the employer’s power to dismiss, generally enforceable by litigation before specialized labour courts. These laws are flanked by complementary measures on income protection in the event of temporary layoffs and by collective information and consultation laws which, to varying degrees, give employee representatives a role in policing and enforcing job security rights. Unfair dismissal laws have a long history in the European context: antecedents for today’s laws can be found in civil law notions of ‘abuse of right’ before the advent of legislation. The unjust dismissal legislation of the Weimar Republic was the first statutory intervention, in the 1920s. Britain, although a comparative latecomer, acquired extensive unfair dismissal laws in the 1970s, at the same time as statutory schemes in several other countries, including France, were significantly strengthened. The only US legislation to incorporate elements of the European approach is the federal ‘WARN’ law, dating from the late 1980s, under which larger employers

50

Regulating for Decent Work

have to give notice of dismissal and make severance payments in the event of downsizing. In the 1960s and 1970s US unemployment was generally higher than in Western Europe, but from the 1980s onwards there was a reversal of this relationship: US unemployment began to fall, and employment grew rapidly by historical standards. In Europe, by contrast, employment growth was sluggish in the aftermath of the oil shocks of the 1970s, and the employment rate fell significantly behind that of the United States. While this process was going on, there were relatively few changes in US employment law that could have been said to be responsible for its faster rate of job creation. In Europe, on the other hand, the 1970s were a period of significant legislative activity in the area of EPL (see Deakin et al. 2007). It was not long before the connection was made between the advance of job security legislation and Europe’s low employment rate, a point reinforced by the influential OECD Jobs Study in the early 1990s (OECD 1994). This view came to have a substantial influence on the content of employment protection laws, as most systems sought to liberalize their EPL regimes by introducing wider exemptions for fixed-term employment, liberalizing the rules on the use of temporary agency work, and encouraging – or at least tolerating – the rise of part-time work below the thresholds at which employment protection rights were triggered. On closer inspection the widely held view that strict EPL has a dampening effect on employment becomes much more nuanced. From a theoretical perspective, EPL can have a number of potentially offsetting effects. In so far as it adds to firms’ hiring costs and slows down the movement of labour from less productive firms and sectors to more productive ones, it will add to unemployment and reduce productivity (Saint-Paul 1997). On the other hand, strict EPL will deter firms from making workers redundant in a downturn, thereby reducing job losses as well as smoothing the impact of the business cycle on demand (Koeniger 2005). Stricter EPL may also increase productivity by providing incentives for investment in training and supporting employeebased innovation (Acharya et al. 2010a, 2010b). Given these various effects, it is perhaps unsurprising that empirical studies have failed to detect a clear role for EPL in structuring European unemployment (Baker et al. 2005; Bertola 2009). Multivariate analyses find, on the whole, that EPL has less influence on employment and unemployment rates than other institutional factors such as the level of unionization (Nickell et al. 2005) and monetary policy (Baccaro and Rei 2007).

Simon Deakin 51

The main data source used to evaluate the effects of employment protection laws is, as we have seen, the OECD’s employment protection index (EPI). Using the 2004 version of the EPI, the OECD’s own analysis found that there were only weak links, on a cross-sectional analysis, between EPL strictness and flows between employment and unemployment, and no correlation between EPL and unemployment levels. Time series analysis suggested that, over time, there was some evidence of a link between EPL strictness and employment flows. However, the main effect of the changes in the law which had taken place in the 1990s and 2000s to allow greater flexibility in the use of temporary and fixed-term work had been to substitute these forms of employment for more regular, permanent work (OECD, 2004, p. 79). A number of recent studies build on the idea of labour law’s endogeneity to national and sectoral contexts in an attempt to arrive at a more finely grained view of the effects of EPL. Dismissal laws emerged out of the practice of permanent employment in firms with bureaucratically organized internal labour markets. The impact of legislation based on this model should be greater in sectors characterized by more irregular and casual employment. There is evidence that EPL has negative effects on productivity in these sectors (Bassanini and Venn 2007; Bassanini et al. 2008). The same studies show that, by contrast, minimum wage legislation has a positive effect on productivity in sectors with low pay, and that parental leave laws, likewise, have a beneficial impact on sectors with a mainly female labour force. This suggests that labour laws can, in principle, operate as a ‘beneficial constraint’ on firms, but that the effect may vary according to the type of law that is adopted, as well as the context in which it is applied. A related theme is the possibility that employment protection laws operate in a way that is complementary to other institutional factors including product market regulation and corporate governance. Thus there is evidence that, in OECD countries, product market deregulation only induces higher economic growth where a high level of EPL is preserved, suggesting that the two are substitutes (Amable et al. 2007). Conversely, EPL appears to be complementary to stakeholder-based corporate governance regimes with concentrated share ownership, and to go against the grain in systems characterized by liquid capital markets and dispersed ownership (Gatti, 2009). The CBR’s longitudinal labour regulation index provides the opportunity to test some of these claims using very long time series which date back to the 1970s, and offers detailed evidence on the effects and interactions of a wide range of legal provisions. Using time series econometrics

52

Regulating for Decent Work

to analyse the CBR dataset, Deakin and Sarkar (2008) find that stronger EPL is associated with productivity increases in Germany after controlling for the growth of GDP. A similar effect can be observed for working time legislation in France. In the case of the United States, the introduction of the WARN laws in the late 1980s is associated with a rise in productivity but a slowdown in the rate of employment growth, and so can be understood in the light of the emergence of ‘jobless growth’ in the US economy at that time. They find no evidence of a positive impact on employment, unemployment or productivity of the deregulatory measures introduced in the United Kingdom in the 1980s and 1990s. Acharya et al. (2010b) also make use of the CBR index. They hypothesize that strong employment protection legislation, as a result of the protection it gives to employees against arbitrary dismissal, should favour employee-orientated forms of innovation. Using a differencein-differences analysis, they find evidence that a strengthening of EPL at country level is associated with increased innovative activity, in the form of higher numbers of patents. They have carried out a similar study for the United States, focusing on the differences between states in the treatment of the employment-at-will rule, with a similar result: judicial narrowing of the employers’ power to dismiss is positively correlated with patent rates (Acharya et al 2010a).

Conclusions: the role of evidence, theory and policy in the evolution of labour law This chapter has provided an overview of recent developments in the empirical study of labour law. It has shown how a growing interest in comparative research has led to the construction of cross-national measures of labour regulation, and how these have been coupled with advances in time series and panel data econometrics to offer new insights into the economic impact of labour legislation in areas including minimum wages and employment protection. Through analysis of this kind, it has been possible to obtain a more complete and arguably more accurate picture of the operation of labour laws. The conventional picture of labour regulation is one that sees it largely as a cost of ‘doing business’, with the potential to increase unemployment and hamper growth. The emerging empirical evidence suggests that this is by no means always, or even predominantly, the case. Minimum wages, if set at an appropriate level for the contexts in which they are applied, can mitigate the effects of low pay without creating unemployment. They may also have productivity-enhancing effects, by forcing firms to use

Simon Deakin 53

labour more efficiently. Similarly, employment protection legislation can have positive effects on productivity, through the encouragement it offers to training and innovation. These effects depend to some degree upon the presence of complementary institutional mechanisms in capital and product markets. Thus the new literature does not predict that labour regulation always and everywhere has beneficial economic impacts. It suggests, instead, that assessments of the economic effects of labour laws must be sensitive to context. Predictions drawn from generalized assumptions of the way labour markets work may turn out to be of limited value. What is needed is, above all, more and better data on the content of labour laws, they way they operate in practice, and the political and economic environments in which they are applied. Despite the progress that has been made recently in attaining a better empirical understanding of the way labour laws interact with the economy, there are difficulties facing the use of evidence to shape policy. One is inherent in the nature of empirical research. Data simply do not exist on some critical variables of interest, and those which are available are incomplete and, to a certain degree, contestable. In addition, the statistical techniques which are used to analyse legal datasets involve limitations and trade-offs which need to be openly acknowledged. The evidence presented to policy makers is, or should be, qualified in the light of these constraints. However, the indeterminacy of social science research is probably not the main reason for the limited use that policy makers make of it. More constraining is the role that theory, relatively uninformed by empirical work, plays in shaping policy perceptions. The conventional neoclassical model of the competitive labour market has a tenacious hold over many parts of the social sciences, even beyond its economics core, and its simplicity and apparent clarity make it appealing to policy makers. For its adherents, the neoclassical model does not need to be realistic. Its main function is to generate propositions which can be tested empirically. Refutation of the theory will not occur through new empirical findings alone. However, empirical work may play a role in shifting some of the theoretical underpinnings of the model. This is beginning to happen with the growing use of transaction cost economics and behavioural approaches to theorize labour market institutions, but the process is slow. Because empirical research is complex, time consuming and resource intensive, the political context plays a role in determining what areas are the subject of research and how those results are used in policy making. The political commitment to the introduction of a national minimum

54

Regulating for Decent Work

wage in Britain in the late 1990s provided the framework within which the Low Pay Commission supported empirical research into the effects of the legislation it was helping to implement. That research influenced the level of the minimum wage which the Commission recommended to government, but the prior decision to introduce a national minimum wage in the first place was determined in the political arena. Similarly, the World Bank’s commitment to the Washington consensus provided the backdrop against which the methodology of the Doing Business reports was developed and was responsible for the uses made of their findings by national governments. The fading of the Washington consensus now assists the process of challenging the methods that underlie the Doing Business approach (Lee et al. 2008). If this policy shift continues, there will be scope in future for further advances in empirical work which should in due course feed into theory.

References Acharya, V.; Baghai-Wadji, R.; Subramanian, K. 2010a. ‘Wrongful discharge laws and innovation’, Working Paper, New York University Stern School of Business. Acharya, V.; Baghai-Wadji, R.; Subramanian, K. 2010b. ‘Labor laws and innovation’, NBER Working Paper 16484 (Cambridge, MA: NBER). Ahlering, B.; Deakin, S. 2007. ‘Labour regulation, corporate governance and legal origin: a case of institutional complementarity?’, Law and Society Review, vol. 41, pp. 865–98. Amable, B.; Demmou, L.; Gatti, D. 2007. ‘Employment performance and institutions: new answers to an old question’, IZA Discussion Paper No. 2731 (Bonn: IZA). Armour, J.; Deakin, S.; Sarkar, P.; Siems, M.; Singh, A. 2009a. ‘Shareholder protection and stock market development: an empirical test of the legal origins hypothesis’, Journal of Empirical Legal Studies, vol. 6, pp. 343–80. Armour, J.; Deakin, S.; Lele, P.; Siems, M. 2009b. ‘How legal rules evolve: evidence from a cross-country comparison of shareholder, creditor and worker protection’, American Journal of Comparative Law, vol. 57, pp. 579–630. Baccaro, L.; Rei, D. 2007. ‘Institutional determinants of unemployment in OECD countries: does the deregulatory view hold water?’, International Organization, vol. 61, pp. 527–69. Baker, D.; Glyn, A.; Howell, D.; Schmitt, J. 2005. ‘Labour market institutions and unemployment: a critical assessment of cross-country evidence’, in Howell, D. (ed.) Fighting Unemployment: The Limits of Free Market Orthodoxy (Oxford: Oxford University Press). Bassanini, A.; Venn, D. 2007. ‘Assessing the impact of labour market policies on productivity: a difference-in-differences approach’, OECD Social, Employment and Migration Paper No. 54 (Paris: OECD).

Simon Deakin 55 Bassanini, A.; Nunziata, L.; Venn, D. 2008. ‘Job protection and productivity growth in OECD countries’, IZA Discussion Paper No. 3555, June (Paris: OECD). Becker, G. 1957. The Economics of Discrimination (Chicago: University of Chicago Press). Bertola, G. 2009. ‘Labour market regulation: motives, measures, effects’, ILO Conditions of Employment and Work Research Series No. 21 (Geneva: ILO). Bertrand, M.; Duflo, E.; Mullainathan, S. 2004. ‘How much should we trust difference-in-differences estimates?’, Quarterly Journal of Economics, vol. 119, pp. 249–75. Botero J.; Djankov, S.; La Porta, R.; Lopez-de-Silanes, F.; Shleifer, A. 2004. ‘The regulation of labor’, Quarterly Journal of Economics, vol. 119, pp. 1340–82. Brown, C.; Gilroy, C.; Kohen, A. 1982. ‘The effect of the minimum wage on employment and unemployment’, Journal of Economic Literature, vol. 20, pp. 487–528. Card, D.; Krueger, B. 1995. Myth and Measurement: The New Economics of the Minimum Wage (Princeton, NJ: Princeton University Press). Deakin, S. 2009. ‘Legal origin, juridical form and industrialisation in historical perspective: the case of the employment contract and the joint-stock company’, Socio-Economic Review, vol. 7, pp. 35–65. Deakin, S.; Lele, P.; Siems, M. 2007. ‘The evolution of labour law: calibrating and comparing regulatory regimes’, International Labour Review, vol. 146, pp. 133–62. Deakin, S.; Sarkar, P. 2008. ‘Assessing the long-run economic impact of labour law systems: a theoretical reappraisal and analysis of new time series data’, Industrial Relations Journal, vol. 39, pp. 453–87. Deakin, S.; Wilkinson, F. 2009. ‘Minimum wage legislation’, in Dau-Schmidt, K.; Harris, S.; Lobel, O. (eds) Encyclopedia of Law and Economics: Labor and Employment Law and Economics (Cheltenham: Edward Elgar). Dickens, R.; Gregg, P.; Machin, S.; Manning, A.; Wadsworth, J. 1993. ‘Wages councils – was there a case for abolition?’, British Journal of Industrial Relations, vol. 31, pp. 515–29. Engle, R.; Granger, C. 1987. ‘Cointegration and error correction: representation, estimation, and testing’, Econometrica, vol. 55, pp. 251–76. Fehr, E.; Falk, A. 1999. ‘Wage rigidity in a competitive incomplete contract market’, Journal of Political Economy, vol. 107, pp. 106–34. Gatti, D. 2009. ‘Macroeconomic effects of ownership structure in OECD countries’, Industrial and Corporate Change, vol. 18, pp. 901–28. Grubb, D.; Wells, W. 1993. ‘Employment regulation and patterns of work in EC countries’, OECD Economic Studies, vol. 21, pp. 7–58. Hall, P.; Soskice, D. 2001. ‘Introduction to varieties of capitalism’, in Hall, P.; Soskice, D. (eds) Varieties of Capitalism: The Institutional Foundations of Comparative Advantage (Oxford: Oxford University Press). Hyde, A. 2006. ‘What is labour law?’, in Davidov, G.; Langille, B. (eds) Boundaries and Frontiers of Labour Law (Oxford: Hart). Iversen, T.; Soskice, D. 2007. ‘Distribution and redistribution: the shadow of the nineteenth century’, presented to the conference on Changing Institutions (in Developed Democracies): Economics, Politics and Welfare, Paris, May 2006.

56

Regulating for Decent Work

Kaufman, B.E. 2010. ‘Institutional economics and the minimum wage: broadening the theoretical and policy debate’, Industrial and Labor Relations Review, vol. 63, pp. 427–53. Kennan, J. 1995. ‘The elusive effects of minimum wages’, Journal of Economic Literature, vol. 33, pp. 1950–65. Koçer, R.; Visser, J. 2009. ‘The role of the state in balancing the minimum wage in Turkey and the USA’, British Journal of Industrial Relations, vol. 47, pp. 349–70. Koeniger, W. 2005. ‘Dismissal costs and innovation’, Economics Letters, vol. 88, pp. 79–85. La Porta, R.; Lopez-de-Silanes, F.; Shleifer, A. 2008. ‘The economic consequences of legal origins’, Journal of Economic Literature, vol. 46, pp. 285–332. Lee, S.; McCann, D. 2007. ‘Measuring labour market institutions: conceptual and methodological questions on “working time rigidity”’, in Lee, S.; McCann, D.; Messenger, J. (eds) Working Time Around the World: Trends in Working Hours, Laws and Policies in a Global Perspective (London: Routledge). Lee, S.; McCann, D.; Torm, N. 2008. ‘The World Bank’s “employing workers” index: findings and critiques: a review of recent evidence’, International Labour Review, vol. 148, pp. 416–32. Machin, S.; Manning, A. 1994. ‘Minimum wages, wage dispersion, and employment: evidence from UK wages councils’, Industrial and Labor Relations Review, vol. 47, pp. 319–29. Manning, A. 2003. Monopsony in Motion: Imperfect Competition in Labor Markets (Princeton, NJ: Princeton University Press). Metcalf, D. 2007. ‘Why has the British national minimum wage had little or no impact on employment?’, CEP Discussion Paper No. 781, LSE. Minford, P. 1985. Unemployment: Cause and Cure, 2nd edn (Oxford: Basil Blackwell). Neumark, D.; Wascher, W. 1995. ‘The effect of New Jersey’s minimum wage on fast food employment: a re-evaluation using payroll data’, NBER Working Paper No. 5224 (Cambridge, MA: NBER). Neumark, D.; Wascher, W. 2006. ‘Minimum wages and employment: a review of evidence from the new minimum wage research’, NBER Working Paper No. 12663 (Cambridge, MA: NBER). Nickell, S.; Nunziata, L.; Ochel, W. 2005. ‘Unemployment in the OECD since the 1960s. What do we know?’, Economic Journal, vol. 115, pp. 1–27. Organization for Economic Co-operation and Development (OECD). 1994. OECD Jobs Study, Evidence and Explanations, Part I: Labour Market Trends and Underlying Forces of Change (Paris: OECD). OECD. 2004. Employment Outlook (Paris: OECD). Posner, R. 1984. ‘Some economics of labor law’, University of Chicago Law Review, vol. 51, pp. 988–1004. Pozen, D. 2006. ‘The regulation of labor and the relevance of legal origin’, Comparative Labor Law and Policy Journal, vol. 28, pp. 43–56. Saint-Paul, G. 1997. ‘Is labour rigidity harming Europe’s competitiveness? The effect of job protection on the pattern of trade and welfare’, European Economic Review, vol. 41, pp. 499–506. Samuelson, P. 1951. ‘Economic theory and wages’, in Wright, D.M. (ed.) The Impact of the Union (New York: Harcourt Brace).

Simon Deakin 57 Stigler, G. 1946. ‘The economics of minimum wage legislation’, American Economic Review, vol. 36, pp. 358–65. Streeck, W. 1997. ‘Beneficial constraints: on the economic limits of rational voluntarism’, in Hollingsworth, J.R.; Boyer, R. (eds) Contemporary Capitalism: The Embeddedness of Institutions (Cambridge: Cambridge University Press). Venn, D. 2009. ‘Legislation, collective bargaining and enforcement: updating the OECD employment protection indicators’, OECD Social, Employment, and Migration Working Papers No. 89 (Paris: OECD). Williamson, O.; Wachter, M.; Harris, J. 1975. ‘Understanding the employment relation: the economics of idiosyncratic exchange’, Bell Journal of Economics and Management Science, vol. 6, pp. 250–78. World Bank (various years) Doing Business Reports (Washington, DC: International Bank for Reconstruction and Development).

3 Precarious Employment and the Problem of SER-Centrism in Regulating for Decent Work* Leah F. Vosko

There is a growing disjuncture between changing patterns of labour force participation in highly industrialized contexts and systems of labour regulation. Policy actors at various levels are attempting to deal with the insecurities associated with ‘new’ forms of employment such as parttime and temporary paid employment and solo self-employment, which have disproportionate effects on women, migrants, older, and younger workers. At the same time, most regulatory responses to precariousness continue to rest on assumptions from an earlier era. Even where they have been modified, most labour laws and policies still centre on the standard employment relationship (SER) and its outdated participation norms, particularly with regard to gender relations, citizenship boundaries and age-related distinctions. Emerging in the post-Second World War era, the SER, defined by a full-time continuous employment relationship between a worker and an employer complete with a social wage, was devised for adult male citizens following a lifecourse divided into discreet segments of education, work and retirement and premised on norms of female caregiving. Since then, there has been a shift in the balance between permanent and temporary international migration for employment, women’s labour force participation rates have risen dramatically, and school-to-work as well as work-to-retirement transitions have become more complex and varied. However, systems * The author gratefully acknowledges the Canada Research Chairs Program and the Social Sciences and Humanities Research Council of Canada for funding the research required to prepare this chapter through its standard research grants programme. I also extend my sincere thanks to John Grundy and Melissa SharpeHarrigan for their able research assistance and to Gerald Kernerman, Sangheon Lee, and Deirdre McCann and the two anonymous reviewers for their critical reflections on early versions of the chapter. 58

Leah F. Vosko 59

of labour regulation have not kept pace with patterns of participation. There is an urgent need to bridge this gap by addressing changing employment relations alongside new participation norms. This chapter charts the problem of SER-centrism and explores possibilities for recasting systems of labour regulation. After a review of several concepts essential to understanding SER-centrism, it begins by describing the limits of adjustments to labour laws and policies centred on the SER by way of a few examples drawn from International Labour Organization (ILO) standards, European Union (EU) directives, and national laws and policies. This aspect of the analysis shows how, by attempting to bring those forms of employment falling just outside the SER within its range, SER-centric approaches to regulation are least likely to improve employment situations deviating sharply from this employment model; the greater the deviation from the SER and its associated participation norms, the lower the level of protection they provide. The focus then shifts to assessing two alternative approaches to SER-centrism – the ‘flexible SER’ proposed by Bosch (2004) and ‘beyond employment’ developed by Supiot (2001). Both regulatory approaches offer promise in terms of de-centring the SER, and addressing changing gender relations in the process. The ‘beyond employment’ approach also offers potential in addressing age-related transitions. Still, more integrated strategies of re-regulation are required. The investigation thus culminates by sketching some principles and policy formulations responding to the disjuncture and associated precariousness by addressing the paid work/membership/care nexus. Through this threefold organization, the chapter contends that regulating for decent work requires both rejecting a singular employment model and replacing axes of differentiation and exclusion from protection along the lines of gender, citizenship, and age with entrées into what I term ‘global labour market membership’ for workers belonging to social groups hitherto marginalized on the basis of the SER’s participation norms. At this point, two further preliminaries are in order: first, the ensuing discussion focuses on SER-centrism as it takes expression in highly industrialized countries since its logic and effects are sharpest therein. However, I take the dynamics of SER-centrism to be inherently global – that is, to be premised on social and spatial divisions of an economically polarized world. In other words, I recognize ‘the shared presence of the North with other places’ (Raghuram et al. 2009, p. 11); at the same time, it remains important to differentiate between places because the SER has ‘far more purchase in some places than others’ (Massey 2004, p. 12). Second, the nature of this intervention is schematic; it represents

60 Regulating for Decent Work

a preliminary attempt to pursue the alternative imaginary I call for in concluding my recent book Managing the Margins: Gender, Citizenship, and the International Regulation of Precarious Employment (2010). In this book, I argue that cultivating this imaginary necessitates addressing normative issues surrounding labour smarket membership because changing gender relations and citizenship boundaries simultaneously challenge divisions between public and private responsibilities, the ‘national order of things’ (Malkki 1995), and the notion of a life-course comprised of unidirectional movements between schooling, work and retirement. They are constitutive of, rather than incidental to, employment norms and therefore must figure centrally in responses to the disjuncture and associated precariousness.1 To this end, the policy formulations and innovations described and assessed in the third part of the chapter are selected on the basis that they emanate from institutions distinct (or departing) from those associated with SER-centric laws and policies.

Key concepts: the SER, the gender contract, and citizenship boundaries In this chapter, I employ three concepts central to understanding SER-centrism: the familiar notions of the SER, the gender contract, and citizenship boundaries. I take the SER to be a normative model of employment reflecting the interplay between social customs and conventions and governance mechanisms that link work organization and the labour supply (Deakin 2002, p. 179). Reaching its apogee in the mid-twentieth century, the SER is a full-time continuous employment relationship in which the worker has one employer, normally works on the employer’s premises under direct supervision, and is in receipt of a social wage (Mückenberger 1989). Its central pillars are standardized working time, permanency, and employee status. In the tradition of scholarship advancing a ‘gendered employment systems’ approach, I see the SER as intertwined historically with a specific gender contract or set of social, legal, and political practices surrounding exchanges between breadwinning and caregiving (O’Reilly and Spee 1998, p. 263; Rubery 1998, p. 23; see also Fraser 1997, and Fudge and Vosko 2001). Feminist scholarship illustrates that at its height, the SER was sustained by a male breadwinner/female caregiver contract. A review of early national and international labour regulations illustrates further that this gender contract helped make the SER possible.

Leah F. Vosko 61

That many early national and international labour regulations were sex-, as well as age-, specific is a familiar story. In Western Europe, North America, and other industrializing contexts many regulations were designed to limit women’s, as well as children’s, working hours and night work, fix minimum wages in female-dominated industries, and establish maternity protections. Prohibitions against women’s night work in industry and the use of white phosphorus in match production were also the first subjects of international labour regulation, beginning in 1906. Such subjects, as well as maternity protection and lead poisoning, were addressed further at the inaugural conference of the ILO in 1919. Although not without contestation nationally and internationally, nor variation nationally, such regulations reinforced women’s responsibility for unpaid work and relegated them to the margins of paid work – to casual work, homework and/or to small workshops (see, for example, on early developments at the national level, contributions to Wikander et al. (eds) 1995; and at the international level, Murray 2001). The male breadwinner/female caregiver contract cultivated by such regulations also helped to prepare the ground for ‘gender-neutral’ labour regulations normalizing pillars of the SER among adult male citizens. For example, early regulations fostering ‘standardized working time’ rested on the assumption that unpaid (presumably female) caregivers would provide for male workers’ reproduction outside the labour force. Such was the case with the ILO Convention on Hours of Work (Industry) (1919), which introduced the eight-hour day and the 48-hour week but covered only wage workers and exclusively those in industry, a designation that effectively encompassed middle-aged workers but which many wage-earning women lacked. In these ways, the SER is gendered at its root as well as targeted specifically at workers in the ‘prime’ wageearning phase of the lifecycle. The emergence of the SER also rested on a particular conception of membership in a community and the rights and obligations attached to this membership (see, for example, Lister 1997, p. 14; Stasiulis and Bakan 2005, pp. 1–2). Assuming a correspondence between membership and territory, the spatial container for the SER was (and remains) the nation-state, asserting its sovereignty by defining populations that can make claims on the state in contrast to those, both within and outside a given national territory, who cannot (Hall and Held 1990, p. 176; Brodie 2002, p. 44; Stasiulis and Bakan 2005, p. 16). The rights and obligations associated with this conception of citizenship emanated from a common nationality, defined typically on the basis of birth, lineage, and/or residence (Sainsbury 2006, p. 231).

62 Regulating for Decent Work

The nation-state-centred notion of citizenship associated with the SER is premised on the fusion of continuity of employment and territorial belonging. At the height of the SER (and prevailing to the present), in highly industrialized countries this meant nationally based employment and immigration laws and policies providing for open-ended employment relationships for workers holding national citizenship, while limiting many non-national citizens to temporary engagements. International labour regulations, such as the ILO and United Nations (UN) instruments, adopted formally beginning in mid-century, but whose roots date to the origins of these organizations, reinforced this conception.

The problem of SER-centrism SER-centrism operates by conflating non-standard and precarious employment, or forms of work for remuneration characterized by ‘dimensions of labour market insecurity’ (Rodgers 1989), such as uncertainty, low income, and limited social benefits and statutory entitlements. Contemporary laws and policies at various scales ‘see’ precarious employment in ‘nonstandard’ – that is, in deviation from the SER, which leads them to seek solutions that minimize deviations. They focus on bringing those forms of employment falling just outside the SER within its range, addressing divergence principally along its central pillars of standardized working time, continuity, and employment status. The result: those in situations deviating sharply from it are least likely to be improved. The greater the deviation from the SER, the less the protection laws and policies tend to provide.2 The gendered, citizenship-coded, and age-mediated margins of the labour market remain intact. At the international level, this logic is evident in several ILO standards and EU directives, originating in the 1970s and 1980s. ‘Precarious employment’ entered the ILO lexicon when an inquiry raised concern about the ‘particularly disadvantaged or precarious … situation’ of categories of workers falling outside the scope of traditional protection measures and the acute problems faced by women and migrant workers (1984a, p. 12, italics added). This inquiry’s findings prompted calls for ILO standards on ‘certain types of economic activity in which normal measures for social protection are particularly difficult to apply … temporary or casual work, seasonal work, subcontracted work and home work’ (1984b, p. 103, italics added). Around the same time, there were parallel efforts to address ‘atypical work’ at the EU level – and they drew attention to the situation of women.

Leah F. Vosko 63

In response, starting in the 1990s ILO standards and EU directives were adopted to address precarious employment – and they focused on nonstandard forms of employment. These regulations were motivated by the idea that, within the context of a nation-state, citizen-workers should not see their employment and occupational opportunities or working conditions limited by barriers erected on the basis of form of employment. There was recognition of the need to allow workers in part-time and temporary paid employment, and the nominally self-employed, the ability to access benefits and entitlements associated with the SER, even if such ‘normal measures’ for labour and social protection are difficult to apply. Three examples are the ILO Part-Time Work Convention, 1994 (No. 175), the EU Directive on Fixed-Term Work (1999/70/EC), and the ILO Employment Relationship Recommendation, 2006 (No. 198).3 The above understanding is reflected most closely in the ILO Convention on Part-Time Work and the EU Directive on Fixed-Term Work. Their approach is to provide for equal treatment on the basis of form of employment among citizens. These regulations pursue this aim in two common ways: first, through the familiar notion of nondiscrimination, interpreted as either equivalent treatment to or treatment no less favourable than a similarly situated worker.4 In both the ILO Convention on Part-Time Work and the EU Directive on Fixed-Term Work, non-discrimination means providing the same level of protection in some areas and proportional protection in others – that is, protection defined in relation to hours in the case of the former and job tenure in the case of the latter. The second way is through the mechanism of the comparable worker whose employment relationship approximates the SER. Both the ILO Convention on Part-Time Work and the EU Directive on Fixed-Term Work seek to encompass forms of paid employment lacking the benefits and entitlements associated with the SER. However, the commitment to equal treatment on the basis of form of employment does not provide for minimum standards. Because of the requirement for a comparator, this approach is also capable only of treating limited labour force insecurities and of addressing the situation of restricted categories of workers. For the limited categories of workers to whom they apply, entitlements are also prorated by proximity to the full-time continuous employment relationship or the SER. The upshot is that the framework provided by the ILO Convention on Part-Time Work only applies fully to permanent part-time wageearners. Similarly, the EU Directive on Fixed-Term Work excludes temporary agency workers.5 This approach also neglects to account for

64 Regulating for Decent Work

the fact that so many women engage in part-time employment because of caregiving responsibilities. Nor does it address the fact that so many migrant workers engage in temporary employment because receiving countries, and the employers within them, seek to externalize the costs of their daily and intergenerational reproduction. These deficiencies are amplified in the 2008 EU Directive on Temporary Agency Work, significantly since, in the EU 15 as elsewhere, higher percentages of women than men tend to stay in such contracts for long periods of time and since young workers and non-national citizens are also overrepresented in this type of temporary employment.6 Under the terms of this directive, workers recruited directly by firms, including fixed-term workers, qualify as ‘standard workers’ vis-à-vis temporary agency workers. Equal treatment also applies only to ‘basic working and employment conditions’ (that is, working time and pay) as opposed to the wider ‘employment conditions’ covered under the Directive on Fixed-term Work. The Recommendation on the Employment Relationship, 2006 (No. 198), goes further along this path. It seeks to combat disguised employment relationships, whose ‘gender dimension’ is acknowledged, and to ensure that standards applicable to all forms of contractual arrangements protect employed workers, including migrant workers. However, the recommendation’s defining challenge is to protect employed workers without interfering with commercial relationships (Paragraph 8). It thus dispenses altogether with the use of a comparator. This entails maintaining the employment relationship as the dividing line between the sphere of commerce and that of the labour market such that only employed workers are to receive labour protection. The corollary is the neglect of workers engaged in work for remuneration falling outside the strictures of the employment relationship, many of whom require protection. Such SER-centric approaches to regulation are also evident at the national level in highly industrialized countries, where, for example, the cases of part-time employment in Australia and self-employment in Canada highlight their tiering effects in practice. In Australia, there is a great deal of part-time employment and it is deeply gendered. Furthermore, among all part-time employees, as of the early 2000s, almost two-thirds were casual. The roots of this situation date back to the early twentieth century, when courts and tribunals recognized the importance of developing an SER for male breadwinners through the introduction of laws and policies fostering weekly hire. Yet, as weekly hire expanded among this group, casual employment became a ‘catch-all’ category for all remaining hiring systems and the means

Leah F. Vosko 65

through which many women accessed paid work (Campbell 1998; O’Donnell 2004). It became common for women to engage in parttime casual employment since the shorter hours required to fulfil care obligations were often available only on this basis (Owens 2001; Pocock et al. 2004). This vexed history shapes manifestations of casual employment in Australia, where all casuals receive ‘loadings’ (wage increments) in lieu of various social benefits and lack access to the full range of protections extended to permanent employees. As a result, a sizeable subset of parttime employment in Australia (that is, part-time casual employment) is defined not only by shorter than ‘normal’ working hours but by uncertainty and limited access to certain benefits (for example, paid vacation, sick leave and public holidays as well as, in practice, unfair dismissal protection). This subset of part-time workers encompasses those casuals engaged on a ‘one-off’ basis (Stewart 1992) for whom a host of protections simply do not apply, those with fixed-term contracts, and those engaged on a more or less ongoing basis, who represent the majority and whose access to protections depends technically upon the length of the term but who are often denied protection on account of regulators’ vernacular interpretations of their casual status (O’Donnell 2004). Many such part-time casual workers are women with caregiving responsibilities. In Australia, there have been efforts to remedy precariousness among part-time workers. However, these tend to follow a dualistic strategy: on the one hand, some regulatory interventions attempt to confine casual employment to intermittent and irregular employment by fostering conversion to permanent status among those casuals (part-time and full-time) who are engaged on a continuous basis. On the other hand, some seek to extend protections and benefits to part-time workers, largely on a pro-rata basis, although these are typically of greatest benefit to part-time permanent workers. Consistent with SER-centrism, the result is the neglect of those who are not only the most sizeable segment of part-time employees but also the ones with the worst employment conditions. A similar set of dynamics operates with respect to self-employment in Canada, a country where the rates are relatively high in comparison with other OECD countries, where growth rates among women far exceeded those of men between 1976 and 2008,7 and where the solo variety, a situation in which the self-employed person does not employ any others, drove its expansion in the post-1980 period.8 SER-centric regulatory approaches uphold the employment relationship as the basis of labour

66 Regulating for Decent Work

protection. They stop short of extending protections to workers who are self-employed and thus fall outside the employment relationship. In Canada, as in other highly industrialized contexts, including in many member states of the EU 15 and Australia, those that are solo selfemployed in their main job are often depicted as choosing independence, freedom and autonomy (Lin, Yates and Picot 1999, p. 6; see also EC 2004; Howard 2004), although in practice many women in this group pursue solo self-employment due to their inability to find suitable paid employment and/or to ‘balance work and family’ and many immigrants resort to it in response to discrimination in the labour force. Low income is a major problem facing this group, particularly women.9 And yet most labour policies, such as minimum employment standards operating at the provincial and federal levels and federally administered income protection policies, still exclude them from coverage. For example, most solo self-employed people lack coverage under provincial minimum labour standards legislation, which includes minimum wages. Most also lack collective bargaining rights under the Canada Labour Code and parallel provincial legislation as well as access to regular unemployment benefits under the federal Employment Insurance programme (Fudge et al. 2002; Cranford et al. 2005; Vosko 2008; MacDonald 2009). The character of solo self-employment in Canada highlights the limits of SER-centric approaches to regulation retaining the employment relationship as the basis of labour protection. More narrowly, it illustrates that precarious self-employment is not limited to the narrow subset of disguised employment. In recognition of this limitation, there have been modest efforts to pursue protections for ‘persons who perform services comparable to those provided by employees and under similar conditions, but whose contractual arrangements with the employer distinguish them from “employees”’ (Canada 2006, p. 64). For example, a 2006 review of the Canada Labour Code proposed extending measures to protect these ‘autonomous workers’ ‘basic right to decent working conditions’ (Canada 2006, Rec. 4.2). However, such efforts still follow an SER-centric logic: even these proposals fail to respond adequately to precarious self-employment because they deal only with self-employed workers that closely resemble employees.

Alternatives to SER-centrism The foregoing examples illustrate the limits of the SER-centric logic of contemporary approaches to responding to precarious employment at

Leah F. Vosko 67

the international and national levels. By conflating non-standard and precarious employment, and thus framing the SER – or close proximity to it – as the solution to precariousness, such approaches mean the greater the deviation from this model, the less protection laws and policies provide. Those workers that are on the furthest margins of the SER and its associated participation norms are the least well served. There is, therefore, a pressing need to address the disjuncture. This process entails disentangling precarious employment, the appropriate target of re-regulation, from non-standard employment. There is a relationship between non-standard and precarious employment. This connection is, however, neither inevitable nor straightforward. When laws and policies treat precarious and non-standard employment synonymously, they obscure the very problems that need to be addressed. At the same time, as the foregoing analysis has shown, changing approaches to employment regulation in this direction is only one part of the solution. Historical exclusions from the SER and its precursors, directed principally at serving the adult male citizen, suggest that another element entails responding to changing gender relations, citizenship boundaries, and related age-based transitions.10 In recent years, two alternative approaches to SER-centrism have emerged that merit scrutiny through such angles – the ‘flexible SER’ (Bosch 2004) and the ‘beyond employment’ (Supiot 2001) approaches. The ‘flexible SER’ The product of a widely cited scholarly intervention, ‘Towards a new standard employment relationship in Western Europe’ (Bosch 2004), and complementary interventions by other scholars (see, for example, Esping-Andersen 2002), the flexible SER approach is a response to debates about successors to the SER. It is an attempt to replace the SER of old with a more decommodified alternative for workers in employment situations formerly falling outside its range. It seeks ‘a flexible framework for self-organized diversity, in which the differing interests of individuals, firms and society are balanced out and the social security system is linked to economic efficiency’ (Bosch 2004, p. 635). The flexible SER encompasses an expanded range of forms of work for remuneration. Although employment is to remain the basis for most labour and many social protections, under this approach its form is to matter less in their content and design. Distinct from the SER-centrism of old, the flexible SER makes way for a new ‘new bargain’ between employers and workers. It pursues two additional avenues to improve the functioning of the SER: first, flexible

68 Regulating for Decent Work

work organization, specifically, increased opportunities for adjusting paid working hours, counterbalanced by a shift to individual from derived rights, with the goal of encouraging all workers to build ‘independent’ social protection through paid work (Bosch 2004, p. 634). The emphasis on flexible work organization also aims to support lifelong learning, where the accent is on ‘active transfers’ for persons experiencing difficulty in achieving labour force integration. The second area of intervention emphasized is the development of public childcare infrastructure for children under six years of age and all-day schooling for the school aged. Through such avenues, at the level of gender relations, the flexible SER approach supports a gender contract defined by dual breadwinning in which certain forms of caregiving (for example, childcare and eldercare) are valued. At the same time, where social protection for men and women is built through labour force activity, proponents of a new flexible SER put their faith in equal employment opportunity. The path to addressing the gender of precarious employment lies in providing mechanisms for both men and women to engage in a range of forms of employment and work arrangements. The flexible SER approach does not treat the ongoing division of domestic responsibilities as an integral matter for public policy. It is assumed that once certain care infrastructure (for example, childcare) is in place and once women no longer derive their rights as dependants, it is sufficient to treat men and women as if they are similarly situated, disregarding persistent gender divisions outside the labour force and, thus, for example, leisure time inequalities (Fraser 1997).11 These limitations pose the risk that a flexible SER approach would contribute to the resort to migrant women care workers. In practice, countering this potential tendency, a flexible SER could accommodate a model of citizenship akin to residence-based denizenship for migrant workers or the extension of considerable employment, social, and civil rights and select political rights, based on legal domicile (Hammar 1985; Lister 1997, p. 48; see also Soysal 1994). However, proponents do not address citizenship boundaries explicitly and thus a nation-state conception remains the default. The commitment to a diversity of forms of employment without a loss of protection characterizing the flexible SER approach, coupled with its support for dual breadwinning and its potential to accommodate denizenship, make it responsive to the disjuncture in significant measure. At the same time, by retaining the notion of an SER, this approach does not have the capacity to dislodge fully either the association between

Leah F. Vosko 69

non-standard and precarious employment or the gender relations and citizenship boundaries integral to an SER (albeit modified). ‘Beyond employment’ The other alternative approach, ‘beyond employment’, pursues a vision of labour and social protection that is inclusive of all people, regardless of their labour force status, from birth to death, in periods of training, employment, self-employment, and work outside the labour force, including voluntary work and unpaid caregiving. It seeks to spread social risks, to be attentive to transitions in the lifecycle, such as movements from paid employment to retirement and from school to work, and to value civic engagement. The approach assumes that every worker should be able, as required, to reduce or increase paid working hours at certain points in his or her lifecycle, while retaining access to protection and income support. A central aim is to normalize working time adjustments to accommodate shorter working hours in periods of weak demand, ongoing voluntary community activities, periodic skills upgrading, and phased-in retirement, as well as extended leaves, such as maternity and parental leaves, and to ensure that workers can maintain longer hours in peak periods of labour force participation. The idea, being encapsulated in the phrase ‘statut professionel’ – translated in English to the notion that ‘an individual is a member of the labour force even if he or she does not currently have a job’ – is to reject a linear and homogeneous conception of working life tied to the employment contract (Supiot 2001, p. x). At the level of employment regulation, adopting this approach would entail moving away from the pillars of the SER as bases for protection. It would mean extending protections to all persons, regardless of their employment status, who were engaged in work for pay and economically dependent on the sale of their capacity to work (Fudge et al. 2002). It would also involve organizing paid working time to better reflect both life’s different phases and changing employment norms. Finally, its notion of ‘statut professionel’ would de-link benefits and job tenure; it would cast gaps in employment, fluctuating levels of employment intensity, and jobs of varying duration as typical (for example, Anxo et al. 2006, p. 94). In this way, the idea is consistent with calls for supporting ‘transitional labour markets’, a notion developed by Schmid (2002, 2005 and 2007) to designate critical phases in an individual’s working life that denote either the beginning or the suspension of continuous employment

70 Regulating for Decent Work

and therefore require a profound adjustment of the labour contract so that a (new) employment relationship can start, or an existing one [can] either be maintained under special conditions, be combined with other useful (unpaid) activities or be (gradually) terminated in favour of retirement. (Hartlapp and Schmid 2008, p. 410) Schmid and others (Auer 2006 and 2007) use this idea to, for example, rationalize calls for a movement from employment insurance to ‘employment ensurance’ – that is, to shift the emphasis of employment benefits away from providing for job security towards employment and/ or labour market security to support workers in transition (for example, moving from school-to-work, work-to-retirement, changing jobs). Social drawing rights are the main mechanism advanced for realizing this approach’s vision for working time adjustments. They are to supplement, rather than replace, entitlements and benefits related to other risks (for example, illness, occupational injury, layoffs and so on), which remain essential to sustaining strong labour and social protections. As I interpret them, social drawing rights are both public and collective, to be distinguished sharply from individual investment accounts. These rights are Supiot’s (2001, p. 56) attempt to enable people to draw on their prior labour force contribution, on the basis of a ‘free decision’ rather than on account of risk, at times when they are required to engage in other forms of labour (such as unpaid caregiving work) or civic participation. He envisions them as ‘a new type of social right related to work in general’ (p. 56) and casts them as social in the way they are established (that is, the process of building up the reserve) and in their aims (that is, the utility of their exercise to the community).12 The beyond employment approach supports a gender contract defined by a more equitable distribution of work (paid and unpaid) among men and women. Like its flexible SER counterpart, it supports a move from derived to individual rights for women, but sees this as insufficient, preferring to recast social rights based on solidarity conceived as a vehicle connecting social rights13 and group-based guarantees to social equality (Supiot 2001, p. 227). In supporting social drawing rights, beyond employment also elevates the value of socially useful activities or work in the public interest. However, publicly provided care services are not a focus of this approach, raising the question of whether moving beyond employment could have the effects of prioritizing income replacement over public infrastructure for socially necessary care work and underrepresenting the importance of labour force participation to women’s equality.

Leah F. Vosko 71

With these qualifiers, the beyond employment approach goes a fair distance towards a universal caregiver model, popularized by Fraser (1997, p. 45) for application in the US welfare state, defined by gender equity. This model entails the development of policies founded on the principles of anti-poverty, anti-exploitation, income equality, leisure-time equality, equality of respect, anti-marginalization, and anti-androcentrism. Under this model, all jobs are to be designed on the assumption that workers are caregivers, shortening hours of work for pay across the board, and extensive employment-enabling services are to be provided. Some informal care work is to be supported publicly, and merged with work for pay under social insurance, whereas other state-supported care work is to be located in civil society (for example, in locally organized institutions) (Fraser 1997, pp. 60, 61). By dismantling oppositions between breadwinning and caregiving and public and private responsibilities, the universal caregiver model offers possibilities for transforming gender relations. Taken to its logical conclusion, it dispenses with the need for a gender contract altogether. The beyond employment approach also assumes (albeit implicitly) a version of citizenship status extending further than the nation-state, while still upholding the fusion of community membership and territory – supranational citizenship whereby rights previously attached solely to national citizenship are supplemented by those tied to citizenship in a larger geopolitical entity comprised of a number of states. The model is EU citizenship, which extends considerable economic, especially work/ employment rights, civil and social rights, as well as select political rights, to citizens of the Union, defined as ‘every person holding the nationality of a Member State’ (EU 1992, Art. 8.1), and lesser rights, tied principally to employment, to third country nationals entitled to become permanent residents of EU member states. One potential outcome is a reduction in certain exclusions contributing historically to the prevalence of precarious employment among migrant workers holding EU citizenship (or EU longterm residence permits) and moving between member states. Another is the extension of elements of residence-based denizenship to third country nationals who are long-term residents of EU member states.14 Still, the overriding logic of a ‘supra-nationality’ (Delanty 1997) reproduces territorially delimited citizenship boundaries at a different scale. Since it is premised on supranational citizenship, much like the flexible SER approach, the version of gender equity informing the universal caregiver model to which beyond employment aspires is also silent on the linkages between gender relations and citizenship boundaries. The universal caregiver model is criticized strongly for neglecting global

72 Regulating for Decent Work

sex/gender divisions, and specifically for focusing on revaluing caregiving work among women holding national citizenship who care for their own locally resident children without paying sufficient attention to the many migrant women care workers in highly industrialized countries who care for other people’s children (Weir 2005, p. 311). Given its lack of attention to gendered processes of global exploitation and economic polarization, the geopolitics of its ‘universal’ vision also merits critical reflection (Beneria 2008; Raghuram et al. 2009; Williams 2008). Thus, even the realization of a beyond employment approach risks perpetuating migrant women’s continuing role in performing a large share of illremunerated paid care work in highly industrialized contexts (Hassim 2008). Additionally, even if the work of paid care workers is fully valued through public care infrastructure in receiving contexts, along with the extension of full membership to migrant care workers, the model addresses neither the effects on the sending economies (Williams 2008) nor women’s limited access to paid employment in the formal economy therein, the very factors compelling many women to migrate in the first place (Beneria 2008).

Regulating for ‘decent work’: creating the conditions for global labour market membership The flexible SER and beyond employment approaches offer promise at the level of employment regulation, especially in severing the link between precarious and non-standard employment. Both approaches also recognize the importance of addressing new participation norms along the lines of gender. Additionally, the beyond employment approach goes a considerable distance in acknowledging the significance of age-related transitions associated with the different phases of life. Still, these alternatives largely neglect the problems associated with citizenship boundaries, so central to the SER and to SER-centrism, and each falls short in terms of offering integrated strategies for re-regulation. As this section will demonstrate, such limits necessitate combining the strongest elements of the flexible SER and beyond employment approaches with mechanisms for reorganising the paid work/membership/care nexus. Doing so would foster entrées for groups of workers hitherto excluded – or partially excluded – from protection due to the SER’s associated participation norms. The notion of labour force membership, or statut professionnel, offers a useful starting point for this exercise. Advanced in connection with the beyond employment approach as a means of reducing the link

Leah F. Vosko 73

between employment status, form of employment, and precariousness, it responds principally to the situation of employed workers in national or supranational contexts facing bouts of unemployment or moving between jobs and employment statuses. However, this conception could be developed to encompass labour force participants defined more broadly (that is, employed, unemployed, and discouraged workers and trainees) and persons moving from one country and/or region to another for work/employment as well as unpaid carers in between and during phases of employment (part- and full-time). Global labour market membership would aim to address processes central to workers’ social reproduction in recognition of the fact that the internal inclusions and exclusions of national citizenship, the male breadwinner/female caregiver contract, and form of employment were intertwined under the SER. It would limit precarious employment through cultivating ‘inclusive citizenship’ (Lister 1997 and 2007), global ‘universal caregiving’ (Fraser 1997; see also Weir 2005; Beneria 2008; Williams 2008), and life-course support for age-related transitions (Schmid 2002, 2005 and 2007; Auer 2006 and 2007) in tandem with de-linking employment status, form of employment, and access to protection. As it is conceived here, building on the work of Lister (1997, see also Lister 2007), inclusive citizenship is a means of rethinking birthplace, lineage and residency as primary bases for accessing protection. It entails five components attempting to free the notion of ‘citizenship’ from the boundaries of nation-states while still recognizing state power in delimiting inclusions and exclusions: non-discrimination so that rules governing entry do not disadvantage particular social groups; observance of basic human rights; autonomous legal status of migrants as individuals regardless of gender or marital status; internationalism, especially highly industrialized countries’ obligations to migrants from economically disadvantaged world regions; and transculturalism, or the affirmation of cultural differences and their fluidity (pp. 63–4). These components complement visions for global universal caregiving by addressing external (that is, status citizenship) as well as internal inclusions and exclusions (Bosniak 2002). Global universal caregiving is, in turn, an attempt to rework the notion of gender equity underpinning the universal caregiver model by enlarging Fraser’s (1997) central notions of anti-poverty, anti-androcentrism, and income equality to support women’s access to paid employment in the formal economy in low-income countries (Beneria 2008), of antiexploitation to respond to the harsh (and discriminatory) treatment

74 Regulating for Decent Work

of women workers employed by foreign-owned firms in such contexts (Hassim 2008), of anti-marginalization to improve low-income states’ capabilities to deliver social supports necessary for limiting gendered poverty, and of equality of respect to strengthen (especially women) workers’ power to shape the content of international regulatory frameworks around the world (Pearson 2004). Research investigating global care chains reveals how, having moved beyond the male breadwinner/ female caregiver gender contract in highly industrialized countries, the terms and conditions and quality of care are now subject to calculative processes of commodification regardless of where it is provided (that is, the public sphere or the home) (for example, Hochschild 2000). To respond to the transnational political economy of care (Williams 2008, p. 12), there is a need to develop supports for care workers in receiving countries meeting the needs of ‘welfare [state] subjects’ at the same time as addressing global interdependencies contributing to escalating crises of social reproduction in economically disadvantaged world regions. Finally, life-course support is envisioned as a response to the increasingly significant risks posed by school-to-work and work-toretirement transitions, as well as shifts between paid and unpaid work on account of caregiving responsibilities and migration-related movements and the interaction between the two (Hartlapp and Schmid 2008). The aim is to embrace the idea of the transitional labour market (Schmid 2002) by providing ‘protected mobility’ (Auer 2006 and 2007) and different permutations and combinations of schooling, work (paid and unpaid), and retirement to, for example, limit income loss and other labour market insecurities due to unpaid caregiving and/or the acquisition of new or renewed job-related skills. Such supports must, however, be distinguished from neoliberal ‘activation’ strategies, which tend to assume a homogenous unemployed population and emphasize rapid return to employment through the use of punitive sanctions and the notion of mutual obligations (Heikkilä 1999, p. 9). Giving effect to these component principles, for persons moving from one country to another for work and/or employment, global labour market membership would contribute to reducing the salience of entry category in shaping access to social and economic rights and protections in particular places. It would also address gaps in rights and protections among migrating workers falling in between different regulatory regimes; that is, it would cultivate protections for migrant workers in the recruitment and placement processes. Among workers with caregiving responsibilities and enduring age-related transitions, by acknowledging the significance of non-labour force work (as well

Leah F. Vosko 75

as educational and training activities) and recognizing its coexistence with employment, it would lessen the compulsion to participate in precarious employment. Policy options towards inclusive citizenship A few emergent policy formulations and innovations aimed at limiting national citizenship as a mode of differentiation and exclusion from protection, and taking root in institutions distinct from those associated with SER-centric laws and policy, share affinities with the notion of global labour market membership. One example of a formulation is Gordon’s (2007, p. 563) proposal for transnational labour citizenship, ‘a new immigration status… entitl[ing] the holder to come and go freely between the sending country [Mexico] and the United States, and to work in the United States without restriction’ based on membership in a transnational workers’ organization instead of ties with a particular employer. By making such membership the basis for attaining this immigration status, Gordon envisions shifting the enforcement of basic employment standards, her focal concern, from ‘the arena of immigration policy in employers’ hands, to the arena of labour solidarity’ (p. 509). Transnational labour citizenship is an interim response15 to a recurring dilemma: on the one hand, the need for a global response to massive inequality compelling people to migrate for employment and, on the other hand, the threat of competition between migrants and national citizen-workers in receiving countries, contributing to xenophobic policies and practices amongst some local workers’ organizations (p. 524). A noteworthy feature of this proposal is the role it accords collective actors in attempt to raise the floor of conditions of work and employment for all workers regardless of immigration status. In this way, the example underscores that principles of global labour market membership will take effect differently than those underpinning the SER’s participation norms. Formulations such as transnational labour citizenship shape and are shaped by policy innovations, emerging at different scales. Like this new immigration status, many such innovations are designed as interim measures, or are short-term oriented. Many are also short-lived. Still, they tend to advance inclusive citizenship in their form and/or institutional bases. One type of innovation is the negotiation of bi- or multinational agreements between unions and employers or workers’ organizations (unions and/or civil society groups) and governments. An example of a binational agreement between unions and employers emerged in 2006 when the United Farm Workers and Global Horizons, a farm labour

76 Regulating for Decent Work

contractor operating in approximately a dozen US states, employing one thousand workers at that time and planning to expand its workforce by a further two thousand, reached an agreement covering terms and conditions of employment as well as aspects of recruitment and placement. Although it was short-lived and was motivated by the decision of the government of Washington State to revoke the firm’s license to do business because of labour standards violations,16 this unique three-year agreement in for example, cluded provisions ranging from protection from retaliation, firing only for a just cause, wage increases above the floor established under temporary migrant worker visa programmes, and paid medical care to workers’ compensation, paid work breaks and seniority protection in hiring and layoffs (Global Horizons–UFW 2006; Greenhouse 2006). One example of a binational agreement between workers’ organizations and government took shape in 2009 when the province of Michoacán in Mexico, the United Food and Commercial Workers of Canada, and the Agricultural Workers Alliance (AWA), a civil society organization, signed a historic agreement to extend services and assistance to Mexican migrant farm workers who were working in Canadian fields and greenhouses. Through this agreement, which is in the process of being extended to other Mexican provinces, nine regionally based centres in four Canadian provinces, operated by the AWA in cooperation with the government of Michoacán, which commits itself to providing human and intellectual resources, deliver services to workers, apprise them of their rights, and provide support for their realization (for example, translation), as well as assisting workers in the retaining of community ties in Mexico and Canada (for example, through providing free long-distance telephone access) (AWA 2009; http:// awa-ata.ca/en/; Alex Barillas, personal communication, UFCW Canada). With respect to labour standards, agreements also exist between subnational arms of government involving labour departments. Among those that are most developed, many are motivated by concerns to attract and retain health care personnel. For example, non-binding agreements exist between the labour ministries in the Canadian provinces of Alberta, British Columbia, Manitoba, and Saskatchewan and the Department of Labour and Employment of the Government of The Philippines, the country of origin of the largest group of migrant health care professionals employed in Canada. Such agreements cover: exchange of information concerning employers and sending agencies; recruitment and selection of workers; cost of recruitment of workers; offers of employment and labour contracts; protection of workers; and human resource development. Provisions regarding workers’ protection aim to prevent

Leah F. Vosko 77

workers from being charged direct and indirect fees incurred in the hiring process, to mandate providing workers copies of their employment contracts, to empower the Philippine Overseas Labour Office to ensure the protection of migrant workers under applicable federal and provincial laws, and to contribute to sustaining health and human resource development in the Philippines. A related area of innovation, with deeper roots, involves agreements between sending and receiving governments providing for the extension of work- and employment-related social security benefits beyond national borders. Currently, it is mainly migrants from and going to highly industrialized countries that have access to and portability of benefits tied to work and employment (such as pension and medical benefits) through such bilateral or multilateral agreements, yet this group represents just 23 per cent of all migrants worldwide (Avato et al. 2009, p. 4). There is, however, growing recognition of the need to extend portable and especially exportable benefits to migrant workers from low-income countries and world regions (see, for example, Yeates 2005; Avato et al. 2009; MacLaren and Lapointe 2009; Van Ginneken 2009).17 While recognizing that ‘within the EU migration flows are typically more symmetric than between developed and developing countries’, many analysts view the EU as a model to emulate because it has the highest standards of portability of work-/employment-related social security benefits across international borders (Holzmann et al. 2005, p. 30). They thereby propose best practices involving the establishment of bilateral agreements for benefits portability, based on multilaterally agreed standards resembling EU regulations on European labour mobility or the agreements between the EU and the Maghreb countries. Furthermore, some commentators suggest that there may be openings in the General Agreement on Trade in Services (GATS) for greater portability of benefits among temporary migrant workers: according to Yeates (2005, pp. 18–19), paradoxically, under GATS Mode 4, the national treatment principle could require governments to eliminate discrimination against temporary migrant workers in their access and use of work-/employment-related social security benefits and the Most Favoured Nation principle could require governments to eliminate discrimination among foreign nationals. Yet, as Yeates also acknowledges, the threat that the rules of GATS as a whole will cultivate a race to the bottom with respect to all such benefits, as well as pressure to open services provision to private actors, moderates this positive prospect significantly (see also Bach 2003).18 A further area of policy innovation encompasses receiving governments’ independent efforts to protect migrant workers from exploitative

78 Regulating for Decent Work

processes of recruitment and placement. Legislation adopted in Canada regulating the recruitment of foreign care workers offers a case in point. In 2009, following on the efforts of other provinces, the province of Ontario introduced Bill 160, the Employment Protection for Foreign Nationals (Live-in Caregivers and Others) Act. Consistent with principles contained in the ILO Multilateral Framework on Labour Migration (2006) and provisions proposed under the ILO’s model trade union agreement on migrant workers’ rights, this Act seeks to protect individuals migrating to Ontario under the federal Live-in Caregivers Program by prohibiting recruiters from charging fees, either directly or indirectly, to migrant care workers, including recruitment and placement fees as well as fees for other services.19 Another example directed at placement and aimed at facilitating the denationalization of labour protections through subnational interventions is a scheme known as ‘Window for the Jobless’, launched in 2002 by the provincial government of Naples, Italy. This initiative fosters secure employment and inclusive citizenship among migrants by providing information, guidance and assistance on employment opportunities and labour rights. Uniquely, it plays a proactive role in assisting undocumented workers in processes of regularization and employment matching pre- and post-regularization. To facilitate regularization, the staff of ‘Window for the Jobless’ distribute regularization kits and act as mediators between state officials (including police), employers, and workers unfamiliar with immigration policies. They also operate a telephone help desk and internet site to address queries and assist workers and employers in filling out regularization forms. This institution was created just before a large-scale amnesty in 2003 when ‘many employers… balked at the prospect of regulation for undocumented foreign employees’, due to administrative costs (for example, the requirement to contribute to social security for migrants awaiting regulation) and hurdles (Girard and Chambers 2005, pp. 201–2). Although the ‘Window for the Jobless’ scheme is highly localized, it exemplifies how innovative practices at the urban scale can give rise to a ‘defacto regime[s] of new rights and identities’ (Holston 2001, p. 340) closely related to principles of global labour market membership. Policy options towards global universal caregiving and life-course support International, regional and local policy initiatives creating entrées for groups hitherto excluded from protection in connection with the SER’s participation norms along the lines of citizenship through

Leah F. Vosko 79

denationalizing employment protections offer an important angle into advancing global labour market membership. However, it is also necessary to introduce policies and practices fostering global universal caregiving and providing supports through age-related transitions. To such ends, social drawing rights, in particular supports for activities in the public interest, interpreted here as public and collective rights, offer one means for realizing global labour market membership. To the extent that they are imagined under the beyond employment approach (for example, as rights for the release of individuals’ time), these rights are to be attainable among all paid workers in a territory on the basis of a prior contribution to the labour force. They are to be accessed by national and supranational citizens or denizens with employment histories in a given national or supranational geopolitical entity. Consequently, were they to materialize, persons lacking reserves on which to draw, such as (largely female) unpaid caregivers, young people, and migrant workers, would have difficulty accessing supports for work in the public interest and/or training. Social drawing rights could, however, be developed to support the unpaid caregiving work, as well as educational and other activities, of women and men lacking sufficient reserves from which to draw due to their age, work/employment, and/or citizenship status. For those with care obligations in the territory in which they are employed, state social security systems offer a logical means of delivering social drawing rights. Policy innovations that could inspire their delivery with respect to the care of children exist in northern Europe, where, in addition to providing for childcare, governments are increasingly extending a package of supports to residents (that is, citizens and denizens), including parental leaves, to engage in caregiving, extending well after the period immediately following childbirth. In addition to providing public subsidized childcare administered at the municipal level, Sweden, for example, provides 13 months of paid parental leave to be shared by the parents after the birth of a child, two of which are reserved for fathers (known as ‘daddy leave’ and inspired by the Norwegian model); gives both parents rights to engage in part-time employment until their child reaches the age of 8 without the loss of social entitlements, offering parents the option of sharing days; and extends generous annual provisions (60 days) for temporary parental leave per child (below 12 years of age) (Björnberg 2002; Nyberg 2004). Such supports have had modest effects in changing gendered patterns of caregiving; this outcome may relate in part to their origins in developing a national welfare state promoting wage labour to increase the tax base and the freedom they give parents for negotiating and utilizing such measures.20 Still, as a

80 Regulating for Decent Work

package they begin respond to new participation norms along the lines of gender- and age-related transitions tied to childbearing and -rearing. In attempt to limit the development of exploitative global care chains, Sweden also extends equivalent working and employment conditions to migrant and non-migrant care workers alongside such supports (Hassim 2008). Yet there remains a vacuum where the social drawing rights of such migrant workers are concerned. Although it does not address exclusions on the basis of citizenship, Belgium’s career break/time credit system represents another model for developing social drawing rights for activities other than childcare – one that is uniquely attentive to shifting participation norms with respect to age. Originating in 1985 in a career break scheme operating through the country’s unemployment insurance programme, and financed by a combination of worker and employer contributions and state subsidies, time credit and career break leaves, provided for private and public sector workers respectively, can be taken at any time during the course of adult life and do not require specific reasons for uptake. Indeed, they operate separately from three coexisting thematic leaves for palliative care, parental leave and medical care.21 A general right for all employees (both part- and full-time), in the private sector the career break system provides for reducing working hours or career interruptions of up to a year full- or half-time and for a maximum of five years for a one-fifth reduction in time whereas in the public sector the time credit system allows employees to reduce their working time by one-fifth, one-quarter, one-third or one-half; in both sectors, the duration of leaves may be extended to five years by collective agreement on an industry or company level (Devisscher and Sanders 2007, p. 122; Vandeweyer and Glorieux 2008, p. 276). Workinghour reductions and career interruptions are available to all public and private sector employees22 who have been employed by the same employer for at least one year (although only 5 per cent of employees in an organization may take this leave simultaneously).23 Eligible employees receive a modest flat rate benefit, whose level is calculated using a formula that takes into account age, the type of time benefit chosen (that is, full-time, half-time, or one-fifth interruption), the number of years worked, and the family situation,24 and maintain all their rights to social security as well as job security (De Lathouwer et al. 2007, p. 178). The career break/time credit system also includes a special scheme available to employees aged 50 or older, who are permitted to reduce their working time by half or one-fifth until retirement. In 2006

Leah F. Vosko 81

approximately 211,000 people were entitled to compensation for career breaks/time credits. Regarding uptake, the one-fifth working time reductions have proved the most popular. Furthermore, among younger age groups, women dominated strongly (Devisscher and Sanders 2007, p. 123). Still, as De Lathouwer et al. (2007, pp. 166–7) demonstrate, new ‘care’ and ‘skill’ risks together motivated the contemporary redesign of this scheme: namely, the intertwined risks of severe income losses and longer-term poverty among (principally women) workers compelled to withdraw from the labour force to engage in unpaid caregiving and among young and older workers unable to access education and training over their careers. For those with care responsibilities outside the territory in which they work, it is also possible to envisage creative modes of delivering social drawing rights. In this respect, bi- and multinational collective agreements between workers and employers offer one potential vehicle. There was, for example, in the UFW-Global Horizons agreement provision for bereavement leave, which covered travel to and from the country of origin while holding open the bereaved worker’s job. Albeit limited, this type of provision offers a starting point for extending measures providing for the release of migrant workers’ time to fulfil caregiving responsibilities with no loss of protection.25 Complementary provisions, designed to enable workers in mobile professions to engage in caregiving, also exist in international labour regulations, such as the ILO Maritime Labour Convention, 2006. Paradoxically, even though this Convention is targeted mainly at a group of workers comprised of significant numbers of men, a report prepared by the ILO (2010, p. 96) on developing a consolidated Convention on domestic work, a field of employment in which women predominate, cites it as a model agreement. One reason for this conclusion is that under the Maritime Labour Convention workers obliged to reside far away from their families at their place of employment are to have opportunities to take periodic paid home leave. Another less individualized and longer-term means for securing support for easing the care obligations on migrant workers, including migrant women workers engaged in paid caregiving work in highly industrialized countries, entails compelling receiving countries to support public physical and social care infrastructure in sending economies. One means is through the establishment of taxes linked to international trade and migration (see, for example, Beneria 2008, p. 17; Pearson 2003), measures that could be advanced through an International Tax Organization.26 There are also prospects for other sorts of investments

82 Regulating for Decent Work

in care infrastructures. For example, in 2004, the 50th World Health Assembly passed a resolution urging member states to ‘establish mechanisms to mitigate the adverse impact on developing countries of the loss of health personnel through migration, including means for the receiving countries to support the strengthening of health systems, in particular human resources development, in the countries of origin’ (WHO 2004, Res. 57.19, Para 1.4). Consistent with this directive, South Africa and the UK operate an on-the-job educational exchange programme that enables clinical professionals to gain experience in NHS facilities in the UK and rural areas of South Africa respectively (Nullis-Kapp 2005, pp. 82–3). Such initiatives offer potential for not only sustaining collective care infrastructure but, given the gendered composition and age-related structure of the global health care workforce (WHO 2006), also simultaneously addressing new participation norms along the lines of gender, citizenship status, and age-related transitions.

Conclusion As has been shown by the foregoing exploration of the historical roots and contemporary dynamics of SER-centrism, a male breadwinner/ female caregiver contract and national citizenship helped lay the basis for the SER, and its exclusions, in early national as well as international labour regulations. The SER was never universal, nor was it ever meant to be. The SER cannot be understood apart from its exclusions; indeed, it rests on them. Therefore, SER-centric approaches to addressing precarious employment that conflate it with non-standard employment are not only least likely to improve employment situations deviating sharply from this model, they are poised to perpetuate long-standing exclusions and partial exclusions from protection. Consistent with the emphases of the flexible SER and beyond employment approaches, in light of this problem, severing the link between form of employment and dimensions of labour market insecurity is essential. Yet so too is responding, via institutions outside those perpetuating SER-centric laws and policies, through moves to denationalize worker-citizenship, assuming that all workers are carers and providing for life-course supports across transitions between, as well as different permutations and combinations of, education, caregiving, and employment. In order to regulate for decent work, integrated strategies directed at global labour market membership are required. Despite their limitations, innovations of the sort canvassed herein are helpful in envisioning the type of institutional and policy arrangements necessary to overcome the problem of SER-centrism.

Leah F. Vosko 83

Notes 1. In outlining the problem of SER-centrism and contemporary responses to it, the first two sections of this chapter refer to the findings of Managing the Margins, specifically those of Chapters 1–3 and 4–6 respectively, whereas the third section advances the preliminary vision delineated therein by sketching a number of promising policy formulations and innovations. 2. This is not to suggest that alignment with the SER does not improve the conditions of some workers. Nor is it to dismiss important functions of the SER, such as access to training, regulatory protections, social benefits, and decent wages. Rather, the goal of this chapter is to make visible the ways in which such vital functions have been fused to a particular form of employment, and their consequences. (On the distinction between the functions and the form of the SER, see especially, Bosch 2004, p. 218.) 3. In another part of this volume, Tomei (Chapter 10) explores a further example: the emergence of an ILO standard on domestic work. 4. The Fixed-Term Work Directive, in addition, offers the technique of aligning fixed-term work to the SER; namely, conversion to an open-ended contract after a certain period or number of contracts to be determined nationally (on this point, see also McCann 2008). 5. The official reason for this exclusion was that there were plans for a separate Directive on Temporary Agency Work. However, the decision did not emerge from a consensus between the social partners negotiating the initial framework agreement on temporary work. Rather, it moved forward to salvage negotiations at the objections of ETUC negotiators and other social actors, who were then concerned about the adoption of an inferior directive on temporary agency work (for a detailed analysis of these negotiations, see Vosko 2009). 6. In 2006, 28 per cent of all women temporary agency workers in the EU 15 engaged in temporary agency work for periods exceeding seven months (vs 20 per cent of men) (Eurostat 2006). In the same year, 11 per cent of all temporary agency workers in the EU 15 were non-national citizens of the countries in which they were employed as opposed to 5 per cent of permanent workers and 7 per cent of all temporary workers. The share of young people (aged 15–24) among temporary agency workers also ranged between 30 and 50 per cent nationally in the EU that year, notably given that young people constitute approximately 10 per cent of the EU labour force (ibid). Similar patterns pertain elsewhere. In Canada in 2004, women represent 53 per cent of temporary agency workers and, among women, racialized immigrants have the highest odds of engaging in this type of temporary employment – 4.5 times higher odds than other women (Fuller and Vosko 2008, p. 44). 7. Between 1976 and 2008, the number of self-employed men roughly doubled (from 873,400 to 1,719,700) whereas the number of self-employed women nearly tripled (from 311,600 to 909,900) (Statistics Canada 2008). 8. Self-employment constituted 16 per cent of total employment in Canada in 2008, 5 per cent of which was comprised of employer self-employment and 11 per cent of which was made up of the solo variety. 9. For example, in 2006, 47 per cent of men reporting solo self-employment as their main job and 57 per cent of women reporting solo self-employment as

84 Regulating for Decent Work

10.

11.

12.

13. 14.

15.

16.

17.

their main job had annual incomes of less than $20,000 Canadian (Statistics Canada, 2006). The ensuing analysis focuses on gender relations, citizenship boundaries, and age-related transitions due to the explicit exclusions characterizing SERcentric regulations historically at the national and international levels for much of the twentieth century, highlighted in the previous two parts. Nevertheless, other exclusions characterizing many SER-centric regulations also merit detailed examination, such as those related to ethnicity, race and (dis)ability. One exception is found in the work of Esping-Anderson (for example, 2002), who does not use the term ‘flexible SER’ but whose policy prescriptions otherwise resemble this approach, who supports explicit incentives for fathers to take leave after the birth of a child (akin to the ‘daddy leaves’ available in Nordic countries, where men’s share of total child leave days is modest but rising). To the limited extent such mechanisms are imagined (or exist, in minor ways, in practice), social drawing rights tend to operate by releasing an individual’s time, normally during an employment contract or following its completion, and they are to be funded principally outside the market; the reserve for these rights is to emanate either from the state directly (that is, for tasks of public interest), from social security (for example, by virtue of having a dependent child), from joint insurance (such as training leave funded by unions and employers’ associations), from firms on the basis of the continuity of an employment contract (for example, parental leave or sabbaticals), or from workers themselves, who may contribute, in whole or in part, to ‘time-accounts’ through various means (including overtime or income reductions) (Supiot 2001, p. 57; see also Schmid 2006). In this way, although their utilization embraces a broad conception of work, employment (and indeed continuity in employment) remains quite central to their operation. Furthermore, the vision for securing the reserves necessary to exercise social drawing rights on the basis of job tenure or self-funding through overtime or salary-sacrificing also make these rights, which aim to be public and collective, vulnerable to privatization and individualization. Supiot (2001) refers specifically to equal access to high-quality services in the general interest, occupational freedom, and lifelong learning. The latter process was initiated in the 2003 Directive Concerning the Status of Third-Country Nationals who are Long-Term Residents, which seeks to enlarge work/employment rights (and economic rights more broadly) as well as social, political, and civil rights among this group. As Gordon (2007, p. 562) herself acknowledges, a long-term solution would focus not just on developing new forms of immigration or entry categories but on global redistribution. According to a UFW official, the union ‘did not resign the contract with Global [Horizons] because the company was debarred from participating in the H-2A program by the US Dept. of Labor for violations that occurred prior to the company signing the [three year] agreement’ (Erik Nicholson, personal communication, 7 November 2009). It is important to distinguish between the portability and exportability of benefits: portability requires cooperation between the social security institutions

Leah F. Vosko 85

18.

19.

20.

21.

22.

23.

24.

25.

26.

of the sending and receiving country and, hence, depends on their existence in the former, whereas exportability only requires that benefits accrued in one country are payable in another country (Avato et al. 2009, p. 6). Although space does not permit elaboration here, bi- and multilateral agreements on credential recognition between sending and receiving governments also exist. These agreements principally take the form of mutual recognition agreements facilitating the portability of skills (ILO 2007; NGO Network of Integration Focal Points 2007; Phillimore et al. 2007). The strength of this legislation is, however, limited because it responds solely to the situation of a small occupational group despite calls to cover all migrants in such situations (e.g., Canadian Migration Institute 2009). It is also diluted by other factors, such as a federal-level practice of permitting border officials to turn migrant workers away upon entry to Canada if they have been given invalid offers of employment by recruitment agencies or employers, a move penalizing workers rather than recruiters or employers making the false offers of employment (CBC 2008). For example, these supports have not significantly increased men’s involvement in caring for young children. Despite the two months of non-transferable daddy leave, many fathers take no leave at all, largely on account of persistent gender income gaps. In this way, they primarily make the lives of women, particularly those cohabiting with men, easier by limiting explicit penalties for caring (Nyberg 2004). Although entitlements from the career break/time credit system may be used for study, they are also distinct from paid educational leaves, which provide more generous support as the worker essentially maintains his/her salary (up to a high ceiling). The time credit is a general right, although employees working in small firms (i.e., with fewer than 10 employees) are excluded (De Lathouwer et al. 2005, p. 12). A further limitation is that in the case of the private sector career break scheme, to access a one-fifth reduction in working time, the employee must have been employed at least five-years with the same employer upon application and be engaged full-time for the previous 12 months (De Lathouwer et al. 2005, pp. 12–13). In 2008, Vandeweyer and Glorieux reported that employees below the age of fifty in the public sector received between a300 and a400 per month for a full-time break, and in the private sector up to a500 (276). Under its terms, bereaved workers received ‘three days paid bereavement leave, paid roundtrip transportation to their country of origin and further leave time if necessary’ (Global Horizons–UFW 2006). An international tax organization was proposed by a High-Level Panel on Financing for Development (Zedillo et al. 2001, v and 15) in the lead up to the Monterrey conference on Financing for Development. However, the proposal failed to receive the endorsement of the Preparatory Committee of the Financing for Development Conference. This idea nevertheless remains in circulation; for example, recommendations made by NGOs in preparation for the Doha development conference in 2008 contained a call for such an organization (Action AID et al. 2008). I am grateful to Lourdes Beneria for explaining to me how an organization of this sort had been proposed.

86 Regulating for Decent Work

References Action AID et al. 2008. ‘Civil society key recommendations for the Doha draft outcome document’, prepared for Doha Preparatory Process on Financing for Development, 30 August (unpublished brief). Anxo, D.; Erhel, C.; Schippers, J. 2006. Labour Market Transitions and Time Adjustment over the Life Course (Amsterdam: Dutch University Press). Auer, P. 2006. ‘Protected mobility for employment and decent work: labour market security in a globalized world’, Journal of Industrial Relations, vol. 48, no. 1, pp. 21–40. Auer, P. 2007. In Search of Optimal Labour Market Institutions (Geneva: ILO). Online: http://www.oit.org/public/english/employment/download/elm/elm07–3.pdf. Avato, J.; Koettl, J.; Wheeler, R.S. 2009. ‘Definitions, good practices, and global estimates on the status of social protection for international migrants’, SP Discussion Paper no. 0909 (Washington, DC: Social Protection and Labor Division, World Bank). Bach, S. 2003. ‘International migration of health workers: labour and social issues’, Working Paper 209 (Geneva: ILO). Online: http://www.ilo.org/global/ About_the_ILO/Media_and_public_information/Feature_stories/lang--en/ WCMS_075617/index.htm. Beneria, L. 2008. ‘The crisis of care, international migration, and public policy’, Feminist Economics, vol. 14, no. 3, pp. 1–21. Björnberg, U. 2002. ‘Ideology and choice between work and care: Swedish family policy for working parents’, Critical Social Policy, vol. 22, no. 1, pp. 33–52. Bosch, G. 2004. ‘Towards a new standard employment relationship in Western Europe’, British Journal of Industrial Relations, vol. 42, no. 2, pp. 617–36. Bosniak, L.S. 2002. The Citizen and the Alien: Dilemmas of Contemporary Membership (Princeton, NJ: Princeton University Press). Brodie, J. 2002. ‘Three stories of Canadian citizenship’, in Adamoski, R.; Chunn, D.E.; Menzies, R. (eds) Contesting Canadian Citizenship: Historical Readings (Toronto: Broadview Press), pp. 43–66. Campbell, I. 1998. ‘Labour restructuring and the growth of casual employment in Australia, 1982–1996’, Doctoral Thesis, Melbourne, University of Melbourne, Australia. Canada. 2006. Fairness at Work: Federal Labour Standards for the 21st Century, prepared by Harry W. Arthurs, Ottawa, Human Resources and Skills Development Canada. CBC. 2008. ‘Mexicans denied entry at Vancouver airport, returned home’, 25 June. Online: http://www.cbc.ca/canada/british-columbia/story/2008/06/24/ bc-migrants-turned-back.html. Cranford, C.J.; Fudge, J.; Tucker, E.; Vosko, L.F. 2005. Self-Employed Workers Organize: Law, Policy, and Unions (Montreal/Kingston: McGill-Queen’s University Press). Deakin, S. 2002. ‘The futures of the contract of employment’, in Conaghan, J.; Fischl, M.; Klare, K. (eds) Labour Law in an Era of Globalization: Transformative Practices (Oxford: Oxford University Press), pp. 117–81. Delanty, G. 1997. ‘Models of citizenship: defining European identity and citizenship’, Citizenship Studies, vol. 1, no. 3, pp. 285–303. De Lathouwer, L.; Debacker, M.; Bogaerts, K. 2007. ‘Care in modern welfare states and social policies: the case of Belgian time credit’, in Anxo, D.; Erhel, C.;

Leah F. Vosko 87 Schippers, J. (eds) Labour Market Transitions and Time Adjustment Over the Life Course (Amsterdam: Dutch University Press), pp. 165–86. Department of Labour and Employment, The Philippines and Government of Manitoba, Canada. Available online at http://www.dole.gov.ph/. Devisscher, S.; Sanders, D. 2007. ‘Ageing and life-course issues: the case of the career break scheme (Belgium) and the life-course regulation (Netherlands)’, in D’Addio, C.; Whiteford, P. (eds) Modernizing Social Policy for the New Life Course (Paris: OECD Publishing), pp. 117–32. Esping-Andersen, G. 2002. Why We Need a New Welfare State (Oxford: Oxford University Press). European Commission (EC). 2004. Second Career: Overcoming the Obstacles Faced by Dependent Employees Who Want to Become Self-employed and/or Start Their Own Business (Brussels: EC Enterprise-Directorate). EUROSTAT. 2006. European Union Labour Force Survey (EU LFS), 1983–2006 (Luxembourg). Fraser, N. 1997. Justice Interruptus: Critical Reflections on the ‘Postsocialist’ Condition (New York: Routledge). Fudge, J.; Tucker, E.; Vosko, L.F. 2002. The Legal Concept of Employment: Marginalizing Workers (Ottawa: Law Commission of Canada). Fudge, J.; Vosko, L.F. 2001. ‘Gender, segmentation and the standard employment relationship in Canadian labour law and policy’, Economic and Industrial Democracy, vol. 22, no. 2, pp. 271–310. Fuller, S.; Vosko, L.F. 2008. ‘Temporary employment and social inequality in Canada: exploring intersections of gender, race and migration’, Social Indicators Research, vol. 88, no. 1, August, pp. 31–50. Girard, L.; Chambers, I. 2005. ‘Naples Italy: a spontaneous model for integration’, in Balbo, M. (ed.) International Migrants and the City (Venice: UN-Habitat), pp. 187–210. Global Horizons–UFW. 2006. Global Horizons–UFW Contract Summary (unpublished collective agreement). Gordon, J. 2007. ‘Transnational labor citizenship’, Southern California Law Review, vol. 80, no. 3, pp. 503–87. Greenhouse, S. 2006. ‘Farmworkers’ union is set to announce first national contract for guest workers’, New York Times, 11 April. Gregoire, D. (ed.) 2006. Newsletter of the Health and Safety Department of ETUIREHS, no. 30–1, October (Belgium: European Trade Union Institute – Research, Education, Health & Safety, HESA). Hall, S.; Held, D. 1990. ‘Citizens and citizenship’, in Hall, S.; Jacques, M. (eds) New Times: The Changing Face of Politics in the 1990s (London: Verso), pp. 173–88. Hammar, T. 1985. European Immigration Policy (Cambridge: Cambridge University Press). Hartlapp, M.; Schmid, G. 2008. ‘Labour market policy for “active ageing” in Europe: expanding the options for retirement transitions’, Journal of Social Policy, vol. 37, no. 3, pp. 409–31. Hassim, S. 2008. ‘Global constraints on gender equality in care work’, Politics & Society, vol. 36, no. 3, pp. 388–402. Heikkilä, M. 1999. ‘A brief introduction to the topic’, in European Foundation for the Improvement of Living and Working Conditions, Linking Welfare and Work (Luxembourg: Office for Official Publications of the European Communities).

88 Regulating for Decent Work Hochschild, A. 2000. ‘Global care chains and emotional surplus value’, in Hutton, W.; Giddens, A. (eds) On the Edge: Living with Global Capitalism (London: Jonathan Cape), pp. 130–46. Holston, J. 2001. ‘Urban citizenship and globalization’, in: Scott, A. J. (ed) Global City-Regions: Trends, Theory, Policy (New York: Oxford University Press), pp. 325–48. Holzmann, R.; Koettl, J.; Chernetsky, T. 2005. Portability Regimes of Pension and Health Care Benefits for International Migrants: An Analysis of Issues and Good Practices, September (Washington, DC: Social Protection Human Development Network, World Bank). Howard, J. 2004. A Stronger Economy, A Stronger Australia: The Howard Government Election 2004 Policy. Protecting and Supporting Independent Contractors (Barton: The Liberal Party of Australia). International Labour Office (ILO). 1984a. Evaluation of the International Programme for the Improvement of Working Conditions and Environment, Report VII, International Labour Conference, 70th Session (Geneva). ILO. 1984b. Record of Proceedings, International Labour Conference, 70th Session (Geneva). ILO. 2010. Decent Work for Domestic Workers. Report IV(1), International Labour Conference, 99th Session (Geneva). Lin, Z.; Yates, J.; Picot, G. 1999. ‘Rising self-employment in the midst of high unemployment: an empirical analysis of recent development in Canada’, Ottawa, Statistics Canada Research Paper Series (Ottawa: Statistics Canada). Lister, R. 1997. Citizenship: Feminist Perspectives (Basingstoke: Macmillan). Lister, R. 2007. ‘Inclusive citizenship: realizing the potential’, Citizenship Studies, vol. 11, no. 1, pp. 49–61. MacDonald, M. 2009. ‘Income security for women: what about employment insurance?’, in Cohen, M.G.; Pulkingham, J. (eds) Public Policy for Women in Canada: The State, Income Security and Labour Market Issue (Toronto: University of Toronto Press), pp. 251–70. MacLaren, B.; Lapointe, L. 2009. ‘Making a case for reform: non-access to social security measure for migrant workers’, Policy Paper, FOCAL, Ottawa, October. Malkki, L.H. 1995. Purity and Exile: Violence, Memory, and National Cosmology Among Hutu Refugees in Tanzania (Chicago: University of Chicago Press). Massey, D. 2004. ‘Geographies of responsibility’, Geografisca Annaler, vol. 86 B, no. 1, pp. 5–18. McCann, D. 2008. Regulating Flexible Work (Oxford: Oxford University Press). Mückenberger, U. 1989. ‘Non-standard forms of employment in the Federal Republic of Germany: the role and effectiveness of the state’, in Rogers, G.; Rogers, J. (eds), Precarious Jobs in Labour Market Regulation: The Growth of Atypical Employment in Western Europe (Geneva: International Institute for Labour Studies/Free University of Brussels), pp. 167–86. Murray, J. 2001. Transnational Labour Regulation: The ILO and the EC Compared (The Hague: Kluwer). NGO Network of Integration Focal Points. 2007. ‘Policy briefing on the assessment of skills and recognition of qualifications of refugees and migrants in Europe’. Online: http://www.ecre.org/files/Policy%20Briefing_Assessment%20of%20 skills%20&%20recongition%20of%20qualifications.pdf.

Leah F. Vosko 89 Nullis-Kapp, C. 2005. ‘Efforts under way to stem “brain drain” of doctors and nurses’, Bulletin of World Health Organization, vol. 83, no. 2, pp. 84–5. Nyberg, A. 2004. ‘Parental leave, public childcare and the dual earner/dual carermodel in Sweden’. Discussion paper, Swedish National Institute for Working Life. Online: http://www.mutual-learning-employment.net/pdf/sweden04/ disspapSWE04.pdf. O’Donnell, A. 2004. ‘“Non-standard” workers in Australia: counts and controversies’, Australian Journal of Labour Law, vol. 17, no. 1, pp. 1–28. O’Reilly, J.; Spee, C. 1998. ‘The future regulation of work and welfare: time for a revised social and gender contract?’, European Journal of Industrial Relations, vol. 4, no. 3, pp. 259–81. Owens, R. 2001. ‘The “long-term or permanent casual”: an oxymoron or a “well enough understood Australianism” in the law’, Australian Bulletin of Labour, vol. 27, no. 2, pp. 118–36. Pearson, R. 2004. ‘The social is political: towards the re-politicization of feminist analysis of the global economy’, International Feminist Journal of Politics, vol. 6, no. 4, pp. 603–22. Pearson, R. 2003. ‘Feminist responses to economic globalisation: some examples of past and future practice’, Gender and Development, vol. 11, no. 1, pp. 25–34. Phillimore, J.; Goodsen, L.; Watts, J. 2007. ‘Recognising migrants’ skills and qualifications: a transnational perspective’. Final report of Working Group 1 of the MEET EQUAL Transnational Partnership, May, University of Birmingham. Pocock, B.; Buchanan, J.; Campbell, I. 2004. Securing Quality Employment: Policy Options for Casual and Part-time Workers in Australia (Adelaide: Chifley Research Centre). Raghuram, P.; Madge, C.; Noxolo, P. 2009. ‘Rethinking responsibility and care for a postcolonial world’, Geoforum, vol. 40, no. 1, pp. 5–13. Rodgers, G. 1989. ‘Precarious work in Western Europe: the state of the debate’, in Rodgers, G.; Rodgers, J. (eds) Precarious Jobs in Labour Market Regulation: The Growth of Atypical Employment in Western Europe (Geneva: ILO), pp. 1–16. Rubery, J. 1998. Women in the Labour Market: A Gender Equality Perspective (Paris: OECD Directorate for Education, Employment, Labor & Social Affairs). Sainsbury, D. 2006. ‘Immigrants’ social rights in comparative perspective: welfare regimes, forms of immigration and immigration policy regimes’, Journal of European Social Policy, vol. 16, no. 3, pp. 229–44. Schmid, G. 2002. ‘Employment insurance for managing critical transitions during the life cycle’, in Auer, P.; Gazier, B. (eds) The Future of Work, Employment and Social Protection, Proceedings of the France ILO Symposium 2002 (Geneva: ILO), pp. 63–82. Schmid, G. 2005. ‘Towards a European social model: managing social risks through transitional labour markets’, in Gijsel, P.; Schenk, H. (eds) Multidisciplinary Economics (The Netherlands: Springer), pp. 33–51. Schmid, G. 2006. ‘Social risk management through transitional labour markets’, Socio-Economic Review, vol. 4, pp. 1–33. Schmid, G. 2007. ‘Transitional labour markets: managing social risks over the life course’, WZB Discussion Paper, SP I 2007-111 (Berlin: Social Science Research Centre). Soysal, Y.N. 1994. Limits of Citizenship: Migrants and Postnational Membership in Europe (Chicago: Chicago University Press).

90 Regulating for Decent Work Stasiulis, D.K.; Bakan, A.B. 2005. Negotiating Citizenship: Migrant Women in Canada and the Global System (Toronto: University of Toronto Press). Statistics Canada. 2006. Survey of Labour and Income Dynamics (Ottawa). Statistics Canada. 2008. Labour Force Survey (Ottawa). Stewart, A. 1992. ‘A typical employment and the failure of labour law’, Australian Bulletin of Labour, vol. 18, pp. 217–35. Supiot, A. 2001. Beyond Employment: Changes in Work and the Future of Labour Law in Europe (Oxford: Oxford University Press). Van Ginneken, W. 2009. ‘Social security and the global socio-economic floor’, Global Social Policy, vol. 9, no. 2, pp. 228–45. Vandeweyer, J.; Glorieux, I. 2008. ‘Men taking up career leave: an opportunity for a better work and family life balance?’, Journal of Social Policy, vol. 37, no. 2, pp. 271–94. Vosko, L.F. 2008. ‘ILO action on “the scope of the employment relationship”: lessons from Canada on fostering social cohesion’, in O’Brien, R. (ed.) Solidarity First: Canadian Workers and Social Cohesion (Vancouver: UBC Press), pp. 169–89. Vosko, L.F. 2009. ‘Less than adequate: regulating temporary agency work in the EU in the face of an internal market in services’, Cambridge Journal of Regions, Economy, and Society, vol. 2, no. 3, pp. 1–17. Vosko, L.F. 2010. Managing the Margins: Gender, Citizenship, and the International Regulation of Precarious Employment (Oxford: Oxford University Press). Weir, A. 2005. ‘The global universal caregiver: imagining women’s liberation in the new millennium’, Constellations, vol. 12, no. 3, pp. 308–30. Wikander, U.; Kessler-Harris, A.; Lewis, J. (eds) 1995. Protecting Women: Labor Legislation in Europe, United States and Australia, 1880–1920 (Urbana, IL: University of Illinois Press). Williams, F. 2008. ‘Theorising migration and home-based care in European welfare states’. Paper presented at the annual conference of the Canadian Political Science Association, 4–6 June, University of British Columbia in Vancouver; Session 5. World Health Organization (WHO) 2004. International Migration of Health Personnel: A Challenge for Health Systems in Developing Countries, World Health Assembly Fifty Seven, Resolution 19, May (Geneva). WHO. 2006. Working Together for Health: The World Health Report 2006 (Geneva). Yeates, N. 2005. ‘The general agreement on trade in services (GATS): what’s in it for social security?’, International Social Security Review, vol. 58, no. 1, pp. 3–22. Zedillo, E.; Al-Hamad, A.Y.; Bryer, D.; Chinery-Hesse, M.; Delors, J.; Grynspan, R.; Livshits, A.; Osman, A.M.; Rubin, R.; Singh, M.; Masayoshi, S. 2001. Recommendations of the High-level Panel on Financing for Development, UN General Assembly document A/55/1000 (New York: United Nations).

4 The Growth of Extended ‘Entry Tournaments’ and the Decline of Institutionalized Occupational Labour Markets in Britain* David Marsden

Introduction A large number of labour markets are characterized by competitive entry. Young workers compete for entry into those firms that provide the best career opportunities, and they do so to become established in particular occupations. When labour markets are characterized by strong institutions, such competition is often concentrated at certain stages in workers’ careers; if they fail to get in, they move on to jobs in less good firms or less prestigious occupations. This chapter explores the dynamics of labour markets in which competition for entry has become prolonged over a considerable period of a person’s working life, and given rise to extended entry tournaments. This can occur because the institutions that previously regulated entry have declined or, in the case of new occupations, may never have existed. Although it will be argued that such tournaments are not a general feature for all occupations, they are of interest because they appear to have developed in line with a decline in other more familiar labour market institutions in recent decades. These extended entry tournaments may enrich our understanding of the increased income inequality within certain

* Earlier versions of this chapter have been presented at the Society for the Advancement of Socio-Economics (SASE), ILO/RDW and the ESRC/SKOPE conferences and I should like to thank participants for their advice and comments, especially Sangheon Lee, Ken Mayhew, François Michon, and Héloïse Petit. I should also like to thank Richard Belfield for making available his matching of the old and new occupational classifications for the NES. The current chapter has revised and extended parts of the analysis that appeared in Marsden (2007). A fuller statistical analysis is available in Marsden (2010). 91

92

Regulating for Decent Work

occupations. For example, Salverda and Mayhew (2009) argue that the decline of inclusive labour market institutions in many countries has contributed to increased pay inequality at the low-pay end. Likewise, Goos and Manning (2007), writing about job polarization in Britain, suggest that the decline of protective labour institutions may explain why the increased employer demand for poor-quality, ‘lousy’ jobs relative to those in the middle has not improved their relative pay. In conclusion, this chapter argues that these developments pose new challenges for labour market regulation. The background to this study is provided by the spread of more flexible forms of economic organization, and the expansion of project-based working which provide unfavourable ground for the traditional models of cost-sharing for training between employer and trainee. The theme was taken up by Osterman (1996) in his edited volume, Broken Ladders, on the decline of large organization internal labour markets in the US, and by Arthur and Rousseau (1996) in theirs on the ‘Boundaryless Firm’. Within the EU, the Supiot Report (1999) addressed a similar theme, and recommended reform of the employment relationship to cater for more transient and fluid employment relationships than in the past few decades, a theme taken up also in the discussion of ‘flexicurity’ (Wilthagen et al. 2007). These represented changes to a pattern that held sway up to the mid1980s. At that time, it was possible to characterize the labour markets for industrial skills in the leading industrial economies as falling into two broad categories: firm-based internal labour markets (ILMs) and occupational labour markets (OLMs). Whereas firms in the US, Japan and France were heavy users of the former kind, those in Britain, like their counterparts in Germany, were heavy users of occupational markets (Maurice et al. 1982; Eyraud et al. 1990; Sengenberger 1987; Jürgens 2003; Marsden 1990). The years up to the late 1970s, roughly speaking, were dominated by what many have referred to as the industrial or ‘Fordist’ model of production, grouped around large production units, often with strongly institutionalized patterns of labour–management relations. By the early 2000s, the industrial labour markets in these countries had undergone profound transformations, and particularly in Britain, where industrial employment slumped, and new models were developing the rapidly expanding services sectors. Whereas the internal and occupational labour markets of the 1970s were characterized by a high degree of institutional regulation, it is not clear that the organization of entry and progression within some of the growing occupations of the service and the ‘knowledge’ economy conform to these models.

David Marsden 93

In contrast to the relatively structured and well-defined ports of entry for established internal and occupational labour markets of the industrial sector, those in many services activities appear to be more open, with more intense and prolonged competition for entry. This chapter argues that ‘entry tournaments’ have flourished in some of these occupations, growing at the expense of more structured entry channels. Their growth is associated in part with the rapid growth of top pay, which serves to attract entry candidates, ‘aspirants’, but it is also connected with the deterioration of conditions for the lowest paid. The growth of project-based employment, with its shorter time commitments, has played a key part in opening up the competition to new aspirants. These ideas are explored by comparing the situation in service occupations in which such changes have been prevalent with those in which more established regulatory structures have been sustained.

The growth of project-based and transient employment relationships Project-based employment is not the norm across the whole economy, as is demonstrated by the data on the stability of job tenure (Auer and Cazes 2000). Nevertheless, it is a method of working that has grown in some sectors and has reinforced its presence in others where it was already established. It has attracted a good deal of attention in recent years because of the pressure on firms to respond more quickly to market changes, and because of the fluidity of projects as an organizational form. The development of this kind of working has been well documented in the media and IT sectors, and, more recently, in the higher education sector (for example, Jones and Walsh 1997; Saxenian 1996; Tolbert 1996; Heery et al. 2004). In the creative and intellectual sectors project-based working has many attractions. As Baumann (2002 and 2003) has observed, the key added value arises in such cases at the stage of conception and design whereas production, and copying, is often relatively cheap. Many of the goods and services are akin to fashion goods, changing rapidly, and with a premium on novelty. In contrast to mass production, in the media, IT and research sectors, there is a predominance of one-off or small batch production, so there are few economies of scale in employment. The lack of economies of scale in employment has profound implications for the institutionalization of training and labour market entry. Internal labour markets required the presence of a sufficiently large workforce so that employers may plan career progression over a reasonable

94

Regulating for Decent Work

period. Likewise, occupational labour markets have usually required a significant input from employers to fund the provision of training for transferable skills. This also requires the organization to display a degree of scale and permanency. Yet the pressures for greater organizational flexibility tend to militate against such organizational patterns, and so provide a less favourable environment for institutionalized patterns of labour market and occupational entry. There are two further reasons for searching within these areas of employment for new patterns of labour market practices. First, the predominant occupational pattern described for the 1970s and 1980s was based on industrial employment, a sector that has now diminished substantially in Britain, as in a number of other countries. In Britain, in the 1970s, it still accounted for about 40 per cent of employment, but by the early 2000s, the figure had dropped to less than 15 per cent, and was still declining. It is therefore unlikely to provide indicators of new models of labour market institutions. Second, the growth of secondary and tertiary education since the 1970s means that employers can recruit people with a much higher level of general education, and, in parallel, better developed skills for self-directed learning. This also affects the supply of workers for such occupations, as it is less constrained by employers’ investments in training than in the more institutionalized internal and occupational labour markets.

Growing use of tournaments to regulate labour market and career transitions The tournament metaphor applied to labour markets comprises a number of key observations which can be drawn from the literature (for example, Lazear and Rosen 1981; Aoki 1988; Lazear 1998; Bognano 2001). It emphasizes the generally high level of competition among workers of similar status for access to higher-status positions. These higher-status positions should be fixed in number, or, at least, the supply of them should be relatively inelastic with regard to the number of aspirants so that there is a degree of rationing of these good positions. Aspirants gain access to the higher-status positions based on their relative merit, a consequence of the limited supply of good positions. The reward structure should be such as to attract a large enough pool of aspirants to maintain the intensity of competition, but also to compensate them for the higher risk of failure than in more structured labour market arrangements. One must consider also the consequences of failure to gain access to the high-status positions. Sometimes former

David Marsden 95

aspirants can move into other occupations, but when these have structured entry paths, they may find they have left it too late. Hence, it can be anticipated that many will gradually slide into low-status positions within their preferred occupation. The internal and occupational labour markets that held sway for many types of work in the 1960s and 1970s regulated a number of key labour market transitions. They regulated entry paths into good jobs: the ‘entry ports’ into firms’ internal labour markets for many blueand white-collar jobs; and through apprenticeship-type arrangements for access to skilled and professional occupations. In contrast to the tournament-oriented model, these structures serve to adjust the supply of entry positions to roughly that of vacancies within the occupation or work area. In the case of ILMs, Becker (1975) argued that firms will usually bear most of the cost of training because the skills are not transferable, and that in order to minimize costs, they will seek to equate the flow of new entrants to their expected future needs, and organize job progression so that skills are accumulated in an orderly fashion. Likewise, in occupational markets, although Becker’s theory predicts that trainees should bear the cost of training because their skills are transferable, in practice, firms often share a substantial part of the cost. They can afford to do this if the skills can be made sufficiently ‘sticky’ – for example, if they are supplemented by firm-specific skills, or there are incentives for staying (see Acemoglu and Pitschke 1999; Stevens 1994). They can also establish regulatory institutions to share the costs among employers and discourage free-riding. Thus, in this case too, employers have an incentive to restrict the number of training and entry positions, consequently limiting competition among aspirant members of the occupation. Structured internal and occupational labour markets provide stable frameworks for investment by both parties. They do this by introducing a degree of closure in the sense described by Clark Kerr (1954) in his account of access to ‘industrial’, or firm-specific, and ‘craft’, or occupational markets. In both cases, closure was associated with a form of regulated transition for new entrants: by recognized ports of entry into firm internal labour markets, and by occupation-specific training positions for occupational markets. In relation to the tournament metaphor sketched above, these structures, by putting a limit on training places, also limit the number of aspirants competing for entry, and they often also compress the time period over which it prevails. Both methods of organizing skills and labour markets came under pressure during the 1980s and 1990s. Firm internal labour markets

96

Regulating for Decent Work

involve a high overhead cost, and the skills in such markets adapt only slowly to changing circumstances. Established occupational markets also take time to adapt because of the large numbers of stakeholders who have to agree the new training norms: for example, the firms that provide training places and hire from such markets, and the workers who invest in occupational skills. Changing technology has put pressure on both forms, as has the increased weight of national and international competition. This has made it more risky for firms both to invest in building up and maintaining their internal labour markets, and to commit on a long-term basis to the skill structures of occupational markets. If one strips away these frameworks, it becomes possible for increased numbers of aspirants to compete for entry into the occupation, and to do so over a longer period of time. During this period, they need to obtain work in order to acquire the necessary skills and develop their network contacts and reputation. The fluidity of employment relationships reduces the incentive for employers to provide systematic training, and so reduces the brakes on excess supply that characterize the more structured labour markets. Thus, whereas organized firm-ILMs, apprenticeships, and established professional training generate a narrow range of entry paths, these more open competitions for occupational entry will be associated with a diversity of career tracks, such as those illustrated in Figure 4.1. Figure 4.1 provides a stylized representation of the entry process into an occupation where there is little pre-established structure in the entry

Occupational threshold

Occupational status

‘Fast track’

Job assignments

‘Slow track’ Time

Figure 4.1 Career tracks for an unstructured occupational entry for a given cohort

David Marsden 97

segment, and aspirants have to find work assignments, represented by the small boxes, that bring them relevant experience, network contacts and help them build up their career portfolios and reputations. The three tracks illustrate different ways in which a given cohort of aspirants may progress, or not, towards gaining entry to the established segment of the occupation. Unlike in organizational careers where jobs are often organized into different career tracks which may be known in broad terms ex ante, in this example, there are few such signposts. Workers know about their previous and present jobs or assignments ex post, but can only surmise where they are leading. The figure illustrates the idea that the tracks are not linearly upward, but contain considerable upwards and downwards variation so that there is always a good deal of uncertainty as to which track one is on, and, therefore, about one’s likely success in gaining access to the established part of an occupation, denoted by crossing the dotted line in Figure 4.1. This may occur by passing qualifications, but it may also be expressed by means of peer group recognition. A notable feature of entry in this example is that it remains open for a prolonged period of a person’s working life. This contrasts with highly structured occupational and internal labour markets where there is usually a clear point at which the aspirant is either in or out. In these more open systems, there is also a risk that aspirants continue to compete for entry long after their opportunities in other occupations have started to close down. Thus, there is a likelihood that some will become trapped in the ‘slow stream’, and have to make do with precarious employment conditions and low pay because it has become too late to enter more structured occupations. Given that some occupations provide institutionalized and predictable channels, why should aspirant employees compete to enter these occupations which offer only unstructured entry transitions? In particular, why should anyone take the risk of becoming stuck on the slow track? In some creative occupations, hedonic motivation can provide a strong pull: the desire to produce ‘art for art’s sake’. High financial rewards may also provide a risk premium to compensate for the increased risk of failure to gain entry. As will be seen shortly, the rewards at the highest levels of some of these organizations have risen considerably in recent years. Those same rewards may also compensate them for the cost of acquiring skills, experience and contacts. The looseness of institutional structures regulating entry may also contribute to aspirants becoming trapped. First, some aspirants may simply be dazzled by the prizes on offer, and, like gamblers, will keep on placing more bets in the hope that one day their luck will change,

98

Regulating for Decent Work

and they will get that elusive break. Secondly, because the tracks shown do not correspond to well-organized career paths, it is difficult for aspirants to know ex ante the track on which they currently find themselves. These tracks are observed ex post. The random element reflected in the fluctuating occupational status of successive assignments on all tracks makes reliable identification more difficult. For example, does the fourth assignment on the slow track mean that one is actually on the middle track, or is it just a piece of temporary good fortune? This means that by the time one realizes one is condemned to life on the slow track, it may be too late to move to other occupations. One is trapped. A third factor contributing to people being locked in relates to the size of the stake associated with each assignment. Because one moves through many assignments rather than entering upon a single training programme, the financial stake at risk at each step is correspondingly smaller. Yet the eventual prize is unchanged. Thus, if aspirants write off their past investments, and focus only on the current one, then many will persist in taking on new assignments even though a hard look at their past experience should tell them they are on the slow track. This process would also tend to encourage aspirants to continue to seek assignments to build up their portfolios until they find themselves trapped. The final piece of the jigsaw concerns the advantages for firms. They benefit from their reduced training costs when employees bear more of the cost of gaining experience, and they also benefit from employees’ apparent willingness to remain in the occupation’s outer orbit long after there has been a decline in their prospects of entry. Such savings are likely to offset the costs of paying for the small minority at the top. Such advantages for employers might apply across many occupations, but there is a further incentive advantage, related to the quality of performance, and which makes such tournaments more appropriate to some occupations than others. This is the subject of the next section. The potential client benefits of tournaments in the less structured occupations There is a particular problem of quality assurance in occupations in which more transient project-based employment has flourished. By the time quality problems have become apparent, the project may have been long completed and the team members dispersed. This contrasts with organizations that provide long-term employment, in which careerpay and promotion provide incentives for long-term performance and for good-quality work when such quality may not be immediately apparent to the employer. It also contrasts with long-established

David Marsden 99

professional work in well-structured occupations. They often have their own licensing arrangements and disciplinary bodies, and their work is also often integrated into long-term positions in organizations – for example, the medical professionals employed by the National Health Service. In the more transient employment relationships of project-based work, these mechanisms are often weak or absent, meaning that other incentive devices are needed. A notable feature of projectbased work is that it involves intensive work at the stage of conception, but its reproduction is often low-cost and involves the work of other categories of employees (Baumann 2003). A piece of investigative journalism, an advertising video, a research project or a new piece of software all require major initial investments, but in contrast to manufactured goods or care services, reproduction and distribution are relatively cheap, and the services of those who carried out the project are no longer required. In a nutshell, the argument is that when a greater role for projectbased employment is combined with serious short-term concerns about the quality of performance, then employers can benefit by restricting the number of key positions, and paying them a higher price, even though there may be many other equally competent people available to fill them. Apart from competition to enter, restricting the number and raising the reward increases the penalty for individuals who let their performance slip once in one of the top jobs. In effect, the top jobs become akin to ‘positional goods’, part of whose benefit derives from their restricted supply (Hirsch 1977). Journalism, the media, academia, software and consulting are interesting areas because they provide illustrations of occupations in which project-based employment and entry tournaments have come to play an increased role, compared with other more structured occupations such as engineering, law enforcement, and health care professions. Although all of these are knowledge-based occupations, they differ in that the first group emphasizes new knowledge and new solutions, whereas the second places greater emphasis on the application of existing knowledge. As Baumann (2003) has argued, the first kind requires more flexible employment structures. The top ‘prizes’ in the first kind of occupations are restricted and often highly visible. The Sutton Trust’s (2006) report on journalism highlighted the growing polarization within that occupation. Top journalists can receive very high rates of pay, and the older institutionalized entry paths have been allowed to decay. The positional good dimension of top jobs can be explained as follows. All journalists depend upon sources for their stories, and news which is regularly of

100

Regulating for Decent Work

great interest, such as in politics, depends upon access to a small number of key sources. Senior politicians and government ministers have their circles of trusted journalists to whom they may provide important but unattributable information, the infamous ‘sources close to the Prime Minister’, and confidential briefings so that journalists have the necessary background to interpret statements and events (The Guardian, 19 June 2006). Such confidences depend upon knowing whom one can trust, and the need for trust restricts the number of people who can enter into these relationships. Indeed, a degree of intimacy is not only a product of the need for trust, but also provides a potential sanction. Journalists who betray that trust know that they are likely to be excluded from the circle. At the opposite end, the crowding effect at entry was highlighted by the Sutton Trust, which noted that, apart from a limited number of graduate internships: ‘The only openings on national dailies are provided by freelance shifts, where money is poor and there is no job security. A lucky few will secure that coveted permanent post’ (Sutton Trust 2006, p. 12). It also observed that similar conditions applied to entry into television journalism. The same report also noted that in earlier decades, newspapers and the broadcast media had provided more systematic entry routes into the occupation, and that it had been possible for a former generation of newspaper editors to rise from non-journalist positions on their newspapers to prestigious editorships (The Guardian, 19 June 2006). Client organizations may also benefit from the low pay and flexible employment of the aspirant members as this releases resources which can contribute, in part, to the funding of higher pay at the top. In the media sector there are also mechanisms that tend to focus potential rewards on a small number of key activities. Although new technology makes it easy to post work directly onto the internet, through sites such as ‘YouTube’, getting a creative work to the right audience, and one that is large enough to cover production costs, is another matter. There are necessarily only a very small number of major creative works of a kind that can gain a wide audience and generate large revenues so the recruitment process for top artists is critical. Of course, many other artists can earn a living in less high-profile activities – for example, in smaller specialist films, or in advertising videos – but their incomes will be closer to the average. High rewards also bring sanctions: a string of disappointing performances will force the former ‘star’ to return to secondary roles. The British Film Institute has conducted a number of surveys on employment and careers in the

David Marsden 101

British film industry. These highlight the precarious employment of new entrants, and the difficult transition to becoming a permanent and recognized member of the sector. It also highlights, in common with journalism, the very low starting rates of pay, and the frequent interruptions to income in between projects (Pettigrew et al. 1997). Similar processes of occupational entry have been observed in the US, in Hollywood, by Jones and Walsh (1997). In the academic world, where career paths were once highly institutionalized, the emergence of similar processes can also be observed, albeit on a more limited scale. For example, the funding of major research projects has become increasingly focused on an elite group of research universities. Lam (2003 and 2005) highlights how major companies that want to conduct joint research with universities target ‘star’ scientists. Such targeting serves a number of functions. The companies know that the research universities themselves provide a degree of peer group quality control through selection and promotion decisions at which they would not be competent. The sums of money are large enough for the scientists to maintain teams of researchers to work on key areas of commercially relevant basic research, but by their very size, they also represent a concentration of resources. Finally, as in the journalist example, concentration provides sanctions: competitor universities and departments can undertake the work if the chosen team fails to deliver, or is careless with commercially sensitive information. The ‘prizes’ for top academic scientists may be financially less attractive than in many other activities, but these are accompanied by alternative incentives such as scientific prestige and the provision of future research resources. A similar process may be at work in university teaching with the growth of a global market for international students. Leading universities attract international students by having famous professors, or at least individuals who are well known in the universities sending students for further study. Ranking of universities in international league tables also tends to highlight the most successful, raising their profile, and enabling them to attract the most able students. Competition for entry means that whatever the quality of the education actually provided, students’ success in gaining entry is a key labour market signal for their future careers. Alumni networks may reinforce this, particularly if they play an active role in job markets. The Parisian Grandes Écoles provide a similar illustration in which reputation enables these schools to cream off the best applicants. In such cases, the fact that potential students and their families can only digest limited information about

102

Regulating for Decent Work

university quality favours the small group with the highest reputation. It therefore becomes essential for individual universities to remain among the leading group if they are to continue to succeed in recruitment, and this puts pressure on the universities to move towards a ‘star system’ for recruitment. At the opposite end of the academic labour market, in Britain, many universities, and particularly the major research universities, now have very high levels of temporary employment among their academic staff – two-thirds of those at Cambridge and Oxford, and just over half at LSE – figures that correlate strongly with the university research ratings.1 Many of these workers are engaged on research projects, as research assistants, postdocs, but also teaching assistants who ‘free up’ the research time of permanent faculty. However, their numbers far surpass what can be absorbed into established university careers. In the occupational areas of software and management consulting similar processes also appear to be at work. It is hard to assess the quality of creative work in advance, and client organizations rely heavily on their contractors to provide top-quality performance. If a client organization hires the services of a top software or management consultant, then it knows that its reputation is at stake if poor-quality work is provided. Thus, for sensitive or important projects, the client has good reason to opt for a supplier with a strong reputation. In such circumstances a system of enhanced rewards for top performers can help to sustain high performance. Thus, in all of these examples, even though the entry tournaments may not have emerged by design, they would appear to fulfil a positive economic function for the organizations that use them. In the less structured careers and looser organizational attachments of project-based employment, the reputations of individuals and teams for the quality of their work shape their future prospects. By focusing resources on those with the highest reputations, client organizations gain additional leverage should performance fall below expectation. Thus the top journalist who betrays the confidences of influential politicians, or the top academic scientist or the leading software engineer who mismanages a major project for a client organization will suffer a loss of reputation and thereby a possible exclusion from future collaborations. This marks a striking contrast with the sample of more structured occupations. These groups were chosen because they continue to benefit from strongly institutionalized labour markets: in the police case, from a strong internal labour market, and in cases of engineers, legal and health care professionals, from the existence of strong occupational

David Marsden 103

markets. Engineers, solicitors and health care professions have a strongly entrenched form of occupational licensing which establishes a wellcodified system of entry into these occupations. The police service also has a well-codified system of rules governing entry into its internal labour market. The partial exception to this picture is that of medical doctors. The Labour Force Survey (LFS) gives the picture for all medical practitioners, but the New Earnings Survey (NES) earnings data (see below) relate only to employees, and are largely concerned with hospital doctors, rather than general practitioners, who are self-employed. Although the occupational rules are clear in respect of the work of fully qualified doctors, a substantial proportion of those working in hospitals are ‘aspirant’ doctors who are still undergoing their training. These are the so-called ‘junior doctors’ who can be pressured into working extremely long hours because they are in a weak bargaining position prior to acquiring their full professional status. However, excess supply in this case is limited because the employer pays a large share of the direct costs of training, although it may arise temporarily from inadequate planning. It is also questionable, in this group of occupations, whether tournaments at the top would play the same kind of beneficial role outlined for those in the previous group. The health care and police services are organized to provide a good quality of service to all citizens on an equal basis. This is a sharp contrast to a drug company that seeks to engage with top scientists in order to develop new market-beating compounds, or to film companies that seek to make a blockbuster film. In these cases there are large rewards for those groups which can develop a superior product.

Employment conditions in ‘tournament’ and in ‘structured’ occupations These issues can be explored more systematically by a comparison of two groups of occupations: Group A, in which entry channels have become more open and careers have become less clearly defined; and Group B, which have by and large retained their established career structures and entry channels. Group A reflects the conditions outlined in Figure 4.1, whereas Group B represents those where structured internal and occupational labour markets have proved more resilient. One crude indicator of difference and change can be found in the evolving employment status of those in such occupations, and notably the growth in self-employment as this implies a radical change in the relationship between individual workers and the organizations hiring their services (Table 4.1). As the self-employed, almost by definition,

104

Regulating for Decent Work

Table 4.1 Changing employment status within selected occupations 1975–2000: self-employed as % of total employment in occupation 1975

1981

1990

1991

1995

2000 Q4

change 75–90/ 91–100

23.4

37.1

42.4

42.6

43.5

+/+

39.3

47.7

45.4

50.7

48.8

+/+

0.0

0.0

0.8

2.7

4.9

=/+

3.5 30.4

6.5 47.8

18.3 34.4

24.1 36.9

20.1 29.5

+/+ +/–

6.3

6.9

11.8

11.5

13.1

10.6

+/=

51.5

52.1

50.8

50.8

39.9

39.7

=/–

0.2 28.8 0.4 8.2

0.0 32.9 0.8 9.6

0.0 34.8 1.4 13.5

0.5 40.1 1.8 13.1

0.0 35.3 0.9 13.7

0.3 34.6 0.9 12.1

=/= +/– =/= +/=

Group A occupations Authors, writers & 22.7 journalists Media (excl. 28.2 journalists & sports) Higher education 0.0 academics Software engineers 1.1 Management consultants na Group B occupations Engineers (excl. software) Legal professionals (judges, barristers & solicitors) Police officers all grades Medical practitioners Nurses and midwives All occupations

Note: *1981–2000 Q4. Note change of occupational classification 1990–1991. For this reason figures are shown for both 1990 and 1991, and the column showing change shows it separately for both periods. Source: Labour Force Survey.

do not have organization-based careers, it also implies a change in the way in which the occupation is organized. Instead of promotion, people have to work on building reputation, peer recognition, and their ability to command improved conditions in the marketplace (Tolbert 1996). Another indicator of changed career practices can be found in the share of part-time employees within an occupation.2 Generally, those in parttime positions find it difficult to stay on the career ladder. In the Group A occupations, the growth of part-time employment has generally been somewhat stronger than in the Group B occupations, the large exception being nurses who are in any case faced by a rather flat career hierarchy. The practice of maintaining employees’ pay during brief absences from work has long been associated with career employment. In the mid-twentieth century, this was one of the key differences between the employment conditions of ‘hourly paid’ blue-collar workers, and the ‘weekly paid’ white-collar workers. Thus an erosion of this principle in

David Marsden 105

qualified white-collar occupations could signal a major change of status for certain groups. Although the evidence is more tentative because of the small sample numbers and year-to-year fluctuations, across the economy as a whole, while the practice of paying employees during short absences has spread overall, it decreased in some of the occupations in group A, notably among journalists, higher education academics, and also in group B, among the medical practitioners. Thus, in the Group A occupations, there appears to have been a significant shift in the nature of career employment, and, consequently, an erosion of those entry paths that used to guide aspirant members. In contrast, the occupations in Group B appear to have maintained their established internal labour markets, as in the case of the police, or their established structures for occupation and entry in the other cases.

Changing opportunity structure in the ‘tournament’ and ‘structured’ occupations A further piece of the jigsaw is provided by earnings data for employees in these occupations over the period under study. If a shift towards the entry tournament model sketched out above captures key elements of change in the occupations in Group A, then one would expect the prospect of increased earnings at the top to go hand in hand with greater uncertainty about entry at the bottom. Indeed, according to tournament theory, high potential rewards for the successful are required to encourage sufficient numbers of new aspirants to compete to enter the occupation. Thus, one would expect the earnings of the aspirants to drift behind those at the top. In contrast, in the occupations where entry is more structured, one would expect the institutions regulating entry to generate a greater degree of solidarity among members of the occupation. Although the NES provides data for employees only, it is likely that there are strong spillover effects between employee and selfemployed earnings as there is a certain amount of mobility between these two statuses. The overall pay picture for Britain between 1975 and 2001 has been one of steadily rising inequalities in pay, with those on the higher levels of pay pulling away from the rest of the employed population (Belfield 2006). The charts in Figure 4.2 summarize the changes for the individual occupations under consideration and for all employees across the whole economy. They show changes in real weekly earnings at selected percentiles, p10, p90, and p95, between 1975 and 2001, adjusted according to 2001 prices. Weekly earnings were chosen for the

106

Regulating for Decent Work Group A occupations Journalists

Media occupations

1400

1400

1200

1200

1000

1000

800

800

600

600

400

400

200

200

0

0

ye a 19 r 75 19 77 19 79 19 81 19 83 19 85 19 87 19 89 19 91 19 93 19 95 19 97 19 99

1600

ye a 19 r 75 19 77 19 79 19 81 19 83 19 85 19 87 19 89 19 91 19 93 19 95 19 97 19 99

1600

Software engineers

HE academics

1600

1600 1400 1200 1000 800 600 400 200 0

1400 1200 1000 800 600 400 200

ye ar 19 75 19 77 19 79 19 81 19 83 19 85 19 87 19 89 19 91 19 93 19 95 19 97 19 99

ye ar 19 75 19 77 19 79 19 81 19 83 19 85 19 87 19 89 19 91 19 93 19 95 19 97 19 99

0

Management consultants 1600 1400 1200 1000 800 600 400 200

ye ar 19 75 19 77 19 79 19 81 19 83 19 85 19 87 19 89 19 91 19 93 19 95 19 97 19 99

0

Figure 4.2 Growth of weekly earnings 1975–2001 at constant 2001 prices for selected percentiles and occupations Note: pnaa (pay not affected by absence); paa (including pay affected by absence; P99 shown for all occupations combined only. Percentiles: p95, p90, and p10 for those whose pay was not affected by absence. Additionally, at p10 shown also for those whose pay was affected by absence. Earnings deflated by Retail Price Index, 2001=100. Full and part-time combined. Source: New Earnings Survey 1975–2001.

charts because they give a better reflection of the earnings attached to the jobs on offer in these occupations, and notably capture important elements of the pay associated with the kind of part-time and shortterm jobs that aspirants often accept in order to get a ‘foot on the ladder’. Table 4.2 summarizes the same data, showing changes between the three-yearly averages at either end of the period, but does so also in the case of hourly earnings, and also both with and without the effects

David Marsden 107 Group B occupations Legal professionals

Engineers 1600 1400 1200 1000

1600 1400 1200 1000 600 400 200 19 75 19 77 19 79 19 81 19 83 19 85 19 87 19 89 19 91 19 93 19 95 19 97 19 99 20 01

0

ye a 19 r 75 19 77 19 79 19 81 19 83 19 85 19 87 19 89 19 91 19 93 19 95 19 97 19 99

800 600 400 200 0

800

Police (all ranks)

Medical practitioners

1400

1400

1200

1200

1000

1000

800

800

600

600

400

400

200

200

0

0

ye ar 19 75 19 77 19 79 19 81 19 83 19 85 19 87 19 89 19 91 19 93 19 95 19 97 19 99

1600

ye ar 19 75 19 77 19 79 19 81 19 83 19 85 19 87 19 89 19 91 19 93 19 95 19 97 19 99

1600

Nurses

All occupations

1400

1400

1200

1200

p99 pnaa

1000

1000

p95 pnaa

800

800

p90 pnaa

600

600

p10 pnaa

400

400

p10 paa

200

200

0

0 ye ar 19 76 19 79 19 82 19 85 19 88 19 91 19 94 19 97 20 00

1600

ye a 19 r 75 19 77 19 79 19 81 19 83 19 85 19 87 19 89 19 91 19 93 19 95 19 97 19 99

1600

Figure 4.2 (continued)

of pay loss from short absences, another indicator of precarious work conditions. The charts also compare weekly pay with and without loss of pay from absence for the lowest decile. The growing gap in absolute real earnings between p95 and p10 is a feature of all of the less structured occupations of Group A, whether we consider weekly or hourly earnings, and whether or not we include the effect of a loss of pay as a result of short absences. The one exception in this group concerns hourly pay not affected by absence among academics. In contrast, among the more structured occupations of Group B, on the whole, those in the bottom decile have narrowed the gap in real pay with those on the 95th percentile: the main exception in this instance appears to have been for weekly pay including pay loss from short absences among salaried doctors and nurses. In the first case,

108 Table 4.2 Summary of growth in real weekly and hourly earnings at selected percentiles 1975–2001 Change on average between three-year periods 1975–77 and 1999–2001, full and part-time combined. ‘Pnaa’ excluding, and ‘Paa’, including those whose pay was affected by absence. Constant 2001 prices. Occupation

Group A occupations Authors, writers & journalists Media (excl. journalists & sports) Higher education academics Software engineers Management consultants* Group B occupations Engineers (excl. software)

P10

P95

P10

P95

Pnaa hourly

Pnaa hourly

++

161.5

206.9

169.7

+

160.2

139.7

152.2

+

131.7 239.1

171.0 314.3

169.2

160.9

Pnaa weekly

Pnaa weekly

146.5

194.9

166.1

gap

gap

P10

P95

Paa weekly

Paa weekly

++

128.4

180.5

175.1

+

145.9

160.2

142.2

–

++ ++

138.5 239.1

174.5 314.3

–

171.9

153.4

gap

P10

P95

Paa hourly

Paa hourly

gap

++

144.1

201.2

++

161.8

+

147.7

165.4

+

19.5

137.4

++

139.7

147.7

+

++ ++

104.4 104.7

187.4 143.6*

++ ++

106.3 104.7

178.1 143.6

++ +

–

133.8

153.7

+

131.2

148.4

+

Legal professionals Police officers all grades Medical practitioners Nurses and midwives All occupations

256.1

191.8

––

256.1

191.8

––

254.1

161.4

––

255.9

154.1

––

186.1

143.0

––

197.9

158.6

––

185.5

142.6

––

194.1

159.0

––

237.1

170.4

––

218.8

166.9

––

108.4

166.1

++

169.8

163.7

–

145.2

150.6

+

203.2

160.7

––

104.3

169.9

++

196.0

182.4

–

107.4

179.4

+

132.7

183.3

++

93.2

173.1

++

127.6

180.7

++

* 1991–2001 only. ‘Management consultants’ were not shown separately for 1975–1990. Key: ‘gap’: The single signs, +/–, denote changes for p95 of up to 25 percentage points greater or less than p10 over the period; the double signs, ++/––, denote changes of more than 25 points. Earnings deflated by retail price index, 2001 = 100. Note: estimates of hourly pay are based on smaller sample numbers because of lesser availability of hours worked. Source: NESPD.

109

110

Regulating for Decent Work

there has been a rise in the share whose pay was affected by absence, and in that of nurses, there has been a large increase in part-time working, and thus, in both cases, probably some erosion of their occupational status since the 1970s. Thus the overall picture is one of less structured careers and an increased exposure to economic uncertainty in the first set of occupations, and of the maintenance of established employment structures and protections in the second group. Accompanying this divergence of employment statuses, there have been parallel changes in earnings inequalities. In the first set of occupations, consistent with the tournament metaphor, there has been a widening of pay inequality, especially with the very top earners surging ahead, and the lowest earners facing stagnating real earnings and increased uncertainty. An analysis of the composition of the lowest-paid 20 per cent shows that there has also been a tendency for the part-time working and exposure to loss of pay to concentrate in the older age groups in the bottom 20 per cent of earners in these occupations, a symptom of being trapped in that position.3 In contrast, the Group B occupations, in which employment and career structures have been maintained, one can see that earnings inequalities have mostly fallen somewhat. An important feature of the argument about tournament occupations concerns the degree to which older workers may become trapped in the low-paying segment. The New Earnings Survey panel data (NESPD) provides some tentative evidence on this. The panel element of the NES provides a set of overlapping career snapshots as some employees remain in the sample over several years. Table 4.3 presents a set of pooled logit regressions for the selected occupations showing the predictors of an individual employee’s being in the bottom or the top quintile range (20 per cent) of weekly earnings for their occupation in a given year. The results for the individual occupations can be found in Marsden (2010, pp. 25–6). To capture the career element, lagged earnings are included for the previous three years. Three was chosen as a compromise between length of career information and loss of observations from attrition. There is considerable inertia in both the bottom and top 20 per cent in both sets of occupations, as shown by the strongly positive coefficients for the employee’s presence in the bottom or top 20 per cent in the preceding years. As expected, being part-time, having the level of pay affected by absence, and being female increase the chance of being in the bottom 20 per cent of earners. The number of yearly observations for an individual is likely to reflect employment stability, and so to correlate with higher pay.

Table 4.3 Summary of logit coefficients: factors predicting the probability of an employee’s pay being in the bottom or top 20% of weekly earnings Group A

Group B

Bottom 20%

Bottom or Top: lag1 yr Bottom or Top: lag2 yrs Bottom or Top: lag3 yrs No of job changes Age > 40 years No job changes age > 40 New hire (last 12 mo) New hire* age> 40 Female Part-time Pay affected by absence No of years in survey Occupation dummies Year dummies N Pseudo R2 Clusters

Top 20%

Bottom 20%

Top 20%

Coef.

sig

Coef.

sig

Coef.

sig

Coef.

sig

2.909 1.362 0.897 –0.139 0.247 0.049 –0.509 0.941 0.257 2.875 3.326 –0.033 yes yes 23,714 0.622 4,818

**** **** **** – *** – * *** **** **** **** ****

2.996 1.604 1.256 0.365 –0.064 –0.239 0.050 0.083 –0.272 –3.104 –1.695 0.005 yes yes 23,714 0.588 4,818

**** **** **** **** – **** – – **** **** **** –

2.116 1.171 0.776 0.059 –0.163 –0.040 0.363 0.277 0.255 2.670 3.116 –0.035 Yes Yes 142,202 0.511 18,987

**** **** **** **** **** – **** **** **** **** **** ****

2.379 1.379 1.094 –0.008 –0.021 0.165 –0.176 –0.133 –0.279 –2.537 –1.665 0.021 yes Yes 142,202 0.509 18,987

**** **** **** – – **** **** – **** **** **** ****

111

Significance levels: **** 1%; *** 2%, ** 5%, * 10%. Robust standard errors used, allowing standard errors to differ between clusters (multi-year observations for individual employees in the panel data set). Unbalanced panel. Note on variables: Number of job changes: computed from the number of times the employee was reported as a new hire in the previous three and current years. New hire: if the employee had been with the current employer for less than 12 months. Pay affected by absence: whether the employer reported that the employee’s pay in the survey period had been affected by absence. Number of years in the survey: computed from the number of years the employee’s id appears in the sample. The full results of these regressions are reported in Marsden (2010, pp. 25–6). Source: New Earnings Survey Panel Data Set, ONS.

112

Regulating for Decent Work

Behind the two groups of occupations lies a contrast between the loosely structured careers of Group A occupations and the organizationally based careers of Group B occupations. As might be expected, career progression in the more structured Group B occupations would carry the lower paid out of the bottom 20 per cent, reflected in the negative relationship between being aged over 40 and low pay. In contrast, in the more loosely structured occupations of Group A, being aged over 40 increases the likelihood of being in the bottom 20 per cent, consistent with the idea of many older workers being trapped. Job changing reveals a similar contrast. The frequency of recent job changes helps those in the loosely structured Group A occupations to reach the top-paid 20 per cent, but it has the opposite effect in the more structured Group B occupations. For these, it seems that frequent job changing leads to exclusion from career progression. Similarly, being newly hired in the loosely structured occupations has little significant effect on pay, whereas in the more structured Group B occupations, the increased likelihood of being low paid is consistent with the practice of hiring into lower-paid entry positions, and at the bottom of career ladders. There is also a revealing interaction between job changing and being aged over 40. Whereas frequent job changes have an overall benefit on those in the loosely structured Group A occupations, the opposite relationship is observed in those aged over 40: frequent job changes for older workers signal exclusion from the top 20 per cent of earners. Likewise, whereas being a new hire is associated with a reduced chance of being low paid overall, for new hires aged over 40, there is an increased chance of being low paid. Being a new hire aged over 40 is also associated with lower pay in the more structured occupations, albeit to a markedly lesser degree. Thus, although there is considerable pay inertia among both groups of occupations as those in the bottom and top 20 per cent of earnings tend to stay unchanged from year to year, the effects of the two types of career systems are revealed by the relationships with age and job changing.

Conclusions Before concluding, it is worth considering some alternative interpretations to that advanced in this chapter. Skill-biased technical change has been a powerful and relatively successful argument used to explain the growing pay inequality in labour markets across a number of countries in the period since the 1970s (Autor, Levy and Murnane 2003). However,

David Marsden 113

the effects of technology on jobs are fairly widely diffused, and do not seem to explain the different experience of the two sets of occupations considered in this chapter. Likewise, increased educational supply could generate a crowding effect at the bottom of the labour market and help to explain why workers on the bottom decile in general have done badly. However, it does not explain why they should have done relatively well in the more structured occupations. The job polarization thesis of Goos and Manning (2007) can explain many of the features observed in this chapter, but would also need to explain the differences between the two groups of occupations. However, there is no obvious reason why jobs in the middle of the Group A occupations should have become more routinized than those in the middle of Group B. The thesis advanced in this chapter does not necessarily contradict any of these models, and could well be complementary. Likewise, in relation to Salverda’s and Mayhew’s (2009) argument about the influence of inclusive industrial relations, the present chapter could be considered to highlight an additional institutional mechanism to those they emphasize such as encompassing collective agreements and extension clauses. The focus of this chapter has been on the changing pattern of labour market careers in selected occupations and its effect on segmentation in Britain since the 1970s. In that decade, it was possible to describe Britain as one of the countries in Europe, alongside Germany, in which occupational structures dominated employment. In this respect it could be contrasted with the importance of firm internal labour markets in countries such as France and Italy. The massive decline of the industrial sector in Britain makes it a less promising area in which to search for newly emerging patterns of labour market segmentation. Consequently, this chapter has focused on service activities, contrasting a selection of those in which less institutionally structured project-based forms of employment have developed with a similar selection of those in which stronger patterns of institutional regulation have remained intact. The special conditions that favour the first group mean that it is unlikely to become a general model for the whole economy. Nevertheless, because its organization responds to growing needs for flexible patterns of production and service provision, and because it is adapted to many aspects of creative activities for which there is a strong demand, one can expect it to grow in these areas. It was argued that the actors in these sectors still have to deal with issues such as skill acquisition and the flows of suitable job applicants, but because of the more transient nature of employment, and of many employing organizations, it was less suited to traditional methods of

114

Regulating for Decent Work

training. It lacks the stability for the provision of internal labour market training, and also some of the foundations for apprentice-type training for occupational markets. It was argued that this discourages employers from funding training for transferable skills, and that therefore more of the burden falls onto the aspirant members of the occupation. Employees have to acquire these skills mostly in while in employment, working on project assignments, and build up their portfolios as they move from one assignment to another. The relatively open entry at this stage leads to a large number of aspirants competing for assignments. Employers can make this work by allowing the pay of the top earners to drift upwards in order to increase the attractiveness of the potential rewards for those who manage a successful transition into the occupation. A rough and ready test of this hypothesis was provided by a comparison of the earnings patterns within broad occupations, some likely to include these tournament-style transitions and some with more traditional routes of entry. It was found that the pay of those on the tenth percentile fared worse in the former than in the latter group of occupations. Clearly, at this stage of the research, these results are suggestive, and more work needs to be done on related aspects of the tournament entry model. For example, Landers et al. (1996), considering entry into US professions, observed that a willingness to work very long hours was also formed a part of the process. Likewise, other indicators of precarious employment among the aspirant entrants, such as job durations, also need to be explored. It is also necessary to explore further the filtering process between those on the fast and medium tracks in Figure 4.1 and those on the slow track who never manage the transition. There is plenty of anecdotal evidence in the occupations considered that after a certain age, a good proportion of those on the ‘slow track’ drift into a sort of secondary labour market within these occupations, working occasionally for low pay and with little security. More work also needs to be done in order to assess how far the entry tournament processes contribute to our understanding of other changes in labour markets over the period. Although one can understand why such tournament-style transitions might develop in these occupations, there is an important question as to whether it is desirable for society as a whole: how far are they compatible with the goal of decent work? Apart from questions about desirable levels of overall inequality between ‘winners’ and ‘losers’, there is a related question concerning the cost of access. One finding of the research carried out by the Sutton Trust was that in recent decades, the growth of entry by long periods of precarious employment has transformed journalism into a much more elitist occupation. Aspiring

David Marsden 115

entrants would often rely on the support of their families while they worked on low and uncertain levels of pay, so that family capital had become a more significant determinant of success. This was visible in the increased proportion of top journalists from independent, fee-paying, schools and from elite universities. Likewise, university-based scientific research requires large numbers of researchers on doctoral student grants and on postdoctoral schemes, all of which involve low levels of income. The ability to sustain this for long periods requires either that family life is postponed, or that families provide a measure of financial support. Students from low-income households who have already built up debts during their years of undergraduate studies would find such burdens extremely hard to bear. The other undesirable face of these developments concerns those who get caught in the secondary segment within their chosen occupation. The link between being aged over 40, job changing and low pay in the loosely structured occupations, shown in Table 4.3, highlights the problem of those who are potentially trapped on low incomes and a succession of temporary assignments. A ‘bohemian’ lifestyle may be attractive in one’s twenties, but it becomes less so when raising a family, or contemplating old age without an adequate pension. To promote decent work for all in these sectors, attention needs to be given to how to address these two problems: social selectivity and trapped aspirants. The irony of the first is that the expansion of higher education has probably contributed to the increased supply of aspirants, which makes this a less promising avenue. On the other hand, given that many of the tournament-type occupations benefit from a good social mix in their membership, as this keeps them close to a wider range of potential customers, one might consider the introduction of a system of bursaries for students from low-income households. Firms might be encouraged to take on gifted trainees from low-income groups, as in previous decades they might have taken on apprentices. For those who have become trapped, the most obvious assistance would seem to be the active use of lifelong learning to assist the transition to a different occupation. On the other hand, if the economic argument for the success of the tournament model is correct, then attempts to stifle it may prove counterproductive as these occupations have been characterized by strong growth in recent years. Finally, decent retirement to follow decent work suggests an urgent need to investigate the provision for post-retirement incomes in these occupations. All of the occupations considered in this chapter are, to varying degrees, ‘knowledge intensive’ in the sense that they require high levels of education and training.

116

Regulating for Decent Work

As noted earlier, in the media, research and consultancy occupations, there is a greater emphasis on the acquisition of new knowledge, which requires more fluid employment structures, such as project-based organization, and this has contributed to their economic success. Therefore, any attempt to reverse these changes is probably undesirable. Instead, we need to consider how to ensure decent work for all within these more flexible systems.

Notes 1. These figures were extracted from the university performance league tables for 2000, published by the Times Higher Education Supplement in its issue of 14 April 2000. The median percentage of non-permanent staff across Britain’s 97 universities was about 40 per cent. Overall, there is a correlation of 0.77 between university research ratings in the national Research Assessment Exercise and the percentage of non-permanent staff. The RAE is one measure of success in the competition for research funds, and it also serves as a quality index in the search to attract overseas students. 2. For details see Marsden (2010, Table 2). 3. Not shown in this chapter, but available from the author.

References Acemoglu, D.; Pischke, J.S. 1999. ‘Beyond Becker: training in imperfect labour markets’, Economic Journal, vol. 109, no. 453, February, pp. 112–42. Aoki, M. 1988. Information, Incentives, and Bargaining in the Japanese Economy (Cambridge: Cambridge University Press). Arthur, M. B.; Rousseau, D. M. (eds.) 1996. The Boundaryless Career: A New Employment Principle for a New Organizational Era (New York: Oxford University Press). Auer, P.; Cazes, S. 2000. ‘The resilience of the long-term employment relationship: evidence from the industrialized countries’, International Labour Review, vol. 139, no. 4, pp. 379–408. Autor, D.H.; Levy, F.; Murnane, R.J. 2003. ‘The skill content of recent technological change: an empirical exploration’, Quarterly Journal of Economics, vol. 118, no. 4, pp. 1279–333. Baumann, A. 2002. ‘Informal labour market governance: the case of the British and German media production industries’, Work, Employment and Society, vol. 16, no. 1, pp. 27–46. Baumann, A. 2003. ‘Path-dependency or convergence? The emergence of labour market institutions in the media production industries of the UK and Germany’, Doctoral thesis (Florence: European University Institute). Becker, G.S. 1975. Human Capital: A Theoretical and Empirical Analysis, with Special Reference to Education (Chicago, University of Chicago Press). Belfield, R. 2006. ‘Developments in pay inequality after 1990: Britain, France, and Germany compared’. Paper presented to the 2006 Conference of the Society for the Advancement of Socio-Economics, Trier.

David Marsden 117 Bognano, M.L. 2001. ‘Corporate tournaments’, Journal of Labor Economics, vol. 19, no. 2, April, pp. 290–315. Eyraud, F.; Marsden, D.; Silvestre, J.J. 1990. ‘Occupational and internal labour markets in Britain and France’, International Labour Review, vol. 129, no. 4, pp. 501–17. Goos, M.; Manning, A. 2007. ‘Lousy and lovely jobs: the rising polarization of work in Britain’, Review of Economics and Statistics, vol. 89, no. 1, February, pp. 118–33. Heery, E.; Conley, H.; Delbridge, R.; Stewart, P. 2004. ‘Beyond the enterprise: trade union representation of freelances in the UK’, Human Resource Management Journal, vol. 14, no. 2, pp. 20–35. Hirsch, F. 1977. The Social Limits to Growth (London: Routledge and Kegan Paul). Jones, C.; Walsh, K. 1997. ‘Boundaryless careers in the US film industry’, Industrielle Beziehungen, vol. 4, no. 1, pp. 58–73. Jürgens, U. 2003. ‘Transformation and interaction: Japanese, US, and German production models in the 1990s’, in Yamamura, K.; Streeck, W. (eds) The End of Diversity? Prospects for German and Japanese Capitalism (Ithaca, NY: Cornell University Press). Kerr, C. 1954. ‘The balkanization of labor markets’, in Bakke, E. (ed.) Labor Mobility and Economic Opportunity (Cambridge, MA: MIT Press), pp. 92–110. Lam, A. 2003. ‘Organizational learning in multinationals: R&D networks of Japanese and US MNEs in the UK’, Journal of Management Studies, vol. 40, no. 3, May, pp. 673–703. Lam, A. 2005. ‘Work roles and careers of R&D scientists in network organizations’, Industrial Relations, vol. 44, no. 2, pp. 242–75. Landers, R.; Rebitzer, J.; Taylor, L. 1996. ‘Human resources practices and the demographic transformation of professional labour markets’, in Osterman, P. (ed.) Broken Ladders: Managerial Careers in the New Economy (New York: Oxford University Press), pp. 215–45. Lazear, E.P. 1998. Personnel Economics for Managers (New York: Wiley). Lazear, E.P.; Rosen, S. 1981. ‘Rank order tournaments as optimum labor contracts’, Journal of Political Economy, vol. 89, pp. 841–64. Marsden, D.W. 1982. ‘Career structures and training in internal labour markets in Britain and West Germany’, Manpower Studies, no. 4, Spring, pp. 10–17. Marsden, D.W. 1990. ‘Institutions and labour mobility: occupational and internal labour markets in Britain, France, Italy and West Germany’, in Brunetta, R.; Dell’Aringa, C. (eds) Labour Relations and Economic Performance (London: Macmillan), pp. 414–38. Marsden, D.W. 2007. ‘Labour market segmentation in Britain: the decline of occupational labour markets and the spread of “entry tournaments”’, Économies et Sociétés , Série Socio-Économie du Travail AB No. 28, 6/2007, pp. 965–98. Marsden, D.W. 2010. ‘The growth of extended “entry tournaments” and the decline of institutionalised occupational labour markets in Britain’, Centre for Economic Performance Discussion Paper no. 989, London School of Economics, London. http://cep.lse.ac.uk/pubs/download/dp0989.pdf. Maurice, M.; Sellier, F.; Silvestre, J.J. 1982. Politique d’éducation et Organization Industrielle en France et en Allemagne (Paris: Presses Universitaires de France). Osterman, P. (ed.) 1996. Broken Ladders: Managerial Careers in the New Economy (New York: Oxford University Press).

118

Regulating for Decent Work

Pettigrew, N.; Willis, J.; Paterson, R.; Dex, S. 1997. British Film Institute Television Tracking Study: Second Interim Report (London: British Film Institute). Salverda, W.; Mayhew, K. 2009. ‘Capitalist economies and wage inequality’, Oxford Review of Economic Policy, vol. 25, no. 1, pp. 126–54. Saxenian, A. 1996. ‘Beyond boundaries: open labor markets and learning in Silicon Valley’, in Arthur, M.; Rousseau, D. (eds) The Boundaryless Career (New York: Oxford University Press), pp. 23–39. Sengenberger, W. 1987. Struktur und funktionsweise von arbeitsmärkten: die Bundesrepublik Deutschland im internationalen vergleich (Frankfurt: Campus Verlag). Stevens, M. 1994. ‘An investment model for the supply of training by employers’, Economic Journal, vol. 104, no. 424, May, pp. 556–70. Supiot, A. 1999. Beyond Employment: Changes of Work and the Future of Labour Law in Europe (Oxford: Oxford University Press). Sutton Trust 2006. The Educational Backgrounds of Leading Journalists (London: The Sutton Trust). Tolbert, P. 1996. ‘Occupations, organizations, and boundaryless careers’, in Arthur, M.; Rousseau, D. (eds) The Boundaryless Career: A New Employment Principle for a New Organizational Era (New York: Oxford University Press), pp. 331–49. Wilthagen, T.; Bekker, S.; Klosse, S. 2007. ‘Making (it) work: introduction to the special issue on the future of the European employment strategy’, International Journal of Comparative Labour Law and Industrial Relations, vol. 23, no. 4, pp. 489–98.

David Marsden 119

Appendix on sources and definitions

Note on sources and definitions: the New Earnings Survey is based on a one per cent sample of employees across all economic sectors of Great Britain, and is carried out by the Office for National Statistics (ONS). The sample is based on national insurance numbers, and returns by employers. Earnings relate to cash earnings before tax and other deductions, and relate to a specific pay period in April each year, and include a proportionate amount of regularly paid bonuses. The tables and charts in this chapter were calculated from NES panel data (NESPD) which comprises a subset of variables for the period 1975–2001. From 2001, the NES was superseded by the Annual Survey of Hours and Earnings, ASHE. The panel element of the dataset arises because every year returns are requested for about 75 per cent of the previous year’s sample. The Labour Force Survey collects information from households in Great Britain and Northern Ireland and therefore covers a wider range of employment statuses, and notably the self-employed, who are excluded from the NES. The LFS does not include data on selfemployed earnings for the years of this study. A major problem for the study of occupations over the period was the change from the ‘KOS’ classification system to the ‘SOC1990’ system from 1991 onwards. For the LFS and the NES a visual matching of the closest occupations was made by the author using detailed information on the categories used in the two classifications. Because of the break, several of the tables show results for the period just before and just after the changeover. For the NES, it was possible also to use the panel element in the survey to assist with matching between the two classifications, and thus use the information on the occupational affiliation of the same individuals before and after the changeover. I am grateful to Richard Belfield for making available his back-mapping of the SOC1990 onto the earlier years for this study. As both methods have their drawbacks, some tables in this paper show results using both methods. Occupations used (SOC 1990): Group A: Journalists: soc1990==380 Media excluding journalists: soc1990>=381 & soc1990<=386 HE Academics soc1990==230 Mgt consultants: soc1990==253 Software engineers: soc1990==214

120

Regulating for Decent Work

Group B: Engineers: soc1990>=210 & soc1990<214 | soc1990>214 soc1990<219 Legal professionals: (soc1990>=240 & soc1990<=242) Police (all ranks): soc1990==152 | soc1990==610 Medical practitioners: soc1990==220 Nurses & midwives: (soc1990>=340 & soc1990<=341)

Part II Recent Country Experiences

This page intentionally left blank

5 Laws or Luck? Understanding Rising Formality in Brazil in the 2000s* Janine Berg

Introduction Until the debt crisis of the 1980s and the lost development decades of the 1980s and 1990s, it was a widely held belief within development circles that developing countries would industrialize over time and that with industrialization, the living standards of the population would rise. Productivity increases would lead to increases in real wages and labour protections, typically restricted to a select few, would be extended to the newly industrialized workforce. But the experience of the 1980s and 1990s, particularly in Latin America, led policy makers to consider otherwise. A series of economic crises and overall stagnant growth gave way to the continued swelling of large informal economies and by the 2000s, the belief that Latin American labour markets could one day resemble those of industrialized countries seemed a bygone dream. Yet during the 2000s, the trend of increasing informalization was reversed in Brazil as well as in Argentina. This chapter discusses the important advances that have been made in Brazil in formalizing its labour market, the reasons for this improvement, and the implications for wider policy debates. Brazil’s labour market performed well in the 2000s, with strong rates of job creation and formal job growth outpacing informal job growth by a three-to-one ratio. A principal objective of this chapter is to assess the extent to which government policies contributed to the improved

* The views expressed in this chapter are those of the author and do not necessarily represent the views of the International Labour Office. I am grateful to Diego Sánchez-Ancochea, Kathyrn Hochstetler, Fabio Bertranou, Sangheon Lee, Deirdre McCann and an anonymous referee for helpful comments on the chapter. 123

124

Regulating for Decent Work

performance of the labour market. This question is important not only for future policy design in Brazil and larger debates within the country about the role of government, but also because it can shed light on two other ongoing debates in economic development. The first debate concerns labour market flexibility – whether labour market regulations impede economic growth and job creation – and the second, somewhat parallel, debate on whether growth, in and of itself is sufficient for improving working conditions, or whether government intervention is also needed.

The debate on labour market flexibility in Brazil The 1990s were a difficult time for the Brazilian economy and an even more difficult time for its workers. Unemployment rates doubled, informality rose sharply, and real wages declined. The poor performance of the labour market led to a flurry of debate in the country about its causes and possible remedies. This debate, unsurprisingly, was highly influenced by international policy makers, particularly the Washington consensus view that in order for economic reforms to be successful, the labour market needed the flexibility to adjust. Only then, it was argued, could workers reap the benefits of the new economic model. Many reforms were made to the Brazilian economy. In 1990, the Collor administration announced the opening of the economy. The programme called for the immediate removal of import prohibitions and non-tariff barriers, followed by a gradual reduction of tariff rates, from an average of 32 per cent in 1990 to 14 per cent in 1993. Following the establishment of the Mercosur Customs Union in 1995, the average import tariff was further reduced to 13 per cent. Brazil’s financial markets were also liberalized, which entailed the removal of prohibitions on foreign investment in the stock exchange as well as the loosening of regulations on the foreign financing of domestic firms. In 1992, investors were given access to options and future markets. The government also undertook drastic measures to ensure price stability and in 1994 unleashed the Real Plan, which was successful in reducing inflation to single-digit levels. Financial liberalization resulted in a surge in portfolio investment which caused the appreciation of the real exchange rate. Unfortunately, this occurred at the same time that domestic companies were being exposed to foreign competition, after previously being protected under the policy of import-substitution industrialization. Domestic industries also suffered from high interest rates, imposed in order to attract capital investment and maintain the value of the exchange rate. From 1993

Janine Berg 125

until the currency devaluation in January 1999, imports grew at an average annual rate of 18 per cent, compared with the 3 per cent annual growth in exports. The backdrop to these reforms was a newly democratic government and a new Constitution from 1988, which, in addition to other measures, extended labour protections. The Constitution increased the cost of dismissal without just cause by raising the penalty from 10 to 40 per cent that employers have to pay on funds accrued in the Guarantee Fund for Time of Service (Fundo de Garantia do Tempo de Serviço – hereafter FGTS). The FGTS is an individual severance fund financed by a monthly employer contribution equivalent to 8 per cent of the employee’s salary, which can be accessed at retirement, upon dismissal as well as for the purchase of a home. If dismissal is without just cause, then the employer must pay a penalty equivalent to 40 per cent of the value of the fund. The 1988 Constitution also reduced working hours from 48 hours to 44 hours per week, raised overtime remuneration from 1.2 to 1.5 times the normal wage, increased vacation pay by one-third of the monthly salary, increased maternity leave from 90 to 120 days, and granted new fathers five days of paternity leave. The Constitution also extended some of these rights to domestic workers, who had previously been excluded from most labour protections. In addition, it removed many restrictions on unionization and on striking that had been placed during the military government. Yet the pressure for labour market reform during the 1990s was strong, and though not as far-reaching when compared with other countries in Latin America, a number of reforms were undertaken to make the labour market more flexible. In 1994, the Law on Cooperatives was passed, which permitted the creation of cooperatives of workers to deliver services to firms without the constitution of a work contract. The law made explicit that there is no employment relationship between cooperatives and their members. Though the law was put forward to help landless peasants, increasingly it was used by employers to avoid their legal obligations towards employees (Marshall 2004). In 1998, the ‘bank of hours’ was instituted, making working hours more flexible, and lowering business costs by allowing working hours to vary, thereby permitting the suppression of the payment of overtime. The measure, however, was restricted to work contracts that are negotiated collectively. Also significant was the end, with the 1994 Real Plan, of the policy of indexing wages to inflation, after almost thirty years of official wage policies. Finally, the government chose to limit increases in the minimum wage and, as a result, it fell by nearly 20 per cent,

126

Regulating for Decent Work

from R$331 in February 1990 to R$269 in February 1999, measured in real terms.1 Thus, although it is true that the Constitution of 1988 raised labour costs, the decision to stop indexing wages and allow the minimum wage to fall contributed to a fall in average real wages from R$978 in 1995 to R$914 in 1999, mitigating some of the increase in overall labour costs.2 Nevertheless, many critics still maintain that it was the rigidity of the Brazilian labour market that led to the poor performance of the 1990s. Bosch et al. (2007), for example, used a model of gross labour flows to analyse the causes of rising informality in six Brazilian cities during the period 1983–2002 and concluded that trade liberalization made only a modest contribution to the increase in informality (21 per cent), whereas strengthened labour rights under the 1998 Constitution explained 76 per cent of the increase.3 They argue that ‘several dimensions of the Constitutional reform, in particular, regulations relating to firing costs, overtime and union power, explain much more … [the] effects work mostly through the reduction in hiring rates, rather than separation rates … overall, the findings confirm the importance of labor legislation to firms’ decisions to create new formal sector jobs in Brazil’ (p. 26). If stricter regulations caused informality in the 1990s as suggested by Bosch et al. (2007), then the logical conclusion for the 2000s is that increased formalization was due to a loosening of labour market regulations. While there was some flexibilization in the 1990s, as mentioned above, the trend in the 2000s was quite the opposite. During the 2000s, the minimum wage nearly doubled in real terms, increasing from R$266 in February 2000 to R$510 in January 2010 (see Figure 5.1). The minimum wage has a significant impact on labour costs in Brazil, as around 15 per cent of the formal labour force earns the minimum wage. It also determines the salaries of other workers on the pay scale as it is often used as a numeraire, with workers’ salaries set as multiples of the minimum wage (Neri and Moura 2006).4 Thus, it is fair to conclude that in the 2000s, there was greater institutional influence on the working of the labour market, not less. And that in spite of – or perhaps, as a result of – this influence, formalization increased.

Labour market trends in Brazil in the 1990s and 2000s In 2008, according to data from Brazil’s national household survey (PNAD), 49.6 per cent of the labour force was formally employed, using the 2002 ILO definition.5 According to this definition, formal employment in Brazil

Janine Berg 127 550 510 500 484 450 409 400 350

419

361 335 300

298

300

280

250

245 252 213

200

258

278

317 281

2002

2004

297

271

290

266

323

150 1990

1992

1994

1996

1998

2000

2006

2008

2010

Figure 5.1 Real minimum wage in Brazil, 1990–2010 Note: Data are for January of each year, measured in January R$ 2010. Source: DIEESE.

includes private salaried workers and domestic workers with a signed labour card (carteira assinada), government workers and military, as well as employers and self-employed workers who contribute to the Brazilian social security scheme (previdência ). Informally employed workers include private sector and domestic workers who do not have a signed labour card (sem carteira assinada), self-employed workers and employers who do not contribute to the social security system, unremunerated workers, and workers who produce and build for their own use. During the 1990s, the performance of the labour market was poor, resulting in a fall in formality rates from 46.4 per cent in 1992 to 43.9 per cent in 1999. In the 2000s, the pattern was reversed, resulting in an increase in formality to 49.6 per cent in 2008 (see Figure 5.2).6 Table 5.1 shows the dramatic difference in the growth of formal and informal jobs between the two periods. During 1992–99, annual formal job growth was a mere 1.3 per cent, whereas informal jobs grew at a rate of 3.0 per cent annually. In the second period, 1999–2008, the patterns were reversed. Formal jobs grew at an annual rate of 5.3 per cent, whereas informal job creation fell to 1.7 per cent annually. The rise in formality rates was driven principally by an increase in the percentage of salaried workers with a signed labour card, who, in 2008, made up 34.5 per cent of the total employed. Between 1999 and 2008, this category grew at an average annual rate of 6.6 per cent, compared with just 0.7 per cent annually between 1992 and 1999. In comparison,

128

Regulating for Decent Work

50.0

49.6 49.5

49.0 48.1

48.0 47.0

46.9 46.4

46.3 46.0

46.0 45.7 45.0

45.5

45.7

45.5 45.2 44.8 44.5

44.0

43.9

43.0 1992 1993 1995 1996 1997 1998 1999 2001 2002 2003 2004 2005 2006 2007 2008

Figure 5.2 Share of formal employment in Brazil (ILO definition), 1992–2008 (per cent) Note: Workers aged 16 years and older. The rural areas of the Northern states (with the exception of Tocantins) are excluded. The PNAD survey was not undertaken in 1994 or 2000. Source: Author’s calculations based on IBGE/PNAD.

during 1992–99, the category of informal salaried workers (sem carteira assinada) grew annually by 3.3 per cent, whereas the rate of growth fell to 2.0 per cent annually during the second period. In 2008, self-employed workers accounted for nearly 21 per cent of the labour force, of whom only 14.5 per cent contributed to the social security system. Job growth in this category also slowed in the second period, with a small improvement in the number of self-employed workers who contributed to the social security system. The more important trend was the sharp slowdown among non-contributing self-employed workers, where job growth declined from an annual rate of 3.9 per cent in 1992–99 to 1.3 per cent in 1999–2008. More employers also contributed to the social security system increasing from 2.0 per cent annually in the period 1992–99 to 3.2 per cent annually between 1999 and 2008. Similarly, the annual growth in the number of non-contributing employers fell from 6.2 to 5.8 per cent, aiding the improvement in formality rates. The two periods witnessed a continued downward trend in the number of unremunerated workers, falling by 0.4 per cent annually during 1992–99 and by 2.6 per cent between 1999 and 2008. Nevertheless,

Janine Berg 129 Table 5.1 Annual employment growth by job category Formal and Informal Annual growth, Annual growth, Percentage Occupational Categories 1992–1999 1999–2008 of occupied workers, 2008 Total Formal Salaried, private-sector workers with a signed labour card Government workers, including military Domestic workers with a signed labour card Self-employed workers who contribute to the social security system Employers who contribute to the social security system Total Informal Salaried, private-sector workers without a signed labour card Domestic workers without a signed labour card Non-contributing, self-employed workers Non-contributing employers Unremunerated workers Work for own consumption and production Total (formal and informal)

1.3% 0.7%

5.3% 6.6%

49.6% 34.5%

3.2%

3.6%

7.7%

11.7%

3.4%

2.0%

–0.6%

0.6%

3.0%

2.0%

3.2%

2.4%

3.0% 3.3%

1.7% 2.0%

50.4% 17.5%

1.6%

2.8%

5.5%

3.9%

1.3%

17.7%

6.2%

5.8%

2.1%

–0.4% –0.2%

–2.6% 2.7%

3.6% 4.0%

2.2%

4.2%

100.0%

Source: Author’s calculations based on IBGE/PNAD.

there was an increase in workers producing for their own use from a fall of 0.2 per cent annually in 1992–99 to an annual increase of 2.7 per cent between 1999 and 2008. An interesting trend that deviates from the pattern of other job categories – and one to which we will return to in a later section – is the

130

Regulating for Decent Work

strong growth of formal job creation among domestic workers (those whom have a signed labour card) during the 1992–99 period. The rate of formal job creation among domestic workers was 11.7 per cent annually whereas job growth among informal domestic workers was 3.4 per cent annually. During 1999–2008, formal job growth among domestic workers still outpaced informal job growth at 3.4 per cent annual growth versus 2.8 per cent annual job growth, but at a much slower pace. Still, by 2008, only 27 per cent of domestic workers had formal jobs.

What explains the rise in formalization? While the rise in formality in Brazil in the 2000s has not gone unnoticed by Brazilian policy makers, there has been little systematic research of the causes, with the exception of Cardoso (2007) and, to a lesser extent, Baltar et al. (2006) and Chahad and Macedo (2003). Chahad and Macedo (2003), in a study of the 1999–2003 period, argue that the increase in formality is due to the introduction of the SIMPLES law, which facilitated registration and lowered the rate of taxation for small businesses. Baltar et al. (2006) attribute the improvement in the labour market to the improved economic environment following the exchange rate depreciation in 1999 and the export boom that began in 2003, as well as to the government’s efforts in ensuring compliance with labour law through increased inspection and jurisprudence. The most detailed analysis to date on the issue is by Cardoso (2007), who studies the years 2001–05. He attributes the growth in formality to the growth and decentralization of public spending, the growth of domestic credit, the growth and diversification of exports, the SIMPLES law as well as improvements in labour intermediation and inspection. This section reviews the principal reasons given as well as some additional reasons and assesses the importance of each to increasing formality. I argue that macroeconomic policies have increased the demand for workers in formal occupations, that demographic shifts and education policies have reduced the supply of youths in the labour market who typically occupy precarious jobs, and that there have been micro-level policy interventions that have altered the behaviour of firms and employers, resulting in greater formalization. These include the introduction of the SIMPLES law, which simplified and lowered taxes for small and medium-sized enterprises, improved labour inspection as well as greater legal awareness among workers, which I illustrate with the example of domestic work.

Janine Berg 131

Increased demand for formal workers There are two favourable aspects about macroeconomic performance in the 2000s when compared with the 1990s that aided the growth of formal jobs. These are: (1) the growth in exports, aided by the boom in commodities as well as the more competitive exchange rate; (2) an increase in internal demand, driven by a more inclusive pattern of growth and the emergence of a sizeable middle class. As mentioned previously, in the early 1990s, Brazil liberalized trade and capital inflows which led to a steady appreciation of the currency during the decade that was not reversed until the devaluation in January 1999 (see Figure 5.3). An incipient liberalization, coupled with an overvalued currency, resulted in a fall in domestic production and a negative trade balance between 1995 and 2000. Jobs were lost. Data from administrative records on formal employment show that between 1990 and 1999, there was an average annual job loss of 2.9 per cent in the mining sector, 8.0 per cent in the agriculture and fishing sectors and 1.9 per cent in the manufacturing sector.7 Indeed, during the 1990s, only the non-tradable sectors (construction, commerce, services and public administration) experienced positive, net formal job growth. In the 2000s, the pattern shifted. Following the devaluation of the real in 1999 and until 2005, the real exchange rate remained highly competitive, boosting exports and protecting domestic industries from import competition. The Brazilian labour market benefited as new jobs were created, particularly in the export sector – where exports grew by 130.0 120.0 110.0 100.0 90.0 80.0 70.0 60.0 50.0

19 90 19 91 19 92 19 93 19 94 19 95 19 96 19 97 19 98 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08

40.0

Figure 5.3 Real effective exchange rate in Brazil, 1990–2008 (2005=100) Source: IPEADATA.

132

Regulating for Decent Work

80 per cent between 2000 and 2008 – but also in the import-competing manufacturing sector. Between 2000 and 2008, the mining industry added formal jobs at an annual rate of 8.1 per cent, the manufacturing industry at a rate of 5.2 per cent and the agriculture and fishing sector at a rate of 3.6 per cent. The experience of Brazil confirms the findings of Frenkel and Ros (2006), in their study of 17 Latin American countries during the period 1990–2002, that there is a significant and positive relationship between the movements in the real exchange rate and the rate of unemployment. For an emerging economy such as Brazil, economic growth during 1992–2008 has not been particularly robust. Comparing the two periods, growth was higher in the period 1999–2008, averaging 4.2 per cent annually, or 2.5 per cent on a per capita basis. During 1992–99, the average rate of growth was 3.2 per cent, one percentage point lower than the second period, and a mere 1.4 per cent annual growth on a per capita basis (see Figure 5.4). The strongest growth years were during 2003–08 when growth averaged 5.2 per cent annually and 3.7 per cent annually per capita. Growth was therefore important during the second period, but the differences in growth rates are not large enough to explain the divergence in labour market performance. There is, however, a characteristic of growth that may help to explain part of the difference. In the second period, growth was inclusive. In 1992, the Gini coefficient for Brazil was 0.58 per cent, rising slightly to 0.59 in 1999. In the 2000s, there has been a steady improvement 7.0 6.0 5.0 4.0 3.0 2.0 1.0

97 19 98 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08

96

19

94

95

19

19

93

19

−1.0

19

19

92

0.0

Figure 5.4 Real GDP growth in Brazil, 1992–2008 Source: Central Bank of Brazil.

Janine Berg 133

in distribution leading to a reduction in the Gini coefficient to 0.54 in 2008. While the Gini coefficient is still extremely high, the reduction during the 2000s demonstrates that the benefits of economic growth are being shared more equitably. The redistribution is principally the result of the doubling of the minimum wage during the 2000s but is also due to the creation and expansion of the Bolsa Família (conditional cash transfer) programme in 2003 as well as the expansion of the rural pension and the Benefíco de Prestação Continuada (social assistance) programme (BPC). Saboia (2007), in an analysis of the impact of the minimum wage on the reduction in inequality during 1995–2005, found that 64 per cent of the improvement in income inequality at the household level was due to increases in the minimum wage. A similar analysis undertaken by IPEA for the 2000–08 period found that 66 per cent of the drop in inequality was attributable to the increase in the minimum wage, 16 per cent was due to the increase in pension benefits, 6 per cent was due to the BPC programme and 12 per cent was due to the Bolsa Família programme (IPEA 2009). The expansion of the social safety net during the 2000s has been important for making growth more inclusive. The BPC is a social assistance programme that provides benefits equivalent to the monthly minimum wage to persons aged 65 or older, or to persons who are unable to work due to disability of any age, whose per capita household income is less than one-quarter of the minimum wage (the extreme poverty line). In 2009, 1.5 million elderly and 1.6 million disabled households received benefits through the BPC. The rural pension, which represents 30 per cent of the benefits paid under the general pension system, covered approximately 8.2 million individuals in the first half of 2010 (up some 30 per cent since 2000). For 2010, this amounts to a transfer of approximately R$50 billion (1.6 per cent of GDP) to rural areas of the country that are usually the neediest. The Bolsa Família programme was created in 2003 and by 2008, 11.1 million families, or 46 million people (approximately one-quarter of the population), benefited from the programme.8 The average benefit was R$80 (US$44) per month and, though meagre, the assistance has been an important source of revenue, particularly in rural areas of the North and Northeast states, where most of the beneficiaries reside. Studies have shown that the benefit has been used primarily for the purchase of food, but also for clothing and shoes (ILO 2009). The benefits received from these social policies have been important for boosting consumption and economic growth in rural and poorer areas and has enabled small retailers to depend upon a steady source of demand for their products (Cardoso 2007). Although these

134

Regulating for Decent Work

small retailers may not necessarily be employing formal workers, the boost in demand affects other parts of the value chain where new job creation is likely to be formal, such as in manufacturing or distribution as well as large retailers such as supermarket chains and hypermarkets. Brazil in the 2000s, and particularly since 2004, has witnessed the emergence of a sizeable middle class. This middle class, defined as having family income of between R$1,064 and R$4,561 per month, grew by 22.8 per cent between 2004 and 2008 to comprise 52 per cent of Brazilian households (Neri 2008). A defining characteristic of this middle class is the consumption of durable goods – refrigerators, washing machines, televisions and cars – that has driven internal demand, contributing to strong and sustainable economic growth and the growth of new, formal jobs. Reduced labour supply Surprisingly little attention has been given to the role of demographics in explaining labour market trends in the 1990s and 2000s. At present, Brazil is in a process of demographic transition from a country characterized by high fertility rates and a large youth population to one of low fertility rates and an ageing population. In 1980–85, the fertility rate was 3.8 births per woman, falling to 2.3 in 2005–10 (ECLAC 2009). Between 1980 and 2008 life expectancy increased from 62.7 to 72.5 years. The demographic transition that Brazil is undergoing meant that during the 1990s, an additional six million youths, aged 15–24, entered the labour market when compared with the 1980s. In the decade of the 2000s, the net entry of youths was reduced by 300,000 over the previous decade (Bercovich 2005). Indeed, between 2005 and 2008, based on PNAD data, there was an absolute reduction of 1.45 million among the 15–24 population, reducing the share of youths in the total population from 18.6 per cent to 17.7 per cent. The reduction has alleviated pressure on the labour market of new entrants, reducing growth in precarious work. Another important policy change that has reduced the number of youths looking for work is the increase in the percentage of adolescents, aged 15 to 17, who study. In 1992, a shockingly high 40.3 per cent of youths in this age group did not attend school. A concerted policy effort to increase secondary enrolment rates resulted in a marked improvement in this indicator. By 1999, only 21.5 per cent of youths aged 15–17 did not attend school and this percentage was further reduced to 17.7 per cent by 2007. The Bolsa Família programme, initiated in 2003, has contributed to the goal of increasing school enrolment rates of youths by requiring that children up to the age of 17 attend school

Janine Berg 135

in order for their families to receive the cash benefit. As a result of these policies and the improved labour market conditions, the number of youths aged 15–17 that were not economically active increased from 57 per cent in 1999 to 65 per cent in 2008. Data from the PNAD give a picture of the type of jobs that economically active youths, aged 15–24, perform and their occupational shifts.9 In 2008, 16.5 per cent of youths were unemployed, slightly lower than the 18.3 per cent rate recorded in 1999. Following the trend of increased formality for the population as whole, 38 per cent held formal jobs compared with 30 per cent in 1999. Another important shift concerned the share of unremunerated workers, which fell from 14.0 per cent in 1999 to 7.5 per cent in 2008. It appears as though some of these unremunerated workers became remunerated, though continued to be informal, as there was an increase from 29.0 per cent in 1999 to 32.7 per cent in 2008 of informal, salaried workers. Self-employment is not as prevalent for youths as it is for the population as a whole. In 2008, 21 per cent of the workforce was self-employed compared with 9.3 per cent for youths aged 15–24. In 1999, 10.8 per cent of employed youths were self-employed. Data on the place of employment corroborate these trends. The share of youths who work in a store or office increased from 58 per cent in 1999 to 68 per cent in 2008. There was a sharp fall in youths who work on a farm from 17.8 per cent in 1999 to 10.7 per cent in 2008. Finally, the number of youths who engage in home work fell from 5 per cent in 1999 to 4 per cent in 2008. At the other end of the age spectrum, longer life expectancy and an adjustment in the laws governing eligibility for retirement benefits meant that an increased number of workers in their fifties have remained in the labour force. In 1999, the government instituted the ‘fator previdenciário’, a formula that calculates pension benefits based on years of contribution and age at retirement. Under the formula, workers younger than the legal retirement age (men under the age of 65 and women under the age of 60) would have their benefits reduced if they opted to retire early, even if they had sufficient years of contribution. As a result of this change in the law, workers have stayed on longer in the labour market. Paiva (2009), analysing household survey data from 2007, found that only 6 per cent of pension beneficiaries in that year were of workers who were younger than the legal retirement age but who had sufficient years of contribution. The SIMPLES Law In 1996, the government instituted a new system of tax exemption and simplification for small and micro enterprises, known as the SIMPLES.10

136

Regulating for Decent Work

The original ceiling set for qualification in the programme was gross annual sales of R$720,000. In 1998, the ceiling was raised to R$1.2 million; and in 2005, it was doubled to $2.4 million (or approximately US$1.3 million). The SIMPLES law has a progressive tax structure, which taxes companies at differentiated, but reduced rates, according to their gross sales.11 The collection of taxes is made simpler through the use of one single form, known as the Darf (Documento de Arrecadação da Receita Federal), which collects the different federal taxes, and, where agreements have been made between municipal, state and federal government, it also provides for the collection of state and municipal taxes. Under the SIMPLES, firms pay a reduced income tax, they are exempted from contributing to the Sistema S, a network of training agencies run by employers’ organizations, and there is also a reduction of the firm’s contribution to the social security system through the PIS/Pasep and the Cofins taxes. Previously, the PIS/Pasep and Cofins taxes were calculated based on total payroll, but it was de-linked from the payroll and substituted for gross sales of the firm, on a progressive scale (González 2006). The overall tax burden of firms under the SIMPLES programme is reduced by up to 8 per cent of annual revenue (Monteiro and Assunção 2006). Firms must still, however, make the necessary social security contributions for each employee and pay the FGTS. Analysis of the SIMPLES law on the formalization of firms (Delgado et al. 2007; Monteiro and Assunção 2006) reveals that it has boosted formality. Delgado et al. estimate that the law contributed to the formalization of close to 500,000 microenterprises during 2000–05, accounting for two million jobs. During the five-year period studied, annual GDP growth was 2.9 per cent, yet there was a 7 per cent annual increase of businesses registered under the SIMPLES. In their analysis, they determine that the typical SIMPLES business is one that employs 2.4 workers, is in the retail trade, and has gross sales of between R$1,000 and R$5,000 per month. Based on their findings, they conclude that ‘without the establishment of the SIMPLES, it is unlikely that the majority of the new establishments created or which already existed informally would have joined the conventional tax and social security system’ (Delgado et al. 2007, p. 39). Monteiro and Assunção (2006) analyse the ten-month period following the enactment of the SIMPLES law. They find that the SIMPLES led to a 13 percentage-point increase in formal licensing among retail firms, when compared with their control group of firms in ineligible sectors. Moreover, they test whether formalization leads to greater

Janine Berg 137

investment and find that while formalization does not change the decision about whether or not to invest, it did result in an increase in the level of investment for those firms that were investing. They find that, on average, licensed firms invested R$5,390 more, which, for firms in their sample, represents 34 per cent of the average annual revenue. This is not surprising, given that one of the principal benefits of being registered is attaining access to formal credit channels. In this respect the SIMPLES created a virtuous circle: firms could register more easily, this gave them access to credit which in turn improved their chances of survival. As a result they thrived and because they were legitimate businesses, they were able to – and had a greater incentive to – register their workers. Improved labour inspection and new approaches to formalization Labour inspection is fundamental in ensuring compliance with labour law and improvements in how inspection is undertaken in Brazil have aided the increase in formality. Improved labour inspection in Brazil is not due to an increase in the number of labour inspectors in the country – the number of labour inspectors has fluctuated at about the 3,000 mark since 1990 and the ratio of labour inspectors to the workforce has actually fallen12 – rather, it is due to changes in the incentive structure as well as new methods for meeting inspection targets, introduced since the mid-1990s. There have been two significant trends. The first was the introduction in the mid-1990s of a bonus system by which a substantial percentage of inspectors’ salaries were tied to individual performance targets as well as the total performance of the labour inspectorate. The new system made explicit the goal of formalizing workers, the collection of employers’ contributions by the FGTS fund as well as the number of workers covered by inspection. The second approach has been the development of teams of inspectors dedicated to tackling specific problems, usually tied to a particular sector. The inspectors who comprise the team are not evaluated by individual performance targets, but rather through team progress reports that demonstrate their ability to address sectorwide problems (see Pires in this volume). The objective is not merely to inspect and sanction (the command-and-control model of labour inspection), but rather to find ways to work with firms so that they can address the problem at hand, be it child labour, work accidents, or pervasive informality. Working in parallel, the two approaches have had an important impact on the success of labour inspection in the country.

138

Regulating for Decent Work

800,000 700,000 600,000 500,000 400,000 300,000 200,000 100,000 0 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Figure 5.5 Number of workers registered as a result of labour inspection in Brazil Note: The Secretariat´s database begins in 1996. Source: Labour Inspection Secretariat, Ministry of Labour.

As Figure 5.5 demonstrates, between 1996 and 2008, the number of workers registered as a result of inspection increased from 268,000 to 669,000. Nevertheless, because of the way the system is structured, the majority of the workers that have been registered have been informal workers working in formal firms. They comprise approximately 2.2 per cent of the total number of workers covered by labour inspection activities. Registering informal workers in formal firms is an important accomplishment and should not be downplayed, but as most of informality concerns informal workers working in unregistered firms, other approaches are also needed.13 Almeida (2008) discusses the experience of three manufacturing clusters located in small, interior cities of the country that were composed almost entirely of informal firms and how the firms in these clusters, working in partnership with the government through their business associations, were formalized.14 In all three cases, the government granted subsidies to the firms which were important for improving the productivity and competitiveness of the clusters. The clusters were also given grace periods for complying with the law (tax, environmental and

Janine Berg 139

labour) so as not to impair the businesses with debilitating sanctions during the transition period. In all three cases, the firms in the clusters came into compliance with the law and achieved greater financial success after the investments were made to improve their productivity. The case of the shoe production cluster in the city of Nova Serrana, Minas Gerais, is illustrative. In 1995, this business community was suffering as a result of the economic crisis and the subsequent restrictions on domestic credit. The industry sought help from SEBRAE,15 a government-run agency that provides assistance to small businesses, as well as the state government. However, neither SEBRAE nor the state government were willing to help the business association due to the high degree of non-registered businesses in the cluster. Instead the association proposed a plan that involved customized courses, given by external consultants, which would be made available to those businesses that would be willing to join the association and comply with the law. SEBRAE would finance the cost of the courses. The objective was to help the businesses find ways to cut costs, in order for them to be able to afford to register their firms and their workers. As many of these firms were underproductive, this was not difficult to do. The business association also negotiated a grace period with the regional labour inspectorate for registering their workers, under the fear that if sanctions were applied incipiently, this could have negative consequences on firm survival. The programme was successful, measured according to a variety of benchmarks. Competitiveness increased and during 2001–05, average annual GDP growth for the city was 9.1 per cent. In 1995, there were 298 registered firms in the shoe production cluster; by 2005, the number of registered firms had increased to 659. Similarly, in 1995, there were 3,121 registered workers in the cluster, jumping to 12,490 by 2005, 93 per cent of whom worked in firms that had fewer than 100 workers. Although it is not possible to say how many experiences such as this have been repeated in the country, Almeida (2008) does provide evidence that 58 per cent of the increase in formal manufacturing employment between 2000 and 2005 occurred in municipalities with fewer than 100,000 inhabitants. But perhaps the more important lesson to be drawn from the case studies is how to go about formalizing firms and their workers. Almeida (2008) argues that for small businesses, many of which are uncompetitive, the command-and-control model runs the risk of incapacitating firms. Inspection is needed, but it needs to be undertaken in conjunction with other government agencies and the business community, so that the support can be given to these

140

Regulating for Decent Work

businesses for developing a high-road competitiveness strategy that ensures respect of the law. Indeed, it seems that this approach is being more widely used in Brazil. Greater legal awareness – The case of domestic workers Despite the overall increase in informality in the 1990s, domestic workers experienced a notable increase in their rate of formalization during this decade. In 1992, according to PNAD data, only 19.4 per cent of domestic workers had a signed labour card. By 1999, there had been a 34 per cent increase in the rate of formalization, with the number of registered domestic workers rising to 26.0 per cent. Formalization continued to increase during the 2000s, but at a much slower rate, reaching a peak of 28.1 in 2007, only to fall to 26.8 per cent in 200816 (see Figure 5.6). Historically, domestic workers have been excluded from labour legislation in Brazil. The Consolidation of Labour Laws (CLT – Consolidação das Leis do Trabalho) of 1943, which governs the employment relationship, specifies in article 7 that the rights do not apply to domestic workers. Similarly, article 5 of the 1949 law regulating weekly rest, excludes domestic workers. In 1972, Law no. 5.859 was approved, entitling

29.0

28.0 27.9

27.0

27.0

26.7 25.5

25.0 23.6

26.0

28.1

27.0 26.6

26.8

24.1

23.0

21.0

19.0

20.7 19.4 19.0

17.0

15.0 1992 1993 1995 1996 1997 1998 1999 2001 2002 2003 2004 2005 2006 2007 2008

Figure 5.6 Registered domestic workers in Brazil (%), 1992–2008 Note: Workers aged 16 years and older. The rural areas of the Northern states, with the exception of Tocantins, are excluded. Source: Author´s calculations based on PNAD.

Janine Berg 141

entitling domestic workers the right to twenty days of paid vacation and making contributions to the social security system, on the part of the employer and the domestic worker, obligatory. The Constitution of 1988 conceded even more rights to domestic workers, although at a more limited level than the rights given to other workers. Of the 34 rights given to workers, only nine were applicable to domestic workers. Nevertheless, it was an important advance as it entitled domestic workers to a series of rights that they had not previously possessed. These included the right to the minimum wage, the 13th month salary, weekly rest, thirty days’ paid vacation, maternity leave (16 weeks), advance notice of one month upon dismissal and pension. Moreover, it gave domestic workers the right to organize. Prior to the Constitution of 1988, the few workers’ organizations that existed were associations, with no legal rights. It is clear that the Constitution made the hiring of domestic workers more costly; thus it is surprising, from this perspective, that there was such a large increase in registration. It thus seems likely, as argued by Chahad and Macedo (2003), that the Constitution engendered greater respect for the law on the part of employers and made them more prudent, and that domestic workers were in a better position to exercise their rights. As Lee and McCann demonstrate in this volume, greater legal awareness is associated with better working conditions as workers are better informed and thus better positioned to negotiate. Cases of domestic workers brought to the Superior Labour Court (Tribunal Superior do Trabalho) have received media attention and since 2002, the outcomes of important cases are posted as press releases on the web page of the Superior Labour Court.17 Although many of these cases concern an infringement of rights of registered domestic workers, there have also been cases of domestic workers suing their employers for not having had a signed labour card, with the court ruling in favour of the domestic worker and demanding back pay from the employer for lost benefits.18 The Constitution of 1988 allowed domestic workers to form trade unions, giving them an additional vehicle by which to demand their rights. Because of the difficulty in unionizing workers – they work in individual households, there is no way to collect union dues – fewer than 2 per cent of domestic workers (101,170 workers in 2005) are unionized. Nevertheless, the various regional unions that do exist – as well as FENATRAD, the National Federation of Domestic Workers – have lobbied successfully for improving rights of domestic workers (Bernandino-Costa 2007). In 2001, after much pressure from the union, the government passed a law giving domestic workers the right to have the FGTS fund,

142

Regulating for Decent Work

though employers’ contributions were made optional. In 1999, the FENATRAD became an affiliate of the Central Workers’ Union (CUT), the largest trade union federation. The FENATRAD often works in partnership with the government, in particular, the Special Secretariat for Women’s Policies (SEPM) and the Special Secretariat for Racial Equality (SEPPIR), as well as various international organizations, to develop policies to improve working conditions for domestic workers.19 They also carry out grassroots campaigns, frequenting areas where domestic workers can be found, so that they can educate them about their rights, raise their self-esteem, let them know about the union, and encourage them to demand from their employer that their labour card be signed.20

Improving upon the advances in formalization Government policies have been instrumental in achieving the increases in labour market formality in Brazil during the 2000s. It is thus reassuring to know that the government has continued to institute laws and policies that can further extend formality. In 2006, the government passed a provisional measure, which was later converted into law, allowing employers of domestic workers to deduct the cost of social security contributions from their taxable income. Employers can only deduct the cost of one worker and are limited to the value of social security contributions made on the minimum wage.21 Due to existing income tax exemptions, the measure is of benefit to those employers whose monthly incomes are above R$2,500 and thus opt for the full income tax declaration. According to calculations from the Secretariat of Policies of the Ministry of Social Security, there are approximately 475,000 domestic workers who could benefit from this policy, which corresponds to roughly 20 per cent of informal domestic workers (Cabanas Guimarães 2008). Although the measure was designed as an incentive for registering workers, the policy has received little visibility as only one publicity campaign was undertaken and it was limited in scope. As a result, there has been no increase in the number of claims made since the measure was first put into effect, suggesting that the employers who are benefiting are those whom had already registered their worker. Still, an effective publicity campaign could help inform households about the measure, thus there remains the potential for this policy to be successful. Also with regard to domestic workers, the government issued a decree in September 2008 that increased the minimum age for domestic employment from 16 to 18 years of age. The decree considers domestic

Janine Berg 143

work under the age of 18 as one of the worst forms of child labour, and makes reference to ILO Convention No. 182, which defines the worst forms of child labour as ‘work which, by its nature or the circumstances in which it is carried out, is likely to harm the health, safety or morals of children’. In issuing the decree the government made reference to the occupational risks in domestic work including intense physical activity, isolation, physical, psychological and sexual abuse and long working hours. At the time of the decree, 245,000 workers aged 16 and 17 were employed as domestic workers. Employers caught employing under-aged workers are subject to a fine up to R$2,012 and must pay back-pay on unpaid benefits.22 Although inspecting homes is difficult and requires a court order, the law makes it less culturally acceptable to employ adolescents as domestic workers. Of adolescents aged 16 and 17 who were employed as domestic workers in 2006, less than 0.2 per cent had a signed labour card.23 In June 2009, the Brazilian government launched the Individual Entrepreneur Law (Lei do Emprendedor Individual), which facilitates the registration of micro businesses with up to one employee and reduces the cost of social security contributions. One of the criticisms of the social security system is that it places an unduly large burden on the selfemployed as they have to pay the contributions of both the employer (12 per cent) and the employee (8 per cent), deterring formalization. Under the new law, microentrepreneurs with annual revenues below R$36,000 per year, can legally register their business and obtain a tax identification number (known as the CNPJ) and will be exempt from federal taxes. They will have to pay social security contributions equivalent to 11 per cent of the minimum wage (or R$52 per month in 2010), but this will entitle them to pension, invalidity and maternity benefits. The government aims to register one million businesses, approximately 3 per cent of the number of uncovered, self-employed workers in the first year. Several publicity campaigns have been planned to help spur registration.

How sustainable are the advances in formalization? During the 2000s, Brazil made important progress in improving working conditions, reversing the 1990s trend of increased informality. But how sustainable are these increases? This chapter has proposed five principal explanations for the increases in formalization since the 2000s: (1) a more favourable macroeconomic environment characterized by a more competitive exchange rate and more inclusive pattern of

144

Regulating for Decent Work

growth; (2) reduced labour supply; (3) the SIMPLES law; (4) improved labour inspection and new approaches to formalization; and (5) greater legal awareness and respect for the law. Although it is not possible to ascertain the relative contribution of the each of the explanations, in this section we attempt to evaluate their importance both for the experience of the 2000s, as well as for the future. There can be improvements in working conditions without economic growth, but sustaining improvements over time requires growth. With the exception of 2003–08 when economic growth averaged 5.2 per cent annually, economic growth in the two periods has been highly volatile and not that robust. But the macroeconomic conditions of the 2000s were more favourable to job creation. During the 1990s, firms were adjusting to the opening up of the economy, but an appreciated exchange rate and restrictions on domestic credit, following the 1995 crisis, hampered their viability causing hiring to nearly halt. In the 2000s, a more favourable exchange rate and a greater availability of domestic credit, albeit at high interest rates, allowed firms to invest and expand. Domestic demand was robust due to the emergence of a new middle class that benefited from the minimum wage increases and the resumption in hiring. For the period 2010–15, the government predicts that growth will average 5.0 per cent as domestic demand continues to expand, fuelled by substantial public investments in infrastructure from the Growth Acceleration Programme (PAC) and other programmes, as well as domestic consumption. The government has taken steps to halt the appreciation of the currency, which has risen since 2004 (with the exception of the first nine months following the onset of the international economic crisis in September 2008). In October 2009, the government levied a 2 per cent financial transaction tax on foreign currency inflows for investment in the Brazilian capital markets. This measure will help, but a sustained appreciation does run the risk of hurting tradable industries, which are important employers of formal labour. The only explanation that qualifies as ‘luck’ is the demographic transition which has decreased the number of youths in the overall population, and by extension, in the labour market. Still, even with regard to youths, the government has increased secondary enrolment and has provided incentives to poor families to keep their adolescent children in school through the Bolsa Família programme. The trend in Brazil is for continued improvement in school enrolment rates, resulting not just in a better-prepared labour force but also a labour market with less supply pressure from youths and fewer precarious jobs.

Janine Berg 145

The government has taken important steps to formalize small and micro-sized businesses. The SIMPLES law of 1996 eased business registration and lowered taxes, encouraging firms that may not have registered under the previous system because of the bureaucratic hassle and expense. As such, the law opened new economic opportunities for these businesses, as they became eligible to access formal credit markets or to pursue work contracts with large, formal firms and the public sector. The 2009 Individual Entrepreneur Law is a step in the same direction. Entrepreneurs will not only be able to access the social security system at an affordable cost, but will also obtain a tax identification number that facilitates access to credit and business transactions in the formal economy. It is an important initiative, although there does remain the risk that firms, rather than hiring a worker, will prefer a contractual relationship with the entrepreneur. Government actions to minimize disguised employment relationships are therefore necessary. Brazil has only half the number of labour inspectors that the ILO recommends and the workforce would be well served by dedicating more resources to inspection. Nevertheless, Brazil’s labour inspectorate has devised strategic ways to increase the number of workers covered and registered through inspection. By resolving problems that are sectorwide, the inspectorate’s work has a far greater impact; firms within the sector may not necessarily be inspected but will learn of any adjustments that are required in order to be in compliance with the law. Efforts to work with the business community and other government agencies to improve working conditions have been fruitful and all indications are that this approach will continue in the future. Although it is difficult to assess whether there is greater respect for the law among a citizenry, the case of domestic worker formalization during the 1990s indicates that inreasing legal awareness can help change cultural patterns and encourage workers to demand their rights. Additional legislation, such as the classification of domestic work under the age of 18 as one of the worst forms of child labour, is also important for changing cultural norms.

Conclusions The experience of Brazil in the 1990s and 2000s highlights the important role of government policy in ensuring that workers reap the gains from economic growth. Brazil is one of the most unequal countries in the world and it was not until the 2000s that the country began making in-roads on reversing this trend. Still, an alarming one-quarter of the

146

Regulating for Decent Work

population lives on less than US$100 per month, despite a GDP per capita of US$8,600 per year. There is a long way to go, but progress is being made. Formal workers in Brazil enjoy important benefits and protections – pension, sickness, disability and death benefits, paid annual leave, parental leave, restrictions on working hours and a guaranteed minimum wage – that informal workers are not guaranteed and must continuously negotiate with their employers if and when they are granted. Even then, they are excluded from social security benefits, which in the case of sickness, disability or death can easily bring about financial ruin in families. It is thus extremely welcome that increasing numbers of workers are employed formally and that measures have been taken to extend social security benefits to the self-employed. The debate about labour market flexibility has been exaggerated, at least in the case of Brazil. The experience of the 1990s and 2000s does not support the claim that labour regulations caused growing informality in the economy. The strong growth in formal jobs in the 2000s, at the same time that the minimum wage nearly doubled in real terms, demonstrates that labour laws are not an impediment and that some policies, such as the minimum wage, can be important for stimulating growth and job creation. Rather than focusing on deregulating the labour market, governments should take steps to encourage firms to register their businesses and their workers, through simplifying registration, lowering taxes or providing incentives to develop high-road competitiveness strategies.

Notes 1. Measured in February 2009 reais for the City of São Paulo. Data available at http://www.dieese.org.br. In February 2010, the prevailing exchange rate was R$1.8 per US$1. 2. Measured in R$ 2007 (OIT 2009). 3. For a critique of Bosch et al. (2007), see Kucera and Roncolato (2008). 4. Interestingly, the minimum wage has also been shown to affect the salaries of informal workers (through the so-called ‘lighthouse effect’), as they negotiate salaries using the minimum wage as a reference. In 2005, 20 per cent of informal, salaried workers earned the minimum wage (Neri and Moura 2006). 5. For the purposes of analysis, I use data from the years 1992, 1999 and 2008. There was a methodological change in the PNAD data prior to 1992 that prohibits the use of earlier years. I use 1999 rather than 2000, because there was no PNAD survey in 2000, as this was a census year. The year 1999 also marks a turning point in the economy as the real was devalued substantially

Janine Berg 147

6.

7. 8.

9.

10.

11. 12.

13.

14.

15. 16.

in January 1999. Finally, 2008 was the latest year available for the PNAD survey at the time of writing. The data are restricted to workers that are 16 years or older, as per the Brazilian legislation, which prohibits the hiring of worker under the age of 16, unless the worker is on an apprenticeship contract (valid for 14–16 year olds). Studies on informality sometimes use other definitions, such as the percentage of workers who contribute to the social security system, typically because of the paucity of detailed labour market data that exists in most developing countries. By this definition, there has also been an improvement in the 2000s, whereas no gains were made in the 1990s. In 1992, 42.5 per cent of workers contributed to the social security system, in 1999, 42.3 per cent were contributing, but by 2008, the percentage had grown to 52.1 per cent. The RAIS (Relação Anual de Informações Sociais) is a registry of formal establishments and jobs, administered by the Brazilian Ministry of Labour. During 2009, the Bolsa Família Programme was expanded to 12.4 million families and average benefits were increased to R$95 per month, or approximately US$53. A 1988 amendment to the Constitution raised the legal working age in Brazil from 14 to 16 years of age; however, children aged 14–16 can be employed as apprentices. ‘Sistema Integrado de Pagamento de Impostos e Contribuções das Microempresas e Empresas de Pequeno Porte’, or in English, ‘Integrated System for the Payment of Taxes and Contributions by Micro and Small Enterprises’. For more details on the Simples Law, see Delgado et al. (2007). In 2008, there were 3,113 labour inspectors for an occupied population of 92 million or one inspector for every 30,000 workers, whereas the ILO standard for an industrializing country is one inspector for every 15,000 workers (ILO 2006). As the number of inspectors did not increase during the time period studied, the ratio of inspectors to working population actually fell from 0.45 in 1992 to 0.36 in 2008 due to the increase in the size of the workforce (OIT 2009). Other categories of informality, such as self-employed workers, do not fall under the purveyance of labour inspection because there lacks the existence of an employment relationship. The work of Almeida (2008) looks at how a wide array of government agencies worked to ensure compliance with environmental, tax and labour regulations. In many of these instances the impetus to comply with the law was not initiated through the actions of the labour inspectorate, but rather through other governmental bodies. The cases are illuminating because they demonstrate how addressing informality cannot be left solely to the labour inspectorate, but require the actions of a wide array of actors. The examples from Pires (2008 as well as chapter 12 in this volume) also support the view that in order to effectively target sector-wide problems, it is essential for the labour inspectorate to work with the business community, labour unions and other groups. Serviço Brasileiro de Apoio as Micro e Pequenas Empresas (Brazilian Service to Support Micro and Small Enterprises). The fall in formality among domestic workers in 2008 is due in part to the increase in workers working on a daily as opposed to a monthly basis. According to data from the PNAD, there has been a steady increase over time

148

17. 18.

19.

20. 21.

22. 23.

Regulating for Decent Work in the percentage of domestic workers who work as day maids, proxied by the percentage of workers who work in more than one house per week. In 1999, 16.4 per cent worked in more than one house compared with 22.8 per cent in 2006 and 25.4 per cent in 2008. In 2004, the Superior Labour Court ruled that an employment relationship can be established only for those domestic workers who work in the same place of employment for more than two days per week, thus day maids are less likely to have a signed labour card. See Diário de Justiça, 2 April 2004, Recurso de Revista 776.500/2001. See http://www.tst.gov.br/ASCS/materias_tematicas/trabalhoDomestico.php. See, for example, the case of the domestic worker who proved that she had been working for the same employer since 1975 and was granted back pay for unpaid benefits (‘Diarista em um dia na semana obtém reconhecimento de vínculo empregatício’, [‘Once-a-week daily worker obtains recognition of employment relationship’] http://ext02.tst.gov.br/pls/no01/NO_NOTICIAS. Exibe_Noticia?p_cod_noticia=8492&p_cod_area_noticia=ASCS&p_txt_ pesquisa=%20domestico]. During the first administration of the Lula government, the President of FENATRAD, Maria Crueza de Oliveria, was a member of the National Council on Women and the National Council on Racial Equality (Bernandino-Costa, 2007). ‘Apenas 544 mil de 1,6 milhão de domésticas têm carteira assinada’, InvestNews (Brazil), 26 April 2006. Social security contributions are 12 per cent of the wage. Because the measure is limited to the minimum wage, the employer would be able to reduce their income subject to taxation by R$735 in 2010 (12 per cent of R$510 per month). ‘Trabalho doméstico está proibido no Brasil para menores de 18 anos’, Folha de São Paulo , 18 September 2008. According to PNAD data for 2006, of employed adolescents aged 16 and 17, a little over 10 per cent were engaged in domestic work.

References Almeida, M. 2008. ‘Além da informalidade: entendendo como os fiscais e agentes de desenvolvimento promovem a fomalização e crescimento de pequenas e médias empresas’ [‘Beyond informality: understanding how insepctors and development agents promote the formality and growth of small and medium enterprises’], IPEA Texto para Discussãon No. 1353. Baltar, P.; Krein, J.D.; Moretto, A. 2006. ‘O emprego formal nos anos recentes’ [‘Formal employment in recent years’], CESIT Carta Social do Trabalho, No. 3, janeiro a abril. Bercovich, A. 2005. ‘Demographic discontinuities: young waves and labour market structure in Brazil’, paper presented at the XXV International Population Conference, Tours, France, 18–23 July. Bernandino-Costa, J. 2007. ‘Sindicatos das trabalhadoras domésticas no Brasil: teorias de descolonização e saberes subalternos’ [‘Domestic worker unions in Brazil: theories of decolonization and alternate knowledge’], Tese de Doutorado, Universidade de Brasília, Departamento de Sociologia.

Janine Berg 149 Bosch, M.; Goni, E.; Maloney, W. 2007. ‘The determinants of rising informality in Brazil: evidence from gross worker flows’, IZA DP, no. 2970, August. Cabanas Guimarães, M. 2008. Políticas para a Expansão da Cobertura dos Trabalhadores e Trabalhadores Domésticas: A experiência do Brasil [Policies for the Expansion of Coverage of Domestic Workers: The Brazilian Experience], Coleção Previdência Social, Vol. 26 (Brasília: Ministério da Previdência Social). Cardoso, J.C. 2007. ‘De volta para o futuro? As fontes de recuperação do emprego formal no Brasil e as condições para sua sustentabilidade temporal’ [‘Back to the future? The sources of recovery in formal employment in Brazil and conditions for its sustainability’], IPEA Texto para discussão, No. 1310. Chahad, J.P.Z.; Macedo, R. 2003. ‘A evolução do emprego no período 1992–2001 e a ampliação do mercado formal brasileiro desde 1999’ [‘The evolution of employment in 1992–2001 and the enlargement of the formal Brazilian labour market since 1999’], in Chahad, J.P.Z.; Pichetti, P. (eds) Mercado de Trabalho no Brasil – padrões de comportamento e transformações institucionais [Brazilian Labour Market – Patterns of Behaviour and Insitutional Transformation] (São Paulo: Editora LTr). Delgado, G.; Querino, A.; Campos, A.; Vaz, F.; Rangel, L.A.; Stivali, M. 2007. ‘Avaliação do SIMPLES: implicações à formalização previdenciária’ [‘Evaluation of SIMPLES: implications and formalization of social security’], IPEA, Texto para Discussão, No. 1277. ECLAC. 2009. Social Panorama of Latin America (Santiago: ECLAC). Frenkel, R.; Ros, J. 2006. ‘Unemployment and the real exchange rate in Latin America’, World Development, vol. 34, no. 4, pp. 631–46. González, D. 2006. ‘Regímenes especiales de tributación para pequeños contribuyentes en América Latina’ [‘Special tax regimes for small contributors in Latin America’], Banco Interamericano de Desarrollo (http://idbdocs.iadb.org/ wsdocs/getdocument.aspx?docnum=703919). Instituto de Pesquisa Econõmica Aplicada (IPEA). 2009. PNAD 2008: Primeiras análises [First Analyses], Comunicado da Presidência num. 30, 24 September. International Labour Office (ILO). 2002. Decent Work in the Informal Economy, Report VI, International Labour Conference, 90th Session (Geneva: ILO). ILO. 2006. Strategies and Practices for Labour Inspection, GB.297/ESP/3, 297th Session of the ILO Governing Body, November (Geneva: ILO). ILO. 2009. Bolsa Família in Brazil: Context, Concept and Impacts, background document prepared for March 2009 Governing Body (Geneva: ILO). Kucera, D.; Roncolato, L. 2008. ‘Informal employment: two contested policy issues’, International Labour Review, vol. 147, no. 4, pp. 321–48. Marshall, A. 2004. ‘Labour market policies and regulation in Argentina, Brazil and Mexico: programmes and impacts’, Employment Strategy Paper, No. 2004/13 (Geneva: ILO). Monteiro, J.; Assunção, J. 2006. ‘Coming out of the shadows: estimating the impact of bureaucracy simplification and tax cut on formality and investment’. Paper presented at the 2006 European Meeting of the Econometric Society. Neri, M. 2008. A Nova Classe Média [The New Middle Class] (Rio de Janeiro: Fundação Getulio Vargas). Neri, M.; Moura, R. 2006. ‘Brasil: La institucionalidad del salario mínimo’ [‘Brasil: The institutionalization of the minimum wage’], in Marinakis, A.; Velasco, J.J. (eds) ¿Para qué sirve el salario mínimo? [What is the purpose of the minimum wage?] (Santiago, OIT).

150

Regulating for Decent Work

Organização Internacional do Trabalho (OIT). 2009. O Perfil do Trabalho Decente no Brasil [Decent Work Profile for Brazil] (Brasília: OIT). Paiva, L.H. 2009. ‘Os impactos redistributivos do projeto de lei que elimina o fator previdenciário’ [‘The redistributive impact of the proposal of law to eliminate the ‘social security factor’], in Schwarzer, H. (org.) Estudos e Pesquisas da Secretaria de Políticas de Previdência Social [Research and Studies from the Secretariat of Social Security Policies] (Brasília: Ministerio da Previdência Social). Pires, R. 2008. ‘Promoting sustainable compliance: styles of labour inspection and compliance outcomes in Brazil’, International Labour Review, vol. 147 (2–3), pp. 199–229. Saboia, J. 2007. ‘Salário minimo e distribução da renda no Brasil no periodo 1995–2005: Fatos e simulações’ [‘The minimum wage and income distribution in Brazil in the 1995–2005 period’: Facts and simulations’]. Paper presented at the seminar on Análisis da PNAD, CGEE/IPEA/MEC/MTE, March, Brasília.

6 Labour Market Regulations and the Welfare of Indonesian Workers Iyanatul Islam

Introduction There is a remarkable – and largely unremarked – asymmetry in the way economists treat the benefits of regulations that seek to protect private property rights and regulations that seek to protect labour rights and standards. The former are seen as essential for growth; the latter are treated as a hindrance. The intellectual imprint of this idea is reflected in both the global and the Indonesian context. It is necessary to understand the Indonesian variant of the global discourse on labour market regulations in order to develop a fuller appreciation of the current and prospective policy agenda on the labour market. This provides the rationale for the current chapter. The discussion focuses on the work of influential critics who contend that the fundamental goal of sustainable employment creation in Indonesia is being starkly compromised by a raft of labour regulations that ostensibly seek to protect the rights of workers. Such regulations ought to be held responsible for poor labour market outcomes in the post-1997 era. One witnesses a combination of relatively high unemployment, persistence of underemployment and informalization of the economy even more than a decade after the end of the 1997 financial crisis. This represents a sharp contrast to the virtuous combination of rising employment, growth in real wages, and structural transformation in the pre-1997 era. The implied policy message is that if the government is serious about resurrecting the economic successes of the past Indonesia should roll back the labour market regulations that it has enacted in recent years. How credible is this narrative in the case of the Indonesian labour market as it is today and as it has evolved over time? The chapter 151

152 Regulating for Decent Work

contends that, on both analytical and empirical grounds, the thesis that current labour regulations entail deleterious consequences on the welfare of Indonesian workers is difficult to substantiate – a conclusion that is in line with the global evidence on the impact of labour market flexibility.1 This does not necessarily mean that current labour laws and institutional arrangements are either appropriate or effective in responding to the particular needs of the Indonesian economy. The challenge is to seek the right balance between the imperatives of growth and employment creation and the goal of protecting and respecting labour rights.

The case against labour market regulations: what the proponents say Those who are concerned about the negative employment consequences of labour market regulations subscribe – implicitly or explicitly – to the paradigm of labour market flexibility. The latter may be regarded as a synonym for the textbook case of a competitive or neoclassical labour market. Workers are free to move in response to shifting relative wage and employment opportunities, while firms are free to adjust the workforce in response to shifting profit opportunities. In such a framework, collective bargaining driven by trade unions, strongly enforced hiring and firing rules, restrictions on casual and flexible modes of employment, severance pay, unemployment benefits, minimum wages and other regulations are regarded as undesirable institutional arrangements as they constrain the free choice of workers and firms. More importantly, arrangements meant to protect the interests of workers paradoxically hurt employment opportunities by raising the cost of labour above what the market will bear. Hence, proponents of labour market flexibility subscribe to the ‘cost-push’ hypothesis in diagnosing poor employment outcomes. Advocates of labour market flexibility also argue that labour regulations can mitigate the capacity of the growth process to reduce poverty on a sustainable basis. Broad-based employment creation is a key mechanism for reducing income poverty, but regulations impede that process by raising the cost of labour. The income-earning opportunities of young, first-time job-seekers and female workers who typically participate in assembly operations in the manufacturing sector are likely to be restrained as they might be forced to seek low-paid employment in the unregulated sector. Gender wage differentials could widen in tandem with rising inter-sectoral differentials.

Iyanatul Islam 153

Labour market flexibility and the Indonesian experience: the conventional interpretation The principles of labour market flexibility in Indonesia were adapted to the requirements of paternalistic authoritarianism that characterized the Suharto regime (1965–98). The emphasis was on the provision of centrally mandated benefits for workers (especially minimum wages and, later, a formal social security system). This was juxtaposed within a political framework that tightly circumscribed labour rights, while informal systems of social protection through the network of friends and families were seen as playing an adequate and complementary role for workers in the informal economy. Such a strategy appeared justified by the rapid growth of the period prior to the 1997 financial crisis and its ability to absorb new entrants to the workforce.2 Some observers argue that the legacy of labour market flexibility inherited from the Suharto era enabled workers to cope with the 1997 financial crisis with a degree of resilience that ensured that the social impact of the crisis was contained. Had labour markets been rigid, one would have witnessed large-scale unemployment and poverty.3 Several scholars are also paying renewed attention to the virtues of labour market flexibility in post-Suharto Indonesia. There is now considerable concern that contemporary Indonesia is typified by onerous labour legislation that has seen the aggressive pursuit of minimum wages and other provisions, such as severance pay. This strategy has raised real wages to unsustainable levels, impeded employment growth in the formal sector and had an adverse impact on the investment climate. These developments are largely seen to stem from the Manpower Act of 2003. The law was the product of a more politically liberal and democratic environment that emerged in the wake of the 1997 financial crisis. It encouraged popular expectations of rising entitlements. Unlike the past, when the political regime was highly centralized, government authorities can no longer simply repress trade unions and civic activities. The current industrial relations system seems to represent an exaggerated reaction to past repressions of labour rights and entitlements. Hence, this is reflected in the vigorous application of minimum wages policy, rather generous severance pay, restrictions on the use of temporary employment contracts and production outsourcing. This creates a high-cost business environment that acts as a brake on much-needed private investment from domestic and external sources. Inadequate investment in turn acts as a brake on formal sector employment generation.

154 Regulating for Decent Work

Here are some typical examples of this viewpoint. A USAID-sponsored study (Aaron et al. 2004, p. 61) concludes: Several recent and pending developments in Indonesia’s labour market policies and practices threaten to raise the costs of hiring new workers and in turn slow modern employment creation. These developments include minimum wage legislation, severance pay regulations, restrictions on use of temporary employment contracts and production outsourcing. These regulations, if implemented in a restrictive manner, will reduce new job opportunities in the modern sector and encourage smaller firms to migrate to the informal sector to avoid excessive compliance costs … While recent developments in minimum wage policy adjustments at the local level have been more rationale [sic] and based on wider stakeholder consultations than in previous years, pending changes to the minimum wage regulations [to be implemented in 2006] threaten to push up wages beyond inflation and productivity gains, and will also contribute to an uncertain and confrontational industrial relations environment. Another study concurs: ‘There has been a dramatic deterioration in the Indonesian labour market performance since the Asian financial crisis mainly due to the decline in investment and a much more regulated labour market’ (Sugiyarto 2005). Yet another observer castigates Indonesia for laws ushering in ‘hiring and firing business practices [that] have created a dysfunctional labour market’ and the solution really is to ‘abolish rigid labour markets’ (Hanke 2007). The following proclamation emanates from one of the most wellinformed scholars of the Indonesian labour market: We find that while the protection of labour freedoms is long overdue, there is mounting evidence that regulation of setting labour standards … benefits the few with ‘better jobs’. It penalises many less fortunate Indonesians in the informal sector and agriculture, and also younger, new, job seekers … [T]he compliance regime in relation to labour standards has altered dramatically in recent years … It is of concern especially in those internationally labour-intensive industries, such as textiles, footwear and clothing … where Indonesia has had a comparative advantage in the past. (Manning 2004, p. 1; emphasis added)4 The ‘cost-push’ hypothesis appears to have influenced the Indonesian government in 2005. This is reflected in the current Medium-Term

Iyanatul Islam 155

Development Plan which includes, as one of its objectives, the restoration of labour market regulations in a more flexible direction. In 2006, the government sought to reform the Manpower Act of 2003, but backed down in response to union-led opposition – much to the dismay of some observers of the Indonesian labour market.

What the proponents of labour market flexibility ignore: the limitations of the competitive model and their implications The textbook model of a perfectly competitive market for labour services that informs the worldview of the proponents of labour market flexibility either ignores or pays insufficient attention to a number of analytical complications and their implications. This stems from various aspects of the basic model. Thus: 1. The model is static, being valid for a full-employment, but zerogrowth, economy with no demographic or structural changes. Hence, labour demand and labour supply can be held constant. 2. The model assumes the absence of any market power exercised by employers. 3. The model ignores efficiency wage considerations. 4. The model ignores the vulnerability of workers to labour risks. 5. There is perfect compliance when regulations are enacted and implemented. 6. No ‘learning’ takes place among key stakeholders when the ‘rules of the game’ change. Relaxation of some or all of these aspects and elements of the basic model will mitigate its strong predictive power. The assumptions of a zero-growth economy and lack of demographic and structural changes are clearly not valid for the Indonesian economy. The (downward-sloping) labour demand curve is derived in the conventional model from the production function. Growth in this framework can be depicted as a shift in the production function due to, say, increases in the investment rate, or increases in factor productivity. This will also cause a rightward shift in the labour demand curve. In the pre-1997 period, as part of an investment-driven process, the Indonesian economy grew at around 7 per cent per annum. In recent years, growth has been in the 5–6 per cent range. The (upward-sloping)

156 Regulating for Decent Work

labour supply curve has also shifted in response to rising labour force growth and urbanization. Under these circumstances, it is important to make a distinction between a movement along a given labour demand curve interacting with a static labour supply curve and shifts of both demand and supply. A perfectly enforced minimum wage and other non-wage labour standards that are above the ‘market-clearing’ level will cause involuntary unemployment and/or increase the incidence of job search unemployment. This can be conceptualized as an upward movement along a given labour demand curve interacting with a given labour supply curve. In a rapidly growing economy undergoing demographic and structural transformation, labour demand and supply will experience progressive shifts, thereby offsetting the regressive impact of minimum wage and other exogenously imposed labour standards. This dynamic framework offers a better representation of the evolution of the Indonesian labour market than the conventional model. It is also well known that monopsonistic employers5 will set wages below the marginal productivity of labour while employment will be below the competitive norm. In these circumstances, a fully enforced minimum wage and other exogenously imposed labour standards can increase wages and employment. One study of Indonesian rural labour markets has claimed that observed wages for the sample that was studied were below the estimated marginal labour productivity of workers – evidence that is consistent with monopsonistic practices.6 The basic model assumes that atomistic, profit-maximizing employers set wages that correspond to exogenously determined productivity. Yet it has been known for some time that productivity can be endogenously determined by wages.7 This is the basis of the efficiency wage hypothesis. Profit-maximizing employers might consider the market-determined wages to be too low and raise them systematically above these market-determined norms in order to motivate workers to perform to higher standards and thus recoup the cost of higher wages through higher productivity. One should not expect any enforced change in the behaviour of employers if minimum wages and other non-wage minima are set at or below the efficiency wage. In this case, involuntary unemployment and search unemployment will be the result of the conduct of profit-maximizing employers rather than government intervention. Efficiency wage considerations are relevant to Indonesia, as they are in a number of developing countries.8 Some observers have noted that

Iyanatul Islam 157

the practice of multinational companies in Indonesia and elsewhere of routinely paying wages that are above market rates for workers with comparable attributes is compatible with the prevalence of efficiency wage norms (Brown et al. 2002). There is an implicit normative message in the basic model of a perfectly competitive labour market that ‘any job is better than no job’ – a message that has powerful adherents.9 Such a view discounts the value of fortifying job security, because the conventional model ignores the vulnerability of workers to labour market risks. In the absence of formal risk-mitigation schemes, workers could be induced to readily accept low-productivity jobs at low wages. This could propel a developing economy into a ‘low productivity, low pay’ equilibrium in which ‘bad jobs’ drive out ‘good jobs’. Insecure workers behave rationally in this framework. In the absence of any job security and legal protection – such as laws against unfair dismissal, the absence of unemployment insurance – workers pay a premium (in the form of low wages and the ready acceptance of any job) to employers in order to reduce the risk of being unemployed. Under such circumstances, the imposition of higher labour standards and various risk-mitigation schemes – such as unemployment insurance – could be both efficient (leading an economy towards ‘high productivity, high wage equilibrium’) and equitable (enabling vulnerable workers to deal with labour market risks). Despite decades of rapid growth in the pre-1997 period, and respectable growth in the post-1997 period, well over a third of Indonesians are still vulnerable to at least a transient episode of poverty. This means that even if the current incidence of the ‘working poor’ is moderate, many more workers are vulnerable to labour market risks entailing spells of unemployment and underemployment engendered by a variety of economic shocks. The 1997 financial crisis as well as the global recession of 2008–2009 has demonstrated that these risks can be very significant. Yet if one uncritically embraces the virtues of labour market flexibility then formal schemes to mitigate labour market risks will not form part of the policy agenda. Issues such as unfair dismissals and unemployment insurance will then be treated as impediments to the job creation process. After all, its adherents would argue, ‘any job is better than no job’. A core assumption of the basic analytical framework that undergirds the thesis of labour market flexibility is that minimum wages and other regulations can be perfectly enforced. Hence, the compliance regime is infallible in its execution. Yet non-compliance, even in developed countries, is widespread. As soon as one introduces the empirically plausible assumption of non-compliance, the strong results that one can

158 Regulating for Decent Work

derive from the basic model will no longer hold. It is thus unsurprising that critics of the current labour market regulations in Indonesia emphasize that the compliance regime has changed ‘dramatically’ in recent years. Yet they also concede that minimum wages and other regulations apply primarily to the formal sector but not the informal sector. In that case, one can show that the impact on overall employment of an economy becomes ambiguous: it depends upon the magnitude of the minimum wage increase and the ‘substitution effect’ reflected in the changing composition of employment between the formal and informal sector (Fields 2004). The issue of non-compliance and its implications have received renewed attention in the analytical literature on minimum wages. One insightful study considers the consequences of inserting imperfect compliance in models of imperfect competition and juxtaposes it with the assumption of imperfect commitment on the part of policy makers. It demonstrates that the wage–employment trade-off becomes more complex (Basu et al. 2007). One of the results derived from analytical model has considerable relevance to the Indonesian case. Consider an economy where there is surplus labour and/or productivity is low. Consider too a political process that yields a government that has egalitarian preferences and thus uses an increase in the minimum wage as a signal to make a credible commitment to enforce the new wage regime. It can be shown that equilibrium wages and employment will both rise in these particular circumstances, a result that hinges on the assumption of imperfect compliance and the credible commitment of a distribution-sensitive government rather than on the assumption of imperfect competition. How does this particular analytical result relate to the Indonesian case? Given its level of development, productivity is still low relative to richer, labour-scarce economies. The political process changed in the post-1997 era. As noted, and as readily accepted by the proponents of labour market flexibility, governments in a newly democratic Indonesia are sensitive to issues of poor labour standards and the ‘working poor’ unlike the pre-1997 era in which the government responded to labour issues largely in response to external pressures. In other words, the degree of enforcement in the post-1997 era is endogenous in nature. Hence, these features of the Indonesian labour market and political economy are thus consistent with some analytical results derived in the aforementioned study. In sum, it is not possible to assert a priori that an increase in minimum wages will inevitably create the wage–employment trade-off that is envisaged in a model of perfect competition and perfect compliance.

Iyanatul Islam 159

The assumption of perfect compliance and infallible enforcement by policy makers entails the corollary that no ‘learning’ takes place among key stakeholders as the ‘rules of the game’ change; they inevitably do when there are major changes in the institutional and regulatory environment. Even the most populist government does not resolutely pursue regulations if the consequences are such that they have an adverse impact on the welfare of large sections of the community. If, as argued, perfectly enforced statutory minima lead to significant unemployment while rewarding formal sector workers, then a distribution-sensitive government will have to manage the trade-off between improving the living standards of formal sector workers and looking after the interests of the unemployed and the informally employed. The tensions that are unleashed in managing this trade-off might well be significant enough to ensure that they attenuate a resolute political commitment to labour market regulations. There are many subtle ways in which a government might ‘signal’ that they are not rigidly committed to statutory minima. One way might be to leave the onerous minimum wage legislation intact on paper, but allow the real value of minimum wages to fall.10 Firms in the formal sector can also become adept at circumventing labour regulations, without necessarily breaching legal limits. Alternatively, firms can seek to recoup the higher cost of complying with regulations by extracting a higher supply of effort from existing workers or by investing in human and physical capital that can boost productivity (see Fraja 1999). In other words, firm-level responses are likely to be much richer than just layoffs and a reduced demand for labour services. Workers too can adapt to changing circumstances by aiming for more realistic wage settlements, if the alternative of pushing for a strict adherence to onerous labour regulations raises the risk of pushing business to bankruptcies. In sum, when one allows for ‘learning’ among key stakeholders to take place as they develop coping mechanisms to deal with large-scale changes in the institutional and regulatory environment, the wage– employment trade-offs might, over the course of time, become muted. As will be argued, such an evolutionary framework is important in understanding the labour market experiences of post-1997 Indonesia.

Labour market regulations in Indonesia and the welfare of workers: an empirical evaluation The discussion so far has delivered the following message: the textbook model of a competitive labour market that provides the intellectual

160 Regulating for Decent Work

inspiration for advocates of labour market flexibility has the appeal of simplicity, but suffers from the inability to recognize many analytical complications. Its strong predictive power is attenuated when one considers the dynamic framework of a growing economy undergoing demographic and structural transformation, the issue of efficiency wages, the adverse implications of job insecurity, the implications of non-compliance, and the fact that a process of ‘learning’ takes place among key stakeholders when the regulatory environment changes. One could argue, despite its limitations, that the conventional model of the labour market can still be treated as a valid characterization of the Indonesian case as long as the evidence is consistent with its key predictions. How robust is the evidence in favour of the view that labour market regulations are the primary determinant of employment outcomes? Consider, first, the case of minimum wages. The authority to set minimum wages is now located at the regional level, although the implementation is expected to take place within national guidelines. It is clear that minimum wages have gone up sharply, especially since the 1997 financial crisis. For example, the provincial minimum wage rose from 61 per cent of the average wage to 69.4 per cent of the average wage over the 2001–6 period. The available data suggest that, during 2005–6, minimum wages grew faster in regions with above-average unemployment rates vis-à-vis regions with below-average unemployment rates (Islam and Chowdhury 2009, chapter 2). In terms of severance pay, it is now one of the highest in the region. A World Bank ‘Doing Business’ survey ranks Indonesia quite adversely in terms of its ‘employment rigidity index’ – it occupied a rank of 154 out of 178 economies. When these findings are combined with rising unemployment (as officially measured) between 2000 and 2005 and secular decline in Indonesia’s share of labour-intensive exports,11 one is tempted to conclude that the predictions of the basic model of a competitive labour market are valid. Such a diagnosis in turn would lead one to support the government’s effort to review and modify the onerous provisions stipulated in the Manpower Act of 2003. Yet the evidence is not as clear-cut as it appears. Hence, it needs to be reviewed with due diligence. The evaluation that follows is offered in several stages. First, it draws attention to the stresses and strains that emerged from the way in which the Indonesian labour market was governed in the pre-1997 period. Second, it questions the view that labour market flexibility acted as a safety valve during the 1997 financial crisis. Third, it revisits the

Iyanatul Islam 161

proposition that the 2003 Manpower Act led to a ‘dramatic’ change in the compliance regime with rather adverse consequences. The pre-1997 characterization of the Indonesian labour market overlooks a number of problems. To start with, some observers, while sympathetic to the notion of labour market flexibility, drew attention to the stresses and strains afflicting the pre-1997 period. One study highlighted the limits of a paternalistic system that offered substantial centrally mandated minimum wage increases in the 1990s to appease workers, while others noted its deleterious consequences on employment creation.12 Another study highlighted the problems caused by the lack of a credible industrial relations system. The mid-1990s were infamous for a sharp increase in labour unrest (see Edwards 1996). These studies argued for the revamping of the industrial relations system and recommended democratic forms of enterprise-level bargaining. Indeed, at the time there was considerable external pressure on the Indonesian government to change its industrial relations system brought about by a combination of exhortations by the US administration, ‘antisweatshop’ activism and international media attention (Harrison and Scorse 2005). As far as the use of the notion of labour market flexibility in explaining the evolution of poverty both during and after the crisis is concerned, one can express reservations about its validity (Dhanani and Islam 2001a). In a so-called rigid (or Keynesian) labour market, the primary vehicle for a rise in poverty following a demand shock (such as the one induced by the Indonesian crisis) is a rise in unemployment (given that wages are rigid). In a neoclassical labour market operating in a dualistic, developing economy – presumably the Indonesian case – the primary vehicle for a rise in poverty because of a demand shock is a decline in real wages reflecting a reallocation of labour to low-productivity activities. Indeed, employment adjustments during the Indonesian crisis of 1997 followed this classic pattern and were accompanied by a steep fall in real wages. Whether the unemployment-driven increase in poverty (the Keynesian case) will be higher than the real wage-driven increase in poverty (the neoclassical case) is ultimately an empirical issue. It cannot be determined a priori. A better way of coming to grips with the understanding of the evolution of poverty during the crisis is to eschew preconceived notions of the virtues of labour market flexibility and carefully re-visit the current state of knowledge on what could have happened to the poor during the 1997 financial crisis and the severe recession that

162 Regulating for Decent Work

followed in 1998. It now appears that the sharp increase in consumption poverty and the plummeting of real wages during the 1998 recession was a tragic manifestation of a collapse in the purchasing power of Indonesians because of the ‘inflation shock’ of 1998 when there was a huge, but temporary, surge in the price of key essentials, including rice. If one combines the 1998 episode with the fact that a significant proportion of Indonesians are clustered around the poverty line, then relatively small shifts in the latter may lead to large changes in the incidence of consumption poverty. However, the sharp rise in poverty (and the steep fall in real wages) during 1998 turned out to be transient as inflation abated (and became deflation for a while), reflecting the restoration of macroeconomic stability. In addition, the government launched a comprehensive, donor-supported, social safety programme. Some elements of the programme worked well, others did not. On balance, it appears that the social safety net interventions managed to reinforce the resilience of ordinary Indonesians in the face of unprecedented adversity. The evidence suggests that social safety net interventions probably prevented an additional 7–11 per cent of the population from sliding below the poverty line.13 What about the deleterious consequences of the 2003 Manpower Act? To start with, there is little evidence that this act and the paraphernalia of regulations attached to it have had a long-term impact on: (i) the aggregate unemployment rate; (ii) the share of informal employment; and (iii) the duration of job search. Admittedly, there is an increase in the official unemployment rate, but this appears to date to be transient – a pattern that has presumably led a World Bank report to suggest that the effects of the 2003 Act might be ‘wearing off’ (World Bank 2007, p. 12). As can be seen in Figure 6.1, the unemployment rate rose sharply between 2000 and 2005, but then declined significantly – despite the global recession of 2008–9. What about the impact of the new labour regulations on informal employment? The advocates of labour market flexibility would argue that onerous regulations affect both the overall unemployment rate and the composition of employment: affected workers who are unable to gain access to formal employment will then be ‘forced’ to participate in the informal economy. Figure 6.2 traces the evolution of informal employment using both the ‘old’ definition and the ‘new’ definition’ as adopted by the Indonesian statistical agencies.14 As can be seen, the share of informal employment rises between 2001 and 2003 and then tapers off between 2006 and 2009. Hence, as in the case of aggregate unemployment rates, the interpretation of the

Iyanatul Islam 163 Indonesia: aggregate unemployment rate, 2000−2009 12 10

%

8 6 4 2 0 2000

2002

2004

2006

2008

Figure 6.1 Indonesia: aggregate unemployment rate, 2000–2009 Sources and notes: Statistics Indonesia (BPS). Note that 2009 refers to the February survey. All others pertain to August.

2001

2003

2006

2009

72 70

%

68 66 64 62 60 58 56 Old def

New def

Share of informal employment (% of total work-force), Indonesia,

Figure 6.2 The changing share of informal employment in Indonesia, 2001–2009 Sources and notes: Extracted from data reported in Nazara (2010, p. 22). ‘Old def’ ⫽ old definition; ‘new def’ ⫽ new definition. The old definition (applicable before 2001) focused exclusively on status of employment (permanent/temporary, paid/unpaid, and so on). The new definition (applicable after 2001) combines status of employment with occupational categories (professional, administrative, clerical and so on).

evidence on labour market performance in Indonesia depends upon the particular subperiod of the 2000s. What about the impact of new labour market regulations on the duration of job search? Once again, under a radically different regulatory

164 Regulating for Decent Work

regime, as proclaimed by its critics, one would expect an increase in the wage premium attached to the formal sector. This would create an incentive for young graduates to engage in a more extensive job search. Yet there is little evidence that, at the aggregate level, the duration of job search has increased in recent years: it is no higher than it was during the 1990s (Dhanani et al. 2009, chapter 3). Critics of the current regulatory regime usually note that it exacerbates wage inequality by penalizing the income-earning opportunities of vulnerable workers, most notably female workers. One study, for example, estimates that the employment elasticity with respect to minimum wages for female workers is the highest compared with economywide estimates (Suryahadi et al. 2003). Yet female unemployment rates are not necessarily higher than males when these rates are measured appropriately. Moreover, gender wage disparities have steadily improved in Indonesia, even though inter-sectoral disparities have widened along some dimensions (Dhanani et al. 2009, chapter 5). Another way to test the claim that the compliance regime with respect to labour standards has increased sharply in recent years is to focus on the perceptions of the business community. How do they perceive the impact of labour regulations on the business climate in Indonesia? How important is it relative to other factors that influence the business climate? Is it getting better or worse over time and relative to other countries? There are now a number of surveys – three of which are global in nature – that allow one to explore these issues in greater depth. Consider the aforementioned World Bank’s ‘Doing Business’ survey. It enlists labour market regulations as only one of the ten factors that affect the overall index of ‘ease of doing business’ in which Indonesia ranks 123rd out of 178 economies. The clear implication is that reforming labour regulations without tackling other impediments will have only a marginal impact on improving the business climate. More importantly, as Table 6.1 shows, the values of the ‘rigidity’ indices that pertain to different components of employing workers are virtually constant between 2003 and 2007. If the proxies for the key explanatory variable, in this case rigid labour market regulations, remain invariant and behave essentially like fixed parameters, then they cannot be marshalled into explaining changes in either unemployment or the structure of employment. While the pertinent indices in terms of the World Bank’s ‘Doing Business’ survey remain constant, a more optimistic picture emerges from another global survey. In terms of a well-known global competitiveness survey as carried out by the World Economic Forum for 2007–8,

Iyanatul Islam 165 Table 6.1 Indices pertaining to employing workers in Indonesia, 2003–2007

Difficulty of hiring index Rigidity of working hours Difficulty of firing index Rigidity of employment index Firing costs (% of salary)

2003

2004

2005

2006

2007

72 0 60 44 10

61 0 60 44 10

72 0 60 40 10

72 0 60 44 10

72 0 60 44 10

Source: World Bank ‘Doing Business’ survey, available at www.doingbusiness.org.

0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 Year [2000−2007]

Figure 6.3 Indonesia: global competitiveness index (1 = lowest rank; higher rank shown as declining fraction), 2000–2007 Source: Calculated from the World Economic Forum, Global Competitiveness Index, various years.

Indonesia ranks 50th out of 131 countries – an improvement of 19 places from the previous survey. Within that aggregate index, Indonesia is ranked 30 out of 131 countries in terms of a ‘labour efficiency’ measure, once again showing substantial improvements over previous years. Figure 6.3 charts Indonesia’s progress between 2000 and 2007 in terms of the World Economic Forum’s ‘global competitiveness index’, where Indonesia’s actual rank is expressed as a fraction of the total number of observations. A declining fraction is shown as an improvement in Indonesia’s relative rank in terms of competitiveness.15 The fact that this steady progress took place despite the contentious Manpower Act of 2003 is quite striking, given the standard allegation that a more regulated labour market implies a less competitive business environment.

166 Regulating for Decent Work Indonesia: Labour market regulations index, 2000−2005 6 5 4 3 2 1 0 Year 2000 2001 2002 2003 2004 2005

Figure 6.4 Indonesia: labour market regulations index, 2000–2005 Note: The index is constructed from data pertaining to minimum wages, restraints on hiring and firing decisions, mandated costs of firing and dismissals, the incidence of centralized collective bargaining and conscriptions or forced labour. This index is one of five sub-indices that make up the overall ‘economic freedom’ index and is measured out of 10 (best score). Source: Derived from data reported in Fraser Institute (www.freetheworld.com).

Yet another global survey produced by the neoliberal Fraser Institute can be utilized to track whether labour regulations have worsened in recent years (see Figure 6.4). The trend that emerges is one in which the overall index of labour market regulations has actually improved between 2000 and 2005, in tandem with the overall improvement in ‘Economic Freedom’ over the same period.16 As in the case of the ‘Doing Business’ survey and the WEF survey, labour regulations represent only one component out of many that affect the overall ‘Economic Freedom’ index that is claimed by its authors to affect growth and living standards. Hence, focusing only on labour regulations as the source of Indonesia’s constraints on competitiveness is inappropriate. It is also worth emphasizing that attempts to establish even simple statistical associations between indices of labour market rigidity and investment performance using a cross-country sample (in which data on Indonesia are included) do not yield robust results. In other words, there is a statistically weak correlation between high degree of labour market flexibility and either net inflows of FDI (as a percentage of GDP) or gross fixed capital formation (as a percentage of GDP). For example, the correlation coefficient between (ranked) indicators of labour market rigidity and FDI is a statistically insignificant –0.31 and a mere –0.07 for gross capital formation (Chatani 2009). One can now turn to Indonesia-specific surveys that seek to elicit pertinent information on changes in the business climate. Surveys of

Iyanatul Islam 167

Summary rating

(A) Level (out of 10) 4.9

5.1

5.1

6.1

6.4

6.4

5.9

5.5

5.8

6.1

6.0

6.3

1970 1975 1980 1985 1990 1995 2000 2001 2002 2003 2004 2005

(B) Ranking 1970 1975 1980 1985 1990 1995 2000 2001 2002 2003 2004 2005

43 of 47 of 64 of 28 of 31 of 45 of 79 of 99 of 89 of 77 of 82 of 86 of 54 72 105 111 113 123 123 123 123 127 130 141

Figure 6.5 Indonesia: economic freedom index Source: Fraser Institute (www.freetheworld.com).

investor perceptions by both the Asian Development Bank (ADB) and the University of Indonesia show that labour problems are not at the top of investors’ concerns. Such problems typically occupy a middleranking position in a list of 12 factors that are seen as key influences on the investment climate. The 2007 survey of the investment climate by the University of Indonesia reveal that only a small proportion of firms (20 per cent and less) experienced various labour-related problems and that this proportion declines between 2006 and 2007. Furthermore, the same survey shows that the costs of handling labour problems increased slightly from 3.5 per cent to 4.5 per cent of total production costs between 2005 and 2006, but declined by 2007 (see Figure 6.6). The view that current labour regulations have created unsustainable cost pressures in the Indonesian economy is difficult to sustain when one takes account of the following evidence. Unit labour costs – a widely used measure of cost competitiveness – have fallen since 2003 and are now at their lowest level since 1993 (see Figure 6.7). This reflects a combination of both rising labour productivity, in manufacturing and elsewhere, and roughly constant labour costs – developments that are contrary to the claims made by the proponents of the ‘cost-push’ hypothesis. One could argue that appealing to the aggregate evidence is inadequate in resolving the validity of the critique that current labour regulations are ultimately prejudicial to the interests of ordinary workers. More robust statistical evaluations are required. One way of conducting such an

168 Regulating for Decent Work (A) Percentage of respondents experiencing labour problems Mid 2005

End 2005

Mid 2006

Mid 2007

25

%

20 15 10 5 0 Demand for higher salary/wage increase

Worker demonstration

Social security problems

Conflict with labour union

Worker strike

%

(B) Cost of handling labour problems (% of total production cost) 5 4.5 4 3.5 3 2.5 2 1.5 1 0.5 0 Mid 2005

End 2005

Mid 2006

Mid 2007

Figure 6.6 Labour problems in Indonesia Source: LPEM-University of Indonesia, Survey on Investment Climate Monitoring, 2007.

evaluation is to consider several econometric studies that have been conducted to identify the employment consequences of minimum wages. Table 6.2 summarizes the findings. While they typically show negative employment consequences, they also find that the minimum wage– employment trade-off is absent in the case of large firms, skilled workers and the export sector. One study also attaches the important caveat that the net impact of minimum wages on employment depends upon macroeconomic conditions. A rapidly growing economy can absorb the rise in labour costs engendered by minimum wages and other institutional factors. In any case, the ‘partial’ elasticity with respect to minimum wages (or its surrogates) at the nationwide level is typically quite small, ranging from 0.1 to 0.03, and some of them are based on

Iyanatul Islam 169

Unit labour index(1993=100)

110

100

90

80

70

60 Agriculture

Manufacturing

50 1993 1994 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

Figure 6.7 Unit labour costs in Indonesia (1993–2006: base year ⫽ 1993) Source: Islam and Chowdhury (2009, chapter 2).

Table 6.2 Minimum wages and employment: summary of findings Author/Study

Key findings

Islam and Nazara (2000)

Statistical findings on the relationship between minimum wages and employment are ambivalent and are influenced by macroeconomic conditions. Employment elasticity with respect to output growth is typically significantly higher than employment elasticity with respect to wages There is no statistically significant impact of minimum wages on employment in the large firms in the formal sector, but the employment creating potential of smaller firms are hurt by minimum wage increases. Doubling minimum wages reduce employment overall by about 10 per cent, but this is not valid for white-collar workers. The negative employment effects are higher for vulnerable workers (young, female, unskilled/less educated). Minimum wages reduce overall employment growth. 15 per cent increase in minimum wages lead to net loss of 298,000 jobs Minimum wages are an ineffective anti-poverty tool

Alatas and Cameron (2003)

Suryahadi et al. (2003)

Widianto(2003)

Bird and Manning (2005)

(Continued )

170 Regulating for Decent Work Table 6.2

(Continued)

Author/Study

Key findings

Rama (1996)

Doubling minimum wages reduce employment by 10 per cent. Updated results show a much more modest impact (5 per cent at best). Once again, the effects are absent in the case of large firms. Plant-level analysis show that minimum wage legislations and ‘antisweatshop’ activism led to sharp real wage increases in 1990–1996 but was accompanied by significant increases in both wages and employment in TFA sector, driven by export-oriented foreign firms. Overall employment fell; estimates suggest that a doubling of real minimum wages lead to 12–18 per cent decline in employment. Firm-level data for 2003–4 show that manufacturing sector employment declines by 2.5 per cent if minimum wage is doubled. Training provided to workers also falls if minimum wages doubled. The reported results are valid only for unskilled workers, but not skilled workers.

Harrison and Scorse (2005)

Sugiyarto and Enriga (2007)

employment functions that barely pass conventional statistical tests of significance. The estimates are thus neither large in magnitude nor robust enough to serve as a solid basis for policy prescriptions. Why, despite the rather ambitious nature of the Manpower Act of 2003, and the media attention that has been given to it, is one unable to discern any significant impact on aggregate labour market indicators and the investment climate as revealed in perceptual data gleaned from the business community? Labour laws, like all other laws, will work – whether in a desirable or undesirable direction – if there is a substantial degree of either voluntary or involuntary compliance. Critics of labour market rigidities usually assert that there has been a major increase in the compliance regime in post-1997 Indonesia in relation to the enforcement of labour standards, but rarely offer pertinent evidence to substantiate their claim other than highlighting the onerous nature of the regulations and emphasizing that the wages of a small number of workers in the formal sector fall below the statutory minimum. One could argue that this evidence alone does not constitute compliance – with its implication of a forced change in behaviour. As has been argued previously, efficiency wage considerations could also be at work if the statutory minimum is at or below the efficiency wage. This is perhaps one reason why econometric studies in the Indonesian case consistently find that

Iyanatul Islam 171

the wage–employment trade-off does not seem to be relevant in the case of large firms. Even if one assumes that the current statutory minima, with respect to both wage and non-wage conditions, deviates from efficiency wage norms, firms, especially large ones, are able to attenuate the stringent contractual obligations to their employees as envisaged by existing labour legislation. This means that the compliance regime, even in the case of the formal sector, is in practice lower than what it seems. It could also mean that the higher costs associated with the new regulations can be recouped through a variety of adjustment measures rather than layoffs and a reduced demand for labour services. Furthermore, while employers become adept at coping with the regulatory environment, workers can also adapt to changing circumstances and are prepared to aim for more realistic wage settlements if the risk is that business will otherwise go bankrupt. It is, at the same time, unrealistic to assert that political actors never learn from their experiences. Why would even a distribution-sensitive government insist on a draconian compliance regime if the risk is significant open unemployment? The Indonesian government has demonstrated clearly that it is highly sensitive to the issue of open unemployment. In sum, when one allows for the plausible proposition that key stakeholders in the economy seek to learn the ‘rules of the game’ in a new institutional and political setting – as in the case of post-1997 Indonesia – a rigid adherence to a regulatory regime becomes unlikely. It is worth restating the point that labour regulations represent only one variable out of a range of factors that influence the business climate. In the case of Indonesia, issues pertaining to macroeconomic stability, corruption and governance, and infrastructure play an even bigger role as determinants of the perceived business climate. An ADB (2005) study summarizes this point rather well: It is crucial to emphasize that … Indonesia’s new labour regulations cannot be the only elements that are constraining the growth of formal sector employment. In fact, available data suggest that factors other than labour regulations are of greater importance. In the first place, [a] survey of manufacturing enterprises … found macroeconomic instability, policy uncertainty, corruption, current tax rates, and costs of financing to be listed more often than labor regulations as major or severe constraints to their business. Second, and in a related vein, Indonesia’s business regulations can by no means be absolved of blame. A World Bank study of such regulations from around the world shows that

172 Regulating for Decent Work

procedures for starting and closing a business in Indonesia are among the most cumbersome in Asia … Of the 15 DMCs covered by the study, the number of days it takes to start a business was 151 – the highest – compared with 41 days in the PRC, and only 8 days in Singapore (the lowest in Asia). Similarly, the time it takes to close a business due to bankruptcy was estimated to be as high as 6 years (the second highest in Asia after India). The corresponding periods for the PRC and Singapore (again the lowest) are 2.4 years and 0.8 years, respectively. Finally, a widely used indicator of governance … shows a deterioration of the quality of governance in Indonesia on five out of six dimensions between 1996 and 2002. Taking all of this other evidence into account, the conclusion that may be drawn is that, while certain elements of new labor regulations may reduce formal firms’ incentives to hire, other factors – ones that a large scale survey of firms’ managers themselves list as more important – are constraining the expansion of the formal sector. It is these other factors that are especially important in dragging down investment and employment in the formal sector. (p. 53, emphasis added) If the thesis of the deleterious consequences of labour market rigidities does not rest on a robust empirical foundation, what alternative hypothesis can be proffered? Employment outcomes are the product of both demand-pull and cost-push factors. Emphasizing the latter to the exclusion of the former risks the adoption of a diagnostic framework that implies monocausality – that is, only labour costs matter in hiring decisions. This is clearly not the case. Policy prescriptions emerging from such a diagnostic framework will be flawed. In order to test the influence of both demand-pull and cost-push factors in influencing employment outcomes, the chapter draws on econometrically estimated employment equations in which employment is a function of GDP growth (representing demand-pull factors) and real product wages (representing cost-push factors) plus a term that represent ‘path dependence’ (that is, current employment is also influenced by past employment). The equations are estimated separately for the pre-1997 period (1993–7) and the post-1997 period (2000–6) across eight major sectors of the economy.17 The results show that, while in the pre-1997 era, the wage–employment relationship was statistically significant in five of the sectors, in the post-1997 era, in only two sectors (mining and utilities and manufacturing) which have relatively small employment shares, the real wage–employment nexus is statistically significant. These econometric estimates also

Iyanatul Islam 173

reveal, for example, that to engender a 1.8 per cent increase in manufacturing employment would entail a substantial cut in average real wages (by about 10 per cent) – a prescription that makes little sense in an economy with significant poverty and vulnerability. In contrast, the employment–GDP relationship is statistically significant in almost all sectors in the post-1997 period. In other words, demandpull factors (via the growth channel) appear to be more significant than labour costs in influencing employment outcomes. This interpretation is consistent with the stylized facts. Employment growth has been faster in 2006–7 at a time when GDP growth has been faster relative to previous periods. Taking a longer-run perspective, structural changes in the Indonesian economy have not been conducive to broad-based employment creation. The export share of labour-intensive products has been on a secular decline since 1993 (see Figure 6.8). The timing of the decline suggests that minimum wages and mandatory benefits were not the key factors, given that concerns about the negative aspects of these regulations are primarily associated with the Manpower Act of 2003. External conditions, most notably Indonesia’s inability to compete with other major exporters of labour-intensive products such as China, could have played a more significant role. Within the non-tradable segment of the economy, the declining GDP share of labour-absorbing sectors has not been compensated by robust employment elsewhere. Investment – a key component of aggregate demand – has generally been subdued in the post-1997 period, and made a negative contribution to overall growth of GDP.

45 40 35

(%)

30 25 20 15 10 5 0 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

Figure 6.8 Export share of unskilled labour-intensive manufacturing in Indonesia Source: The figure showing the export share of labour intensive manufacturing exports was supplied by Dr Andin Hadiyanto, Director of the Business Climate Research and Dev. Centre, Ministry of Trade.

174 Regulating for Decent Work

Concluding remarks This chapter evaluated the influential view that the regulatory environment is a key determinant of labour market outcomes. Those who extol the virtues of labour market flexibility draw attention to the virtuous combination of rising employment, sustained real wage growth and structural change that characterized the pre-1997 period. Workers paid a price in terms of the attenuation of labour rights, but benefited from the material gains engendered by a flexible labour market. Others argue that a competitive labour market acted as a safety valve during the 1997 financial crisis. The current policy discourse is characterized by the narrative that the various regulations that are embedded in the governance of the Indonesian labour market has altered dramatically to the detriment of the fundamental goal of broad-based employment creation. This view is in tune with a global agenda in which labour market reforms are seen by influential international organizations as a key element of a pro-growth, pro-jobs, pro-poor agenda. The analytical and empirical foundation of the thesis that labour market regulations represent the primary determinant of labour market outcomes is not robust – a finding that is in line with the global evidence. The worldview of those who seek labour market reforms overlooks the many analytical innovations that challenge the strong assumptions underpinning the basic model of a fully flexible and competitive labour market. When one allows for growth and demographic and structural changes that can lead to shifts in labour demand and supply, monopsonistic practices, efficiency wages, vulnerability to labour market risks, imperfect compliance, and a process of ‘learning’ by key stakeholders, it is easy to show that the strong predictive capacity of the basic model can no longer be relied on to make policy prescriptions. It is perhaps unsurprising, given the analytical complications that emerge as soon as one deviates from the basic model of labour market flexibility, that the Indonesian experience needs to be interpreted with considerable care and caution. The successes of the pre-1997 period cannot hide the fact that stresses and strains began to emerge in the 1990s in an industrial relations system that was subservient to the imperatives of paternalistic authoritarianism. Even those who are sympathetic to the cause of the agenda of labour market flexibility argued the need for a change in direction. The proposition that a competitive labour market can be credited for acting as a safety valve during the crisis overlooks alternative explanations, and gives insufficient credit to the role that

Iyanatul Islam 175

government interventions played in mitigating the consequences of the crisis. The dire prognosis of the ramifications of labour regulations in the post-1997 seems to be valid when the current labour laws – as encapsulated in the 2003 Manpower Act – are seen in isolation. The key conclusion of this chapter is that there is little evidence that either basic labour market indicators or the investment climate has changed in any adverse fashion at least as perceived by stakeholders in the business community. Some surveys indicate an improvement in the investment climate – and even an improvement in a generic labour market index. The aggregate official unemployment rate and the share of informal employment display a transient change; the duration of job search remains roughly constant; gender wage gaps have improved despite predictions from econometric models that they should worsen. Data on both unit labour costs and the wage–productivity relationship for 2000–2007 do not show any ‘pro-labour’ bias. It is likely that those who express alarm at the deleterious consequences of labour regulations in Indonesia are unduly influenced by observations drawn from a particular period (2003–2005) while ignoring alternative evidence and developments beyond 2005. The critics of current labour regulations seem to assume that the compliance regime with respect to enforcement of labour standards has gone up sharply in recent years and also imply that labour costs are the primary factor in hiring decisions, thus overlooking the importance of other variables in affecting the business environment and in determining labour market outcomes. Both these positions are not tenable. The chapter concluded that labour regulations might appear rather onerous on paper, but might mean little in practice. Large organizations in the formal sector can circumvent stringent contractual obligations to workers while staying within the limits of the legal framework. According to several studies, minimum wages do not represent a major factor in the hiring decisions of large firms – evidence that seems to be consistent with efficiency wage considerations. In addition, firms that operate in agriculture as well as the informal economy remain outside the realm of formal regulations. Hence, if compliance is limited in practice, the adverse effects from labour regulations are not expected to materialize in any systematic way. The chapter also noted that the importance of labour regulations should be assessed in relation to other variables that affect the investment climate and growth. When judged within a relative framework, one detects that business stakeholders ascribe considerable importance

176 Regulating for Decent Work

to issues such as the quality of governance as binding constraints on growth. The clear implication is that, even if were possible to adopt a fully flexible labour market, one cannot expect the resurrection of a growth miracle in Indonesia if other structural constraints – poor business regulations, low-quality governance, poor infrastructure – remain in place. Evidence extracted from estimated employment functions also suggest that demand-pull factors – emanating from GDP growth and structural change – are often more important as determinants of employment outcomes than labour costs. What policy implications follow from the above findings? The chapter clearly does not endorse a ‘minimal regulations’ approach, but neither does it mean an endorsement of a ‘business as usual’ scenario. It is necessary to recognize that stringent labour legislation seems to be irrelevant for the many millions of Indonesians who work in the informal economy. It is important to appreciate that even as large firms develop coping mechanisms to deal with an ambitious agenda of enforcing labour regulations, they incur some transactions costs for doing so. The challenge is to find institutional arrangements that can strike the right balance between the imperatives of growth and employment creation and the need to protect and respect labour rights and fortify the capacity of ordinary workers to cope with labour market risks. Possible innovations include a well-designed and well-funded unemployment benefits scheme and a less coercive approach to minimum wages that can largely replace the currently contentious and costly severance pay scheme and defuse the tensions surrounding the regime of minimum wages. These are issues that are worthy of exploration in future research.

Notes 1. The cross-country evidence is reviewed in Berg and Kucera (2008). 2. Manning (1998), Wiebe (1996), Mason and Baptist (1996) provide a generally positive evaluation of labour market performance under the Suharto regime. 3. As Manning (2000, p. 108) observes: ‘I conclude that labour markets remained highly flexible despite the rapid economic transformation. I suggest that this flexibility is a key explanation for why unemployment and poverty did not rise more than they did during the crisis’ (emphasis added). See also Tubagus (2000). 4. Manning’s views on the impact of wages on competitiveness reflect a widely held position among other economists. For example, Brown et al. (2002, p. 10) note: ‘In our judgment ... there is reason to believe that labor-intensive manufacturing in developing countries is relatively sensitive to changes in wage levels. This is particularly true for the production of apparel and footwear, which are prototype “footloose” industries.’

Iyanatul Islam 177 5. See Manning (2005) for a treatise on monopsonistic employers and their implications for labour market outcomes. Manning argues that in a monopsonistic labour market, the ‘free’ market can no longer be regarded as an ideal and should not be used as the sole basis for policy prescriptions. 6. Ezelea-Harrison (2005). One should attach the caveat that the presence of monopsonistic employers is the not the only explanation of the misalignment between wages and marginal productivity. 7. See Yellen (1984) and Akerlof and Yellen (1986, introduction). 8. See Abbas and Zaman (2005) for the Pakistani experience. 9. Harrison and Scorse (2005, p. 1) include Jagdish Bhagwati, one of the luminaries of the economics profession, and Nicholas Kristoff, influential New York Times op-ed writer, among the ranks of those who are wary of imposing higher labour standards in developing countries because of the ‘any job is better than no job’ mindset. 10. See Freeman (1993) who was among the first to assemble evidence to show that real minimum wages fell in the 1980s across a wide range of developing countries with diverse regulatory arrangements. 11. The evidence is reviewed in more detail at a later juncture. 12. See Aggrawal (1995). See Rama (1996) who claims that a doubling of minimum wages leads to a 10 per cent decrease in employment. Manning and Junankar (1998, p. 89) contend that the observed increase in aggregate unemployment in the mid-1990s is ‘probably related to the minimum wage policy’, but concede that it has ‘much to do with changing survey definitions’. The employment consequences of minimum wages are discussed more fully in the main text. 13. See Dhanani and Islam (2001b) for a thorough review. 14. See notes to Figure 6.2 for clarification of alternative definitions. 15. Of course, a relative rank as measured here can show an improvement even if there is no change in the absolute rank. For example, if a country has a rank of 75 out of 100, its relative position will improve simply by increasing the number of observations, say, 75 out of 120 countries, if the additional observations are ranked below the country. In the case of Indonesia, an improvement is observed in both the absolute and relative ranking. 16. The key ingredients of economic freedom, as conceptualized by the Fraser Institute, are: personal choice, voluntary market exchange, freedom to enter and compete in markets, protection of persons and property from aggression by others. The index is made up of five sub-indices: size of government; ‘sound’ money; legal structure and security of property rights; free trade; regulations on labour, credit and business. The indices have a maximum value of 10, representing the best score. The late Nobel Laureate Milton Friedman, the doyen of neoliberals, was associated with the inaugural work of the Institute. 17. See Dhanani et al. (2009, chapter 6, appendix 6.2) for details on methodology and results.

References Aaron, C.; Kenward, L.; Bird, K.; Desai, M.; Aswicahyono, H.; Basri, C.M.; Tubagus, C. 2004. ‘Strategic approaches to job creation and employment in Indonesia’. Report prepared for USAID (Jakarta: USAID).

178 Regulating for Decent Work Abbas, S.K.; Zaman, A. 2005. ‘Efficiency wage hypothesis: the case of Pakistan’, The Pakistan Development Review, vol. 44, no. 4, pp. 1051–66. Aggrawal, N. 1995. ‘Indonesia: labor market policies and international competitiveness’, World Bank Policy Research Working Paper No. 1515 (Washington, DC: World Bank). Akerlof, G.A.; Yellen, J.L. 1986. Efficiency Wage Models of the Labour Market (Cambridge: Cambridge University Press). Alatas, V.; Cameron, L. 2003. ‘The impact of minimum wages on employment in a low-income country: an evaluation using the difference-in-difference approach’, World Bank Policy Research Working Paper No. 2985 (Washington, DC: World Bank). Asian Development Bank (ADB). 2005. Asian Development Outlook (Manila: ADB). Basu, A.K.; Chau, N.H.; Kanbur, R. 2007. ‘Turning a blind eye: costly enforcement, credible commitment and minimum wage laws’, IZA Discussion Paper No. 2998 (Bonn: Institute for the Study of Labour (IZA)). Berg, J.; Kucera, D. (eds) 2008. In Defence of Labour Market Institutions: Cultivating Justice in the Developing World (London and Geneva: Palgrave and ILO). Bird, K.; Manning, C. 2005. ‘Minimum wages and poverty in a developing country: simulations from Indonesia’s household survey’. Departmental Working Paper (Canberra: Research School of Pacific and Asian Studies, Australian National University). Brown, D.K.; Deardorff, A.V.; Stern, R.M. 2002. ‘The effects of multinational production on wages and working conditions in developing countries’, RSIE Discussion Paper no. 483. Presented at NBER/CEPR/SNS Conference, International Seminar on International Trade (ISIT), ‘Challenges to Globalization’, Höberge Gård, Stockholm, 24–5 May. Chatani, K. 2009. ‘Labour market regulations and investment’, Policy Note, October (Jakarta: ILO). Dhanani, S.; Islam, I. 2001a. ‘Labour market adjustments to Indonesia’s economic crisis: a comment’, Bulletin of Indonesian Economic Studies, vol. 37, no. 1, pp. 1211–31. Dhanani, S.; Islam, I. 2001b. ‘Poverty and vulnerability in a period of crisis: the case of Indonesia’, World Development, vol. 3, no. 7, pp. 1211–31. Dhanani, S.; Islam, I.; Chowdhury, A. 2009. The Indonesian Labour Market: Changes and Cchallenges (London and New York: Routledge). Edwards, S. 1996. ‘Labour regulations and industrial relations in Indonesia’. Paper presented to the Ministry of Manpower–World Bank workshop on Indonesian Workers in the 21st Century, Jakarta, 2–4 April. Ezelea-Harrison, F. 2005. ‘Two-tier wage systems in agriculture: evidence from Indonesian microdata’, South Western Economic Review, vol. 32, no. 1, pp. 1–13. Fields, G.S. 2004. ‘Dualism in the labour market: a perspective on the Lewis model after half a century’, The Manchester School, vol. 72, no. 6, pp. 724–35. Fraja, D.G. 1999. ‘Minimum wage legislation, work conditions and employment’, CEPR Discussion Paper No. 1524 (London: Centre for Economic Policy Research). Freeman, R. 1993. ‘Labour market institutions and policies: help or hindrance to economic development?’, Annual Proceedings of the World Bank Conference on Economic Development (Washington, DC: World Bank), pp. 117–56. Hanke, S.H. 2007. ‘Abolish rigid labour markets’, July (Washington, DC: Cato Institute).

Iyanatul Islam 179 Harrison, A.; Scorse, J. 2005. ‘Improving the conditions of workers? Minimum wage legislation and anti-sweatshop activism’, California Management Review, vol. 48, no. 2, pp. 144–60. Islam, I.; Chowdhury, A. 2009. Growth, Employment and Poverty Reduction in Indonesia (Geneva: ILO). Islam, I.; Nazara, S. 2000. ‘Minimum wages and the welfare of Indonesian workers’, Occasional Discussion Paper No. 3 (Jakarta: ILO). Manning, A. 2005. Monoposony in Motion: Imperfect Competition in Labour Markets (Princeton, NJ: Princeton University Press). Manning, C. 1998. Indonesian Labour in Transition: An Economic Success Story? (Cambridge: Cambridge University Press). Manning, C. 2000. ‘Labour market adjustments to Indonesia’s economic crisis: Context, trends and implications’, Bulletin of Indonesian Economic Studies, vol. 36, no. 1, pp. 105–36. Manning, C. 2004. ‘Legislating for labour protection: betting on the weak or the strong?’, Working Paper in Trade and Development 04/08, Division of Economics (Canberra: Research School of Pacific and Asian Studies, Australian National University). Manning, C.; Junankar, R. 1998. ‘Choosy youth or unwanted youth? A survey of unemployment’, Bulletin of Indonesian Economic Studies, vol. 42, no. 2, pp. 143–70. Mason, A.; Baptist, J. 1996. ‘How important are labour markets to the welfare of the poor?’ Paper presented to the Ministry of Manpower–World Bank workshop on ‘Indonesian Workers in the 21st Century’, Jakarta, 2–4 April. Nazara, S. 2010. ‘The Indonesian informal economy: a case study’. Report prepared for the ILO (Jakarta: ILO). Rama, M. 1996. ‘The consequences of doubling the minimum wage: the case of Indonesia’, World Bank Policy Research Working Paper no. 1643 (Washington, DC: World Bank). Sugiyarto, G. 2005. ‘Creating better and more jobs in Indonesia: a blueprint for policy action’, ERD Policy Brief no. 3 (Manila: ADB). Sugiyarto, G.; Enriga, B. 2007. ‘Does minimum wage reduce employment and training?’ (Manila and Los Banos: ADB and University of Philippines). Suryahadi, A.; Widyanti, W.; Perwira, D.; Sumarto, S. 2003. ‘Minimum wage policy and its impact on employment in the urban formal sector’, Bulletin of Indonesian Economic Studies, vol. 39, no. 1, pp. 29–50. Tubagus, F. 2000. ‘The impact of the crisis on the labour market in Indonesia’. Report prepared for the ADB, Manila, 22 March. Widianto, B. 2003. Making the Most of Minimum Wages Policy (Jakarta: BAPPENAS). Wiebe, F. 1996. ‘Income insecurity and underemployment in the Indonesian informal sector’. Paper presented to the Ministry of Manpower–World Bank Workshop on ‘Indonesian Workers in the 21st Century’, Jakarta, 2–4 April. World Bank. 2007. Indonesia: Economic and Social Update (Jakarta: World Bank). Yellen, J.L. 1984. ‘Efficiency wage models of unemployment’, American Economic Review, vol. 74, no. 2, pp. 200–5.

7 The Enactment of Three New Labour Laws in China: Unintended Consequences and the Emergence of ‘New’ Actors in Employment Relations Fang Lee Cooke

Introduction In the last two decades economic globalization and changing ideologies on the role of the state have led to different responses from governments in respect of their regulatory role in employment relations (for example, Martínez Lucio and Stuart 2004; Bamber et al. 2010). In western economies, this has often taken the form of de-regulation or ‘re-regulation’, as MacKenzie and Martínez Lucio (2005, p. 500) argued. There is also an inclination to move away from a hard, i.e. regulatory, approach towards a softer, that is, a voluntary, approach to managing employment relations through the adoption of innovative schemes such as social partnership (for example, Martínez Lucio and Stuart 2004). In China, the change of government leadership to Premier Wen Jiabao and President Hu Jintao in 2003 marked the beginning of the pursuit of an economic development policy that emphasizes the areas of social justice, social harmony and environmental protection. This is an important departure from the efficiency-driven economic development policy pursued by their predecessors typically influenced by the economic thinking of Deng Xiaoping – the architect of modern Chinese economic development. The primary objective of the state intervention in employment relations and human resource management (HRM) practices is twofold: to provide a greater level of protection to the workers, and, relatedly, to facilitate enterprises to establish harmonious employment relations as part of its agenda to build a harmonious society (see Li and Xiang 2007; Warner and Zhu 2010). 180

Fang Lee Cooke 181

In 2007, labelled the legislative year of China, the government stepped up its legislative activities, passing three major pieces of employment-related laws to take effect from 2008: the Labour Contract Law (LCL) enacted on 1 January 2008; the Employment Promotion Law (EPL) enacted on 1 January 2008; and the Labour Dispute Mediation and Arbitration Law (LDMAL) enacted on 1 May 2008. These three pieces of new labour legislation are interconnected and cover different stages of the employment relationship, that is, from recruitment to the termination of the employment contract. More specifically, the EPL aims to promote employment and secure the employment rights of workers; the LCL seeks to regulate the rights and responsibilities of employer/labour user organization and the worker; and the LDMAL is a procedural law that aims to ensure the fulfilment of these rights. The promulgation of the LDMAL is to support the implementation of the LCL. The legislative necessity and urgency of the above laws arises after a gap of more than 12 years since the introduction of China’s first major piece of labour legislation – the Labour Law (passed in 1994 and enacted on 1 January 1995). During this period, significant changes have taken place in the economic structure and employment environment. These changes have led to the divergence of goals and interests of actors in employment relations – a heightened level of conflicts between these interests and the surge of power on the part of the employers. The promulgation of the new laws signals the government’s renewed and stronger determination to raise the level of protection to its workforce to counterbalance the prerogative of employers. Employees are afforded greater power to seek justice through the legal channel when these laws are violated by employers. Together, these labour laws and their supplementary regulations provide a legal framework within which employment relationship is governed and labour market regulated in principle (see Figure 7.1). The primary objective of their implementation is to achieve a more efficient and equitable labour market. In parallel, a system for labour dispute resolution is formed, albeit far from being robust. It has been argued that, with ‘the major exception of freedom of association’, the labour standards established by the series of labour laws and regulations of China ‘are not markedly inferior to those of comparable countries and indeed many developed nations’ (Cooney 2007, p. 674). What remains most problematic is the lack of effective enforcement (Taylor et al. 2003; Cooney 2007). While implementation failures are characteristic of all regulatory systems (Cooney 2007), the

182 Regulating for Decent Work

Chinese system is frustrated by the multiplicity of employment-related laws, directive regulations and administrative policies issued at central, provincial and municipal government level, the ambiguous status of some of these regulative instruments, and the confusing channels through which workers can seek to secure compliance of laws (ibid.). Scholars on labour disputes in China have also argued that regional institutional arrangements, managerial styles and the characteristics of the workforce influence the level of labour disputes (for example, Chan 2001; Lee 2007). The promulgation of the three new laws by the central government therefore raises an important set of questions that have not yet been addressed. What is the legislative environment for introducing employment regulations in China, and what institutional barriers may impede its implementation? Given the brief history and limited experience of the Chinese state in relation to employment legislation, does it have sufficient legislative capacity to provide an effective regulatory framework for the labour market and employment relations? Are these laws able to bring to an end the fragmented authoritarianism that has been an enduring and key feature in the economic and social development parallel with the central state control in the post-Mao era? Similarly, will these new laws be tight enough to circumvent the growing ability of employers, particularly since the 1990s, to bypass or override regulatory constraints in pursuing their business objectives? What may be the impact of these regulations on the workers and other institutional actors? What new actors may have emerged in the regulatory process and how do the interactions of these institutional actors shape the legislative outcomes? This chapter contributes to the existing understanding of the role of employment regulations in employment relations in China by addressing the above questions. It explores the dynamics of state policy, regulation and interactions amongst institutional actors in relation to the implementation of the new legislation at the organizational and institutional level against a broader social and politico-economic context in China. Using the LCL and LDMAL as the focal point for discussion, the chapter examines the response of employers, workers and the trade unions to the new regulations and the emergence of ‘new’ institutional actors in the regulating process. These include: employers’ associations, employment agencies, HR consultancy firms, foreign client firms and non-governmental organizations (NGOs). In one sense these ‘new’ actors are not really ‘new’ because some of them have existed for a number of years. What is new is their more direct and active role in employment relations at

Fang Lee Cooke 183

the workplace/organizational level as an unintended consequence of the enactment of the new labour laws in 2008. The chapter adopts Bellemare’s (2000, p. 386) definition of an actor in an industrial relations (IR) environment as ‘an individual, a group or an institution that has the capability, through its action, to directly influence the industrial relations process, including the capability to influence the causal powers deployed by other actors in the IR environment’. It also adopts Michelson’s (2008, p. 27) argument that these actors do not necessarily have to be influential at all three levels – that is, the workplace, organization and institutional levels – at all times. Given the nature of the study, a qualitative method was adopted in order to capture in-depth information. Information came from two main sources: first-hand empirical data collected by the author between September 2007 and April 2009 and secondary data. Semi-structured interviews were conducted with 71 informants from Beijing, Wuhan, Shanghai and Guangdong Province of China. They include: 43 (owner) CEOs and HR directors/managers of state-owned, private and foreign-owned firms, seven HR consultants, three labour authority officials, two union officials, two leading employment relations/labour law scholars in China, and 14 migrant workers from seven factories. In addition, informal interviews were held between the author and five workers who were involved in three labour disputes arbitration cases.

Background, key focus and intended legislative progress of the three new laws Employment Promotion Law (EPL) The dual reasons for the promulgation of the EPL are the needs to promote employment and regulate the labour market (also see chapters by Berg, Islam, and Frey in this volume on similar pressure for the governments of other developing countries). First, there is mounting pressure for the government to create employment opportunities for the newcomers in the labour market (for example, rural migrant workers, school leavers and university graduates) and for those who have been displaced by their employer as a result of technological advancement, intensifying global competition, industrial structural change and ownership reforms. Creating employment and enhancing employment security is a crucial way to secure the livelihood of the workers and their family because of the lack of social security provision for the majority of the Chinese population. A second reason for the need for the law is that the absence of laws on employment rights, which is a precondition for other labour

184 Regulating for Decent Work

rights, has led to a disorderly labour market management situation. In the absence of a central law, there are a proliferation of local administrative policies and regulations. These are often confusing, contradictory and unprofessional and have a low level of enforceability (Han and Liu 2008). The EPL was therefore promulgated in 2007 to take effect from 1 January 2008. The EPL is essentially a regulation that promotes employment security. It contains nine chapters and 69 articles. Key aspects covered include: policy support, fair employment, employment services and administration, occupational education and training, employment assistance, monitoring and inspection, and legal liability. There are four key focuses in the EPL (see Figure 7.1). First, it emphasizes the role of local governments in promoting employment, vocational education and skill training through a range of measures and mechanisms. These include, for example, funding vocational education and training institutes, initiating training programmes for rural migrant workers, monitoring the training budgets and activities of enterprises and imposing penalties on non-compliant firms. Second, the EPL seeks to create a healthy labour market by cleaning up the employment agency/job centre industry. A number of rules are introduced to define the scope of services of employment agencies and to regulate the behaviour of employment agencies and user firms. Forced closure and/or financial penalties may be applied to violating agency operators. A third focus of the EPL is fairness in employment – a point that is also specified in the LCL. The EPL specifies that employers cannot discriminate on the grounds of ethnicity, race, gender, religious belief and disability, or against groups such as carriers of infectious pathogens or candidates of rural origin. Although equal opportunity rights have been stipulated in the Constitution (1954, 2004) and the Labour Law, it is the EPL that makes the most specific and comprehensive statement, relatively speaking, for the first time in China’s employment law history, about the types of workers who should not be discriminated against. The expanded categories of vulnerable groups of workers indicate that China’s employment law is not only becoming more liberal in recognizing changing social characteristics of its labour force, but also converging to the international trends of equal opportunity and diversity management (Cooke 2011b). In addition, a remedial channel is outlined for legal liability. A fourth focus of the EPL is to provide employment assistance through the provision of favourable policy conditions for entrepreneurship (self-employment) (for example, Han and Liu 2008; Liu 2008).

Fang Lee Cooke 185 Employment Promotion Law

Labour Contract Law

Labour Law

• Role of local governments in promoting employment, vocational education and skill training • Regulating employment agencies to clean up the industry • Anti-discrimination in recruitment • Encourage entrepreneurship (self-employment) through favourable conditions

• Formation of contracts • Circumstances for the termination of contract and compensation • Emphasis on non-fixed term contract • Social security contribution

•

People’s court (For discrimination cases)

•

Applicable to those who have formal employment relationship with employers Focusing on the termination but not the formation of contract

Enforcement connections/ implementational impacts of the laws on each other Administrative channel for labour disputes settlement • Labour inspection and monitoring authority

Labour Disputes Mediation and Arbitration Law • Emphasis on mediation instead of arbitration • Single arbitration as final ruling for certain labour disputes to speed up settlement • Extension of the period in which one can apply for arbitration (from 60 days to 12 months) • Shortening the time for conclusion of arbitration (within 60 days) • No fees charges for arbitration • Burden of proof shifting towards employers

Legal procedure for labour disputes settlement Mediation

Arbitration

Litigation

• Labour disputes mediation committee at the enterprise level;

• Labour disputes arbitration commission at district, county and municipal level

• People’s court

• People’s (civil) mediation body legally set up at the grassroots level; • Labour disputes mediation body at community and district level

Figure 7.1 Major labour market and employment-related laws in China and key elements

Labour Contract Law (LCL) Similarly, two main reasons have led to the introduction of the LCL. First, the legal system that governs employment relations has been flimsy, relying largely on the Labour Law which applies mainly to those in the formal employment sector with formal employment relationships (Hu 2004; Cooney et al. 2007; Dong 2008). It fails to address the growing tensions emerging from the privatized economy and the growing informalization of employment, such as casual work and agency work (known as labour dispatch in China). There is considerable ambiguity as to whether certain laws and regulations should apply to the informal sector and workers in informal employment. Employers also tend to

186 Regulating for Decent Work

take advantage of these regulatory loopholes and argue for exemption (Cooke 2008). Despite the lack of consensus on its definition, it is agreed that informal employment has become a significant form of employment in China, employing an estimated 20 per cent of the total workforce (for example, Peng and Yao 2004; Shi and Wang 2007; Wu 2008). Again, the legislative void is filled in part by a range of administrative regulations and legislative instruments adopted by local governments (Cooney et al. 2007). Whilst these interventions have had some effects, they are essentially administrative regulations that have limited consistency, authority and enforceability (Hu 2004). A second related reason is that in recent years there have been increases in both the number of labour dispute cases and the number of workers involved (see NBS 2008; also see Cooke 2008 for a detailed analysis). A notable parallel development is the increasingly individualistic nature of the disputes. In 2001, collective labour disputes cases (that is, cases that involved three or more people) made up 6 per cent of the total cases and 61 per cent of the total number of workers involved in the disputes. In 2007, collective labour disputes cases made up 3.7 per cent of the total cases and 42 per cent of the total number of workers involved in the disputes (NBS 2008). Main forms of non-compliance on the part of the employers include: the non-provision of an employment contract, the under payment of wages, wage arrears, excessive overtime and failure to make contributions to social insurance premium. According to the statistics from labour inspections, less than 20 per cent of the small and medium-sized private firms had signed contracts with their workers. This figure was much lower in the self-employed business sector (cited in Chang 2008). Since the Labour Law only covers those in formal employment, this leaves the majority of workers in de facto employment relationship unprotected. For those who have signed employment contracts, the majority of contracts were just one year in duration. The short-term nature of the contractual relationship renders employment relations unstable at the macro level. According to a survey conducted by the labour authority in April 2005, nearly 13 per cent of the workers were paid below the local minimum wage level. Some firms reduced the real wage by unilaterally reducing the unit price of production or raising the production targets which forced workers to work unpaid overtime to complete their tasks. Wage arrears and the unlawful deduction of wages were the most common violation cases revealed in the labour inspections (cited in Chang 2008). A primary objective of the LCL is therefore to create a formal and

Fang Lee Cooke 187

stable employment relationship between the workers and employing organizations through a tighter definition of labour contract arrangements so that employers cannot easily evade their responsibilities at the expense of the workers. The LCL encompasses articles on probationary period, redundancy, liquidated damage, severance pay, covenant not to compete and labour dispatching (agency work), and so on. It offers a more comprehensive guidance than the Labour Law on the employment relationship, from the formation of contracts to circumstances for the termination of contract and compensation (see Figure 7.1). It restricts employers’ autonomy to dismiss workers at will by introducing a non-fixed-term labour contract. It is anticipated that the tighter regulation will lead to the formalization of employment relations and reduce the number of those in informal employment, particularly agency workers and hourly rate workers. In the meantime, workers are given increased freedom of mobility by forbidding the use of deposit for bondage – a strategy commonly deployed by employers to tie their workers to the firm. The LCL also requires employers to make contribution to the social insurance premium for the employee (a requirement that is also stipulated in the Labour Law). In addition, the regulatory role of the trade unions has been strengthened through the right of joint decision in management practices. The legislative intent is to provide greater protection of workers’ substantive rights and increase the costs of non-compliance for employers. From this perspective, LCL is seen as a progression from the Labour Law. Local governments have been given more power, or, rather, more responsibility to monitor the compliance of the LCL. Labour Disputes Mediation and Arbitration Law (LDMAL) The labour disputes arbitration system of China was resumed in 1987. With the promulgation of the ‘Regulations on Enterprise Labour Disputes Treatment’ in 1993 and the Labour Law in 1994, a system of consultation, mediation, arbitration and litigation has been established, albeit a rather fragile one. In order to support the enactment of the LCL, the LDMAL was designed, approved and enacted within a year. Such a speed was rare in China’s law-making history but was deemed necessary. The LDMAL contains several important changes from the previous disputes resolution system, which in principle makes justice more accessible to the workers. These changes include: • ‘One Arbitration, Final Ruling’ for certain labour disputes (for example, a claim for wage arrears, medical expenses relating to work injuries,

188 Regulating for Decent Work

• •

•

• •

unpaid overtime and social insurance) instead of going through the ‘one arbitration and two trials’ procedure which is time consuming. Compared to the procedures of other types of disputes in China, the mediation and arbitration of labour disputes involved the most complicated procedures, a fact that has been widely criticized. The Mediation and Arbitration Law is intended to reduce this procedural complexity (Liang 2008; Tong 2008; Wang and Zhang 2008). Extending the application for arbitration from 60 days to one year to give workers more time to understand their situation and take action. Once the arbitration commission decides to accept the case, it should be dealt with within 45 days or with an extension of up to 15 days for a complicated case. This has shortened the time for concluding arbitration. Shifting the burden of proof as well as the consequences for failing to provide certain evidence to the employer in circumstances ‘where the evidence related to the disputed matter is in the form of records kept by the employer’ (Article 6). No fees charges for arbitration and reduced fees for litigation. Adding the mediation stage prior to arbitration in order to shift from a judicial to a voluntary approach to labour disputes resolution.

One new emphasis of the LDMAL is the role of mediation. By adding the mediation stage prior to the arbitration stage in the labour disputes resolution procedure, the government hopes to resolve labour disputes in an efficient and peaceful manner in line with its ideological objective of building a harmonious society. More specifically, the LDMAL (Article 10) has expanded the provision of the labour disputes mediation bodies to include three types: 1. Labour disputes mediation committee at the enterprise level; 2. People’s (civil) mediation body legally set up at the grassroots level; 3. Labour disputes mediation body set up at the community and district level where there is mediation capacity (but what qualifies ‘having mediation capacity’ is not defined). This multilayered labour disputes mediation mechanism provides workers with several channels to seek settlement. However, there are still some loopholes in the design of the law. Firstly, there is no specific or legal requirement for the establishment of a mediation committee at the enterprise level. Employers are left to make their choice whether they ‘have the conditions’ to set up a committee or not, and, if so, how

Fang Lee Cooke 189

it is to operate. As a result, many employers have not set up a mediation committee. Where one exists, it is observed more in name than in substance. The absence of an independent third party in the committee is a common phenomenon. Secondly, the second and third types of mediation bodies lack authority, legal position and power – a problem exacerbated by the deficiency in legal competence of the mediators whose rulings are often perceived unfair and challenged. This situation is not conducive to the promotion of mediation as the main means of labour dispute resolution (Lin 2008). The promulgation of the three major pieces of employment-related laws has, in principle, significantly streamlined China’s legal framework for employment protection and labour disputes resolution. However, the implementation of these laws has to date been met with a number of serious challenges due to conceptual ambiguities in the clauses, structural problems in legal channels, system abuse by both employers and workers and resource constraints in handling disputes. These issues are discussed in the next section.

Challenges to implementation and the role of/impact on the main institutional actors Local authorities, labour dispute resolution bodies, employers, workers and trade unions are the traditional main actors in the enforcement of labour regulations. This section investigates the impact of the laws on them, their tactical responses and dynamic interactions (see Figure 7.2). The EPL is a much less influential and controversial law than the LCL and LDMAL, as it is directed mainly at the local governments. By contrast, the consequences triggered by the implementation of the LCL and LDMAL, many of which were not foreseen, are unfolding and creating strong impacts on the existing positions of the institutional actors (also see Frey in this volume on the role of institutional actors). Employment Promotion Law (EPL) The EPL has been criticized for being ambiguous in its wording and sounding more like a government policy statement than a serious piece of legislation. There is little clarification in terms of what legal responsibility should apply to unlawful behaviour, making it impossible to enforce (see, for example, Han and Liu 2008; Ma 2008). For example, Article 62 stipulates that ‘In the event of any employment discrimination in violation of the provisions of this Law, the relevant worker(s) shall be entitled to initiate legal proceedings in the people’s court.’

190 Regulating for Decent Work

MNC client firms/NGOs

Workers

Trade unions

Employers

Employment agencies

Local governments

HR consultancy firms

Employers’ associations

Figure 7.2 Inter-relations between institutional actors

However, there is no clear guideline on what constitutes discrimination, what and how evidence should be used, and what compensation/ remedial actions should/can be taken to facilitate its enforcement. Similarly, there is no specification on what types of fiscal policies and favourable conditions should be in place to promote entrepreneurship and self-employment (Han and Liu 2008), leading to a wide variety of local interpretations and implementation styles. Labour Contract Law (LCL) The content of the LCL has been highly controversial from the outset. During the consultation period in 2006–07 on its draft versions more than 80,000 comments were received. Most of the feedback comments were submitted by employers and their pressure groups, both domestic and international. These include the European Union Chamber of Commerce and the American Chamber of Commerce. Foreign firms also threatened to withdraw their investment from China if the draft law were not amended. Numerous seminars and workshops were held with different interest groups, including government officials, international business lobby groups, trade unions/labour organizations, employers’ associations and scholars, to discuss and refine the drafts. The enacted version of the LCL ‘represents a compromise between the competing demands of these many interest groups’ (Cooney et al. 2007, p. 788).

Fang Lee Cooke 191

One key point of unresolved tension in the LCL is the clause on open-ended contract. It was considered to be the most inappropriate by many business owners and some economists. Critics held the view that this would mean the return to ‘iron rice bowl’ and ‘job for life’. It would end the freedom of enterprises in labour deployment and ultimately lead to the destruction of China’s economy (Xu and Luo 2008). A survey revealed that 70 per cent of enterprises had asked for amendment (Dong 2008). And the uncompromising attitude of the government on this clause has led to a panic wave of nationwide retrenchment in the months prior to the enactment of the LCL in order to (partially) pre-empt its impact. Among those joining the redundancy rush were many multinational corporations (MNCs) and flagship Chinese firms. Avoidance of entering formal employment relationships with the workers is the common tactical response of employers to bypass the legal constraints. This typically takes two forms. One is hiring the workers through employment agency firms (see below for discussion). The other is not giving a formal contract other than a verbal agreement. Some employers also offer a small sum of money to the workers on top of their wage instead of the full amount of social security premium as a ‘win–win’ solution (see below for discussion on the problems of social insurance). It is believed that the enactment of the LCL has a number of implications for employers in their HRM. First, labour cost is likely to increase significantly due to the cost associated with social insurance premium, redundancy pay, the reduction of flexible forms of employment, and compensation and penalties when employers are caught violating the law. Second, employers will face a higher level of employment risk as a consequence of stricter rules on probationary employment, greater freedom for employees to resign (abolition of unlawful private agreements that bond the employee to the employer) and the resultant risk of exposure of commercial confidentiality. Third, as noted above, the stable and long-term employment relationship promoted by the LCL will lead to difficulties in labour deployment due to the lack of labour mobility and skill redundancy, particularly of the longer-serving employees. This renewed ‘iron rice bowl’ labour deployment system is likely to reduce the efficiency of the firm – a problem that has plagued the state-owned enterprises. Fourth, the LCL presents a stricter requirement on firms’ HRM policy in principle. Not only should the policy be lawful, but also the process of policy formulation needs to be democratic through the consultation with workers’ representatives and/or trade unions. Company policies formulated in the absence of democratic participation is deemed invalid (also see Qiao 2008).

192 Regulating for Decent Work

Social security premiums are another major issue in the implementation of LCL that has led to a range of creative compliance from the local governments, employers and workers. In areas where economic development is slow, local governments issue favourable policy regulations and offer concessions to attract external investment. For example, in an inner city of Guangdong Province, businesses registered in Hong Kong enjoy an exemption period of five years during which the social security contributions for their employees are voluntary instead of statutory. This has encouraged mainland Chinese entrepreneurs to register their business in Hong Kong and operate on the mainland to take advantage of the privileged policy conditions. Similarly, some local governments specify that participation in social insurance schemes is not compulsory for non-local firms. This has enabled firms with subsidiaries in adjacent cities to make their employees register with a subsidiary in one city but live and work in another in order to save employment costs. Ironically, not all workers want to sign an employment contract with their employers. This is particularly the case for migrant workers. Contribution to social insurance is the crucial reason for workers to avoid signing the contracts. According to the LCL, both the employer and the employee need to pay monthly social insurance premiums. However, social insurance schemes are designed and launched on a local level. Many local governments also specify that the workers can only enjoy the insurance benefits (such as pensions) if they live in the same city where they pay in the contribution. The lack of coordination and transferability of social insurance schemes nation-wide means that migrant workers only have the opportunity to pay into the scheme. They are unlikely to benefit from it, as they tend to roam around the country for work and return to their hometown when work is unavailable or when they are unable to work. While they are allowed to withdraw from the scheme if they leave the city, they can only be reimbursed their contributions without any interest. Nor can they benefit from the contributions of their employer. This is in fact a tactic deployed by some local governments to fill the holes in social security funds, created in part by the compulsory early retirement programme implemented in the state sector during the mid-1990s and early 2000s. Labour Disputes Mediation and Arbitration Law (LDMAL) It was anticipated that the number of labour dispute cases will surge following the enactment of the LCL and the LDMAL. This is largely to do with the abolition of arbitration fees. In some places, the surge of labour dispute cases has become a source of concern for social stability.

Fang Lee Cooke 193

Interviews with labour authorities, trade union officials and labour law scholars have highlighted a number of problems in the design and implementation of the LDMAL (also see Hou 2008; Wang and Zhang 2008). First, there are insufficient guidelines on mediation (only seven articles in Chapter 2 for mediation compared to 35 articles in Chapter 3 for arbitration). This is a significant drawback given the fact that the overall mediation ability of grassroots mediators is extremely weak, with little training and professional background to facilitate them to carry out their tasks. Second, the ‘One Arbitration, Final Ruling’ (Article 47) system does not work well because the low competence of arbitrators leads to unfair ruling and appeal. Even if the ruling is fair, employers who lost the case can still appeal to the people’s court and cause delay in the implementation of the decision made by the arbitration commission. Third, the new procedure of ‘mediation–arbitration–first trial–second trial’ that replaces the original ‘one arbitration–two trials’ procedure has in fact led to the lengthening of the dispute resolution period because the lost party is entitled to appeal to the higher level. Employers may abuse this system as a tactic to delay payment to victim employees, as noted above. Fourth, the abolition of fees charges for arbitration and only ten yuan (less than US$2) fee to the law court for trial encourage the abuse of the system and leads to a flood of cases, many without merits. While arbitration is more likely to be abused by employees, employers may use the law court stage to wear down their employees. Interviews with informants show that since there are no fee charges, employees are more likely to take their employers to the arbitration commission without careful consideration of whether the employer has acted unlawfully or not. Some of the complaints involve the workers’ misinterpretation of their employment contracts and wage payment structure. Some employees who file complaints do so simply to seek revenge when they decide to leave their employing organization. Fifth, the abolition of fees charges, the sharp rises in the number of cases and the shortened time for concluding the cases have led to serious resource constraints. The LDMAL specifies that the funding of arbitration bodies should be covered by the fiscal budget of the local government. However, arbitration bodies cannot apply for the fiscal budget on their own. Many arbitrators are legal professionals who are brought in as part-time arbitrators. Without any fees income, it is difficult to secure their services. As a result, labour authorities force the dispute cases back down to the enterprises and ask them to achieve settlement at the enterprise level.

194 Regulating for Decent Work

In addition, since the arbitration commission is set up within the labour authority department, it is prone to political interference. Local governments may sacrifice labour inspection and protection for the sake of economic growth, as the saying goes: ‘law and order enforcement needs to give way to economic development’. Certain enterprises are the ‘no go area’ for labour inspection, or indeed any inspection, due to their (economic) bargaining power over the local government. Some labour officials can be co-opted by employers and become negligent. Inevitably, the procedural standards governing the conduct of hearings and the quality of rulings vary considerably between arbitration commissions and across the regions. Sixth, the simultaneous presence of both administrative (that is, labour inspection and monitoring authority) and legal (that is, arbitration and litigation) channels through which workers can seek to settle their disputes with the employers causes confusion and puts the labour authority in a very awkward position as their function straddles both approaches. The labour inspection authority sometimes finds itself between a rock and a hard place, as aggrieved workers are seeking innovative ways to win their cases. For example, Article 9 of the LDMAL stipulates that workers can complain to the labour authority when their employer has violated the state regulations, delayed the payment of wages, underpaid wages, delayed the payment of medical expenses for work-related injuries and so on. The labour authority should deal with the complaint in accordance with the law. Instead of using legal channels to resolve their disputes, workers opt to use the administrative channel – that is, they make a direct complaint to the local labour inspection and monitoring authority and request that the authority deals promptly with the disputes. If their request is rejected or if they are dissatisfied with the recommendation, they will submit administrative litigation proceedings to the people’s court suing the labour authority for their inaction or inappropriate action. Given their low capacity for handling the disputes, labour authorities lose in the majority of cases. Faced with a rising number of litigation proceedings, local authorities exert pressure on enterprises to settle the disputes internally and promptly. Therefore, workers and their legal representatives mobilize this method because they believe that it is an efficient and effective way to settle the disputes (Zhai 2008). Seventh, the LDMAL has strengthened the regulatory role of the trade unions, whose involvement in labour disputes resolution at all levels is legitimized and central. As enterprises are encouraged to resolve their disputes internally for ideological and practical reasons, this poses a serious challenge to the trade unions. Only one trade union, the

Fang Lee Cooke 195

All-China Federation of Trade Unions (ACFTU), is recognized by the Chinese government and led by the Communist Party. The ineffectiveness of the trade unions in representing workers’ rights and interests has been widely observed (for example, Taylor et al. 2003; Clarke 2005). Local branches of ACFTU have met with persistent resistance to union recognition from employers in the private and informal sector, where the majority of migrant workers are employed and the labour disputes level is relatively high compared with that in the state sector. Paradoxically, there is a relatively low level of workers’ demand to establish a trade union in the private and informal sector, partly because workers are unfamiliar with the concept of workplace representation but more so because of the perceived inefficacy of the trade union in advancing workers’ interests (Cooke 2011a). None of the 19 workers interviewed declared any allegiance with the trade unions. Nor would they seek the union’s support if/when they encounter disputes with their employer. None of the 34 private enterprises in this study have established a trade union organization, although 26 of them claimed to have established some sort of workers’ representative committee to play an extended management function.

The emergence of (new) institutional actors As touched upon above, new institutional actors are emerging in the process as employers are grappling their way around the new laws, particularly the LCL. Notable ‘new’ actors include: employers’ associations/pressure groups, employment agencies, foreign client firms, international NGOs, and HR consultancy firms. It should be noted that not all of these actors can be gegarded are new. Some of them, for example employment agencies and employers’ associations, have existed for a considerable amount of time but have gained power through ‘episodic intervention’ (Michelson 2008, p. 27) in the light of the promulgation of the new labour laws. Others (such as HR consultancy firms) seize the opportunities created by the regulations to establish themselves or to form a more constant source of institutional influence (as with international NGOs). In this section, we examine the emergence and role of these (new) actors in the enforcement of the labour legislation. The growing power of employers’ pressure groups Employers’ associations in China are generally neither well established nor independent. Similar to the ACFTU, the China Enterprise Confederation–China Enterprise Directors Association (CEC–CEDA) is

196 Regulating for Decent Work

the only official employers’ association that the state recognizes at the national level as the sole representative of employers’ interests. The state forms some kind of unequal partnerships with these organizations which often act on behalf of the state and help implement government policies (see Unger and Chan 1995). Facilitating the development of a harmonious relationship between the stake holders and providing training and human resource development activities are amongst the listed responsibilities of the CEC–CEDA (CEC–CEDA website, accessed on 6 May 2009). The subordination to state control means that the CEC–CEDA have limited autonomy beyond state-sanctioned activities. Nevertheless, it is important to note that the lobbying power of Chinese employers is rising outside the CEC–CEDA. They are able to form pressure groups rapidly to exert pressure on the government if forthcoming regulations and policies are likely to have a significant negative impact on their business environment. The approval process of the LCL is an example of their episodic interventions – the final version was watered down from the draft version as a result of employers’ lobbying (see, for example, Cooney et al. 2007; Cooke 2011a). Employers’ pressure groups have lobbied for the amendment of the LCL since shortly after its enactment. For example, in March/April 2008 individual entrepreneurs/business leaders used their opportunities as representatives on the National People’s Congress and the National People’s Political Consultative Conference to raise their requests for amendments during the plenary sessions of the congress/conference. This is in spite of the fact that the National People’s Congress and the Ministry of Labour and Social Security had stressed repeatedly that the discussion of the LCL should focus on its effective implementation arther than on making amendments (Ma 2008). Regularization of employment agencies? Despite being in existence for more than two decades, employment agencies represent a relatively new actor in employment relations in China. They are supposed to play a transitional role that helps provide employment as a government function on the one hand, and facilitates the government to change its function through the marketization of employment services on the other (Li et al. 2006). As such, there are intricate relationships between the employment agencies, local governments and employers (see Figure 6.2). The majority of the employment agencies and job centres have been set up by, or under the auspices of, the local governments in the last decade to provide services at the lower end of the labour market. Between 2001 and 2007 there had been a

Fang Lee Cooke 197

40 per cent increase in the number of employment agencies (NBS 2002, 2008); some 70 per cent of them were funded by the local government. This significant growth was a response to the large-scale downsizing in the state sector, the continuous inflow of rural migrant workers to urban areas to seek employment and the growing number of unemployed school leavers and college graduates (Li et al. 2006; Cooke 2011a). It was anticipated (by the state) that the enactment of the LCL would see the reduction of those hired by employment agencies, promoting a more direct and stable employment relationship between the worker and the firm. The reality so far has been a stark contrary. As noted earlier, to pre-empt the negative impact of the new law on employment cost, many employers dismissed their long-tenured workers and rehired them under new temporary contracts. Others dismissed their workers and rehired them as agency workers through employment agencies. As a result, employment agencies have prospered and the number of workers registered with them is growing. Agency workers often receive lower wages and much less social security protection than employees of the user firms. Managers interviewed disclosed that employment agency firms have some sort of agreement with the local labour authority and the social security insurance company which allows them to provide only partial social security coverage to a certain number of workers. This is how they drive down the employment cost and make a profit. Despite the central government’s recent instruction requesting local governments to withdraw themselves from employment agency services in order to avoid compromising their monitoring role, many local governments continue to run employment agencies under the pretence of reducing involvement. In some cases, they do so through pseudoprivatization or subcontracting to those who enjoy close personal ties with local government officials and profit from the business. Interviews with the local labour officials revealed that a continuous link with the employment agencies is necessary for the local government to fulfil its responsibility set out in the EPL. With the protection, albeit somewhat hidden, of the local governments on the one hand, and the rising business demand from user organizations on the other, employment agencies are likely to become a more institutionalized actor in employment relations. And agency employment will be a thriving form of labour deployment, contrary to the objective of the LCL. In fact, the policy stance of the central government in 2008 suggests a partial retreat of the government in its attempt to regulate and shrink the employment agency sector. Early 2008 saw the publication of the ‘Detailed Regulations on the Implementation of Labour Contract Law’

198 Regulating for Decent Work

(hereafter Detailed Regulations). It specified that an agency worker can only be deployed for a period of no more than six months. This was a remedial regulation to prevent employers’ abuse of the agency employment market as a result of the enactment of the LCL. In September 2008, the ‘Regulations on the Implementation of the Labour Contract Law’ was promulgated, removing the restrictions previously imposed on the agency employment sector by the Detailed Regulations. Foreign client firms and international NGOs as monitors Foreign client firms operating at the upper end of the product market and international NGOs have been an increasing source of influence on the labour standards in China, particularly in the export-oriented manufacturing sector. As there is an intensification in global business competition, MNCs are under increasing pressure to reduce their costs on the one hand and are under growing demand on the other hand to review their sourcing strategy, the labour standards of their supplier firms, and their wider role in the economic and social development in developing countries (Frenkel 2001; Chan and Ross 2003). The corporate social responsibility (CSR) of MNCs in the global economy is becoming an important aspect in the evaluation of corporate performance (e.g. ILO 1999; Cooke 2011b). Interviews with owner CEOs/managers of private manufacturing firms that are export-oriented revealed two scenarios. Those who operate at the lower end of the product market reported that they had not been under any pressure from their foreign client firms in relation to CSR issues, including labour standards. By contrast, those who operate at the upper end of the product market by producing brand-name products admitted that compliance to CSR requirements, including the labour laws and other business regulations of China, is paramount for securing business contracts with foreign clients. This suggests that the LCL does have a regulatory impact on certain firms. It is effectively implemented where there are sufficient business incentives to do so. And foreign MNCs and international NGOs are playing an indispensible role in ensuring its compliance. Nevertheless, the cost of compliance is largely borne by the Chinese manufacturers. As remarked by an owner CEO of a firm manufacturing computer speakers: ‘The foreign client firms all come to inspect the factory when they negotiate new contracts. They demand top quality products and world-class labour standards from us but are only prepared to pay pedlars’ price for it.’ Similarly, despite operating within immense political constraints and in limited locations and sectors, international NGOs and domestic ones

Fang Lee Cooke 199

under international patronage have been playing a role in monitoring the labour standards and legislative compliance. They also provide financial, medical, legal, educational and emotional support to the workers in sweatshop plants through some forms of organizing primarily outside the workplaces (see, for example, Murdoch and Gould 2004; Pun and Yang 2004). The promulgation of the LCL and related regulations undoubtedly provide them with more legal instruments to carry out their work, although gaining a wider political recognition from the state and operational legitimacy from the local governments remains a formidable challenge for the foreseeable future. The rise of consultancy firms and HR outsourcing providers as advisers Another unforeseen change following the enactment of the LCL and the LDMAL is the growth in HR outsourcing. This is due in part to the sharp increase in the number of labour disputes. Managerial informants admitted that they are increasingly looking to external experts to handle their labour disputes and employee benefits/social security administrations and to design their staffing policy to bypass the constraints of labour laws. They do so with a similar rationale to that of deploying agency workers from employment agency firms – ‘to save the hassles’. According to the seven HR consultants from five consultancy firms interviewed in this study and information from the secondary data, the enactment of the new labour legislation has undoubtedly brought new business opportunities. These include legal training and advice, employee benefits/social insurance administration, and businesses on other aspects of the HRM/employment practices. It becomes clear that despite still being a small market and largely in its embryonic stage of growth, the HR consultancy industry is set to become a more established institutional actor, involving in – and therefore influencing – the design of organizational procedures and HR policies. Arguably these consultancy firms play an important role in raising the legal awareness of employers and the quality of employment relations through educational effect, given the fact that a large number of employers may violate regulations as a result of ignorance of the legal requirements. Collectively, they also contribute to raising the professional standard and the competence level of HRM of the country. However, it would be naïve to think that the educational function is the primary raison d’être of consultancy firms. Profiting from the regulations, in the form of financial gains and/or social and political capital gains, remain the key motive of the emergence of these entities, as the HR consultants interviewed admitted.

200 Regulating for Decent Work

Discussion and conclusions This chapter critically analyses the legislative background and impacts of the enactment of the three new labour laws in China in 2008. It illustrates how the promulgation of these laws, particularly the LCL and the LDMAL, has led to the emergence of new actors in the labour market and in employment relations. These new actors, such as employers’ associations, employment agencies, labour dispute arbitrators, HR consultancy firms, MNC client firms and NGOs, are largely external to the employing organization. Whilst not all of these actors are new, or equally influential, in the governance of the labour market and employment relations, the enactment of the new labour legislation has introduced new dynamics in the evolving role of these actors and that of the traditional actors. These actors operate across multiple sites and spaces, overlapping as well as supplementing each other’s functions, and often interacting with each other in rather subtle and complex ways that can neither be seen nor officially endorsed. Despite the central government’s strong desire to create a more humanistic employment environment for the workers through a tightening of legislative governance, the commitment to the legislation is much less evident at the local government level. A unique feature of the Chinese laws and regulations is that the central government provides the broad framework. It is the responsibility of the local governments to devise their localized regulations based on these national frameworks. This flexibility is arguably necessary in a vast country such as China with significant economic disparities both across and between the regions. But the decentralization of interpretation and enforcement also opens up opportunities for implementation slippage, as the power and determination of the local governments and labour authorities may be circumvented by the priority of economic development. Some officials may even be co-opted by employers and other actors. Existing evidence clearly shows that local governments have been largely responsible for the implementation slippage due to their micropolitical and economic role. Grappling between the need to enforce the laws on the one hand and the need to stimulate economic growth on the other, local governments invent policy regulations, which are often worded loosely, to offer a flexible interpretation and implementation of the laws. The lack of adequate resources in the arbitration and court systems to cope with the sharp increase in the labour disputes caseload further contributes to operational pragmatism and muddling. Nevertheless, the policy regulations introduced by the local governments

Fang Lee Cooke 201

to supplement the labour laws promulgated by the central government should not necessarily be seen as a total deviation from the state’s objective. To some extent, these policy regulations provide a transitional period needed for firms and local governments to cushion the immediate impact of the laws. The silence of the central government in respect of the localized versions of labour laws indicates a form of tacit approval of this local compromise. It allows the central state to demonstrate its commitment to improving the labour standards and building a harmonious society on the one hand, and offers implementation flexibility at the local level on the other. Aided by local governments’ administrative flexibility, both formally and informally, employers continue to find ways to bypass legal constraints, including forming alliances with employment agencies. The latter offers employers an efficient way to curb employment costs because of the resources they devote to developing relationships with the local government, which enables them to overcome regulatory constraints in the labour market. At the legislative level, employers continue to form alliances and exert pressure on the state to amend the LCL to minimize the impact it may cause on their business. With respect to the workers, whilst there may have been a widening of their access to justice, the quality of justice is by no means guaranteed, contingent upon the local jurisdiction and the attitude of other actors. Some workers simply tolerate employers’ blatant noncompliance with the labour laws in silence in exchange for employment opportunities. It must be noted, however, that workers are neither entirely powerless nor all innocent. Instead, some of them are able to navigate through, or even manipulate, the system effectively to their advantage. Whilst the trade unions’ regulatory role has been strengthened in the labour legislation, their efficacy at the workplace level remains rather limited in terms of union recognition, involvement in consultation, collective bargaining and labour disputes resolution. To some extent, the space created by the absence of the trade unions in the private and foreign-funded enterprises is filled by foreign client firms and (international) NGOs, if in a somewhat temporary and discrete manner. The emerging monitoring role of the foreign client firms and the growing presence of NGOs may undermine the legitimacy of unions and their ability to fulfil the range of functions prescribed to them by the state. That said, the institutional position of NGOs is far from secure, as they rely on external resource support on the one hand and the tolerance of the state on the other, and can easily be isolated through political

202 Regulating for Decent Work

discourse. Similarly, foreign client firms may look to other countries for their suppliers when the price of Chinese suppliers becomes comparatively uncompetitive. The somewhat transient and precarious nature of foreign client firms and NGOs in their labour standards monitoring role may regenerate sites which formal monitoring channels have not been able to reach. This study shows that the working of the LCL and the labour disputes system has implications for the broader employment relations system in China. According to Cooney et al. (2007, p. 802), the promulgation of the LCL ‘attests to the increased openness of the Chinese legislative process to a wide variety of external influences’ and is a ‘clear improvement on the legal position that prevailed prior to its enactment’. Acknowledgement needs to be made of the efforts by the state in terms of raising its legislative standards and the protection of workers. However, the conceptual, procedural and practical problems exhibited in the three new laws suggest that the legislative capacity of China is still very weak, at least in the employment sphere. In particular, the new labour disputes resolution system appears rather ineffectual due to the conceptual ambiguity in the law, the structural problem in resolution routes, and resource constraints in the resolution process. The enactment of the LCL has actually promoted non-compliance or creative compliance to the disadvantage of workers. The somewhat ironic and certainly unexpected outcomes of the LCL and LDMAL reveal further that the efficacy of these laws has been compromised by the social, political and economic goals of actors at the local level. Collectively, they reconstruct and implement the laws at the local sites in the light of a range of dilemmas and constraints that have emerged and in the process of doing so, reconfigure each other’s responsibilities and boundaries in a pragmatic and mutually dependent manner. Interactive dynamics amongst actors have been noted in other societal contexts (for example, Martínez Lucio and MacKenzie 2004; Osterman 2006; Michelson 2008), but have been much less understood or even expected in the Chinese context where obedience to authoritarian instructions has been a more familiar story. A major contribution of this chapter is therefore to provide knowledge on the challenges to introducing new regulations in China and the role labour regulations play in employment relations. Importantly, it brings to the fore the role of a number of new actors, and in particular their dynamic interactions, which, despite demonstrating a high level of complexity, have remained largely unnoticed in the studies of employment relations in China.

Fang Lee Cooke 203

References Bamber, G.; Lansbury, R.; Wailes, N. (eds). 2010. International and Comparative Employment Relations, 5th edn (London: Sage). Bellemare, G. 2000. ‘End users: actors in the industrial relations system?’, British Journal of Industrial Relations, vol. 38, no. 3, pp. 383–405. Chan, A. 2001. China’s Workers under Assault: The Exploitation of Labour in a Globalising Economy (New York: M. E. Sharpe). Chan, A.; Ross, R. 2003. ‘Racing to the bottom: international trade without a social clause’, Third World Quarterly, vol. 24, no. 6, pp. 1011–28. Chang, K. 2008. ‘On the legislative basis and legal orientation of the Labour Contract Law’, Legal Forum, vol. 23, no. 2, pp. 5–14. China Enterprise Confederation–China Enterprise Directors Association (CECCEDA) ( http://www.cec-ceda.org.cn, accessed on 6 May 2009). Clarke, S. 2005. ‘Post-socialist trade unions: China and Russia’, Industrial Relations Journal, vol. 36, no. 1, pp. 2–18. Cooke, F.L. 2008. ‘The dynamics of employment relations in China: an evaluation of the rising level of labour disputes’, Journal of Industrial Relations, vol. 50, no. 1, pp. 111–38. Cooke, F.L. 2011a. ‘Employment relations in China’, in Bamber, G.; Lansbury, R.; Wailes, N. (eds) International and Comparative Employment Relations, 5th edn (London: Sage and New South Wales, Australia: Allen & Unwin Pty Ltd), pp. 307–29. Cooke, F.L. 2011b. ‘Social responsibility, sustainability and diversity of human resources’, in Harzing, A.; Pinnington, A. (eds) International Human Resource Management, 3rd edn (London: Sage), pp. 583–624. Cooney S. 2007. ‘China’s Labour Law, compliance and flaws in implementing institutions’, Journal of Industrial Relations, vol. 49, no. 5, pp. 673–86. Cooney, S.; Biddulph, S.; Li, K.G.; Zhu, Y. 2007. ‘China’s new Labour Contract Law: responding to the growing complexity of labour relations in the PRC’, Legal Studies Research Paper, Melbourne Law School (http://ssrn.com/ abstract⫽1115550, accessed on 2 June 2009). Dong, B. H. 2008. ‘From the Labour Law to the Labour Contract Law’. Paper presented in at the international conference ‘Breaking down Chinese Walls: The Changing Faces of Labor and Employment in China’, Cornell University, 26–8 September (Ithaca, NY). Frenkel, S. 2001. ‘Globalization, athletic footwear commodity chains and employment relations in China’, Organization Studies, vol. 22, no. 4, pp. 531–62. Han, G.J.; Liu, J. 2008. ‘An examination of the Employment Promotion Law in the context of social harmony’, Legal Studies Analysis (faxue pinglun), no. 5, pp. 65–76. Hou, X.B. 2008. ‘An analysis of the labour disputes treatment model and its weaknesses in the Labour Disputes Mediation and Arbitration Law’, Commercial Intelligence: Finance and Economic Study, no. 8, pp. 70 and 100. Hu, X.J. 2004. ‘On the legal system of China’s labour market’, Journal of Anhui University of Technology (Social Sciences Edition), vol. 21, no. 5, pp. 19–20. International Labour Office (ILO) 1999. International Labour Conference, 87th Session Report of the Director-General: Decent Work (Geneva: ILO).

204 Regulating for Decent Work Lee, C.K. 2007. Against the Law: Labor Protests in China’s Rustbelt and Sunbelt (Berkeley, CA: University of California Press). Li, B.A.; Xiang, S.Q. 2007. ‘Harmonious labour relations: an important foundation of building a harmonious society’, Labor Economy and Labor Relations, no. 6, pp. 40–2. Li, X.J.; Xu, Y.D.; Zhu, J.X. 2006. ‘Employment relationship under the form of employment leasing’, Labor Economy and Labor Relations, no. 1, pp. 5–8. Liang, Y.L. 2008. ‘New developments in the mediation and arbitration of labor disputes in China’, King and Wood PRC Lawyers, Internet source: hppt://www. kingandwood.com, accessed on 3 June 2009. Lin, L.Y. 2008. ‘A comparison of the labour dispute resolution system between Hong Kong and Mainland China: from the angle of the new Labour Disputes Mediation and Arbitration Law’, Legal System and Society, no. 11, pp. 59–60. Liu, Q.T. 2008. ‘Interpreting the Employment Promotion Law’, Journal of Chongqing Institute of Technology (Social Science Edition), vol. 22, no. 2, pp. 6–9. Ma, J.J. 2008. ‘A review of the labour laws legislation in 2007’, Lingnan xuekan, no. 4, pp. 60–3. MacKenzie, R.; Martínez Lucio, M. 2005. ‘The realities of regulatory change: Beyond the fetish of deregulation’, Sociology, vol. 39, no. 3, pp. 499–517. Martínez Lucio, M.; MacKenzie, R. 2004. ‘“Unstable boundaries?” Evaluating the “new regulation” within employment relations’, Economy and Society, vol. 33, no. 1, pp. 77–97. Martínez Lucio, M.; Stuart, M. 2004. ‘Swimming against the tide: social partnership, mutual gains and the revival of “tired” HRM’, International Journal of Human Resource Management, vol. 15, no. 2, pp. 410–24. Michelson, G. 2008. ‘New actors in Australian employment relations’, in Michelson, G.; Jamieson, S.; Burgess, J. (eds) New Employment Actors: Developments from Australia (Oxford: Peter Lang), pp. 1–31. Murdoch, H.; Gould, D. 2004. Corporate Social Responsibility in China: Mapping the Environment, Global Alliance for Workers and Communities Publication Series (US). National Bureau of Statistics of China (NBS). 2002. China Labour Statistical Yearbook 2002 (Beijing: China Statistics Press). National Bureau of Statistics of China (NBS). 2008. China Labour Statistical Yearbook 2008 (Beijing: China Statistics Press). Osterman, P. 2006. ‘Community organizing and employee representation’, British Journal of Industrial Relations, vol. 44, no. 4, pp. 629–49. Peng, X.Z.; Yao, Y. 2004. ‘Clarifying the concept of informal employment and promote the development of informal employment’, Social Science, no. 7, pp. 63–72. Pun, N.; Yang, L.M. 2004. ‘The Chinese working women’s network’, Against the Current (http://www.solidarity-us.org/atc/113luce.html, accessed on 13 November 2005). Qiao, J. 2008. ‘Labor Contract Law in China: Changes and implications’. Paper presented at the international conference ‘Breaking down Chinese Walls: The Changing Faces of Labor and Employment in China’, Cornell University, 26–8 September (Ithaca, NY). Shi, M.X.; Wang, B.Q. 2007. ‘Modelling of labour relations in informal employment’, China Labour, no. 11, pp. 22–4.

Fang Lee Cooke 205 Taylor, B.; Chang, K.; Li, Q. 2003. Industrial Relations in China (Cheltenham: Edward Elgar). Tong, W.D. 2008. ‘Progress in the labour disputes system’, Law and Life, no. 2, pp. 22–3. Unger, J.; Chan, A. 1995. ‘China, corporatism, and the East Asian model’, The Australian Journal of Chinese Affairs, vol. 33, no. 1, pp. 29–53. Wang, H.; Zhang, L. 2008. ‘An analysis of the new problems in the implementation of the Labour Disputes Mediation and Arbitration Law’, Market Modernization, October, pp. 367–8. Warner, M.; Zhu, Y. 2010. ‘Labour–management relations in the People’s Republic of China: whither the ‘harmonious society’?’, Asia Pacific Business Review, vol. 16, no. 3, pp. 285–98. Wu, W. 2008. ‘An estimation of the size of informal employment in the urban area in China’, Journal of Chongqing Technology Business University, no. 1, pp. 79–83. Xu, K.; Luo, J.Q. 2008. ‘Softening the Labour Contract Law’, Development and Management of Human Resources, no. 12, pp. 86–8. Zhai, Y.J. 2008. ‘The difficulty and challenge of labour inspection in China’, Administration and Law, no. 8, pp. 75–9.

This page intentionally left blank

Part III Old Challenges, New Techniques

This page intentionally left blank

8 Short Time Compensation as an Employment Stabilization Policy Robert M. LaJeunesse

For all of its celebrated labour market flexibility, the United States (US) has been unable to protect its labour market from the global financial contraction of 2008–09 through the rapid creation of jobs. Moreover the US is now experiencing a ‘jobless recovery’ as the economy slowly returns to trend growth levels. Meanwhile the ‘stodgy’ labour markets of central Europe have fared much better despite limited fiscal sovereignty under the euro. The global financial crisis has shown that labour market flexibility is a double-edged sword that may cut more deeply on the downstroke. If firms are relatively free to slash jobs and shift the costs of unemployment onto society rather than to negotiate with their employees and the state over survival tactics such as work-sharing arrangements, then serious macroeconomic effects will mount and exacerbate the contraction. As with past economic crises, the prospect of using work-sharing to insulate labour markets has experienced a revival following the onset of the global financial crises. In times of economic contraction the total number of hours worked typically falls and there is some arbitrary reshuffling of full-time and part-time employment. The greater the shift to part-time employment, the fewer the number of workers forced into ‘full-time unemployment’. For centuries now, governments across the globe have repeatedly turned to work-sharing strategies as a way of stabilizing their economy against the uncertainties of market-based growth (Cross 1989). The time-honoured use of such mechanisms begs the question of why many Anglo-Saxon governments (namely Australia, the UK and the US) have not been more aggressive in their regulation the supply of labour in an effort to achieve a more efficient, extensive, and less arbitrary distribution of work hours in order to stabilize employment and, by extension, economic activity. This chapter seeks to further 209

210 Regulating for Decent Work

clarify how unemployment insurance (UI) compensation can be used to promote hours flexibility rather than employment flexibility to stabilize labour markets and economic output. Understanding the macroeconomic importance of work-sharing suggests that short time compensation (STC) and other work time regulation policies should assume a greater prominence in stabilization policies, particularly in a post-industrial world of fewer jobs and growing ecological constraints.1

A comparative history of work-sharing The practice of work-sharing as a means of spreading the burden of unemployment predates unemployment compensation in some countries. With the establishment of formal unemployment compensation, these countries were able to enhance the attractiveness of work-sharing by allowing workers to claim partial unemployment benefits following a work time reduction. In Germany, part-time unemployment compensation dates back to the 1927 Placement and Unemployment Insurance Act. The programme was used most prominently in the 1970s as a way of mitigating rising unemployment in Germany related to stagflation and globalization. At its height in 1975, some 12,548 firms and 773,334 workers participated in the short time compensation programme, averting between 170,000 and 224,000 layoffs (Best 1988; Levitan and Belous 1977). The most generous compensation benefits have been paid by the German programme, which may explain the reduced opposition from senior workers in Germany relative to the US (addressed below). Observing the success of the German programme, many European countries now grant part-time unemployment benefits (Van Audenrode 1994). Considering that part-time unemployment is essentially a more egalitarian social response to a reduction of total work hours, it is unsurprising that European nations have embraced the concept more eagerly than Australia, the UK and the US. The greater solidarity shown by Europeans has allowed them to maintain very competitive rates of labour productivity while also enjoying the fruits of a post-industrial economy in the form of greater leisure time (Cette 2007). Canada has also used work-sharing on a relatively large scale to reduce the reliance on layoffs. In 1981, the Canadian Minister of Employment and Immigration authorized the temporary establishment of a national STC programme. The unanticipated success of the programme led to an increase of the original $10 million budget to more than $440 million by 1983. Eligibility for the Canadian programme was modelled after the German programme. It therefore required an initial workweek reduction

Robert M. LaJeunesse 211

of 20 per cent, and the reductions could not go below 10 per cent or above 60 per cent for the length of the agreement. Moreover, firms needed to avoid industrial disputes during the reductions, provide evidence of future economic prospects, document previous work hours, attest that workweek reductions were both necessary and temporary, maintain at least 40 per cent of their workers during the reduction in hours, and provide the written consent of a collective bargaining agent or employee representative. Two important macroeconomic aspects of the funding of the Canadian STC system were that the past layoff history of individual firms had no effect on their insurance premiums/contributions and that roughly one-fifth of the programme costs were covered from the general revenues of the federal government. Thus, the Canadian programme illustrates that a national STC programme does not have to be ‘pre-funded’ via payroll taxes and could be financed by a variety of federal fiscal arrangements. Although a state government would have to be somewhat concerned with revenue-neutrality when spending on payroll expansion incentives, federal programmes are not revenueconstrained under a ‘functional finance’ view of macroeconomic relations. As addressed below, federal financing of STC programmes becomes important during the transition from temporary short hours to longer hours or to permanently reduced hours. We will see that STC would not have to come to an abrupt end at the end of the unemployment period and could serve as more than a temporary stability measure by easing the burden of structural adjustment to a post-industrial economy of fewer aggregate work hours. Since a sovereign issuer of a currency can make near limitless amounts of money available to achieve a full utilization of labour (Lerner 1943; Wray 1998), many developed nations could easily finance a broad-based STC programme as the Canadian, French and German governments have done from time to time, albeit to a limited extent. Despite being fiscally restrained, many US states are rediscovering the virtues of work-sharing as an employment stabilization policy. The first significant experiments with work-sharing in the US took place during the Great Depression. President Hoover championed the idea of work-sharing in an effort to abate unemployment, but never passed any legislation to mandate it. Nevertheless work-sharing under Hoover did preserve some jobs. Hoover claimed that a reduction in the manufacturing workweek from 44.2 to 38.3 hours between 1929 and 1932, combined with private relief from employers, saved more than two million jobs. President Roosevelt promoted work-sharing by establishing voluntary codes on minimum wages and maximum hours in the

212 Regulating for Decent Work

National Recovery Act (NRA). Due in part to the NRA, the workweek fell further in 1934 to 34.6 hours – 22 per cent below its 1929 level. Since the notion of spreading work in lieu of layoffs predated unemployment compensation in the US, politicians lacked a mechanism to maintain the income of those placed on shorter hours to bolster their spending power. Yet, even without short time compensation, work-sharing achieved some success. Nemirow (1984a) argues that despite rising hourly productivity from less fatigued workers, the more efficient use of workers, and improved plant utilization, the jobs to output ratio increased from .78 in May 1933 to .93 in February 1934. Clearly work-sharing was only one of many tools needed to abate the severe unemployment of the 1930s, and served as a supplement – rather than a replacement – for vital public spending programmes designed to address such a deep contraction. During the halcyon days of capitalism in the postwar period from 1945 to 1972, there was little interest in work-sharing as unemployment rates were low and the length of the workweek declined gradually. With the stagflation, globalization and feminization of the workforce in the 1970s and 1980s, however, individual states began to investigate the STC programmes operating abroad as a way of guarding against rising unemployment. In the early 1980s, a few US states including California, Arizona and Oregon, served as STC pioneers. Since the passage of the Tax Equity and Fiscal Responsibility Act in 1982, US states have been allowed to distribute Federal Social Security Administration funds through the Unemployment Insurance programme to workers placed on part-time hours. Over roughly three decades, the number of states offering part-time unemployment compensation has grown, in fits and starts, to a current total of 17. Relative to Europe and Canada, work-sharing is not particularly prevalent or generous in the US. Unemployment compensation in the US typically replaces around one-half of the gross wages and lasts for 26 weeks. With so few states paying the partial benefits at such a low replacement rate, work-sharing has never amounted to more than 2 per cent of yearly total unemployment benefits (Balducchi and Wandner 2007). Nevertheless, the establishment of part-time unemployment compensation benefits has enhanced the effectiveness of contemporary work-sharing efforts to stabilize employment and economic activity.

Assessment of recent STC programmes: A work-sharing revival The work-sharing lessons from the Great Depression and many subsequent recessions indicate that employers prefer to utilize work-sharing

Robert M. LaJeunesse 213

in order to avoid layoffs. The evidence routinely shows that the number of involuntary part-timers typically rises before unemployment begins to increase. Bednarzik (1983, p. 3) observes that ‘the number of persons involuntarily working part-time is generally just a few steps ahead of changes in overall unemployment’. That is, the number of workers on part-time hours for economic reasons typically increases before unemployment begins to rise and then falls as employers restore the hours of those on short workweeks before expanding their payrolls. Generally, employers are not looking for an excuse to lay off workers but tend to reduce hours of work when possible before resorting to layoffs. Since the costs of employee turnover (that is, recruitment, screening and training costs) have arguably increased in the high-tech economy, contemporary employers may have an even greater incentive to avoid layoffs during economic contractions. Thus, work-sharing tends to occur automatically during periods of economic contraction, albeit on a highly arbitrary basis. Despite their limited availability and use, part-time unemployment benefits have expanded rapidly during recessions to cushion US workers and the economy from the pernicious effects of layoffs. Balducchi and Wandner (2007) investigated the use of STC during the 2001 recession in a range of US states. Their analysis shows a clear increase in the use of work-sharing in the majority of states that offer a STC programme during economic contractions. Figure 8.1 illustrates the cyclical spurt of

120000

Number of beneficiaries

100000 1997 (Pre-recession)

80000

2001 (Recession) 2005 (Post-recession)

60000

40000

20000

0

US Total

Arizona California Missouri New York

Texas Washington

Figure 8.1 Cyclical use of part-time unemployment compensation in the US and for the six most active states

214 Regulating for Decent Work

STC participants for the six largest STC states. Although the scale of STC use is minuscule as a percentage of total UI benefits, the data suggest that the policy has been used in a countercyclical fashion. Indeed, research by Van Audenrode (1994) concludes that STC programmes used in Canada and Europe have yielded work-sharing solutions that are every bit as flexible as the layoff pattern that has typically been employed by US firms. Van Audenrode (1994) has identified a link between the generosity of STC programmes and the responsiveness of hours to fluctuations in labour demand. Analysing the effectiveness of work time variations in achieving labour market adjustments in ten OECD countries with STC programmes, Van Audenrode (p. 97) concludes that, ‘countries with the most generous STC programs have large and quick working time responses to variations in the need for labour. In those countries, overall labour adjustments end up being as flexible as in the United States because working time adjustments compensate for restrictions on firings.’ In other words, hours flexibility has tended to offset the greater rigidity that European firms face in expanding and contracting their payrolls, suggesting that a more solidaristic method of labour market adjustment is possible. Indeed, the greater participation in the Canadian and European STC programmes could be related to a stronger culture of solidarity than prevails in the United States. A variety of evidence emerging in the wake of the global financial crisis suggests that employers are still keen to retain workers on short workweeks. The Bureau of Labour Statistics reports that approximately 9,084,000 US workers were working fewer than 35 hours a week due to ‘slack work’ or ‘lack of full-time work’ in May 2009 – a 72 per cent increase over the 5,290,000 who were part time for economic reasons in May 2008 (BLS Employment Situation, Table A-5, June 2009) (see Figure 8.2). This strong cyclical response has seen the number of involuntary part-time workers increase to 6.5 per cent of the workforce, surpassing the 1982 peak of 6 per cent. Figure 8.3 illustrates that most of the increase in parttime work has been a result of hours reduction rather than a structural growth of part-time employment (ibid., Table A-5, June 2009). Meanwhile the number of those working part-time for non-economic reasons (by choice) has fallen slightly, from 19.7 million in May 2008 to 18.8 million in May 2009, reflecting the contraction in general employment during the recession. What these figures suggest is that for every one involuntary part-time worker there are three workers who are working part-time through choice. Importantly, evidence suggests that even more workers are willing to shift to part-time status. In a survey of

Robert M. LaJeunesse 215 10000 9000

Thousands of workers

8000 7000 6000 5000 4000 3000 2000 1000

Jan−08

Jan−04

Jan−06

Jan−02

Jan−00

Jan−96

Jan−98

Jan−94

Jan−90

Jan−92

Jan−88

Jan−86

Jan−84

Jan−82

Jan−80

Jan−78

Jan−76

Jan−74

Jan−72

Jan−70

Jan−68

Jan−66

Jan−64

Jan−62

Jan−60

Jan−56

Jan−58

0

Figure 8.2 Number of workers who are working part-time due to slack work or lack of full-time jobs in the US

25,000 Voluntary part-time

Involuntary workweek reduction

May−2008 Jan−2009

Feb−2009 Mar−2009

Only part-time available

20,000

15,000

10,000

5,000

0 Apr−2009 May−2009

Figure 8.3 Part-time work status by choice in the US

757 full-time workers, the Opinion Research Corporation (26–30 March 2009) found that 94 per cent of employees were willing to save jobs by changing or reducing their schedule (Yost 2009). If their job was at stake, 8 out of 10 full-time workers would move to a ‘compressed’ four-day week with the same number of hours. Sixty per cent of

216 Regulating for Decent Work

workers would accept additional unpaid vacation days, and nearly half (48 per cent) would reduce their hours with reduced pay. Although the spectre of job loss may have motivated desire for more free time, the survey results are consistent with surveys conducted during tight labour market conditions (Golden and Gebreselassie 2007). As a result of the global financial crisis, a diversity of employers – varying from private, to public, to the non-profit sector – have chosen to reduce hours as a means of mitigating layoffs. They have also employed a variety of tactics. A survey of 141 human resource executives at US-based companies found that 22 per cent have reduced workweeks, 24 per cent have implemented mandatory shutdowns, and 17 per cent have required furloughs, while 11 per cent have offered voluntary furloughs (Watson Wyatt Worldwide 2009). A bigger survey of 518 large companies by Hewitt Associates (2009) found that 20 per cent of companies had cut work hours or imposed furloughs, a much greater fraction than the survey has revealed in previous recessions. In addition to large public employers such as the State of California implementing reduced workweeks and furloughs – which essentially amounts to unpaid vacation leave – many private employers across the US are also realizing the benefits of work-sharing (Yost 2009).2 Yet only those employers and employees residing in one of the 17 states that offer STC can avail themselves of partial federal unemployment benefits to help offset the income loss resulting from their reduction in hours. Even in those states with programmes in place the largest hurdle may be a lack of awareness regarding the availability of such programmes. Past experience suggests that STC arrangements were almost always a result of employer initiative (Best 1988). Since work-sharing has proven popular with both employers and employees, public policy should seek to make the reduction and reorganization of hours a more attractive, viable, and popular alternative to layoffs.

A macroeconomic theory of employment stabilization The principal macroeconomic benefit of work-sharing programmes that spread the burden of an economic contraction is the salutary impact they have on job security and, consequently, on aggregate consumption patterns. The economic benefits were summarized cogently by the 1995 Advisory Council on Unemployment Compensation, ‘By adjusting hours of work instead of the number of workers, the cost of unemployment, both financial and otherwise is broadly shared, rather than being borne by a relatively small proportion of workers.’ Work-sharing results

Robert M. LaJeunesse 217

in a superior distribution of employee sacrifices relative to payroll contractions achieved through layoffs and unpaid furloughs as all workers maintain some degree of job attachment and fringe benefits coverage. Under a work-sharing plan rather than subjecting 20 per cent of the workforce to a complete layoff, a representative employer could cut the weekly hours of the entire workforce from 40 to 32 in order to achieve a similar 20 per cent payroll reduction. With partial unemployment compensation benefits in place, the workers receive 80 per cent of their salary from their employer and roughly half of the lost 20 per cent from the STC programme for a period of 26 weeks or more. After adjusting for lower income tax liabilities, the typical worker will have gained a 20 per cent increase in their leisure time for an income loss of less than 10 per cent. Such a modest income reduction makes it much more likely that workers will be able to maintain their debt obligations and critical consumption purchases, as well as a good portion of their discretionary spending.3 The importance of long-term job attachment must also be emphasized. In addition to retirement savings, seniority wages, and a wide array of human capital investments, job tenure also affects access to unemployment compensation in many countries as applicants are judged to be ineligible because of their inadequate earnings resulting from sporadic employment histories. Since it is easier for employees to cover a smaller percentage income loss – by drawing down savings or borrowing – being unemployed on a ‘part-time’ basis is likely to stabilize aggregate consumption relative to ‘full-time’ unemployment. Workers may have to make some alterations to their spending patterns while working shorter hours, but their greater leisure hours may reduce or offset their need for certain purchases such as daycare, commuting, drycleaning, and so on. Like any social insurance programme, the process of spreading the risk of a calamity has important effects on behaviour and well-being. In the case of work-sharing, enhancing the job security of all workers can have a stabilizing effect on consumption patterns, which will in turn attenuate macroeconomic fluctuations. Although the costs to an unemployment compensation scheme are relatively similar whether benefits are granted to a small number of beneficiaries or a larger group receiving partial benefits, the social costs of ‘part-time’ unemployment are intuitively lower than under full-time unemployment (Kerachsky et al. 1997). When individual workers are forced to bear the brunt of layoffs, a rash of social costs emerge that could be avoided with a broader implementation of short time compensation. Full-time unemployment has been linked with a greater

218 Regulating for Decent Work

incidence of alcoholism, drug abuse, child and spousal abuse and other anti-social behaviour, which would likely be less prevalent under worksharing. It should be noted that the socialization of the costs of full-time unemployment represents a public subsidy to the firm and senior employees (insiders). In the US context, the social costs of full-time unemployment are significant due to the reliance on employer-provided health care. Nemirow (1984a, p. 37) points out that distributing reduced employment among many workers on a four-day workweek will ‘help public policy deal in a more rational way with the problem of health insurance for the unemployed’. Although sovereign currency governments are not fiscally restrained in achieving full employment, there are fiscal advantages to work-sharing vis-à-vis layoffs. Full-time unemployment entails a greater usage of public support programmes such as welfare payments, food stamps, and rent assistance than would be needed under ‘part-time unemployment’. Since they accumulate over time, the highest macroeconomic costs of full-time unemployment may be those related to human capital loss and hysteresis. The path-dependency effects of full-time unemployment are well documented, but there may also be ideological or psychological elements to the phenomenon that have been overlooked – that is, society may become more accepting of higher unemployment rates over time. In this case, it is not the skills of the worker or the commitment of the employer that is lacking, but rather the social desire to reduce unemployment. Intuitively, relying on work-sharing would eliminate much of the hysteresis effects related to the skill and behaviour of workers as well as those related to the attitudes of employers and governments regarding ‘equilibrium’ or ‘natural’ rates of unemployment. In addition to serving as a countercyclical labour market policy, Balducchi and Wandner (2007, p. 5) argue that work-sharing has benevolent effects on human capital development ‘by helping employers retain valuable employees and keeping those employees connected to their jobs during periods of reduced demand for labour’.

STC as an automatic stabilizer and structural adjustment mechanism Even if the workers never return to their erstwhile hours regime, parttime employment can help provide a mechanism for maintaining and developing important human capital investments during an economic transition. A large portion of workers placed on four-day weeks could use their day off to attend classes or training. Such investments would

Robert M. LaJeunesse 219

reduce the path-dependent drag on employment when economic output rebounds or improve the flexibility of the labour force in adapting to structural economic changes. Liberating unemployment compensation programmes from the notion that they are ‘pre-funded’ programmes would assist governments in pursuing such broad socioeconomic objectives. As a corollary to the macroeconomic issues mentioned above, the role of the federal financing of STC is imperative in order to allow it to serve as both a short-run economic stability plan as well as a long-run structural adjustment programme. Drawing on Abba Lerner’s (1943) observation that governments are not fiscally restrained up to the level of full employment allows STC programmes to abandon any lingering notions that a UI system has to be pre-funded by insurance premiums. When financed by a federal government using a sovereign currency, STC programmes can pursue broad objectives related to labour market solidarity (seniority payments to older participants), environmental impacts (throughput restraints) and socioeconomic development (structural adjustment). Operating under the notion of a government budget constraint has led to the inaccurate belief that UI systems need to be self-funded. Since high-wage workers in the US do not typically contribute more in the form of UI premiums, it has been argued that it would be inequitable to offer them higher wage replacement rates through a STC programme. US Representative Patricia Schroeder (1984, p. x) described her federal STC legislation, in terms of a pre-funded risk pool, ‘the bill was designed to protect the solvency of the UI trust fund by requiring that employers certify that the use of STC is in lieu of an equivalent number of hours of full-time layoffs …’. From a ‘state money’ or ‘functional finance’ view, it is impossible for any programme funded by a federal government with a sovereign currency to become insolvent. Although the US UI system is structured as a quasi-state and federal programme, the ability of states to draw low-interest loans from the federal government upon the depletion of their pre-funded pools has allowed many state systems to remain solvent throughout a deep and potentially protracted recession. The spending prowess of the government also affords a wider array of STC objectives over the long run. Governments might choose to fund STC as a work–family balance initiative, for example, rather than a countercyclical measure that has to guard against employers ‘abusing’ the system by placing workers on short hours. US policy makers might design STC programmes that encourage labour hoarding of employees on full health benefits by having some of the UI benefit flowing to the

220 Regulating for Decent Work

employer to offset fringe benefits. Subsidizing both the fixed and variable costs of labour would likely result in a STC to layoff conversion ratio greater than one – that is, 40 hours of work time reduction on STC may avert more than one full-time equivalent layoff. Having a ‘soft budget constraint’ is critical to the success of STC programmes during the transition out of an economic contraction. Recognizing the sovereign spending power of federal governments allows policy makers more flexibility in weaning employers and employees off UI benefits as the economy recovers from a downturn. During stable economic conditions, workers and employers may find that they prefer the short hours arrangement, due in part to higher levels of private and social productivity rates. Rather than abruptly ending STC assistance, governments could choose to phase out benefits as productivity gains gradually allow employers to cover the income gap of those remaining on short hours. In a sense, short hours could become the objective of STC. Governments could then subsidize the transition to shorter hours in the short run while long-run productivity gains would be made available to maintain wages as UI assistance was withdrawn gradually.4

Labour market equality Concerns over the ‘solvency’ of UI programmes have resulted in STC programmes that are largely unattractive to senior workers and their unions. The notion of seniority-based compensation – whether specified in union contracts or corporate practices – has led to misguided opposition to STC programmes. Unions have typically allocated job security on the basis of seniority, fostering opposition to workweek reductions lacking seniority-based wage offsets. Most US UI programmes set maximum compensation benefits at a relatively low rate, requiring significantly larger salary sacrifices for high-paid workers on STC programmes. Yet a functional finance view of fiscal management allows governments to make seniority-based STC payments to encourage solidarity between different generations of workers. Since UI programmes do not have be self-funding, STC compensation benefits could be made more attractive to senior employees, who have traditionally resisted work-sharing as an interruption of the quid pro quo they have established through their loyalty to their employer. In addition to shifting the impact of an economic downturn from younger to older workers, a federally financed programme could shift the financing burden from state governments to the federal government. There is no need for strict STC benefit

Robert M. LaJeunesse 221

limits; workers could simply be compensated at a proportion of their prevailing wage rate. Nostrums of fiscal responsibility have led to the false notion of a trade-off between seniority protection and layoffs. On the contrary, there is nothing that states that seniority perquisites have to take the form of hours protection. Senior workers could simply be placed on short hours at attractive replacement rates. Under the existing UI system, many US workers make contributions to UI programmes throughout their working life yet never draw on their UI benefits. The inequity of the funding of the present US system could be ameliorated by structuring UI in a similar manner to Social Security insurance. Under such a system, UI premiums and benefits could still be collected and distributed in a progressive manner, but the expanded use of the programme through STC would increase its familiarity and the level of social acceptance. STC programmes that offer benefits that are attractive to a wider array of workers could increase their political palatability and diminish any resentment that high-paid, permanent workers have towards paying slightly higher premiums. It bears repeating that the UI system does not need to be pre-funded or self-financing as the reduced social costs of unemployment will reduce government outlays in other areas and a federal government does not face a hard budget constraint when addressing unemployment. In this sense the funding of an STC programme would be designed with an emphasis on promoting equity and use rather than solvency. The government may want to structure the premiums paid to the UI system proportional to pay rates and potential benefits as an equity issue, but the solvency of a federal programme is never a concern when it is implemented under a sovereign currency. To ameliorate the divisive structure of existing US UI programmes, STC replacement rates should be geared to individual wages and benefits rather than flat rates contrived out a concern for the solvency of the UI programme. It is possible to configure a system that offers ‘worksharing incentive’ benefits to senior workers who otherwise would not lose their job, thereby fostering solidarity among the workforce. Such a structure is not a fantasy under a functional finance approach to economic development which acknowledges the government is not restrained in its spending when it comes to fighting unemployment. Although European workers have been more collectivist and egalitarian than their US counterparts, a lack of coordinated fiscal policy in the Euro zone currently undermines the government’s ability to encourage shorter hours. Conversely, Australia, the UK, the US and other currencyissuing countries are resurrecting the government’s responsibility for sufficient spending as an economic stabilization measure. However,

222 Regulating for Decent Work

this time one important difference exists; eliminating unemployment must be accomplished in a climate-constrained environment that can ill afford an expansion of the current composition of economic activity. Therefore, spending to encourage working time reduction represents a more sustainable use of stabilization funds than a growth-oriented employment policy. Since the unemployment compensation insurance premiums for US employers are based on the firm’s past layoff record, US employers could benefit from reduced insurance premiums by utilizing part-time unemployment rather than layoffs. Indeed, higher insurance premiums for those that shift the social costs of full-time unemployment on to society could encourage firms to utilize short time compensation instead of layoffs. Since firms will also benefit from the productivity improvements of work-sharing, it may not require significant financial encouragement for firms to recognize the advantages of retaining a larger share of their workers on their payroll at reduced hours during economic contractions or structural adjustment periods (Facer and Wadsworth 2008; Best 1988). Evidence from Europe – where severance payments and other costs make flexibility over work hours a more cost-efficient option than layoffs – shows that if society simply made it more expensive for firms to shift the social cost of unemployment onto taxpayers, STC programmes would become more attractive (Nemirow 1984b). Akin to their indifference concerning the distribution of income, many mainstream economists argue that the critical employment variable is total hours of employment, rather than the distribution of those hours. But just as heterodox economists have argued that normative and macroeconomic concerns over the distribution of income are indispensable, so is the distribution of work hours. Since the distribution of hours is one of the main conduits of the income distribution, economists cannot concern themselves with income and hours growth without also considering the composition of work hours. Just as the viability of focusing on income maximization alone is constrained by social and ecological implications, so to is a narrow focus of maximizing total work hours through a work fetish that lets the hours fall where they might. Another important equity concern is the ability of work-sharing to protect employment gains made by minorities during tight labour market conditions. Recessions do not have an equal effect on all segments of the workforce. Most economic contractions are not ‘equal opportunity dis-employers’ as employers tend to make use of a ‘last in, first out’ layoff policy. Since minorities are typically the last to be hired during tight labour market conditions, they tend to shoulder the bulk

Robert M. LaJeunesse 223

of the costs of unemployment when labour market adjustment relies on layoffs rather than an egalitarian sharing of work hours. Instead of the costs of unemployment falling on a small group of disadvantaged individuals, work-sharing allows a larger group to shoulder the costs of output reductions with only a marginal reduction in their total disposable income, thereby stabilizing economic activity and protecting the employment gains that were made by minorities during previous economic expansions.

Conclusion During a 1960 campaign address to a steelworkers’ union, President Kennedy cited the communist threat as an excuse for not reducing hours, ‘In the face of the Communist challenge, a challenge of economic as well as military strength, we must meet today’s problem of unemployment with greater production rather than by sharing the work.’ Given that today’s paramount challenges are now unemployment, labour market inequality, and climate change, a new progressive US president cannot make similar excuses. Since Kennedy’s abrogation, Europe has exceeded the US in expanding the leisure time of the working class (as well as other social benefits). Despite its celebrated flexibility, the US is left with a financial holocaust, a haemorrhaging labour market, and a moribund economy. Meanwhile, a lack of adequate federal fiscal policy in Europe has forced many countries to become increasingly experimental with work hours and the strategy has been somewhat successful in insulating labour markets. A telling tale of Europe’s relative advantage is the fact that unemployment in the US (10.2 per cent in October 2009) is now higher than in France (9.8 per cent in September 2009) and double that of the Netherlands (5 per cent in September 2009), where weekly hours are the shortest in the developed world. There is a lesson in these changing fortunes for all countries. The implications are that labour supply adjustments are effective in protecting employment and that fiscal sovereignty can be useful in expediting and enhancing employment stabilization via short time compensation.

Notes 1. Japan’s massive government spending on infrastructure projects of questionable social worth and environmental stewardship suggests that traditional ‘Keynesian’ economic growth policies need to be abandoned for a new form of distributing the ‘unspent social surplus’ (Ayres 1944).

224 Regulating for Decent Work 2. For most public employees in California, the furloughs will result in having the first and third Friday of each month off (a 9.5 per cent reduction in the 40-hour week) and 9 per cent reduction in pay from February 2009 until June 2011. Interestingly, public sector layoffs in the 1980s were one of the main motivations behind California establishing a STC programme during Governor Brown’s first term in office (Ittner 1984). 3. The social importance of maintaining debt obligations has been highlighted by the spate of home foreclosures linked to job losses and the ‘neighbourhood effects’ that have blighted entire communities in the US. 4. This was the theoretical rationale of payroll subsidies granted to French employers during the transition to the 35-hour workweek.

References Ayres, C. 1944. The Theory of Economic Progress (Chapel Hill, NC: The University of North Carolina Press). Balducchi, D.; Wandner, S. 2007. ‘Worksharing policy: power sharing and stalemate in American federalism’, The Journal of Federalism, vol. 30, pp. 1–26. Bednarzik, R.W. 1983. ‘Short workweeks during economic downturns’, Monthly Labour Review, vol. 106, June, pp. 3–11. Best, F. 1988. Reducing Workweeks to Prevent Layoffs: The Economics and Social Impacts of Unemployment Insurance-supported Work Sharing (Philadelphia: Temple University Press). Best, F.; Mattesich, J. 1980. ‘Short-time compensation systems in California and Europe’, Monthly Labour Review, vol. 103, July, pp. 13–22. Cette, G. 2007. ‘Europe – États-Unis: qui est le plus productif?’, Alternatives Économiques , no. 260, Juillet–Août, pp. 76–7. Cross, G. 1989. A Quest for Time: The Reduction of Work in Britain and France, 1840–1940 (Berkeley: University of California Press). Facer, R.L.; Wadsworth, L.L. 2008. ‘Alternative work schedules and work–family balance: an assessment from a growing municipality’, Review of Public Personnel Administration, vol. 28, no. 2, pp. 166–77. Golden, L.; Gebreselassie, T. 2007. ‘Overemployment mismatches: the preference for fewer work hours’, Monthly Labour Review, April, pp. 18–37. Hewitt Associates LLC. (2009). Cost Reduction and Engagement Survey 2009. Available online at http://www.hewittassociates.com/_MetaBasicCMAsset Cache_/Assets/Articles/2009/Hewitt_Survey_Highlights_Cost_Reduction_and_ Engagement_042009.pdf. Ittner, L. 1984. ‘The federal response to short-time compensation’, in MaCoy, R.; Morand, M. (eds) Short-Time Compensation: A Formula for Work Sharing (New York: Pergamon Press), pp. 158–82. Kerachsky, S.; London, R.; McCanne, D.; Needels, K.; Nicholson, W.; Walsh, S. 1997. Evaluation of Short-time Compensation Programs (Washington, DC: US Department of Labour. Lerner, A. 1943. ‘Functional finance and federal debt’, Social Research, February, pp. 38–51. Levitan, S.A.; Belous, R. 1977. Shorter Hours, Shorter Weeks: Spreading the Work to Reduce Unemployment (Baltimore: Johns Hopkins University Press).

Robert M. LaJeunesse 225 Nemirow, M. 1984a. ‘Work-sharing approaches: past and present’, Monthly Labour Review, September, pp. 34–9. Nemirow, M. 1984b. ‘Short-time compensation: some policy considerations’, in MaCoy, R.; Morand, M. (eds) Short-Time Compensation: A Formula for Work Sharing (New York: Pergamon Press), pp. 158–82. Schroeder, P. 1984. ‘Foreword’, in MaCoy, R.; Morand, M. (eds) Short-Time Compensation: A Formula for Work Sharing (New York: Pergamon Press), pp. i–xi. Van Audenrode, M. 1994. ‘Short-time compensation, job security and employment contracts: evidence from selected OECD countries’, The Journal of Political Economy, vol. 102, no. 1, pp. 76–102. Watson Wyatt Worldwide. 2009. Effect of the Economic Crisis on HR Programs (Online: Watsonwyatt.com), April. Wray, L.R. (1998) Understanding Modern Money: The Key to Full Employment and Price Stability (Cheltenham: Edward Elgar). Yost, C. 2009. Work ⫹ life Fit Reality CheckSummary www.worklifefit.com/pdf/ wlf_realitycheck_summ09.pdf.

9 Gender and the Minimum Wage* Jill Rubery and Damian Grimshaw

Introduction A statutory minimum wage is an important element of a gender equality policy programme. While much of mainstream economics still assumes that minimum wages distort the functioning of labour markets, there is accumulating evidence that labour markets without regulation do not operate efficiently. Differential rent sharing across industries, imperfect labour mobility and idiosyncratic employer strategies mean there is no tendency towards the law of one price in labour markets. Moreover, these features are gendered: more women than men are likely to be employed in industries where employers have a limited ability to pay, to be hindered by immobility caused by family position and access to welfare benefits and, as a consequence, to be vulnerable to monopsonistic employer power. A minimum wage can act as a strategic instrument in countering some of these distortionary effects and smoothing out some of the imbalances that result from the interactions between sex segmentation of labour supply and sex segregation in employment with processes of wage-setting. This chapter draws on cross-national empirical evidence to highlight the positive role that can be played by minimum wages, particularly in conjunction with an ‘inclusive’ industrial relations system, in reducing those distortions in the labour market that are the combined result of * This chapter draws on a range of related work, including work undertaken in 2002 for the ILO on pay equity and minimum wages and more recent work during 2007–10 for the EC on minimum wages and collective bargaining (DG Employment), for the ILO on minimum wages and pay equity and for the World Bank on the labour market flexibility debate and women’s employment. We are grateful to all three organizations for their support. 226

Jill Rubery and Damian Grimshaw 227

institutional arrangements and gendered norms. The chapter is organized as follows. The first section discusses the pervasive notion in labour economics of the law of one price. The chapter then reviews international evidence on gender pay gaps among the low paid and identifies an association with country wage-setting systems including minimum wages. The following two sections develop an institutional analysis of women’s relative disadvantage in the labour market, from both a supplyand a demand-side perspective. The conclusion sums up the debate and argues for more attention to gender equality issues in minimum wage analysis.

Labour markets and the law of one price The law of one price still provides the central organizing principle for labour economists’ theoretical framework of wage determination. The principle assumes that, in a competitive labour market, workers with similar skill levels and who are employed in similar jobs receive the same wage. The bulk of empirical research on wages takes this principle to measure a variety of ‘wage gaps’ to signal the degree to which labour markets are distorted from competitive outcomes. Of course, few economists would expect the law of one price to hold 100 per cent in real world labour markets, and there are several notable mainstream contributions that have fostered a more nuanced approach to understanding wage structures (see Oswald and Trostel 2000). These include studies that identify wage variations linked to: (i) rent sharing, where firm profitability has a positive impact on pay (Slichter 1950; Blanchflower et al. 1996); (ii) unions and other institutions such as minimum wage legislation (Bruno and Sachs 1985; Card and Krueger 1995); (iii) imperfect labour mobility (especially arising from job search costs, firm-specific training and household decisions, see Mortensen 2003); and (iv) firms’ payment systems (Lazear 1995). Despite the fact that the law of one price fails to hold in several areas of labour market functioning, the core principle is still retained. In a presidential address, labour economist Barry Hirsch defended its retention in the following terms: There are numerous reasons why Adam Smith’s theory of compensating wage differentials and the law of one wage should not and does not strictly hold in real world labor markets. Yet the law of one wage provides a fundamental and remarkably useful approach not only to describe market wage determination, but also to identify and interpret

228 Regulating for Decent Work

existing wage gaps. It remains our single most powerful tool and the necessary starting point for most wage analyses. This is no small thing. (Hirsch 2008, pp. 931–2) Mainstream empirical research on wages may set out to establish how far empirical reality lies from theoretical predictions but international cross-country research suggests that the size of unaccounted-for wage gaps varies inversely with the presence and strength of labour market institutions. For example, Teulings and Hartog (1998) find that the unexplained variation in wage equations (when wages are regressed on individual human capital characteristics) is greater in the deregulated US and the UK than in Norway and Sweden at the other end of the spectrum. Moreover, Schettkat (2002, table 3.1) demonstrates that these residuals correlate negatively (–0.84) with the centralization of wage-setting institutions (using the Calmfors/Driffill ranking). These findings imply that centralized and coordinated wage-setting institutions generate wage structures more in line with the textbook model of competitive, efficient labour markets. While a surprising result for mainstream economists, Schettkat argues it is in fact ‘consistent with the view that labour markets are fundamentally imperfect markets and that the transaction costs to achieve the market-clearing equilibrium wage are high in decentralised bargaining systems. From a theoretical perspective, centralised bargaining may collect information and substitute for the auctioneer’ (2002, p. 13). Research also documents a significant unexplained gap between male and female earnings, which again implies the absence of a law of one price in labour market functioning. In this chapter, we are interested in the extent to which the presence or absence of a statutory minimum wage and strong or weak collective bargaining coverage is associated with the size of the gender pay gap across countries. Our research review focuses in particular on gender gaps among the low skilled and low paid as this is where the impact of a minimum wage is greatest. Our expectation is that labour markets without a minimum wage are less likely to conform to the competitive law of one price than those with stronger low wage protection.

Evidence in the earnings data The international evidence suggests that although in most countries gender pay gaps are larger among higher-skilled and more-educated workers, they are still positive at the bottom of the pay hierarchy. In this section, we interrogate the available data to explore the nature of these

Jill Rubery and Damian Grimshaw 229

lower-level gaps and their linkages to minimum wage and collective bargaining systems. Gender gaps at the lowest decile OECD earnings data reveal considerable variation in gender pay gaps among workers paid at the lowest wage decile, ranging from a positive wage premium of 10 per cent in Hungary to a pay gap of 33 per cent in Germany; the average gap among the 21 countries in the study is 14 per cent. Part of the inter-country variability is explained by differences in earnings data – some countries report hourly pay data, others report weekly or annual data. Also, for most countries the data only cover full-time employees and therefore do not present a fair representation of the gender pay gap, especially in countries with a high share of part-time working. Moreover, for three countries – Denmark, France and the Netherlands – the OECD earnings data include part-time workers and are therefore likely to overestimate the gap compared to the other countries (see Appendix, Table A9.1). Nevertheless it is instructive to explore the relationship between size of gender gap at the lowest pay decile and the character of institutions for regulating low-wage work. We divide countries into three groups; those having a relatively high minimum wage (40 per cent or higher of average earnings), a strong and inclusive collective bargaining system (at least 80 per cent coverage) and having neither of these (see Table 9.1). Two countries, France and Belgium, appear twice with both a high minimum wage and a strong collective bargaining system, but otherwise the lists are mutually exclusive. The main result is that, on average, the lowest decile gender pay gap is smallest (8 per cent) for the group of countries with a high relative level of statutory minimum wage. For the nine countries with strong collective bargaining coverage, the average pay gap at the lowest decile is 15 per cent. The pay gap is highest at 21 per cent among the group with neither strong collective bargaining coverage nor a high-value statutory minimum wage. This approximate pattern provides some support for the argument that institutional arrangements for regulating low-wage work can reduce women’s vulnerability to low pay. They complement the more general finding that more coordinated and centralized wage bargaining institutions generate a more egalitarian wage structure and contribute to closing the gender pay gap (see, most recently, the regression results in EC 2008, pp. 93–4). Nevertheless, this simple clustering veils considerable variation within the groups. For example, Austria stands out for having a gender pay gap at the lowest pay decile of 31 per cent – the second-largest behind

230 Regulating for Decent Work Table 9.1 Gender pay gaps at the lowest decile (D1) (full-timers only) by type of wage system, 18 OECD countries, 2005

Average gender pay gap

High relative minimum wage (40%+ of average)

Strong collective bargaining coverage (80%+)

AS 2% NZ 5% IE 6% FR* 9% BE* 16%

FR* 9% DK 10% IT 10% SE 11% FI 12% ES 15% BE* 16% NL 17% AT 31%

8%

15%

Weak CB coverage AND No or low minimum wage US UK CA SZ JA DE

12% 15% 19% 19% 27% 33%

21%

Note: * France and Belgium appear twice since they both have strong collective bargaining coverage and a high relative minimum wage. Data for Central and Eastern European countries are excluded. See Appendix Table A9.1 for details. Source: OECD earnings database for inter-decile wage gap, ICTWSS for collective bargaining coverage and OECD for relative minimum wage data.

Germany among the 18 countries under consideration – despite near complete collective bargaining coverage estimated at 99 per cent. Part of the explanation is to be found in differential minimum rates in sectoral collective bargaining agreements, with lower rates in female-dominated sectors than in male-dominated sectors. However, in January 2009 social partners established a new cross-sectoral minimum of a1,000 per month as a result of union campaigns. Reports suggest that ‘virtually the entire private sector’ has been covered by the new cross-sectoral minimum which may reduce the gender pay gap at the bottom (Eironline 2009). Gender differences in the incidence of low-wage work International studies reveal considerable cross-country diversity in the incidence of low-wage work (for example, Gautié and Schmitt 2010; Grimshaw 2011; Vaughan-Whitehead 2010). OECD data for 2006 point to a high incidence of low pay among men (17 per cent or more) in Hungary, Poland, the US and the Republic of Korea, yet less than 9 per cent in Denmark, Japan, Finland and Sweden. When we consider low pay among women, the picture of cross-country diversity is even stronger; the Republic of Korea, Japan, Germany, the US and the UK

Jill Rubery and Damian Grimshaw 231

register a low pay incidence of 29 per cent or more, while Finland and Sweden score less than 10 per cent. Several studies have linked wage-setting institutions to the incidence of low-wage work. Salverda and Mayhew’s (2009) recent study of 13 European countries plus the US identifies a central role for statutory minimum wages and inclusive labour relations in limiting low pay. Schettkat has also found a negative correlation (–0.73) between the incidence of low pay and the Calmfors/Driffill ranking of centralization, for a sample of 14 countries (2002, table 3.1). Less research has been done on the impact of these institutions on the relative incidence of low pay among women. Figure 9.1 shows the incidence of low pay among women relative to men from 2000 to 2006. Women are more likely than men to be low paid in all countries, with the exception of Hungary where there are doubts about the robustness of the data. Especially large gaps in representation are found in the Republic of Korea, the Czech Republic, Germany and Japan, where data for 2000 suggest women are more than three times more likely than men to be in low-wage work. In countries with a relatively high statutory minimum

Low pay incidence of women relative to men

5.0 2000

2006

4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5

C Z D E JA

FI N L KO

D K

U K SW

IR C A ES

IT U S FR

AS N Z

H U PO

0.0

Figure 9.1 Low pay incidence of women relative to men, full-timers, 2000–2006 Note: See Appendix Table A.9.2 for details. Source: OECD earnings database.

232 Regulating for Decent Work Table 9.2 Low-pay incidence of women relative to men (full-timers only) by type of wage system, 2000 High relative minimum wage (40%+) NZ AS FR IE

Average low pay incidence of women compared to men

1.4 1.6 1.7 2.0

1.7

Strong collective bargaining coverage (80%+) IT FR SE FI ES NL DK

1.5 1.7 2.1 2.1 2.1 2.6 2.6 2.1

Weak CB coverage AND No or low minimum wage US CA UK DE JA

1.6 2.0 2.3 3.9 4.9

2.9

Note: Low pay defined as two-thirds median earnings. Missing low-pay data for Austria, Belgium and Switzerland. Central and Eastern European countries excluded. France appears under both high relative minimum wage and strong collective bargaining; low pay data for France is from a separate source and therefore not strictly comparable. See Appendix Table A9.2 for details. Source: OECD earnings database for low-pay data. All other sources same as Table 9.1.

wage or with strong collective bargaining coverage, there is a lessening of the greater risk for women compared to men (Table 9.2): on average, in both country groups, women are approximately twice as likely to be low paid as men (a multiple of 1.7 in high minimum-wage countries and 2.1 in strong collective bargaining coverage countries). In countries with neither form of protection, women are almost three times as likely to be low paid compared to men (a multiple of 2.9). Gender pay ratios in elementary occupations The pattern in most countries is that the gender pay gaps widen as one moves up the skill or education hierarchy. Data for the EU25 show an average gender pay ratio of 77 per cent in elementary occupations compared to 71 per cent among managers. Similarly, women’s earnings drop from 87 per cent of men’s earnings among the low educated to 69 per cent among the high educated. Nevertheless, the gender gaps among those in low-skill occupations or with low levels of education are still significant and subject to considerable cross-national variety. Figure 9.2 presents the average gender pay ratio (hourly earnings) in three low-skilled occupations, for a selection of EU member states. The gender pay ratios range from 70–80 per cent in Austria and the UK to

Jill Rubery and Damian Grimshaw 233 100 Sales & Services

Machine Operators

Elementary

95

Average gender pay ratio

90 85 80 75 70 65 60 55 50 AT

UK

DE

IE

ES

IT

FI

NL

DK

SE

FR

Figure 9.2 Average gender pay ratio by low-skill occupation in selected EU member states, 2002 Source: Eurostat Structure of Earnings Survey (hourly earnings).

90–100 per cent in Denmark, Sweden and France. The sample of countries is smaller compared to our above exploratory analyses, making the identification of wage system effects more difficult; Nevertheless, there appear to be similar patterns. The UK and Germany are both members of our category of countries with weak collective bargaining coverage and either no or a low minimum wage, and both have wide pay gaps among these relatively low-skilled occupations. This provides further evidence of the risks for women’s pay in countries without strong institutions for regulating low wages. Overall, the empirical evidence highlights that even within low-paid or low-graded jobs women fare less well than men and have a significantly higher risk of working in a low-paid job. Both factors contribute significantly to the aggregate gender pay gap. Most importantly for our analysis, we also find that the presence of a relatively high statutory minimum wage or extensive collective bargaining coverage influences the size of gender gaps at the bottom of the labour market. Countries without either are more likely to register wide gender gaps in the indicators examined here. In the next section, we explore in more detail factors on both the supply and the demand side that shape women’s vulnerability to low pay.

234 Regulating for Decent Work

The social structuring of labour supply and the reservation wage The debate on the impact of labour market regulation on women’s labour market position focuses on their tendencies to be labour market outsiders as the result of discontinuous participation (Rubery 2010). From this perspective employment regulations including minimum wages are said to act against the interests of women (Bertola et al. 2001; Botero et al. 2004), even though it is mainly women who are recipients of any rise in minimum wages. It is argued that this improvement in pay is offset by decreases in employment opportunities, particularly for those trying to enter the labour market as those who are insiders are protected by the transaction costs of hiring and by investments in their skills (Lindbeck and Snower 2002). This deregulation debate treats the discontinuity of women’s employment as a universal gender characteristic, rather than as a product of particular family, employment and welfare regimes, so that the policy imperative is not to harm the prospects for women as outsiders, rather than to change policy regimes to encourage women to become insiders/continuous employees. One element of the latter approach would be to improve pay so that it is more ‘worthwhile’ for women to work. Much of the variation in labour supply patterns between men and women is not necessarily related to the biological act of childbirth but to the ways in which the family, employment, tax and welfare regimes treat women and mothers. These conditions may account for not only gender differences but also cross-country differences by gender in key labour supply characteristics such as ‘reservation wages’ or commitments to continuous employment. Moreover, as the new literature on monopsony suggests, it is women’s vulnerability related to their responsibilities in respect of family and childcare that may leave them exposed to employer ‘exploitation’, due to labour supply inelasticity. We take these three features – reservation wages, employment continuity and job mobility – to develop a gendered labour supply perspective on the need for minimum wages as an element in equal opportunity policy. Gender and reservation wages There are two reasons why women may have lower reservation wages; first they are less likely to be in receipt of state support – in the form of unemployment benefits; and second they may be socialized to expect to be at least partially economically dependent and to make only a marginal rather than a full proportional contribution to total family

Jill Rubery and Damian Grimshaw 235

income. Because family systems, welfare state arrangements and social norms vary between countries, so will the extent and intensity of these effects. That is, these are not ‘essential’ gender characteristics, but so long as they persist there is a need to protect women against taking up wage offers that may reflect women’s reservation wages rather than their productive potential. The importance of unemployment benefits as an alternative means of providing a wage floor to minimum wages has been pointed out by Gregg (2000). He notes the patchy coverage of unemployment benefits even in welfare states with well-developed unemployment benefits systems. In the absence of minimum wages, benefits paid to those not in work are likely to act as wage floors. The variation in eligibility and generosity means that they are individual wage floors not society wide ones. Hence they do not mimic a minimum wage but create an upward sloping labour supply curve, whereby a higher wage broadens the proportion of the workless willing to work at the offered wage. In the case of earnings related benefits, this profile may mimic the individuals’ potential wage quite well. In other circumstances there is no reason why it should. (Gregg 2000, p. 137) The groups most likely to be excluded from benefits are those with short and discontinuous work histories, those in non-standard jobs (especially where these are outside social protection) and, where access to benefits is means tested, those living in households with at least one wage-earner. These groups include young people (who may implicitly be expected to rely on family support) and women. Table 9.3 indicates the much lower level of access of women to benefits among the unemployed in some countries. The EU15 countries with the largest gender gaps among unemployed benefit recipients are the UK, Ireland and Spain (as well as the Netherlands according to Gallie and Paugam 2000), while Italy and Greece have low coverage for all (see also Grimshaw and Rubery 1997; Leschke 2007). Where there is fairly uniform coverage by gender the rate may be both high and low. Where male coverage is low, as in Italy and Greece, then men as well as women may rely on intra-family support when they are unemployed, a system made possible by the continuation of more extended family systems based on income pooling in these countries (Karamessini 2008). However, other systems of support may be available to those laid off – such as compensation for workers, predominantly men, in large companies in Italy – so that these data may exaggerate

236 Regulating for Decent Work Table 9.3 Share of unemployed in receipt of benefits and the gender gap Country Spain Greece Italy France Belgium Luxembourg Germany Denmark Portugal Finland Austria Ireland UK

Male (%)

Female (%)

Gender gap (%)

34.6 13.6 4.3 51.0 79.9 22.2 68.7 85.8 26.9 79.7 59.5 87.9 33.3

15.9 9.4 3.3 40.6 74.0 17.9 69.4 83.7 23.4 75.4 43.5 44.9 17.2

54 31 23 20 7 19 –1 2 13 5 27 49 48

Note: ECHPS – the question asked is ‘Do you receive unemployment benefit or assistance?’ Source: Azmat et al. (2004).

uniformity by gender (Simonazzi et al. 2009). Low reservation wages may not lead to low-wage employment if wage systems are still protected. However, where employment and wage protection is limited low reservation wages may lead to low-wage employment. Differences in eligibility for unemployment benefits by gender are often not taken into account even in studies testing for the impact of unemployment benefits on the employment of women and men. For example, in the cross-country study by Bassanini and Duval (2006) for the OECD fairly strong negative impacts on employment for women from high unemployment benefits were found. However, the authors were surprised that the effects on full-time and part-time employment were similar as they hypothesized that high benefits would have a greater impact on lower-paid part-timers. However, the data were based on benefit levels rather than on the eligibility of women in general or part-timers in particular for benefits – conditions that can be expected to vary between countries. The consequence of ineligibility for benefits may be that many women who are without work but wish to work are likely to consider themselves as inactive rather than unemployed. This status of being without an independent income may impact on the willingness to take low-waged work. Manning (2003) has found that more women enter into employment from non-employment and that these earn less than those entering jobs from employment after the usual controls including experience, thereby contributing to the gender pay gap. This tendency to receive lower wages is likely to be exacerbated

Jill Rubery and Damian Grimshaw 237

where women do not have access to unemployment benefits as an alternative to low-wage employment. Men in societies with developed welfare states tend to rely on the state when without employment but in many societies for women their key source of support when not in employment is the family. Indeed, many societies expect women to take time away from wage employment to care for children and may not provide substitute income through highly paid parental leave, thereby presuming that women can and will be dependants – at least for a period. This presumption of income pooling reinforces the tendency for women to have low reservation wages when they return to employment. The potential for women to undercut the wages paid to male workers led to the Webbs (1920) referring to women as parasitic workers and the industries in which they were employed as parasitic industries. As Power explains, this approach locates women’s lower wages not in their lower potential productivity but in their position in the domestic economy which leads to their use by employers as a source of cheap labour: This parasitic-industries analysis is interesting for a number of reasons. First, parasitic-industries arguments began the analysis of wages not with an assumption of freely choosing individuals and clearing markets, but rather with a vision of a disorganized and predatory economic environment, with large and damaging differentials in power and information between workers (particularly women and children) and employers. Second, the analysis presumed that workers’ productivity was much more an outcome of the organization of jobs by employers than of workers’ own behavior. If employers could get workers cheaply (because of their relative powerlessness), they would not bother to use workers productively. Third, the analysis focused on the effects of low wages on the community as a whole, putting priority on the macro-efficiency of labor market outcomes. (Power 1999, p. 63) This argument in relation to parasitic workers and industries emerged in the original debates on establishing a minimum wage in the US and resurfaced again in living wage campaigns (Power 1999). The lack of an effective social welfare system in the US coupled with a higher predominance of racial in addition to gender inequalities has heightened the importance of low pay in poverty prevention (Nolan and Marx 2009) and also extended concerns over low pay to men as well as women. Two features of the debate around minimum wages as living wages are of particular importance for gender equality. First a living wage

238 Regulating for Decent Work

presumes that participation in full-time work should enable an adult to cover their living costs. To the extent that they worked full-time this approach would reduce women’s dependence on their partners. Second, living wage campaigns target the clients of low-paying employers – that is, those organizations further up the value chain that typically have a greater ability to pay. This approach recognizes that firms in competitive sectors may be exploited by more powerful clients to provide services at low wages, causing them to seek out and exploit those workers with the weakest bargaining power – such as migrant workers or women. The scenario has analogies with the debate on how less developed countries can secure a fair reward for their contribution to global value chains; powerful clients extract value while the weaker bargaining position of workers and firms further down the value chain allows a redistribution of value to wealthy economies (Heintz 2007). Seguino (2000) adds a gender dimension by arguing that export growth in some developing countries is dependent upon gender inequalities. Both the developed world receiving the exports and the less developed exporting societies benefit at the expense of the underpayment of female workers employed in export processing zones. Export growth may be beneficial for domestic investment, but the process is subsidized by women through their high efforts relative to wages. Thus the low wage received by women in such contexts is regarded as reflective of their weak bargaining power – particularly in the face of mobile capital – rather than their weak productivity or productivity potential. The reliance of the export growth model on gender inequality is related to the reduced likelihood that gender inequalities will lead to overt political conflict. The economic effects of conflict … are transmitted through the negative impact on investment and macroeconomic policy. These differing results may be reconciled by recognizing that inequality is less likely to produce social conflict if the burden is born by women, a group traditionally socialized to accept gender inequality as a socially acceptable outcome. (Seguino 2000, p. 1212) The presumption of acquiescence by women in the receipt of low pay may also lie behind other social and employment policies. For example, the UK has adopted a policy of providing state subsidies to almost all adults aged over 25 in low-wage work, except if the individual is in a relationship with another person who is in relatively high-paid work. Apart from young people the most important low-wage group not subsidized under this scheme are women from couple households.

Jill Rubery and Damian Grimshaw 239

This policy was designed to encourage unemployed ‘breadwinners’ to take up low-wage work since in the 1990s the majority of jobs available to the unemployed were part-time and low-paid and were being taken by the inactive – primarily women – in households with at least one income-earner. This approach re-establishes the notion of a family wage – albeit at an implied low level of subsistence – provided in part by employers and in part by the state, and it leaves unchallenged the payment of wages below subsistence or living wage levels to those deemed to be supported by the family. However, one rather unintended consequence of the spread of in-work benefits has been an explicit policy of maintaining and improving the level of the national minimum wage in the UK to prevent employers from taking too much advantage of the subsidy. The possibility that the wages of young people and women are suppressed because of family subsidies is not, however, acknowledged directly. Gender and labour market outsiders One principal basis for the claim that employment regulation, including minimum wages, is harmful is that the costs are borne primarily by labour market outsiders in the form of foregone earnings for those remaining unemployed or through lower earnings for those who consequently work in the informal sector. This argument raises two main issues for gender equality. First, instead of focusing on the problems for outsiders, it may be more productive to aim to reduce the gap between men as insiders and women as outsiders – by offering opportunities for paid leave, flexible working and affordable childcare to enable women to retain their position in the labour market. Moreover, women’s status as ‘outsiders’ or ‘marginal’ participants results not only from work–family conflicts but also from specific tax regimes that favour labour market withdrawal or engagement in marginal or unstable jobs. These tax restrictions arise out of specific welfare regimes that reinforce women’s position as outsiders. Second, UK evidence suggests the costs of women’s overrepresentation in labour market flows has increased as there has been a widening of the gap between entry wages and wages for those in continuous employment (including those who have changed jobs) (Gregg and Wadsworth 2000). More entry-level jobs are located at the bottom of the labour market and in the absence of policies to reduce women’s risk of being outsiders, minimum wage regulation can protect against any further widening of gaps between entry wages and continuing employment wages. Research into this widening of the entry wage gap found

240 Regulating for Decent Work

little difference between men and women, but the implication is that employers treat a spell out of the labour market for childrearing as being equivalent to the same period of unemployment (the main cause of most men being outsiders). The same study shows, however, that the wage penalty of part-time work, which correlates with women’s employment in the UK, is significantly higher for entry wages than for continuing jobs (Gregg and Wadsworth 2000, table 7). There are two different interpretations of policies to protect wages for new entrants. For some this may help to reduce the penalties of discontinuous employment which in the UK are sufficiently large that those entering on these low wages may never make up the gap. Alternatively, this entry wage gap may be regarded as reflecting real losses in human capital consequent upon employment interruption. If so, as Gregg and Wadsworth argue, the deterioration in human capital – whether from education or from employment experience – must be much more rapid than any estimates of its formation. To reduce gender wage inequalities it is likely that a combination of measures is necessary: minimum wages to prevent exploitation of re-entrants and measures to help women retain their footholds in the labour market. Gender, job mobility and monopsony Monopsony implies that employers have power in the labour market and are able to use that power to reduce wage levels below ‘competitive’ levels. This power derives in part from the restricted options available to the labour they employ. Women have less access to unemployment benefits than men and may therefore have a lower reservation wage. In addition, women have more limited job options because of their family responsibilities. This family position may mean they are only able to take jobs within a certain geographical area or with certain working hours due to commitments. Manning (2003) argues that women in the UK are increasingly long-term and committed workers, with a reduced tendency to work intermittently. This increases the inelasticity of labour supply to the employer as they may have limited incentives to seek jobs with better pay or longer hours due to both specific tax incentives and restrictions on geographical mobility. These may be intensified by limited local transport, internal family dynamics (for example, who has access to the family car, who has childcare responsibility) and by welfare and education arrangements, such as the length of the school day and the availability and costs of childcare. These vary across societies. For example, there is a short school day in Germany and a much longer school day in France which may help to explain the greater incidence

Jill Rubery and Damian Grimshaw 241

of women in short hour local jobs in Germany. Conditions also change over time; from the 1990s onwards there were increases in women’s access to private transport in the UK, but the gender differences remain (Smith et al. 2006). Women benefit more than men from improvements in local transport rather than long-distance commuting, so that different policy priorities will impact on monopsony effects. Women’s limited access to alternative income sources and job choices means that they are more exposed than men to the risk of monopsony power. Their labour supply to an individual employer is less price elastic than men’s, even if women at a more aggregate level are regarded as having a more price-elastic demand curve.1 A distinction needs to be made between general rises in real wages, which may attract more women into the labour market by changing the trade-off between domestic work and wage work (Mincer 1962), and the extent of labour market choices for women who are already committed to employment. Although we have identified that vulnerability to monopsony power may be reduced by a range of policies, the most direct protection is the introduction of legal minimum wages. Under monopsony, minimum wage rises may not only raise pay but also improve employment opportunities. There are disputes over whether there is robust evidence of employment increases following minimum wage hikes – Neumark and Wascher (2006), for example, question Card and Kreuger’s (1994) results – but there is near consensus that the employment effects of minimum wage rises are small (Freeman 2005, 2007), at least within the range of minimum wage increases experienced in most OECD countries over recent years. However, to a large extent the degree of monopsony, or vulnerability to low wages that is experienced by women, is also influenced by the organization of the demand side, the topic to which we now turn.

Organization of the demand side The argument we have developed is that minimum wages are needed to help to create a more level playing field for fairer treatment. So far we have made the case primarily in respect of supply-side reasons and vulnerability to monopsony. Here we turn to demand-side factors and labour market institutions that create or sustain gender inequalities which could be reduced by the application of effective minimum wage policies. Two primary arguments need to be made here; first, in the absence of a minimum wage, we would not expect firms or sectors to pay uniform minimum wage levels. Instead wage levels, even for the lower skilled, are likely to reflect differences in sectoral ability to pay

242 Regulating for Decent Work

(economic rents). Not only do women tend to be located in sectors where rents are low but they may also benefit less from the sharing of rents in sectors where these are sizeable. National minimum wages may reduce disparities between male- and female-dominated sectors and workplaces, thereby contributing to a narrowing of the gender pay gap. Second, there are significant variations between societies – linked to their industrial relations institutions and their approach to gender equality – that influence the extent of women’s vulnerability to low pay and the need for minimum wages to improve their protection. Labour markets can be categorized as relatively inclusive or exclusive (Gautié and Schmitt 2009), but even within these categorizations, some differences in gender effects are also likely to exist. Gender segregation and wage-setting arrangements Recent empirical research as summarized above has established that wages in the absence of regulation do not tend towards the law of one price. Instead weak regulation may increase wage dispersion while more coordinated wage-setting arrangements may reduce wage dispersion. Although industry effects are observed widely, their size varies not only between countries but also for men and women. In their exploration of interindustry differences in six countries, Gannon et al. (2007) find that, simple t-tests … show that between 43 per cent and 71 per cent of the industry wage disparities are significantly different (at the .10 level) for women and men. For Ireland and the UK, this percentage is quite high at 68 per cent and 71 per cent respectively, compared to only 43 per cent in Belgium. Moreover, Chow tests indicate that sectoral wage differentials are significantly different (at the .01 level) as a group for both sexes in all countries. (p. 146) Moreover this research found that inter-industry wage differentials explain important but variable shares of the gender pay gaps between countries. Variations were found in the importance of both differences in gender distribution across sectors, and in gender differences in industry wage premia. Wider dispersion of industry wage premia is associated with more decentralized or less extensive wage bargaining. Ireland has the largest industry effects related to large differences in gender wage differentials by sector; indeed, wage dispersion by industry is much larger for women, indicating that men have been able to escape low pay – even in low-paying sectors. These results all relate to 1995 and Ireland provides a potentially interesting case for further research as the

Jill Rubery and Damian Grimshaw 243

implementation of a national minimum wage after 1995 might have narrowed gender wage differentials within sectors and narrowed wage dispersion for women relative to men. The lack of evidence for a law of one price under de-regulation suggests that minimum wages could increase comparability in wages between sectors. It is not only gender segregation across workplaces that matters, but also the size of wage penalties for gender segregation. Coordinated bargaining systems may do more to reduce these penalties than minimum wages as minimum wages can at most affect the wage floor, rather than the overall level of inequalities. Nevertheless it may be easier to implement a minimum wage than to develop coordinated wage-setting institutions with high coverage. Moreover, not all coordinated systems promote gender equality as evidence above with respect to Austria suggests. A second important conclusion of the Gannon et al. (2007) study is that different industrial relations systems provide for different interactions with the gender pay gap and that targeted policies are needed. This is supported by other research into the interactions between the gender pay gap and forms of wage setting. For example, a UK study shows that once gender segregation at the workplace is included the impact of gender as an individual characteristic on wages became insignificant (Forth and Millward 2001). This suggests it was the opportunity to pay low wages to all provided by the patterns of gender segregation that was most significant in generating low pay for women. These organizations were almost all outside the collective bargaining systems and were thus most likely to be affected by the introduction of a national minimum wage. However, the impact of wage floors depends upon the starting points. In some contexts they may provide a new floor to the labour market and reduce gender pay gaps, but in others they may provide opportunities to pay minimum rates rather than observe those in collective agreements (Grimshaw and Rubery 2010). The availability of a range of minimum rates may lead to new strategies to pay lower wages. For example, Finland does not have a statutory minimum wage but it does have a hierarchy of collective agreements; research suggests that sex segregation by collective agreements is a separate and increasing cause of the gender pay gap, particularly for part-time workers (Laine 2008). Also, in both Germany and Austria companies have been evading highpaying collective agreements – in Germany companies switch to no collective bargaining or make use of the new but low-wage agreement for temporary workers (Mitlacher 2007), while in Austria there has been a process of switching from high-paying to low-paying agreements

244 Regulating for Decent Work

(Hermann and Flecker 2009). It is this wide variation among collective agreements in Austria that prompted unions to press for a new €1,000 monthly minimum wage while in Germany the issue of whether or not to introduce a statutory national minimum wage is still unresolved (Bosch and Weinkopf 2010). Women, particularly those in mini jobs, are the most likely to be low-paid and outside collective regulation but this applied even before the recent decline in collective coverage, such that the fastest rate of increase among the low paid is among men (Bosch and Kalina 2008). It is perhaps the increasing effects of deregulation on men that have promoted an active debate on minimum wages in Germany when previously the low wages paid to women excited less interest among trade unions. Other institutional arrangements and practices interact with the industrial relations systems to shape the impact of minimum wages on women’s vulnerability to low pay. A key example of the differences in the importance of the minimum wage in protecting women applies to the care sector across EU countries. Simonazzi’s (2009) comparative study makes clear that in the UK the introduction of a statutory national minimum wage had a major impact on care workers’ wages just at the time when the majority of care work was outsourced to private providers. In contrast, there is less influence in countries where care work is either organized as a skilled profession (Sweden and Germany) or relies on the informal sector (Italy). The extent to which women’s vulnerability to low pay is actually manifest in particular sectors, firms, occupations or contract types is thus highly dependent upon a range of institutional factors, including the industrial relations systems. The importance of minimum wage protection as a means of countering vulnerability varies across countries and over time, but industrial relations systems are also changing constantly so that women may face new risks of being employed on low wages even within regulated labour markets.

The need to refocus minimum wage debates on gender equity The analysis presented has suggested how low-wage segments may develop along gendered lines, with women facing particular forms of discrimination and disadvantage. These tendencies towards low pay and pay discrimination among women are far from inevitable. The size and form of low-wage labour supply groups is historically and culturally contingent and the impact of the segmentation of labour supply on

Jill Rubery and Damian Grimshaw 245

the structure of jobs in the economy is mediated through the effects of regulations, collective bargaining, industrial structure, state policy and the like. However, the dominance of the de-regulation debate means that many of these mediating and protective factors are being called into question. Although our analysis provides a general approach to understanding why labour standards are required (see also Deakin in this volume), in this final section we consider the need for minimum wage debates to focus more on the issue of gender pay equity. Table 9.4 outlines the advantages and disadvantages of minimum wages for promoting gender pay equity. A refocusing of the minimum wage debate on issues of gender discrimination is needed for three main reasons. First, debates about minimum wages typically focus on their impact on household poverty and are dismissive of their policy relevance where secondary earners (often women who are dependent on their male partner’s earnings) dominate. However, there is a major danger in ignoring the growth of low-wage and non-standard employment (Vosko in this volume) as it may herald a destabilization in conditions for higher-wage workers, calling into question custom and practice in working hours, pay premia and other benefits as well as wage levels. Such problems are likely to be more acute where the weakness of minimum wage regulation and collective bargaining has allowed for wide wage inequality (as in the US, the UK and Spain). Moreover, even when household income is raised above poverty levels through the wages of other household members, problems in relation to poverty may emerge; intra-household distribution may mean that not only women’s standard of living but also that of children may be at or below poverty levels. Furthermore, reliance on low wages may lead to future poverty due to household dissolution. Dependence upon another adult does not just contravene human dignity but also introduces major economic risks for women that are shown in the poverty distributions for single parents and older workers. Second, minimum wages provide an effective and important means of reducing pay inequities between male and female workers by establishing a fair price for individual effort among low-paid workers. The case for action to reduce gender pay discrimination should stand in its own right and not be dependent upon other arguments, such as the impact on household income distribution. The principle of equal pay for equivalent work is in fact assumed within the mainstream economics framework; therefore the argument against minimum wages as a means of reducing pay discrimination implies that current wage structures are independent of the sex of the worker taking the job. However, most

246

Table 9.4 Advantages and disadvantages of minimum wage policies for promoting gender pay equity Minimum wage policies

Coverage

Effectiveness

Costs of implementation

Side effects/spin-offs

Advantages

Covers all low-paid women where costs of discrimination are highest

Changes external market wage for women

Simple to understand and can be enforced by employee awareness, even in the informal sector

Reduces incentives to subcontracting

Extends to groups not included within collective bargaining or organizational pay hierarchies

Provides a mechanism for regular pay increases

Costs may be borne by whole community through higher prices; may be skewed towards wealthy consumers of services and is therefore a good redistributive tool

Complements gender pay equity policies; reduces wage dispersion and gender pay gap.

Applies to all disadvantaged groups, not just women

Can be used as a numeraire for social benefits, thereby integrating wage and social security policy

Provides basis for reducing long working hours necessary to achieve subsistence income By setting a universal norm delegitimizes the ideology that pay can differentiate by social group

Disadvantages

Some low-paid women may be excluded for administrative and other reasons – homeworkers, part-timers, informal sector

May be set at a very low level if not differentiated by sector or skill

Risk of wage inflation if leads to compensating claims

Employers may reduce working hours to offset costs

Effectiveness over time depends upon mechanism for indexation

Underpaid workers may not have the power to complain

May be used as a tool of macroeconomic management; this objective may conflict with gender pay equity

May be difficult to enforce in small firms; may promote informal sector growth Link to social security may lead to erosion of minimum wage value if there are public spending cuts

Workers close to minimum may bear the cost if skill differentials squeezed

Source: Rubery (2003).

247

248 Regulating for Decent Work

mainstream studies still find evidence of discrimination by gender in the labour market as a whole and it would be remarkable if this was not found and indeed concentrated among the lowest-paid groups. One problem is that the presence of some low-paid men may be taken as evidence that low pay reflects low skill rather than gender, but where women dominate the low-paid sector, disadvantaged men may find themselves tied to rates which reflect the low valuation of women’s work. The third reason for minimum wages is that this form of protection is typically more encompassing, in terms of the coverage of labour force groups, than many other welfare and labour market institutions and therefore is more likely to cover all, including women, at risk of working at very low wages. Even here there are still groups such as domestic workers that may not be covered (see Tomei in this volume). Other labour market floors such as unemployment benefits are less encompassing and introduce significant indirect gender bias through rules governing entitlement and the level and duration of payments. This exclusion applies to other disadvantaged groups such as the young and migrants. Moreover, even in countries categorized as having an inclusive system of industrial relations, collective bargaining and union representation may still be weaker in many low-wage sectors where women are overrepresented, particularly personal services and domestic work. Nevertheless, a minimum wage is only one element in ensuring gender pay equity and is not a substitute for developing an inclusive approach to wage-setting, since union representation and widespread collective bargaining coverage are the main means for developing a transparent wage structure that provides for fair comparison above the lowest-paid jobs. Indeed, in the absence of strong institutions to regulate the pay structure above the minimum, women in senior or skilled positions in sectors affected by minimum wage hikes may face wage compression – a scenario that appears consistent with changes in the UK earnings distribution in recent years (Denvir and Loukas 2007). It is also possible that a statutory minimum wage substitutes for a higher minimum in collective wage agreements in some contexts. Thus minimum wages do not offer a panacea for gender pay inequity but still constitute an essential element of any comprehensive policy to tackle the persistent gender pay gap. The starting points for effective policy have to be first a rejection of the notion that it is institutions which distort fair and efficient labour market outcomes and, second, the development of a universal rather than a partial wage floor. Any comprehensive

Jill Rubery and Damian Grimshaw 249

policy, however, must be a targeted policy aimed at mitigating the specific effects of the national system of welfare, employment and industrial relations on gender equality.

Note 1. It is also arguable that women’s labour supply is only price elastic in an upward direction; thus once women have entered the labour market they are less likely to withdraw even in a downturn (Humphries and Rubery 1984; Rubery 1988).

References Azmat, G.; Guell, M.; Manning, A. 2004. ‘Gender gaps in unemployment rates in OECD countries’, Working Paper for the Centre for Economic Performance, LSE, London. Bassanini, A.; Duval, R. 2006. ‘Employment patterns in OECD countries: reassessing the role of policies and institutions’, OECD Social, Employment and Migration Working Papers, no. 35 (Paris: OECD). Bertola, G.; Blau, F.; Kahn, L. 2001. ‘Comparative analysis of labor market outcomes: lessons for the US from international long-run evidence’, NBER Working Paper No. 8526 (Cambridge, MA: National Bureau of Economic Research). Blanchflower, D.G.; Oswald, A.J.; Sanfey, P. 1996. ‘Wages, profits and rent-sharing’, Quarterly Journal of Economics, vol. 111, no. 1, pp. 227–51. Bosch, G.; Kalina, T. 2008. ‘Low-wage work in Germany: an overview’, in Bosch, G.; Weinkopf, C. (eds) Low-Wage Work in Germany (New York: Russell Sage Foundation), pp. 19–112. Bosch, G.; Weinkopf, C. 2010. ‘Minimum wage systems and changing industrial relations: the case of Germany’ Report for the European Commission (www. mbs.ac.uk/research/europeanemployment/projects). Botero, J.; Djankov, S.; La Porta, R.; Lopez-de-Silanes, F.; Shleifer, A. 2004. ‘The regulation of labor’, Quarterly Journal of Economics, vol. 119, no. 4, November, pp. 1339–82. Bruno, M.; Sachs, J. 1985. Economics of Worldwide Stagflation (Cambridge, MA: Harvard University Press). Card, D.; Krueger, A. 1994. ‘Minimum wages and employment: a case study of the fast-food industry in New Jersey and Pennsylvania’, American Economic Review, vol. 84, no. 4, pp. 772–93. Card, D.; Krueger, A. 1995. Myth and Measurement: The New Economics of the Minimum Wage (Princeton, Princeton University Press). Denvir, A.; Loukas, G. 2007. ‘The impact of the national minimum wage: pay differentials and workplace change’, available at www.lowpay.gov.uk/low-pay/ research. Eironline. 2009. ‘New monthly minimum wage for private sector’, http://www. eurofound.europa.eu/eiro/2009/02/articles/at0902039i.htm (accessed June 2009).

250 Regulating for Decent Work European Commission (EC) 2008. Industrial Relations in Europe, DGV (Brussels: EC). Forth, J.; Millward, N. 2001. ‘The low-paid worker and the low-paying employer: characteristics using WERS98’, NIESR Discussion Paper No. 179 (London: National Institute of Economic and Social Research). Freeman, R. 2005. ‘Labour market institutions without blinders: The debate over flexibility and labour market performance’, NBER Working Paper No. 11286 (Cambridge, MA: National Bureau of Economic Research). Freeman, R. 2007. ‘Labor market institutions around the world’, NBER Working Paper No. 13242 (Cambridge, MA: National Bureau of Economic Research). Gallie, D.; Paugam, S. 2000. ‘The experience of unemployment in Europe: the debate’, in Galllie, D.; Paugam, S. (eds) Welfare Regimes and the Experience of Unemployment in Europe (Oxford: Oxford University Press), pp. 1–22. Gannon, B.; Plasman, R.; Ryck, F.; Tojerow, I. 2007. ‘Inter-industry wage differentials and the gender wage gap: evidence from European countries’, The Economic and Social Review, vol. 38, no. 1, Spring, pp. 135–55. Gautié, J.; Schmitt, J. (eds) 2010. Low-Wage Work in the Wealthy World (New York: Russell Sage Foundation). Gregg, P. 2000. ‘The use of wage floors as policy tools’, OECD Economic Studies, no. 31 (Paris: OECD). Gregg, P.; Wadsworth, J. 1995. ‘Gender, households and access to employment’, in Humphries, J.; Rubery, J. (eds) The Economics of Equal Opportunities (Manchester: EOC), pp. 345–63. Gregg, P.; Wadsworth, J. 2000. ‘Mind the gap, please: the changing nature of entry jobs in Britain’, Economica, vol. 67, pp. 499–524. Grimshaw, D. 2011. ‘What do we know about low wage work and low wage workers? Analysing the definitions, patterns, causes and consequences in international perspective’, ILO Conditions of Work and Employment Series No. 28 (Geneva, ILO). Grimshaw, D.; Rubery, J. 1997. ‘Workforce heterogeneity and unemployment benefits: the need for policy reassessment in the European Union’, Journal of European Social Policy, vol. 7, no. 4, pp. 291–315. Grimshaw, D.; Rubery, J. 2010. ‘Minimum wage systems and changing industrial relations in Europe: comparative report’. Report for the European Commission (www.mbs.ac.uk/research/ europeanemployment/ projects). Heintz, J. 2007. ‘Rethinking global labor standards: controversies, constraints and possibilities’, The Good Society, vol. 16, no. 2, pp. 63–72. Hermann, C.; Flecker, J. 2009. ‘Is institutional continuity making a creeping paradigm shift in the Austrian social model?’ in Bosch, G.; Lehndorff, S.; Rubery, J. (eds) European Employment Models in Flux: A Comparison of Institutional Change in Nine European Countries (Basingstoke: Palgrave Macmillan), pp. 131–54. Hirsch, B. 2008. ‘Wage gaps large and small’, Southern Economic Journal, vol. 74, no. 4, pp. 915–33. Howell, D.; Okatenko, A. 2008. ‘By what measure? A comparison of French and US labor market performance with new indicators of employment adequacy’, New School Working Paper, available at http://www.newschool.edu/cepa/ publications/workingpapers/SCEPA%20%20Working%20Paper%202008-2.pdf (accessed June 2009). Humphries, J.; Rubery, J. 1984. ‘The reconstitution of the supply side of the labour market: the relative autonomy of social reproduction’, Cambridge Journal of Economics, vol. 8, no. 4, pp. 331–46.

Jill Rubery and Damian Grimshaw 251 Karamessini, M. 2008. ‘Continuity and change in the Southern European social model’, International Labour Review, vol. 147, no. 1, pp. 43–70. Laine, P. 2008. ‘Segregation, gender pay gap and cross-sectional heterogeneity across collective agreements’, Discussion Paper No. 241 (Helsinki: Center of Economic Research). Lazear, E. 1995. Personnel Economics (Cambridge, MA: MIT Press). Leschke, J. 2007. ‘Gender differences in unemployment insurance coverage – a comparative analysis’, SP 2007-106 (Berlin: WZB). Lindbeck, A.; Snower, D. 2002. ‘The insider–outsider theory: a survey’, Discussion Paper 534 (Bonn: IZA). Manning, A. 2003. Monopsony in Motion: Imperfect Competition in Labour Markets (Princeton, NJ, Princeton University Press). Mitlacher, L. 2007. ‘The role of temporary agency work in different industrial relations systems – a comparison between Germany and the USA’, British Journal of Industrial Relations, vol. 45, no. 3, pp. 581–606. Mincer, J. 1962. ‘Labor force participation of married women’, in Lewis, H. Gregg (ed.) Aspects of Labor Economics (Cambridge, MA: NBER). Mortensen, D. 2003. Wage Dispersion: Why Are Similar Workers Paid Differently? (Cambridge, MA: MIT Press). Nolan, B. and Marx, I. (2009) ‘Economic inequality, poverty and social exclusion’, in Salverda, W.; Nolan, B.; and Smeeding, T. (eds) The Oxford Handbook of Economic Inequality (Oxford: Oxford University Press). Neumark, D.; Wascher, W. 2006. ‘Minimum wages and employment: a review of evidence from the new minimum wage research’, NBER Working Paper No. 12663 (Cambridge, MA: National Bureau of Economic Research). Oswald, A. J.; Trostel, P.A. 2000. ‘What can labour research teach us about macroeconomics?’, in Backhouse, R.E.; Salanti, A. (eds) Macroeconomics and the Real World – Vol. 2: Keynesian Economics, Unemployment, and Policy (Oxford: Oxford University Press). Power, M. 1999. ‘Parasitic-industries analysis and arguments for a living wage for women in the early twentieth-century United States’, Feminist Economics, vol. 5, no. 1, pp. 61–78. Rubery, J. (ed.). 1988. Women and Recession (London: Palgrave). Rubery, J. 2003. ‘Pay equity, minimum wage and equality at work’, Working Paper 19 (Geneva: ILO). Rubery, J. 2010. Labour Market Flexibility and Women’s Employment, Report to the Gender Unit (Washington, DC: The World Bank). Salverda, W.; Mayhew, K. 2009. ‘Capitalist economies and wage inequality’, Oxford Review of Economic Policy, vol. 25, no. 1, pp. 126–54. Schettkat, R. 2002. ‘Institutions in the economic fitness landscape: what impact do welfare state institutions have on economic performance?’, WZB discussion paper Berlin, FS I 02 – 210. Seguino, S. 2000. ‘Gender inequality and economic growth: a cross-country analysis’, World Development, vol. 28, no. 7, pp. 1211–30. Simonazzi, A. 2009. ‘Care regimes and national employment models’, Cambridge Journal of Economics, vol. 33, no. 2, pp. 211–32. Simonazzi, A.; Villa, P.; Lucidi, F.; Naticchioni, P. 2009. ‘Continuity and change in the Italian model’, in Bosch, G.; Lehndorff, S.; Rubery, J. (eds). European Employment Models in Flux: A Comparison of Institutional

252 Regulating for Decent Work Change in Nine European Countries (Basingstoke: Palgrave Macmillan), pp. 201–22. Slichter, S.H. 1950. ‘Notes on the structure of wages’, The Review of Economics and Statistics, vol. 32, no. 1, pp. 80–91. Smith, N.; Beckhelling, J.; Ivaldi, A.; Kellard, K.; Sandu, A.; Tarrant, C. 2006. ‘Evidence base review on mobility – choices and barriers for different social groups’, CRSP Working Paper no. 554 (Loughborough: Department for Transport). Teulings, C.; Hartog, J. 1998. Corporatism or Competition? Labour Contracts, Institutions and Wage Structures in International Comparison (Cambridge: Cambridge University Press). Vaughan-Whitehead, D. 2010. ‘Minimum wage revival in the enlarged EU: explanatory factors and policy issues’, in Vaughan-Whitehead, D. (ed.) The Minimum Wage Revisited in the Enlarged EU (Geneva: ILO), pp. 1–54. Webb, S.; Webb, B. 1920. Industrial Democracy (London: Longmans, Green & Co).

Jill Rubery and Damian Grimshaw 253

Appendix Table A9.1 Country data on collective bargaining coverage, relative level of statutory minimum wage and gender pay gap at the lowest pay decile % Collective bargaining coverage (20051) US JA NZ CA HU PO UK IE CZ SZ AS DE IT ES DK NL FI SE FR BE AT AVERAGE

14% 16% 30% 32% 35% 35% 35% 44% 44% 48% 60% 63% 80% 81% 82% 82% 90% 92% 95% 96% 99% 60%

Weak Weak Weak Weak Weak Weak Weak Weak Weak Weak Weak Weak Strong Strong Strong Strong Strong Strong Strong Strong Strong

Minimum wage as % of average wage (2007) 0.24 0.30 0.49 0.36 0.35 0.33 0.38 0.40 0.32 — 0.45 — — 0.31 — 0.38 — — 0.51 0.47 — 0.38

Low Low High Low Low Low Low High Low — High — — Low — Low — — High High —

Lowest decile gender pay gap for full-timers2 (20053) 12% 27% 5% 19% –10% 2% 15% 6% 23% 19% 2% 33% 10% 15% 10% 17% 12% 11% 9% 16% 31% 14%

Notes: 1. 2005 data except New Zealand (2000), Poland and the Czech Republic (2006). 2. Earnings data for full-timers only except Denmark (all employees), France and the Netherlands (full-time equivalent). 3. 2005 data except Finland, Ireland, Poland, Sweden and Switzerland (2004), Spain (2002), Italy (1996) and Belgium (1995). No inter-decile earnings data for Portugal. Source: OECD database for minimum wage and inter-decile wage data; ICTWSS for collective bargaining coverage (except Ireland from Eironline).

254 Regulating for Decent Work Table A9.2 Low pay incidence by gender Men

2000 AS CA CZ DK FI FR* DE HU IE IT JA KO NL NZ PO ES SE UK US

12.1% 15.6% 7.0%1 5.5% 4.6%2 11.5% 6.3% 20.4% 13.2% 8.9% 6.7% 15.6% 11.1%1 13.3% 15.0%1 12.0%3 4.3% 14.0% 19.3%

Women

2006 5

16.0% 15.4% — 8.2%5 4.6%4 9.2% 9.2%5 24.7% 13.9%4 — 8.0% 16.6% — 12.6% 20.9%4 — 4.7%4 15.7% 19.7%5

Women’s low pay incidence as a multiple of men’s

2000

2006

2000

2006

18.9% 31.3% 24.4% 14.1% 9.5% 19.6% 24.8% 26.5% 25.9% 13.4% 33.1% 48.1% 29.2% 19.1% 23.6% 25.0% 8.9% 31.5% 31.7%

20.1% 27.5% — 16.3% 9.6% 16.2% 31.0% 21.6% 24.0% — 33.8% 41.9% — 17.3% 26.2% — 8.9% 29.1% 29.4%

1.6 2.0 3.5 2.6 2.1 1.7 3.9 1.3 2.0 1.5 4.9 3.1 2.6 1.4 1.6 2.1 2.1 2.3 1.6

1.3 1.8 — 2.0 2.1 1.8 3.4 0.9 1.7 — 4.2 2.5 — 1.4 1.3 — 1.9 1.9 1.5

Note: *Data for France from a separate source – and are therefore not comparable — since missing from OECD database. 1. 1999 data. 2. 2001 data. 3. 2002 data. 4. 2004 data. 5. 2005 data. Same years for women. Source: OECD earnings database. For France see Howell and Okatenko (2008, figure 4).

10 Decent Work for Domestic Workers: An Achievable Goal or Wishful Thinking?* Manuela Tomei

Introduction The International Labour Conference (ILC) of the International Labour Organization (ILO) has engaged, as of June 2010, in a discussion on decent work for domestic workers with a view to the possible adoption of international instruments to protect domestic work in June 2011. Domestic work remains almost invisible as a form of employment and is often excluded, either totally or in part, from the scope of labour law and social security regimes. It is synonymous with informal and low-paid work and is perceived as lacking in value because it mirrors work traditionally performed by women at home for other women without pay. The knowledge and skills associated with it are usually acquired and developed, within the family, through experience, rather than through formal training. The preponderance in this workforce of women from historically disadvantaged communities, such as racial and ethnic minorities, or with immigrant status, and with generally lower than average years of education, exacerbates further the perception of this work as being lacking in worth. Furthermore, domestic workers often have no identity as workers, while their employers do not view them as employees with rights, but as aides whom they are doing a favour by offering them a few hours of work. Furthermore, employers in this sector tend not to consider themselves as employers (Domestic Workers United and Data Center 2006). This further devalues the labour of care, making it a second-class job, and contributes to its stubborn informality. * The views expressed in this chapter are those of the author and do not necessarily represent the views of International Labour Office. 255

256 Regulating for Decent Work

The ILO’s concerns about this category of workers is not new,1 but only recently the Organization has decided to develop an international norm for domestic workers, thereby confirming ILO’s commitment, as embedded in its Decent Work Agenda, to bring workers once deemed to be outside of its constituency into its mainstream work. A focus on decent work for all means recognizing that domestic workers are real workers and acknowledging the personal and domestic nature of the work, while reaffirming its compatibility with the employment relationship. Domestic work is to be treated both as ‘work like any other and work like no other’ (Blackett 2005). The standard-setting item on decent work for domestic workers advances ILO earlier efforts to establish – through the Part-Time Work Convention, 1994 (No. 175) and the Home Work Convention, 1996 (No. 177) – some minimum standards of protection for atypical and highly feminized, precarious work. These two Conventions mark a departure from an earlier focus on the standard employment relationship centred on the male breadwinner model working all year round for an indefinite duration and for a single employer. Convention No. 177, in particular, by viewing home as a worksite subject to inspection and other forms of regulation, broadens the boundaries of the worksite beyond conventional industrial set-ups, thus recognizing the changing and gendered nature of the employment relationship in a globalizing economy (Vosko 2006). The ILO’s renewed interest in regulating domestic work is associated with the rise worldwide in the demand for paid care services as a coping strategy to tackle the growing work–family tensions (ILO 2009a). This has exposed the need for addressing the legal vulnerability of domestic work and improving the working conditions therein in order to also enhance the quality of the services rendered to households. The lack of legal protection and social acknowledgement of domestic work reflects a substantial paradox: households entrust domestic workers with what they hold most dear – their homes and their loved ones – and expect from them the same unconditional availability of the mothers or wives whom these workers replace. Trust and loyalty are inherent aspects of the labour of care, and yet households, in the absence of a system of certification of skills or previous work experience, cannot verify the abilities and integrity of the people working for them nor are they often able or willing to compensate these workers in a manner which is commensurate with the responsibilities and the emotional bonding involved in domestic work, especially in respect of childminding or caring for frail adults. Low wages, volatile and unpredictable working

Manuela Tomei 257

hours, an absence of social benefits and an ill-defined employment status are the norm instead. This has been found to generate turnover, discourage training and contribute to the mismatch between the skills demanded and those on offer (Boris and Klein 2007). The promotion of decent work for domestic workers is therefore to be seen as a means of improving the quality of both the working lives of these workers and also the services they provide, thereby enhancing the well-being of the households they service. In recent years, a number of countries, particularly in Europe, have introduced new public schemes that encourage the use by households of paid domestic services, such as cleaning, housekeeping, childminding and other services, through simplified payment procedures, sometimes through the supply of state incentives, and also through the introduction of standard contracts. Another related strategy has been the ‘industrialization’ of domestic work, whereby domestic workers are employed by licensed enterprises rather than directly by households who pay a fee to these enterprises for the work performed. This chapter seeks to explore whether growing public and private investment in the development of paid domestic work results in a higher social and economic value being given to paid housekeeping and care work, improved working conditions for domestic workers and quality services for households. To this end, several national schemes in industrialized countries and one developing country are examined from the viewpoint of their impact on: (i) the quantity and quality of jobs in the households’ services sector; (ii) the incidence of informality; and (iii) the professionalization of domestic work. This expression is to be understood as involving the recognition of the informal learning process prevalent in this sector; the description of the tasks and responsibilities associated with domestic work; the definition of the content of the requisite knowledge and skills; formal training, including lifelong training which permits the updating and expansion of the required knowledge to keep up with technological innovation; and skills’ certification (European Economic and Social Committee 2010). The analysis is based on a review of secondary literature, ranging from government and other official reports to academic articles. The chapter is structured into three sections: the first part examines the trends in and characteristics of domestic work; the second part reviews the causes behind the expansion of state support to the outsourcing of women’s unpaid domestic choirs and care work, and provides a comparative analysis of the features and impact of such public schemes on a number of variables; the third and concluding section puts forward

258 Regulating for Decent Work

some suggestions about approaches and strategies to help raise the living and working standards of domestic workers, while contributing to quality domestic services for households.

Heterogeneity and specificity of domestic work Accurate and comparable international data on the extent of domestic work are unavailable. The different definitions of ‘domestic work’ in national statistical surveys and the fact that national statistics often do not count ‘domestic workers’ as a separate category, but place them under broader statistical headings, are part of the problem. High levels of under-reporting, as a result of the high incidence of undeclared or illegal work, such as child labour, is another major factor. When data are available nationally, however, significant numbers of workers are involved, including in industrialized countries (ILO 2009a). The significance of paid domestic work is linked to the growing difficulties that working families face in combining paid work with family responsibilities. These difficulties originate from a number of factors: the steady increase in women’s participation in the labour force, which reduces their availability for unpaid care work; the collapse of the extended family; the ageing of societies, in parallel with the shift from nursing homes to home care in the name of greater efficiency and better-tailored services (Boris and Klein 2007),2 changes in the structures of families; and the absence or inadequacy of family-friendly public and workplace policies. Some analysts suggest that the increase of domestic work in industrialized countries is also associated with the widening of income inequalities in recent years (Milkman et al. 1998). In middle-income countries, the expansion of the sector and its growing heterogeneity are ascribed to the demand for domestic services on the part of the middle and the lower-middle classes, coupled with changes in the demographics, characteristics and needs of domestic workers (Rodgers 2008; Thomson 2009). In countries characterized by low-income agrarian/informal economies, where welfare and care provisioning is left to families and communities, domestic work constitutes an important pillar of care-related work, whose significance has acquired further prominence especially in countries ravaged by the HIV/AIDS pandemic (Razavi and Staab 2008). The term ‘domestic work’ encompasses a wide range of occupations, the number and nature of which vary significantly from country to country. While certain countries, such as Brazil, place under the label ‘domestic work’ a broad spectrum of occupations, ranging from cleaning

Manuela Tomei 259

and laundry personnel, housekeepers and childminders to gardeners, private drivers and security guards, other countries, such as France, opt for narrower definitions, since rural domestic workers, gardeners and guards of private property are covered by collective agreements other than those relating to domestic workers (ILO 2009a). The list of occupations associated with domestic work may be open-ended, and where the employer’s household hosts the family’s commercial activity, domestic workers may also be required to ‘help’ occasionally on the commercial undertaking. Nonetheless, housework (cleaning, cooking, laundry, shopping) and taking care of children or other dependent family members seem to be the most common occupations across countries. Equally varied is the employment status of these workers: they may work as wage workers for one or more households – whether on a full-time or a part-time basis. They may also be employed by licensed institutions or agencies for delivering services in and for private households. Further, they may be considered as self-employed. The composition of the domestic workforce also changes by country and over time. In some countries, like those of the Middle East and the Gulf States, the vast majority of domestic workers are adult migrant women working full-time for a single employer. These workers tend to work and live in their employers’ household. In industrialized countries, the most prevalent situation is that of domestic workers working occasionally or on a part-time basis for multiple households, who may recruit them directly or obtain their services through a third-party agency that employs these workers. In developing countries, the landscape is more varied, ranging from live-in domestic workers, often very young, working for just a single employer, to live-out domestic workers working full-time or part-time for one or more private households. The relative weight of each of them in the total domestic workforce also tends to vary over time. Several Southern American countries, for instance, have witnessed an important decline in the live-in to live-out ratio over the past decade furthering addition to changes in both the demand and supply of domestic labour (Rodgers 2008; Thomson 2009). Shared features among domestic workers Despite the variety of their circumstances and work arrangements, domestic workers share some common features that distinguish them from other workers: they are primarily women belonging to disadvantaged groups, although men prevail in a few occupations, such as that of ‘gardener’ or ‘security guard’, who work in or for a household, which means working in isolation. Domestic workers have no peers to turn

260 Regulating for Decent Work

to for support or guidance about their conditions of employment. Nor are there any problem-solving or dispute settlement mechanisms in place to address possible tensions that may arise from the employment relationship (Dresser 2008). Isolation, vulnerability to less favourable treatment and abuse tend to be particularly marked among the live-in workers, especially when they are migrants. Living in the employer’s household may be the worker’s free choice aimed at saving on housing costs in order to send more generous remittances back home. Often, however, it is an imposed decision. This is the case, for instance, with the worker sponsorship systems whereby expatriate workers can obtain an entry visa and a residence permit in particular countries only if employed by a national citizen or institution. During the employment contract period, the sponsor-employer assumes full economic and legal responsibility for the employee who, in turn, can only work for that sponsor and is allowed to stay in the country as long as the contract remains in force. In these instances, the privacy of domestic workers, contacts with friends and relatives and other freedoms, such as deciding if and when to interrupt an employment relationship, may be severely constrained. As a result, domestic workers become vulnerable to abusive treatment, including physical and sexual abuse, the withholding of passports and forced labour (ILO 2009b). A number of bilateral agreements or memoranda of understanding have been signed by sending and receiving countries in recent years to address these issues in part (Ghosheh 2009). It has been suggested that the bargaining status of domestic workers working for multiple employers may be stronger than that of those servicing just one employer, as they can compare the standards in their different situations (Dresser 2008). Moreover, working for various employers permits a minimization of the offal in income that may result from the loss of one employer. However, it may also be costly and stressful, particularly when households are scattered and transportation problematic. Another distinguishing trait of domestic work is the degree of proximity and intimacy with the employer, and the emotional involvement that may be entailed, especially when it involves childcare or the care of the elderly. Devetter and Rousseau (2009) identify four types of employment relationships characterized by different degrees of personalization and power balance. They contend, however, that a highly personalized relation may not necessarily be a source of subordinated dependence; it can be associated with a relatively strong employee’s bargaining power when replacement cost is high.

Manuela Tomei 261

Domestic work: undervalued work Traditionally, neoclassical economy has viewed paid domestic work as lacking in value and exogenous to the labour market. By rendering services to the household – so the argument goes – domestic work does not generate profits and, hence, does not fit the ‘productive’ economy. The low value assigned to paid domestic work is linked to the fact that it mirrors work which women have traditionally performed at home without pay and which is taught, informally, in the domestic sphere. Housework and caring for children or frail older adults is perceived as necessitating limited, if any, competencies and skills; entailing minor responsibilities; being relatively safe and demanding limited effort. The fact that women, and women from disadvantaged groups, typically do this work – whether or not for pay – exerts downward pressure on wages. One additional important factor is that the women performing paid domestic work tend to face serious obstacles in finding better-paid jobs in other domains because of their usual lower than average years of education. Moreover, the perception of domestic workers, like other women workers, as ‘secondary earners’ making marginal contributions to household income also explain their lower reservation wages (Rubery and Grimshaw in this volume). Therefore it comes as no surprise that in countries where minimum wages are set by sector or occupation – whether by law or through collective bargaining – domestic workers are often not covered, and, when they are, their minimum wage rates tend to feature amongst the lowest minima (ILO 2008). When the minimum wage rate for domestic workers was first introduced in Chile in the early 1970s, for instance, the lower rate was justified on the grounds that part of their remuneration was paid in kind in the form of food and lodging. In the past decade, however, this argument has lost relevance as the live-in to live-out ratio has reversed, the latter constituting the large majority of the total domestic workforce to date (Reinecke and Valenzuela 2008). In view of this development, the Government of Chile has recently decided that, as of January 2011, domestic workers will be paid the same minimum wage rate which applies to low-skilled workers.

State-supported use of paid domestic and care work: a recent phenomenon Recognizing that the price of housekeeping and care services has an influence on households’ demand for them, a number of countries in Europe and middle-income countries such as Brazil have put in place

262 Regulating for Decent Work

public schemes that make it easier and less costly for households to access such services, thereby tackling the ‘invisibility’ and informality of domestic workers. State involvement occurs in a market which has been characterized traditionally by an unstructured system of demand and supply of domestic services and which has relied on informal, ‘word of mouth’ channels. As stressed earlier, employers of domestic workers constitute a unique group of employers who may not consider themselves to be employers or their domestic workers as employees, may be unaware of their obligations under the law and tend to resort to friends and relatives in order to set wages and other working conditions. The state regulation of the domestic work market marks a departure from earlier policy stances that regarded the organization of households as falling under the exclusive jurisdiction of families. This policy shift acquires an important symbolic value as it legitimizes the use by households of paid domestic services to satisfy housekeeping and caring needs that had been previously expected to be dealt with by the women of the households, especially mothers (Windebank 2007). This explains why the outsourcing of women’s care work is sometimes promoted in Europe as part of a broader public agenda aimed at helping women to articulate better their professional and family lives, thereby contributing to gender equality in the labour market. By emancipating women from the burden of daily housekeeping choirs – the argument goes – they are able to engage in paid work or training activities on a more equal footing with their male peers, while allowing less well-qualified women to secure a job and an income (European Foundation 2007). It has been argued, however, that these schemes, which continue to resort primarily to female labour, reinforce the perception that care work is women’s work, thus failing to challenge both the low value attributed to this type of work and also the unequal division of household work – whether paid or unpaid – between men and women. The imbalance in the sharing of family responsibilities is destined, in turn, to erode in the long run any gains that better-qualified women may be able to attain relative to their male peers in the labour market (Windebank 2007). Moreover, the transfer of low-valued housekeeping tasks from better-off women to women from historically disadvantaged groups with low qualifications contributes to deepening intra-gender, socioeconomic inequalities along ethnic, nationality and class lines (Dussuet 2005). The industrialization of domestic work, a phenomenon confined at present to the industrialized countries, through the establishment of ‘triangular’ employment arrangements involving the licensed enterprises,

Manuela Tomei 263

the employees of such enterprises and the households, or end-users, should lead, nonetheless, to greater labour protections to domestic workers and more reliable and professional services to households. The supply of domestic services through employees of licensed enterprises has the merit of identifying unambiguously who the employer is and making it clear that domestic workers are labour, not family. In principle, it should also encourage more standardized employment relationships and reduce the volatility characterizing domestic work. It is also plausible to expect employees of licensed enterprises to have a stronger bargaining power relative to workers recruited directly by a household (Dresser 2008), especially when replacement cost is low (Devetter and Rousseau 2009). On the other hand, licensed enterprises that are subject to government control and approval should have greater incentives to invest in the professionalisation of their employees with a view to offering better-quality services. The provisioning of state incentives, such as subsidized jobs or fiscal incentives, should also promote greater employer compliance with the law, considering households’ inability or unwillingness to pay for such services. The notion of ‘affordability’ is a highly subjective one, and it is plausible to assume that, as long as society attaches a low value to unpaid care work, the cost of its replacement by paid services will never be considered to be cheap enough. The supply of public funding also raises concerns about how equitably state incentives are distributed among households belonging to different income levels, and whether or not the use of paid domestic services continues to be the preserve of the richest social classes or has also been extended to lower-income strata. This section examines Titre-Service (hereinafter TS) in Belgium, the Chèque Emploi Service Universel (CESU) in France, the Chèque Service (CS) in the Canton of Geneva (Switzerland)3 and a set of measures recently introduced by the Brazilian government to encourage the registration of domestic workers. All of these national schemes share in common the introduction of simplified systems for the registration of domestic workers and the payment of employers’ contributions. An additional common feature of the TS, CESU and Brazil’s policy measures is the supply of fiscal incentives to the participating employers and household users. These schemes, however, differ in many other respects, including the type of employment arrangements concerned, the number and type of institutions involved and the ultimate social and economic goals that they seek to attain. There are also variations in the size, patterns and characteristics of domestic work between Europe and Brazil. In the former countries,

264 Regulating for Decent Work

domestic work is predominantly part-time, involves working for different households and still absorbs a relatively limited proportion of the female workforce. In Brazil by contrast, full-time domestic work for a single household remains significant and constitutes the main occupation for Brazilian women. In both instances, however, domestic workers comprise mainly low-income migrant women – whether national or international migrants – often belonging to disadvantaged racial groups and with low levels of formal qualifications. The cheque-system: The European approach towards tackling informality The TS, CESU and CS schemes share a common goal: reducing the incidence of undeclared work among domestic workers. Adherence to these programmes entails the automatic application of a set of clearly defined labour norms, namely the standard employment contract for the domestic work sector in the case of the CS, in the Canton of Geneva, and the sectoral collective agreement in the case of Belgium and France. However, in Belgium and France, consistent with the EU employment strategy adopted in the early 2000s, the search for better-quality jobs is pursued in conjunction with the promotion of job creation in order to curb the long-term unemployment of people with low levels of employability, including low-skilled, low-income, migrant women (European Foundation 2007). Work and family reconciliation also feature in both programmes, albeit less prominently. In Belgium and France, policies are driven by a desire to develop household services as a clear-cut sector with considerable potential for job creation. This explains the introduction of innovative organizational and legal forms, such as social cooperatives, an increased acceptance of methods of work that are prevalent in the for-profit sector, a greater reliance on public–private partnerships and the promotion of a plurality of employment arrangements and actors beyond the conventional direct recruitment of domestic workers by individual households. Indeed, the Belgian TS covers only licensed enterprises that employ domestic workers rendering services to client households, while the French CESU covers licensed businesses, as well as direct employment by individual households and employment agencies that place domestic workers in private households alike. Conversely, sustaining higher rates of employment growth is an absent goal from the CS of the Canton of Geneva, its primary concern being the protection of the efficiency of the Canton’s social security system through the expansion of its contributory basis. This is consistent with the fact that immigrants with irregular status, who seem to constitute

Manuela Tomei 265

an important share of total employment in the sector, are also allowed to join the CS, in contrast to the French CESU which restricts coverage to nationals and documented immigrants. Neither does the CS pursue the development or modernization of the sector as reflected by its exclusive focus on households directly recruiting domestic workers. These three schemes envisage the simplification of both the administrative recruitment formalities and the payment for services rendered by workers, including the calculation of the employers’ social security charges. In contrast to the Canton of Geneva, TS and CESU also provide incentives to employers in the form of deductions to employers’ social security contributions, reduced VAT or tax exemptions. Simplified payment of social security contributions Belgium, France and the Canton of Geneva have all adopted systems of cheques/vouchers. However, there are significant differences between the individual schemes in respect of the type of employment relationships permitted, the services rendered, the social benefits provided, the institutions involved and the whole modus operandi (see Table 10.1). For instance, while all three systems provide compensation for employment injuries; retirement pensions; maternity protection; family allowances and unemployment insurance, only the Belgian and French systems also cover health insurance. With regard to the range of services that are made available through these systems, only CESU in France provides, in addition to housekeeping services – whether in or for a household –, care services for children and people with limited autonomy (mainly elderly people). The TS was introduced by law in 2001 in order to promote the domestic services sector or ‘services de proximité ’ and became operational in 2004. It permits private households to pay a licensed enterprise 4 for the services provided by a worker under contract with the enterprise. The employment relationship is therefore triangular: there is an employer (enterprise), a user of the service (private individual/household), and a worker who is employed and remunerated by the enterprise. Private individuals pay the enterprise for the services rendered through cheques that are issued by an authorized enterprise. The cheques are given to the workers who, in turn, hand them on to their employing enterprises, while the latter give them back to the company issuing the cheques. Direct employment in which the private individual acts as the employer with no mediation is not allowed under the TS system. Encouraging the creation of enterprises supplying domestic services – what might be termed the ‘industrialization’ of paid domestic services – responds to

Country Name of measure Belgium

France

Fiscal or other incentives

Services covered

Social security Public benefits covered subsidies

Institutions involved

Titre-service 30 per cent (2004). fiscal deduction for the final user (cheque is purchased at a7.50; final cost after fiscal deduction is a5.25).

Services de proximité including: (i) inside the household: home-work services; (ii) outside the user’s household: dairy shopping; tran sport of disabled persons; ironing for the end user in the employee’s household.

General health care, workers’ compensation for employment injuries, retirement pension, maternity protection, unemployment insurance.

Yes (fully subsidized by the state)

Workers, service users and licensed enterprises-whether public, from the social sector or profit-oriented.

Chèque Emploi Service Universel (CESU).

Services à la personne : (a) inside the household: housework services; gardening; bricolage; childcare; school support; assistance to persons in need of care (old-age, sick and disabled persons); administrative assistance; (b) outside the household: childcare (crèches, guarderies, etc.); various deliveries (food, shopping, ironed clothes); assistance in transport and travel of persons in need of care (old-age and people with disabilities and children).

General health care, workers’ compensation for employment injuries, retirement pension, maternity protection, unemployment insurance.

Yes, but on a costsharing basis with the institutions involved

(i) direct employment: Individual employers and workers; (ii) mode préstataire : Individual employer, broker agency, and worker; (c) mode mandataire: licensed enterprise – whether public, from the social sector or profitoriented, employee, and service user.

Tax and social security contributions’ rebate for both final users and institutions involved

Childcare, bricolage and gardening excluded.

266

Table 10.1 Main features of the ‘cheque’ systems in Belgium, France and the Canton of Geneva (Switzerland)

Canton of Geneva

Chèque Service

No

Only home-based services, including: housework services; childcare; and old-age care; gardening.

Compensation No for nonemployment injuries (if working more than eight hours per week); compensation for employment injuries; retirement pension; maternity protection; family allowances; unemployment insurance.

Individual employer, workers, not-for-profit organizations and public bodies.

267

268 Regulating for Decent Work

the double aim of professionalizing and improving the quality of jobs in this sector (Devetter and Rousseau 2009). The wages of service providers under TS contracts are fixed by collective agreements stipulated by the sous-commission paritaire 322.01 (minimum rate is a9.11 per hour). Licensed enterprises can include profit-seeking private enterprises, public enterprises, and also enterprises operating in the social sector, which may vary in terms of size and also in terms of their previous experience in the sector (Defourny et al. 2009). In France, CESU5 was introduced in 2006 as a simplified instrument for hiring, remunerating and paying social charges to workers supplying services à la personne , including housekeeping. CESU is one part of a broader national plan which sees care work as a sector with substantial potential for employment creation. The CESU system covers both the services provided by a licensed enterprise (mode prestataire) and those rendered by domestic workers who are employed by private households. Workers can be hired directly or through the intermediation of a broker agency (mode mandataire). Depending upon the type of services provided, there is some variation in the procedures and requirements that the enterprises must meet, with stricter and more demanding rules being in place when relating to care for children or for clients with limited personal autonomy (CERC 2008). The enseignes, institutions recently introduced into the system, seek to boost the demand for the services delivered through the mode prestataire, by putting the service providers and potential service users in contact. CESU cheques encompass two types of cheques, namely the CESU bancaire and the CESU pré-financé . The CESU bancaire, which constitutes a simplified way to pay social security contributions, can be used only to pay workers employed by the private individual/household whether directly or through an intermediation agency. It can be obtained at banks and is composed of two declaration forms: one for declaring the remuneration of the workers (paid in the traditional form: cash, cheques, and so on) and another for declaring their social charges to the Centre National Chèque Emploi Service Universel (CNCESU). The workers will be paid at least the minimum applicable wage (SMIC). Upon receipt of the declaration form from the CNCESU, employers’ and workers’ contributions are calculated and automatically deducted from the employers’ account. As for casual work, the declaration form acts as an employment contract. The CNCESU addresses to the worker a certification which exonerates the employer from the obligation of releasing a pay slip. The CESU pré-financé is used instead to pay services rendered by all workers, whether employed by private households or by licensed

Manuela Tomei 269

enterprises (mode prestataire). Childcare – whether by individual workers or by institutions such as crèches and nurseries – outside the home of the service users can also be remunerated with this type of cheque. CESU préfinancé is usually paid as a salary complement, comparable to a food voucher, and may be financed either wholly or in part by the employer. In France, direct employment for personal services is predominant with more than 83 per cent of the workers in the sector engaged in direct employment, of which 10 per cent are supplied by placement organizations (CERC 2008). Workers in direct employment are covered by the Labour Code and by a Collective Agreement (Convention Collective nationale des salariés du particulier employeur ) negotiated by Fédération des Particuliers-Employeurs (FEPEM) and the main national trade unions (CFDT, CGT, CFTC and FOU). The existence and content of this agreement, however, is often unknown to employers and workers and its enforcement difficult to monitor (Sansoni 2008). In the Canton of Geneva, direct employment by private households is the only modality envisaged. By adhering to CS, the employer pays the net salary to the worker and communicates the salary paid to the CS, while disbursing an advance on the social security charges. CS then calculates the social charges due and deducts them from the employer’s advance. CS also provides the worker with an annual wage statement. Moreover, adherence to CS entails the automatic application of the provisions embedded in the standard employment contract applicable to the sector, namely the ‘Employment contract between the employer and the worker using Chèque Service ’. According to this model contract, payment of the minimum wages, among others, has been compulsory since May 2005. This contract refers to the model contract for full-time and part-time domestic workers, which entered into force in 2004 and is compulsory in respect of the domestic work employment relationships in the Canton of Geneva. The CS is generally run and managed by public bodies and not-for-profit associations. Fiscal and social incentives In Belgium the cost per cheque is a7.5 per hour, but, when the tax rebate is included, the cost to the user declines to a5.25.6 For each hour worked, the company issuing the cheques reimburses a20.80 to the registered enterprise, one-third being the user payment (a7.5) and twothirds of the state subsidy (a13.30). In addition, the user of the services benefits from a tax rebate. In France, both CESU bancaire and CESU préfinancé offer important fiscal and social advantages. In respect of direct employment, the

270 Regulating for Decent Work

employer benefits from a tax credit7 or a tax reduction8 up to 50 per cent of payments (wages and social security contributions) When opting for the services of a licensed enterprise, the applicable VAT rate to the service user is only 5.5 per cent. Moreover, licensed enterprises, in contrast to the enterprises supplying the same services but being out of the system, are exonerated from the payment of social security contributions of their employees for an amount equivalent to the social charges that correspond to the SMIC (CERC 2008). The employers, who cost-share the programme though the pre-funded CESUs, do not pay social security contributions (up to an annual ceiling of a1,830 per employee) and benefit from a tax credit up to 25 per cent of the profits generated through their financial contributions to the scheme. Since January 2009, following the introduction of a family tax credit, the cost for the employer has declined even further: After all fiscal benefits have been deducted, a a100 CESU voucher costs the employer just a42, making pre-funded CESUs much less expensive than a direct salary increase.9 In 2008, France devoted a8 billion to support personal service job development, including direct allowances, tax deductions, and reduced social contributions (ibid.). In Switzerland, no fiscal deductions or other benefits are provided for the users of the CS. This is consistent with the latter’s exclusive concern with raising social security incomes and broadening the Canton’s tax base, the promotion of the outsourcing of household work not being one of its expressed policy goals. The CS is likely to remain confined to French-speaking Cantons and the Canton of Berne. Recently, the Federal Parliament has rejected parliamentary proposals aimed at encouraging and supporting the CS system also in the German-speaking Cantons on grounds of the high cost associated with the management of the CS system at the federal level, and its overlap with the new procedure introduced by the Federal Act against Informal Work.10 This Act, which entered into force in January 2008, provides for a simplified and centralized procedure for calculating and paying social security charges and taxes relative to unspecified less-remunerated activities. Domestic work, as low-wage work, is thus covered. Cantons are free to maintain or introduce new CS systems, as they deem best. The employer is also free to choose the system of her/his preference (new simplified procedure or CS) for paying social security contributions and taxes regarding services rendered by her/his domestic employees. The Law in question, however, entails lesser protections than those guaranteed through the CS system, since it does not provide for maternity protection which is of crucial importance to a workforce where women prevail.

Manuela Tomei 271

Assessing the impact on the quantity of employment Job creation is an explicit goal pursued by the Belgian and French public policies geared towards the development of the domestic service sector. In Belgium, between 2006 and 2007, the number of TS workers increased by over 40 per cent, while that of registered users grew by 44 per cent. Reimbursed cheques rose by 53 per cent (nearly 50 million service cheques were reimbursed to enterprises in 2007), while the number of licensed enterprises grew by about 30 per cent, reaching a total of 2,050 enterprises. In terms of job creation, about 87,152 people were employed in 2007, representing a 40 per cent increase relative to the previous year. Of this total, 98 per cent were women, 46 per cent previously unemployed and 39 per cent had limited qualifications.11 In France, the National Agency for Personal Service, established in 2005 by the government to coordinate public policies in this sector, reported the creation of 1,925,400 jobs in 2007 compared with 1,765,400 in the previous year. Of the total number of jobs created in 2007, one-third related to services à la personne . In this sector, the number of full-time jobs reached 46,000 – compared with 33,000 in 2006. By 2007, 1,404,754 individual employers used CESU bancaire and 733,375 workers had been registered. As per CESU préfinancé , the amount of cheques issued in 2007 doubled relative to 2006. In January 2008, 85,898 subjects (natural or legal entities) accepted payment with CESU préfinancé . Of these, 73,366 were private individuals (especially assistantes maternelles), and 9,532 were institutions such as crèches. In 2007, housework featured as the most requested amongst services à la personne at around 41 per cent of the total, followed by ‘other’ (various non-specified activities) at 17.5 per cent, gardening at 15.5 per cent, home-based childcare at 12 per cent, computer and internet assistance at 7.5 per cent and petit bricolage at 6.5 per cent.12 Workers in the services à la personne sector are predominantly women. In 2006, around 58 per cent of the total performed domestic services, 31 per cent were assistantes maternelles and 11 per cent worked in home-based care activities such as care of oldaged persons. Compared with other sectors, workers providing services à la personne are more likely than others to work for several employers (47 per cent in 2006). This seems to be the case, in particular, for assistantes maternelles and home-based childcare (68 per cent) and workers doing housekeeping (40 per cent) (CERC 2008). In the Canton of Geneva, the few available data show that, despite the lack of fiscal incentives, the system of CS has contributed to a substantial increase in the number of registered workers within 18 months of its introduction.13 The entry into force of the Federal Act against

272 Regulating for Decent Work

Informal Work, in January 2008, seems to have given an additional boost to CS: the number of persons affiliated to CS jumped from 1,392 in September 2007 to 3,430 in June 2008.14 Assessing the impact on the quality of employment Under the TS system in Belgium, workers do not require any special qualifications. This is consistent with the scheme’s concern which is reducing long-term unemployment. As a result, the beneficiaries of unemployment benefits and/or social aid, often with no previous work experience in this sector, make up a significant proportion of the workers involved in the system. Workers may be granted short-term or fixed-term contracts for fulltime or part-time work. Originally, two different employment contracts were issued depending upon whether or not the workers were already on unemployment or social aid schemes.15 To date all workers, without distinction, are entitled to fixed-term contracts after three months of short-term contract. This modification was introduced by the government in response to the criticism formulated by the Court of Auditors. The latter, as well as other analysts (Nassaut et al. 2007), signalled a number of additional flaws such as the lack of clear parameters to establish what is to be considered a ‘job’. The absence of clear criteria results in the overestimation of the number of jobs created, as all the workers who transit through the system, even temporarily, are counted for. Another problem relates to the non-observance of the Law on Labour Contracts which requires workers not to work less than 600 hours per year. The Social Security Office puts the estimates at 63 per cent of the total, while the annual evaluation reports, produced as part of the monitoring system of TS, register a decline from 20 per cent, in 2005, to almost 11 per cent, in 2007. While different sources produce different data, in all events the number of workers below this threshold remains significant. In addition to working for short hours, often against their will, many TS workers also work for a number of different employers. This makes it more difficult for workers to have a sense of loyalty in relation to both their employers and end-users, while deepening the isolation that is typically experienced by domestic workers. In response to these criticisms, the government, after introducing one single contract for all workers, is in the process of drafting a law which envisages the indexation of TS exchange value and regulates minimum working time for TS workers.16 In France, the quality of jobs of domestic workers varies depending upon whether household services are provided through the mode mandataire, in which domestic workers are placed in individual households by a broker

Manuela Tomei 273

agency, or the mode prestataire, in which licensed enterprises put the services of their employees at the disposal of the end-users. It has been reported that many workers hired initially through the mode mandataire shifted, eventually, towards direct employment (Croff 2007). The reason appears to lie in that, while the mode mandataire may be a useful means to increase the number of hours worked, the costs associated with it, namely the intermediation costs and the lower net wages, outweigh the benefits. Moreover, the fact that the intermediary agency tends not to intervene when the agreed job assignments are disregarded by the client creates strong incentives for the worker to bypass the broker agency. The provision of services through the mode prestataire form allows for increased control by workers over their daily workload, and for an improved adjustment of their schedules to the demands of private households. Opportunities for training are also more readily available and, overall, the working conditions of employees of licensed enterprises appear to be better than those of workers directly employed by private individuals. The quality of jobs, however, is uneven across enterprises. This is the result of the heterogeneity of the licensed enterprises, the growing competition of commercial companies over not-for-profit organizations and the lack of standardized control of the quality of their operations. Enterprises from the social sector are more likely to issue employment contracts without time limits and to secure their employees a greater number of hours of work than either public enterprises or profit-seeking private companies (Defourny et al. 2009). They also provide services that are better tailored to the needs of their clients: home visits are conducted with a view to ascertaining the client’s requirements, thereby making a more realistic assessment of the tasks to be performed and the workload of their employees, and identifying possible sources of occupational hazards for both the client and the employee. These enterprises are also found to provide workers with opportunities for regular informationsharing and training activities that help to reduce their isolation, while enhancing their productivity. (This shows the crucial importance of establishing a serious monitoring system that makes regular assessments of the impact of the operations of the participating enterprises against a transparent and common set of benchmarks.) From the perspective of the services’ user, the advantages of the mode prestataire lie in that recruitment and selection, as well as layoffs, annual paid leave and sick leave are all taken care of by the licensed enterprises. The overall cost, however, as seen later, may be higher than in the event of direct employment. The large recourse to the latter, as pointed out

274 Regulating for Decent Work

earlier, explains the high incidence of part-time jobs and short-term contracts under the CESU system (part-time absorbs 70 per cent of housework and domestic help, while only 19 per cent of domestic workers and 23 per cent of home-based care providers have fixed-term contracts) (Sansoni 2008). The persistence of informality among domestic workers Despite some progress, informality persists. For many workers, the benefits of social security coverage are perceived to be of marginal value, whilst the cost in the form of lowered net income after taxation is more tangible. This may hold true especially for migrant workers, particularly when they have no residence permit, as well as for workers who may already avail themselves of social benefits, although indirectly, through their spouses. From the employers’ viewpoint, informal work continues to be more attractive for those services with a higher intermittent nature such as occasional childcare. Furthermore, opting for formal employment may, in some instances, exceed the ceiling for benefiting from fiscal advantages (CERC 2008). This has contributed to the creation of a grey area in which only part of the hours worked are declared to CESU (Devetter and Jany-Catrice 2009), thus increasing net wages for the workers and lowering the social security charges to be paid by the employers. This grey area is very difficult to monitor because households fall outside the outreach of labour inspection. Moreover, the fragmented nature of domestic work, with the same worker often working for various employers for a variable number of hours a week (it is not unusual to find workers who work for around a dozen different employers every week), adds to the challenge. This situation resembles that of the Canton of Geneva where, since the introduction of the CS, the number of declared workers has increased substantially, but the number of hours worked did not increase in a commensurate manner (Flückiger and Pasche 2005). A plausible explanation for this mismatch is that the CS, in contrast to the French CESU, allows foreign workers with no regular residence permit to be paid through these cheques and be granted all the corresponding entitlements. The workers’ participation in the scheme, however, does not alter their irregular status. This deters workers from declaring the full amount of hours worked. Should they do otherwise, the daily income would be smaller and they would have not guarantee to withdraw the accumulated benefits in case of deportation. Moreover, as already mentioned, many domestic workers may already be covered by social security as dependent family members and therefore

Manuela Tomei 275

have less incentive to secure social benefits through their paid work. Furthermore, the existence in a number of industrialized countries of a generalized old-age minimum pension benefit, independent from the contributory life of an individual, conditional only to income tests, may also constitute a disincentive for workers to participate fully in the above-mentioned schemes. The gradual improvement of both the scope of coverage to other benefits, such as health coverage and family allowances, and the entitlements, conditions and the levels of benefits may constitute one way of overcoming the present constraints. Professionalizing domestic work Along with wages and other working conditions, workers’ skills and qualifications are an important determinant of the quality of the services rendered. Low education levels are a shared feature of domestic workers in Belgium, France and the Canton of Geneva. In the latter, for instance, domestic workers have very poor formal educational attainments in comparison with workers in occupations for which no particular qualifications are required (Flückiger and Ferro-Luzzi 2009). In Belgium, the heavy reliance on the long-term unemployed has had the effect of lowering the quality of the services on offer, especially care-related services that – prior to the introduction of the TS – were provided by qualified personnel (Sansoni 2008). The deterioration of the quality of services has become a source of concern for the government and a potential threat to the legitimacy of the TS in view of its high operational costs and heavy public subsidies. In order to respond to this concern, in 2007 the government established a dedicated training fund to improve the workers’ skills and competencies. Under the Fund, licensed enterprises can request the partial reimbursement of training fees17 through the Belgium National Employment Office. In France no such dedicated fund exists. However, with the consent of the employer(s), domestic workers may benefit from different forms of vocational training for a maximum of 40 hours per year.18 The FEPEM, an employers’ organization that represents over three million individual employers employing domestic workers – or 12 per cent of all French households – promotes a rich menu of training activities, including long-distance learning, on a range of subjects from preventing and addressing violence against elderly people to childminding for pre-school or pre-adolescent children (FEPEM 2008). No training opportunities are envisaged under the CS in Switzerland, as the scheme is not interested in either the development or the modernization of the household services as a distinct sector.

276 Regulating for Decent Work

Specifying the content of domestic work and describing the corresponding tasks; identifying and certifying the range of skills and competencies required to perform it in a professional manner, including the capacity to detect and address the potential sources of occupational risk and hazard; and developing training programmes for domestic workers are necessary to challenge the entrenched view of domestic work as work which anyone can do and thus must be unskilled and relatively worthless. In the case of migrant domestic workers, improving their proficiency in the national language of the host country would also improve their communication with their employer, thereby preventing unnecessary misunderstanding and tensions, and facilitate their integration in the local community. The identification of the multiple occupations and the corresponding set of requisite knowledge and skills, comprised under the generic term ‘domestic work’, would help overcome the present blurred frontiers among such occupations (a cleaner may also be requested to undertakes childcare, while a childminder may also be expected to deal with cleaning and ironing). This is likely to encourage the specialization of domestic workers, while contributing to the emergence of a professional identity which is missing at present. Brazil: an integrated approach towards domestic work Brazil is a middle-income country which in recent years has been experimenting with tax deductions to employers as a means of encouraging them to make regular payments of the social security charges of their domestic workers and pay them at least the minimum wage. At this stage it is too early to draw any conclusions about the effectiveness of such fiscal policies in reducing the incidence of undeclared domestic work. Brazil, nonetheless, is an interesting case because these subsidies connect to a broader public agenda aimed at improving the overall performance of labour markets through, among others, new approaches to formalization and greater respect for the law (Berg in this volume). The attention paid to domestic work has also to be seen as the government’s determination to narrow the huge, historical divide between the richest and whiter stratum of society and the poorest and darker bottom of the social ladder. This public agenda challenges the portrayal of Brazil as a ‘racial democracy’ and recognizes that race, rather than simply class, is an important determinant of social and economic disadvantage, and that gender and racial discrimination reinforce each other, domestic work being paradigmatic of the disadvantages stemming from multiple discriminations. Until 1888, when slavery was abolished in Brazil, domestic work was a slaves’ job and ever since this time it has continued to represent the

Manuela Tomei 277

main occupation of Afro-Brazilian women. The exclusion of domestic work from the scope of labour law until the early 1970s and the low educational endowments of the black population, especially black women, have contributed to this state of affairs (Gomes and Bertolin 2010). The measures introduced by the government of President Lula to lend visibility and dignity to domestic work are therefore to be seen as recognizing a long overdue societal obligation towards the country’s black population. In 2007, of the more than six million domestic workers, 94 per cent were women, 61 per cent black and 39 per cent white. Domestic work represents more than 16 per cent of women’s total employment. Although black women constitute the majority of domestic workers, white women show higher levels of formality (the percentage of white women with carteira assinada was 30 per cent, compared with 25 per cent of black women). Disparities between black and white women also exist in respect of their labour earnings, but this gap has been declining since 1996 (Sanches 2009). A turning point in the country’s process of recognition of the value of domestic work and of domestic workers’ rights was the inclusion in the 1998 Constitution of a number of important labour guarantees to these workers, including social security rights and domestic workers’ right to organize. The enactment of these Constitutional provisions was an essential, first step in the formalization and effective protection of domestic workers, considering the obstacles that they and their employers face in the exercise of their right to collective bargaining (ILO 2009a). The classification, in 2008, of domestic work under the age of 18 as one of the worst forms of child labour has been another key legal measure challenging the perception of domestic work as risk-free and raising public understanding of the many occupational risks with which it is associated. In 2005, for the first time the government launched a pilot, targeted programme at the federal level – the Programa Trabalho Domestico Ciudadao (Solange Sanchez 2009). The programme brings together the Ministry of Labour, the Domestic Workers’ Federation (Federaçao Nacional das Trabalhadoras Domésticas (FENATRAD)), other trade unions, the Special Department for Promoting Racial Equality Policies (SEPPIR), the Special Department for Women Policies (SPM), the Human Rights Department, and the ILO. The programme comprises three main axes of work: the first one is aimed at improving the schooling and skills of domestic workers; the second is geared towards strengthening the capacity of domestic workers and their organizations to exert their representational

278 Regulating for Decent Work

rights and influence public policies; and the third component aims at revaluing domestic work through awareness-raising campaigns around a range of issues, such as human rights, violence against women, occupational safety and health and child domestic work. This multi-pronged approach tackles several determinants of the low status of domestic workers, namely the low educational levels and the lack of or weak collective representation. The programme, however, has not managed to move beyond mere experimentation and expand its (limited) geographical scope (Gomes and Bertolin 2010). A further improvement of the status of domestic workers was brought about by Act No. 11.324 in 2006. This recognizes a number of new rights for domestic workers and provides fiscal incentives towards the formalization of domestic workers. The main provisions include: (i) a right to 30 days’ annual paid leave and to public holidays; (ii) employment protection in case of maternity; (iii) prohibition of salary deductions for meals and accommodation; and (iv) fiscal incentives for employers registering their domestic employees with the National Social Security Institute. This Act was adopted within the framework of a broader national policy seeking to expand the social security coverage to the most vulnerable categories of workers. Until 2012, the employer of a domestic worker may deduct from her income tax the social security contributions paid for the worker. This fiscal incentive is limited to one domestic worker per employer. Deductions shall not exceed the employer contribution calculated on the basis of at least the minimum wage (including 13th month payment and holidays), and are subject to the verification of the employer’s compliance with her social security obligations.19 It is as yet too early to assess whether these fiscal incentives will result, eventually, in an increased formalization of domestic work. Nevertheless, the data of the Ministry of Social Security concerning the period 2005–2007 show only a one percentage point increase in the numbers of domestic workers aged between 15 and 59 years who are in possession of a ‘carteira assinada’ (namely workers with a signed labour card which entitles them to full protection). This suggests that the employers availing themselves of these incentives were those who had already registered their domestic workers with the competent authorities. This development is not particularly encouraging considering that the Secretariat of Policies of the Ministry of Social Security had estimated that approximately 20 per cent of informal domestic workers should have benefited from this fiscal policy measure. A wider and better-planned promotional campaign, nonetheless, could help to raise, at least in part,

Manuela Tomei 279

the proportion of domestic workers with a signed labour card (Berg in this volume). It is worthwhile signalling, however, that a trend towards the formalization of domestic work had been registered as early as the 1990s and continued, albeit at a lower pace, in the 2000s. This is possibly the result of a combination of socially inclusive policy measures, some of which had been initiated during the previous administration, such as sustained upward real adjustments in the national mandatory minimum wage, which covers domestic workers, and conditional cash transfers to low-income households in order to, among others, encourage them to keep their adolescent children in school thereby reducing labour supply pressure. This suggests that integrated interventions, which combine better legal protection for domestic workers through awareness raising, improvements of skill levels and fiscal incentives, stand a good chance of bringing about sustained improvements in living and working conditions of domestic workers, when they are framed within broader public policies that seek to reduce the incidence of low-wage jobs and narrow racial and gender inequalities.

Concluding remarks The historical exclusion of domestic workers from workers’ status and fundamental employment guarantees will not be overcome overnight. Several different factors contribute to this situation: the home as the workplace; the personal and intimate nature of domestic work; the systematic undervaluation of this work because it largely involves work performed by low-income and migrant women for other women; and the geographical fragmentation and isolation of domestic workers. Nonetheless, growing attention, both nationally and internationally, is being paid to their particular circumstances, and different experiments to uplift their status are underway. This chapter has reviewed national public schemes in Belgium, France, the Canton of Geneva (Switzerland) and Brazil that encourage, through various incentives, the outsourcing of unpaid female care and housekeeping work. While in Brazil only domestic workers working for an individual employer are involved, in the European context, the reviewed programmes encompass workers working for multiple households and different employers. These may comprise individual employers as well as licensed enterprises – whether public or profit-oriented – that provide domestic services to private households. A common feature of all

280 Regulating for Decent Work

these schemes is the active involvement of the state and formal market institutions in a domain which has traditionally been perceived as a private affair, and the ‘sacred’ preserve of families and households. Another shared trait of these programmes is their emphasis on reducing transaction costs for employers through simplified recruitment formalities and simplified payment procedures. In all cases, except for the CS in Canton of Geneva, fiscal incentives are also provided to reduce the cost of hiring domestic workers. An incentive-based approach appears to be of particular relevance to a sector in which the conventional systems of verification of compliance with labour law do not easily and automatically apply to the home as a workplace. The fact that adherence to these programmes entails the automatic application of a set of clearly defined labour norms, namely the standard employment contract for the domestic work sector in the case of the CS, in the Canton of Geneva, and the sectoral collective agreements in the case of Belgium and France, helps employers to think of themselves as employers and departing from conditions of employment set arbitrarily through ‘word of mouth’. The review shows that these schemes, whether or not accompanied by fiscal subsidies, have helped to reduce the incidence of informal domestic work, as illustrated by the considerable increase in the number of workers registered with the Social Security Systems of all the countries concerned. Nonetheless, other types of underreporting have emerged. In Europe, for instance, the growth in the number of registered workers has not been matched by a proportional growth in the number of declared hours of work. This suggests that the incentives put in place are either insufficient or not well targeted. Closer attention should be paid to the factors and circumstances that may encourage domestic workers to forgo under-the-counter payments and prefer higher cash income today than social benefits in the long run, especially when undocumented migrant workers are involved. One possible way out could consist of providing higher and short-term, rather than long-term, benefits, such as health insurance. One hypothesis of this chapter was that the ‘industrialization’ of domestic work, through the creation of salaried posts within service agencies, would result in greater standardization of the employment relationships, and in lower levels of arbitrariness and volatility. This chapter has shown that, while the low overall quality of such jobs remains an issue, workers employed in enterprises that provide services to private households, as in the case of TS Belgium and CESU mandataire in France, stand a better chance of seeing their worker status recognized

Manuela Tomei 281

also because the identity of the employer is clear. In general, these workers have higher and more stable hours of work, greater training opportunities and may be less isolated (although this is not necessarily the case) than workers employed by an individual employer either directly or through an agency. In the CESU scheme in France and the CS scheme in the Canton of Geneva, for instance, workers engaged in one-to-one relationship and working for multiple employers face greater difficulties in adjusting their schedules to that of the employing households, greater volatility in terms of hours and wages, and receive no compensation for the long distances that they may have to cover in order to move from one household to another. Not all licensed enterprises, however, guarantee a comparable level of conditions of work and quality of service. Enterprises from the social sector are found to be better employers than profit-seeking enterprises; in particular, temporary employment agencies tend to pass on to the worker the entire costs of the flexibility demanded by the service user. This raises the question of whether or not such double standards should be accepted in a system which receives generous subsidies from the state and provides what can be seen as an essential service to society. A related issue which this chapter has not examined in any detail, but which is worth mentioning, relates to the distribution of state subsidies across households by income level. The available literature suggests that better-off households – those that can afford to pay for domestic services in any event – tend to derive the greatest benefit from such schemes. Moreover, affluent households do not necessarily comprise dual-earning couples. This suggests that the assumption that these schemes would enable well-qualified women to better articulate their work and family responsibilities, thereby improving their relative position to men in the labour market, was misplaced (Windebank 2007). Quality domestic services, especially caring for children or frail adults, require proper training, regular supervision and decent working conditions to secure workers’ loyalty. In the past few years some important measures have been taken to overcome these problems. These include, for instance, the creation of the National Agency for Personal Service and also of the enseignes, institutions responsible for guaranteeing a better matching of the demand and supply of domestic services in France. But a well-functioning monitoring system that certifies and controls, against a clear set of benchmarks, the quality of jobs and services rendered by all participating institutions – both public and private – and penalizes underperformance, is essential to resolving the present disparities in both the quality of jobs and the services on offer.

282 Regulating for Decent Work

While employees of not-for-profit licensed enterprises may fare better than those employed by private individuals, the ‘industrialization’ of domestic work has not managed, to counter the undervaluation of domestic work and its characterization as a low-wage job. The depersonalization of the employment relationship and the shift towards more standardized employment relationships has, paradoxically, eroded the pressuring power that workers engaged in direct employment can exert on their employers, especially when labour supply is relatively limited, and made competition among employees fiercer. As a result, employment in the domestic sector remains fundamentally precarious. Moreover, given the fragmented and isolated nature of domestic work, the creation of a more formal relationship has not resulted either in a growth of trade union membership in the sector. The creation of domestic workers’ cooperatives may be more conducive to breaking the isolation of these workers and reinforce solidarity among them, as shown by a number of countries (ILO 2009a). The persistent low valuation of domestic work in the Canton of Geneva is consistent with the CS’s exclusive concern with expanding the Canton’s tax base in order to increase social security incomes, while in Belgium and France it is related to the tensions between the different goals pursued by the TS and CESU, respectively. The urgency to generate large numbers of new jobs, especially for groups of workers of low employability, cannot be easily reconciled with the proclaimed goal of improving the quality of jobs that are highly informal. Granting domestic workers higher and more stable wages, less volatile working hours and social benefits has the effect of raising the cost of domestic services that state incentives can offset only in part, especially if the quality of the services provided do not meet households’ demands or expectations. This, in turn, jeopardizes the viability of significant improvements in domestic workers’ working conditions. Moreover, these schemes have perpetuated the image of domestic work as the work of women, and, more specifically, low-income and low-qualified women, thus entrenching the view of domestic work as work of little value. This has been compounded by the absence in the original design of the TS and CESU of a coherent strategy to professionalize the sector through the establishment of a skills certification system and the offer of a well-articulated menu of training courses. Nonetheless, the measures recently adopted in Belgium and France to redress this situation, including through the supply of public funding for encouraging employers of domestic workers to invest in the training of their employees, are encouraging. The Brazilian Programa Trabalho Domestico Ciudadao,

Manuela Tomei 283

despite its limitations in terms of scope and duration, is particularly interesting as it combined the offer of training specifically related to the occupation in question with literacy programme and para-legal training on labour rights. It has been argued that vocational training for domestic workers may, at best, secure career progression within the sector – a ‘maid’ in a private household may take up a job as a cleaner in a commercial office (Lautier 2003). While this may be the case, this move, nonetheless, may grant workers with social security and pension benefits that, otherwise, would not be easily made available to them. But, most importantly, occasions to get together, discuss and share experiences help forge a common professional identity which is often weak because, among others, their assignments often cut across several occupations. Developing such an identity is an indispensable precondition for collective representation and action for better working and living conditions. In instances where fiscal space is narrow, systems like that of the TS or CESU in Belgium and France are not viable. Furthermore, they are less suited to realities in which full-time domestic work is still prevalent, domestic labour supply unrestricted and industrial relations systems less well developed. In these countries, as evidenced by the Brazilian experience, legal action that recognizes that domestic workers have fundamental employment guarantees on a par with other workers is an essential, first step towards their effective protection. Laying down the conditions for effective law enforcement is equally important, especially considering the legal and administrative obstacles facing labour inspection in verifying compliance in private households. Another major challenge lies in the fact that domestic workers and their employers often fail or may even resist perceiving themselves as such. It is therefore important that they understand that domestic work is work and that domestic workers have labour rights the respect of which is a guarantee of greater reliability for and accountability vis-à-vis the household. Awareness-raising campaigns, such as that in Brazil, can help to lower the level of tolerance towards abusive practices against domestic workers and upgrade their social status. Minimum wage policies, when not discriminatory and enforced, may not only secure higher labour earnings, but may also improve other non-wage working conditions such as (lesser) hours of work (Hertz 2004). They can also prove to be an important means to reduce pay inequities between men and women at the lower end of the wage scale (Rubery and Grimshaw in this volume). Acknowledging the common challenges that are faced by domestic workers – whether full-timers or part-timers, live-ins or live-outs – because

284 Regulating for Decent Work

of both the site in which they work and the gendered nature of domestic work, while stressing their contribution to the functioning of contemporary society, are key elements of this collective identity.

Notes 1. The ILC adopted, in 1948 and 1965 respectively, a Resolution Concerning the Conditions of Employment of Domestic Workers and a Resolution calling for normative action. In 1970, the ILO published the first survey ever produced on the status of domestic workers worldwide. 2. For a review of the history, rationale and implications of the contracting out of care labour and services by the State in the US, see Boris and Klein (2007). 3. The Chèque S ervice was first introduced in the Canton of Valais in 1999; this was followed by the Canton of Geneva in 2004 and subsequently the other Cantons of French-speaking Switzerland. The Canton of Berne is the only non-French speaking Canton that has also introduced this scheme. 4. Licences are awarded by the Federal Ministry of Employment on the basis of a set of legal and administrative criteria. 5. CESU has been introduced by Act No. 2005–841 of 26 July 2005 regarding the development of personal services and different measures toward social cohesion. CESU unifies and substitutes the previous Chèque Emploi Service and Titre Emploi Service. 6. Recently, the legal basis for creating a new social TS has been approved. Social TS will allow low-income people to acceed to TS at a minimum price under a4. See http://www.emploi.belgique.be/defaultNews.aspx?id=23650. 7. For those exercising a liberal profession or those inscribed in the job-seekers list for at least three months in the year of reference. 8. For taxpayers and those not having any right to tax credit. 9. See: ACCOR Services, Ticket CESU. Un complément de revenue défiscalisé Online at http://www.ticket-cesu.fr/cesu.aspx?CatId=Entreprise_cesu&PageId= Entreprise_cesu_Avantages. 10. See http://www.parlament.ch/f/suche/pages/geschaefte.aspx?gesch_id= 20083372; http://www.parlament.ch/f/suche/pages/geschaefte.aspx?gesch_ id=20073041. In German-speaking Cantons there have been some initiatives for regularizing the employment of domestic workers with regard to social security coverage. With the exception of Berne, these initiatives were managed by private enterprises. 11. http://www.emploi.belgique.be/WorkArea/showcontent.aspx?id=18852. 12 http://www.servicesalapersonne.gouv.fr/Public/P/ANSP/Actus/RA_2007_ global.pdf. 13. http://etat.geneve.ch/des/site/master-content.jsp?componentId=kmelia518& themeId=2963&pubId=6281. 14. http://www.geneve.ch/dse/doc/communiques/2008-07-16–cheque-service. pdf. 15. Workers of category A (receiving unemployment benefits), unlike those of category B (all other workers), were subject to a minimum working time of three hours per day. In both categories, the employer had the obligation to

Manuela Tomei 285

16. 17. 18.

19.

offer to the worker who worked under a short-term contract a fixed-term contract (at the seventh month for category A and at the fourth month for category B). http://www.rva.be/D_new/Press/Mededelingen/2009_02_03/PersFR.pdf. http://www.emploi.belgique.be/defaultTab.aspx?id=10148; http://www.emploi. belgique.be/publicationDefault.aspx?id=23260. Training can take place during or outside hours of work and is conditional upon the prior authorization of all employers in the case of those workers working for various employers. FEPEM is present in the 22 administrative regions of France and provides training activities through a national network of 160 certified vocational training institutions. Social security contributions are 12 per cent of the wage. Because the measure is limited to the minimum wage, the employer would be able to reduce their income subject to taxation by R$735 in 2010 (12 per cent of R$510 per month).

References Blackett, A. 2005. ‘Promoting domestic workers’ human dignity through specific regulation’, in Fauve-Chamoux, A. (ed.), Domestic Work as a Factor of European Identity: Understanding the Globalization of Domestic work, 16th–21st centuries (Bern: Editions scientifiques européennes). Boris, E.; Klein, J. 2007. ‘We were the invisible workforce: unionizing home care’, in Cobb, D.S. (ed.), The Sex of Class: Women Transforming American Labour (Ithaca, NY: Cornell University Press). Croff, B. 2007. ‘La GRH demeure une question cruciale’, Développements , no. 45, février, pp. 6–11, cited in CERC (2008). Conseil de l’Emploi des Revenus et de la Cohésion Sociale (CERC). 2008. Les Services à la Personne (Paris: La Documentation Française). Online at http:// www.myobase.org/GEIDEFile/rapport-service-personne-2008.PDF?Archive=19 2415791069&File=rapport%2Dservice%2Dpersonne%2D2008_PDF. Defourny, J.; Henry, A.; Nassaut, S.; Nyssens, M. 2009. ‘Les titre services: quelle qualité d’emploi et d’organisation du service?’, Regards Economiques, no. 69, Avril, pp. 6–7. Devetter, F.X.; Rousseau, S. 2009. ‘The impact of industrialization on paid domestic work: the case of France’, European Journal of Industrial Relations, vol. 15, no. 3, pp. 297–316. Domestic Workers United and Data Center. 2006. Home is Where the Work Is: Inside New York’s Domestic Work Industry, 14 July. Dresser, L. 2008. ‘Cleaning and caring in the home: shared problems? shared possibilities?’, Industry Studies Conference Paper. Online at http://ssrn.com/ abstract=1123565. Dussuet, A. 2005. Travaux des Femmes: Enquête sur les services à domicile (Paris: L’Hartmann). European Economic and Social Committee. 2010. ‘The professionalisation of domestic work’, Opinion of the Section for Employment, Social Affairs and Citizenship (SOC/372), Brussels, 12 May.

286 Regulating for Decent Work European Foundation. 2007. Employment in Household Services (Luxembourg: Office for Official Publications of the European Communities). Fédération des Particuliers-Employeurs (FEPEM). 2008. Catalogue Formation 2008 des Salariés du Patriculier Employeur: Professionnaliser les emplois familiaux (Paris: FEPEM). Flückiger, Y.; Pasche, C. 2005. Analyse du secteur clandestin de l’économie domestique à Genève, Final report, Observatoire Universitaire de l’Emploi (Geneva: Laboratoire d’économie appliquée). Flückiger, Y.; Ferro-Luzzi, G. 2009. Le travail domestique en Suisse, Calcul d’un seuil de salaire en usage en vue de l’édiction d’un contrat-type de travail au sein du secteur des Services domestiques en Suisse (Genève: Observatoire Universitaire de l’Emploi). Ghosheh, N. 2009. ‘Protecting the housekeeper: legal agreements applicable to international migrant domestic workers’, The International Journal of Comparative Labour Law and Industrial Relations, vol. 25, no. 3, pp. 301–25. Gomes, A. and Bertolin, P. 2010. ‘Regulatory challenges in domestic work: The case of Brazil’ (Unpublished paper). Hertz, T. 2004. ‘Have minimum wages benefited South Africa’s domestic service workers?’ Working paper presented at the forum on African Development and Poverty Reduction: The Macro-Micro linkage, Somerset West, 13 to 15 October. International Labour Office (ILO) 2008. Global Wage Report 2008/9: Minimum Wages and Collective Bargaining – Towards Policy Coherence (Geneva: ILO). International Labour Office (ILO) 2009a. Decent Work for Domestic Workers, Report IV(1), International Labour Conference, 99th session 2010 (Geneva: ILO). ILO. 2009b. The Cost of Coercion, Global report under the follow-up to the ILO Declaration on Fundamental Principles and Rights at Work, Report I(B), International Labour Conference, 98th session 2009, (Geneva: ILO). Lautier, B. 2003. ‘Las empleadas domésticas latino-americanas y la sociología del trabajo, algunas observaciones acerca del caso brasileño’, Revista Mexicana de Sociología, Vol. 65, no. 4, pp. 789–814. Milkman, R.; Reese, E.; Ross, B. 1998. ‘The macro-sociology of paid domestic labour’, Work and Occupations, vol. 25, no. 4, pp. 483–507. Nassaut, S.; Vermer, M.C.; Nyssens, M. 2007. ‘Le titre-service, un état de la situation’, Regards économiques , no. 50, pp. 20–5. Razavi, S.; Staab, S. 2008. ‘The social and political economy of care: contesting gender and class inequalities’. Paper prepared for the Expert Group Meeting on ‘Equal sharing of responsibilities between men and women, including care giving in the context of HIV/AIDS’, 6 to 8 October (Geneva: UNRISD). Reinecke, G.; Valenzuela, M.E. 2008. Distribución y mercado del trabajo: un vínculo ineludible, April (Santiago: ILO). Online at http://www.oitchile.cl/pdf/publicaciones/ele/08-19.pdf. Rodgers, J. 2008. ‘Cambios en el servicio doméstico en América Latina’, in Valenzuela, M.E. (ed.), Trabajadoras domésticas: un largo camino hacia el trabajo decente (Santiago: ILO). Sansoni, A.M. 2008. ‘L’utilisation des vouchers en Belgique et en France dans le domaine des services à la personne: quelles leçons on peut en tirer?’, Working Paper Adapat, No. 62/2008. Sanches, S. 2009. ‘El trabajo doméstico en Brasil’, in Valenzuela, M.E. (ed.), Trabajadoras Domésticas: Un largo camino hacia el trabajo decente (Santiago: ILO).

Manuela Tomei 287 Thomson, M. 2009. ‘Workers not maids – organizing household workers in Mexico’, Gender and Development, vol. 17, no. 2, pp. 281–93. Vosko, L. 2006. ‘Gender, precarious work and the international labour code: the ghost in the ILO closet’, in Fudge, J.; Ownes, R. (eds), Precarious Work, Women, and the New Economy: The Challenge to Legal Norms (Oxford and Portland, OR: Hart). Windebank, J. 2007. ‘Outsourcing women’s domestic labour: the chèque emploiservice universel in France’, Journal of European Social Policy, vol. 17, no. 3, pp. 257–70.

This page intentionally left blank

Part IV Elements of Effective Regulation

This page intentionally left blank

11 The Impact of Labour Regulations: Measuring the Effectiveness of Legal Norms in a Developing Country Sangheon Lee and Deirdre McCann* Introduction It is paradoxical that the recent flourishing of research on the economic impact of labour regulations has exposed the simplistic understandings that pervade this literature and that underpin deregulatory policy agendas. The debate among economists is conducted between two competing schools, which might be characterized as ‘distortionists’ and ‘institutionalists’. These accord different weights to the economic costs and benefits of labour regulation and have embarked upon divergent quests for optimal institutions (for a review, see Bertola 2009 and Freeman 2009). Both strands of this research are problematic, however, especially in their relevance to developing countries, in that most of the studies assume the de jure and de facto reach of labour laws to be coterminous. In consequence, they neglect to consider the effectiveness of legal regulations (Aherling and Deakin 2007; Lee and McCann 2008). This chapter contends that analyses of the economic and employment impacts of labour regulations in developing countries will inevitably be misleading unless effectiveness is integrated into the analysis. In an effort to help shape such an approach, this work draws on new survey data from Tanzania to explore certain of the key dimensions of legal effectiveness, namely workers’ awareness of legal standards and the actions they take in response to the breach of these norms. Our rather optimistic estimates show that although the majority of workers in Tanzania are familiar with the statutory standards on working time, minimum wages and maternity leave, a substantial number, just less than * The authors have benefited from excellent research assistance by Ana Choi and insightful comments from Carola Nolte. 291

292 Regulating for Decent Work

one-third, are unaware of these legal standards. Further, workers’ awareness has little to do with the existence of a written employment contract, which suggests that employment contracts in Tanzania often lack the necessary substance. At the same time, workers’ awareness is strongly associated with better working conditions, particularly in respect of working time and maternity leave. While the channelling mechanism for this association is yet to be fully understood, this chapter will argue that this effect can take place through influencing workers’ job satisfaction and/or changing their behavioural patterns, especially vis-à-vis employers. These findings suggest strong potential for improving working conditions through increasing workers’ awareness, without substantial institutional reform.

The effectiveness of labour law in low-income settings Rather schematically, two central positions on the operation of labour law can be identified in economic impact literature. The first, the ‘formalist’ narrative, adheres to a mechanistic view of labour laws by assuming their application to be both comprehensive (laws protect all workers who fall within their scope) and complete (workers are entitled to the full array of legal protections outlined in legal texts, to the maximum permissible extent). Perhaps unsurprisingly, this crude notion of regulation has a muted presence in the scholarly literature. It has, however, found favour in the policy realm, where it underpins much of the deregulatory rhetoric that has in recent years been extended to developing economies. This account is most visible in the work of the World Bank’s Private Finance Initiative (PFI), and, in particular, in the assertions on the economic impact of labour standards made in the PFI’s Doing Business reports (see in particular World Bank 2005). While the research that underpins the Doing Business project offers a more sophisticated understanding of the operation of labour laws (Botero et al. 2004; although see Aherling and Deakin 2007), the Bank’s policy guidance has posited a relationship between labour laws and economic outcomes that is grounded in formalist assumptions and has generated simplistic conclusions about the damaging economic impact of labour standards (Berg and Cazes 2007; Lee and McCann 2008). The formalist discourse has fuelled an accompanying narrative of the urgency and promise of labour law reform that has been influential in many developing countries. Governments have been prompted or encouraged to embark on sometimes semi-continuous projects of reform that are carried out in the absence of empirically grounded understandings of the operation of the existing legal measures (on Asian countries, see Lee and Eyraud 2008).

Sangheon Lee and Deirdre McCann 293

The World Bank literature also illustrates the second key narrative of labour law’s operation in low-income settings. This ‘pessimist’ account depicts legal regulation as broadly irrelevant to the developing world for a variety of reasons, whether in the assumption that laws are so ineffectively enforced as to be irrelevant, so economically inappropriate to be damaging, or transplants from colonial ‘parent’ regimes that have inevitably failed to seed. One version of the pessimist narrative suggests a clear-cut dichotomy between ‘formal’ and ‘informal’ economies, with labour standards unknown in, or irrelevant to, the latter (World Bank 2005). Also identifiable is a degree of despair about legal and institutional change that contrasts with the optimistic account of reform harboured by the formalist narrative. This position in part reflects a suspicion about the worth of conventional enforcement mechanisms that is being challenged by the evolving literature on labour inspection in low-income countries (for example, Piore and Shrank 2008; Pires in this volume). It also questions the relevance of conventional legislated standards by constructing a dichotomy between these and other, ‘soft’, forms of regulation, primarily Corporate Social Responsibility (CSR) initiatives, and positing the latter as the preferred, rather than a complementary, mode of regulation. By rejecting both these narratives, this chapter is tied to literatures on the role of labour market regulation in low-income countries that are of a contrasting impetus and origin to the economic impact research. One is the evolving search, often animated by recourse to Sen’s capability theory, for a rationale for labour market regulation that can support both social and economic development (see, for example, Browne et al. 2005; Lee and McCann 2006; Langille 2007). A nascent interdisciplinary literature is also rejecting conventional assumptions about the role of regulation in developing countries and instead endeavouring to elicit the details of law’s operation in these settings and to suggest techniques for more effective regulation (Fenwick et al. 2007; Berg and Kucera 2008). These literatures are now sufficiently evolved to support investigations of the role and intersection of regulatory forms in low-income settings that can parallel the level of detail and contextual awareness that characterizes analyses of advanced industrialized economies. The chapter can therefore be situated in relation to a number of debates reflected in the literature on labour market regulation in lowincome countries. It is feasible, for example, to configure this analysis, at least in part, as an exploration of the role of legal regulation in the ‘informal economy’, in that an element of its enquiry is whether legal norms are, or have the potential to be, relevant in contexts in which

294 Regulating for Decent Work

their enforcement is unlikely. These contexts can therefore be subsumed within conceptions of informality, and in particular by the ILO’s reigning characterization of this phenomenon as de jure or de facto resistance to legal regulation.1 The chapter examines whether the law can exercise some degree of influence on firms and working relationships that have, at least in certain of their dimensions, a tentative relationship with state regulatory norms. In this regard, the analysis builds on a notion of ‘legal observance’, which has been elaborated elsewhere to capture compliance with the central substantive dimensions of legal standards in the absence of the likelihood of enforcement (Lee and McCann 2008; see also Frey in this volume). The notion of observance is drawn on in this chapter in both its descriptive and normative dimensions: in the former guise, to measure legal awareness; in the latter, to question whether formalization strategies can be tied to an influence of statutory standards that is distinct from the formal apparatus of the legal system. This endeavour also implicitly responds to the ongoing deliberations within labour law scholarship on the future of the field (see, for example, Davies et al. 2000; Conaghan et al. 2004; Davidov and Langille 2007; Arup et al. 2006). In this regard, it suggests one of the possible avenues towards eliciting a richer understanding of the present state of labour law, and therefore more effectively to predict or guide its future, is to engage in a detailed analysis of the operation of specific legal measures that is attentive both to their requirements and regulatory and national contexts. As a result, the chapter carves a narrow focus of analysis along the axes of regulatory technique, substantive subject and national context. With respect to regulatory techniques, statutory standards are singled out for examination: substantive requirements on working relationships that are imposed by the state (and that contrast with the ‘procedural intervention’ of collective bargaining) (Langille 2007). The selection of statutory standards may appear rather quaint. In recent scholarship, legislated norms often appear more often as a relic of labour law’s past than a symbol of its future. The shift towards CSR, for example, as both policy initiative and research preoccupation, is substantially a response to the recognized deficiencies of state-level regulation. In labour law’s reception of regulation theory, statutory standards have been associated with the discredited ‘command and control’ model of mandatory standards backed by sanctions.2 The literature on legal indicators has inspired protests over its preoccupation with statutory regulation to the exclusion of collective bargaining (Lee and McCann 2008). And some authors question the promise of the very project of statutorily mandated

Sangheon Lee and Deirdre McCann 295

labour rights. As Arthurs (2006) has suggested, ‘The greatest gains for workers might perhaps come, not from securing access to a fixed array of legally enforceable protection and benefits, but from being regarded as an indispensable component of every calculus of public policy which might affect them’ (p. 388). Against the grain of this disillusionment with statutory standards, the chapter makes a case for a careful examination of the role, significance and potential of legislated norms in developing countries. In part it is inspired by the turn to regulation theory in labour law scholarship, which in the process of rejecting ‘command and control’ models points to a future for statutory norms distinct from their conventional role. In such novel regulatory designs, statutory standards are aligned with the array of other techniques suggested to constitute ‘innovative’ regulatory frameworks (financial incentives, training, information campaigns and so on) (Arup 2001; Fenwick et al. 2007). As Fenwick et al. (2007) suggest in their study of laws on micro and small enterprises: ‘[I]nnovative approaches to regulation do not require the State to abandon traditional methods of labour regulation. … [L]egally binding standards must remain the touchstone of any regulatory system’ (p. 15). Such insights on the potential for statutory standards to form part of a broader array of regulatory techniques confirm the distinction between the substantive requirements of such standards and the techniques adopted for their implementation or enforcement, and thereby reinforce the contention that reform efforts should be centred on the latter (Lee and McCann 2008). They also help to situate labour standards in relation to evolving conceptions of formality and informality, in which working relationships and firms, rather than constituting distinct formal and informal ‘sectors’, occupy points along a continuum of formality that is shaped by the degree of compliance with legal requirements across a range of cognate fields (ILO 2002; Chen 2006). This in turn suggests the merits of investigating whether protective standards have more influence on ‘informal’ working relationships than is suggested by the formalist narrative identified earlier. To justify its faith in the exploration of statutory regulation, this chapter examines the relevance of legislated standards to workers themselves. In part, this approach is in line with the literature on labour law indicators, which has taken into account the actions of a number of labour market actors. These actors have been called on, for example, to assess the depiction of national laws in the relevant indices. Chor and Freeman (2005), for instance, have found the World Bank’s Employing Workers Index (EWI) to be out of line with understandings on

296 Regulating for Decent Work

the part of both trade unionists and employers on the requirements of a number of the domestic labour laws. The role of the judiciary has also been explored. In the economic literature, the intensity of labour inspection and magnitude of penalties for violation tend to be taken to account for ‘observance’ or compliance (for example, see Squire and Suthiwart-Narueput 1997). Other research, however, reveals a far more complex picture. A number of studies highlight the critical role of the judiciary in determining legal requirements (Bertola and Cazes 2000; Bhattacharjea 2006; Deakin et al. 2007); and reveal the difficulties in capturing this judicial role in the form of indices to be exacerbated when complex relationships exist between federal, state or provincial laws (see, for example, Bhattacharjea 2006; Mitchell et al. 2010). The impact of labour regulations has also be recognized to depend to some degree upon the response of enterprises to their requirements. Edwards et al. (2004), for instance, investigating the employment impacts of minimum wage legislation in small firms in the UK, found its effects to vary according to ‘competitive conditions’, which proved to be more influential than the regulations themselves. There is substantial evidence, then, of the role of executive, judiciary and enterprise behaviours in determining the impact of labour regulations.3 The response of workers, however, is largely absent from this literature. Clearly, worker reactions to legal norms can be either individual or collective, and outcomes will vary accordingly. The intent of this chapter is to capture what is essentially an individualistic response, in its examination of legal awareness, including according to key variables such as gender, age and education. On the axis of regulatory subject, the chapter suggests attentiveness to the diversity among substantive standards within the labour law canon, to suggest the variations in effectiveness among different legal norms. In line with this approach, the chapter is centred exclusively on what can be termed ‘working conditions’ laws. This chapter therefore impinges on the rapidly evolving field of research on job quality (see the central contribution of Green 2006); and, in particular, on the contribution of job quality to economic development (see, for example, Deakin and Wilkinson 2000). It is increasingly recognized that although the quantity of employment generated in the recent decades of economic growth has been considerable, the quality of this employment has been overlooked (see, for example, Chen et al. 2004; Lee and Eyraud 2008). In this regard, the particular focus of this chapter is on traditional components of quality of work, namely the legal standards that govern wages, hours of work and maternity leave.

Sangheon Lee and Deirdre McCann 297

Finally, in line with its call for detailed analyses of labour law’s operation in specific national contexts,4 the chapter focuses on the key East African economy of Tanzania. This setting is particularly apt for an analysis of the influence of working conditions laws in a developing country. Tanzania has witnessed unprecedented economic growth in recent years, of more than six per cent per annum (ILO 2009b). Employment growth has been rapid: Labour Force Survey figures indicate an increase of 484,201 from 1990–2001 and 816,176 from 2000–05 (Kahyarara and Rutasitara 2009). This job growth has embraced an increase in employment relationships of the kind that are governed by conventional labour law frameworks: a sectoral shift towards services has produced an increase in wage and salaried workers (ILO 2009b). Yet, more than 80 per cent of jobs are categorised as informal (ILO 2009b). Further, Tanzania’s rapid growth has spurred policy consideration of the potential for integrating job quality into employment and development strategies (see further Ackson and McCann forthcoming). Developments at the pan-African level have in part inspired this policy shift. In April 2004, the African Union Extraordinary Summit on Employment and Poverty Alleviation in Africa endorsed the ILO’s Decent Work Agenda, placing a particular emphasis on the creation of high-quality jobs.5 In April 2009, a National Tripartite Workshop was convened to address job quality in Tanzania. The resulting National Policy Statement on Job Quality in the United Republic of Tanzania confirmed job quality as a goal of Tanzanian employment policy (ILO 2009b).

Data and methodology: the Working Conditions Survey Global Module and Survey on Working Conditions in Tanzania Reliable information on working conditions in developing countries is limited. It is even more challenging to find data on workers’ legal literacy, whether on employment law entitlements generally or on the subset of working conditions laws. Given these constraints, empirical testing of the speculation in the literature on the influence of labour laws in developing countries has so far been impossible. The programme of research on which this chapter reports has developed one method of responding to these challenges. A Working Conditions Survey Global Module was jointly developed by a team of economists and lawyers from the Conditions of Work and Employment Programme of the International Labour Organisation (ILO) and the European Foundation for the Improvement of Living and

298 Regulating for Decent Work

Working Conditions (EFILWC). The Global Module was adapted from the survey instrument of the European Working Conditions Survey (EWCS), which has been conducted periodically in European countries since 1991 and investigates a broad range of dimensions of the quality of working life (on the EWCS, see EFILWC 2007). The Global Module is aimed at improving the capacity of ILO constituents to monitor changes in working life.6 It is designed to take into account labour market conditions in developing countries, and in particular the massive presence of informal work, and to be adjusted to take account of the socio-economic and institutional contexts of countries in which a survey is conducted. The data in this chapter are drawn from the Survey on Working Conditions in Tanzania (SWCT), which was the first national survey to be designed in line with the Global Module and was intended to pilot-test its operation. A household survey conducted in urban centres in seven regions across Tanzania7 the SWCT comprised interviews with a nationally representative sample of 1,249 individuals in 1,240 households (for full details of the survey, see Kahyarara and Rustasitara 2009). The SWCT has therefore generated the first nationally-representative sample of detailed data on conditions of work in a low-income setting. In line with the strategy outlined in the Global Module, changes were made to the survey methodology in the SWCT to tailor it to the economic life of a developing country and the specificities of the Tanzanian labour market (see further Kahyarara and Rutasitara 2009). These adjustments were made by a Tanzania-based research team led by the University of Dar es Salaam. In designing the SWCT, it was decided in consultation with representatives of the Tanzanian government and national-level trade unions and employers’ organizations, to target the survey on urban areas and on the tourism and informal sectors. As a different sampling framework was used for the tourism sector, this paper excludes it and focuses on non-tourism sectors, including on informal employment. The informal sector has been singled out by government policy in Tanzania as among the most promising sources of future employment, and therefore of poverty reduction. This sector features in policy efforts to alleviate unemployment and poverty, which have included in recent decades macroeconomic policy reforms, increased investment in the private sector, the promotion of skills training, increased financial services from both the government and private sector, and legislative and policy reforms intended to create an attractive environment for investment (Kahyarara and Rustasitara 2009). Formalization has also featured in employment and development policies: it has been identified as a priority in the National Employment Policy 2008, for example, and

Sangheon Lee and Deirdre McCann 299

elaborated in more detail in the Sustainable Industrial Development Policy (1996–2020) . These policies have not, until recently, strongly embraced compliance with labour laws. This dimension of formalization has more recently been advanced with respect to conditions of work laws in the National Policy Statement on Job Quality (see further Ackson and McCann forthcoming). The inclusion of the informal sector in the SWCT permits an investigation of working conditions in relationships conventionally defined as beyond labour law’s reach. In Tanzania, informal jobs, defined to encompass the self-employed and casual workers and those in small enterprises, are estimated to account for 70 per cent of the workforce in urban areas (Labour Force Survey figures). Significant efforts were made in the design of the Global Module to develop an effective method of measuring informal employment. This was achieved primarily by introducing a set of multilayered filtering questions which identify informal employment according to conventional definitions by establishment size (less than five people usually work in the establishment), location (on ‘a footpath, street or open space’) and the absence of a written contract or agreement. This technique permitted the SWCT to cover a range of activities, ranging from simple service activities to manufacturing and technical services, and to embrace a wide range of occupational classifications.

Awareness of working conditions laws in Tanzania Worker awareness and enforcement of legal measures has emerged as a subject for research in industrialized countries in part due to a recognition of growing vulnerability within increasingly individualized regulatory frameworks (Pollert 2005; Meager et al. 2002). As collective mechanisms decline, and legislated individual rights become more significant, the enforcement of these entitlements has been displaced from its rather peripheral position within the labour law debates and ushered to centre stage. This growing preoccupation with the significance and role of individual legislated entitlements has spawned a literature on the enforcement of legal standards by individual workers. This literature is particularly developed in the UK, a regime in which a straightforwardly deregulatory government was replaced in the late 1990s by one that remained suspicious of collective regulation of the labour market, but was devoted, at least initially, to enacting a set of individual employment rights (for example, Meager et al. 2002; Holt and Grainger 2005; White and Croucher 2007; on the ‘New Labour’ labour

300 Regulating for Decent Work

law project in the UK see further Davies and Freedland 2007). In this context, a literature developed that is centred on workers’ awareness of employment rights and the techniques they use to assert their legal entitlements. Primarily empirical in nature, this line of work has developed and refined methodologies to pursue these issues, and has been drawn on in designing the ‘legal awareness’ elements of the Global Module. Beyond the UK, this line of research is not as advanced, although Eamets and Masso (2005) have highlighted the weak enforcement of employment protection laws in Baltic countries, which they found partly attributable to poor awareness, and Cooke (2005) has reported the findings of a survey in China that found most workers and employers to be unaware of legal standards. Further, the significance of awareness raising to improving working conditions in developing countries attracted the attention of Fenwick et al. (2007) in a study of micro and small enterprises that represents one of the few attempts to consider the extent of legal awareness and its potential impact on actual working conditions in low-income economies. Despite the rather limited research, a preoccupation with individual legal rights can be considered highly relevant to developing countries, if the focus is on certain of their parallels with deregulated regimes in the industrialized world. As labour markets in which collective norms are not a forceful form of regulation, and legislation has the most prominent role among formal regulatory techniques, these industrialized and developing regimes have shared preoccupations, in particular with the regulation of informal and precarious work (on industrialized countries, see Ram et al. 2004; Fudge and Owens 2006; Vosko 2007). These parallels emerge in Tanzania, where the collective representation of workers is minimal and legal rights primarily enforced by way of individual complaints-based mechanisms (Ackson 2009). One of the unique aspects of the survey is that, in addition to questions about actual working conditions (for example, employment status, wages/incomes, working time and work organization, physical work environment, work and family and so on), it includes a set of questions centred on whether the workers are aware of their legal entitlements in relation to working conditions. Specifically, respondents are asked if they are aware of key standards on minimum wages, working time and maternity leave (see Table 11.1).8 The survey questions on awareness are novel; in that they are not present in the EWCS survey instrument. In part, this omission is attributable to the complexities of identifying the relevant legal standards across all EU Member States, in particular due to the diversity of the regulatory

Sangheon Lee and Deirdre McCann 301 Table 11.1 Survey questions about workers’ awareness Q9-1 1 2 Q9-2

1 2 Q9-3 1 2 88

Are you aware that most workers are entitled to be paid a specific minimum wage (which is currently set at ____ *)? Yes No Are you aware that most workers should not be required to work more than ___ hours* per week (beyond which overtime premium should be paid)? Yes No Are you aware that most workers are allowed ___ weeks* of maternity leave off work to have a baby? Yes No Not applicable

Note: *The final survey questionnaire used in the fieldwork specifies relevant statutory standards. Source: The Global Module for Working Conditions Survey (draft).

systems (such as the degree of reliance on collective measures, level of articulation of legal norms in federal systems etc.). These problems were overcome in the SWCT, because the survey was conducted in a single jurisdiction9 and also because legal standards tend to be articulated in legislative form at the national level. While the survey questions offer a reasonable method of capturing workers’ awareness (especially given the paucity of such data), they are subject to some limitations. Most significantly, the questions have the potential to overestimate the extent of awareness, since they specify the relevant statutory standards. Such reminders may tempt the respondents to claim falsely that they are aware of these standards (Meager et al. 2002). The resulting estimates could therefore be expected to be much higher than those obtained from other types of questions such as those that ask respondents to specify the existing standards.10 This potential upward bias in estimation should be taken into account in interpretation and analysis. Statutory norms and workers’ awareness: level and variations Table 11.2 shows the extent of workers’ awareness of working conditions standards.11 It indicates a significant level of awareness, just less than 70 per cent in total. This finding is comparable to that of research in the industrialized world (Meager et al. 2002). It therefore indicates higher levels of legal awareness than are often assumed for developing countries.

302 Regulating for Decent Work Table 11.2 Workers’ awareness (% of employees: excluding tourism sector: N=666) Minimum wages

Working time

Maternity protection

All Gender

73.9

67.5

61.9

Male Female Age

73.6 73.9

66.3 69.8

55.8 77.5

16–20 21–25 26–30 31–35 36–40 41–50 51 and above Education

33.3 63.4 76.3 78.6 71.2 87.3 90.5

30.6 59.7 69.1 74.5 64.4 78.2 81.0

41.7 56.2 68.3 65.0 59.2 54.7 55.0

None Primary school Secondary school Technical school College, university and above Region

54.8 69.1 75.8 92.9 87.8

48.4 59.8 70.4 92.9 83.8

57.1 56.1 61.2 85.7 81.1

Dar es Salaam (capital) Mwanza Mbeya Covered by collective agreements

71.7 78.1 83.5

61.8 77.2 83.5

52.6 82.2 82.4

Yes No Having a written contract or agreement

80.7 72.3

71.6 66.7

69.2 59.3

Yes No

64.0 79.6

59.9 71.4

55.8 65.4

Source: Working conditions survey in Tanzania (2009).

Yet a significant proportion of workers lack awareness of the key working conditions entitlements. Overall, around 30 per cent were not aware of statutory entitlements on working time, minimum wages and maternity leave. Further, there is considerable variation in awareness of the different working conditions standards. Awareness of minimum wages is high

Sangheon Lee and Deirdre McCann 303

(73.9 per cent). The incidence is much lower for awareness of the maternity leave entitlement (61.9 per cent), which is primarily attributable to lower levels of awareness among male workers (55.8 per cent). Awareness of the weekly working time limit was higher than that of the maternity leave entitlement, but lower than that of minimum wages (67.5 per cent). Unsurprisingly, the incidence of awareness is positively correlated with age and educational attainment, with the exception of the relationship between age and maternity protection, which does not show any clear pattern. The influence of age is particularly strong in the case of working time and minimum wages, perhaps suggesting that work experience provides a critical opportunity for workers to learn about their legal rights. For example, only one-third of young workers aged 16–20 were aware of the statutory hours limits and of minimum wages, but the incidence more than doubled for the age group of 26–30. Educational attainment displays a similar pattern. It is plausible to assume that collective agreements and employment contracts are useful tools for helping workers to learn about their legal rights. There are grounds for such an assumption in the Tanzanian case. One of the primary legislative measures, the Employment and Labour Relations Act, 2004 (ERLA), provides legal support for union recognition and collective bargaining.12 It also requires that a written list of specified particulars be provided to employees,13 which can be included as part of their contracts.14 These particulars should include an employee’s hours of work15 and remuneration.16 The ELRA further requires that if an employee does not understand the written particulars, the employer must ensure that the information is explained in a manner that the employee understands.17 Despite the statutory requirements, however, these assumptions are not strongly supported by the data. As Table 11.2 indicates, the incidence of awareness is higher among workers who reported that they were covered by collective agreements (for example, in the case of minimum wages, 80.7 per cent versus 72.3 per cent for workers who are not covered by collective agreements), but the difference is relatively small (about 5–10 percentage points). In the case of employment contracts, the result is more surprising. Table 11.2 shows that the conclusion of a written contract or agreement reduces the likelihood that workers are aware of the major working conditions rights. These gaps are large and significant. This paradoxical result, however, needs to be interpreted with great care. For instance, it is noteworthy that in Tanzania, despite the provisions of the ELRA, written

304 Regulating for Decent Work Table 11.3 Employment contracts without teeth?: % of written contract or agreements which provide details on wages, working time and maternity leave (N=553) Wages 49.5

Working time

Maternity leave

52.2

21.9

Source: Working conditions survey in Tanzania (2009).

70.0 Aware

Not aware

62.9

62.7

% of respondents

60.0 49.7

50.0 40.0

37.0

30.0 20.0

17.8

17.1

10.0 0.0 Working no longer than the statutory maximum of 48 hours per week

Earning more than the minimum wages

Opportunities for maternity leave (women only)

Figure 11.1 Can awareness make a difference? Working time, minimum wages and maternity leave Source: Working conditions survey in Tanzania (2009).

contracts tend to lack what can be considered essential elements. As Table 11.3 demonstrates, about half of these contracts do not include any details about wages or working hours, and only one-fifth contained specific references to maternity leave. Does awareness matter? Awareness and actual conditions The key question for this chapter is whether workers’ legal awareness plays a significant role in improving actual conditions of work. In particular, we are interested in the extent to which workers’ awareness of their legal entitlements can contribute to increasing compliance with, or observance of, statutory regulation. The Tanzanian survey suggests that it matters considerably. To illustrate this point, Figure 11.1 examines how actual conditions vary depending upon workers’ awareness of relevant regulations in the area of working time, minimum wages and maternity protection. In the

Sangheon Lee and Deirdre McCann 305

case of working time, 37.0 per cent of workers who are not aware of the statutory maximum hours (48 hours per week) work more than this legal limit. However, the proportion is much lower – 17.1 per cent – for those who are aware of the statutory maximum. Similarly, although pay below the minimum wage is widespread in Tanzania, its incidence is much smaller among workers who are aware of minimum wages (49.7 vs 62.9 per cent). The difference is most striking in the case of maternity protection. About 63 per cent of women workers who are aware of their legal entitlement to maternity leave believe that they will benefit from it when needed, but the proportion falls to 17.8 per cent for workers who are unaware of this statutory benefit. To test the impact of awareness on actual working conditions in a more rigorous manner, a series of logistic regressions controlling for standard demographic variables have been undertaken. The results are reported in Table 11.4 in which all of three coefficients show the expected signals. The first column reports the relationship between two binary variables: working more than the statutory standard (48 hours) and awareness of this standard. The reported coefficients indicate that workers who are aware of the statutory standard are 1.6 times less likely to work more than this standard (holding constant other standard demographic variables). In the case of maternity protection, the impact of awareness is much greater. Women workers who are aware of maternity protection law are 2.6 times more likely to believe that they will benefit from maternity leave. The likelihood of earning no less than the minimum wage is positively related to the awareness of the applicable minimum wage rates. It is estimated that workers’ awareness would increase this likelihood by 35 per cent, but the coefficient is not statistically significant. This is due Table 11.4 Can awareness make a difference? The results of logistic regressions Working time (all) Coefficient Odds ratio

–1.59 (.28)* .20 (.06)*

Minimum wage (all)

Maternity protection (women only)

.35 (.23) 1.42 (.33)

2.6 (.43)* 13.42 (5.83)*

Notes: 1. *Significant at 5%. 2. Standard errors are in parentheses. 3. Logistic regressions which include awareness as an independent variable, along with control variables such as gender, education, age, industry, occupation, and size of the establishment. 4. Dependent variables: working more than 48 hours (working time); being paid more then minimum wages (minimum wages); and benefiting from maternity leave (maternity protection). Source: Working conditions survey in Tanzania (2009).

306 Regulating for Decent Work

in part to the fact that this statistical exercise is less relevant to minimum wages. For instance, high-wage-earners are not particularly interested in the statutory minimum wage, because their level of wages is high above it (Meager et al. 2002). How can awareness improve actual working conditions? Job satisfaction and workers’ behavioural patterns The analysis so far has shown that increased awareness can contribute to an improved observance of working conditions laws. A related question, then, is how this outcome is generated, or, more specifically, the channelling mechanism through which awareness translates into actual improvements in the working environment. While this is a complex question that needs to take into consideration a variety of factors, in this chapter we will examine two possible, and related, factors: job satisfaction and workers’ behavioural patterns in relation to noncompliance with labour laws. First, the awareness of legal rights and the resulting recognition that they are not being complied with can make workers view their working environment in an unfavourable light, and therefore may reduce the levels of overall job satisfaction. Such an impact on job satisfaction can prompt efforts to remedy the situation. Indeed, it is known that job satisfaction can determine workers’ performance and future behavioural patterns, such as turnover (see Green 2006 for a review; Lee 2005). It is feasible to examine this possibility with regard to working time in Tanzania, as the SWCT includes a question on the respondents’ satisfaction with current working hours.18 Table 11.5 reports the results of a simple regression on job dissatisfaction which controls for actual working hours. It is clear from this table that workers’ dissatisfaction with working hours increases with their levels of awareness of statutory maximum working hours, once their actual working hours are controlled for. Secondly, it cannot be assumed that workers are passive once they recognize the gaps between legal entitlements and their actual conditions of work. In fact, awareness may tend to lead to corrective actions, even if they may not always be successful. This point is demonstrated clearly from the responses to the SWCT question on techniques to address divergence from the legal standards: ‘Suppose you had a problem at work because your working conditions were not in line with your legal rights. What would be the most effective way of solving the problem?’ (See Table 11.6.) The majority of respondents (57.8 per cent) preferred the ‘direct method’ of complaining to their employers. Passive

Sangheon Lee and Deirdre McCann 307 Table 11.5 Awareness and job dissatisfaction: working hours – regression coefficients Constant

Dissatisfaction with working hours (1=very satisfied, …, 4= not satisfied at all) All Men Women

1.818 (.102)* 1.877 (.126)* 1.779 (.180)*

Awareness (statutory maximum hours)

.196 (.060)* .103 (.036)* .050 (.118)

Controlling for Actual hours (five categories)

.057 (.020)* .034 (.024) .236 (.070)*

Notes: 1. Figures in brackets refer to standard errors. 2. * Significant at 1% level. 3. Other controlled variables (see Table 11.4) are not reported, except actual hours. Source: Working conditions survey in Tanzania (2009).

Table 11.6 How would workers react? (% of respondents) Suppose you had a problem at work because your working conditions were not in line with your legal rights. What would be the most effective way of solving the problem? Do not complain 10.4

Complain to your employer

Complain to trade union

57.8

3.3

Complain to colleagues

Others*

14.8

13.8

Source: Working conditions survey in Tanzania (2009). * Including ‘complain to a government agency’, ‘complain to an NGO’, ‘contact a lawyer’, ‘bring a claim in a court’, and ‘others’.

measures such as ‘do not complain’ or ‘complain to colleagues’ were minority options. It is also striking that resorting to trade unions, lawyers or government authorities were not popular options in Tanzania.

Implications and conclusions This chapter has revealed a relatively high level of awareness of statutory entitlements on working conditions in a low-income setting. In doing so, it has challenged the dominant narratives of labour law’s influence that were outlined at the outset of the chapter. Our findings counter the formalist account’s calls for ongoing law reform, by suggesting the

308 Regulating for Decent Work

need for a proper empirical understanding of the operation and impact of existing laws. The findings also belie the pessimist narrative by suggesting that ‘informal economies’ are not entirely or inescapably situated beyond the influence of labour regulations. Central to both these refutations, and a key finding of this chapter, is that fully to understand labour law’s operation it is necessary to examine its significance to workers. This research strategy can therefore accompany efforts to map the role of the judiciary, labour inspectorates and employers in shaping the influence of statutory norms. Further, this research on Tanzania has suggested a strong link between workers’ legal awareness and their actual working conditions. The logistic regression results for working hours and maternity protection bolster our hypothesis that legal awareness will make some difference to actual working conditions. This effect appears to be channelled, at least in part, by individual workers’ resort to complaining to their employers when they recognize that their working conditions diverge from minimum statutory standards. Hence, it can be suggested that learning about legal rights is akin to providing a different perspective for workers on acceptable conditions of work, which they can then use to re-consider their current working environment or any dissatisfaction that they may previously have experienced. Although further research is needed, this process implies that legal awareness has the potential to become an effective instrument of improving working conditions, increasing job satisfaction and encouraging proactive behavioural patterns that ensure legal observance. These effects suggest a policy need for initiatives that can improve legal awareness. The research reported in this chapter suggests avenues that are currently neglected in Tanzania, and perhaps elsewhere, namely the role of trade unions and of written contracts of employment in conveying information on labour law entitlements. Indeed, in the case of Tanzania, policy actors could act on requirements on the provision of written information on working conditions that are already embedded in statute. This chapter has therefore provided important insights into the ongoing debates on the labour market impact of labour regulations. The SWCT confirmed the presence of ‘observance gaps’. These gaps are massive, with the result that the relevance of common statistical exercises that regress the indicators of labour regulations on labour market variables becomes questionable. This finding also implies that the available indicators on labour regulations (and in particular the World Bank’s Employing Workers indicators) do not properly capture actual ‘constraints’ that are encountered by employers. In other words,

Sangheon Lee and Deirdre McCann 309

the constraints on business summarized in these indicators tend to suffer from overestimation. The finding that observance gaps may be narrowed, to some degree, by improving workers’ knowledge of their legal entitlements, supports the view that debates on labour market regulation should not be configured as a choice between ‘regulation’ or ‘de-regulation’, but rather as a question of ‘how to regulate’ (Fenwick et al. 2007; Lee et al. 2008). Finally, it suggests that policy actors may be able to improve legal outcomes without profound institutional change or substantial outlay, allowing developing country governments to channel their limited resources into cost-effective strategies. In this way, policy-actors in low-income countries may be able to avoid the wasteful process of continuous labour law reforms that are carried out in the absence of any opportunity to evaluate their impacts.

Notes 1. See para. 3 of the Resolution concerning decent work and the informal economy, adopted by the International Labour Conference, Geneva, in 2002. 2. For an exploration of the implications of regulation theory for labour law scholarship, see Arup et al. (2006). 3. Another strand of literature on the enforcement of labour standards has examined the role of trade in weakening or strengthening legal measures as developing countries shift to more open economies. See Brown (2001) and Berik and Rodgers (2008). 4. This call reflects Saavedra and Tommasi’s (2007) observation that policies on informality should be country-specific. 5. The Summit’s Declaration on Employment and Poverty Alleviation in Africa explicitly endorses the ILO’s Decent Work Agenda, Paragraph Four, available at http://www.africa-union.org/EMPLOYMENT/EMPLOYMENT.htm [last accessed 19th Feburary 2011]. 6. See further the Global Module at http://www.ilo.org/travail/areasofwork/ lang—en/WCMS_145738/index.htm (under construction). 7. Dar es Salaam, Mwanza, Mbeya, Aruha, Kilimanjaro, Tanga and Morogoro. 8. This module is available from the authors on request. 9. The SWCT was confined to mainland Tanzania. 10. Some surveys on awareness have posed more direct questions such as: ‘Can you tell me what you think is the current hourly rate of the minimum wage for adult workers?’ (e.g., White and Croucher 2007). 11. For a review of the regulatory framework in Tanzania, see Ackson (2009). 12. Part VI. 13. Section 15(1). 14. Section 15(2). 15. Section 15(1)(g). 16. Section 15(1)(h). Details should also be provided on the method of calculation of remuneration and on any benefits or payments in kind.

310 Regulating for Decent Work 17. Section 15(3). Employers are also required to display a statement of their employees’ ELRA rights in a conspicuous place (Section 16). 18. Such an analysis is not feasible for minimum wages and maternity protection.

References Ackson, T. 2009. ‘Regulating working conditions in the globalizing world: recent trends in Tanzania’. Paper presented at the Regulating for Decent Work Conference, Geneva, 8–10 July. Ackson, T.; McCann, D. forthcoming. ‘Working life and development in Tanzania: approaches to legal regulation in a low-income setting’, International Labour Review. Aherling, B.; Deakin, S. 2007. ‘Labour regulation, corporate governance and legal origin: a case of institutional complementarity?’, Law & Society Review, vol. 41, pp. 865–908. Arthurs, H. 2006. ‘What immortal hand or eye? – who will redraw the boundaries of labour law?’ in Davidov, G.; Langille, B., Boundaries and Frontiers of Labour Law (Oxford: Hart), pp. 373–89. Arup, F. 2001. ‘Labour law as regulation: promises and pitfalls’, Australian Journal of Labour Law, vol. 14, pp. 1–13. Arup, C.; Gahan, P.; Howe, J.; Johnstone, R.; Mitchell, R.; O’Donnell, A. 2006. Labour Law and Labour Market Regulation (Sydney: Federation Press). Berg, J.; Cazes, S. 2007. ‘The doing business indicators: measurement issues and political implications’, ILO Economic and Labour Market Paper 2007/6 (Geneva: ILO). Berg, J.; Kucera, D. (eds). 2008. In Defence of Labour Market Institutions: Cultivating Justice in the Developing World (Geneva and London: ILO and Palgrave). Bertola, G. 2009. ‘Labour market regulation: motives, measures, effects’, ILO Conditions of Work and Employment Series no. 21 (Geneva). Bertola, G.; Cazes, S. 2000. ‘Employment protection in industrialized countries: the case for new indictors’, International Labour Review, vol. 129, no. 1, pp. 57–72. Berik, G.; Rodgers, Y. 2008. ‘Options for enforcing labour standards: lessons from Bangladesh and Cambodia’, Journal of International Development, vol. 22, no. 1, pp. 56–85. Bhattacharjea, A. 2006. ‘Labour market regulation and industrial performance in India: a critical review of the empirical evidence’, Indian Journal of Labour Economics, vol. 49, no. 2, pp. 211–32. Botero, J.; Djankov, S.; La Porta, R.; Lopez-de-Silanes, F.; Shliefer, A. 2004. ‘The regulation of labour’, Quarterly Journal of Economics, vol. 119, no. 4, pp. 1339–82. Brown, D. 2001. ‘Labor standards: where do they belong on the international trade agenda’, Journal of Economic Perspectives, vol. 15, no. 3, pp. 89–112. Browne, J.; Deakin, S.; and Wilkinson, F. 2005. ‘Capabilities, social rights, and European integration’, in Salais, R.; Villeneuve, R. (eds), Towards a European Politics of Capabilities (Cambridge: Cambridge University Press). Chen, M.A. 2006. ‘Rethinking the informal economy: linkages with the formal economy and the formal regulatory environment’, in Guha-Khasnobis, B.; Kanbur, R.; Ostrom, E. (eds) Linking the Formal and Informal Economy: Concepts and Policies (Oxford: Oxford University Press).

Sangheon Lee and Deirdre McCann 311 Chen, M.; Vanek, J.; Carr, M. 2004. Mainstreaming Informal Employment and Gender in Poverty Reduction (London: Commonwealth Secretariat). Chor, D.; Freeman, R. 2005. ‘The 2004 global labor survey: workplace institutions and practices around the world’, NBER working paper 11598 (Cambridge, MA: NBER). Conaghan, J.; Fischl, R.M.; Klare, K. 2004. Labour Law in an Era of Globalization: Transformative Possibilities and Practices (Oxford: Oxford University Press). Cooke, F. 2005. HRM, Work and Employment in China (London: Routledge). Davidov, G.; Langille, B. (eds) 2007. Boundaries and Frontiers of Labour Law (Oxford: Hart). Davies, P.; Collins, H.; Rideout, R. 2000. Legal Regulation of the Employment Relation (Amsterdam: Kluwer Law International). Davies, P.; Freedland, M. 2007. Towards a Flexible Labour Market? (Oxford: Oxford University Press). Deakin, S.; Lele, P.; Siems, M. 2007. ‘The evolution of labour law: calibrating and comparing regulatory regimes’, International Labour Review, Vol. 146, pp. 133–162. Deakin, S.; Wilkinson, F. 2000. ‘Labour law and economic theory: a reappraisal’, in Davies, P.; Collins, H.; Rideout, R. (eds), Legal Regulation of the Employment Relation (Amsterdam: Kluwer Law International). Edwards, P.; Monder, R.; Black, J. 2004. ‘Why does employment legislation not damage small firms?’, Journal of Law and Society, vol. 31, no. 2, pp. 245–65. Eamets, R.; Masso, J. 2005. ‘The paradox of the Baltic states: labour market flexibility but protected workers?’, European Journal of Industrial Relations, vol. 11, no. 1, pp. 71–90. European Foundation for the Improvement of Living and Working Conditions (EFILWC) 2007. Fourth European Working Conditions Survey (Dublin: EFILWC). Fenwick, C.; Howe, J.; Marshall, S.; Landau, I. 2007. ‘Labour and labour-related laws in micro and small enterprises: innovative regulatory approaches’, ILO SEED Working Paper no. 81 (Geneva: ILO). Freeman, R. 2009. ‘Labor regulations, unions and social protection in developing countries: market distortions or efficient institutions?’, NBER Working Paper 14789 (Cambridge, MA: NBER). Fudge, J.; Owens, R. 2006. Precarious Work, Women and the New Economy: The Challenge to Legal Norms (Oxford, Hart). Green, F. 2006. Demanding Work: The Paradox of Job Quality in the Affluent Economy (Princeton, NJ: Princeton University Press). Holt, H.; Grainger, H. 2005. Results of the Second Flexible Working Employee Survey, DTI Employment Relations Research Series no. 39 (London: Department of Trade and Industry). International Labour Office (ILO). 2009a. Key Indicators of the Labour Market (Geneva). ILO. 2009b. National Profile of Working Conditions in the United Republic of Tanzania (Geneva). Kahyarara, G.; Rutasitara, L. 2009. ‘Final report on working condition survey in Tanzania’, report prepared for ILO (unpublished). Langille, B. 2007. ‘What is international labour law for?’, Law & Ethics of Human Rights, vol. 3, no. 1, pp. 47–82. Lee, S. 2005. ‘Working conditions and satisfaction at work: determinants and interactions in new EU member states and candidate countries’, in

312 Regulating for Decent Work Vaughan-Whitehead, D. (ed.), Working and Employment Conditions in New EU Member States: Convergence or Diversity? (Geneva: European Commission and ILO). Lee, S.; Eyraud, F. (eds) 2008. Globalization, Flexibilization and Working Conditions in Asia and the Pacific (Geneva and Oxford: ILO and Chandos). Lee, S.; McCann, D. 2006. ‘Working time capability: towards realizing individual choice’, in Boulin, J.Y.; Lallement, M.; Messenger, J.C.; Michon, F. (eds), Decent Working Time: New Trends, New Issues (Geneva: ILO). Lee, S.; McCann, D. 2008. ‘Measuring labour market institutions: conceptual and methodological questions on “working-hour rigidity”’, in Berg, J.; Kucera, D. (eds), In defence of labour market institutions: cultivating justice in the developing world (Geneva and London: ILO and Palgrave). Lee, S.; McCann, D.; Torm, N. 2008. ‘The World Bank’s “employing workers index”: findings and critiques – a review of recent evidence’, International Labour Review, vol. 147, no. 4, pp. 416–432. Meager, N.; Tyers, C.; Perryman, S.; Rick, J.; Willison, R. 2002. Awareness, Knowledge and Exercise of Individual Employment Rights (London: Department of Trade and Industry). Mitchell, R.; Gahan, P.G.; Stewart, A.; Cooney, S.; Marshall, S.D. 2010 ‘The evolution of labour law in Australia: measuring the change’ Australian Journal of Labour Law, vol. 23, no. 1, pp. 61–93. Panagariya, A. 2001. ‘Labor standards and trade sanctions: right end, wrong means’, report prepared for the conference ‘Towards an agenda for research on international economic integration and labor markets’. Piore, M.; Shrank, A. 2008. ‘Towards managed flexibility: the revival of labour inspection in the Latin world’, International Labour Review, vol. 147, no. 1, pp. 1–23. Pollert, A. 2005. ‘The unorganised worker: the decline in collectivism and new hurdles to individual employment rights’, Industrial Law Journal, vol. 34, no. 3, pp. 217–238. Ram, M.; Edwards, P.; Jones, T. 2004. Informal Employment, Small Firms and the National Minimum Wage (London: Low Pay Commission). Saavedra, J.; Tommasi, M. 2007. ‘Informality, the state and the social contract in Latin America: a preliminary exploration’, International Labour Review, vol. 146, nos. 3–4, pp. 279–309. Squire, L.; Suthiwart-Narueput, S. 1997. ‘The impact of labour market regulations’, World Bank Economic Review, vol. 11, no. 1, pp. 119–43. Vosko, L. 2007. Precarious Employment: Understanding Labour Market Insecurity in Canada (Montreal and Kingston: McGill-Queen’s University Press). White, G.; Croucher, R. 2007. Awareness of the Minimum Wage in the Hairdressing Industry: An Evaluation of the DTI/HMRC Targeted Campaign. Research report for the Low Pay Commission. World Bank. 2005. Doing Business 2006 (Washington, DC: World Bank).

12 Governing Regulatory Discretion: Innovation and Accountability in Two Models of Labour Inspection Work* Roberto Rocha C. Pires

Introduction Contemporary debates on the role of labour market regulation have been strongly influenced by the neoclassical economic tradition that considers this type of state intervention to be inimical to economic growth (Lee and McCann in this volume). Mainstream development economics has long advocated for an unavoidable trade-off between the extension and enforcement of labour regulations and the ability of firms to compete in increasingly globalized markets. A now vast literature (for example, Johnson et al. 1998; Schneider and Enste 2000; Friedman et al. 2000; Batra et al. 2003) argues that extensive workers’ rights and job protection regulations are associated with firms’ inefficiency, growing unemployment rates, and the expansion of the informal sector.1 In this view, not only do labour regulations ‘distort’ the market, but their enforcement is often depicted as legalistic, bureaucratic and ineffective, and the discretion enjoyed by law enforcers is often portrayed as a major source of corruption and economic inefficiencies, particularly in developing countries.2 In contrast to this perception, a growing number of development scholars have been emphasizing the possibility of a ‘high-road’ path in the international economy, demonstrating how firms’ growth, profitability and production, on the one hand, and social rights, higher labour standards and wages, on the other, might become mutually reinforcing processes (Rodrik 1997; Stiglitz 2000; Bazan and Navaz-Aleman 2004; * I would like to thank the Department of Labour Inspection (Secretaria de Inspeção do Trabalho ) at the Brazilian Ministry of Labour for the cooperation and feedback at various stages of this research. 313

314 Regulating for Decent Work

Posthuma 2004; Abrami 2005; Milberg and Houston 2005; Bernard and Boucher 2007).3 Sharing this alternative perspective on the relationship between labour standards and development, recent studies on the enforcement of labour regulation (for example, Piore and Schrank 2008; Pires 2008; and Silbey et al. 2009) have documented instances in which inspectors have nonetheless operated effectively and used a systemic perspective to produce technological, legal and managerial solutions to the obstacles preventing firms from complying with the law. Labour inspection in countries as diverse as France, the Dominican Republic and Chile has been described as a key element in the reconciliation of workers’ protections with firms’ competitiveness and productivity. In Brazil, for example, improvements in labour inspection have also been associated with the growth of formality in the labour market (Berg in this volume). These studies have been drawing attention to the fact that variations in the way labour regulation is enforced are important to development outcomes. While in some instances labour inspectors limit themselves to the narrow boundaries of their formal mandate and to the strict implementation of law as written, in other cases they refuse to be confined by narrow legalistic interpretations of legislative intent and instead adopt innovative strategies, working collaboratively with other organizations, including the subjects of regulation, to solve complex business and regulatory problems. What accounts for this variation? What are the factors leading to the emergence of innovative practices in the enforcement of labour regulations? This chapter contributes to this ongoing debate by calling attention to the role of labour inspection in mediating between social protection and economic development. It examines the complexities of the operation of labour inspectorates with the goal of understanding the conditions under which labour inspectors learn, innovate and promote compliance, departing from the dominant characterization of labour inspection as inescapably legalistic and bureaucratic. The analysis centres on one specific set of variables of repeated policy relevance: organizational models for managing the discretion and performance of labour inspectors. I argue that, under certain conditions, structures and strategies for managing discretion and performance create opportunities and incentives for innovative enforcement approaches and for the solution of problems preventing firms from complying with labour regulations. The chapter proceeds by first reviewing the ‘fear of discretion’ that has characterized debates in legal studies and political science since the mid-twentieth century. I describe briefly how two contemporary models in public administration – namely, the new public management and

Roberto Rocha C. Pires 315

the experimentalist governance approaches – differ from traditional responses promoting administrative law and oversight to respond to the problem of discretion. I compare these two models and identify variations in the strategies they offer to reconcile organizational performance and capacity with accountability and controls on the misuses of discretion. After contrasting the variations in these theoretical models, I present a series of comparisons between these two models within the same organizational setting. Using data collected through ten months of fieldwork in the Brazilian Department of Labour Inspection, I show how each model shapes labour inspection work and its outcomes. Finally, I outline some of the most salient aspects of these comparisons, proposing in conclusion more general hypotheses about the relationship between forms of accountability and staff motivation, and between collaborations, problem solving, and responsiveness.

The fear of discretion and contemporary responses in public administration The discretion enjoyed by bureaucrats in daily decision-making processes has long been treated as residual category by analysts of the modern state (Davis 1969; Hawkins 1992), as the Weberian view of bureaucracy as a system of impersonal and dispassionate rule-oriented behaviour prevailed as the hegemonic framework of analysis, with just a few dissenting voices (Blau 1970; Crozier 1964; Gouldner 1954).4 Research in different fields – such as socio-legal studies (Wilson 1968; Bittner 1967, 1990; Van Maanen 1973, 1978; Brown 1981; Silbey and Bittner 1982) and policy studies (Leonard 1977; Lipsky 1980; Wilson 1989; Maynard-Moody and Musheno 2003) – has empirically demonstrated that bureaucratic discretion is pervasive and possibly indispensable in legal and administrative systems. Despite what appears to be the inevitability of discretion, debates have more often been characterized by the fear of bureaucratic tyranny and the risks of unchecked decision making than by the potential benefits of responsible exercise of discretion. Scholars within the tradition of liberal legalism interpreted discretion as a threat to the rule of law, a breach of the ‘social contract’, creating space for inconsistency and arbitrariness, and, consequently, the potential for injustice. According to this liberal tradition, discretion needs to be confined, structured and checked by administrative law – procedures and rules regulating the conduct and practices of administrative agents (Davis 1969; Handler 1986; Bryner 1987; Hawkins 1992). For political scientists more concerned about democratic structure than legal process,

316 Regulating for Decent Work

discretion also posed a challenge to the idea of political accountability (between bureaucrats and elected officials), and called into question the liberal structure of constitutional separation of powers (checks and balances). Within this tradition of constitutional democracy, much attention has been devoted to limiting bureaucratic discretion by instituting procedural mechanisms and oversight on agency performance by Congress, the President, and civil society (Calvert et al. 1989; McCubbins et al. 1987; McCubbins and Schwartz 1984). Retrospective evaluations have demonstrated that legal procedures and oversight do reduce levels of discretion. However, they also showed that the remedy has been as bad as (if not worse than) the disease. For example, Bryner (1987) and Handler (1986) documented how excessive and misdirected actions to reduce discretion have damaged the capability of public sector organizations to accomplish delegated tasks, by making administrative processes more confusing and, in general, reducing the ability of agencies to function effectively. As a result, additional attention to bureaucratic procedures has undermined parallel attention to helping organizations to make the complex decisions necessary for the implementation of policies and regulation. In the past few decades, as the efforts to limit discretion at the expense of organizational capacity have proved increasingly inefficient as well as ineffective, two bodies of literature in public administration have offered models that purport to balance the control of bureaucrats’ discretionary decision making with a concern for bureaucratic capacity and competence. In contrast to the earlier efforts, these new models for public administration emphasized organizational structure and managerial practices rather than rules and legal procedures, administrative law and oversight, as more pragmatic and effective means for managing discretion. First, the new public management (NPM) paradigm became one of the mantras of public sector reform throughout the world in the 1980s and 1990s. Against the breakdown of bureaucratic capacity in previous decades and the widespread discontent with government performance, NPM raised the prospects of improving bureaucracies’ efficiency and responsiveness to political principals and citizens, with its orientation towards outcomes and the optimization of the public budget. The literature on the topic identifies three main characteristics of public sector reforms categorized under the rubric of NPM: (i) decentralization, with the disaggregation of subnational government actors, the splitting up of large hierarchical structures, and the separation of core vs other functions of government; (ii) privatization and competition, with deregulation, the creation of quasi-markets for most public services, and the establishment

Roberto Rocha C. Pires 317

of public–private partnerships (PPP); and (iii) performance management, with the institution of targets and output indicators to measure the performance of organizations and their bureaucrats, and a strong emphasis on pecuniary-based, specific performance incentives such as pay-forperformance schemes (Osborne and Gaebler 1992; Dunleavy and Hood 1994; Pollit 1995; Bresser Pereira and Spink 1999; Barzelay 2001). The NPM solution to the problem of discretion claimed to avoid the mistakes of the past by emphasizing the measurement of outputs rather than control of means via legal and administrative procedures. Under this model, public sector organizations should define a short list of performance targets that can be narrowly defined, quantified and measured. Each and every bureaucrat is assigned a piece of the overall target. Supervisors constantly monitor bureaucrats in terms of their performance in meeting these targets, in reference to quantitative output indicators. In order to provide the right incentives, managers administer bonuses (pay-forperformance schemes) on the salaries of only those workers who meet the target periodically. Thus, the NPM solution restricts bureaucrats’ discretion primarily by providing strong incentives (significant increases in salaries) only for the desired actions/outcomes without severely limiting the capacity of bureaucrats to pursue policy goals (that is, less paperwork, a greater latitude in ways to deal with problems, and so on). The second model, the experimentalist governance (EG) approach, emerges as a criticism by EG scholars (such as C. Sabel, J. Zeitlin, M. Dorf, and W. Simon) of the untenable nature of the economic, rational framework at the heart of NPM proposals, most specifically the principal–agent relationship. Drawing from institutional economics, NPM models in general separate conception from execution and assume the existence of principals (such as civil society actors, political parties, or elected officials) who already know what needs to be done to solve collective problems. Supposedly, these principals are ready to translate public goals into detailed performance targets – such as, for example, a 50 per cent increase in the formalization of labour, or a 20 per cent decrease in the school drop-out rates. In contrast, EG scholars argue that there are no such principals in the polity with the robust and panoramic knowledge, nor unchallenged consensus assumed for this directive role. Therefore, the main problem for reform is not to determine performance targets and the right incentive system, but to determine ways actors can interact, discover, and learn together what needs to be done, and how to do it (Sabel 2004, 2005). Thus, in this second managerial model of regulation, the solution to achieving regulatory efficacy requires experimentalist organizations

318 Regulating for Decent Work

‘that assume the provisionality of their goals and institutionalize social learning by routinely questioning the suitability of their current ends and means, and by periodically revising their structures in light of the answers’ (Sabel 2004, p. 4). Experimentalist organizations display the attitude of constantly detecting and correcting errors at the lowest levels, and then adjusting the higher-level structures to generalize successes and encourage more refined error detection. Through constant reflexive adjustment, EG scholars argue, public sector organizations can simultaneously: (i) expand their capacities to solve complex problems by adapting to rapidly changing conditions and by tailoring their responses to diverse clienteles; and (ii) heighten accountability of the front-line bureaucrats to their supervisors and the larger public consistent with the rule of law. As the example of state child protective service systems reform in the United States indicates: The reforms do not achieve accountability by constraining frontline decisions through rules. Rather, frontline discretion is increased, but joined to the requirement that, in the course of establishing and adjusting plans for children, frontline workers and the professionals and stakeholders with whom they collaborate explain the choices they make in terms of the governing values of the program. Review of these explanations in turn allows administrative superiors and outside oversight bodies to detect and begin considering how to correct misjudgments by individual case workers, systemic flaws in operating routines at the local office or program level, and even ambiguity or mistake in the agency’s own conception of its key commitments and plans for achieving them. Thus, the agency learns to improve while monitoring what it does, and the same process that makes customization of services effective makes it accountable as well. We call such learning-by-monitoring institutions ‘experimentalist’. (Noonan et al. 2008, p. 3) Recent developments in experimentalist institutions have been documented in different countries and areas of public service.5 In all of these cases, analysts attributed the successful outcomes observed to the greater autonomy of front-line bureaucrats to adapt policy/project goals during their implementation in each specific situation, and to the establishment of mechanisms for continuous error detection and correction based on arguments and reports from the front line (usually peer review, benchmarking, and so on), culminating in periodic revisions of framework goals and procedures.

Roberto Rocha C. Pires 319

These models offer two very different solutions to the notion of the unchecked autonomy of government agents. The NPM solution to managing bureaucrats’ discretion emphasizes narrowing programmes and holding bureaucrats accountable to the attainment of specific and quantifiable performance targets. The EG solution suggests a process through which bureaucrats are constantly required to give reasons through peer and/or public review procedures for their discretionary decisions in resolving problems. The debate concerning the efficacy of these models is ongoing and the volume of research continues to increase. However, efforts to make a side-by-side comparison of these two models (NPM vs EG) in terms of their implications for governance and the management of regulatory bureaucracies remain scarce. This chapter seeks to fill this gap by expanding our knowledge of how these two models are enacted by comparing: (i) how they organize the work of regulatory inspectors; (ii) the impacts of each approach on inspection outcomes; and, finally, (iii) the strengths and weaknesses of different solutions to the problem of managing discretion.

One organization and two models for managing labour inspectors’ discretion This research is based on the observation of the operation of both managerial models working in the same government agency: the Department of Labour Inspection (DLI) within the Brazilian Ministry of Labour. The agency’s mission is to assess compliance with and enforce the national labour regulation, including both wage and hour laws and health and safety norms. The authority to enforce labour regulation is established at the federal level, but its implementation takes place through a decentralized system, through which approximately 3,000 labour inspectors are distributed across 27 state-level offices. These inspectors have jurisdiction over more than 78 million workers employed across the formal and informal labour markets, in 2.7 million registered firms, and untold unregistered firms, across all 5,564 Brazilian municipalities. Even though the agency is under-staffed and under-resourced given the magnitude of the task,6 the career of labour inspectors has been reformed significantly since the country’s re-democratization in 1985. The recruitment of inspectors through competitive public service exams and a rewarding career path – one of the best-paid jobs in the federal civil service – gradually resulted in higher organizational capacity and professionalization.

320 Regulating for Decent Work

I first started investigating this organization in December 2006, conducting fieldwork for ten months (over a period of two years) in three Brazilian states (Bahia, Minas Gerais and Pernambuco), and trying to understand the variation in regulatory styles (coercion or assistance) adopted by labour inspectors and the impacts of such variation on regulatory outcomes (Pires 2008). As I carried out data collection, observation, and interviews,7 I began to realize that there were two different systems operating simultaneously through which management (at the central level) supervised or monitored the work of the inspectors. The first system follows closely the dictates of NPM models and is based on individual and territorially circumscribed inspections monitored on the basis of individual performance targets (for example, the number of workers formally registered). It included a pay-for-performance compensation system (reaching up to a 45 per cent bonus on an inspector’s salary, with one-third being tied to individual performance and twothirds tied to the collective performance of the inspectorate). I also noticed a second system, resembling the EG approach; this was based on teams of inspectors working on projects organized around themes, sectors, or problems (for example, child labour, illegal subcontracting, or silicosis in the mining sector), monitored on the basis of team progress reports and their ability to address sector-wide problems. The coexistence of the two models offered a unique opportunity to compare these two forms of organizing inspection work. On the one hand, many important variables are held constant: the same organization and group of professionals (that is, same career, status, legal mandate, salaries, and so on), enforcing the same regulations in the same country and state (for example, Pernambuco), while dealing with the same specific issues (cases involving both wages and hours and health and safety regulations). On the other hand, under this relatively constant organizational setting, there are two different methods for organizing and supervising the street-level work of inspectors (that is, different strategies for managing discretion – NPM and EG). Therefore, the comparisons allow for the ‘isolation’ of the effects of the independent variable ‘management models’ on inspectors’ routines of work and on the outcomes of their actions.

Managing labour inspectorates: discretion, work routines, and outcomes Looking closely at the inspection agents in the state of Pernambuco, I compare two groups of labour inspectors, one managed following the

Roberto Rocha C. Pires 321

NPM model and the other through EG. I look at how the groups handled a common set of issues: severance payments commonly referred to in Brazil as FGTS (Fundo de Garantia Por Tempo de Serviço ), and fraudulent cooperatives and illegal subcontracting. For each issue, I compare (i) how the different management models organized inspection work (formal inspection procedures and enforcement strategies); and (ii) the impacts of these organizational similarities and/or differences on the labour inspection outcomes in Pernambuco.

FGTS collection In Brazil, national labour law establishes that formal workers have the right to severance payments8 when dismissed or retired, through their access to a special job security fund: FGTS (Fundo de Garantia por Tempo de Serviço ). Every month, employers contribute 8 per cent of a worker’s wage to this fund, which accumulates while the worker is still employed by the firm (that is, proportional to the worker’s tenure). As an important source of revenue for the federal government,9 the FGTS was instrumental in the fiscal adjustment of the 1990s. The Ministry of Planning put pressure on and provided incentives for labour inspectors to focus on the violations of FGTS payments, especially under-payments by firms, in order to raise federal revenues. Thus, for more than a decade, since the mid-1990s, the Department of Labour Inspection (DLI) defined FGTS collection as one of the main priorities for labour inspection in Brazil. Nonetheless, collecting these contributions through labour inspection has been organized in two different, yet coexisting, ways in DLI’s Pernambuco State Office. The first strategy took shape as soon as FGTS collection became a national priority, and following NPM reforms, DLI determined that every individual inspector anywhere in the country had to meet performance targets in terms of collecting such revenue. DLI instructed inspectors to verify conformity with FGTS in every single inspection, even in the cases in which inspection was motivated by other types of labour law violation. The impacts of defining FGTS collection as a priority and establishing performance targets were considerable – the collection of such revenue by inspectors in the entire country increased fourfold between 1996 and 2005 (Table 12.1). At the same time, since the mid-1990s, inspections in Pernambuco (as well as in other Brazilian states) have been organized according to a zoning system. The state-level office assigns pairs of labour inspectors to a geographical district within the state. Inspectors are expected to

322 Regulating for Decent Work Table 12.1 Total FGTS collection by labour inspection in Brazil, 1996–2005 Year

Amount (US$)

Year

Amount (US$)

1996 1997 1998 1999 2000

114,202,231.20 225,119,264.87 275,295,590.83 307,418,537.60 411,332,339.08

2001 2002 2003 2004 2005

368,500,063.09 480,284,704.85 398,969,690.00 414,483,525.00 411,443,815.00

Source: MTE/SIT (Ministério do Trabalho e Emprego/Secretaria de Inspeção do Trabalho).

cover their area by going in effect door to door, business to business, searching for firms violating labour regulations in their jurisdiction. In the absence of any special form of planning (for example, diagnostic instruments, investigation schedules), workplace inspections and investigation routines lack a strategic focus and are diverse, varying by each pair of inspectors. Inspections are triggered primarily in response to complaints received from individual workers and unions. In addition to having a FGTS collection target, each individual inspector is also expected to inspect a minimum number of firms each month. These performance targets (monetary and number of inspections) create incentives for inspectors to meet their goals by focusing their enforcement efforts on many small firms with small FGTS debts, because these are easier and quicker to process, leaving aside larger firms with potentially larger, but more complicated debts. As a result, a large number of inspectors, almost every one of them, have been investing most of their time on one single issue: collecting FGTS. The result is not very efficient in terms of FGTS collected (in US$) per inspection when compared to the strategy employed by a small group of inspectors under an alternative management system (see Table 12.2). In 2006, DLI authorized the creation of a pilot project in the Pernambuco State Office: the FGTS Operational Group or GO-FGTS.10 Four of the 145 inspectors in the State Office were assigned to the GO-FGTS. As these inspectors formed the group, they were automatically discharged from meeting the performance requirements assigned to ordinary inspectors. DLI classified them as performing ‘special activities’, and therefore being immune from typical NPM performance measurements. By grouping these inspectors together and freeing them from predefined quantitative performance targets and inspection procedures, they were also no longer confined to geographical districts. In effect, they were given more organizational space for devising enforcement efforts with a strategic focus on economic activities and larger firms with potentially higher FGTS debts.

Table 12.2 Comparing the outcomes of FGTS by labour inspectors in the Pernambuco State Office, 2007 Number of Inspected inspectors firms Zoning System (individual performance targets) GO-FGTS Total Pernambuco Total Brazil

Total FGTS collected % of firms % of total Avg. FGTS Avg. FGTS and notified by inspected FGTS collected per collected per inspection (US$) collected firm (US$) inspector (US$)

141

12,959

12,583,883.16

98.57%

34.81%

971.05

89,247.40

4 145 3,174

188 13,147 285,462

23,568,255.16 36,152,137.96 566,486,244.08

1.43% 100% –

65.19% 100% –

125,363.05 2,749.84 1,984.45

5,892,063.79 249,325.09 178,477.08

Source: MTE/SIT (Ministério do Trabalho e Emprego/Secretaria de Inspeção do Trabalho) and Pernambuco State Labour Inspection Office (SRTE-PE).

323

324 Regulating for Decent Work

The first step taken by the newly created group was to interact with CEF (Caixa Econômica Federal – the federal bank that administers FGTS deposits). The group of inspectors requested access to information with which to develop a system capable of identifying those firms with larger unpaid debts and the sectors with a greater propensity to have indebted firms. With a data analysis system in place, the GO-FGTS identified a short list of 1,000 firms with a higher potential for FGTS collection out of the universe of 62,000 firms in Pernambuco. These firms together employ approximately 40 per cent of all formal workers in the state; individually, they have relatively large workforces of their own or operate in sectors that have traditionally violated severance payment norms (for example, beverages, hotels, sugarcane processing). Targeting these firms made the task more manageable and, according to a member of the operational group, ‘… we can not only inspect but also monitor compliance in 1,000 firms’. In addition to monitoring, focusing on this target group allows the operational group to tailor enforcement strategies to each economic sector, adapting procedures to take advantage of particular circumstances, in order to produce greater impacts in terms of bringing a large number of firms into compliance and collecting as much revenue as possible in each intervention. Before the development of this informational system, it would take up to eight months for one inspector to audit a large firm with approximately 3,000 employees, for example, and then to identify only the FGTS debts and irregularities. As a result, large firms with complex debts were frequently either ignored or under-inspected. With the new data analysis tools, it takes only a few hours to identify FGTS irregularities in a firm. After obtaining such information, inspectors can perform workplace audits, having in hand the numbers and documentary evidence indicating the amount and potential causes of debt. In the course of these inspections, firms can choose to pay the debt right away, negotiate a payment schedule, or refuse to pay and bear the respective sanctions. In order to increase the coercive power of their operations, the group of inspectors reached out and partnered with the federal treasury attorneys, who can bring lawsuits resulting in heavy fines against debtors of the national treasury. Finally, in contrast to the responsive approach that often deals with one firm at a time employed by the pairs of inspectors working under the district/zone system, the operational group developed a standardized procedure for inspection. Through repeated, almost continuous conversations among its members and partners about different strategies for inspection, the group developed a repertoire of tactics that proved

Roberto Rocha C. Pires 325

efficient and effective. The group has periodic meetings to constantly discuss results and revise these practices as situations change. This relative standardization has had positive impacts for firms: it creates predictability and a sense of justice to the degree that firms discover that inspectors are using the same procedures for all firms in the same sector. This relative consistency, in comparison with the district-pair system, also creates greater legitimacy and more positive decisions when cases are appealed in court because inspection work appears coherent and uniform. The GO-FGTS procedures go beyond the strategic targeting of large firms. They also involve continued interactions with relevant partners, such as banks and federal treasury attorneys, leading to even better diagnostic information and a customized approach for each economic sector, which cumulatively produce a more effective enforcement. When compared to the outcomes of the zone system (Table 12.2), the GO-FGTS, which employs only 3 per cent (that is, four) of the inspectors in the Pernambuco office, collected 65 per cent of the total FGTS collected by all inspectors in the state. When the group was created, Pernambuco FGTS collection doubled between 2005 and 2006. In 2007, the Pernambuco GO-FGTS collected the highest absolute amount of FGTS (higher than in the most industrialized states) and benefited the largest number of workers among Brazil’s state offices. As the members of the group were freed from meeting predefined performance targets and had increased latitude to develop more complex actions by collaborating with other government agencies, they were able to be more productive using minimal internal resources.

Fraudulent cooperatives and illegal subcontracting Cooperatives of workers or producers have existed in Brazil since 1891, but they first received legal status in the 1970s; the 1988 Constitution further consolidated and stimulated this arrangement for work and production. However, in 1994, seemingly minor changes introduced in a paragraph of the labour law created ambiguity and uncertainty. Very quickly, cooperatives sprang up across the country as a low-cost method for firms’ outsourcing activities that has high labour costs (ranging from cleaning and maintenance services to administrative staff, doctors and nurses in hospitals). In the context of a worldwide restructuring of production, firms have been systematically ending historical legally contracted employment relationships and re-contracting the same group of workers through labour cooperatives.11 For firms, outsourcing to cooperatives represented a way to bypass labour regulations, avoiding

326 Regulating for Decent Work

labour costs and the payment of employees’ benefits, as they established service, rather than labour, contracts with the group of workers in a cooperative. For workers, these cooperatives represented the possibility of continuous employment, albeit with the loss of all prior employment rights and benefits; they became members, not employees, of service provision cooperatives. Because the cooperatives seemed to exploit an unintended legal loophole, and because they signalled a decline in both work protections and FGTS collections, current judicial interpretations of labour law in Brazil disallow the use of cooperatives for outsourcing ‘end-activities’ (for example, a software designer in a software development firm) and for the mere intermediation of labour (such as cooperatives that produce nothing but the labour force of its members). As a result, these fraudulent cooperatives have been the subject of interventions by labour inspectorates. And, as in the previous example of FGTS collection, inspectors have been dealing with the problem of fraudulent cooperatives simultaneously through two different approaches in the Pernambuco State Office. The first approach is also based on the organization of inspection work in pairs of inspectors according to a geographical zoning system. Following the same lines of FGTS inspection (above), inspectors are expected to meet performance goals with regard to the formalization of employment relationships. As firms have been resorting increasingly to outsourced labour from cooperatives, several inspectors in the Pernambuco office started to notice the frequency with which workers who had previously had formal and direct jobs were being pushed into these service provision cooperatives, thereby undermining the office’s targets for increasing formalization rates (that is, the number of jobs created under formal contracts and with all of the legally mandated rights and benefits). However, these fraudulent cooperatives are not easy to deal with under the quick and mechanized inspections anticipated by the performance measurement system of predefined and quantified goals. Thus, once they spot such frauds, pairs of inspectors have dealt with them with non-uniform understandings and inspection procedures – for example, the collection of accounting records on-site or by formal request from the office, interviews with workers during work hours or outside the workplace, the investigation of subcontracted cooperatives or only hiring firms – resulting in sanctions (notifications and fines) being easily overruled when they are appealed by firms. In addition to the growing number of complaints from workers and unions about the spread of

Roberto Rocha C. Pires 327

these fraudulent arrangements, labour prosecutors (MPT – Ministério Público do Trabalho ) have also been demanding more effectiveness from labour inspectors in dealing with the issue (that is, developing detailed investigations and producing the evidence necessary for prosecutors to file lawsuits against firms). In response to these external demands and also to internal pressures from an informal pioneering group of inspectors struggling with the issue of fraudulent cooperatives in the Pernambuco State Office since 2000, state and federal managers authorized the creation of Equipe de Combate à Fraude nas Relações de Emprego (ECOFREM – Special Group for Combating Employment Relation Frauds) – a group of seven inspectors dedicated to the investigation of frauds in employment relationships. Recognizing the complexity of the problem and the need for a special approach, federal managers granted ‘special activity’ status to the members of the group, exempting them from the standard performance measures. In order to be effective, the work of the group required a more open-ended process and detailed investigations to produce the documentary proof that characterizes the fraudulent employment relationship. To create sustainable cases against employers and fraudulent cooperatives, the group used diverse tactics including affidavits from workers, negotiation with firms, and partnerships with unions, professional associations and government organizations (including judges and state attorneys). To check the misuse of this open-ended mandate, state-level managers monitor ECOFREM’s performance on the basis of periodic written progress reports that are then used to justify the continuation of the project. Since its creation, ECOFREM has maintained intense dialogues with MPT prosecutors (for the characterization of frauds and employment relationships) and with labour unions (exchanging specific information about hiring practices in each sector) in order to devise common strategies and procedures for intervention and monitoring. The application of such standardized procedures provided the same treatment for firms in the same sector and strengthened the consistency of the regulatory efforts. Since the creation of ECOFREM, no single fine has been overruled in Pernambuco labour courts. Moreover, the investigation strategies and practices are reassessed periodically by inspectors together with MPT prosecutors, who strengthen the coercive power of the group by filing lawsuits against firms and cooperatives that fail to comply with consent decrees. As a result of these exchanges, the group started to undertake sector-wide operations for each economic sector. The main goal was to promote change in hiring practices in entire sectors, sector by

328 Regulating for Decent Work

sector, especially those sectors with large numbers of firms traditionally engaged in illicit forms of subcontracting. From 2001 to 2002, ECOFREM launched an operation to tackle fraudulent hiring arrangements in the software industry in Recife (Pernambuco), as they realized that 32 per cent of all complaints filed about ‘cooperatives’ were in the local IT industry. According to the coordinator of the group: The union [SINDPD] came after us and the MPT saying they had identified that the sector was growing but formal employment and the wage mass were decreasing. There was something happening. They had heard from some workers about the growth of cooperatives in the sector. We knew by experience that, in previous years, workers were mostly formal in this sector. Firms were resorting to cooperatives as a strategy to cut production costs in the face of fierce competition from IT firms in India. As the investigations progressed, the group realized that nearly every firm in the sector had some kind of arrangement involving the subcontracting of software designers, systems engineers and other professionals, in the form of ‘cooperatives’. These workers laboured every day in the same office, answering to the same manager, all of which constitute the employment relationship, according to Brazilian regulations. The inspectors and their partners were aware that the cost of formally hiring all these workers (retroactively) was so high for the mostly small and medium-sized firms facing international competition that it could have put them out of business. Nonetheless, they could not ignore the situation. To meet the various interests of the firms, the workers, and the state, the group of inspectors held a series of meetings with 35 firms. Over the course of eight months and more than 50 meetings, they negotiated a compliance schedule through which firms gradually re-hired workers on a direct basis, in line with increases in the demand for production. Between 2001 and 2003, the operation led firms to re-hire 2,215 workers who had been previously involved in fraudulent cooperatives. Between 2002 and 2006, ECOFREM developed an operation in the health care sector. In addition to receiving a significant number of complaints from workers, the group of inspectors had already diagnosed an acute problem in the sector through the analysis of official data on the termination of formal labour contracts in hospitals and clinics, especially for doctors, nurses and other health care professionals. In Recife, hospitals,

Roberto Rocha C. Pires 329

physiotherapy clinics and laboratories have long been misclassifying workers as members of service provision cooperatives. As a result of this ‘hiring culture’ in the sector, health care professionals were unprotected by laws restricting excessive overtime or granting rights to vacations. Medical workers suffered from sleep deprivation as the result of double night shifts, drug addiction, and mental problems, which cumulatively undermined the quality of the health care treatment for patients. In an effort to change the traditional hiring practices in this sector, ECOFREM administered a series of workshops attended by more than 195 professionals to explain the law and what firms’ managers could – and should – do to comply with regulations. They inspected 64 health care facilities. In collaboration with labour prosecutors (Ministério Público do Trabalho ), ECOFREM secured 177 consent decrees with firms and unions. Over a four-year period, they formally registered 2,067 formerly unregistered workers, including doctors, nurses and medical assistants. In addition to the concrete results of thousands of newly registered workers, these sector-specific operations became demonstration models with spillover effects in other sectors. For example, following on from the success of ECOFREM in the computer and health care sectors, the DLI created a national operation in 2006 to investigate similar frauds in banks in seven different states, all designed to draw on the experience of Pernambuco inspectors. Based on information from workers and from the government database of formal employment contracts (CAGED), inspectors report that many firms comply with the regulation simply because they had heard of other firms being inspected and punished. For example, a group of hospitals that had not yet been inspected registered more than 300 doctors, in the months immediately following ECOFREM’s operation. In addition to creating an effective procedure for dealing with such a complex problem as fraudulent cooperatives that bypass labour protections and regulations, the members of the group fare better than ordinary inspectors even when measured in terms of standard individual productivity indicators. For instance, while, on average (for 2007), each ordinary inspector formalized 15 jobs per month, each ECOFREM member formalized 25 jobs per month.

Discussion: management approaches, bureaucratic behaviour, and outcomes The comparisons across the cases described above makes clear that the two different methods for organizing inspection work (NPM and EG) involve significantly different tools through which supervisors control

330 Regulating for Decent Work

the work of inspectors, as well as different inspection strategies and procedures. The comparisons also suggest a plausible causal association between these strategies and the outcomes of inspection work. It is possible to draw three main conclusions. First, each managerial model offers a different strategy for supervisors to control the performance of front-line inspectors. As is indicated by each of the cases, the different ways through which supervisors monitor the work of inspectors also seem to affect their motivation and job performance. After more than two decades, even sympathetic analysts acknowledge NPM-inspired reforms have failed to meet expectations for improved public administration. Perhaps NPM reforms have simply become ‘middle-aged’, being unable to sustain their early energy (Hood and Peters 2004; Dunleavy et al. 2006); yet abundant criticism cites the paradoxical and dysfunctional effects of predefined quantitative performance measures (Bouckaert and Balk 1991). My interviews and observations confirm the claims by inspectors that the introduction of quantitative performance indicators from higher-level management interferes with professional autonomy, undermining both commitment and performance. The predefinition of specific and narrow goals by managers far removed from inspection routines and fieldwork favours mechanistic, bureaucratic check-list inspections. In effect, these official indicators tell inspectors a priori what they should consider relevant and what they should ignore, proscribing other potentially important observations and actions. Some inspectors reported their frustrations in not being able to develop cases more complexly (over long time periods, and in partnership with other knowledgeable and interested organizations), and having to move on before achieving noticeable improvements in business practices. In contrast, inspectors working on teams or special groups emphasized their ability to develop more contextual and sector-specific understandings of violations, business practices and legal norms, and, therefore, to mobilize both formal and informal institutions in the promotion of compliance (Frey in this volume). As one inspector stated, ‘inspection activities become less about law enforcement and more about how to stimulate employers and workers to continually improve work environments’. Moreover, it seems that special groups and teams have a different relationship with the administrative centres. Instead of primarily reporting achievements or failure according to predetermined numerical goals, they are granted the freedom to argue for the redefinition of goals, as well as procedures, and strategies as they develop their cases. Although subject to central and local supervision, the work of groups is also subject

Roberto Rocha C. Pires 331

to other control mechanisms: peer pressure from inside the labour inspectorate and external pressure from partners who – through their collaborations – build positive expectations concerning inspectors’ performance. These elements have already been identified in other studies as important sources of government workers’ motivation even under adverse conditions (Justice 1986; Tendler 1997). Second, the strategies and related procedures developed by inspectors under each model influence the pattern of inspectorial interventions and their outcomes. As the cases indicate, unplanned and complaintdriven inspections organized by the zone system of geographical jurisdictions employ non-uniform procedures for firms in the same sector/condition and usually fail to produce detailed investigations or legal evidence of wrongdoing. Even though the pay-for-performance system yielded improved outcomes in one case, FGTS collection, the process of establishing quantifiable targets, measures to monitor its attainment, and rewards for those bureaucrats who meet the goals failed to reduce unregistered workers and illegal cooperatives. Critics suggest that the definition of narrow and quantifiable performance targets where regulatory enforcement is necessarily fragmented into several agencies reduces the scope of action and will likely push bureaucracies away from addressing complex and interrelated problems.12 Conversely, the organization of inspection work through teams, groups and special projects eliminates some of the obstacles to the development of sector-wide operations and favours continued interaction between inspectors and diverse, but relevant partners. Sector-wide operations demand diagnostic information about the underlying causes of noncompliance, that is, contextual understanding of violations, while encouraging the customization of enforcement actions to sectorspecific social and productive dynamics. As articulated by a number of inspectors, they begin to move away from thinking about how to catch increasing numbers of lawbreakers as they gain greater latitude to consider why firms break the law in the first place. In addition, enforcement teams with sector-wide orientations push inspectors towards addressing more complex problems and towards practicing relational interdependence through open-ended processes. As the cases demonstrated, groups are more prone to seek collaborations within and across organizations, as they recognize their actions cannot by themselves deal effectively with a complex problem. These ‘open-ended conversations’ (Piore 2009; Lester and Piore 2006) between inspectors and other government and non-government actors are sources of innovation leading to effective problem-solving in sectors

332 Regulating for Decent Work

as diverse as health care and information technology, for example. Therefore, in contrast to the ‘technical’ specification of outputs and the mimicking of market performance incentives, the EG approach emphasizes ‘deliberative administration’, bringing in elements of dialogue, negotiation, sequential agreements and collaborations across different units of the administration and external partners as key features promoting creative solutions for complex collective problems (Fischer and Forester 1993; Evans 2002, 2005; Brugué 2004; Baccaro and Papadakis 2009). Additionally, case comparisons in the previous section suggest that inducing improved performance might be more associated with valuing bureaucrats’ autonomy to innovate and learn from reflection (justification) on their practices than with creating formal incentives and pressures for greater productivity on a narrow set of outcomes. The third and final conclusion adds a cautionary note to the benefits of organizing inspection through groups and special projects. Since the planning and execution of sector-wide operations, and the respective interactions within and across organizations, take time to hit the ground, the work of groups and special projects becomes unresponsive to the more immediate demands of workers, as well as of policy makers and politicians. Even though in the medium-to-long term the work of groups is more likely to solve complex and relevant problems, in the short run, hazardous and illegal situations experienced by workers may remain unnoticed and unremediated (Silbey 1984). In contrast, under the zone system, inspectors are free to respond immediately to workers’ complaints, even if the intervention is less likely to promote long-term changes in business practices or affect the underlying causes of non-compliance. For public sector bureaucracies such as the labour inspectorate, responsiveness is an important attribute for building a good reputation and public image, as well as for harvesting political support. Therefore, the possibility of combining both models under the same service seems promising as a way of reconciling problem solving with responsiveness, and reaching a desirable balance, as described by March (1991), between the organizational functions of the exploration of new possibilities (experimentation/innovation) and the exploitation of old certainties (efficiency/mass production). In other words, a division of labour between units focused on innovation and units focused on processing routine work efficiently, in which the success of each branch is dependent upon the other branch doing its part, might be desirable for overall organizational performance. In the Brazilian experience, we currently observe a gradual shift away from the prevalence of

Roberto Rocha C. Pires 333

the pay-for-performance system and towards a greater emphasis on the organization of inspection work based on groups and special projects. Responsive and geographically bounded inspections should still remain, but as a residual category, only sufficient to cover for emergency complaints received from workers in vulnerable situations (for example, the non-payment of wages, imminent risk of death of accident, and so on).

Concluding remarks The comparative analysis developed in this study indicates that the variable ‘management model’ (or the strategies adopted by management to control the discretion and performance of street-level officers) has important implications when it comes to explaining why regulatory inspectors perceive possibilities for innovation in the development of their work. The cases analysed indicate that certain features of the management models and the ways they organize inspection work – such as predefined performance targets vs open-ended processes and the constant revision of goals, performance measures and inspection procedures; individual vs team work; and so on – affects the extent to which inspectors see the relevance and possibilities for working collaboratively within and across organizations in the development of effective solutions for compliance problems. The empirical material also provides supporting evidence for the claim that improving bureaucratic performance is not only about defining the right incentive system but should focus primarily on: (i) setting in motion processes for the constant revision of goals and their measures; and (ii) redefining the mechanisms and procedures to alter work routines every time they become hostile to the achievement of desired goals; both of which necessarily require interactions with a wide array of potential partners. Therefore, a deeper understanding of how management practices evolve in regulatory bureaucracies and how inspectors incorporate such practices in their routines should be an indispensable aspect of the quest for explaining the outcomes of labour inspection.

Notes 1. The World Bank’s World Business Environment Survey (Batra et al. 2003; Doing Business 2006) identifies labour market regulation as chief among the obstacles towards building enabling conditions for the operation and growth of enterprises. For example, econometric studies such as Friedman et al. (2000) have shown that countries with more regulation tend to have a higher

334 Regulating for Decent Work

2.

3.

4.

5.

6.

share of the unofficial economy in total GDP – a one-point increase in the regulation index (ranging from 1 to 5), ceteris paribus, is associated with a 10 per cent increase in the share of the unofficial economy, when controlled for the GDP per capita, for 76 developing, transition and developed countries. For example, in Brazil – one of the most heavily regulated labour markets in the world (Botero et al. 2004; Doing Business 2006) – those who share the perception of the trade-off have advocated for deregulation and flexibilization as the most effective way to attract investment, promote firms’ competitiveness, and increase formal job growth rates (see Berg in this volume). Some of these authors provide evidence, mainly from OECD countries, that government spending on social protection, employment protection regulations, and income security in the labour market are significantly correlated with positive changes in international economic performance and competitiveness – that is, trade and inward foreign investment. This is due primarily to the wide acceptance and relatively narrow interpretations of Max Weber’s work on bureaucracy. For Weber, bureaucracy represented the organizational form of a sociopolitical system (system of domination) that stood in sharp contrast to other ideal-types of organization, namely charismatic and patriarchic domination. In the latter two systems, the exercise of power was legitimated, respectively, by the extraordinary characteristics of the leader or by tradition. In contrast to these historically more prevalent systems of domination, Weber described bureaucracies (and rational-legal domination) as the rule of law, in which formal rules establish a clear line of command through hierarchal structure; prescribe the eligibility criteria, duties and competences attached to positions within the organization; promote a division of labour (specialization); define the procedures and scope of decision-making processes at all levels of the organization (including the decisions about changes in rules); and specify the processes for succession in power. These features of rational-legal organization cumulatively lead to the control of individual and personal inclinations, desires and opinions, and the minimization of their effects in the machine-like functioning of the organization (Weber 1946). Weber’s ideal-typical conceptualization of bureaucracy, which – only as an ideal type – envisioned a gapless institutional framework dictating how agents act under all possible circumstances and thereby making state agents impersonal cogs within a preprogrammed organizational machine, provided the theoretical framework for interpretations of actual bureaucratic organizations as having ‘rule of law rather than personal discretion at their heart’ (Lange and Rueschemeyer 2005, p. 241). Examples include: reform of public schools and rolling rule regimes (metaregulation) in the regulation of food safety in the US (Sabel 2004); reform of state child protective services systems in Alabama and Utah (Noonan, Sabel and Simon 2008); welfare services in Denmark, Ireland and the Netherlands (Sabel 2005); and systems of social protection, occupational health and safety, drug and food safety, telecommunications, electricity, maritime safety, and financial services in the European Union (Sabel and Zeitlin 2008). The number of inspectors in Brazil is only half that recommended by the ILO and lower (per 100,000 workers rate) than in some of its Latin American neighbours such as Argentina, Chile and Uruguay (Piore and Schrank 2007). However, even constrained by these resource limitations, Brazil’s labour

Roberto Rocha C. Pires 335

7.

8.

9.

10. 11. 12.

inspection service has received international recognition for its outstanding and innovative programmes to eliminate forced labour and child labour. The interviews added up to a total of 114, averaging two hours in length. I conducted approximately half (49) of the interviewees with labour inspectors in the three states and at the central level in Brasilia. I complemented and cross-checked (triangulation) the stories and data collected from these labour inspectors by interviewing another 65 actors who were involved in specific cases, including firm owners, managers, workers and representatives of business associations, trade unions and government agencies (for example, the National Health and Safety Institute, the Attorney General’s Office, and development banks). In Brazil, all formal employment relationships must be recorded by employers on the employee’s work permit (carteira de trabalho). This permit entitles the worker to several wage and non-wage benefits paid for by the employer, such as retirement benefits, unemployment insurance, and severance payments (see also Berg in this volume). In Brazil, FGTS is the main source of funding for housing, sanitation and infrastructure projects, as well as social policies. And, in public accounting terms, this fund plays a major role in balancing federal debts vs revenues. In 2007, the project is scaled up and operational groups become mandatory for all state offices. See Berg (in this volume) for an analysis of the growth of informality in the 1990s and its decline in the following decade in Brazil. Critical reactions to NPM reforms come not only from scholars but also from public sector workers and professionals themselves. Current criticism on NPM reforms (and its disaggregation of organizations – core vs other functions – and narrowing of programmes and tasks to the extent that they can be written into a contract) recognizes its inability to deal with complex, interrelated or cross-cutting problems, such as preventative health, school reforms, childcare services, and social assistance programmes, all of which require the coordination of local knowledge with a range of different services provided by different government organizations.

References Abrami, R. 2005. ‘On the high road: trade, international standards, and national competitiveness’, Accountability Forum, vol. 4, pp. 39–46. Baccaro, L.; Papadakis, K. 2009. ‘The downside of deliberative public administration’, Socio-Economic Review, vol. 7, no. 2, pp. 245–76. Barzelay, M. 2001. The New Public Management: Improving Research and Policy Dialogue (Berkeley, CA: University of California Press). Batra, G.; Kaufmann, D.; Stone, A.H.W. 2003. Investment Climate around the World: Voices of the Firms from the World Business Environment Survey (Washington, DC: World Bank). Bazan, L.; Navas-Aleman, L. 2004. ‘The underground revolution in the Sinos Valley: a comparison of upgrading in global and national value chains’, in Schmitz, H. (ed.), Local Enterprises in the Global Economy: Issues of Governance and Upgrading (Cheltenham, UK, and Northampton, MA: Edward Elgar), pp. 110–39.

336 Regulating for Decent Work Bernard, P.; Boucher, G. 2007. ‘Institutional competitiveness, social investment, and welfare regimes’, Regulation & Governance, vol. 1, no. 3, pp. 213–29. Bittner, E. 1967. ‘The police on skid row: a study of peace keeping’, American Sociological Review, vol. 32, no. 5, pp. 699–715. Bittner, E. 1990. Aspects of Police Work (Boston, MA: Northeastern University Press). Blau, P. 1970. ‘A formal theory of differentiation in organizations’, American Sociological Review, vol. 35, no. 2, pp. 201–18. Botero, J.C.; Djankov, S.; La Porta, R.; Lopez-de-Silanes, F.; Shleifer, A. 2004. ‘The regulation of labor’, Quarterly Journal of Economics, vol. 119, no. 4, pp. 1339–82. Bouckaet, G.; Balk, W. 1991. ‘Public productivity measurement: diseases and cures’, Public Productivity & Management Review, vol. 15, no. 2, pp. 229–35. Bresser Pereira, L.C.; Spink, P. 1999. Reforming the State: Managerial Public Administration in Latin America (Boulder, CO: Lynne Rienner). Brown, M. 1981. Working the Streets: Police Discretion and the Dilemmas of Reform (New York: Russell Sage Foundation). Brugué, Q. 2004. ‘Modernizar la administración desde la izquierda: burocracia, nueva gestión públicay administración deliberativa’, Revista del CLAD Reforma y Democracia, vol. 29, June, pp. 1–16. Bryner, G. 1987. Bureaucratic Discretion: Law and Policy in Federal Regulatory Agencies (New York: Pergamon Press). Calvert, R.; McCubbins, M.; Weingast, B. 1989. ‘A theory of political control and agency discretion’, American Journal of Political Science, vol. 33, no. 3, pp. 588–611. Crozier, M. 1964. The Bureaucratic Phenomenon (Chicago: The University of Chicago Press). Davis, K.C. 1969. Discretionary Justice (Baton Rouge: Louisiana State University Press). Doing Business. 2006. ‘Doing Business 2006: Creating Jobs’. Report co-published by the World Bank and International Financial Corporation. http://www. doingbusiness.org/. Dorf, M.; Sabel, C. 1998. ‘A constitution of democratic experimentalism’, Columbia Law Review, vol. 98, no. 2, pp. 267–473. Dunleavy, P.; Hood, C. 1994. ‘From old public administration to new public management’, Public Money and Management, vol. 14, no. 3, pp. 9–16. Dunleavy, P.; Margetts, H.; Bastow, S.; Tinkler, J. 2006. ‘New public management is dead – Long live digital-era governance’, Journal of Public Administration Research and Theory, vol. 16, no. 3, pp. 467–94. Evans, P. 2002. ‘Hybridity as an administrative strategy: combining bureaucratic capacity with market signals and deliberative democracy’. Plenary Lecture at VII International Congress of CLAD on State Reform and Administrative Development, Lisbon, Portugal, 11 October. Evans, P. 2005. ‘Harnessing the state: rebalancing strategies for monitoring and motivation’, in Lange, M.; Rueschemeyer, D. (eds) States and Development: Historical Antecedents of Stagnation and Advance (London: Palgrave), pp. 26–47. Fischer, F.; Forester, J. (eds). 1993. The Argumentative Turn in Policy Analysis and Planning (Durham, NC: Duke University Press). Friedman, E.; Johnson, S.; Kaufman, D.; Zoido-Lobaton, P. 2000. ‘Dodging the grabbing hand: the determinants of unofficial activity in 69 countries’, Journal of Public Economics, vol. 67, pp. 459–93.

Roberto Rocha C. Pires 337 Gouldner, A. 1954. Patterns of Industrial Bureaucracy: A Case Study of Modern Factory Administration (New York: The Free Press). Handler, J. 1986. Conditions of Discretion: Autonomy, Community, and Bureaucracy (New York: Russell Sage Foundation). Hawkins, K. (ed.). 1992. The Uses of Discretion (Oxford: Oxford University Press). Hood, C.; Peters, G. 2004. ‘The middle aging of new public management: into the age of paradox?’, Journal of Public Administration Research and Theory, vol. 14, no. 3, pp. 267–82. Johnson, S.; Kaufman, D.; Zoido-Lobaton, P. 1998. ‘Regulatory discretion and the unofficial economy’, American Economics Review, vol. 88, no. 2, pp. 387–92. Justice, J. 1986. ‘The invisible health worker: the peon’, and ‘Sources of information’, in Policies, Plans & People: Culture and Health Development in Nepal (Berkeley, CA: University of California Press), pp. 101–6. Lange, M.; Reuschemeyer, D. (eds). 2005. States and Development: Historical Antecedents of Stagnation and Advance (New York: Palgrave Macmillan). Leonard, D.K. 1977. Reaching the Peasant Farmer: Organization Theory and Practice in Kenya (Chicago and London: University of Chicago Press). Lester, R.; Piore, M. 2006. Innovation: The Missing Dimension (Cambridge, MA: MIT Press). Lipsky, M. 1980. Street Level Bureaucracy: Dilemmas of the Individual in Public Services (New York: Russell Sage Foundation). March, J. 1991. ‘Exploration and exploitation in organizational learning’, Organization Science, vol. 2, no. 1, pp. 71–87. Maynard-Moody, S.; Musheno, M. 2003. Cops, Teachers, Counselors: Stories from the Front Lines of Public Service (Ann Arbor, MI: The University of Michigan Press). McCubbins, M.; Noll, R.; Weingast, B. 1987. ‘Administrative procedures as instruments of political control’, Journal of Law, Economics, and Organization, vol. 3, no. 2, pp. 243–77. McCubbins, M.; Schwartz, T. 1984. ‘Congressional oversight overlooked: police patrol versus fire alarms’, American Journal of Political Science, vol. 28, no. 1, pp. 165–79. Milberg, W.; Houston, E. 2005. ‘The high road and the low road to international competitiveness: extending the neo-schumpeterian trade model beyond technology’, International Review of Applied Economics, vol. 19, no. 2, pp. 137–62. Noonan, K.G.; Sabel, C.; Simon, W. 2008. ‘Legal accountability in the servicebased welfare state: lessons from child welfare reform’, Columbia Public Law Research Paper No. 08-162; Law of Social Inquiry, 2009. Available at SSRN: http://ssrn.com/abstract=1088020. Osborne, D.; Gaebler, T. 1992. Reinventing Government: How the Entrepreneurial Spirit is Transforming the Public Sector (Reading, MA: Addison-Wesley Publ. Co.). Piore, M. 2009. ‘SASE annual meeting 2008 Presidential Address, San José, Costa Rica: second thoughts: on economics, sociology, neoliberalism, Polanyi’s double movement and intellectual vacuums’, Socio-Economic Review, vol. 7, no. 1, pp. 161–75. Piore, M.; Schrank, A. 2007. ‘Norms, regulations and labor standards in Central America’, Serie Estudios y Perspectivas, CEPAL, Sede subregional en México, México.

338 Regulating for Decent Work Piore, M.; Schrank, A. 2008. ‘Toward managed flexibility: the revival of labour inspection in the Latin world’, International Labour Review, vol. 147, no. 1, pp. 1–23. Pires, R. 2008. ‘Promoting sustainable compliance: styles of labour inspection and compliance outcomes in Brazil’, International Labour Review, vol. 147, no. 1, pp. 199–229. Pollitt, C. 1995. ‘Justification by works or by faith? Evaluating the new public management’, Evaluation, vol. 1, no. 2, pp. 133–54. Posthuma, A. 2004. ‘Taking a seat in the global marketplace: opportunities for “high road” upgrading in the Indonesian wood furniture sector?’, in Keister, L.A. (eds) Globalism/Localism at Work (Research in the Sociology of Work), vol. 13 (Bingley: Emerald Group Publishing Limited), pp. 181–201. Rodrik, D. 1997. Has Globalization Gone Too Far? (Washington, DC: Institute of International Economics). Sabel, C. 2004. ‘Beyond principal–agent governance: experimentalist organizations, learning and accountability’, in Engelen, E.; Ho, M.S.D. (eds) De Staat van de Democratie. Democratie voorbij de Staat, WRR Verkenning 3 (Amsterdam: Amsterdam University Press), pp. 173–95. Sabel, C. 2005. ‘Globalisation, new public services, local democracy: what’s the connection?’, in OECD, Local Governance and the Drivers of Growth (Paris: OECD), pp. 111–31. Sabel, C.; Zeitlin, J. 2008. ‘Learning from difference: the new architecture of experimentalist governance in the EU’, European Law Journal, vol. 14, no. 3, May, pp. 271–327. Schneider, F.; Enste, D.H. 2000. ‘Shadow economies: size, causes, and consequences’, Journal of Economic Literature, vol. 38, no. 1, pp. 77–114. Silbey, S.S. 1984. ‘The consequences of responsive regulation’, in Hawkins, K.; Thomas, J.M. (eds) Enforcing Regulation (Boston, MA: Kluwer-Nijhoff Publishing), pp. 147–70. Silbey, S.S.; Huising, R.; Coslovsky, S. 2009. ‘The sociological citizen: recognizing relational interdependence in law and organizations’, Annee Sociologique, vol. 59, no. 1, pp. 201–29. Silbey, S.; Bittner, E. 1982. ‘The availability of law’, Law and Policy Quarterly, vol. 4, no. 4, pp. 399–434. Stiglitz, J.E. 2000. ‘Democratic development as the fruits of labor’, Perspectives on Work, Vol. 4, No. 1, pp. 31–8. Tendler, J. 1997. Good Government in the Tropics (Baltimore, MD: Johns Hopkins University Press). Van Maanen, J. 1973. ‘Observations on the making of policemen’, Human Organization, vol. 32, no. 4, pp. 407–18. Van Maanen, J. 1978. ‘The asshole’, in Manning, P.; Van Maanen, J. (eds) Policing: A View from the Street (New York: Random House), pp. 221–38. Weber, M. 1946. Essays in Sociology, with Gerth, H.H.; Mills, C.W. (Oxford: Oxford University Press). Wilson, J.Q. 1968. Varieties of Police Behavior: The Management of Law and Order in Eight Communities (Cambridge, MA: Harvard University Press). Wilson, J.Q. 1989. Bureaucracy: What Government Agencies Do and Why They Do It (New York: Basic Books).

13 A Diagnostic Methodology for Regulating Decent Work Diane F. Frey

Introduction This chapter presents a diagnostic methodology to provide insight into the interventions and processes necessary to progressively achieve decent work as it is defined in international law. Substantial progress has been made in developing Decent Work Indicators1 and in bringing them into spheres beyond the International Labour Organization (ILO), for example, as targets for the Millennium Development Goals (MDGs). Indicators will no doubt play a critical role in measuring progress in achieving the ILO Decent Work Agenda. Yet indicators have limitations. They are not intended to be diagnostic and they do not tell us a great deal about the underlying processes associated with improving or deteriorating performance. Further, they do not generally help to clarify or build consensus about which policy interventions will improve progress towards achieving decent work for all. The diagnostic methodology presented here intends to complement decent work indicators by addressing these limitations. Drawing on institutions theory from political economy the chapter reframes labour market regulations as holistic institutions, comprised of rules, norms and actual behaviours, the so-called ‘rules of the game’ of employment. Based on the understanding that institutions reinforce and contribute to labour rights violations, the chapter presents an approach to improving labour market regulation and compliance based on Harold Koh’s compliance theory from international law. The chapter uses the example of work-time limits and constraints on obligatory overtime for purposes of demonstrating the diagnostic methodology. The approach could equally be applied to other areas and scales of inquiry such as labour inspection (Pires in this volume) 339

340 Regulating for Decent Work

or large-scale reforms such as Chinese labour law reform (Cooke in this volume). Methodologies based on institutions theory have already guided examinations of work-time regulation. For example, they are implicitly used by the ILO Committee of Experts on the Application of Conventions and Recommendations (CEACR) when it examines the effect given in law and practice to Conventions such as Nos 1 and 30 on work time for purposes of assessing compliance (ILO 2005). Lee and McCann have also emphasized the importance of examining not just labour laws but their impact and actual outcomes. Their notion of observance of statutory labour regulation is a composite encompassing enforcement by labour ministries and courts, as well as the influence of regulation through norms even in the absence of enforcement (Lee and McCann 2008, p. 47). The present effort also emphasizes outcomes; however, it examines more closely the institutional basis of employer observance of work-time limits as compliance, the situation where the employer’s behaviour conforms to specified rules. Following this introduction, I briefly introduce institutions theory and illustrate it with the example of work-time limits and obligatory overtime. Using this institutional framework, it is possible to identify problems and combinations of problems that arise in implementing work-time limits. For example, there may be rule-based problems in which the written rules contradict or fail to adequately give effect to normative goals and therefore undermine them. Alternatively, well-written rules may not be enforced effectively. Additionally, formal institutions interact with informal institutions such as corruption, bribery or blacklisting, and in the face of weak formal institutions, they may prevail in structuring social behaviour. Finally, institutions beyond the immediate sphere of work-time rules may impact on the observance of work-time limits. These include, for example, immigration institutions interacting with labour institutions. Then the chapter explains how institutional arrangements can be systematically mapped to correspond to dimensions of Harold Koh’s compliance theory from international law. In this way, institutional shortcomings can be matched to compliance interventions. Compliance theory is well suited to institutional approaches because it, like institution theory, treats norms, rules and behaviours as critical components in achieving change and compliance. The contention is that, to be successful, interventions must be integrated, multiple and mutually reinforcing to achieve effective work-time limits in line with decent work. In the longer range, the goal of these interventions is to create circumstances

Diane F. Frey 341

where actors adopt norm-based behaviours because they have internalized them into their own system of values (Koh 1997, 1998). The chapter then moves on to illustrate the diagnostic methodology with an examination of work-time limits and obligatory overtime constraints in Honduras. Based on evidence taken from the reports of the ILO, non-governmental organizations such as the International Trade Union Confederation (ITUC),2 and US State Department reports, it presents an analysis of the formal, informal and multiple institutional arrangements associated with the non-observance of work-time limits and obligatory overtime constraints. It also links the institutional arrangements to interventions recommended to improve compliance. Finally, some conclusions about the diagnostic methodology will be drawn.

Institutions and work-time limits Labour institutions are the formal as well as informal rules, practices and policies that structure all of the conditions in which people work (Berg and Kucera 2008). Work-time limits and obligatory overtime, like minimum wages, vocational training, social security and collective bargaining, are examples of different sets of labour market rules, practices and policies here called labour market institutions. Labour institutions such as work-time limits are the outcome of a complex system of social relations, production and national laws (ibid., p. 11). Power and political compromise play a key role in the establishment of these institutions, as aptly described by Cooke in this volume (see also Knight 1992; Lukes 2005). The contention and compromise that gives rise to institutions on work-time limits is illustrated, for example, by the debate over whether a national law should mandate overtime premium pay after 40 or 48 hours in a week. The adoption of either rule will affect the absolute and relative income of employers and workers. The effectiveness of formal institutions The first dimension of this analysis is the effectiveness of the formal institutions related to work-time limits and constraints on obligatory overtime. Formal institutions are effective to the degree that the obligatory overtime and work time are actually constrained and governed as opposed to being widely ignored or circumvented (Amable 2003; Helmke and Levitsky 2006). Formal institutions are composed of interrelated elements, including the substantive content of formal rules along with their interpretation

342 Regulating for Decent Work

and enforcement. Rules are institutional statements and include: (i) designations identifying to whom the statement applies; (ii) verbs that permit, obligate or forbid; (iii) the particular actions or outcomes to which the verbs apply; (iv) conditions defining when, where, how and to what extent that action or outcome in question is permitted, obligatory or forbidden; and (v) the ‘or else’ statement defining the sanctions to be imposed for not following the rule (Crawford and Ostrom 1995, p. 584). The substance of the rules on obligatory overtime and work-time limits, whether in a national law or ILO Convention, can be analysed using this institutional grammar. Institutional grammar could be used to systematically identify and analyse the regulatory rules underpinning the discretion exercised by labour inspectors in Brazil as presented by Pires in this volume. In this framework, problems in regulation do not arise simply from inadequate enforcement. Instead, the formulation of the rules can, alone, as well as in combination with their interpretation and enforcement, undermine observance of work-time limits. In other words, rules ostensibly created to limit work-time can actually undermine its effective limitation. Similarly, rules ostensibly prohibiting obligatory overtime can make it easier for employers to impose obligatory overtime. Some of the ways that formal rules do this are by containing contradictory verbs, some of which prohibit while others allow work beyond nationally established limits. Substantive rules can also be limited by narrowly naming the categories of workers for whom the rule applies or equally by exempting categories of workers from application of the rule (Frey 2010, p. 141). Alternatively, rules may clearly prohibit obligatory overtime but then contain inadequate or entirely absent procedures and sanctions to address violations of the rules (ibid., p. 142). Interpretation is an authoritative decision on whether the rule has been violated. The process often involves a challenge, for example, in an administrative or judicial proceeding, followed by a decision, defining and elaborating on the rules and their definitions in relation to particular circumstances (Koh 1998). The enforcement of a rule is the application of the corresponding sanction, which raises the cost of non-compliance by eliminating potential profit as well as imposing additional costs as punishment (Scholz 1997; Fisher 1981). Enforcement mechanisms include monitoring and penalties, such as fines and prison sentences, as well as the procedural rules through which the mechanisms are applied, which may increase or decrease the probability of conviction (Becker 1968). Alternatively, it is possible to imagine that rules might establish positive incentives and rewards for conformance, such as tax breaks.

Diane F. Frey 343

Interactions between formal and informal institutions The second dimension of institutional analysis is the interaction between formal and informal institutions. Informal institutions are not created through official channels such as legislatures or courts, and penalties imposed for their breach are not official but are, nonetheless, real and include social exclusion, loss of reputation and retaliation (Helmke and Levitsky 2006). Informal institutions include social norms and social conventions (ibid.). Social norms are values that are observed irrespective of the behaviours of other and correspond to ideas of prudential behaviour (North 1990, p. 4; Ostrom 2005). For example, it may be considered virtually unthinkable to schedule work on highly important national or religious holidays. In contrast with the prudential values associated with norms, social conventions are about shared expectations (Amable 2003). Adherence to social conventions depends upon the fact that others also adhere to them. Bribery is a common example of a social convention. Informal institutions interact with formal institutions and contribute to the observance or non-observance of work-time limits and constraints on obligatory overtime. They can contribute to observance by strengthening incentives to comply and can even substitute for deficits in ineffective formal institutions such as ambiguous substantive rules (Helmke and Levitsky 2006). As Lee and McCann (2008, p. 47) point out, ‘laws can have an effect on practice … even in the absence of enforcement’ (see also Browne, Deakin, and Wilkerson 2002). Alternatively, informal institutions can undermine the observance of work-time limits by displacing formal institutions and thereby contributing to their irrelevance (Helmke and Levitsky 2006). Even when formal institutions are not openly violated, informal institutions can undermine observance through, for example, creative compliance, meaning situations where actors comply with the letter of the law while violating its normative spirit (ibid.). Institutions beyond those directly related to work-time limits The third dimension of institutional analysis is the influence of institutions in spheres of economic and social life that are not directly related to work-time limits and constraints on obligatory overtime. Institutions in different spheres interact and reinforce each other in several different ways (Amable 2003; Hall and Soskice 2001). One example is where workers in general are protected against obligatory overtime by specific institutions, but have difficulty gaining effective access to protection because citizenship institutions put them at risk of deportation if

344 Regulating for Decent Work

they file complaints. Another example is when one set of institutions imposes its logic on another institution. Market-oriented institutions can undermine labour protection institutions as, for example, when governments respond to the World Bank’s ‘Doing Business’ indicators by weakening their enforcement of labour regulations. The main contribution of this last dimension is to caution regulators to be mindful of and attempt to identify influences on observance that may be seemingly unrelated or tangential. Power and contention in institutions can be seen in all three dimensions. In the settlement over the establishment of formal institutions, power asymmetries result in differential formulations of institutional rules and enforcement regimes. Once the formal rules are established there may not be sufficient power to bring about a fundamental change in an unwanted formal institutional settlement, but there are still opportunities for contentious skirmishing to subvert or strengthen formal institutions (Mershon 1994). Finally, institutional hierarchies reflect the imposition of one set of institutional rules over others based on power. Examples of the role of power can be found elsewhere in this volume, as, for example, when Cooke (in this volume) points to continued skirmishing among labour relations actors in China while its relatively new labour law reforms are being implemented.

Compliance theory As Berg and Kucera (2008, p. 11) point out, labour institutions, such as work-time limits, are the outcome of social relations, production and national laws. As a result, national labour institutions may diverge significantly from internationally recognized definitions and standards for ‘decent work’, such as those established by the ILO and international human rights law. The ILO has defined decent work as consisting of four pillars: (i) employment promotion; (ii) social protection; (iii) social dialogue; and (iv) rights at work (ILO 1999). All of these pillars are grounded in ILO Conventions and interpretations by ILO Committees and its governing body. Decent work has also been defined by the United Nations Committee on Economic Social and Cultural Rights (CESCR) in its General Comment 18. In this document, the Committee defines decent work as composed of three parts: (i) a general right to work, protecting the right to freely choose employment (article 6); (ii) an individual right to work, covering the rights to just and favourable conditions at work (article 8); and (iii) the collective right to work, including the rights to form and join trade unions (CESCR 2006).

Diane F. Frey 345

National institutional arrangements concerning work-time limits and constraints on obligatory overtime can be analysed systematically for their compliance with international norms, such as ILO Conventions on hours of work, weekly rest, the organization of work time, night work and holidays with pay (ILO 2005). Outside the ILO framework, the same institutions could be assessed for their compliance with the provisions of the International Covenant on Economic, Social and Cultural Rights, including those protecting the right to just and favourable conditions of work (Article 7), the right to safe and healthy working conditions (Article 7b), and the right to rest, leisure and reasonable limitation on working hours and periodic holidays with pay (Article 7d). International law provides a meaningful and common metric when looking at the many diverse aspects of decent work, whether examining obligatory overtime in a small Central American country, labour law reform in China (Cooke in this volume), the largest society in the world, or a singular aspect of labour inspection in Brazil (Pires in this volume). Through the use of common metrics like ILO Conventions, research and narratives on decent work in one country or in one area of inquiry can be more usefully comparative. Compliance is a state of conformity between a specified rule and an actor’s behaviour (Fisher 1981, p. 20). According to Koh (1998, p. 4), the best chance of achieving compliance is to create circumstances in which actors, such as states and employers, adopt norm-based behaviours because, over time, they have internalized the norms into their own systems of values. In the interim, building towards this state of compliance is made possible by multiple complementary mechanisms corresponding to the diverse motivations that drive compliance. These include: (i) being coerced into adopting and enforcing rules; (ii) acting out of self-interest by making non-compliance costly and irrational; (iii) enhancing the legitimacy of the norms by reinforcing them as company, community, national and international values; and (iv) internalizing international legal norms into domestic legal, social and political processes (Koh 1997, 1998). Internalization takes place in different dimensions, which correspond usefully to institutional components. Legislative internalization occurs when ‘international legal norms become embedded into binding domestic legislation’ (Koh 1998, p. 643). National legislation can be assessed in terms of whether its institutional grammar is consistent with the international legal norm. It is also possible using institutional grammar to locate specific problems such as contradictory substantive rules, overly broad exemptions from substantive rules, and inadequate procedures and sanctions.

346 Regulating for Decent Work

Judicial internalization3 is the incorporation of an international norm into the domestic legal system via judicial interpretation or, in some cases, executive action, legislative action or some combination of all three actions (Koh 1997). Institutional frameworks can help in assessing the congruence between domestic laws and international legal norms. They can help locate the potential sources of incongruence in formal institutions such as the formulation of rules, their interpretation and enforcement. They can also help to explain or diagnose the role of informal institutions, such as corruption, in contributing to failed implementation and non-observance of international legal norms. Koh’s (1997) ultimate goal is social internalization, which is the condition in which a norm acquires such a degree of public legitimacy that there is widespread general obedience with it. A lack of social internalization is to be expected where there is widespread non-observance of international legal norms. In such cases, institutional frameworks provide useful diagnostics. It is possible to assess (or at least ask) whether there is evidence that the legislature, judiciary or executive branches work actively to enhance, or alternatively undermine, the normative legitimacy of the international legal norm. Similarly, it is possible to assess (or at least ask) whether there are social norms counter to the international legal norm that are championed by branches of government and other social actors. Cross-cutting these three dimensions of internalization is Koh’s concept of political internalization, which occurs when political elites accept an international norm and incorporate the norm into domestic policy. This, in many ways, is the most difficult assessment to make. Here too institutional theory may help guide an assessment. For example, it is helpful to consider whether the government itself observes the relevant international legal norm in its role as employer. If the government does not adhere to international legal norms on labour then it may well be less likely to enforce these norms against other employers. Similarly, it is helpful to assess the government’s success in addressing incongruities that are brought to its attention. If the government takes no action, for example, this may indicate a de facto lack of acceptance of the legal norm despite its formal ratification. Finally, it is important to ask and assess, with the input of the social partners, whether power relations in establishing and maintaining domestic institutions have resulted in their congruence with international legal norms. There is no particular sequence in which legislative, judicial, social and political internalization must occur in order to achieve compliance. Instead, according to Koh (1998, pp. 643–4), they come about through

Diane F. Frey 347

the actions and interactions among a set of transnational actors. The interactions unfold in four distinct phases: interaction, interpretation, internalization and obedience. These phases are reminiscent of the work of the ILO itself, especially in the processes of the ILO Committee on Freedom of Association and ILO Committee of Experts on the Application of Conventions and Recommendations (CEACR). The CEACR is responsible for monitoring compliance with ILO Conventions, including those on work-time limits and obligatory overtime. Interaction The interaction phase involves a provocation by domestic and transnational actors working to mobilize popular opinion and support for an objective by framing it as a moral issue beyond merely national particularistic politics (Koh 1998, p. 647; Nadelmann 1990, p. 482). For example, domestic trade unions and federations around the world work with ITUC in publishing the Annual Survey of Trade Union Rights Violations. The survey elevates attention to violations and seeks to mobilize support for the universal human rights of trade unionists. The audience for the Annual Survey is both international and domestic, including domestic government ministries responsible for protecting trade union rights. Domestic unions and ITUC have been called ‘norm entrepreneurs’ (Nadelmann 1990). Norm entrepreneurs also enlist actors within the government to support and act as allies, thereby becoming ‘governmental norm sponsors’ (Koh 1998, p. 648). As allies, the governmental norm sponsors are able to promote inside governmental forums the changes advocated by the norm entrepreneurs from the outside. Some governments, and intergovernmental agencies and actors within them, themselves become norm entrepreneurs. Drawing on work done by Sikkink, Koh explains that norm entrepreneurs and their governmental norm sponsors work together in developing transnational networks of mutual assistance and expertise (ibid., p. 649). Interpretation Furthermore, critical to Koh’s compliance process are interpretive communities defined by their competence to declare both general norms of international law (for example, treaties) and specific interpretations of those norms in particular circumstances. Through interpretation of norms in specific circumstances these communities are a ‘lawdeclaring fora’ and include private as well as public stages such as treaty regimes, domestic, regional, and international courts, ad hoc tribunals,

348 Regulating for Decent Work

domestic and regional legislatures, executive entities, commissions and non-governmental organizations. To be considered an ‘interpretive community’, an entity must be cable of ‘receiving a challenge to a nation’s conduct, then defining, elaborating, and testing the definitions of particular norms and opining about their violation’ (Koh 1998, p. 650). ILO’s CEACR is an interpretive community. It interprets the meaning and application of ILO Conventions in its process of monitoring and supervising a country’s compliance with the ILO Conventions it has ratified. The Committee examines reports submitted by governments as well as communications, or ‘shadow reports’ often from trade unions or other non-governmental organizations that challenge the government’s assertions of compliance with ILO Conventions. The CEACR publishes neither the government reports, nor communications from trade unions critical of government compliance. Instead, the Committee reports its findings to the annual International Labour Conference. In addition, the Committee reports its interpretations and findings in Observations and Individual Direct Requests referenced to specific countries. The Committee also undertakes an annual General Survey in which it examines country reports from ILO member states in respect of one specific labour standard which the country may or may not have ratified. Here again, the Committee does not publish the government report nor the shadow reports and instead submits its findings and interpretations of the labour standards to the Governing Body of the ILO. Internalization Koh argues that domestically based bureaucratic compliance procedures exist and respond to interpretive communities (Koh 1998, p. 651). Once a ruling has been made interpreting and applying an international rule to specific domestic circumstances, domestic governments adopt structures, operating procedures, and other mechanisms to help achieve and maintain habitual compliance with internalized international norms (ibid., p. 652). Koh further points out the long-standing existence of domestic and intergovernmental organizations with the responsibility to assure that domestic government policy conforms to the international legal standards that are ‘embedded in domestic law’ (ibid.). The interchange between international interpretations and domestic bureaucracy produces an enmeshing of the domestic with the international as the international law becomes entrenched in domestic legal and political processes (Keohane 1992 cited in Koh 1998). The process of seeking the enmeshing of domestic and international law is often seen in CEACR observations of domestic legal and political

Diane F. Frey 349

processes. For example, in a 2009 Individual Observation concerning Guatemala, the CEACR noted a court decision concerning payment of overtime that was incongruent with Guatemala’s obligations under Hours of Work (Industry) Convention, 1919 (No. 1) (ILO 2009, p. 1). The Observation also noted that a Guatemalan Tripartite Subcommittee on Legal Reforms would be discussing amendments to the Labour Code necessary to harmonize Guatemalan law with Convention No. 1 (ibid.). In continuing interaction with Guatemala, the CEACR requested that the government inform the Committee of the conclusions reached by the Tripartite Subcommittee. Finally, Koh (1998, pp. 653–4) argues that internalization is also promoted by creating linkages across distinct issue areas. In this way, non-compliance in one area is connected with frictions and repercussions in other issue areas. This, in turn, further reinforces the role of domestic bureaucracies in developing government conduct in compliance as a matter of default (David 1994 cited in Koh 1998 p. 654, FN 136). Koh argues that avoidance of such frictions provides domestic bureaucracies with strong incentives to embrace compliance generally as a matter of policy (ibid., pp. 654–5). This aspect of internalization is less evidenced by CEACR supervision. CEACR supervision is based on each Convention ratified by a government and is generally separate and distinct from other ILO Conventions and non-ILO treaty obligations. Obedience Koh (1998, p. 655) argues that through repeated patterns of interaction, interpretation and internalization compliance becomes habitual and self-interested rather than coerced and externally driven. Unfortunately, as many trade unionists and human rights activists know, habitual, selfinterested compliance does not always occur despite repeated interactions and interpretations. For example, the 2009 CEACR observation on Convention No. 1 in Guatemala ‘notes with regret’ that harmonization of Guatemalan domestic law with Convention No. 1 ‘has been raised for many years without any progress being noted’ (ILO 2009, p. 1). In other words, despite repeated interaction and interpretation, internalization does not occur.

Institutions and compliance theory as diagnostic methodology Assessments based on institutional and compliance theories can help identify and analyse continuing gaps between international legal norms

350 Regulating for Decent Work

and their implementation as domestic institutions. Koh (1998, p. 652) points out that the same actors and processes necessary for compliance can be active in resisting and preventing compliance. He argues, for example, that in the United States domestic bureaucracies exist to justify non-compliance with international legal norms as evidenced by recent US contentions that its treatment of prisoners conforms to its reinterpretation of the Geneva Conventions (ibid.). In this way, non-compliance may result either from the absence of processes and actors or with their presence and efforts to resist compliance. To make sense of noncompliance as well as compliance, it is helpful to return to institutional frameworks. Compliance within institutional arrangements, as with their establishment in the first place, represents a ‘compromise’ and is the result of conflict over disparate interests and power asymmetries (Amable 2003, p. 10; Knight 1992). Assessments based on institutional and compliance theories are not only useful for criticizing non-compliance and encouraging domestic internalization of international legal norms. The systematic comparison of domestic institutions to international legal norms can also help in identifying gaps and shortcomings in the international legal norms themselves. For example, it is possible to identify practices imposing obligatory overtime that are not currently addressed in the CEACR’s treatment of obligatory overtime and forced labour. Similarly, practices imposing obligatory overtime may be compliant with CEACR legal norms yet produce perverse outcomes that are counterproductive to achieving decent work for all. Systematic comparison of domestic institutions to international norms is a potential source of iterative improvement to both domestic institutions and the international norms. In sum, well-selected indicators can be extremely useful in measuring progress towards achieving decent work but invariably reduce the breadth and qualitative complexity of the information on which they are based. They must do so in order to reduce extensive and complex bodies of information to a limited group of decent work components. As a result, even well-chosen indicators may contribute little to identifying the underlying impediments to achieving decent work. Lee and McCann (2008) have noted the need to study the processes of observance and all of the factors that play a role in observance of labour law regulations. Without systematic identification and analysis of the obstacles, there is little diagnostic information upon which to base recommendations and debates for reform. Institutional and compliance theories each have contributions to make in assessing and improving compliance with labour market

Diane F. Frey 351

regulation but in order to do so they must be operationalized. This chapter next illustrates the application of the institutional and compliance frameworks as a diagnostic methodology to assess labour market regulation and practice holistically and to inform discussion of reform interventions. The overarching goal of the present diagnostic methodology is to buttress and complement decent work indicators with assessments to help to identify obstacles and inform debates about reforms needed to achieve decent work. Its objectives are: (i) to analyse compliance by modelling a methodology to both mirror Koh’s concepts and amplify the work done by the CEACR; (ii) to produce meaningful diagnostic categorizations and typologies of obstacles to effectively achieve observance of decent work-related regulation; and (iii) to identify possible interventions aimed at achieving decent work that could be the basis for policy agenda setting and substantive discussions. Table 13.1 shows the correlation between institutions and compliance frameworks and thereby provides the guidelines for this diagnostic methodology.

An illustration The chapter uses the example of work-time limits to illustrate the methodology for several reasons. First, the limitation of work-time is among the original concerns at the founding of the ILO and central to the enjoyment of life in, as well as beyond, the workplace. Work-time limits are also closely related to decent work since they are intended to provide ‘protection against undue fatigue and ensure reasonable leisure and opportunities for recreation and social life’ as well as promote health and safety (ILO 2005, para. 24). Limits on work time and obligatory overtime remain a source of tension between employers and workers and are linked to many other aspects of decent work, such as employment security and minimum wages (ibid.; McCann 2005, p. 12). Finally, work-time limits and obligatory overtime also demonstrate the porous boundaries between the ‘pillars’ of Decent Work and fundamental human rights. Obligatory overtime can, in some circumstances, amount to a violation of the Forced Labour Convention, 1930 (No. 29) (ILO 2007, para. 132, FN 314–16). The case of work-time limits and obligatory overtime in Honduras provides a good example to illustrate the many institutional combinations that prevent compliance with labour market regulations such as work-time limits. In the interactive and interpretive processes of the CEACR, reports and shadow reports are submitted to the Committee and

Obstacles

Description

Institutional theory explanation

Compliance theory intervention

Effect

Text

Rules are poorly written

Rules are ambiguous or contradictory – sanctions are ineffective

Legislative internalization of international rules

Align domestic rules and sanctions with international norms

Interpretation Well-written rules are interpreted poorly by courts

Defections from rules are formally supported and legitimized

Judicial internalization of international rules enacted in domestic law

Align interpretations of domestic rules and sanctions with international norms

Enforcement

Well-written and faithfully interpreted rules are poorly enforced

Defections from rules are tolerated becoming de facto social conventions

Political internalization of follow through on enforcement intended by international rules

Align enforcement of domestic rules with international norms

Social norms

Well-written and faithfully interpreted rules are undermined by non-supportive social norms

Formal institutions are undermined by informal institutions – Actors pursue socially-legitimate behaviours counter to formal institutions

Social internalization of legitimacy of norms underlying the rules

Align social norms with domestic rules by enhancing the normative legitimacy of the rules

Conventions

Well-written and faithfully interpreted and enforced rules are undermined by social conventions

Formal institutions are undermined by informal institutions – actors pursue ‘expected’ but not socially approved behaviours counter to formal institutions

Social internalization of legitimacy of norms underlying the rules

Align social conventions with domestic rules by enhancing the normative legitimacy of the rules

Other Institutions

Well-written and faithfully interpreted and enforced rules are undermined by other institutions

Formal institutions undermine (or reinforce) each other – institutional complementarities and hierarchies

Enmeshment and linking of issues ensuring institutional hierarchy in line with international rules

Align multiple institutions to create hierarchies intentionally reinforcing compliance outcomes

352

Table 13.1 An institutional approach to compliance

Diane F. Frey 353

form the basis of its interpretations and judgements about Honduras’ compliance with ILO Conventions. In the resulting CEACR Direct Requests and Individual Observations, the Committee often mentions allegations contained in shadow reports and government responses to them. Yet these interactive exchanges occur off-stage and remain off-stage since the CEACR does not publish the reports themselves. To mirror the process of the CEACR, the methodology includes a variety of perspectives on work-time limits in Honduras, including narrative documentary evidence from US State Department Human Rights reports and non-governmental organization reports from ITUC and the Asociación Servicios de Promoción Laboral (ASEPROLA). A central source of narrative evidence comes from Individual Direct Requests and Individual Observations of the CEACR. The use of these distinctive perspectives in documentary form also mirrors the interactions envisioned by Koh. The distinct perspectives in these documents represent at least some of the collective voices in debates over decent work. Ultimately, a transparent assessment based on these sources explicitly allows independent evaluators to identify where they disagree within the assessment methodology (Moran 2005). In the end, the goal is not a technical debate over decent work monitoring but a wider debate about how to involve all sectors of society toward progressively achieving decent work for all. The diagnostic assessment involves three interrelated steps. First, evidence of work-time limits and constraints on obligatory overtime is analysed in relation to institutional outcomes and arrangements – formal institutional components, social norms and conventions, and institutions from other realms of social and economic life. Second, the institutional outcomes and components are compared to compliance criteria derived from international legal norms such as ILO Convention No. 29. This allows for the identification of at least some of the complex and often multiple interactions and combinations of conditions and causal paths to the observance and non-observance of work-time limits (Ragin 1987). Third, using Koh’s compliance theory framework, it is possible to link institutional deficits with compliance interventions, or, alternatively, to potential gaps in the compliance criteria. Work-time limits and constraints on obligatory overtime in the ILO The imposition of extra hours outside of normal daily working hours, or ‘obligatory overtime’, is the subject of numerous complaints by Central

354 Regulating for Decent Work

American unions (ILO 2007, para. 132, FN 314–16). Work hours have historically been considered a matter of working conditions under numerous ILO Conventions, including the Hours of Work (Industry) Convention, 1919 (No. 1), and the Hours of Work (Commerce and Offices) Convention, 1930 (No. 30) (ILO 2007). The CEACR has clarified that obligatory overtime is not forced labour in violation of Convention No. 29 as long as it remains within nationally legislated limits or collectively bargained agreements (ILO 1998). When obligatory overtime remains within these limits, it amounts to ‘bad working conditions’ rather than forced labour (ILO 2007, para. 134). According to the CEACR, when obligatory overtime extends beyond legislatively or collectively bargained limits, further inquiry is needed to determine whether or not it is forced labour in violation of Convention No. 29 (ILO 2007). The CEACR identified two conditions in which obligatory overtime violates Convention No. 29. First, forced labour violations exist when workers work in excess of nationally established limits because they fear dismissal from their job. Second, when pay systems are based on productivity rather than time spent working, they violate Convention No. 29 if workers are obligated to work in excess of nationally established limits in order to earn the minimum wage. In these two circumstances, obligatory overtime is not merely a matter of poor working conditions; it is forced labour in violation of Convention No. 29 (ILO 2007). Based on this two-step definition, this section of the chapter examines evidence of forced obligatory overtime in Honduras. Through this examination, institutional patterns emerge that are matched with compliance interventions. Compliance interventions may be either legislative, based on problems with rules, or judicial, based on problems with interpretation and enforcement. Social compliance interventions are necessary when informal institutions interfere with the effectiveness of the formal rules. Political compliance interventions are necessary when political elites have not accepted the international norms and have not incorporated them into their domestic policy. Maquilas in Honduras According to the US State Department Human Rights reports, there are prohibitions against excessive compulsory overtime in Honduras (US State Department 2006, p. 11). Despite these rules, employers in Honduras are broadly depicted by both the US State Department and the ITUC as ignoring nationally established work-time limits. Obligatory overtime is among the working conditions that maquila workers complain

Diane F. Frey 355

about in Honduras and employers frequently ignore maximum workweek limits and prohibitions against compulsory overtime (ICFTU 2005, pp. 1–2; US State Department 2006, p. 11). Honduran workers who refuse to work extra hours are fired immediately. The ITUC provides an example of a group of maquila workers fired in a textile factory for having complained about working conditions including compulsory overtime (ASEPROLA 2004, p. 1; ICFTU 2005 p. 1). There is no evidence of obligatory overtime being used as a means to earn the minimum wage in Honduras. The effectiveness of formal institutions Obligatory overtime in Honduras results from multiple mechanisms associated with the formal institutional arrangements. Substantive rules establish ‘normal’ daily and weekly hours of eight and 44 respectively (ILO 2008, p. 37). These ‘normal hours’ do not, however, limit overtime, which is subject to separate rules and, by definition, is work performed in excess of the normal hours of work and is formally recognized as hours that are in excess of the statutory ‘maximum’ (ILO Database of Conditions of Work and Employment Laws (henceforth, ILO Database)).4 Work-time including overtime is limited to 12 hours per day and the workweek is limited to six days, making for a statutory maximum workweek of 72 hours (ILO 2008). These limits are not explicitly stated as rules and instead must be derived from calculations. Additionally, ‘normal’ hours do not limit work-time because there are other rules establishing categories of workers for whom the ‘normal’, ‘regular’ work-time limits do not apply. These exemptions are based on job titles such as manager, supervisor, domestic worker, driver and private chauffeur (ILO Database). Another exemption is established based on qualities of the job – such as ‘intermittent work’ for hairdressers and hotel staff. Finally, there is a more vaguely articulated category, which is work that ‘by its nature cannot be subject to the hours limits’ such as agriculture and stockbreeding. Maquila workers are not specifically listed in any of the exempt categories. It is unclear whether the government considers 44 or 72 hours to be the relevant limit. The lack of clarity in the rules is aggravated by complaint procedures established by the rules that are slow and often fail to result in any action by the Ministry of Labour (ILO Database; US State Department 2006, p. 11; ASEPROLA 2004, p. 25; ICFTU 2005, p. 1). In addition, even if an employer is found to be in violation, the substantive rules lack specific sanctions for violations (ASEPROLA 2004, p. 25). Unsurprisingly, ‘normal’ or ‘regular’ work-time limits are rarely enforced. In sum, what

356 Regulating for Decent Work

looks like a restrictive labour market regulation – namely the 44-hour workweek – is not the applicable regulation limiting work time in Honduras. Interactions with informal institutions In Honduras, harsh social conventions supplement ineffective rules and the non-enforcement of work-time limits. Employers fire workers for joining unions and complaining about working conditions, as well as for refusing to work overtime. Maquila employees, who are mostly women, are victims of physical and verbal abuse as well as mandatory pregnancy testing. Employers set extremely high volumes of work on strict deadlines and use production quotas. If the quota is not reached, the bonus is not paid, even if the employee worked overtime in an effort to meet the quota. Employers offer time off in lieu of overtime pay and work schedules are based on employer needs and the ups and downs of supply and demand (ICFTU 2005, pp. 1–2; ASEPROLA 2004, pp. 21–25). Employers and workers expect these behaviours, which makes them social conventions. Social norms also interact with formal institutions and reinforce obligatory overtime practices in Honduras. Neither business nor government embraces the normative legitimacy of work-time limits. Businesses in Honduras, as elsewhere in the region, maintain that the ups and downs they face in terms of production should take precedence over rules on work-time limits (ASEPROLA 2004, p. 20). Similarly, the government promotes the acceptance of obligatory overtime as a means to modernize and generate economic development leading to a culture of tolerance for labour rights violations within the government. These social norms help to crowd out opposing social norms that could promote the legitimacy of work-time limits. In sum, already inadequate formal institutions are further undermined and those who consider complaining about obligatory overtime lack normative legitimacy (ICFTU 2005, pp. 1–2; ASEPROLA 2004, pp. 20–5). Institutions in other realms of social and economic life In Honduras, wage-setting institutions contribute directly to workers’ low pay relative to the costs of basic needs. As a result, basic needs, such as food, cost three times more than workers earn for salaries. Institutions governing fiscal, monetary and trade policy also affect worker pay relative to basic needs, especially when they combine to lower the value of currency workers earn. The US State Department attributes the Honduran employers’ behaviour of ignoring overtime limits to high

Diane F. Frey 357

levels of unemployment and underemployment. The government’s economic development policy is based on making concessions to foreign investors in order to attract foreign investment. Among the concessions is tolerance for labour rights violations (ASEPROLA 2004, p. 25; US State Department 2006, p. 11). The patterns of interactions between these distinct sets of institutions raise the possibility that structural adjustment policy institutions are hierarchically superior to labour market regulation institutions. Institutions governing fiscal, monetary and trade policy arising from structural adjustment policy prescriptions increase worker poverty relative to basic needs, especially when they combine to lower the value of currency workers earn. Similarly government policy to attract foreign direct investment hierarchically displaces labour rights institutions simply by conceding to employers that the government tolerates labour rights violations. Notably, it is not necessary to reform laws to make structural adjustment policies hierarchically superior to labour rights because formal and informal institutions provide ample mechanisms to accomplish it (ASEPROLA 2004, p. 25). Compliance Evidence indicates that Honduran maquila employers force their employees to work beyond the nationally established 44-hour normal workweek and do so by threatening to fire them should they refuse. Further, there is evidence that employers do indeed fire workers for refusing to work obligatory overtime hours. In this case, maquila employers do not appear to be constrained by the 44-hour normal workweek. This situation cannot be characterized as decent work. Whether it is also forced labour requires further analysis. Obligatory overtime practices in Honduran maquilas violate CEACR norms prohibiting forced labour if the imposed work hours extend beyond national work-time limits. If the ‘regular’ 44-hour weekly work-time limit is considered such a ‘national work-time limit’, then employer threats and actual dismissals of workers for refusing to work beyond 44 hours rises to the level of forced labour. If, on the other hand, the national work-time limit is 72 hours, then compliance would depend upon whether the obligatory overtime and dismissal threats for refusing to work it are within or beyond the 72-hour limit. Inescapably, the fundamental right to be free from forced labour depends upon the establishment of national work-time limits. In the case of Honduras, formal institutions establishing national work-time limits and constraints on obligatory overtime are problematic

358 Regulating for Decent Work

not just because they are unenforced. Instead the formal institutional rules meant ostensibly to prevent forced labour and encourage decent work have an opposite and detrimental effect. To address this incongruence between Honduran formal institutions and the compliance criteria established by the ILO, several compliance interventions are necessary. First, legislative interventions are necessary to clarify whether the 44- or 72-hour weekly limit is the relevant national work-time limit for maquila workers. In line with compliance theory, the national worktime limit should comply with ILO standards. Second, substantive rules should be assessed and amended based on whether their grammar is clear and congruent and whether their sanctions are sufficiently dissuasive and persuasive. Third, procedures need to be assessed based on whether they sufficiently increase the likelihood of potential rewards and or penalties for employers based on their observance of the substantive rules. For example, a procedural rule could establish a default that if the employer does not appear in court to contest the worker’s claim, the court will make a determination based only on the evidence given by the worker. This procedural default counters a common social convention in which employers delay proceedings simply by not participating. Fourth, substantive rules, procedures sanctions and rewards are also necessary to address the dismissal of workers who lawfully refuse to work obligatory overtime beyond nationally established limits. For example, in addition to payment of overtime wages and reinstatement of the worker, an employer could face additional penalties, in effect contributing additional resources for the Labour Ministry to monitor their compliance with work-time limits. Finally, legislative interventions are also necessary to create or enhance rules that challenge other social conventions, such as physical or verbal abuse and production quota systems that enable employers to avoid overtime rules. These legislative interventions could, for example, reach across areas of regulation enmeshing labour regulation with rules about obtaining licenses to export, receive tax incentives and remain incorporated. In sum, considering the myriad reasons that actors have to comply with rules, compliance interventions help guide questions for discussion among social partners such as, ‘Do our formal institutions make compliance more or less likely based on coercion, self-interest, and legitimacy?’ Maquila operators work in a very competitive industry and formal institutions currently reward operators who rely on obligatory overtime as part of their competitive strategy. Yet it is possible to envision

Diane F. Frey 359

formal institutions that would make compliance with work-time limits a winning rather than losing strategy. For example, if violations were met with costly sanctions and compliance positively rewarded, competition among operators would continue but reliance on obligatory overtime would become a losing strategy. In effect, obligatory overtime would be taken out of competition among operators without disadvantaging them. Judicial interventions are necessary to effectively implement these new substantive and procedural rules and sanctions through judicial interpretation and enforcement. When a court decision appears to be incongruent with international law, social partners could be empowered to engage in analysis and discussion of the decision and to identify further legislative interventions. Social interventions are necessary to enhance the norms underlying the legitimacy of work-time and overtime limits and to challenge and replace tolerance for labour rights violations with norms consistent with respect for labour rights. Social interventions include communications from courts, legislatures, labour ministry personnel and employers in which they name, discuss and reinforce the underlying intention and value of the rule itself. Taken together, legislative, judicial and social compliance interventions implicate a larger discussion about the fundamental political internalization of international norms on work-time limits and obligatory overtime. Debates about political internalization can be informed by analysis of the efficacy of actual legislative, judicial and social interventions. Debates about compliance can then centre on whether domestic institutions are aligned with international legal norms rather than relying simply on how well a government articulates its commitment. Political internalization is also evidenced by government activities to champion or undermine the normative legitimacy of work-time and overtime limits, as well as in its own observance of work-time limits in its role as employer. Lastly, political internalization is demonstrated by a government’s commitment to and effectiveness in confronting obstacles to observance of work-time limits that arise in other areas of social and economic life. The analysis of obligatory overtime in Honduran maquilas demonstrates the tangle of multiple institutional interactions that support forced obligatory overtime. In this light, it is questionable whether compliance interventions narrowly aimed at work-time and overtime institutions alone can bring about institutions more fully compliant with Convention No. 29. Instead, compliance intervention strategies must also challenge institutional influences from other spheres of social life that

360 Regulating for Decent Work

undermine compliance, including minimum-wage setting and economic development institutions. As Koh argues, in these circumstances, multiple and mutually reinforcing compliance interventions are necessary. Moreover, compliance interventions must enhance negative and positive incentives across institutions. Discussion of compliance of Honduran work-time and obligatory overtime institutions with international legal norms also helps to frame some of the shortcomings, ambiguities and gaps in the international legal norms. First, it is disturbing that a 72-hour weekly work-time limit could be considered consistent with the enjoyment of life and providing ‘protection against undue fatigue and ensuring reasonable leisure and opportunities for recreation and social life’ as well as promoting health and safety (ILO 2005, para. 24).

Conclusion The diagnostic methodology presented in this chapter aims to enhance the assessment of the obstacles to achieving observance of labour market regulations essential to decent work and fundamental human rights. The examination of work-time limits and constraints on obligatory overtime as institutions helps to guide analysis of the influence of formal institutional arrangements such as rules, their enforcement and interpretation. At the same time, it acknowledges that formal institutions alone cannot explain observance of work-time limits and constraints on obligatory overtime. Observance of work-time limits also depends upon informal social norms and social conventions as well as institutional influences from other realms of social and economic life. Finally, in addition to the necessity of considering all the three dimensions of institutions to understand institutional outcomes, we need to be mindful that power is necessary to establish and maintain institutions. Institutional patterns associated with non-observance of work-time limits can, in turn, be matched to Koh’s compliance interventions. Critically, this compliance assessment is anchored in authoritative interpretations of international legal norms, not policy goals. Compliance problems may include rules, interpretation and enforcement, social conventions and social norms, as well as the combined effects of multiple institutions. Compliance deficits, once identified, do not lead to specific, easily identifiable interventions. Instead they lead back to how institutions are formed and changed in the first place, by interaction, conflict and power. Identifying compliance deficits can, however, frame these interactions and conflicts in a context of international legal norms and

Diane F. Frey 361

create an agenda for discussion and consensus building among social partners. Institutional frameworks and compliance theory also help move discussion of work-time limits and obligatory overtime beyond its articulated commitment through ILO Convention ratification alone. Using institutional frameworks and compliance theory focuses attention and discussion on how domestic institutions actually align with international obligations. Failure of governments to observe and to address incongruities between domestic institutions and international legal norms raise legitimate questions about whether or not a government has politically internalized the commitments they articulate. The examination also brings to the fore a number of problems with the CEACR compliance criteria. First, the narrow CEACR definition of forced obligatory overtime creates perverse outcomes. Within nationally established work-time limits, threats of dismissal and overtime as a means of earning the minimum wage are permitted. These rules do not appear to violate the norms established by the CEACR. There is considerable ambiguity about what constitutes reasonable national worktime limits and to whom they apply. Moreover, domestic workers, who have the longest work hours, are entirely exempt from work-time limits. The institutional and compliance analysis presented here is based on ILO, US State Department and NGO documents. They are imperfect, incomplete and invariably political and controversial. Since the CEACR process does not allow us a window on the conflicting perspectives and facts in its monitoring processes, the methodology works to mirror the interactions and perspectives that occur within the CEACR process and to amplify and extend the CEACR interpretive work establishing compliance criteria. In line with Koh’s compliance processes, identifying compliance deficits helps to create interactions in which actors can challenge a government’s claim to be compliant with an international norm. Similarly, analysis of compliance based on imperfect and incomplete information requires caution and openness to correction. For example, it is possible that obligatory overtime occurs in Honduras as a means of earning the minimum wage despite the fact that it is not present in the evidence examined. In this light, and consistent with Koh’s work, the findings presented, based on the evidence, are intentionally offered to provoke correction and amendment in an iterative process of interaction and interpretation. This is precisely the kind of engagement that the CEACR seeks with governments and social partners in dialogue over how to improve compliance with ILO Conventions. Koh’s approach to

362 Regulating for Decent Work

internalized obedience to international norms also involves an iterative process between the different actors involved at international, domestic and local levels. These interactions comparing and justifying institutions with international legal norms are necessary to bring about real and lasting protection of labour rights and to achieve decent work for all.

Notes 1. For details, see http://www.ilo.org/integration/themes/mdw/lang—en/index. htm. 2. ITUC was formerly called the International Confederation of Free Trade Unions (ICFTU). 3. Koh terms this ‘legal internalization’. 4. http://www.ilo.org/dyn/travail/travmain.home.

References Amable, B. 2003. The Diversity of Modern Capitalism (Oxford: Oxford University Press). Asociación Servicios de Promoción Laboral (ASEPROLA). 2004. ‘An examination of six basic labor rights in Honduras’, no. 3, San José, Costa Rica. Becker, G.S. 1968. ‘Crime and punishment: an economic approach’, Journal of Political Economy, vol. 76, no. 2, pp. 169–217. Berg, J.; Kucera, D. (eds). 2008. In Defence of Labour Market Institutions: Cultivating Justice in the Developing World (New York: Palgrave Macmillan and ILO). Browne, J.; Deakin, S.; Wilkerson, F. 2002. ‘Capabilities, social rights and European market integration’, ESRC Centre for Business Research Working Paper 253, University of Cambridge. Committee on Economic, Social and Cultural Rights (CESCR) 2006. ‘General Comment 18, The right to work’, ICESCR, UN Doc. E/C.12/GC/18. Crawford, S.E.S.; Ostrom, E. 1995. ‘A grammar of institutions’, American Political Science Review, vol. 89, no. 3, pp. 582–600. David, P.A. 1994. ‘Why are institutions the “Carriers of History”? path dependence and the evolution of conventions, organizations and institutions’, Structural Change and Economic Dynamics, vol. 5, pp. 205–20. Fisher, R. 1981. Improving Compliance with International Law (Charlottesville, VA: University Press of Virginia). Frey, D.F. 2010. ‘An institutional approach to labor-related human rights compliance: a case of forced labor in Nicaragua and Honduras’, in Lewin, D., Kaufman, B.E. and Gollan, P.J (eds) Advances in Industrial and Labor Relations (AILR), vol. 17 (Bingley: Emerald Group Publishing Limited). Hall, P.; Soskice, D. 2001. Varieties of Capitalism: The Institutional Foundations of Comparative Advantage (Oxford: Oxford University Press). Helmke, G.; Levitsky, S. (eds) (2006) Informal Institutions and Democracy: Lessons from Latin America (Baltimore: John Hopkins University Press). International Confederation of Free Trade Unions (ICFTU). 2005. ‘Reports on core labour standards in the countries applying for the GSP-Plus’. Available at www.icftu.org/www/PDF/GSP.pdf.

Diane F. Frey 363 International Covenant on Economic, Social and Cultural Rights (ICESCR). 1966. Available at http://www2.ohchr.org/english/law/cescr.htm. International Labour Office (ILO). 1998. Report of the Committee of Experts on the Application of Conventions and Recommendations: General Report and Observations Concerning Particular Countries, Report III (Part 1A), International Labour Conference, 86th Session, Geneva. ILO. 1999. Decent Work, Report of the Director-General, 87th Session, International Labour Conference, Geneva. ILO. 2005. Hours of Work: Fixed or Flexible, Report of the Committee of Experts on the Application of Conventions and Recommendations: General Survey concerning the Hours of Work (Industry) Convention, 1919 (No. 1, and the Hours of Work (Commerce and Offices) Convention, 1930 (No. 30), Report III (Part 1B), International Labour Conference, 93rd Session, Geneva. ILO. 2007. Eradication of Forced Labour, Report of the Committee of Experts on the Application of Conventions and Recommendations: General Survey concerning the Forced Labour Convention, 1930 (No. 29), and the Abolition of Forced Labour Convention, 1957 (No. 105), Report III (Part 1B), International Labour Conference, 95th Session, Geneva. ILO. 2008. Working Conditions Laws 2006–2007: A Global Review (Geneva: ILO). ILO. 2009. ‘Individual Observation concerning Hours of Work (Industry) Convention, 1919 (No. 1): Guatemala’, in Report of the Committee of Experts on the Application of Conventions and Recommendations, Report III (Part 1A), International Labour Conference, 97th Session, Geneva. Keohane, R.O. 1992. ‘Compliance with international commitments: politics within a framework of law’, 86th Annual Meeting, American Society of International Law Proceedings 176 (2, 7, 9). Knight, J. 1992. Institutions and Social Conflict (Cambridge: Cambridge University Press). Koh, H. 1997. ‘Why do nations obey international law?’, Yale Law Journal, vol. 106, pp. 2599–659. Koh, H. 1998. ‘The 1998 Frankel lecture: bringing international law home’, Houston Law Review, vol. 35, pp. 623–81. Lee, S.; McCann, D. 2008 ‘Measuring labour market institutions: conceptual and methodological questions on working hours rigidity’, in Berg, J.; Kucera, D. (eds), In Defence of Labour Market Institutions: Cultivating Justice in the Developing World (London and Geneva: Palgrave Macmillan and ILO). Lukes, S. 2005. Power: A Radical View, 2nd edn (New York: Palgrave Macmillan). McCann, D. 2005. ‘The role of work/family discourse in strengthening traditional working time laws: some lessons from the on-call debate’, Law in Context, vol. 23, no. 1, pp. 127–47. Mershon, C. A. 1994. ‘Expectations and informal rules in coalition formation’, Comparative Political Studies, vol. 27, no. 1, pp. 40–79. Moran, T.H. 2005. ‘Monitoring compliance with international labor standards: how can the process be improved, and what are the implications for inserting labor standards into the WTO?’, Journal of Business Ethics, vol. 59, nos 1–2, pp. 147–53. Nadelmann, E.A. (1990) ‘Global prohibition regimes: the evolution of norms in international society’, International Organisations, vol. 44, no. 4, pp. 479–526. North, D.C. 1990. Institutions, Institutional Change and Economic Performance (Cambridge: Cambridge University Press).

364 Regulating for Decent Work Ostrom, E. 2005. Understanding Institutional Diversity (Princeton, NJ: Princeton University Press). Ragin, C.C. 1987. The Comparative Method: Moving Beyond Qualitative and Quantitative Strategies (Berkeley, CA: University of California Press). Rihoux, B. 2006. ‘Qualitative comparative analysis (QCA) and related systematic comparative methods: recent advances and remaining challenges for social science research’, International Sociology, vol. 21, no. 5, pp. 679–706. Scholz, J.T. 1997. ‘Enforcement policy and corporate misconduct: the changing perspective of deterrence theory’, Law and Contemporary Problems, vol. 60, no. 3, pp. 253–68. US State Department Human Rights Report 2006. Honduras, available at http:// www.state.gov/g/drl/rls/hrrpt/2006/78896.htm.

Index active transfers 68 Agricultural Workers Alliance (AWA) 76 All-China Federation of Trade Unions (ACFTU) 195 Annual Statistics on Hours and Earnings 119 Annual Survey of Trade Union Rights Violations 347 apprenticeships 95, 96, 147 arbitration 187–8, 193–4 Asian Development Bank (ADB) 167 atypical work 7, 62, 256 Australia collective bargaining 253 gender pay gap 253 low pay 254 female workers 231 minimum wage 253 part-time workers 64 Austria collective bargaining 253 gender pay gap 253 gender pay ratio 233 unemployment benefit 236 autocorrelation 44 autonomous workers 66 autoregressive distributed-lag approach 44

Bolsa Família (BF) programme 133 Caixa Econômica Federal 324 carteira assinada 127, 130, 276, 278 Consolidation of Labour Laws 140 Department of Labour Inspection 319, 321 domestic work 140–2, 263–4, 276–9 Equipe de Combate à Fraude nas Relações de Emprego (ECOFREM) 327–9 employment growth 129 exchange rate 131 fertility rate 134 formal employment 128, 130 improved labour inspection 137–40 improvements in 142–3 increased worker demand 131–4 legal awareness 140–2 reduced labour supply 134–5 sustainability of conditions 143–5 fraudulent cooperatives 325–9 Fundo de Garantia Por Tempo de Serviço 125, 321–5 GDP growth 132 Gini coefficient 132–3 Growth Acceleration Programme (PAC) 144 illegal subcontracting 325–9 Individual Entrepreneur Law 143, 145 labour card see carteira assinada labour market flexibility 124–6 reform 125 trends 126–30 Mercosur Customs Union 124 minimum wage 127

behavioural economics 34, 53 Belgium collective bargaining 253 domestic work 263, 265, 266, 268, 272 gender pay gap 253 minimum wage 253 Titre-Service 263, 265, 266, 268 unemployment benefit 236 beyond employment approach 59, 67, 69–72, 79, 82 Brazil 123–50 bank of hours 125 365

366

Index

Brazil – continued Ministério Público do Trabalho 327, 329 Programa Trabalho Domestico Ciudadao 277 RAIS (Relação Anual de Informações Sociais) 147 self-employment 128 SIMPLES law 130, 135–7, 145 social assistance program (Benefíco de Prestação Continuada) 133 social security 127, 128 youth employment 134–5 Broken Ladders (Osterman) 92 bureaucratic discretion 315–19 fear of 315–16 limitation of 316 Cambridge Centre for Business Research (CCBR) index 40–1, 44, 51–2 Canada collective bargaining 253 Employment Protection for Foreign Nationals (Live-in Caregivers and Others) Act 77 gender pay gap 253 low pay for female workers 231 minimum wage 253 self-employment 64–5 work-sharing 210–11 care workers see domestic work career breaks 80–1 career tracks 96–8 Chèque Emploi Service Universel (CESU) 263, 266, 268–9 Chèque Service (CS) 263, 267, 269 cheque-system for domestic work 264–76 Chèque Emploi Service Universel (CESU) 263, 266, 268–9 Chèque Service (CS) 263, 267, 269 fiscal and social incentives 269–70 impact on employment quality 271–4 social security contributions 265, 268–9 Titre-Service 263, 265, 266, 268

childcare 260, 266–7 affordability of 239 domestic workers and 274, 276 gender and 234, 240 public 67, 78–9, 269, 271 see also domestic work China 180–205 All-China Federation of Trade Unions (ACFTU) 195 arbitration 187–8, 193–4 consultancy firms 199 deposit for bondage 187 Detailed Regulations law 197–8 employer pressure groups 195–6 employment agencies 191, 196–8 employment contracts 191–2 Employment Promotion Law (EPL) 181, 183–5 elements of 184 implementation 189–90 foreign client firms 198–9, 202 human resource outsourcing 199 informal employment 186 institutional actors 190, 195–9 Labour Contract Law (LCL) 181, 182, 185–7 elements of 185, 187 implementation 190–2 Labour Dispute Mediation and Arbitration Law (LDMAL) 181, 182, 187–9 elements of 185, 187–8 implementation 192–5 One Arbitration, Final Ruling clause 187–8, 193 labour disputes 186 Labour Law 181 legislative environment 182 mediation 188, 193 NGOs 198–9 social security system 192 trade unions 201–2 China Enterprise Confederation–China Enterprise Directors Association (CEC–CEDA) 195–6 citizenship 6, 58 boundaries 60–2, 67, 68–9 inclusive 73, 75–8 supranational 71

Index 367 collective bargaining 152, 166, 201, 211, 226, 228–30, 232–3, 243, 245, 248 by country 253 Committee on Economic Social and Cultural Rights (CEACR) 344, 351, 353, 361 competitive model, limitations of 155–9 compliance theory 339, 340, 344–9 as diagnostic methodology 349–51 Honduras 357–60 institutional approach 352 interaction 347 internalization 348–9 interpretation 347–8 obedience 349 consultancy firms 199 Conventions see International Labour Organization; domestic workers coordinated market economics 37 corporate social responsibility 198, 293 cost-push hypothesis 152, 154, 167, 172 Czech Republic collective bargaining 253 gender pay gap 253 low pay 254 female workers 231 minimum wage 253 decent work 1, 3–5, 8, 350, 351, 353, 362 as goal of 114 for domestic workers 255–87 indicators 339, 351 regulating for 1, 59, 72–5, 82, 339, 340, 344–5, 357–8, 360 unstructured occupations and 115–16 see also individual countries; Decent Work Agenda; Regulating for Decent Work Network Decent Work Agenda 4, 256, 297, 339 demand-side organization 241–4 Deng Xiaoping 180 Denmark collective bargaining 253

gender pay gap 253 gender pay ratio 233 low pay 254 female workers 231 unemployment benefit 236 deposit for bondage 187 deregulation 180 narrative 1–3 reforms 51 developing countries domestic work 259 economic/social development 198 efficiency wages 156–7 export growth 238 industrialization of 123 labour laws 291–312 loss of health personnel 82 diagnostic methodology 339–64 compliance theory see compliance theory institutional theory 92, 341–4, 349–51 difference-in-differences approach 34, 43, 46, 52 discretion in decision-making 315–19 domestic work 140–2, 255–87 abuse of workers 260 affordability 263 Belgium 263, 265, 266, 268, 272 Brazil 140–2, 263–4, 276–9 cheque-system 264–76 features of 266–7 fiscal and social incentives 269–70 impact on employment quality 271–4 social security contributions 265, 268–9 Convention 81 developing countries 259 employment status of workers 259 France 263, 266, 268–9, 272–4 heterogeneity and specificity 258–61 industrialization of 257, 262–3, 280–1 informality among workers 274–5 living in 260

368

Index

domestic work – continued occupations 256–7, 258–9 professionalization of 275–6 shared features 259–60 state-supported use 261–4 Switzerland 263, 267, 269 undervaluing of 261, 282 see also childcare earnings gap 107–9 gender-related 15, 164, 227, 228–33 elementary occupations 232–3 low-wage work 230–2 lowest decile 229–30 see also gender pay gap Economic Freedom index 166, 167 efficiency wage 156–7 employability 11, 12 employer pressure groups 195–6 employer response 20 Employing Workers Index (EWI) 2, 3, 18, 39, 295–6 employment domestic see domestic work informal China 186 Indonesia 162–4 low-productivity 157 precarious 58–90 project-based 92–3, 98, 99, 102, 113, 116 social value of 11 structured occupations 103–6 tournament (unstructured) occupations 103–6 transient 92–3, 98, 99, 113 see also entries under job categories employment agencies 191, 196–8 employment contracts 191–2 employment ensurance 70 employment growth 129 employment protection legislation (EPL) 39, 42, 49–52 see also EPL index; EPL-binding industries Employment Relationship Recommendation, 2006 (No. 198) 63, 64 employment stabilization 216–18

endogenous laws 35 enforcement 342 entrepreneurship 14 entry ports 95 entry tournaments 6, 93–102 client benefits 97–102 EPL see employment protection legislation (EPL) EPL index 39–40 EPL-binding industries 49 European Union 6, 59, 190 citizenship 71 Directive Concerning the Status of Third-Country Nationals who are Long-Term Residents (2003) 84 Directive on Fixed Term Work (1999/70/EC) 63, 83 Directive on Temporary Agency Work (2008) 64, 83 third country nationals 71 European Working Conditions Survey 298 EWI see Employing Workers Index (EWI) exchange rates in Brazil 124, 130–2, 144 experimentalist governance 317–18 FDI see foreign direct investment (FDI) female workers low pay 67, 231–2, 254 migrant care workers 72, 78 unemployment benefit 236 wage inequality see gender inequality; gender pay gap film industry, British 100–1 Finland collective bargaining 253 gender pay gap 253 gender pay ratio 233 low pay 254 female workers 231 unemployment benefit 236 firing costs index 39 flexicurity 21, 92 Fordism 92 foreign client firms 198–9, 202

Index 369 foreign direct investment (FDI) Honduras 357 Indonesia 166 formal employment Brazil 128, 130–42 improved labour inspection 137–40 improvements in 142–3 increased worker demand 131–4 legal awareness 140–2 reduced labour supply 134–5 sustainability of conditions 143–5 formal institutions 341–2 Honduras 355–6 France Chèque Emploi Service Universel (CESU) 263, 266, 268–9 collective bargaining 253 domestic work 263, 266, 268–9, 272–4 gender pay gap 253 gender pay ratio 233 labour laws 41 low pay 254 female workers 231 minimum wage 45, 253 unemployment benefit 236 fraudulent cooperatives 325–9 furloughs 216 GDP see gross domestic product (GDP) gender contract 60–2 gender inequality 226–52 job mobility 240–1 labour market outsiders 239–40 law of one price 227–8 and political conflict 238 and reservation wage 234–9 wage setting 242–4 gender pay gap 164, 227, 228–33 by country 253 elementary occupations 232–3 low-wage work 230–2, 254 lowest decile 229–30 minimum wage 244–9

unemployment benefit eligibility 236 see also earnings gap, gender-related gender pay ratio 232–3 gender-neutral labour laws 61 gendered poverty 74 General Agreement on Trade in Services (GATS) 77 Germany collective bargaining 253 gender pay gap 253 gender pay ratio 233 labour laws 41 low pay 254 female workers 231 Placement and Unemployment Insurance Act (1927) 210 unemployment benefit 236 work-sharing 210 Gini coefficient of Brazil 132–3 global labour market membership 59 conditions for 72–5 globalization 11 Granger causality 44 Greece, unemployment benefit 236 gross domestic product (GDP) Brazil 132, 133 Indonesia 166, 173 Home Work Convention, 1996 (No. 177) 256 Honduras Asociación Servicios de Promoción Laboral 353 compliance 357–60 formal institutions 355–6 informal institutions 356 maquilas 354–5 wage-setting institutions 356–7 work-time limits 351–3 Hours of Work (Commerce and Offices) Convention, 1930 (No. 30) 354 Hours of Work (Industry) Convention, 1919 (No. 1) 61 Hu Jintao 180 human resource management 180 human resource outsourcing 199

370

Index

Hungary collective bargaining 253 gender pay gap 253 low pay 254 female workers 231 minimum wage 253 illegal subcontracting 325–9 ILO see International Labour Organization (ILO) inclusive citizenship 73 policy options 75–8 indeterminacy, regulatory 17–23 India, labour laws 41 Indonesia 151–81 business regulations 171–2 compliance 158, 159 Economic Freedom index 166, 167 efficiency wage 156–7 employment indices 165 employment rigidity 160 FDI 166 global competitiveness index 165 industrial relations 161 labour intensive manufacturing, export share 173 labour market flexibility 153–5 performance 154 labour market regulations 159–73 and informal employment 162–4 and wage inequality 164 labour market regulations index 166 labour problems 168 Manpower Act (2003) 153, 155, 160, 162, 170, 175 Medium-Term Development Plan 154–5 minimum wage 153, 156, 158, 160, 169–70 paternalistic authoritarianism 153 poverty 157, 161–2 unemployment rate 163 unit labour costs 169 wage-employment trade-off 158–9 workers’ welfare 159–73

industrial relations China 183 Indonesia 161 inflation shock 162 informal economy 22 informal employment China 186 Indonesia 162–4 informal institutions 343 Honduras 356 innovation 14 institutional theory 92, 341–4 as diagnostic methodology 349–51 formal 341–2, 343 informal 343 market-orientated 344 power and contention in 344 see also labour laws institutional variables 38–45 interaction 347 internal labour markets 91, 94 internalization 348–9 judicial 346 legislative 345 political 346, 359 social 346 International Covenant on Economic, Social and Cultural Rights 345 International Development Association 2 International Labour Conference 255 International Labour Organization (ILO) 61, 294, 297–8, 334, 339, 342, 347–9, 358, 361 Committee of Experts on the Application of Conventions and Recommendations 340 Database of Conditions of Work and Employment Laws 355 Decent Work Agenda 4, 255–6, 297, 339, 344–5 Employment Relationship Recommendation, 2006 (No. 198) 63, 64 Home Work Convention, 1996 (No. 177) 256 Hours of Work (Commerce and Offices) Convention, 1930 (No. 30) 354

Index 371 Hours of Work (Industry) Convention, 1919 (No. 1) 61 Maritime Labour Convention (2006) 81 Multilateral Framework on Labour Migration 78 Part-Time Work Convention, 1994 (No. 175) 63, 256 standard on domestic workers 7, 83, 284 standards 59, 62–3, 358 work-time limits 351, 353–4 International Tax Organization 81, 85 International Trade Union Confederation 341 interpretation 342, 347–8 Ireland collective bargaining 253 gender pay gap 253 gender pay ratio 233 low pay 254 female workers 231 minimum wage 253 unemployment benefit 236 Italy collective bargaining 253 gender pay gap 253 gender pay ratio 233 low pay 254 female workers 231 unemployment benefit 236 Window for the Jobless 78 Japan collective bargaining 253 gender pay gap 253 low pay 254 female workers 231 minimum wage 253 job attachment 217 job changing 111 job creation 13, 152 domestic work 271 job mobility 234 and gender 240–1 job protection 13, 157 see also employment protection legislation (EPL) jobless recovery 209

journalism 99–100 judicial internalization 346 judicial interventions 359 Koh, Harold 339, 340 Korea low pay 254 female workers 231 Labour Force Survey 102, 118 labour index 39–40 labour inspection 137–40, 313–38 Brazilian FGTS 125, 321–5 bureaucratic discretion 315–19, 320–1 fraudulent cooperatives 325–9 illegal subcontracting 325–9 models of 319–20 work routines and outcomes 320–1 labour laws 8–23, 41 awareness of 299–307 importance of 304–6 improvements in working conditions 306–7 levels and variations 301–4 as beneficial constraints 12–14 benefits 8–16 case against 152 case for 31–57 China 180–3, 200–2 advisors on 199 employers’ pressure groups 195–6 employment agencies 196–8 Employment Protection (EPL) 183–5, 189–90 Labour Contact Law (LCL) 185–7, 190–2 Labour Disputes Mediation and Arbitration Law (LDMAL) 187–9, 192–5 monitoring of 198–9 compliance see compliance theory cross-regional variation 40–1 developing countries 291–312 diagnostic methodology 339–64 economic and social impact 32–8 endogenous 35, 42 enforcement 342

372

Index

labour laws – continued gender-neutral 61 imperfections in 33, 34 indeterminacy 17–23, 36 indices of 18–19 Indonesia 159–73 and informal employment 162–4 interpretation 342 low-income settings 292–7 maternity protection 61, 302, 304, 305 minimum wage see minimum wage moral hazard 12 research and policy 19–23 as second-best instrument 10–12 and social values 14–16 Tanzania 291–2, 297–309 underregulation bias 16–17 and wage inequality 164 working time 13, 21, 302, 304, 305 labour markets Brazil 123–50 decline of 91–118 flexibility see labour market flexibility insecurity 62 internal 92, 95 regulation see labour laws labour market equality 220–3 labour market flexibility 12, 39 Brazil 124–6 Indonesia 153–5 limitations of competitive model 155–9 USA 209–10 labour market institutions see institutional theory labour market outsiders 239–40 labour market regulations see labour laws labour regulation see labour laws labour supply, social structuring of 234–41 labour unions see trade unions last-in, first-out policy 222 law of one price 227–8, 243 legal effectiveness 19, 22 legal origins hypothesis 31, 36, 38–9, 42

legal rules 23, 31, 35–6, 37, 38, 42–3 legal variables 38–45 legislative internalization 345 legislative interventions 358 Lerner, Abba 219 liberal market economics 37, 38 life-course support 73, 74, 78–82 longitudinal data 43 low pay 66 by country 254 by gender 254 female workers 66, 231–2 and job changing 111, 114 Low Pay Commission 47, 53–4 low-wage work, gender pay gap 230–2 Luxembourg, unemployment benefit 236 male workers low pay 254 unemployment benefits 236 management consulting 102 mandatory shutdowns 216 Maritime Labour Convention (2006) 81 maternity protection 61 awareness of 302, 304, 305 media 100–1 mediation 188, 193 Mercosur Customs Union 124 Millennium Development Goals (MDGs) 339 minimum wage 237–8 awareness of 302, 304, 305 by country 253 Brazil 127 Indonesia 153, 156, 158, 160, 169–70 gender equity 244–9 minimum wage legislation 2, 18, 31 adverse effects 7, 13, 33, 34–5 analysis of 15, 33 employment protections and 45–9 gender discrimination and 15, 21, 61, 226 labour market impacts of 13 monopsony and 9 productivity and 51–2

Index 373 self-employment and 66 see also under individual countries monoposonistic employers 156, 177, 226 monopsonistic practices 174 monopsony 234, 240–1 ‘dynamic’ 34 employer 46–8 gender and job mobility 240–1 minimum wages and 9 see also monopsonistic employers; monopsonistic practices; monopsony effects monopsony effects 43 moral hazard 12 Most Favoured Nation status 77 Multilateral Framework on Labour Migration 78 multinational corporations 198 National Minimum Wage, UK 20, 47, 239, 243, 244 Netherlands collective bargaining 253 gender pay gap 253 gender pay ratio 233 low pay 254 female workers 231 minimum wage 253 New Earnings Survey 103, 110, 119 new public management 316–17, 319, 330 New Zealand collective bargaining 253 gender pay gap 253 low pay 254 female workers 231 minimum wage 253 NGOs see non-government organizations (NGOs) non-government organizations (NGOs) in China 182, 198–9 non-wage labour costs index 39 norm entrepreneurs 347 obedience 349 obligatory overtime constraints on 353–4 Honduras 359–60

occupational labour markets 92, 95 occupational licensing 102 OECD see Organisation for Economic Co-operation and Development (OECD) Office for National Statistics 118 Organisation for Economic Co-operation and Development (OECD) Employment Outlook (2004) 42 employment protection legislation (EPL) index 40–1 Jobs Study 49 oversight 316 paid parental leave 79, 85 panel data cointegration 44 parasitic workers 237 parental leave 237 paid 79, 85 part-time employment 65, 103, 209 by choice 215 changes in 214–15 slack work 215 social costs 217 see also work-sharing part-time unemployment benefit see short-time compensation Part-Time Work Convention (1994) 63, 256 path dependence 172 pay differentials see earnings gap pay inequalities 106, 107–8 Philippine Overseas Labour Office 77 Poland collective bargaining 253 gender pay gap 253 low pay 254 female workers 231 minimum wage 253 political internalization 346, 359 Portugal, unemployment benefit 236 positional goods 99 poverty, gendered 74 precarious employment 58–90

374

Index

project-based employment 92, 93–4, 98–9, 102, 113 protected mobility 74 public–private partnerships 317 Regulating for Decent Work Network 1, 3–5 regulatory indeterminacy 17–23 rent sharing 10–12 rent-seeking 10, 33 reregulation 180 reservation wage 234–41 and gender 234–9 reverse causation 42 rigidity of employment index 39, 165 risk sharing 10–12 rules of the game 339 Schroeder, Patricia 219 self-directed learning 94 self-employment 12, 58, 63, 64, 65–6, 69, 83, 259, 299 Britain 103–4, 119 Brazil 127–9, 135, 143, 146–7 China 184–6, 190 SER see standard employment relationship (SER) SER-centrism 58–90 alternatives to 66–7 beyond employment approach 59, 69–72 problem of 62–6 severance pay 153 short time compensation (STC) 213, 216 as automatic stabilizer 218–20 benefits of 217 benefits to employers 222 as structural adjustment mechanism 218–20 and work flexibility 214 see also work-sharing SIMPLES law 130, 135–7, 145 simultaneous causation 42 Smith, Adam 227 social drawing rights 70, 79–80, 81, 84 social internalization 346

social interventions 359 social security Brazil 127, 128 China 192 under cheque-system 265, 268–9 see also unemployment benefit social values 14–16 software industry 99, 102, 104, 105, 108, 119, 326, 328 Spain collective bargaining 253 gender pay gap 253 gender pay ratio 233 low pay 254 female workers 231 minimum wage 253 unemployment benefit 236 standard employment relationship (SER) 6, 58, 60–2 flexible 59, 67–9, 70, 71, 72, 82, 84 see also SER-centrism standardized working time 60–1, 62 statut professionel 69, 72 structured occupations 97, 99, 102, 103–6, 107, 112, 113 Supiot Report 92 Survey on Working Conditions in Tanzania 298 Sutton Trust 99, 100, 114 Sweden collective bargaining 253 gender pay gap 253 gender pay ratio 233 low pay 254 female workers 231 paid parental leave 79, 85 Switzerland Chèque Service 263, 267, 269 collective bargaining 253 domestic work 263, 267, 269 gender pay gap 253 synthetic indices 38, 40 Tanzania Employment and Labour Relations Act (2004) 303 labour laws 291–2, 297–309

Index 375 time credits 80–1, 85 time series 32, 39, 41, 42, 44, 45, 46, 51–2 Titre-Service 263, 265, 266, 268 tournaments see entry tournaments; unstructured careers trade unions 75, 84, 152–3, 344, 347–8, 354, 356 Brazil 277, 322, 326–7, 329, 335 China 182, 187, 189–91, 194–5, 201–2 domestic workers and 141, 147 France 269 job security and 220, 227 Tanzania 298, 307–8 see also collective bargaining; industrial relations transaction cost economics 34, 53, 228, 234, 280 transient employment 93–4, 99 transnational labour citizenship 75 UFW–Global Horizons agreement 75–6, 81 UK collective bargaining 253 gender pay gap 253 gender pay ratio 233 labour laws 41 low pay 254 female workers 231 minimum wage 253 National Minimum Wage 20, 47, 239, 243, 244 unemployment benefit 236 underregulation bias 16–17 unemployment benefit 13 eligibility for 235–6 exclusion from 235 part-time workers see short time compensation unemployment insurance 210, 214, 219, 220–1 see also short time compensation unemployment rate Indonesia 163 youth 47 unions see trade unions unit labour costs in Indonesia 169

United Food and Commercial Workers of Canada 76 United Nations Committee on Economic Social and Cultural Rights 344, 351, 353, 361 universal caregiving 71, 72–3, 78–82 universities 4, 115, 116 international students 101–2 temporary employment 102 university teaching 101–2 unstructured careers 96–7 changing opportunities 106–12 client benefits 98–103 employment conditions 103–6 USA collective bargaining 253 gender pay gap 253 labour laws 41 low pay 254 female workers 231 minimum wage 45–6, 253 National Recovery Act 211–12 WARN laws 49, 52 work-sharing 211–12 varieties of capitalism approach 31, 37 wages gender inequality see gender pay gap minimum see minimum wage reservation 234–41 wage setting 156 gender inequality 242–4 and productivity 156 wage-setting institutions 356–7 Weber, Max 334 Wen Jiabao 180 Window for the Jobless 78 women see female workers work-sharing 209 history 210–12 macroeconomic benefits 216–18 and productivity 222 revival of 212–16 STC see short time compensation (STC) see also part-time employment

376

Index

work-time limits 341–4 constraints on obligatory overtime 353–4 Honduras 351–3 see also working time regulation working conditions 296 and awareness of labour laws 306–7 Working Conditions Survey Global Module 297–8 working time regulation 13, 21 awareness of 304, 305

World Bank Doing Business survey 2, 3, 39, 54, 160, 164–6, 292, 333 Employing Workers Index (EWI) 2, 3, 18, 39, 295–6 Internal Evaluation Unit 3 Private Finance Initiative 292 World Business Environment Survey 333 youth unemployment 42 Brazil 134–5