European Welfare States and Supranational Governance of Social Policy Ailish Johnson
St Antony’s Series General Editor...
31 downloads
723 Views
884KB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
European Welfare States and Supranational Governance of Social Policy Ailish Johnson
St Antony’s Series General Editor: David Faure (2003–) Fellow of St Antony’s College, Oxford Recent titles include: Ailish Johnson EUROPEAN WELFARE STATES AND SUPRANATIONAL GOVERNANCE OF SOCIAL POLICY Archie Brown (editor) THE DEMISE OF MARXISM-LENINISM IN RUSSIA Thomas Boghardt SPIES OF THE KAISER German Covert Operations in Great Britain during the First World War Era Ulf Schmidt JUSTICE AT NUREMBERG Leo Alexander and the Nazi Doctors’ Trial Steve Tsang (editor) PEACE AND SECURITY ACROSS THE TAIWAN STRAIT C.W. Braddick JAPAN AND THE SINO–SOVIET ALLIANCE, 1950–1964 In the Shadow of the Monolith Isao Miyaoka LEGITIMACY IN INTERNATIONAL SOCIETY Japan’s Reaction to Global Wildlife Preservation Neil J. Melvin SOVIET POWER AND THE COUNTRYSIDE Policy Innovation and Institutional Decay Juhana Aunesluoma BRITAIN, SWEDEN AND THE COLD WAR, 1945–54 Understanding Neutrality George Pagoulatos GREECE’S NEW POLITICAL ECONOMY State, Finance and Growth from Postwar to EMU Tiffany A. Troxel PARLIAMENTARY POWER IN RUSSIA, 1994–2001 A New Era Elvira María Restrepo COLOMBIAN CRIMINAL JUSTICE IN CRISIS Fear and Distrust Julie M. Newton RUSSIA, FRANCE, AND THE IDEA OF EUROPE
Ilaria Favretto THE LONG SEARCH FOR A THIRD WAY The British Labour Party and the Italian Left Since 1945 Lawrence Tal POLITICS, THE MILITARY, AND NATIONAL SECURITY IN JORDAN, 1955–1967 Louise Haagh and Camilla Helgø (editors) SOCIAL POLICY REFORM AND MARKET GOVERNANCE IN LATIN AMERICA Gayil Talshir THE POLITICAL IDEOLOGY OF GREEN PARTIES From the Politics of Nature to Redefining the Nature of Politics E. K. Dosmukhamedov FOREIGN DIRECT INVESTMENT IN KAZAKHSTAN Politico-Legal Aspects of Post-Communist Transition Felix Patrikeeff RUSSIAN POLITICS IN EXILE The Northeast Asian Balance of Power, 1924–1931 He Ping CHINA’S SEARCH FOR MODERNITY Cultural Discourse in the Late 20th Century Mariana Llanos PRIVATIZATION AND DEMOCRACY IN ARGENTINA An Analysis of President–Congress Relations Michael Addison VIOLENT POLITICS Strategies of Internal Conflict Geoffrey Wiseman CONCEPTS OF NON-PROVOCATIVE DEFENCE Ideas and Practices in International Security Pilar Ortuño Anaya EUROPEAN SOCIALISTS AND SPAIN The Transition to Democracy, 1959–77
St Antony’s Series Series Standing Order ISBN 0–333–71109–2 (outside North America only) You can receive future titles in this series as they are published by placing a standing order. Please contact your bookseller or, in case of difficulty, write to us at the address below with your name and address, the title of the series and the ISBN quoted above. Customer Services Department, Macmillan Distribution Ltd, Houndmills, Basingstoke, Hampshire RG21 6XS, England
European Welfare States and Supranational Governance of Social Policy Ailish Johnson Foreword by Allan Larsson
in Association with St Antony’s College, Oxford
© Ailish Johnson 2005 Foreword © Allan Larsson 2005 All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No paragraph of this publication may be reproduced, copied or transmitted save with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, 90 Tottenham Court Road, London W1T 4LP. Any person who does any unauthorized act in relation to this publication may be liable to criminal prosecution and civil claims for damages. The author has asserted her right to be identified as the author of this work in accordance with the Copyright, Designs and Patents Act 1988. First published in 2005 by PALGRAVE MACMILLAN Houndmills, Basingstoke, Hampshire RG21 6XS and 175 Fifth Avenue, New York, N.Y. 10010 Companies and representatives throughout the world. PALGRAVE MACMILLAN is the global academic imprint of the Palgrave Macmillan division of St. Martin’s Press, LLC and of Palgrave Macmillan Ltd. Macmillan® is a registered trademark in the United States, United Kingdom and other countries. Palgrave is a registered trademark in the European Union and other countries. ISBN-13: 978–1–4039–3995–1 ISBN-10: 1–4039–3995–0 This book is printed on paper suitable for recycling and made from fully managed and sustained forest sources. A catalogue record for this book is available from the British Library. Library of Congress Cataloging-in-Publication Data Johnson, Ailish M. European welfare states and supranational governance of social policy / Ailish Johnson; foreword by Allan Larsson. p. cm. — (St. Antony’s series) Includes bibliographical references and index. ISBN-13: 978–1–4039–3995–1 ISBN-10: 1–4039–3995–0 1. European Union countries – Social policy. 2. Welfare state – European Union countries. 3. European Union. 4. International Labour Organization. I. Title. II. Series. HN373.5.J63 2005 320.6’094 — dc22 10 9 8 7 6 5 4 3 2 1 14 13 12 11 10 09 08 07 06 05 Printed and bound in Great Britain by Antony Rowe Ltd, Chippenham and Eastbourne
2004048936
To my family
Contents List of Tables and Figures
ix
List of Abbreviations
xi
Acknowledgements
xiii
Foreword by Allan Larsson
xv
1 European Welfare States and Supranational Governance of Social Policy Defining social policy First explanatory variable: welfare state regimes in the EU Second explanatory variable: institutions and new forms of governance Third explanatory variable: when is ideology relevant? Method and sources 2 Governance by Law: Health and Safety in the Workplace Welfare state preferences in health and safety policy Health and safety regulation in the EU: 1952–85 Negotiations on qualified majority voting (QMV) in art. 118 of the SEA The Framework Directive on the introduction of measures to encourage improvements in the safety and health of workers at work (Directive 89/391/EEC) Conclusion: the unanticipated consequences of qualified majority voting in the SEA 3 Governance by Collective Bargaining: The Social Dialogue Welfare state preferences for Social Dialogue Social Dialogue at the EU level: 1955–90 Negotiations on the enhancement of Social Dialogue in the Maastricht Treaty The Framework Agreement on part-time work: negotiations and national implementation Conclusion vii
1 3 7 14 22 23 28 29 31 34
42 58 61 65 71 74 86 95
viii Contents
4 Governance by the Open Method of Coordination: The European Employment Strategy Welfare state preferences in employment policy Employment policies in the ECSC, EEC and EU: 1954–94 Amsterdam Treaty negotiations The European Employment Strategy: the anticipated and unanticipated consequences of the European Employment Title Conclusion 5 Global Governance of Social Policy: EU Member States at the International Labour Organization The structure of social policy legislation and rights in the global system National ratifications of ILO Conventions by EU Member States EU–ILO relations: formal and informal inter-institutional cooperation Social policy at the global level: five specific issue areas of cooperation Conclusion: global governance and the EU
97 98 103 108
123 138 142 145 151 155 165 183
6 Conclusion Supranational forms of governance compared Leaders, resisters and passive Member States in supranational governance What’s in and what’s out: the evolution of supranational governance in social policy
186 187
Notes
207
Appendix: List of Interviews
227
Bibliography
232
Index
260
196 201
List of Tables and Figures Tables 1.1 1.2 2.1 2.2 3.1 3.2 3.3 3.4 4.1 4.2 4.3 4.4 5.1 5.2 5.3 5.4 6.1
Spending on social policy in EU Member States, 1998 Qualified Majority Voting in the Council of Ministers, 1981–2004 Member State political affiliations, 1985 Factors determining the evolution of health and safety policy at the EU level Collective bargaining systems in EU-15 Member States Member State political affiliations 1991–2 Factors determining the evolution of Social Dialogue at the EU level Part-time work as a percentage of total employment, and women as a percentage of part-time workers, 1998 Allocation of labour market expenditures: 1990, 1996 Unemployment rates: 1990, 1996 Member State political affiliations 1995–7 Factors determining the evolution of employment policy at the EU level Total number of ILO Convention ratifications by EU-15 Member States ILO Convention ratifications by EU-15 Member States before and after EU membership Ratifications by EU-15 Member States of ILO Conventions proposed 1987–2001 National positions in ILO policy areas Evolution of regional governance in social policy
9 16 36 50 68 75 85 88 100 101 110 129 152 153 154 167 191
Figures 1.1 5.1
Overall research design Social rights in the global and European system
ix
26 146
List of Abbreviations ACP ACP ADEDY ALMPs BEPGs CAP CEEP CBI DGB DG V ECJ ECOFIN ECSC EEC EES EFTA EMU EP ESC ESF ETUC EU GATT GSEE GSP ICITO IGC ILO IMEC IOE ITO LO
Approved Code of Practice African, Caribbean and Pacific countries Civil servants union (Greece) Active labour market policies Broad Economic Policy Guidelines Common Agricultural Policy European Centre of Public Enterprises Confederation of British Industry German Confederation of Trade Unions Directorate General V, now DG Employment and Social Affairs European Court of Justice Economic and Financial Affairs Council of the European Union European Coal and Steel Community European Economic Community European Employment Strategy European Free Trade Association European Monetary Union European Parliament European Economic and Social Committee European Social Fund European Trade Union Confederation European Union General Agreement on Tariffs and Trade Greek General Confederation of Labour Generalized System of Preferences Interim Commission of the International Trade Organization Intergovernmental Conference International Labour Organization Industrial Market Economy Countries group (at the ILO) International Organization of Employers International Trade Organization (proposed; replaced by the GATT 1947) Swedish Trade Union Confederation
xi
xii List of Abbreviations
NAP
National Action Programme (of the European Employment Strategy) ND New Democracy Party (Conservative party, Greece) OECD Organisation for Economic Co-operation and Development OMC Open Method of Coordination PASOK Socialist Party (Greece) PERMREP Permanent Representation to the European Union (national PERMREPs) QMV Qualified Majority Voting SACO Swedish Confederation of Professional Associations SAF Swedish Employers’ Confederation (now the Confederation of Swedish Enterprise) SAP Social Democratic Party (Sweden) SAP Social Action Programme SEA Single European Act SEM Single European Market SEV Federation of Greek Industries SMEs Small and Medium Sized Enterprises TACIS Technical Assistance to the Newly Independent States (former USSR) TCO Swedish Confederation of Professional Employees TEC Treaty establishing the European Community (Treaty of Rome) TEU Treaty on European Union (Maastricht Treaty) TUC Trades Union Congress (UK) UEAPME The European association of craft, small and medium-sized enterprises UNICE Union of Industrial and Employers’ Confederations of Europe WTO World Trade Organization
Acknowledgements A journey cannot begin, nor will it ever be finished with joy, without the support of friends. For their strong friendship ‘The Ladies’ of Oxford, London, Dublin and Toronto have my deepest thanks: Adjoa, Bec, Brooke, Chloé, Clare, Cleo, Kathleen, Monika, Porscha, Sarah P, Sarah R, Siobhan, Shelley, Tanya, Tina and Vanessa. Other friends who helped along the way were Robert Van Schaik, Sam Becker, Andrew Laidlaw, David and Lindsay Smith, Nick Gallus, Robert Gerwarth, Rick and Anna Boomgaardt, John Booth and especially Dan Campbell. As is the case with many an adventure, this one benefited from the help of several key backers. I am forever indebted to the following for the funding of the research of this book: the Harrison McCain Scholarship of University College, and the Department of International Relations, Oxford, the British Council, the Scatcherd European Scholarship, the Europaeum Travel Bursary and the Government of Canada. Kalypso Nicolaïdis, my supervisor at the University of Oxford, has my special thanks. For additional guidance during my extended research trips I would like to thank Panos Tsakloglou of the Athens School of Economics and Business and Richard Blackhurst of the Graduate Institute of International Studies, Geneva. I am also indebted to Sylvia Ostry of the University of Toronto who inspired the topic of my research. For assistance with interviews in Athens, Brussels, Geneva, London and Stockholm, and for their valuable discussions of my research, I would like to express my appreciation to: Loukas Tsoukalis, Gregory Papanikos, Lord Butler of Brockwell, Allan Larsson, Karin Johansson and Anders Lindberg. The following colleagues who read and commented on various drafts have my great thanks: Ngaire Woods, Christopher McCrudden, David Hine, Ben Rosamond, Peter Leslie, Randle Wilson, Mark Rhinard, Helen Callaghan, Richard Pan, Sarah Percy, Vivien Collingwood, Nicole Evans, Charles La Salle, Isolda Guevara, Aaron Fowler, David Perdue and Alex Muggah. Of course any errors or omissions remain my own. I would also like to thank all those who were willing to be interviewed for this book, many of whom took extra time to discuss my research and point me in the direction of further sources.
xiii
xiv Acknowledgements
Jill Lake and Alison Howson of Palgrave Macmillan brought the work to print, and for that I am very grateful. Denyse Goulet agreed to have her wonderful artwork used on the cover. Thanks, too, to Colin Dexter for introducing me to the ‘Oxford comma’, which I used throughout the draft of this book, that my copy-editor Keith Povey dutifully removed! Finally, my family has been the biggest source of support and love throughout my adventures, intellectual and otherwise. I would like to dedicate this book to Clarke, Mary, Brendan, Aidan and May. AILISH JOHNSON
Foreword The subject of this book – the interplay between national and supranational interests, ideas, and institutions in the field of social policy – is one that I have been closely involved in as a practitioner. In the early 1990s employment policy was seen by many governments of that era as a matter of national concern, and sometimes a residual, a policy in the margins of economic policy. It was not a matter of common European concern. Entrepreneurship by both leader Member States and by Commission officials is key to explaining the expansion of the social policy agenda in the EU. Commission President Jacques Delors is one of these key figures, having been instrumental in both encouraging the development of the Social Dialogue in the EU, and later in fostering the White Paper on Growth, Competitiveness and Employment. Delors paved the way for Sweden’s push to include employment in the Amsterdam Treaty. Other key national players, such as Prime Minister Jean-Claude Juncker of Luxembourg, and EU officials such as Commissioner Flynn, helped move the process embodied in the Treaty to reality. The Luxembourg Process, now known as the European Employment Strategy (EES), is an established process for the review and improvement of employment policies in Europe. The work of several Presidencies has also underpinned and expanded the process available in the Amsterdam Treaty. As this example shows, the interrelationship between the Commission and Member States, both during moments of Treaty reform and in the policy-making process, is a unique and complex phenomenon, but one that can bear substantial fruit when these forces work together. Indeed, each of the chapters of this book demonstrates that the national-level story is essential to the European one, and that creative new forms of policymaking will emerge that permit integration without supranationalism. The EES, which is but one variation of the open method of coordination (OMC), demonstrates that harmonization is not the goal. Rather, a common strategy aids and bolsters national-level reform and creates a process of continuous learning and improving in which even the most advanced Member States have their assumptions and policy prescriptions put to the test. Today, there is more integration than ever in Europe. The fear that EMU would lead to social policy deregulation has not occurred. Quite xv
xvi Foreword
the contrary, it seems that EMU triggered the resurgence of national social pacts and provided a window of opportunity for agreement over the European Employment Strategy. One of the interesting conclusions of this book is that successive waves of enlargement of the EU also encourage the development of high standards in the social field, promoting a ‘race to the top’. The enlarged Union of 25 Member States thus promises a complex yet rich opportunity for cooperation in the interests of EU citizens. We still have a long way to go. In particular, we must continue to strive for greater participation in the labour market, especially among women, and we must ensure that workers are provided with access to lifelong training. Some, more pessimistically, may note that the goals we set ourselves at Lisbon in March 2000 remain a long way off. We are in danger of not meeting our targets. With these goals, I would respond, we have a common vision of where our journey should take us, and, by striving for these goals at both national and regional levels, we have motivated EU members to start on a path of collective action that ten years ago would not have seemed possible. I look forward to the new challenges and directions EU social policy will take, and I trust that the research community that is represented so well in the pages of this book will be there to both document and inspire progress. ALLAN LARSSON Former Director-General, DG Employment and Social Affairs EU Commission and former Swedish Minister of Finance
1 European Welfare States and Supranational Governance of Social Policy
Governance is the transformation of a diverse set of actor preferences into policy choices and coordinated action (Eising and Kohler-Koch 1999, p. 5). Governance in the social policy field has traditionally been located at the level of the state. Increasingly, however, new forms of governance are emerging at supranational levels. Policy-making in the social policy field at the regional level in Europe now includes three distinct forms of governance by the European Union: governance by EU law, governance by EU-level bargaining between employers’ and workers’ associations in the Social Dialogue, and the governance of employment policy in the form of the European Employment Strategy. These diverse and complex modes of regional governance in the social policy field have emerged over a short period of time. While governance by EU law was included in the Treaty of Rome on the basis of unanimity voting in a highly limited number of issue areas, qualified majority voting was not introduced until the Single European Act (1987). Subsequent Treaty revisions of Maastricht (1993), Amsterdam (1999), and Nice (2002) have each resulted in the expansion of EU competence in the social policy field, and in new forms of governance. Policymaking, while necessarily continuing to involve states, also extends to the global level through the activities of the International Labour Organization (ILO). Supranational governance remains largely limited, however, to a narrow range of social policy best described as ‘social regulation’ which, while providing basic rules for the operation of the labour market and employment contract, does not involve large fiscal transfers to the individual. Even given the limitations to regional and global governance, the fact that supranational policy cooperation has emerged in the social field is 1
2
European Welfare States
surprising, considering that states have a long history in social policy provision. As social policy, along with taxation, is one of the key ways governments distribute resources in society, why would states be willing to share this competence? And given the diversity of welfare states, why do states agree to cooperate at supranational levels, and how may cooperative mechanisms be formed at the supranational level? The emergence of new forms of supranational governance indicates that under intense forms of economic integration states may demand social policy responses beyond their own borders. This suggests that some aspects of social policy may be particularly vulnerable to the pressures of economic integration. Regional and global governance may balance or offset, then, some of the real and perceived effects of economic competition and integration on social policy. Even under conditions of weaker economic integration, such as exists between EU Member States and the rest of the world, global institutions are increasingly called upon to establish basic minimum social policies. While this book takes as its field of study a specific issue area, an examination of EU- and global-level governance in social policy is of interest to the broader international relations research community. This is because much of the thinking about the processes of regional integration relies more generally on a reading of the European experience. The EU is the most developed regional institutional system for the provision of economic and social policy. An analysis of the EU indicates what states at the most advanced level of economic integration may be capable of in the social policy field, and signals a path for other regional associations. The EU is also held up as a model for global cooperation and a ‘powerful reason for optimism about the prospects for a fair global labour market’ (Mehmet et al. 1999, p. 217). If this is the case, EU models of governance must persist beyond the regional level. An assessment of the global forms of governance available to the ILO, and the activity of the EU and its Member States at the ILO, is the subject of the penultimate chapter of this book. This chapter suggests that the EU cannot be treated as a unitary actor at the global level. The preferences of Member States for cooperation in different sub-issue areas of social policy have a direct impact on the EU’s strengths and weaknesses at the global level. In addition, enforcement mechanisms are much weaker at the global level when compared to EU enforcement mechanisms. This chapter also suggests that as a result of occupying similar ‘regulatory space’ (Murray 2001, p. vii), regional and global governance are not always complementary. Despite the weaknesses of global governance, and the tensions
European Welfare States and Supranational Governance 3
between global and regional governance, the ILO’s work in core labour standards is an important and necessary complement to regional governance in a global economy. Finally, tracing the relationship between the EU and the ILO over time is a valuable way of exploring whether the interaction of regional and global governance is stable over time, or if changes in relative policy-making capacity and policy leadership between the regional and global levels can be detected. A study of regional and global social policy demands a methodology that is sensitive to both national and supranational contributions to governance. The research method employed in this book is inspired by studies that combine the national roots of policy-making with an understanding of EU institutions and decision-making procedures (EspingAndersen 1996; Hine and Hussein 1998; Rhodes 1998; Scharpf 1999; Ferrera et al. 2000). This book proceeds by analysing the creation of new forms of governance by combining detailed information about national welfare state regimes with an analysis of the ways in which regional and global institutions present opportunities and incentives for cooperation. Each chapter on regional and global forms of governance thus takes, as its starting-point, national explanations for initial Member State preferences before turning to the institutional setting of decision-making. This book hopes to make both an empirical and an analytical contribution. Empirically, evidence is provided on the evolution of social policy at all three levels of governance (national, regional, global), so often neglected in international relations. This evidence serves to suggest how new forms of supranational governance emerge, and the relative contribution of national, regional, global, and non-state actors to the creation of supranational governance. The comparative analysis of several Member States also contributes to the understanding of the effects of supranational governance on national policies. The inclusion of Greece is of special importance, given its absence, or cursory treatment, in even the most comprehensive comparative research done on social policy to date (Esping-Andersen 1990; Falkner 1998; Leibfried 1992; Leibfried and Pierson 2000; Scharpf and Schmidt 2000).
Defining social policy T. H. Marshall noted the imprecise boundaries of social policy. Social policy is ‘a term which is widely used but does not lend itself to precise definition’ (Marshall 1975, p. 11). As a result, Marshall suggested a ‘conventional approach’ for defining social policy. This approach asks which areas of government action are generally accepted as belonging to the
4
European Welfare States
social policy sphere, using conventional practice to set the limits of a definition (Marshall 1975, p. 11). The ILO surveyed the social policies of its members in 1952, and suggested that social policy includes the following measures: maintaining income during sickness, unemployment allowance, old age allowance (old age pension) and family allowance. Housing, education, community services and health care also form part of the broad remit of social policy (Marshall 1975, p. 11). This broad conception of social policy has been embraced by contemporary scholars of social policy, who define the core of social policy as consisting of ‘social insurance, public assistance [social assistance], health and welfare services, and housing policy’ (Majone 1993, p. 158). Social policy may thus be described as consisting of several components. Benefit provisions during illness, unemployment, time off due to accidents in the workplace, and old age may be termed ‘social insurance’ (or ‘social security’) and are benefits built up by the individual during time in employment. The term ‘social assistance’ refers to grants made on the basis of a means test, and includes minimum incomes provided to the long-term unemployed, the disabled and the elderly who may not have worked, or whose social-insurance-provided pension is insufficient. Housing and family benefits, provided on the basis of a means test, may also be thought of as social assistance. Governments also enact regulations affecting the labour market, such as rules concerning working time, employment contracts, and employment policies designed to affect labour market entry and exit. Other regulations may concern occupational health and safety and other working conditions (Leibfried 1992, p. 99). These latter forms of government intervention have been termed ‘social regulation’, and may be considered a subset of social policy (Majone 1993). Social regulation may also be seen as policies at the interface of welfare-providing social policy and market-correcting economic policy. Additional components of social policy include health care and education. Values are also embedded in the definition and provision of social policy, and have led to the development of a variety of regimes for the definition and delivery of social policy. For example, British conceptions of social policy are preoccupied with reconciling universal social rights available to all citizens with market efficiency. Continental social policies, in contrast, are informed by the concept of solidarity. The main thinkers behind French social policy, namely Emile Durkheim, Léon Bourgeois and Pierre Laroque, emphasized the concept of interdependence (solidarism) that created a system of solidarity out of social pluralism. The dominant value in this system was the protection of
European Welfare States and Supranational Governance 5
the individual from risks associated with work – illness, accidents, unemployment – through occupation-based schemes of social security. The individual, by doing as much as possible for his or her needs in cooperation with similar efforts by others (mutualité ), ensured mutual aid (entraide) (Laroque 1953; Bussemaker 1999, pp. 32–3). In Germany, the first EU Member State to have a well-developed system of social security, the first social policies were based on solidarity within occupational categories (Esping-Andersen 1990, p. 26). Social services were not a result of rights of citizenship, but were attached to work. The tensions between values can be seen today in the debates about the kinds of social policy that should be provided at the regional and global levels, in particular as the marketplace expands and as economic integration deepens. The phrase ‘welfare state’ is one that emerged after the Second World War, and was first used to describe the complex of social policies enacted under the Labour government in Britain after 1945. From Britain, ‘the phrase made its way round the world’ (Briggs 1961, p. 221).1 The welfare state possesses a unique institutional structure for the administration of social policy (Ashford 1986). In Weberian terms, the development of the welfare state may be linked to the processes of rationalization, and the development of bureaucratic authority and the emergence of social clients (Cousins 1987). The expansion of the services provided by the welfare state in the twentieth century to cover a widening range of risks has also been described as a ‘general phenomenon of modernization’ (Flora and Heidenheimer 1981, p. 23). The growth of the number and kinds of social services provided by the state was a function of its ability to create procedures for defining social policy, raise the funds (through taxation and contributions) to pay for these policies, identify recipients and distribute benefits and services. The welfare state may thus be conceived as a very advanced and developed state apparatus, functioning in a political space that recognized its legitimacy, and in an economic space that provided the means of paying for its outputs. Through its evolution and expansion, the welfare state has partially incorporated and partly supplanted other institutions that formerly accepted responsibility for welfare: the individual, the family, the church, local organizations and charities, and the firm (Flora and Heidenheimer 1981, p. 51). Thus, the welfare state represents an institutionalization of risks and responsibilities formerly borne by other actors. This has led some to label those countries, including the EU Member States, with well-developed welfare services that continue to change and develop over time, as ‘welfare societies’ or ‘social states’ (Titmuss 1968, p. 127; Leibfried 1992).
6
European Welfare States
The welfare state, in its historical and institutional context, must also be linked to one other broad phenomenon: the development of capitalism and the market economy. As the linkage between the welfare state and modernization suggests (Flora and Heidenheimer 1981, p. 23), social policies emerged as industrialization came to dominate the production structure and as markets became increasingly regulated by state bureaucracies. The state became involved in mediating the market mechanism, which ‘organizes the exchange of economic resources and commodities’, through three processes: ‘regulation (labour legislation), supplementation (social security systems), and replacement (social services)’ (Flora and Heidenheimer 1981, p. 40). Harold Wilensky, who theorized that the welfare state would only emerge and expand in tandem with economic development, called this ‘the process of humanizing industrial society’ (Wilensky 1975, p. 119). Political power is ‘deliberately used’ to ‘modify the play of market forces’ (Briggs 1961, p. 228). The development of the welfare state should thus be thought of as part of a nexus of state-building that intersects and overlaps with the demand for increasing social security provisions by politically enfranchised populations, the supply of the means to fulfil these demands through the wealth generated by capitalism, and the creation of administrative structures to manage complex policies. As this nexus developed most fully in the post-war period, it has been described as the post-war ‘Keynesian compromise’ between capitalist production and the socialist state (Strange 1999, p. 188). In Germany, ‘social market economy’ (Soziale Marktwirtschaft)2 and in Britain ‘welfare capitalism’ are used to refer to the mutual existence and organization of capitalism and the welfare state (Streeck 1999, p. 48). Interestingly, the draft EU Constitution approved by the heads of government of the enlarged EU in June 2004 states that the union shall work for a ‘highly competitive social market economy, aiming at full employment and social progress, a high level of protection’ and shall ‘combat social exclusion and discrimination’ (EU 2004, art. I-3). While welfare state regime types provide the broadest way of comparing Member States, the policies examined in detail in this book – health and safety, social dialogue, employment – could be more specifically described as policies relating to the labour market. The interaction between the welfare state and the labour market is incredibly close and complex. Policies regulating the labour market are, in large part, described as social regulation. These polices are the key interface between welfare policies legislated and/or provided by the state (health and safety legislation, worker training)3 and the organization of work as part of the
European Welfare States and Supranational Governance 7
capitalist system of production (working time, worker information and consultation). Importantly, some key labour market policies are not examined in detail, such as pensions and disability benefits, which are typically grouped under the category of ‘social insurance’. The emphasis here is on labour market policies that are grouped under ‘social regulation’: those policies that involve legislation by governments to improve social welfare but which do not require large-scale fiscal transfers to the level of the individual. It is of interest that those aspects of social policy integration that have occurred beyond the state are concentrated in social regulation rather than the other forms of social policy, that require much larger-scale financing from national and sub-national authorities (including municipalities) and from workers and employers.
First explanatory variable: welfare state regimes in the EU Welfare state regime types, as developed by Gøsta Esping-Andersen and others, identify four clusters of welfare states: liberal, conservativecorporatist, social-democratic, and southern (Esping-Andersen 1990; Esping-Andersen and Kolberg 1992; Esping-Andersen 1996; Ferrera 1996; Ferrera 1998). Each EU Member State may be described as resembling most closely one of these regime types: Germany, Belgium, France, the Netherlands, Luxembourg and Austria are conservative-corporatist, the UK and Ireland are liberal, Sweden, Denmark and Finland are socialdemocratic, and Greece, Spain, Italy and Portugal are of the southern welfare state regime. Esping-Andersen’s work on welfare state regime types also suggests why Member States express such diversity in their preferences for EU social policy cooperation. Member States with social-democratic and conservative-corporatist welfare state regimes have social and labour market policies that provide a large range of services at a high level of benefit. They are expected to want such policies to be adopted by other Member States, and thus should be policy leaders at the EU level. If such polices were adopted across the EU, they would no longer form a competitive disadvantage in the cost of locating production and hiring labour in the region. Historically well-developed welfare states also possess strong domestic interests, especially unions, who demand the continuation of welfare state benefits and government intervention in the labour market. Further, if other Member States adopted such services and high wage replacement levels, there would be less of a tradeoff
8
European Welfare States
between competitiveness and achieving social justice inside the EU, although, of course, competitive pressures would still remain from outside. Thus, social dumping inside the EU – or the relocation of investment and production to areas where social standards are low in order to lower costs – would not occur. However, leader states are well aware that not all other Member States will be willing or able to implement policies at a high level. Their goals are thus to set a policy minimum, to create incentives for policy-making in ways that raise standards in other Member States, and to implement procedures (such as the consultation of unions and employers) closer to their own. Table 1.1 illustrates the range of spending on social policy in EU Member States. Concomitantly, Member States with liberal and southern welfare state regimes that provide fewer services at a lower level of benefit will not want to see their competitive advantage disappear. Further, under conditions of European Monetary Union (EMU), and the necessary fiscal discipline that is part of the Stability and Growth Pact, Member States with less-developed welfare states are constrained (but certainly not prevented) from expanding their spending in any case. Lower wages and lower levels of benefit provided to those in work may also reflect the differences in productivity between Member States, providing yet another reason for Member State governments to be unwilling to price their working citizens out of the market. Member States with liberal and southern welfare state regimes should be expected, on the basis of a welfare state regime analysis, to resist EU-level cooperation in the social policy field. In addition, southern Member States also possess weaker administrative cultures, less capable of both raising funds to pay for expanded social services and implementing these services, which would suggest additional resistance to supranational policies (Symeonidou 1996, pp. 68–9). Although the above description states the case most generally, it must be recalled that welfare states represent unique and complex bargains. Each Member State will exhibit its own politics of the welfare state; in other words, given the historical and complex arrangement of the policies that exist (or do not), there will be national differences in the entrenchment of specific policies in terms of their popular support, the actors that would be affected by reform, and the nature of their financing and administration (Pierson 2001a, p. 14; Pierson 2001b, p. 455). Thus, sub-issue areas must be examined at the national level in detail in order to avoid oversimplifying welfare state regimes.4 This analysis could be called a ‘welfare state explanation’ of social policy cooperation; Member States will form their preferences with regard
39 42 34 25 42 41 43 48 44 64 53 43 46 44 25
33 30 27 30 29 29 28 28 24 25 25 23 22 27 16
Source: Eurostat 2002, pp. 111, 122.
2
46
28 35 36 37 41 36 40 33 35 37 30 30 46 37 37 41
35
Spending on sickness, health care and disability, % of total
9 9 12 15 9 7 13 5 4 3 5 5 13 4 15
7
3
11 10 13 13 10 5 9 10 14 4 8 5 2 9 13
8
Spending on policies for family and children, % of total
4
5 3 4 5 3 7 3 1 1 0 4 2 1 7 5
4
Liberal
Spending on housing and social exclusion, % of total
Southern welfare state regimes;
Spending on unemployment, % of total
Conservative-corporatist welfare state regimes;
Spending on old age and survivors benefits, % of total
1 Social-democratic welfare state regimes; welfare state regimes.
EU-15 average Sweden1 Denmark Finland France2 Germany Netherlands Belgium Austria Luxembourg Italy3 Greece Portugal Spain UK4 Ireland
Spending on social policy, % of GDP
Table 1.1 Spending on social policy in EU Member States, 1998
9
10 European Welfare States
to EU policies in the social and labour market field on the basis of the policies that form their national welfare state. Member States will act to promote social and labour market policies where these are present at the national level, and suggest their adoption at the EU level. Member States will resist EU cooperation where such policies do not already exist, or exist at a lower standard, at the national level. The hypothesis tested is that Member States should act to promote or resist cooperation in the social policy field on the basis of their national welfare state regimes: states with high-standard and high-cost welfare state regimes are expected to be policy leaders, while low-standard and low-cost welfare state regimes are expected to be policy resisters. An examination of national welfare state histories, and the administration of specific policies, is necessary in order to explain the variance in national preferences for social policy integration at supranational levels. However, there is a gap between how Member States are expected to behave based solely on their national welfare state regimes and their observed behaviour. In many instances, Member States with low standards do not resist social policy cooperation, as is expected, but rather they agree to cooperate. Some states even behave passively in the design of new forms of governance in the social policy field, and in the creation of specific policies that result. Policy leaders, policy resisters and passive states The research gathered here shows that rather than falling into two simple categories of leaders and resisters of supranational social policy, as the research hypothesis initially suggests, EU Member States in fact have a tendency to fall into one of three categories of states with very different approaches to social policy integration at supranational levels: policy leaders, policy resisters and passive states. The first group of states are policy leaders. These Member States wish to see the EU create a more level playing field in social policy as a result of their highly institutionalized welfare states that provide a wide range of benefits, often at a high level, and that have highly regulated working conditions; the qualification ‘more level’ playing field is necessary, as even leader Member States are not always seeking common across-theboard standards. The approach of France in the 1950s – which argued for the harmonization of pay equality between men and women in the Treaty of Rome, and for the harmonization of working time, benefits and holiday periods in order to export many of its own welfare state policies to the emerging European Economic Community – has been abandoned. This is because it was clear that, even among the original
European Welfare States and Supranational Governance 11
six Member States, there was an inability to agree on common standards owing to different productivity and wage levels (ILO 1956). Despite having abandoned demands for policy harmonization, in specific issue areas, and under certain conditions, leader Member States would like to see agreement coalesce at a higher-than-average level, and certainly higher than the lowest common denominator as represented by the most reluctant Member State. Policy leaders push other Member States to agree to social policy integration, and they may express preferences for forms of governance other than governance by law in order to overcome past policy failures, or in order to reach agreement with reluctant states. Member States that are best typified as leaders are those with social-democratic welfare state regimes that emphasize redistribution, that provide a range of social and labour market benefits at a high level, and that have high union density. Such states are Denmark, and, since its membership in 1995, Sweden. Belgium, one of a group of conservative-corporatist welfare states that has high levels of benefits based on occupation groups and labour market policy concertation (either bipartite or tripartite), is also a consistent policy leader. A second group of Member States resists EU-level social policy. These Member States are opposed to contracting at the EU level because of the nature of their welfare states. Member States that resist EU-level policies often have liberal welfare state regimes that provide less extensive benefits to their citizens at lower levels of wage replacement, and that emphasize the individual’s responsibility for his or her own well-being over concepts of occupational or group rights. Resistant Member States do not want to see themselves bound to provide their citizens with more costly services as a result of integration at the regional level. Resistance may also be linked to government ideology that is sceptical about the project of European integration more generally. Ideological opposition to social policy integration suggests that cooperative efforts at the regional level should be limited to the economic field, and should leave the competency of national governments over social policy intact. The key social policy resister is the UK. Resisters have prevented policy cooperation in several instances, such as on proposals for worker consultation in the 1980s and 1990s. Resisters have also objected to the Commission’s use of Treaty articles to propose social policy legislation, such as the UK’s appeal to the ECJ regarding the Commission’s use of art. 137 (ex 118a) on health and safety to propose working time legislation (UK 1996a). Of course, resisters also eventually do participate in some sub-issue areas of social policy. Their agreement to coordinate action at the regional
12 European Welfare States
level often requires specific limits to be set on formal cooperation. More specifically, resistant Member States will often not agree to expand the competence of the EU in areas governed by law. As a result, new forms of governance involving new actors, and in some instances with fewer implementation costs and enforcement sanctions, are suggested to overcome their resistance. However, resistant Member States are not able to control Treaty reform fully owing to the requirements of complex bargaining, nor are they able to control policy-making owing to committeestyle decision-making that aims for consensus. As a result, treaty reforms and policy outputs often occur at a higher level than desired by resistant Member States. In addition, one of the unanticipated consequences of cooperation is that policy entrepreneurs, such as leader Member States and the European Commission, push the limits set by resistant states. There is also a third group of states that are passive Member States. Passive states possess southern welfare state regimes, characterized by historically underdeveloped policies and lower benefit levels (Ferrera 1996). The underdevelopment of national policies means that these states have far less developed preferences for EU social policy integration than either the leaders or resisters. What is notable about the passive Member States is the fact that they do not resist EU social policy-making, as explanations based solely on national welfare state histories or on social dumping and the preservation of competitive advantage would assume. More specifically, a social dumping theory regarding EU integration would suggest that over time, high-cost countries will have to lower their social standards in order to remain competitive, or risk having production and investment shift to low-cost countries. Costs are measured by both wage and non-wage costs such as social security contributions, working conditions including the rules associated with hiring and firing labour, and health and safety provisions. Regulation will thus coalesce around the lowest common denominator, namely the standards of the low-cost countries (Eichener 1992; Mirka and Ruhm 1997; Pelkmans 1997; Blanpain 1998). As will be demonstrated, the social dumping theory is not borne out by evidence from the EU. Given the presence of EU incentives, including funding, Member States agree to cooperate at the EU level despite their relative underdevelopment. Passive Member States include Greece, Italy, Portugal and Spain. Member states with a low level of social policy provision also have additional reasons to agree to cooperate: the need for domestic policy modernization, the desire for catch-up with more advanced Member States, and the hand-tying provided by the EU. Supranational governance,
European Welfare States and Supranational Governance 13
it will be shown, has the greatest impact on states where national policy histories are underdeveloped. However, while the passive Member States ultimately agree to new EU regulatory systems, their ability to implement EU outputs is curtailed by their low level of national institutional development. Uneven implementation has serious implications for the ultimate results of EU policy-making, and attempts to construct a common European social model. Certain Member States are consistently to be found in one of the three categories – leader, resister, passive – across the EU social policy issue areas we examine: health and safety, the Social Dialogue on part-time work, and the European Employment Strategy (EES). These include Denmark, Belgium and Sweden as policy leaders, the UK as a resister, and Greece as a passive Member State. Other Member States exhibit a greater variety in their stance towards EU-level social policy-making, depending on the issue area. The most important examples of inconsistent Member States are France and Germany. Given that the UK often resists EU social policy, that France and Germany have an ambiguous relationship with supranational governance in the social policy field, and that EU regulation presents great challenges for relatively underdeveloped Member States, the fact that social policy has developed at the EU level at all is of great interest to those examining the ways in which institutions can shape preferences and encourage agreement, and close gaps between the behaviour expected based solely on national preferences and actual outcomes. A welfare state explanation of preferences and positions in bargaining can thus explain, by careful analysis of specific areas of social policy, which Member States should be expected to be the most and least supportive of cooperation. The five Member States chosen for more in-depth consideration – France, Germany, Greece, Sweden and the UK – represent the diversity of welfare state regimes to be found in the EU, and the resulting diversity of positions taken with regard to supranational social policy integration. Given the 1 May 2004 enlargement to the new Member States of the East, the case of Greece is of particular interest in demonstrating how Member States with weaker domestic social policy structures behave under conditions of supranational cooperation.5 However, on its own, such an analysis cannot explain why states that should be expected to resist social policy cooperation on the basis of their national welfare state regime eventually do agree to Treaty reform that permits cooperation and, in some cases, are largely passive in the process of policy-making. Even more importantly, a welfare state explanation cannot account for the variety of new governance forms that
14 European Welfare States
have developed at the EU level. While national preferences can explain quite a bit about the relative enthusiasm of a state for cooperation in a specific area of policy-making, such an explanation is insufficient to explain the emergence of highly diverse and complex modes of governance over a very short period of time in the European Union. There are thus two sets of issues that an analysis using national welfare state regimes as its foundation cannot explain: the first are the gaps between expected and observed behaviour in Member States with regard to social policy cooperation; the second are the specific features of new forms of supranational governance in social policy, which have grown more flexible in terms of their implementing requirements over time.
Second explanatory variable: institutions and new forms of governance As Paul Pierson has explained, the key claim to be derived from ‘historical institutionalism’ is that actors may be in a strong initial position, seek to maximise their interests, and nevertheless carry out institutional and policy reforms that fundamentally transform their own positions ... in ways that are unanticipated and/or undesired. (Pierson 1998, p. 30)6 Southern Member States should be expected to have strong initial positions against cooperation, and yet agree to policy-making in common. Despite a preference for non-cooperation in the social policy field, the UK has also agreed to participate in EU governance of specific social issues. In addition, there are instances where France and Germany have expressed disinterest in cooperation, and yet have eventually agreed to cooperation. The fact that Member States with diverse national welfare state regimes agree to contract in the social policy field indicates that the supranational level of decision-making, namely the opportunities for bargaining and incentives provided by EU institutions, must factor in to the analysis of how social policy integration occurs.7 Institutions encourage cooperation in a number of different ways. Institutions can solve collective action dilemmas. Institutions may provide actors with greater certainty about the behaviour of other actors through the provision of information, and may supply enforcement mechanisms to ensure compliance with behavioural standards (Hall and Taylor 1996, p. 939). Strong enforcement mechanisms are supplied in the EU by the oversight of EU legislation by national courts and the
European Welfare States and Supranational Governance 15
Commission, and by referral to the European Court of Justice (ECJ). The ECJ’s legal supremacy gives it the power to strike down national legislation found to be in conflict with EU laws (Deakin 1997; Hix 1999; Slaughter et al. 2000).8 Such an assessment is commonly found in intergovernmental explanations of EU-level treaty bargains (Moravcsik 1998). Elements drawn from a historical institutionalist analysis of the EU, on the other hand, concentrate on explaining why gaps – ‘significant divergences between the institutional and policy preferences of memberstate governments and the actual functioning of institutions and policies’ – emerge, and why they cannot be closed (Pierson 1998, p. 34). In order to explain the gap between the expected and the actual behaviour of Member States, and the emergence of new forms of governance over time, several institutional factors must be considered: formal rules during both bargaining and policy-making, Commission entrepreneurship, the effects of institutional learning, the financial incentives and disincentives provided by cooperation, and unanticipated consequences. Formal rules during bargaining and policy-making Institutions set out the formal rules by which states may create common policies. A key example of a formal rule in the EU context is the use of unanimity or qualified majority voting (QMV) in the Council of the European Union when making EU law. Some key areas of regional social policy are governed by unanimity voting rules, giving each Member State a veto over the legislation proposed. Still, there have been significant moves since the SEA to increase QMV in specific sub-issue areas under art. 137 (ex art. 118a) that allow EU law to be created even if not all Member States vote in favour of specific legislation. The extension of QMV to some aspects of social policy was a key change in the federal rules governing decision making in the EU that has allowed for common policy-making to be facilitated. In a constantly enlarging EU it is impossible for a single large Member State (UK, Germany, France, Italy) to block EU law proposed under QMV, although in practice policy proposals are adapted so as to try and achieve consensus rather than forcing a vote (Hayes-Renshaw and Wallace 1996). From 1995 to 2004, the smallest blocking minority had to contain at least three Member States. Table 1.2 outlines the evolution of QMV in the EU. Under QMV in an enlarged EU of 25 Member States, which is currently governed by the Nice Treaty, passing legislation by a Council vote under QMV is more complex than in previous periods. 232 out of 321 votes are needed, plus a majority of Member States with 62 per cent of the EU’s population (EU 2000a).9 Blocking minorities could be as little as the
16 European Welfare States Table 1.2 Qualified Majority Voting in the Council of Ministers, 1981–2004 Distribution of votes 1981–5
1986–94*
1995–2004*
France Germany Italy Belgium Netherlands Luxembourg UK Denmark Ireland Greece Portugal Spain Austria Sweden Finland
10 10 10 5 5 2 10 3 3 5
10 10 10 5 5 2 10 3 3 5 5 8
10 10 10 5 5 2 10 3 3 5 5 8 4 4 3
Total Number needed to pass under QMV
63 45
76 54
87 62
* UK opt-out on Social Protocol 1991–7. Source: Adapted from Felsenthal and Machover (1997, p. 34).
three largest Member States (Germany, France, UK) that have enough of the total population (44 per cent) to block a vote by QMV. Alternatively, by the number of votes needed, the smallest blocking minority could be four Member States. In the proposed Constitution for Europe, which is to replace earlier treaties and provisions and consolidate them into one ‘EU’ legal entity, the qualifications for a successful vote by QMV have been increased even from the Nice Treaty rules (EU 2004). The proposed Constitution is not yet in effect, and has yet to be ratified by Member States. The purpose of this discussion of QMV is to point out that social policy is one area where changes to formal rule making have had a significant impact on the approval of legislation, although some aspects of social policy continue to remain under unanimity rules, such as social security (social insurance). Importantly, the difficulty of passing EU law under these formal rules of decision-making, even under QMV, caused the Commission and leader Member States to suggest the creation of alternative and more flexible forms of governance.
European Welfare States and Supranational Governance 17
Forms of social policy governance in the EU do not often bind Member States to one explicit standard, and in some instances do not even provide for formal enforcement and sanctions, for example in the case of the European Employment Strategy and the open method of coordination (OMC). In such cases Member States have made a tradeoff between cooperation and enforcement. In other words, leader Member States could have expressed a preference to cooperate only under the conditions of governance by law, enforceable by the ECJ, such as is provided in the ‘Community Method’ of cooperation which results in governance by law in the form of Council Directives and Regulations (Wallace 2000). In such an instance, it is probable that resistant Member States would have chosen not to cooperate at all and thus blocked cooperation. Therefore, leader Member States have agreed to cooperation under weaker enforcement mechanisms, or even in the absence of enforcement mechanisms, trading off robust enforcement mechanisms for an agreement to cooperate. Other formal rules in the EU in fact give power to the Commission, and not the Member States. In the EU, formal rules on the creation of legislation under the Community Method give the Commission the right of proposal. The Commission may consult widely in the initial design of proposed legislation and choose the Treaty Article under which to propose legislation. As a result, legislation is sometimes proposed that does not have unanimous or even majority support in the Council. The Commission is also able to propose legislation to fill policy gaps left by previous legislation. Control of new proposals does not rest with the Member States in the form of the Council, and thus no Member State can guarantee that regulation falling under EU competence will not be subject to proposals for new or enhanced legislation. EU decision-making also contains rules regarding the participation of certain actors, notably unions and employers in a decision-making capacity via the Social Dialogue and on a consultative basis in the Economic and Social Committee (ESC), and national experts in numerous advisory committees. The limited and structured opportunities for participation by groups at the EU level also means that some groups will have the benefit of participating in formal decision-making, notably unions, while other and more disparate groups in the EU, such as unemployed persons or the poor, will not have such opportunities. Commission entrepreneurship In addition to the formal rules of policy-making and enforcement that empower supranational actors, supranational assessments of institutions
18 European Welfare States
note the ability of institutions to move actors beyond narrow considerations of their respective interests. The role of the Commission cannot be ignored in an explanation of the outcome of policy-making in the EU. Arguments about the ‘common good’ may be introduced by the European Commission, for example, and to encourage policy-making on the basis of criteria other than pure national interest (Pierson 2000). Concerned with policy-making in the Union as a whole, the Commission has often used the rhetoric of the need for ‘balance’ by linking the development of social policy at the EU level to the Single European Market (SEM) and European Monetary Union (EMU).10 Careful process-tracing will reveal that economic policy-making has not automatically led to increased involvement in the social policy field, as neo-functional theories of integration would suggest (Lindberg 1963). Rather, social policies must be carefully promoted by the Commission over time, via Social Action Programmes or specific proposals, in advance of Member State agreement to cooperate (Commission 1974; Council 1974; Commission 1989; Commission 1995a). Often, new and alternative forms of EU governance, designed to overcome past policy failures and to expand cooperation to new areas, are the result of policy entrepreneurship by the European Commission. Specifically, officials from DG Employment and Social Affairs (formerly DG V) promote new forms of governance as a way of overcoming opposition to the use of the traditional Community Method. Even during Intergovernmental Conferences (IGCs), where the Commission does not have a formal role, Commission contacts with the government holding the Presidency of the Council of Ministers and with other governments have enabled it to promote the inclusion of new forms of governance in Treaty reform bargains.11 In addition, as the European Commission has the right of proposal under the Community Method of governance, specific policies are promoted that support cooperation at the supranational level. Even in modes of governance where the Commission does not have the right to propose legislation, activism by the Commission, and in some cases by the European Court of Justice (ECJ), has closed gaps and promoted cooperation on specific policies at levels unanticipated by Member States at the time of initial contracting. Complex decision-making rules and committee decision-making often reinforce the position of the Commission and isolate the most reluctant Member States during policy-making. Key individuals in the Commission have also played the role of ‘policy entrepreneur’ (Kingdon 1984), developing policy proposals, playing the role of policy sherpa around national governments and key national
European Welfare States and Supranational Governance 19
PERMREP officials, and building coalitions of interested parties, such as in the case of Allan Larsson in employment policy.12 In the context of Treaty and policy negotiations the Commission may help to provide new and innovative versions of texts for bargaining because of its intellectual resources and ability to gather information on Member State positions, as in the case of the Maastricht Social Protocol negotiations (Ross 1995b; Stubb 1998; Dehousse 1999). Commission officials are often more ‘risk-oriented ... open minded and innovative’ than national officials and are encouraged to present ‘new ideas’ and ‘innovative regulation which attempts to go beyond everything which can be presently found in the Member States’ (Eichener 1992, pp. 53–4). As a result of their role as guardian of the Treaties and the general European interest, and with self-interest in promoting European integration, Commission policy entrepreneurs often promote solutions that go well beyond amalgams or averages of national policies. In day-to-day policy operation, the role of the Commission as an agenda-setter (Garrett and Tsebelis 1996; Pollack 1997), and as a process manager of the committee system that prepares legislation (Eichener 1992), enables it to drive through policies that do not represent either the lowest common denominator or some kind of rough average of Member State positions, such as in the case of health and safety legislation. The results are policies that have implications even for policy leaders, although Member States with lower standards require the greatest adjustment to fulfil the policy outputs of EU modes of governance. The Commission, therefore, helps to ‘shape the intellectual framework in which ... national governments think about social policy’ (Hine 1998). The role of the Commission as a policy entrepreneur before, during and after Treaty change and in instances of specific policy-making is one of the key factors of an institutionalist analysis.
Institutional learning Institutions ‘shape the direction of the acquisition of knowledge and skills’ (North 1990, p. 78). Institutional learning effects include perceptions of the advantages of cooperation, and the expectations of future cooperative opportunities (North 1990, p. 94). On the positive side, learning effects create greater awareness of the interests and needs of other institutional partners, so that effective issue-based coalitions may emerge, together with the development of best-practice models that involve the transfer of skills from one partner to another. Institutional learning is actively promoted by the Open Method of Coordination
20 European Welfare States
(OMC) in employment policy, and by the ILO in its Global Employment Initiative. On the negative side, learning effects from the broader experience of EU integration have made Member States extremely careful in their negotiations on social policy. Policy learning over time has also made Member States wary of being bound to policies enforceable by the ECJ, the financial implications of integration (especially for those Member States that are net contributors to the EU budget), and the ability of the Commission to take what may appear to be a limited Treaty basis and use it to expand EU competencies in unexpected areas. All of these learning effects come in to play when Member States are considering the development of new forms of regulation. A vital function of institutions is thus in providing a stable setting for cooperation over time. By providing a forum for multiple iterations of cooperation over time, institutions allow Member States to learn what outcomes may be possible from the rules as they stand. Frustration with the resulting amount of cooperation from these rules will cause leader Member States to suggest, or at the least be disposed to, alternative forms of cooperation. Learning effects of policy cooperation will also work to empower resistant Member States as well. Over time, Member States have become increasingly familiar with the possible effects of social policy cooperation, most notably the cost of implementing EU legislation. Policy resisters will thus work to limit cooperation explicitly during Treaty reform, and demand forms of governance and specific policy outputs that will have limited implementation costs.
Financial incentives and disincentives The presence of funding in an institution, and the potential for members to access these funds, can act as a powerful incentive to cooperation. In the case of the EU, while budget lines associated with social legislation are minuscule in comparison with those that support agricultural production, they may in fact represent significant sources of funding for poorer Member States, especially when compared with national spending on specific policy areas such as active labour market policies (ALMPs). Financial incentives, in the form of funds to aid social development and cohesion, are necessary to explain the agreement of Greece to the SEA, for example. The availability of EU financing for the implementation of aspects of the European Employment Strategy (EES) is also a key institutional incentive that helps to explain gaps between expected and actual state behaviour.
European Welfare States and Supranational Governance 21
Alternatively, the funding of policies at the regional or global level can provide a disincentive to wealthy Member States who fear increasing contributions. The UK is especially effective in setting financial limits to the price of supranational governance, such as placing limits on the permitted financial impact of legislation on business. Germany, too, has placed explicit limits on the financial implications of new forms of governance. Unanticipated consequences One of the key elements of historical institutionalist theory is that actors are not fully aware of the implications of their choices, or the way future events will structure or force a reinterpretation of their choices. This means that while actors may be in a strong initial position, [and] seek to maximise their interests ... [they] nevertheless carry out institutional and policy reforms that fundamentally transform their own positions (or those of their successors) in ways that are unanticipated and/or undesired. (Pierson 1998, p. 30)13 Member States are assumed to be rational in terms of knowing their underlying preferences, and articulating these in national positions during bargaining. However, uncertainty about the relationship of actions to outcomes makes Member States ‘risk-adverse’, and as a result they may adopt policies that emphasize immediate collective gains over expectations of possible long-term payoffs (Knight 1992, p. 44). Evidence demonstrates that there is a high degree of uncertainty about the possible outcomes that result from placing social policy in EU Treaties, not least because of the leadership of the Ministry of Foreign Affairs in negotiations, and not Ministries for Labour, Employment, and Social Security, and because of a lack of expertise in particular Member States. In a ‘complex’ and ‘constantly changing’ bargaining environment, individual negotiators ‘act within the boundaries of their limited knowledge, capacities, and ways of action’ to select their preferred options (Stubb 1998, pp. 30–1).14 High degrees of uncertainty are identified, most notably in the Greek case across issue areas. There is also continued uncertainty with regard to the interaction and supremacy of EU versus ILO legislation in areas such as health and safety, both during negotiations over ILO legislation and at the ratification stage (Frid 1995). The volume of policies that exists at the EU level heightens the potential for unanticipated consequences. This is because as policy areas increase, as new forms of decision-making proliferate, and as the number
22 European Welfare States
of members of the EU increases, limits are placed on the ability of individual governments to control the evolution of policy (Pierson 1998, p. 39). The best example may be the evolution of employment policy, which was eventually included in the Treaties despite initial German and UK opposition because of the support of many smaller Member States, and because of the creation of a very flexible form of coordination that permitted resistant Member States to agree to contract. A great uncertainty about long-term consequences may be said to have existed when policies were initiated in 1957 with the Treaty of Rome, but policy-makers are increasingly sensitive to the logic of unanticipated consequences. One of the results of policies such as the Common Agricultural Policy (CAP) and the development of trade competencies by the European Commission has been the increasing sensitivity of policymakers to the consequences of placing policy competencies at the EU level; national policy-makers expect the European Commission, supported by ECJ Decisions, to be an activist policy-maker in those areas where it has competence, and to be a policy entrepreneur (Pierson 1998, p. 35). As suggested above, policy learning as developed over time has made Member States highly sensitive to issues of competence and implementation. Thus, national decision-makers with an interest in supranational policy-making in a particular area are faced with a dilemma: contract under circumstances that contain greater uncertainty as the time horizon extends, or exist in a state of non-contractual, non-formal relations in the absence of rules and enforcement mechanisms. Institutions are thus essential to explain how policy cooperation may begin, how formal rules are set down for cooperation that do not always reflect the positions of the most powerful Member States, and how policy outcomes result that are above the lowest common denominator. Institutions thus provide formal rules and learning effects that promote cooperation along certain paths, but which may also encourage the search for new forms of cooperation. Policy entrepreneurship by the European Commission and leader Member States, as well as the demands of resistant Member States, are also vital to explaining how new and more flexible forms of governance emerge. Over time, policy cooperation often expands in ways unanticipated by states.15
Third explanatory variable: when is ideology relevant? One other factor exogenous to a historical institutionalist framework must be considered in order to complete an explanation of gaps in expected and observed behaviour of Member States, namely the ideology
European Welfare States and Supranational Governance 23
of the political party in power towards EU integration more generally.16 Ideology is most relevant in the case of the UK. Even when national policies would suggest a preference for supranational cooperation, under Conservative governments since Thatcher the UK resists social policy integration for ideological reasons. Regime change in the UK, from Conservative to Labour governments, is vital to explain why the UK has agreed to cooperate in the social policy field in the period under study in this book. However, while the settlement area of the UK expands from non-cooperation to include possible cooperation under a Labour government, the overall pattern of resistance persists; Labour governments are still more resistant to EU competence in social policy than their continental Social-Democratic counterparts. The UK has consistently expressed one of the most resistant bargaining positions in the EU during both Treaty reform and policy-making, and at the global level is the most reluctant to see enhanced Commission leadership and a more robust relationship between the International Labour Organization (ILO) and the World Trade Organization (WTO). Notably, the left–right political divide on social policy or on EU integration is not highly relevant in explaining when Member States will cooperate over time except in the case of the UK. Governments of both the left and the right in EU Member States have acted as policy leaders as a result of deeply embedded national preferences on social policy that have emerged, in some instances, over more than a hundred years of history. Such national preferences are, in most instances, not susceptible to a large shift as a result of change in government. National policy histories are a more effective way of predicting national preferences in supranational social policy cooperation than the ideology of national governments. This explanation contrasts with liberal governmental explanations of national preference formation that emphasize economic interests, government ideology, and relative wealth and bargaining power of the Member State (Moravcsik 1991; Moravcsik 1997; Moravcsik 1998; Moravcsik and Nicolaïdis 1999). It will be demonstrated that, in the case of social policy, national policy histories are more effective in identifying policy leaders than intergovernmental theory that concentrates on government ideology at key moments of bargaining and the relative bargaining strength of Member States.
Method and sources Process-tracing is the method by which the explanatory variables named above may be clearly identified. Process-tracing involves examining how
24 European Welfare States
the policy issue was constructed by the actors that initiated it, outlining their preferences, coalition-building efforts, and bargaining positions, and subsequently tracing the development of the corresponding ‘governance regime’ in the policy issue area (Armstrong and Bulmer 1998, p. 56). This involves identifying the actors involved in the policy-making process, the national, regional and global level institutions through which these interests bargain and are mediated, the resulting legislation, the enforcement mechanisms and secondary legislation that develops, and national implementing measures. In order to verify the interests of various groups involved in social policy design, national peak employer and employee groups were interviewed. In order to verify positions in the bargaining process, interviews were conducted with national government negotiators and experts. Interviewees were asked about the role of the Commission relative to Member States, specific personalities in the negotiating process, and implementation issues. In this way, first-hand impressions of the role of the Commission relative to Member States were recorded, and the role of states versus institutions in influencing Treaty bargaining and policy outcomes was tested. In order to assure the anonymity of some of the interviewees, each interviewee has been assigned a number and is referred to by this number when cited. A full list of the interviews conducted and their corresponding number, some with name and exact position withheld, is in the Appendix. For interviewees working at the ILO, where institutional mechanisms as an explanatory variable are weaker than at the EU-level, the role of large Member States was examined as well as the ILO Secretariat. Interviewees also gave their accounts of experiences with the implementation of EU and ILO policies. It is hoped that these more detailed aspects of the research design will answer concerns that institutionalist methodology is often weak in verifying the role of institutions relative to other explanations (Peters 1999, pp. 145, 150). In addition to interviews, primary documents available in English and French were analysed.17 Primary sources consulted include: position papers and policy statements by national governments, social partners, the Commission and the Council; national government and Commission submissions to the Intergovernmental Conferences (IGCs), and to Presidencies (such as those to the Luxembourg Presidency summit on employment); minutes of Social Affairs Council meetings; internal correspondence (such as internal memoranda of the International Labour Office, and correspondence between the Council of Ministers and the
European Welfare States and Supranational Governance 25
Commission); official correspondence (such as between the ILO and the EU, or letters from UNICE to its member organizations); and Presidency conclusions.18 Secondary literature on social policy, national welfare state histories, national administrative systems and EU- and global-level institutions was also widely consulted, and is referenced. The main challenge posed by the use of historical institutionalism as a method is the complexity of tracing the simultaneous development of national, EU-level and global-level developments. There is an integrated story to be told about the development of supranational forms of social policy cooperation across national, regional and global systems of governance. Only through careful process-tracing and timing may we be able to discern whether a policy was influenced by EU or global legislation, or whether policies were already in the national pipeline and happened to coincide with regional or global initiatives. The repetition of these processes – historical analysis using secondary sources, interviews and primary document analysis – for each of the Member States under consideration, and at the EU and global levels, allows a cross-national comparison of preference formation, bargaining, and implementation. It is also hoped that the detailed research carried out will correct for the criticism that historical institutionalist analysis is ‘rarely based on a detailed, primary-source analysis of national preference formation’ (Moravcsik 1998, p. 491). The overall research design can also be represented diagrammatically, and is presented in Figure 1.1. National histories are examined for the purposes of identifying the underlying preferences of Member States in the social policy field (stage 1). An explanation of national positions based on national welfare state histories is then tested during the bargaining process to examine if national welfare state histories are sufficient to explain national positions, and the relative influence of states and EU-level institutions on bargaining is examined (stage 2). The outcome of Treaty negotiations is then assessed (stage 3). An examination of the process from Treaty reform to specific policy proposals and implementation is also examined so that the application of policy over time can be assessed (stage 4). This book seeks to demonstrate that by combining detailed information about national welfare state regimes with an analysis of the ways in which institutions present opportunities and incentives for cooperation, the creation and specific features of new forms of supranational governance can be explained. In essence, the overall research design marries national explanations for preference formation with an
Figure 1.1 Overall research design
National welfare state context Kinds of social policy in place (social assistance, insurance, and social regulation) Groups favoured by present social policy National configurations preferences Form of relations between unions and employers Standards for working conditions Employment conditions: minimum wage, working time, benefits, annual leave/ holidays, parental leave Financing of social Government policies ideology Levels of benefits and towards access to benefits EU integration
Stage 1: National preferences
Financial implications Policy entrepreneurship Implementation costs
EU institutions Formal rules: proposals of policy, actors involved, voting rules Institutional learning
+ Entrepreneurship
National positions during bargaining Leaders, resisters, and passive states
Stage 2: Positions and bargaining
ILO: Convention
EU: Treaty reform
OUTCOME
Stage 3: Treaty reform
+ Unanticipated consequences
Policy outcome over time: implementation Interaction of national, EU and ILO legislation
ECJ decisions National implementation, decisions, case-law
Committee decisionmaking, entrepreneurship Directives Framework agreements The EES and OMC
Policy-making
Stage 4: Policy-making and implementation
26
European Welfare States and Supranational Governance 27
institutionalist awareness of the processes by which formal rules and informal practices guiding decision-making may structure the choices available to governments for cooperative action. As a result of institutional features, policy choices emerge ‘that cannot be predicted from the preferences and relative power of the Member States alone’ (Pollack 1996, p. 430). Both actor preferences and institutions are taken seriously (Scharpf 1997, p. 1).19 It is to this task we now turn in the following chapters.
2 Governance by Law: Health and Safety in the Workplace
The regulation of health and safety in the workplace sets the tone for the permissible basis upon which competition can occur in a common market. Competition between states on the basis of health and safety standards in order to lower costs and attract investment and production is one of the key fears embodied in the social dumping hypothesis: under conditions of economic integration, high-cost countries will be forced to lower their standards in health and safety in order to stay competitive with low-cost countries (Eichener 1992; Mirka and Ruhm 1997; Blanpain 1998). EU1 legislation attempts to regulate the conditions under which productive activity takes place, and the tools such as machinery and protective equipment used in this activity, in order to create common minimum standards below which competition should not take place (Smismans 2001, p. 73). Lest this topic seem abstract, it is worth noting that in 1996 there were over 4.5 million accidents at work that resulted in more than three days’ absence from work. Just over 5,500 employees were killed in that year, concentrated in the construction, manufacturing, transport, and agricultural sectors. The cost to insurance schemes across the Union is estimated to be €20 billion per year (Agency 2001, pp. 8–9). This chapter demonstrates that policy leadership by Denmark was necessary in order to introduce new rules to facilitate decision-making in health and safety and overcome past policy failures. At the time of Treaty revision, the UK was the most resistant state. The UK finally agreed to qualified majority voting (QMV) in health and safety in order for it to realize gains in other issue areas. Resistance reasserted itself, however, at the policy-making stage. This chapter also traces how the institutional structure of the EU, most importantly an activist Commission, helped to structure agreement. Jacques Delors, the former French finance minister, 28
Governance by Law 29
accepted the post as the new Commission president in July 1984. During the negotiations Delors advanced an extension of QMV and an enhancement of EU social powers. Using the rhetoric of ‘balancing’ social policy against ‘the perceived centripetal effects of the single market’, the Commission and leader Member States were able to build a coalition of governments willing to make advances in EU-level social policy (Corbett 1998, p. 235). This pattern repeats itself in successive IGCs. This chapter also traces how institutional features of the EU, including decision-making by committee, and QMV, created a policy output well above the lowest common denominator once new competences were agreed. The Commission was a ‘key actor’ in the process of creating health and safety legislation at a high standard (Eichener 1992, p. 1). The Commission, as the manager of the governance process established at the EU level, created a ‘dominant position’ in the various consultative arenas that create policy proposals and the committees that debated the proposals (Eichener 1992, p. 56). This chapter focuses first on assessing the preferences of the Member States in health and safety regulation based on national histories in this policy area. The second section of the chapter examines the early history of health and safety policies in the ECSC (European Coal and Steel Community) and EEC. Given the weak powers of the Commission in the social field and the requirement of unanimity voting, legislation was slow in developing and clearly outpaced by technological innovation. The third section of the chapter details the introduction of QMV in decisionmaking for health and safety in the Single European Act (SEA). The enlargement of the EU to Spain and Portugal, as well as the 1981 membership of Greece, added a southern dimension to the EU where lower wages, social benefits, and health and safety standards were the norm. The creation and implementation of the Framework Directive on health and safety (89/391/EEC) detailed in section four, were greatly influenced by the desire of northern Member States to preserve national standards as a result of the southern enlargement. Although largely passive in the policy creation phase, southern welfare states may be said to reassert their national preferences for lower levels of regulation through poor implementation of the Directive.
Welfare state preferences in health and safety policy Differences in welfare state organization, rooted in varying philosophies regarding the division of responsibilities between the individual, the
30 European Welfare States
employer and the state, resulted in very different health and safety policies in the EU Member States. Relative levels of economic development also contributed to different standards in health and safety. First, there is a distinction between those states where public authorities have the leadership in setting general rules for industrial medicine and the health of workers, such as in France, Greece, Portugal and Spain and those states where sector-specific and very technical systems were developed, such as in Denmark, Germany and Sweden. In general, southern welfare state regimes are especially reliant on statutes to provide regulation of health and safety, whereas corporatist (with the exception of France) and social-democratic Member States rely in part on collective agreements that are often sectoral in nature to create rules. In the UK, general standards are set by national legislation, and much discretion is left to the employer to cover the specific risks of the workplace. Second, in terms of the financing of health and safety, the state funds health and safety measures from the central budget in the socialdemocratic and liberal welfare states, whereas the corporatist welfare states often partially fund health and safety from employer and employee contributions. Occupation-based insurance funds are also important in funding health and safety in Austria and Germany. As a result of these differences, benefits for injured workers vary widely between Member States, with low levels of benefits provided by liberal and southern welfare states, and more generous provisions provided by social-democratic and corporatist welfare state regimes. Third, social-democratic and corporatist welfare state regimes provide for worker participation in the development and oversight of health and safety policies. In liberal and southern welfare state regimes, legislation for worker participation covers large enterprises, but enterprises with no union organization may not give effect to this legislation. Much more discretion on the issue of worker participation is left to the employer in small and medium enterprises (SMEs). There were thus very different forms of organization of health and safety standards in the EC in the mid-1980s. These different organizational forms also correspond to a wide range of standards from high levels in Denmark, Germany, the Netherlands and Sweden,2 medium levels in Belgium, France, Ireland, Italy, Luxembourg and the UK, and low levels in Greece, as well as future members Spain and Portugal. Corresponding with these differences were varying welfare state regimes: Member States with social-democratic welfare state regimes have high standards in workplace health and safety and are expected to support raising standards via EC legislation so as to ensure that these standards are removed
Governance by Law 31
from competition; conservative-corporatist and liberal welfare states have relatively medium to high levels of standards and are expected to support the creation of EC standards so as to avoid competition on the basis of health and safety standards with the low-cost south. However, resistance to the creation of higher standards than those found in their national systems is expected. Member States with southern welfare state regimes are expected to resist high standards. More specifically, although improvements in worker health and safety may be welcomed by southern trade unions (as long as jobs are not threatened), the imposition of these standards by EC law and quick transition periods to higher standards will be resisted by southern governments and employers who bear the costs of implementation.
Health and safety regulation in the EU: 1952–85 The origins of the EU in the ECSC meant that health and safety in the workplace has a 50-year history at the regional level. The illness of miners as a direct result of their work underground, and deaths in the workplace due to industrial accidents in the coal and steel industries, were apparent even before the founding of the ECSC. Indeed, the ILO began its legislative programme with health and safety initiatives in the 1920s (covered in Chapter 5). Article 55 of the Treaty of Paris obliged the High Authority of the ECSC to encourage technical and economic research on the coal and steel industries, including issues relating to worker safety. The Treaty gave no authority for common policies to be enacted, however. The working method of the High Authority was ‘the revision of new safety techniques and preventative measures … [using] the traditional approach of international cooperation in social matters’ through research programmes, a scientific advisory committee, and an ECSC committee of worker and employer representatives (Collins 1975a, p. 87). A serious mining disaster in France 1956 in which 136 miners were killed (many of them Italian) led to the adoption of over 300 recommendations regarding mine safety and the creation of a tripartite Mines Safety Commission that included representatives from the UK, then not a member of the ECSC, and the ILO (ECSC 1957, p. 487). The Mines Safety Commission was to act by majority vote and was considered an independent intergovernmental organization, although its secretariat services were supplied by the High Authority of the ECSC (Collins 1975a, p. 84). After the death of over 300 miners in two separate incidents in Germany in 1962, the terms of reference for the Commission
32 European Welfare States
were expanded to include industrial health in coalmines, but no additional powers beyond the ability to issue recommendations were extended to it (Council 1965). In the Treaty of Rome, the European Commission was given the ‘task of promoting close cooperation between Member States in the social field’ on issues including the ‘prevention of occupational accidents and diseases’ and ‘occupational hygiene’ (ex art. 118, now covered by arts. 137 and 140). Cooperation was to be achieved under art. 118 via studies, opinions and consultation with national and international organizations working on health and safety. The process of creating goods and services was thus not subject to formal EEC law-making, but to a looser and more consultative form of information exchange and recommendations. In addition to art. 118, national restrictions on the trade of products created in the EEC were permitted on the basis of ‘the protection of health and life of humans’ in art. 36 (now art. 30). The creation of common EEC product standards so as to reduce potential barriers to trade based on human health permitted by art. 36 was thus the focus of EEC legislation proposed in the late 1970s and 1980s under art. 100 (now art. 94).3 Article 100 gave the Commission the competence to propose directives ‘for the approximation of such laws, regulations, or administrative provisions of the Member States as directly affect the establishment or functioning of the common market’ and required unanimity voting. Legislation before the SEA thus concerned the harmonization of product standards in order to achieve a common market and was proposed under art. 100, and not art. 118, although such laws had the effect of regulating the health and safety of workers using these products (Smismans 2001, pp. 73–4, 78). In 1962 an Industrial Health and Safety Division was established in Directorate General V of the European Commission (now DG Employment and Social Affairs). Recommendations on company medical services and on conditions for the recognition and notification of occupational diseases were the first soft law provisions – in other words, legally recognizable but not enforceable and lacking the power of possible sanctions in the case of non-implementation – on health and safety adopted by the EEC.4 However, Member States resisted Commission activism in the social field by rejecting Commission recommendations falling under art. 118. In addition, no meetings of the Social and Labour Affairs Council were held for close to three years, from late 1963 to 1966 (Gold and Matthews 1996, p. 4). The Commission was also criticized for making direct contact with ‘non-public authorities and for
Governance by Law 33
the creation of committees not directly specified by the Treaty’ (Collins 1975a). Constraints on Commission activity were compounded by the French boycott of the Council in 1965–6 (the ‘empty chair crisis’), and were only solved when the Luxembourg Compromise assured the right of any Member State to veto proposals in cases where vital national interest was at stake. External factors that caused the EC to re-examine the social dimensions of integration included the industrial disputes of the late 1960s, the 1973 economic crisis and its aftermath, and competition with Japan and the US that highlighted labour market rigidity in the EC (Sandholtz and Zysman 1989, p. 110; Vogel 1993, p. 67). Internal factors such as uneven growth in the EC and rising unemployment also contributed to a subtle shift in emphasis towards social aspects of economic integration. The Paris summit of 1972 stated that the heads of government ‘attached as much importance to vigorous action in the social field as to the achievement of economic union’ but did not enhance the weak regulatory powers of art. 118 (EC 1972). The Social Action Programme (SAP) followed in 1974 (Council 1974), and the Advisory Committee on Safety, Hygiene and Health Protection at Work (ACSH) was created to help the Commission develop specific legislative proposals. The first community action programme on safety and health at work began in 1978 with a series of directives proposed under art. 100 that focused on legislative harmonization in the area of product safety standards (Commission 1978). The aim of legislative harmonization, however, met with ‘resistance from employers and the majority of the Member States’ which, under unanimity voting, led to ‘a fairly rapid whittling down of the proposed action areas’ of the SAP (Vogel 1993, pp. 67–8).5 Legislation in the period 1970–85 – on safety signs, lead, asbestos, noise, and chemical, physical and biological agents6 – was thus at a low level of protection despite Commission activism. The creation of common EC technical standards also affected health and safety in cases of the use of dangerous products. Again, however, EC-level regulation on subjects such as electrical equipment, tractors and agricultural machinery, lifting devices and protective equipment was slow in development and was outpaced by technological change (Pelkmans 1987). The ‘New Approach to Technical Harmonization and Standardization’ was introduced in a Council Resolution in 1985 in order to aid in the completion of the internal market while taking into account ‘essential health and safety requirements’ (Council 1985). The New Approach distinguishes ‘areas requiring harmonization from those where mutual
34 European Welfare States
recognition of national regulations and standards suffices’ (Baldwin and Daintith 1992, p. 6), and delegated ‘regulation of technical specifications to private law standardization bodies’ (Eichener 1992, p. 19).7 EC legislation should thus provide only essential requirements ‘beyond which Member States’ rules ought to be considered as mutually equivalent’ (Nicolaïdis 1996, p. 5). This reorientation to mutual recognition in order to overcome the problems besetting approaches based on product standard harmonization is ‘perhaps the best example of policy learning taking place at both the national and European levels’ (Majone 1996b, p. 268). The 1988 revision of the Directive on chemical, biological and physical agents followed the New Approach. Still, the Commission could only propose legislation based on product standards, and not on the conditions under which products and services were produced.
Negotiations on qualified majority voting (QMV) in art. 118 of the SEA Background to negotiations: the coming enlargement and national governments The negotiation of the SEA took place in the context of impending enlargement to Spain and Portugal, set to occur on 1 January 1986. Several issues had already been brought to the fore as a result of the southern enlargement, such as the insufficiency of the Community’s financial resources, the distribution of the budget and the British rebate, and CAP reform. These issues were resolved in 1984 at the Fontainebleau summit. Other implications of the southern enlargement remained unresolved, however. Belgium, France, Denmark and Germany ‘feared the accession of Spain and Portugal in 1986 would lead to wider inequalities in social provision among Member States’ (Compston 2001a). Still, it was not clear that QMV held the key to the answer, since QMV could be used to lower standards just as easily as it might raise them. The coming enlargement was also a factor in the UK’s desire to see the Single European Market (SEM) approved, and the necessary institutional mechanisms created to ensure its completion by 1992. The new members were expected to be protectionist in their economic orientation, and thus the approval of the ‘1992’ project before enlargement was critical (Budden 2002, p. 88). The UK held out hope, however, that the SEM could be completed without calling an Intergovernmental Conference (IGC) on treaty reform.
Governance by Law 35
The completion of the SEM was a further reason for high-standard countries to fear that social standards would be eroded, as ‘national protective regulations, which functioned as trade barriers, would be removed by the SEM’s deregulatory measures, leaving a regulatory vacuum that would lead to community-wide competition for the reduction of cost-intensive safety and health standards’ (Eichener 1992, p. 2). However, although specific interests in Member States, such as trade unions in Denmark and Germany, expressed fears about the negative effects economic integration would have on social policies, Member States did not push for widely expanded EU competence in the social policy field. As will be seen in subsequent chapters, social policy is never the main focus of Treaty negotiations. Rather, the SEM and EMU, as well as institutional reform, have been the main focus of successive negotiations.8 The other vital background to the negotiations was the majority of right and centre-right governments in the ten Member States, listed in Table 2.1. An intergovernmental analysis of the negotiations would predict resistance to new social policy competences given the dominance of parties of the right, and also predict resistance on the part of the poorest Member States, namely Greece and Ireland.9 Intergovernmental theory does a good job, therefore, of identifying the most resistant Member States on the issue of QMV – namely Greece because of its relative poverty, and the UK because of its Conservative government. However, intergovernmental theory cannot identify the policy leader, Denmark, which had a government of the Right yet eventually pushed for an extension of QMV to health and safety policy. A consideration of the Danish proposal for QMV under art. 118 is missing from Moravcsik’s study of the SEA, and Moravcsik repeatedly states that Denmark was opposed to QMV (Moravcsik 1998, pp. 366, 376). In addition, an intergovernmental analysis cannot explain the high standard of regulation that later resulted from policy-making under art. 118a. A welfare state explanation of preferences that emphasizes the particular features of the policy area under consideration does a good job of identifying leaders in this case of Treaty reform. However, a welfare state explanation cannot tell us why policy resisters eventually agreed, or how the features of specific policy proposals emerged. Institutional factors, such as the complex bargaining environment during Treaty reforms, financial incentives, and Commission entrepreneurship at key moments, are needed in order to explain outcomes, as is government ideology opposing EU integration in the case of the UK. Over time, the
36 European Welfare States Table 2.1 Member State political affiliations, 1985 Member State
Rich or poor*
Party at head of government
Head of government
Political alignment of government: right (Conservative, Gaullist, Christian Democrat), or left (Socialist, Labour)
Belgium
Rich
CVP
Wilfried Martens
Right
Denmark
Rich
KF
Poul Holmskov Schlüter
Centre-Right
France
Rich
PS
President Mitterrand
Left
PS
PM Laurent Fabius
Germany
Rich
CDU
Helmut Kohl
Right
Greece
Poor
PASOK
Andreas G. Papandreou
Left
Ireland
Poor
Fine Gael
Garret FitzGerald
Centre-Right
Italy
Rich
PSI
Benedettino Craxi
Centre-Left coalition
Luxembourg
Rich
CSV/PCS
Jacques Santer
Right
Netherlands
Rich
CDA
Ruud Lubbers
Right
UK
Rich
Conservatives
Margaret Thatcher
Right
* 1989; determined on the basis of GDP per capita at PPP (purchasing power parity); rich ⫽ above average; poor ⫽ below average. Source: Eurostat (2001, p. 24).
unanticipated consequences of the changes made to art. 118 also expanded regional competence in social policy in ways not predicted at the time of agreement. The IGC to amend the Treaty of Rome was called at the Milan European Council in June 1985.10 The IGC meetings started in September 1985 under the Luxembourg Presidency, that, although Conservative, was clearly a pro-integrationist government. The final agreement was reached at the Luxembourg European Council in December 1985 and signed by the foreign ministers in February 1986. This tight schedule of negotiations was necessary, in addition to the enlargement of 1 January 1986, because of national elections scheduled for early 1986 in France, the Netherlands and Spain.
Governance by Law 37
Debates over QMV: the splitting of the minimalist camp The Dooge Committee Report on Institutional Affairs to the June 1985 Council in Milan was very similar to the 1984 European Parliament Draft Treaty that suggested QMV in all areas of ‘common action’. However, it was clear that there was a split at the start of negotiations between the original six plus Ireland and the Commission, who generally agreed with this formula, and the UK, Greece, and Denmark, who were opposed to a formal extension of QMV. The UK, Greece and Denmark therefore represented a minimalist view of the extension of QMV. The Commission had come out early, at the Milan meeting, in favour of QMV as a general working method. Germany had favoured extending QMV since its 1984 preparations for the Dooge Committee, but also wished to retain a veto for ‘the government with the highest standards’ in cases of QMV under art. 100 (Moravcsik 1998, pp. 331, 366). France also supported a limited extension of QMV, and Mitterrand had expressed support for a ‘European Social Area’ that would include common rules on issues such as working time. Notably, however, the FrancoGerman draft Treaty of June 1985 proposed QMV under art. 100, as well as for ‘articles governing services, transport, relevant external tariffs, capital movements, and right of establishment’, but not for social policy (Moravcsik 1998, p. 365). The resistant Member States on QMV were, at the start of the negotiations, Denmark, Greece and the UK. The conservative Danish government was opposed to QMV over fears of a loss of sovereignty, and an unwillingness to delegate competence to the EC. The Greek government also wished to retain its veto. Greece preferred unanimity voting in order to avoid being subject to the higher standards of the other Member States that could well result if QMV was introduced. The Greek prime minister and leader of the PASOK socialist party, Andreas Papandreou, had been against EC membership. Now, his main negotiating aim was to strengthen the redistributive elements of the EC while retaining Greece’s veto on new policies (Lavdas 1997, pp. 170–6). The UK’s main aims in participating in the IGC was ‘to export its global “neo-liberal” deregulatory philosophy to the European level’ while avoiding the extension of QMV to areas including social policy (Budden 2002, p. 86). Thatcher was an ardent anti-integrationist and opposed to QMV, but she was also convinced of the economic benefits the SEM held for the UK (Thatcher 1993, p. 553). UK Foreign Secretary Geoffrey Howe, on the other hand, ‘advocated positive but pragmatic engagement in the IGC’ to ensure the SEM and enhanced political cooperation (Budden 2002, p. 79), and was ‘willing to settle slightly above the lowest common
38 European Welfare States
denominator’ in return for these goals (Budden 1994, p. 302). The tension in the UK position was thus clear from the beginning: how was sovereignty, symbolized by the national veto, to be preserved at the same time as pushing for QMV, which was necessary to ensure the completion of the SEM?11 The clearest supporter of a more robust social policy throughout the negotiations was the Commission. The Commission presented a package linking the SEM with three flanking policies: monetary cooperation, technology policy and ‘communautaire’ policies that were in essence economic and social cohesion (Commission 1985, p. 10). The accompaniment of economic and political union by a European form of solidarité – in this case through cohesion funding for less developed Member States – was a constant leitmotif of the Delors Presidency of the EU (Interview 33). The Commission also proposed a ‘new art. 118b providing for a social dialogue with the possibility for contractual relationships, at European level’ (Corbett 1998, p. 238). Other accounts of the SEA negotiations give credit for the introduction of art. 118b to the French delegation (De Ruyt 1989, p. 192), although the close relationship between Delors, the former French minister of finance and now president of the Commission, and the Mitterrand government may explain the confusion over the introduction of the proposal. Vague Commission proposals to include reference to a ‘European social space [espace social européen]’ in art. 117 were rejected (De Ruyt 1989, pp. 192–3). The European Parliament also insisted on an improvement of its role in decision-making, threatening to reject the 1986 draft budget, fire the Delors Commission, or withhold its opinion on the IGC if no advances were made (Budden 2002, p. 93). The new cooperation procedure was approved in the SEA, as governments were especially anxious to avoid reopening budgetary debates that had poisoned the EU atmosphere in the early 1980s. However, it was Denmark – initially resistant to an expansion of EC competencies and under a conservative government – that proposed an expansion of QMV to art. 118. Despite a resistance to the expansion of European competences more generally, and a strong desire to preserve state sovereignty, Denmark was a leader in health and safety policy as a result of its well-developed national preferences for high standards in this field of social policy. After national consultations the Danish government, up until then a consistent minimalist on future cooperation, made a surprising contribution to the IGC in October 1985 ( —— 1985, pp. 3–4). The Danes suggested a chapter on ‘Labour market and working environment’.12 A revised art. 118 would permit the adoption of
Governance by Law 39
Directives on minimum health and safety standards by a Council vote of an unspecified majority. The focus of the proposed Article was not the equalization of working conditions, but rather the creation of minimum standards (Baldwin and Daintith 1992, p. 11). Member States with higher standards would be allowed to retain them. The Danes were supported in this initiative by the French (De Ruyt 1989, p. 73). Only the UK explicitly opposed the Danish proposal for art. 118, and neither the British Labour Party nor the TUC mobilized to back a stronger EC social policy. However, while maintaining a reserve on social policy, the UK tried to limit the damage of possible majority voting on social policy by splitting art. 118 into those areas where majority voting would be permitted – art. 118a on health and safety – and those where it would not be (Falkner 1998). The Luxembourg Presidency linked the acceptance of the Danish proposal to approval of the SEM, and, in an interesting turn of events, two of the initially most minimalist states now had opposing interests: the Danes wanted QMV on health and safety to be approved as a condition of their approval for the SEM, while the UK was opposed to any extension of QMV in the social field, seeing labour market policy in isolation from completion of the SEM. The Danes also proposed that the Council monitor unemployment, but these suggestions for a modification to art. 122 were later dropped by the Luxembourg Presidency only to emerge 10 years later in discussions on the Treaty of Amsterdam (Corbett 1998, p. 238). By November 1985 the Danes had stated their intention to adopt art. 118a Directives by QMV. Danish trade unions, long proponents of the retention of the national veto, supported QMV once it became clear that other Member States also wanted to stress the importance of the social dimensions of integration, and that the ECJ would rule to uphold national practices such as binding collective agreements. In addition, the ‘unilateral UK rejection of determining even minimal basic rules in the labour market field’ (Norgaard 1992, p. 210), and the ‘unsatisfactory experience with harmonization negotiations’ over product standards in the early 1980s made it clear that QMV would facilitate more efficient decision-making (Moravcsik 1998, p. 317). Under new rules for QMV, a blocking minority could not be formed by Greece, Ireland, Portugal and Spain. Only with the help of the UK could the poorest and lowest-standard Member States form a blocking minority. The smallest successful blocking minority would consist of at least three Member States (Spain, the UK, and one of either Greece or Portugal).13 Although the UK continued to reserve its opinion on art. 118, the UK clearly took a pragmatic view of negotiations. Seeing the other
40 European Welfare States
Member States lined up in favour of QMV, UK negotiators made further attempts at damage limitation. The UK insisted that Directives undertaken under art. 118a ‘shall avoid imposing administrative, financial, and legal constraints in a way which would hold back the creation and development of small and medium-sized undertakings’. Convinced that UNICE would continue to block substantive agreements from arising, the UK did not seem to have mounted much opposition to the weak Social Dialogue procedures supplied by art. 118b. In addition, the new art. 100a (now art. 95) to prevent distortions of the internal market permitted QMV, but excluded from QMV ‘fiscal provisions, those relating to the free movement of persons’ and ‘those relating to the rights and interests of employed persons’ (art. 100a, para. 2). This clause in art. 100a was originally suggested by Belgium and the Netherlands, and later supported by Germany. The UK was not the only Member State hesitant to accept the art. 118 provisions. German acceptance of QMV in health and safety ‘owed much to limits on EU legislation on other parts of the Agreement’ (Rhodes 1999, p. 139), including a consideration of the needs of SMEs, the subsidiarity clause, and the art. 100a provision that the Commission would take ‘a high level of protection’ as its standard in proposals. Even more problematic than German hesitancy was stubborn Greek resistance to approve the whole package of negotiations without an agreement on the transfer of resources to poorer Member States. Andreas Papandreou had first raised the issue of funding for an Integrated Mediterranean Programme (IMP) as a condition of Greek approval for the southern enlargement (Thatcher 1993, p. 545).14 The ‘Papandreou Memorandum’ envisioned a significant transfer of resources to Greece, and a number of derogations from the acquis communautaire (De Ruyt 1989, p. 197). Funding of economic and social cohesion in the EC became ‘more than a simple side-payment’ in negotiations, ‘for it appeared to be the one issue on which Athens might use its art. 236 veto’ on the IGC (Budden 1994, p. 307).15 Ireland jumped aboard the Greek bandwagon, and proposed specific changes to the Treaty that would make the impact on poorer Member States and regions a consideration in all common EU policies, without specifically demanding transfers from the richer states. As Budden points out, Greece and Ireland ‘revealed their settlement areas’ – namely cohesion in exchange for their acceptance of the SEM and QMV – and in doing so divided themselves from the lowest-common-denominator position represented by the UK position (Budden 1994, p. 316). The strategy seemed especially risky given German concerns about increased funding, since the southern enlargement would further increase the number of cohesion states.
Governance by Law 41
The eventual compromise reached on cohesion was that the structural funds (the European Social Fund, the European Agriculture Guidance and Guarantee Fund, and Regional Fund) would be increasingly coordinated, their decision-making structures gradually changed from unanimity to QMV, and the financial principles guiding their payouts targeted towards cohesion. There was also an implicit recognition of the need for a ‘substantial increase’ in the structural funds at the next budget revision (Corbett 1998, pp. 235–6). Despite the insistence of participants that the SEA was approved without issue linkages (Cockfield 1994, pp. 45–6),16 it seems more likely that an implicit forward linkage was made with structural funding and that Greek demands for an absolute increase in the structural funds was left for future debates (Budden 1994, pp. 346, 400). Others have put this linkage more bluntly, describing it as a ‘payoff from the rich north to the poorer south for its sustained participation in the project’ (Sandholtz and Zysman 1989, p. 121). Calling the cohesion funding a ‘side-payment’ is problematic, however. If funds further economic development, and help pay for the social costs of modernization, their creation is better considered a central and legitimate aim of the Community, rather than a ‘side-payment’ or side issue. Such funding was more likely to be considered a key payment or necessary payment by poorer countries, rather than a ‘side-payment’ as reflects the interests of the majority of rich Member States. For its part, the UK finally agreed to QMV under art. 118a only at the very end of negotiations in February 1986. The UK agreed for two reasons. The first is that its primary negotiating objective was achieving the SEM. In light of the compromise reached on the vast majority of British objectives for the SEM, throwing the deal away on the issue of social policy was clearly not in the UK’s interest. As one British observer noted, Mrs. Thatcher accepted the Delors package [‘1992’] not because she was enamoured of its immediate returns but because over the long haul she believed that the overall outcome lay well within its [the UK’s] settlement area. (Wallace 1990, p. 225) Second, it is clear that the implications of art. 118a were underestimated. To the British government in the Thatcher era the Single Act was little more than a vague expression of intent and the procedures were the mechanism provided to frustrate that intention. (Cockfield 1994, p. 66) With QMV limited to health and safety, and with SMEs explicitly sheltered from costly obligations, it seemed certain that the UK could avoid interference in social policy from the EU level. Still, given that the
42 European Welfare States
primary UK working document for the SEA negotiations, ‘Europe – The Future’ did not contain any mention of social policy, and given the consistent UK opposition to any expansion of EU competency in the social policy field, the extension of QMV in art. 118a is clear evidence that the results of the SEA were higher than the lowest common denominator as represented by the UK position. The reluctance to view the agreement as a path-breaking moment was symbolized by the decision to call the final agreement the ‘Single European Act’ rather than the ‘Treaty of [or on] European Union’ as suggested by earlier drafts (EC 1987). Perhaps the more interesting question is not why QMV was extended to art. 118a, but rather why QMV was not extended to other areas of social policy in light of the southern enlargement. Social policy had few champions at the IGC. Most surprisingly, Socialist France did not present social policy proposals, highlighting the inconsistency between President Mitterrand’s rhetoric on this subject and French inaction. Given the much higher standards of the 10 EC Member States in 1985 (with the sole exception of Greece) than those of Portugal and Spain, the limited areas in which QMV was extended illustrate that the desire to prevent social dumping by exporting the costs of labour market regulation to new Member States did not drive the negotiations. It is reasonable, then, to suggest that given the already high level of British health and safety standards, very little ground had been conceded by the UK in placing QMV in art. 118a for health and safety. Thatcher herself felt that she had ‘surrendered no important British interest’ (Thatcher 1993, p. 555). However, over time, the little ground that was conceded opened up an area of unanticipated proportions.
The Framework Directive on the introduction of measures to encourage improvements in the safety and health of workers at work (Directive 89/391/EEC) Shortly before the SEA, one of the most insightful writers on the EU characterized it as being in a ‘joint decision trap’ that would prevent effective policy-making (Scharpf 1985; Scharpf 1988). This inaction was the result of unanimous voting rules under which governments, notably the UK in cases of social policy, could veto any agreement that did not suit their national interest (Budden 1994, p. 156). Concomitantly, the delegation of powers from the national to the EU level and the pressures of international competition also rendered national policy-making
Governance by Law 43
impossible, hence the ‘joint decision trap’. Further, even if some Member States could preserve their own national standards, at the EU level ‘a roll-back, at least a long-term standstill of innovation and improvement in the area of social regulation was expected’ (Eichener 1992, p. 5). At most, standards were expected to reach a lowest ‘common denominator of the nine or ten states needed to form a qualified majority’ (Eichener 1992, p. 33). As a result of the SEA a limited window of QMV had been opened up in health and safety legislation. The SEA ‘was intended to break the logjam on social policy by clarifying the Community’s competence to adopt social legislation and by extending … qualified majority voting in the Council’ (Deakin 1997, p. 122). Would this new rule in decisionmaking help to overcome the trap and prevent standstill? And, if the trap was overcome, would decision-making occur at a low level, as the social dumping hypothesis would predict? Creation of the Framework Directive In November 1987 the Commission submitted its initial draft proposal for several Directives in the health and safety field to the Advisory Committee on Safety, Hygiene and Health at Work (ACSH). This committee is composed of workers, employers, experts, government officials and representatives from national standardization bodies. German delegations may also include a representative of the Länder governments (Eichener 1992, p. 28). The Commission’s draft Directive was ‘consistent with Nordic practice’ in going beyond the merely physical and technical aspects of work (Johnson 1996, p. 195). Initially full of proposals on a wide range of subjects, the text was eventually pared down to a Framework Directive to improve health and safety in the workplace, while other issues were hived off to become separate draft Directives.17 Because of the inadequacy of the Commission’s initial texts, the Workers’ Group of the Advisory Committee put forward several amendments that were ‘partially incorporated’ into the first official proposal (Vogel 1993, p. 74). According to an ETUC document, the workers’ group suggested several amendments: a more robust concept of prevention that would anticipate risks rather than merely try to make some risks safer, a definition of health that included physical and psychological factors, and a clear statement on worker consultation in the workplace on health and safety issues (ETUC 1988). The ETUC thus desired a Directive aimed at a high level, with a broad definition of workplace health and safety similar to the Nordic conception. The ETUC Secretariat kept its affiliates up to date on the progress of the legislation after the
44 European Welfare States
advisory committee stage, although its influence declined after the advisory stage (ETUC 1989). In 1988 the Commission and EFTA (European Free Trade Association) states devoted funds to create the Trade Union Technical Bureau (TUTB) to provide research and observation capacity to the ETUC. TUTB officers also observe EU and CEN committees and may coordinate the work of national trade union experts (Eichener 1992, pp. 85–6). UNICE, for its part, aimed for a Directive at a medium or average level of regulation with a minimum amount of specific obligations for employers. UNICE was heavily influenced in its statement on the Directive by the CBI, which held the chair of the Health and Safety Working Party of UNICE. Examples of the CBI’s influence on the UNICE statement included the suggestion of British legal wording for the text, such as prevention ‘as far as is reasonably practicable’, and a focus on the cost of the legislation to SMEs. UNICE viewed articles on the responsibilities of the employer and preventative services as too detailed for a Framework Directive, especially the obligation of employers to work to prevent health and safety risks by appointing a worker or external service to review potential risks (Vogel 1993, pp. 79–84). UNICE suggested the elimination of these articles and their replacement with a simple statement on employer responsibility that did not lay down any specific methods (UNICE 1988).18 Notably, the detailed legislation suggested by the Commission remained in the proposed Directive despite UNICE opposition. The other highly controversial issue was worker information and participation in health and safety protection, resisted by UNICE because of fears about worker consultation procedures entering EU policies via the back door of health and safety provisions. UNICE agreed to worker participation on the basis of ‘current practice in the Member States’ (UNICE 1988). The wording ‘in accordance with national laws and/or practices’ was adopted, in effect freezing information and consultation at national levels. Other issues in addition to worker information and consultation were also left to ‘national laws and/or practices’, such as the form of reporting of occupational accidents (art. 9, para. 1(d)), the right of employees to appeal ‘to the authority responsible’ in cases where they consider the measures taken to ensure safety to be inadequate (art. 11, para. 6), and health surveillance (art. 14, para. 1). The most controversial issues after the drafting stage remained the obligation of employers and workers for health and safety at work, and the liability of employers. The Directive states clearly the responsibility of the employer to follow the legislation set out for his workplace by national and EU regulators, to be informed about the latest practices in
Governance by Law 45
health and safety, to improve his workplace, and to anticipate and prevent injuries in his workplace. The standard of health and safety required is the latest state of the art,19 and not simply the articulated standards. For UK and Irish legal systems, where ‘legal requirements are generally interpreted literally’ and where the duties of employers are qualified by a ‘so far as reasonably practicable’ clause, such detailed prescriptions were resisted (Baldwin 1992, p. 219).20 Owing to a UK government demand, the final text permits Member States to ‘provide for the exclusion or limitation of the employers’ responsibility where occurrences are due to unusual and unforeseeable circumstances, beyond the employers’ control, or owing to exceptional events’ (art. 5, para. 4). Also as a result of UK insistence, the Commission proposal for the Framework Directive was accompanied by a ‘Fiche Financière’ and ‘Fiche d’impact’ (Commission 1988a). Surprisingly, Greece was very passive in the process of Directive design. The Greek Presidency in the second half of 1988 defined as its priorities the implementation of arts. 118a and 118b concerning the health and safety of workers and the promotion of the social dialogue (Rapacciuolo 1990, p. 123). This indicates that despite the acrimony over QMV and the higher standards of the rest of the Community, Greece was willing to recognize, at least rhetorically, that the EC had competence to legislate. Importantly, the moment of high politics had passed; although at the IGC Greece had threatened a veto, Greek actions in technical discussions may be described as passive. Explanations for this passivity are not rooted in the welfare state explanation of preferences, which clearly indicate that Greece should resist the imposition of higher standards that would entail heavy costs for its agricultural and industrial sectors, and the imposition of standards without long transition periods. Instead, the explanation for this passivity rests in both Greek administrative culture and the consensual style of EU committee decision-making that dominates the creation of technical legislation, including that for health and safety. Greek administrative culture does much to explain Greek passivity. The politicization of the staffing at the Greek PERMREP in Brussels (rather than staff selection through a merit-based civil service) means that officials often do not possess relevant expertise.21 Insufficient ‘preparation of policy alternatives’ and negotiation strategies all mean that Greek influence on draft Directives is limited in highly technical areas (Spanou 2000, pp. 170–2). The lack of English and French language skills also hampers intervention; when officials make interventions in Greek, in the words of one official who has participated in a number of committees,
46 European Welfare States
‘the meaning is often lost’ and dual translation makes ‘the meaning so diluted as to be pointless’ (Interview 16).22 Finally, poor domestic policy consultation in Greece means that despite the clear upgrade in standards that EU legislation represented for Greece, employers did not register large-scale opposition to the health and safety legislation as it was being negotiated. SMEs in particular were not well informed about the negotiations, or about the resulting law. A second element in Greek passivity is the consensual style of decision-making in EU committees that rewards technical expertise and often ignores unconstructive opposition. The personalities of individual committee members and the experience of cooperation over time – in which case some members may be termed ‘leaders’ and others ‘obstructionist’, while others are viewed as ‘disinterested’ – gives committee decision-making a unique dynamic that is not able to be reduced to national bargaining power as expected by an intergovernmental analysis (Eichener 1992, p. 66). Committees are not ‘little Councils’ where national voting power is reproduced; rather, charismatic and innovative persons from the Commission and national delegations can have a great influence on the outcome, as in more than 50 per cent of advisory committees formal voting does not take place (Eichener 1992, pp. 54–5). Specifically, delegations from ‘technologically advanced countries have an advantage, because they usually have a greater potential for expertise than the low-level countries’ (Eichener 1992, pp. 51–2). In this highly technical setting, best practices, even from then non-EU Member State Sweden, were absorbed in the policy-making process when introduced by Commission officials or national delegates looking to upgrade their own national systems via Brussels (Pierson 1998, p. 54). While Member States vary greatly in their influence in committees, passivity is a general feature of states with a southern welfare state regime, suggesting a link between relatively underdeveloped welfare states and administration more generally. A 1989 study of comitology proceedings indicated that French, German and UK delegations had the most influence, Denmark, Italy and the Netherlands had a medium level of influence, and that Greece, Portugal and Spain had the least (IEP 1989). Compared with their populations, the middle-influence countries, especially the Netherlands and Denmark, have disproportionately high levels of influence, Italy and Spain lower ones (Eichener 1992, p. 52). This pattern was reproduced in the negotiations on the machinery Directive that also took place in 1987–8. Greece did not even participate in the CEN committee that drew up safety standards for machinery, and Italy also had very low representation in this instance
Governance by Law 47
despite being the second-largest manufacturer of machinery in the EU (Eichener 1992, p. 68). The disproportionate influence of Member States carries on at the level of PERMREP discussions. An analysis of the minutes of nine Social Affairs Group meetings on the draft Framework Directive from March to November 198823 demonstrates several of the trends suggested so far: leadership by high-standard countries and the Commission, southern passivity, and UK and Irish resistance alongside effective national interventions so as to lower the level of agreement. The minutes of nine meetings of the Social Affairs Group were analysed by counting the number of reservations expressed by Member States and the nature of these reservations (whether in order to object to the inclusion of text or to request further clarification), and also the number of proposals made in the form of suggestions for new or differently worded text. The UK, Denmark and Ireland expressed most of the reservations, although in many instances Danish reservations demanded clear definition of terms whereas the UK and Ireland continued to reserve on the employer responsibility, prevention system, and worker consultation clauses. The UK and Ireland also submitted the most proposals for alternative texts, including a joint text on the insertion of a clause regarding employer responsibility ‘as far as is reasonably practicable’ in order to water down the detailed provisions of this chapter of the Directive (Ireland 1988). The UK and Ireland were not opposed to worker information and consultation, but wanted a very specific and circumscribed article and made many proposals for this chapter. Belgium, for one, rejected that the UK and Irish legal systems were so different from continental systems that they could not accept the articulation of minimum standards, and the UK and Irish positions on employer responsibility were not accepted in the Social Affairs Committee. Denmark, in particular, pushed for a role for the social partners in the Directive, although ultimately the chapter on worker information and consultation endorsed existing national practices. The Commission was also very active in proposing alternative texts. The Commission proposed more alternative texts than all other Member States, except the UK and Ireland, in most instances presenting compromise texts. The EP under the cooperation procedure also worked to expand the definition of preventative services, and attempted to define a common form of worker participation, although in the end the result was a reference to national systems (Vogel 1993, pp. 81, 83).
48 European Welfare States
Greece and Portugal were by far the most passive Member States. Even in the number of reservations expressed (as differentiated from presenting alternative texts), Greece only expressed reservation on 5 occasions, and Portugal on 3. Spain expressed reservations on 12 occasions and Italy on 31, especially in the case of Italy as the discussions reached the final stages. Italy proposed the one explicit limit on scope that was adopted in the Directive, the non-application to ‘domestic servants’ (art. 3a). Still, even the most active southern Member State in terms of reservations (Italy) expressed only half as many reservations as the UK did. Notably, however, Spain was the most active of the southern Member States in proposing alternative texts. Spain proposed as many alternative texts as Germany, marking it clearly in the middle range of influence in suggesting new texts after the UK, Ireland, Belgium and Denmark. Greece, Italy and Spain only became active as a group in proposing additions to the text when pushing for a delay of five years in the application of the Directive. They were joined in this proposal by the UK. The problem of applying the Directive to SMEs was given as the reason for the delay. Ultimately, the Directive was to be applied by 31 December 1992, a delay of three years. In later Directives, such as 89/ 654/EEC on minimum safety and health requirements for the workplace, the Directive was to be applied three years after signing, yet Portugal negotiated for an extra year and Greece negotiated for a fiveyear derogation. The final Directive provided a broad definition of the working environment, including working time and social relations at work, not simply the machinery and equipment used during production and traditional safety and health considerations such as fire exits and cleanliness (EEC 1989). An example is the relief from monotonous work provided in art. 6(2d). The concept of occupational health ‘is extended to cover the protection of both physical and mental health’, including stress (Eichener 1992, p. 6). Article 6 on ‘General obligations on employers’ contains detailed obligations for employers on prevention of risks, evaluation of risks, adapting the work to the individual worker, adapting to technical progress, and replacing dangerous forms of work with the ‘non-dangerous or less dangerous’ forms. Article 6 also details risk assessment and worker consultation on the introduction of new technologies and dangerous work. Member States were left to define the size of the undertakings (most probably SMEs) where the employer rather than a worker representative could be responsible for protective and preventative services (art. 7).
Governance by Law 49
The Directive was passed unanimously. QMV was thus not the direct instrument that secured passage of the Directive, but rather QMV provided the indirect threat that forced Member States to the negotiating table to work out mutually agreeable outcomes (Hayes-Renshaw and Wallace 1996). The development of health and safety Directives in EU committees – first the tripartite Advisory Committee on Safety, Hygiene and Health at Work and in later stages in the Social Affairs Committee composed of national officials – means that the influence of active and resistant Member States is countered by constant Commission entrepreneurship and leadership by high standard Member States.24 Passivity on the part of low-standard southern Member States also permits agreement at a higher level than would be expected on the basis of relative bargaining power, or by a social dumping hypothesis. In specific instances, EU health and safety regulation is not higher than national regulation, but the overall standards adopted are similar to Sweden, Denmark, and the Netherlands, and higher than that of Germany (Eichener 1992, pp. 6–8).25 The standards supplied in the Framework Directive are thus ‘capable of being constructed as affording protection at a very high level, certainly not a minimum one’ (Baldwin and Daintith 1992, p. 11). In this case, while Member States give up sole regulatory authority at the national level, the power to set and enforce standards is not lost but rather enacted through reregulation at the EU level (Majone 1996a). Table 2.2 summarizes the development of health and safety legislation from QMV to the creation of the Framework Directive. National implementation of the Directive Article 249 (ex 189) of the Treaty states: Directives are ‘binding, as to the result to be achieved, upon each Member State to which it is addressed, but shall leave to the national authorities the choice of form and methods’. Implementation for all EU legislation thus relies on Member States, who must transpose Directives into national law and who must, in the first instance, ensure compliance with EU law. The ECJ may rule on national lacunae in implementation only if the case is referred from a national court or brought to it by the Commission. The first stage in the implementation of a Directive is the transposition of the EU legislation into detailed national law. Member States may choose to transpose Community health and safety Directives via a number of different legal forms, including statutes, collective agreements, and industrial injury mutual insurance associations in the German case (a list of the different forms of implementing methods used by Member States is in the endnotes).26 By the early 1990s, however, not even
50 Table 2.2 Factors determining the evolution of health and safety policy at the EU level Time
SEA negotiations 1985
Design of Framework Directive 89/391/EEC
Leader member state and interests
Denmark
ETUC, Denmark, Belgium
Resistant member states and interests
UK, Greece, UNICE
UK, Ireland, Spain, UNICE
Commission, entrepreneurship
Supported QMV for all of art. 118. Proposed new art. 118b on social dialogue (approved). Proposed cohesion policy (approved). Proposed art. 117 on a European social space (rejected).
Adopted a broad definition of health and safety, a robust list of preventative requirements, and pushed for the broad scope of the Directive and explicit employer responsibilities.
Unanimity had blocked proposals in Council. QMV was proposed for art. 118, then limited to health and safety by negotiations. QMV was also to be applied to the European Social Fund (ESF).
QMV formally available in the last instance, but development of the Directive by debate and attempt to reach consensus in committees reduced the influence of low-standard states.
Factor
Other institutional factors: Rules
Learning effects in policy design
Need for QMV in light of slowness to develop Directives, UK opposition under unanimity voting, and southern enlargement.
Financial implications
Limits on application of H&S legislation to SMEs. Structural funding. Postponed debate on ESF to future.
Cost assessment to be carried out by the Commission. Size of undertaking taken into consideration in the Directive.
Unanticipated consequences
Creative use of art. 118a by Commission. Extension of QMV to other areas of art. 118 (covered in concluding section of this chapter).
Created a high level of health and safety requirements, rather than the lowest common denominator.
Governance by Law 51
this first stage of implementation had occurred in all Member States. As the author of a comprehensive review of the implementation of the Directive in the early 1990s wrote: By the deadline prescribed for the entry into force of the framework Directive – unanimously agreed to by the States concerned – fewer than half the Community States had legislation on the statute book to transpose the framework directive, and it will be seen that even these legislative instruments generally fail to satisfy the basic legal criteria laid down in the case law of the Court of Justice … concerning the obligations of Member States to transpose Community directives. (Vogel 1993, p. 3) Difficulties in implementing the Directive also result from the fact that the Directive covers a wide range of issues and concepts, rather than a limit value whose implementation can be much more easily gauged – such as the maximum exposure to a dangerous substance permitted in a workplace, or the length of parental leave – thus leaving a good deal of discretion to the Member States to interpret what fulfilling the minimum criteria will entail. The flexibility of implementation provided by a Directive (rather than a Regulation, which is more detailed) is one of its appeals. However, ‘its result is to create serious, perhaps insoluble problems of measurement of the … impact of the Community norm’ (Baldwin and Daintith 1992, p. 14). Once EU legislation has been transposed, implementation relies on sub-national actors such as employers who will give the law effect in their workplaces and national labour inspectorates that police health and safety regulation. As inspection services vary greatly in the Member States on issues such as the frequency of inspections, information given to employers to help them comply with the law, and sanctions (fines, closures, prosecution), it is unsurprising that results also vary (Gold and Matthews 1996, p. 37). Member States rejected the idea of a ‘Euroinspectorate’ (Baldwin 1992, p. 221), although senior labour inspectors do meet twice annually to exchange information (Commission 1993c). As a result, the European Commission simply does not have the capacity to monitor independently the implementation of Directives, and may have an interest in not doing so, since weak oversight of implementation may reduce the number of conflicts in the Council at the decision-making stage (Eichener 1992, p. 63). Oversight by the Commission encounters further problems in the collection and interpretation of data from the Member States (Baldwin 1992, p. 245).
52 European Welfare States
Even 10 years after the Directive was to come into effect, aspects of it are not fully implemented by those Member States who went into the negotiations with very low standards. Low-standard countries are not the only ones with implementation problems, however, for even in countries such as the Netherlands and Germany it appears that formal legal rules are ‘softened’ by weak enforcement practices (Baldwin 1992, pp. 229–30). According to a recent study, implementation of the Framework Directive on health and safety varies widely in the EU, ‘being available to 25–100 per cent of working people, depending on the country’ (Hamalainen et al. 2001, p. 82). As is demonstrated by analysis of implementation in specific Member States, coverage is very close to 100 per cent in Member States with social-democratic welfare state regimes, high in the conservative–corporatist and liberal states, and low in the southern states. Sweden and Denmark: high national standards, few problems yet subtle changes In Denmark, and later Sweden, the transposition of the Directive ‘raised no major problems’ (Vogel 1993, p. 137). In the words of one Swedish legal official in charge of the transposition of health and safety Directives, ‘We have made an assessment that health and safety is a nonissue for Sweden’ (Interview 19). However, this same official also noted that some specific technical aspects of national legislation have had to be adjusted to meet ‘specific Commission requirements’ (Interview 19).27 Cooperation between employers and workers in preventing accidents at work is well developed and is formally integrated into industrial relations. Specific risks are also addressed by collective agreements. However, owing to the limitation on the governments in both Denmark and Sweden to extend the provisions of collective agreements to all sectors (erga omnes), and the fact that collective agreements do not guarantee protection over time (as they expire), national laws have become increasingly important relative to collective agreements in implementing EU Directives. In Denmark, sectors previously excluded from general health and safety services such as local government, teaching and social services, were included in a 1993 Ministerial Order so as to comply with the general coverage of the Directive. Maritime workers and the armed forces had also to be included. Obligations to provide for written risk assessment in the Directive went beyond national provisions, but these were fulfilled in cooperation between the employers and unions (Vogel 1993, pp. 148–9). The Danish experience with the Framework Directive and
Governance by Law 53
the Machinery Directive helped to convince unions that the EU was committed to setting standards at a high level (Norgaard 1992, p. 213). In Sweden, the main challenge presented by the Directive was that preventative services were left to the discretion of the employer. Under Swedish legislation companies with less than 2000 employees must set up their own company preventative service, with the result that preventative services were developed with highly sector-specific requirements (such as for the construction industry, agriculture and forestry). Changes to the Work Environment Act aimed to strengthen explicit employer responsibility for preventative services and to ensure compliance by SMEs (Vogel 1998, pp. 57–65). A representative of a Swedish union stated: ‘we were worried that we would be forced to lower or converge our standards, but in general we have noticed that we work with the minimum standards set by the EU.’ Concerns were expressed, however, that the Commission ‘is very interested in the transposition of Directives, but not so much in the enforcement of Directives’, permitting low standards to persist ‘at the other end of Europe’ (Interview 19). Germany and France: new requirements for national legislation In Germany the major difficulty in implementing the Directive is that its national system emphasizes insuring risk rather than prevention. The Federal government confronted a dual challenge in implementation, as the interests of the Länder and the industrial accident associations had to be considered. In the context of reunification the Bundesrat (the parliamentary chamber of the Länder) proposed that a new health code be drawn up, which was approved at the end of 1993. Thus, the Directive came at a time of upheaval in the German system and there were attempts from many national interests – employers, insurance associations, Länder – to limit the impact of EU regulation. The Machinery Directive, proposed and approved simultaneously with the Framework Directive, was also highly controversial and generated great resistance from national standard-setting bodies (Wank 1992, pp. 73–5). Although the new legislation conforms to much of what is stated in the Directive, the preventative system in place has not been altered; no new rules about the conditions under which employers assume the responsibility for prevention were enacted, and the complex division between technical regulations and worker health has been preserved (Vogel 1993, pp. 291–6). As a result, the Commission took Germany to the ECJ over incorrect transposition of the Directive in 1999, and the ECJ ruled that changes
54 European Welfare States
to German legislation were in fact needed in order to ensure proper risk assessment in workplaces with fewer than 10 workers (ECJ 2002). Perhaps the biggest impact of the Directive in Germany was that it convinced trade unions that EU legislation did not entail the necessary weakening of national standards. In specific instances, EU legislation may extend national rules, for example by providing coverage to all workers. Alternatively, it may enhance national provision, for example by explicitly stating the obligation of the employer to observe ‘state-ofthe-art’ standards (Wank 1992, p. 72). In France, the Prevention of Occupational Risks Act (91-1414) of 31 December 1991 implemented the Directive. For France, the biggest challenge was reorienting its emphasis on the supervision of risks to the preventative approach of the Directive. Specifically, the Directive ‘goes considerably beyond the administrative and criminal avenues of occupational risk prevention which are the hallmark of French law’ by providing detailed obligations for employer and employees (Chaumette 1992a). The obligation to carry out a risk assessment was new to French legislation, as were many of the preventative services to be implemented. Another controversial aspect of the Directive in France was the liabilities placed on workers by art. 13 of the Directive, especially since the Directive was ambiguous with regard to how the employer was to inform workers of their duties. The CGT union opposed any attempt to allow criminal fault to fall on the employee (Vogel 1993, p. 176). Debates have arisen on the conditions that would allow workers the right to stop work because of a threat to their health and safety, contained in art. 8 para. 4. In short, a number of domestic debates were opened as a result of the Directive. The most fundamental impact was the obligation not merely to oversee and minimize risks but to anticipate and prevent them. Overall, implementation has attempted to conform to the Directive ‘whilst retaining the traditional principles of French employment law by adapting existing texts’ (Chaumette 1992b, p. 44). The UK: little substantive difficulty, great symbolic resistance Owing to the similarity with the existing health and safety system already in place, the Directive was not difficult to implement in the UK. However, the ‘symbolic scope’ of the Directive evoked opposition from the Conservative government (Vogel 1993, p. 238); it could be expected that the Conservatives would oppose any EU regulation of the workplace that might impose new obligations on employers. In particular, the government was sensitive to legislative changes to the health and
Governance by Law 55
safety regulations already in place, the cost of new regulation, and requirements for the intervention of state authorities (UK 1991, p. 2). The regulatory aim of the UK continued to be to place as much discretion as possible in the hands of employers. Transposition of the EU Directive was covered, in part, by the existing legislation. Additional regulations were also passed by Parliament, and other aspects of the Directive were covered in an Approved Code of Practice (ACP). As the ACP does not provide for binding legal duties, trade unions questioned whether such an instrument was the appropriate national instrument for the transposition of the Directive (Vogel 1993, p. 259). The biggest controversy surrounding implementation concerned the employers’ liability, which had been carefully circumscribed by case law limiting liability in cases, for example, of negligence by the employee or in cases where a fellow worker caused the injury (Baldwin 1992, p. 210). Given that the Directive has been transposed in a piecemeal fashion, it is arguable that the Directive has not extended the liability of UK employers. The other controversial issue was worker participation, although in practice the EU Directive recognized ‘national laws and/or practices’ in this area and did not extend worker information and consultation in the UK. Despite government-level resistance to EU legislation, the UK Health and Safety Commission has pushed the European Commission to ensure enforcement in other Member States so that Directives are more evenly applied throughout the EU (UK 1989–90, p. 21). Such actions demonstrate that despite government resistance, there may be a ‘divergence between [national] agency and government policies’ (Baldwin 1996, p. 99). Employers’ reactions to the implementation of the Framework Directive were, on the whole, positive (Byre 1993, p. 19). Indeed, over time employers have come to accept that ‘in areas such as health and safety, common standards based on medical and scientific evidence are necessary in support of the Single Market’ (CBI 1996). This reaction can largely be explained by the fact that implementing EU legislation required very little adjustment. Employers noted that the EU ‘should formulate legislation which set reasonable “achievable” standards, and which was not over-detailed or unduly rigid’ (Byre 1993, p. 19). Despite the UK’s best attempts to limit EU legislation in the preparatory stages, the chairperson of the UK Health and Safety Commission has stated: ‘The [Single European] Act in effect paved the way for a shift from national to EC primacy in policy-making in the area’ (UK 1989–90,
56 European Welfare States
p. 17). It is thus not the substantive but the symbolic effects of the Directive that are most troubling to UK governments. Greece: low standards persist, indicating the weakness of EU Directives in effecting wholesale change The Framework Directive was transposed into national legislation by a series of Presidential Decrees from 1987 to 1996. The Framework Directive thus became operational in legislative terms seven years after its creation at the EU level and four years after the deadline for transposition. The Framework Directive, and the daughter Directives that followed it, are the main legal basis for the regulation of health and safety in Greece and extended health and safety legislation in prevention, risk assessment, the training of workers, worker participation and the representation of workers in decision-making on health and safety issues. In short, there ‘was a need to adjust in many areas as a result of the 1989 Framework Directive’ (Interview 76). The most positive impact of the Directive was in encouraging a national discussion of health and safety issues, albeit ‘not a flourishing one, but better compared to before [EU legislation]’ (Interview 76). A more negative effect of the Directive was that EU minimum standards became the maximum policy objective for adjustment. EU legislation ‘is the maximum legislation we will get in the absence of EU pressure or new legislation and without a push from the social partners’ (Interview 76). In practical terms, there are two main obstacles to the effective enforcement of the Directive in Greece. The first is insufficient administrative and technical infrastructure to implement the Directives, which results in poor coverage, monitoring and control (Spanou 1998, pp. 479–80). The weak technical expertise in the government was already seen in the formative stages of the Directive. At the implementation phase this is compounded by the lack of qualified labour inspectors. As of 2002, the Ministry of Labour and Social Security was still recruiting labour inspectors, who are needed in much greater numbers, perhaps hundreds more, in order to ensure enforcement (Interview 54). The second problem is the difficulty of ensuring compliance owing to the large amount of production that takes place in SMEs and workshops. Owing to lack of inspectors and poor information supplied to employers in SMEs, EU health and safety legislation has been transposed but not yet fully implemented. The comprehensive nature of EU legislation has met with some resistance from employers at the implementation stage, especially where union representation is weak, namely in SMEs (Interview 54).
Governance by Law 57
In order to improve implementation, the Hellenic Institute for Occupational Health and Safety (ELINYAE) was founded in 1993. ELINYAE is funded by social contributions and the European Social Fund. ELINYAE is the first institution to support the implementation of health and safety legislation outside the Ministry of Labour. Its primary purpose is to help ensure the implementation of health and safety policy in all companies, regardless of sector and size, as set out in the 1989 Framework Directive.28 ELINYAE was modelled on the Delors concept of the social dialogue, in which employers and employees engage on issues of concern and build solutions in common, as well as the health and safety institutions of the Nordic countries (Interview 76). The quantitative impact of EU legislation is more difficult to ascertain. There has been a decreasing trend in industrial accidents in Greece, but data are imperfect and in some cases unreliable. There is a vast problem with under-reporting in the Greek case, due in part to the fragmentation of social security whereby statistics are available for only the 45 per cent of the working population that is covered by the general social insurance fund (IKE). IKE data indicates a downward trend in industrial accidents, but structural changes in employment such as the decline of employment in the mining sector and manufacturing account for some, if not most, of the downward trend (Interview 76). Owing to the lack of data for non-IKE workers, and for the informal labour market that persists in agriculture and construction and that accounts for up to 10 per cent of the Greek labour market, the grounds for asserting a correlation between improved legislation and the downward trend in industrial accidents are tenuous.29 While there has been almost total harmonization of Greek law with EU law via transposition, the problem remains that ‘in Greece, there is a difference between the law and its application’ (Interview 54). This problem is well known – the creation of a special legal service under the authority of the Minister of National Economy, ‘including a European law division responsible for legal incorporation and transposition of EC law’, was a result of the concern about the rising number of noncompliance cases in the mid-1980s (Spanou 2000, p. 168). The experience of Greece illustrates the longstanding ‘implementation deficit’ of the EU that may be described as an ‘incapacity to elicit reliable compliance with supranational regulations and directives’ owing to ‘its almost exclusive reliance upon agencies of national and subnational governments for the enforcement of supranational norms’ (Schmitter 1997, p. 411). In addition, low-standard countries may engage in ‘selective defection (read, ‘cheating on specific obligations’)’
58 European Welfare States
from Directives (Schmitter 1997, p. 411). EU Directives are thus ‘highly vulnerable to evasion’ where unions are weak and where ‘unregulated labour markets are the norm’ (Rhodes 1991b), indicating the weakness of Directives as instruments of harmonization in cases where national standards are low. In 1994 the European Agency for Health and Safety at Work was founded in Bilbao, Spain (Regulation 2062/94/EC), with the aim of creating network of health and safety organizations across the EU in order to disseminate information and good practice. The poor implementation record of states was one of the key motives behind the founding of the Agency.30 Overall, EU legislation has had minimal effects in high-standard countries such as Denmark and Sweden, very specific effects, especially in preventative services and employer responsibility, in Germany, France and the UK, and a profound effect on health and safety legislation in low-standard countries such as Greece. Poor national enforcement of the Directive means, however, that despite formal harmonization in the EU to a common floor of health and safety regulation at a high level, low standards persist in the EU. While health and safety standards have been removed from a free-for-all competition among states, workers still continue to be exposed to dangerous working conditions. Serious accidents at work, resulting in more than three days’ lost work, are on the decline in the EU-15, from an average of 4,539 per 100,000 workers in 1994 to 4,088 in 1998. However, Spain and Portugal had the highest rate of accidents (7,073 and 6,180, respectively) while Sweden had the lowest (1,329) (Eurostat 2002b, p. 125). Great disparities continue in both national enforcement of supranational legislation and numbers of accidents.
Conclusion: the unanticipated consequences of qualified majority voting in the SEA There have been several unanticipated consequences of the introduction of QMV in art. 118a of the SEA. The first is that regulation, instead of coalescing at a low level, has been established at a high level. In the Framework Directive, the Commission (after consultation with trade unions and national experts) introduced concepts from high-standard countries, such as Sweden. The support of high-standard countries such as Denmark, the isolation of resistant Member States and the passivity of the southern states in EU committees ensured that these standards became the EU standard.
Governance by Law 59
The second unanticipated consequence of QMV in art. 118a has been the extension, over time, of QMV to other areas of social policy such that the scope of policy subject to QMV has been substantively enlarged over time. As subsequent chapters will show, the experience of QMV under the SEA convinced Member States that cooperation at the EU level did not entail policy-making at the lowest common denominator. Leader Member States joined the Commission in pushing for an extension of QMV to new areas of social policy in subsequent Treaty reforms. The introduction of QMV also encouraged the Commission to test the scope of art. 118a, with the result that regulatory space has opened up in areas ‘traditionally considered to be outside the field of health and safety at work’ (Majone 1996a, p. 75). Although challenged by the UK, the use of art. 118a to introduce legislation on pregnant workers and on working time has been upheld by the ECJ (ECJ 1994). Clearly the idea that QMV would be extended in subsequent treaty revisions within the next decade was not anticipated by the UK in 1986. This is not surprising, given that the SEA was the first major overhaul of the Treaty of Rome. Margaret Thatcher avoids any mention of art. 118a in her summary of the SEA negotiations, save to mention a reservation she had on social policy (Thatcher 1993, p. 555). Thatcher does mention art. 118a in her comments on the dangers of the Maastricht Treaty. Here she states that art. 118a as ‘amended by the Single European Act’ has permitted ‘a means of imposing the high social costs of Germany and France on Britain by the back door’ (Thatcher 1995, p. 484). Later UK documents are even clearer on the unanticipated consequences of art. 118a: The UK’s intention in agreeing to the incorporation of art. 118a into the Treaty was to allow the adoption, by QMV, of minimum requirements specifically related to the health and safety of workers … The ECJ has ruled … that the scope of art. 118a is much broader than the UK envisaged when the article was originally agreed. (UK 1996a) In addition, art. 118b was to develop over time into a distinct form of governance, examined in the next chapter on the Social Dialogue. The seeds for the further isolation of the most resistant Member State, the UK, were thus sown in the SEA. As the UK Chairperson of the UNICE Health and Safety Working Party, speaking to a conference in Ireland, summarized: It is however now becoming clear to us that under majority voting conditions among the twelve Member States, not only is the
60 European Welfare States
European Commission in a much more powerful position vis-à-vis the Members, but the problems of minority States can be ignored and the process of collapse of a minority position is much swifter. The position of a government on its own is much weaker than it used to be. This has very clear implications for the position of states such as yours [Ireland] and mine [the UK]. Our legal and administrative arrangements, and indeed economic structure, differ quite substantially from those of many other Member States … Our champions at the Council table are few. There is a risk that under majority voting our cause will be less clearly heard, and less well understood. We are wholeheartedly in support of QMV where it is a matter of harmonizing specifications, but where, as here, it bears on our social and legal and judicial systems, we are wary of it. (UNICE 1989) Although health and safety has passed a difficult hurdle in the SEA negotiations, the UK’s resistance to an expanded social policy at the EU level was to be the major sticking point in both the day-to-day operations of the Union and the next IGC.
3 Governance by Collective Bargaining: The Social Dialogue
The creation and implementation of social policy, in particular labour market policies, by governments in cooperation with unions and employers’ groups is a feature of several national systems in Western Europe, such as those of Belgium and Austria. Tripartite policy-making (governments, unions, employers) is also a feature of the International Labour Organization (ILO), created in 1919. In other national systems, unions and employers (the social partners) are left to bargain independently of the government. In these bipartite cases, longstanding social-democratic governments implicitly support national-level collective bargaining and the implementation of collective agreements, such as in Sweden and Denmark, or there is a long tradition of sectoral (industry) collective bargaining where key sector negotiations lead bargaining that is used as the national standard, as in Germany. The development of governance by collective bargaining at the EU level thus takes its inspiration from international social policy-making, and even more so from national traditions in EU Member States. There are two aspects to Social Dialogue at the EU level: a tripartite macroeconomic dialogue created by the Cologne European Council of 1994, and an autonomous bipartite dialogue between employers and unions at the EU level that was first introduced in the SEA and enhanced by the Maastricht Treaty of 1991 (ratified 1993). As a result of the Maastricht Treaty, Social Dialogue became a distinct process for the creation of EU-level legislation, and as a result of the conclusion of several Framework Agreements, emerged as a distinct form of governance. It is the latter process of bipartite Social Dialogue, defined as ‘meetings and exchanges of views between management and labour at the Community level, which may eventually lead to contractual relations including collective agreements’ on subjects included in art. 137 (ex art. 118), such 61
62 European Welfare States
as working conditions and hours, that is the focus of this chapter (Engles and Salas 1998, p. 85).1 Social policy proved to be one of the most contentious issues negotiated during the 1990–1 IGC. Further, it was not clear until October 1991 what position union and employer groups would take on the expansion of the Social Dialogue as a formal and distinct avenue for policy creation, nor, until the final European Council meeting at Maastricht in December 1991, how the intransigent opposition of the Conservative UK government under John Major would be accommodated. Negotiations on social policy thus existed in a highly uncertain position throughout 1991; at no time before the final negotiations at Maastricht could the substantive expansion of social policy competency at the EU level and the strengthening of the Social Dialogue as a new form of governance be assured. How, then, was agreement finally reached? First, national preferences must be taken into consideration. National preference formation in this case is based on three factors: (1) the strength of national unions, as measured by union density; (2) the normal level of bargaining over wages, working time and working conditions, whether at the firm, sectoral (also known as industry or branch) or national level; and (3) the organization of national collective bargaining into bipartite or tripartite bargaining structures. Union density will be defined here as: high (60–100 per cent), medium (20–59.9 per cent), and low (under 20 per cent). Member States with high union density and a strong tradition of collective bargaining at the national level, whether tripartite or bipartite, should be expected to support the extension of such procedures at the EU level for reasons primarily related to exporting their national model. Examples of such countries are Belgium, Denmark and Sweden. It should be noted that although Austria, Finland and Sweden were not yet members of the EU in 1990, their national confederations of unions and employers were members of the ETUC (European Trade Union Confederation) and UNICE (Union of Industrial and Employers’ Confederations of Europe). National social partners of these states were included in UNICE and ETUC discussions regarding Social Dialogue. Other Member States with a medium level of union density but strong national traditions of collective bargaining at the sectoral level, such as Italy and Germany, or state involvement in wage setting via tripartite structures, such as Ireland and the Netherlands, are also expect to support Social Dialogue. Member States with medium to low union density and a weak history of national collective bargaining should be expected to resist Treaty competency in the field of EU-level bargaining, such as
Governance by Collective Bargaining 63
Greece, France and the UK. Although unions in these Member States may be in favour of Social Dialogue, it is expected that the lack of welldeveloped histories of policy-making via national collective bargaining or tripartite negotiations will structure government preferences towards a rejection of labour market regulation via Social Dialogue. An uncertain or intermediary position is expected from those countries where tripartite structures were in flux in the late 1980s – Spain (where tripartite structures were in decline from 1986) and Portugal. The preferences of Member States alone, however, cannot explain the presence of Social Dialogue on the IGC agenda, nor the agreement to empower Social Dialogue as an alternative to the Community Method of regulation. Three institutional factors are a vital part of the explanation. The first is Commission entrepreneurship. Without the activism of the Commission in creating opportunities for social partner involvement in EU institutions, and the entrepreneurship of Jacques Delors as Commission President in suggesting a new form of governance based on EU-level collective bargaining, Social Dialogue would not have been introduced at the EU level. Further, by supporting the development of EU-level confederations of the social partners, most notably the ETUC, the Commission helped to create the very interests that would participate in the Social Dialogue and eventually call for its enhancement (Dolvik and Visser 2001). The second institutional factor is the failure of unanimity voting to permit agreement on social policy in the EU. Given blockages in Council, Member States in favour of cooperation looked to extend QMV and were disposed to consider new regulatory paths. By the late 1980s there was a ‘considerable backlog of draft legislation’ (Addison and Siebert 1997, p. 12), and ‘considerable dissatisfaction with the lack of progress on social policy’ (Deakin 1997, p. 124). Legislation that could not be passed by Council included various drafts of the ‘Vredeling initiative’, the draft Fifth Directive on company law, the European Company Statute, working time, 1982 and 1983 proposals on atypical work, 1982 proposals on temporary employment agencies, and failed attempts to propose parental leave provisions. An enhanced EU-level Social Dialogue could, therefore ‘circumvent or break social policy stalemates in the Council’ (Falkner 1998, p. 72). Notably, however, it was not Member States but Jacques Delors who, as Commission President, consistently pushed the Social Dialogue ‘as a solution to the stalemate in Council over social policy’ (Addison and Siebert 1997, p. 14). Breaking such stalemates took on a greater sense of urgency for high standard states in light of the completion of the SEM, which highlighted the
64 European Welfare States
differences in national labour market regulation. Such differences were amplified in the EU by the post-SEA enlargement to Spain and Portugal, which, along with Greece, were low-wage countries. Gerda Falkner has also noted the role of the posted workers case (the Rush Portuguesa case of 1989), and the relocation of multinationals such as Renault and Hoover, in highlighting the differences in labour market regulation in Europe around the time of the completion of the SEM (Falkner 2002, pp. 105–06). The third institutional factor is learning effects as a result of reiterative cooperation at the EU level. Actors involved in social policy – governments, unions and employers – learned over time that the Commission would continue to propose social policy regulation at a high level. Such legislation would have the opportunity to pass in Council if QMV was extended. Therefore, even those governments and employer organizations that would be expected to reject Social Dialogue on the basis of national preferences may in fact support Social Dialogue as a ‘second-best’ solution (the ‘first-best’ being no regulation at all), with the knowledge that the Commission would continue to propose social regulation that might be approved if QMV was extended to new areas. Support for the Social Dialogue is thus also explicable for those who wanted to circumvent, and even prevent, policy regulation led by the European Commission. As the Secretary-General of UNICE explained: Experience so far in EC social legislation is that the legislator is getting it wrong. The legislator is being too detailed, too prescriptive, is trying to do everything from Brussels … We felt we would be better custodians of subsidiarity than the legislator. We were also convinced that the unions would be too. The reason is that our members and theirs will be breathing down our necks because they do not want to lose national sovereignty. That is very healthy. They will also allow us to go along with broad framework type agreements rather than the prescriptive and detailed type of legislation that we are getting from the Commission. (Dolvik 1997, p. 211) Finally, in the case of France and the UK, political ideology, a factor exogenous to the insitutionalist analysis, must be examined. Governments of the left are responsible in both cases for agreeing to enhance cooperation in the social policy field despite national histories that would suggest resistance. In the French case, agreement occurred at Maastricht in 1991. In the UK case, agreement occurred in 1997 when the Labour government opted-in to the Social Protocol.
Governance by Collective Bargaining 65
The next section of this chapter presents welfare state preferences for Social Dialogue in greater detail. The second section details the development of Social Dialogue at the EU level from 1955 to 1990, and the third focuses on the Maastricht Treaty negotiations. A national welfare state explanation of national preferences and institutional factors, most notably Commission leadership, are traced through these two sections. The positions of national confederations of employers and unions, and the European confederations UNICE and CEEP (employers) and ETUC (unions), are also considered alongside those of governments and the Commission. The interests of UNICE, CEEP and the ETUC were as important as those of Member States in determining the Treaty outcome on Social Dialogue. In this chapter the European-level social partners are thus treated as actors in their own right. Finally, a specific example of Social Dialogue, negotiations on part-time work, is examined. In this section, the role of social partners in the new governance system created by Social Dialogue is assessed, and national implementation of the EU Framework Agreement on part-time work (given effect in EU law by a Council-approved Directive) is examined. By carefully tracing preferences and bargaining positions this chapter will illustrate that Member States were surprisingly passive in the development of the Social Dialogue and that the entrepreneurship of the Commission is a key explanatory variable. The chapter also corrects confusion about Greece’s preferences in the IGC, which one expert has said was in favour of an enhanced social policy (Ross 1995a, p. 103), and which another has said was opposed (Moravcsik 1998, p. 453). The sidepayments explanation of poor-country agreement to the Social Protocol put forward by Lange (Lange 1993) is also challenged. National histories in collective bargaining predisposed several poor states towards EU level collective agreements, while for others the entrepreneurship of the Commission and the desire to emulate more developed Member States made them willing to accept Social Dialogue as a new regulatory method.
Welfare state preferences for Social Dialogue National patterns of Social Dialogue are in flux in Western Europe. Strong traditions of national level bargaining resulting in collective agreements have either broken down or been reduced (in the sense of the number of workers covered) in the 1990s. This has been the case in both Germany and Sweden. In Ireland, social partnership agreements via government coordinated wage agreements where none traditionally
66 European Welfare States
existed have been in effect since 1987 (O’Donnell and Thomas 2002). Social pacts have also experienced resurgence in the Netherlands, and have been attempted with various forms of success in Greece and Portugal. It is vital to note, then, that although preferences may be derived from traditional patterns of national bipartite or tripartite bargaining, such traditions were under great pressure in the period of the Maastricht Treaty negotiations. Before examining national preferences, as embedded in national welfare state regimes, the contradictory goals of various actors in promoting Social Dialogue as a form of social policy governance must be explored. Several governments and the European Commission viewed Social Dialogue as an alternative route to social policy legislation given the blockage of much social policy legislation in the Council. Social Dialogue would enable labour market regulation via the actors most affected by such policies (namely workers and employers), and would recreate national bargaining patterns at the EU level. Social partner participation was also a means of securing a peaceful route to economic and monetary union. The ETUC viewed Social Dialogue as a desirable way of reaching agreement, backed up by the possibility of Commission legislation in case of negotiating failure. The UNICE view was that negotiation could water down legislation, or that it could even prevent agreement if it refused to negotiate. Further, a breakdown of collective bargaining would signal employer resistance, and might be picked up by the European Council should the Commission propose further legislation. It would also be a mistake, however, to take the view that all employers were against an enhanced Social Dialogue. Employers from countries where social partnership is well developed, such as Belgium and the Netherlands, pushed for a UNICE agreement with the ETUC. Further, not all national unions were initially supportive of EU-level Social Dialogue. Nordic and German unions were especially wary of undercutting their importance in national bargaining by signing up to an EUlevel Social Dialogue that would be less robust than national bipartite agreements. As a result of these very different interests, the positions of national employers and unions during bargaining, in addition to governments, will be followed as closely as possible. As outlined in the introduction to this chapter, Member States with highly developed systems of collective bargaining at the national or sectoral level are expected to support the adoption of Social Dialogue at EU level; those with weak national systems, and bargaining devolved to the level of the firm, are expected to resist the creation of EU-level structures
Governance by Collective Bargaining 67
for bargaining. An explanation based on the costs of corporatist patterns of bargaining may at first appear to be central to the argument – highly unionized Member States with patterns of corporatist bargaining at the national level have higher labour costs and would like to see these exported to other Member States. For example, in 1990 average hourly labour costs for manual and non-manual workers in industry (in ECU) in strongly corporatist Member States were 19.30 in Belgium and 17.19 in Denmark (Eurostat 2002a, p. 16).2 Low-wage Member States would probably be made worse off (in terms of increased costs) by a shift to legislation via EU-level collective agreement, and to QMV in social policy more generally (Lange 1993, p. 21; Falkner 2002, pp. 101–2). States with low wages included Greece (6.97/hour in 1992) and Portugal (3.57/hour in 1990), but also Spain (11.30/hour in 1990), Ireland (11.64/hour in 1990) and the UK (12.20/hour in 1990). However, given that wages and social security were explicitly excluded from the social-policy field at the EU level, national governments should not be expected to be concerned solely with relative labour costs. Rather, where national institutions for collective bargaining were being developed, as in the case of Ireland, or where their development was seen to be of benefit to the normalization and modernization of economic and political life, as in Greece and Spain,3 even poorer Member States with low wage costs will support Social Dialogue at the EU level. Such an analysis rests on the ‘goodness of fit’ between national institutions and those proposed at the EU level, and a desire to emulate more developed Member States, rather than a cost analysis. Table 3.1 outlines the collective bargaining systems found in the EU Member States. On the basis of three factors (union density, the level of bargaining, and the involvement of the state in bargaining through tripartite structures), Belgium, Denmark, Germany, Ireland, Italy, Luxembourg, the Netherlands and Portugal should support the enhancement of EU Social Dialogue, while France, Greece, Spain and the UK are expected to oppose. Although not Member States in 1990, confederations of trade unions and employers from Austria, Finland and Sweden were members of the ETUC and UNICE and were involved in consultations with the European Commission. Later, they also took part in Social Dialogue negotiations, and as members of the European Economic Area (EEA) were bound to observe ‘at least the minimum content of Community labour law’ that included the Social Protocol (Deakin 1997, p. 124). These states became full members in 1995. On the basis of their national systems, Austria, Finland and Sweden are also
Bargaining at national, sectoral, or firm level; kind of negotiations Bipartite (tripartite negotiations on economic policy may occur); national collective agreement provides framework for sectoral ones. Bipartite; sectoral agreements since 1990. Tripartite incomes policy; national collective agreements, plus sectoral. Tripartite (depending on issue); general national agreement and sector or firm level elaborates. Tripartite; sectoral. Bipartite; sectoral. Tripartite (depending on issue); national framework agreements, plus regional and sectoral. Tripartite; sectoral and firm.
Member State and welfare state regime type*
Denmark (SD)
Sweden (SD)†
Finland (SD)†
Belgium (C)
Netherlands (C)
Germany (C)
Austria (C)†
Luxembourg (C)
Table 3.1 Collective bargaining systems in EU-15 Member States
FEDIL
BKW, VOI
VNO, NCW BDA
FEB/VDO
STK (TKL)
SAF
DA, IR
Main employers’ association(s)
CGT, LCG, FEP
OGB
DGB, DBB, DAG
FNV, CNV, RMHP
CSC, ABV/FGTB, CGSLB
STTK Akava
LO, SACO, TCO SAK,
LO, FTF
Main labour association(s)
50% medium
54.1% medium
38% medium
22.3% medium
69.3% high
72.5% high
82.4% high
75.3% high
Union density
68
Bipartite; national tripartite framework negotiations broke down 1986, thus sectoral. Bipartite (but state involved in framework negotiations); national framework negotiations, sectoral. Bipartite (but links to political parties still strong); national collective agreements in 1990–1, then sectoral and firm levels. Legislation plays key role. Tripartite; national framework agreements, sectoral. Voluntary tripartite; national level wage agreements regulate firm-level bargaining. Firm level.
Spain (S)
Italy (S)
Greece (S)
Portugal (S)
Ireland (L)
United Kingdom (L)
CBI
IBEC
AIP, CIP
SEV
Confindustria
CEOE
CNPF
TUC
ICTU
UGT, CGTP
GSEE, ADEDY
CGIL, CISL, UIL
UGT, CCOO, ELA-STV
CGT, CFDT, CGT-FO, CFTC, CGC, FEN
37% medium
48.2% medium
40% medium
25% medium
39.1% medium
12.1% low
9.2% low
Sources: Lecher (1994); Ebbinghaus and Visser (2000); Commission (2000b).
Not EU members until 1995, but members of ETUC and Social Dialogue agreements could be extended to them via their membership in the European Economic Area (EEA).
†
* C, conservative-corporatist; L, liberal; S, southern; SD, social-democratic.
Bipartite; national general agreement for minimum standards, then sectoral and firm levels. Relies heavily on the role of central legislation.
France (C)
69
70 European Welfare States
expected to support EU-level Social Dialogue, albeit with less influence at the Treaty bargaining stages than the Member States. Given these expectations, the main gaps in behaviour occur in the cases of Greece, France and Spain, who all agreed to Social Dialogue despite national histories that would suggest resistance. These gaps are explained by political ideology in the case of France, and by institutional factors that helped shift bargaining positions in the case of Spain and Greece. Another gap in expected behaviour occurred when the UK, which was expected to oppose the Social Chapter, acceded to it in 1997. The pro-integrationist ideology of the Blair Labour government (relative to the Major Conservative government) explains the gap in the UK case. Notably, given that in the final Agreement both poor Member States (Greece, Spain) and governments of the right (Denmark, Germany, the Netherlands) supported the expansion of social policy at Maastricht, and that governments at times took positions at odds with the preferences of employer groups (France, Greece, Portugal after Maastricht4 and UK under Labour), liberal intergovernmental theory is not highly relevant in this case, as it predicts that rich governments of the left will be the most supportive of social policy expansion during Treaty bargaining (Moravcsik and Nicolaïdis 1999).5 Understanding national traditions in social policy is thus vital to explaining the positions of Member States. National welfare state histories also help to identify the policy leader (Belgium), and primary policy resister (the UK), while also highlighting Member States whose eventual agreement cannot be explained by national histories alone (Greece). Despite the ‘goodness of fit’ between EU Social Dialogue and a majority of national systems, in particular those of Belgium, Italy and the Nordic countries,6 Member States were not the key to the development of the Treaty section on Social Dialogue. While the ultimate design of the Social Dialogue may have been acceptable to the majority of Member States, and indeed accorded with the interests of Belgium as the leader Member State in this sub-issue area, activism to include enhanced provisions for the Social Dialogue came largely from the European Commission, and later from unions and employers and not from Member States. The next section traces the development of Social Dialogue at the EU level from 1955 to 1990. Commission entrepreneurship in steadily enhancing the ability of the social partners to participate in EU decision-making is the key explanatory variable in this period, especially in the critical 1985–90 period, leading up to the Maastricht negotiations.
Governance by Collective Bargaining 71
Social Dialogue at the EU level: 1955–90 Union and employer groups have played a consultative role in EU-level decision-making since the creation of the ECSC, which had a consultative committee of unions and employers from 1955. During the conferences to create the EEC, the Dutch government proposed the creation of a bipartite committee that the EEC could consult on labour market and employment issues (EC 1999, arts. 257–62). Today, the Economic and Social Committee (ESC) provides an important network of contacts between national representatives and top-level Commission officials.7 However, the ESC is solely an advisory body, and is largely seen more as ‘a sounding board for the Commission and Council’ than a route for the development of social regulation (Compston 2001a, p. 3). The most important actors in the governance of Social Dialogue are the social partners: ETUC on the workers’ side, and UNICE on the employers’ side. The European Centre of Public Enterprises (CEEP) also bargains on the employers’ side of public services.8 UNICE was founded in 1958, CEEP in 1961 and the ETUC in 1973 (Lanzalaco 1992; Rath 1994; Cowles 2002; Greenwood 2002). The ETUC in particular has benefited from sustained financial assistance from the European Commission, and from the European Parliament (Dolvik 1997, p. 239 fn. 161; Dolvik and Visser 2001, pp. 16–17).9 All three of the largest and most important social partner organizations – UNICE, CEEP and the ETUC – are each composed of national affiliates who help to guide policy through annual meetings, a committee system, and a Secretariat based in Brussels. Notably, ETUC and UNICE membership was broader than EU membership; thus, although not EU members until 1995, Austrian, Finnish and Swedish unions and employers were party to the Social Dialogue discussion, and as members of the EEA (European Economic Area) would be bound by the provisions of the Social Protocol. Social partner participation in governance at the EU level has a very mixed history, however. Commission attempts to enhance social partner participation in the 1970s and 1980s demonstrate the conflict between employer and union interests in creating a form of governance based on collective bargaining. In the mid-1970s the Commission called a series of Tripartite Conferences to discuss issues such as vocational training, youth unemployment, equal opportunities and the insertion of the long-term unemployed into the labour market. These conferences met six times between 1974 and 1978, at which stage the ETUC ‘withdrew its support under complaints about lack of progress and also
72 European Welfare States
confronted with rising dissent within its ranks’ (Schmitter and Streeck 1991, p. 139). The tripartite Standing Committee on Employment, created in 1970, had little substantive policy output except on helping to resolve contentious parts of the collective redundancies Directive of 1975 (75/129/EEC). More successful was the integration of social partners in the boards of new European agencies, such as the European Foundation for the Improvement of Living and Working Conditions in Dublin, which created ‘cooperative experiences’ that ‘facilitated the development of “social dialogue” later on’ (Falkner 1998, p. 72). Consultative committees were also used to try and help smooth the way for EU legislation. Tripartite consultations (including the Commission and Member State governments) on introducing common standards on working time, and the reduction of working time with the hope of boosting employment, took place in 1983. Employers, however, objected to EU-level regulation. When these negotiations broke down, the Commission did not gain the support of the Council for a draft recommendation. Subsequent Commission recommendations on parttime and temporary work were also blocked in the Council (Crijns 1987, p. 59). During the 1980s, employers as represented by UNICE were opposed to participation at the EU level for fear that outputs of tripartite discussions might be used ‘by the Commission as the basis for social legislation’ (Rhodes 1991a, pp. 18–19). Employers refused to participate in the Val Duchesse discussions, launched by Jacques Delors in 1985, in anything other than a consultative role on general issues affecting the labour market. There was a feeling among national employer confederations that still persists today that the Commission, and specifically DG Employment and Social Affairs, favours an interventionist approach to regulation with a bias in favour of unions and workers (Interview 52). Tripartite discussions with the Commission were bound to be tipped in favour of union demands. Still, as a result of Delors’s initiative, joint working groups were created on macroeconomic policy and new technologies that produced common non-binding opinions (UNICE et al. 1987). Bipartite bodies were also created in 1985 at ‘cross-industry level, extending gradually to a number of sectors’, in particular in the services sectors, including telecommunications and banking (Commission 1995b; Commission 2000b). The belief in the superiority of negotiated settlements between capital and labour over technocratic rule setting was at the heart of Delors’s personal political philosophy (Ross 1995a, p. 45). Delors made the enhancement of Social Dialogue ‘one of his own personal projects’
Governance by Collective Bargaining 73
(Gold 1998, p. 112). In a speech to the European Parliament Delors stated: ‘A European collective agreement is not just an empty slogan. It would provide a dynamic framework, one that respected differing views – a spur to initiative, not a source of paralysing uniformity’ (Delors 1985). Delors’s Val Duchesse initiative to reinvigorate tripartite discussions at the EU level was presented to Member States as enhancing ‘the “social dimension” of the single market’ [la ‘dimension sociale’ du marché intérieur] (Venturini 1988, p. 11), and later as the social corollary to EMU. Social integration had clearly not progressed at the same rate as economic integration. The development of a ‘European social space’ [un espace social européen] was integral to fulfilling the original motivation of the Community to provide for both economic and social progress (Venturini 1988, pp. 11–12). Such statements were accompanied by caveats. The first was that wages and social protection would remain at the national level but that issues of working time and worker information and consultation were possible areas for EU-level Dialogue. The second was that the Commission would continue to retain its ability to propose legislation in case the Social Dialogue reached an impasse on any particular issue (Venturini 1988, pp. 62, 65). Delors’s overall vision was that as the SEM progressed, the Community needed a social policy to demonstrate a ‘balanced approach to integration’ and ‘solidarity as expressed by the Structural Funds’ (Interview 33). Ultimately, social policy was an integral part of Delors’s vision of economic and political union in Europe (Interview 33). During the SEA negotiations the Commission proposed a new article in social policy that would provide ‘for a social dialogue with the possibility for contractual relationships, at the European level’ (Corbett 1998, p. 238). The SEA gave the Val Duchesse formulae of social dialogue explicit backing in the new art. 118b, and obliged the Commission to ‘endeavour to develop the dialogue between management and labour at European level which could, if the two sides consider it desirable, lead to relations based on agreement’ (EC 1987). The effective intervention of the Roundtable of European Industrialists during the SEA negotiations also built up employer contacts at the EU level (Cowles 1995). The SEA provisions proved insufficient, however, to institutionalize ‘real codetermination of EU policy’ (Compston 2001b). From 1987 to 1990, the Commission attempted to upgrade Val Duchesse initiatives from tripartite discussions to substantive policymaking. Article 12 of the non-binding 1989 Charter of Fundamental Social Rights of Workers states that ‘contractual relations’ between ‘the two sides of industry at European level’ are possible (EC 1989).
74 European Welfare States
The Social Action Programme, launched after the 1989 Social Charter, created a ‘small ad hoc group’ that met to discuss training and labour market issues (Ross 1995b, p. 377). The Commission, as per the SEA, was to consult the social partners before proposing legislation, but was stymied by UNICE refusal to participate in discussions. The only agreements reached during this time were on training schemes in the railway and energy sectors, and were signed by CEEP and the ETUC but not by UNICE. Social Charter initiatives were also blocked in Council by the UK, and in specific instances by Portugal (European Works Council, posting of workers), Italy (pregnant workers, young workers), Greece (young workers) and Spain (young workers) (Falkner 1998, pp. 68–9). Still, the Commission pressed on with the idea that the intensification of the EU-level Social Dialogue presented a potential ‘instrument of regulation’ that would operate via negotiation and consensus, and that would give the social partners a stake in the creation of a Europe without frontiers (Commission 1988b; Venturini 1988, pp. 63–5, 70). During this time the Commission was what could be termed a policy incubator, keeping the idea of a Social Dialogue alive despite conflicting interests in its development. Council presidency conclusions from 1988 and 1989 (German, Greek and Spanish) mirrored Commission thinking, each stating variations on the theme that the completion of the internal market ‘must be achieved in a climate of close cooperation between employers and workers’ (Council 1989a) by ‘the systematic pursuit of a constructive dialogue between management and labour at the Community level’ (Council 1989b). In February 1991 the Commission proposed that the ETUC, UNICE and CEEP use the ad hoc group to submit their own proposal to the IGC that would lead to the Maastricht Treaty. What is notable about the negotiations is the fact that the Commission continued to exert leadership in the field of Social Dialogue, and social policy more generally, in the context of the IGC. Although backed by Belgium, the Commission acted as the key advocate for the enhancement of art. 118b during the negotiations.
Negotiations on the enhancement of Social Dialogue in the Maastricht Treaty Background to negotiations: national governments and policy initiatives since the SEA The Maastricht Treaty was negotiated over a one-year period from late 1990 to December 1991 by a series of intergovernmental conferences (IGCs) that culminated at Maastricht under the Dutch presidency. Two IGCs
Governance by Collective Bargaining 75
operated in parallel, one on economic and monetary union and the other on political union that included institutional reform and social policy. Monthly ministerial meetings were held (except during August), as were informal ministerial sessions and weekly meetings of personal representatives of the Ministers, most often the permanent representatives in Brussels (Corbett 1998). Governments in power during this time were mostly of the right, although France and Spain had governments of the left (see Table 3.2).
Table 3.2 Member State political affiliations 1991–2 Member State
Rich or poor* (in 1991)
Party at head of government
Head of government
Political alignment of government: right or left†
Belgium
Rich
CVP
Martens (until 7 March 1992); Dehaene
Right
Denmark
Rich
KF
Schlüter
Centre–Right
France
Rich
President: PS PM: PS; SPS
President: Mitterrand PM: Cresson (to 2 April 1992); Bérégovoy
President: Left Prime Minister: Left
Germany
Rich
CDU
Kohl
Right
Greece
Poor
ND
Mitsotakis
Right
Ireland
Poor
Fianna Fáil
Haughey (to 11 February 1992); Reynolds
Centre–Left
Italy
Rich
DC; PSI
Andreotti (to June 1992); Amato
Centre coalitions
Luxembourg
Rich
CSV/PCS
Santer
Right
Netherlands
Rich
CDA
Lubbers
Right
Portugal
Poor
PS
Soares
Left
Spain
Poor
PSOE
González
Left
UK
Poor
Conservative
Major
Right
* Determined on the basis of GDP per capita at PPP; rich ⫽ above average; poor ⫽ below average (Eurostat 2001, p. 24). † Right (Conservative, Gaullist, Christian Democrat), or left (Socialist, Labour).
76 European Welfare States
The most important event prior to the negotiations was German reunification in October 1990.10 German government, business and union opinions on social policy were not greatly affected in negotiations by reunification, however, suggesting the dominance of the Kohl administration and longstanding West German preferences in social policy in determining the bargaining position. The most important issues under negotiation in 1990–1 were the criteria and timetable for economic and monetary union. These negotiations formed the backbone of the IGC. It was not until June 1990 that a second and parallel IGC on political union was agreed to, upon the insistence of Germany and France. A joint letter from Kohl and Mitterrand suggested priorities for the IGC on political union to be the strengthening of democracy, institutional reform and the creation of a common foreign and security policy (CFSP) (Kohl and Mitterrand 1990). Social policy reform was not suggested in the joint letter but was later added to the IGC agenda. The focus of this section is on negotiations to enhance provisions for Social Dialogue included in the political union IGC, and, more briefly, on the extension of QMV to social policy. Other issue areas are considered only insofar as they explain the agreement reached on social policy. With regard to social policy, activity had been slowly increasing in the 1987–91 period in light of QMV first introduced in the SEA. A number of Directives in health and safety, designed to create minimum standards, had resulted. Other Commission proposals on working time and worker information and consultation, however, were stuck in Council. In addition, the Community had adopted a non-binding Social Charter in 1989. Despite the inclusion of art. 118b in the SEA, and Commission efforts to promote the Social Dialogue, no EU-level negotiations among the social partners had occurred. Still, the fact that UNICE had not exited the tripartite discussion process from 1987 to 1990 was considered ‘significant’ by the ETUC and the Commission (Dolvik 1997, p. 202). The 1990–1 IGC In late 1990 the only Member State openly supportive of social dialogue at the EU level was Belgium, supported by the Commission and the EP (Commission 1990; Parliament 1990, paras. 13–15). Belgium proposed that EU-level negotiations between workers and employers could be consolidated by a Council decision taken by QMV. Belgium also proposed that a tripartite European Committee on Labour be created to serve three functions: (1) a forum for negotiating Framework
Governance by Collective Bargaining 77
Agreements at the EU level that would then be implemented, after Council approval, by national legislation or collective agreements; (2) a consultative body for new legislative proposals; (3) a source of new EU legislation if it was to request that the Commission draw up draft proposals in spheres of its competence (Falkner 1998, p. 90–1). The Netherlands also agreed that an EU-level ‘industrial policy’ or ‘increased coordination of national policies’ could be considered by the IGC (Netherlands 1990, p. 133). The Belgian delegation, in its social policy proposal of October 1990, noted that EU law was in certain instances too ‘dirigiste’, and that the social partner route to legislation would correct this trend (Falkner 1998, p. 90). Commission proposals in March 1991 closely mirrored Belgium’s, although it also went beyond them in several ways. The Commission proposed that it should trigger social partner negotiations, that social partners could also be involved in implementing agreements at the national level and that in instances where agreement failed to be reached via social partner dialogue that the Commission would enact the traditional regulatory system of governance by law (the so-called ‘negotiate or we will legislate’ clause) (Coldrick 1991; Rhodes 1991a, p. 23; Falkner 1998, p. 91).11 Although these two texts were not used in draft Treaties, they served as a template for negotiations among the social partners six months later. On the issue of social policy more generally, interest for strengthening QMV was expressed by Belgium, Denmark, France, Germany, Greece, Italy and Portugal (Portugal 1992; Ross 1995a, p. 150). The Greek proposal ‘asked for extension of QMV to all legislative acts’, excluding a few articles but not art. 118 (Greece 1992). In the words of a New Democracy Member of Parliament who was a Minister at the time of Maastricht, ‘we [New Democracy] were in favour of deeper social policy in Europe’ (Interview 1). Member States that supported discussions about extending QMV, although their positions on social policy were reserved, were Luxembourg, the Netherlands, Spain and the UK (UK 1990; Falkner 1998, p. 87). By December 1990 consideration of the ‘social dimension, including the need for social dialogue’, was added to the IGC agenda (Council 1990). In April 1991 the Luxembourg Presidency presented a first proposal for a draft treaty on political union. On Social Dialogue the draft hardly went beyond what was already in the SEA (Council 1991a, art. 118b). However, on the basis of a Commission proposal (Commission 1991b), a revised draft treaty of June 1991 stated that the Commission should consult management and labour ‘on the advisability of
78 European Welfare States
Community action’, and that ‘should management and labour so desire, the dialogue between them at the Community level may lead to relations based on agreement which shall be implemented in accordance with the procedures and practices peculiar to each Member State’ (Council 1991b, art. 118b).12 The threat of Commission use of the traditional route to EU law, in cases of the failure of negotiations, was not included. The Dutch Presidency transposed this text in their new draft treaty of September 1991 (Council 1991c), although in other areas the new Dutch draft proved unacceptable. Due to disagreement over the Dutch draft, which contained a unified institutional structure for all Treaty areas rather than hiving off more contentious issues such as defence and immigration into separate intergovernmental structures, negotiators went back to using the Luxembourg draft. In the case of social policy this did not represent a significant change. Fearing that the UK would strike a compromise deal and agree to an extension of QMV in some limited areas of social policy, and that a prointegrationist Labour government might be elected to power in the upcoming UK election, increased legislation in the social policy field appeared a ‘realistic threat’ in the eyes of UNICE (Falkner 1998; Branch and Greenwood 2001; Tyszkiewicz 2002a). Vasso Papandreou, the Greek commissioner for DG V (now known as DG Employment and Social Affairs), also suggested to the social partners at a meeting in early 1991 that ‘[a]n extension of QMV in the realm of social policy was a virtually inevitable outcome of the IGC’ (Branch and Greenwood 2001, pp. 53–4). As a result of the Dutch Draft Treaty, and uncertainty over a possible British compromise, UNICE agreed to submit a joint statement with CEEP and the ETUC on European-level negotiations. The Commission made it clear that a joint statement would be highly influential in cementing agreement on the form of Social Dialogue to be adopted in the treaty negotiations. The October 1991 Agreement on Social Dialogue Using the Ad Hoc Group struck by the Commission, intense negotiations between UNICE, CEEP and the ETUC took place at the end of October. Contentious issues included whether sectoral agreements would be included, implementation, the coverage of SMEs and how the final agreements would be extended to bind national affiliates (Dolvik 1997, p. 202). In UNICE, Belgian, Dutch and Italian employers’ representatives broke ranks with the hardline stance against EU-level Framework Agreements
Governance by Collective Bargaining 79
(Rhodes 1991a, p. 23), and Danish, French and Spanish employer groups also pushed for a joint statement (Dolvik 1997, pp. 202, 211). In the words of a member of the UNICE Secretariat: ‘As viewed from my Danish perspective … I think that it is fair to say that there were a few organizations who effectively took the lead’ (Trampe 2000, p. 185). German business remained ambivalent (Dolvik 1997, p. 202; Moravcsik 1998, p. 396), despite having issued a joint statement with DGB in 1989 supporting increased powers of the social dialogue committees formed out of Val Duchesse (DGB and BDA 1989). The British CBI decided it ‘had more to lose both externally and internally’ if it walked out of negotiations, and agreed to sign a joint statement (Dolvik 1997, p. 203). Portuguese and Greek employers were ‘the most reluctant’, while the Irish ‘followed for more or less practical reasons, fearing increased Community legislative interventionism’ (Dolvik 1997, p. 211). On the ETUC side, national affiliates were generally supportive of enhanced Social Dialogue, having long held the ambition of improved cross-border coordination of wages and working conditions, and industrial action. The Danish, German and Swedish unions were more reluctant given their strong national positions and fears of being replaced by the ETUC in Brussels. Social Dialogue was highly attractive, however, for those unions in France and the UK where their role was diminishing (Tyszkiewicz 2002a). The UK’s TUC, for example, had shifted its position towards a more favourable opinion of EU cooperation in 1988 after consultations with Jacques Delors (Best 1994, p. 258). Given a longstanding interest in EU action on a number of issues, the ETUC was able to present a united position in support of the Social Dialogue. National histories thus explain the positions of national employers’ representatives; those with a history of negotiated legislation – particularly the Belgian, Danish and Dutch employers – were the leaders in this instance. Greek reluctance is explicable by the very weak tradition of bipartite bargaining, and the lack of expertise on EU issues in the SEV. The UK’s CBI adopted a highly pragmatic approach, as it often does, preferring to have input into the Agreement (and thus stand firm in voicing minimalist proposals) rather than walking out and watching from the sidelines. The problem continued, however, as to whether employers wanted the Social Dialogue as a means to create less burdensome legislation, or as a means to block legislation altogether. This tension is highlighted in the words of the UNICE Secretary-General: Employers wisely took the pragmatic view: experience showed them that the EU had an insatiable appetite for social legislation, much of
80 European Welfare States
it very detailed, very restrictive and too costly for companies. Employers were convinced the IGC would expand the range of subjects in this field, on which the Council could decide by qualified majority vote, thus making it likely that many more social policy directives damaging to business interests would be passed through the Council. There was only one solution: to acquire the right to step in and replace legislators. (Tyszkiewicz 2002b) Certainly it is clear that without supranational pressures suggesting alternative paths to regulation approved by unanimity voting, namely an extension of QMV, the most reluctant national affiliates would not have been willing to empower Social Dialogue as an alternative form of governance. One such national social partner was the German BDA (employers), which adopted a surprisingly reluctant stance given its national expertise in bipartite bargaining. In the final stages of the October negotiations, the German BDA suggestion that in cases where social partner negotiations failed, the Council could only act by unanimity to approve new legislation. This suggestion almost blocked a final compromise text. The Commission mediator, Jean Degimbe, constantly reminded UNICE and the ETUC that the treaty would be drafted with or without their input and finally created a draft ‘cobbled together with bits of texts from both sides’ that was signed on 31 October (Dolvik 1997, p. 202). Notably, the final text was drafted by the Commission mediator, although agreement was subject to the agreement of the delegates from the ETUC, CEEP and UNICE. The joint statement included the innovation of setting a nine-month time limit for negotiations, and provided for implementation of social partner agreements via a Council decision (CEEP et al. 1991). The Ad Hoc Group signed off on the compromise text without going back to their national organizations for approval.13 In the view of one commentator, UNICE was ‘hedging its bets’ with the joint statement, knowing that a UK veto could prevent any new Treaty competencies in Social Dialogue (Ross 1995b, p. 380). The joint submission to the IGC was thus the product of ‘tactically coinciding but strategically contradictory interests’ (Dolvik 1997, p. 209); the ETUC was genuinely interested in negotiations, while UNICE was interested, at the very least, in preventing stronger wording in the text, such as ‘negotiate or legislate’ and extended QMV, and at the very most was hoping for a UK veto of all new social policy provisions.
Governance by Collective Bargaining 81
December 1991: the final negotiations at Maastricht The Dutch presidency forwarded the joint ETUC–CEEP–UNICE statement to the IGC. The inclusion of the text of the joint statement in the subsequent draft treaty was a significant reward for the entrepreneurial efforts of the Commission, and consistent ETUC demands for EU regulation via collective agreement (Council 1991d). In the final negotiations at Maastricht in December, the Dutch presidency introduced a new negotiating text that watered down the extension of social competencies and QMV so as to get UK approval (Ross 1995b, p. 380). Germany, Ireland, Portugal and Spain14 also ‘worked over the year leading up to Maastricht to water down the draft Social Chapter’ (Compston 2001b, p. 121). This would indicate that several low-standard countries and Germany were active in reducing areas where QMV would be accepted, and that their position was adopted by the Dutch presidency to bring the UK to agreement. The Dutch proposal was, in any case, rejected by the British delegation, as were proposals that would have involved British opt-outs on specific elements of new legislation. On the last evening of the negotiations, Major remained unmoved, offering only to agree to an expansion of social policy if it remained under unanimity voting. For his part, John Major has stated that he rejected the Social Protocol largely for domestic economic reasons, namely the costs new social legislation would have placed on employers and concerns about unemployment. He has rejected the notion that his refusal was on the basis of internal Conservative Party opposition (Interview 6).15 However, given sources that state the CBI would have been willing to accept Social Dialogue (Dolvik 1997, p. 203; Forster 1998), and suggestions by other researchers that Employment Minister Howard did threaten to resign (contrary to Major’s own statement) (Falkner 2002, p. 108), the evidence is mixed between national social policy history and ideological explanations for the British rejection. Major’s standstill proposal was at once rejected by Kohl (Moravcsik 1998). As a result of the UK rejection of all proposed versions of a social policy agreement including QMV, and German insistence on a higher level of agreement, negotiations looked poised to fail altogether. Commission intervention at this point was thus crucial, as Delors proposed a Social Protocol that would allow eleven Member States to move forward in limited areas using QMV, and an enhanced Social Dialogue on the basis of the 8 November Dutch Draft Treaty. The UK would opt out of the Social Protocol in its entirety (Ross 1995b, p. 380;
82 European Welfare States
Pierson 1998, p. 55). Kohl advanced the Commission proposal to the heads of government, and it was adopted. Thus, the agreement on the Social Protocol at Maastricht, significantly moving 11 Member States along a cooperative path while excluding the UK, was an awkward compromise between policy leaders and the most resistant Member State brokered by an entrepreneurial Commission. Even an intergovernmental analysis of the Maastricht negotiations concedes that Delors’s intervention ‘moved the resulting policy slightly in the direction of a more active social policy’ (Moravcsik 1998, p. 454). In the final Agreement, art. 2 of the Social Agreement extended QMV to a host of new issue areas, such as working conditions, the information and consultation of workers, gender equality in the labour market and the integration of persons excluded from the labour market. In the absence of the UK, a blocking minority would require at least three Member States. The Social Dialogue was included in arts. 3 and 4 (now arts. 137–9) as a means of both negotiating and implementing EU legislation in areas covered by the Treaty.16 The European Commission is to consult the social partners in advance of proposing social policy so as to give them the option of negotiating. As Paul Pierson has succinctly explained: ‘Member state governments, which had exploited Britain’s expected position to engage in cheap talk, suddenly found themselves exposed’ (Pierson 1998, p. 55) and locked in to agreement.17 The French government ‘tried to toughen the text [further], but this was vigorously opposed by the Iberian countries’ (Dolvik 1997, p. 205). Addressing the issue of side-payments at Maastricht: why did southern Member States agree? There is a gap between expectations and outcomes in the case of the behaviour of two southern EU Member States, namely Greece and Spain. Greece had no history of effective national-level bargaining, and Spain’s had collapsed in 1986. Further, union density in Spain was very low. These southern Member States contrast with Italy, which has a tradition of national level agreements and medium union density, and Portugal, which has a history of tripartite bargaining and medium union density. Lange, in an influential article on the Maastricht negotiations, described agreement on the Social Protocol as the result of ‘sidepayments’ – economic transfers to the poorer Member States to help with the costs of integration and combat domestic political opposition (Lange 1993, pp. 22–3). However, while his argument has relevance for poor Member State agreement for EMU, it is not at all evident that poor
Governance by Collective Bargaining 83
Member States agreed to QMV in the social policy field or to Social Dialogue on the basis of side-payments. In the case of Greece, the southern welfare state chosen as a case study for this book, this explanation does not hold true for several reasons. The first is that Greece had suggested the extension of QMV to the IGC as part of its initial submission in May 1990, before Spanish suggestions for enhanced structural funding (Greece 1992). Second, given that a government of the centre-right was in power (New Democracy), and that the opposition was socialist (PASOK), domestic opposition would not result from agreeing to the social aspects of Maastricht. Indeed, PASOK supported the Treaty when it was put to Parliament for ratification, although it did accuse the ND government of using EMU as an excuse for adopting unpopular economic and social measures ‘while avoiding the political costs’ (Koliopoulos 1994, p. 122).18 Given relatively strong domestic support for EU social policy (52 per cent in favour in 1990), the electoral costs of agreeing to the Social Protocol would not have been high (Compston 2001b, p. 104). Third, an interview with a member of the then ND government suggests that ND agreed to Social Dialogue because it thought that art. 118(b) was ‘a means for adapting to modernization, including working conditions, employment, and social security’ (Interview 1). Greece wanted to emulate other EU Member States as a development and modernization strategy. An adviser to the Greek Ministry of Labour stated that Prime Minister Mitsotakis was in favour of the watered-down Dutch proposal at Maastricht, but ‘the Greek delegation was not unhappy with the Social Protocol either’. He further stated: ‘I don’t believe ND approval for the Social Protocol was tied to any other policy’ (Interview 17). Side-payments were simply not necessary in order to get Greek agreement to Social Dialogue, or to the perhaps more contentious QMV procedures in the Social Protocol of 1991. Lange himself admits that his ‘side-payments’ argument may better explain EMU than QMV and the inclusion of Social Dialogue in the Social Protocol at Maastricht (Lange 1993, p. 23). Falkner also uses a side-payment argument, but in order to explain union agreement for EMU rather than poor country agreement for social policy (Falkner 1998, p. 179). The idea that Social Dialogue was a necessary sidepayment to unions (and not poor Member States) has also been suggested by former UNICE Secretary General Tyszkiewicz (Tyszkiewicz 2002b, p. 17). Lange also rejects the view that poor countries acted in their longterm economic interest by pursuing a ‘developmentalist strategy’ (Lange
84 European Welfare States
1993, p. 21), but does not marshal any evidence to suggest why this explanation may be rejected. The difficulty of proving or disproving that Greece acted on the basis of its long-term economic interest is acknowledged, but certainly it cannot be dismissed outright. The desire of poor Member States to improve their national systems, rather than stay underdeveloped relative to other EU Member States, is constantly underestimated in EU literature (Pelkmans 1997; Moravcsik 1998; Scharpf 1999). Interviews in Athens indicate that the desire for modernization via the adoption of EU-level policies was indeed a factor in the Greek decision-making process, and evidence from the Spanish case indicates that Spanish governments have agreed to successive social policy developments in the Treaties because of a desire to modernize the Spanish labour market and the availability of European Structural Funds to aid this transition (Argandona 1997, p. 207). The relationship between EMU, Cohesion Funding and the Social Protocol remains murky. It is, admittedly, reasonable to suppose that poorer Member States would not have vetoed the whole Treaty, as the UK did, given the additional cohesion funding promised (Compston 2001b). However, the central argument here is that a direct sidepayment – Cohesion Funds for Social Protocol – did not occur. As Falkner has explained, as a result of ‘consistent communicative action’ by working groups and the Commission ‘even those governments with adverse economic interests came to Maastricht ready to strengthen the “social dimension” … which was … perceived as a legitimate and necessary corollary to the Single Market’ and to EMU (Falkner 1998, p. 96). Social Dialogue was also viewed as an extension, in some cases, of national systems (Ireland, Portugal), and as a means of developing national social partnership in others (Greece). Table 3.3 summarizes the explanatory factors for the development of the Social Dialogue as a new mode of regional governance, both pre-Maastricht and during the IGC. The passivity of most Member States in the face of Commission entrepreneurship is notable. With national histories that had familiarized them with patterns of tripartite and bipartite bargaining, most Member States were willing to go along with a new regulatory method provided the social partners themselves could agree. The UK was the notable exception, as it was resistant throughout negotiations to the idea of extended QMV and to Social Dialogue. In such a context the consistent role of the Commission in shaping and leading discussions on Social Dialogue was the key explanatory variable.
85 Table 3.3 Factors determining the evolution of Social Dialogue at the EU level Time
The development of Social Dialogue, 1955–90
Maastricht negotiations on Social Dialogue, 1990–1
Leader Member States
Netherlands proposed ESC.
Belgian government ETUC Dutch, Danish, Belgian employers
Resistant Member States and interests
UNICE
UK Portuguese, Greek, UK employers
Commission entrepreneurship
1. Tripartite Conferences 1974–8 2. Standing Committee on Employment, 1970 (under control of Council) 3. 1983 tripartite consultations 4. Val Duchesse discussions 5. Proposed art. 118b for SEA 6. Social Action Plan adhoc group
Commissioner Papandreou and Delors suggested Treaty working group to social partners via Social Action plan ad hoc group. Delors proposal of the Social Protocol to the IGC, forwarded by Helmut Kohl.
Other institutional factor Rules
Failure of unanimity to pass legislation. Highly limited area where QMV applied (only on health and safety under 118a).
Employers and unions empowered to negotiate agreements; regulatory alternative to Commission and Council. QMV extended.
Learning effects in policy design
Commission would continue to propose legislation. Possibility of extending QMV.
Factor
Financial implications Unanticipated consequences
Cohesion funding?
Val Duchesse became the basis for a formal Social Dialogue.
UK opt-out not total; implications for UK firms. Future ECJ Decisions.
86 European Welfare States
The Framework Agreement on part-time work: negotiations and national implementation There was great initial confusion surrounding the Social Protocol. Some experts argued that the Agreement was not part of the corpus of EC law but rather an intergovernmental agreement (Vogel-Polsky 1994). Others argued that the Agreement permitted legally binding action at the EC level (Bercusson 1994). Additional legal questions, such as how a conflict of law between an EU agreement and national law would be determined, and whether the Commission and Council had the power to amend agreements once concluded, remained (Dolvik 1997, p. 222), demonstrating a high degree of uncertainty as to the consequences of the Protocol. Once contentious referenda on the Treaty had been concluded (notably the Danish rejection in June 1992 and the close French referendum in September 1992), the key question for the future was whether the Social Dialogue would enable agreement to occur at the EU level, and allow implementation at the Member State level. Would the Social Dialogue overcome policy deadlock? Negotiations on the part-time work Framework Agreement The areas chosen for Social Dialogue from 1994 to 1998 are ones in which agreement in the Council was blocked. Social Dialogue was thus seen as an alternative avenue to regulation in areas where resistant states had enacted a veto on EU-level legislation. Importantly, although the UK had opted-out of the Social Dialogue, the TUC continued to participate actively in the ETUC (Lea 1998, p. 136). The CBI also continued to participate in UNICE, although with very unfavourable results for the first Social Dialogue on European works councils. This first Social Dialogue negotiation failed largely owing to the opt-out of the CBI in mid-negotiations and the subsequent collapse of the UNICE position. The Commission then proposed a Directive that was approved by QMV under the Social Protocol and in the absence of the UK (Directive 94/45/EC). Once the Directive was passed, it was clear that British firms had obligations whether their government had opted-out of the Social Protocol or not, for example in the provision of works councils in subsidiaries of British multinational firms located in other EU countries.19 Subsequent to this poor start, it was agreed at UNICE ‘that the CBI would continue to participate but not have veto rights’, nor be bound by Framework Agreements of which it did not approve (Falkner 2000, p. 9). The Social Protocol had thus left ‘tremendous room for unanticipated consequences’, and already by 1994 it was clear that the
Governance by Collective Bargaining 87
UK opt-out was not as ‘airtight’ as John Major had expected (Pierson 1998, p. 55). The UK government had vetoed a Commission proposal on parental leave in September 1994, and this was the subject of the second Social Dialogue and the first Framework Agreement between the social partners. The Agreement provided a ‘low’ standard ‘if compared to what the more advanced Member States already had before, but constituted significant progress … for Ireland, and advances in detail for several other countries’ (Falkner 1999a, p. 93). However, the details of the Agreement require adjustment even in high-regulation states, such as Sweden, where the new EU rules allowed a parent to take time off before the birth of their child. The Council approved the Directive unanimously in March 1996, again without the UK (Directive 96/34/EC). Over time, it has thus become clear that the Council cannot amend the Framework Agreement as negotiated by the social partners, but rather approves it, or not, in a kind of ‘take-it-or-leave-it situation’ (Falkner 2000, p. 17). The third topic for Social Dialogue negotiations was various forms of atypical work. Negotiations on atypical work – including part-time, agency and temporary work contracts – proved contentious, coming as they do at the nexus between the security of full-time work and the flexibility of employers to hire and fire workers. By extension, labour market policies on the rights of atypical workers to partly employerfunded insurance and health schemes also affect their takeup of other forms of social policies such as social assistance, in the case of long-term unemployment or old age, and state-funded health services. Definitions and rules on part-time work, and the treatment of parttime workers, vary considerably in the Member States. In Greece, legislation approved in 1990 gave part-time workers the right to a minimum pension, unemployment benefits and severance pay, and made employer social contributions for part-time workers proportional to earnings (Kottis and Kottis 1996, p. 8). Still, owing to several factors, the most important of which is the low wages paid for full-time work, part-time work is not popular among men or women. However, the issue of pay would be explicitly excluded from any EU-level discussions. In Sweden and Denmark part-time work was restricted by collective agreement, something employers were eager to change. Many unions saw part-time and other forms of atypical work as a threat to full-time jobs. In the UK, although minimum standards for maximum working time have been established for health and safety reasons, the regulation of working time is ‘left entirely up to firm-level or individual agreements’
88 European Welfare States
(Lodovici 2000, p. 39). Part-time work is regulated by agreement, often at firm level, and not by legislation (Lecher 1994, pp. 100–1). In Germany, part-time work was regulated by both legislation and collective agreement (Lecher 1994, pp. 100–1). The most controversial issue for German employers concerned incentives for hiring part-time workers on the basis of reduced social security contributions that they did not want to see eliminated. In France, legislation had been passed in 1993 to allow greater flexibility in working hours. This legislation included requirements to reduce working time, but allowed working time limits to be assessed over a period as long as one year. The implementation of the law was ‘left to the social partners through collective bargaining’ (Gautié et al. 2000, pp. 256–7). Table 3.4 illustrates the growing incidence of part-time work in the EU as well as the percentage of part-time workers who are women as of 1998. By 1998 almost one-third of all working women (32 per cent) were engaged in part-time work, and one in six workers was part-time (Commission 2000b, p. 4). The incidence of part-time work has a great range in the EU, however, from under ten per cent of employment in Greece and Italy to over 20 per cent in the UK (Clauwaert 2001, p. 11). In the UK, nearly half of all women (45 per cent) work part-time and hourly earnings of part-time workers are 70 per cent of the average hourly earnings of full-time workers (Labour Research 1999). The ECJ had ruled on the conditions of part-time work in the 1980s, in so far as they refer to gender equality, in a series of cases involving Gabrielle Defrenne, an employee of the Belgian airline Sabena who complained of working time policies that discriminated against women (ECJ Table 3.4 Part-time work as a percentage of total employment, and women as a percentage of part-time workers, 1998 PT work as % of total employment, 1990
PT work as % of total employment, 1995
PT work as % of total employment, 1998
Females as % of part-time workers, 1998
France Germany Greece Sweden UK
12.2 13.4 6.7 14.5 20.1
14.2 14.2 7.7 15.1 22.3
14.8 16.6 9.2 13.5 23.0
79.3 84.1 63.6 97.3 80.4
EU average
13.3
14.9
16.0
81.8
Source: OECD (1999, p. 240).
Governance by Collective Bargaining 89
cases Defrenne I, II, III; also J.P. Jenkins v. Kingsgate Clothing Productions Ltd). The 1989 Social Charter reaffirmed the right of non-standard workers to receive equitable remuneration (EC 1989, art. 5). This wording was watered down, however, in the EU Charter of Fundamental Rights (2000).20 The Commission had proposed nine Directives21 relating to atypical work between 1982 and 1990, but only one Directive on health and safety for temporary workers proposed under art. 118a was adopted by QMV (91/383/EC). Two proposals on atypical work made under art. 100a were opposed by Germany and the UK (Rhodes 1991a, p. 17). The UK had also opposed the adoption of the 1993 Directive on maximum working time (Directive 93/104/EC), and took its objection to the legal base under which the Directive was proposed (art. 118a) to the ECJ, where it subsequently lost. The ILO had also proposed a Convention on part-time work in the early 1990s, but the idea that EU Member States would adopt this Convention in common proved an abject failure (this issue is covered in more detail in Chapter 5 on the ILO). When a proposal by the German presidency was vetoed in 1994, again by the UK, the Commission decided to consult the social partners about a possible Social Dialogue route to regulation of atypical work. Notably, the German proposal represented ‘the minimum of the minimum, below which nothing is conceivable in social protection’ in the words of DG V Commissioner Padraig Flynn (Jeffery 1998). The negotiations thus had to deal with the weight of past proposals and ECJ rulings on part-time work, the growing importance of part-time work as a form of employment, and the 1994 ILO Convention on PartTime Work (No. 175) which the ETUC expected the Social Dialogue to at least match, if not exceed, in providing for equality of treatment and job security for atypical workers. The Commission also stated in its opening consultations with the social partners that the 1994 German proposal was not to be taken as the starting-point for negotiations. In addition, there was fear among the social partners that governments might label Social Dialogue a failure given that it had only produced one Directive on parental leave (Lapeyre 2000, p. 188). In the shadow of law and the shadow of upcoming Treaty reform, therefore, the social partners started negotiations on ‘flexibility of working time and security for workers’ in October 1996. Initial negotiating positions were ‘very antagonistic’ (Falkner 1999b, p. 7), as the ETUC wanted to negotiate on all forms of atypical work including agency work, homework and telework, although in separate agreements. UNICE wanted to bargain on permanent part-time workers
90 European Welfare States
only. The ETUC also wanted social security included in discussions of non-discrimination, whereas UNICE/CEEP ‘were reluctant to include any reference to social security in the agreement at all’ (Falkner 1998, pp. 133, 136). By the fifth meeting of social partners it was agreed that the first subject to be negotiated on would be part-time work more generally, and not limited to permanent part-time workers (Falkner 1999b, p. 7). Bargaining on other forms of atypical work would come later. The key exchange of concessions during negotiations was that employers agreed that part-time workers should benefit from equal treatment compared with full-time workers, based on the percentage of a full-time job worked by the employee, while unions agreed to permit legal and contractual obstacles to the use of part-time work to be identified and eliminated (Branch and Greenwood 2001, p. 59). The importance of concluding a second agreement before treaty negotiations in Amsterdam in June 1997 was a huge spur to compromise. Both sides wanted to retain their ability to regulate in social policy, and concluding an agreement was the best proof that Social Dialogue represented a viable route to regulation. The final draft European Framework Agreement on parttime work was accepted on 15 May 1997 (Directive 97/81/EC) and approved by the Council, on the basis of a Commission proposal, in December (EC 1997). After the UK signed the Social Chapter the Directive was also extended to the UK in Directive 98/23/EC of 7 April 1998 (EC 1998). The Directive defines a part-time worker as ‘an employee whose normal hours of work calculated on a weekly basis or on average over a period of employment up to one year, are less than the normal hours of work of a comparable full-time worker’ (clause 3:1). The self-employed are excluded from the Agreement, and casual workers may be excluded ‘in conformity with national industrial relations practice’ (clause 2:2). The Directive that turns the Agreement into binding EU law may be implemented by national legislation or collective agreement (EC 1997). National implementation of the Directive on part-time work Opinions on the impact of the part-time work Directive vary, and depend on national situations before the Directive. The German DGB union confederation viewed the Agreement as being too weak. The Agreement avoided the issue of equality of treatment between full-time and part-time workers in social security provisions – an issue of special importance in Germany where, the Commission estimated, the employment of a part-time worker cost 36 per cent less to employers than in the Netherlands despite almost identical rates of pay owing to
Governance by Collective Bargaining 91
different thresholds at which employer contributions to social insurance schemes started (Commission 1998b, p. 3). The DGB and the ETUC Women’s Committee opposed ratification (as did the European federation of the building sector, the Luxembourg Christian confederation and the French FO), but were out-voted inside the ETUC, which had adopted its own version of QMV procedures in 1991 (Dolvik 1997, pp. 341–2). The European Parliament, in its opinion on the Directive, criticized the numerous exceptions permitted to the principle of nondiscrimination, and noted that the Directive fell short of the ILO Convention of 1994 (Parliament 1997). On the other hand, the Agreement will ‘have the most far-reaching implications’ for the UK ‘because of the absence of current labour law protection for most part-time workers’ and because the Agreement will bring the national social partners (TUC and CBI) into discussion on the implementation of the Directive (Dolvik 1997, p. 341). The new legislation on part-time work thus has had a variety of impacts in the Member States. The UK: wide implications of EU legislation narrowed by national efforts The Agreement has the biggest substantive impact in the UK, where parttime workers were not covered by statute and where the Agreement ‘provides a number of new rights for workers and consequent obligations for employers’ (Clauwaert 2001, p. 13). However, owing to the restrictive definition of part-time workers provided in the Agreement, which require comparison to a full-time worker doing similar work, one expert has suggested that ‘only 7 per cent of the 6 million part-time workers concerned in the UK would benefit from the Directive’s protection’ (Clauwaert 2001, p. 14). UK unions feel that ‘the [Labour] government strategy is to minimize the Agreement and reduce it as much as possible’ (Interview 62). The Agreement does, however, represent a sea in the thinking of employers about equitable treatment for part-time workers. Whereas in the early 1990s employers viewed EU regulation as ‘likely to inhibit their flexibility to determine the labour requirements and conditions most suited to their particular circumstances’ (Byre 1993), by the time the Framework Agreement was negotiated, ‘UK employers had accepted that non-discrimination was a good idea’ (Interview 52). The European Commission has estimated that the Framework Directive on part-time work covers only 30 per cent of workers in the UK, the lowest coverage in the EU next to Luxembourg at 68 per cent (Commission 2000b, p. 11). The wide disparity between the UK and the rest of the EU demonstrates how flexible
92 European Welfare States
forms of integration will allow a very determined Member State to limit the national effects of EU legislation. Greece: explicit national standards and empowering national actors Given that self-employed persons are excluded from the Agreement, and that over 30 per cent of Greek workers are self-employed (Commission 1999), there is a large gap in the coverage of the Directive in the Greek case. However, the legislation did prove beneficial in stating that parttime workers are to be compared to full-time workers on a pro rata basis, something that was accepted but not formally part of Greek law (Interview 17). If anything, unions were surprised at the amount of flexibility left to national measures (Interview 17). Still, the trade unions feel that the experience of European Social Dialogue has strengthened their role vis-à-vis the state and employers (Interview 54). Employers, on the other hand, feel that EU-level Social Dialogue ‘pushes us to higher standards that are difficult to meet’ because of the presence of ‘other Member States that are very progressive’ (Interview 48). A national Social Dialogue on flexible forms of work was attempted around the time of the Directive, but the national social partners failed to reach an agreement on the government’s text. Efforts have been made to establish a national Social Dialogue to discuss labour market issues, most notably in the creation of OKE (Economic and Social Council of Greece) in 1994. The establishment of new bipartite and tripartite institutions in Greece ‘was heavily influenced by the Delors Presidency of the EU Commission’ (Ioannou 2000, p. 221). However, owing to traditional weakness and politicization, the social partners ‘were simply not ready to participate’ in a national Social Dialogue on labour market reform 1997 (Interview 65). Others call OKE ‘a toothless talk-shop’ that is powerless in the face of continuing patterns of clientelistic politics (Interview 71). National social partners continue to be very divided on labour market reforms. EU legislation is thus able to impose minimum standards ‘geared towards greater efficiency’ in the face of continuing national deadlock (Interview 17). Greek politicians even wish ‘for more EU rules on national social dialogue’ that would ‘mandate the inclusion of certain groups’ so as to bypass fractious national debates (Interview 1). Sweden and Denmark: stabilizing Social Dialogue while opening debate Nordic unions saw the Agreement as ‘a positive step’ towards ‘reducing the scope for social dumping’ (Dolvik 1997, pp. 340–1). As one Swedish
Governance by Collective Bargaining 93
union official noted: ‘Directive by Directive we are levelling out working standards in the EU upwards [and are] diminishing the space for companies to compete on the basis of bad working conditions’ (Interview 55). The Agreement also presented the opportunity for national collective agreements to push for the abolition of threshold provisions for national occupational schemes even if these were not explicitly covered in the Agreement (Dolvik 1997, pp. 340–1). In the Nordic countries the Agreement has kicked off a national debate on the benefits and drawbacks of encouraging part-time work, given that the Agreement obliges the social partners to eliminate barriers to parttime work. The impact on Sweden and Denmark may be less substantive in the short term, but greater in the long run, as unions are forced to rethink their objections to atypical work. In Sweden, the Social Dialogue has proven a surprise for Swedish employers who thought they would get a ‘union-free [European] Union’ (Interview 60). Social Dialogue created an ‘even bigger problem’ than getting out of national-level collective bargaining, as participating in the EU-level Social Dialogue was a requirement of membership (Interview 60). However, while employers are not big supporters of the Social Dialogue because ‘we have the experience of the 1970s’, employers have ‘little to worry about with EU rules and regulations because we already have such legislation in Sweden’ (Interview 50). The EU Social Dialogue provides a stable role for union and employer concertation in the implementation of EU Directives via collective agreement even while national agreements are in flux. France: national legislation surpasses EU-level efforts The most extensive reregulation of working time occurred in France, shortly after the introduction of the 1997 part-time work legislation. The 35-hour work week legislation was promised in the 1997 election by Jospin. The introduction of the Aubry Law, adopted in May 1998, left the details of its implementation to local negotiations, and offered employers the opportunity to reorganize their labour force via weekly working time reduction or an extension of holidays in order to conform to the new legislation (Casey and Gold 2000, p. 47). The legislation was implemented by 1 January 2000 for larger firms, and by 1 January 2002 for firms with fewer than 20 employees (1999). France thus has the most regulated system of working time in the EU (Lodovici 2000, p. 40), and is also the only Member State to have exhibited higher regulatory standards in the areas of working time and conditions (rather than status quo or deregulatory trends) over the past decade (OECD 1999, p. 61).
94 European Welfare States
There is no evidence that the Aubry Law was inspired by EU legislation, but rather by attempts to boost national employment. Unanticipated consequences: using Social Dialogue to push the boundaries of EU competence Given the ambiguity of much of the text of the Agreement, and the different national interpretations that will result, there is much scope for future ECJ Decisions, further regulation, and thus for unanticipated consequences. A level of ambiguity is necessary for agreement, but this comes at the cost of a certain level of uncertainty regarding the future interpretation of the Agreement. Other Framework Agreements, such as that on fixed-term work concluded since the part-time work Directive, (Directive 99/70/EC) also fail to define ‘objective grounds’ for discrimination against atypical workers. However, the ECJ’s flexibility to interpret the meaning of ‘objective grounds’, if the Agreement is challenged on this basis, will run up against common patterns of justification in Member States (Bercusson and Bruun 1999, pp. 112–14). The annex to the Directive also notes: Without prejudice to the role of national courts and the Court of Justice [ECJ], the parties to this agreement request that any matter relating to the interpretation of this agreement at European level should in the first instance be referred by the Commission to them for an opinion. (Preamble to Framework Agreement, annexed to the Directive, EC 1997) The social partners are thus anxious that they be given the right of first clarification in light of any disputes over the meaning of the Agreement. One trade unionist stated that, in asking the Commission for guidance on the temporary-agency work negotiations, ‘the Commission reported to us verbally but not on paper that pay issues are in fact included in conditions related to equal treatment’ (Interview 62). Unions may thus use the Social Dialogue to slowly introduce competencies excluded by the EU Treaties. However, in a statement annexed to the fixed-term work Directive, the UK government underlined that the Framework Agreement does not apply to pay, the right of association, the right to strike or the right to impose lock-outs, which are excluded from the Treaty under art. 137 (UK 1999). It is clear that both social partners and governments wish to assert a stake in delineating where ‘the shifting sands of moving texts’ will take the EU, leaving less room for the ECJ to fill gaps (Bercusson and Bruun 1999, p. 62).
Governance by Collective Bargaining 95
Conclusion Evidence from EU-level activity in the period 1970–90, and during the IGC leading to the Maastricht Treaty, demonstrates that the Commission played the key role in introducing and developing Social Dialogue as a form of social policy governance. Predisposed to collective bargaining by national histories, convinced by the Commission that Social Dialogue was a vital corollary to economic integration, and in many instances frustrated with deadlock on social policy in the Council, Member States were willing to let the Social Dialogue develop largely through Commission initiatives to bring EU-level social partners to agreement on the form this cooperation might take. The evolution of Social Dialogue as a form of regional governance must therefore be seen as part of a back-and-forth process between Commission entrepreneurship and the pressure of interests, namely labour, for binding collective bargaining. Using its financial resources, but more importantly its resources as an agenda-setter for daily activities and legislative action plans in the Union, the Commission helped to empower trade unions at the EU level, who then requested Treaty reforms including a formal procedure for EU-level collective bargaining. A clear Treaty base in the SEA prompted further activism by providing a facilitating role for the Commission. Despite great resistance on the part of the UK, the EU-11, most notably Kohl and Mitterrand in the final negotiations at Maastricht, were convinced of the need to move ahead on social policy. Again the Commission stepped in, at the intergovernmental moment of the Maastricht IGC, to suggest the compromise of the Social Protocol and the UK opt-out. Since the Maastricht Treaty, the social partners themselves have played the key role in developing specific legislation, although the constant pressure of possible legislative action by the Commission is clearly necessary for UNICE’s participation in the process (Interview 62). Still, UNICE views its direct involvement in negotiations to be preferable to Commission activity, and its presence in the Social Dialogue is one that allows it ‘to bypass, slow down, and manage social policy rather than letting the Commission legislate’ (Interview 37). UNICE’s reluctance is understandable. Trade unions were losing ground in Europe in the 1980s and 1990s, owing to membership loss and the decline of national-level bargaining, and the Social Dialogue has helped unions to gain at the EU level at the very moment that they were losing ground in Member States (Interview 55, 56). As one author has
96 European Welfare States
stated, ‘Unions did not regain in Europe the power they lost at home. Yet … borrowing from Alan Milward … Europe has come to the rescue of national unionism in economic and political lean times’ (Dolvik and Visser 2001, p. 39). The key national effect of the EU-level Social Dialogue is the re-empowerment of unions where national collective bargaining was on the decline, such as in Sweden, and the enhancement of union participation in social policy where it was weak, such as in the UK and Greece. At a time when national industrial relations are in great flux, the Social Dialogue anchors national social partners in supranational decision-making and perpetuates governance by collective bargaining in the social policy field. The effects of the policy outputs of governance by collective bargaining depend on the national context before Framework Agreements are enacted by Council Directives. Again, legislation has effects in all Member States, even in the case of policy leaders. However, it is clear that EU legislation has the greatest effect on policy resisters. The flexibility provided by national implementation means, however, that resisters act to limit the effects of EU legislation. The enforcement mechanism of the ECJ is thus essential to counteract lacunae in national implementation, should national test cases be brought before it. The attempt by the social partners to limit the ECJ’s influence by requesting the right of opinion ‘in the first instance’ (that is, before the ECJ) indicates that the boundaries of the application of the Directive are clearly contested. Perhaps the best description of the Social Dialogue is as ‘a moving feast, it is constantly changing’ (Interview 52). Specific policy outcomes are dependent on a highly complex set of variables that include the specific issue at hand, the legal traditions of Member States, the preferences of unions and employer groups, and the ways in which EU-level confederations mediate the interests of their diverse members. The most significant backdrop to the Social Dialogue remains, however, the Commission’s ability to propose legislation in cases of the failure of the Social Dialogue, and the Commission’s ability to work away at closing gaps left after the negotiators have concluded.
4 Governance by the Open Method of Coordination: The European Employment Strategy
What caused an issue that had been floating around various EU-level policies for four decades to make the leap onto the Treaty agenda, and finally into the Treaty of Amsterdam? The research presented here suggests that a welfare state explanation of preferences, based on national spending on labour market policies and unemployment rates, does a good job of identifying the Member States who led the creation of an EU employment policy, namely Denmark and Sweden. The Danish presidency of 1993 helped put employment policy on the agenda of the EU, and Sweden consistently pushed for an employment strategy in the 1996 IGC. High unemployment rates in southern welfare states (with the exception of Portugal) also partially explain why Member States with low rates of expenditure on labour market policies may have found EU-level cooperation appealing, and were thus passive supporters of an employment title rather than policy resisters. A welfare state explanation of preferences also does a good job of identifying the most resistant state, the UK, whose low spending on employment policies and lower-than-average unemployment rate would suggest non-cooperation in employment policy. However, with a change in government from Conservative to Labour leadership, the UK agreed to cooperate on employment policy. The ideology of governments concerning European integration must, therefore, also be considered in order to explain fully Member State preferences. A combination of a national welfare state explanation and government ideology explains Member State preferences. However, to explain the outcome of agreement to include an Employment Title in the revised Treaty, and the particular features of the governance regime it created, several factors suggested by historical institutionalism must be considered. Three factors – Commission entrepreneurship, learning 97
98 European Welfare States
effects, and the role of unanticipated consequences – are necessary in order to explain fully the specific features of employment policy. Ongoing Commission work to promote cooperation in employment, and policy entrepreneurship by key Commission officials, is especially important. The evolution from policy proposal to Treaty base to concrete policy is thus a complex one. Only careful tracing of the development and timing of employment initiatives allows an assessment of the national and institutional factors that explain the features of the new form of governance embodied in the European Employment Strategy (EES) and the open method of coordination (OMC). Process-tracing also allows answers to several puzzles, such as how integration proceeded without backing from the largest Member States (Germany, France and the UK), and why employment policy was initially suggested by non-EMU countries, thereby suggesting that, at its impetus, employment policy should not be considered a direct spillover from EMU. This chapter is divided into five sections. The first reviews employment policies at the national level, and presents a welfare state explanation of employment policy preferences at the EU level. Sections two, three and four examine the evolution of employment policy at the EU level in different stages: 1954–94, during which time informal cooperation in employment policy took place at the EU level; the 1996–97 IGC leading to the Treaty of Amsterdam and the Employment Title; and 1997–2001 during which time the EES solidified and the OMC was developed. In each of these sections the key explanatory variables will be traced: leader and resistant Member States, and institutional factors such as Commission entrepreneurship, learning effects and unanticipated consequences. Section five returns to the national level to assess the impact of the EES on national policy. By careful process-tracing of the timing of national and EU-level policies, the influence of the EES on national policies will be assessed.
Welfare state preferences in employment policy Employment policies are designed to influence inflows and outflows to the labour market; in other words, employment policies aim to influence the numbers of individuals taking or giving up employment. Policies used to influence labour markets are described as either passive measures – such as income maintenance programmes, early retirement, reductions in working time, regulatory interventions such as those to set
Governance by the Open Method of Coordination 99
rules for the hiring and firing of workers, and wage policies – or active labour market policies (ALMPs) – such as job brokering and placement services, worker training, subsidized employment schemes, direct job creation in the public sector, and the targeting of employment policies towards specific groups, such as youth, women, or the long-term unemployed (Calmfors and Skedinger 1995, p. 91; de Koning et al. 2001, pp. 1–2). Each member state has its own unique mix of passive and active labour market policies and labour market institutions. Sweden and Germany have long national histories of ALMPs, often implemented via collective agreements between unions and employers but also by government legislation and programmes. Sweden is especially notable for its development of ALMPs and full employment as a key aspect of the economic policy mix. In Germany, the federal division of responsibilities for employment policy make the Länder key actors in policymaking. The UK and France have a much weaker record in ALMPs. The UK especially has emphasized those kinds of ALMPs that help jobseekers find employment, rather than skill development or training. In the French case, government legislation emphasizes passive labour market policies such as the reduction of social security contributions paid by employers for new employees, and the reduction of working time to boost employment. Greece has almost no national history in the domestic development of ALMPs. Greece’s use of ALMPs in the 1980s was the result of EU initiatives, initially via the European Social Fund (ESF). The welfare state regime of Greece, and of the southern welfare states more generally, are thus very weak in the area of ALMPs. The use of passive measures, notably early retirement (especially in the public sector) is more common, although increasingly regulated in the 1990s. A national welfare state explanation for national preferences would suggest that those Member States with high levels of spending on employment policies would be in favour of EU cooperation so as to promote the adoption of employment policies by other Member States. Although it is far from clear that higher than average spending on employment policies should be viewed as a competitive disadvantage, and thus something to export to other Member States, higher levels of spending indicate a greater commitment by these Member States to intervening in labour market entry and exit and, as such, they will want to see the approach of more laissez-faire states mitigated by common rules. Table 4.1 illustrates the allocation of labour market spending in 1996, and 1990 for comparison. On the basis of 1996 data on expenditure on
100 European Welfare States
employment measures, we should expect that the following Member States with higher than average levels of expenditure to be supportive of employment policy at the EU level: Belgium, Denmark, Finland, Germany, Ireland, Netherlands and Sweden. A second-cut analysis of the national level also considers how unemployment rates may have promoted EU-level cooperation. Table 4.2
Table 4.1 Allocation of labour market expenditures: 1990, 1996 Member State and welfare state regime
Total expenditure on active and passive labour market policies, % of GDP
Denmark (SD) Netherlands (C) Finland (SD) Sweden (SD) Belgium (C) Ireland (L) Germany§ (C) France (C) Spain (S) Italy (S) Austria (C) Portugal (S) UK (L) Luxembourg (C) Greece (S) EU–15 average
Active expenditure (ALMPs*), % of GDP
1990
1996
1990
1996
5.66 3.22 2.16 2.57† 3.89 4.39 3.13 2.67†† 3.18 1.53 1.28† 1.14 1.57† 1.09 1.05 2.57
5.93 5.49 5.31 4.62¶ 4.24 4.07 3.92 3.13 2.69 1.94 1.78 1.74 1.70¶ 0.95 0.89 3.23
1.26 1.04 1.01 1.69† 1.23 1.54 1.35 0.80†† 0.76 0.69 0.31† 0.72 0.62† 0.36 0.55 0.93
1.78 1.51 1.69 2.36¶ 1.47 1.66 1.43 1.34 0.66 1.07 0.36 0.85 0.45¶ 0.28 0.45 1.16
SD ⫽ social-democratic; C ⫽ conservative; L ⫽ liberal; S ⫽ southern. The thick line in the column shows spending on employment policies in 1996 separates the countries that are above the EU average (above the line) from those below it. * ALMPS defined as: public employment services and administration, labour market training, youth measures, subsidized employment, measures for the disabled. Passive labour market measures are defined as unemployment compensation and early retirement for labour market reasons. †
1990–1;
††
1991;
§
1991, for united Germany;
Sources: OECD (1995); OECD (2000).
¶
1995–6.
Governance by the Open Method of Coordination 101
illustrates unemployment rates in 1996, and 1990 for comparison. On the basis of higher-than-average unemployment rates in 1996 the following Member States may favour EU cooperation: Belgium, Finland, France, Germany, Greece, Ireland, Italy, Spain and Sweden. Of countries with high unemployment rates, Belgium, Finland, Germany, Ireland and Sweden also have higher than average rates of spending on employment policies. These states should thus be the most enthusiastic about cooperation at the EU level. France, Greece, Italy and Spain have a more mixed national picture, with lower than average spending on employment policy but higher than average unemployment. These four Member States (France, Greece,
Table 4.2 Unemployment rates: 1990, 1996 Member State and welfare state regime
Spain (S) Finland (SD) France (C) Ireland (L) Italy (S) Belgium (C) Greece (S) Sweden (SD) Germany (C) UK (L) Portugal (S) Denmark (SD) Netherlands (C) Austria (C) Luxembourg(C) EU-15 average
Unemployment rate 1990
1996
Change, % (⫹ indicates expansion of unemployment)
16.3 3.2 9.0 13.4 9.0 6.7 6.4 1.7 4.2 7.1 4.6 7.7 6.2 … 1.7
22.2 14.6 12.4 11.7 11.7 9.7 9.6 9.6 8.9 8.2 7.3 6.8 6.3 4.4 3.0 8.87*
⫹5.9 ⫹11.4 ⫹3.4 ⫹1.7 ⫹2.7 ⫹3.0 ⫹3.2 ⫹7.9 ⫹4.7 ⫹1.1 ⫹2.7 0.9 ⫹0.1 … ⫹1.3
SD ⫽ social-democratic; C ⫽ conservative; L ⫽ liberal; S ⫽ southern. The thick line separates countries above the EU average for unemployment from those below it. * Without Spain, as a high outlier. The EU-15 average is 9.76 with Spain. Sources: OECD (1995); OECD (2000).
102 European Welfare States
Italy, Spain) may favour EU cooperation if they believe EU-level activity could provide strategies for reducing unemployment and be beneficial to improving the success of national employment policies. In the French case, government ideology towards cooperation is important to explain the final agreement to go ahead on employment policy cooperation after the election of Lionel Jospin and the Socialist Party (PS) in June 1997. Of the southern welfare states (Greece, Italy, Portugal, Spain), Spain had a Conservative government, Greece and Portugal had governments of the left, and Italy had a centre-left government, yet all four states were supportive of an EU-level employment policy. Institutional factors at the EU level will therefore be especially important in order to explain their positions, such as the cost and benefits of such a policy, and the implementation obligations placed on Member States. Notably, approval of the Employment Title was not linked to the provision of additional or special funds. A side-payments argument to explain the acceptance of Treaty change on the part of southern Member States cannot, therefore, be used (Lange 1993). Also on the basis of these two variables – employment policy spending and unemployment rates – it should be expected that Member States with low spending on passive and active labour market policies and low unemployment rates should resist EU policies. These states are Austria, Luxembourg, Portugal and the UK. In these four cases it will be expected that variables other than welfare state histories will be necessary to explain the gap between expected and actual behaviour, namely institutional factors such as Commission entrepreneurship – that convinced Member States over time of the value of cooperation on employment policy – and government ideology towards EU integration. All four of these Member States had pro-integrationist governments at the time of final negotiations in Amsterdam, three of the left (in Austria, Portugal and the UK), and one of the right (in pro-integrationist Luxembourg). The gap in the UK case thus does not appear under the Conservative Major government, which behaved as predicted by a national welfare state explanation of preferences in resisting EU employment policy, but rather under the Labour Blair government that agreed to the Employment Title. A welfare state explanation of preferences is thus very good at explaining why policy leaders behaved as they did. There is a high correlation between social-democratic welfare state regimes and a procooperation stance. However, a welfare state explanation is not sufficient to explain why states that should be policy resisters, especially the UK and Portugal with low employment spending and low unemployment
Governance by the Open Method of Coordination 103
rates as well as Liberal (UK) and southern (Portugal) welfare state regimes, eventually agreed to the policy. It also does not fully explain the reluctance of France or Germany. Institutional factors, the priority of EMU over other EU-level policies in bargaining, and government ideology towards EU integration must also be considered. The factors identified by a historical institutionalist analysis – Commission entrepreneurship, learning effects that attuned Member States to the possible budgetary and implementation costs of cooperation, and unanticipated consequences – are especially important in order to explain the features of the EES and the emergence of the OMC. Each of these institutionalist factors, as well as the leadership provided by Member States, will be traced in the following sections on the evolution of employment policy at the EU level. The explanation of outcomes provided in this chapter thus combines a welfare state explanation of preferences with an institutional analysis over time. Only by combining leadership by key Member States, often during their role holding the Presidency of the European Council, with institutional factors can the emergence of the European Employment Strategy as a new and highly flexible form of regulation be explained.
Employment policies in the ECSC, EEC and EU: 1954–94 Early EU-level activity concentrated on vocational training. Between 1954 and 1960 the High Authority of the ECSC ‘had granted 42,518 million units of account1 towards help given to 115,085 workers of whom 80 per cent were miners’ (Collins 1975a, p. 40). With regards to vocational training there was no direct mandate given to the High Authority, but art. 69 of the ECSC Treaty was interpreted by the High Authority as permitting activities to advance the employment of coal and steel workers. Art. 69 also concerned the mobility of workers. Having no right to impose common standards, the High Authority attempted to encourage the adoption of ‘new ideas’ through ‘a constant programme of committees, and study groups, seminars and study visits, publication of reports and the dissemination of information about courses’ (Collins 1975a, p. 60). The instruments used in the 1960s ‘relied upon the exchange of experience and information at both industrial and governmental levels as a means of making progress’ (Collins 1975a, p. 60). EEC employment policies were promoted through the European Social Fund (ESF), founded in arts. 123–7 of the Treaty of Rome. ESF financing may fund up to one half of a Member State project
104 European Welfare States
addressing redundancy, training, resettlement and help for migrant workers. In later ESF funding periods, this amount was increased to 65 per cent for projects in Objective 1 regions (with a per capita GDP close to or less than 75 per cent of the Community average), and 45 per cent elsewhere. A contribution by a member state government is therefore required in order to release an ESF grant. The majority of ESF funding (90 per cent) is concentrated on providing co-financing for vocational training, with a small percentage for employment assistance schemes (Commission 1991a). The first ESF (1958–71) financed the re-employment of over 1 million workers, the bulk of them Italians, through travel grants and language classes (Harrop 1996, p. 137). Guidelines for ESF funding have been revised several times to include programmes focused on women (1971), young people (1975) and the long-term unemployed (Collins 1983, pp. 36, 39). The focus on such groups was underlined in the Maastricht Treaty (1993). The Maastricht Treaty permitted QMV under the Social Protocol for policies to aid ‘the integration of persons traditionally excluded from the labour market, without prejudice to art. 127 [on the ESF]’ (art. 2, para. 1 of the Social Protocol, annexed to the Maastricht Treaty). The third ESF (1983–8) and fourth ESF (1989–93) continued the targeting of groups, and also provided a ‘regional concentration of assistance’ to sub-national areas where unemployment was particularly high, including regions in the UK (Commission 1986; Harrop 1996, pp. 137–8). By ESF IV, the ESF was integrated with structural and cohesion funds so as to concentrate on underdeveloped regions in Greece, Ireland, Portugal and Spain.2 Commission activism in employment policy greatly increased in the 1980s. Beginning with personnel exchanges between national employment ministries, the Commission began a systematic collection of information on national employment policies in 1982 called Mutual Information System on Employment (MISEP) (Silletti 1987, p. 14).3 At the 1985 IGC that was to lead to the creation of the Single European Act, the centre-right Danish government proposed that the Commission present an annual report on employment policy under a revised art. 122. If employment dropped below a certain percentage (not named), the Commission could be authorized to present proposals to the Council on policies to boost employment rates. However, the Danish proposal failed to be endorsed (Corbett 1998, p. 238). Despite the lack of a formal Treaty competence, informal cooperation in employment policy continued.
Governance by the Open Method of Coordination 105
Shortly after the completion of the SEA the Council asked for more information-sharing on employment policies in ECOFIN (Council Resolution 86/C340/02 of 22 December). This was partly in response to a May 1986 initiative by the UK, Ireland and Italy to refocus social policy on labour market and employment issues but with no new competencies for the EU level. The UK–Italian–Irish document proposed action in small businesses, flexible forms of employment, worker training and long-term unemployment (Gold and Matthews 1996, pp. 21–3). The Commission and other Member States, however, continued to see the social policy agenda in much broader terms, with employment policy as but one issue. The LEDA programme (local labour market development programme) of 1986–7 carried out a series of pilot projects on local employment initiatives in order ‘to generate … know-how from the experience of different types of [geographic] areas’ and ‘to disseminate this know-how to Community and national policy-makers and administrators and to local participants … across the Community’ (Commission 1997a). A Social Action Programme was launched to give effect to the 1989 Charter, and a DG V Steering Group was formed to discuss education, training and labour market policies (Commission 1989). The first annual Employment in Europe report was written in 1989 for dissemination to the Council, Parliament, social partners and the Standing Committee on Employment (Commission 1989). A fundamental shift thus occurred in the late 1980s, from regulation to promote the mobility of workers across frontiers to a two-prong strategy of regulation as well as an increase in the analysis and dissemination of information on national policies on employment. The sharing of information and best practice (or ‘good practice’) in employment policy is thus not new to the OMC method introduced by the Lisbon Summit (which is discussed in greater detail below). Denmark again raised the idea of ‘promotion of cooperation between the Member States to combat unemployment’ in 1992, during discussion of the Maastricht Treaty, but no Treaty competence emerged (Council 1992). Employment rates (having a certain percentage of the working population in employment, as opposed to measuring the numbers of unemployed) had also been discussed as possible convergence criteria for EMU in the Maastricht Treaty negotiations, but this suggestion ‘had been refused’ (Dehousse 1999, p. 37). The interest of Denmark in promoting supranational cooperation in employment policy was clear, and the Commission used the 1993
106 European Welfare States
Danish Presidency of the European Council to suggest the creation of a common framework for employment cooperation. At a meeting of the Employment and Social Affairs Ministers on 3–4 May 1993, Padraig Flynn, the Commissioner of DG V, presented a proposal for a ‘framework for actions’ at all levels within the Community in relation to employment. The response from the Council was that while recognizing that responsibility for employment action lies primarily with themselves, the Ministers underlined the need for supplementary coordination and action at the Community level in order to address the common problems of employment and unemployment. (Commission 1993a, p. 1) The Community-Wide Framework for Employment was the result. It committed the Community institutions and the Member States to a ‘structured, cooperative and systematic, process of analysis of policy reflection on possible solutions to the employment problem with a view to concerted policy action’ over an 18-month period (Commission 1993a, p. 7). Commission activism was thus empowered by a national government (the Danish presidency) with a history of active involvement in the labour market.4 Areas suggested for ‘analysis and action’ by the Employment Framework included ‘adaptability at the workplace’, ‘new working time structures’, ‘training systems’, ‘reducing labour costs … notably by modifying the incidence of taxation systems’, and ‘local initiatives … to create employment’ and the promotion of ‘Community pilot experiences’ (Commission 1993a, pp. 8–9). The May 1993 proposals were thus the critical starting-point for employment policies at the EU level and not the December White Paper, as is commonly suggested (Ross 1995a; Ross 1995b; Colucci 1998; Goetschy 2001). The main actor behind the May 1993 paper, and the suggestion of an ongoing work programme on employment to 1994, was DG V Commissioner Padraig Flynn. Flynn had joined the Commission from Dublin in January 1993, and was more influential than Jacques Delors in the initial stages of policy development (Interview 36). As for where the policy ideas in the May 1993 paper came from, one of its authors stated: ‘The ideas came from international thinking that was out there in economic circles – the OECD, ILO’ and ‘the conventional centre-left ideas of a network of Commission officials who worked on these issues’ (Interview 36). Commission officials working on draft papers ‘had to try to limit the reference to both sloppy Delors notions – service
Governance by the Open Method of Coordination 107
de promité, troisième secteur – and the draconian notions of Thatcher fed through Secretary General [of the Council] Williamson’ (Interview 36). Still, Delors’s thinking managed to make it into Part A of the paper via Jerome Vignon, a French Commission official. After the successful presentation of the Community Wide Framework for Employment, the Danish presidency gave the Commission a mandate to develop a White Paper on competitiveness and employment. The Community-Wide Framework for Employment had shown Delors ‘that there was political support out there’ for an EU-level employment policy. Delors, who had been reluctant to launch a European employment strategy to the Council, and worried that it might focus attention away from EMU, was willing in light of Flynn’s success to present his own plan for employment policy5 that then outpaced and overshadowed the original proposal (Interview 36). Commissioner Padraig Flynn went to national capitals in support of Delors’s new initiative, where he visited ministers of finance, employment and education, in order to collect ideas for the White Paper, and to sell the concept of greater cooperation in the employment field (Interview 36). The White Paper, presented in December under the Belgian presidency, emphasised issues that had been raised in the Community Wide Framework for Employment, such as ‘lifelong education and training’ and ‘greater flexibility’ in labour markets (Commission 1993d).6 Member states were to draw suggestions for the design of their employment policies from the Commission White Paper. The 1994 Essen European Council meeting followed up the White Paper, and identified the ‘fight against unemployment and equality of opportunity for men and women’ as ‘the paramount tasks of the EU and its Member States’ (Council 1994). Five ‘key areas’ were identified for national action, taking into account ‘the specific features of their [individual Member State] economic and social situation’: (1) investment in human resources through vocational training, (2) the organization of work in more ‘flexible’ ways (working time, wages, social security), (3) the reduction of non-wage labour costs, (4) increase incentives to work and adopting ALMPs, and (5) targeting groups such as women and the long-term unemployed (Council 1994). Labour and Social Affairs Councils, ECOFIN and the Commission were charged with monitoring employment trends and the policies of the Member States and were to report annually to the European Council, starting in December 1995. The first national employment reports were submitted in spring 1995. At the Madrid European Council of December 1995 a joint report from the Council (ECOFIN and Labour and Social Affairs) and the
108 European Welfare States
Commission was received and the creation of a second joint report was set for December 1996. The Council also stated that the five Essen criteria constituted an ‘employment strategy’ and a ‘framework’ for the development of Member States’ employment policies on a ‘multiannual’ basis (Council 1995, annex 2), and that ‘it is necessary to establish as soon as possible the mechanisms envisaged in the joint report’, namely a ‘stable structure and common indicators’ (Council 1995). Employment policy at the EU level from 1989 to 1995 was thus the result of Commission entrepreneurship, supported by successive presidencies and validated by European Councils. In this period, ‘no one talked about the process becoming an annual one, but it just started happening’, as the Council replied to Commission communications and reports, and as successive Councils tried to put their own stamp on the employment discussion (Interview 40). Thus, even in its informal stages, cooperation on employment policy had led to the unanticipated result of a multiannual process of reporting on national employment policies at the EU level on the basis of commonly articulated guidelines.
Amsterdam Treaty negotiations Background to negotiations: European context and national governments The Amsterdam Treaty was signed in June 1997. It was the product of two years’ worth of preparatory meetings: a 1995 reflection group including MEPs, the 1996 Intergovernmental Conference (IGC) of regular meetings of the foreign ministers, and several European Councils. The terms of reference of the IGC were ‘to bring the EU closer to the citizen’ (Parliament 1996, p. 11), complete the single market and EMU, and encourage ‘supplementary action’ [in addition to the SEM and EMU] ‘to fulfil the objective of a high level employment while ensuring social protection’. The Union was to ‘examine’ how it ‘could provide the basis for better cooperation and coordination in order to strengthen national policies’ in employment (Council 1996a, point 1). At the EU level, a new European Commission was struck in January 1995. Jacques Santer, the former PM of Luxembourg, was the president of the Commission. His approach to the 1995–7 negotiations has been described as ‘pragmatic’, but lacking in the ‘ambitious proposals’ present in the Delors presidency participation in IGCs (McDonagh 1998, p. 209). Allan Larsson, the former Swedish minister of finance with vast high-level contacts in the Nordic countries and among European social democrats, became director-general of DG V in May 1995.7
Governance by the Open Method of Coordination 109
Patrick Venturini, who had worked in Delors’s cabinet on social dialogue and employment policy, was one of Larsson’s senior advisers. The presence of Larsson as director-general with the integral Employment and Social Affairs portfolio in the 1995–7 period is critical. As a member of the Commission, Larsson was well-placed to receive information about the negotiating positions of Member States, pass this information along to other national delegations and Presidencies working on draft positions, and also use the information to draft Commission papers that tried to bridge the gap between enthusiastic and reluctant member states.8 The other vital background to the Treaty negotiations was the balance between national governments of the left and right. It will be illustrated that this divide was less salient than might be expected by intergovernmental theory, which predicted that socialist governments in wealthy countries would be the most supportive of an EU employment policy, and that conservative governments in poor countries would be the least supportive (Moravcsik and Nicolaïdis 1999, p. 63).9 In the end, both poor and rich countries supported the Employment Title, as did governments of the left and right. Intergovernmental theory does a good job of identifying policy leaders (socialist, rich Member States), but it cannot explain the agreement of poor and conservative countries. A national welfare state explanation, by going deeper into an analysis of specific national policies, can explain Irish cooperation and goes part way to explaining why Spain and other poor countries with high unemployment levels, such as Greece, agreed to cooperation. In June 1995, when the reflection group first met, there were eight governments of the left and six of the right; in June 1997, at the time of the final Treaty negotiations, there were nine governments of the left, five of the right, and one cohabitation government in France (Chirac– Jospin). Regime changes from 1995 to 1997 occurred in Portugal, Spain, the UK and France, as outlined in Table 4.3. The proposal for a separate title on employment: Swedish leadership As early as 1994, the Swedish government under social-democratic leadership had declared its intention to pursue ‘additional measures to fight unemployment’ at the EU level10 (Sweden 1994; Miles 1997, p. 265). The idea that the Swedish model of society could contribute positively to western European cooperation, especially in light of the failure of oldstyle social democracy in France in 1981–3, circulated in union and government circles (Bieler 2000, pp. 70–1). In other words, could the EU be ‘Swedenized’? (Olsen 1998, p. 361). Swedish unions, too, saw the need
110 Table 4.3 Member State political affiliations 1995–7 Member State
Rich or poor (1996)*
Party at head of government
Head of government
Political alignment of government, right or left †
Austria
Rich
SPÖ
Vranitzky to 28 January 1997; Klima
Left
Belgium
Rich
CVP
Dehaene
Right
Denmark
Rich
SD
Rasmussen
Left
Finland
Poor
SDP
Lipponen (from 13 April 1995)
Left
France
Rich
President: RPR; PM: RPR to 3 June 1997, then PS
President: Chirac (from 17 May 1995); PM: Juppé (from 18 May 1995 to 3 June 1997), Jospin (from 3 June 1997)
President: Right Prime Minister: Right (to 3 June 1997); Cohabitation
Germany
Rich
CDU
Kohl
Right
Greece
Poor
PASOK
Papandreou; Simitis (from January 1996)
Left
Ireland
Poor
Fine Gael
Burton (to 26 June 1997)
Right
Italy
Rich
‘Non-Party’
Dini (to 18 May 1996); Prodi
Centre-left
Luxembourg
Rich
CSV/PCS
Juncker
Right
Netherlands
Rich
PvdA
Kok
Left
Portugal
Poor
PSD
Silva (to 28 October 1995);
Right
PS
Guterres
Left
Spain
Poor
PSOE; PP (from 3 May 1996)
González (to 3 May 1996); Maria Lopez
Left Right
Sweden
Rich
SAP
Carlsson to 21 March 1996; Persson
Left
UK
Poor
Conservative to 1 May 1997; then Labour
Major (to 1 May 1997); Blair
Right to 1 May 1997; then left
* Determined on the basis of GDP per capita at PPP (purchasing power parity) in 1996; rich ⫽ above average; poor ⫽ below average (Eurostat 2001, p. 24). † Right ⫽ Conservative, Gaullist, Christian Democrat; left ⫽ Socialist, Labour.
Governance by the Open Method of Coordination 111
to join EU-wide union movements in order to counter the increasing bargaining strength of multinational firms. In July 1995 the Swedish government issued a note on its negotiating objectives for the 1996 IGC that contained the idea of a separate chapter for employment and the concept of an ‘employment union’ for Europe (Sweden 1996b).11 Common principles for employment policy would include worker training and the promotion of active labour market policies, improving employment in underdeveloped regions, equal opportunities for men and women, and strengthening the Social Dialogue (Sweden 1996a, p. 3). The Social Affairs and ECOFIN Councils would cooperate in preparing draft guidelines for the formation of national employment policies. A new employment committee was to advise on the formation of guidelines and to monitor progress (Sweden 1996a; Miles 1997). The Swedish proposals also stressed the importance of national governments in directing cooperation on employment issues (Lindahl 1996). Swedish thinking on the inclusion of employment policy in the new treaty is best summarized by Allan Larsson: There is no way in which Europe can have one single employment policy. There is no room for harmonization. There must be many different policies, but they have to be based on the understanding that employment is a matter of common concern, that there should be common objectives and targets and a common strategy, while maintaining the responsibilities of the Member States for national employment policies. (Larsson 2001) At the same time as a new European employment policy was mooted in Sweden, the Corfu European Council of 1994 agreed to the creation of a Reflection Group to prepare the IGC. The Reflection Group was composed of 15 personal representatives of the foreign ministers, a member of the Commission, and 2 MEPs (Corbett 1998, p. 371). It met from June to December 1995. Its purpose was not pre-negotiation, but rather to provide an opportunity ‘to think freely and openly and identify issues that could be dealt with by the IGC’ (Corbett 1998, p. 372). In September 1995 Gunnar Lund,12 the Swedish representative to the Reflection Group, proposed a separate employment title. The proposal included an employment committee and a statement that employment is a matter of common concern (Miles 1997, pp. 280–1). ‘The original Swedish proposals were workmanlike, stressing general principles and procedural mechanisms whereby national governments would retain
112 European Welfare States
competency and leadership’ rather than ‘specific policy objectives’ to which opposition would have been more pointed (Miles 1998, p. 17). Notably, the policy leader had anticipated that competency would be a key issue, and proposed a method for cooperation that would retain competence at the national level. Sweden wanted cooperation in employment policy, but was well aware that the Member States would not agree to supranational governance by law. Elisabeth Guigou, representing the EP’s Party of European Socialists and a former member of the Mitterrand government, supported the proposal ( Johansson 1999, p. 94). Guigou’s adviser, Richard Corbett, supplied the UK Labour Party’s IGC working group with Reflection Group documents (Corbett 1998, pp. 375, 406 n. 1 to chapter 15). Elmar Brok, a Christian Democrat MEP with close ties to Chancellor Kohl, reportedly had ‘a green light’ for the employment chapter, but divisions in the German government prevented outright support ( Johansson 1999, pp. 95–6). In the end, the reflection group agreed that while the main responsibility for employment policy lay at the national level, ‘the Union also has a responsibility for setting the right conditions for job creation’ (Council 1996e, p. 23). A separate employment chapter, or title, thus became a part of a developing agenda for the IGC negotiations, but strong support for it was still limited to the policy leaders, Sweden and Denmark.13 Alongside IGC preparations, the European Council continued to meet according to its normal schedule. At Madrid in December 1995, the European Council stated the EU’s commitment to ‘the fight against unemployment and for equal opportunities’ and called for an EU job creation strategy (Council 1995). The 1996 IGC: small Member States for; large Member States against By the start of 1996, national positions on employment policy were becoming clear. Belgium, Denmark, Greece, Italy, Luxembourg, Netherlands, Portugal, Spain and Sweden stated that they supported expanding the EU’s capacity in employment policy with a new and specific reference in the Treaty (Belgium 1996; Denmark 1996; Greece 1996; Italy 1996; Portugal 1996; Spain 1996).14 Finland, the only Nordic socialdemocratic state committed to the third stage of EMU,15 made an explicit reference to the interdependence of a new employment policy and EMU and the importance of including a high level of employment as an objective of the Union (Finland 1996). Other states, such as Ireland, mentioned support for EU involvement in employment policy
Governance by the Open Method of Coordination 113
without detailing new provisions (Ireland 1996), perhaps anticipating their role as broker during their presidency in the second half of 1996. Austria also expressed general support for an employment policy (Austria 1996). The above group of states who supported a new treaty basis for employment is notable for its eclectic mix of poor countries with governments of the left (Greece, Portugal – as of October 1995) and right (Spain – as of 3 May 1996), and rich countries of the left (Austria, Denmark, Netherlands, Sweden) and right (Belgium, Luxembourg). Also notable was the absence of the Franco-German axis in pushing for an EU employment policy. The German government in its position papers leading up to the IGC had emphasized the ‘implementation of the subsidiarity principle, including transferring powers back’, a ‘clear separation of powers between the EU and the Member States’, and the involvement of the Länder in developing the German negotiating position (Germany 1996). The German Bundesrat ‘had agreed on a set of Länder demands for the IGC according to which the principle of subsidiarity as a rule for the division and execution of competencies has to be improved’, including a clear separation of competencies (Kompetenzabgrenzung) (Algieri and Janning 1996, p. 15). The Länder, notably Bavaria, wanted to protect job subsidies from EU oversight, and avoid the possibility of paying for employment policies in the rest of Europe (Hyde-Price and Jeffery 2001, pp. 709–10). It was also clear that Germany did not want to see an expanded budget; the inclusion of employment incentive provisions would be difficult for Germany to agree to. The ‘budgetary implications of proposals’ were thus ‘a key factor in the negotiations’ (Dehousse 1999, p. 26). The cost of German unification had been high, and a further drain on resources to the EU level was unwelcome.16 This ‘fear of financing’ is an element in the German reluctance to agree to the Employment Title that historical institutionalism may also describe as an example of institutional learning – after repeated experiences of policy creation and implementation over time Germany was well aware that unexpected financial consequences could result from the creation of new policies at the EU level, such as the Cohesion Funds that had accompanied EMU. Allan Larsson has said that he felt that: there were two camps in the [German] government [during the negotiations], one consisting of people in the Ministry of Economic Affairs and in the Ministry of Foreign Affairs (both led by FDP-ministers), the
114 European Welfare States
other camp was led by Chancellor Kohl in close cooperation with Mr. Brok, one of the two EP-representatives in the IGC. (Interview: 32) This first camp had fears that ‘the inclusion of an Employment Title would lead to new demands on huge structural funds, and that Germany had to pay for it’ (Interview: 32). The issues of competence and financing are thus key factors in explaining the gap in the expected preferences of Germany and its position during bargaining, and explain the inclusion of specific limits on funding in the Employment Title. It seemed possible that France would support an employment chapter, despite the lack of explicit support for such an initiative in the official government paper on the IGC, given high French unemployment rates. However, no French leadership on the Employment Title was forthcoming (Cohen-Tanugi 1996, p. 28; Parliament 1996). The French had concerns that a new European Employment Strategy might take attention and resources away from the main priority of launching EMU by 1999 (Interview 8; Cohen-Tanugi 1996, p. 23). The French were also concentrating on a national election, held in June 1997. With limited amounts of government resources dedicated to the IGC, then, France prioritized EMU over employment. One of the difficulties of the conference, therefore, was ‘that the traditional Franco-German partnership of which so much had been expected never really materialised’ (Petite 2001, p. 3). One participant in the Amsterdam negotiations noted, among factors at play in the IGC, ‘the loss of inspiration of the Paris–Bonn axis’ as well as the lack of grand visions for the future of the EU (Papadopoulos 2002). The UK opposed any development of the EU’s powers in the field of employment and would not give up its opt-out of the Social Protocol (UK 1996). Treaty change required unanimity, and it had been clear from the outset of negotiations that the Conservative UK government had an ideological opposition to further European integration in the social field and would not accept the Employment Title. As a result of UK stonewalling, most of the negotiations ‘were conducted by the other 14 Member States, in the hope that the election would break the deadlock’ (Petite 2001, p. 3). ‘Everybody knew the IGC was heading for Tony Blair or a crisis’ (Dehousse 1999, p. 13). By the opening of the 1996 IGC, then, the idea of separate treaty provisions on employment policy looked promising, but had substantial hurdles in the disinterest of France and the potential opposition of Germany – concerned with subsidarity and the opinion of the Länder – and the outright opposition of the UK.
Governance by the Open Method of Coordination 115
The Italian government hosted the opening of the IGC and favoured a wide agenda, including matters ‘related to the goal of a Treaty which the public can understand, [and] which strengthens the Union’s democratic character’ (Parliament 1996, p. 95). The Turin European Council confirmed that employment policy should be examined ‘in order to fulfil the objective of a high level of employment while ensuring social protection’ and that the IGC ‘should examine how the Union could provide the basis for a better cooperation and coordination in order to strengthen national policies’ (Council 1996a). Commission President Jacques Santer presented a ‘European Confidence Pact for Employment, Growth and Competitiveness’ as his own initiative for a ‘common strategy’ to unite EU institutions, national governments and the social partners on these issues (Commission 1996). It never proved the uniting force that he had hoped, however, unlike the multi-annual action programme on employment. Work on a specific draft article continued under the Irish presidency in the second half of 1996. At an information meeting of representatives in Cork on 6 July 1996 the Danish government presented a draft working text on ‘Title VII: Employment Policy’ containing four articles: (1) instauration of employment policy; (2) coordination of employment policy monitoring and guidelines; (3) incentive measures (excluding harmonization); (4) a consultative committee on employment (IGC 1996a). This draft text set out a formal structure for reporting on employment policy on the basis of common guidelines, annual national reports, a joint Council–Commission report on employment and recommendations to Member States issued by the Council on the basis of a qualified majority vote (QMV) on Commission recommendations. Although not in the draft text, the introduction to the report noted that the convergence criteria for EMU and the Broad Economic Policy Guidelines would continue to be respected, that the new title would not require an expansion of the Community budget, and that employment remained a national responsibility and did not change the division of competencies (IGC 1996a). Later that year the Swedes further refined ‘Title VII: Employment’ with a draft ‘art. 109n’ that stated: The Member States and the Community shall conduct their policies concerning employment with a view to contributing to the achievement of the objectives of the Community as defined in art. 2, and in the context of the broad guidelines referred to in art. 102(2) [later art. 99(2)] and in compliance with the principles set out in art. 3a. (IGC 1996b)
116 European Welfare States
While eventually divided between different Articles in the final version, the language relating the new employment policy to the Broad Economic Policy Guidelines (BEPGs) and overarching Treaty objectives did not change substantially. The BEPGs were to come first in the Treaty, and employment policy was to be consistent with these objectives. Still, an explicit reference to the BEPGs ‘was highly contentious, given the extreme reluctance to risk reopening any aspect of that debate’ (Petite 2001, p. 11). The Danes, for their part, proposed QMV for decisions made under the Title (IGC 1996a).17 By mid-1996 the sticking points were clear: division of competencies, and incentive measures that might prompt new programmes or budgetary expansion. By the end of September 1996, the Irish presidency had consolidated draft texts into a draft ‘New [Chapter] [Title] on Employment’ (Council 1996d). The draft contained much of the Danish and Swedish proposals and aimed at ‘the upper end of realism’ of what might be agreed to by all Member States, save those most deeply opposed to a new initiative, namely the UK and Germany (McDonagh 1998, p. 85). Three ‘questions of a more political character’ were highlighted for comments from delegations: a) whether specific provisions on the coordination of national employment policies at Community level, along the general lines proposed, should be included in the TEC; b) if so, where in the Treaty such provisions should be located. Two alternatives have been considered: – a separate Title in the Treaty immediately following Title VI (economic and monetary policy); or – a specific chapter within a new Title VIII bringing together provisions on social policy, employment and vocational training; c) whether or not to make explicit in the Treaty the need for coherence between the coordination of employment policies and the coordination of economic policies under 103(2). (Council 1996d)18 The presence of square brackets in the contentious part of the text reassured reluctant Member States, giving them ‘a more tentative flavour’. As a result, none of the Member States ‘rejected the text out of hand’ (McDonagh 1998, p. 86). The Irish Draft Treaty of December 1996 included a new Title on Employment (Council 1996d). It contained a revised art. B (TEU) and art. 3 (EEC) that included the goal of sustaining a high level of employment. The Draft Treaty called for ‘a common strategy for employment’
Governance by the Open Method of Coordination 117
and recognized that the promotion of employment was ‘a matter of common concern’ (art. 2(1) Dublin II). While the wording changed, the text on employment that solidified during the Irish Presidency ‘remained at the heart of the evolving text and provided the basis for ultimate agreement’ (McDonagh 1998, p. 87). The Dublin Declaration on employment was also issued, as per the mandate given to the Council at Madrid in December 1995 to review a second joint report on employment. The Council Declaration described the reports of the Member States to the Council and Commission as ‘multiannual employment programmes’ (Council 1996c). Building on the success achieved at the Dublin summit, and the inclusion of an Employment Title in the draft Dublin Treaty, the European Commission issued a summary document that labelled the 1994–6 European Council conclusions, member state reports, and joint employment reports called ‘The European Employment Strategy’ (EES) (Commission 1997a). With or without agreement on including a formal Employment Title in the Treaty, the Commission was laying the groundwork for future cooperation. At the end of 1996 the inclusion of an employment title in the revised Treaty looked possible but was by no means assured. Several Member States did not want to commit financial resources to combat unemployment, including the UK, Germany, and the Netherlands (Petite 2001, p. 11). The Irish presidency had repeatedly stressed that any incentive measures included in the new title should not involve harmonization of national laws nor new spending programmes, and this viewpoint had been reiterated in a Commission preparatory document for Amsterdam on the Social Protocol (Commission 1997c). For those who wanted to see the Employment Title included in the revised TEU, the 1 May 1997 election of Tony Blair and a Labour government in the UK was welcome news. As a result of ‘remarkable preparation by the British Foreign Office and the Labour shadow ministers, it was possible to launch … [the] third phase [of negotiations] more or less immediately’ (Dehousse 1999, p. 14). Labour Party contacts had also been in constant touch with the Party of European Socialists (PES) and Allan Larsson in DG V during the IGC (Interview 7). By the end of May it became clear that the two Member States whose opposition could block the Employment Title – the UK and Germany – might agree to the new provisions subject to certain conditions. For the UK this meant that any new EU-level activity would be directed to reform of the labour market, and not to new regulation. At the end of May the UK suggested that the following language (in italics) be inserted into the first Article of the draft Title: ‘Member states and the
118 European Welfare States
Community shall work towards developing a coordinated strategy for employment and particularly for promoting a skilled and adaptable workforce and flexible labour markets which are responsive to economic change’ (UK 1997). This critical wording was adopted for the first Article of the Employment Title, with the only addition of ‘a skilled, trained, and adaptable workforce’ in the final version. Swedish proposals for a more detailed list of principles for employment guidelines were rejected (Sweden 1997). In the absence of more specific guidelines, it could be assumed that the five Essen guidelines of 1994 would be an important reference to future EU-level employment policy (Council 1994). The change in the UK’s preferences on social and employment policy is fully attributable to the ideological stance of the Labour Party favouring greater European cooperation in this issue area, and cannot be attributed to a national welfare state explanation that illustrated the low spending of the UK on employment policy and its lower than average unemployment rates. That regime change in the UK was essential to the ultimate success of the Employment Title, and moved the negotiations to a new level of possible agreement, is borne out by comments of participants in the 1995–7 process (Dehousse 1999; Petite 2001; Interview 32). Not all assessments of the new Labour government’s involvement were positive, however. The lack of endorsement for more specific guidelines, and the orientation of the first article of the Title to adaptability and flexibility, caused one observer to comment: ‘the Blair government … collaborated with the German government to dilute the provisions of the employment chapter’ (Dehousse 1999, pp. 36–7). Germany also weighed in with its suggestions for the Title at the end of May. These were that the ‘special responsibility of the social partners in employment’ be included in the second article of the Title and that they be consulted in the process of cooperation. Germany also suggested the insertion of wording stating that the ‘competences of Members States in employment … are not affected’ by cooperation, and that the EP, the ESC, and the Committee of the Regions should give their ‘opinions’ on employment policy (Germany 1997a). While Germany’s suggestions for the inclusion of the responsibility of the social partners, earlier iterated by Sweden, were not included at the time, those regarding the EP, the ESC and the Committee of the Regions were adopted.19 Still, at the end of May, France was reserving its position on a number of issues (UK 1997, see marginalia). Spain (now under a Conservative leadership) proposed changing the basis of Council adoption of both guidelines and recommendations from QMV to unanimity (Spain 1997), although it had explicitly supported the creation of an employment
Governance by the Open Method of Coordination 119
policy in the Spanish government’s bipartisan IGC White Paper of 1996 (Spain 1996). One participant of the negotiations has stated that QMV was ultimately accepted as a way of ‘weakening the general case for a “flexibility clause” as the threat of national veto could push the remaining Member States to use a “flexible” approach’ (Papadopoulos 2002). Although QMV was not ideal for some Member States, it was preferable to a multi-speed EU. Clearly resistant Member States could not control outcomes, even during Treaty bargaining. QMV in the Title is also of interest as it indicates that despite their reluctance to give up competence, Member States were still interested in creating procedures that would facilitate regional cooperation. During the 1996 IGC process the role of leader Member States in keeping employment on the agenda, and the support of other smaller Member States, was thus especially critical in the face of UK and German opposition and French lack of interest. Will employment fly? The 1997 Amsterdam Treaty negotiations The final negotiations were held on 15–17 June20 in Amsterdam. Having elected a new Socialist PM on 3 June, France was now more disposed to a new employment initiative at the EU level. Germany was thus the only Member State that was still opposed. The German position at the IGC may have been a bargaining chip – ‘[o]fficially it may have been against; unofficially it was not’ ( Johansson 1999, p. 96). Still, the German Ministry of Finance was opposed, as this chapter has traced, to several key aspects of the final Employment Title. In order to win German approval, a clear statement of the retention of competence by the Member States, a consultative voice for the Committee of the Regions,21 and no new budgetary provisions were inserted into the Title. However, incentive measures may be adopted by QMV (pursuant to art. 251). Other governments who are net contributors to the budget were also ‘afraid of new expenditures’ and supported German amendments to reduce the scope of possible financial incentives (Dehousse 1999, pp. 37–8). The use of incentive measures was clarified by two declarations added to the Treaty of Amsterdam including their maximum duration (5 years), the maximum financing, and their place under Heading 3 of the financial perspective that consists of 6.2 per cent of the European budgetary resource and is separate from the Structural Funds (Council 1997a). Poor-country support for the Employment Title was thus not tied to any financial provisions. ‘We were very realistic in thinking that we
120 European Welfare States
would not get additional funds’, one Greek treaty negotiator has stated. Further, ‘the Germans were adamant that this [the inclusion of an Employment Title] should not lead to new appropriations to the Community budget’ (Interview 8). Approval of the Employment Title was not linked to approval of additional funds, despite the fact that the Employment Title might signal a shift in emphasis from structural development to unemployment, and from underdeveloped Member States to areas of unemployment in the broader Union. Ultimately, the role of Chancellor Helmut Kohl in the final negotiations pushed Germany to agree to employment policy when faced with a reluctant Finance Ministry and opposition from the Länder. The success of the IGC process, and European integration more generally, was ‘a personal goal of the highest priority’ for Kohl (Algieri and Janning 1996, p. 20). The final Employment Title was inserted after the Title on EMU and not as part of the Social Title (the integrated Social Protocol from Maastricht), which appears later in the revised TEU (Title XI). The linkage between employment and economic policy is thus emphasized by the Treaty placement. The Title states that the promotion of employment is ‘a matter of common concern’ and commits the Member States to ‘coordinate their action in this respect within the Council’ on the basis of Commission proposed guidelines, the production of national reports, and a joint annual report by the Commission and Council. All decisions, including the issuing of recommendations to Member States, are by QMV. An Employment and Labour Market Committee was also created, replacing the redundant Employment Committee of the 1970s (EC 1999). The Treaty thus ‘only changed in a very limited way’ from the Dublin draft as a result of the Amsterdam negotiations (Dehousse 1999, p. 36). The most significant absence in the final document is the lack of any mention of the social partners, despite the original proposals from Sweden that suggested their inclusion, and despite German support. This is because the Employment Title followed the economic rather than social policy model, where social partners only have a consultative role. Second, as the Social Protocol was being integrated into the Treaties for the first time, it was seen as ‘too complicated’ to insert the social partners in employment policy at the same time (Interview 36). This detailed examination of the 1995–7 negotiating process provides several key conclusions about the form of governance created by the new Treaty provisions. It had been clear from the outset that in order to get agreement among EU Member States a new form of cooperation that
Governance by the Open Method of Coordination 121
did not lead to uniform policy goals would have to be created. The final design in the Title borrows several aspects of traditional Community policy-making, notably the proposal of guidelines and recommendations by the Commission with approval by the Council based on QMV. The weight of institutional tradition – whereby the Commission has the right of proposal – and the informal practice of employment reporting that had been ongoing since 1995 were of great influence in the final result. Policy learning over time also indicated to Member States that should unanimity be used for the proposal of guidelines, resistant Member States could hold the Title in stasis. However, competency is explicitly retained at the national level, and harmonized policies are not to result (art. 129). No legislation is to be produced, and there are no enforcement procedures as in the case of the traditional ‘Community Method’ of regulation. Rather, there are incentive measures. Learning effects from past policy designs thus played a key role in determining the bargaining positions of Member States. The tracing of the Employment Title over time also presents a corrective to the intergovernmentalist assessment that emphasizes the period of final Treaty negotiations at the level of heads of state. Moravcsik and Nicolaïdis summarized the results in Employment policy as follows: Predictably, a Socialist government in wealthy France sought special provisions on employment; a Christian Democrat government in wealthy Germany promoted cautious progress; a Tory government in poor Britain opposed all concessions. Even the advent of a Labour government, while predictably reversing British opposition to membership of the social charter, did not lead to major substantive shift in preferences on labour market policies. (Moravcsik and Nicolaïdis 1999, p. 63) Process-tracing reveals the following: (1) France did not seek special provisions on employment, but rather joined other member states rather late in the IGC in supporting a well-developed draft employment title; (2) the German government did not oppose employment policy on the basis of being Christian Democratic, as their contribution at Essen in 1994 illustrated. The intergovernmental emphasis on Germany as a rich state provides a partial, but not full, explanation of their objections to the Title. National welfare state factors, such as the organization of employment policy in federal Germany and the concerns of the Länder, as well as institutional factors, such as the ‘fear of financing’ poorer Member States, all combined to create significant hurdles; (3) poor countries, most actively Greece, supported employment policy, as did
122 European Welfare States
Portugal, Finland, and even Spain and Ireland under conservative governments. On the other hand, a welfare state explanation of preferences, which emphasizes national policy history and unemployment rates in determining Member State positions in negotiations, provides good predictive results. Sweden was the most enthusiastic supporter, followed by Denmark. Swedish and Danish leadership is predicted by national spending on employment policies. Poor Member States with high unemployment rates also proved agreeable to cooperation, as did rich Member States with high levels of spending on employment policies (such as Belgium and the Netherlands). Other rich Member States with low unemployment rates and low spending (Austria, Luxembourg), which would be expected to resist cooperation on the basis of a welfare state explanation, evidently thought the policy held little cost of agreement for them, especially after explicit provisions on financing were attached. Of those Member States that were the greatest resisters, the UK opposed the Employment Title under the Conservative government of Major, but approved it under the Labour government of Blair. The main problem for a welfare state explanation of preferences is thus not why the Conservatives opposed the Title, given lower than average unemployment rates and spending on employment policies in the UK, but why the Labour Party agreed. Ideology has more salience as an explanation than national institutional history in the UK case. France, despite a high unemployment rate and almost average spending on employment policy, did not approve the Employment Title until the final stages of the negotiations. France wanted to concentrate on EMU under a conservative government (Chirac–Juppé). The cohabitation government was key to increased support for the Employment Title. National institutional history has less salience than political factors in the case of France, as well. Germany was opposed to the Employment Title, and this also cannot be accounted for solely by a welfare state explanation. German Länder opposed because of fears of losing competency in employment policy, but Germany had a high unemployment rate and relatively high spending on employment policy. Especially under the conditions of EMU, greater support from Germany should have been expected. The additional factors of the government’s ‘fear of financing’ (that is an expanding budget because of new competencies) and emphasis on subsidiarity must be included to explain the outcome. Greece, as well as Portugal and Spain, approved the Employment Title, although it was not linked to any significant new funding. However,
Governance by the Open Method of Coordination 123
interviews with Greek officials made it clear that they knew no new funding would be forthcoming from the EU, especially in the light of German opposition to new lines of financing. A Greek IGC delegate stated that Greece supported the Employment Title because of high unemployment rates and because of the need to provide a policy to balance the economic focus of EMU (Interview 8). Further research is needed on the motives of the poorer Member States, especially poor and conservative-led Spain which defies explanation by intergovernmental theory, and low-spending, low-unemployment Portugal which the historical institutionalist analysis cannot account for. In the case of Portugal, government ideology and the leadership of Prime Minister Guterres, who had been involved in Party of European Socialists (PES) planning on employment policy since 1994, are suggested as possible explanatory factors. Finally, neither intergovernmentalism nor a welfare state explanation adequately accounts for the role of the Commission. The Commission played a vital role in preparing employment policy before Amsterdam via the employment Framework and White Paper process launched in 1993. The Commission first put employment on the agenda, and provided substantive guidelines for the debate. The process of information sharing got the idea of employment coordination circulating among national officials as a kind of ‘icebreaker’ or familiarization process, and emphasized the ties between EMU and employment. Policy entrepreneurs such as Padraig Flynn and Allan Larsson built coalitions of support for an EU employment policy. Flynn was able to use his role as Commissioner to visit national capitals and gather information about national positions, while Larsson was tied in to transnational policy networks of Nordic countries, socialist Member States and trade unions ( Johansson 1999, p. 97). A historical institutionalist analysis, highlighting the role of Commission entrepreneurship and learning effects, is a necessary supplement to an explanation of preference formation and explains the development of the institutional features of employment policy.
The European Employment Strategy: the anticipated and unanticipated consequences of the European Employment Title From Amsterdam to Luxembourg: from a Title to a Process One assessment of the Employment Title suggested: ‘there was no chance of coordinating financial transfers or fundamental labour market
124 European Welfare States
reforms, meaning that cooperation could, in the end, be little more than symbolic’ (Moravcsik and Nicolaïdis 1999, p. 63). This section of the chapter outlines how the Employment Title has resulted in more than symbolic cooperation by institutionalizing the informal process of employment policy reporting that had gone on before the Amsterdam Treaty into a new form of regional governance. Careful process-tracing shows that as a direct but unanticipated result of the Employment Title, the OMC was developed as an alternative to the traditional Community method of decision-making, or governance by law, and to the Social Dialogue. The OMC as a new mode of governance has had an impact on national employment policies that will be assessed in the following section. The Employment Title was quickly transferred into a detailed working process by a combination of Commission activism and the support of the Luxembourg presidency in the second half of 1997. The ‘very close synergy of Member State [from Luxembourg] and Commission activity’ came from two sources: first, the DG V team under Allan Larsson developed a draft Joint Employment report and proposals for employment policy guidelines by mid-September, only three months after the Amsterdam Council; and second, Jean-Claude Juncker, the prime minister of Luxembourg, was also the minster of finance and the minister for labour and employment. This unique portfolio of positions provided him with the opportunity to combine the two main policy areas associated with employment, labour markets and economics, with his capacity as Prime Minister in order to put forward proposals as the President of the European Council (Interview 40). Luxembourg thus took the agenda-setting initiative of putting employment policy at the forefront of Council meetings in the second half of 1997, while the Commission took the leadership in providing content for the agenda. A good example of this was the Presidency request that the Commission present the first set of employment guidelines to the Luxembourg ‘Jobs Summit’ in November 1997 (Flynn 1997). The guidelines cover a wide range of themes including taxation, training, ALMPs, adaptability, lifelong learning and long-term employment. These guidelines can be traced back to the Commission White Paper of 1993 that was given substance at the Essen Council meeting of 1994. Policy practice, when it did emerge in the EES, ‘was found to be shaped by the prevailing institutional structure’ – in this case the Essen process and national reporting (Bulmer 1998, p. 380). In the Commission-developed 1998 guidelines for Member States’ employment policies, 4 themes were presented, namely entrepreneurship,
Governance by the Open Method of Coordination 125
employability, adaptability and equal opportunities, and under them 12 specific guidelines were suggested (Commission 1997b; Flynn 1997). Although we do not delve into the pillars and guidelines in great detail, a brief explanation of each of the main pillars is warranted. Entrepreneurship is the ‘capacity to create jobs’ by reducing non-wage labour costs and other forms of taxation on employment, and reducing the red tape involved in business start-ups. Employability concerns ‘enhancing people’s capacity to be employed’ through targeting youth and the long-term unemployed, and addressing the skills gap. Adaptability refers to both enterprises and individuals through tax incentives, especially for training. The equal-opportunities pillar targets female participation in the labour market (Flynn 1997; Commission 1997b, Commission 1998b). The conclusions of the Luxembourg Job Summit only slightly modified the Commission proposals on guidelines. All four themes, now referred to as ‘pillars’, were kept (Council 1997b). Only the guideline for the creation of a pan-European secondary capital market by the year 2000 was taken out, and a guideline concerning exploiting the opportunities for job creation at the local level was put in its stead. A guideline on promoting the integration of people with disabilities into working life was also added to under the equal-opportunities theme (Council 1997b). Each of these guidelines were to be addressed in follow-up National Action Plans (NAPs), written at the national level. What is remarkable about the period from June to November 1997 is how rapidly the institutional structure around employment policy coalesced. The speed of this process indicates several things: (1) that employment policy was already well articulated at the EU-level, with goals and major policy thrusts agreed upon well before the Luxembourg Summit, although at an informal level of cooperation; (2) that the Commission had shown leadership in creating the content and structure of the employment guidelines, and had delivered a coherent package in a short period of time; (3) that Member States were supportive of the process, and that they agreed on its content and focus as few changes to the Commission’s proposed guidelines were made. In other words, there was already a high degree of convergence of opinion on the main principles for an employment strategy at the EU level in light of informal practices that had come before the Amsterdam Treaty. A timetable of approving guidelines (December–January), presenting National Action Plans (May–June), presenting implementation reports (Summer), and the creation of a summary Joint Employment Report (Autumn) followed by the issuing of revised guidelines was created at
126 European Welfare States
Luxembourg. By 1998 the term ‘European Employment Strategy’ had been created to describe the process as a whole (Commission 1998b). What was missing from the EES in its first year, however, was the issuing of recommendations provided for in art. 128 (4) of the revised TEU, although comments on NAPs are scattered throughout the 1998 Joint Employment Report (Commission 1998b). Recommendations, proposed by the Commission and approved by QMV in the Council, have been ‘the most contentious aspect of the EES’ (Interview 40). Recommendations were first issued by the Commission at the end of 1999 for approval in Council. Member States ‘were caught off guard’ as ‘they thought the recommendations might be like the economic and finance recommendations of the Stability and Growth Pact that have only been used once, in the case of Ireland’ (Interview 40). Recommendations have proved to be the most political part of the process as they are bargained over and adjusted in Council meetings, although, owing to earlier consultation, the 2000 recommendations proved far less contentious (Goetschy 2001, p. 411). Perhaps the greatest sign of political commitment to the EES is that recommendations have gone forward at all. However, without any punitive damages attached, the issuing of recommendations has less tangible effects for Member States than sanctions and fines under the Growth and Stability Pact or under governance by law. The EES: from a strategy to the Open Method of Coordination (OMC) Successive European Councils have all stamped their unique mark on the EES. At Cardiff in 1998, the emphasis was on structural reform of the larger economy, and a reform strategy for making the Single Market a motor for new jobs was presented (Council 1998, paras. 17–26). At Cologne in 1999 the emphasis was on macroeconomic coordination and the idea of macroeconomic dialogue involving the social partners.22 The combination of a highly dynamic Presidency at the start of 2000 and an activist European Commission was to have a deep impact on European governance. At the Lisbon Summit (March 2000), the method of national coordination and inter-institutional cooperation developed by the EES was ‘baptised’ the Open Method of Coordination (OMC) (Larsson 2002). The OMC takes the specific features of the EES – Commission-proposed and Council-approved guidelines, national reporting, a Joint Commission–Council report, and Commission-proposed recommendations – and identifies them as a mode of governance that may be used in other areas. The Lisbon summit conclusions of
Governance by the Open Method of Coordination 127
23–24 March 2000 endorsed the OMC as the key method to guide various policies on employment, social exclusion, and including such issues as poverty, long-term unemployment, social protection and pensions (Council 2000b). The OMC thus elevates the governance regime developed by the EES to a general method of cooperation that may be adopted in other areas, but does not add anything new to it. However, the EES is only one example of how the new OMC may work. The OMC as applied to other areas may not replicate all aspects of the EES, such as recommendations. The Treaty of Nice, negotiated under the French Presidency in late 2000, incorporated the OMC into art. 137 for the areas of social exclusion and the modernization of social protection systems (EU 2000a, art. 137 (1) (j, k) and art. 137 (2) (a)). Revised art. 137 is almost an exact replica of art. 129 of the Employment Title of 1997, adding only the phrase ‘improving knowledge’ to the description of the process of information exchange and the identification of best practice. Thus, soft law, involving information sharing, the setting of common objectives and guidelines, the diffusion of best practice, and the naming of good and bad policy performers, has been elevated to a formal regulatory system in social policy by the inclusion of the OMC in the Treaty of Nice (Bruun 2001). Under the heading ‘modernizing the European Social Model by investing in people and creating an active welfare state’, the Lisbon European Council also endorsed a target for the overall employment rate of 70 per cent by 2010, and one for the female participation rate of 60 per cent by 2010 (Council 2000b). Annual spring summits, first held in Stockholm in 2001, are to review progress towards the diverse set of targets and goals set at Lisbon, and are to update priorities. The development of targets at Lisbon is of interest in view of German opposition to them (Germany 1997b, p. 7; Germany 2000, p. 5). As early as 1997 the target of a 70 per cent employment rate was raised by the Commission in its submission to the Luxembourg Summit (Commission 1997b). The Commission again suggested the 70 per cent target for EU employment rates by 2010 in its paper for the Lisbon Summit, along with other targets for female participation rates (60 per cent by 2010), the reduction of poverty, and investment in human resources (Commission 2000a). These targets were developed in a meeting of key DG V and Prodi cabinet officials two weeks before the Lisbon Summit. ‘Perhaps had there been more time for Member States to oppose they [the targets] would not exist!’, noted one Commission official who helped to develop them (Interview 36).
128 European Welfare States
The 2001 employment guidelines reflect the increasing complexity of the EES combined with the goals and targets set out in Lisbon to achieve a ‘knowledge-based society’ (Council 2001a). Five ‘horizontal’ objectives have been added, including the 2010 goals of 70 per cent employment and 60 per cent female participation, as well as improving lifelong learning, common statistics, and the focus of NAPs on all four pillars (Council 2001a). The role of the social partners has also been strengthened, giving them responsibility for the implementation of adaptability guidelines concerning work organization and lifelong learning, and allowing them to propose new guidelines (Council 2001b, para. 34). This corrects for the weak role for social partners in the Employment Title. Table 4.4 summarizes the activity of both leader and resistant Member States, and Commission entrepreneurship and other institutional factors, in the evolution of employment policy in the three stages of EU cooperation examined. The next section examines implementation at the national level. Consequences at the national level: the interaction of the EES with national employment policies While at the EU level the Employment Title has led to the creation of the EES, and to the unanticipated consequence of creating a new working method (OMC), what are the national consequences of the EES? Interviews with national officials in charge of writing the National Action Plans (NAPs), other employment policy officials, and social partners in Sweden, the UK and Greece reveal diverse consequences. For those Member States with a long tradition of ALMPs, the EES has had the surprising result of the issuing of recommendations in areas long thought to be above reproach. For Member States with almost no institutional history of ALMPs the EES has had a much deeper impact, providing them with a stable framework for the creation of new policies whose full impact cannot yet be assessed. For these latter Member States, EU-level policies have been absorbed and help to guide domestic policy. Sweden: are you looking at me? As Sweden has a long tradition of ALMPs and a low unemployment rate, it did not expect the EES to result in much adjustment of its policies. Compared with other Member States, Sweden’s employment policies have garnered much praise from the Commission, and one of the smallest number of recommendations. However, ‘the fact that we got any recommendations at all was of surprise to some people at first’ (Interview 55).
129 Table 4.4 Factors determining the evolution of employment policy at the EU level Time
Informal cooperation: Amsterdam 1993–6 negotiations on an Employment Title: 1996–7
European Employment Strategy: 1997–2001
Leader Member States
Danish Presidency of 1993 Belgian Presidency of 1993 Germany Presidency of 1994
Sweden: representative to IGC placed employment on the Treaty agenda. Denmark: proposed Treaty text.
Luxembourg Presidency of 1997: leadership of J-C Juncker in hosting the Luxembourg Jobs Summit and activating the Employment Title before Amsterdam was ratified.
Resistant Member States
Insufficient information to identify resistant Member States.
UK Germany France
Germany did not want explicit targets in employment policy.
Commission entrepreneurship
1980s: exchanges of employment ministry personnel organized by Commission officials, EURES, MISEP, LEDA. 1993: Flynn’s CommunityWide Framework 1994: Delors’s White Paper; development of Essen criteria in cooperation with German Presidency.
Allan Larsson in DG V played coordinating role among Commission and national officials. Coalition – building in support of the Employment Title.
Allan Larsson and Commission officials in DG V worked with the Luxembourg Presidency to design EES.
No formal rules on employment policy cooperation.
Guidelines proposed by the Commission and approved by QMV. No Formal sanctions.
EES suggested a new form of governance for other issue areas.
Subsidiarity: competence to be kept explicitly at national level. Fear of budgetary
Very flexible implementation of guidelines. No formal sanctions
Factor
Other institutional factors: Rules
Learning effects in policy design
continued
130 European Welfare States Table 4.4
continued
Time
Factor
Informal cooperation: Amsterdam 1993–6 negotiations on an Employment Title: 1996–7
Learning effects cont.
expansion: explicit limits on provisions and new budgetary lines. QMV proposed for issuing of recommendations.
Financial implications
Use of European Social Fund (ESF) to fund specific objectives.
No new budget lines. Incentive measure may be introduced by QMV.
Unanticipated consequences
A multiannual process of reporting on employment policy at the EU level, organized by the Commission.
Coalesced into a formal process (EES) very quickly.
European Employment Strategy: 1997–2001 for nonimplementation. ECJ is cut out.
Recommendations more intrusive; EES became the model for the OMC; possible resentment over time by poorly performing Member States?
One of the most surprising recommendations has been under the equal-opportunities pillar (Council 2000a; Council 2001b). With one of the highest rates of female participation in the EU (71 per cent),23 Sweden is at the forefront of including women in the labour market. However, as the Commission pointed out, women are highly segregated within the Swedish labour market in public services and at low and medium levels of authority within firms. The Commission had recommended that Sweden examine and reduce the ‘occupational and sectoral segregation’ of women (Council 2000a). Highlighting labour market information from a new perspective is one benefit of the EES. The EES may also have the effect of countering the decentralization of ALMPs that was desired by the Swedish Employers’ Confederation. Swedish employers initially expressed concerns ‘over the Employment chapter’s future role’ if it would mean implementing specific policies designed by the European Commission (Miles 1998, p. 22). Having had long experience with ALMPs, and having acquired a healthy amount of scepticism about their rewards relative to cost, employers wanted firmlevel bargains to be struck over investment in training, and linked to
Governance by the Open Method of Coordination 131
productivity gains. As the international spokesman for the Swedish employers noted: ‘We want EU policy to avoid our mistakes’, specifically in using ALMPs to shelter the unemployed (Interview 50). The EES has repoliticized a debate over who will set the terms for training and the finance of programmes, and refocused attention on the Swedish tradition of ALMPs at a time when employers wish to move away from government-designed nationwide policies (Interview 56). Overall, the EES has had little impact on the orientation of Swedish employment policy. However, the fact that the EES has highlighted areas formerly thought of as beyond reproach illustrates that EU policies require adjustment in some form from all Member States, irrespective of their level of development. Such criticisms also provide fodder to domestic opposition parties, notably the conservative Moderate Party, which argues for lower taxes on labour. Still, although not all recommendations will be acted on, they can serve to shake up a self-contented Sweden. In the long-term, as well, Swedish employers and unions are pleased with the development of the OMC as it permits EU-level discussions but retains ultimate policy competency at the national level. Swedish unions appreciate that OMC provides a way of permitting cooperation that would otherwise be blocked by unanimity voting (Interview 55). Indeed, the LO has conceded that ‘the age of binding laws belongs to the SEM’ (Interview 55). UK: no policy U-turn The UK had begun moving towards a greater emphasis on ALMPs in the 1990s. The Blair government reinforced this trend with the introduction of the New Deal policy in 1997. The New Deal provides training and subsidies for employers taking on workers for more than six months. It also guarantees work for all unemployed 18–24-year-olds unemployed for more than six months, and offers them a choice of training, subsidized private sector work, voluntary sector work or work with public agencies. The UK was thus poised to place a greater emphasis on ALMPs irrespective of the EES. In the words of one UK official, the links between the EES and national policies should not be overstated, since the EES is ‘simply a mechanism through which national Governments take forward a shared set of objectives’ (Interview 24). The NAP is prepared by an International Unit that crosses the Department of Work and Pensions and the Department of Education and Skills, with wide consultation across Whitehall (including the Department of Trade and Industry, the Women’s Unit and the Treasury)
132 European Welfare States
(Interview 24). Less traditionally, the CBI and the TUC have also prepared responses for the UK implementation report in the form of a joint statement. These reports are of ‘some significance, given the absence of institutional and political support for national-level social dialogue in the UK’ (Foden 1999). On a procedural level the EES has created an impetus for national-level social dialogue where none existed previously. In the absence of national institutions for bipartite and/or tripartite social dialogue, the specific guidelines of the EES require national-level dialogue on issues such as training. It is difficult to evaluate, however, whether there is much more than ‘cosmetic’ cooperation (Foden 1999, p. 532). It seems that traditional patterns persist, and that ‘employers (and the government) consider these matters are best dealt with at the level of the organization or the workplace’ rather than at the national level (Foden 1999, p. 537). The UK performs very well on the two basic indicators suggested at Lisbon, with a high employment rate (71.2 per cent) and a high female participation rate (64.6 per cent), both predating the EES. The Commission’s recommendations therefore focused on ‘structural challenges’, namely the large percentage of women in part-time work due to the lack of childcare provision by the state and occupational gender segregation, a significant number of long-term unemployed persons hidden inside the unemployment rate, a low skill level, and the need for greater encouragement of the ‘partnership approach’ (that is social dialogue) on work modernization (Council 2000a; Council 2001b). More specifically, an expansion of the Jobseeker’s Allowance was recommended (Council 2001b), and a National Childcare Strategy has been developed (Commission 2002, p. 26). The UK ‘views the process of recommendations very positively’ (Interview 24), perhaps because it is a Member State that performs very well on structural labour market indicators. One area where the UK may benefit from the EES over time is in skill development, for this is one area ‘where we, comparatively, do not do very well’ (Interview 52). UK employers may, in the long run, be influenced to change their traditional attitude, which views training as a cost rather than as an investment (Byre 1993, p. 25). The only danger of the EES is that the process is now so time-consuming and ‘over-burdened’ that ‘the weight of the process will squeeze out the policy thrust’ where changes are most needed (Interview 24). Greece: policy absorption and modernization The experience of Greece may be an example of what Goetschy has described as the catalytic nature of the EES, which boosts the efficiency
Governance by the Open Method of Coordination 133
of national employment policies ‘by establishing external constraints and targets’ and through ‘tough monitoring and a real evaluation culture’ (Goetschy 2001, p. 403). The Greek experience is more mixed, however, than first appears. Officials react negatively to being ranked lowly in the benchmarking aspect of the EES, and politicians have not taken an active interest in the NAPs. Still, now that the Greek government has signed on to the process, the EES locks-in employment policies across time, and focuses attention on groups in the labour market that were previously ignored. Greece was almost totally passive in the creation of the EES, although it supported both the Employment Title and the Luxembourg Process. This was because political resources and expertise in 1995–2000 were devoted almost exclusively to entering EMU. This explains to a large extent why Greece was willing to accept a policy design that included many measures that are not a traditional focus of Greek policy, such as the disabled and women. In addition, there was a feeling among Greek officials that they wanted to use their objections in ECOFIN to concentrate on monetary policy and not waste their time objecting to employment policies where, to paraphrase several interviewees, Greek officials would probably have been ignored anyway, given Greece’s poor employment record and limited experience in ALMPs (Interviews 14, 18). The 1997–2000 NAPs ‘lacked strategy’ and were not positively reviewed by the Commission. One reason for the low quality of Greek NAPs has already been suggested, namely the concentration on EMU. In addition, the Ministry of Labour and Social Security is very low in the hierarchy of ministries and has not attracted high-level government ministers in the past. The Ministry of Labour and Social Security has personnel that have been described as ‘extremely weak and poor’, in the words of one official.24 Both political parties exhibit little interest in the NAPs. Parliamentary review of the NAPs has produced a low quality of debate, and MPs argue not for or against the broad strategy of the NAP, but rather for more money to narrow segments of the population that are usually constituents (Interview 18). The Commission has become ‘much more demanding in its accounting and we [Greek officials] have becoming more careful in the administration of funds for training’ (Interview 16). However, national monitoring of the NAP is its weakest element in the Greek case. Its programmes tend to fall into various Community Framework programmes and are monitored by administrators of these individual programmes and not as a whole. Further, poor collection of statistics by the Ministry of Labour
134 European Welfare States
and Social Security has compounded the difficulties of developing and monitoring policies. Data provision is improving, however, which will help adapt future NAPs. The appointment of a reform-minded general secretary, Evi Christopolou, to the Ministry of Labour and Social Security in 2001 has led to the cleaning up of OEAD (the Greek Manpower and Employment Organization) policies, notably the training programmes. OAED was restructured over the period 2000–2, a reform that has attracted crossparty protests as it is staffed by appointees of the political parties in numbers equivalent to their representation in Parliament. Unless this reform is successful, however, ‘there will always be a black hole in the implementation of Greek NAPs’ (Interview 18). The challenge is to transform OAED from an informal benefit agency to a public employment office, a service which it has performed in the past but very inefficiently (Interview 73). The main benefit of the NAP process in the Greek case is that it creates continuity and accountability in the provision of policies to combat unemployment, countering the short-termism that typified government thinking in this area (Interview 18).25 The NAP clarifies priority areas, helps with planning efforts, and to some extent depoliticizes debate,26 locking in goals over time rather than leaving them open to change on a clientelistic basis. The EES has also had an impact on the attitudes of civil servants who are coming to believe that the goals of the NAPs, and ALMPs more specifically, can and should be implemented efficiently (Interview 18). Membership in the EU, bringing both access to the ESF and participation in the EES, has been ‘the cornerstone of implementing ALMPs in Greece’ (Interview 73). The EES has also had the effect of creating policies in areas where the Greek government was previously inactive. Greek NAPs clearly focus on pillar I – employability – and few resources are allocated to pillar IV – equal opportunities for men and women. Despite the uneven spread of resources across the pillars, without EU initiatives the unilateral development of policies for women, minorities and the disabled would have been unlikely. Because of the additionality principle (where the ESF contributes to policies once national budgets are approved), national spending is generated in areas that would not normally receive national funding in order to qualify for EU funds. Further, as the pillar on equal opportunities is part of the EES, the Greek government ‘must be seen to be doing something’ in these areas and so is forced to develop policies, however weak (Interview 18). Even low-budget policies have
Governance by the Open Method of Coordination 135
the effect of building constituencies that are now mobilized and encouraged to lobby the government for additional funds. There is also an unfulfilled side to the EES process, however. Despite the projected benefits of information sharing and the identification of best practice, ‘policy emulation has been the neglected part of the process’ in Greece (Interview 18). Greece does not have ‘even one policy for emulation’, and this means that the process of exchange of best practice effectively excludes Greek officials who must listen to recommendations, but do not contribute their own examples to the process. As a result, Greek officials view the sharing of best practice ‘rather negatively’. There is a feeling of frustration at being at the bottom of measurement tables, and a feeling that benchmarking may be ‘unfair’ and ‘biased against Greece’s social system’ (Interview 16). EU benchmarking uses the European Household Panel data as the basis for indicators which do not capture many of the more informal aspects of the southern welfare state model, such as unofficial employment, family care of the elderly, and home ownership. Notably, Greek negotiators at Amsterdam expected that the ‘related surveillance mechanism would not be too intrusive’ (Papadopoulos 2002). However, Commission recommendations for Greece delve much deeper than general employment issues, and concern equality issues, and youth and long-term unemployment; the Commission has also recommended regulatory and fiscal change, an examination of the interrelationship of the tax and benefit system in discouraging labour market participation, and a greater role for the social partners. Statistical services should be improved. Greece was also criticized for lacking an overall strategy for the implementation of the guidelines across all four pillars (Council 2000a; Council 2001b). It is not surprising that constantly sitting at the bottom of various ranking systems creates a negative viewpoint among Greek officials. As OMC evolves to other policy areas it is important to note the uneven participation of Member States; if OMC consistently cites the need for policy adjustment in the less developed Member States, a backlash to its extension may be the result. Certainly the intrusiveness of the surveillance mechanism should be cited as one of the unanticipated consequences of the Employment Title as far as Greek negotiators are concerned. In addition, the possible negative side-effects of OMC may prove to be an unanticipated consequence of the process more generally. There has also been ‘resistance to outcome measurement by politicians of all political parties’, none of whom are eager to be benchmarked
136 European Welfare States
across time ‘on objective criteria.’ In the words of one official, ‘no one [that is, politicians] believes the EES can have an impact. They just see it as a means of distributing funds … if the money came from another policy they would take it as well’ (Interview 18). As the EES guidelines are very broad, there is room to distribute funds to many social groups that may also be clients of the government. However, Greek governments have talked for so long about the need for common social policies and cohesion spending that the Greek government ‘finds itself in a trap’ and ‘cannot but agree to concrete steps’ to measure the application and effect of policies (Interview 16), again highlighting the lock-in effects of the EES. Finally, there is some sense of being overburdened by EMU and the EES. As employment policies are traditionally absent from the Greek policy-making spectrum, and as intellectual and political resources have been channelled into the EMU project, it cannot be expected that the EES will help to generate new thinking over a short period of three years (1997–2000). There has been improvement, however, as the 2001 NAP garnered a much more positive response from the Commission. For Greece there are thus positive results – learning effects, policy lock-in to counter clientelism, and access to EU funding to develop ALMPs – and negative side-effects – disappointment leading to disinterest at always being ranked low in comparative tables, and a lack of best practice to share – from the EES. Only an assessment over time can demonstrate which results will dominate, but it is clear that even in the short term the EES has led to a reorientation of Greek employment policy. France: national policy domination The 1998 NAP experience was criticized by the social partners who felt they had not been adequately consulted (Casey and Gold 2000, p. 113). Indeed, the resulting NAP was drawn up ‘in interministerial meetings’ (Casey and Gold 2000, p. 113). The most radical restructuring of the French labour market during the period 1997–2001 was the result of a passive measure – the introduction of the 35-hour workweek in the Loi Aubry of June 1998. The law is notable for two reasons: first, the concept of the 35-hour workweek was developed at the national level in advance of the introduction of the EES, although its implementation as a national law coincided with the start of the Luxembourg Process;27 second, the law was proposed without prior consultation of the social partners and against the wishes of the national employers’ federation (CNPF) (Lafoucrière 2000, p. 10).
Governance by the Open Method of Coordination 137
A basic EU-level principle – social partner consultation – is still weak in France, despite EES recommendations that call for strengthening the partnership method (social dialogue) (Council 2000a). The national pattern of industrial relations, characterized by weak unions and central government intervention, continues as the dominant pattern. However, the French government assessment of the impact of the EES is more upbeat. The French five-year report noted that the EES enriched the national social dialogue, providing a venue for dialogue on new issues such as adaptability (France 2002). The EES has also fostered greater coherence among the national ministries that administer employment policy, and has focused attention on the preventative approach to unemployment, with the creation in France of the ‘New Start’ and ‘New Services Job for Youth’ programmes of 1998 (Commission 2002, p. 25). Germany: keeping labour market reform an active issue After several rounds of talks with the social partners, ‘covering general economic policy, social and labour market policy and vocational training’ in 1997–8, the German government produced its first NAP in April 1998 (Casey and Gold 2000, p. 113). No joint statement was made, however, between the government and the social partners. The DGB called the 1998 NAP ‘all in all disappointing and in some points a “provocation” (such as the government’s presentation of cuts in sickness benefit or the weakening of dismissal legislation as employment promotion)’ (Casey and Gold 2000, p. 113). Employers also expressed ‘scepticism’ that governments could create jobs other than via cutting taxes and labour costs (Casey and Gold 2000, p. 113). Germany received a high number of recommendations (10) in the period 1999–2000, many of them quite detailed. They include reducing passive employment policies that favour early retirement of older workers, reducing taxes and social security contributions on labour, introducing a labour force survey in line with a Council Regulation (577/98), improving incentives for lifelong learning, and better monitoring of the long-term unemployed in order to provide early retraining (Council 2000a; Council 2001b). The recommendations indicate that the German apprenticeship scheme is not sufficient to produce skilled workers, especially among older members of the workforce. The EES thus has the effect, as in Sweden, of opening up debates in policy areas that were traditionally thought to be sufficient. The major German employment policy initiative Alliance for Jobs was planned before the EES. However, while its content may have already
138 European Welfare States
been largely in place, the innovations in the design of the policy, such as a group of experts working specifically on benchmarking and individual working groups on taxation, have been influenced by the method and guidelines of the EES. The EES also serves to reinforce a highly contentious dialogue between German unions, employers and the government on issues such as taxation, training and lifelong learning that several of the social partners would have liked to avoid.28 Experience of the EES: diversity inside of convergence As the consideration of national-level impacts illustrates, Member States have been varyingly affected by the EES, depending on their own national policy history. This was noted in the five-year review of the EES by the Commission, which stated that ‘while a few Member States were already implementing policies largely in line with [the EES guidelines] … clear convergence can be noted for other Member States – albeit at different paces’ (Commission 2002, p. 10). In particular, instances of convergence towards ALMPs, gender equality policies, and social inclusion were recognized across Members. One of the most significant changes may be the structural shift in perception in the EU from ‘managing unemployment, towards managing employment growth’, with the attendant shift of focus on labour market participation, training, lifelong learning, and the reduction of taxes on labour (Commission 2002). The synchronization of the employment guidelines with the Broad Economic Policy Guidelines (BEPGs) beginning in 2003 was also very encouraging in structural terms. EMU, and the BEPGs that underpin monetary union and fiscal convergence, provide a macroeconomic framework within which the EES guidelines can be more effectively translated into specific labour market outputs, such as wage restraint, worker training and an increase in availability of flexible forms of employment.
Conclusion This chapter has summarized the diverse influences and events that created the Employment Title, and later the European Employment Strategy and the Open Method of Coordination. A close examination of cooperation before 1997 reveals that Commission entrepreneurship, alongside Council Presidency interest in the subject of employment by Denmark in 1993, helped to create an informal process of employment policy coordination at the EU level. Leadership from Sweden and
Governance by the Open Method of Coordination 139
Denmark during Treaty bargaining then suggested creating a formal process for European cooperation. The support of successive Council Presidencies, and a bargaining process that left many decisions to the final negotiation among heads of government, explains how the initial opposition of Germany and the UK to the Employment Title, and French lack of interest in it, were overcome. This illustrates that an analysis of the preferences of large Member States is not sufficient in order to understand the emergence of new forms of supranational governance. The appeal of employment policy was not solely rooted in national policies, but in the case of Luxembourg, also in the belief of the benefits of EU cooperation. Other Member States may move in and out of a leadership role depending on the circumstances, notably Germany. Germany pushed cooperation during its 1994 Presidency but exhibited great reluctance when formal cooperation was suggested. Only explicit limits on the competencies of the European Commission, a highly flexible reporting method, explicit limits on budgetary implications and the absence of formal sanctions convinced Germany and the UK to agree to employment cooperation. Finally, although described as passive in policy design, Member States such as Greece provided valuable support to the policy during the Treaty reform stage. The Commission laid the groundwork for an ongoing process of cooperation in the employment field before the Treaty negotiations. Thus, while employment policy was elevated to a formal process at Amsterdam, it is clear that EU-level action would have persisted even had the Employment Title not been agreed to. Less expected was the creation of a new system of EU policy-making, the OMC, which follows the EES model of reports and National Action Plans, but may or may not include Commission recommendations. When other areas of coordination are more developed (social exclusion, pensions), a comparison with the EES will allow further testing of the conclusions presented here. The first of these is that all Member States have to adjust to some extent as a result of governance by coordination, even leader Member States. The second is that Member States with weak institutional national histories are more affected by OMC than might be expected by its critics. The third is that benchmarking may result in a backlash if the same Member States are constantly criticized, although, given the passivity of Member States with lower levels of national policy development, perhaps quiet resentment on the part of the consistently criticized is a more realistic expectation. It will be interesting to follow whether there will be a shift in
140 European Welfare States
focus of spending in the ESF and other structural funds from regional development to overall indicators such as unemployment when the budget comes up for renegotiation in 2006, after eastern enlargement of the EU. The issue of a shift in focus of EU spending from regional development to overall indicators is therefore flagged here as a possible source of future unanticipated consequences of the Employment Title and the EES. According to Eurostat data, there has been an improvement in both unemployment rates (down from 8.1 per cent in 2000 to 7.6 per cent in 2001) and employment rates (up from 63.2 per cent in 2000 to 63.9 per cent in 2001) (Eurostat 2001). However, it is very difficult to judge the effectiveness of the EES via employment statistics, as it is difficult to isolate the effect of the EES among a host of variables. In addition, the effects of employment policies on employment rates are very difficult to measure in national contexts. A recent study concluded that ALMPs have had a positive but not significant impact on changes in flows from long-term unemployment to employment (de Koning et al. 2001, pp. 13, 296). This study also noted the difficulty of gathering accurate information, and of making assessments across time, given the constantly changing nature of government policy in the employment policy field. Perhaps the EES will help future research by ‘locking in’ guidelines from year to year. The most important concrete observations we can make to date on the effects of regional governance of employment policy concern the impact of the EES on national policy debates, and the role of national social partners. For policy leaders there is little change in national policies as a result of EES. However, the EES has served to open up policy debates that leader Member States may have thought were above reproach. For policy resisters the EES has a similar effect in opening up issues for debate. In the UK, the EES has also strengthened the national tripartite social dialogue where such structures were weak. The EES thus further embeds national social partners in policy-making, even where union and employer groups may have little experience of such participation. Passive Member States with weak policy traditions are the most affected by the EES, which requires them to activate policies in areas where policy history may be non-existent. The EES thus means different things to different Member States, each with their own national histories and specific problems. The EES, and the OMC more generally, permit diverse Member States to interact on an issue where harmonization and hard law are simply not possible. Given the extension of OMC to new areas in the Nice Treaty, it is
Governance by the Open Method of Coordination 141
possible to speculate that the experience of the EES may have shifted Member State positions with regard to regional cooperation in a number of sub-issue areas of social policy (EU 2000a). Interestingly, it was a method of discussion, information sharing, and coordination in order to prevent market distortion that Balassa recommended in the case of social policy in his seminal work on economic integration (Balassa 1961, p. 229). The parallels with the OMC of today are striking.
5 Global Governance of Social Policy: EU Member States at the International Labour Organization
Apart from this shadowy promise of peace, there is in any case no assurance … that such unions would advance social wellbeing. The advocates of continental unions have often argued that a universal system skips a stage in the political evolution of the world. But in our time the world has grown into a loose but living social system. If continental unions would tie up certain strands they would tear up many more, and more promising [strands of cooperation]. David Mitrany, ‘A War-time Submission (1941): Territorial, Ideological, or Functional International Organisation’, in The Functional Theory of Politics (1975), p. 109 David Mitrany, born at the end of the nineteenth century and writing about international cooperation in the middle of the twentieth, had very clear doubts about the positive effects of a European union on global cooperation. His thesis was that such regional unions would pull the energy of leaders and the capacity of international law away from common solutions towards particular constellations of interests, and prevent more fruitful forms of global cooperation from developing. The 60 years since his comments were written have seen the creation and development of the EU. Have Mitrany’s reservations come to pass? This chapter will demonstrate that although many of his doubts about the creation of a ‘continental union’ in Europe as a wholly positive force for international cooperation were well founded, the European Commission and EU Member States are very active in promoting cooperation in social policy at the global level. 142
Global Governance of Social Policy 143
An analysis of social policy-making at the global level is a necessary complement to the previous chapters on regional governance for several reasons. The first is to permit an analysis of the forms of global governance. How are policies proposed and enforced at the global level, and how does this process differ from regional forms of supranational governance? The first section of this chapter presents an overview of the institutions that have a remit in social policy rights and legislation at the global level. More detail is then provided on the creation, working method and policy outputs of the International Labour Organization (ILO). The second purpose of this chapter is to assess how EU Member States behave at the ILO. The EU is not a member of the ILO. Only states are members of the ILO, and all EU Member States, including the 10 new Member States of 2004, are also members of the ILO. Chapter 1 suggested that welfare state histories have a large bearing on the decision of states to cooperate at supranational levels of governance. The argument presented there was that states should act to extend their national systems under conditions of deep economic integration. States with high cost and benefit policies will lead proposals for the adoption of such policies by other states so as to reduce the possible competitive disadvantages of such policies. States with low costs are expected to resist higher standards. We expect, therefore, that states with social-democratic and conservative-corporatist welfare state regimes should be the most enthusiastic about cooperation at the global level. States with liberal and southern welfare states should be expected to be more reluctant about cooperation. Does the argument presented in Chapter 1 also hold true for the global level? In order to assess a national welfare state explanation of preferences for global cooperation, the number of ratifications of ILO Conventions by EU Member States are compared in the second section of this chapter. The third purpose of this chapter is to analyse the interaction between global and regional forms of governance in the social policy field. In what areas does the ILO have the competency to legislate, and what occurs when ILO and EU competences overlap? On a less descriptive and more analytical level, is global governance at the ILO a necessary accompaniment to regional social policy under conditions of globalization, and, if so, is it effective in plugging the gaps left by regional regulation so as to reduce competition among states on the basis of social policy? By considering such questions we assess whether European and global forms of governance are mutually supportive, or if they may work against the effective operation of one another.
144 European Welfare States
The third section of this chapter thus explores both the formal and informal relations between the EU and the ILO from a historical perspective. The formal relationship between the ILO and the EU, as represented by the Commission, is highly circumscribed given that the EU is not a member of the ILO. ECJ Decisions state that the Commission and the Member States are to cooperate and express a common view on areas where the EU shares competence with the Member States. Still, Member States continue to resist sharing formal competence at the ILO with the Commission. The diverse reasons for state resistance to formal Commission authority will be examined. However, in practice, in dayto-day operations of the ILO, the Member States and the Commission in fact coordinate their positions on a range of issues via several different mechanisms: the EU Presidency, regional groups, and the Industrial Market Economy Countries group (IMEC). Informal cooperation at the ILO is thus widely accepted by EU Member States. In the fourth section of this chapter, five specific issue areas of global governance are examined. By examining five areas in detail it is clear that the EU cannot be treated as a unitary actor at the global level. EU coordination varies considerably by issue area, and the level of coordination is related to both EU competence, and to Member State preferences. The EU acts most effectively as a group on political issues, has a mixed record on technical issues, and is weakest on budgetary matters. This chapter, taken as a whole, explores the interaction of regionaland global-level policy-making in much greater detail than is at present provided in the literature. The multilevel governance literature describes how states no longer control the process of policy-making, as subnational interests have mobilized themselves at other levels of governance to ‘pressure state executives into particular actions’ (Marks and Hooghe 1996, pp. 341, 256). The analysis does not consider global-level institutions, however, capping itself at the EU level. Leibfried and Pierson, two of the leading scholars on EU social policy, state: ‘[I]f Member States have lost considerable control over social policy in the EU this is primarily because of processes other than the efforts of Union officials to develop social policy for a new European universe’ (Leibfried and Pierson 2000, p. 275). The implication is that exogenous forces, namely competitive pressures brought to bear by globalization as a process of liberalization and market-opening, have eroded national capacity for policy-making. No attempt is made, however, to discern what activities may be enacted beyond the EU level to counteract this trend. Discussions about globalization and its effects on welfare states include factors such as trade competition and the export of low-skill jobs, tax
Global Governance of Social Policy 145
competition to reduce the non-wage labour costs for employers, and the mobility of investment capital to low-standard countries (Rhodes 2001). The focus here is not on globalization writ large, but rather on how governance at the global level may seek to address some of these challenges and how global approaches to these problems interact with regional ones. However, even when the distinction between globalization and global governance is clear in the literature (Deacon 2001), little detail is provided about the specific operating methods of the EU Member States or the interaction of EU and ILO legislation in specific policy areas.
The structure of social policy legislation and rights in the global system Global division of labour in social policy The global division of labour in social policy is complex. First, a distinction between international social policy legislation and social rights must be made. International social policy legislation entails the creation of obligations via nationally ratified Conventions by a globallevel institution. Such agreements often cover very specific areas of policy, such as employment policy for youth, working conditions for certain types of workers, or health and safety guidelines for a particular job. At the global level the ILO has the remit to create social policy legislation. International social rights, in contrast, fall under the jurisdiction of several global-level institutions and of European regional institutions. Social rights consist of the treatment to be expected by every individual regarding working conditions, entry and exit from the labour market, representation (if so desired) by a union, and wage-setting. At the global level such social rights are embodied in the UN Covenant of Fundamental Economic, Social and Cultural Rights (1976). At the regional level in Europe, social rights are embodied in the EU Charter of Fundamental Social Rights of Workers (1989), the EU Charter of Fundamental Rights (2000) and the European Social Charter (1965) overseen by the Council of Europe.1 Figure 5.1 sets out these institutions diagrammatically. Importantly, most economic and social rights contained in various international agreements are not self-executing in the sense that they need further legislative action at the national level in order to be effective (Betten 1993b, p. 385). Even if social rights have been accepted by a state via national ratification of an international charter or Convention, this does not automatically give an individual citizen of that state the right to invoke those rights in a national court unless they have been
European Court of Human Rights
European Convention on Human Rights (1950)
Organisation for Economic Co-operation and Development (OECD) Guidelines for Multinational Enterprises (2000)
International Court of Justice (ICJ)
Key:
formal link;
European Court of Justice (ECJ)
European Union (EU) Treaties Directives and Regulations Charter of Fundamental Social Rights of Workers (1989) Charter of Fundamental Rights (2000)
Dispute Settlement Body
World Trade Organization (WTO) GATT art. XX(e) on forced labour
link that may be invoked for information or for appeal.
Figure 5.1 Social rights in the global and European system
Conference on Security and Cooperation in Europe (CSCE) Helsinki Final Act (1975) Basket III: Human Rights
European Social Charter (1965) Supervised by the Committee of Ministers
Council of Europe
International Labour Organization (ILO) (1919) 1948 Declaration of Philadelphia, 1998 Declaration on Fundamental Rights, Conventions
Economic and Social Committee (ECOSOC)
Global Compact (2000)
United Nations (UN) Covenant of Fundamental Economic, Social and Cultural Rights (1976)
146
Global Governance of Social Policy 147
applied via national legislation or collective agreements (Betten 1993b, pp. 386, 392). Rather, in the absence of national implementing legislation, various international agreements permit only those who are an official party to the agreement – in most cases Member States, but in some instances in ILO legislation also national social partners – to invoke the rights contained therein via international investigations. The focus of this chapter is on the social policy legislation of the ILO, rather than the institutions that house and oversee international social rights. International Labour Organization During the First World War trade unions in western European countries, notably England, France and Germany, as well as in the United States, adopted resolutions calling for the creation of an international congress of worker and employer representatives to oversee the creation of common standards on working time, occupational health and safety and child labour. The American Federation of Labour (AFL) called for such an initiative to be part of ‘any peace congress to be held at the close of the war’ ( Joyce 1980, p. 27). The 1919 Peace Conference at Versailles appointed a commission to examine the creation of international labour legislation, and the final Treaty created the International Labour Organization (ILO) as a permanent section of the League of Nations (Treaty of Versailles 1919, pt XIII).2 The ILO’s main purpose is to raise and protect the standards of national labour legislation by building a code of international labour legislation. International labour legislation on working conditions aims to protect the individual worker, prevent competition among nations on the basis of working conditions and provide social justice as part of creating a ‘permanent peace’ among nations (ILO 1919, Preamble). The motivations that founded the ILO were thus humanitarian (improve working conditions for all workers), economic (provide a common standard of labour legislation so as to enable competition to exist on bases other than working conditions) and political (respond to union demands, provide a forum for regulated interaction with worker and employer groups and set common standards in industries that had been used for the production of materials of war). All of these motivations are referenced in the Preamble of the ILO Constitution and serve as its guiding objectives (ILO 1919). The ILO was the only element of the League of Nations to survive the Second World War, and in 1945 the ILO became a specialized agency of the new United Nations. The 1944 Declaration of Philadelphia
148 European Welfare States
reaffirmed the ILO’s basic principles, including that labour is not a commodity, that freedom of expression and association are essential to sustained progress, that workers, employers and governments enjoy equal status in the ILO, and that ‘poverty anywhere constitutes a danger to prosperity everywhere’ (ILO 1944). The ILO has three main bodies. The International Labour Conference is the body to which national delegations are sent once a year in order to vote on proposed legislation that may take the form of a Convention or a recommendation. Each national delegation consists of two government representatives, and one worker and one employer representative. The Governing Body is elected from the International Labour Conference and sets the agenda of the yearly Conference. The Governing Body is also tripartite, with 28 government representatives, 14 employer representatives and 14 worker representatives. The ILO Constitution provided that 10 Member States from states ‘of chief industrial importance’ are permanent members of the Governing Body (ILO 1919). These 10 states have traditionally included France, Germany, Italy and the UK. The other 18 government members of the Governing Body are elected. Employer and worker representatives to the Governing Body are elected by worker and employer delegates to the International Labour Conference and do not have to be from the same states as the government representatives on the Governing Body. The International Labour Office conducts research, aids in investigations, provides technical assistance to Member States, and creates proposals for Conventions or recommendations that are vetted by the Governing Body and put before the annual Conference. The International Labour Office is staffed by international civil servants from ILO Member States. All ILO members are states. Article 12 of the ILO Constitution permits institutions of international public law to participate in its deliberations without the right to vote. The Commission of the EEC, and subsequently the EU, was granted observer status at the ILO Conference and technical meetings where Conventions are prepared without a vote. This stands in contrast to the EU’s membership of the WTO, where it is not an observer (as is typical of its status at international organizations) but rather a full member. If necessary, the European Commission may vote in the WTO by casting the number of votes equal to the number of the Member States of the EU (Frid 1995, pp. 170, 179, 203).3 Because of the resistance of American labour representatives to binding legislation at the time of the ILO’s creation, the ILO’s legislative capacity is ‘a compromise between binding and advisory opinions’ (Betten 1993b, p. 9). The International Conference takes binding decisions only
Global Governance of Social Policy 149
to the extent that ratification by national legislatures is necessary to make a proposed ILO Convention binding. Any Conventions adopted by the International Conference of the ILO by a two-thirds majority are to be put before the appropriate national body, which will decide either to make the Convention binding on the Member State or not, through national ratification. Member States are obliged to submit all approved Conventions and recommendations before the appropriate national authority for possible ratification ‘in no case later than 18 months from the closing session of the Conference … for the enactment of legislation or other action [such as implementation via collective agreement]’ (ILO 1919, art. 19). Thus, rather than producing a corpus of commonly endorsed international labour law, the ILO serves as the forum for the development and review of legislation that is adhered to on a kind of pick-and-choose basis by its Member States. Core labour standards are an exception, in that all ILO members regardless of whether they have ratified the eight Conventions associated with core labour standards are charged with the ‘respect’ of these Conventions. The Conventions themselves are also of interest, as they offer ‘various forms of flexibility’ to the ratifying Member State (Betten 1993b, p. 21). The ILO adopts Conventions on specific issues within its remit that set minimum standards of performance, or that concern principles and guidelines for policy. Conventions are usually accompanied by recommendations that contain guidelines for implementing measures that would give effect to the Convention. Many Conventions suggest minimum standards, for exposure to dangerous substances in the workplace for example, but Member States are able to adopt higher standards if they so wish under art. 19 (8) (ILO 1919). Less-developed countries (LDCs) may apply lower norms than stipulated in the case of some Conventions. The development of industrial organization or ‘other special circumstances’ is to be taken into account in both the creation and review of ILO Conventions (ILO 1919, art. 19 (3)). Finally, some conventions offer the possibility of adopting one or more parts rather than the entire convention (Betten 1993b, p. 22). ILO Conventions and recommendations are ‘regularly and systematically monitored’ on the basis of national government reports submitted to the Committee of Experts and the Conference Committee on the Application of Conventions and Recommendations that reports to the Governing Body (Swepston 1992, p. 100). ILO supervisory procedures also allow for the submission of representations from worker and employer groups on ratified conventions, and on freedom of association (the right to form a union) irrespective of whether the state has ratified
150 European Welfare States
Conventions nos. 87 and 98 on the basic principles of freedom of association. The first form of representations are covered by arts. 24 and 25 of the ILO Constitution, and the latter by the special supervisory mechanism concerning freedom of association established in 1951. An exchange of letters on the representation shall occur, and if a government statement concerning the Convention under consideration is not received or is ‘not deemed to be satisfactory by the Governing Body’ then the exchange of letters will be published (ILO 1919, art. 25). Should an ILO Member State or delegate request it, complaints regarding non-application of ratified laws may then be submitted to the International Labour Office and the Governing Body will create a formal Commission of Inquiry (ILO 1919, art. 26). In extreme cases where information is not forthcoming, direct contact missions may be sent to the Member State under examination. The naming and shaming of states that fail to implement Conventions is usually the ultimate remedy available to the ILO, for, as the former assistant Director-General described, the ILO operates through ‘persuasion’ rather than the ‘imposition of mandatory measures’ (Valticos 1982, p. 381).4 In an instance of non-acceptance of ILO Commission of Inquiry recommendations, a state may refer the case to the International Court of Justice (ILO 1919, art. 31). In 2000 the ILO adopted, for the first time in its history, an action under art. 33 of its constitution (failure to carry out recommendations of Commission of Inquiry or ICJ) against Myanmar/Burma on the issue of forced labour. The resolution recommended that all ILO members review their relations with Myanmar and take ‘appropriate measures’ such that its practice of forced and compulsory labour might be stopped (ILO 2000b). The relation between the action and WTO rules against products of prison labour (art. XX (e)) ‘is yet to be clarified’ as WTO art. XX(e) has never been invoked (Commission 2001a, p. 6). Lest we judge the ILO system of supervision to be insufficient, it should be remembered that the ILO enforcement mechanism is not the weakest method available in the global system. Indeed, ‘the system of supervision by … the ILO, is more likely to encourage compliance than is the mere issuance of “views” by the UN Human Rights Committee’ (Hannum 1992, p. 29). However, it is clear that ILO enforcement mechanisms are much weaker than those available to WTO dispute settlement panels, which may recommend changes to national laws, and also compensation or retaliation to injured parties (Hoekman and Kostecki 1995). The enforcement capacity of the ECJ of the EU is also far more robust. The ECJ is able to strike down national legislation in cases where
Global Governance of Social Policy 151
it conflicts with EU law, and impose fines on states that do not comply (Craig and De Burca 2003).
National ratifications of ILO Conventions by EU Member States This section will examine the number of ILO Conventions ratified by EU-15 Member States. Is an explanation based on national welfare state histories sufficient to explain their behaviour at the ILO? Also, since joining the EU, has the number of ratifications by Member States declined, stayed stable, or increased? Member State ratification of ILO Conventions Given the historically well-developed institutional structures of the welfare state in social-democratic and conservative-corporatist welfare state regimes, it should be expected that these states have ratified the most Conventions. Liberal welfare states have also been very well developed since the end of the Second World War, and since ILO legislation is often set as framework legislation without benefit levels it is expected that they will have ratified legislation, especially since 1945. States with southern welfare state regimes are expected to have the fewest number of ratifications, given the lower level of development of their institutional structure and the fact that they were under dictatorships (Portugal until 1974, Greece until 1974 and Spain until 1975) which, it is speculated, will not have promoted international cooperation. Table 5.1 illustrates the total number of ILO Conventions ratified by Member States. The table provides some surprises. The first is that welfare state regime types have very little predictive value regarding the number of ratifications of ILO Conventions by states. Regime types are not clustered together. Although two states with social-democratic regime types are near the top and over the EU average (Sweden and Finland), one is near the bottom (Denmark). Two states with liberal welfare regimes are both below the EU average, but the UK with 85 ratifications is very near the average and clearly has not resisted ratifying ILO legislation. The biggest surprise, however, is that southern welfare states have a very mixed ratification result, with two at the top of the EU in the number of ratifications (Spain at 128 and Italy at 108) and two below the EU average (Portugal at 75 and Greece at 69). National welfare state regimes thus have a very low explanatory power in ILO ratification records among EU Member States. The national labour laws of the Member State – which may be more or less predisposed to international
152 European Welfare States Table 5.1 Total number of ILO Convention ratifications by EU-15 Member States Member State
Welfare state regime type
Spain France Italy Netherlands Finland Sweden Belgium UK Germany Portugal Luxembourg Ireland Greece Denmark Austria EU-15 average
Southern Conservative-corporatist Southern Conservative-corporatist Social democratic Social democratic Conservative-corporatist Liberal Conservative-corporatist Southern Conservative-corporatist Liberal Southern Social democratic Conservative-corporatist
Total number of ILO Conventions ratified (of 184)* 128 116 108 101 96 91 89 85 76 75 75 73 69 67 52 86.7
* 1919–2001; includes Conventions later revised or shelved.
legislation in setting standards in the labour market as opposed to other legal forms, such as national statutes and collective bargaining – is one possible explanation for the gap between expected and actual behaviour.5 Another explanation may be that southern welfare states have turned to the global level to provide minimum standards in order to modernize weak national systems. Table 5.2 illustrates the total number of ILO Conventions ratified by Member States, and the percentage of ILO Conventions ratified before and after the state became a member of the EU (or the then EEC). Even though the ILO began proposing Conventions in 1919, most EU-15 Member States have ratified fewer Conventions since joining the EU. Even among the original 6 states that joined the EEC in 1958, 3 ratified more ILO Conventions from 1919 to 1957 than from 1957 to 2001. The fact that more Conventions were proposed from 1919 to 1957 (107) than from 1957 to 2001 (77) is clearly one reason why more ratifications occurred before 1957. A more accurate picture of the challenge presented by EU membership for ILO ratifications is presented by Table 5.3, where we control for the
Global Governance of Social Policy 153 Table 5.2 ILO Convention ratifications by EU-15 Member States before and after EU membership EU Member State (by date of membership in the EEC, then EU) France (1 January 1958) Italy (1 January 1958) Netherlands (1 January 1958) Belgium (1 January 1958) Germany (1 January 1958) Luxembourg (1 January 1958) UK (1 January 1973) Ireland (1 January 1973) Denmark (1 January 1973) Greece (1 January 1981) Spain (1 January 1986) Portugal (1 January 1986) Finland (1 January 1995) Sweden (1 January 1995) Austria (1 January 1995)
ILO Conventions ratified before EU membership as % of total proposed*
ILO Conventions ratified after EU membership as % of total proposed
63.0 52.7 42.6 63.0 43.4 36.0 77.6 67.1 76.1 58.0 89.8 86.7 89.6 93.4 92.3
37.0 47.3 57.4 37.0 56.6 64.0 22.4 32.9 38.9 42.0 10.2 13.1 10.0 6.6 7.7
* 1919–2001; includes Conventions later revised or shelved.
number of ratifications proposed in a period and the date of EU membership by examining only one period. Table 5.3 examines the number of ratifications of ILO Conventions proposed from 1987 to 2001. A total of 22 ILO Conventions were proposed from 1987 to 2001, during which time 4 EU treaty revisions also took place (SEA, Maastricht, Amsterdam, Nice) that each expanded the competency of the EU in social policy. Table 5.3 indicates that EU-15 Member States have performed rather poorly in ratifications since 1987. Of the 3 Member States that joined the EU in 1995, 2 have the most ratifications during the period 1987–2001 (Finland and Sweden), and Austria, which had the fewest number of overall ratifications, jumped to sixth place in the period 1987–2001.6 The other surprise is that the original 6 Member States are among those who ratified the least number of Conventions from 1987 to 2001, with France and Belgium tied for the worst record. Despite the fact that in overall ratifications the original 6 Member States have a higher-than-average record of 94.1 ratifications, from 1987 they have an average of 2.8 ratifications, well below the EU-15 average of 4.5.
154 European Welfare States Table 5.3 Ratifications by EU-15 Member States of ILO Conventions proposed 1987–2001 Member State (and date joined the EU)
Finland (1995) Sweden (1995) Spain (1986) Ireland (1973) Netherlands (1958) Austria (1995) Denmark (1973) Italy (1958) Germany (1958) Luxembourg (1958) Portugal (1986) UK (1973) Greece (1981) Belgium (1958) France (1958) EU-15 average
Number of Conventions ratified (out of 22) proposed since 1987
12.0 10.0 9.0 7.0 5.0 4.0 4.0 4.0 3.0 3.0 2.0 2.0 1.0 1.0 1.0
Conventions ratified as % of total proposed since 1987
54.0 44.5 41.0 31.2 22.7 18.2 18.2 18.2 13.6 13.6 9.1 9.1 4.5 4.5 4.5
Rank in the EU for 1987–2001: 1 ⫽ most successful (overall rank for total number of ratifications) 1 2 3 4 5 6 6 6 7 7 8 8 9 9 9
(5) (6) (1) (11) (4) (14) (13) (3) (9) (10) (10) (8) (12) (7) (2)
4.5
The EU-12 have an average of 3.5 ratifications, while the EU-15 average jumps to 4.5 ratifications. As the EU legal regime increasingly comes to resemble a federal legal system with federal and state (in this case EU and national) laws, ILO ratifications may decline. Federal systems have lower records of ILO ratifications, for example the United States with 14, Canada with 30, Switzerland with 56, and Germany with 76 as compared with an EU-15 average of 86.7 ratifications. This may indicate that as the EU legal regime becomes more entrenched, Member States find it harder to ratify global Conventions. A more robust conclusion cannot be drawn, however, unless the demands of ILO legislation over the period 1987–2001 are compared with earlier periods on some kind of legal and cost basis in order to assess whether increasingly detailed, technical or costly-toimplement ILO legislation was also an explanation for the low number of ratifications.
Global Governance of Social Policy 155
EU–ILO relations: formal and informal inter-institutional cooperation In this section the relationship between the EU and the ILO is explored from several perspectives. The first sub-section explores the formal agreements that have developed between the ILO and the EU, and ECJ case law that has added further detail, although perhaps not clarity, to the formal agreements. The second sub-section explores the day-to-day coordination among EU Member States at the ILO via the Presidency of the EU, and the Industrial Market Economy group (IMEC) in Geneva. The relative importance of these groups (the EU via the Presidency or IMEC) depends on whether the issue at hand is of a political, technical or budgetary matter. EU coordination is most effective on political matters, more mixed in areas of technical cooperation and very weak on budgetary matters. Relations between the ILO and the EU as represented by the European Commission: Formal Agreements and ECJ Decisions In 1953 the ILO and the ECSC concluded their first agreement on cooperation in the promotion of social and economic progress, the improvement of living and working conditions, and the development of employment. The 1953 Agreement allowed for mutual consultation, for the exchange of information and statistics, and for the ILO to be both consulted and contracted for technical assistance by the ECSC. The High Authority of the ECSC was also invited as an observer to the meetings of the Committee of the Coal Industry and the Committee of the Steel Industry of the ILO (ECSC 1953).7 An ILO expert committee was asked in the mid-1950s to advise founding governments on the social competencies to be given to the European Economic Community (ILO 1956). The Ohlin Report of 1956 noted: so long as we confine our attention to international differences in the general level of costs per unit of labour time, we do not consider it necessary or practicable that special measures to ‘harmonize’ social policies or social conditions should precede or accompany measures to promote greater freedom of international trade as such differences reflect variations in productivity (ILO 1956, pp. 40–1).8 The persistent themes of the debate about social aspects of
156 European Welfare States
economic integration are present in the Ohlin Report: the fear that competition from countries with lower labour standards and lower labour costs might be unfair, that the social costs of integration might be spread unequally in society, and that countries with more advanced labour standards might find it difficult to raise wages or social benefits in light of international competition (ILO 1956, pp. 3–4). Many of the Ohlin Report recommendations were included in the EEC Charter one year later, such as in art. 48 (now art. 39) on the free movement of labour, art. 118 (now art. 137) on provisions for cooperation in social security, and art. 119 (now art. 141) on equal pay. The Report also suggested the use of ILO Conventions by European countries in order to solve ‘certain of the social problems connected with closer European economic co-operation’ (ILO 1956, p. 116). The idea of using ILO Conventions as a basis for social policy in the EEC was proposed by Belgium (Belgium 1955). The following ILO Conventions were suggested for possible common ratification by each of the six Member States: Conventions nos. 1 (Hours of Work in Industry), 52 (Holidays with Pay), 63 (Statistics of Wages and Working Hours), 81 (Labour Inspectors), 87 (Freedom of Association and Protection of the Right to Organize), 88 (Employment Services), 95 (Protection of Wages), 96 (Fee Charging Employment Agencies), 97 (Migration for Employment), 98 (the Right to Organize and Collective Bargaining), 100 (Equal Remuneration), and 102 (Minimum Standards for Social Security). The Common Market sub-committee on social issues examined the Belgian proposal, although in the end this method was not used and an independent European Commission was created to promote cooperation on social issues among the six new Member States (Sous-Commission 1955). From the beginning of the EEC, then, using the ILO to create common standards among the Member States was rejected in favour of an independent regional system. In 1958, an accord signed by Walter Hallstein, president of the EEC, and David A. Morse, Director-General of the ILO, established the modalities for their cooperation. The agreement was made on the basis of a combination of arts. 117 and 229 of the Treaty of Rome (the promotion of living and working standards, and the maintenance of useful contacts with the UN and Specialized Agencies, respectively) (EEC 1958). Consultation between the International Labour Office and the Commission, the possibility of issuing invitations to representatives of the ILO and the EEC to express opinions in the various committees of both organizations, the exchange of information, and terms for technical assistance were established. Notably, however, no specific agreement was made on the
Global Governance of Social Policy 157
possible observer status of the EEC at the ILO in the final Agreement, as had been made for the High Authority of the ECSC at certain Committee meetings of the ILO in areas of its competence (ECSC 1953). An early draft of the ILO–EEC Agreement found in the ILO archives did contain provisions for reciprocal representation in the form of mutual observer status, but for unknown reasons these were crossed out and not included in later drafts of the Agreement (ILO 1957?). A draft article permitting an ILO representative to be invited to the Economic and Social Committee (ESC) was also deleted from the draft Agreement (ILO 1957?). This situation was not formally rectified until 1989, when the standing orders of the ILO were amended to provide the Commission, acting on behalf of the EEC, observer status (Commission, Exchange of letters OJC 24/8, 1 February 1990). In the 1958 Agreement the Commission was given the gatekeeping function of inviting the ILO to any ‘appropriate organ of the EEC subordinate to it’ (EEC 1958, art. 5).9 Contact between the ILO and the Commission from 1952 to 1972, for which there is open access to all records, is marked by an early period of intense cooperation and then a gradual decline to ad hoc arrangements marred by poor communication and highly dependent on the personalities involved. From 1952 to 1962 the importance of the ILO in providing information to the ECSC, especially on health and safety questions but also in the area of vocational training, and to the EEC on social security and labour mobility questions, is evident. In 1961 a permanent contact committee (comité permanent de contact) between the ILO and the EU was created, consisting of two European commissioners (Social Affairs and External Relations), two members of the ILO (one of which was traditionally the Director-General), and other functionnaires assigned to it on an ad hoc basis (ILO 1961). The committee was to meet up to two times annually. It also became practice for the European commissioner for Social Affairs to meet with the ILO Director-General once a year. The ILO established an office in Brussels to facilitate cooperation and information exchange with the EU, and the Mission of the European Communities to the International Organizations in Geneva has served as the contact point for ILO–EU relations for the European Commission. The high-level meetings ‘continue to be the most important source of communicating information from the EU to the ILO’, followed by information gathering by the ILO office in Brussels, and then service-level meetings between ILO and Commission officials in both Brussels and Geneva (Interview 34). In the period 1952–72 the ILO was the senior institution of the two, with a well-developed network of researchers and experts that had taken
158 European Welfare States
the lead in developing international legislation in the social field. By the 1970s, however, the ambitions of the EEC had begun to surpass those of the ILO. Regional initiatives took precedence over ILO activity in the eyes of Member States, and new Commission initiatives, such as the 1970 Tripartite Conference on employment and the development of product safety standards, took little consideration of what was being done at the ILO. Where personal relations were good between the ILO representative in Brussels and the contact point in DG V, information was exchanged in a timely fashion. When such ‘lynch pins’ (chevilles ouvrières) of the relationship disappeared, communication could be delayed and patchy (ILO 1970b).10 The importance of person-to-person contact for EU–ILO cooperation remains, as, in the words of an ILO official in an interview in 2002: ‘many contacts between the ILO and EU officials are based on personal contact and not a systemic relationship’ (Interview 43). In the mid-1970s the inter-institutional balance between the two institutions shifted almost completely. The ILO could provide views on highly technical matters, notably the harmonization of the social security regimes of the new Member States in the case of migrant workers, but the idea that the ILO could serve as ‘an adviser and mediator in the process of mutual accommodation inside a widening community’ retreated as the politics among the nine EEC Member States sidelined global-level intervention in their highly complicated bargains (ILO 1971b, p. 8). When the first enlargement of the EEC was negotiated in 1973, a note to the Director-General of the ILO stated: ‘the Communities have both the expertise and resources, and the direct means of action to promote and implement a dynamic social policy, which far exceed those of the ILO’ (ILO 1971b, p. 4). ILO officials thus recognized that the EEC Treaty, combined with the intellectual resources available to the Commission and the legal capacity of the ECJ, had surpassed the policy capabilities of the ILO. Concrete examples of the interaction between EEC and ILO legislation support the observation of the growing dominance of EEC over ILO legislation. In 1979 the Council, acting on the basis of a Commission proposal (Commission 1979), granted the Commission the authority ‘to participate in the negotiations concerning the revision of ILO Convention no. 67 on hours of work and rest periods’ as Council Regulation 543/69 partially covered areas which were up for renegotiation in the ILO. The important role of the EEC in adopting or rejecting amendments to the Convention during technical discussions, and the poor representation of developing countries, led to the new Convention (no. 153) being ‘by and large compatible with Community legislation’ but also to the
Global Governance of Social Policy 159
unwillingness of developing countries ‘to vote in favour of such strict provisions’ (Council 1979a). Exceptions were therefore added to the Convention. However, Member States could not agree to ratify Convention no. 153 in common via the Council. In the end, only Spain ratified. There was also ‘considerable political reaction from employers’ and workers’ delegations at the Conference’ to Commission participation. Member States requested that the Commission adopt ‘a low profile’ in order ‘to avoid the impression that it was trying to dominate the Conference by voting as a bloc’ (Council 1979b). In 1991 the ECJ ruled that the ILO Convention no. 89 on Night Work (Women) was incompatible with art. 141 (ex 119) of the Consolidated Treaties and Directive 76/207/EEC on equal treatment in employment of 1991 (ECJ 1991).11 The ILO Convention was struck down by the nine Member States who had ratified it without any prior notification or discussion with the International Labour Office. Notably, in light of EU enlargement, ILO Members from Central Europe have also denounced Convention No. 89 in 2001 (Czech Republic, Slovakia). Member States also proved unwilling to submit a common response to the ILO in 1990 on the preparation of Convention No. 170 on safety in the use of chemicals. The Commission referred the matter to the ECJ with the view that Member States were, in the case of Convention No. 170, required to inform the ILO that the European Community institutions had exclusive competency in this matter under art. 118a (now art. 137) (Betten 1993a, p. 246). The European Commission also asked the ECJ for an advisory opinion on the competence of the EU to ratify this Convention. This would, ideally, create a situation where ILO Conventions had common effect across all EU Member States with the possibility of ECJ enforcement should they fail to implement them. Making ILO Conventions part of enforceable EU law ‘would prevent Member States from evading or denouncing their provisions by retreating behind their own constitutional systems’, since the ILO must rely on national courts to implement Conventions (Rhodes 1991a, p. 16). Ratifying ILO Conventions in common via an ECJ-enforceable Directive would not only provide a pan-EU enforcement mechanism, but would also create coherence among the Member States in regard to their international obligations in the social policy field. At present, there is very patchy ratification of ILO Conventions by Member States. Such a complex process, it should be noted, would have to be supervised by the Commission, as the ILO would ‘not concern itself with disciplining any Member States which failed to respect these Community arrangements’ (ILO 1994, para. 7).
160 European Welfare States
The ECJ ruled that ‘ILO Convention No. 170 is a matter which falls within the joint competence of the Member States and the Community’ and that the institutions of the EU and the Member States have ‘the requirement of unity in the international representation of the Community’ (ECJ 1993). In cases where Member States have retained joint competence at the EU level ‘the right to speak [at an international organization] will be exercised by the Presidency [of the EU] and not by the Commission’ (Frid 1995, p. 195). Given that any legislation composed of minimal rules, as ILO legislation is, is to be considered an area of joint competence according to the ECJ opinion, the assignation of competencies with regard to leadership, representation, coordination, and ratification between the Commission and the Member States on ILO Conventions remained ambiguous (ECJ 1993). In practice, however, the Member States have proven unwilling both to submit a common response to ILO Conventions in areas where they share competence with the EU, and to permit the use of EU mechanisms to ratify ILO Conventions. Interviews with national representatives at the ILO revealed that Member Sates have a range of opinions on the enhancement of formal EU coordination at the ILO. Member States such as France, Germany and the UK have cited legal reasons (competence, and the fact that the EU is not a member of the ILO) for their opposition to formal Commission leadership on ILO issues (ECJ 1993, para. IV; Frid 1995, p. 302). France and Denmark have also opposed formal Commission coordination of submissions to the ILO because of the autonomous role of the social partners in developing ILO legislation provided in the ILO Constitution (Council 1986, p. 3 annex II). Greece, however, is willing to explore greater EU coordination at the ILO (Interviews 25, 27). This may be explained by both its traditionally weak administrative practices and the weak development of social partners that would oppose such a delegation of powers. Parallel to the unresolved difficulties of formal EU representation at the ILO, problems of inter-institutional cooperation between the ILO and the EU have persisted. An ILO review of technical cooperation in 1988 noted three main problems hindering cooperation: the ‘strong preference’ of the Commission for executing agencies based in the EEC, the complexity of EEC rules and procedures that often conflict with those of the ILO, and the decentralization and fragmentation of decision making in the EEC between different directorate-generals (DGs), which the ILO found hard to cope with (ILO 1988a). This review led to an initiative to revitalize the ILO–EU relationship through the enhancement of the activities of the permanent contact committee and the
Global Governance of Social Policy 161
formal recognition of the EU’s observer status, as represented by the Commission (Commission 1990). Still, more than a decade after this review, an ILO official cited the difficulties of tracking issues between Commission DGs and the lack of budgetary and financial compatibility as two continuing problems in ILO–EU relations (Interview 41). Preenlargement technical assistance, and advice on adaptation to the labour market sections of the acquis communautaire to Central and Eastern European states, was one example cited by the ILO as an area where joining up ILO technical assistance and EU initiatives (such as PHARE and TACIS funding) was frustratingly complex (Interview 41). An exchange of letters in 2001 produced the same rhetoric on the importance of ILO–EU cooperation, and yet indicated the more advanced stage of social policy legislation at the regional level. Specific initiatives of ILO–EU cooperation are an exchange of information and experience on the EES with the ILO Employment and Development Office, ‘targeted collaboration’ on social protection issues, and ‘the possible dissemination of the lessons from the European experience with social dialogue to other regions of the world’ (Commission 2001b). In employment, social protection and social dialogue, then, the EU has advanced to a deeper stage of integration than the ILO, facilitated by a more cohesive and much smaller membership, and by a more robust (though not necessarily less complex) set of institutions. Where the ILO maintains leadership is in the promotion of labour standards, an area in which it is well poised to exercise leadership given its global membership. The 2001 exchange of letters noted a ‘strong common interest’ in promoting ‘labour standards and human rights … alongside economic development and trade liberalisation’ via the ILO’s 1998 Declaration on Fundamental Principles and Rights at Work and the EU Charter of Fundamental Rights (Commission 2001b). In the words of one Commission official: ‘states are jealous of their international role and they don’t want to share it with us [the European Commission]’ (Interview 34). In other words, despite the enhancement of the EU’s remit in social policy in successive Treaty revisions, Member States resist transplanting formal regional cooperation in social policy to the global level.12 Given the opacity of the ECJ’s ruling in 1993 and the ‘unsavoury’ option of further ‘litigation over their [Commission vs. Council] competence in matters relating to the ILO’, only negotiation between all parties concerned will create effective solutions to the challenges of EU representation at the ILO (ESC 1995). In practice, this is exactly what is occurring on an informal and case-by-case basis in Geneva, with coordination of the Member States led by the EU Presidency.
162 European Welfare States
Informal coordination at the ILO: the Presidency, regional groups and IMEC The ILO Constitution would not prevent several Member States from expressing their views through a regional representative, but no formal agreement on EU representation of Member States at the ILO has been forthcoming (Frid 1995, p. 295). The reluctance of Member States to cede formal competence to the Commission contrasts with day-to-day practices at the ILO. In fact, the Member States coordinate via three mechanisms: the EU Presidency, formal regional groups and the Industrial Market Economy Countries group (IMEC). New ILO legislation is also discussed in Brussels via the coordinating unit in DG Employment and Social Affairs. In comparison with formal instructions from Brussels via the 133 Committee on trade matters (so named after the Treaty article on trade), however, coordination on ILO social policy in Brussels is very ad hoc. As a result of the absence of formal mechanisms, there is also little contact between the ILO and the European Parliament (Interview 34). Most of the coordination of the EU Member States at the ILO thus occurs in Geneva. The country holding the Presidency of the EU ‘has a great role to play in coordinating positions and reading out the position of the EU to the ILO meetings’ (Interview 27). Presidency leadership of EU coordination is also necessary because at times the right of the Commission representative to speak for the EU has been questioned by social partners from the Member States, including those of the UK and France (Frid 1995, pp. 299–300). The Presidency chairs meetings of Member States held in Geneva. These are considered to be Council, or intergovernmental meetings. Such regular meetings are held either just before or just after IMEC meetings (discussed below). The Presidency can also take the initiative to put issues on the agenda for discussion by the Member States. At a technical level, for example, the proposed ILO Maternity Convention (No. 183 of 2000) was discussed heavily in internal Commission meetings and among Member States to create proposals for the Convention. Effective EU technical cooperation may have backfired with regard to the ILO’s goals, however, for the resulting legislation ‘was too detailed for the majority of ILO Member States’ (Interview 53). Italy is the only EU Member to have ratified Convention No. 183. A similar outcome occurred when EU Member States tried to coordinate on the proposed ILO Convention on Part-Time Work (No. 174 of 1994). Coordination got ‘bogged down in the Social Affairs Committee’ over disagreements as to whether an ILO Convention could be effectively enforced (Interview 53). In the end,
Global Governance of Social Policy 163
only four Member States ratified the Convention (Finland, Italy, Luxembourg, the Netherlands). Still, the EU-level social partners used the ILO Convention as a key reference document during Social Dialogue negotiations, as did the European Parliament (Parliament 1997). Presidency coordination of a common EU position is most effective on high-level political questions, such as the treatment of trade unions in Colombia, sanctions against Myanmar/Burma, and support for the implementation of core labour standards, especially on child labour. A coordinated text is presented on these issues by the Presidency, after consultation with the Commission in Brussels and/or Geneva, and Member States suggest amendments. EU meetings are also held during the International Labour Conference to adjust the statement to any new developments. Other ILO and IMEC members have rated EU cooperation as ‘most often effective on issues of foreign policy and political issues’ (Interview 28). The EU Accession states, some of which as of 1 May 2004 are now Member States of the EU, also often associate themselves with the EU position on political matters. EU Accession states are ‘consulted but do not take part in coordination meetings [among Member States]’ in Geneva (Interview 34). Consultation takes place before the International Labour Conference, and ‘it is up to them [the representatives of the Accession states at the ILO] to associate themselves with us’ by stating their association with the EU position during their national intervention. All candidate countries were ‘treated equally’, according to the Commission, with no division between first-round and second-round enlargement states (Interview 34). Other EU Member States feel that the Commission ‘pressures the CEECs’ (Central and Eastern European Countries) and ‘gives them their position rather than consulting or sharing information with them’ (Interview 25). Still, nothing holds the Member States to this common position for, in practice, the Member State holding the Presidency of the EU presents a common view on behalf of all EU governments, but voting by Member States is ‘not always coherent’ with that view (ILO 1993a, p. 2). In addition, after EU statements are read by the Presidency, government representatives from EU Member States may stand to support the EU position but may also add their own comments and priorities. The prospect that Presidency leadership could somehow be reversed to make way for sole Commission leadership at the ILO is thus ‘gloomy’, as Member States still enjoy having the ability to make their own independent interventions at the International Labour Conference (Interview 34). Much depends on the leadership of the state holding the Presidency. Belgium promoted improved cooperation among EU states at the ILO
164 European Welfare States
during its Presidency of the EU Council in the second half of 2001. The Belgian Presidency met bilaterally with representatives from other IMEC states to inform them that ‘IMEC needed to take a back seat to EU coordination at the ILO’ (Interview 31). Belgium’s status as a small state that could be enhanced via increased regional cooperation, and the prointegrationist ideology of successive Belgian governments, explain its interest in greater EU coordination at the ILO. Given that the Belgian Presidency (and indeed any Presidency) is only six months long, however, this initiative in Geneva had little effect on the daily practice of relations in which positions are sketched out on an issue-by-issue basis and which leaves Member States free to amplify various other issues in addition to the EU position. Formal regional groups also exist in the ILO (Asia and Pacific, Europe, Africa, the Americas), as is permitted under art. 38 of the ILO constitution and again have tripartite representation. Regional groups may make resolutions or submit reports to the Governing Body. In the European Regional Group, the Commission sends a delegation, is recognized to speak in its own right, and promotes the EU agenda concerning labour market reform (ILO 2000a). The Industrial Market Economy Countries (IMEC) crosscuts these regional groups. Given its diverse and economically powerful membership, the IMEC position is usually the most influential of the groups permitted in the ILO. IMEC was formed at the ILO in the late 1970s in reaction to the threatened departure of the US from the ILO.13 A smaller group of states to provide direction and management to the agenda of the ILO was proposed by Canada, and a representative from Canada chairs the meetings. IMEC is an informal group that, nonetheless, has developed recognition because of its membership. IMEC has 27 members including all EU-15 Member States, the United States, Canada, Australia, Japan, and New Zealand. The group meets ‘to take a common position in various ILO committees, on policy areas, and organizational issues’ (Interview 31). IMEC meets once a month at the Mission level in between Governing Body sessions which take place in March and November. It meets every day at the ILO during the Governing Body, and at the UN during the once-yearly International Labour Conference. At the Governing Body and Conference session the numbers reach as high as 100 persons as senior Heads of Delegation and their supporting staff take part. The chair of the IMEC group makes interventions and common statements in the Governing Body on political issues, and on technical issues to committees such as on standard setting. It is in the Governing Body
Global Governance of Social Policy 165
of the ILO that IMEC proves to be more effective than the EU (which has four permanent members of the ILO Governing Body – France, Germany, Italy and the UK – in addition to any elected members). On programme, budgetary and administrative issues dealt with in the Governing Body ‘IMEC is more influential than any regional groupings’ (Interview 28). The relative importance of IMEC is also affected by the fact that not all Member States are equal in the Governing Body; the four largest EU Member States have much greater influence in the Governing Body and are less inclined to cooperate on budgetary issues. There is ‘no EU coordination on finance or budgetary issues’ in the Governing Body (Interview 25). There is also no EU coordination in the Geneva Group, which brings together the largest donor states to the UN in order to discuss budget priorities for the UN institutions, including the ILO. EU coordination is thus weakest on financial or budgetary questions. Often the IMEC and the EU position is very similar such that the positions ‘are not duplicate, but overlap’ (Interview 27). This can present difficulties for EU Member States who ‘are faced with two coordinating bodies’ (Interview 25). However, the UK ‘would rather work with US and Japan rather than as a [European] bloc’ on certain issues (Interview 25). Ultimately, informal cooperation and coordination among EU Member States depends on the issue; there is great variation in the level of cooperation between EU Member States, depending on the subject involved. On political questions, EU Member States are quite well organized with a common position, while on technical interventions the picture is more mixed and reflects national priorities and preferences for social policy organization. This is reflected in the high variation of ILO Convention ratifications among Member States. Finally, on financial questions there is little EU coordination, reflecting the systemic weakness of the EU as a result of its small budget that can only support global ventures, such as the promotion of the ILO Declaration on Fundamental Rights, in a very limited way.
Social policy at the global level: five specific issue areas of cooperation Both the legal and informal modes of EU coordination at the ILO have been examined in the above section, but what of specific instances of cooperation? What roles do the EU Member States and the European Commission play in specific areas of policy-making? This section examines five areas of policy-making at the ILO. Three of these correspond
166 European Welfare States
to policy areas of EU governance and were explored in the chapters on regional governance, namely health and safety, Social Dialogue, and employment policies. Core labour standards and possible links between labour standards and trade will also be examined. Social policy emulation had begun in western Europe in the late 1800s with social insurance schemes. By exploring five different areas where the global level has been involved in social policy-making, the transmission mechanisms for social policy – which run both up from Member States and downwards from the global level – will be highlighted. Such a review also suggests a pattern to cooperation on social policy beyond the state. Areas of a more technical nature, such as health and safety, prove initially fruitful, as do minimum standards for working conditions. Over time, it is possible for members of supranational institutions to make normative statements in common concerning equality in the workplace, and worker rights, that may be followed up by legislation. National positions in these various policy areas are often missing in the discussion, and so a summary of Member State activity for France, Germany, Greece, Sweden and the UK in each area is provided in Table 5.4. As ILO Conventions are only approved at the International Labour Conference, it is up to each Member State to ratify proposed Conventions, or not. In each specific area, then, the ratification record of each state may be taken to indicate their preference for enacting an ILO policy or not. National positions on greater EU coordination at the ILO, and on cooperation between the WTO and ILO on labour standards, are also noted. Health and safety standards The original call for international legislation on labour issues emerged from the recognition that working conditions should be improved in order to raise the standard of living, improve public health and benefit production. Such a raise in standards, it was recognized by early activists, could ‘be brought about only at a multilateral level to avoid distortion of competition’ (Betten 1993b, pp. 4, 8, 187). Of the first 20 ILO Conventions, 6 concern health and safety policies exclusively (such as work with lead paints, or medical examinations for young seafarers), while a further 8 contain important aspects of health and safety legislation (such as hours of rest, minimum working age and the regulation of night work). Health and safety has been the starting-point for both global and EU governance in social policy; transnational governance structures may
Has not ratified C155, C170 or C174.
Has ratified C144 but has not ratified C154 on Collective Bargaining. Has ratified C122 on Employment Policy (1964).
Has ratified all core labour standards.
The French government is in favour of a formal link between the ILO and the WTO.
No.
Health and Safety C155 Occupational H&S (1981), C170 Chemicals Conventions (1990), C174 Prevention of Major Industrial Accidents (1993)
Social Dialogue C144 on Tripartite Concertation and C154 on Collective Bargaining
Employment C122 Employment Policy (1964) and C168 Employment Promotion and Protection Against Unemployment (1988)
Core labour standards (8 Conventions)
Links between trade and labour standards
Should the EU represent Member States at the ILO in areas of exclusive competence?
France
No.
The German government is not in favour of a formal ILO–WTO link. The Ministry of the Economy is willing to introduce social elements in economic and aid agreements but not trade sanctions.
Has ratified all core labour standards but for Convention 192 on Child Labour. 192 should be ratified by the end of 2002.
Has ratified C122 on Employment Policy (1964).
Has ratified C144 but has not ratified C154 on Collective Bargaining.
Has not ratified C155, C170 or C174.
Germany
Table 5.4 National positions in ILO policy areas
Yes.
The Greek government supports some form of ongoing dialogue between the ILO and WTO.
Has ratified all core labour standards.
Has ratified C122 on Employment Policy (1964).
Has ratified both Conventions.
Has not ratified C155, C170 or C174.
Greece
No.
Swedish Foreign Ministry: there should be a forum within the EU where trade and social issues can be discussed. This is a minimalist approach. (Interview 55).
Has ratified all core labour standards.
Has ratified C122 on Employment Policy and C168 on Employment Promotion and Protection.
Has ratified both Conventions.
Has ratified all 3 conventions (and is the only member state to have done so).
Sweden
No.
Not in favour of any link between the ILO and the WTO. The ILO is the place to discuss labour market issues at the global level. Reluctantly agreed to labour standards in the EU’s GSP regulations.
Has ratified all core labour standards.
Has ratified C122 on Employment Policy (1964).
Has ratified C144 but has not ratified C154 on Collective Bargaining.
Has not ratified C155, C170 or C174.
UK
Only the following have not ratified all 8 Conventions: Estonia, Latvia, Czech Republic.
9 have ratified C122; only Malta has not. None have ratified C168.
8 have ratified C144, but only 4 have ratified C154.
6 have ratified C155, none have ratified C170, and only 1 (Estonia) has ratified C174.
10 new Member States (2004)
167
168 European Welfare States
find it possible to begin social policy work in highly technical areas where human health in concerned, and that may be justified by resulting gains in productivity. In addition, it is very difficult for Member States to oppose improvements in working standards where the results of legislation aim to reduce deaths in the working population, improve the health of workers, and reduce the burden on other social policies, such as health services for injured workers. ILO Conventions on Health and Safety were among the earliest Conventions promoted by the Commission, including Convention No. 115 Radiation Protection (1960), ratified by 12 EU-15 Member States14, and No. 119 Guarding of Machinery (1963), ratified by 5 states (Denmark, Finland, Italy, Spain, Sweden). During the period 1920–70, ILO health and safety standards were quite influential in developing national practices in Europe, and its legislation even formed the backbone of Greek health and safety legislation into the 1980s (Interview 76). However, as national legislation advanced in the 1960s and 1970s, ILO legislation lost its influential status in most EU Member States. Officials from EU Member States such as Germany, Sweden and the UK agreed with the proposition that ILO legislation complements national legislation, but that national legislation is now much more developed than ILO Conventions (Interviews 19, 25, 27). In addition, the advance of EU legislation in the health and safety field has also presented complications for Member States wishing to develop and ratify ILO legislation in this field. ILO Convention No. 170 (1990) concerns Safety in the Use of Chemicals at Work. Aspects of this Convention are covered by EEC Directives.15 As a result, both Member States and the Community (in the form of a joint reply) submitted comments to the International Labour Office during the preparation of this Convention. As a result of debates concerning EU competence at the ILO, only Sweden has ratified Convention No. 170, and did so before its membership in the EU. These issues were raised again during the discussions for an ILO Convention on the prevention of major industrial accidents. The resulting ILO Convention on Occupational Health and Safety of 1981 had only 2 EEC-10 signatories out of a total of 37 (Denmark, Italy), and the 1993 Convention No. 174 on the Prevention of Major Industrial Accidents had only 1 of the EU-12 Member States as a signatory (Netherlands) out of a very low total of 7 signatories. The poor ratification record of Convention No. 174 by EU Member States is all the more surprising since the Commission was willing to take a ‘pragmatic approach’ to the ‘representation of the Community’ on this
Global Governance of Social Policy 169
Convention, allowing Member States to make independent submissions (Commission 1994). Social dialogue The poor record of the EEC in social dialogue up until the 1990s was commented on in several ILO memos that note that the ‘moral authority’ and ‘moral prestige’ of the ILO are intimately linked with its tripartite structure (ILO 1971c, pp. 2, 4). ILO officials are proud to point out that the ILO has ‘a comparative advantage [over the EU] in social dialogue’ having practised social dialogue in an institutional setting since 1920 (Interview 43). The ILO has produced several Conventions to bolster the foundations of social dialogue at the national level. Conventions no. 87 (1948) and 98 (1949) cover criteria regarding the right to join a trade union and the right to collective bargaining. These two Conventions are considered to be two of the fundamental rights of all workers. All EU-15 Member States have ratified these two Conventions. The ILO has also produced Convention No. 144 Tripartite Consultation (1976) and No. 154 Collective Bargaining (1981) that may both be implemented in Member States in a number of manners from formal agreements to once-yearly consultation sessions. All EU-15 Member States save Luxembourg have ratified Convention No. 144, while only six (Belgium, Finland, Greece, Netherlands, Spain and Sweden) have ratified No. 154. Still, despite Social Dialogue at the EU, no policy or process at the global level demonstrates more clearly the resilience of national traditions than the practice of social dialogue. The UK was singled out for reprimand during the Thatcher government for its anti-union practices on the basis of a TUC complaint to the ILO under Convention No. 87 (freedom of association).16 The UK’s awkward position as the EU Member State most removed from concertation or corporatist practices (trilateral or bilateral), even under Labour governments, is thus replicated at the ILO level. French labour representatives are also notorious at the Conference for being very outspoken against employers, preserving ‘une certaine tradition de syndicalisme révolutionnaire’ that was first described in the 1970s (ILO 1971a). The more militant attitude of French and Italian unions, some with communist roots, contrasts greatly with German and Scandinavian models of concertation (ILO 1971b, p. 6; ILO 1972b, p. 1) and with the even more ‘diverging’ tradition of weak social partner concertation in the UK (ILO 1972b, p. 3). As a result of these highly divergent traditions, national trade unions have proved unwilling to speak with a single voice at the ILO as they do
170 European Welfare States
at the EU level via the ETUC, nor are national trade unions willing to feed a common position to the Commission. This is perhaps surprising given the support the Commission has given to trade unions through the EU Social Dialogue. Union representatives have explained, however, that given the very weak nature of EU-level concertation, and the limited areas in which Social Dialogue has led to legislation, they are not willing to delegate policy-making to the Commission at the global level (Interviews 55, 56). Strong national unions have also noted that they do not want their position diluted by European coordination (Interviews 55, 56). Employers have also resisted giving UNICE a formal role in the ILO employers’ group. National employer representatives who are both ILO delegates and members of UNICE have informal contact in Geneva. The ILO employers’ group also liases with the International Organization of Employers (IOE), which has many members that are also members of UNICE. European members of the IOE meet as a group once a year, and the UNICE Secretariat also attends. Informal links that transmit information are thus certainly present, although UNICE members have made it clear that they do not want the UNICE Secretariat acting in their common interest at the ILO. Employers are ‘only in Geneva to the extent that the ILO may inspire the Commission and European legislation’ and so wish to keep tabs on ILO activities (Interview 53). For UK employers this is no doubt reflective of their desire to avoid exposing themselves to the possibility that UNICE coordination would lead to calls for a single employers’ position at the ILO, where the CBI would once again often be isolated. The great resistance of unions and employers to presenting a single voice at the global level in the ILO is perhaps the clearest indication of the continuing diversity of national practices in the social field, and the resilience of national traditions in the representation of labour market participants. The lack of EU social partner coordination at the global level also indicates the weakness of EU coordination beyond the regional level, and underlines the importance of regional learning effects on cooperation at the global level. Employment The aims of the ILO in employment policy are its promotion as a key means to combat gender inequality and social exclusion, and the creation of international standards to combat unemployment more generally (Betten 1993b, p. 329). Before the Second World War, information on job availability and assistance in getting a job through coordinated employment services were the focus of ILO activities in employment
Global Governance of Social Policy 171
(Betten 1993b, pp. 343–6). ILO Convention No. 2 obliged ratifying states to create free public employment agencies under a central authority, and required governments to collect and report on unemployment statistics. Convention No. 88 Employment Services (1948) concerned the establishment of employment services. Fourteen Member States ratified Convention No. 88 – all save Austria, although Italy and the UK have since denounced it. In the 1960s the ILO was invited to comment on European Social Fund initiatives to retrain agricultural, coal and steel, and maritime workers, for which the expertise of the ILO-associated Turin Centre proved valuable (ILO 1965; ILO 1970a). ILO Convention No. 122 on Employment Policy Convention (1964) required Member States to pursue full employment policies, and to ensure freedom of choice in employment and equal opportunities for all workers. Convention No. 122 also advocated ALMPs, such as education, vocational training and health services as means to prepare job seekers for work. Notably, ILO measures coincided with the promotion of ALMPs at the OECD (OECD 1964), and, albeit in a very limited way, in the EEC. Fourteen EU-15 Member States ratified Convention No. 122, all save Luxembourg. It should be noted that many of the priorities adopted by the EES and incorporated into its four pillars – adaptability, employability, entrepreneurship and equal opportunities – were picked up from OECD and ILO studies and reports that date back more than 20 years. Since the 1964 Employment Policy Convention, however, the ILO’s initiatives in employment promotion have met with far less enthusiasm from EU-15 Member States. Convention No. 168 Employment Promotion and Protection Against Unemployment (1988) was ratified by only two Member States, Finland and Sweden, which were not EU members when they ratified the Convention (both in 1990). In 1995 the Commission also denounced ILO Convention No. 96 on Fee-Charging Employment Agencies (Revised). As of 2001, 7 EU-15 Member States had denounced the Convention, while 4 (Belgium, France, Ireland, and Luxembourg) had not. Again, these developments reflect the general trend that saw interest in the ILO as a means of creating common standards in social policy diminish among EU Member States as regional powers expanded. The Global Employment Agenda (GEA) was launched by the ILO at an Employment Forum in Geneva in November 2001, and adopted by the ILO’s Governing Body in March 2002. The GEA agenda gives a clear indication that EU policies have surpassed those of the ILO. Although still in its development phase, the GEA is to call for global institutions,
172 European Welfare States
including the World Bank, the IMF and UN agencies, to recognize the importance of employment strategies to reducing poverty, preventing discrimination and social inclusion, and promoting skill development and adaptability. There ‘are a lot of common features between’ the GEA, the EES and the OECD Jobs Strategy (ILO 2002, see para. 1.1). In the words of one Commission official ‘the ILO Employment Agenda will be quite different than the EES, but we can inspire principles and values of the ILO Agenda, although perhaps not specific policies’ (Interview 34). Certainly the National Action Plans have inspired the thinking of ILO officials who hope to include such reporting in the GEA to supplement the Country Employment Policy Reviews and the World Employment Report already produced biennially by the ILO’s Employment Strategy Department. The GEA leadership also has an overlap with the EES in one key personality, Allan Larsson, former Swedish Minister of Finance, former DG at the Social Affairs Directorate in Brussels, and a leader in the development of the GEA. One official speculated: ‘Mr Larsson may be trying to do for the ILO what he did for Europe’ (Interview 41). Notably, however, the EU Member States have not coordinated their position on this initiative, viewing it as one that will largely affect developing countries and provide a link to the UN Millennium goals on poverty reduction (Interview 27). In the view of EU Member States, while the ILO may use the European Employment Strategy (EES) as inspiration for the Global Agenda on Employment, they are ‘separate matters’ (Interview 27). The present state of employment policy at the global level demonstrates that the EU has surpassed the ILO in producing concrete initiatives. The fact that EU Member States are not yet coordinated on employment issues at the global level is no doubt linked to EU Treaty competence in this issue area where, even more so than in the case of health and safety policy, states retain competence. Core labour standards The 1919 Constitution of the ILO does not identify any specific human rights or labour standards as of special or primary importance. In the 1970s, the European Commission pushed the ILO to identify certain Conventions as ‘fundamental norms’ in order to get ILO approval for a linkage between a respect for ILO Conventions and trade preferences for less-developed countries (Commission 1978; ILO 1984). This was in the context of EEC negotiations with the ACP (African, Caribbean and Pacific) countries over the Lomé Convention. It was not until the 1980s, however, that the ILO identified the freedom of association, the
Global Governance of Social Policy 173
prohibition of forced labour, and equal pay as core labour rights. Up until then ‘we [the ILO] had avoided designating any of our Conventions as a right’ (Interview 43). Beginning in the mid-1980s, a debate began both inside the ILO and among its members about the possible benefits of highlighting specific Conventions as fundamental norms or human rights, and about which Conventions might be considered as such (Betten 1993b, p. 66). The external impetus to identify certain Conventions as ‘core labour standards’ gained momentum at the 1995 World Summit for Social Development in Copenhagen, which identified specific Conventions as basic worker rights. In the Programme of Action, governments committed themselves to safeguarding and promoting respect for basic workers’ rights, including the prohibition of forced labour and child labour, freedom of association and the right to organize and bargain collectively, equal remuneration for men and women for work of equal value, and nondiscrimination in employment, fully implementing the conventions of the International Labour Organization (ILO) in the case of States parties to those conventions, and taking into account the principles embodied in those conventions in the case of those countries that are not States parties to thus achieve truly sustained economic growth and sustainable development. (UN 1995, para. 54 (b)) The shift in language from one of standards to one of rights was ‘of fundamental importance’ to shifting the debate over which core labour standards should be singled out from one based on efficiency arguments and concerns about trade protectionism to one based on the rights that are needed to ensure equality in the pursuit of justly compensated work (Langille 1997, p. 34). A shift to the language of human rights also helped to overcome the resistance of the United States, which has a poor record of ratification of ILO Conventions, and others to the idea of collective rights that might belong to a group rather than solely to an individual. The 1995 Copenhagen Summit, the 1996 OECD study on international trade and core labour standards (OECD 1996) and the 1996 Singapore WTO Ministerial Declaration acknowledged the existence of ‘internationally recognised core labour standards’ and identified the ILO as ‘the competent body to set and deal with these standards’ (WTO 1996b). These statements reinforced the ILO’s work on a Declaration outlining the core labour standards.
174 European Welfare States
The 1998 ILO Declaration on Fundamental Principles and Rights at Work identified 5 core labour standards covered by 8 ILO Conventions. These are: freedom of association and the right to bargain collectively (Conventions Nos. 87 and 98), the elimination of all forms of forced or compulsory labour (Conventions Nos. 29, 105), the effective abolition of child labour, especially in its worst forms (Conventions Nos. 138, 182), and the elimination of discrimination in respect of employment and occupation (Conventions Nos. 100, 111) (ILO 1998, art. 2 (a–d)). Note that some refer to ‘four core standards’, with freedom of association and the right to bargain collectively combined as one core standard. The ILO is to assist its members in attaining these standards by offering technical assistance and advice (ILO 1998, art. 3). Importantly, the Convention obliges ‘all Members, even if they have not ratified the Conventions in question … to respect, to promote and to realize … [these] fundamental rights’ (ILO 1998, art. 2). All EU-15 Member States have ratified the Conventions that cover the core labour standards (Germany and Belgium being the last to do so, in early 2002). Of the 10 new Member States of 2004, 7 have ratified all 8 Conventions (Cyprus, Hungary, Lithuania, Malta, Poland, Slovakia, Slovenia), while 3 have not (Czech Republic, Estonia, Latvia). Interestingly, Convention 138 on the abolition of child labour is one that all the last 3 have yet to ratify. The draft Constitution approved by heads of government in 2004 contains the substance of the ILO core labour standards, although they are not referred to explicitly as such, including in art. I-3 a reference to the EU’s role in ‘relations with the wider world’ in contributing to ‘free and fair trade … and the protection of human rights, in particular the rights of the child’ (EU 2004). Notably, however, national practices rather than any single Union standard apply to the right of collective bargaining and action, including strike action, in the ‘Solidarity’ section of the new draft Constitution (EU 2004, Title IV). The legal status of the draft Constitution is still unclear as it has yet to be ratified by Member States. The initiative to have as many states as possible ratify the 1998 ILO Declaration has been very successful. In the words of one ILO official, ‘Although I have a philosophical objection to some Conventions being named “rights” and thus more important than others, I have no marketing objection’ (Interview 43). Extra-budgetary financing has been provided by the UK and Germany, among others, for both technical assistance relating to the promotion of the Declaration and programmes to reduce and eliminate the worst forms of child labour.
Global Governance of Social Policy 175
While forced labour may evoke images of Burmese workers, there are examples that indicate that the EU has only recently brought its own practices in line with the ILO Charter, and still has some way to go in its enforcement. European militaries, including those of Belgium, Ireland and Greece, permitted forced labour as a disciplinary punishment, and penal codes in France, Germany and Spain permitted forced labour in prisons (Betten 1993b, p. 147). Complaints against the UK on union recognition occurred in the 1980s, and Sweden had a complaint filed against it for non-compliance of the Freedom of Association Conventions and Merchant Shipping Convention No. 147. Further, there is a malignant problem of forced labour, in particular the traffic of women into prostitution, throughout the EU and especially in large urban centres such as Berlin, Brussels, London, Rome that remains to be addressed. Trade and labour standards A historical examination of the treatment of labour standards in the international trading system reveals that labour standards have been on the agenda of international trade agreements since the late 1940s. The original proposal for an International Trade Organization (ITO) was made by the United States in 1946 as a complement to the Bretton Woods institutions for monetary policy (IMF) and development finance (World Bank). From the start, proposals for a reciprocal exchange of market access and equal (or ‘national’) treatment noted several ‘General Exceptions’ (USA 1946). The US proposals were discussed during a series of ‘United Nations’ Conferences on Trade and Employment’ held in London, Geneva, New York and Havana from October 1946 to March 1948. These conferences were attended by a diverse group of states including western European states, Brazil, China, Colombia, Egypt, Mexico and the US. The ILO was also represented as a participating member of the conferences’ committees, although it could not vote on amendments, nor was it a party to the final agreement. The International Chamber of Commerce and the World Federation of Trade Unions were also granted permission to make written submissions. Simultaneous to the development of a draft charter for the ITO, a separate drafting group of the UN Conference on Trade and Employment began work in February 1947 on a General Agreement on Tariffs and Trade (GATT) that could be incorporated into the ITO Agreement, or serve as an interim agreement on some basic rules for the trade in goods while a more detailed and broader set of proposals was being hammered out. The first ‘Tentative and Non-Committal Draft suggested by the
176 European Welfare States
Delegation of the United States’ to the GATT Drafting Committee included ‘art. XVIII: General Exceptions’, which mirrored those in its original 1946 proposal for an ITO (UN 1947b, p. 28). The list of exceptions, ‘subject to the requirement that measures are not applied in such a manner as to constitute a means of arbitrary or unjustifiable discrimination’, included ‘(h) relating to the products of prison labour’, first proposed by the US and tightened up by a Canadian proposal (UN 1947b, p. 28). This proposal for an article concerning the discrimination against products made by prison labour as a formal exception to the GATT was incorporated into the first draft of the GATT as art. XX (h) (UN 1947c). The article remained unopposed by governments during several drafting committee meetings from February to October 1947. Article XX (h) was included in the GATT Agreement of October 1947 as art. XX (e), when several of the original US proposals for art. XX were moved to a new art. XXI on ‘Security Exceptions’ (UN 1947d). There are thus two lists of exceptions in the GATT (1947): General Exceptions (art. XX), which include exceptions necessary ‘to protect public morals’, ‘to protect human, animal or plant life or health’, and the trade in such goods as gold, silver, and ‘national treasures’, and Security Exceptions (art. XXI) that cover trade in weapons and trade during times of war. Government delegates met to compose a Final Act proposing the creation of the ITO during a conference in Havana from November 1947 to March 1948. The proposed ITO would administer both the GATT and a broader agreement that would include employment and commercial policy, economic development and international investment, restrictive business practices, and an international commodity agreement (ICITO 1948c). Article 7 of the Final Act committed members ‘to eliminate substandard conditions of labour in production for export’, and included consultation and cooperation with the ILO in improving labour standards (ICITO 1948c, art. 7 (1–3)). The Final Act, or Havana Charter, created an Interim Commission of the International Trade Organization (ICITO) to elaborate on the work of the conference. The ICITO, based in Geneva, developed proposals for the elaboration of the Havana Charter. Proposals included those for relations with non-governmental organizations (NGOs), the Bretton Woods institutions and other international organizations including the ILO (ICITO 1948b). The document ‘Relations Between the International Labour Organization and the International Trade Organization’ was developed in consultation with the ILO (ICITO 1948a). It notes that in order to provide for ‘effective co-operation and the avoidance of unnecessary
Global Governance of Social Policy 177
duplication’ between international organizations the ITO agrees to ‘consult and co-operate with the ILO’ on ‘all matters relating to labour standards that may be referred to the ITO’ (Preamble). The draft formal agreement for ITO–ILO cooperation also provides for ‘attendance’ of each others’ meetings, allowing ILO and ITO representatives to ‘participate without vote in deliberations’ (art. IV), the possibility of creating joint committees on ‘questions of common interest’ (art. V), the exchange of information and documents (art. VI), and cooperation between their respective statistical services (art. VIII, ICITO 1948a). The formal agreement proposed between the ITO and the ILO never came to fruition, however, as the US Congress voted against ratifying the Havana Charter that would have enabled the creation of the ITO (Diebold 1952). The GATT was then left as the only institution to emerge from the UN Conferences on Trade and Employment from 1946 to 1948.17 Relations between the GATT (later the WTO) and the ILO were never formally established, although much later formal agreements were created between the WTO and the IMF and the World Bank that allow for mutual observer status (WTO 1996a).18 Because of the lack of any labour-related clauses in the GATT 1947 save for art. XX(e), there was no mandate for GATT–ILO relations. Proposals in the mid-1980s for limited cooperation with the ILO, suggested by the United States, were rejected by the WTO membership (Interview 47). No disputes have been launched on the basis of art. XX(e), not even in the egregious case of politically isolated Myanmar/Burma, and thus there has been no elaboration of the article and possible linkages to ILO Conventions or core labour standards by WTO dispute settlement jurisprudence. The recent ILO art. 33 Action against Myanmar/Burma has taken place exclusively in the ILO. The idea of a linkage between trade and labour was raised again during the final stages of the Uruguay Round of negotiations (1986–94). These discussions were part of the launch of the new World Trade Organization (WTO) in 1995, created by an expansion of the GATT to many of those areas covered by the original ITO, although excluding any mention of employment policy. The US, under Democratic President Bill Clinton, took the lead along with France on the issue of a GATT/WTO–ILO working party that could explore deeper links between the two organizations. However, no labour provisions were included in the final agreement to create the WTO because of developing country opposition (notably from India, Pakistan and Malaysia) (Hoekman and Kostecki 1995, p. 263). In one ILO internal report, it was noted that the ‘EC delegation at the GATT … kept a low profile regarding the
178 European Welfare States
US proposals to set up a working party’ (ILO 1988c). The ILO is invited, as a matter of informal courtesy, to observe the WTO Ministerial Meetings, but in practice even this has raised opposition from Member States. In 1996 the Director-General of the ILO, Mr Michael Hansenne, was ‘uninvited from the Singapore Ministerial’ because of Member State opposition (Interview 46). Partially in response to the failure to include a reference to labour provisions in the WTO, and encouraged by renewed US and public interest, the ILO founded the Working Party on the Social Dimensions of Trade Liberalization in 1994, renamed the Working Party on the Social Dimensions of Globalization in 2000. The WTO Secretariat submitted a paper on trade and employment policy to the ILO Working Party on Globalization, indicating that informal cooperation between secretariats had been established. The United States and the EU again raised the idea of a formal ILO–WTO working party during the preparations for the 1996 WTO Ministerial (Commission 1995c). This idea was rejected by the WTO membership at the 1996 Singapore Ministerial, as was the grander and more vague EU idea of a World Body of all international organizations (Interview 35). Still, the 1996 Singapore Ministerial Declaration stated: ‘We renew our commitment to the observance of internationally recognised core labour standards’ and that ‘the WTO and ILO Secretariats will continue their existing collaboration’ (WTO 1996b). At the 1999 Seattle Ministerial, the EU advocated granting the ILO observer status in the WTO and creating a ‘Joint ILO/WTO Standing Working Forum on trade, globalization and labour issues’ (Commission 2001a, annex 1). No conclusions were reached on this issue, however, and a proposed trade round failed to be launched. In the Commission’s assessment, ‘the apparent desire of at least one major WTO Member [the US] for linking labour standards to trade sanctions contributed to the failure of that Conference [to launch a new trade round]’ (Commission 2001a, p. 8). With the change in US government in 2001 (from the Democratic Clinton administration to the Republican administration under George W. Bush), the EU took over ‘the forefront’ of the promotion of some form of ILO–WTO cooperation. The EU ‘was definitely the leader’ in calling for an enhanced place for labour standards in the Doha Declaration of November 2001, although it was supported by the US in asking for ‘stronger’ language on labour (Interviews 34, 46). The European Commission ‘put a lot of political weight behind their submission on labour’ and Pascal Lamy (then EU commissioner for trade) was ‘furious’ when the Chair ultimately rejected the strengthening of the aspects of
Global Governance of Social Policy 179
the Singapore Ministerial Declaration that concerned labour (Interviews 46, 35). The Doha Declaration, which guides the new WTO ‘Development Round’ of trade negotiations, reaffirmed the Singapore Declaration and took note ‘of the work under way in the ILO on the social dimensions of globalization’ (WTO 2001c). At the same time as the Doha WTO Ministerial, a new ILO World Commission on the Social Dimensions of Globalization was created, perhaps in partial compensation for the lack of a strengthening of the labour language at the WTO. The ILO World Commission reported in 2004 (ILO 2004). The ILO World Commission has advocated stronger action to ensure respect for global labour standards that provide a minimum level of social protection, and work with the WTO and other international organizations in order to reach this goal. Some kind of link between trade and labour continues to be ‘the hottest issue’ at the WTO – in other words, the most contentious (Interview 47). Importantly, despite US and EU support for a link between trade and labour standards, the form of such a link has not been fully spelled out. First of all, there is a divide on the use of trade sanctions as an instrument to promote core labour standards. The US government under Bill Clinton had suggested the use of trade sanctions, while the EU firmly opposes ‘any sanctions-based approaches’ (Commission 2001a, p. 11). The EU supports cooperation between the ILO and WTO, a ‘regular dialogue’ between these two institutions, and ‘positive measures to encourage respect of core labour standards’ (Commission 2001a, p. 11). The EU has as its ultimate objective ‘some kind of formal cooperation’ between the ILO and WTO (Interview 35). It is clear, however, that even EU Member States are not in agreement about what form possible ILO–WTO cooperation could take. The UK has expressed ‘scepticism’ about the value of labour standards in the WTO, and went so far as to say, privately, to the European Commission that the EU statement on labour at the Doha Ministerial (2001) was ‘not 100 per cent acceptable’ (Interview 25). It is no secret that the UK is not enthusiastic about the EU’s stated position on enhancing WTO–ILO cooperation (Interviews 28, 29). The German government is not in favour of a formal ILO–WTO link, but the Ministry of Economy is increasingly willing to introduce social elements, such as respect for core labour standards, into international economic and aid agreements (Interview 27). For those EU Member States that do agree that some form of ILO–WTO cooperation should occur, the most minimalist approach – requiring no formal relations at the global level – is embodied in the Swedish
180 European Welfare States
proposals for a forum within the EU where trade and social issues can be discussed (Interviews 55, 56). The formal or informal nature of the ‘forum’ and its operating mechanisms are unclear. More recent statements from the Swedish government note that ‘concrete WTO support for the ILO is required’, but again, details are lacking (WTO 2001e). The Greek PASOK government supported the creation of ‘a wide-ranging permanent dialogue between all interested parties in the ILO and all relevant organizations, including the WTO’ (WTO 2001b). Proposals for a Social clause represent a maximalist approach that is usually associated with the French. A Social clause would form a separate article in the WTO Agreement and would cover minimal social standards and cultural issues ‘presumably as a condition for market access’ (Hoekman and Kostecki 1995, p. 263). What this means in detail is, again, unclear. The French have formally suggested a ‘permanent dialogue’ between the WTO and the ILO on matters of ‘social development’ (WTO 2001d). During their Presidency of 2001 the Belgians attempted to hold a series of meetings in Geneva among ‘like-minded countries’ that approve of the idea of a Social clause in the WTO, including France, Greece, Italy and Sweden, but ‘the Commission shut it down’ with the insistence that Member States should not take the leadership on this issue (Interview 30). National welfare state history provides a partial explanation of these national preferences. Member States with social-democratic (Sweden), and corporatist (Belgium, France) philosophies of the welfare state are more inclined to seek linkages between trade and labour policies than those with an individualistic animating philosophy of social policy (UK). Again, however, Germany occupies an intermediary and far less predictable place on the scale of EU Member States. Member State preferences for increased coordination at the global level, perhaps to support weak national coordination, such as in the case of Greece, are also important. The fact that a common EU submission on labour standards was made to the Doha Ministerial is a clear example of European Commission ‘upgrading’ of the Member States’ positions; taken individually Member State opinion is varied in its opposition and support for WTO–ILO cooperation, but as coordinated in Brussels by the Commission via the weekly 133 committee meeting, the position of the EU Member States is greater than the sum of its parts. Via Commission coordination and leadership of the Doha negotiations, the EU presented a clear desire for enhanced cooperation between the ILO and the WTO, taking the lead in the WTO membership on this issue (Interviews 28, 29, 46).
Global Governance of Social Policy 181
Less enthusiastic Member States, such as the UK, permitted the Commission to voice this opinion with the knowledge that the WTO membership would not approve of a strengthened WTO–ILO relationship. Although a full-blown Social clause in the WTO, or enhanced enforcement mechanisms for the ILO akin to the WTO, will not be forthcoming in the medium term (and may, in some eyes, not be the most desirable way forward19), there is one mechanism at the global level that links trade and labour in an informal manner. This is the monitoring of national labour standards through the trade policy review mechanism (TPRM) already in place at the WTO. The TPRM has been used as a vehicle by the US to request information on India’s adherence to ILO core labour standards, for example, although the Member State under review is under no formal obligation to report on labour standards (WTO 1998, paras. 68–70). In addition, on a day-to-day and informal basis, the director-generals of the two organizations are in contact by phone, and ‘ILO–WTO Secretariat meetings happen all the time at an informal level’ (Interview 46). From the WTO perspective, the ILO should be encouraged to develop its initiatives promoting core labour standards, for ‘the more the ILO develops an effective agenda on labour the more we [that is the WTO] are off the hook’ (Interview 46). As ILO oversight of labour standards becomes more effective, the negotiating burden on the WTO may be lightened. The 1998 ILO Declaration on Fundamental Principles and Rights at Work, the success of the ratification initiative, and policies to encourage implementation of the Declaration (notably on child labour) are signs that this may be happening. Given the impasse on WTO–ILO cooperation, the new frontier in trade and labour standards is at the sub-global level. The crystallization of ILO core labour standards in the 1998 Declaration enables their incorporation into regional and national agreements. At the EU level, ILO core labour standards have been incorporated into the new generalized system of tariff preferences (GSP) for 2002–4 by a qualified majority vote in the General Affairs Council of 10 December 2001 with only Portugal voting against.20 The GSP allows market access at a rate even more favourable than that guaranteed by most-favoured-nation treatment on a range of goods from developing countries. Any developing country may apply for special GSP ‘incentives by ensuring that its national law incorporates the substantive content of the standards laid down in the 8 ILO Conventions on fundamental principles and rights at work’. This GSP measure allows for the ‘possibility of withdrawing preferences under the GSP in the event of serious and systemic violations
182 European Welfare States
of ILO standards’ at ‘the end of a procedure which, inter alia, grants the country concerned the opportunity to avoid such withdrawal by undertaking to adopt the measures necessary to comply with the standards’ (Council 2001c, p. 12). Regional mechanisms for integrating core ILO labour standards are not uncontested, however. The EU’s integration of the promotion of ILO standards as one of the avenues for qualification for GSP incentive measures, as well as special incentives for combating drug trafficking (Council 2001e), resulted in a request for consultations by Thailand and India at the WTO in 2001.21 Without a dispute settlement forum for non-EU members provided in the EU’s institutions, developing country governments took their complaints about the EU’s GSP regime to the global level. In 2002 the first step in a possible formal complaint to the WTO Dispute Settlement Body was taken (WTO 2001a; WTO 2002a). Specifically, India argued that GATT/WTO rules only permit setting qualifications for GSP that ‘facilitate and promote the trade of developing countries’ and that ‘respond positively to the development, financial, and trade needs of developing countries’, and that EU GSP tariff preferences that introduce ILO core labour standards and other forms of differentiation ‘cannot be reconciled with these two requirements’ and amount to extra-GSP ‘discrimination’ (WTO 2002a). After consultations, India decided to limit its formal complaint to GSP schemes granted by the EU under its Drug Arrangements (WTO 2002b). These provide preferential market access for developing countries fighting the war on drugs, and many countries benefit from Drug Arrangements, unlike special incentive arrangements for the protection and promotion of workers’ rights under which only Moldova has been granted preferences (although others have applied). The application of ILO core labour standards as a means of access to the EU’s GSP incentives were thus not part of the WTO panel report, nor the subsequent appeal that eventually ruled, among other issues, that objective criteria that have the effect of differentiation in GSP schemes (including, potentially, the promotion of ILO core labour standards) are permissible under WTO rules (WTO 2004a). That said, the WTO ruled that current EU Drug Arrangements do not fully comply with these criteria and must be brought into conformity with WTO rules no later than 1 July 2005 (WTO 2004b). Still, it can be expected that tensions between regional attempts to expand and enforce labour standards, and global-level uncertainty regarding the interaction and enforcement of international agreements in their various legal forms, will persist. India has made it clear that it may launch a new request for consultations at the WTO when the current EU GSP
Global Governance of Social Policy 183
scheme expires at the end of 2004, should different criteria for developingcountry qualification for preferential market access persist.
Conclusion: global governance and the EU This chapter has demonstrated that some of Mitrany’s fears about the impact of a European Union have indeed come to pass. Since the 1970s, EU legislation has surpassed that of the ILO in importance to EU Member States, with the effect that national contributions at the ILO are in part determined by the existence of regional legislation. In addition, ILO legislation has been struck down in cases where it is incompatible with EU legislation. Regional governance, in a legal sense, has proved much more robust than global governance. Still, where EU and ILO legislation coexist there is a complicated legal tangle that ‘includes questions of international, community, as well as national constitutional law’ (Betten 1993b, p. 53). These questions have not been resolved in the 10 years since this assessment was made. A review of fundamental rights in the EU noted that the role and impact of ILO Conventions ‘remain uncertain’ (Commission 1999b). Given the trend of decreasing ratification of ILO Conventions by EU Member States (Table 5.3), it seems that the emphasis has shifted from global to regional standard setting in social policy. Part of the tension between EU and ILO legislation results from the position of the Commission at the ILO as a non-voting observer. Forced, ultimately, to stand back and watch Member States decide whether to ratify Conventions, or not, and in light of weak ILO enforcement mechanisms, the Commission has continued to press ahead with its own agenda of social policy legislation. The issue of EU competency and day-to-day leadership has been much more effectively resolved in the case of the WTO, where political and ECJ decisions have resolved uncertainty over competency on new issues. In contrast, issues continue to be managed on a case-by-case basis at the ILO. The informal state of relations is complicated by the expanding agenda of the EU in the social field in each successive Treaty revision, and new forms of regulation, such as the OMC, that sit at an uncertain divide between Member State and regional authority. Social partner objections to Commission leadership, and national sensitivity about Commission leadership, indicate that a resolution of competency should not be expected. The uncertain and awkward divide between EU and ILO legislation will continue. The Governing Body of the ILO and its Legal Committee, as well as International Labour Office
184 European Welfare States
officials, have shown great resolve in consistently stating that the relationship of EU Member States at the ILO is up to them to solve. On the normative side, regional and global governance have proved more compatible. The concrete policy activity of the early days of the ECSC and the EEC were inspired directly by the ILO, and the goals of EU policy-making hold much in common with the Preamble of the ILO constitution. Activity in the areas of labour mobility, worker retraining, social security and health and safety were all connected to the ILO agenda in their earliest stages. The development of worker rights, equality between men and women, and non-discrimination in the workplace in EU Treaties also have resonance with the earliest goals of the ILO and the 1998 Declaration. In the ILO’s view, the EU has promoted many of its core values: respect for core worker rights, an ever-increasing level of standards in social policy, equality among men and women, nondiscrimination in all its forms, and social dialogue (ILO 1988b, p. 8). In the area of core standards, the ILO has ‘reinvented itself as the primary and unique universal institution’ (Murray 2001, p. 219). The EU supports the ILO in this effort, no doubt hoping to prove that it is not unilaterally applying core labour standards as a protectionist measure. Still, tensions between regional and global levels of governance persist when the EU pushes for an agreement that is at a much higher level than other member of the ILO can accept. The most critical observer notes that: the EU destroys this place [the ILO] by creating a bloc position that is very difficult to move. Developing countries don’t really like it, and in the end, ratifying ILO conventions does not seem to be a priority of EU Member States. (Interview 53) In the words of another ILO representative: The danger must be avoided of falling into the temptation – which experience has shown to be not imaginary – of shifting from a concern for compatibility to an insistence on conformity, by projecting Community solutions onto the international level, with all the undesirable consequences which this may entail. (ILO 1994, para. 5) As the representative of the government of Denmark expressed it: ‘too many recent Conventions have been modelled on Community standards so that no one could ratify them, not even the EC States’ (ILO 1993b, p. 3). This is a great problem for the ILO’s mandate to create universally applicable standards in social policy.
Global Governance of Social Policy 185
Certainly it would make governance in the EU more coherent to have regional positions on issues replicated at the global level, but this would clearly make it more difficult for states to come to agreement in international fora. There is some evidence to indicate that at the global level, where the goal is to assert general principles and develop minimum or framework legislation, the presence of a bloc of welldeveloped nations having worked out a highly detailed negotiating position makes it more difficult to come to agreement with other states and may threaten effective compromise. The desire for coherence in EU decision-making thus meets the issue of finding effective compromise at the multilateral level head-on, and it is not clear that these are compatible goals. At present, the loose association of the EU Member States at the ILO, while certainly contrasting with their working at the WTO, provides a messy but pragmatic solution to the difficult challenge presented to EU Member States operating in a global forum. More formal EU coordination at the ILO might help to solve the problem of coherence, but the challenge of compatibility between regional and global forms of governance in social policy would remain.
6 Conclusion
This concluding chapter has three main goals. The first is to assess the creation of supranational forms of governance at the regional and global levels. By drawing upon the observations provided in individual chapters, conclusions will be made about the conditions under which supranational governance emerges. Having assessed different forms of governance, we are now able to make some comparisons between these various forms, and suggest how they interact. Each new form of supranational governance has empowered a different range of actors, and results in different policy outputs. What are the implications of increasingly diverse, increasingly flexible forms of integration on traditional forms of governance by law? And what implications do regional forms of governance have for the global level? The second goal of the concluding chapter is to reassess the argument presented in chapter one. What happens to leader, resistant and passive Member States over time? Do they continue to exhibit similar bargaining positions, or does EU cooperation shift their positions, and perhaps even underlying preferences, over time? Although Member State positions are shown to shift during bargaining, over the past 20 years Member State preferences have remained remarkably stable. This suggests the resilience of national welfare state regimes despite the effects of deep regional economic integration and globalization. However, it is clear that supranational governance has a variety of implications for national welfare state regimes. The transforming effects of supranational governance are the greatest where national welfare state regimes are weak, or underdeveloped, but policy transformation as a result of supranational governance takes place even in the most advanced Member States.
186
Conclusion 187
The third goal of this chapter is to assess the point to which supranational governance has evolved in social policy, and suggest where it might go in the next decade. In what sub-issue areas of social policy is supranational governance most evolved, and why? What issues are not covered, and might they be in future? The EU enlarged in 2004 and now includes 25 Member States, and will likely enlarge several more times in the coming decade.1 What implications will enlargement, and the resulting increase in the diversity of welfare state regimes, have for regional governance of social policy? The conclusions presented here are tentative, but are grounded in the empirical observations made in this study.
Supranational forms of governance compared Conditions under which supranational forms of governance emerge Chapter 1 suggested three explanatory variables for the creation of supranational forms of governance: the diversity of welfare state regimes, the ways in which institutions present opportunities and incentives for cooperation, and government ideology. The purpose of selecting several case studies of supranational governance in different sub-issue areas was to draw out generalizations on the relative importance of these causal variables, and the relative importance of national, supranational, and non-state actors in the creation of new forms of governance. The evidence indicates that the relative influence of national policy histories and institutional factors changes depending on the stage of policy-making that is being analysed. At different stages of the policymaking process the causal variables shift in importance. Still, it is clear that national and institutional factors are at each stage necessary to account for outcomes. At the preference-formation stage, national welfare state explanations are at the height of their explanatory power. In each of the three issue areas examined at the EU level, national policy histories were the most salient factor determining the policy leaders. In the case of Denmark and Sweden especially, robust national welfare states and a longstanding commitment to government oversight of social and employment policy are a more important factor than government ideology in determining preferences for supranational cooperation; despite being some of the most anti-integrationist states, Denmark and Sweden have been consistent policy leaders by introducing QMV in health and safety policy, creating a high standard of health and safety legislation and introducing
188 European Welfare States
cooperation in employment policy. At the preference-formation stage, government ideology must also be considered. Ideology is most relevant in explaining the UK’s resistance to supranational social policy cooperation and, in particular, the change in the UK’s position on the Social Protocol in 1997 under the Blair Labour government. At the bargaining stage, national preferences must be joined by institutional factors in order to explain bargaining positions and outcomes. Leader Member States pushed for QMV in health and safety, and developed the draft chapter on Employment Policy, while Commission entrepreneurship was essential to the creation of the Social Dialogue and the Social Protocol at Maastricht. Both institutional factors and national bargaining positions must be considered to explain the exact features of new forms of governance. Policy learning over time has caused Member States to seek ways to reduce implementation costs, and to limit the cession of national authority to the regional level. The UK is especially effective in setting limits to cooperation during moments of Treaty bargaining, such as the splitting of art. 118 into sections permitting QMV and those where unanimity would continue, placing limits on the financial impact of legislation, and inserting labour market guidelines into the EES. Germany, too, has placed explicit limits on the financial implications of new forms of governance, suggesting the heightened influence of large Member States at moments of high politics. Institutional factors take on their greatest salience at the bargaining stage when accounting for the agreement of passive Member States. Financial incentives, in the form of funds to aid social development and cohesion, are necessary to explain the agreement of Greece to the SEA. The linkage between financial incentives and advances in social policy is unclear in the case of Maastricht, especially given the complexity of this Treaty bargain that included EMU. In the case of the Amsterdam Treaty, financial incentives were not apparent, although ESF funding continues to be aimed at unemployment. Rather, the development of more flexible forms of integration that do not cede competence to the EU, the relaxation of enforcement mechanisms, and the interest in policy emulation explain the agreement of southern Member States at Amsterdam. At times of Treaty bargaining, non-state actors have been essential to the outcome only at Maastricht. The enhancement of the Social Dialogue as a new form of governance was owing to consistent entrepreneurship by the Commission, and to the agreement of the EU-level social partners to hammer out a common text. Policy learning suggested to the social partners, especially employers, that if they did not step into a regulatory role,
Conclusion 189
the Commission was sure to continue with new legislation that stood an increased chance of passing in the Council under QMV. As a result of the competence provided to them by the Treaty, non-state actors are the key to governance by the Social Dialogue. At the stage of specific policy-making (or policy design), institutional factors are the most salient. Commission entrepreneurship is especially important to explain the creation of a health and safety Framework Directive well above the lowest common denominator. Commission entrepreneurship was also vital to the evolution of the EES, along with the pro-integrationist Luxembourg presidency. Decision-making in committees isolates the most resistant Member States, forcing them to concentrate on a small number of issues that are of greatest importance to them while the general level of the policy output rises. Passive Member States are also isolated during policy design, owing to poor preparation of officials and lack of national expertise. At the implementation stage, Member State policy histories again exhibit high explanatory power, especially in the case of less-developed Member States. Weak and underdeveloped national policies are easily replaced by the transposition of EU legislation, but the effective application of EU legislation is much more difficult. Commission oversight and the possibility of ECJ sanctions encourage Member States to evaluate constantly the implementation of EU policy. Given the weaker enforcement mechanisms in the OMC, we may expect that application will also be weaker in employment policy unless greater linkages are created to other forms of EU governance with enforceable provisions, either through case law or other agreements with legal standing such as the Stability and Growth Pact (although its enforceability is also uncertain). Policy resisters also reassert their reluctance to cede regulatory authority to the supranational level at the implementation stage. The UK has used the ambiguity of language in Directives to limit the application of EU law, as in the case of the Directive on part-time work. It is also questionable whether UK procedures for worker participation in health and safety conform to the Directive. In the EES, the UK has focused on the entrepreneurship pillar, demonstrating its interest in specific subsets of policy cooperation under the OMC. Resistance is not simply limited, then, to the bargaining phase, but may re-emerge at other stages of the policy process. An examination of the conditions under which supranational governance emerges at the regional level suggests that successive forms of governance were approved in order to provide more flexibility, in terms
190 European Welfare States
of national implementation instruments, and to reduce implementation costs while permitting cooperation to expand. A comparison of the forms of EU governance draws out the increasing flexibility of new forms, and suggests their impact on one another. EU governance in the social policy field: three forms of governance Each form of regional governance empowers different actors, requires different inputs from these actors, and results in unique policy outputs. Table 6.1 summarizes the emergence of the three forms of supranational governance: governance by law, governance by collective bargaining via the Social Dialogue, and governance by the open method of coordination. The table suggests that there are three periods of social policy development in the EU, each interlinked with the forms of governance available to policy-makers: a declaratory phase when governance by law approved by unanimity voting was the only form of governance and outcomes were minimal, a period of gradual strengthening of regional social policy associated with the extension of QMV to health and safety, and a period of substantive politics linked to QMV on a wider (but still limited) number of issues, governance by Social Dialogue, and the OMC. Having traced in detail the emergence of these forms of governance, we may ask, how do these different forms interrelate and affect one another? First, the research suggests that each form of governance inspired, in some way, the creation of other forms in order to overcome past policy failures and the reluctance of some Member States to agree to cooperation. Policy failures (in other words, the inability to successfully agree to proposed legislation) were especially salient in the proposal of QMV in health and safety policy so as to prevent the UK, and possibly southern Member States, from vetoing cooperation. Policy failures were also part of the history behind the Commission’s proposal of the Social Dialogue; when several issues were stymied by national vetoes, the Commission turned to the EU social partners as an alternative route to regulation. While resistant member states may have banked on the CBI trying to play a veto role similar to the UK government in the Social Dialogue, other national social partners asserted their interest in producing legislation via this new form of governance. To some extent, the CBI played the veto role in the failed Social Dialogue negotiations on European Works Councils, and yet the subsequent proposal of EU legislation by the Commission using the traditional ‘Community Method’ (and approved by QMV in the absence of the UK in Directive 94/45/EC of 22 September 1994) showed that this veto was not absolute.
Unanimity.
Low (constrained by Council).
Activism of the Commission in the the social policy field
1986 SEA: art. 118a QMV on health and safety; art. 118b concept of Social Dialogue introduced.
Title III, arts. 48–51 free movement of workers. Article 51 social security provisions. Part III Social Policy arts. 117–28.
Medium 1972–84. High 1985–90 (Delors).
1986 SEA: start of QMV in health and safety legislation (art. 118A).
1989: Social Charter.
1972 Paris Summit Declaration. 1974 Social Action Plan.
Article 3: free movement.
1972–93: gradual breakthrough
Treaty of Rome: Preamble.
Voting rules in Council
Treaty provisions and major initiatives
1957–71: declaratory politics
Table 6.1 Evolution of regional governance in social policy
High 1993 (Delors). Medium 1993–2004.
continued
(1) Governance by law: QMV is extended in each treaty revision. (2) Governance by Social Dialogue: Council approves Framework Agreement according to voting rules of art. 137. (3) Governance by coordination (OMC): Council helps develops guidelines, approves NAPs, approves Commission recommendations, writes Joint Employment Report.
2004 proposed Constitution includes previous provisions; Part I art. 3 ‘social market economy’; Part II Charter of Fundamental Rights incl. labour standards.
2002 Nice, included OMC as a working method for social exclusion and other issues.
1999 Amsterdam Title VIII Employment.
1993 Maastricht Protocol on Social Policy, Social Dialogue as new form of governance (now art. 139).
1993–2004: substantive but limited politics
191
Consultative role. EP not directly elected.
ECJ.
Consultative, through European Social Committee (ESC).
Creation of customs union: limited social policy harmonization for gender equality, holidays. Creation of ESF. 1961 Social Charter at Turin. 1966 Veldkamp. Memorandum. Gender equality legislation.
Enforcement mechanisms
Role of the social partners
Relation of social policy to economic policies
Other important EUlevel initiatives
1957–71: declaratory politics
Role of the European Parliament in social policy-making
Table 6.1 continued
1974 Social Action Plan; 1977 77/187/EEC on worker consultation; 1985 Val Duchesse Dialogue.
SEM: QMV in social policy only extended to health and safety policies.Cohesion funding to aid adjustment of least-developed Member States.
Consultative, through European Social Committee and through art. 118b on social dialogue.
(1) Governance by law: ECJ can sanction member states with fines, and backdate application of legislation.
EP directly elected from 1979. 1986 SEA introduced cooperation procedure under QMV; consultation when unanimity voting in Council.
1972–93: gradual breakthrough
1989 Health and Safety Framework Directive; 1994 Cohesion Funds; 1995 first social partner agreement.
EMU: Social Dialogue promoted by Commission alongside EMU. Employment Title promoted at Amsterdam by non-EMU states. Increasing relation between Broad Economic Policy Guidelines (BEPGs) and EES.
(1) Governance by law: consultative. (2) Governance by Social Dialogue: legislative role through EU-level collective bargaining. (3) Governance by coordination (OMC): consultative, may also propose new guidelines.
(2)Governance by Social Dialogue: ECJ. (3) Governance by coordination: naming and shaming.
(1) Governance by law: co-operation or co-decision for QMV; consultation for unanimity voting. (2) Governance by Social Dialogue: opinion. (3) Governance by coordination (OMC): consultation.
1993–2004: substantive but limited politics 192
Conclusion 193
In addition, while the CBI was able to drop out of the Social Dialogue pre-1997, UK-owned firms operating in the rest of the EU were still bound by the outputs of Social Dialogue. QMV rules were eventually adopted by UNICE and ETUC to ensure that no one national social partner could hold its confederation ‘hostage’ with a veto. In the case of employment policy, leader Member States were not interested in giving the EU exclusive competence, nor could they expect that resistant Member States would agree to using governance by law in employment policy. A form of governance with limited competence implications, and no legislative or financial sanctions, was thus proposed. There are other, more complex ways, that these forms of governance affect one another given their simultaneous activities. Two main effects should be mentioned. The first is that Social Dialogue has empowered national social partners at the regional level of governance, and has suggested their use in other policy areas for both functional – technical and legitimacy reasons. Social partners had participated in the Economic and Social Committee since 1958, and as part of health and safety advisory committees at the EU level since 1978. Social Dialogue, however, has particularly affected the EES. Although not initially included in the Employment Title, the social partners are to be consulted on issues of lifelong learning, and they may propose new guidelines. Second, the use of the OMC as a method in employment policy has suggested its use elsewhere. The use of OMC is to be applied to pensions and social exclusion (EU 2000a). Germany, along with Britain and the Netherlands, originally rejected the Commission programme to fight against social exclusion in 1994. In 1995 a draft Directive on social exclusion was rejected, again by Germany, on the basis of subsidiarity and an ‘alleged misuse of the legal basis of art. 235’ (Rhodes 1999, p. 143). The experience with the EES may have demonstrated that cooperation using soft law modes of governance ‘creates no legal rights or obligations for legal or natural persons in the Member States, but is merely … a tool for promotion of cooperation of various policies within the Community’ (Bruun 2001, p. 310). In this sense, OMC returns to the cooperative method of governance suggested in the Treaty of Rome (EEC 1957). The OMC may also be applied to issue areas where other forms of governance have been used. Health and safety policy is one example where governance by law is the norm, but where OMC may be increasingly applied. The European Agency for Safety and Health at Work was founded in 1994, and works according to principles that closely resemble the OMC.2 The Agency gathers information on occupational health and
194 European Welfare States
safety in order ‘to identify and validate examples of good practical solutions’, and to allow Member States ‘to learn and improve from each other’s experiences’ (Agency 2001, pp. 4–5, 20). Such activity is in danger of shifting governance in the health and safety issue area ‘from mainly regulatory policy-making into mainly persuasive policy-making’ (Smismans 2001, p. 88), or as Wolfgang Streeck predicted in 1995, from law towards ‘voluntarism’ (Streeck 1995, pp. 430–1). The key trend that must be analysed in the coming years is whether the success of the OMC results in a declining use of governance by law and governance by Social Dialogue, both of which result in hard law outputs. Can the OMC be applied fruitfully to areas where Member States have expressed a desire for the creation of common standards, or does hard law remain the necessary governance method? It may be suggested that the three European forms of governance embody different ideas about social policy cooperation, and at times may not so much complement as compete with one another, much as EU and ILO legislation may sometimes compete for similar regulatory space. The creation of governance by law reflects the interests of highstandard Member States who wish to export their national policies to low-standard states so as to prevent policy competition. The Social Dialogue embodies the Commission’s desire to see social policy evolve alongside economic integration, and is an endorsement of the socialdemocratic and conservative-corporatist traditions of consulting social partners in national decision-making. Finally, the EES and the OMC allow states to inject a range of values into policy-making, such as an emphasis on full employment, active labour market policies, worker training and gender equality from the social-democratic states, and labour market flexibility from liberal states. Especially in the case of the OMC, where a range of guidelines is acted upon, there may be tension between the values embodied in the guidelines. Such tensions are to be expected in a complex, multi-member polity. Regional and global governance: the challenges of occupying similar regulatory space The above section compared three forms of regional governance. Global forms of governance combine elements of all three forms – governance by law, governance by Social Dialogue, and governance by coordination – in a unique combination that reflects the institutional characteristics of the ILO. The ILO has, as its main goal, policy coordination among a diverse number of states, and suggests the emulation of best practice, much as the OMC does. Policies are developed by tripartite discussions, not unlike
Conclusion 195
the Social Dialogue. The policy outputs are often Conventions, ratified by states, which form part of national labour law once ratified. In terms of outputs, the ILO replicates governance by law, although the choice of which laws apply is determined by individual members rather than possessing a unified code or system. However, whereas Member States and the ECJ enforce regional governance by law, the ILO possesses only weak enforcement capabilities. Global governance is much more dependent on states than regional governance for the effective oversight of its policies. As explained by a former Assistant Director-General of the ILO, ‘the greatest difficulty in regards to a system of supervision [at the ILO] founded on documents [provided by Member States] is to be able to assess the effective application of Conventions in practice and not merely legislative conformity’ (Valticos 1982, p. 373). In this sense, ILO policies are more akin to governance by the OMC, which relies on a national system of reporting and on the ‘naming and shaming’ of states to encourage implementation. As demonstrated in Chapter 5, there are both positive and negative consequences of the EU and the ILO ‘occupying much the same regulatory space’ (Murray 2001, p. vii). Where the EU and the ILO occupy similar regulatory space in the case of minimum standards, EU legislation will effectively trounce ILO Conventions; in this sense, regional governance has more impact on national policies than global governance. On normative issues, though, the two institutions are highly compatible. Global governance could be said to fill a void left by regional governance in the case of core labour standards, which the EU can regulate only for its own members. Given its global membership, the ILO permits the development of internationally agreed labour standards that aim to provide a floor of worker rights. The results, if properly applied, are social justice and the prevention of competition among states on the basis of basic labour rights. When comparing systems of global governance, however, it is clear that labour legislation still does not enjoy the same robust consultation and enforcement mechanisms as trade policy. This is for a host of reasons, ranging from the legal structure of the ILO (Conventions ratified on an individual national basis as opposed to a single body of law ratified by all members), to the preferences of members and the lack of a core group of states that would agitate for such enforcement mechanisms to be created and provide the necessary funding. Alternatively, enforcement mechanisms based on fines and/or sanctions and compensation for injured parties may not be the appropriate way to encourage the implementation of basic labour rights. In this case, a robust series of
196 European Welfare States
oversight mechanisms, such as formal reviews of national labour systems and a system for complaints, and the availability of technical assistance and grants to improve the implementation of labour rights, must be encouraged.
Leaders, resisters and passive Member States in supranational governance The previous section has explored how the various forms of supranational governance interact with one another. But what of the argument presented in Chapter 1, suggesting the importance of national welfare state regimes in predicting preferences for supranational governance, and the existence of leader, resistant and passive Member States in the bargaining and policy-making process? Have these categories of states retained their validity throughout the case studies? Supranational governance and a welfare state explanation of preferences Research into social policy cooperation over time at the EU level reveals both the strengths and weaknesses of a welfare state explanation. Such an explanation is very good at identifying policy leaders, especially among the social-democratic states such as Denmark and Sweden, and at suggesting that conservative-corporatist states, such as Belgium, will also often be among policy leaders. The explanation also identifies the Member State most resistant to EU cooperation in the social policy field, the UK. At the two extremes of Member State opinion on social cooperation – leaders and resisters – a welfare state explanation is highly relevant. An investigation of social policy cooperation also reveals gaps between the expected behaviour of Member States based on welfare state regimes and their actual behaviour during both moments of intensive intergovernmental bargaining on Treaty reform, and in instances of specific policy-making. An explanation based solely on welfare state regimes does not explain the middle range of Member States that act neither as explicit policy leaders nor as resisters. Southern states do not resist policy, as would be expected, but rather are best typified as passive; Greece, Italy, Portugal and Spain may not propose new forms of cooperation or lead the building of coalitions to approve policy, nor do they actively oppose an expansion of cooperative efforts in social policy. Instead, they often join coalitions of Member States willing to create new avenues for policy-making, and agree to specific EU-level policy outputs.
Conclusion 197
A welfare state analysis of underlying preferences also fails to explain why France and Germany have shown such inconsistent leadership in the social policy field, despite being characterized as the motor of European integration (Szukala and Wessels 1997; Webber 1999). Even though they possess conservative-corporatist welfare state regimes (although Germany is a much more classic example than France), these two large Member States have not been the driving force behind social policy. Given the findings of Danish and Swedish leadership in the social policy field, despite their relatively low interest in encouraging deeper EU integration writ large (most notably in monetary policy), studies involving specific policy areas are valuable in pinpointing how Member States shift in their relative influence and leadership depending on the issue area. Member States may colonize certain issue areas that they identify as closest to their national priorities, and concentrate their resources, such as expertise, Council Presidency activity, and bargaining power, in these issue areas. Member States may behave against ‘typecasting’ in these issue areas. The behaviour of the UK in promoting the single market and a European defence policy may be another example of a Member State behaving against ‘typecasting’. Such behaviour may also help to explain the passivity of southern Member States. With limited administrative resources available to them, states such as Greece, Portugal and Spain may concentrate on issue areas such as the Structural Funds, fishing policy, CAP and EMU rather than dedicate finite government resources to resisting social policy. Member States exhibit remarkably stable positions with regard to social policy cooperation in the European Union over the period 1985–2000. In both the macro-level Treaty changes and micro-level policy-making, most Member States act consistently across sub-issue areas of social policy. Exceptions are certainly to be found, such as the suggestion of a ‘risk assessment philosophy’ that followed the ‘innovative Scandinavian health concept and risk assessment approach’ in the Machinery Directive by the UK expert to the Health and Safety Advisory Committee, which has been explained as a national official using the EU level to overcome resistance in his own Member State (Eichener 1992). Ireland presents another puzzle, since given its liberal welfare state regime it should be expected to resist social policy and yet, in a number of cases since 1991, it has favoured cooperation. The problem may rest less with Irish national policy histories and institutional factors than with the typology of welfare states (Esping-Andersen 1990). Ireland sits at an awkward divide between a traditional liberal welfare state
198 European Welfare States
system of benefits and a corporatist organization of the labour market. In most cases, however, Member States act in accordance with their national policy histories and in reaction to the incentives provided by institution-based cooperation. The consistent behaviour of Member States is also of interest in isolating those cases where ideology is an explanatory factor. Governments of the left and right have not proved salient in cases such as Greece and Ireland. Governments of the right in Belgium, Denmark and Luxembourg have repeatedly supported social policy expansion. The conservative German government of the period 1986–97 was led by Kohl, and, as was shown in Chapter 3 on the Social Dialogue and the Social Protocol, his leadership in the final Treaty bargain was decisive in expanding social policy. Ideology is most relevant in the case of the UK, for the UK opt-in to the Social Chapter can only be explained if the regime change from Conservative to Labour in 1997 is considered. However, even under a Labour government, the UK has demonstrated policy resistance, such as its active work to limit the application of the part-time work Directive and its reluctance to apply the parental leave Directive. Ideology may also be relevant in the Spanish case, where a change in government from left to right in May 1996 resulted in greater reluctance to cooperate in employment policy. Still, ideology as embodied by governments of the left and right does not effectively predict government bargaining positions in the social policy field. The consistent behaviour of Member States across time, issue area and government may be the strongest indicator of both the resilience of national welfare states despite conditions of regional and global integration, and the ability of institutional incentives to structure preferences towards cooperation in the case of passive Member States. It is to these issues that we now turn. Supranational governance as welfare state transformation Supranational governance has the greatest impact on states where national policy histories are underdeveloped. Careful process-tracing reveals that in the case of Greece, supranational governance has been the primary influence in setting new minimum standards for health and safety, re-establishing social dialogue as a form of governance at the national level, and has led to the introduction of ALMPs. The case of Greece also indicates that regional standards have surpassed global standards in their influence; in health and safety policy, where ILO standards were once taken as the policy standard, EU standards now serve this purpose (Interview 76), and where the ILO encouraged the
Conclusion 199
preservation of trade unions during the dictatorship, the EU now encourages them to engage actively in policy-making. These observations suggest that for Member States with underdeveloped welfare states, the EU may in fact act to transform the welfare state. The case should not be over-stated, for benefit levels still remain relatively low in Greece, and many services, especially for women and the unemployed, are absent from the Greek welfare model. However, given that research often fails to focus on southern Member States, such as Greece and Portugal in particular, the transformative effects of EU policy are underrated in present literature. As one Greek politician noted: ‘Greece is modernizing and this [impetus] is coming from outside’ (Interview 1). Marks and Hooghe are therefore incorrect when they state: collective decision-making among states involves a significant loss of control for individual state executives … Decisions concerning rules to be enforced across the EU (for example harmonizing regulation of product standards, labour conditions, etc.) have a zero-sum character, and necessarily involve gains or losses for individual states. (Marks and Hooghe 1996, p. 346) In fact, in the case of social regulation, regional-level policies may empower the Greek government to modernize its welfare state while also permitting high-standard welfare states to export their preferences to the Union. Research on south-eastern European countries indicates the transformative effects of EU trade, defence, and justice and home affairs policies as transmitted through EU aid policies and the Stabilization and Association Agreements ( Johnson 2001). Weak states may build entire sections of national administrative and legal regimes, such as in competition policy, trade law and investment, around EU practices. Further research on the transformative effects of EU social policy in the southern Member States, and in Central Europe, would bolster this tentative claim. Even in high standard states, supranational governance requires adjustment. In the case of Sweden, EU Membership has required some changes to Swedish legislation. For example, the 1991 Directive on informing the employee of the terms which apply for employment (91/533/EEC) was integrated in the new 1994 Act on Employment Protection, and the working time Directive (93/104/EC) contained requirements that were more comprehensive in scope than Swedish legislation, which permitted exceptions in negotiated collective agreements (Sigeman 2000,
200 European Welfare States
pp. 156, 159). The Framework Agreement on Parental Leave also required changes to Swedish legislation to permit time off both before and after birth or adoption (96/34/EC). Governance by law, rather than by national collective agreement, also ‘implies a clear break with tradition within the regulation of Swedish working life’ (Sigeman 2000, p. 157). Even for high-standard Sweden, joining the EU has had transformative effects, although they have not been as extensive as those in the Greek case. Resilience of diverse national welfare state regimes Specific transforming effects of supranational governance have been identified. However, a contrary trend is also evident. Research suggests that while supranational governance has resulted in convergence in Member States in specific sub-issue areas of social policy, national welfare state institutions and labour market characteristics are still highly resilient under conditions of both European economic integration and globalization. Despite a willingness to cooperate – itself a product of diverse reasons – states bring with them national welfare state regimes that are difficult to change in a short period of time. The resilience of national models is also explained by the limited scope of supranational policies, the flexible forms supranational cooperation takes, the lack of financial resources at the EU that would permit redistributive projects, and weak implementation and enforcement mechanisms. EU policy in particular continues to be concentrated in social regulation, rather than policies involving redistribution; EU policy focuses on labour markets and not the major areas in which states spend their money on social policy, namely pensions, unemployment benefits and health. One general observation of this book is that spillover from economic integration to social policy at supranational levels has been highly limited.3 Under conditions of economic integration at the regional and global levels, social policy integration does not automatically follow despite myriad pressures and logics for integration. The lack of enforcement capacity in the OMC, and in the ILO at the global level, is an additional factor that enables the national models to persist. Without any sanctions other than naming and shaming, these forms of governance provide few incentives to encourage, or force, Member States along the path to policy convergence. The influence of supranational policy is, ultimately, ‘highly dependent upon domestically existing structures of governance, interest organization, and their attendant normative conceptions’ (Lyberaki and Tsakalotos 2000, p. 207). There is a danger of being too optimistic about
Conclusion 201
the impact of European integration on convergence in the face of deeply rooted national cultures and highly complex forms of policy administration (Petmesidou 2000, p. 325). Historical institutionalism, although applied most robustly to the regional level, suggests that as a complex set of rules, interests and policies, national welfare states are highly path-dependent and given to slow and incremental change over time. The setting of Lisbon-type criteria and benchmarking in the OMC may, in fact, be the best way of measuring over longer time periods how successful EU policies are in encouraging change and convergence to common European standards.
What’s in and what’s out: the evolution of supranational governance in social policy Two final observations suggest how supranational governance in social policy may evolve in the next decade. The first observation concerns Treaty competence, and the ways in which complex forms of governance constantly test the formal boundaries Member States have placed on cooperation. The second observation concerns the effects of EU enlargement on national preferences for social policy cooperation at the regional level. What effects will an enlarged EU, including 25 Member States, have on regional governance of social policy? On the first issue of supranational competence, the experience of the ILO and the EU suggests that governance begins in technical areas, such as health and safety, where opposition on the part of national actors and governments is often minimal. Highly technical areas of cooperation may prove the most fruitful in initial stages of integration. At the ILO level, social policy cooperation has evolved across a wide range of social policies. The creation of Conventions that cover a wide range of issues centred on work and employment, but that also cover benefits and equality issues, is the result of two factors: one, it is up to states to ratify these Conventions, so their production at the global level does not automatically bind members to apply them; two, there are no financial sanctions on states who do not apply ratified Conventions. One area where the ILO has had recent success is in the promotion of a set of Conventions as core labour rights. The global membership of the ILO suggests that it is the proper institution for the articulation and monitoring of international labour standards. It is clear, however, that the WTO membership, operating on consensus-based principles and having many members from low-wage and low-standard countries, will not endorse a linkage between ILO standards and trade policy in the
202 European Welfare States
next decade in the absence of fundamental institutional change. The experience of the EU suggests that making funding available for the improvement of labour standards is one way in which resistant Member States may be encouraged to cooperate. As regional governance has evolved, it has expanded most quickly in areas closely related to the labour market, or in those policies falling under the category of social regulation. This indicates that the EU is, at present, an underdeveloped system of governance in social policy if one believes that the remit of national social policies should be replicated at the EU level. The Treaties continue to put explicit limits on social policy cooperation at the regional level. Article 137 states: ‘The provisions of this article shall not apply to pay, the right of association, the right to strike or the right to impose lock-outs’ (EC 1999, art. 137(6)). These explicit exceptions are also replicated in the proposed Constitution of 2004 (EU 2004). Importantly, the core labour rights as defined by the ILO must be implemented by national legislation (ILO 1998). This is because the EU is not a member of the ILO, and EU-level law does not cover one of the core labour standards, the right to strike. Although coordination of social security is possible under art. 137, governance by law in this field is limited by the application of unanimity voting that gives each Member State a veto on any draft legislation. The limited financial resources of the EU suggest that the convergence, or harmonization, of benefits to the elderly, poor, unemployed, disabled, and other groups is also effectively excluded from cooperation. Even in those issue areas where cooperation does occur, such as health and safety, benefit levels are not the subject of legislation. There is, for example, no common benefit level for those injured at work even if they sustain identical injuries; a severed limb does not result in equal compensation in France and Greece, for example. There is also no equal access to services; the EU, for example, does not guarantee the right of workers to training schemes. It is vital to note, however, that supranational competencies tend to evolve slowly, and that, as was the case in all three sub-issue areas examined at the EU level, Treaty competence emerged only after a period of informal or non-Treaty based cooperation. EU Treaty competence does not result in cooperation, but is itself the result of several years of activity. Given the evolution of policy over time, it is possible to note the issues to which regional governance may expand despite their present exclusion from the consolidated version of the Treaties. The first is a set of issues that will be governed at the regional level by the open method
Conclusion 203
of coordination: pensions, and poverty and social exclusion. As was the case in employment policy, cooperation at the EU level began in 2001 before the Treaty of Nice was even ratified. National Strategy Reports on pensions were submitted in September 2002, and a joint Commission and Council report was drawn up on pensions in Spring 2003 (Council 2002, para. 22). A similar process also began for reporting on social exclusion. Both of these new uses of the OMC do not, however, contain the issuing of Commission-proposed and Council-approved recommendations. Still, the standards to be met in each issue area are clear: pensions are to be ‘both financially sustainable and meet their social objectives’, and ‘a high level of social protection’ is to be provided in the Member States (Council 2002, para. 22, 25). Although the EU is not involved in financing benefits, it has issued objectives to Member States in two of the categories of social policy that require spending: social insurance and social assistance. The close relations of spending on welfare state policies to government budget deficits, regulated for EMU Members by the Stability and Growth Pact, is no doubt pushing cooperation on pensions in particular. The best example of policy spillover from economic integration to social policy governance may thus emerge in the early years of the 2000s. It is possible, too, that the OMC may apply to health care, and to housing in the coming years (Council 2001d, para. 28; Council 2002, para. 12). In promoting cooperation on housing, the EU returns to an issue of social policy it concentrated on in the early years of the ECSC (Collins 1975a). Other areas of potential cooperation are linked to the EES. A new goal of providing ‘child care facilities … by 2010 to at least 90 per cent of children between 3 years old and the mandatory school age and at least 33 per cent of children under 3 years of age’ has been proposed as part of removing disincentives to female labour force participation (Council 2002, para. 32). Even wage levels are now increasingly part of the EU agenda, especially in the context of EMU that provides transparency in wage levels. It should be noted that the guarantee of a fair wage for coal and steel workers, such that firms could not lower wages in an attempt to offset the costs of economic adjustment or lower wages as a means of lowering costs so as to become more competitive, was included in the ECSC Treaty under art. 68 (ECSC 1951). Ex-Article 119 of the EEC guaranteed equal wages for equal work between men and women (EEC 1957). Wage levels more generally, however, are explicitly excluded from art. 137 (EC 1999). A discussion on wages at the EU level has nonetheless proceeded, but largely on the relationship of wages to inflation. The
204 European Welfare States
broad principles that have been set out include the need for productivity to exceed wage increases, and the role of wage settlements by the social partners in contributing to non-inflationary growth and employment (Germany 1997b). The idea of a ‘decent reference wage’ was included in drafts of the 1989 Charter of Fundamental Rights of the Worker (EU), but was removed after objections from the British, Spanish and Portuguese delegations (Rhodes 1991a, p. 14). The experience of southern enlargement is that ‘the raising of real wages is a slow process, in line with internal productivity trends rather than an abstract European “norm” ’ (Commission 2000b, p. 3). Resistance to cooperation on wages and non-wage social costs is also suggested by the fact that they are a key way by which Member States can ‘determine competitiveness in an integrated economy with a single currency’ (Rhodes 1998, 40). As for the issue areas into which supranational governance expands, it must be recalled that social policy is not rationally allocated to various levels on an efficiency or equality basis – there is no overall decisionmaking model that assigns policy to the levels of statecraft, community craft and world craft. Instead, there is ‘a messy and ambiguous vertical division of labour between national and EU levels’, and between these levels and the global one (Wessels 1997, p. 279). The historically sensitive methodology employed suggests that the inefficiencies, tensions and mismatches between the demands created by work and welfare needs, and the actual policies provided, mirror the history of the development of the welfare state at the national level. Welfare states evolve slowly over time, do not cover all economic and social risks, and are not evenly spread out in terms of resource allocation to groups or individuals in society. We should not expect EU or global-level social policies to be possessed of a policy rationale missing from the state level. Much depends, of course, on the interests of Member States in seeing EU competencies evolve. A general observation is that EU Member States have agreed to cooperation in expanding numbers of issues related to social policy as the number of Member States has increased. The enlargement of the early 1980s, first to Greece and then to Spain and Portugal, was part of the background to the Single European Act negotiations and helps to explain, in part, why Member States with higher social standards wanted qualified majority voting to be applied for the first time in limited areas of social policy; the composition of EU Member States was about to change, and perceptions regarding the desirability of changing formal decision-making rules in certain areas of social policy were therefore shifted. As illustrated by Table 6.1, more new forms of
Conclusion 205
social governance have been created and produced policy outputs in 1993–2002, when the EU expanded to 15 members, than in any other period. It may be suggested, therefore, that the increasing diversity of welfare state regimes in the EU actually contributes to supranational integration; as the diversity of Member States increases, competitive pressures on social policy also increase, raising demands from high-standard Member States for more cooperation. However, while the demands for social policy may increase from leader Member States, the difficulties of ensuring implementation of common or minimum-level policies also increase. There is no doubt that the corpus of social legislation that exists at present will be difficult to implement in the new Member States of the enlarged EU, not only because of weak policy histories but also because of the underdevelopment of employer groups and the recent nature of privatization (Commission 2000b, p. 23). There will be some areas, however, where the new enlargement states of 2004 will exceed EU standards, such as female participation rates (Commission 2002, p. 18). Funding was already extended to the applicant countries to develop national focal points for the collection and dissemination of data on occupational health and safety (Agency 2001, pp. 5, 17), and, as in the case of southern Europe, EU funding for the development of social policies will no doubt be critical, especially as it may be predicted that EU legislation will become the national standard in many instances. A close examination of national welfare state histories and of state behaviour across sub-issues areas is needed in order to assess properly resilience and transformation in the newest Member States. As demonstrated by the case studies, much of the future of supranational governance in the social policy field depends on the policy entrepreneurship of institutional officials, on the leadership provided by the most enthusiastic states and on the financial incentives provided to passive and resistant states. With the development of the OMC, it is possible that an effective compromise has been drawn between the desire of states to retain ultimate authority in social policy and the realities of economic integration that increase demands for policy cooperation at the supranational level. States may be far more willing to cooperate at supranational levels, and to cede some authority over policy objectives to supranational authorities, if they are assured that their legal competence is not under threat. Policy entrepreneurship is likely not to be in short supply; valuable ideas on possible methods of social policy cooperation are constantly circulating from a range of sources. Given this, the restrictions on supranational governance,
206 European Welfare States
especially if poor implementation is considered a limitation, are most likely to be linked to the weakness of financial resources and oversight mechanisms. As institutions have been proven to influence Member State positions on supranational cooperation, both institutional incentives and lacunae are key to understanding the limits of supranational governance.
Notes 1 European Welfare States and Supranational Governance of Social Policy 1. The phrase ‘welfare state’ developed a pedigree; it goes back to the 1875 British legislation to ensure collective bargaining by trade unions and the right to strike, and to the first programme of compulsory social insurance introduced by Bismarck in Germany in the 1880s (Marshall 1975, pp. 21, 45). From its largely German beginnings, the concept of social insurance for the working population spread to Belgium, Denmark, France, Norway and Sweden. Other policy innovations followed, including the introduction of family allowance by Belgium in 1930 and France in 1932. These policies were emulated by other states, creating a ‘movement of convergence … as the example of the pioneers was generally followed’ (Marshall 1975, p. 78). 2. Soziale Marktwirtschaft should be differentiated from Sozialstaat, the term used under Bismarck as ‘a prescription against socialism and a means to win the new proletariat’s loyalties for the autocracy’ (Esping-Andersen 1990, p. 108). More generally, Sozialstaat refers specifically to the concept of state-provided social insurance, while Soziale Markwirtschaft locates the individual in a system of capitalist production and state provided social services intended to enable the individual to retain control over his or her entry into, and exit from, the labour market. 3. Defining worker training as one of the three categories of social policy (assistance, insurance, regulation) is an especially tricky one – is it a benefit in the form of education, a form of social insurance to the unemployed, or a way of meeting a skills gap in the labour market? Here worker training is treated as a policy that seeks to remedy inefficiencies in the labour market; but, depending on the long-term histories of individuals receiving retraining, it could be viewed as one of the other components of social policy. 4. This statement follows Pierson: ‘Understanding the contemporary politics of the welfare state requires examination of the distinctive problems and dynamics which characterize particular sectors’ (Pierson 2001a, p. 11). 5. Such a comparison between EU Member States of the south and east cannot be taken too far, and in further work the welfare states of the East merit individual analysis. For an extension of the welfare state regime analysis to central-eastern Europe see Ferge (2001). 6. The presentation and use of historical institutionalism also draws heavily upon Addison and Siebert (1997), Scharpf (1997), Armstrong and Bulmer (1998), Bulmer (1998), Pierson (1998), Aspinwall and Schneider (2001). 7. The term ‘institution’ is used in the work to refer to the EU level, namely the European Commission, Council, Parliament and ECJ. The complex of national social and labour market institutions and policies will be referred to as ‘national welfare state regimes’ in order to avoid confusing terms. 207
208 Notes 8. A limit on the power of the ECJ is that national courts must refer cases to the ECJ, although the Commission, as guardian of the Treaties, may also take cases to the ECJ in cases of non-implementation. 9. The votes in an enlarged EU under the Treaty of Nice are distributed as follows: France, Germany, Italy, UK have 29; Poland and Spain have 27; Netherlands 13; Belgium, Czech Republic, Greece, Hungary, Portugal 12; Austria and Sweden 10; Denmark, Finland, Ireland, Lithuania, Slovakia 7; Cyprus, Estonia, Latvia, Luxembourg, Malta, Slovenia 4. Under the proposed Constitution, a successful vote by QMV would require support by at least 55 per cent of the members of the Council, at least 15 Member States, and Member States representing at least 65 per cent of the EU’s population (EU 2000a; EU 2004). This reform of QMV in the EU-25 backs away from suggestions to move to a more simple system of a ‘double majority’ of Member State support and population, and from the draft European Convention where it was proposed votes passed by QMV would consist of the majority of Member States and that those Member States must represent at least three-fifths (60 per cent) of the population of the Union (EU 2003). 10. An example of Commission rhetoric is the following statement: ‘[W]e have to strike the right balance between economic and social considerations. We [the Commission] have firmly rejected the view that only economic considerations should be taken into account and that social policy is a luxury that can be afforded in good times but thrown out in bad times’ (DG V official) (van Zonneveld 2000, p. 14). The Commission’s use of rhetoric to create an ‘image’ of the policy problem at hand, and its ability to create ‘venues’ for policy cooperation, led one author using the term ‘image-venue entrepreneur’ to describe the Commission (Wendon 1998). 11. As a recent analysis of IGCs on the basis of historical institutionalist analysis explained: ‘[S]cholars should not easily discard supranational actors from their research designs on treaty reform – although looking at more factors is demanding and supranational entrepreneurship is not always easily visible’ (Falkner 2002, p. 113). In addition, ‘there is a need to relate developments in the periods between IGCs to the treaty reform process’ that can only happen if the analysis spans policy-making over time (Christiansen et al. 2002, p. 28). 12. Observations about the critical importance of policy entrepreneurs has given rise to well-deserved criticism that institutionalist theories subsume individual actors inside a more mechanistic view of political interaction. Some writers have attempted to ‘put the individual back in’ through actorcentred approaches to institutionalist theory that marry rational choice with institutionalist paradigms (Scharpf 1997). 13. The phrase ‘unanticipated consequences’ is from Pierson (Pierson 1998). Other writers on historical institutionalism have used the phrase ‘unintended consequences’. Although the difference between these two phrases may be minimal, ‘unanticipated consequences’ is preferred. This is because the rationalist underpinnings of my work assume that Member States will attempt to anticipate the consequences of cooperation and yet, given the long time horizon in which policy-making will occur, will be unable to do so. Consequences of new bargains at the time of Treaty reform, or in
Notes 209
14.
15.
16.
17.
18.
19.
instances of policy-making, are thus not so much unintended as they are unanticipated. This comment is especially relevant, coming as it does from an academic and Amsterdam Treaty negotiator for Finland who was (in 2002) a member of Commission President Romano Prodi’s cabinet. Like the literature that emphasizes governance in the EU ( Jachtenfuchs 2001), historical institutionalism is eclectic in its emphasis on both actors (Member States) and interests (such as unions and employers), and the ways in which supranational institutions constrain actor preferences and shape change (Marks and Hooghe 1996, p. 355). Indeed, many of the elements of historical institutionalism are folded into the multilevel governance literature, notably the constraint of principles by institutional rules and the role of unanticipated consequences (Marks and Hooghe 1996, pp. 353–6). While Pierson subsumes changes in government as part of ‘shifts in COG [Chiefs of Government] policy preferences’, here a change in government is treated as exogenous to the historical institutionalist model (Pierson 1998, pp. 41–3). The Secretariat of OKE, the Economic and Social Council of Greece, also translated a few Greek sources (mostly legal statutes) not available in English or French. For this they have my thanks. The 30-year rule preventing researchers from accessing national, EU and ILO documents until three decades have passed is slowly dissolving in the wake of calls for greater transparency, notably at the EU level by Nordic Member States. At the European Council Archives I was given access to a range of files, some as late as 2001, and permission was granted for all of my requests to quote these documents. Documents released to me will also go on the web as part of the Council’s work to promote access to documents for researchers who cannot travel to Brussels. At the ILO I was given access to documents from the Director-General’s office up to 1992. Full transcripts of yearly International Labour Conference meetings are public. At the WTO, many official documents are available, as are summaries of various Council meetings, although the record is, by its nature, less detailed when compared to summaries of European Council meetings that give the position of each Member State on the issue under discussion. Fritz Scharpf’s actor-centred institutionalism represents an integration of ‘rational-choice and institutionalist’ paradigms that is very similar to the research design presented here (Scharpf 1997, p. 36). As Scharpf reminds us, actor-centred institutionalism is ‘a framework … not a theory’, but provides ‘guidelines for the search for explanations’ (Scharpf 1997, p. 37). His framework ‘emphasizes the influence of institutions on the perceptions, preferences, and capabilities of individual and corporate actors and on the modes of their interaction’ (Scharpf 1997, p. 38). There is a danger in Scharpf’s analysis, however, of overstating the extent to which preferences are shaped by institutions. Underlying preferences are best considered as expressing the outcome of national processes, while institutions may shape bargaining positions and perceptions of possible bargaining strategies relative to possible outcomes.
210 Notes
2 Governance by Law: Health and Safety in the Workplace 1. The use of the term EU is especially problematic in a chapter concerning the SEA when the European Union was not yet in existence, but rather the European Economic Community (EEC), EURATOM and the European Coal and Steel Community (ECSC). Collectively these institutions (EEC, EURATOM and ECSC) were referred to as the ‘European Communities’ or ‘European Community’ (EC). The term EU will be used in general discussion for the sake of consistency, but where necessary for accuracy the term EC will be used. 2. Although Sweden was not an EU Member State in 1985, a Swede – Pehr Gyllenhammer, CEO of Volvo – was the head of the influential European Roundtable of Industrialists (ERT) that pushed for R&D tax credits, a reduction of barriers to mergers and a host of other measures to ease trade and labour flows in Europe. The ERT became a key supporter of the Commission White Paper on the Internal Market (Cowles 1995), and advocated QMV to ease the passage of market-liberalizing legislation (Moravcsik 1998, p. 356). EFTA states including Sweden were then (1985) also members of the European standardization bodies (CEN, CENELEC and the ETSI) and, as EFTA members, aimed ‘to influence the Commission to set the minimum standards [in health and safety] to correspond to the presumably higher requirements in the EFTA countries’ (EFTA 1988). In addition, health and safety policy had been an area of cooperation between the Nordic Council of which Denmark was also a part. Despite being a non-member, Swedish preferences were fed into the decisionmaking process via the ERT and Commission experts. 3. Article 100a (now art. 95) was added to the Treaty under the SEA to permit the approximation of laws for the purpose of completing the internal market. It, too, has been used to propose product and technical standards that have an effect on health and safety in the workplace. 4. See OJ EC 31 August 1962. 5. The European Parliament (appointed rather than elected until 1979) also resisted setting high levels of worker protection (Vogel 1993, p. 68, fn. 15). In the 1990s the EP acted consistently in favour of higher levels of protection for workers. 6. Directives 77/576/EEC (safety signs), 82/605EEC (lead), 83/1107/EEC (asbestos), 86/188/EEC (noise), Directive 80/1107/EEC (exposure to chemical, physical, and biological agents), also known as the 1980 Framework Directive, as well as Directive 78/610/EEC on exposure to vinyl chloride monomer. 7. Mutual recognition had previously been used in EFTA. 8. The Commission played an important entrepreneurial role in making concrete proposals for the completion of the SEM and helped to mobilize prointegrationist business interests in the form of the European Roundtable of Industrialists (Sandholtz and Zysman 1989; Cockfield 1994; Cowles 1995; Ross 1995a). Other explanations focus on the convergence of national economic preferences held by governments of the right (Moravcsik 1991). It is clear that national preferences and Commission activism are both necessary explanatory variables to account for Treaty outcomes, including grand economic projects such as the SEM and EMU.
Notes 211 9. This analysis is based on the predictions provided in (Moravcsik and Nicolaïdis 1999) which argues that rich countries of the left should be the most in favour of approving social policy, while poor governments of the right should be the most resistant. This prediction is presaged in (Moravcsik 1998) where Moravcsik states: ‘Much of Thatcher’s opposition reflected scepticism about EC social regulation from a relatively right-wing government in a (relatively) low-income country’ (p. 346). Although definitions of ‘poor’ and ‘rich’ are not provided, it seems reasonable to assume that countries below the EU average should be considered relatively poor. Moravcsik and Nicolaïdis do not state which variable (political party or relative wealth) is the more salient. The behaviour of poor countries with governments of the left and rich countries with governments of the right is thus not clearly predicted by their analysis. 10. For an account of the extraordinary use of majority voting to call the IGC, and more detail on the Milan Summit, see (Budden 1994). 11. The UK government was not, however, in full support of the SEM as proposed by Commissioner Lord Cockfield, although the SEM was clearly the main driving interest of the UK in engaging in the IGC. Thatcher herself described Cockfield as having gone native once in Brussels (Thatcher 1993, p. 547). For a detailed analysis of the UK position see Budden (1994), especially pp. 318–23 on the differences between Cockfield and the UK government. 12. Budden attributes the Danish proposal of QMV under art. 118 to the need to secure the agreement of the Social-Democratic opposition for the SEM project (Budden 1994, p. 314). The consensual and coalition style governance that typifies the Danish and Swedish approach to relations with the EU is underappreciated in this analysis. 13. Under QMV rules from 1986–94 a formal blocking minority needed 23 votes. Voting weights were as follows: 10 votes each for France, Germany, Italy and the UK; 8 for Spain; 5 for Belgium, Greece, the Netherlands and Portugal; 3 for Denmark and Ireland; 2 for Luxembourg. For more on the evolution of QMV see Chapter 1. In practice, however, Directives have tended not to come under a strict QMV vote but rather have been guided by a desire for consensus or a kind of informal unanimity (Hayes-Renshaw and Wallace 1996). 14. Papandreou’s intransigence was all the more frustrating as Greek negotiators to the EC in the 1970s repeatedly emphasized that Greece is not seeking enormous financial rewards from Community members, that its economy is strong enough to stand the strains of entry without added support and that its impact upon the balance of budgetary advantages within the Community will therefore be small. (Wallace 1979, p. 33) The application of increasingly tough pre-accession criteria for new Member States, notably the countries of eastern Europe, is partially attributable to the difficulties experienced in enlargement to the South. 15. An adviser attached to the prime minister, Tzochatzopoulos, told the EP’s delegation: Greece’s reservations on the veto could be explained in psychological terms by its limited experience as a Member State and its particular
212 Notes problems: once the issue of the IMPs [Integrated Mediterranean Programmes] was solved, the government’s position might become more flexible. (Corbett 1998, p. 198) 16. Lord Cockfield stated in his memoirs: And on social policy, the other main area in which demands for ‘linkage’ [with the Single Market Programme] were put forward, it took more than four years before the Social Charter saw the light of day. Had ‘linkage’ been accepted it would have resulted in intolerable delay being imposed on the Internal Market Programme. Lord Cockfield also notes: The agreement I made with my colleagues from the South was that in return for their support for my Internal Market Programme, I would support them in their demands that the Structural Funds should be doubled. But ‘linkage’ I would not accept at any price … Ultimately the success of the line I had taken was sealed by the fact that at the second Brussels Summit in February 1988 when the success of the Internal Market Programme was clear, the Heads of Government agreed to doubling the Structural Funds. It is worth recalling that once again this was bitterly opposed by the United Kingdom. They wanted the Single Market but were not prepared to pay for it. (Cockfield 1994)
17.
18.
19.
20.
My reading of this statement is that linkage did, in fact, occur. However, given the vagueness of what constitutes linkage, and the differences between both explicit and implicit issue linkage and between immediate and forward linkages (in time), opinions on this subject vary. In addition, the notion of side-payments is vague. Minimum health and safety requirements for the workplace (Directive 89/654), use of machines and equipment (Directive 89/655), personal protective equipment (Directive 89/656), handling heavy loads (Directive 90/269), visual display units (Directive 90/270), carcinogens (90/394), biological agents (Directive 90/394), exposure to asbestos (Directive 90/679). UNICE proposed the following statement replace arts. 4 and 6: ‘The employer shall be responsible for the safety and health of the workers in the undertaking/establishment, regardless of the methods he uses.’ Because of the dynamic nature of technology, health and safety standards may often lag behind technological innovations and new working methods. As a result, standards may be outdated very soon after they are created. Introducing a reference to the state of the art is meant to correct for this. Revising regulation on a regular basis is also important given technological change. In Germany, reference to privately developed standards is also used to keep standards updated (Eichener 1992, pp. 12–13). In the EU the ‘state of the art’ is defined by the European standardization bodies, and so the level of protection to be provided is decided not only by the formal Regulations approved in the Council but also by the technical committees of the European standardization bodies. In other Member States such a consideration by the Courts is referred to as ‘the principle of proportionality’ ( James 1993, p. 149).
Notes 213 21. Mr Pangalos, under-secretary for foreign affairs responsible for European affairs of the PASOK government (1993), had this to say about politicization of staffing in Brussels: When the time comes for the implementation of agreed policies with regard to the personnel in the administration … the Minister, any Minister, on personal or party considerations chooses the people to be sent to Brussels or that will become heads of the EU-relevant departments within these ministries. As a result there are people in charge with no knowledge of the subjects involved whatsoever or that cannot even speak a foreign language. (Quoted in Tsimisizelis 1996, p. 220) 22. Dual translation often occurs in specialist meetings owing to the unavailability and cost of securing translators that can work directly in all nine official Council working languages. A practical example would be that no translator would be present at a committee meeting that can translate directly from Greek to Danish, so a translator waits for the translation from, say, Greek to English, or Greek to French, and then translates from English or French to Danish. According to an official interviewed, Greek committee members who speak more than one Community language often ‘listen in’ on other channels while their colleagues are talking to see what the translators are saying, and that the meaning is ‘often lost’ owing to poor translation. In this official’s experience interventions are most effective if done in English, and Greek delegates who speak English ‘often speak English when they feel they really need to make a point’ (Interview 16). This last comment was confirmed in other interviews with Greek committee members (Interviews with 13, 14, 18). The danger of interventions in English, however, is that ‘if you stop speaking Greek the number and quality of translators may suffer even further.’ The difficulties of translation can only increase exponentially with EU enlargement to eastern Europe. 23. The following ‘Resumés des Travaux du Groupe des Questions sociales’ from the Council archives were analysed: 8 avril 1988 (5504/88 RESTREINT), 13 avril 1988 (5562/88 RESTREINT), 4 mai 1988 (5968/88 RESTREINT), 12 juillet 1988 (7411/88 RESTREINT), 16 septembre 1988 (8079/88 RESTREINT), 13 octobre 1988 (8607/RESTREINT), 14 novembre 1988 (8915/ 88 RESTREINT), 24 novembre 1988 (8916/88 RESTREINT), 28 novembre 1988 (9716/88 RESTREINT). As a result of the request to release these documents, they will soon be scanned and made available via the Council Archive web page. 24. The leadership of the Commission in drafting legislation at a high level of protection was also in evidence in the Machinery Directive (89/392/EC) and the Display Screen Equipment Directive (90/270/EC), supported by the Economic and Social Committee who advised on the Directives, and by the European Parliament that was also interested in a high level of protection for workers (Eichener 1992, pp. 58–52). The Commission has even drafted proposals that met with criticism from both leader and large Member States, such as in the case of the Machinery Directive where the definition of machinery was opposed by Denmark, France, Germany and the UK (Eichener 1992, p. 57). 25. In one notable case in the creation of the Machinery Directive, however, it was not the Swedes or Danes who managed to convince the Commission of
214 Notes the value of the broader concept of occupational health but rather a ‘reformist member of the British Labour Inspection Service’ who, in drafting the Machinery Directive ‘brought his innovative ideas to the European Community’ after ‘he had no success [in reforming the health and safety philosophy] in his own country’ (Eichener 1992, pp. 36, 54). Notably, Germany was the only state to vote against Directive 89/392/EC on the Safety of Machinery, probably because this would mean that national patterns of decision-making involving very powerful bipartite bodies (such as the insurance injury organization Berufsdenossenschaften) would lose sole regulatory power (Eichener 1992, p. 39). 26. The diversity of tools used for implementation is apparent in the diversity of national legislation notified to the Commission by Member States in fulfilling the transposition of Directive 80/391/EEC: Presidential decrees in Greece and new laws that drew heavily on the wording of the Directive [Presidential Decrees 1975 (no. 61), 1987–96 and Laws N.1837, 03/89, and No. 22224, 05/07/94], a new national statute in France that reproduced almost verbatim much of the Directive (The Prevention of Occupational Risks Act (91–1414) of 31 December 1991), a revision of several legal acts in Sweden (Arbetsiljög 1977: 1160, Arbetsmiljöförordning 1977: 1166, Kungörelse om skyddsåtgärder mot skada genom fall (1981), Kungörelse om arbetsställningar och arbetsrörelser (1983), Kungörelse om takarbete (1983), Kungörelse med föreskrifter om första hjälpen vid olycksfall och akut sjukdom (1984), Kungörelse med föreskrifter om personalutrymmen (1992); (AFS 1983–92 and Arbetarskyddsstyrelens Författningssamiling 11/09/90, 14/09/81)) and Germany (Unfallverhütungsvorschrift (1974–93) (Bundesgesetzblatt 1957, 1975, 1986–96)), and a diversity of legislation that, collectively, is said to implement the Directive in the UK (Health and Safety at Work etc. Act 1974, Industrial Relations Order 1976, Safety Representatives and Safety Committee Regulations 1977, The Health and Safety at Work (northern Ireland) Order 1978, Employment Protection (consolidation) act 1978, Safety Representatives and Safety Committees Regulations, 1979, The Health and Safety (First-Aid) Regulations 1981, The Reporting of Injuries, Health and Safety (First-Aid) Regulations (Northern Ireland) 1982, Fire Service Order 1984, Diseases and Dangerous Occurrences Regulations 1985, The Relevant Statutory Provision for northern Ireland Reporting of Injuries, diseases and Dangerous occurrences Regulations, The Management of Health and Safety at Work Regulations 1992, The Personal Protective Equipment at Work Regulations, 1992, Management of Health and Safety at Work Regulations, 1992, Employment Rights Act 1993, Personal Protective Equipment at Work Regulations 1993, The Health and Safety (consultation with employees) Regulations 1996, Factories Ordinance, Factories (safety) Regulations 1996, Fire precautions (workplace) Regulations 1997 (Statutory Rules, Statutory Instruments, Statutory Rules of northern Ireland 1976–97)) (Commission 1998a, pp. 87–96). 27. Although not covered here, it may be worth noting that Directives concerning chemical products and machinery standards have required more substantive technical adjustments to Swedish legislation than Directives concerning minimum health and safety standards (Interview 19). Even high-standard countries thus have some adjustments to make as a result of EU legislation.
Notes 215 28. ELINYAE’s mandate includes fulfilling the services necessitated by national and EU legislation. In the 1988 and 1991 National General Collective Agreements, unions agreed to lower pay rises in exchange for the creation of LAEK (Account for Employment and Vocational Training). It is LAEK that manages ELINYAE, and not the Ministry of Labour. ELINYAE was thus founded not by statute but by the terms of the National Collective Agreements, making it a highly unusual organization in Greece. 29. In the words of one official, We are not happy with the downward trend because the downward trend may be significantly correlated to structural change. Up to 80 per cent of the improvement [in the declining number of industrial accidents] may be linked to structural change, and 20 per cent of the improvement from better legislation … but we simply have no data to test this assertion (Interview 76). 30. A detailed examination of the Agency is outside the scope of this chapter. However, it is interesting to note that repeated attempts by the author to reach the national focal point in Greece (identified on the Agency website) via email while in Athens went answered, while emails to the UK and Swedish focal points resulted in responses within two working days. New EU Agencies cannot solve weak national administration.
3
Governance by Collective Bargaining: The Social Dialogue
1. Social Dialogue could be further subdivided into the Inter-sectoral Social Dialogue conducted by UNICE, CEEP and the ETUC and the sectoral social dialogues. The Inter-sectoral Social Dialogue is covered here. For more detail on the inter-sectoral social dialogues, see Commission (2000b), Weber (2001). 2. The earliest information on hourly wages in industry (manual and nonmanual workers) given for Germany is 26.12 ECU/EUR in 1995, and for France at 19.12 in 1992 (Eurostat 2002a, p. 16). 3. One expert on Spain noted the following about the effects of EU legislation: The Spanish government has emphasized its commitment to bringing Spanish legislation into line with Community legislation. It seems to have accepted that implementation of the ESC [European Social Charter] in Spain would increase production costs, relying on the modernization of the economy and the generosity of the European Structural Funds to provide the foundations for restoring competitiveness to the Spanish economy. (Argandona 1997, p. 207) 4. French business did not strongly support the ratification of Maastricht because of its social policy provisions, which ‘business believed to be unnecessary and disadvantageous’ in light of the British opt-out (Moravcsik 1998). 5. Liberal intergovernmentalism seems to equate employer interests with economic interests (Moravcsik 1997, 1998). The role of unions is not considered, and it is far from clear how conflicting national union and employer group opinions on the Social Dialogue could be reconciled and presented as ‘economic interests’ of a Member State.
216 Notes 6. Dolvik and Visser, two of Europe’s leading experts on national collective bargaining and unions, note that the final mechanisms for EU Social Dialogue most closely resemble those of Belgium (Dolvik and Visser 2001, p. 30). The Commission has stated that ‘the development of agreement-based law at European level is to a large extent inspired by the experience and practice of the Nordic countries, Belgium, and Italy’, and these countries are those that have most often implemented Directives by means of collective agreement (Commission 2000b, pp. 28, 37). Although the EU Social Dialogue is described as ‘unique in Europe’ (Dolvik and Visser 2001, p. 30), it may be seen to ‘fit’ better with national traditions of bipartitie bargaining, followed by tripartite bargaining systems. 7. Such networks, while hard to document, demonstrate the overlap between national officials, trade unionists and the Commission. One early example of the interconnection of the ETUC and the European Commission was Carlo Savoini, a former Italian trade union official who became the secretary of the European Secretariat of ICFTU (which later became the ETUC) in 1964, an official in DG V in 1973, and head of the DG V unit on Social Dialogue in 1991. A later example was Patrick Venturini, a French trade unionist who was the key adviser on Social Dialogue in Jacques Delors’s cabinet in the 1980s, and later head of ESC in 1998. Jaques Delors, who had encouraged bipartite bargaining in France in the 1970s in his role as adviser to French PM Jacques Chaban-Delmas, voted in favour of the Tripartite Employment Committee for the EU and was a key advocate of the Social Dialogue when he became Commission president in 1995. Later, Jacques Santer, an author of the Werner Plan, advocated social partner participation in all stages of EMU. Santer became the Luxembourg PM 1984–95, and succeeded Delors as president of the Commission. 8. The ETUC was preceded by the European trade union secretariat (SSE) and the European Confederation of Free Trade Unions (CESL). The European Association of Craft, Small and Medium-Sized Enterprises (UEAPME) has also pushed for participation in the Social Dialogue but, although consulted on issues, has not been invited to be a formal member of the process. A UEAPME–UNICE cooperation agreement may open up UEAPME participation in future negotiations (Branch and Greenwood 2001, p. 42). Other social partners have been recognized to negotiate voluntary sector-specific agreements since January 1999. These social partners may be cross-industry or sector specific union and employer groups with recognized national affiliates in several Member States, and must have the capacity to effectively participate in the process. For a list of such sectoral social partners, see Commission (2000b, p. 16). This chapter concentrates on the larger Framework Directives that are signed by the ETUC, UNICE and CEEP. 9. Dolvik describes the ETUC as extremely weak, without the experience of joint European action, deprived of financial and staff resources, lacking authority over or commitment from its affiliates, with ill-defined and often contested relations between sectoral and trans-sectoral union activities and responsibilities, and utterly dependent upon its sponsors in the Commission, and in DG V. (Dolvik and Visser 2001, p. 16)
Notes 217 10. It has been suggested that German agreement to enter EMU was linked to European, and most critically French, approval for reunification (Garrett 1993). Other authors have demonstrated, however, that German preferences favouring monetary union were in place before 1990 and were surprisingly unchanged despite the monetary havoc created by the economic conditions of reunification (Moravcsik 1998). 11. Falkner goes so far as to suggest, on the basis of several interviews, that the Commission may have colluded with the Belgian social ministry in developing these proposals, and allowed Belgium as a Member State government to make the initial proposal to the IGC (Falkner 1998, p. 92, 2002). 12. Luxembourg Presidency ‘Draft Treaty on the Union’ 18 June 1991: art. 118b 1. Should management and labour so desire, the dialogue between them at Community level may lead to relations based on agreement including agreement which shall be implemented in accordance with the procedures and practices peculiar to each Member State. 2. In the field referred to in art. 118, where management and labour so desire, the Commission may submit proposals to translate the agreement referred to in paragraph 1 into Community legislation. The Council shall act as laid down in art. 118. 13. A more detailed consideration of the composition of the Ad Hoc Group and the October 1991 negotiations can be found in Dolvik (1997). Dolvik emphasizes the division between national delegates appointed by the ETUC to the Ad Hoc Group, and national affiliates. It seems that neither employers nor unions initially took the Ad Hoc group very seriously, and had not given it a clear bargaining mandate (Dolvik 1997, pp. 207–8). It should be noted that the ETUC and UNICE secretariats have an even stronger impetus to European-level agreement than national affiliates, as their specific purpose is representation at the EU level. The ETUC secretariat, and to a lesser but significant extent that of UNICE, had ‘vested interests in defending the legitimacy of their institutionalized role’ (Dolvik 1997, p. 215). The ETUC and UNICE secretariats thus each became an additional actor in the process of bargaining, and these organizations should be viewed as more than conglomerates of national interests. For a more detailed look at UNICE consultation of national members see Tyszkiewicz (2002a). Notably, the ETUC introduced QMV for internal decision-making in 1991, and UNICE in 1999. 14. Germany’s proposal of a form of reinforced QMV (with a higher number of votes need for agreement under QMV) sits alongside Kohl’s activism in the final negotiations. Germany’s concerns about subsidiarity, competence, and the costs of deeper integration, combined with Kohl’s highly pro-integrationist attitude, create this ambiguous position. 15. In an interview with John Major, he suggested that his three primary reasons for rejecting the Social Protocol were: (1) concerns about reversing supplyside changes made to the British economy in the 1980s; (2) new burdens placed on employers; (3) losing competitiveness relative to the United States. In response to the question ‘Is there a fourth reason [that you did not sign the Social Chapter] which is perhaps that you would not have gotten the eventual majority in Parliament that you did for approving Maastricht?’,
218 Notes Major replied: No, it was not true, because I would have always had the support of the Labour Party and the Liberal Party if I had signed it. The arguments, because I know what you are referring to … [pause] there were newspaper articles subsequently that I didn’t sign the Social Chapter because one or two members of my Cabinet threatened to resign [if the Social Chapter was included]. If they did threaten to resign I didn’t know it at the time I had negotiated it, because they had certainly not said that to me in Cabinet, or anywhere else. They may have whispered it behind their hands to newspapers who reported it subsequently, but not to me. So, in my mind, it [threats by Cabinet members to resign if the Social Chapter was signed] never was a factor, and neither do I believe that if I had so signed the Social Chapter that they [my Cabinet members] would have resigned … .suggestions to the contrary were always newspaper talk, and complete tittle-tattle. (Interview 6) On 23 July 1993 a vote of confidence in the Major government as a result of debates concerning the Maastricht Treaty and the Social Protocol was won by 339 votes to 229. 16. EU legislation had also been previously transposed into national law via a collective bargaining agreement between management and labour at the national level. In Belgium, the Directives on collective redundancies (75/129) and on acquired rights in case of transfer of enterprises (77/187) ‘have been (at least partially) transposed into Belgian law by way of (extended) [that is, national and binding on all employers and workers] collective bargaining agreements’ (Engles and Salas 1998, p. 84). The national systems of France, Germany and the Netherlands also permit implementation by collective bargaining in certain instances. 17. Still, it is not totally clear how Delors could have forced Member States to agree to the old November draft on social policy had this not been agreeable to them. Paul Pierson called this early draft a ‘bargaining chip in the expectation that Britain would eventually accede to a contentless compromise’ (Pierson 1998, p. 55). It is not explained how the social partners would have reacted to art. 118b being changed after their October agreement. I interpret Pierson’s comments as meaning that the Dutch government had presented a watered-down text in the hopes of getting the UK to agree, and when it did not, Member States were not able to rely on the watered-down draft as it would have named them as reluctant cooperators in social policy. In any case the Dutch presidency tried to dilute the agreed compromise after Maastricht, according to Jon Erik Dolvik, an effort that ultimately failed (Dolvik 1997, p. 217). 18. Papandreou’s inconsistent view of the EU continued during his Parliamentary speech, where, despite having pledged his party’s support for the ratification of the Treaty, he noted: ‘[The Treaty of Maastricht] is for us an invitation to a tough and unfair game … . Unfair because the Treaty of Maastricht reflects almost entirely the interests and views of the rich North’ (Koliopoulos 1994, p. 117). Ultimately the Greek Parliament ratified the TEU by a vote of 286 to 6.
Notes 219 19. Regulation agreed to under the Social Protocol would still be binding on UK firms operating elsewhere in the EU if they had a certain number of subsidiaries and employees in the EU and the EEA (European Economic Area). EEA Members such as Austria, Finland, Norway and Sweden were also bound to the Social Protocol. Other member firms of the CBI extended works councils voluntarily – such as United Biscuits and BP Oil (Falkner 1995, p. 11). 20. ‘Everyone residing and moving legally within the EU is entitled to benefits and social advantages in accordance with Community law and national law and practices’ (art. 34). 21. Two drafts of 1982 (OJ C 62/7 and OJ C 128/2) were amended in 1983 (OJ C 18/5) and 1984 (COM(84) 159 final). Three new drafts were proposed in 1990 (COM(90)228 final) and two were amended later that year (OJ C 305/8). Of the three proposed and amended Directives, only one was adopted (Directive 91/383/EC).
4
Governance by the Open Method of Coordination: The European Employment Strategy
1. In the period immediately following the Second World War there was no clear exchange rate between ECSC currencies and the European Payments Union was being formed. Accounts were thus registered in a ‘unit of account’; the exact dollar, pound, franc or DM equivalent I have been unable to find. 2. For more detail on ESF, structural and cohesion funds, see Shackleton (1993), Hooghe (1996), Morata and Muñoz (1996), Nanetti (1996) and Tsoukalis (1997). 3. A precursor to MISEP was SEDOC, a common European system for advertising vacancies, which was created in 1972 and superseded by EURES (the European Employment Service) in 1993 (Commission 1993b). 4. The Danish minister who served as the chairperson of the Employment Ministers’ Informal Council was described as ‘supportive’ of Flynn’s initiative, in contrast to ‘negative interference from the Council secretariat’ (Interview 36). 5. ‘There is no way that Delors would have allowed the Flynn-led Employment Initiative [The Community Wide Framework for Employment], with its specific 18-month timetable, to go forward if, at that time, he had other plans’ (Interview 36). 6. The White Paper has been described as ‘like the Bible – there’s something in it for everyone, and there are contradictory bits, too’ (Interview 36). 7. Allan Larsson had chaired a 1992 SAMAK (Nordic Social Democratic) report ‘Sätt Europa i arbete’ (Make Europe Work) ( Johansson 1999, p. 91). This report, which had brought him into contact with all Nordic Finance and Employment Ministers, was later adapted into a further report by the Party of European Socialists (PES) on a European Employment Initiative in 1994. António Manuel de Oliveira Guterres chaired the PES initiative. Guterres was the prime minister of Portugal during the influential 2000 Portuguese Presidency of the EU. 8. It may also be argued that his presence in the Commission is an overlap of Swedish national interest – his place at DG V was targeted by the Swedes as a place to put a top official – and Commission entrepreneurship.
220 Notes 9. Moravcsik and Nicolaïdis do not state whether political affiliation of the government is more important than the relative wealth or poverty of the Member State, although political affiliation seems to be more important in their analysis. Further, they do not make it clear how they have defined poor and rich states (Moravcsik and Nicolaïdis 1999, p. 63). Here I have used GDP per capita at purchasing power parity for the year 1996 to determine rich (above the EU average) and poor (below the EU average) Member States. The Moravcsik and Nicolaïdis explanation (Moravcsik and Nicolaïdis 1999) is quite different from the explanation one might have expected from the liberal intergovernmental school for the Employment Title. Given the emphasis on economic interests in liberal intergovernmentalism (Moravcsik 1997), one might have expected a hypothesis that stresses the salience of economic interests – namely employers – in influencing the government’s position. In this case, one would hypothesize that Germany would be in favour of the Employment Title in order to export costs, France and Sweden would support the Employment Title (for the same reasons as Germany) and Greece and the UK would oppose. This set of predictions is very close to the one suggested by a welfare state explanation of preferences, although the welfare state explanation also considers unemployment rates and not only spending on labour market policies. 10. Other priority issues during negotiations to join the EU included transparency (access to documents and the minutes of meetings of Councils of Ministers, for example), improving the democratic legitimacy of the EU, preserving neutrality under CFSP, cooperation with the Baltic states and Russia, and the environment ( Johnson 1998). These priorities were reflected in the ‘3 Es’ of the 2001 Swedish presidency of the European Union ( January– June): Employment, Environment and Enlargement. The announcement of the ‘3E’ presidency caused some critics to remark that there were two ‘Es’ missing – equality (of men and women) and EMU. Still others jokingly referred to the three ‘Es’ of the Swedish presidency as ‘efficiency, efficiency, efficiency’. 11. The government also hoped that suggesting an EU employment policy would ‘promote greater unity [in the SAP] by stressing a common theme for all Social Democrats. It was hoped that the anti-EU grass roots would be appeased. It was also argued that the unemployment issue would make the benefits of EU membership more real to a hostile Swedish public’ (Miles 1997, p. 280). 12. Gunnar Lund, formerly state secretary in the Finance Ministry under thenMinister Allan Larsson, was at the time (1995) state secretary in the Foreign Ministry and became the Swedish representative in the 1996 IGC. 13. As a French negotiator noted of the reflection group: [E]ach foreign ministry appointed a representative who was supposed to be independent. In practice the Group consisted largely of those people who would ultimately negotiate the new Treaty, plus two representatives of the European Parliament. As a result, the Group’s report cast its net wide: all of the opinions voiced by one or other of its members are included, with no way of ranking them in order of priority or ruling out the most improbable. (Petite 2001, p. 1)
Notes 221 14.
On employment the memorandum [on the IGC from the Governments of Belgium, Luxembourg and the Netherlands, 7 March 1996], while not denying the primary responsibility of Member States, proposes that the Treaty should contain explicit reference to the complementary role of the Union in promoting employment, with new provisions to this effect. In particular, it is proposed that employment policy coordination between Member States should take the form of annual Commission recommendations to be adopted by the Council. … It is also considered that the Community should make a greater contribution to job-creating investment in the Member States, and that a committee for employment should be set up to examine the subject and make suitable proposals. (Belgium 1996) While considering that the main efforts will still have to be made at national level, the Danish government believes that the new Treaty should include a specific section on employment, intended to strengthen the objective of achieving a higher employment level and emphasizing that employment is a common responsibility requiring a coordinated approach. (Denmark 1996) It [the Greek government] therefore supports incorporating a special title on employment into the revised Treaty. (Greece 1996) [Spain] advocates inclusion in the TEU of specific provisions in employment policy. (Spain 1996)
(NB: This report was written by the Socialist government (PSOE) but also had the backing of the Popular Party that formed the government in May 1996 (Rodrigo 1996, p. 28).) Italy believes that the revised Treaty must include a chapter on employment committing the Member States to undertake closer coordination of their employment policies on the basis of the strategic guidelines adopted at the Essen and Cannes European Councils. (Italy 1996) 15. The third stage of EMU was to begin on 1 January 1999 and fixed the conversion rates of participating countries against the euro in anticipation of the full changeover to euro notes and coins in January 2002. 16. Germany’s economic growth had slipped from 4.5 per cent to 2.2 per cent in 1994, financial transfers from former West to East Germany were, at their peak, 5 per cent of West German GDP, and a hard monetary policy followed by the Bundesbank had reduced growth, investment and tax revenue (Szukala and Wessels 1997, p. 80). 17. QMV, rather than unanimity, was the ‘normal’ Council voting procedure to suggest for the draft Title since ‘it deals in general policy issues and co-operation [among Member States]’ rather than ‘minimum social standards … where there are precise competitive issues at stake’ (Interview 36). 18. The complexity of the Treaty revisions is shown by the confusion, over arts. 102 and 103, between this text and the Swedish draft text. Arts. 102 and 103 of the Treaty on European Union are to be found in the consolidated Treaty Establishing the European Community in arts. 97, 98, 99 and 100. Art. 99 concerns the formulation of the Broad Economic Policy Guidelines (BEPG).
222 Notes 19. The exact wording regarding Member State competence is: The Community shall contribute to a high level of employment by encouraging cooperation between Member States, supporting, and if necessary, complementing their action. In doing so, the competences of the Member States shall be respected. (art. 127(1)) 20. Negotiations had no doubt already started on the Sunday night, 15 June. The Treaty negotiations concluded at 3:35 a.m. on Wednesday 18 June (McDonagh 1998, pp. 188, 194). 21. The Committee of the Regions, modelled on the Economic and Social Committee, was created in the Maastricht Treaty. 22. It should be noted that the German Presidency agenda was overwhelmed by the events in Kosovo, which took up valuable meeting time in the Council. 23. Second only to Denmark at 71.6 per cent (Commission 2001c). 24. One Greek official also noted that the Commission has a ‘bad effect’ of crowding out work that should be done by national governments and outbidding the government for capable staff. In the case of the NAPs, it may pay more (literally!) to be part of the national review teams of the European Commission than to take part in national working groups that create the NAPs in the first place. There have been cases of experts refusing work on creating NAPs only to show up writing the Commission critiques of the NAPs (Interview 16). 25. The attitude of the government, and of national officials and those participating in the labour market programmes, was described by one official as a ‘take the money and run mentality’ inspired by the instability of the previous decades (Interview 73). 26. Anyone familiar with Greek politics may question the use of the word ‘depoliticize’ in any context. It is employed here in a relative fashion to indicate that however much politicians may bluster about policies, the guidelines to be followed in the NAPs are set in Brussels and cannot be changed. 27. The Socialist government promised to introduce legislation to reduce the working week on 10 October 1997 as voluntary implementation was too slow (Blanpain 1998, p. 6). 28. More recently, the German Federal Labour Offices (the national agency for job-seekers and unemployment benefits) were found to have been grossly exaggerating statistics on success rates for finding employment. A review of five offices found that success rates for job-seekers as a result of the intervention of the Labour Offices was 18 per cent rather than the reported 51 per cent. Offices had claimed credit for jobs that individuals had found themselves. As there are 181 offices it is not known to what extent this problem may be endemic (FT 2002). The weakness of EU-level policies is highlighted by the dependence on national data which is not verified by Eurostat.
5
Global Governance of Social Policy: EU Member States at the International Labour Organization
1. Other regional mechanisms for the protection of social rights include the Inter-American Human Rights System under the Organization of American
Notes 223
2.
3.
4.
5. 6.
7.
States (OAS), the African Charter on Human and Peoples’ Rights of the Organization of African Unity (OAU), the Permanent Arab Commission on Human Rights of the League of Arab States and the labour clauses of the North American Free Trade Agreement (NAFTA). The United States, not a member of the League, joined the ILO in 1934 during the presidency of Franklin D. Roosevelt. Although the US left the ILO in 1977 during the Carter administration, it rejoined at the beginning of the Reagan administration in 1980. Other international organizations in which the EU is also a member, rather than an observer, include the Food and Agriculture Organization (FAO), the North West Atlantic Sea Fisheries Organization (NAFO) and international commodity agreements, such as the International Coffee Agreement and the Sugar Council. The ILO formerly had a ‘blacklist’ of worst-performing Member States. This was eliminated in 1980 as it was not seen to represent an incentive for poorly performing Member States to improve (Betten 1993b, p. 403). I am grateful to Christopher McCrudden, professor of International Labour Law at the University of Oxford, for this point. Norway, a non-EU Member State that may provide a point of comparison, ratified the same number of Conventions as Finland and Sweden (10) in 1987–2001. Norway also voted on becoming an EU Member State, but the proposition was rejected by a nationwide referendum in 1994. Norway has a total of 109 ratifications as of 2001. Le Conseil d’Administration du Bureau International du Travail invitera la Haute Autorité à se faire représenter au titre d’observateur aux réunions de la Commission de l’industrie charbonnière et de la Commission du fer et de l’acier de l’Organisation Internationale du Travail. (para. 7) The Agreement was signed by Jean Monnet, president of the High Authority of the ECSC, and David A. Morse, director-general of the ILO.
8. The chairman of this report, Bertil Ohlin (former Swedish Minister of Commerce 1944–5 and former Leader of the Liberal Party), went on to develop the highly influential Heckscher – Ohlin model of international trade, for which he won a Nobel prize. The other members of the Group of Experts on Social Aspects of Problems of European Economic Integration were: Mr Maurice Byé (France), Mr T. U. Matthew (UK), Mr Helmut Meinhold (Germany), Mr Pasquale Saraceno (Italy) and Mr Petrus J. Verdoorn (Netherlands). Mr Byé (France) wrote a dissenting opinion at the end of the report questioning the assumptions of the report, notably the immobility of capital and labour. Byé described this as ‘the most unrealistic of all [the possible scenarios]’, and suggested that the relative immobility of persons (labour) and certain goods combined with perfect mobility of capital and imperfect competition would suggest different conclusions (ILO 1956, pp. 123–4, 134). Byé noted that the threat of a flight of capital which has already been held over earlier attempts at social progress would nullify the efforts of the trade unions and governments. Under such circumstances it would be impossible to expect willing support for plans to integrate Europe … In practice, the least
224 Notes ambitious social legislation would tend to be taken as the standard. (ILO 1956, p. 129) 9. The invitation of the ILO to EU Committees has not been without contention. The ILO was invited to participate as an observer in several highly specific technical committees in the late 1960s and early 1970s, such as the consultative committee on social problems in maritime fishing and the consultative committee on social problems in internal transport. However, the formation of the Permanent Employment Committee in the 1970s caused a difference of opinion between the Commission, which wanted the ILO to be invited as an observer, and the Council of Ministers, which did not. A 1993 exchange of letters on ILO participation in Group of 24 meetings regarding eastern and central Europe, which are coordinated by the EU, between the ILO and Sir Leon Brittan (vice president of the Community and commissioner for DG1 Trade) was ‘far from satisfactory’ in the director-general’s (ILO) opinion (ILO 1993b). Sir Leon had noted that the ‘ILO will hence be actively involved in the deliberations and efforts of the Group of 24 concerning social aspects of reform programmes in the framework of countrylevel coordination’, but gave no indication as to how this might come about (ILO 1993b, Appendix: letter from Sir Leon Brittan dated 27 May 1993). The UK government representative to the Governing Body, as written in the minutes of the Governing Body session, noted that her government had approved ILO participation and that it was ‘extremely irritating to receive a letter of this kind from the Commission’ (ILO 1993a). 10. The ‘key figure’ identified was Mr Vinck, functionnaire at the Commission (ILO 1970a). In other notes, Mr Paul Finet of the High Authority is mentioned as a ‘personality’ who developed the cooperation between the ECSC and the ILO (ILO 1971b). Social Affairs Commissioner Coppé was described as more ‘conservative’ in his views of social policy (ILO 1971c). 11. It should be recalled that it is ‘standing case law of the Court that EC Member States cannot undertake international commitments which would be in conflict with (or even affect) Community law’ (Betten 1993a, p. 255). While art. 141 and judgements by the ECJ hold that a ban on night work for women in industry should no longer be permitted as part of rules on equal treatment, the 1958 ILO Convention No. 89 on Night Work (Women) (Revised) sets different guidelines for the work of women at night than those for men (Valticos 1993, p. xix). This ruling ‘was later softened’ in Case C-158/91 Levy (Betten 1993a). 12. It is interesting to note that Member States have also resisted expanding the competence of the Commission on new trade issues at the WTO, where operating as a bloc is far more established, on issues such as trade in services (Meunier and Nicolaïdis 1999). Exceptions to the Commission’s leadership of trade policy at the WTO are negatively listed, however, and are very few in number compared with the lack of any areas of exclusive EU competence in the social field at the ILO. 13. I am grateful to Mr Don MacPhee at the Canadian Mission in Geneva, chair of IMEC (2002), for his detailed explanation of IMEC procedures. It is interesting to note that another Canadian, John Mainwaring, then Director of International Relations at Labour Canada, was largely responsible for the conception and creation of IMEC.
Notes 225 14. Belgium, Denmark, Finland, France, Germany, Greece, Italy, Netherlands, Portugal, Spain, Sweden, UK. ILO Convention 115 was not ratified by Austria, Ireland and Luxembourg. 15. Directive 80/1107/EEC of November 1980 on the protection of workers from the risks related to exposure to chemical, physical and biological agents at work, amended by Directive 88/642/EEC, Directives 79/831/EEC and 88/379/ EEC on the labelling and packaging of dangerous substances, and the Framework Directive 89/391/EEC. 16. The 1984 complaint by the TUC to the ILO under Convention No. 87 (Freedom of Association) concerned Thatcher’s decision that trade union membership at the British Government Communications Headquarters at Cheltenham (GCHQ) posed a threat to national security, making it a condition of employment that all personnel at GCHQ did not belong to a union or staff association other than the internal and approved staff association (Betten 1993, p. 74). The Thatcher government also rescinded various ‘fair-wage’ resolutions (breaking the terms of ILO Convention No. 94, 1982) and (in 1986) modified minimum wage-fixing machinery, removing employees under 21 from cover and ending control over working conditions (entailing the deratification of ILO Convention No. 26). (Rhodes 2000, p. 46) A case by the DGB (German Confederation of Trade Unions) was also brought to the ILO on restrictions to the right to strike by teachers and postal workers. This was appealed by the German government, but not brought before the ICJ as, in the opinion of the German government, the ICJ had not yet taken a position on the right to strike and ILO Conventions (ILO 1991). 17. Strangely, the ICITO continued to exist until the creation of the WTO in 1995 in order to entitle its members and the GATT Secretariat to certain benefits accrued to the ICITO during its two years of existence, such as a contractual personality and staff pension benefits under the UN system. 18. It should be noted that the 1996 Agreement on cooperation with the World Bank and IMF was hard-fought by the then WTO Director General Renato Ruggerio (Interview 46), and that cooperation between the World Bank and the WTO Secretariat, such as on Trade Policy Reviews of WTO Member States and on development issues related to the Doha Development Agenda, are not consistent. Inter-institutional relations at the global level are still highly complex, and by no means well sorted out. 19. From the ILO perspective, using trade sanctions to enforce core labour standards ‘may not be the best way to promote standards’ in the first instance, and in addition, ‘the international community is not ready to grant sanctions to any institution’ other than the WTO (Interview 43). The route to ratification and respect of core labour standards through trade-related measures will not happen in the foreseeable medium term via either the WTO or unilateral ILO action in trade-related areas; for example, ILO Officials were ‘shot down in flames’ by ILO members when the idea of ILO leadership on the labelling of products to indicate their compliance with production processes that respect core labour standards was suggested. In
226 Notes short, the trade route to enforced labour standards is blocked and, in the view of many officials, is not the most desirable road anyway. 20. The 2000 Cotonou Agreement between the EU and the AfricanCaribbean-Pacific (ACP) 77 also suggested that EU–ACP cooperation should aim at ‘encouraging the promotion of participatory methods of social dialogue as well as respect of basic social rights’ (Commission 2001a, p. 12 and Annex 6). 21. The initial request for consultations on the EU’s requirements for GSP was submitted by Brazil in 2000 concerning drug law enforcement and special treatment for soluble coffee, but was later dropped (WTO 2000). The Thai and Indian request for consultations with the EU on labour and environmental standards, and drug law enforcement requirements to qualify for GSP, was joined in 2001–2 by several Central American governments: Costa Rica, Guatemala, Honduras and Nicaragua.
6
Conclusion
1. The first ‘wave’ of enlargement in May 2004 was to Cyprus, Czech Republic, Estonia, Hungary, Latvia, Lithuania, Malta, Poland, Slovakia and Slovenia. The next wave of countries may include Bulgaria and Romania. Turkey is also a candidate country. The EU has negotiated Stabilization and Association Agreements with several Balkan states (Albania, Croatia, FYROM) and may do so for others (Bosnia-Herzegovina, Serbia and Montenegro). The limits of the EU are set by the requirements of new Member States to possess market economies, democratic political systems, a respect for human rights, and the rule of law and their ability to absorb and implement the acquis communautaire. 2. Representatives to the Agency’s Administrative Board are ‘Member State representatives from each of the three key stakeholder groups – governments, employers and worker organizations, as well as three European Commission representatives’ (Agency 2001, p. 6). In 2000 there was no Greek member of staff at the Agency, and 17 per cent of its staff were British. There was one Swede in the 35-member Agency. While no strong correlation can be made between Member State success in implementing legislation and members of staff at the Agency, it is still interesting to note that Greek officials continue to be underrepresented or absent from health and safety aspects of EU decision-making. 3. The research presented did not set out to test neo-functional assertions about spillover (Haas 1964; Wessels 1997; Jensen 2000). However, it is clear from the observations of Chapters 2 through 4 that spillover has been highly limited in the social policy field in the EU, and this may be of interest to researchers attempting to refine theories of the cases in which spillover occurs, and how this happens.
Appendix: List of Interviews I. National level (including national officials in Brussels) (i) Politicians Greece 1. Ms Marietta Giannakou-Koutsikou, Member of the Greek Parliament, New Democracy Party, and former minister.
Sweden 2. Goran Lenmarker, Member of Parliament (Riksdagen) and spokesman on foreign affairs for the Moderate Party. 3. Martin Nilsson, Member of Parliament (Riksdagen) and member of the Labour Market Committee, Social Democratic Workers’ Party. 4. Soren Lekbert, Member of Parliament (Riksdagen) and chairman of the Committee on EU Affairs, Social Democratic Workers’ Party. 5. Susanne Eberstein, Member of Parliament (Riksdagen) and member of the Social affairs Committee, Social Democratic Workers’ Party.
UK 6. John Major, British prime minister 1990–7. 7. Anthony Smith, Member of Parliament (Labour Party) and former liaison to the Party of European Socialists.
(ii) National bureaucrats, civil servants (a) Treaty negotiators Greece 8. Mr Constantinos Papadopoulos, adviser for European affairs, EFG Eurobank, Athens, and member of the Greek delegation to the Reflection Group and the 1996–7 Intergovernmental Conference (IGC). 9. Dr Takis Ioakimidis, adviser to the Ministry of Foreign Affairs and the Prime Minister’s Office on the European Union, negotiator for the Greek government at the Nice Treaty (2000) and president of the Hellenic Centre for European Studies (EKEM).
(b) Permanent representatives (Brussels) Sweden 10. Fredrik Sjogren, labour law and employment officer, Swedish Permanent Representation to the EU. 11. Johanna Möllerberg, health and safety officer, Swedish Permanent Representation to the EU. 227
228 Appendix
Germany 12. Eckehart Hagen, labour and social affairs officer, German Permanent Representation to the EU.
(c) EU committee members Greece 13. Mr Dimitrios Politis, member of the board of directors of GSEE (Greek General Confederation of Labour) and vice-president of OKE (Economic and Social Council of Greece), and member of the Advisory Committee on Hygiene, Health and Safety. 14. Former member of the Economic Policy Committee (EPC) of the EU and the Monetary Policy Committee (MPC) of the EU.
Sweden 15. Lennart Ahnstrom, chairman of the EU Council Working Group on Machinery under the Swedish Presidency of the EU and head of the Machinery Department at the Swedish Work Environment Authority.
(d) Other civil servants and government advisers Greece 16. Dr Platon Tinios, special adviser to the prime minister of Greece on social policy. 17. Professor Nikos Koutsiaras, adviser to the Ministry of Labour and Professor at the University of Athens. 18. An adviser to the Ministry of Labour and Social Security.
Sweden 19. A Swedish official. 20. Kerstin Wahlberg, Chemistry and Microbiology Division at the Swedish Work Environment Authority. 21. Bengt Johansson, principal administrative officer, Machinery and Personal Protective Equipment Division of the Swedish Work Environment Authority. 22. Ingela Ronn, press secretary, Swedish Work Environment Authority. 23. Anders Lindberg, political adviser for EU Affairs to the Social Democratic Party.
UK 24. An official in the EU Division of Department for Education and Employment.
(iii) National representatives to international organizations UK 25. A UK official. 26. A UK official.
Other states 27. A German official. 28. An American official.
List of Interviews 229 29. A Canadian official. 30. A Canadian official. 31. A Canadian official.
II. EU level: commissioners, EU civil servants 32. Allan Larsson, former Directorate-General of DG V (now DG Employment and Social Affairs), and former Swedish finance minister [via email]. 33. Patrick Venturini, secretary-general of the Economic and Social Committee and former European Commission adviser on Social Dialogue and Social Policy in Jacques Delors’s cabinet. 34. A Commission official. 35. A Commission official. 36. A Commission official. 37. A Commission official. 38. A Commission official. 39. A Commission official. 40. A Commission official.
III. Global level: civil servants of international organizations ILO 41. 42. 43. 44. 45.
An An An An An
ILO ILO ILO ILO ILO
official. official. official. official. official.
WTO 46. A WTO official. 47. A WTO official.
IV. Interests: the social partners (i) Business and employer groups Greece 48. Ms Anastasia Koutsivitou, head of Social Affairs Department, SEV (Federation of Greek Employers). 49. Mr Motsos, president of the Hellenic Confederation of Professionals, Craftsmen and Merchants [formerly the Greek Confederation of Small and Medium Sized Enterprises] (GSEVEE).
Sweden 50. Christian Bratt, director, international affairs, SAF (Swedish Employers Confederation, now the Confederation of Swedish Enterprise).
230 Appendix 51. Olof Allgardh, director, Brussels office, SAF (Swedish Employers Confederation, now the Confederation of Swedish Enterprise).
UK 52. Susan Anderson, head of social and labour market policy, Confederation of Business Industry (CBI), and member of the UNICE negotiating team for Social Dialogue on part-time work.
UNICE (Brussels) Contacted but could not arrange an interview time.
ILO Bureau for Employers’ Activity (Geneva) 53. Deborah France, deputy-director, Bureau for Employers’ Activity, International Labour Office.
(ii) Labour and employee groups Greece 54. Mr Dimitrios Politis, member of the board of directors of GSEE (Greek General Confederation of Labour) and vice-president of OKE (Economic and Social Council of Greece) [note, also listed as an EU Committee member, #13].
Sweden 55. Leif Hagg, International Department, LO (Swedish Trade Union Confederation). 56. Bo Ronngren, ombudsman, LO (Swedish Trade Union Confederation). 57. Anders Schaerstrom, programme secretary, SACO (Swedish Confederation of Professional Associations). 58. Ake Zettermark, labour market policy, SACO (Swedish Confederation of Professional Associations). 59. Sven Erik Jansson, press officer, SACO (Swedish Confederation of Professional Associations). 60. Leif Dergel, labour market secretary, TCO (Swedish Confederation of Professional Employees). 61. Keth Zettervall-Thapper, international secretary, TCO (Swedish Confederation of Professional Employees).
UK 62. David Feickert, Trades Union Congress (TUC) (UK) representative to the ETUC, and member of the ETUC negotiating team for Social Dialogue on part-time work.
ETUC (Brussels) 63. David Foden, research officer, European Trade Union Institute (EUTI), Brussels. 64. Amparo Serrano, research officer, European Trade Union Institute (ETUI), Brussels.
List of Interviews 231
(iii) Other members of social partner organizations 65. Dr Gregory Papanikos, secretary-general, OKE (Economic and Social Council of Greece). 66. Mr Yannis Koyzis, scientific consultant to the Labour Institute of GSEE-ADEDY and professor of industrial relations.
V. Academics 67. Dr Mats Benner, University of Lund. 68. Dr Lena Gonas, expert on gender equality in the workplace, Arbetslivsinstitutet (National Institute for Working Life). 69. Kerstin Ahlberg, researcher on the EU Employment Strategy, Arbetslivsinstitutet (National Institute for Working Life). 70. Dr Nicklaus Bruun, head of labour law research, Arbetslivsinstitutet (National Institute for Working Life). 71. Dr Panos Tskagloglou, professor, Athens School of Economics and Business (expert on Greek welfare policies, social exclusion, social assistance for the elderly). 72. Dr George Pagalotos, professor, Athens School of Economics and Business (expert on the history of the Greek welfare state). 73. Professor Manos Matsaganis, former special adviser to the prime minister on social policy and professor at the University of Crete. 74. Dr Calliope Spanou, professor, Athens School of Economics and Business (expert on Greek public administration). 75. Dr Dimitris Sotiropoulos, professor, University of Athens (expert on the history of the Greek welfare state). 76. Dr Christos Ioannou, director of research for the Hellenic Institute for Occupational Health and Safety, Athens.
Bibliography —— (1985) Agence Europe. Brussels. —— (1999) French Cut Hours to Create Jobs. Labour Research (March). Addison, J. and W. S. Siebert (eds) (1997) Labour Markets in Europe: Issues of Harmonization and Regulation. London, The Dryden Press. Addison, J. T. and W. S. Siebert (1999) Regulating European Labour Markets: More Costs than Benefits? London, Institute of Economic Affairs. Agency (2001) Annual Report 2001 of the European Agency for Safety and Health at Work. Luxembourg, European Agency for Safety and Health at Work: 63. Algieri, F. and J. Janning (1996) The German Debate, in F. Algieri, H. Grabbe, K. Hughes et al. (eds). The 1996 IGC – National Debates (2): Germany, Spain, Sweden and the UK. London, Royal Institute of International Affairs: 1–21. Anxo, D., S. Carcillo and C. Erhel (2001) Aggregate Impact Analysis of Active Labour Market Policy in France and Sweden: A Regional Approach, in H. Mosley and J. de Koning (eds). Labour Market Policy and Unemployment: Impact and Process Evaluations in Selected European Countries. Cheltenham, Edward Elgar: 49–76. Argandona, A. (1997) Spain and the European Social Charter: Social Harmonization with Unemployment and High Wage Growth, in J. Addison and W. S. Siebert (eds) Labour Markets in Europe: Issues of Harmonization and Regulation. London, The Dryden Press: 191–211. Armstrong, K. and S. Bulmer (1998) The Governance of the Single European Market. Manchester University Press. Ashford, D. E. (1986) The Emergence of the Welfare States. Oxford, Blackwell. Aspinwall, M. and G. Schneider (2001) Institutional Research on the European Union: Mapping the Field, in M. Aspinwall and G. Schneider (eds) The Rules of Integration: Institutionalist Approaches to the Study of Europe. Manchester University Press: 1–18. Austria (1996) Guidelines of the Austrian Government on the Subjects Likely to be Dealt with at the 1996 IGC (submitted in June 1995). White Paper on the 1996 Intergovernmental Conference: Summary of Positions of the Member States of the European Union with a view to the 1996 IGC. Brussels, European Parliament: II, 124–34. Axt, H. -J. (1997) Greece’s Accession to the EC and its Economic Consequences: A Comparative View, in H.-J. Axt (ed.) Greece and the European Union: Stranger Among Partners? Baden-Baden, Nomos Verlagsgesellschaft: 23–43. Bahrena, L., L. Delander and H. Niklasson (2001) Towards Intensified Local Level Co-operation in the Design and Implementation of Labour Market Policies: An Evaluation of Some Swedish Experiments and Reforms, in J. de Koning and H. Mosley (eds) Labour Market Policy and Unemployment: Impact and Process Evaluation in Selected European Countries. Cheltenham, Edward Elgar: 256–90. Balassa, B. (1961) The Theory of Economic Integration. London, Allen & Unwin.
232
Bibliography 233 Baldwin, R. (1992) The United Kingdom, in R. Baldwin and T. Daintith (eds) Harmonization and Hazard: Regulating Workplace Health and Safety in the EC. London, Graham & Trotman: 203–21. Baldwin, R. (1996) Regulatory Legitimacy in the European Context: The British Health and Safety Executive, in G. Majone (ed.) Regulating Europe. London, Routledge: 83–105. Baldwin, R. and T. Daintith (1992) The European Framework, in R. Baldwin and T. Daintith (eds) Harmonization and Hazard: Regulating Workplace Health and Safety in the EC. London, Graham & Trotman: 1–17. Banoutsos, I. (1994) Legislative Provisions and Experience in Greece of Participation in Health and Safety, ETUC. Banting, K. (1995) The Welfare State as Statecraft: Territorial Politics and Canadian Social Policy, in S. Leibfried and P. Pierson (eds) European Social Policy: Between Fragmentation and Integration. Washington, Brookings Institution: 269–300. Belgium (1955) Note du Gouvernement Belge sur les Ratifications des Conventions Internationales du Travail. MAE. 128f/55 iv, European Council Archives. Belgium (1996) Memorandum on the IGC from the Governments of Belgium, Luxembourg and the Netherlands, 7 March 1996. White Paper on the 1996 Intergovernmental Conference: Summary of Positions of the Member States of the European Union with a View to the 1996 IGC. Brussels, European Parliament: II, 25–9. Benner, M. and T. Bundgaard Vad (2000) Sweden and Denmark: Defending the Welfare State, in F. W. Scharpf and V. A. Schmidt (eds) Welfare and Work in the Open Economy: Diverse Reponses to Common Challenges. Oxford University Press: II, 399–466. Bercusson, B. (1994) The Dynamics of European Labour Law after Maastricht. Industrial Law Journal 23(1): 1–31. Bercusson, B. and N. Bruun (1999) The Agreement on Fixed-Term Work: A First Analysis, in K. Ahlberg, B. Bercusson, N. Bruun and C. Vigneau (eds) FixedTerm Work in the EU: A European Agreement Against Discrimination and Abuse. Stockholm, National Institute for Working Life (Arbetslivsinstituet): 51–131. Berger, S. (2002) Germany in Historical Perspective: The Gap Between Theory and Practice, in S. Berger and H. Compston (eds) Policy Concertation and Social Partnership in Western Europe: Lessons for the 21st Century. Oxford, Berghahn: 125–38. Best, G. (1994) The United Kingdom and the Ratification of the Maastricht Treaty, in F. Laursen and S. Vanhoonacker (eds) The Ratification of the Maastricht Treaty: Issues, Debates and Future Implications. Maastricht, European Institute of Public Administration: 245–78. Betten, L. (1993a) Court of Justice of the European Communities, Opinion 2/91 in Relation to Convention No. 170 of the International Labour Organisation Concerning Safety in the Use of Chemicals at Work. International Journal of Comparative Labour Law and Industrial Relations (Autumn): 244–57. Betten, L. (1993b) International Labour Law: Selected Issues. Boston, Kluwer. Beveridge, W. H. (1942) Social Insurance and Allied Services. London, HMSO Cond. Senes. Bieler, A. (2000) Globalisation and Enlargement of the European Union: Austrian and Swedish Social Forces in the Struggle over Membership. London, Routledge.
234 Bibliography Blanpain, R. (1998) The European Union, Employment, Social Policy and the Law, in R. Blanpain (ed.) Institutional Changes and European Social Policies after the Treaty of Amsterdam. The Hague, Kluwer: 1–66. Bonoli, G. and B. Palier (1999) Framework to citizenship? Current transformations in the French welfare state, in J. Bussemaker (ed.) Citizenship and Welfare State Reform in Europe. London, Routledge: 42–57. Branch, A. and J. Greenwood (2001) European Employers: Social Partners, in H. Compston and J. Greenwood (eds) Social Partnership in the European Union. Basingstoke, UK, Palgrave: 41–70. Brian, A. W. (1989) Competing Technologies, Increasing Returns, and Lock-In by Historical Events. Economic Journal, 99: 116–31. Briggs, A. (1961) The Welfare State in Historical Perspective. European Journal of Sociology 2: 221–58. Brown, T. (1995) Sweden in the European Union. Dublin, Institute of European Affairs. Bruun, N. (2001) The European Employment Strategy and the ‘Acquis Communitaire’ of Labour Law. The International Journal of Comparative Labour Law and Industrial Relations 17(2): 309–24. Budden, P. (2002) Observations on the Single European Act and the ‘Relaunch of Europe’: A less ‘Intergovernmental’ Reading of the 1985 Intergovernmental Conference. Journal of European Public Policy 9(1): 76–97. Budden, P. M. (1994) The Making of the Single European Act: The United Kingdom and the European Community, 1979–1986, Faculty of Modern History, University of Oxford. Bulmer, S. J. (1998) New Institutionalism and the Governance of the Single Market. Journal of European Public Policy 5(3): 365–86. Burda, M. C. (1997) Persistently High Irish Unemployment: A Comparison with the UK, in A. W. Gray (ed.) International Perspectives on the Irish Economy. Dublin, Indecon Economic Consultants: 85–111. Bussemaker, J. (1999) Against Social Solidarity and Citizenship: Justifying Social Provision in Britain and France, in J. Bussemaker (ed.) Citizenship and Welfare State Reform in Europe. London, Routledge: 27–41. Byre, A. (1993) Trade Unions and Employers in the Single Europe. London, Policy Studies Institute. Calmfors, L. and P. Skedinger (1995) Does Active Labour Market Policy Increase Employment? Theoretical Considerations and Some Empirical Evidence from Sweden. Oxford Review of Economic Policy 11(2): 91–109. Casey, B. and M. Gold (2000) Social Partnership and Economic Performance: The Case of Europe. Cheltenham, Edward Elgar. CBI (1996) Building a Europe that Works: An Agenda for the IGC. London. CEEP, ETUC, UNICE (1991) Agreement of 31 October 1991. Brussels. Chaumette, P. (1992a) ‘Commentaire sur la loi du 31 decembre 1991 relative aux obligations de l’employeur et du salarie en matière de securité au travail.’ Droit Social 4. Chaumette, P. (1992b) France, in R. Baldwin and T. Daintith (eds) Harmonization and Hazard: Regulating Health and Safety in the European Workplace. London, Graham & Trotman: 19–45. Christiansen, T., G. Falkner and K. E. Jorgensen (2002) Theorizing EU Treaty Reform: Beyond Diplomacy and Bargaining. Journal of European Public Policy 9(1): 12–43.
Bibliography 235 Clauwaert, S. (2001) Survey on the Implementation of the Part-Time Work Directive/agreement in the EU Member States and Selected Applicant Countries (Interim Report). Brussels, European Trade Union Institute. Clogg, R. (1993) Introduction: The PASOK phenomenon, in R. Clogg (ed.) Greece, 1981–89: The Populist Decade. London, Macmillan: viii–xiii. Cockfield (1994) The European Union: The Single Market. Chichester, Chancery. Cohen-Tanugi, L. (1996) The French Debate, in L. Cohen-Tanugi, S. Fagiolo, M. Kwast-van Duursen and A. C. G. Stubb (eds) The 1996 IGC – National Debates (1) Finland, France, Italy and the Netherlands. London, Royal Institute of International Affairs 1: 23–35. Coldrick (1991) European Trade Unions and the Protection of Workers’ Rights, LSE Research Seminar, paper quoted in Rhodes (1991a) on fn. 358, p. 23. Collins, D. (1975a) The European Communities: The Social Policy of the First Phase, the European Coal and Steel Community 1951–1970. Vol I. London, Martin Robertson. Collins, D. (1975b) The European Communities, The Social Policy of the First Phase: The European Economic Community 1958–72. Vol II. London, Martin Robertson. Collins, D. (1983) The Operation of the European Social Fund. London, Croom Helm. Colucci, M. (1998) Searching for a European Employment Strategic Initiative. Institutional Changes and European Social Policies after the Treaty of Amsterdam. R. Blanpain. London, Kluwer: 101–56. Commission (1974) Social Action Programme. Bulletin of the European Communities Supplement 2/74. Brussels, Commission of the European Communities: 11–36. Commission (1978) La Coopération au Developpment et le Respect de Certaines Normes Internationales en Matière de Conditions de Travail. Brussels. COM (78) 492 final, Commission of the European Communities. Commission (1979) Recommendation for a Council Decision Authorizing the Commission to Participate in the Sixty-Fifth International Labour Conference on the Negotiations for the Revision of ILO Convention No. 67 on Hours of Work and Rest Periods in Road Transport. Brussels. COM (79) 301 final, Commission of the European Communities. Commission (1985) EC Bulletin. No. 9. Brussels, Commission of the European Communities. Commission (1986) Guidelines for the Management of the European Social Fund for 1987–89. Social Europe. Luxembourg, Directorate-General for Employment and Social Affairs of the European Commission 3/86: 49–52. Commission (1988a) Proposition de Directive du Conseil Concernant la Mise en Oeuvre de Mesures pour Promouvoir l’Amélioration de la Securité et de la Santé des Travailleurs sur le Lieu de Travail. Brussels. COM (88) 73 final. Commission (1988b) Rapport ‘La Dimension Sociale du Marché Interieur’. EC Bulletin 1988/12: 9–10 Brussels; also reproduced in Venturini (1988) as Annex 7. Commission (1989) Communication from the Commission Concerning Its Action Programme Relating to the Implementation of the Community Charter of Basic Social Rights for Workers. Brussels, Commission of the European Communities. COM (89) 568 final. Commission (1990) Exchange of Letters with the ILO, OJ C 24/8, 1 February 1990. Brussels, Commission of the European Communities.
236 Bibliography Commission (1991a) Keys to the European Social Fund. Social Europe. Luxembourg, Directorate-General for Employment and Social Affairs of the European Commission, Vol. 2: 10. Commission (1991b) Proposals During Initial Phase of the Political Union IGC (Selection), D. European Commission Proposal on the Social Dimension and the Development of Human Resources (in documents annex), in R. Corbett (ed.) The Treaty of Maastricht, From Conception to Ratification: A Comprehensive Reference Guide. Harlow, Longman: 218–43. Commission (1993a) Community-Wide Framework for Employment. Brussels, Commission of the European Communities. COM (93) 238 final. Commission (1993b) Commission Decision 22 October 1993 on the creation of EURES (the European Employment Service) (93/569/EEC). Brussels, Commission of the European Communities. OJ L 274 06.11.93, p. 32. Commission (1993c) Committee of Senior Labour Inspectors. Social Europe. Directorate-General for Employment and Social Affairs of the European Commission. 3/93, Vol. 3. 13: Brussels, Commission of the European Communities. Commission (1993d) White Paper on Growth, Competitiveness and Employment: The Challenges and Ways Forward into the 21st Century. Brussels, Commission of the European Communities. COM (93) 700 final. Commission (1994) Proposal for a Council Decision on the Exercise of the Community’s External Competence at International Labour Conferences in Cases Falling Within the Joint Competence of the Community and its Member States. COM (94) 2 final. Brussels, Commission of the European Communities. Commission (1995a) Medium Term Social Action Programme. Brussels, Commission of the European Communities. COM (95) 134. Commission (1995b) Major Landmarks in the Social Dialogue. Social Europe 2. Luxembourg, Directorate-General for Employment and Social Affairs of the European Commission, Vol. 2: 13–14. Commission (1995c) The Trade System and Internationally Recognized Labour Standards (non-paper). Brussels, Commission of the European Communities. Commission (1996) Labour Market Studies: Sweden. Brussels, DG Employment and Social Affairs (Swedish Institute for Social Research and ECOTEC Research and Consulting). Commission (1997a) The Way Forward: The European Employment Strategy, Contributions to and Outcome of the Dublin European Council, 13 and 14 December 1996. Brussels, DG Employment and Social Affairs. Commission (1997b) Commission Communication: Proposal for Guidelines for Member States Employment Policies. Brussels, Commission of the European Communities. Commission (1997c) Transmission Note, Subject: Commission Contribution on the Integration of the Social Protocol, Incentive Measures in the Social Area. Brussels, Commission of the European Communities. CONF/3904/97. Commission (1997d) Etude sur le marche du travail: France. Brussels, DG Employment and Social Affairs (Bernard Brunhes Consultants). Commission (1997e) Labour Market Studies: UK. Brussels, DG Employment and Social Affairs (Peter Robinson, Centre for Economic Performance, LSE). Commission (1997f) The 1998 Employment Guidelines. Council Resolution of 15 December 1997. Brussels, Directorate-General for Employment, Industrial Relations and Social Affairs.
Bibliography 237 Commission (1998a) National Transposition Measures: Situation at 1st January 1998 (Secretariat General of the Directorate-General for Employment, Industrial Relations and Social Affairs). Luxembourg, European Communities. Commission (1998b) Employment Policies in the EU and in the Member States: Joint Report 1998. Brussels, European Commission Directorate-General for Employment and Social Affairs and the European Council (Social Affairs and ECOFIN). Commission (1998c) Brussels, Commission of the European Union. COM (90) 228 final. Commission (1999a) Employment in Europe. Brussels, DG Employment and Social Affairs, Unit V/A.1. Commission (1999b) Affirming Fundamental Rights in the European Union: Time To Act, Report of the Expert Group on Fundamental Rights. Brussels. DG for Employment, Industrial Relations and Social Affairs Unit V/D.2. Commission (2000a) Contribution de la Commission Européenne au Conseil Euopeen Special de Lisbon des 23 et 24 Mars 2000. Brussels. Commission (2000b) Industrial Relations in Europe 2000. Luxembourg, European Commission (DG Employment and Social Affairs, Directorate D). Commission (2001a) Communication from the Commission to the Council, the European Parliament and the Economic and Social Committee Promoting Core Labour Standards and Improving Social Governance in the Context of Globalization. Brussels, Commission of the European Communities. COM (2001) 416 final. Commission (2001b) Letter from Commissioner Diamantopoulou to Mr Juan Somavia, Director General of the ILO. Brussels, DG Employment and Social Affairs: 4.EMPL991388495. Commission (2001c) Employment in Europe 2001: Recent Trends and Prospects. DG Employment and Social Affairs. Commission (2002) Taking Stock of Five Years of the European Employment Strategy: Communication from the Commission to the Council, the European Parliament, the Economic and Social Committee, and the Committee of the Regions. Brussels, Commission of the European Communities. Compston, H. (2001a) Introduction, in H. Compston and J. Greenwood (eds). Social Partnership in the European Union. Basingstoke, Palgrave: 1–10. Compston, H. (2001b) The Intergovernmental Dimension of EU Social Partnership, in H. Compston and J. Greenwood (eds). Social Partnership in the European Union. Basingstoke, Palgrave: 98–128. Corbett, R. (1998) The European Parliament’s Role in Closer EU Integration. London, Macmillan. Council (1965) Council decision of 11 March 1965, Journal Officiel, no. 46. Council (1974) Council Resolution Concerning a Social Action Programme. 21 January 1974. Brussels: Council of the European Communities. OJ (Official Journal) C 013 12/02/1974, pp. 1–4. Council (1979a) Information Note by the General Secretariat for the Ad Hoc Joint Working Party on Transport and Social Questions, Subject: Hours of Work and Rest Periods (Road Transport) Convention and Recommendation, 1979. Brussels, General Secretariat of the Council of the European Communities (7890/79). Council (1979b) Outcome Meeting of Ad Hoc Joint Working Party on Transport and Social Questions, Subject: Hours of Work and Rest Periods (Road Transport)
238 Bibliography Convention and Recommendation, 1979. Brussels: Council of the European Communities (8219/79). Council (1985) New Approach to Technical Harmonization and Standardization (Council Resolution 7.5.1985 COM (85) 310 final). Brussels, Council of the European Communities. Council (1986) Note Point ‘A’ du Comite des Representants Permanents (1ère partie) (réunion du 19 decembre 1986) au Conseil (session du 22 decembre 1986). Brussels, Council of the European Communities (10862/1/86). Council (1989a) Madrid European Council Conclusions. European Council. June 1989. Council (1989b) Rhodes European Council, Greek Presidency Conclusions. European Council. December 1989. Council (1990) Conclusions of the European Council Meeting Held in Rome, 14–15 December 1990 (in documents annex), in R. Corbett (ed.) The Treaty of Maastricht, From Conception to Ratification: A Comprehensive Reference Guide. Harlow, Longman: 101–3. Council (1991a) Luxembourg Presidency ‘Non-paper’, Draft Treaty with a View to Achieving Political Union, 12 April 1991 (in documents annex), in R. Corbett (ed.) The Treaty of Maastricht, From Conception to Ratification: A Comprehensive Reference Guide. Harlow, Longman: 267–88. Council (1991b) Luxembourg Presidency ‘Draft Treaty on the Union’, 18 June 1991 (in documents annex), in R. Corbett (ed.). The Treaty of Maastricht, From Conception to Ratification: A Comprehensive Reference Guide. Harlow, Longman: 293–320. Council (1991c) Dutch Presidency Draft Treaty ‘Towards European Union’; 24 September 1991 (in documents annex), in R. Corbett (ed.) The Treaty of Maastricht, From Conception to Ratification: A Comprehensive Reference Guide. Harlow, Longman: 329–340. Council (1991d) Dutch Draft Union Treaty 8 November 1991 (in documents annex), in R. Corbett (ed.) The Treaty of Maastricht, From Conception to Ratification: A Comprehensive Reference Guide. Harlow, Longman. Council (1992) Denmark and the Treaty on European Union, Edinburgh European Council, Conclusions of the Presidency, Part B. 11 and 12 December. European Council. Council (1994) Essen European Council Presidency Conclusions. European Council. 16 December 1994 (00300/94). Council (1995) Madrid European Council Presidency Conclusions. European Council. 16 December 1995 (00400/95). Council (1996a) Turin European Council Presidency Statement. European Council. 29 March 1996 (SN/00/1/96). Council (1996b) Presidency Note of the Irish Presidency of the European Council held in Luxembourg on 1 October 1996 on Employment. 25 September 1996. Brussels, Council Archives CONF/3923/96 CAB. Council (1996c) Dublin European Council. European Council. 16 December 1996. Brussels (SN 401/1/96 and 00401-x/96). Council (1996d) Dublin II: Conference of the Representatives of the Governments of the Member States, The European Union: Today and Tomorrow. Dublin (CONF 2500/96). Council (1996e) 1996 Intergovernmental Conference (IGC ‘96): Reflection Group Report and Other References for Documentary Purposes. Luxembourg, General Secretariat of the Council of the European Union.
Bibliography 239 Council (1997a) Declaration to the Final Act on Incentive Measures. Amsterdam, European Council. June 1997. Council (1997b) Luxembourg Extraordinary European Council Meeting on Employment. Luxembourg, European Council. 20 November 1997 (SN 300/97). Council (1998) Cardiff European Council: Presidency Conclusions. European Council. 16 June 1998 Cardiff (00 150/1/98 REV 1). Council (2000a) Council Recommendation of 14 February 2000 on the Implementation of Member States’ Employment Policies. Brussels, Council of the European Union. Council (2000b) Lisbon European Council Presidency Conclusions: European Council. 24 March 2000 (100/1/00). Council (2000c) Santa Maria Da Feira: Presidency Conclusions. European Council. 8 December 2000 (400/1/00). Council (2001a) Council Decision of 19 January 2001 on Guidelines for Member States’ Employment Policies for the year 2001. Brussels, Council of the European Union. Council (2001b) Council Recommendation of 19 January 2001 on the Implementation of Member States’ Employment Policies. Brussels, Council of the European Union. Council (2001c) 2397th Council Meeting, General Affairs. Brussels, European Council (15078/01). Council (2001d) European Council Meeting Laeken: Presidency Conclusions, 14 December 2001 (00300/1/01). Council (2002) Presidency Conclusions: Barcelona European Council. 16 March. Council of the European Union (100/1/02). Cousins, C. (1987) Controlling Social Welfare: A Sociology of State Welfare, Work and Organizations. Brighton, Wheatsheaf. Cowles, M. G. (1995) Setting the Agenda for a New Europe: The ERT and EC 1992. Journal of Common Market Studies 33(4): 501–26. Cowles, M. G. (2002) Large Firms and the Transformation of EU Business Associations: A Historical Perspective, in J. Greenwood (ed.) The Effectiveness of EU Business Associations. Basingstoke, Palgrave: 64–78. Craig, P. and G. De Burca (2003) EU Law: Text, Cases, and Materials. Oxford University Press. Crijns, L. H. (1987) The Social Policy of the European Community. Social Europe 1(87): 51–62. Daly, M. (2000) A Fine Balance: Women’s Labour Market Participation in International Comparison, in F. W. Scharpf and V. A. Schmidt (eds) Work and Welfare in the Open Economy: Diverse Responses to Common Challenges. Oxford University Press, 2: 467–510. David, P. (1975) Technical Choice, Innovation and Economic Growth. London, Cambridge University Press. Davidson, A. (1989) Two Models of Welfare: The Origins and Development of the Welfare State in Sweden and New Zealand, 1888–1988. Stockholm, Almqvist & Wiksell. de Koning, J., H. Mosely and G. Schmid (2001) Introduction: active labour market policies, social exclusion, and transitional labour markets, in J. de Koning and H. Mosely (eds) Labour Market Policy and Unemployment. Cheltenham, Edward Elgar: 1–16. De Ruyt, J. (1989) L’Acte unique Européen: Commentaire. Brussels, Editions de l’Université de Bruxelles.
240 Bibliography Deacon, B. (2001) International Organizations, the EU and Global Social Policy, in B. Palier, P. Prior and R. Sykes (eds) Globalization and European Welfare States. Basingstoke, Palgrave: 58–76. Deakin, S. (1997) Integration through Law? The Law and Economics of European Social Policy, in J. T. Addison and S. W. Siebert (eds) Labour Markets in Europe: Issues of Harmonization and Regulation. London, Dryden Press: 118–51. Deakin, S. and H. Reed (2000) River Crossing or Cold Bath? Deregulation and Employment in Britain in the 1980s and 1990s, in G. Esping-Andersen and M. Regini (eds) Why Deregulate Labour Markets? Oxford University Press: 115–47. Dehousse, F. (1999) Amsterdam: The Making of A Treaty. London, Kogan Page. Delors, J. (1985) ‘The Thrust of Commission Policy’, Statement to the European Parliament. Strasbourg, Bulletin of the European Communities, Supplement No. 1. Denmark (1996) Bases for Negotiations: An Open Europe. The 1996 IGC. Memorandum of the Danish Government, 11 December 1995. White Paper on the 1996 Intergovernmental Conference: Summary of Positions of the Member States of the European Union with a view to the 1996 IGC. Brussels, European Parliament: II, 32–5. DGB and BDA (1989) Joint Declaration of the DGB and BDA (signed at Frankfurt), in J. T. Addison and W. S. Siebert (eds) Labour Markets in Europe: Issues of Harmonization and Regulation. London, Dryden Press: 189–90. Diamandouros, P. N. (1993) Politics and Culture in Greece, 1974–91: An Interpretation, in R. Clogg (ed.) Greece, 1981–89: The Populist Decade. London, Macmillan: 1–25. Diebold, W. (1952) The End of the I.T.O. Princeton, NJ, International Finance Section, Dept. of Economist Social Institution. Princeton University Press. Dolvik, J. E. (1997) Redrawing Boundaries of Solidarity? ETUC, Social Dialogue and the Europeanisation of Trade Unions in the 1990s. Oslo, ARENA and FAFO. Dolvik, J. E. and J. Visser (2001) ETUC and European Social Partnership: A Third Turning Point?, in H. Compston (ed.) Social Partnership in the European Union. Basingstoke, Palgrave: 11–40. Donges, J. B., C. Krieger et al. (1982) The Second Enlargement of the European Community: Adjustment Requirements and Challenges for Policy Reform. Tubingen, J. C. B. Mohr (Paul Siebeck). Dorey, P. (2002) Britain in the 1990s: The Absence of Policy Concertation, in S. Berger and H. Compston (eds) Policy Concertation and Social Partnership in Western Europe: Lessons for the 21st Century. Oxford, Berghahn: 63–76. Dupré, D. (2000) Accidents at Work in the EU in 1996. Luxembourg, Eurostat. Eardley, T., J. Bradshaw et al. (1996) Social Assistance in OECD Countries: Synthesis Report. London, Social Policy Research Unit, on behalf of OECD and Department of Social Security: 195. Ebbinghaus, B. and J. Visser (2000) The Societies of Europe: Trade Unions in Western Europe since 1945: A Series of Historical Data Handbooks on the Development of Europe from the Nineteenth to the End of the Twentieth Century. London, Macmillan. EC (1972) EC Bulletin 10/1972. Brussels. EC (1987) Single European Act. OJ No. L 169, 29 June 1987. EC (1989) Community Charter of the Fundamental Social Rights of Workers (Adopted by the Heads of State or Government of 11 Member States at Strasbourg). Brussels. EC (1997) Council Directive 97/81/EC of 15 December 1997 Concerning the Framework Agreement on Part-Time Work Concluded by UNICE, CEEP and the
Bibliography 241 ETUC Annex: Framework Agreement on Part-Time Work. Brussels: OJ L 014, 20/01/1998, pp. 0009–0014. EC (1998) Council Directive 98/23/EC of 7 April 1998 on the Extension of Directive 97/81/EC on the Framework Agreement on Part-Time Work Concluded by UNICE, CEEP and the ETUC to the United Kingdom of Great Britain and Northern Ireland. Brussels: OJ L 131, 05/05/1998, p. 0010. EC (1999) Treaty Establishing the European Community, Consolidated Version. ECJ (1991) Case 345/89 Stoeckel [1991] (ECR I-4047). Luxembourg. ECJ (1993) Opinion 2/91 ILO Convention No. 170 of the ILO Concerning Safety in the Use of Chemicals at Work. Luxembourg. ECJ (1994) Case 84/94 UK v. Council of the European Union, Judgement of 12 November 1996 [1996] ECR 5755 al. 16 and 17. Luxembourg. ECJ (2002) ‘Commission v. Germany’, Judgement of the Court (Fifth Chamber) C-5/00. ECSC (1951) Treaty Establishing the European Coal and Steel Community. ECSC (1953) Accord concernant la collaboration entre l’Organisation Internationale du Travail et la Communauté européenne du charbon et de l’acier. Brussels: 167–68. ECSC (1957) 6th General Report; Decision of the Council of 9 July 1957 Setting Down the Responsibilities and Internal Organization of the Permanent Commission for Safety in Coal Mines. Journal Officiel no. 28, 31 August 1957. Brussels: II. EEC (1957) Treaty of Rome. Luxembourg, Office for Official Publications of the European Communities. http://www.europe.eu.int/eur-/ex/en/search/treaties_ founding.html [accessed 20 September 2004]. EEC (1958) Accord concernant la liaison entre l’Organisation internationale du travail et la Communauté Économique Européenne. Brussels: 521–29. EEC (1989) Framework Directive 89/391/EEC: Council Directive on the Introduction of Measures to Encourage Improvements in the safety and health of workers at work: pp. 0001–0008. EFTA (1988) Note by the Secretariat of the EFTA Committee on Technical Barriers to Trade: Report on the Social Policy of the Community, Occupational Safety. Eichener, V. (1992) Social Dumping or Innovative Regulation: Processes and Outcomes of European Decision-Making in the Sector of Health and Safety at Work Harmonization. Florence, European University Institute. Eising, R. and B. Kohler-Koch (1999) Introduction: Network Governance, in R. Eising and B. Kohler-Koch (eds) The Transformation of Governance in the European Union. London, Routledge: 3–13. Engles, C. and L. Salas (1998) Labour Law and the European Union After the Amsterdam Treaty, in R. Blanpain (ed.) Institutional Changes and European Social Policies after the Treaty of Amsterdam. London, Kluwer Law International: 67–100. ESC (1995) Opinion on the Relations between the EU and the International Labour Organisation (ILO), Economic and Social Committee. Brussels. (95/C 102/03). Esping-Andersen, G. (1990) The Three Worlds of Welfare Capitalism. Cambridge, Polity Press. Esping-Andersen, G. (1996) After the Golden Age? Welfare State Dilemmas in a Global Economy, in G. Esping-Andersen (ed.) Welfare States in Transition: National Adaptations in Global Economics. London, Sage: 1–31. Esping-Andersen, G. (1999) Social Foundations of Post-industrial Economies. Oxford University Press.
242 Bibliography Esping-Andersen, G. and J. E. Kolberg (1992) Welfare States and Employment Regimes, in J. E. Kolberg (ed.) Between Work and Social Citizenship. Armonk, NY: M.E. Sharpe: 3–35. ETUC (1988) Resolution sur les propositons de la Commission en matière de santé et de securité (Annexe: Déclaration du Groupe Travailleurs du CCHS – Session plenière du 12 janvier 1988). Brussels, Le Comité Executif. ETUC (1989) Letter to the Affiliated Organisations, Industry Committee, and Worker Members of the CCHS on the Framework Directive on Health and Safety. Brussels. EU (2000a) Treaty of Nice. Brussels, Commission of the European Communities (OJ C80/1 10.3.2001). EU (2000b) Charter of Fundamental Rights. Official Journal of the European Communities. (OJ 2000c/364/01). EU (2003) Draft Treaty Establishing a Constitution, Official Journal OJ C 169, 18 July. EU (2004) Draft Treaty Establishing a Constitution for Europe. CIC-87/04, ADD1 and ADD2. Brussels: European Council. 18 June 2004. European Parliament (1990) Martin II Report to the European Parliament. Brussels, Parliament of the European Union. European Parliament (1996) White Paper on the 1996 Intergovernmental Conference: Summary of Positions of the Member States of the European Union with a View to the 1996 Intergovernmental Conference. Brussels. European Parliament (1997) Report on Commission’s Proposal for a Council Directive Incorporating the Framework Agreement on Part-time Work (Rapporteur: Karin Jons). Brussels. Eurostat (2001) Eurostat Yearbook 2001. Luxembourg, Office for Official Publications of the European Communities. Eurostat (2002a) Statistical Yearbook. Luxembourg, Office for Official Publications of the European Communities. Eurostat (2002b) The Social Situation in the European Union: 2001. Luxembourg. Fakiolas, R. (1987) Interest Groups: An Overview, in K. Featherstone and D. K. Katsoudas (eds) Political Change in Greece: Before and After the Colonels. London, Croom Helm: 174–88. Falkner, G. (1995) The Maastricht Protocol on Social Policy: Theory and Practice. Brighton, Sussex European Institute, University of Sussex: 16. Falkner, G. (1998) EU Social Policy in the 1990s: Towards a Corporatist Policy Community. London, Routledge. Falkner, G. (1999a) European Social Policy: Towards Multi-Level and Multi-Actor Governance, in R. Eising and B. Kohler-Koch (eds) The Transformation of Governance in the European Union. London, Routledge: 83–97. Falkner, G. (1999b) Interest Groups in a Multi-level Polity: The Impact of European Integration on National Systems. Florence, European University Institute. Falkner, G. (2000) The Council or the Social Partners? EC Social Policy between Diplomacy and Collective Bargaining. Paper presented at IVSA World Conference XVIII. Quebec City, Canada. August 2000. Falkner, G. (2002) How Intergovernmental are Intergovernmental Conferences? An example from the Maastricht Treaty. Journal of European Public Policy 9(1) (February): 98–119. Fargion, V. (1996) Social Assistance and the North-South Cleavage in Italy. Southern European Society and Politics 1(3): 135–54.
Bibliography 243 Felsenthal, D. S. and M. Machover (1997) The Weighted Voting Rule in the EU’s Council of Ministers, 1958–95: Intentions and Outcomes. Electoral Studies 16(1): 33–47. Ferge, Z. (2001) Welfare and ‘Ill-Fare’ Systems in Central-Eastern Europe, in R. Sykes, B. Palier and P. Prior (eds) Globalization and European Welfare States. Basingstone, Palgrave: 127–52. Ferrera, M. (1994) EC Citizens and Social Protection. Luxembourg, Commission of the European Communities. Ferrera, M. (1996) The ‘Southern Model’ of Welfare in Social Europe. Journal of European Social Policy 6(1): 17–37. Ferrera, M. (1998) The Four ‘Social Europes’: Between Universalism and Selectivity, in M. Rhodes and Y. Meny (eds) The Future of European Welfare: A New Social Contract? London, Macmillan: 81–96. Ferrera, M., A. Hemerijck and M. Rhodes (2000) The Future of Social Europe: Recasting Work and Welfare in the New Economy. Report for the Portuguese Presidency of the EU. Finland (1996) Finland’s Starting-Points and Objectives for the 1996 IGC – Report of the Finnish Government, 27 February 1996. White Paper on the 1996 Intergovernmental Conference: Summary of Positions of the Member States of the European Union with a View to the 1996 Intergovernmental Conference. Brussels, European Parliament: II, 146–150. Flora, P. and A. Heidenheimer (eds) (1981) The Development of Welfare States in Europe and America. London, Transaction Books. Flynn, P. (1997) Speech by Commissioner Padraig Flynn on the Commission Proposals to the Special Jobs Summit 20–21 November. Brussels, European Commission. Foden, D. (1999) The Role of the Social Partners in the European Employment Strategy. Transfer 5(4): 522–41. Forster, A. (1998) Britain and the Negotiation of the Maastricht Treaty: A Critique of Liberal Intergovernmentalism. Journal of Common Market Studies 36(3): 347–68. France (2002) Evaluation de la stratégie européenne pour l’emploi. http://europa. eu.int/comm/employment_social/news/2002/may/eval_fr.pdf[accessed September 2004]. Frid, R. (1995) The Relations Between the EC and International Organizations: Legal Theory and Practice. London, Kluwer. Fry, J. A. (ed.) (1979) Industrial Democracy and Labour Market Policy in Sweden. Toronto, Pergamon. FT (2002) German Labour Office Under Fire. Financial Times 7 February: 2. Fuchs, S. and R. Schettkat (2000) Germany: A Regulated Flexibility, in G. Esping Andersen and M. Regini (eds) Why Deregulate Labour Markets. Oxford University Press: 211–44. Garrett, G. (1993) The Politics of Maastricht. Economic and Politics 5(2): 105–24. Garrett, G. and G. Tsebelis (1996) ‘An Institutional Critique of Intergovernmentalism.’ International Organization 50(2): 269–300. Gautié, J., M. A. Malo and L. Toharia (2000) France: The Deregulation that Never Existed, in G. Esping-Andersen and M. Regini (eds) Why Deregulate Labour Markets? Oxford University Press: 245–270. Gelauff, G. and M. Pomp (2000) Deregulation and Labour Market Reforms: The Role of the Social Partners, in G. Galli and J. Pelkmans (eds) Regulatory
244 Bibliography Reform and Competitiveness in Europe: Horizontal Issues. Cheltenham, Edward Elgar: I, 381–429. George, V. and P. Taylor-Gooby (1996) European Welfare Policy: Squaring the Welfare Circle. London, Macmillan. Germany (1996) Germany’s Objectives for the IGC, 26 March 1996. White Paper on the 1996 Intergovernmental Conference: Summary of Positions of the Member States of the European Union with a view to the 1996 IGC. Brussels, European Parliament: II, 39–44. Germany (1997a) D- Anderungswunsche zu Kapitel 3 BESCHAFTIGUNG (document in French: ‘Chapitre 3. Emploi’), Deutsche Delegation to the IGC. Council of the European Union CAB Archives in Brussels, SN 540/97 (C41) Seiten 1–4, 26 May 1997. Germany (1997b) German Action Programme for the Employment Policy in Europe. Council of the European Union Archives, Brussels (12160/97) LIMITE SOC 358 ECOPIN 180. 10 November 1997. Germany (2000) Comments of the German Government on the Portuguese Presidency Document on Employment, Economic Reforms and Social Cohesion – Towards a Europe Based on Innovation and Knowledge (Council Document 5256/00) for the Special Meeting of the European Council to be held on 23/24 March in Lisbon (translation). Brussels, Council Archives. Goetschy, J. (2001) The European Employment Strategy from Amsterdam to Stockholm: Has it Reached its Cruising Speed Yet? Industrial Relations Journal 32(5): 401–18. Gold, M. (1998) Social Partnership a the EU Level: Initiatives, Problems and Implications for Member States, in D. Hine and H. Kassim (eds) Beyond the Market: The EU and National Social Policy. London, Routledge: 107–33. Gold, M. and D. Matthews (1996) The Implications of the Evolution of European Integration for UK Labour Markets. London, National Institute of Economic and Social Research (Department for Education and Employment). Research Series No. 73. March 1966. Goldstein, J. (1993) Ideas, Interests, and American Trade Policy. Ithaca, NY, Cornell University Press. Gough, I. (1996) Social Assistance in Southern Europe. South European Society and Politics 1(1): 1–23. Greece (1992) Greek Memorandum on Contribution to the Discussions on Progress Towards Political Union (Aide-Mémoire of 15 May 1990), in F. Laursen, S. Vanhoonacker and R. Wester (eds) The Intergovernmental Conference on Political Union – Institutional Reforms, New Policies, and International Identity of the European Community. Maastricht, European Institute of Public Administration: 277–82. Greece (1996) Memorandum of the Greek Government of 24 January 1996 on the IGC: Greece’s Position and Comments. White Paper on the 1996 Intergovernmental Conference: Summary of Positions of the Member States of the European Union with a view to the 1996 IGC. Brussels, European Parliament: II, 53–5. Greece (2001) National Action Plan for Social Inclusion 2001–2003. Athens, Greek Ministry of Labour and Social Security (Coordinating Ministry). Greenwood, J. (2002) Inside the EU Business Associations. London, Palgrave. Haas, E. (1958) The Uniting of Europe: Political, Social and Economic Forces 1950–1957. Stanford University Press.
Bibliography 245 Haas, E. B. (1964) Beyond the Nation State. Stanford University Press. Hall, P. A. and R. C. R. Taylor (1996) Political Science and the Three New Institutionalisms. Political Studies 44(5): 936–57. Hamalainen, R.-M., K. Husman, J. Rantaen, K. Rasanen and P. Westerholm (2001) Survey of the Quality and Effectiveness of Occupational Health Services in the European Union and Norway and Switzerland. Stockholm, National Institute for Working Life and Finnish Institute of Occupational Health: 232. Hannum, H. (1992) Implementing Human Rights: An Overview of Strategies and Procedures, in H. Hannum (ed.) Guide to International Human Rights Practice. Philadelphia, University of Pennsylvania Press: 19–38. Hantrais, L. (2000) Social Policy in the European Union. London, Macmillan. Harrop, J. (1996) Structural Funding and Employment in the European Union: Financing the Path to Integration. Cheltenham, Edward Elgar. Hayes-Renshaw, F. and H. Wallace (1996) The Council of Ministers of the European Union. London: Macmillan. Healy, S. and B. Reynolds, Eds. (1998) Social Policy in Ireland: Principles, Practice and Problems. Dublin, Oak Tree. Hilbert, C., G. Schmid and S. Speckesser (2001) Does Active Labour Market Policy Matter? An Aggregate Impact Analysis for Germany, in J. de Koning and H. Mosley (eds) Labour Market Policy and Unemployment: Impact and Process Evaluations in Selected European Countries. Cheltenham, Edward Elgar: 77–114. Hine, D. (1998) Introduction: The European Union, State Autonomy and National Social Policy, in D. Hine and H. Kassim (eds) Beyond the Market: The EU and National Social Policy. London, Routledge: 1–13. Hine, D. and K. Hussein (eds) (1998) Beyond the Market: the EU and National Social Policy. London, Routledge. Hix, S. (1999) The Political System of the EU. New York, St Martin’s Press. Hoekman, B. and M. Kostecki (1995) The Political Economy of the World Trading System. Oxford University Press. Hooghe, L. (1996) Cohesion Policy and European Integration: Building Multi-level Governance. Oxford University Press. Hooghe, L. (2001) Top Commission Officials on Capitalism: An institutionalist Understanding of Preferences, in M. Aspinwall and G. Schneider (eds) The rules of Integration: Institutionalist approaches to the Study of Europe. Manchester University Press: 152–76. Hyde-Price, A. and C. Jeffery (2001) Germany in the European Union. Journal of Common Market Studies 39(4): 689–718. ICITO (1948a) Executive Committee, Second Session: Item 4 of the Provisional Agenda: Relations Between the International Labour Organization and the International Trade Organization including the Annex ‘Draft of Agreement Between the International Labour Organization and the International Trade Organization’. Geneva, Interim Commission of the International Trade Organization. ICITO (1948b) Executive Committee, Second Session: Relations with Non-Governmental Organizations. Geneva, Interim Commission of the International Trade Organization. ICITO (1948c) Final Act and Related Documents of the United Nations Conference on Trade and Employment, held at Havana, Cuba from November 21, 1947 to March 24, 1948 (or the Havana Charter). Lake Success, New York, Interim Commission for the International Trade Organization.
246 Bibliography ICITO (1948d) Second Session of the Interim Commission for the International Trade Organization, Provisional Agenda (Revised). Geneva, Interim Commission of the International Trade Organization. IEP (1989) Comitology: Characteristics, Performance and Options. Bonn, Institut für Europäische Politik. IGC (1996a) Proposition de modification du traité instituant la Communauté européenne (presented at Cork informal meeting of IGC representatives by the Danish delegation to the IGC. Brussels: Council archives. Reference (12.07) (OR.dk) CONF 3864/96., 6 July). IGC (1996b) Employment: Proposals for Treaty Amendments, Treaty on European Union. (Presented by Swedish delegation to the IGC). Brussels: Council archives. No number assigned. ILO (1919) Constitution of the International Labour Organization. ILO (1944) Declaration Concerning the Aims and Purposes of the International Labour Organization. Philadelphia, General Conference of the International Labour Organization. ILO (1956) Social Aspects of European Collaboration (Ohlin Report). Geneva, ILO Studies and Reports (New Series). ILO (1957?) Projet préliminaire d’Accord concernant la collaboration entre l’Organisation internationale du Travail et la Communauté économique européenne. From ILO archives, Geneva File IGO-051-3. ILO (1961) Letter from the President of the Commission of the European Economic Community to the Director-General of the International Labour Office and Letter from the Director-General of the International Labour Office to the President of the Commission of the European Economic Community. Brussels and Geneva: 533. Brussels and Geneva, Official Bulletin Vol. XLIV No. 7, p. 533. ILO (1965) Troisième réunion du Comité permanent de contact entre le B.I.T. et la Commission de la C.E.E. Geneva. From ILO archives, Geneva. 11 March 1961. File IGO 051-2-1 (J.2)(D. 29. 1965). ILO (1970a) The European Communities – ILO 051-2-1 (J.4). (Note: undated document. 1970 is suggested given the chronology of other documents where this document was found.) From ILO archives, Geneva. ILO (1970b) Minute Sheet of Mr. Jacques Lemoine. From ILO archives, Geneva. File IGO 051-2-1 (J.4). ILO (1971a) Conference Internationale du Travail: Compte rendu d’entretiens. From ILO archives, Geneva. File Z 15/15/1. ILO (1971b) Co-operation between the ILO and the European Communities (Note for the Director-General). From ILO archives, Geneva. File Z 15/15/1. ILO (1971c) Considérations complementaires sur les relations entre le BIT et les Communautés. From ILO archives, Geneva. No file reference. ILO (1972a) Perspectives de collaboration avec les Communautés européennes: Conclusions des entretiens de Bruxelles du 4 mai 1972. From ILO archives, Geneva. File Z 15/15/1. ILO (1972b) Préparatifs en vue de la Septième Réunion du Comité de Contact BIT/Communautés Européennes: Réunion interdépartementale du 2 octobre 1972. From ILO archives, Geneva. File Z 15/15/1. ILO (1976) Revision des Accords entre l’OIT et les Communautés Européennes. Geneva. From ILO archives, Geneva. File IGO 099-2. ILO (1984) Contenu social et culturel de Lomé III. Geneva.
Bibliography 247 ILO (1988a) Brief on ILO/EEC Technical Cooperation. From ILO archives, Geneva. File Z 15/15 Jacket 4. ILO (1988b) Note relative a la visite de M. Jacques Moreau, Secretaire general du Comité économique et social européen. From ILO archives, Geneva. File Z 15/15 Jacket 4. ILO (1988c) Visit to the ILO of Mr. Moreau, General Secretary of the European Social Economic Council. Geneva. ILO (1991) Committee on Freedom of Association Case 1528, 27th Report, Approved by the Governing Body at the 249th Session. ILO, Geneva. ILO (1993a) Eleventh item on the agenda: Report of the Committee on Standing Orders and the Application of Conventions and Recommendations. Geneva, International Labour Office/ Governing Body. ILO (1993b) Fifth Supplementary Report: ILO Participation in meetings of the Group of 24 (256th Session). Geneva, ILO Administrative Council. ILO (1994) Fourth Item on the Agenda: The external competence of the European Community and the ILO’s standard-setting activities, following the Opinion delivered by the Court of Justice of the European Communities. Geneva, Governing Body Committee on Legal Issues and International Labour Standards. ILO (1998) ILO Declaration on Fundamental Principles and Rights at Work. Geneva. ILO (2000a) Report and conclusions of the Sixth European Regional Meeting; Report of the Director General for the Sixth European Regional Meeting. Geneva. ILO (2000b) Resolutions submitted to the 88th Session of the International Labour Conference. Geneva. ILO (2002) ILO–European Commission High-Level Meeting (Draft Joint Annotated Agenda version 08/2/02). Brussels: 14 February. ILO (2004) A Fair Globalization: Creating Opportunities for All. Geneva: World Commission on the Social Dimensions of Globalization. Ioannou, C. A. (1999) Trade Unions in Greece: Development, Structures and Prospects. Athens, Ebert. Ioannou, C. A. (2000) Social Pacts in Hellenic Industrial Relations: Odysseys or Sisyphus? Social Pacts in Europe: New Dynamics. 219–36. Ireland (1988) Problème évoqué par les Délégations Irlandaise et Britannique (Ireland and UK government submission). Brussels, ‘Groupe des Questions sociales’ of the Council of the European Communities. Ireland (1996) White Paper on Foreign Policy: External challenges and opportunities, 26 March 1996. White Paper on the 1996 Intergovernmental Conference: Summary of Positions of the Member States of the European Union with a view to the 1996 IGC. Brussels, European Parliament: II, 92–5. Italy (1996) Position of the Italian Government on the IGC for the revision of the Treaties, 18 March 1996. White Paper on the 1996 Intergovernmental Conference: Summary of Positions of the Member States of the European Union with a view to the 1996 IGC. Brussels, European Parliament: II, 101–4. Jachtenfuchs, M. (2001) The Governance Approach to European Integration. Journal of Common Market Studies 39(2): 245–64. James, P. (1993) Occupational Health and Safety, in M. Gold (ed.) The Social Dimension: Employment Policy in the European Community. Basingstoke, UK, Macmillan: 135–52. Jeffery, M. (1998) Not Really Going to Work? Of the Directive on Part-Time Work, ‘Atypical Work’ and Attempts to Regulate It. Industrial Law Journal 27(3): 193–213.
248 Bibliography Jensen, C. S. (2000) Neofunctionalist Theories and the Development of European Social and Labour Market Policy. Journal of Common Market Studies 38(1): 71–92. Johansson, K. M. (1999) Tracing the Employment Title in the Amsterdam Treaty: Uncovering Transnational Coalitions. Journal of European Public Policy 6, 1 (1999): 85–101. Johnson, A. M. (1998) Neutrality and the 1995 Enlargement: Extending EU Membership and the CFSP to Neutral States. M.Sc. Thesis, Department of International Relations. London School of Economics. Johnson, A. M. (2001) Albania’s relations with the EU: on the road to Europe? Journal of Southern Europe and the Balkans 3(2): 171–92. Johnson, D. (1996) The Impact of the Nordic Countries on EU Social Policy, in L. Miles (ed.) The European Union and the Nordic countries. London, Routledge: 186–202. Joyce, J. A. (1980) World Labour Rights and Their Protection. London, Croom Helm. Katrougalos, G. S. (1996) The Southern European Welfare Model: The Greek Welfare State in Search of an Identity. Journal of European Social Policy 6(1): 39–60. Katseli, L. (1990) Economic Integration in the Enlarged European Community: Structural Adjustment of the Greek Economy, in C. Bliss and J. Braga de Macedo (eds) Unity with Diversity in the European Economy: the Community’s Southern Frontier. Cambridge University Press: 235–309. Keller, B. (1998) National Industrial Relations and the Prospects for European Collective Bargaining, in W. Lecher and H.-W. Platzer (eds) European Union – European Industrial Relations? Global Challenges, National Developments and Transnational Dynamics. London, Routledge: 47–64. Keohane, R. (1984) After Hegemony: Cooperation and Discord in the World Political Economy. Princeton University Press. Keohane, R. O. (ed.) (1989) Neoliberal Institutionalism: A Perspective on World Politics. International Institutions and State Power: Essays in International Relations Theory. Boulder, CO, Westview Press: 1–20. Kingdon, J. W. (1984) Agendas, Alternatives, and Public Policies. Boston, Little, Brown. Knight, J. (1992) Institutions and Social Conflict. Cambridge University Press. Kohl, H. and F. Mitterrand (1990) Kohl-Mitterrand Letter (in Document Annex), in R. Corbett (ed.) The Treaty of Maastricht, From Conception to Ratification: A Comprehensive Reference Guide. Harlow, Longman: 126. Koliopoulos, K. (1994) Greece and the ratification of the Maastricht Treaty, in F. Laursen and S. Vanhoonacker (eds) The Ratification of the Maastricht Treaty: Issues, Debates and Future Implications. Maastricht, European Institute of Public Administration: 113–28. Korpi, W. (1978) The Working Class in Welfare Capitalism: Work, Unions and Politics in Sweden. London, Routledge & Kegan Paul. Korpi, W. (1992) Welfare State Development in Europe Since 1930: Ireland in a Comparative Perspective. Dublin, Economic and Social Research Institute. Kottis, A. P. and G. C. Kottis (1996) Greece. Luxembourg, The Research Centre of the Athens University of Economics and Business for the European Commission, DG Employment and Social Affairs. Krasner, S., Ed. (1983) International Regimes. Ithaca, NY, Cornell University Press.
Bibliography 249 Kratochwil, F. and J. G. Ruggie (1986) ‘International Organization: A State of the Art on an Art of the State.’ International Organization 40(4): 753–75. Labour Research (1999) Full Rights for Part Timers. Labour Research 88(3): vol. 88, issue 3. Lafoucrière, C. (2000) The European Employment Strategy: The Third Pillar, Adaptability (A Critical Evaluation of the Situation in France and Belgium, or: Positive Social Dialogue Versus Negative Social Dialogue). Brussels, European Trade Union Institute Discussion and Working Papers. Lange, P. (1993) Maastricht and the Social Protocol: Why Did They Do It? Politics and Society 21(1): 5–36. Langille, B. (1997) Eight Ways to Think about International Labour Standards. Journal of World Trade 31(4): 27–53. Lanzalaco, L. (1992) Coping with Heterogeneity: Peak Associations of Business within and across Western European Nations, in J. Greenwood, J. R. Grote and K. Ronit (eds) Organized Interests and the European Community. London, Sage: 173–205. Lapeyre, J. (2000) Nordic Labour Markets and European Social Policy: The ETUC Perspective, in A. C. Neal (ed.) European Social Policy and the Nordic Countries. Aldershot, Ashgate: 186–95. Laroque, P. (1953) Réflexions sur le problème sociale. Paris, Editions Sociales Françaises. Larsson, A. (2001) A Turning Point for Employment Policy. Manuscript provided by the author. Larsson, A. (2002) The New Open Method of Co-ordination: A Sustainable Way Between a Fragmented Europe and a European Supra State? A Practioner’s View. Lecture given at Uppsala University. Lavdas, K. A. (1997) The Europeanization of Greece: Interest Politics and the Crises of Integration. London, Macmillan. Lea, D. (1998) European Social Dialogue and Industrial Relations: The View of the TUC, in W. Lecher and H.-W. Platzer (eds) European Union – European Industrial Relations? Global Challenges, National Developments and Transnational Dynamics. London, Routledge: 133–8. Lecher, W. (1994) Trade Unions in the European Union: A Handbook. London, Lawrence & Wishart. Leibfried, S. (1992) Europe’s Could-Be Social State: Social Policy in European Integration, in W. J. Adams (ed.) Singular Europe: Economy and Polity of the European Community After 1992. Ann Arbor, University of Michigan Press: 97–118. Leibfried, S. and P. Pierson (1997) Social Policy, in H. Wallace and W. Wallace (eds) Policy-Making in the European Union. Oxford University Press: 185–207. Leibfried, S. and P. Pierson (2000) Social Policy, in H. Wallace and W. Wallace (eds) Policy-Making in the European Union. Oxford University Press: 267–92. Lind, J. and I. Hornemann Moller (1999) European Convergence or National Specificities?, in J. Lind and I. Hornemann Moller (eds) Inclusion and Exclusion: Unemployment and Non-standard Employment in Europe. Aldershot, Ashgate: 219–30. Lindahl, R. (1996) The Swedish Debate, in F. Algieri, H. Grabbe, K. Hughes et al. (eds) The 1996 IGC – National Debates (2): Germany, Spain, Sweden and the UK. London, Royal Institute of International Affairs: 37–50.
250 Bibliography Lindberg, L. (1963) The Political Dynamics of European Economic Integration. Oxford University Press. Lodovici, M. S. (2000) The Dynamics of Labour Market Reform, in G. EspingAndersen and M. Regini (eds) Why Deregulate Labour Markets? Oxford University Press: 30–65. Lyberaki, A. and E. Tsakalotos (2000) Reforming the Economy without Society: Social and Institutional Constraints to Economic Reform in post-1974 Greece. Athens, Department of International and European Economic Studies, Athens University of Economics and Business: 13. Majone, G. (1993) The European Community Between Social Policy and Social Regulation. Journal of Common Market Studies 31(2): 153–70. Majone, G. (1996a) The European Commission as Regulator, in G. Majone (ed.) Regulating Europe. London, Routledge: 61–79. Majone, G. (1996b) The Future of Regulation in Europe, in G. Majone (ed.) Regulating Europe. London, Routledge: 265–83. Manow, P. and E. Seils (2000) Adjusting Badly: The German Welfare State, Structural Change, and the Open Economy, in F. W. Scharpf and V. A. Schmidt (eds) Welfare and Work in the Open Economy. Oxford University Press. II: 264–307. Marinkakou, M. (1998) Welfare States in the European Periphery: The Case of Greece, in R. Sykes and P. Alcock (eds) Developments in European Social Policy: Convergence and Diversity. Bristol, Policy Press: 231–47. Marks, G. (1993) Structural Policy and Multilevel Governance in the EC, in A. W. Cafruny and G. G. Rosenthal (eds) The State of the European Community. Harlow, Longman: 391–410. Marks, G. and L. Hooghe (1996) European Integration from the 1980s: State Centric v. Multilevel Governance. Journal of Common Market Studies vol. 34, no. 3: 341–78. Marshall, T. H. (1964) Class, Citizenship and Social Development. Garden City, NY Doubleday. Marshall, T. H. (1975) Social Policy in the Twentieth Century. London, Hutchinson. Mavrogordatos, G. T. (1993) Civil Society under Populism, in R. Clogg (ed.) Greece, 1981–1989: The Populist Decade. London, Macmillan: 47–64. McDonagh, B. (1998) Original Sin in a Brave New World: An Account of the Negotiation of the Treaty of Amsterdam. Dublin, Institute of European Affairs. Mehmet, O., E. Mendes and R. Sinding (1999) Towards a Fair Global Labour Market: Avoiding a New Slave Trade. London, Routledge. Metcalfe, D. (1994) Transformation of British Industrial Relations? Institutions, conduct and outcomes 1980–1990. The UK Labour Market: Comparative Aspects and Institutional Developments. Cambridge University Press: 126–57. Meunier, S. and K. Nicolaïdis (1999) Who Speaks for Europe? The Delegation of Trade Authority in the EU. Journal of Common Market Studies 37(3): 477–501. Mielke, S., P. Rutters and K. Tudyka (1994) Trade Union Organization and Employee Representation, in W. Lecher (ed.) Trade Unions in the European Union. London, Lawrence & Wishart: 127–234. Miles, L. (1997) Sweden and European Integration. Aldershot, Ashgate. Miles, L. (1998) ‘Diamonds are Forever’: Sweden and the IGC. University of Hull Press. Milner, H. V. (1997) Interests, Institutions, and Information: Domestic Politics and International Relations. Princeton University Press.
Bibliography 251 Milner, S. (2002) France in Historical Perspective: The Impossibility of Partnership, in S. Berger and H. Compston (eds) Policy Concertation and Social Partnership in Western Europe: Lessons for the 21st Century. Oxford, Berghahn: 97–109. Mirka, J. and C. J. Ruhm (1997) The European Social Charter and North American Free Trade Agreement: Perspectives from the United States, in J. T. Addison and W. S. Siebert (eds) Labour Markets in Europe: Issues of Harmonization and Regulation. London, Dryden: 212–28. Mitrany, D. (1965) ‘The Prospect of European Integration: Federal or Functional? Journal of Common Market Studies 4(2): 123–34 (Also reprinted in J. S. Nye (ed.) (1968) International Regionalism. Boston, Little Brown). Mitrany, D. (1975) The Functional Theory of Politics. London, Martin Robertson, on behalf of the London School of Economics. Morata, F. and X. Muñoz (1996) Vying for European Funds: Territorial Restructuring in Spain, in L. Hooghe (ed.) Cohesion Policy and European Integration: Building Multi-Level Governance. Oxford University Press: 195–218. Moravcsik, A. (1991) Negotiating the Single European Act: National Interests and Conventional Statecraft in the European Community. International Organization 45(1): 19–56. Moravcsik, A. (1997) Taking Preferences Seriously: A Liberal Theory of International Politics. International Organization 51:4 (Autumn 1997) 538–41. Moravcsik, A. (1998) The Choice for Europe: Social Purpose and State Power from Messina to Maastricht. London, UCL Press. Moravcsik, A. and K. Nicolaïdis (1999) Explaining the Treaty of Amsterdam: Interests, Influence, Institutions. Journal of Common Market Studies 37(1): 59–85. Mosher, J. J. and D. M. Trubek (2003) ‘Alternative Approaches to Governance in the EU: EU Social Policy and the European Employment Strategy’, Journal of Common Market Studies 41(1): 63–88. Moussis, N. (1969) Greek Industrialists and the Association of Greece with the Common Market. Athens, Skouras. Murray, J. (2001) Transnational Labour Regulation: The ILO and the EC Compared. The Hague, Kluwer. Nanetti, R. Y. (1996) EU Cohesion and Territorial Restructuring in the Member States, in L. Hooghe (ed.) Cohesion Policy and European Integration: Building Multi-Level Governance. Oxford University Press: 59–88. Netherlands (1990) Dutch Government 1st Memorandum, May 1990, Possible Steps Towards European Political Union (in documents annex), in R. Corbett (ed.) The Treaty of Maastricht, From Conception to Ratification: A Comprehensive Reference Guide. Harlow, Longman: 127–33. Nicolaïdis, K. (1996) Mutual Recognition of Regulatory Regimes: Some Lessons and Prospects. Paris, OECD. Nielsen, R. and E. Szyszczak (1993) The Social Dimension of the European Community. Denmark, Handelshojskolens Forlag. Norgaard, G. (1992) Industrial Relations, in L. Lyck (ed.) Denmark and EC Membership Evaluated. London, Pinter: 204–14. North, D. C. (1990) Institutions, Institutional Change and Economic Performance. Cambridge University Press.
252 Bibliography O’Donnell, R. and D. Thomas (2002) Ireland in the 1990s: Policy Concertation Triumphant, in S. Berger and H. Compston (eds) Policy Concertation and Social Partnership in Western Europe: Lessons for the 21st Century. Oxford, Berghahn: 167–89. OECD (1964) Recommendations on an Active Manpower Policy. Paris, Manpower and Social Affairs Committee of the OECD. OECD (1995) OECD Employment Outlook. Paris. OECD (1996) Trade, Employment and Labour Standards: A Study of Core Workers’ Rights and International Trade. Paris. OECD (1998) Labour Force Statistics 1977–1997. Paris. OECD (1999) Employment Outlook. Paris. OECD (2000) OECD Employment Outlook. Paris. Olsen, G. M. (1998) Re-modelling Sweden: The Rise and Demise of Compromise in a Global Economy, in G. M. Olsen (ed.) Power Resources Theory and the Welfare State: A Critical Approach, essays collected in honour of Walter Korpi. University of Toronto Press. Pagoulatos, G. (1999) Perceptions of Governance in Greek State Retreat: Implementing Policy Change Against Policy-making Persistence, in R. Eising and B. Kohler-Koch (eds) The Transformation of Governance in the European Union. London, Routledge: 189–207. Papadopoulos, C. A. (2002) Greece and the 1996–97 EU Intergovernmental Conference, in F. Laursen (ed.) The Amsterdam Treaty: National Preference Formation, Interstate Bargaining and Outcome. Odense University Press. Parsons, N. (2002) France in the 1990s: Struggling with the Weight of History, in S. Berger and H. Compston (eds) Policy Concertation and Social Partnership in Western Europe: Lessons for the 21st Century. Oxford, Berghahn: 111–23. Pelkmans, J. (1987) The New Approach to Technical Harmonization and Standardization. Journal of Common Market Studies 25(3): 249–69. Pelkmans, J. (1997) European Integration: Methods and Economic Analysis. Harlow, Addison Wesley Longman. Pereirinha, J. A. (1996) Welfare States and Anti-Poverty Regimes: The Case of Portugal. Southern European Welfare States 1(3): 198–218. Peters, B. G. (1999) Institutional Theory in Political Science: The ‘New Institutionalism’. London, Pinter. Petite, M. (2001) ‘Treaty of Amsterdam.’ Draft paper: 1–31. Petmesidou, M. (1991) Statism, Social Policy and the Middle Classes. Journal of European Social Policy 1(1): 31–48. Petmesidou, M. (1996) Social Protection in Greece: A Brief Glimpse of the Welfare State. Social Policy and Administration 30(4): 324–47. Petmesidou, M. (2000) Social Protection in Greece in the Nineties: Reforming the ‘Weak’ Welfare State, in E. Mossialos and A. Mitsos (eds) Contemporary Greece and Europe. Aldershot, Ashgate: 303–30. Pierson, P. (1998) The Path to European Integration: A Historical-Institutionalist Analysis, in W. Sandholtz and A. Stone Sweet (eds) European Integration and Supranational Governance. Oxford University Press: 27–58. Pierson, P. (2000) The Limits of Design: Explaining Institutional Origins and Change. Governance: An International Journal of Policy and Administration 13(4): 475–99.
Bibliography 253 Pierson, P. (2001a) Introduction: Investigating the Welfare State at Century’s End, in P. Pierson (ed.) The New Politics of the Welfare State. Oxford University Press: 1–14. Pierson, P. (2001b) Coping with Permanent Austerity, in P. Pierson (ed.) The New Politics of the Welfare State. Oxford University Press: 410–56. Pipkin, C. W. (1931) Social Politics and Modern Democracies. New York, Macmillan. Pollack, M. (1996) The New Institutionalism and EC Governance: The Promise and Limits of Institutional Analysis. Governance 9(4): 429–58. Pollack, M. A. (1997) Delegation, Agency, and Agenda Setting in the European Community. International Organization 51 (Winter 1997): 99–134. Pollack, M. A. (2001) International Relations Theory and European Integration. Journal of Common Market Studies 39(2): 221–44. Pontusson, J. (1992) The Limits of Social Democracy: Investment Politics in Sweden. Ithaca, NY, Cornell University Press. Portugal (1992) Portuguese Memorandum on Political Union with a View to the Intergovernmental Conference, in F. Laursen, S. Vanhoonacker and R. Wester (eds) The Intergovernmental Conference on Political Union – Institutional Reforms, New Policies, and International Identity of the European Community. European Institute of Public Administration, Maastricht: 304–12. Portugal (1996) Portugal and the IGC for the revision of the Treaty on European Union – Foreign Ministry document, March 1996. White Paper on the 1996 Intergovernmental Conference: Summary of Positions of the Member States of the European Union with a view to the 1996 IGC. Brussels, European Parliament: II, 134–40. Putnam, R. D. (1988) Diplomacy and Domestic Politics: The Logic of Two-Level Games. International Organization 42(3): 427–61. Rapacciuolo, D. (1990) Progress Report for 1989. Social Europe. vol. 190. pp. 122–7. Rath, F. (1994) The Co-ordinates of Trade Union Policy for Europe, in W. Lecher (ed.) Trade Unions in the European Union. London, Lawrence & Wishart: 237–74. Rhodes, M. (1991a) The Politics of the Social Dimension: National Versus Transnational Regulation in the Single European Market. Manchester, Department of Government, Victoria University of Manchester, UK. Rhodes, M. (1991b) Social Dimensions of the Single European Market. European Journal of Political Research 12: 147–69. Rhodes, M. (1996) Southern European Welfare States: Identity, Problems and Prospects for Reform. South European Society and Politics 1(3): 1–22. Rhodes, M. (1998) Defending the Social Contract: The EU Between Global Constraints and Domestic Imperatives, in D. Hine and H. Kassim (eds) Beyond the Market: The EU and National Social Policy. London, Routledge: 36–59. Rhodes, M. (1999) An Awkward Alliance: France, Germany and Social Policy, in D. Webber (ed.) The Franco–German Relationship in the European Union. London, Routledge: 130–47. Rhodes, M. (2000) Restructuring the British Welfare State: Between Domestic Constraints and Global Imperatives, in F. W. Scharpf and V. A. Schmidt (eds) Work and Welfare in the Open Economy. Oxford University Press: II, 19–67. Rhodes, M. (2001) The Political Economy of Social Pacts: ‘Competitive Corporatism’ and European Welfare Reform, in P. Pierson (ed.) The New Politics of the Welfare State. Oxford University Press: 164–94.
254 Bibliography Rodrigo, F. (1996) The Spanish Debate, in F. Algieri, H. Grabbe, K. Hughes et al. (eds) The 1996 IGC – National Debates (2): Germany, Spain, Sweden and the UK. London, Royal Institute of International Affairs: 22–35. Rosamond, B. (2000) Theories of European Integration. Basingstoke, Macmillan. Ross, G. (1995a) Jacques Delors and European Integration. Cambridge, Polity. Ross, G. (1995b) Assessing the Delors Era and Social Policy, in S. Leibfried and P. Pierson (eds) European Social Policy: Between Fragmentation and Integration. Washington, DC: Brookings Institution: 357–88. Ruggie, J. G. (1982) ‘International Regimes, Transactions and Change: Embedded Liberalism in the Postwar Economic Order.’ International Organization 35(4): 379–415. Sandholtz, W. and A. Stone Sweet (1998) Integration, Supranational Governance, and the Institutionalization of the European Polity, in W. Sandholtz and A. Stone Sweet (eds) European Integration and Supranational Governance. Oxford University Press: 1–26. Sandholtz, W. and J. Zysman (1989) 1992: Recasting the European Bargain. World Politics 42(1): 95–128. Scharpf, F. (1985) The Joint Decision Trap: Lessons from German Federalism and European Integration. Berlin, Wissenschaftszentrum. (Discussion Paper). Scharpf, F. W. (1988) ‘The Joint Decision Trap: Lessons from German Federalism and European Integration.’ Public Administration 66(3): 239–78. Scharpf, F. (1997) Games Real Actors Play: Actor-Centered Institutionalism in Policy Research. Boulder, CO, Westview Press. Scharpf, F. W. (1999) Governing in Europe: Effective and Democratic? New York, Oxford University Press. Scharpf, F. W. and V. A. Schmidt (eds) (2000) Welfare and Work in the Open Economy: From Vulnerability to Competitiveness. Oxford University Press. Schmitter, P. C. (1997) The Emerging Europolity and Its Impact Upon National Systems of Production, in R. Boyer and J. R. Hollingsworth (eds) Contemporary Capitalism: The Embeddedness of Institutions. Cambridge University Press: 395–430. Schmitter, P. C. and W. Streeck (1991) National Corporatism to Transnational Pluralism: Organized Interests in the Single European Market. Politics and Society 19(2): 133–64. Scholte, J. A., A. M. Goetz, R. O’Brien and M. Williams (2000) Contesting Global Governance: Multilateral Economic Institutions and Global Social Movements. Cambridge University Press. Shackleton, M. (1993) The Community Budget After Maastricht, in A. W. Cafruny and G. G. Rosenthal (eds) The State of the European Community. Harlow, Longman: 2, 373–90. Sigeman, T. (2000) Labour Law in Sweden – Characteristic Features in an International Perspective, in A. C. Neal (ed.) European Social Policy and the Nordic Countries. Aldershot, Ashgate: 145–72. Silletti, D. (1987) Specific Measures Taken Under the Community Policy Relating to the Employment Market: Exchanges of Officials from the National Employment Services: An Instrument for Cooperation and Arriving at a Common Policy in Matters of Employment. Social Europe 1/87: 13–15. Slaughter, A.-M., A. Stone Sweet and J. H. H. Weiler (eds) (2000) The European Court and National Courts – Doctrine and Jurisprudence: Legal Change in Its Social Context. Oxford, Hart.
Bibliography 255 Slomp, H. (2000) European Governments and Labour Relations: Convergence or Divergence, in H. Goverde (ed.) Global and European Polity? Organizations, Policies, Contexts. Aldershot, Ashgate: 105–17. Smismans, S. (2001) Functional Participation in European Occupational Health and Safety Policy: Democratic Nightmare or Additional Source of Legitimacy? Florence, European University Institute. Sous-Commission des Problèmes Sociaux. (1955) Projet: Conclusions Sommaires relatives à l’harmonisation des règlementations en vigueur, Comité Intergouvernemental crée par la Conférence de Messine, Sous-comité des problèmes sociaux. Council of the European Union Archives: MAE/CIG Document No. 121. Spain (1996) Elements for a Spanish position at the 1996 IGC, 28 March 1996. White Paper on the 1996 Intergovernmental Conference: Summary of Positions of the Member States of the European Union with a view to the 1996 IGC. Brussels, European Parliament: II, 74–84. Spain (1997) Spanish Proposals on Employment, Spanish Delegation to the IGC. Council of the European Union Cabinet (CAB) document. Spanou, C. (1998) European Integration in Administrative Terms: A Framework for Analysis and the Greek case. Journal of European Public Policy 5(5): 467–84. Spanou, C. (2000) Greece, in H. Kassim, G. P. Peters and V. Wright (eds) The National Co-ordination of EU Policy: The Domestic Level. Oxford University Press: 161–81. Steinmo, S. and K. Thelen (1992) Historical Institutionalism in Comparative Politics, in F. Longstreth, S. Steinmo and K. Thelen (eds) Structuring Politics: Historical Institutionalism in Comparative Analysis. Cambridge University Press: 1–32. Strange, S. (1999) The Future of Global Capitalism, or, Will Divergence Persist Forever?, in C. Crouch and W. Streeck (eds) Political Economy of Modern Capitalism: Mapping Convergence and Diversity. London, Sage. Streeck, W. (1995) From Market Making to State Building? Reflections on the Political Economy of European Social Policy, in S. Leibfried and P. Pierson (eds) European Social Policy: Between Fragmentation and Integration. Washington, DC, Brookings Institution: 389–431. Streeck, W. (1996) Neo-Voluntarism: A New European Social Policy Regime? in G. Marks, F. Scharpf, P. Schmitter and W. Streeck (eds) Governance in the EU. 64–94. Streeck, W. (1999) German Capitalism: Does It Exist? Can It Survive?, in C. Crouch and W. Streeck (eds) Political Economy of Modern Capitalism: Mapping Convergence and Diversity. London, Sage: 33–54. Stubb, A. C.-G. (1998) Flexible Integration and the Amsterdam Treaty: Negotiating Differentiation in the 1996–97 IGC (PhD Thesis). London School of Economics and Political Science, Department of International Relations. Sweden (1991) Parliamentary Briefing by the Minister for Foreign Trade, Ms. Anita Grandin, concerning EFTA-EC relations (18 January ). Documents on Swedish Foreign Policy 1990. Stockholm, Ministry of Foreign Affairs: 219–22. Sweden (1994) Speech to the Swedish Parliament on 23 November by the Prime Minister, Mr. Ingvar Carlsson, with reference to the outcome of the referendum on membership of the European Union. Stockholm, Ministry of Foreign Affairs. Sweden (1996a) Strengthening the Fight Against Unemployment: An Issue for the Intergovernmental Conference. Stockholm, Swedish Ministry of Finance.
256 Bibliography Sweden (1996b) Note of July 1995 on the Fundamental Interests of Sweden with a View to the 1996 Intergovernmental Conference. White Paper on the 1996 Intergovernmental Conference: Summary of Positions of the Member States of the European Union with a View to the 1996 Intergovernmental Conference. Brussels, European Parliament: II, 150–52. Sweden (1997) Add to art. 2, page two (Swedish proposals for principles to serve as guidelines for the formulation of Member States’ and the Community’s employment policy). Council of the European Union Cabinet (CAB) document. Swepston, L. (1992) Human Rights Complaint Procedures in the International Labour Organization, in H. Hannum (ed.) Guide to International Human Rights Practice. Philadelphia, University of Pennsylvania Press: 99–116. Symeonidou, H. (1996) ‘Social Protection in Greece.’ South European Society and Politics 1(3): 67–86. Szukala, A. and W. Wessels (1997) The Franco-German Tandem, in G. Edwards and A. Pijpers (eds) The Politics of European Treaty Reform: The 1995 Intergovernmental Conference and Beyond. London, Pinter: 74–99. Teague, P. (1999) Economic Citizenship in the European Union: Employment Relations in the new Europe. London, Routledge. Thatcher, M. (1993) The Downing Street Years. London, HarperCollins. Thatcher, M. (1995) The Path to Power. London, HarperCollins. Timmins, N. (2001) The Five Giants: A Biography of the Welfare State. London, HarperCollins. Titmuss, R. M. (1968) Commitment to Welfare. London, Allen & Unwin. Titmuss, R. M. (1976) Essays on the ‘Welfare State’. London, Allen & Unwin. Toft, C. (1997) German Social Policy, in M. Mullard and S. Lee (eds) The Politics of Social Policy in Europe. Cheltenham, Edward Elgar, 144–69. Trampe, N. (2000) Nordic Labour Markets and European Social Policy: A Perspective from UNICE, in A. C. Neal (ed.) European Social Policy and the Nordic Countries. Aldershot, Ashgate: 179–85. Tsimisizelis, M. (1996) Greece, in D. Rometsch and W. Wessels (eds) The European Union and Member States: Towards Institutional Fusion? Manchester University Press: 216–52. Tsoukalis, L. (1997) The New European Economy Revisited. Oxford University Press. Tyszkiewicz, Z. (2002a) National Members and their EU Associations, in J. Greenwood (ed.) The Effectiveness of EU Business Associations. Basingstoke, Palgrave: 171–81. Tyszkiewicz, Z. (2002b) Review of Hugh Compston and Justin Greenwood (eds.) Social Parternship in the European Union, 2001. EUSA Review, vol. 15, no. 1: (2002 Winter): 16–17. UK (1989–90) Annual Report. London (HMSO), HSC/E. UK (1990) Extract from House of Lords Select Committee on the European Communities Twenty-Seventh Report: Economic and Monetary Union and Political Union, 30 October 1990 (in documents annex), in R. Corbett (ed.) The Treaty of Maastricht, From Conception to Ratification: A Comprehensive Reference Guide. Harlow, Longman: 187–90. UK (1991) Proposals for Health and Safety (General Provisions) Regulations and Approved Code of Practice, Consultative Document. London, Health and Safety Committee.
Bibliography 257 UK (1996a) Proposal for Amendment of Articles 54, 57, and 118A, and for a Protocol Concerning Directive 93/104/EC (Working Time): Memorandum by the United Kingdom. Cabinet (CAB) document, CONF/3978/96 (Annex) CAB. UK (1996b) UK White Paper of 12 March 1996 on the IGC: ‘An Association of Nations’. White Paper on the 1996 Intergovernmental Conference: Summary of Positions of the Member States of the European Union with a View to the 1996 Intergovernmental Conference. Brussels, European Parliament: II, 159–64. UK (1997) Suggested EU Amendment (Article 1). Council of the European Union Cabinet (CAB) document. UK (1999) Annex I/A item note from General Secretariat of the Council. Brussels. Council of the European Union Cabinet (CAB) document (8946/99 Soc 236 ADDI). UN (1947a) Drafting Committee of the Preparatory Committee of the International Conference on Trade and Employment: Administrative Provisions, Suggestions by Canadian Delegation, United Nations. GATT/WTO Archives, Geneva. E/PC/T/C.6/W.58. UN (1947b) Drafting Committee of the Preparatory Committee of the UN Conference on Trade and Employment, Tentative and Non-Commital Draft Suggested by the United States for a General Agreement on Trade and Tariffs, United Nations. UN (1947c) Drafting Committee of the Preparatory Committee of the United Nations Conference on Trade and Employment: Draft Agreement on Tariffs and Trade, United Nations. GATT/WTO Archives, Geneva. E/PC/T/C.6/85. UN (1947d) Final Act Adopted at the Conclusion of the Second Session of the Preparatory Committee of the United Nations Conference on Trade and Employment, General Clauses of the General Agreement on Tariffs and Trade (vol. 1 of 4). Lake Success, New York. UN (1995) Copenhagen Declaration on Social Development and Programme of Action of the World Summit for Social Development. Copenhagen, World Summit for Social Development. UNICE (1988) Observations and Proposals for Amendments Regarding the Proposal for a Directive on the Introduction of Measures to Encourage Improvements in the Safety and Health of Workers at the Workplace. Brussels. UNICE (1989) Speech by R. F. Eberlie, Chairman, Health and Safety Working Party of UNICE on ‘Health and Safety Directives: The Employer’s View’, speech given in Dublin and Cork. (17, 18 May) Brussels. UNICE, CEEP and ETUC (1987) Avis commun du groupe de travail ‘nouvelles technologies et dialogue social’ concernant la formation et la motivation, l’information et la consultation: avis commun du groupe de travail ‘macroeconomique’ sur la rapport economique annual 1987–1988, in P. Venturini (ed.) Un Espace Social Européen à l’Horizon 1992. Luxembourg, Office des publications officielles des Communautés européennes: annexe 4. USA (1946) Suggested Charter for an International Trade Organization of the United Nations. Washington, DC, US Department of State. Valticos, N. (1982) The International Labour Organization, in P. Alston and K. Vasak (eds) The International Dimensions of Human Rights. Paris, UNESCO: 363–99. Valticos, N. (1993) Foreword, in L. Betten (ed.) International Labour Law: Selected Issues. Boston, MA, Kluwer: xvii–xxii.
258 Bibliography van Zonneveld, H. (2000) The Nordic Countries and the European Commission’s Medium-Term Social Policy Objectives, in A. C. Neal (ed.) European Social Policy and the Nordic Countries. Aldershot, Ashgate: 13–17. Venieris, D. N. (1996) Dimensions of Social Policy in Greece. South European Society and Politics. Issue 1.3 London, Frank Cass Publishers: 260–9. Venturini, P. (1988) Un Espace social européen à l’horizon 1992. Luxembourg, Office des publications officielles des Communautés européennes. Verney, S. (1993) From the ‘Specialist Relationship’ to Europeanism: PASOK and the EC 1981–89, in R. Clogg (ed.) Greece, 1981–89: The Populist Decade. London, Macmillan: 47–63. Vogel, L. (1993) Prevention at the Workplace: An Initial Review of How the 1989 Community Framework Directive is Being Implemented. Brussels, European Trade Union Technical Bureau for Health and Safety. Vogel, L. (1998) Prevention at the Workplace: The Impact of Community Directives on Preventive Systems in Sweden, Finland, Norway, Austria and Switzerland. Brussels, TUTB (European Trade Union Technical Bureau for Health and Safety). Vogel-Polsky, E. (1994) Maastricht ou la voie étroite de social, in M. Telo (ed.) Quelle union social européenne? Editions de l’Universié de Bruxelles: 79–96. Wallace, H. (1990) Making Multilateral Negotiations Work, in W. Wallace (ed.) The Dynamics of European Integration. London, Pinter: 213–28. Wallace, H. (2000) The Institutional Setting: Five Variations on a Theme, in H. Wallace and W. Wallace (eds) Policy Making in the European Union. Oxford University Press: 3–37. Wallace, W. (1979) Grand Gestures and Second Thoughts: the response of Member Countries to Greece’s Application, in L. Tsoukalis (ed.) Greece and the European Community. London, Saxon House: 21–38. Wank, R. (1992) Germany, in R. Baldwin and T. Daintith (eds) Harmonization and Hazard: Regulating Health and Safety in the European Workplace. London, Graham & Trotman: 49–75. Webber, D. (ed.) (1999) The Franco-German Relationship in the European Union. London, Routledge. Weber, T. (2001) The European Sectoral Social Dialogue, in H. Compston and J. Greenwood (eds) Social Partnership in the European Union. Basingstoke, Palgrave: 129–53. Wendon, B. (1998) The Commission as Image-Venue Entrepreneur in EU Social Policy. Journal of European Public Policy 5(2): 339–53. Wessels, W. (1997) An Ever Closer Fusion? A Dynamic Macropolitical View on Integration Processes. Journal of Common Market Studies 35(2): 267–99. Wilensky, H. H. (1975) The Welfare State and Equality. Berkeley, University of California Press. Williams, C. (2002) Britain in Historical Perspective: From War Concertation to the Destruction of the Social Contract, in S. Berger and H. Compston (eds) Policy Concertation and Social Partnership in Western Europe: Lessons for the 21st Century. Oxford, Berghahn: 51–61. Wood, S. (2001) Labour Market Regimes Under Threat? Sources of Continuity in Germany, Britain and Sweden, in P. Pierson (ed.) The New Politics of the Welfare State. Oxford University Press: 368–409. WTO (1996a) Agreement Between the WTO and the IMF and the World Bank. Geneva.
Bibliography 259 WTO (1996b) Singapore Ministerial Declaration. Singapore. WTO (1998) Trade Policy Review: India, Minutes of Meeting (Chair: Mr Ali Said Mchumo, Tanzania). Geneva. WTO (2000) European Communities – Measures Affecting Soluble Coffee. Geneva. WTO (2001a) European Communities – Generalized System of Preferences, Request for Consultation by Thailand. Geneva, WTO. WT/DS242/1. WTO (2001b) Statement by H. E. Yannis Zafeiropoulos, Deputy Minister for Foreign Affairs, Greece. Doha Ministerial Declaration. Geneva, WTO. 20 November 2001. WT/MIN(01)/DEC/1. WTO (2001c) Doha Ministerial Declaration. Geneva, WTO. 20 November 2001. WT/MIN(01)/DEC/1. WTO (2001d) Statement by H. E. Mr Laurent Fabius, Minister for Economy, Finance and Industry, France. Doha. WTO (2001e) Statement by H. E. Mr Leif Pagrotsky, Minister for Trade, Sweden. Doha. WTO (2002a) European Communities – Conditions for the Granting of Tariff Preferences to Developing Countries, Request for Consultation by India. Geneva, WTO. WT/DS246/1. WTO (2002b) European Communities – Conditions for the Granting of Tariff Preferences to Developing Countries, Request for the Establishment of a Panel by India. Geneva, WTO. 9 December 2001. WT/DS246/4. WTO (2004a) Appellate Body Report – European Communities Conditions for the Granting of Tariff Preferences to Developing Countries. Geneva, WT/DS246/AB/R. WTO (2004b) European Communities – Conditions for the Granting of Tariff Preferences to Developing Countries – Arbitration under Article 21.3(c) of the Understanding on Rules and Procedures Governing the Settlement of Disputes. Geneva, WTO. 22 September 2004. WT/DS246/15. Yataganas, X. A. (2001) The Treaty of Nice: The Sharing of Power and the Institutional Balance in the European Union – A Continental Perspective. European Law Journal 7(3): 242–91.
Index ACP (African, Caribbean and Pacific countries) 172 active labour market policies (ALMPs) 20, 99–100, 107, 128–38, 140, 171, 198 Amsterdam Treaty 1, 39, 97, 108–23, 129, 188, 191 Australia 164 Austria 9, 16, 30, 61–2, 67–8, 71, 100–1, 122, 152–4, 171 Balassa, Bela 141 Belgium 9, 16, 30, 34, 36, 40, 47, 75, 100–1, 107, 122 behaviour at the ILO 152–4, 163–4, 169, 171, 174–5, 180 policy leader in EU 11, 47–8, 50, 70, 74, 76–7, 85, 112–13, 129, 196 Presidency of the EU 2001 163–4 role of unions in 61–2, 66–8, 78 bench marking see open method of coordination best practice see policy learning Blair, Tony 102, 114, 117, 122 Bourgeois, Léon 4 Bretton Woods 175 Broad Economic Policy Guidelines (BEPGs) 116, 138, 192 Brok, Elamr 112 Canada 154, 164, 176 capitalism, development of 6–7 CEEP see European Centre of Public Enterprises Charter of Fundamental Rights of the Worker (EU) see Social Charter child labour see International Labour Organization Clinton, Bill 177, 179 Cohesion Fund 20, 38, 40–2, 50, 82–5, 188, 192 Colombia 163
Common Agricultural Policy (CAP) 22, 197 Community Method 17–18, 63, 121, 190 see also governance by law competitive advantage 12, 28, 30–1, 33, 67, 143, 156, 203–4 Confederation of British Industry (CBI) 44, 69, 79, 81, 82–92, 170, 190, 193 Conference on Security and Cooperation in Europe (CSCE) 146 conservative corporatist welfare state regime 7–9, 11, 52, 67–9, 98–103, 143, 151–4, 180, 194, 197 see also welfare state regime Corbett, Richard 112 Council (ECOFIN, General Affairs, Social Affairs) 24, 32–4, 36, 37, 39, 43, 46, 50–1, 60, 61–6, 71–8, 81–2, 85–8, 90, 95–6, 104–5, 107–8, 111–12, 117, 120, 124–6, 133, 181 Council of Europe 145–7 Cyprus 174 Czech Republic 159, 167, 174 Degimbe, Jean 80 Delors, Jacques 28–9, 38, 41, 57, 63, 72–3, 81–2, 85, 106–8, 129 Denmark 9, 16, 30, 34–6, 49, 67–8, 70, 75, 77, 79, 86–7, 92–3, 100–1 behaviour at the ILO 151–4, 160, 168, 184 national implementation of EU law 52–3 policy leader in EU 11, 35, 37–40, 47–8, 50, 58, 85, 97, 104–6, 112–13, 115, 129, 138–40, 187, 196–8
260
Index 261 Denmark – continued Presidency of the EU 1993 97, 105–7 role of unions in 35, 61–2 Directorate General Employment and Social Affairs (formerly DG V) 18, 32, 72, 78, 89, 105–6, 108, 124, 127, 129, 158, 160–2, 172 Durkheim, Emile 4 Economic and Social Committee (ESC) 17, 71, 118, 193 entrepreneurs see European Commission Esping-Andersen, Gøsta 7–8 Estonia 167, 174 EU Charter of Fundamental Rights 89, 145–7 EU Constitution (2004) 6, 16, 202 EU Enlargement 15–17, 187, 201–6 1973 158 1981 29, 204, 1986 29, 34–7, 40–2, 58, 64, 204 2004 159, 163, 201, 204 EU Presidency 161–5 see also individual Member States EU voting rules see qualified majority voting; unanimity voting rules European Centre of Public Enterprises (CEEP) 65, 71–6, 78–82, 90 European Coal and Steel Community (ECSC) 29, 31–2, 103, 155–6, European Commission as agenda setter 19 oversight of national implementation 51–2 as policy entrepreneur 17–19, 29, 32–5, 37–8, 43–4, 49–50, 58–60, 63, 65, 72, 77, 79–82, 85, 95–6, 97–8, 103–4, 123–8, 138–41, 178–81, 188–9, 191–2, 205 as process manager 19 as promoter of EU cooperation 32 right to propose legislation 17, 32, 42, 77, 96, 121 see also Delors, Jacques; Directorate General Employment and Social Affairs
European Convention on Human Rights 145–7 European Court of Human Rights 145–7 European Court of Justice (ECJ) 11, 15, 17–18, 20, 22, 39, 49, 53–4, 59, 85, 88–9, 94, 96, 130, 145–7, 150–1, 155, 158–61, 183, 189, 195 European Economic Area (EEA) 67, 71 European Economic Community (EEC) 29, 31–4, 103–4, 155–8 European Employment Strategy (EES) 17, 20, 97–141, 188, 189, 191, 203 Community-Wide Framework for Employment 106–7 Essen criteria 107–8, 118, 124 history of 103–29 and interaction with ILO policies 161, 167, 170–2 National Action Plans under 125–6, 128–38 pillars or themes of 124–5, 130, 134–5 European Monetary Union 8, 18, 35, 82–4, 105, 107–8, 112, 122, 138, 188, 192, 197, 203 employment rates as possible convergence criteria for 105 European Parliament 37–8, 73, 91, 105, 111–12, 118, 162–3, 191–2 European Social Fund (ESF) 20, 50, 57, 103–4, 130, 134, 188, 192, 197 European social model 13, 73 European Trade Union Confederation (ETUC) 43–4, 50, 62–3, 65–7, 71–82, 85–90, 95–6, 170, 193 see also regional governance European Works Council 190 Falkner, Gerda 64 Finland 9, 16, 62, 67–8, 71, 100–1, 112, 122, 151–4, 163, 168–9, 171 Flynn, Padraig 89, 105–7, 129
262 Index France 9, 15, 16, 30–1, 34, 36–7, 64, 75, 87, 99–103, 109 as inconsistent in reaction to EU policy 13, 42, 69, 77, 82, 98, 112–14, 118–23, 129, 139–40, 197 behaviour at the ILO 147, 152–4, 160, 162, 165–71, 175, 180 harmonization of social policy 10 implementation of EU policy 53–4, 58–9, 93–4, 136–7 role of unions and collective bargaining in 62, 65, 67, 69, 79, 88, 169 see also conservative corporatist welfare state regime General Agreement on Tariffs and Trade (GATT) 175–7 see also World Trade Organization Germany 9, 15, 16, 30–1, 49, 59, 75, 87, 90, 99–103 as inconsistent in reaction to EU policy 13, 22, 34–7, 40, 48, 52, 70, 77, 81, 89, 98, 112–14, 116–23, 127, 129, 139–40, 193, 197 behaviour at the ILO 147, 151–4, 160, 165–70, 174–5, 179 concerns about financial costs of EU policy 21, 113–14, 119–20, 121, 188 development of welfare state 5 implementation of EU policy 53–4, 137–8 role of unions and employer groups in 61–2, 67–8, 79–80, 87, 91 unification 76 see also conservative corporatist welfare state regime; Kohl, Helmut Global Employment Agenda see International Labour Organization, employment global governance 142–85 enforcement of 149–51, 195–6 importance of funding to 202
relationship with regional governance 21, 186, 142–5, 151–85, 189–96, 204 see also ILO, supranational governance, WTO globalization 144–5, 178, 179 governance by collective bargaining relationship to other forms of EU social policy 190–4 see also Social Dialogue governance by law 11–12, 28–9, 190–5, 200 relationship to other forms of EU social policy 190–4 governance by open method of coordination relationship to other forms of EU social policy 190–4 see also open method of coordination (OMC) Greece 9, 16, 29, 30, 36, 74–5, 77, 87–8, 96, 99–103, 109, 112–13, 121–3, 139–40 administrative culture 45–6 behaviour at the ILO 151–4, 160, 165–70, 175, 180 and EU funding 20–1, 40–2 and 1990–1 IGC 65 implementation of EU policy 56–8, 92, 132–6 modernization effect of EU policies on 67, 82–4, 132–6, 199 as passive towards EU social policy 12–13, 45–8, 70, 82–5, 119–20, 196–7 Presidency of the EU in 1988 45 resistance to QMV 37, 40–2, 50 role of unions and collective bargaining in 62, 66–9, 79 transformation of national policy by supranational governance 82–4, 92, 168, 198–201 see also southern welfare state regime GSP Scheme (Generalized System of Preferences) 167, 181–3 Guigou Elisabeth 112
Index 263 Hallstein, Walter 156 Hansenne, Michael 178 health and safety 28–60 art. 137 (ex 118a) 11, 15, 32–42, 50, 58–60, 159, 188 and the European Agency for Safety and Health at Work 58, 193–4 EU laws as new standard 49–52, 56–8, 198, 205 Framework Directive (89/391/EEC) 29, 42–58, 189 in ILO 166–9 in new member states (of 2004) 205 Machinery Directive 197 for temporary workers 89 see also governance by law Helsinki Act 146 historical institutionalism 14–22, 25–7, 123, 200, 206 financial incentives and disincentives of 20–1, 206 formal rules 15–17, 50, 85, 204 in explaining gaps in Member State behaviour 13–15, 50, 85, 129–30, 196 in explaining new forms of governance 14, 16–18 institutional learning 19–20, 188 process tracing 23–6, 98 summary of impact on EU social policy 186–90 see also European Commission – as policy entrepreneur; unanticipated consequences Hungary 174 ideology, role in social policy cooperation 22–3, 35, 64, 70, 97, 102, 118, 122, 187–9 India 177, 181–2 Industrial Market Economy Group (IMEC) 155, 162–5 institutionalism see historical institutionalism institutions 14–15, 19–22, 206 see also historical institutionalism Intergovernmental Conferences (IGCs) 18, 24
1985 IGC 34–42, 104 1990–1 IGC 62, 74–85 1996 IGC 106, 111–19 intergovernmental theory 35, 46, 70, 82, 95, 109, 121–3 International Labour Organization (ILO) 4, 89, 106, 142–85, 91 child labour 163, 167, 174, 181 Constitution 147–8 core labour standards 161, 167, 172–5, 178, 181–2, 184, 195, 201–2 Declaration on Fundamental Rights 146, 161, 165, 181 employment policies of 161, 170–2, 201 enforcement mechanisms of 149–51, 183, 195–6, 200–2 evolution of governance 201–2 Global Employment Initiative 20 and health and safety 31–2, 166–9, 198–9, 201 history of 147–51 Ohlin Report (1956) 155–7 regional groups in 162–5, 184–5 relationship with EU 21, 144, 151–85, 194–6 relationship with WTO 167, 175–83, 185, 201–2 Social Dialogue at 167, 169–70, 184 tripartite structure of 61, 148 World Commission 179 International Monetary Fund (IMF) 175 International Organization of Employers (IOE) 170 International Trade Organization (ITO) 175–7 Ireland 9, 16, 30, 36, 59, 62, 65, 67, 69, 75, 79, 81, 87, 100–1, 105, 109, 112, 116–17, 122, 152–4, 171, 175, 197–8 Italy 9, 12, 16, 30, 36, 37, 39–40, 46–8, 50, 62, 67–9, 74–5, 77–8, 88, 100–1, 105, 112, 115, 147, 151–4, 162–3, 165, 168–9, 171, 180, 196
264 Index Japan 33, 164–5 joint decision trap 42–3 Jospin, Lionel 102 Junker, Jean-Claude 124 ‘Keynesian compromise’ 6 Kohl, Helmut 36, 75, 81–2, 85, 95, 112, 120, 198 labour market 6–7, 33, 39, 61, 57, 87, 105, 200, 202 informal 57 Lamy, Pascal 178 Laroque, Pierre 4 Larsson, Allan 19, 108–9, 111, 113–14, 117, 124, 129, 172 Latvia 167, 174 League of Nations 147–8 liberal welfare state regime 8–9, 30–1, 52, 68–9, 98–103, 143, 151–4, 180, 194 see also welfare state regime Lisbon criteria 126–8, 201 Lithuania 174 Lomé Convention 172 Lund, Gunnar 111–12 Luxembourg 9, 16, 30, 36, 67–8, 75, 77, 91, 100–1, 112, 122, 139, 151–4, 162, 169, 171 Presidency of EU in 1985 36, 39 Presidency of EU in 1997 124–6, 129, 189 Luxembourg Compromise 32–3 Maastricht Treaty 1, 19, 59, 61–2, 70, 74–85, 104, 188 Major, John 62, 81, 87, 102, 122, 217–18 Malaysia 177 Malta 167, 174 market economy, development of 6 Marshall, T.H. 3–4 Milward, Alan 96 Mitrany, David 142, 183 Mitterrand, François 37, 38, 42, 76, 95 Moldova 182 Myanmar 150, 163, 175, 177
national experts role in EU committees 17, 43, 46, 49, 58 Netherlands, the 9, 16, 30, 36, 40, 46, 52, 62, 66–8, 70–1, 74–82, 85, 90, 100–1, 112, 117, 122, 152–4, 163, 169, 193 New Approach, the 33–4 New Zealand 164 Nice Treaty 1, 15–16, 127, 140, 191 Ohlin Report (1956) 155–7 Open Method of Coordination (OMC) 17, 126–41, 183, 191 applications of 127 as example of institutional learning 19–20 benchmarking 201 enforcement of 189, 200 evolution of 97, 140, 203–5 possible expansion of 203–5 see also governance by open method of coordination Organisation for Economic Co-operation and Development (OECD) 106, 146, 171, 173 Pakistan 177 Papandreou, Andreas 36, 37, 40 parental leave, Directive on (96/34/EC) 63, 87, 198, 200 part-time work 86–94 see also Social Dialogue pay see wages Pierson, Paul 14 Poland 174 policy entrepreneurs see European Commission policy failures 11, 28, 190 policy learning 34, 50, 64, 85, 105, 121, 127, 129–30, 135–6, 188, 194 see also open method of coordination Portugal 12, 16, 34, 39, 42, 46, 48, 58, 62, 66–7, 69–70, 74–5, 77, 79, 81–5, 97, 100–1, 109, 112, 122, 151–4, 196–7, 201, 204 Presidency of EU in 2000 126–8
Index 265 qualified majority voting (QMV) 15–17, 29, 34–42, 49, 50, 58–60, 63–4, 76–8, 80–6, 116, 119, 129, 187–8, 190–4, 204 regional (EU) governance benefit levels (absence of harmonization in) 202–3 effect on enlargement on 201–6 employers’ participation in 17, 24, 44–5, 61–96, 188–9, 192–4 enforcement and oversight mechanisms 14–15, 17, 20, 22, 95–6, 189, 192, 200, 206 evolution of 11–12, 22, 189–94 (esp. 191), 201–6 financing of 50, 85, 202, 205–6 implementation of 13, 20, 49–58, 90–4, 188–9, 205–6 increasing flexibility of 190–4 informal cooperation in 107–8, 202 limits to 67, 86, 200–6 policy leaders, resisters, passive states 10–14, 50, 85, 129, 196–8, 205 possible expansion of 201–6 relationship to global governance 21, 142–5, 151–85, 186, 194–6, 204 role of EU committees 33, 43, 49, 189 transformation of national policies 198–200, 205 union participation in 17, 24, 44–5, 61–96, 188–9, 192–4 regulatory space 2 Roundtable of European Industrialists 73 Santer, Jacques 108, 115 side payments 40–2, 65, 82–4, 102 Single European Act (SEA) 1, 15, 20, 29, 32, 34–42, 59, 61, 63–4, 73–4, 95, 104–5, 188, 192 Single European Market (SEM) 18, 34–5, 73, 108, 191–2 Slovakia 159, 174 Slovenia 174
Social Action Programmes 18, 74 Social Charter 73, 89, 145–7, 161, 192, 204 social-democratic welfare state regime 7–9, 11, 30–1, 52, 68–9, 98–103, 143, 151–4, 180, 194, 196 see also welfare state regime Social Dialogue 17, 61–96, 111, 188–9, 191 European Works Councils 86–7 Directive on Fixed Term Work (99/70/EC) Directive on Parental Leave (96/34/EC) 63, 87, 89, 198, 200 Directive on Part Time Work (97/81/EC) 86–94, 198 at ILO 167, 169–70, 184 see also governance by collective bargaining social dumping 8, 12 social justice 147 social partners, the 61, 70–1, 94–6, 105, 120, 163, 169–70, 188, 192, 193 see also regional governance social policy child care 132, 203 definition of 3–7, 145–7 enforcement mechanisms 14–15, 17, 20, 22, 189, 192 equality (equal treatment) 107, 156, 159, 170, 173, 184, 200, 203 health 87, 203 housing 203 inconsistent reaction to 13 leaders in 7–8, 10–11, 20, 193, 196–8 passive in reaction to 12–13, 189, 196–8 pensions 193, 203 poverty 203 resisters of 8, 10–12, 20, 189, 193, 196–8 right of association 172, 202 right to strike 149–50, 156, 202 right to impose lock-outs 202 social assistance 4, 87, 203
266 Index social policy – continued social exclusion 193, 203 social insurance 4, 87, 203 social market economy (Soziale Marktwirtschaft) 6 social regulation 1, 4, 6–7, 200, 202 social security (see also social insurance above) 16, 67, 156, 202 values in 4–5, 147–8, 184, 194 see also health and safety, Social Dialogue, and European Employment Strategy; global governance, regional governance; social dumping; Welfare State Social Protocol 19, 64, 67, 71, 81–2, 85–6, 90, 104, 117, 120, 198 social rights 145–7 soft law 127, 193 solidarity 4–5, 129 southern welfare state regime 8–9, 12–14, 30–1, 52, 68–9, 98–103, 143, 151–4, 180, 201 and policy emulation 65, 67, 140, 188 see also Greece; welfare state regime Spain 9, 12, 16, 34, 39, 42, 46, 48, 58, 62, 67, 69–70, 74–5, 77, 79, 81–4, 100–1, 109, 112, 118–19, 122, 151–4, 159, 168–9, 175, 196–8, 204 spillover 98, 200, 203 Stability and Growth Pact 8, 189, 203 Structural Funds 119 subsidiarity 64, 193 supranational governance 1–3, 111, 186–206 effect on national policy 186, 198–201 emergence of 189–94 limits to 200–6 spillover 200, 203 see also regional governance; global governance
Sweden 9, 16, 30, 46, 49, 59, 71, 87, 96, 99–103 behaviour at the ILO 151–4, 165–71, 175, 179–80 EU policy as requiring changes in 199–200 implementation of EU policy 52–3, 92–3, 128–31 policy leader in EU 11, 97, 109–19, 129, 138–40, 187, 196–8 role of unions and collective bargaining in 61–2, 65–8, 79, 109, 111 see also social-democratic welfare state regime Switzerland 154 Thailand 182 Thatcher, Margaret 23, 37, 41–2, 59–60, 107, 169 Trades Union Congress (TUC) 69, 79, 86–92 Treaty of Amsterdam see Amsterdam Treaty Treaty of Nice see Nice Treaty Treaty of Paris (ECSC) 31 Treaty of Rome 22, 32, 103, 156, 158, 191, 193 Treaty on European Union (TEU) 82, 117 see also Maastricht Treaty unanticipated consequences 12, 14, 18, 21–2, 50, 58–60, 85, 86, 94, 123–8, 130, 135 United Kingdom 9, 15, 16, 30, 36, 67, 75, 77, 99–103, 105 as resister of EU social policy 11, 22, 34–42, 45–8, 50, 59–60, 62, 70, 74, 81, 84, 86–92, 94–6, 97–8, 102–3, 114, 116, 129, 139–40, 189, 190–3, 196, 198, 204 behaving against ‘typecasting’ 197 behaviour at the ILO 148, 151–4, 160, 162, 165–71, 175, 179 concerns about financial costs of EU policy 21
Index 267 United Kingdom – continued development of welfare state 5 ideology towards social policy 23, 64, 188 implementation of EU policy 54–6, 91–2, 131–2 opt-in to EU social policy 64, 70, 90, 117–18, 122, 198 opt-out of EU social policy 81–2, 87, 95, 190 role of unions in 62, 69 why it agrees to EU social policy 14 see also liberal welfare state regime; Blair; Major; Thatcher UN Covenant of Fundamental Economic, Social and Cultural Rights 145–7 UN Millennium goals 172 UNICE 40, 44, 50, 59–60, 62, 64–7, 71–91, 95–6, 170, 193 see also regional governance union density 62, 68–9 United States of America 33, 147–8, 154, 164–5, 173, 175–8, 181 Val Duchesse tripartite discussions 72–3, 79, 85 Venturini, Patrick 109 Versailles Peace Conference (1919) 147 Vredeling initiative 63
wages (pay) 67, 73, 155, 202–4 decent reference wage 204 fair wage, concept of 203 welfare capitalism 6 welfare state 5–10 and industrialization 6 and state building 6 resilience over time 200–1, 204 transformation by supranational governance 198–200 see also welfare state regimes welfare state regimes 7–14, 25–7, 30–1, 65–70, 97–103, 121–3, 143, 151–4, 180, 186–9, 194, 196–201 Wilensky, Harold 6 women in part time work 88–9 participation rates (in labour market) 127 World Bank 175 World Trade Organization (WTO) 146, 148, 150, 167, 173, 175–83, 185, 201–2 Article XX (e) on prison labour 146, 150, 176–7 and challenge of GSP scheme of the EU 181–3 proposals for a Social Clause in 180–1 Trade Policy Review Mechanism in 181 see also General Agreement on Tariffs and Trade