The Governance of Active Welfare States in Europe (Work and Welfare in Europe)

The Governance of Active Welfare States in Europe

Tomáš Sirovátka

Work and Welfare in Europe Series Editors: Denis Bouget, University of Nantes, France, Jochen Clasen, University of Edinburgh, UK, Ana Guillén Rodriguez, University of Oviedo, Spain, Jane Lewis, London School of Economics and Political Science, UK and Bruno Palier, Sciences-po Paris, France Titles include: Sigrid Betzelt and Silke Bothfeld ACTIVATION AND LABOUR MARKET REFORMS IN EUROPE Challenges to Social Citizenship Sonja Drobnic and Ana Guillén Rodriguez WORK-LIFE BALANCE IN EUROPE The Role of Job Quality Colette Fagan, Maria Gonzalez Menendez and Silvia Gomez Anson WOMEN IN MANAGEMENT European Employment Policy Neil Fraser, Rodolfo Gutierrez and Ramon Pena-Cassas WORKING POVERTY IN EUROPE Paolo Graziano, Sophie Jacquot and Bruno Palier THE EU AND THE DOMESTIC POLITICS OF WELFARE STATE REFORMS Europa, Europae Karl Hinrichs and Matteo Jessoula LABOUR MARKET FLEXIBILITY AND PENSION REFORMS Flexible Today, Secure Tomorrow? Trudie Knijn WORK, FAMILY AND TRANSITIONS TO ADULTHOOD IN EUROPE Birgit Pfau-Effinger and Tine Rostgaard CARE, WORK AND WELFARE IN EUROPE Rik van Berkel, Willibrord de Graaf and Tomáš Sirovátka THE GOVERNANCE OF ACTIVE WELFARE STATES IN EUROPE

Work and Welfare in Europe Series Standing Order ISBN 978–0–230–28026–7 (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

Tomáš Sirovátka

The Governance of Active Welfare States in Europe Edited by

Rik van Berkel Utrecht University, The Netherlands

Willibrord de Graaf Utrecht University, The Netherlands

and

Tomáš Sirovátka Masaryk University, Czech Republic

Tomáš Sirovátka

Selection and editorial matter © Rik van Berkel, Willibrord de Graaf and Tomáš Sirovátka 2011 Individual chapters © their respective authors 2011 All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No portion of this publication may be reproduced, copied or transmitted save with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, Saffron House, 6-10 Kirby Street, London EC1N 8TS. Any person who does any unauthorized act in relation to this publication may be liable to criminal prosecution and civil claims for damages. The authors have asserted their rights to be identified as the authors of this work in accordance with the Copyright, Designs and Patents Act 1988. First published 2011 by PALGRAVE MACMILLAN Palgrave Macmillan in the UK is an imprint of Macmillan Publishers Limited, registered in England, company number 785998, of Houndmills, Basingstoke, Hampshire RG21 6XS. Palgrave Macmillan in the US is a division of St Martin’s Press LLC, 175 Fifth Avenue, New York, NY 10010. Palgrave Macmillan is the global academic imprint of the above companies and has companies and representatives throughout the world. Palgrave® and Macmillan® are registered trademarks in the United States, the United Kingdom, Europe and other countries. ISBN 978–0–230–25200–4 hardback This book is printed on paper suitable for recycling and made from fully managed and sustained forest sources. Logging, pulping and manufacturing processes are expected to conform to the environmental regulations of the country of origin. A catalogue record for this book is available from the British Library. Library of Congress Cataloging-in-Publication Data The governance of active welfare states in Europe / edited by Rik Van Berkel, Willibrord de Graaf, Tomáš Sirovátka. p. cm. — (Work and welfare in Europe) Includes bibliographical references and index. ISBN-13: 978–0–230–25200–4 (hardback : alk. paper) ISBN-10: 0–230–25200–1 (hardback) 1. Welfare state—Europe. 2. Europe—Social policy. 3. Europe—Politics and government—1989– I. Berkel, Rik van. II. Graaf, Willibrord de. III. Sirovátka, Tomáš. IV. Title. V. Series. HN373.5.G68 2011 361.6 5094—dc22 2011007806 10 9 8 7 6 5 4 3 2 1 20 19 18 17 16 15 14 13 12 11 Printed and bound in Great Britain by CPI Antony Rowe, Chippenham and Eastbourne

Tomáš Sirovátka

Contents

List of Tables and Figures

vii

Notes on the Contributors

ix

Acknowledgements

xiv

RECWOWE Book Series: Work and Welfare in Europe Funding

xv xvii

1 The Governance of Active Welfare States Rik van Berkel, Willibrord de Graaf and Tomáš Sirovátka 2 The Administration of Income Provision and Activation Services Rik van Berkel

1

22

3 Accelerating Governance Reforms: the French Case Anne Eydoux and Mathieu Béraud

38

4 Fragmented Governance Continued: the German Case Irene Dingeldey

62

5 Steering with Sticks, Rowing for Rewards: the New Governance of Activation in the UK Sharon Wright 6 The Governance of Activation Policies in Italy: from Centralized and Hierarchical to a Multi-Level Open System Model? Paolo R. Graziano and Annelies Raué 7 The Liberal Governance of a Non-Liberal Welfare State? The Case of the Netherlands Rik van Berkel and Willibrord de Graaf 8 Marketization in a Federal System: New Modes of Governance in Unemployment Insurance and Social Assistance in Switzerland Franziska Ehrler and Fritz Sager v

Tomáš Sirovátka

85

110

132

153

vi Contents

9 Governance of Activation Policies in the Czech Republic: Uncoordinated Transformation Tomáš Sirovátka and Jiˇrí Winkler 10 Decentralization and Back to Centralization: the Swedish Case Renate Minas 11 Governance of Integrated Activation Policy in Finland Vappu Karjalainen and Peppi Saikku

173

195 216

12 The Governance of Active Welfare States in Europe in a Comparative Perspective Rik van Berkel, Willibrord de Graaf and Tomáš Sirovátka

237

Index

264

Tomáš Sirovátka

List of Tables and Figures

Tables 1.1 Performance indicators in welfare-to-work programmes 1.2 Potential effects of governance reforms 1.3 Public expenditure on labour-market policies in 2007 (percentage of GDP) 2.1 Main actors and agencies involved in administrating Unemployment Benefit (UB) and Social Assistance (SA) schemes (2010) 2.2 Main actors and agencies in the provision of activation services (2010) 3.1 The public service for placement and unemployment compensation (prior to 2005) 3.2 The three circles of public employment services (since 2005) 3.3 Unemployment compensation (insurance and solidarity schemes) and social assistance 4.1 The implementation structure for income protection and activation in Germany (1990s and 2010) 5.1 Implementation structure for income protection and activation in the UK (late 1980s and 2010) 6.1 Employment and social policy governance models: a typology 6.2 Passive and active labour-market policies (ALMP) expenditure in selected EU countries (percentage of GDP, 2005) 6.3 Share of persons with an equalized disposable income below the poverty threshold in the EU 6.4 Relative household consumption-based poverty in Italy, regional distribution 2006–9 6.5 Distribution of tasks before and after the 1997 reforms 6.6 Main Italian trade unions’ internal composition (2009) 7.1 Implementation structure for income protection and activation in the Netherlands (late 1980s and 2010)

vii

Tomáš Sirovátka

13 17 19

28 35 42 43 45 65 91 112

114 115 115 124 129 134

viii List of Tables and Figures

7.2 Volume of started activation trajectories (1999–2005) 7.3 Target groups of Work First (percentage of municipalities in which the named target group is the main target group of Work First programmes, 2005–6) 8.1 Main governance reforms in Switzerland 9.1 Trends in governance of activation after accession to the EU (2004) 10.1 Main governance reforms in Sweden 11.1 Ended unemployment periods in LAFOS (2004–7) 11.2 Network governance model in activation policy

148

148 159 178 196 230 231

Figures 6.1 Total unemployment rate (percentages) 6.2 Expenditure for services targeted to families, elderly people and the disabled (1980–2000) 11.1 Position of Labour Force Service Centres (LAFOS) within the administrative structure for activation policy

Tomáš Sirovátka

114 116

223

Notes on the Contributors

Mathieu Béraud is Senior Lecturer in Economics at the University of Nancy 1 (Henri Poincaré), France, Researcher at 2L2S (Institute of Sociology, Lorraine, France) and GREE (Group of Research on Education and Employment). His fields of research are employment and social policies, as well as the organization of work and working conditions. He is participating in several national and international research programmes in these topics. Willibrord de Graaf is Associate Professor at the Department of Interdisciplinary Social Science, Utrecht University, the Netherlands. His research is in welfare state development, in particular labour market and activation; his interests include the theme of a multicultural society, and in particular the transition from school to work of young migrants. He is also involved with the homeless. Irene Dingeldey is Head of the research unit ‘Changes in the Working Society’, Institute of Work and Labour at the University of Bremen, Germany. She holds a PhD in social science and has worked on various comparative EU research Projects on family tax systems and labourmarket policies. Most recently she has edited a book on ‘Governance of Welfare State Reform’ and contributed to several comparative books on labour-market reforms and the development of human rights. A comparison of Activating Welfare States will be published in Germany in 2011. Her articles have been published in peer-reviewed journals like the European Journal of Political Research, Journal of Social Policy, Feminist Economics and Politische Vierteljahresschrift. Franziska Ehrler studied Political Science and Economics at the Universities of Bern (Switzerland) and Bordeaux (France), and now works as a scientific collaborator at the Swiss Conference for Social Aid and Welfare and as a research assistant at the Centre of Competence for Public Management at the University of Bern. Her main research interests lie within the fields of social policy and family policy. ix

Tomáš Sirovátka

x

Notes on the Contributors

Anne Eydoux is Senior Lecturer in Economics, Researcher at the CRESSLessor, University of Rennes 2, France, and Associate Researcher at the Centre for Employment Studies. She is also a member of the European Network of Experts on Gender, Social Inclusion, Health and Long Term Care (EGGSI). Her fields of research are employment and social policies, social inclusion and gender studies. She has coordinated several national research programmes and is the author of many articles and book chapters on employment, social policies and gender. Paolo R. Graziano is Assistant Professor at Bocconi University, Italy and Visiting Fellow at the Centre for European Studies, Sciences-Politiques, Paris. He has collaborated on a number of comparative projects on topics relating to social policy, Europeanization and social integration. He has published two books in Italian on Europeanization (2004, Il Mulino) and CSR promotion policies (Franco Angeli, 2009, with Matteo Bassoli), has co-edited a book on Europeanization (Palgrave, 2007, with M.P. Vink). He has also published in the following international journals: European Journal for Political Research, European Political Science, European Journal of Social Security, International Journal of Comparative Labour Law and Industrial Relations, Global Social Policy, Journal of Social Policy, Regional and Federal Studies, Social Policy and Administration. Vappu Karjalainen is Senior Researcher at the National Institute for Health and Welfare (THL) in Helsinki, Finland and a Docent in Rehabilitation Science (University of Lapland, Finland). Her fields of interest are long-term unemployment, activation and rehabilitation policy, with a special focus on active social policy and network approaches in organization of welfare services. She has contributed to a recent book on governance of activation policy in Yuri Kazepov (ed.): Rescaling social policies. A comparative study on the path towards multi-level governance in Europe (London: Ashgate, 2010). Renate Minas is Associate Professor of Social Work at Stockholm University, Sweden, and a Researcher at the Institute for Futures Studies in Stockholm. Her fields of interest are welfare state reforms in Sweden and Europe, with a special focus on social assistance schemes and activation policies from a governance perspective. Some recent publications include ‘Social expenditures and public administration – are local social assistance costs in Sweden a matter of organisation?’, International Journal of Social Welfare (2010), as well as ‘The territorial organization of

Tomáš Sirovátka

Notes on the Contributors

xi

European social assistance schemes’, ‘Actors and governance arrangements in the field of social assistance’ and ‘Rescaling development in Europe: convergence or divergence?’, all published in Y. Kazepov (ed.) (2010) Rescaling of social welfare policies. A comparative study on the path towards multi-level governance in Europe (London: Ashgate, 2010). Annelies Raué currently works as Research Assistant at the Institute of Social and Economic Research and Policy (ISERP) at Columbia University, USA, where she focuses on the use and efficacy of social programmes, in particular education, in developing-country settings. She holds a Bachelor of European Studies from the University of Maastricht, the Netherlands (2008) and continued her studies as a Fulbright Fellow at Columbia University in New York, where she received a Master of International Affairs (2010). During her studies she worked with Professor Graziano in Italy at Bocconi University’s Department of Institutional Analysis and Public Management. Fritz Sager is Professor of Political Science at the Centre of Competence for Public Management at the University of Bern, Switzerland. He is specialized in administrative studies and theory, policy research, programme evaluation, organizational analysis and Swiss politics. He has acted as external expert for the Federal Government in various policy fields such as health, transport, environment, security, vocational education and international relations. His research has been published in the Public Administration Review, Governance, Public Administration, Policy & Politics, Political Studies, West European Politics, Journal of Urban Affairs, Public Money & Management, Evaluation, Revue Internationale de Politique Comparée and the Swiss Political Science Review, among others. In 2010 he won the Marshall E. Dimock Award for the best lead article in Public Administration Review during the volume year 2009. Peppi Saikku (M.Soc.Sc.) is Researcher at the National Institute for Health and Welfare (THL) in Helsinki, Finland. She has participated in a number of national evaluation projects on integrated service structures in the fields of social and health care, employment services and rehabilitation. She has had several national publications in these topics. Tomáš Sirovátka is Professor of Social Policy at the Faculty of Social Studies, Masaryk University, Brno, Czech Republic. His research is focused on the labour market, employment and social policy. He has published numerous books in the Czech Republic and contributed

Tomáš Sirovátka

xii Notes on the Contributors

to several comparative books on employment and social policies – recently in A. Cerami and P. Vanhuysse (eds) Post-Communist welfare pathways. Theorizing social policy transformations in Central and Eastern Europe (Palgrave, 2009); in R. van Berkel and B. Valkenburg (eds) Making it personal. Individualising activation services in the EU (The Policy Press, 2007); and in A. Serrano Pascual and P. Magnusson (eds) Reshaping welfare states and activation regimes in Europe (Peter Lang, 2007). He has published in Journal of Comparative Policy Analysis, European Journal of Social Security, International Review of Sociology, International Journal of Sociology and Social Policy, Social Policy and Administration, Journal of Marriage and the Family, Czech Sociological Review and Prague Economic Papers. Rik van Berkel is Associate Professor at the School of Governance of Utrecht University, the Netherlands. His research interests are situated on the interface of social policy, public administration and public organization studies, and include active welfare state transformations, the introduction of new models for the provision of social services, and processes of change in public organizations responsible for the implementation of social policy and governance reforms. He has published several edited volumes and articles in journals such as Public Administration, Journal of Social Policy, Journal of European Social Policy, Social Policy and Society and European Journal of Social Work. Jiˇrí Winkler is Associate Professor in Social Policy and social work at the Faculty of Social Studies, Masaryk University, Brno, Czech Republic. He researches and lectures on the design, management and implementation of public programmes in social policy and employment, and on organizational development in public administration institutions. He has coordinated or has taken part in many international and national research projects, in recent years focusing on labour-market and employment policy or childcare. Recently he has worked as a government expert for the Roma Decade Programme (integration in the labour market) and currently as national expert in the Peer Review Programme (for disability management in the labour market) of the European Commission. Aside from 70 articles in peer-reviewed journals, he is the author of several Czech monographs such as Implementation: institutional aspect of the analysis (2002), Theory of decision-making and social policy dynamic (2007) and co-editor of the Czech monographs Czech labour market and the European employment strategy (2004) and Labour market insecurities (2009).

Tomáš Sirovátka

Notes on the Contributors

xiii

Sharon Wright is Lecturer in Social Policy at the University of Stirling, UK. Her research has included an ethnographic street-level study of employment services implementation from adviser and user perspectives; a 3-year European research project comparing employment advice services for out-of-work adults in five countries; and an in-depth study, based in Australia for six months, comparing welfare reforms and the contracting-out of employment services with the UK. She was Managing Co-editor of the journal Social Policy & Society and member of the UK Social Policy Association Executive Committee (2006–11). Recent publications include Contracting out employment services: lessons from Australia, Denmark, Germany and the Netherlands (Child Poverty Action Group, 2008) and, with Dr Tess Ridge, Understanding Inequality, Poverty and Wealth: policies and prospects (The Policy Press, 2008).

Tomáš Sirovátka

Acknowledgements

This book is the result of one of the activities undertaken in the context of the EU-funded sixth Framework Programme’s Network of Excellence ‘Reconciling Work and Welfare’ (RECWOWE; see http://recwowe.eu). The central objective of RECWOWE is to create a European research network capable of overcoming the fragmentation of existing research on questions of work and welfare in Europe; its aim is to promote the research excellence required in identifying and understanding the multiple tensions between work and welfare, as well as strategies for managing or resolving these. The network was begun in 2006, for a period of 5 years. We would like to thank the RECWOWE participants, and especially the participants and coordinators of the working group ‘Toward employment-friendly welfare states’, for providing an inspiring and stimulating environment for working on this book.

xiv

Tomáš Sirovátka

RECWOWE Book Series: Work and Welfare in Europe

Series Preface Since the late 1970s, both labour markets and welfare regimes have been under intense economic pressure and have been challenged by profound changes in social and demographic structures, as well as in social norms. The work–welfare relationship has always been key to the modern state, but has undergone substantial change over recent decades. On the one hand, approaches to social provision have become more explicit in recognizing the importance of the relationship to employment, such that it is possible to suggest that social policies have become ‘employmentled’. On the other hand, it is possible to argue that there is a profound ‘disconnect’ between labour markets and welfare systems, and that this partly explains the socio-economic problems that some countries are facing in terms of high unemployment and low activity rates in particular. Changes in labour markets, which have often tended towards more flexibilization, often pose challenges to the protective goals of social policies; thus the financial and programmatic configurations of social policies are often held to have impeded job creation. These arguments over the nature and effects of the relationships between labour markets and welfare systems have been, and are, intense and need to be further analysed. These relationships between labour markets and welfare systems constitute the backdrop for this series, which takes as its starting point the tensions that now characterize this centrally important relationship between ‘work and welfare’. Among these tensions, one might mention the tension between firms’ demands for more labour market flexibility and citizens’ needs for economic security; the tensions between the increased participation in paid work and the importance of family life; the greater fluidity in family relationships; the greater flexibility in the labour markets; the friction between quantity and quality of the jobs to be created, between job creation and maintaining or improving the quality of employment; and finally, the conflicts raised by the need to adapt (industrial) social protection systems to new labour market structures. xv

Tomáš Sirovátka

xvi

RECWOWE Book Series: Work and Welfare in Europe

This book series has been created within RECWOWE, a European network of excellence created within the 6th Framework Programme of the European Commission (FP6). The full title for the network’s activities is ‘Reconciling Work and Welfare in Europe’. Such a reconciliation is an important political objective for most EU member states. In order to meet it we need to improve our understanding of the relationship between work and welfare in very diverse national settings across member states. The RECWOWE series publishes books that analyse work and welfare, and that pay particular attention to the tensions that now characterize this relationship. Beyond the main tensions mentioned above, the series is open to any other forms of analysis of the relationship between work and welfare. It is committed to publishing work that focuses on the gender dimension of this relationship, on the impact of the relationship on migrants and, at its multi-level – European, national and local institutional dimensions. Denis Bouget, University of Nantes Jochen Clasen, University of Edinburgh Ana Guillen, University of Oviedo Bruno Palier, Sciences-po Paris Jane Lewis, London School of Economics and Political Science

Tomáš Sirovátka

Funding

This book has been published thanks to the European research project RECWOWE (Reconciling Work and Welfare in Europe), 2006– 2011, co-funded by the European Commission, under 6th Framework Programme for Research – Socio-economic Sciences and Humanities (contract nr 028339–2) in the Directorate-General for Research. The information and views set out in this book are those of the authors and do not necessarily reflect the official opinion of the European Union. Neither the European Union institutions and bodies nor any person acting on their behalf may be held responsible for the use which may be made of the information contained therein.

xvii

Tomáš Sirovátka

This page intentionally left blank

Tomáš Sirovátka

1 The Governance of Active Welfare States Rik van Berkel, Willibrord de Graaf and Tomáš Sirovátka

Introduction All over Europe, activation or active labour-market policies have become important instruments to promote the transition from welfare to work, that is, to (re)integrate unemployed people dependent on social insurance or social assistance in the labour market. Activation policies, that is programmes and services aimed at promoting the employability and labour-market participation of unemployed people, play a key role in making welfare states more activating and in reducing welfare state dependency. The European Employment Strategy, which contains specific guidelines with respect to the introduction of activation programmes, clearly endorses this orientation of social policies. European countries reveal significant variety in their activation strategies and policies: in the public expenditure on activation,1 in the nature and content of activation programmes, in the coverage of programmes and in the treatment of various target groups, and so on (Barbier, 2004; Dingeldey, 2007). Nevertheless, the social problems of unemployment and welfare state dependency are dealt with in quite different ways nowadays in the European countries compared with, say, two decades ago. Social policy analysts have written extensively on this new trend in welfare states, comparing and analysing the content of activation programmes, the consequences of the activation of welfare state arrangements for unemployed people, the effects of programmes on labour-market (re)entry, etcetera (see, for example, Lødemel and Trickey, 2001; Gilbert, 2002; Van Berkel and Hornemann Møller, 2002; Handler, 2004; Serrano Pascual, 2004; Hvinden and Johansson, 2007; Serrano Pascual and Magnusson, 2007). These content-related aspects of active welfare state reforms are not the core topic of this book. Instead, in 1

Tomáš Sirovátka

2

The Governance of Active Welfare States

this book we will look at another type of welfare state reforms, namely those that some have called reforms of operational policies (Carmel and Papadopoulos (2003) distinguish operational and formal policies, where the latter regulate substantial, content-related aspects of social policies and the former the organization of policy delivery). These reforms affect the ways in which the provision and delivery of social policies are organized or, put differently, the governance of social policies (Borghi and Van Berkel, 2007). The concept of governance has been widely discussed in the social sciences, and has been defined in a variety of ways (Rhodes, 2000; Newman, 2001; Hill and Hupe, 2002). For some authors, governance is practically synonymous with government. For example, for Kettl (2002, p. ix) governance ‘is the way government gets the job done’. Others argue that governance refers to ‘a change in the meaning of government, referring to a new process of governing; or a changed condition of ordered rule; or the new method by which society is governed’ (Rhodes, 2002, p. 209). In this context, governance may be an umbrella concept referring to a variety of forms of governance. Rhodes (2002), for example, distinguished six uses of the concept of governance. It may also have a more narrow and specific definition, such as Jessop’s (2003, p. 101): ‘the reflexive self-organisation of independent actors involved in complex relations of reciprocal interdependence, with such self-organisation being based on continuous dialogue and resource-sharing to develop mutually beneficial joint projects and to manage the contradictions and dilemmas inevitably involved in such situations’. For the purpose of this book, an encompassing rather than specific definition is more adequate, as we want to be able to distinguish various types, forms, models or regimes of governance. Rather than developing our own definition, the following definition of Kooiman and Bavinck (2005, p. 17) covers the way in which governance will be used in this book quite accurately: ‘Governance is the whole of public as well as private interactions taken to solve societal problems and create social opportunities. It includes the formulation and application of principles guiding those interactions and care for institutions that enable them’. Both the activation of unemployed people and providing them with income support are actions taken to solve societal problem and create social opportunities – irrespective of questions regarding the desirability or effectiveness of these actions. In providing activation and income support, a range of both public and private actors is involved: national and municipal authorities; public or semi-public agencies such as benefit agencies, Public Employment Services (PES), welfare agencies; private actors such as reintegration companies, NGOs, employers’

Tomáš Sirovátka

Rik van Berkel et al.

3

and employees’ organizations, and so on. The principles guiding these interactions are laid down in the operational policies mentioned before. They may regulate the tasks and responsibilities of both national, regional and local state actors; they structure interactions between public and private agencies (for example, through quasi-market or network principles) and between various public agencies (for example, through principles guiding inter-agency cooperation and coordination). Care for institutions that enable them points at the importance of supporting and facilitating public and private actors in contributing to the solving of societal problems, and to the processes of capacity building for policy implementation that, as we will see, is a recurrent issue in active welfare state reforms. Governance reforms have become increasingly important aspects of public sector and welfare state reforms in general, and of the delivery of income protection and activation programmes for unemployed people specifically. Elsewhere it has been argued that, although reforms of the governance of income protection and activation programmes cannot be seen in isolation from wider public sector reforms, the transformation of passive into active welfare states has created its own governance reform urgencies (Van Berkel, 2010). Following the definition of governance presented above, activation involves new actions to solve societal problems and create social opportunities, and these ask for a reconsideration of public as well as private interactions, for the principles underlying these interactions and for the organizational capacities that these actions require. For example, income provision and activation are integrated in active welfare states as part of the conditionality of income provision. In contexts where both functions are organized separately, inter-agency cooperation and coordination are considered necessary. In addition, activation services are an important instrument in promoting unemployed people’s employability and labour-market participation. In particular where the long-term unemployed are concerned, individualized services are considered necessary to make activation successful, which may ask for local decision-making space in designing services or for the involvement in service-provision processes of a variety of providers of social services. In recent years, social scientific researchers have begun paying attention to these governance reforms in the policy areas of activation and income support programmes for unemployed people of working age. This has stimulated interdisciplinary approaches integrating insights and theories from the social policy, public administration and public management research fields. By now, publications are available on

Tomáš Sirovátka

4

The Governance of Active Welfare States

a range of specific governance-related issues in these policy areas (see Van Berkel and Borghi, 2008), such as marketization (Struyven and Steurs, 2005; Van Berkel and Van der Aa, 2005), contractualism (Sol and Westerveld, 2005), inter-agency cooperation (Lindsay and McQuaid, 2008) and individualized service provision (Van Berkel and Valkenburg, 2007), whereas other publications focus on governance in a broader sense (Considine, 2001; Bredgaard and Larsen, 2005; Larsen and Van Berkel, 2009). This book aims to contribute to this relatively new strand in social scientific research.

Research questions This book analyses and compares governance reforms in the policy areas of activation and income protection for unemployed people in nine European countries. The reason for focusing on these policy areas is that – as mentioned before – they have become more closely linked and integrated, partly as a logical consequence of the increasing conditionality of entitlements to income provision. The book covers the UK, Germany, France, Italy, Sweden, Finland, Czech Republic, the Netherlands and Switzerland; these countries represent Europe’s diversity in various ways. The countries include both large and small European countries, and they also include social-democratic (Sweden, Finland, Netherlands), liberal (UK), conservative (Germany, France, Switzerland), southern European (Italy) and post-communist (Czech Republic) welfare states. In addition, they cover countries with above and below EU average public expenditures on labour-market policies (see note 1). And from a public administration point of view, the selection of countries discussed in this book includes countries with a conservative trajectory of public management reform, modernizers, as well as marketizers (Pollitt and Bouckaert, 2000). The book is the result of joint work by the authors in the context of a network of European researchers funded by the EU’s 6th Framework Programme: Reconciling Work and Welfare (RECWOWE). Two sets of research questions will be addressed in this book, the first addressing the nature of governance reforms. What governance reforms have been taking place in the countries? In terms of the definition of governance used in this book, this question refers to the public and private actors involved in the provision of activation and income support, on the structure of their interactions and the principles guiding these interactions, as well as the ways in which these actors are equipped to carry out their tasks and responsibilities. In addition, these reforms

Tomáš Sirovátka

Rik van Berkel et al.

5

will be analysed in terms of various guiding principles of new modes of governance: – new distributions of centralized and decentralized policy making and policy implementation responsibilities; – the introduction of quasi markets, based on the principle of competition, in the provision of activation services and programmes; – the promotion of inter-agency cooperation and partnerships involving the agencies traditionally responsible for these policy areas (social insurance funds, local welfare agencies and the PES), as well as other service providers; – the introduction of new public management instruments affecting interactions between various levels of the state and interactions between national and local offices of public agencies. At a more abstract level, the question can be raised whether we see shifts of governance regimes in the countries involved in the study. The literature on governance distinguishes several of these governance regimes, which will be elaborated upon below. The second set of questions concerns the intended and unintended effects of the reforms. These questions can be approached from two points of view. First, one can evaluate whether or not governance reforms promote the labour-market participation of unemployed people: many of these reforms are justified by arguing that they support the reforms aimed at making welfare states more activating, and that they contribute to promote welfare-to-work transitions. In other words, governance or operational policy reforms are expected to support and promote the effective implementation of the new active welfare state paradigm. Secondly, the question can be raised on how governance reforms affect the nature and content of welfare state services, as well as the treatment and approach of the unemployed. Rather than interpreting new governance systems and structures as merely technical ways to get things done, that is, to administer benefit systems and effectively provide activation and employment services, we will address the issue of the impact of governance reforms on substantial characteristics of social policies: the nature, quality and responsiveness of services, the treatment of unemployed people in service provision processes, etcetera. The literature on active welfare state reforms has paid considerable attention to the ways in which substantial policy reforms affect the rights and obligations of unemployed people, and to the impact of modern welfare states on social citizenship (Peck, 2001; Gilbert, 2002; Handler,

Tomáš Sirovátka

6

The Governance of Active Welfare States

2004). This book wants to explore whether governance reforms may have similar effects. The remaining part of this chapter is structured as follows. In the next section, some of the main characteristics of, and guiding principles underlying governance reforms will be discussed. Then, attention will be paid to the literature on governance regimes. These sections provide the framework for analysing governance reforms in the following chapters. The fifth section of this chapter focuses on effects of governance reforms. In the final section, the structure of the book will be presented.

Governance reforms: principles, actors, capacities In this section we will discuss some of the main principles underlying governance reforms that are discussed in the literature and are relevant for the social policy areas analysed in this book. In discussing these principles, attention will also be paid to the increasing diversity of actors involved in these policy areas, as well as to the issue of organizational capacities that welfare state reforms require. Decentralization In their comparative study of public management reform, Pollitt and Bouckaert (2000) deconstructed the concept of decentralization by distinguishing three strategic choices: – Political decentralization versus administrative decentralization. In the first case, policy authority shifts from national government to regional or local elected political representatives. In the second, authority is shifted to some (formally independent) public agency. In the context of activation and employment policies, Mosley (2009) described this distinction in terms of a decentralization of policy responsibilities versus a decentralization of managerial discretion. – Competitive versus non-competitive decentralization. This refers to the means by which authority is passed to an agency and whether or not competition is involved. – Internal versus external decentralization. This refers to the issue of whether authority is passed within the organization or to an independent entity outside the organization. Focusing specifically on activation or employment services, Greffe (see OECD, 2003) has listed several reasons why national governments decide to opt for decentralization, most of which relate to societal

Tomáš Sirovátka

Rik van Berkel et al.

7

changes and policy responses to these changes that require policy making and/or policy delivery processes in proximity to the contexts and collective and individual actors involved: the multifaceted nature of unemployment, the increasing volatility of the labour market in a globalizing and knowledge-based economy, the aim to develop innovative solutions through local partnerships, the transformation of passive into active welfare states and – for programmes aimed at promoting social inclusion in a broader sense than paid employment – the identification and development of activities that can promote social participation. In the literature on decentralization in the context of activation or labour-market policies, the strategic choices distinguished by Pollitt and Bouckaert (2000) are clearly present (for example, Giguère, 2005). First, internal decentralization takes place in public agencies involved in benefit administration and providing activation services. For example, national agencies responsible for PES may decide to give regional/local agencies more room in making decisions regarding programme design and programme delivery, in order to increase their ability to adapt national policies to regional/local circumstances. Secondly, political decentralization occurs when national governments devolve policy competences to regional or local governments. Similar motives may be at stake here, namely providing regional or local actors with the authority to develop services and policy programmes that are adjusted to local labour-market circumstances or to the characteristics of local target populations. Thirdly, competitive decentralization takes place when quasi-markets are established for the provision of publicly financed services such as activation and employment services. We will return to this later in the chapter. Capacity building is considered to be a major issue in decentralization processes. Mosley argues that the decentralization of responsibilities for social policy programmes to regional or local levels requires, among others, ‘being able to coordinate local actors, analyse local needs, develop appropriate strategies, implement programmes, monitor, control and evaluate performance, and comply with accountability standards that may be required by higher level authorities’ (Mosley, 2009, p. 37). In discussing the various strategic choices concerning decentralization, Pollitt and Bouckaert (2000, p. 85) make an important observation that is relevant in the context of our social policy areas as well: Centralization is also part of the picture: as some authority has been decentralized, simultaneously there have been significant instances of a tightening of central control and oversight.

Tomáš Sirovátka

8

The Governance of Active Welfare States

In other words, when analysing governance reforms, we should focus on both decentralization and centralization trends. These trends may occur diachronically: periods of decentralization may alternate with periods of centralization, and vice versa (Pollitt, 2008). They may also take place synchronically, for example, when decentralization is accompanied by measures that should ensure that regional/local actors act in accordance with national policy objectives. In this context, the introduction of what has been called an incentive paradigm (Van der Veen and Trommel, 1999) in public administration and the use of new public management instruments (such as performance indicators, benchmarking processes, funding mechanisms) may influence local decision-making processes. Thus, even though decentralization does increase the room for local decision making and for tailoring activation programmes to local needs, the conditions under which decentralization takes place may put considerable constraints on the decisions local actors make, and may in practice function as a process of implicit centralization. Centralization may also take place as part of the capacity-building process. Municipalities may decide or be obliged to share resources with other municipalities in order to professionalize service provision processes. In several countries (for example, Denmark, the Netherlands), public administration reforms have been accompanied by mergers of municipalities in order to increase the scale at which local services are provided.

Marketization and competition Another new form of governance in welfare-to-work programmes is the introduction of principles of subcontracting, marketization and competition in service provision, accompanied by splitting the roles of service purchaser and service provider. Although subcontracting does not necessarily have to take place in a competition context (Sultana and Watts, 2004), our focus here will be on – using Pollitt and Bouckaert’s (2000) terminology – competitive decentralization. Markets for providing activation and employment services often reveal characteristics of quasi-markets: purchasing power does not come from the consumers of services (unemployed people) but from the state and state agencies that act as purchasers on the market (LeGrand, 1991). The idea behind the introduction of quasi-markets in service provision is that promoting competition will have a positive impact on the efficiency and effectiveness of services, on their quality and flexibility or responsiveness to local and individual needs, and on their price.

Tomáš Sirovátka

Rik van Berkel et al.

9

Markets of activation services may differ in numerous aspects. Firstly, the nature of the service purchasers may vary. Purchasers may be PES agencies or benefit agencies, and purchasing may take place at central, regional or local levels. In some cases, unemployed people act as purchasers (even though their purchasing power still comes from the state) through instruments such as vouchers, personal reintegration budgets or individual reintegration agreements (Sol and Westerveld, 2005; Kaps, 2009). Secondly, the nature of the providers may vary as well: they may involve (semi-)public agencies, private non-profit organizations and NGOs or private for-profit companies. Thirdly, the nature of contracts between purchasers and providers can be different. For example, Struyven et al. (2002) distinguish between classic contracts and trust contracts. Trust contracts are concluded for a longer time period than classic contracts, are less detailed in specification and involve more intense cooperation between purchaser and provider. Contracts may also differ in their underlying incentives for performance, for example, in terms of whether they are based on no cure, no pay or no cure, less pay mechanisms (Bruttel, 2005). Fourthly (partly related to the former), competition can vary in strength and in nature (for example, competition on price or on quality). Fifthly, the degree and nature of state regulation of the market can be different: states can rely on market selfregulation or decide to introduce regulations concerning, for example, market access, price, or nature of services. In addition, states can oblige the agencies that act as purchasers to outsource services fully or partially, or can leave this decision with the agencies. Finally, the roles of, and distribution of tasks between, purchasers and providers can be different. For example, purchasers can act as referral agencies, leaving decisions about the nature of services to the providers; or they may intervene more directly in individual activation processes of clients (Jewell, 2007). These examples of market diversity make clear that apart from the question of whether countries introduce quasi-markets for the provision of activation services, the characteristics of these markets are also an interesting research topic. Once again, building organizational capacities to enable public agencies to act as purchasers in the service market is an important topic. These agencies will have to gain insight into the market, the providers and the kind of services that their clients need, to develop tendering procedures and monitor the performance of providers, and to decide about how they want to design relationships with external providers. Of the various new forms of governance, marketization has probably been discussed and investigated most (Considine, 2001; Bredgaard

Tomáš Sirovátka

10

The Governance of Active Welfare States

and Larsen, 2005; Sol and Westerveld, 2005; Struyven and Steurs, 2005; Van Berkel and Van der Aa, 2005). These studies make clear that quasimarkets for the provision of activation and employment services are not a typical liberal welfare state phenomenon. Even though the USA and Australia were the first to introduce quasi-markets, they can now be found in all welfare regime types, as later chapters will illustrate. Inter-agency cooperation and partnerships Inter-agency cooperation and partnerships are the third principle underlying new forms of governance. We will use inter-agency cooperation to denote cooperation between public agencies traditionally involved in benefit administration and employment services: benefit agencies, local welfare agencies and PES. Strictly speaking, it is not a new form of governance (Newman, 2001), but nevertheless, major reforms aimed at promoting inter-agency cooperation have been taking place across the EU. Cooperation between these agencies can be promoted for several reasons: – The increasing conditionality of income support requires exchange of information regarding the activation activities of unemployed people and their compliance with obligations. – Coordination between income protection schemes may be needed to promote efficient and effective activation of unemployed people and to prevent transferrals between income protection programmes. – Service coordination and integration may also be promoted in order to make services more accessible and user friendly. Inter-agency cooperation can take a variety of forms. For example, agencies can bring their front offices together under one roof, or establish a new agency in which their front office activities are shared and direct client contacts take place. These so-called one-stop agencies should prevent the requirement for clients to be referred from one agency to another, or from one counter to another. More far-reaching forms are possible too, and may involve full mergers of formerly independent agencies. Partnerships involve more far-reaching forms of cooperation and coordination than inter-agency cooperation, in the sense that they include a large range of public and private actors in service provision. They are a new form of governance proper: according to Jessop (2000), they are a response to the failures of (quasi-)markets without a return to the state which, in his view, was politically and ideologically

Tomáš Sirovátka

Rik van Berkel et al.

11

unacceptable. Geddes and Benington (2001, p. 2) argued that partnership is a concept ‘to promote integration both vertically (between different tiers of government – European, national, regional, local and grass roots) and horizontally (between different spheres of society – public, private, voluntary and civil)’. Partnerships are attempts to govern complex social problems, and in the context of social security and activation are considered a fruitful approach in providing services for the long-term and most vulnerable groups of unemployed, who – at least in a significant number of cases – are not only confronted with problems regarding labour-market participation, but also with problems in areas such as housing, drug addiction, health, debts, or child care. A study of partnerships carried out in the 1990s by the European Foundation for the Improvement of Living and Working Conditions (Geddes, 1998) identified several factors that contribute to building partnerships and directly refer to the capacity-building issue: an identification of the benefits gained by cooperation; strong leadership; skilled management and staff; strong shared local identity; active participation of all partners in strategic decisions and implementation; seeing new solutions to problems; obtaining new resources, as well as maintaining a solid resource base; appropriate skills training and development. New public management In many publications, new public management is defined in a broad sense as a new public administration paradigm that includes several of the guiding principles underlying interactions between public and private actors, as discussed above. For example, Carroll and Steane (2002, p. 196) write that new public management is a form of public management 1. whose proponents affirm that private sector management models and techniques can be applied in the public sector; 2. that is associated with a commitment to plural models of the provision of public services by a mixture of business, the non-profit sector and government actors, emphasizing the importance of cost, choice and quality in the precise mix of service providers; 3. that is associated with a revised role for government in the provision of public services, characterized as ‘steering not rowing’; 4. that expresses a strong belief in the role of the market and quasimarket mechanisms in coordinating the supply and demand for public services and the use of contractual mechanisms for the governance of provision;

Tomáš Sirovátka

12

The Governance of Active Welfare States

5. that attempts a separation of the political decision-making processes from the management of public services. In this book, we will use the concept of new public management in a somewhat narrower sense, focusing on the principles guiding the ways in which national governments steer public agencies and manage their conduct and, more specifically, their performance. Newman (2001) used the concept of politics of performance in this context, highlighting the pivotal role of results and effects in the new public management. Van der Veen and Trommel’s (1999) concept of incentive paradigm emphasizes that active welfare states are based on an image of human conduct – which not only includes the conduct of unemployed people but also of collectivities such as public agencies and other actors involved in social policy making and implementation – that is driven by economic and rational calculations: Citizens and organizations are motivated by self-interest (or organizational interest), and will always weigh the costs and benefits of conformist behavior. In such a model, the public interest requires that these calculations are influenced in such a way that self-interest (or organizational interest) corresponds as far as possible with public goals. Van der Veen and Trommel (1999, p. 294) In their volume on contractualism in employment services, Sol and Westerveld emphasize the importance of management by objectives and what they call internal public contracts ‘which are aimed at managing public actors by means of agreements on outputs and are directed at demarcating the playing fields for PES and lower-level public authorities such as municipalities’ (Sol and Westerveld, 2005, p. 384). This literature shows that new public management is characterized strongly by performance steering and by using incentives and disincentives as steering mechanisms. One of the core issues in debates about new public management is the issue of how to define performance indicators and what to measure in performance measurement. In their famous book on the entrepreneurial state, Osborne and Gaebler (1992, pp. 356–7) provide the following overview of performance indicators in the context of welfare-to-work programmes, that clearly show the political nature of decisions regarding selected indicators (Table 1.1).

Tomáš Sirovátka

Rik van Berkel et al. Table 1.1

13

Performance indicators in welfare-to-work programmes

Indicator

General definition

Welfare-to-work

Output (or process)

Volume of units produced

Numbers of people trained

Outcome (or result)

Quality/effectiveness of production: degree to which it creates desired outcomes

Numbers of people placed in jobs, working and not in receipt of welfare after six months, one year and beyond. Impact on their lives

Programme outcome

Effectiveness of specific programme in achieving desired outcomes

Numbers placed in jobs, working and not in receipt of welfare after six months, one year and beyond. Impact on their lives

Policy outcome

Effectiveness of broader policies in achieving fundamental goals

Percentage of potential work force unemployed, on welfare and in poverty; percentage of welfare population in receipt of welfare more than one year, five years

Programme efficiency

Cost per unit of output

Cost per job trainee; placement; retained job, etcetera

Policy efficiency

Cost to achieve fundamental goals

Cost to achieve desired decrease in unemployment, poverty rate, welfare caseload

Programme effectiveness

Degree to which programme yields desired outcomes

Numbers placed in jobs, working and not in receipt of welfare after six months, one year and beyond. Impact on their lives

Policy effectiveness

Degree to which fundamental goals and citizens needs are met

Effect on larger society: for example, poverty rate, welfare caseload, crime rate, later spending to remediate poverty

Derived from Osborne and Gaebler (1992), Appendix B.

Governance regimes Another, though not completely different, theoretical perspective for analysing governance reforms comes from the literature on governance regimes. These regimes can be interpreted as ideal-type governance

Tomáš Sirovátka

14

The Governance of Active Welfare States

models and can be helpful in identifying broader governance reform trends. Newman (2001) distinguished four models of governance: the hierarchy model, the rational goal model, the open-system model and the self-governance model. These models are described as follows (Newman, 2001, p. 38): – The hierarchical model corresponds with what others (for example, Denhardt and Denhardt, 2000) called the old public administration, and is oriented towards control, predictability and accountability and based on formal authority. – The rational goal model captures the new public management reforms in a broad sense, with their strong emphasis on performance and (dis)incentives. It is oriented at maximization of output and outcome as well as economic rationalism, and based on managerial power. – The open-system model is most closely related to the partnership model mentioned before. Its focus is on flexibility, expansion and adaptation, and it is based on flows of power within networks (network governance is a concept more or less akin to this model of governance). – The self-governance model is a governance model that is absent in most of the other governance literature, although some of its characteristics will sound familiar. It is oriented towards devolution, participation and sustainability, and based on citizen and community power. Considine’s (2001) study of the public management of welfare-towork is one of the studies that does not specifically distinguish a self-governance model. This exemplifies the concept that citizen or community power is particularly underdeveloped in governing the social problems of unemployment and social exclusion, which in the context of active welfare states are so often constructed in terms of individual responsibilities and obligations rather than individual entitlements and empowerment – in as far as the concept of empowerment is present in the activation discourse, it refers mainly to the (rather paternalistic) notion that labour-market participation will empower citizens rather than putting them in charge of their lives or, more specifically, their activation (Van Berkel and Valkenburg, 2007). Considine’s study distinguishes four governance models. In this book, we will use these models as a heuristic typology in analysing and comparing the broader trends that characterize the often numerous specific governance reforms that have been taking place in the countries in this

Tomáš Sirovátka

Rik van Berkel et al.

15

study. Considine described these governance models in terms of four dimensions: the sources of rationality, forms of control, primary virtues and service delivery focuses that characterize each model. These four dimensions can be seen as an elaboration of the principles guiding the interactions between (both public and private) actors mentioned in the definition of governance discussed in the above. First, he distinguished procedural governance: its source of rationality is the law, the form of control used are rules, its primary virtue is reliability, and the service provision focus is on universal treatment. In other words, procedural governance resembles the old public administration or what Newman called the hierarchical model of governance. Newman’s rational goal governance model is represented in two of Considine’s types of governance: corporate governance and market governance. The sources of rationality are management in the corporate governance type, and competition in the market governance model. The forms of control used are plans and contracts respectively. Whereas the primary virtue in the corporate governance type is that it is goal-driven, being cost-driven is the primary virtue of the market governance model. This distinction is also present in the service delivery focuses: targets in the corporate governance model, prices in the market governance type. Interestingly, on the basis of his empirical research Considine concludes that the corporate and market governance types are in fact representing a single strategy: ‘these two governance orientations, derived from the reform literature and previous research, are a single strategy. Officials do not distinguish them as alternate approaches, rather they successfully synthesize them into a new form’ (Considine, 2001, p. 149). Considine’s fourth governance model is network governance, which more or less resembles Newman’s open-system model of governance. Its source of rationality is relationships, its form of control co-production. The primary virtue of this type of governance is flexibility, and the service delivery focus is brokerage. Considine distinguishes two forms of network governance which are of relevance in the context of social security and activation policies. The first refers to collaboration between officials and organized interests, which in these policy areas focuses attention towards the role of the organized interests of employers and employees, which used to be – and in several cases still is – significant in many welfare states. In the second form of network governance, ‘clients, suppliers and producers are linked together as coproducers’ (Considine and Lewis, 2003, p. 134). In general, we expect to find a trend away from the traditional procedural governance model. At the same time, it is more difficult to predict to what governance regime the reforms will lead us. Here, we need to bear in mind that in practice, we may find elements of all regimes, as

Tomáš Sirovátka

16

The Governance of Active Welfare States

Martin (2001, p. 209) concluded: ‘This suggests not the arrival of a new, hegemonic “outcome-focused paradigm” but a more gradual transition characterized by the co-existence and interaction of hierarchical, market based and collaborative frameworks for coordinating service delivery.’

Effects of governance reforms The literature on new forms of governance often contains extensive lists of advantages or promises that proponents and advocates of these new forms of governance expect will result from their introduction. Decentralization is seen as promoting policy flexibility, the adaptability and responsiveness of policies and programmes to local needs and circumstances, and the capacities of local partnerships to develop integrated and tailor-made solutions to local problems (OECD, 2003; Mosley, 2009). Creating holistic approaches that deal with problems of silo management are seen as major advantages of networks and partnerships. They are also expected to be more innovative than traditional forms of governance, to improve effectiveness and efficiency of policies and programmes by mobilizing resources from a variety of actors and avoiding overlap in services, and to increase accessibility of services and client friendliness (Newman, 2001). The main advantages of quasi-markets for providing public services are expected to be effectiveness and efficiency. Introducing a split between purchasers and providers of services and competition between providers is expected to make services not only cheaper, but also of higher quality, less standardized and more client focused and tailor-made (Struyven and Steurs, 2003). Gains in terms of effectiveness and efficiency are also expected from the introduction of new public management in the public domain (Osborne and Gaebler, 1992). Of course, the real story will probably be a lot more complex than these rather straightforward hypotheses regarding the results of specific forms of governance. First of all, no such thing as a ‘pure’ form of governance exists in practice: as we shall see, the governance of income protection and activation policies consists of a mix of governance regimes in all countries in our study. Secondly, as we saw above, forms of governance come in a variety of sorts. As Pollitt (2002) argued, although there may be convergence in talk, decisions, practices and (potentially) results concerning administrative change, ideas, practices and preferences differ considerably from country to country. Thirdly, governance practices and the implementation of governance reforms may deviate from governance models introduced by governments. From

Tomáš Sirovátka

Rik van Berkel et al.

17

the implementation literature we know that a variety of factors may explain this: for example, local actors’ interpretation of the reforms, the conditions under which they have to implement them, and their capacities to do so. In other words, the independent variable in studies of the effects of governance reforms is extremely complex and dynamic. As far as the dependent variable is concerned, that is, the effects of governance reforms, things are not much simpler, because the expected effects are various and there is considerable discussion on how they should be operationalized in effect studies; in addition, it is very hard to establish robust causal relationships between forms of governance and effects – a common problem, of course, in many social scientific effect studies that take place in a complex and dynamic social environment. This book does not intend to present a sophisticated evaluation of effects of new forms of governance. Our scope is far more modest and explorative: we will try to map what insights into the effects of new forms of governance currently exist in the countries discussed in this volume. We will see that while there are indications for a large variety of effects, our knowledge of what mechanisms cause these effects is still very limited. In this book, not all possible effects will be studied; we will not focus on efficiency effects. Our main interest is in three types of potential effects of governance reforms. First, we are interested in the employment effects of the reforms (outcome effects), which includes (gross and net) impact on job placement, job retention and job quality. Secondly, we will look at what could be called output effects, including coverage and reach of programmes and services (which target groups are served and what proportions of target groups acquire access to services), service

Table 1.2

Potential effects of governance reforms

Type of effect

Indicator

Outcome effects

Job placement Job retention Job quality

Output effects

Coverage and reach Service content and quality Range and variety

Process effects

Voice and choice Individualized service provision Rights, obligations, responsibilities

Tomáš Sirovátka

18

The Governance of Active Welfare States

content and quality, and range and variety of programmes. Thirdly, we will distinguish process effects, which focus on the treatment and approach of unemployed people as well as their social status: voice and choice of people involved in activation, the individualized nature of the service provision process, as well as the rights, obligations and responsibilities of the unemployed. Table 1.2 summarizes these potential effects.

Structure of the book Chapter 2 will present an overview of the main actors and agencies involved in the administration of income protection schemes and the provision of activation services for unemployed people in the countries discussed in this book. This chapter provides comparative background information on how social security and employment services are currently governed in the various countries. Chapters 3 to 11 are the core of the book. In these chapters, governance reforms in the areas of social security and activation will be discussed and analysed. Subsequently, the chapters will focus on the following countries: France, Germany, the UK, Italy, the Netherlands, Switzerland, Czech Republic, Sweden and Finland. Rather than providing an overview of all relevant reforms in the various countries, the chapters will focus on some main reform processes that have been taking place in the policy areas on which this book focuses. Thus, the chapters will provide an in-depth analysis of governance reforms and reform trends, rather than pretending completeness. In addition, each chapter will try to make an assessment of the effects of the reforms, in as far as insight into these effects is available in these countries. Chapter 12 will conclude the book with a comparative analysis of the reforms analysed in the previous chapters. Here, we will return to the framework that was outlined in this introduction to see what similar as well as dissimilar reform trends can be found in the various countries, what this tells us about the ways in which governance regimes have developed and are developing, and what – from a comparative perspective – can be said about the effects that these reforms have had.

Note 1. According to Eurostat data, the EU countries spent 1.7 per cent of GDP labourmarket policies in 2007 (public expenditure). Individual EU countries’ public expenditure ranged from 0.154 per cent of GDP in Estonia to 3.294 per cent of GDP in Belgium. Public expenditure (2007) in the countries discussed in this volume is summarized in Table 1.3.

Tomáš Sirovátka

Rik van Berkel et al.

19

Table 1.3 Public expenditure on labour-market policies in 2007 (percentage of GDP) Country Czech Republic Finland France Germany Italy Netherlands Sweden Switzerland UK EU 27

Services & measures

Support

Total

0.255 0.821 0.914 0.747 0.406 1.093 1.078 n/a 0.321 0.663

0.204 1.427 1.239 1.625 0.712 1.394 0.665 n/a 0.157 1.020

0.458 2.248 2.154 2.399 1.118 2.487 1.741 n/a 0.478 1.683

References Barbier, J.C. (2004) ‘Systems of social protection in Europe: Two contrasted paths to activation and maybe a third’, in J. Lind, H. Knudsen and H. Jørgensen (eds) Labour and Employment Regulation in Europe (Brussels: Peter Lang). Borghi, V. and R. van Berkel (2007) ‘New modes of governance in Italy and the Netherlands. The case of activation policies’, Public Administration, 85/1, 83–101. Bredgaard, T. and F. Larsen (2005) Employment Policy from Different Angles (DJOeF: Copenhagen). Bruttel, O. (2005) Contracting-Out and Governance Mechanisms in the Public Employment Service (Berlin: Wissenschaftszentrum Berlin). Carmel, E. and T. Papadopoulos (2003) ‘The new governance of social security in Britain’, in J. Millar (ed.) Understanding Social Security: Issues for Social Policy and Practice (Bristol: Policy Press). Carroll, P. and P. Steane (2002) ‘Australia, the New Public Management and the new millennium’, in K. McLaughlin, S. Osborne and E. Ferlie (eds) New Public Management: Current Trends and Future Prospects (London: Routledge). Considine, M. (2001) Enterprising States: The Public Management of Welfare-to-Work (Cambridge: Cambridge University Press). Considine, M. and J. Lewis (2003) ‘Bureaucracy, network, or enterprise? Comparing models of governance in Australia, Britain, the Netherlands, and New Zealand’, Public Administration Review, 63/2, 131–41. Denhardt, R.B. and J.V. Denhardt (2000) ‘The new public service: Serving rather than steering’, Public Administration Review, 60/6, 549–59. Dingeldey, I. (2007) ‘Between workfare and enablement: The different paths to transformation of the welfare state’, European Journal of Political Research, 46/6, 823–51. Geddes, M. (1998) Local Partnership: A Successful Strategy for Social Cohesion? (Luxembourg: Office for official publications of the European Communities). Geddes, M. and J. Benington (eds) (2001) Local Partnerships and Social Exclusion in the European Union: New Forms of Local Social Governance? (London: Routledge).

Tomáš Sirovátka

20

The Governance of Active Welfare States

Giguère, S. (2005) ‘Local employment development, decentralisation, governance and the role of government’, in S. Giguère and Y. Higuchi (eds) Local Governance for Promoting Employment. Comparing the Performance of Japan and Seven Countries (Tokyo: The Japan Institute for Labour Policy and Training). Gilbert, N. (2002) Transformation of the Welfare State: The Silent Surrender of Public Responsibility (Oxford: Oxford University Press). Handler, J. (2004) Social Citizenship and Workfare in the United States and Western Europe: The Paradox of Inclusion (Cambridge: Cambridge University Press). Hill, M. and P. Hupe (2002) Implementing Public Policy (London: Sage). Hvinden, B. and H. Johansson (eds) (2007) Citizenship in Nordic Welfare States: Dynamics of Choice, Duties and Participation in a Changing Europe (London: Routledge). Jessop, B. (2000) ‘Governance failure’, in G. Stoker (ed.) The New Politics of British Urban Governance (Basingstoke: Macmillan). Jessop, B. (2003) ‘Governance and meta-governance: on reflexivity, requisite variety and requisite irony’, in H. Bang (ed.) Governance as Social and Political Communication (Manchester: Manchester University Press). Jewell, C. (2007) Agents of the Welfare State. How Caseworkers Respond to Need in the United States, Germany and Sweden (New York: Palgrave Macmillan). Kaps, P. (2009) ‘Marketization of placement services as a double recommodification of labour’, in F. Larsen and R. van Berkel (eds) The New Governance and Implementation of Labour Market Policies (Copenhagen: DJOeF). Kettl, D. (2002) The Transformation of Governance: Public Administration for TwentyFirst Century America (Baltimore and London: Johns Hopkins University Press). Kooiman, J. and M. Bavinck (2005) ‘The governance perspective’, in J. Kooiman, M. Bavinck, S. Jentoft and R. Pullin (eds) Fish for Life. Interactive Governance for Fisheries (Amsterdam: Amsterdam University Press). Larsen, F. and R. van Berkel (eds) (2009) The New Governance and Implementation of Labour Market Policies (Copenhagen: DJOeF). LeGrand, J. (1991) ‘Quasi-markets and social policy’, The Economic Journal, 101/408, 1256–67. Lindsay, C. and R. McQuaid (2008) ‘Inter-agency co-operation in activation: Comparing experiences in three vanguard “active” welfare states’, Social Policy and Society, 7/3, 353–65. Lødemel, I. and H. Trickey (eds) (2001) An Offer you Can’t Refuse: Workfare in International Perspective (Bristol: The Policy Press). Martin, S. (2001) ‘Implementing “Best Value”: Local public services in transition’, Public Administration, 78/1, 209–27. Mosley, H. (2009) ‘Decentralisation and local flexibility in employment services’, in F. Larsen and R. van Berkel (eds) The New Governance and Implementation of Labour Market Policies (Copenhagen: DJOeF). Newman, J. (2001) Modernising Governance: New Labour, Policy and Society (London: Sage). OECD (2003) Managing Decentralisation: A New Role for Labour Market Policy (Paris: OECD). Osborne, D. and T. Gaebler (1992) Reinventing Government: How the Entrepreneurial Spirit is Transforming the Public Sector (London and New York: Penguin). Peck, J. (2001) Workfare States (New York: The Guildford Press).

Tomáš Sirovátka

Rik van Berkel et al.

21

Pollitt, C. (2002) ‘The new public management in international perspective. An analysis of impacts and effects’, in K. McLaughlin, S. Osborne and E. Ferlie (eds) New Public Management: Current Trends and Future Prospects (London: Routledge). Pollitt, C. (2008) Time, Policy, Management: Governing with the Past (Oxford: Oxford University Press). Pollitt, C. and G. Bouckaert (2000) Public Management Reform: A Comparative Analysis (Oxford: Oxford University Press). Rhodes, R. (2000) ‘Governance and public administration’, in J. Pierre (ed.) Debating Governance: Authority, Steering and Democracy (Oxford: Oxford University Press). Rhodes, R. (2002) ‘The new governance: governing without government’, in S. Osborne (ed.) Public Management: Critical Perspectives (London: Routledge). Serrano Pascual, A. (ed.) (2004) Are Activation Policies Converging in Europe? The European Employment Strategy for Young People (Brussels: ETUI). Serrano Pascual, A. and L. Magnusson (eds) (2007) Reshaping Welfare States and Activation Regimes in Europe (Brussels: Peter Lang). Sol, E. and M. Westerveld (eds) (2005) Contractualism in Employment Services: A New Form of Welfare State Governance (The Hague: Kluwer). Struyven, L. and G. Steurs (2003) ‘Towards a quasi-market in reintegration services: First assessment of the Dutch experience’, Australian Journal of Labour Economics, 6/2, 331–55. Struyven, L. and G. Steurs (2005) ‘Design and redesign of a quasi-market for the reintegration of jobseekers: Empirical evidence from Australia and the Netherlands’, Journal of European Social Policy, 15/3, 211–31. Struyven, L., G. Steurs, A. Peeters and V. Minne (2002) Van aanbieden naar aanbesteden. Marktwerking bij arbeidsbemiddeling en -reïntegratie in Australië, Nederland, Verenigd Koninkrijk en Zweden (Leuven: Acco). Sultana, R. and A. Watts (2004) ‘Career guidance in Public Employment Services across Europe’, International Journal for Educational and Vocational Guidance, 6, 29–46. Van Berkel, R. (2010) ‘The provision of income protection and activation services for the unemployed in “active” welfare states. An international comparison’, Journal of Social Policy, 39/1, 17–34. Van Berkel, R. and V. Borghi (2008) ‘Review article: The governance of activation’, Social Policy and Society, 7/3, 393–402. Van Berkel, R. and I. Hornemann Møller (2002) Active Social Policies in the EU: Inclusion through Participation? (Bristol: Policy Press). Van Berkel, R. and B. Valkenburg (2007) Making it Personal. Individualising Activation Services in the EU (Bristol: Policy Press). Van Berkel, R. and P. van der Aa (2005) ‘The marketisation of activation services: A modern panacea? Some lessons from the Dutch experience’, Journal of European Social Policy, 15/4, 329–45. Van der Veen, R. and W. Trommel (1999) ‘Managed liberalization of the Dutch welfare state: A review and analysis of the reforms of the Dutch social security system, 1985–1998’, Governance: An International Journal of Policy and Administration, 12/3, 289–310.

Tomáš Sirovátka

2 The Administration of Income Provision and Activation Services Rik van Berkel

Introduction As a prelude to the following chapters, this chapter will present an overview of the main actors and agencies that are involved in the administration of income protection systems for the unemployed on the one hand, and in the provision of employment and activation services on the other. As far as income protection schemes are concerned, our focus will be on the agencies responsible for the administration of unemployment benefits and social assistance. Although it is not the objective of this chapter to go into the characteristics of national social policies in any detail, some of these need to be elucidated as social policy characteristics evidently constitute one of the factors determining how administrative and service provision structures appear. This is most clearly the case where income protection systems for the unemployed are concerned: a multi-tiered income protection system often implies an articulated structure for the administration of various income protection schemes. So in those cases where an elaboration of social policy characteristics is necessary to present an overview of relevant agencies and organizations, these characteristics will be highlighted. The national administrative structures for providing income protection and activation services for the unemployed and the actors and agencies involved in service provision processes are all but stable – if they were, this book would probably not have been written. As was argued in Chapter 1, this is partly related to the fact that the social policy arrangements that are supposed to be administered and implemented through these structures are being reformed; and partly the result of new ideas about governance and public administration. The unavoidable consequence of these institutional dynamics in the policy fields of income protection and activation is that the state of the art 22

Tomáš Sirovátka

Rik van Berkel

23

presented in this chapter shows a snapshot at a certain point in time (2010), and may no longer give an adequate description of the situation at the point of time when the reader is reading this book. However, the more detailed analyses and accounts of governance reform processes presented elsewhere in this book will compensate for this potential lack of accuracy, in that they will help in understanding the directions of, and the challenges underlying, future governance reform processes. This chapter is divided into three sections. In the first section, agencies involved in the administration of income provision systems for the unemployed are discussed. The second section focuses on the actors involved in the provision of activation services. In the final section, some conclusions are drawn.

The administration of income support systems Most countries in our study combine unemployment benefit systems with national social assistance systems in providing income support for unemployed people: the UK, Finland, Sweden, Germany, the Netherlands, France and the Czech Republic. The two remaining countries (Switzerland and Italy) do not have national social assistance systems. In the UK, social assistance for the unemployed (Income Support) was integrated with the unemployment benefit system in 1996 into the Jobseeker’s Allowance (JSA). This Allowance consists of two parts: a flat-rate, contribution-based JSA, lasting a maximum of six months; and a means-tested, income-based JSA that is not limited in duration. The latter is more or less similar to continental social assistance systems, although not in the way in which it is administered, as we will see below. In Italy and Switzerland, regional social assistance systems do exist. Italy has experimented with a national RMI system, but this never resulted in national social assistance legislation. Italian regions have introduced social assistance schemes, but these are very fragmented, partly because they depend on regionally available resources for meeting regional needs – both of which are very unequally distributed across the country. In Switzerland, the cantons have their own social assistance legislation, and differences between cantons are significant, both in terms of level of services and in organizational issues, although attempts to harmonize cantonal systems are taking place. Although all countries have unemployment benefit systems, differences between them are significant. It has been mentioned already that the UK has no system of wage-related unemployment benefits. Both Sweden and Finland have voluntary unemployment insurance

Tomáš Sirovátka

24

The Administration of Income Provision and Activation Services

systems that provide their unemployed members with an earningsrelated benefits, and a system of basic unemployment allowance for those who did not join unemployment insurance schemes. Finland also has a system of labour market support, which offers a means-tested benefit for those who have exhausted their unemployment benefit entitlements. Another country where a similar dual unemployment compensation system exists is France, which has a social insurance system and a so-called solidarity scheme for those who have exhausted their insurance entitlements. In some countries, similar income provision systems in between unemployment benefits and social assistance previously existed but have been abolished (for example, Germany and the Netherlands). Germany and the Netherlands have an income-related system of unemployment benefits. German and Dutch unemployment insurance contributions are obligatory, which is also the case for the Swiss and Czech unemployment benefit systems. In Italy, a variety of unemployment benefit arrangements exists. There are ordinary unemployment benefits, an obligatory social insurance scheme. Apart from that, short-term unemployment benefit systems exist: an ordinary version that provides earnings supplements to temporarily unemployed workers; and a special version for people who are unemployed due to sector- or region-specific restructuring processes. Whether or not these benefits are provided depends on negotiations between the Labour Ministry and social partners. Finally, Italy has a mobility benefit that aims at facilitating labour-market transitions of dismissed workers. In several countries, reforms of the administrative structures for income protection systems have been taking place: these reforms will be analysed in more detail in the following chapters. These governance reforms may be directly related to changes in the income protection systems themselves, as was the case in Germany after the Hartz reforms. But governance reforms may also take place without simultaneous changes in income protection schemes, as was illustrated by reforms of the administration of unemployment compensation in France (unemployment insurance and the solidarity scheme) and the Netherlands (unemployment benefits). Here, we will discuss the administrative structure of income protection schemes for the unemployed as they exist in 2010. The UK is the only country in our study having a national social assistance scheme (the income-based JSA) that is not administered by municipal agencies but by a national agency, Jobcentre Plus. In the other countries with national social assistance systems, municipalities play an important role in administration and implementation;1 in France,

Tomáš Sirovátka

Rik van Berkel

25

however, the departments rather than the municipalities are the primary actors in the social assistance system (RSA, the successor to the RMI). In financing social assistance, there are considerable differences between the countries: funding can be national, regional, local or a combination of these. In Sweden, social assistance is funded by local taxes. Compared with other countries with national social assistance systems, Sweden’s system is highly decentralized. The Social Service Act of the early 1980s stipulated that social assistance should ensure a reasonable standard of living, but did not provide specified social assistance payment standards. Since then, attempts to harmonize social services have been made (through legislation as well as through recommendations issued by the National Board for Health and Welfare), resulting, among others, in uniform benefit standards for social assistance in 1998. Nevertheless, considerable inter-municipal variation remains. In Finland, strengthening local responsibilities for dealing with long-term unemployment has been an important objective of reforms. A reform introduced in 2006 shared the responsibility for funding both social assistance and labourmarket support for the long-term unemployed between municipalities and employment authorities, that is, Public Employment Services (PES). In the Netherlands, social assistance administration is a municipal responsibility, though funding comes from national government. Recent reforms increased municipal financial responsibilities: nowadays, the national funding of social assistance payments is budgeted, which implies that in cases of budget deficits, local resources will have to be used to fund social assistance payments. In the Czech Republic social assistance is administered by the municipalities as well; funding comes from national government. In the French social assistance system, local agents are involved in administration, but financing consists of a mix of national, departmental and local resources. Germany has witnessed the most far-reaching reform process, as the Hartz reforms combined social policy and governance reforms. The reforms integrated social assistance for those considered capable of working (that used to be administered by the municipalities) and unemployment assistance (that used to be administered by the Federal Employment Agency) into unemployment benefit II. In administrative terms, a complex situation emerged. In the mainstream administrative system of unemployment benefit II, new local agencies were established (the Arbeitsgemeinschaft or ARGE), which are jointly managed by the municipalities and the Federal Employment Agency. At the same time, an experimental administrative model was introduced in 69 municipalities, which have full responsibility for the administration of income support for the long-term unemployed (that

Tomáš Sirovátka

26

The Administration of Income Provision and Activation Services

is, unemployment benefit II). In 23 municipalities, the tasks of the Federal Employment Agency and the municipalities remain to be carried out separately. As mentioned above, Switzerland and Italy have regional social assistance systems. In Switzerland, some non-binding coordination/ cooperation at national level takes place through the Swiss Welfare Conference (SKOS) that issues recommendations to federal authorities, cantons, municipalities and private welfare organizations. Public social assistance is financed by cantons or municipalities, but private actors play a role as well, and provide income support as well as services. Larger municipalities have social assistance offices that administer social assistance; in other parts of the country, a process of regionalization of the administration of social assistance is taking place. Italy’s social assistance is most fragmented. Social assistance provision is completely left to the regions, without any national legal framework or guidelines for coordination or resource distribution. The administration of regional social assistance programmes is highly dependent on third-sector agencies and actors, voluntary organizations and private players, which adds to the already considerable diversity. As far as unemployment benefits are concerned, in Sweden and Finland unemployment insurance is administered by the unemployment funds, and not by the state Social Insurance Agency (which administers other social insurance schemes). The Swedish funds are independent formally, but in reality are run by trade unions and supervised by the Unemployment Insurance Board (IAF). Decisions concerning benefits for the insured unemployed are the responsibility of the funds. In Finland, the basic unemployment allowance for the unemployed who are not members of an unemployment fund, as well as labour-market support, are administered by the National Social Insurance Institution (KELA). Both the state and the municipalities are responsible for funding labour-market support of people who are dependent on this support for more than 500 days, which implicates the municipalities in dealing with long-term unemployment. In Germany, unemployment benefit I is administered by the Federal Employment Agency, which witnessed a serious reduction in the role and influence of social partners. At the local level, the Agency’s Service Centres provide services to the benefit recipients. These Centres are located in Job Centres that also house the ARGE agencies (see above). The two French unemployment compensation systems (social insurance and the solidarity scheme) are financed in different ways (through contributions and the state budget, respectively). Governance of both systems is different

Tomáš Sirovátka

Rik van Berkel

27

as well: social insurance is governed by social partners through the National Insurance fund Unédic, and the solidarity scheme by the state. Nevertheless, the administration of both benefit systems at the local level is the responsibility of the same agency. Since 2009, this takes place in the so-called local Employment Counters (Pôles emplois), a merger of the local offices of the national insurance fund (Assedics, which before 2009 were responsible for the local administration of both social insurance and the solidarity scheme) and the French PES (ANPE). Compared with the other countries, Switzerland has a more diversified system of unemployment funds, in the sense that some are private initiatives of employee and employer associations, whereas others are cantonal. Unemployment insurance is obligatory, but people are free to decide of which fund they wish to be a member. At the national level, the equalization fund collects all resources for covering benefit expenses and administering unemployment benefits. The Equalization Board is the central body for implementing unemployment insurance, and is supervised by a tripartite body. The tasks of the funds are to test eligibility and administer payments. In the UK and the Netherlands, Jobseeker’s Allowance and unemployment benefits, respectively are administered by a public agency: Jobcentre Plus in the UK and the UWV (Agency for the Administration of Employees’ Insurance) in the Netherlands. In both countries, social partners are not involved in the administration of unemployment benefits. In the Czech Republic, unemployment benefits are administered by the PES. In contrast to the situation in the UK and the Netherlands, where benefit agencies have the status of an executive agency of the Department of Work and Pensions (UK) and an autonomous administrative agency (ZBO, Netherlands), respectively, the administrative body in the Czech Republic is part of the Ministry of Labour and Social Affairs. The PES administers benefits through local employment offices. In Italy the main actors in unemployment benefits are the Ministry of Labour and the National Institute for Social Security (INPS), which is responsible for the collection of contributions and the payment of benefits. Table 2.1 summarizes the above by schematically presenting the main actors and agencies involved in the administration of income protection schemes for the unemployed in the countries involved in our study.

The organization and provision of activation In the countries included in our study, activation has very different histories. For example, Sweden introduced active labour-market policies

Tomáš Sirovátka

28

The Administration of Income Provision and Activation Services

Table 2.1 Main actors and agencies involved in administrating Unemployment Benefit (UB) and Social Assistance (SA) schemes (2010) UB

SA

Switzerland

Unemployment funds, Equalization Board

Cantons, regions, municipalities; private actors

Czech Republic

Public Employment Services

Municipalities

Germany

Federal Employment Agency

(UB II) ARGE (Federal Employment Agency + municipalities) or municipalities (in 69 municipalities)

France

Unédic, Employment counters

Departments and municipalities

Finland

Ministry of Employment and the Economy, Unemployment funds, KELA

Municipalities

Italy

Ministry of Labour, National Institute for Social Security

Regions, municipalities, private actors

Netherlands

UWV

Municipalities

Sweden

Unemployment Funds

Municipalities

UK

(Contribution-based JSA) Jobcentre Plus

(Income-based JSA) Jobcentre Plus

in the 1940s, whereas in the Czech Republic activation – as well as the institutions involved in the provision of activation services – is a typical post-communist phenomenon. Nevertheless, with the attempts to activate welfare state arrangements during the 1990s and 2000s, all countries were confronted with issues regarding the organization and provision of activation services, as well as regarding the coordination of income protection administration and activation. When comparing the organization and provision of activation in the various countries, three issues emerge: – the degree of integration of administering income protection and providing activation; – the degree of integration of the provision of activation services targeted at unemployment benefit and social assistance recipients; – the involvement of other actors (other than PES, benefit agencies and municipal social assistance agencies) as service providers in the area of activation.

Tomáš Sirovátka

Rik van Berkel

29

In the UK, Jobcentre Plus is responsible for the activation of the unemployed. As Jobcentre Plus also functions as the agency administering Jobseeker’s Allowance, the level of integration of income provision and activation – and of servicing short-term and long-term unemployed people – is larger than in any of the other countries involved in this study: the UK has one agency responsible for providing income protection and activation for all unemployed able to work. The role of municipalities in the UK is modest in a comparative perspective, which is mainly a consequence of the absence in the UK of a locally administered safety net provision for unemployed people (although, as the Czech case shows, the presence of a locally administered safety net provision does not guarantee an important role of municipalities in activating social assistance recipients – see below). The responsibility of Jobcentre Plus for activation does not imply that Jobcentre Plus is the sole service provider; other actors, such as private providers and municipal actors, are involved in providing activation services as well. In recent years in the UK, the role of external providers operating in a market of providers of activation services has been increasing. In the Czech Republic integration of activation services aimed at unemployment benefit and social assistance recipients is high, in the sense that Czech PES provide activation for both groups. In addition, as we saw in the preceding section, the PES also administers unemployment benefits so that no separate agencies exist for the administration of unemployment benefits and the activation of benefit recipients. Czech municipalities have no primary role in activating social assistance recipients, which – as we will see below – is quite exceptional when we look at countries with national social assistance schemes. Characteristic for the Czech case is that the formation of democratic labour-market and employment institutions is a relatively recent phenomenon – it started in the early 1990s. The main tasks of the Czech PES in the area of activation are the design of National Action Plans for Employment, the implementation of activation policies and the monitoring and evaluation of these policies. The actual service provision takes place in local employment offices, which depend on the Ministry for funds and staff allocation. At the same time, the role of European Social Fund (ESF) funding of activation has increased during the first decade of the twentyfirst century, which has reduced the impact and control of the national Czech PES on regional and local activation. Though the involvement of private providers in service provision is not a new phenomenon, their role has become more important, and they play an especially prominent role in the implementation of activation programmes funded by the ESF.

Tomáš Sirovátka

30

The Administration of Income Provision and Activation Services

Compared with the UK and the Czech Republic, the other countries have more clearly separated systems for the activation of benefit and social assistance recipients – or for the easy and hard to reintegrate unemployed. At the same time, the backgrounds of these dual systems of activation are different. For example, in Germany, Sweden and the Netherlands, the role of municipalities in activating social assistance recipients increased because municipalities – who are responsible for the administration of social assistance, and sometimes also (partly) for funding social assistance expenses – were dissatisfied with the activation services provided by the PES for social assistance recipients. In Italy and Switzerland, the absence of national social assistance systems seems to play a role in understanding why both groups of unemployed are served by different activation systems, since benefit recipients are the responsibility of national institutions (although in Italy, this is only partly the case), whereas social assistance recipients are served by regional or local agencies that do not operate in the context of a national legal social assistance framework. In Sweden, the PES plays a crucial role in the provision of activation. The PES comes under the responsibility of the Ministry of Employment. Yearly, the government formulates an assignment that defines policy objectives, target groups, policy measures and financial conditions. By the 1990s external service providers were involved in activation; however, this mainly concerned preparatory education activities. In recent years, the Swedish national government has started to stimulate the involvement of external actors in activation, particularly in matching and coaching activities. Local discretion in activation is low in the sense that the national assignment provides detailed aims and finances. At the same time, it is high because individual frontline workers decide in individual cases what offer is most suitable, and because the range of available offers is quite large. Nevertheless, in 2006 the Swedish government decided to reduce the number of activation programmes, and to focus on shorter programmes. In the area of social assistance, municipalities have assumed increasing responsibility for activating social assistance recipients. Sweden is one of the countries where a dual system of activation exists, although this dualism has decreased recently (see below). We have already seen that inter-municipal diversity is significant where social support is concerned, and the same applies to activation. For example, local variation exists regarding the extent to which people are defined as unemployed, the extent to which the unemployed are referred to activation programmes, and the extent to which the unemployed are sent to national

Tomáš Sirovátka

Rik van Berkel

31

(PES-run) or local programmes. Municipalities can involve external providers, but most municipal activation programmes are carried out by the municipalities themselves. Interestingly, recent years have seen a shift of responsibilities for the activation of the young unemployed (traditionally, one of the core target groups of municipal activation) from the municipalities to the (nationally governed) PES. Because of this, the Swedish PES is not only responsible for the activation of unemployment benefit recipients, but also for activating an increasing proportion of social assistance recipients. In Germany, dividing lines in the organization of activation run between the unemployed recipients of Unemployment Benefits I and II (UBI and II). At the same time, for each of these groups income provision and activation are integrated, that is, provided by the same agency. As far as UB-I is concerned the German situation resembles the Czech, in as far as the German PES (the Federal Employment Agency) is responsible for both benefit administration and the activation of benefit recipients. As mentioned in the preceding section, local Service Centres provide income protection and activation to UBI recipients. UB-II recipients receive income protection and activation from local agencies jointly managed by the Federal Employment Agency and the municipalities (ARGE agencies; see preceding section), at least in a majority of municipalities. Thus, the activation of the long-term unemployed is the shared responsibility of a national agency and the municipalities. Traditionally, external service providers play a role in the provision of employment and activation services in Germany. The Hartz reforms, however, provided an opportunity to strengthen the element of competition in service provision, involving the Federal Employment Agency itself in the competition process. External providers can be involved in service provision in two ways: either by a process of outsourcing, where the Federal Employment Agency contracts external parties for service provision; or by so-called placement vouchers, which unemployed people may use to buy services from public or private providers. In the Netherlands, the provision of income support and activation are integrated, though separately for the two groups of benefit recipients on the one hand, and social assistance recipients on the other. Whereas the benefit agency (UWV) is responsible for the insured, the municipal local welfare agencies are responsible for social assistance recipients. At the local (or, in the case of smaller municipalities, supra-local) level, co-location and, increasingly, coordination and cooperation of services provided by the benefit agency and municipalities

Tomáš Sirovátka

32

The Administration of Income Provision and Activation Services

are promoted. Originally, this took place in the so-called Centres for Work and Income; since the merger between the Centres and the benefit agency in 2009, these agencies for co-location, coordination and cooperation are called Work Plazas (Werkpleinen). Nevertheless, these attempts at service integration do not mean that the primary responsibilities for activation are not as described above. External, often private, providers play a major role in the provision of activation in the Netherlands. Outsourcing activation is obligatory for the benefit agency, and optional for the municipalities (since 2005: prior to that it was obligatory for the municipalities as well). Because of this, providers of activation services for unemployment benefit recipients are mainly private providers. The number of private providers is considerable, partly because unemployment benefit recipients are allowed – under certain conditions – to select their own service provider. The situation is more mixed where the activation of social assistance recipients is concerned: some municipalities outsource most services to external providers, some municipalities rarely contract external providers, and many municipalities use a mix of outsourced and in-house activation. In Finland the PES has a crucial role in providing employment services for the insured unemployed. The PES is also an important actor in providing activation for labour-market support and social assistance recipients, but here cooperation with municipal welfare agencies takes place. Finland is another example of a country where different activation regimes exist for the short-term and long-term unemployed. As a matter of fact, in Finland the concept of activation is exclusively used in the context of promoting labour-market participation of the long-term unemployed. For the short-term unemployed, the PES introduced regional Job-seeking Centres that should provide labour-market information and coach jobseekers in finding jobs. The activation of long-term unemployed people – who may receive social assistance, labour-market support or a combination of these two forms of income support – has received specific attention in Finland. To this end, a network approach has been promoted, which materialized in the so-called Labour Force Service Centres (LAFOS). Municipalities were free to establish a local LAFOS unit, which is based on a form of network governance in which national actors (especially PES) and local actors work together on the basis of a contract between national employment authorities and municipalities. Co-funding is one of the pillars of this form of network governance: national and local authorities provide the local LAFOS units with the same numbers of workers and share costs for service provision on an equal basis. LAFOS combines a variety of disciplines considered relevant for servicing the long-term unemployed: employment services,

Tomáš Sirovátka

Rik van Berkel

33

social welfare services, health care, rehabilitation and social insurance. External providers are involved in service provision in Finland, but they do not play a similar primary role as in some of the other countries involved in our study. In terms of the two forms of integration mentioned at the start of this section (integration of income protection and activation on the one hand, and the activation of unemployment benefits and social assistance recipients on the other), Switzerland reveals low levels of integration. The absence of national social assistance legislation draws a clear dividing line between responsibilities for the insured (national/federal) and social assistance recipients (cantonal level). In addition, the administration of unemployment benefits (the responsibility of the unemployment funds) is organized separately from the activation of unemployment benefit recipients. A series of reforms introduced in 1995 put the activation of unemployment benefit recipients high on the Swiss political agenda and introduced new structures and institutions for implementing activation, separate from the existing unemployment funds and their administrative agencies. The reforms introduced Regional Employment Centres (REC), at least one in each canton. These RECs replaced the local employment centres that existed previously. With the establishment of the RECs, a new way of financing was also introduced: whereas the local employment centres were financed by the municipalities, the RECs are financed by the unemployment insurance funds. The tasks of the regional centres include advising jobseekers, imposing sanctions, allocating available jobs, checking employability, acquiring new jobs and allocating labour-market measures. At the national level, the State Secretariat for Economic Affairs is responsible for the functioning of the Employment Centres, and has a steering role. The same reforms also created a quasi-market for activation measures. So-called Logistics Offices (one in each canton) act as purchasers on this market and, together with the RECs, these Offices determine the need for activation measures, plan offers, award contracts to providers and evaluate the services of providers. Providers may be either public or private agents. As Swiss social assistance is a cantonal responsibility, differences between the cantons are significant, including differences regarding the activation of social assistance recipients, even though the increasing emphasis on activating social assistance recipients is a national trend. Activation measures targeted at social assistance recipients include career orientation measures, assistance in integration within the regular labour market, placement and employment programmes, supplementary employment placements and therapeutic or pedagogical offers.

Tomáš Sirovátka

34

The Administration of Income Provision and Activation Services

Private social assistance plays a role in Switzerland as well, which may not only involve individual financial assistance but also measures aimed at work or social integration. Nevertheless, marketization of the provision of activation services did not boom to a similar degree as in the case of the insured unemployed. Cooperation in the areas of activating the insured and social assistance recipients is a relatively recent phenomenon: it has received increasing attention in recent years. The administrative integration of income support and activation has been strengthened in France, especially where the activation of recipients of unemployment insurance benefits and the solidarity scheme are concerned. An important step in this process was the above-mentioned merger between the French PES (ANPE) and the local offices of the insurance agency into local Employment Counters. These counters now play an important role in the activation of unemployment insurance and solidarity scheme recipients. Two other institutions play a core role as well: first, the Association for Adult Vocational Training (AFPA), and secondly the Regional Direction of Firms, Competition, Consumption, Work and Employment (DIRECCTE), which plays a role in defining and implementing labour-market policies in the French regions. The involvement of external service providers through network (co-contractor) or market (sub-contractor) arrangements is common in France. External providers include NGOs as well as for-profit organizations such as temping agencies and private placement agencies. There is a trend to strengthen the role of market mechanisms in involving external service providers, which challenges the position of traditional not-for-profit actors in the provision of employment and activation services. As far as the activation of recipients of French social assistance (the Active solidarity income, RSA) is concerned, other actors are responsible than in the case of recipients of the insurance and solidarity schemes. Departmental General Councils have a primary task in organizing activation services for social assistance recipients, though municipalities have competencies in this area as well and play an increasing role in activating this group of the unemployed. The General Councils may enter into partnerships with the Employment Counters in organizing the provision of activation services, but they may also decide to cooperate with not-for-profit associations in doing so. The integration of income protection and activation on the one hand, and of providing activation services for the insured and social assistance recipients on the other, seems to be most weakly developed in Italy. Whereas the National Institute for Social Security is responsible for the administration of benefits, a decentralization process in the

Tomáš Sirovátka

Rik van Berkel

35

1990s delegated important functions in the area of employment and activation services to the regions and provinces. National activation programmes still exist, but regional, provincial and local actors now have considerable competencies in the area of activation (for example, professional training, counselling and job mediation), resulting in a multi-level model for the governance of activation and employment services. Besides public agencies (the PES), private employment agencies also play a role in providing activation services. Opening the provision of employment services to private actors has partly been a consequence of the flexibilization of the Italian labour market, which was accompanied by the establishment of (private) temporary work agencies for matching supply and demand where temporary work and work on fixedterm contracts is concerned. Whether activation is provided in-house or outsourced is decided by public agencies. In providing activation services to social assistance recipients, a variety of actors is involved: municipalities, the PES and a range of private and public actors. Given the fragmentation of Italian social assistance, and the large differences across the country regarding both level of needs and the availability of resources to meet these needs, it is impossible to provide a general picture concerning the activation services available for Italian social assistance recipients. In Table 2.2 we present a summarizing overview of the main actors and agencies involved in the organization and provision of activation services in the nine countries. Table 2.2

Main actors and agencies in the provision of activation services (2010) UB

SA

Switzerland

Regional Employment Centres, Logistics Offices (purchasing), private providers

Cantons/regions/municipalities, private providers

Czech Republic

PES, public and private providers

PES, public and private providers

Germany

Federal Employment Agency (Service Centers), private providers

ARGE, private providers

France

Employment Counters, Unédic, AFPA, DIRECCTE, private providers

Departments and municipalities, private providers

Finland

PES, LAFOS (long-term unemployed), private providers

Municipalities, LAFOS (long-term unemployed), private providers

Tomáš Sirovátka

36

The Administration of Income Provision and Activation Services

Table 2.2

(Continued) UB

SA

Italy

Regions, provinces, private providers

Regions/municipalities, public (including PES) and private providers

Netherlands

UWV, private providers

Municipalities, public and private providers

Sweden

PES, private providers

Municipalities, private providers, PES

UK

JobCentre Plus, private and public providers

JobCentre Plus, private and public providers

UB, unemployment benefit; SA, social assistance.

Conclusion This chapter provided an overview of the main actors and agencies involved in providing income protection and activation services to unemployed people in nine European countries. As will be elaborated in the following chapters, these administrative and implementation structures are the result of extensive reform processes in most countries, some of which are very recent. Although the differences between the countries are significant, the overview presented in this chapter reveals that all countries are coping with some similar issues related to the institutional consequences of the attempts to make welfare states more activating. First, the increasing emphasis on activation raises issues concerning the tasks, roles and positions of the agencies traditionally responsible for employment services – with the exception of the Czech Republic, all countries established employment services agencies long before the activation paradigm became dominant in social policies. In all countries, reconsidering the position of public employment services agencies not only affected the competencies and capacities of these agencies, but also strengthened the role of external providers (public and private, for-profit and not-for-profit) in the provision of activation services. Secondly, activating welfare state arrangements has made coordination of the activities of public employment agencies and unemployment benefit agencies an important topic. Whereas Germany, with its integrated agency for the administration of unemployment benefits and the provision of employment services, used to be exceptional, we now see integrated agencies in the UK, France, the Netherlands and the Czech Republic.

Tomáš Sirovátka

Rik van Berkel

37

Thirdly, most countries with national or regional social assistance schemes – the exceptions are the UK and the Czech Republic – still have separate institutional structures for the activation of social assistance recipients, although in Sweden we see a decreasing role of municipalities and an increasing role of the PES. The traditional important role of municipalities (and, in some cases, regional authorities) in social assistance schemes plays a role in explaining this situation, as does the failure of public employment agencies in several countries to provide adequate services for the hard to employ, which stimulated and subsequently strengthened the role of municipal activation. Nevertheless, all countries show efforts to improve the cooperation of municipal (or regional) and national actors in promoting the activation of social assistance recipients, even though these efforts may vary in intensity and may take different forms. As will be elaborated later, LAFOS agencies in Finland are an example of intensive cooperation between local and national institutions. Finally, we have already mentioned the increasing involvement of external providers of activation services, through either market arrangements or networks and partnerships. There are some indications – and we will return to these in later chapters – that the introduction of quasimarkets in the provision of activation services has been more important in the activation of unemployment benefit recipients than in providing services for social assistance recipients. A possible explanation for this might be that the ‘activation needs’ of social assistance recipients are more complex, and that meeting these needs requires the integration of a larger variety of services that are not readily available on the market but ask for the cooperation of various specialized service providers. In this book, our main focus is on income support and activation services. However, it is important to emphasize that promoting the (labour-market) participation of more vulnerable groups of unemployed people – many of whom are recipients of income safety net provisions such as social assistance – will often ask for services beyond income support and activation, and that this will have an impact on the implementation and service provision structures that are developed to govern the service provision process.

Note 1. British municipalities do have a role in the administration of Housing Benefit and Council Tax Benefit.

Tomáš Sirovátka

3 Accelerating Governance Reforms: the French Case Anne Eydoux and Mathieu Béraud

Introduction The French welfare state regime is deemed to be corporatist-conservative in usual comparative typologies (Esping-Andersen, 1999), while its activation regime is considered as a hybrid one, a mix between the neoliberal regime, relying on market efficiency as well as on individual rationality, and the social-democratic one, imposing the burden of supporting employees’ return to work on the state (Barbier, 2002). Recent changes in activation policies and Public Employment Services (PES) have not fundamentally altered the characteristics of France’s income protection and activation regime (Béraud and Eydoux, 2009). Nevertheless, it has been through an accelerated process of reform that has led some experts to speak of a ‘brutal acceleration of history’ (Dupeyroux, 2005). Many reforms imply substantial changes in activation policies, such as the extension of activation to income support recipients and the development of a making-work-pay (neoliberal) strategy since the late 1990s, or (albeit in a more social-democratic fashion) the most recent redeployment of active measures to meet the current economic slowdown (Erhel, 2010). Some reforms also concern the steering of activation policies and the governance of PES, in line with a general trend in French governance reforms aimed at modernizing public administration and services, as well as at improving and controlling their performance (Perret, 2008). Activation policies and services are in the front line and have been through the various dimensions/trends of renewed governance (discussed in Chapter 1 of this volume): decentralizing competences to the Regional or Departmental level; developing New Public Management (NPM) practices within the national employment agency ANPE;1 outsourcing services to a wide range of providers increasingly subject to competition (quasi-markets); promoting inter-agency cooperation and the merging of the ANPE with 38

Tomáš Sirovátka

Anne Eydoux and Mathieu Béraud

39

the local benefit agencies (Assedics)2 of the national insurance fund (Unédic3 ) into a new institution, the Employment Bureau (Pôle emploi). Despite these major changes, few research publications deal with the governance of employment and integration policies or with the organization of PES in France, where these topics are matters of policy if not of politics. While official communications may act as a screen masking actual practices, fieldwork is sometimes delicate to conduct (Barbier, 2007). As a consequence, the main materials dealing with governance reforms are official public reports ordered by successive French governments during the reform process, as well as dispersed research that attempts either to assess quantitatively the outcomes of one reform or another, or, relying on fieldwork, to enlighten some aspect of the changes at stake. The diversity and institutional complexity of French labour market intermediaries, whether public or private, have long been underlined (Baron et al., 1994). However, it is only recently that this complexity has been reasserted in official reports, putting the governance of activation policies and PES institutions on the agenda (Balmary, 2004; Marimbert, 2004; Cahuc and Kramartz, 2005; CERC, 2005; Cour des Comptes, 2006; Boulanger, 2008). These reports have pointed out the insufficient coordination or the lack of evaluation in the steering of activation policies, or in PES, as sources of inefficiencies or unequal treatment. Recent governance reforms can be seen as somehow reflecting these preoccupations. Nevertheless, their impact on the French governance regime has not been studied. This chapter will address the governance of PES and activation policies by relying on the comparative literature of governance regimes, qualifying the French activation regime as a mix of procedural rules, corporate regulations, market competition and partnership relations (in the sense of Considine (2001), see Chapter 1). We will first present the main institutions in charge of activation policies and of administering income protection schemes in their relevant context of changing activation strategies. Then, the main features of governance reforms will be explored in detail, and the shifts at stake in the French governance regime will be qualified. Finally, the main impacts of the reforms will be discussed drawing on available reports and research.

Today’s challenge: monitoring changes in activation strategies The changes that have occurred in the steering of employment policies and in the governance of PES have to be analysed in their relevant

Tomáš Sirovátka

40

Accelerating Governance Reforms: the French Case

context: the intensification and shifts of activation policies in the face of the rise of massive unemployment (during the 1980s), and in line with the European employment strategy launched in 1997. Intensifying and shifting activation policies Substantial changes have occurred in French activation policies that, if not modifying drastically the French activation regime, can be seen as reflecting a new compromise between the liberal conception emphasizing individual responsibility and the social-democratic approach concerned with universalism and state responsibility regarding employment matters. Early changes were more adaptive (in the face of the rapid increase in unemployment rates) rather than taking an explicit direction. In contrast, liberal arguments have been developed and discussed since the late 1990s, as activation policies have extended their scope to deal with income support recipients (and not solely with the registered unemployed). Liberal arguments based on individual economic rationality have been mobilized in both policy and institutional reforms. The first one favours economic incentives and the generalization of personalized support and counselling for both income support recipients (whether jobseekers or not) and the unemployed. It primarily focuses on economic incentives in order to sustain the supply of work: to make work pay has become the new leitmotiv that governs the redesigning of unemployment and assistance (income support) benefits. The second liberal argument rather focuses on employment institutions and frontline workers. The assumption is that PES and frontline workers do not behave optimally and that new procedures (profiling) and organizational change (marketization, appealing to private providers) are liable to improve the economic efficiency of the system and the unemployed return to stable jobs (Ferracci, 2007). Institutional change in the French governance regime (decentralization, NPM and contractualism,4 outsourcing, marketization) have thus developed in parallel to formal procedures and routines (profiling) and to performance assessment procedures that make frontline workers increasingly responsible for the rate of return to employment. French activation strategy has tended to increase not only the duties of the unemployed and income support recipients in looking for jobs, but also frontline workers’ responsibility for employment. The institutional governance of PES Generally speaking, the procedural governance of French PES is both centralized and corporatist, with a mix of state regulations and

Tomáš Sirovátka

Anne Eydoux and Mathieu Béraud

41

negotiations between social partners. While activation policies are essentially steered by the state, several labour-market institutions are administered jointly by social partners who contribute together with the state to set tripartite employment regulations. Describing the governance of French PES in detail is not an easy task because of its complexity and accelerated evolution. Major changes have occurred since the passing of Law No 2005-32 of 18 January 2005 on Social Cohesion, which provides the first legal definition of an extended PES. Before, only a public service for placement existed formally, apart from unemployment compensation institutions (Table 3.1). Before 2005, the public service for placement included three major institutions under the authority of the Ministry of Labour at the national level, with decentralized services or agencies at the regional/departmental/local level: (1) the Directorate General for Labour, Employment and Vocational training (DGTEFP) in charge of the definition and implementation of labour-market policies; (2) the National Employment Agency (ANPE), holding a monopoly for job placement; and (3) the Association for Adult Vocational Training (AFPA),5 dealing with vocational training and guidance. The unemployment compensation system was not included in the public service for placement but relied on a separate institution, the National Insurance Fund (Unédic), administered by the social partners. Since 2005, job placement and unemployment compensation institutions have constituted the hard core of a new, extended PES that encompasses job placement, integration into employment, vocational training and job-search support, and which includes new actors situated within three circles (Table 3.2). The first circle corresponds to hard-core PES institutions and brought together the DGTEFP, the AFPA, the ANPE, and the Unédic (formally recognized as being part of the unemployed integration into employment process) and their decentralized services or local agencies. Two major changes have occurred within this circle since 2005: after the Law No 2008-126 of 13 February 2008, the ANPE and the Assédics (local agencies of the Unédic) were merged into a single institution, the Employment Bureau,6 while the Decree No 2009-1377 of 10 November 2009 replaced the DDTEFP and DRTEFP by a new, single, regional institution, the Regional Directorate of Firms, Competition, Consumption, Work and Employment (DIRECCTE).7 The second circle is composed of organizations that contribute to the PES, that is, local authorities and municipalities. The third circle draws together a range of organizations that may participate in PES – through co-contracting or subcontracting procedures – despite the fact that they sometimes develop, or mainly

Tomáš Sirovátka

42

Table 3.1

The public service for placement and unemployment compensation (prior to 2005) The public service for placement under the authority of the Ministry of Labour

Unemployment compensation Social partners

Institution at the national level

DGTEFP (1997) (State service)

ANPE (1967) (Public institution)

AFPA (1949) (Tripartite organization)

Unédic (1958) (Joint system)

Role

Definition, implementation and control of public labour-market policies

Job placement, reclassification, information and advice for jobseekers and firms, labour-market statistics

Training, vocational guidance and human resources advice

Registration of all unemployed

Regional level (Regional offices)

DRTEFP (1995)

DR-ANPE

DR-AFPA

DR-Assédic

Departmental level (administrative district)

DDTEFP (1992)

DD-ANPE Local AFPA agencies

Assédic agencies

Local level

Tomáš Sirovátka

Local ANPE agencies (ALE)

Administration of the insurance scheme Benefit-paying agency

Table 3.2

The three circles of public employment services (since 2005) First circle

Second circle

Third circle

Hard core institutions of PES

Organizations that contribute to the PES

Organizations that may participate in the PES

Main governance principles

Procedural (central steering by the state/Ministry of Labour) and corporatist (social partners participate in the administration of the Unédic, the AFPA and the Employment bureau).

Procedural and partnership

Partnership and quasi-market

Institutional actors

Until 2008

Since 2009

DGTEFP, DRTEFP, DDTEFP ANPE/ALE

DIRECCTE

Municipalities and local authorities

Co-contracting institutions: Local Missions, APEC, Employment course

Unédic/Assédics

Unédic

AFPA/AFPA centres

AFPA

Employment bureau

Subcontracting associations, temporary work agencies, human resource agencies New private placement agencies 43

Tomáš Sirovátka

44

Accelerating Governance Reforms: the French Case

focus on, other activities. The co-contractors acting as partners of hard-core institutions are Local Missions in charge of the social integration of vulnerable young unemployed, the APEC8 dealing with managers’ job search support and the network Employment Course,9 which is in charge of the integration of the disabled. Subcontractors are NGOs (mainly associations dedicated to the reintegration of vulnerable groups), private temping agencies (for example, Adecco) or new private placement agencies (for example, Altedia), which are subcontractors for institutions in the first circle, such as the ANPE, the Unédic or now the Employment Bureau. As a consequence of the extended definition of French PES, new forms of governance have been made visible alongside the mainstream procedural (state and corporatist) one: partnership governance between hardcore institutions, local authorities and co-contractors, and quasi-market governance between the hard core and the subcontractors (see below). The institutional governance of income protection schemes The procedural governance of the French income protection system, reflecting the differentiated nature of benefits, is largely corporatist and tripartite, involving social partners and/or the state in different combinations according to the income protection scheme: insurance, solidarity or assistance (Table 3.3). The French income protection system is primarily a contribution-based, Bismarckian system linking social protection to employment (of male breadwinners) and to the family (for the spouse and dependent children), rather than attaching it to citizenship. As it became particularly vulnerable to the development of new social risks such as family break-ups or long-term unemployment that led to emerging forms of poverty, new social entitlements, essentially funded by the state, developed in the 1970s and 1980s. The Lone Parent Allowance (API) created in 1976 and the Minimum Integration Income (RMI) created in 1989 were established on a more universal basis for poor (often unemployed) households; both were replaced in June 2009 by the Active Solidarity Income (RSA). A Solidarity Scheme was created in 1984 for the unemployed with long working experience and who had exhausted their entitlements to insurance benefits. The unemployment compensation system is a dual system, divided into two schemes that are governed according to different principles. The insurance scheme relies on corporatist principles: financed through workers’ and employers’ contributions, it is administrated by social partners at the head of the National Unemployment Insurance Fund, Unédic (created in 1958). It provides the insured unemployed with

Tomáš Sirovátka

Table 3.3

Unemployment compensation (insurance and solidarity schemes) and social assistance Unemployment compensation system (passive labour-market policies) Insurance scheme

Solidarity scheme

Benefit-paying agency

Local insurance funds (Assédics) before 2009; Employment Bureau (Pôle emploi) since 2009

Administration

National unemployment insurance fund, Unédic

Solidarity fund created in 1984 The state

Social partners

Assistance and integration policies Income support scheme Local family funds (CAF) for the Département General councils (Departmental level) in charge of the integration of the income support recipients

Decision making

Tripartite

The state

The state

Financing

Employers’ and workers’ contributions

State budget

Département, local taxes and state budget

45

Tomáš Sirovátka

46

Accelerating Governance Reforms: the French Case

conditional insurance benefits relative to their previous occupational earnings and experience. The solidarity scheme follows procedural governance principles: it is funded and governed by the state that offers conditional, means-tested and lump-sum benefits to the unemployed who have exhausted their rights to insurance benefits and still have sufficient employment records (5 years out of the previous 10). Despite these different governance principles, both schemes are strongly interrelated: social partners’ decisions regarding unemployment insurance must be agreed by the state since they have an impact on the solidarity scheme (when rules to access to insurance compensation harden, many unemployed who no longer qualify for insurance benefits turn to the solidarity scheme). Both schemes are integrated within a single benefit paying agency network: the Assédics (local agencies of Unédic) until 2009, and the local agencies of the Employment Bureau since then. Social assistance (allowances defined according to households’ income and structure) to those who belong to a poor household also obeys procedural governance principles. It is institutionally separated from the unemployment compensation system, but plays the role of a third stage in social protection of the unemployed. From an institutional point of view, income support schemes are defined at the national level by the state, and funded by both the state and the Départements.10 The allowance-paying agencies are the Local Family Funds (CAF) that belong to the National Family Fund (CNAF). Corporatist regulations: the role of social partners in PES administration Social partners play a central role in the governance of public employment policies and services in France, through their contribution to employment negotiations as well as to the administration of PES institutions (Freyssinet, 2010). They administrate a range of employment institutions: the Unemployment Insurance Fund (Unédic), the Association for Adult Vocational Training (AFPA), the Association for Managers’ Employment (APEC) and so on. As such, they contribute to the definition of employment services. As administrators of Unédic (together with the state), they have been in charge of designing insurance benefits since 1958. Since the 1980s, they have also participated in activation policies aimed at encouraging the return to employment through the definition of the rules allowing the combination of a job (called ‘reduced activity’) with insurance benefits, or through the development of activation programmes dedicated to the insured unemployed. In 2001, social partners went a step further in their involvement in activation policies,

Tomáš Sirovátka

Anne Eydoux and Mathieu Béraud

47

by signing the tri-partite convention of 13 June 2001 with the ANPE and the state, regarding the implementation of the Plan to Support Return to Employment (PARE). Through this convention, they committed themselves to participate in the financing of the additional human resources needed by ANPE to implement this activation programme. After 2001 and the generalization of job-search support by ANPE, the Unédic also gradually increased its participation in vocational training for the insured unemployed, to become a major contributor in this field along with the state and the regions, until the late 2000s, and the merger of ANPE and the Assédics into the new Employment Bureau (see below).

Governance reforms Since the early 1980s, many changes have occurred in the governance of French PES and employment policies. Major employment institutions were created at a time when the unemployment rate was very low (the 1950s and 1960s) and subsequently had to adapt to the rise of massive unemployment, as well as to the reinforcement of activation strategies from the 1980s onwards. Governance reforms have affected the centralized steering of employment policies (through decentralization processes), as well as the management (NPM procedures, contractualism) and the organization of PES (outsourcing, inter-agency cooperation). An unsynchronized decentralization process and a limited devolution to local government In France, where the steering of activation policies is centralized, decentralization and devolution are at work, but proceeding in a disorganized and unsynchronized way.11 Decentralization primarily takes the form of competency transfers for vocational training and social inclusion matters. Employment policies and PES remain responsibilities for central government, with only limited devolution of powers. The French decentralization process essentially concerned income support policies and vocational training. It began in the early 1980s, when 40 laws and 300 decrees gave new competencies to local government to favour local initiatives. It received a new impulse in the early 2000s (Cour des comptes, 2009). Constitutional Law No 2003-276 of 28 March 2003 regarding the decentralized organization of the Republic recognized the principle of free administration for local authorities and provided for resource allocations in accordance with their new competencies and associated expenses. The Constitutional Bylaw No 2004-258 of 29 July 2004 gave a certain degree of financial autonomy to local

Tomáš Sirovátka

48

Accelerating Governance Reforms: the French Case

authorities. Since the 1980s, the decentralization of vocational training, social assistance and the social integration of income support recipients has taken place at three levels: regional, departmental and local (communal and inter-communal). Today, France’s regions play an important role in local vocational training policies: they have competencies (since the mid-1960s) in economic development, as well as in workers’ and youth vocational training since 1983 and 1993, respectively. Regional competencies now also encompass vocational training for the unemployed (since 2004) and extend to the governance of AFPA programmes (since December 2008). The aim of the decentralization laws of 2003 and 2004 was to reinforce the regions as the relevant level for conducing decentralized policies. This purpose has, however, been impeded by later policy decisions, for instance entrusting the implementation of the Active Solidarity Income (RSA) and the development of systematic job search support for income support recipients to the departmental level. Since 1988, the Départements have had competencies in local development and in the social integration of income support recipients. Since 2004, General (departmental) Councils have had responsibility for the activation programme Minimum Activity Income (RMA12 ). Since 2009, they have implemented the Active Solidarity Income (RSA) and have organized job-search support for all RSA recipients, either by establishing partnerships with the Employment Bureau or by contracting with associations in the non-profit sector. At a local level, communes (or groups of communes) have local competencies regarding social assistance and contribute more and more to local development in coordination with other actors in the field of employment, vocational training or assistance. All in all, the decentralization process appears unsynchronized and partly inconsistent (Cour des comptes, 2009). At a time when income support recipients are increasingly considered as ordinary unemployed, their enrolment in activation programmes could have legitimated their affiliation to the Employment Bureau. However, their integration into employment has been increasingly delegated to departmental institutions (General Councils), usually in charge of social aid (Eydoux and Tuchszirer, 2010). The devolution of responsibility for employment and activation policies reflects two alternative logics: a downward logic consisting of central steering and undifferentiated implementation at the local level; and an upward logic aiming at encouraging local development and initiatives coming from local actors (Berthet et al., 2002). Upward processes are at work in employment policies but remain limited: development

Tomáš Sirovátka

Anne Eydoux and Mathieu Béraud

49

initiatives from local actors have been gradually encouraged, while PES was compelled to participate in such initiatives. At the beginning of the 1980s, the creation of Local Missions as well as the implementation of local employment committees13 aimed at favouring local cooperation and initiatives. At the end of the 1990s, several active employment measures, based on local diagnosis and aimed at responding to both national and local needs, were delegated to local action plans. Since 2005, local actors’ cooperation has been encouraged within Employment Houses that bring together employment intermediaries and which are presided over by locally elected representatives. However, the coordination of local actors remains the Achilles’ heel of devolution: local actors do not have the same level and degree of decisional competencies. The employment zones they cover are often defined differently at the sub-regional as well as at the sub-departmental level, and their competencies may be redundant. Also, their institutional logics or vocational cultures may be conflicting (Berthet and Cuntigh, 2002; Marimbert, 2004). French employment policies remain dominated by a downward logic, implying central steering and territorial implementation, while PES remains centralized with only limited internal decentralization (Marimbert, 2004). Unemployment benefits and income support are still centrally defined at the national level, by social partners (insurance schemes) or by the state (solidarity and assistance schemes), while benefit-paying agencies mainly check whether or to what extent claimants qualify to benefit from unemployment compensation or income support. Similarly, the public effort dedicated to activation programmes is a centralized prerogative and service provisions are defined in a standardized manner – except for income support recipients. Some competencies of the National Employment Agency (ANPE, now the Employment Bureau) have been gradually delegated to the regional level or to local employment agencies, enabled to decide how they meet national requirements and to what extent they delegate service provisions to other providers. This limited decentralization of PES came with the development of limited NPM practices.

New Public Management and State contractualism: slowly, softly, paradoxically The promotion of NPM in France dates back to the end of the 1980s, when the government of Prime Minister Rocard launched a modernization process in the public sector (Decree of 23 February 1989 and Decree of 25 January 1990). The aim was to improve the quality of services for

Tomáš Sirovátka

50

Accelerating Governance Reforms: the French Case

the public, to reform human resource management and to develop performance management, policy evaluation and accountability. However, this process remained quite slow and ambivalent in France, even with the General Revision of Public Policies (RGPP) launched in 2007 (Perret, 2008). In PES, it has essentially concerned substantive activation policies with the generalization of impact assessment procedures. The promotion of NPM in the governance of French PES has remained limited and has taken the form of much softer contractualism than in either the Netherlands or the UK. The introduction of performance contracts (so-called ‘progress contracts’) between the state and the National Employment Agency (ANPE) since 1990 is the best illustration. Barbier (2007), relying on the analysis of the first three progress contracts (1990–2003), has shown that French contractualism has been essentially formal. The ANPE was not fully independent from the state and not as autonomous, in order to be able to negotiate a contract with the state – or to be evaluated, penalized or rewarded for its results. Moreover, when contracting with the ANPE, the state was not able to predict the evolution of unemployment and the future adaptations of employment policy programmes that were not written into the contract. As a consequence, the first three progress contracts (1990–3, 1994–8, 1999–2003) were not always honoured by either party, while policy orientations and change were decided as if the contracts did not exist. Policy evaluation and accountability did not really develop and remained limited, even in the fourth performance contract for 2006–10. With hindsight, as Barbier (2007) puts it, major change in employment and activation policies as well as in the governance of the ANPE took place outside the scope of performance contracts. This was again the case when the merger of the ANPE and the Assédics was decided in 2008. If the contracting process between the state and the National Employment Agency remained limited, other forms of contracting, between the ANPE and other providers, have expanded since the 1980s. Contracting-out and outsourcing: from partnership to quasi-markets? The increasing outsourcing of activation services in France has not primarily followed explicit (neoliberal) policy orientations aimed at institutionalizing efficient quasi-markets. But things have changed since the mid-2000s, when official reports began to examine it and to perceive it as a means of extending the scope and (possibly) the efficiency of the French PES. The outsourcing of activation services originated in the development of a public activation strategy in the face of massive unemployment

Tomáš Sirovátka

Anne Eydoux and Mathieu Béraud

51

(Balmary, 2004, 2006). It began at the end of the 1980s with the implementation of programmes favouring the redeployment of laidoff workers and the integration into employment of vulnerable groups. Outsourcing intensified with the reinforcement (at the end of the 1990s) and generalization (in the 2000s) of personalized support in job search for every unemployed person, as well as with the development of consultancy services for employers. In 1998, programmes that were targeted at young unemployed with no qualifications were extended to all longterm unemployed youth (within the New Start Programme). In 2001, the ambitious PARE (Plan to Support Return to Employment) generalized individualized job-search support to all registered unemployed, by means of a Personalized Action Plan (PAP) implemented by the ANPE. In 2006, the PAP was replaced by the Personalized Project for Returning to Employment (PPAE), implying Monthly Personalized Support (SMP). The ANPE’s human resources were expanded between 1990 and 2005, when the number of employees rose from 11,000 to 28,000. Its economic resources also increased as subsidies from the state and Unédic rose. However, ANPE was unable to implement the reinforced activation strategy alone and had to delegate activation services to other labourmarket intermediaries: by the mid-2000s, two-thirds of service provision was outsourced (Balmary, 2006). Since the mid-2000s, the contracting-out policy became more conscious and organized, gradually contributing to the construction of a quasi-market for job search support and placement. In accordance with Communication 641 of the European Commission (1998), the Law on Social Cohesion of 18 January 2005 put an end to ANPE’s formal public monopoly on job placement.14 Henceforth, every provider, whether public, non-profit or private (such as temporary work agencies), has been able to participate in job placement, as long as services to the unemployed remain free and non-discriminatory. In turn, the ANPE has been allowed to create affiliated companies delivering specific, chargeable services to firms and competing with other private intermediaries like temporary work agencies, human resource or recruitment agencies, although it has not yet used this opportunity. Since 2005, the public call for tender procedure regarding services aiming to favour labourmarket integration has become increasingly competitive. These changes have raised debates opposing proponents of increased privatization and marketization of employment services in the name of economic efficiency, and advocates of public services in the name of social cohesion and equity. It has also provoked conflicts within employment institutions: on 29 March 2007, more than 43 per cent of the 28,000 ANPE agents staged a one-day strike, fearing the deconstruction of PES.

Tomáš Sirovátka

52

Accelerating Governance Reforms: the French Case

The 2005 Law on Social Cohesion extended the scope (circles) of PES, to include the many intermediaries involved in service provisions (see Table 3.2). The three circles, however, do not reflect the intermediaries’ activities or responsibilities as purchasers or providers – or purchaserproviders like ANPE (today, the Employment Bureau). Instead, they are defined according to their institutional proximity to the state (Balmary, 2006). The third circle combines providers whose relationships to ANPE or to the hard-core PES are very different. Some of them (co-contractors), mainly in the non-profit sector, are contributing to PES through partnerships and cooperation, receiving public subsidies for their activities: Local Missions, the APEC and the Employment course. Others, called subcontractors, are more or less competing in providing services (purchased by the Employment Bureau, other hard-core PES institutions or General Councils) in the quasi-market for job-search support, training or consultancy: private non-profit providers dealing mainly with vulnerable groups as well as private for-profit providers such as temporary work agencies or human resource agencies that may either provide services to the unemployed for job search or to employers for human resource consultancy. According to Balmary (2006), the 2005 Law had a strong political meaning: by extending the definition and scope of PES, it established that the state does not assume the exclusive responsibility for the labour market situation and that employment is a matter of public (that is, general) interest, justifying the common effort of various actors. It does not however institutionalize a regulated market for service provision: while competition between subcontractors is current in France, subcontractors’ activities are not systematically evaluated, providers are only paid exceptionally according to their results (CERC, 2005) and ‘clients’ (the unemployed) do not choose their provider. Moreover, the Law reasserts public service principles in guaranteeing equal access to PES, such as non-discrimination, free job placement for the unemployed, public control and information about job placement. The quasi-market thus remains limited: competition mainly applies to subcontractors, essentially in territories where they are sufficiently numerous, while co-contractors mostly escape from it. However, the gradual extension of PES and the involvement of new private providers, together with the new rules applying to public invitations to tender in the past few years, have broadened the quasi-market. Private actors in the profit sector, such as temporary work or human resources agencies, entered the market to diversify their activities, thus challenging traditional associations in the non-profit sector whose position now appears

Tomáš Sirovátka

Anne Eydoux and Mathieu Béraud

53

more delicate (Béraud and Eydoux, 2009). Moreover, ANPE’s wish to draw together its subcontractors led the national agency to put an end to its existing contracts in December 2007 (which would normally ran until December 2008) and to launch new, extended calls for tender (concerning 20,000 to 30,000 unemployed) in broader geographical zones, thus making it hard for small providers to satisfy the new conditions. Recently, competition has also been extended to co-contractors: the APEC in charge of unemployed managers is now increasingly challenged by private providers and is worried about its future (Bissuel, 2010).

Inter-agency cooperation and the merging of benefits and activation services Governance reforms have also promoted inter-agency cooperation between the unemployment insurance agencies (Assédics) and the National Employment Agency (ANPE). An institutional merger, first evoked in 1966 just before the creation of ANPE, has been placed on the agenda again since the beginning of the 2000s, especially following the official report by Marimbert (2004) on the merger of employment services. The Law on Social Cohesion of 2005 provided a first step in this direction, by encouraging the setting up of single gateways offering diverse institutional services at the local level, in order to prevent the unemployed from going from one institution to another. It was at first experimented through new institutions called Employment Houses (Maisons de l’emploi) gathering (in principle in the same building) intermediaries from the first and second circles, most often Local ANPE agencies, the Local Missions and sometimes the Assédic or AFPA agencies, or possibly also temporary work agencies. This joint location had a tool, a single electronic file (DUDE) containing all relevant information on unemployed persons for the use of various authorized institutions (ANPE and Assédics). The Law of 13 February 2008 went a step further, providing for a merger between the two major institutions of the first circle, ANPE and the Assédics, into a single one, the Employment Bureau as of 1 January 2009. The objective was first to increase the efficiency of PES by developing one-stop agencies for the unemployed and centralizing the information regarding unemployment benefits and job search; and secondly to reduce the existing segmentation in the treatment of jobseekers who, according to their benefit status, relied on different institutions and activation programmes provided by ANPE (all unemployed) or Unédic (insured unemployed only). In other words, the aim was to

Tomáš Sirovátka

54

Accelerating Governance Reforms: the French Case

‘cut the umbilical cord between the income protection system and the system supporting the unemployed job-search’ (Tuchszirer, 2010).

A changing governance mix? The accelerated reform process currently operating in activation policies, as well as in the organization of PES, makes it difficult to characterize the changes in the French governance regime. Reflecting the institutional segmentation of activation policies and income protection institutions, the French governance regime is a mixed one, combining state and corporatist procedural governance with corporate governance (mainly in hard-core PES institutions), partnership and market (in the contractual relationships between hard-core institutions and actors in the second and third circles). Nevertheless, adaptations are at work in the governance mix: long-term decentralization, outsourcing and marketization processes have led to three major changes. First of all, if the steering of activation policies and the governance of PES remain mainly procedural and centralized, dominated by the state’s decisions and control, proceeding by law and tripartite negotiations and asserting public service values, there is a limited and ambivalent decentralization process. Competency transfers to local authorities have been restricted in their scope to vocational training and to the reintegration into employment of income support recipients, while the devolution of employment policies has mainly responded to a downward logic, leaving little room to local initiatives. The internal decentralization of the National Employment Agency (ANPE, now the Employment Bureau) also remained limited, relying on soft new management procedures. Secondly, the governance of the many intermediaries who operate in the extended PES is now largely decentralized, obeying mainly a partnership model, as far as the relationships between the Employment Bureau and its co-contractors are concerned, and increasingly to a quasimarket model. In the past decade, changing rules have imposed more publicity and competition in the call for tender procedures regarding activation and social inclusion services. Increased marketization is at work, exposing traditional non-profit subcontractors, and even some former partners of the national employment agency (co-contractors), to the challenges of new private providers. Thirdly, if the French governance regime is partly corporatist, in the sense that unions’ regulations play an important role, social partners who have jointly administrated Unédic since 1958 have recently lost some of their institutional and decisional prerogatives in the field of activation and training programmes for the insured unemployed. The

Tomáš Sirovátka

Anne Eydoux and Mathieu Béraud

55

decision in 2008 by the French government to merge ANPE and the Assédics to create the new Employment Bureau has led to both transfer of Unédic’s activation programmes (with corresponding funds) to the new institution and confinement of social partners to the enforcement of the unemployment insurance convention: they played only a limited role in the reform of unemployment compensation during the negotiation of the 2008 national agreement on the modernization of labour market. From this point of view, the merger represents a major change (Tuchszirer, 2008) that reduces the participation of Unédic in activation policies and limits social partners’ role in unemployment insurance.

The blurred impacts of governance reforms In contrast to the substantive activation reforms and employment programmes that are now systematically assessed (the rate of return to employment being the main indicator), PES governance reforms have been evaluated much less. Measuring the impacts of governance reforms is thus a delicate task. Not only have these reforms been complex and liable to produce a wide range of (desirable, as well as unexpected) effects, but the indicators that might be used for that purpose are lacking because policy objectives were not always explicitly formulated or were changing or even ambivalent – as diverse as improving the quality of services for the unemployed, providing low-cost services, increasing the rate of return to employment. Experts and official reports often agree in asserting that the assessments of the French governance regime are scarce, not always carried out by independent experts, and hardly disseminated to the public because of a weak culture of accountability. The lack of evaluations in PES governance reforms mainly concerns the process of the state’s internal decentralization and contractualism, as well as the role of social partners: whereas Barbier (2007) demonstrates that the assessments of performance (progress) contracts between the state and ANPE have long remained partial and confidential, Freyssinet (2010) emphasizes that there has been no real impact assessment of the social partners’ actions on macroeconomic employment regulations, as well as on equity and security in the labour market. We will thus focus here on the impact of the decentralization processes, the subcontracting of employment services and increasing inter-agency cooperation. Decentralization in practice: the diversity of local implementation A recent report on the ‘State’s implementation of decentralization’ (Cour des comptes, 2009) points out the unsystematic character of

Tomáš Sirovátka

56

Accelerating Governance Reforms: the French Case

evaluation regarding the exercise of competencies delegated to local authorities: statistics on decentralized assistance and integration policies are insufficient, shared diagnoses and information are missing, while experimentation remains timid. Evaluations and experimentation are, however, increasingly developing at a decentralized level, bearing essentially on substantive activation reforms in the field of social inclusion policies: the recent implementation of the Active Solidarity Income boosted experimentation and assessments in Départements before and after its generalization in 2009. Procedural and institutional reforms have scarcely been assessed. In recent empirical research bearing on the implementation of local social policies regarding care for the elderly and the social integration of income support recipients, Gramain and Neuberg (2009) analysed the impact of the mid-2000s decentralization. First of all, they show that the decentralization of job-search support to the General (departmental) Councils has been accompanied by the re-centralization of competencies previously delegated to a sub-Departmental level. Secondly, analysing the relationships between General Councils in charge of the activation of RSA recipients and their subcontractors, they underline the variety of local practices, pointing out the fact that despite the wish of public authorities to develop quasi-markets through new call-for-tender rules, selection procedures do not systematically proceed through call-for-tenders. Call-for-tender procedures often seem complex to local authorities, while competition turns out to be irrelevant when local providers are few or when some of them appear as key players. Moreover, local authorities tend to circumvent call-for-tenders that place providers in a precarious situation: market rules requiring that providers should be paid according to their outcomes would make them vulnerable to the local economic circumstances, while a payment based on effective provisions would expose them to cash problems. The local governance of integration policies thus tends to be based on negotiations between local authorities and their providers, rather than on market or competitive relationships, so that the variety of local arrangements exposes income support recipients to differentiated treatment. Contracting-out: the public provider performs better than private ones The outsourcing of activation and employment services since the 1980s has primarily developed without any assessment of its effects (Balmary, 2004). It is only recently, with the public objective of reforming PES

Tomáš Sirovátka

Anne Eydoux and Mathieu Béraud

57

governance and delegation of job-search support to private providers deemed to be both cheaper and more efficient, that an experiment has been carried out. The so-called OPP-CVE experimentation is aimed at assessing and comparing the efficiency (in terms of costs and rate of return to work) of public service provision by the National Employment Agency (ANPE) and private services from new providers (like Altedia or Ingeus) subcontracting for Unédic, called ‘private providers for job placement’ (OPP). For the purpose of the experiment, ANPE developed a new service provision called ‘on Course Towards Enterprises’ (CVE) that could be compared to the services of private new providers (advisors offering a comparable service to the same number of unemployed with similar characteristics). Within its methodological limits, the quantitative evaluation proved that ANPE did better than private providers in terms of reintegration into employment (Béhagel et al., 2009). Meanwhile, empirical research has shown that coaches working for these new providers do not develop innovative methods, but rather use selective procedures (such as creaming) in order to improve their outcomes (Divay, 2009). More generally, the marketization at work tends to draw new private providers into the quasi-market for placement and job-search support. These new providers in turn challenge non-profit associations specialized in the reintegration of the unemployed, creating precarious conditions for traditional subcontractors (Béraud and Eydoux, 2009). Whatever these outcomes, which could have favoured cautiousness in the marketization and outsourcing process (if not direct limits), the Employment Bureau now submits more and more providers to market rules and, given the rise in unemployment, turns increasingly to private providers. Inter-agency cooperation and the merging of ANPE and Assédics: a treacherous process The impact of inter-agency cooperation and the effects of the merger of ANPE and the Assédics are still blurred and difficult to comprehend. Employment Houses where inter-agency cooperation has developed since the 2005 Law on Social Cohesion have not been designed as experimental tools. In practice, their organization, functioning and policies have revealed great local disparities (Anciaux, 2008). Fieldwork in Employment Houses has shown that the implementation of interagency cooperation between the Assédics and the local agencies of ANPE has turned out to be very complex and delicate (Béraud and Eydoux, 2009). First, technical and practical problems have limited cooperation: the electronic DUDE file for sharing information was not available to

Tomáš Sirovátka

58

Accelerating Governance Reforms: the French Case

all personnel in Employment Houses. Secondly, institutional problems created organizational conflicts. For instance, the director of a local ANPE agency complained about the fact that ANPE agents at the common reception of the Employment House were essentially dealing with benefits matters, and thus paid to do the work of Assédics employees. More fundamentally, difficulties were statutory: Assédic and ANPE employees had different qualifications and competencies as well as different working and employment conditions (in particular, different wage levels), but were increasingly asked to do the same work. When interviewed in 2007, directors of both Assédics and local ANPE agencies did not believe in the feasibility of the merger between both institutions at the national level because of these existing statutory differences. Nevertheless, the merger and the creation of the new provider have been effective since January 2009; in a conflictual context, aggravated by the economic slowdown and the increasing number of unemployed registering at the new Employment Bureau. The impact of the merger has not yet been assessed, but so far seems rather negative in terms of organizational efficiency: press articles often relate the organizational difficulties of the new institution in facing both its organizational change (implying conflicts, strikes, statutory changes as well as massive vocational training programmes for employees) and the new rush of unemployed brought on by the current economic crisis. All in all, the overall impact of French governance reforms remains difficult to assess, notably because the reform process has been accelerated in the past few years, blurring the evaluation processes. In terms of efficiency, the reforms have not solved the coordination problems faced by the many actors involved in the implementation of activation policies. At the local level, actors sometimes find solutions to improve their coordination (Béraud and Eydoux, 2009) or to escape from marketization (Gramain and Neuberg, 2009), but the complexity of the landscape is still the same. At the national level, the restructuring of PES has been twofold. First, the merger of the Assédics and ANPE has slightly reduced the institutional complexity in the first circle, but at the cost of major organizational difficulties, at least in the transition period. Secondly, the outsourcing and marketization processes have included new actors in the third circle competing for subcontracts with the Employment Bureau, without any empirical evidence regarding the efficiency of the quasi-market: experimentation as well as fieldwork rather suggest that much cautiousness is needed in market regulations.

Tomáš Sirovátka

Anne Eydoux and Mathieu Béraud

59

Notes 1. 2. 3. 4. 5. 6.

7. 8. 9. 10.

11.

12. 13. 14.

Agence nationale pour l’emploi. Associations pour l’emploi dans l’industrie et le commerce. Union nationale pour l’emploi dans l’industrie et le commerce. Contractualism here primarily refers to the fact that the state sets performance contracts with public organizations. Association pour la formation professionnelle des adultes. Like ANPE before, the Employment Bureau is a national institution with local agencies (local employment bureaux) in charge of the registration of the unemployed, as well as of job search support and benefit payments at the local level. While Assédic agencies have been merged into the Employment Bureau, the joint institution (Unédic), which defines insurance compensation rules at a national level, still exists. Direction régionale des entreprises, de la concurrence, de la consommation, du travail et de l’emploi. Association pour l’emploi des cadres. Cap emploi. Since 2004, the state has transferred the funds and the responsibility for financing the RMI to the Départements. Today, the Départements still finance the basic share of the RSA allowance, while the state pays for the additional (‘active’) share designed to make work pay. The French term ‘décentralisation’ refers to the transfer of power from central government to France’s regions or departments. Its meaning is closer to that of devolution, but for linguistic reasons the English term ‘decentralization’ is used as a translation here. RMA was a subsidized contract dedicated to RMI and to ASS (unemployment solidarity scheme) recipients. Comités de bassin d’emploi. This monopoly was shared with municipalities and other institutions working with ANPE; it was essentially formal because many actors used to participate in job placement.

References Anciaux, J.-P. (2008) Les maisons de l’emploi, mission d’évaluation du dispositif (Paris: Ministère de l’économie, des finances et de l’emploi, La documentation française). Balmary, D. (2004) Rapport de l‘instance d’évaluation de la politique de l’emploi et recours à des opérateurs externes (Paris: Commissariat Général au Plan, La documentation française). Balmary, D. (2006) ‘Un nouveau service public de l’emploi’, Droit Social, 6, 594–8. Barbier, J.-C. (2002) ‘Peut-on parler de l’activation de la protection sociale en Europe?’, Revue française de sociologie, 43/2, 307–32. Barbier, J.-C. (2007) ‘La gestion et l’évaluation du service public de l’emploi en France dans la décennie 1990: Matériaux pour une réflexion comparative

Tomáš Sirovátka

60

Accelerating Governance Reforms: the French Case

internationale’, Documents de travail du Centre d’Economie de la Sorbonne, 2007.01. Baron, C., M.-C. Bureau, E. Le Dantec and P. Nivolle (1994) ‘Les intermédiaires de l’insertion’, Dossiers du CEE, No1. Béhagel, L., B. Crépon and M. Gurgand (2009) Evaluation d’impact de l’accompagnement des demandeurs d’emploi par les opérateurs privés de placement et le programme Cap vers l’entreprise (Paris: CREST). Béraud, M. and A. Eydoux (2009) ‘Activation des chômeurs et modernisation du service public de l’emploi: Les inflexions du régime français d’activation’, Travail et emploi, 119, 9–21. Berthet, T. and P. Cuntigh (2002) ‘Politiques d’emploi et territoires’, Bref Céreq, 182, 1–4. Berthet, T., P. Cuntigh, C. Guitton and O. Mazel (2002) ‘1982–2002: La territorialisation progressive des politiques de l’emploi’, Premières informations et premières synthèses, No 24.2 (Paris: Dares). Bissuel, B. (2010) ‘Concurrencée, l’APEC cherche sa raison d’être’, Le Monde, 3 May. Boulanger, J.-M. (2008) Contribution à la préparation de la convention tripartite entre l’Etat, l’Unédic et la nouvelle institution créée par la loi du 13 février 2008 (Paris: La documentation française). Cahuc, P. and F. Kramarz (2004) De la précarité à la mobilité : vers une sécurité sociale professionnelle (Paris: La documentation française). CERC (2005) Aider au retour à l’emploi, Rapport du Conseil Emploi Revenus Cohésion sociale No.6 (Paris: La documentation française). Considine, M. (2001) Enterprising states. The Public Management of Welfare-to-Work (Cambridge: Cambridge University Press). Cour des comptes (2006) L’évolution de l’assurance chômage: de l’indemnisation à l’aide au retour à l’emploi (Paris: La documentation française). Cour des comptes (2009) La conduite par l’Etat de la décentralisation (Paris: La documentation française). Divay, S. (2009) ‘Nouveaux opérateurs privés du service public de l’emploi: les pratiques des conseillers sont-elles novatrices’, Travail et emploi, 119, 37–49. Dupeyroux, J.-J. (2005) ‘Avant propos’, Droit Social, 12, 1081–3. Erhel, C. (2010) ‘Les politiques de l’emploi en Europe: le modèle de l’activation et de la flexicurité face à la crise’, Document de travail CEE. Esping-Andersen, G. (1999) Les trois mondes de l’Etat providence (Paris: Presses Universitaires de France). European Commission (1998) Modernising Public Employment Services to Support the European Employment Strategy, Communication from the Commission of 13 November 1998, COM(98) 641 final. Eydoux, A. and C. Tuchszirer (2010) ‘Du RMI au RSA: les inflexions de la solidarité et de la gouvernance des politiques d’insertion’, Document de travail du CEE. Ferracci, M. (2007) ‘Améliorer le service public de l’emploi: ce que disent les faits’, Revue française d’économie, 21/3, 75–135. Freyssinet, J. (2010) Négocier l’emploi, 50 ans de négociations interprofessionnelles sur l’emploi et la formation (Paris: Editions Liaisons). Gramain, A. and S. Neuberg (2009) ‘Réagencements territoriaux récents et conduite des politiques sociales à l’échelle locale’, Travail et emploi, 119, 77–91.

Tomáš Sirovátka

Anne Eydoux and Mathieu Béraud

61

Marimbert, J. (2004) Le rapprochement des services de l’emploi (Paris: La documentation française). Perret, B. (2008) L’évaluation des politiques publiques (Paris: La découverte). Tuchszirer, C. (2008) ‘La réforme du service public de l’emploi: suite et fin?’, Regards sur l’actualité, 343, 65–73. Tuchszirer, C. (2010) ‘France. An unemployment system ill-adapted to the economic crisis’ in F. Lefresne (ed.) Unemployment Benefits Systems in Europe and North America: Reforms and Crisis (Brussels: ETUI).

Tomáš Sirovátka

4 Fragmented Governance Continued: the German Case Irene Dingeldey

Introduction The importance of the insurance principle for the provision of status-oriented benefits and corporatist governance arrangements (inclusion of social partners and welfare production by third sector organisations) were highlighted by Esping-Andersen (1990), when he characterized Germany as a conservative corporatist welfare state. These features, however, do not entirely reflect the rather complex governance of the German welfare system. Another important aspect, in particular in the area of employment policy, is institutional fragmentation. Since the 1960s a three-tier system provided functionally differentiated benefits to the unemployed and the poor. Those with sufficient employment records were directed to the primarily contributionbased unemployment insurance system, receiving an earnings-related unemployment benefit (UB; Arbeitslosengeld), as well as employment promotion addressed to maintain the achieved employment status. The long-term unemployed received unemployment assistance (UA; Arbeitslosenhilfe), still paid in relation to prior income (although at a lower rate) but tax financed (from the federal budget) and means tested. Finally, the social assistance (SA; Sozialhilfe) system was designed to prevent poverty through the provision of a tax-financed (by municipalities), flat rate and strictly means-tested benefit. According to the different tiers, but also within the single tiers, responsibilities for labour-market programmes, benefit payments and employment assistance were subject to multi-level governance. Hence, the main feature of the governance of German labour-market policy may be its fragmentation. Most recent reforms of the governance structure under the label of activating labour-market policy therefore not only aimed for 62

Tomáš Sirovátka

Irene Dingeldey

63

marketization and the introduction of New Public Management (NPM) structures but primarily for a homogenization of the institutional structure, namely the improvement of inter-agency cooperation and coordination, including full organizational mergers. As a result of these reforms in Germany we see a decreasing role of the social partners, a much higher importance for NPM, a slight increase in the market as well as the rather mixed trends of decentralization and (re-)centralization. Most significant, however, is the partial failure of reforms intended to homogenize the institutional structure, which has resulted in an even more fragmented governance regime than before. This argument will be developed in this chapter in several steps: First, a brief overview will be given of the main actors and agencies involved in administering income protection schemes and activation services for the unemployed at the end of the 1990s (the start of active welfare state reforms in Germany) and at present (summer 2010). Secondly, various changes of governance will be analysed in more detail. Finally, I will give an interpretation of the effects of the governance reforms and a possible shift in the governance regime.

The Hartz reforms and beyond: a path break in the governance of labour-market policies With respect to labour-market activation, Germany may be called a late reformer. While other European countries had started to reshape their unemployment assistance systems by the end of the 1990s, Germany transplanted its system to the east (Dingeldey, 2007, 2009). During the reunification process one of the major goals of labour-market policy was to cushion the enormous job losses through various labourmarket programmes. Thus, it was not until 1998 that labour-market policy was activated (still under the Christian Democratic–Liberal coalition), encompassing a re-regulation of the rules for the unemployment insurance system; reform of the Federal Employment Office (FEO; Bundesanstalt für Arbeit); and changes to the instruments of active labour-market policy. Within the law employability was formulated as a major goal of labour-market policy and the process of reforming the governance of the FEO began (Finn et al., 2005). At the beginning of 2002 the so-called placement scandal happened, following a Federal Audit Office report in 2002 which revealed that the FEO had exaggerated its job placement data. As a reaction the government imposed the commission on ‘Modern Services in the Labour

Tomáš Sirovátka

64

Fragmented Governance Continued: the German Case

Market’ (briefly the Hartz Commission, named after its chairman), which may be regarded as a strategy to overcome reform deadlock by externalizing the preparation of political decisions.1 The job of the commission was not to develop an entire concept for employment creation, but to prepare a reform of labour market administration and services. The commission’s proposals (Hartz, 2002) have been enacted, with some revisions, by four subsequent legislative acts on Modern Services on the Labour Market between 2002 and 2005, briefly called the Hartz Reforms I–IV. Although the reform of job placement services had begun prior to the Hartz Commission, the process was greatly intensified under Hartz I and II, although Hartz III (which came into force on 1 January 2004) provided the major impulse to change the FEO into a modern provider of labour-market services. Under Hartz IV, local employment agencies were supposed to become Job centres responsible for all labour-market services to the unemployed. The organization of Job centres as one-stop shops was supposed to end the division of responsibility between the FEO and municipalities, the former being responsible for the administration of unemployment insurance benefits and employment promotion measures and the latter for payment of social assistance and other social services. The combination of formal and operational reform elements, respectively the integration of unemployment assistance and social assistance into one benefit system for all those capable of working in 2005 and the respective change of organizational structures, may be seen as the most drastic reform in the field of labour-market policy during the last 20 years in Germany. The comparison of the respective governance structures, however, shows that it hardly displayed a reduction of complexity and fragmentation. At the end of the 1990s, just before the starting point of active welfare state reforms in Germany, the concrete implementation structure for income protection and employment services for the unemployed was as outlined in Table 4.1. The FEO was established as the central body of the unemployment insurance system. Although contributions and benefits are determined politically, it enjoyed considerable autonomy in relation to implementation. Its organizational structure reflected Germany’s federal system of government. The FEO was organized as a self-governing corporation under public law. Its Council and Management Board were dominated by a corporatist governance structure, including representation from the social partners. This structure was replicated in the other

Tomáš Sirovátka

65 Table 4.1 The implementation structure for income protection and activation in Germany (1990s and 2010) Late 1990s

2010

Unemployment Insurance Benefit

Unemployment Benefit (UB)

Unemployment Benefit I (UB I) regulated by SGB III

Benefit Administration

Federal Employment Office (Bundesanstalt für Arbeit)

Federal Employment Office (Federal Agency for work)

10 regional employment offices (Länder)

10 regional directories

181 local FEOs (660 branch offices)

178 local Agencies for Work (610 branch offices)

Funding

employer’ and employees’ contributions; deficit financing by federal budget

employer’ and employees’ contributions; deficit financing by bank credits as proposed in 2010

Activation

FEO, local offices

FEO, local offices

Private providers

Private providers

Unemployment Assistance

Unemployment Benefit II (UBII) regulated by SGB II

Benefit Administration

equal to UB

from merger of Unemployment Assistance and Social Assistance (for those capable of working)

Funding

taxes out of federal budget

Activation

equal to UB

Social Assistance Benefit Administration

Municipalities

Funding

Taxes, municipalities’ budget

Activation

Municipalities; private providers

356 ARGE consortia (joint bodies of FEO and municipalities) or municipalities (69 opted-out municipalities with sole responsibility and 23 municipalities/local FEOs with complete separation of responsibilities) Taxes: benefit from federal budget; housing costs from municipalities’ budget

Tomáš Sirovátka

66

Fragmented Governance Continued: the German Case

Table 4.1

(Continued) Late 1990s

2010 Activation as co-financing from FEO (contribution funded) and federal budget plus additional measures funded by municipalities Equal to UB I; but promotion of 1-Euro Jobs dominated by municipal employment providers

Source: Own design.

two administrative levels that consisted of a network of ten regional employment offices at Land (state) level and 181 local or regional public employment offices, with 660 branch offices, which covered districts of varying sizes and usually connected with several municipal districts (Finn et al., 2005). The FEO was responsible for the implementation of active labourmarket policy and two main benefit payments: Unemployment Insurance Benefit was paid for a maximum of 12 months (since 1987, for a longer duration of 18 (32) months only for those older than 42 (53) years). Unemployment Assistance was unlimited, but means tested. Both benefits were related to prior earnings, although the replacement rate for the latter was much higher than that for the former (UB: 60 per cent, 67 per cent with dependants; UA: 53 per cent, 57 per cent with dependants). While UB was funded mainly by payroll contributions paid in equal parts by insured employees and their employers to the FEO, UA was financed by taxes out of the federal budget. Labourmarket programmes for both UB and UA recipients were funded from payroll contributions. When unemployment rose, an increase in both government funding and contributions was necessary. The municipalities administered a third, means-tested benefit, namely Social Assistance, formed as a last and separate safety net for those in need but ineligible for UB or UA.2 It provided a flat-rate payment for adults with additional payments for children and other adult dependants living in the same household. Additional financial assistance was provided towards housing costs. The means test for SA was much stricter than for UA and included also the wider family, such as parents in separate households (Clasen, 2005; Steffen, 2009).

Tomáš Sirovátka

Irene Dingeldey

67

Hence, a general feature of the German labour-market policy was its multi-level governance, briefly summarized as follows: At federal, Land (state) and local government levels, institutions are either wholly or partially responsible for the development, implementation, and application of labour market policy instruments and measures. Their activities are guided by federal legislation that specifies general employment policy and the labour market policy instruments to be applied nationally by the FES (Bundesanstalt [now Bundesagentur] für Arbeit). The 16 Länder have no direct influence on these policies with the exception of their participation through the Bundesrat. Until 2003, the Länder were also represented, together with social partners, in tripartite committees controlling the middle tier of the FES. The Länder supervised LAs which delivered SA, the lowest level of ‘income support’. Finn et al. (2005, p. 7) Although many changes have occurred prior to 2010, the aspect of multi-level governance still persists: since reforms in 2002 the FEO operates as the Federal Employment Agency (Bundesagentur für Arbeit) and local offices have become Agencies for Work (AfW). The middle-tier Land level of administration has been reconstituted as a number of Regional Directorates. The 178 local AfWs still have a unified structure for three main services, namely placement, active labour-market policy and the administration of the insurance-based unemployment benefit (renamed as UB I; Arbeitslosengeld I) (Eichhorst et al., 2008). Entitlement to UB I is still restricted to the unemployed with an adequate contributions record and is limited to one year (with exceptions only for older workers). The greatest organizational changes were therefore related to the merger of the two benefit systems UA and SA to form Unemployment Benefit II (UB II), thereby creating a single benefit for all needy people capable of working but not eligible for UB I.3 The new benefit is paid as a flat-rate basic income support scheme (Grundsicherung) for jobseekers and other adults in the same household, with additional supplements for children as well as for housing costs (Eichhorst et al., 2008). The resulting two-tier system of unemployment protection is regulated by two legislative frameworks within the Social Code Book (SGB): the insurance-based UB I, with most of the employment promotion measures being regulated in SGB III, while the means-tested Unemployment Benefit II is regulated by SGB II (Dingeldey, 2011).

Tomáš Sirovátka

68

Fragmented Governance Continued: the German Case

In order to administer this new benefit and to design employment services for the respective clients, 356 joint bodies of local AfW and municipalities – the ARGE consortia – were created (being responsible for 80 per cent of territorial units serving 85 per cent of customers). In 23 municipalities a division of responsibilities between the AfW and the municipalities has continued, which means that local FEOs cooperate with municipal administrations but do not have formal agreements. A further 69 so-called opted-out communes (Optionskommunen) took over sole responsibility for implementing UB II and activating the respective benefit recipients (Brussig and Knuth, 2009; Kirsch et al., 2009). In spite of a homogenization of the benefit system through the creation of a two-tier system (Dingeldey, 2011), the governance of the German unemployment protection system remains complex with respect to regulation within two legal frameworks and the distribution of responsibilities, both concerning the payment of the various components and the provision of services to the unemployed.

Main trends in governance reforms Major reforms to the unemployment protection system in Germany have taken place since the end of the 1990s, as indicated in the previous section and in Table 4.1. The placement scandal in 2002 and the subsequent Hartz reforms accelerated the reform process, combining formal and operational policy reforms. In this section, I will discuss in more detail the impact of these reforms on the different modes of governance detailed in Chapter 1. Diminishing role of the social partners In 2002, as an immediate reaction to the placement scandal, the corporatist boards of directors at all three levels of the FEO were abolished and replaced by three-person Executive Boards, with a Chair, initially on a renewable five-year contract (Kannengießer and Gundel, 2003; Finn et al., 2005; Eichhorst et al., 2008). Furthermore, the competence of the still-corporatist governing board was reduced to a mere controlling function, curtailing indirectly the role of the social partners (Konle-Seidl, 2003). Even more important was the indirect marginalization of the social partners that took place with the introduction of UB II. In contrast to the self-management structures of the German unemployment insurance system, the ARGEs, or municipal administrations, are not legally obliged

Tomáš Sirovátka

Irene Dingeldey

69

to develop co-determination structures. The majority of administrations responsible for UB II voluntarily created some advisory councils which, however, do not only include the social partners but also other social and welfare associations (Deutscher Bundestag, 2008). Within the emerging network structures the representation of interests of the unemployed through trade unions became more difficult, and in administrations without councils it does not exist at all. In the entire unemployment protection system, therefore, the structures of tripartite codetermination and the influence of the social partners were weakened, whereas the role of third-sector organizations as provider of employment promotion measures increased, also in the area of interest representation (Klenk, 2009a, b). In consequence it is not corporatist tripartism but network governance that dominates the implementation of UB II at the local level (see also below). Inter-agency cooperation It was intended that the local FEOs be turned into Jobcentres that were to serve as one-stop shops to administer all labour-market services and unemployment benefits – an idea partly modelled on the British Jobcentre Plus that indeed created a unified single gateways for all unemployed people receiving a single, flat-rate benefit (Knuth, 2009). However, while the UK reform took place within a centralized political system, the respective reform plans in Germany intended not only the merger of a federal and a municipal benefit but also the reshaping of a rather complex and multi-tiered governance system. The motivation for this fundamental organizational reform in Germany was at least twofold: – One of the long-standing critics of the old, three-tiered unemployment protection system in Germany was that it created incentives for institutional actors to transfer the unemployed from one system to the other in order to save costs within the particular budgets instead of investing in their reintegration into the labour market. This practice, being characterized as a marshalling yard, was intended to end through the merger between UA and SA as a new framework for integrated provision of benefits and labour-market services. – Furthermore it was the attempt to solve problems – understood as cross-cutting or wicked issues – hitherto fragmented among the responsibilities of various public authorities in line with ideas of NPM theory through joined-up or holistic forms of governance (6, 1998; 6 et al., 2002).

Tomáš Sirovátka

70

Fragmented Governance Continued: the German Case

In spite of these goals the external reorganization of the unemployment protection system and employment services resulted in an even more fragmented financial and organizational structure than before, due to the institutional setting and protracted negotiations within a federal state (Eichhorst et al., 2008; Brussig and Knuth, 2009). First, the organizational division of benefit administration remaining as UB I is still administered by the local AfW and financed through the unemployment insurance fund. In contrast, UB II for all other people capable of working is financed by taxes paid out of federal and municipalities’ budgets and administered by different types of organizations, namely by ARGEs (consortia of AfW and municipality), by 69 (opted-out) municipalities in sole responsibility and – in 23 cases – separately by local AfWs and municipalities. Secondly, even within the ARGE organizations a clear division of labour between the local AfW and the municipalities remains. The local Agency for Work is responsible for financing and implementing active labour-market measures and the administration of the benefit for all those categorized as being capable of working (359 ¤ per month in 2010; 90 per cent of it to partners considered as being capable of working) as well as supplements for children below the working age of 15 (Sozialgeld). The municipalities are in charge of the rent subsidies4 and traditional social services like debt, drug, social-psychological counselling and childcare facilities (Eichhorst et al., 2008). The character of the ARGE as a joined body is still questioned, because there is a clear tendency within the consortia to develop into organizations of their own that use only staff and facilities of the partners who established them (Brussig and Knuth, 2009). This means that, for example, variation in case managers’ contractual employment conditions regarding working time and salaries is a function of being employed either by the local AfW or the municipality. Hence, with respect to just one case, the process of placement, employment promotion and benefit administration is still based on the division of work between different organizational entities with different promotion cultures. In the 23 cases that still have an entirely separate fulfilment of administering UB II by local authorities and the AfW, hardly any integration of the different employment services and the administration is to be found. Hence, only the 69 opted-out municipalities are fully responsible for all services. In consequence, the orientation towards case management is much stronger and the various functions are much more integrated here (Schütz, 2009).

Tomáš Sirovátka

Irene Dingeldey

71

The continued fragmentation of the reformed unemployment system is summarized as follows: Instead of the desired ‘single gateways’ or the uniform ‘Job-Centre’, Germany now has two separate regimes of social protection and employment services for workless people as before, but the second tier has now become much larger [ . . . ] and comes in three variants. In the 23 territories with ‘separate fulfilment’ not only two gateways exist, but the majority of workless people actually have to visit both of them in order to collect their means of subsistence. Brussig and Knuth (2009, p. 78) Thirdly, the fragmented financial responsibilities still create incentives to transfer the unemployed from one benefit system to the other. When the new benefits were created, the municipalities used the classification ‘being capable of working’ in a way that meant that 90 per cent of former SA recipients were transferred to UB II, also including people with severe disabilities, as it was in the interest of the municipalities for their benefits to be partly financed through the federal budget (Eichhorst et al., 2008). At the same time, incentives were maintained to avoid labourmarket activation for the hard-to-place unemployed receiving UB I, as they were not expected to be integrated easily into the labour market, but would soon transfer to another system, namely UB II. In order to combat that trend, since 2008 half of the costs of employment promotion and administration for the long-term unemployed have to be paid as an integration fee (Eingliederungsbeitrag) out of the insurance fund of the FEO (the other half is financed out of the federal budget). Decentralization and (re-)centralization In Germany the FEO decides the structure and content of public employment services. The provision of the respective services takes place within a unified federal structure, where decentralized implementation units cooperate with private service providers (Botzem et al., 2009). Within that multi-level governance system various processes of decentralization and recentralization have occurred, so that no strong emphasis on either can be recognized. A process of decentralization began in 1998 with the introduction of the SGB III (see above). Since then, the local FEOs are able to decide on the allocation of their funds to various programmes financed by a global reintegration budget for active measures regulated by SGB III. Furthermore, they may allocate 10 per cent of the budget to developing

Tomáš Sirovátka

72

Fragmented Governance Continued: the German Case

their own innovative programmes (Schütz, 2009), a regulation that has been expanded in 2010 to include the organizations implementing SGB II. In toto, this increased the competencies of the local employment offices to decide on the mix of instruments they want to use in line with the development of local labour markets. Since then the German unemployment insurance system is now regarded as a system with high budget flexibility at the local level. However, it is equally argued that in the German system extensive powers were given to the regional level in conjunction with re-centralization (Schneider et al., 2006; Mosley, 2009). One indicator may be the centralization and standardization of the purchase of labourmarket services through the creation of five regional purchasing centres (Regionale Einkaufszentren, REZ) that replaced the former process of negotiations of vocational training and employment promotion programmes between local FEOs and providers, including the contracting-out of placement services (Kaps, 2009b). Each of these regional centres specializes in purchasing a particular type of labour-market service, although the awarding of a contract is done by the regional department located in the respective district. This new practice is criticized, because the purchase of programmes is subject to a long bureaucratic process and the centralized tendering favours cheap suppliers without sufficient integration into local labour markets (Oschmiansky, 2010). Furthermore, the current practice through the REZ is generally seen to be detrimental to delivering integrated and very specialized services to those very distant from the labour market (Goerne, 2010). Observers therefore see a (re-)centralization of competencies in the way that local priorities within programmes may be performed only additionally, as the priority is the performance of nationally set ratios and goals. Among these, cost effectiveness is being defined as the key criterion when choosing programme contents and participants (Eichhorst et al., 2008). In this context also the centrally introduced practices of NPM may be seen as an element of centralization and standardization (see below). With respect to the governance of SGB II this judgement, however, has to be differentiated with respect to the various organizational models. As the local AfW is subject to guidance, management by objectives and control by the federal ministry of employment, the ARGE consortia are widely attached to this hierarchical model of governance while the municipal lead organizations are not (Schütz, 2009). Hence, the autonomy of local decision making is much higher for municipal providers than for the ARGEs (Institut für Angewandte Wirtschaftsforschung, 2009). A disadvantage of this organizational model, however, is seen

Tomáš Sirovátka

Irene Dingeldey

73

in the fact that a standardization of controlling and statistic software across the various institutions is still missing. Thus, even the various grades of centralization or decentralization with respect to the different institutional models may be seen as another element indicating the high degree of fragmentation in the German system. New Public Management The introduction of New Public Management in the German FEO included new forms of guidance, control and management by objectives. Since 2004, national priorities have been imposed through the planning of targets for the management by objective system that links budgets in respect of costs of particular programmes (input) with integration results (output). Cost-effectiveness became a key criterion when choosing programme contents and participants. The measurement of output is based on newly introduced IT systems for control available to all local employment agencies (Schütz, 2009). Thus, quantitative goals are set for each local office taking into account special problems of regional labour markets (Eichhorst et al., 2008). The core element in the new organization of the local FEO was the conversion of the formerly hierarchically organized employment offices into service centres (Kundenzentrum). The new organizational design is based on case management and has been closely related with the goal of reducing caseloads. The profiling of jobseekers constitutes an essential element of the new placement process, which is controlled centrally. It serves as a tool for customer segmentation and the determination of individual assistance, but also as an instrument for the allocation of resources. Hence, market clients (a) are considered to be job ready and have the best chances of finding employment. Clients for counselling and activation (b) are adjudged as needing minor adjustments of skills through limited training. Clients for counselling and qualification (c) need more attention and are likely to be directed to training programmes and other measures to increase flexibility. Incentive assistance clients (d) require special attention because they face the poorest re-employment prospects and bear the risk of long-term unemployment. The assignment to one of these four categories determines participation in particular action programmes and resource allocation, namely quick and sustainable placement (a), a change of perspective (b), reduction of employment barriers and qualification measures (c) and preservation of marketability and case management (d) (Eichhorst et al., 2008; Oschmiansky, 2010). The individual targeting of active measures includes the active cooperation of the person seeking assistance. At the end of the assessment

Tomáš Sirovátka

74

Fragmented Governance Continued: the German Case

process an integration agreement between the caseworker and the jobseeker is signed, specifying individual integration objectives. Since 2006, integration plans must be renewed every six months, three months in the case of younger claimants. The placement process is monitored and finally controlled along process ratios within the local agencies. Beyond the econometrically measured effects on the individual reintegration chances, the monitoring results from the centrally organized controlling system providing information on the achievement of the agency’s operational goals (Eichhorst et al., 2008). Marketization Processes of privatization or contracting-out were traditionally of minor importance in Germany, because the provision of labour-market programmes was always conducted by third parties, mostly by private, non-profit, third-sector organizations while private for-profit organizations were not very important (Trampusch, 2002; Dingeldey, 2009). The only exception to that rule was the public monopoly for job placement services, which was abolished in 1994. At the end of the 1990s, job placement conciliation through private organizations, however, still made up a share of only 5 per cent of all registered placements (Konle-Seidl, 2003). In line with the ideas of NPM, competition between public and private providers was increased and elements of marketization were introduced in order to empower the unemployed in their role as clients. In line with these ideas, contracting out of placement services to non-public agencies was enforced and vouchers, both for placement services and training measures, were introduced in 2002. The effect of these transformations may be regarded as limited in relation to both market share and effectiveness (Oschmiansky, 2010). Since 2002 the FEO has been able to commission the delivery of the placement service in total to third parties, namely private providers5 (Kaps, 2009b). In 2009 this was used for 213,978 cases (entrants, whole year), about 7 per cent of all unemployed UB I and UB II benefit recipients (Bundesagentur für Arbeit, 2010a, own calculation). While the contracting-out of placement services was aimed at promoting competition among private agencies for contracts, the voucher system was intended for the promotion of competition between private and public services. Furthermore, it was intended to increase self-responsibility and choice and strengthen the role of the unemployed as customers (see below).

Tomáš Sirovátka

Irene Dingeldey

75

In order to strengthen the use of temporary employment as a placement instrument, Personnel Service Agencies (PSAs) were set up as publicly funded but privately organized temporary work agencies. The creation of local PSAs was subject to tendering and regulated by contract. As private service providers, PSAs receive payments from the public employment agency in exchange for employing jobseekers as temporary workers (Kemmerling and Bruttel, 2006). A more intense focus on the compensation of individual integration deficits led to a re-regulation that introduced the individual placement budget in 2009 (§ 46 SGB III). This joined various instruments of mobility aids, that is, travel expenses for job interviews, daily travel costs or relocation assistance in line with the acceptance of a new job, in order to make access less complicated (Bundesagentur für Arbeit, 2010b). Together with the centralization of the purchase of labour-market services, the introduction of vouchers for training measures curtailed the capability of local AfWs to design training measures in response to short-notice demands. This was combined with an increased control of the quality of suppliers as well as outcomes with respect to the final labour-market integration of clients (Schütz, 2009). For UB II recipients the dominant form of employment promotion is the so called 1-Euro-job, which means working for benefit plus a supplement. Although this programme created strong controversies because of the supposed substitution of regular work, it was used extensively by all the governing institutions of the SGB II. Guidelines for the use of these programmes were elaborated by the above-mentioned newly created networks of local labour-market actors, mostly designed as advisory boards of the ARGE (Kaps, 2009a). The municipal employment provider (Beschäftigungsgesellschaften), created by municipalities in order to integrate hard-to-place SA recipients in employment, still plays a dominant role in implementing the 1-Euro-jobs policy. In the founding contracts of the ARGE it usually has a particular role. In the opted-out municipalities it is even more important, as relevant parts of service provision are often delegated to it (Kaps, 2009a). Hence, marketization within the area of employment promotion for UB II recipients may be regarded as limited.

Effects of governance reforms Besides controlling and monitoring, evaluation of policy interventions is part of effective governance within the NPM approach. In Germany

Tomáš Sirovátka

76

Fragmented Governance Continued: the German Case

the Hartz legislation made allowance for this demand, requiring evaluation research of single labour-market instruments and of the performance of the different implementation institutions of SGB II, namely the ARGE and municipal providers. Measurement was orientated towards the global objectives formulated in the legislation itself, namely integration into the labour market, improvement of employability and social stabilization of clients (§1, section 1, SGB II). In the following section I will discuss only evaluation results concerning process, outcome and output effects (see Chapter 1) related to the governance structures discussed above, namely particular elements of marketization as well as effects concerning the reform and organizational differences of provider institutions of SGB II.

Outcome effects With respect to the general labour-market performance we have seen a considerable decrease in the numbers of registered unemployed people since 2005, until the worldwide financial crisis of 2008. Apart from a positive macroeconomic development, this may to some extent be explained by increased job-search activities by the unemployed motivated by the ‘threat’ of a transfer to UB II. In general, however, simulation studies show only marginal effects on labour supply (Eichhorst et al., 2008). And overall for the long-term unemployed, incentives to move from benefit receipt to work were as limited as the demand for their qualifications in the labour market. Thus – in spite of all the latest reforms – long-term unemployment6 in Germany was still 45.5 per cent in 2009. This was rather high compared with other European countries like Denmark (9.1), the UK (24.6) or France (32.7) (OECD, 2010). The results of the evaluation studies indicated that labour-market services provided by ARGEs were oriented strongly towards direct labourmarket integration, including a stricter practice concerning sanctions. In contrast, the municipal providers gave greater priority to social integration, enabling access to additional social services and the promotion of qualifications (Institut für Angewandte Wirtschaftsforschung, 2009). Thus in general, the ARGE performed better with respect to labour-market integration, in effect the end of means-tested benefit receipt, while the municipal providers were better at improving employability (Institut für Wirtschaftsforschung and Institut für Angewandte Wirtschaftsforschung, 2009).

Tomáš Sirovátka

Irene Dingeldey

77

Output effects The quality of private placement services was not assessed in general to be better than that of public ones. While there were some positive results for the placement services voucher concerning shortening the duration of unemployment, the third-party delegation did not have any significant positive effects in the first years (Kaps, 2009b). Due to marketization, however, the use of private placement services increased. In 2009, 56,455 persons (entrants, whole year), about 1.8 per cent of all unemployed UB I and UB II benefit recipients, asked for a voucher for private placement. Successful placement, however, was much lower (Bundesagentur für Arbeit, 2010a; own calculation). The importance of the PSAs, namely the use of agency work as a labour-market integration instrument, turned out to be much lower than initially intended by the legislator. In 2007, the number of PSA participants was below four thousand, of which less than 30 per cent entered the regular labour market (Bundesregierung, 2006a). Hence, even negative effects concerning the integration into the first labour market were shown for PSA participants (Bundesregierung, 2006b). Early evaluation studies showed an improvement in placement services within the FEO. This may be related both to an improved ratio, from 409 to 256 clients per placement officer in the area of SGB III due to the introduction and separate treatment of SGB II clients between 2004 and 2005 (WZB and Infas, 2006), and the reorganization of the placement process within the newly introduced customer centres. Overall, the profiling process created a creaming of labour-market promotion. ‘Incentive assistance clients’, considered as hard to place, scarcely got access to high-quality labour-market promotion measures7 (Schütz, 2009). This negative development was further enforced by the new financial structure concerning unemployment protection and employment promotion that continued to create incentives for the local employment agencies to deny employment promotion measures that would burden the integration budget of the insurance fund and to ‘wait’ until the hard-to-place clients passed on to UB II.8 In terms of the quality of activation measures, the marketization of further training programmes was evaluated in 2006 to have positive results concerning the integration of participants into the labour market (IZA/DIW/infas, 2006). It is, however, critically recognized that this effect may be due to the reduction of the overall number of participants and a ‘positive’ preselection of the participants. The suppliers of training measures need certification given by separate organizations, and only programmes that expect a re-integration of 70 per cent of

Tomáš Sirovátka

78

Fragmented Governance Continued: the German Case

participants by six months after the end of the programme get public funding. This restrictive regulation, in conjunction with the introduction of the training voucher, is suspected of provoking a strategy of risk minimization from the side of the providers, which in consequence leads to a standardization of programme supply and a reduction in the quality of programmes (Bundesregierung, 2006b; Oschmiansky, 2010). With respect to innovation and programme quality, this development has been seen as ambivalent (Baethge-Kinsky, 2007) or even negative (Oschmiansky, 2010). Thus the service orientation in Germany is still criticized as being overly concerned with optimizing only the organization of the placement process. Hence the differentiation of services in line with profiling categories is supposed to create greater efficiency in labour-market integration. Within that framework the individual customers’ satisfaction is given low priority (Franck, 2006; Hielscher and Ochs, 2009). Process effects The evaluation of the different organizational models, namely ARGE and municipal providers (opted-out municipalities), is differentiated between a general case management approach and a specialized case management approach. In 2007 only one-third of all provider institutions used a general case management approach (for clients older than 259 ), while all the others applied the specialized case management approach in limiting it to the difficult-to-place clients. With respect to the case management approach, however, variations between the different organizational traditions were striking. While 77 per cent of the ARGEs used case management only for specialized groups, 81 per cent of municipal providers used a case management approach for all clients (Institut für Angewandte Wirtschaftsforschung, 2009). When case management was practised, placement services were included to a major extent (65 per cent of all institutions; ARGE 65 per cent; municipal provider 61 per cent).10 In contrast, benefit provision continued to be organized as a separate task. Hence 93 per cent of all institutions did not integrate benefit provision with case management, and the practice of integration was even lower for the ARGE (6 per cent) than for municipal providers (17 per cent). Furthermore, the networking with other social services like drug and debt counselling or childcare services was performed better by municipal providers (Institut für Angewandte Wirtschaftsforschung, 2009). It is therefore concluded that the placement process is much more individualized within municipal providers than within the

Tomáš Sirovátka

Irene Dingeldey

79

ARGE organizations (Institut für Angewandte Wirtschaftsforschung, 2009). Some elements of marketization may have increased choice concerning specialized services, but also created new problems. For example, a lack of market transparency creates difficulties for jobseekers in finding the most suitable private service and constrains their ability to act like customers. This is compounded since the FEOs are not allowed to advise jobseekers on the quality of private placement agencies (Kaps, 2009b). Overall, the development of rights, obligations and responsibilities in this organizational context must be regarded as dispersed. For the short-term unemployed, a high level of social protection, employment promotion and service quality has been maintained, while benefit cuts, tightened conditionality and a higher flexibility concerning sanctions have been applied to UB II recipients (Eichhorst et al., 2008). Thus, a dualization of not only governance structures but also of social rights was produced (Dingeldey, 2010).

A shift towards a new governance regime? In the German case the question concerning a shift towards a new governance regime according to the various types discussed in Chapter 1 may be even more difficult to answer than for other countries. The traditional unemployment insurance regime provided a status-oriented unemployment benefit that was based on procedural governance. The merger of the former UA, a benefit strongly linked to the unemployment insurance system, with SA created a flat-rate basic income scheme that tended to develop into a universal working-age benefit. Due to the abolition of status protection for former UA recipients, the fierce conditionality of benefits and the expansion of work-for-benefit programmes as employment promotion measures, the installation of the new regime was considered as a pathbreaking change or paradigm shift (Oschmiansky et al., 2007; Mohr, 2008; Knuth, 2009). The question concerning the emergence of a new governance regime, therefore, has to be answered in the light of a dualization of the German unemployment protection system (Dingeldey, 2010), with respect to both the new regime and changes within the old, insurance-based regime. The old, insurance-based system was overwhelmingly characterized by continuity, although some changes within the procedural governance regime were implemented; the principle of elements of selfgovernance via tripartite corporatism at the management level was partly withdrawn, although not generally questioned (Klenk, 2009b).

Tomáš Sirovátka

80

Fragmented Governance Continued: the German Case

Old forms of rule-based governance were combined with NPM elements such as management by objectives. Output- and efficiency-oriented goals, however, are still implemented within standardized forms of procedural governance. As control now refers to both output targets and processes, compliance with formal regulation is even stronger than before. Furthermore, stronger uniformity of placement models and elements of marketization curtailed the scope of action for local employment agencies and may be interpreted as re-centralization (Schütz, 2009). Elements of decentralization, however, can be seen in an increase of budget flexibility given to local AfWs (Schütz, 2009). The newly created basic income regime is governed differently: it is entirely tax financed, mixing federal and municipal budget resources. Furthermore, in the management of the local implementation institutions we see the emergence of network governance through the voluntarily created advisory boards replacing the formal regulation of tripartism (Klenk, 2009b). Beyond these general features, there are questions over the extent to which the implementation structure of UB II can be characterized by a single governance regime, when its main feature is the fragmentation of responsibilities concerning both the area of service provision and financial support of the clients between the federal employment agency and the municipalities. Furthermore, the organizational variety between – and even within – the implementing institutions is huge (Knuth, 2009). Overall, we may therefore conclude that the dominant feature of the entire German governance regime of labour-market activation seems to be the continued fragmentation of responsibilities and the mix of governance forms.

Notes 1. Like its chairman, most of the 15 members of the Hartz Commission were representatives from the business sector. The different groups of representatives of the social partners, of government and public administration and scientific experts had only a few places each (Schmid, 2003). 2. It was also used by UB and UA recipients who needed additional support (as a top-up), because the payments of the respective benefits were too low to maintain the household. 3. SA therefore remains as a residual system to cover needy persons not capable of working. 4. In order to avoid the overburdening of municipalities, a share of housing cost that may not exceed 49 per cent of total costs (calculated by a complicated formula reflecting the number of households receiving UB II) is financed out of the federal budget (SGB II § 46).

Tomáš Sirovátka

Irene Dingeldey

81

5. Earlier versions of partial commissioning had already been introduced in 1998 (Kaps, 2009b). 6. Persons unemployed for more than 12 months as a percentage of all unemployed. 7. As a reaction to this practice of exclusion of incentive assistance clients from the ordinary measures of employment promotion, between 2005 and 2008 additional programmes for the hard-to-place were established (Oschmiansky, 2010). 8. This happened between 2005 and 2008, when the FEO avoided costs for the labour-market promotion of the hard-to place unemployed in order to pay only a penalty tax to the ARGE claimed when clients passed from UB I to UB II (Eichhorst et al., 2008). 9. For clients younger than 25, the general case management approach was more often used within all institutional forms. 10. In the other institutions case managers are not responsible for processes concerning the integration of clients into the first labour market, but may arrange access to employment promotion measures.

References 6, P. (1998) ‘Problem-Solving Government’, in I. Hargreaves and I. Christie (eds) Tomorrow’s Politics: The Third Way and Beyond (Trowbridge: Demos). 6, P., D. Leat, K. Seltzer and G. Stoker (2002) Towards Holistic Governance: The New Reform Agenda (Houndmills/New York: Palgrave). Baethge-Kinsky, V. (2007) ‘Transformation statt Erosion arbeitsmarktpolitischer Qualifizierung – Lehren aus der Evaluation der Hartz-Gesetze’, WSIMitteilungen, 60/6, 312–9. Botzem, S., J. Hofmann, S. Quack, G. Schuppert and H. Strassheim (eds) (2009) Governance als Prozess: Koordinationsformen im Wandel (Baden-Baden: Nomos). Brussig, M. and M. Knuth (2009) ‘Alternative styles and roles for local governments’, in F. Larsen and R. van Berkel (eds) The New Governance and Implementation of Labour Market Policies (Copenhagen: DJOeF). Bundesagentur für Arbeit (2010a) ‘Der Arbeitsmarkt 2009’, Amtliche Nachrichten der Bundesagentur für Arbeit, 58/2. Bundesagentur für Arbeit (2010b) Der Arbeitsmarkt in Deutschland: Jahresrückblick 2009 (Nürnberg: Bundesagentur für Arbeit). Bundesregierung (ed.) (2006a) Bericht 2005 der Bundesregierung zur Wirkung der Umsetzung der Vorschläge der Kommission Moderne Dienstleistungen am Arbeitsmarkt. Drucksache 16/505 (Berlin: Deutscher Bundestag). Bundesregierung (ed.) (2006b) Bericht 2006 der Bundesregierung zur Wirkung der Umsetzung der Vorschläge der Kommission Moderne Dienstleistungen am Arbeitsmarkt (Berlin: Deutscher Bundestag). Clasen, J. (2005) Reforming European Welfare States – Germany and the United Kingdom Compared (Oxford: Oxford University Press). Deutscher Bundestag (2008) Bericht zur Evaluation der Experimentierklausel nach § 6c des Zweiten Buches Sozialgesetzbuch. Drucksache 16/11488 (Berlin: Deutscher Bundestag).

Tomáš Sirovátka

82

Fragmented Governance Continued: the German Case

Dingeldey, I. (2007) ‘Between workfare and enablement: the different paths to transformation of the welfare state’, European Journal of Political Research, 46/6, 823–51. Dingeldey, I. (2009) ‘Changing forms of governance as welfare state restructuring: Activating labour market policies in Denmark, the UK and Germany’, in I. Dingeldey and H. Rothgang (eds) Governance of Welfare State Reform: A Cross National and Cross Sectoral Comparison of Policy and Politics (Cheltenham: Edward Elgar). Dingeldey, I. (2010) ‘Activating labour market policy in a “Bismarckian” welfare state – Old and new divisions of social rights and citizenship in Germany’, in A. Neville (ed.) Human Rights and Social Policy: A Comparative Analysis of Values and Citizenship in OECD Countries (Cheltenham/Northampton: Edward Elgar). Dingeldey, I. (2011) ‘Germany: Moving towards integration while maintaining segmentation’, in J. Clasen and D. Clegg (eds) Unemployment Protection Systems and Labour Market Change in Europe (Oxford: Oxford University Press). Eichhorst, W., M. Grienberger-Zingerle and R. Konle-Seidl (2008) ‘Activation policies in Germany: From status protection to basic Income Support’, in W. Eichhorst, O. Kaufmann and R. Konle-Seidl (eds) Bringing the Jobless into Work? Experiences with Activation Schemes in Europe and the US (Heidelberg: Springer-Verlag). Esping-Andersen, G. (1990) The Three Worlds of Welfare Capitalism (Cambridge: Polity Press). Finn, D., M. Knuth, O. Schweer and W. Somerville (2005) Reinventing the Public Employment Service: The Changing Role of Employment Assistance in Britain and Gemany (London: Anglo-German Foundation for the Study of Industrial Society). Franck, M. (2006) ‘Zur Qualität der Arbeitsvermittlung der BA: Kundenzufriedenheit durch Dienstleistungsquaität oder Dienstleistungsqualität durch Kundenzufriedenheit?’, Soziale Sicherheit, 55/4, 110–15. Goerne, A. (2010) Personalisation of Active Labour Market Policies? From Social Assistance to Unemployment Benefit II, unpublished manuscript 16/06/2010. Hartz, P. (2002) ‘Der Bericht der Kommission’, Soziale Sicherheit, 51/8–9, 254–60. Hielscher, V. and P. Ochs (2009) ‘Das prekäre Dienstleistungsversprechen der öffentlichen Verwaltung’, in S. Bothfeld, W. Sesselmeier and C. Bodegan (eds) Arbeitsmarktpolitik in der sozialen Marktwirtschaft – Vom Arbeitsförderungsgesetz zum Sozialgesetzbuch II und III (Wiesbaden: VS Verlag für Sozialwissenschaften). Institut für Angewandte Wirtschaftsforschung (2009) ‘Organisatorische Umsetzung des SGB II vor Ort – Ausgewählte Ergebnisse aus Untersuchungsfeld 1 der §-6c-SGBII-Evaluation’, IFO Schnelldienst, 62/1, 22–30. Institut für Wirtschaftsforschung and Institut für Angewandte Wirtschaftsforschung (2009) ‘Makroanalyse und regionale Vergleiche (Untersuchungsfeld 4) in Rahmen der Evaluation der Experimentierklausel nach §-6c-SGBIIEvaluation’, IFO Schnelldienst, 62/1, 40–57. IZA/DIW/infas (2006) Evaluation der Maßnahmen zur Umsetzung der Vorschläge der Hartz-Kommission. Modul 1b: Fördrung Beruflicher Weiterbildung und Transferleistungen (Bonn: IZA). Kannengießer, C. and E. Gundel (2003) ‘Der Reformprozess der Bundesanstalt für Arbeit – vom bürokratischen Monolith zum Dienstleister am Markt’, Sozialer Fortschritt, 52/8, 207–13.

Tomáš Sirovátka

Irene Dingeldey

83

Kaps, P. (2009a) ‘Die Rolle der Kommunen in der Arbeitsmarkt- und Beschäftigungspolitik’, in S. Bothfeld, W. Sesselmeier and C. Bodegan (eds) Arbeitsmarktpolitik in der sozialen Marktwirtschaft – Vom Arbeitsförderungsgesetz zum Sozialgesetzbuch II und III (Wiesbaden: VS Verlag für Sozialwissenschaften). Kaps, P. (2009b) ‘Marketization of placement services as a double recommodification of labour’, in F. Larsen and R. van Berkel (eds) The New Governance And Implementation of Labour Market Policies (Copenhagen: DJOeF). Kemmerling, A. and O. Bruttel (2006) ‘New politics’ in German labour market policy? The implications of the recent Hartz reforms for the German welfare state’, West European Politics, 29/1, 90–112. Kirsch, J., M. Knuth, G. Mühge and O. Schweer (eds) (2009) Können wir nicht einfach gute Freunde bleiben...?Getrennte Aufgabenwahrnehmung in der Grundsicherung für Arbeitsuchende bietet keine Zukunft. IAQ – Report 2009–04 (Essen Duisburg: Universität Essen Duisburg). Klenk, T. (2009a) ‘Die korporatistische Arbeitsverwaltung’, Aus Politik und Zeitgeschichte, 27, 34–9. Klenk, T. (2009b) ‘Vom Arbeitsförderungsgesetz zum Sozilagsetzbuch II und III: Pfadwechsel in der korporatistischen Arbeitsverwaltung?’, in S. Bothfeld, W. Sesselmeier and C. Bodegan (eds) Arbeitsmarktpolitik in der sozialen Marktwirtschaft – Vom Arbeitsförderungsgesetz zum Sozialgesetzbuch II und III (Wiesbaden: VS Verlag für Sozialwissenschaften). Knuth, M. (2009) ‘Grundsicherung “für Arbeitsuchende”: ein hybrides Regime sozialer Sicherung auf der Suche nach seiner Governance’, in S. Bothfeld, W. Sesselmeier and C. Bodegan (eds) Arbeitsmarktpolitik in der sozialen Marktwirtschaft – Vom Arbeitsförderungsgesetz zum Sozialgesetzbuch II und III (Wiesbaden: VS Verlag für Sozialwissenschaften). Konle-Seidl, R. (2003) ‘Von der Anstalt zum ersten Dienstleister am Arbeitsmarkt? Möglichkeiten und Grenzen der Reform der Bundesanstalt für Arbeit’, Gesundheits- und Sozialpolitik, 1/2, 22–9. Mohr, K. (2008) ‘Creeping Convergence – Wandel der Arbeitsmarktpolitik in Großbritannien und Deutschland’, Zeitschrift für Sozialreform, 54/2, 187–207. Mosley, H. (2009) ‘Decentralisation and local flexibility in employment service’, in F. Larsen and R. van Berkel (eds) The New Governance and Implementation of Labour Market Policies (Copenhagen: DJOeF). OECD (2010) Employment Outlook: Moving Beyond the Jobs Crisis (Paris: OECD). Oschmiansky, F. (2010) ‘Aktive Arbeitsförderungsinstrumente seit Einführung des SGB III: Rückblick und Ausblick’, Sozialer Fortschritt, 1, 16–23. Oschmiansky, F., A. Mauer and K. Schulze (2007) ‘Arbeitsmarktreformen in Deutschland – Zwischen Pfadabhängigkeit und Paradigmenwechsel’, WSIMitteilungen, 60/6, 291–7. Schmid, G. (2003) ‘Gestaltung des Wandels durch wissenschaftliche Beratung: Das “Bündnis für Arbeit” und die “Hartz-Kommission” ’, in S. Ramge and G. Schmid (eds) Managment of Change in der Politik? Reformstrategien am Beispiel der Arbeitsmarkt- und Beschäftigungspolitik. Ein Werkstattbericht (Münster/New York/München/Berlin: Waxmann). Schneider, H., W. Eichhorst and K. Zimmermann (2006) ‘Konzentration statt Verzettelung: Die deutsche Arbeitsmarktpolitik am Scheideweg’, Perspektiven der Wirtschaftspolitik, 7/3, 379–97.

Tomáš Sirovátka

84

Fragmented Governance Continued: the German Case

Schütz, H. (2009) ‘Neue und alte Regelsteuerung in der deutschen Arbeitsverwaltung’, in S. Bothfeld, W. Sesselmeier and C. Bodegan (eds) Arbeitsmarktpolitik in der sozialen Marktwirtschaft – Vom Arbeitsförderungsgesetz zum Sozialgesetzbuch II und III (Wiesbaden: VS Verlag für Sozialwissenschaften). Steffen, J. (2009) Sozialpolitische Chronik, http://www.arbeitnehmerkammer.de/ sozialpolitik/doku/02_politik/chronik/chronik_gesamt.pdf (accessed 1 November 2010). Trampusch, C. (2002) Die Bundesanstalt für Arbeit und das Zusammenwirken von Staat und Verbänden in der Arbeitsmarktpolitik von 1952 bis 2001 (Köln: Max Planck Institute). WZB and Infas (2006) Evaluation der Maßnahmen zur Umsetzung der Vorschläge der Hartz-Kommission, Modul 1a, Neuausrichtung der Vermittlungsprozesse, Ergebnisbericht 2005 (Berlin/Bonn: WZB).

Tomáš Sirovátka

5 Steering with Sticks, Rowing for Rewards: the New Governance of Activation in the UK Sharon Wright

Introduction The past decade has seen significant changes to both the content and governance of benefits and services for unemployed people in the UK. Between 1997 and 2010, Labour governments broke a new path following bearings set, to a large extent, by their Conservative predecessors. An extensive programme of welfare reform was justified primarily on the grounds of cost saving, in an uncompromising effort to ‘promote work as the best form of welfare’ (DWP, 2008a). This form of activation was designed for ‘moving people from welfare into work’ (DWP, 2003, p. 3), without a precise official definition of what this work-first approach meant (Lindsay et al., 2007; see below). What emerged was a series of reforms to social security law that offered relatively small carrots and threatened strong sticks in an explicit attempt to reduce the number of people claiming benefits. This focus on stimulating rates of job entry (often at the lower end of the labour market) also influenced the new forms of provider–contractor relationship that have evolved during several stages in the marketization of employment services. The relationships between different providers (in statutory, voluntary and private sectors) have been increasingly permeated by market values and competitive impulses, controlled by the central state through incentives that favour short-term employment placements. The UK has played an interesting role on the international stage of governing activation, appearing initially as a pioneer in applying competitive market forces to the administration of benefits and services for out-of-work adults, at first in transforming the civil service, then through incorporating incentive-based financial reward systems 85

Tomáš Sirovátka

86

Steering with Sticks, Rowing for Rewards

for providers in prototype Employment Zones (in 2000). After the first forays into marketization, new governance arrangements developed incrementally and the UK kept further backstage whilst countries like Australia, the Netherlands and Denmark transformed employment services more extensively. Currently, the UK has regained prominence in designing new contracts that are billed as being capable of avoiding the pitfalls experienced elsewhere (DWP, 2008a). However, concerns remain about the extent to which it is possible to avoid risks such as: questionable efficiency gains/cost savings; widespread creaming and parking of the most disadvantaged jobseekers; lack of innovation; and concentration on short-term, cheaper job placement techniques at the expense of longer-term training, education or support (Bredgaard and Larsen, 2008; Van Berkel and Borghi, 2008; Wright, 2008). 2010 marks a significant point in these ongoing developments, with work-first activation tested by recession in the unfamiliar political territory created by an incoming Liberal–Conservative coalition government at UK level. This chapter outlines the implications of the trends towards combining work-first activation with increasing marketization in the development of new modes of governance. First, the work-first emphasis of UK policy is discussed, before outlining current governance structures for the administration of social security for the unemployed and employment services. The main trends in the reform of the governance of activation are identified, highlighting processes of decentralization and (re-)centralization, marketization and contractualization. Finally, the effects of these reforms are considered.

A liberal work-first type? The UK approach to activation is not simply the type of liberal workfare model that is epitomized by the US (Handler, 2003). Rather, UK policy making has, if only to a limited extent, been informed by ethical (White, 2000) and client-centred (Lindsay and Mailand, 2004) approaches (Lindsay et al., 2007). The broader policy context has included more social democratic attempts to make work possible (with a National Childcare Strategy to increase the availability of affordable, high-quality childcare and the Child Tax Credit, which can help some parents with the cost of childcare) and to make work pay, with the introduction of the National Minimum Wage and Working Tax Credit to top up low wages (see Millar, 2002, 2009). On balance, however, the New Labour approach can be viewed as predominantly work first in character because of its overarching goal of increasing short-term job entry,

Tomáš Sirovátka

Sharon Wright

87

achieved through interviews to promote job search activities. An alternative human capital development approach would give greater weight to the development of qualifications and skills that could lead to more fulfilling long-term employment, with greater concern for retention and advancement (Lindsay et al., 2007).

Governing activation in the UK: an overview Perhaps surprisingly, the UK has retained a highly centralized system, in one form or another, for administering benefits and providing employment information, advice and job matching for more than a century. Although the past ten to fifteen years have seen significant changes to the content of policy, to the target group for employment services and in the governance of employment services, core functions for the majority of job-seeking benefit recipients have remained consistently within the control of central government, delivered within the realm of the civil service. In hindsight, the initial creation of the UK public employment service in 1909 was a centralization of an existing open-market mix of private and charitable labour bureaux (Showler, 1976, p. 21). It was this very early step that established the state monopoly over job matching that was met shortly afterwards (in 1911) by what was to be an irreversible involvement in insurance-based benefits. In long-term perspective, the UK norm was to provide integrated financial assistance and job search support to unemployed people. This situation continued for the majority of the twentieth century and was only reversed a generation ago, when the roles of benefit administration/policing and job matching were separated by the creation of Jobcentres in 1973 (Fletcher, 1997; Price, 2000). This was the context for the introduction of activation and new governance arrangements from the late 1980s onwards. Mass unemployment in the 1980s placed the system under immense pressure, and in 1987 the dual roles of unemployment benefit administration and job matching were again joined together. Despite these developments, it was not until 1990 that the first move was made away from traditional civil service monopoly, when the Employment Service was one of the first areas of civil service to be reformed into an executive agency (Price, 2000). As one of a new type of government Next Step agencies, this meant functioning semi-autonomously from central government, with greater operational freedom offered in exchange for new forms of financial accountability through management systems (led by a chief executive) and performance measured against indicators

Tomáš Sirovátka

88

Steering with Sticks, Rowing for Rewards

(Price, 2000). Most policy making continued to be done at government department level, with the Employment Service mainly concerned with matters of implementation. The new Employment Service was at the vanguard in applying business principles to public service practice and values (Horton and Jones, 1996; Foster and Hoggett, 1999) and cost cutting came to be a key priority. It was not long after this first flush of managerialism that marketization also began to creep in. In practice, elements of marketization were evident in the early 1990s, but not extensive – involving market testing, contracting-out (mainly of training services) and cost reviews. These first experiments with using the private sector to provide training for unemployed people created a degree of service fragmentation and were considered by some commentators as amounting to quasi-privatization (Tonge, 1999), but in terms of marketization this step barely registered on the scale in comparison with the developments of the late 1990s and early twenty-first Century. By then, it was the delivery arrangements for New Labour’s welfare-to-work policies (starting in 1997) that developed substantial roles for non-state agents in the direct delivery of employment services (see below). Organizationally, between 1990 and 2002, there were separate central government departments1 and networks of local offices for dealing with social security claims (latterly named the Benefits Agency) for all those who were not required to look for work, in parallel with a network of local Jobcentre offices (which dealt with benefits and job-search assistance for unemployed people). In 2002, these separate agencies were merged to form Jobcentre Plus, as a one-stop shop for the delivery of benefits to working-age adults, with increased work-related conditionality for non-Jobseeker’s Allowance claimants, including inactive groups such as lone parents and ill or disabled people who had previously been exempt from work-related conditionality. Since then, all income maintenance has been administered centrally, except for housing-related benefits, which have always been administered by local authorities. This means that, unlike many European countries, local authorities (the equivalent of municipalities) do not have any role in administering core benefits or employment services for people out of work. This new governance of social security and employment services coincided with the advent of activation. It was the late 1990s that marked a watershed in the UK approach to dealing with unemployment. The outgoing Conservative government implemented the controversial Jobseeker’s Allowance (JSA) Act just before leaving office in 1997. Blair’s

Tomáš Sirovátka

Sharon Wright

89

New Labour party had condemned the harshness of the new jobseeker regime on several grounds whilst in opposition. They argued that introduction of JSA undermined the principle of earned entitlement. Similarly, there was an increase in the proportion of unemployed people claiming means-tested benefits. Young people were particularly disadvantaged by JSA because the value of the benefit was reduced for 18–24-year olds, and 16- and 17-year olds were excluded from entitlement entirely (except for exceptional cases of severe hardship). In addition to this, there were increased obligations for unemployed people to prove they were actively seeking work, which were policed more closely in accordance with new back-to-work plans (Jobseeker’s Agreements). There were also tough penalties for non-compliance with procedures. On gaining power, however, Blair’s New Labour party revised their views and moved towards increasing conditionality and compulsion, creating an ever harsher regime than the one they previously criticized. Job preparation and training services were expanded and fragmented rapidly through the multiple New Deal programmes, which operated via local partnership arrangements (Theodore and Peck, 1999). Similarly, there was incremental development of a range of small area-based initiatives, which pushed the boundaries of marketization, promoted decentralization and incorporated inter-agency working (see below). An intensive phase of reform was undertaken by successive Labour governments (1997–2010) and, in 2010, the Department for Work and Pensions was engaged in a major project to develop a UK-wide quasimarket in employment services. The clearest statements of the intended governance and delivery arrangements for this new way of working were set out in the Department for Work and Pensions (DWP) Commissioning Strategy (2008a) and the welfare reform white paper ‘Raising expectations and increasing support: reforming welfare for the future’ (DWP, 2008b). Policy makers have consciously adjusted the role of government in the field of social security and employment services from rowing to steering. The, then, Secretary of State for Work and Pensions, James Purnell, who oversaw these changes, clearly viewed the desired effect as moving ‘away from a mindset of top-down control into a world where we devolve initiative and innovation’ (2008b, Ministerial Foreword). Similarly, the DWP Commissioning Strategy (2008a, p. 7) presented the department as a ‘key enabler’ in ‘delivering the Government’s ambition of becoming a world leader in skills and tackling concentrations of worklessness at a local level’. The rowing is mainly to be done by an army of frontline workers in private and voluntary sector agencies, with

Tomáš Sirovátka

90

Steering with Sticks, Rowing for Rewards

preferential prime contractor positions available for large and successful private companies (including multi-national corporations). Helping people to find work has become highly profitable for some private agencies. Prime contracts have been won mainly by large for-profit agencies, which subcontract local delivery to a range of local providers. In the case of the Flexible New Deal 1, this amounted to 179 public (including local authorities), private, voluntary sector and mixed-sector organizations, many of which were already providing employment services in their localities. There has been criticism of this contracting model, including concern over the ethics of private companies profiting from the misfortune of vulnerable people (Piggott and Grover, 2009). The implications of these changes are far reaching for all concerned. These new delivery arrangements have been laid on top of pre-existing principles for the provision of social protection. There has been no fundamental challenge to the prevailing character of UK intervention, which is currently less generous than at any time in the post-war period, favouring residual, stigmatizing and punitive approaches to the support of adults who leave the labour market (even when, in the case of JSA claimants, this exit was involuntary). Broadly, benefits have become increasingly conditional upon active job search requirements for a widening client group, with only minimal narrowly defined forms of human capital development (Lindsay et al., 2007). Cost cutting has been a persistent concern for successive governments, increasingly achieved through incentives and directives to monitor and control the behaviour of advisers and benefit recipients. The timing of these developments is also interesting, because the construction of benefit dependents as a drain on national resources intensified during a period when economic conditions had allowed for a relatively consistent spell of low unemployment (around 5 per cent) over the first decade of New Labour power (1997–2008); such low unemployment had been achieved only once before in UK history during the immediate post-war period (Nickell, 1999). Perhaps even more significantly, UK employment rates were exceptionally high in both historic and international comparison in the period preceding the global economic downturn in late 2008. Welfare-to-work policies were not, therefore, as might easily be assumed, a response to high unemployment. Quite the reverse, they were conceived at a time of unusually high employment. Not only did the employment rate remain extremely high, but it peaked at 74.9 per cent in the first quarter of 2008 (National Statistics, 2008), making the UK one of only a handful of EU and OECD countries to achieve such success.

Tomáš Sirovátka

Sharon Wright

91

Work-first policy proposals and quasi-market delivery arrangements were conceived of during a stable period of healthy economic conditions, but have been implemented in an adverse environment. This means that policy makers have chosen to base reforms on a model of human action that emphasizes individual behavioural explanations for unemployment at a time when economic projections suggest that the global downturn will result in less secure employment possibilities for a substantial proportion of citizens. Added to this are ongoing changes to eligibility conditions for other benefits, including the abolition of social assistance in the form of Income Support and the introduction of the Employment Support Allowance, which will re-allocate benefit recipients (particularly lone parents and ill and disabled people) to Jobseeker’s Allowance. The end product is that unemployment (measured by both ILO and claimant count) are set to rise at a time when benefit recipients are charged with greater responsibility for the circumstances in which they find themselves and are required to take specified action to find jobs. The following sections explore the trends in the development of welfare governance in the UK (summarized in Table 5.1) that have led to this current situation.

Table 5.1 Implementation structure for income protection and activation in the UK (late 1980s and 2010) Late 1980s

2010

Unemployment Insurance/benefit administration

Jobcentres/Department of Social Security (later Benefits Agency)

Jobcentre Plus

Social assistance administration

Department of Social Security (later Benefits Agency)

Jobcentre Plus

Activation of insured

Jobcentres

Jobcentre Plus

Various programmes

Various programmes delivered by a range of agencies

Manpower Services Commission Activation of uninsured

No compulsory measures

Jobcentre Plus (via Work-ocused interviews) Various programmes delivered by a range of agencies.

Tomáš Sirovátka

92

Steering with Sticks, Rowing for Rewards

Main trends in governance reforms Governance reforms in the field of social security and employment services mainly took place from the 1990s onwards, intensifying in the early twenty-first century. As in the Netherlands (Van Berkel, 2010), the speed of change was rapid and gained momentum in the absence of evidence that the chosen methods would result in the desired effects (Bredgaard and Larsen, 2008). In the UK at least, the motivations for reform were clear from the outset, with explicit goals of cost cutting by reducing entitlement (alongside goals such as the eradication of child poverty) and introducing competition to increase efficiency, reinforced by an ideological commitment to end so-called benefit dependency (Freud, 2007). Paradoxically, policy makers’ increased faith in market mechanisms came at a time when the foundations of the economy were shaken by market failure in global finance, which led to rises in unemployment not witnessed in a generation. Rather than leading to a reflection on the ability of competitive forces to deliver improvements in well-being for vulnerable citizens, the recession of 2008 appears to have redoubled efforts to cut costs. There is still little evidence that independent agencies operate any more efficiently or effectively than public bureaucracies in assisting people to find employment. In fact, since the majority of jobseekers return to employment without contact with private providers (Harker, 2006), there can be little justification for the routine dismissal of bureaucracy as synonymous with inefficiency in the field of employment services (Davies, 2008). This section explores the main trends in governance reforms, in relation to processes of de- and re-centralization, marketization and contractualization. Strong central control, decentralization and (re-)centralization In the UK, there is a legacy of central control over political processes and policy making in the field of social security and employment services. This is important because it constrains action at local level, which is mediated by standardized Department for Work and Pensions policies, priorities and procedures. This provides the context for local delivery, which, although delivered by a range of subcontracted agencies for certain programmes, is controlled strongly via contract compliance and standardized delivery frameworks. Centralized policy making leaves little room for the development of innovative or localized responses. This means that central control coexists alongside ongoing developments in the decentralization of service delivery and simultaneous re-centralization through contract control, targets,

Tomáš Sirovátka

Sharon Wright

93

incentives and outcome payments. These seemingly conflicting processes actually appear to be mutually reinforcing in some instances. First, a strongly controlled central political and policy-making state has survived, despite devolution in 1999 to Scotland and Wales. Secondly, a form of decentralization is evident in the development of executive agencies like the Employment Service in the 1990s and the evolution of the quasi-market in employment services from the late 1990s through to 2010, which has fragmented delivery arrangements and propagated a new amalgam of agencies resting between the citizen and the state. Finally, re-centralization has undoubtedly occurred in two dimensions: (a) through the organizational merger of the Benefits Agency and Jobcentres into a one-stop-shop, Jobcentre Plus, in 2002; and (b) the compliance and control measures for managing the work of frontline advisers via standardized priorities and procedures. The following sections explore these processes in greater depth. Centralized control One distinctive aspect of the governance of activation in the UK, in contrast to many other European countries, is the unusual power of the central state to mandate change. UK political systems, organizational structures and policy-making practices all reinforce powerful centralizing tendencies, with limited roles for sub-national or local governments, social partners or interest groups (see Barlow and Röber, 1996). This means that the Westminster government has retained a long-term ability to dictate the content and delivery mechanisms for policy in the field of social security and employment services that is almost unparalleled among Western democracies. There are relatively few veto points in the UK system and barely any meaningful negotiations or compromises to be reached with interested parties. These conditions allow for radical changes to be implemented quickly, with limited resistance. This position has survived the ideological challenges of neo-liberalism during the 1980s and political devolution in the twenty-first century – the new administrations in Scotland and Wales have no jurisdiction over this policy sphere. The Department for Work and Pensions is one of the richest and most powerful departments in the UK government, within which decisions about the content and governance of social security and employment services have developed in relative isolation. This central mandate for reform is also related to the limited role of social partners in political or policy-making processes. This absence of social partners from debate or delivery means that the boundaries of consultation are tight and can be expected to defer to employers’

Tomáš Sirovátka

94

Steering with Sticks, Rowing for Rewards

interests (which have been disproportionately represented through forums such as the National Employment Panel) rather than the concerns of employees or unemployed people (see Finn et al., 2005). However, even without such exclusive conditions for policy making, there now appears, in any case, to be a strong cross-party consensus (amongst the three main political parties: Conservatives, Liberal Democrats and Labour) on the main principles of reform in relation to: increases in work conditionality for all claimant groups (Theodore and Peck, 1999); defining unemployment, its causes and politically acceptable solutions; the use of competitive market forces and contracting-out (see below). Decentralization The story of the governance of activation in the UK is not, however, just about the strong central state. A parallel process of decentralization has been evident, especially since the constitution of the Employment Service in 1990 as a semi-autonomous executive agency, organizationally distinct and distanced from the traditional structures and practices of the central state. This first phase of decentralization was connected tightly with the application of new public management practices (including incentive management) as an alternative to state bureaucracy rather than devolving decision-making, since the Employment Service (like its successor, Jobcentre Plus) remained tightly controlled by central government (Horton and Jones, 1996; Clarke and Newman, 1997). The purpose of decentralization was primarily to save money by restructuring, removing a tier of management and changing frontline practice in line with private sector values (Clarke and Newman, 1997). Although this represents a significant form of decentralization, the extent of the process was limited by the continuing central control over standardized approaches, design of policy instruments and the dictation of performance measurement priorities and techniques (which focused on monitoring and rewarding short-term job outcomes). The second form of decentralization emerged most concertedly with the evolution of the quasi-market in employment services from the late 1990s onwards. This fragmented and marketized delivery arrangements, but again reinforced the role of the central state in setting the terms of contractual engagements. Contracting-out was developed within particular programmes, especially for long-term unemployed people claiming the means-tested, assistance version of benefit (Income-based Jobseeker’s Allowance). This began to intensify with the introduction of partnership arrangements for New Deal programmes (from 1998

Tomáš Sirovátka

Sharon Wright

95

onwards). New Deal programmes were designed at UK level by the New Deal Task Force (which included representatives from private, public and voluntary sector agencies, employers and trade unions). The details of delivery arrangements varied amongst the 144 delivery units, with the Employment Service (from 2002 Jobcentre Plus) working with local agencies from all sectors (including, for example, local authorities and colleges) to develop annual ‘District Delivery Plans’ (Finn, 2009). Separate programmes existed for target groups including young people (aged 18–24, unemployed for six months plus), adults (aged 25 and over), lone parents and disabled people. These programmes were designed at national level, but administered locally via a two-tier system beginning with series of gateway interviews at a Jobcentre (later Jobcentre Plus) office, after which referrals were made to one of a number of available options, such as paid employment or training. This process established the expansion of contractualization and decentralization in terms of the large-scale engagement of non-state actors. This trend developed with the introduction of prototype areabased Employment Zones targeted at specific local areas of high long-term unemployment in 2000. These involved extensive use of private providers and rewarded success (counted as 13-week job placements) with outcome payments. This initiated the trend, now embedded, of the acceptability of profiting financially from job-matching services for benefit recipients. Similar area-based initiatives, such as Working Neighbourhoods Pilots, increased the engagement of private providers and increased financial rewards for job placements in particular neighbourhoods. Overall, though, the use of private agencies was seen as supplementary to the core centralized and standardized provision, because most unemployed people continued to find employment before being referred to a private provider. Following the Freud Review (2007), large-scale decentralization was introduced in a UK-wide quasi-market of employment services, establishing major players in prime contractors and subcontracting arrangements for the delivery of the Flexible New Deal. The Flexible New Deal replaced all previous New Deal programmes that differentiated between user groups (for example, lone parents, disabled people, young people, older people and the longest-term unemployed). Participants engaged in a series of interviews with Jobcentre Plus advisers for the first phase of the programme, before being referred to contracted-out providers in later phases. This was the centrepiece of the new quasi-market, and yet the role for non-state actors in dealing face-to-face with benefit recipients was still confined to long-term unemployed after one year of

Tomáš Sirovátka

96

Steering with Sticks, Rowing for Rewards

claiming benefits. Until that point, Jobcentre Plus retained the majority of contact with jobseekers through a series of standardized interviews at different stages of the claim. Interestingly, however, Jobcentre Plus has also been reformed to minimize face-to-face contact with benefit recipients. Decentralization has not come full circle, since the central state has retained and strengthened its role in dictating every level of reforms from the overarching policy goals, through the governance arrangements to the precise instruments used by frontline advisers. Even the most right-wing thinkers have not suggested a complete withdrawal of government intervention in assisting people into employment. In this sense, central government intervention in employment transitions (especially those of the lowest-paid and most vulnerable workers) is more or less universally accepted. This is not simply a question of ensuring fair play through market regulation. The Department for Work and Pensions (incrementally, especially between 1997 and 2010) has operated to establish compliance and control with a new incentive paradigm. Re-centralization As has been demonstrated above, the strong central UK state has brought about changes that have strengthened the centre and decentralized delivery – pulling inwards and pushing outwards simultaneously. In addition to these processes, there is also evidence of re-centralization, particularly in relation to the organizational merger that formed Jobcentre Plus in 2002. The creation of Jobcentre Plus involved joining the Benefits Agency (which had main responsibility for administering income maintenance) with Jobcentres (which already had a role in activating all Jobseeker’s Allowance claimants). This shift towards onestop shops has occurred in several countries (Clasen et al., 2001; Van Berkel, 2010). This development went hand in hand with a widening of the activation net to involve benefit recipients previously exempt from compulsory job search conditions (that is, ill or disabled people and lone parents) in work-focused interviews. This re-centralization allowed for more concentrated application of the work-first and work-for-all approach. Also, in a similar way to the creation of the Employment Service in 1990 (representing decentralization, see above) the driving principle was cost cutting. The Jobcentre Plus organizational merger also involved 30,000 job cuts. The second element of re-centralization in the UK involved the use of contracts to impose central control over the management and frontline

Tomáš Sirovátka

Sharon Wright

97

delivery of services delivered by a multiplicity of actors. The anticipated dismantling of Jobcentre Plus following the Freud Review (2007) did not transpire fully, because it remained necessary to deal with the administration of benefits centrally and because most jobseekers return to work anyway before they reach the point of referral to external agencies. Jobcentre Plus retained a substantial role in administering benefits and job search support, continuing to deliver the majority of employment services for all short-term unemployed people. Both the Department for Work and Pensions and Jobcentre Plus continue to re-centralize through marketization and contractualization, based on particular assumptions about the motivation and actions of out-of-work citizens. Marketization Increased marketization has been closely connected with the process of decentralization outlined above, and is an increasingly significant trend in the new governance of activation in the UK. For more than a decade, the UK Government has increased its efforts to contract-out advice and job-matching services for benefit recipients. A decisive moment came in these developments when, in 2006, investment banker David Freud was commissioned to review welfare-to-work programmes and recommend changes that were reputedly intended to increase the employment rate (although, as noted above, this was already exceptionally high by historical and international standards). Freud’s (2007) proposals for a large-scale marketization of advice services for benefit recipients were far-reaching and contentious. Market-based reforms were subsequently adopted. Freud’s personal influence on these developments was very strong, since, following his figure-heading of Labour welfare reforms, Lord Freud changed allegiance and was appointed in 2010 as Parliamentary Under-Secretary of State for the Conservative–Liberal Democrat coalition government, again playing a key role in welfare reform. In practice, the development of the quasi-market in employment services was dominated quickly by a small number of prime contractors, with a subordinate role for other providers, including voluntary agencies and local authorities, as well as smaller-scale private providers. For this reason, some analysts have interpreted this change largely as a privatization (Grover, 2009; Piggott and Grover, 2009) rather than simply a marketization, within which private sector values and practices are generalized to all providers via contract design and performance management (see below). However, the dominance of the market by largescale profit-making agencies does not necessarily mean big changes in what is available at local level, because subcontracting arrangements

Tomáš Sirovátka

98

Steering with Sticks, Rowing for Rewards

often result in the use of existing services. This means that change is mainly in the terms of the contract – mediated by profit-making multinationals. Private providers dominate this new market and squeeze out not only competitor organizations, but alternative ways of working, values and beliefs. In this way, the impact of reforms in the governance of activation may impact on benefit recipients indirectly in the longer term, if the conditions of marketization and contractualized funding mechanisms come to threaten the operation, and ultimately the organizational survival, of agencies that have the capacity to represent and advocate on behalf of vulnerable citizens (either through policy campaigning or face-to-face support). That is to say, the consequences of a reorientation of state support for advice, towards services that offer work-first advice, may leave voluntary sector agencies without financial support for the aspects of their work that do not fit this mould. Contractualization New governance reforms of decentralization and marketization are closely connected with developments in contractualization. Contracts are now used extensively to regulate cooperation within the public sector or between public sector and private and voluntary sector agencies. Delivery arrangements for the New Deal programmes (1997–2010) marked a step towards contractualization and fragmentation of delivery for specific groups, but also contributed towards a form of network governance (Considine, 2001). It is difficult to discern the extent to which the rhetoric of joined-up government and partnership working shaped real relationships between different types of provider organizations. The design of contracts (see DWP, 2009) imposed standardized practices that limited discretion and were underpinned by a logic that assumed an understanding of the problem of being out of paid employment that was not necessarily shared by user organizations, who may find it more appropriate to prioritize supporting the well-being of users in relation to needs other than employment (for example, housing, health, family responsibilities).

The shift towards a new hybrid governance regime The evolution of the UK quasi-market in employment services has taken place over almost 20 years. As Newman (2001, 2004, 2005) has argued, it is not possible to track a straightforward shift over time from hierarchy to networks, because these forms of governance coexist and interact.

Tomáš Sirovátka

Sharon Wright

99

However, it is possible to identify changing priorities in the logic of governance reforms. The first such change was to begin to apply market values to the procedural governance (Considine, 2001) that characterized centralized, standardized and rule-bound monopoly state provision of employment information and advice to benefit recipients (approximatel 1990–7). The second collection of reforms was to pioneer the contracting of private providers to deliver area-specific programmes for long-term unemployed people (approximately 1997–2002). The third phase was the current large-scale development of the quasi-market in employment services (designed 2006–8 and implemented from 2009 onwards). These changes can be related to Considine’s (2001) analysis of procedural, corporate, market and network governance. The main change in governance was to apply corporate governance to this new organizational formation through the introduction of competitive business principles, values and practices (Horton and Jones, 1996; Foster and Hoggett, 1999). A lasting concern that was embedded during this period was the pressure to save money by operating more efficiently, implemented through middle-management redundancies, downgrading of terms and conditions for the employment of frontline staff and increasing use of performance targets (Finn and Taylor, 1990; Fletcher, 1997). The emphasis on management by incentives contributed to what Van der Veen and Trommel (1999) call the incentive paradigm, which is based on a particular understanding of human motivation and behaviour that has become increasingly popular as a point of reference in reforms of the content and delivery arrangements in this policy field. However, there is empirical evidence from the late 1990s to suggest that the penetration of new forms of corporate governance may have been limited, with street-level staff acting mainly in relation to a vast system of rules and regulations, signalling deferral to the older procedural form of governance (Considine, 2001; Considine and Lewis, 2010). For the majority of jobseekers, their interactions with advisers were governed mainly by the rules, alongside new systems of numeric job placement targets (Considine, 2001; Wright, 2006). This would seem to suggest that, rather than a clear transition from procedural to corporate governance, strong aspects of procedural governance resisted the new vogue for corporate values and practices, leaving very little room for discretion or enterprise. In fact, it may be that the primary influence of corporate governance in the late 1990s was rhetorical rather than being strongly realized in practice, with its true significance being to open the door for future marketizing tendencies (Considine, 2001).

Tomáš Sirovátka

100

Steering with Sticks, Rowing for Rewards

The simultaneous operation of procedural governance with the targets of corporate governance thus represents a hybrid type (Wiggan, 2009). Changes were clear in both governance and activation. Despite a general embedding of corporate governance during the 1990s and early twenty-first century, there was still no wholesale shift away from hierarchy or bureaucracy, with the vast majority of jobseekers being assisted back into employment by civil servants before reaching the duration of claim to enter an intensive programme such as one of the New Deals.

Effects of governance reforms It is very difficult to find evidence that isolates the effects of changes in the content of policy (for example, formal reductions in eligibility for benefits, compulsory job-search activities and requirements to accept any job offers) from the effects of changes in governance (such as marketization), particularly since area-based initiatives (such as Employment Zones) operated concurrently with standard Jobcentre Plus provision and other programmes (for example, New Deal programmes). Establishing, for example, whether quasi-markets in employment services offered higher-quality services or better value for money is difficult because of a general lack of ‘valid and systematic knowledge on the outcomes and effects the system is able to deliver’ (Bredgaard and Larsen, 2008, p. 345). Furthermore, these ongoing processes are, to some extent, interconnected, even if they may appear contradictory (as in the case of simultaneous decentralization and re-centralization in the UK case). The intricacy of each of the processes captured by the term governance reform further complicates the task of unravelling causes and effects. In addition to these methodological difficulties (see Hasluck and Green, 2007; Stafford, 2007), it is impossible to establish the counterfactual – that is, to compare how effective public provision could have been if the resources used for marketization (for example, incentive payments for providers and employers, transaction costs) had been invested instead in the existing services. Outcomes In terms of straightforward job placement, standard Jobcentre Plus services claim the greatest success rate of assisting nine out of ten jobseekers back into employment within a year (Harker, 2006). However, since Jobcentre Plus provides only limited support with minimal direct contact or assistance, and ‘the great majority of customers leave benefit without having participated in any of the major Jobcentre Plus

Tomáš Sirovátka

Sharon Wright

101

interventions’ (Hasluck and Green, 2007, p. 4), it would perhaps be more accurate to attribute a proportion of this success to individual jobseekers’ efforts. When data for 2001–6 were compared with five-year periods in the 1980s and 1990s, they showed that the number of people claiming benefits for unemployment had reduced (as had the average duration of claim), and that there had been no increase in the number of people making repeat benefit claims (NAO, 2007b). Overall, official evaluation evidence concludes that initiatives such as the New Deal programmes, Employment Zones and Working Neighbourhoods Pilots were successful in assisting more benefit recipients into employment than would have got a job otherwise (Hasluck and Green, 2007; NAO, 2007a). Jobseekers, rather than other client groups (for example, lone parents or disabled people), have benefitted most from employment programmes and relatively few people from workless households have participated (Dewson et al., 2007; NAO, 2007a). Between 1998 and 2007, 1.8 million people entered employment from a New Deal programme (NAO, 2007b). However, the independent effects of programmes that pushed disadvantaged groups towards the labour market are difficult to identify, since at the same time a thriving economy offered job opportunities, which paid better because of minimum wages and in-work tax credits and were made more possible through help with the costs of childcare, which increased in availability (Gregg et al., 2006). It could be that the solution to high levels of benefit receipt amongst inactive groups lies with meeting needs (through, for example, adequate remuneration for employment and childcare provision) rather than activation. Overall, the cost effectiveness of programmes such as New Deal and Employment Zones was variable, with cost effectiveness dependent on longer-term employment outcomes (NAO, 2007b). Longitudinal cohort data showed a minority (12–21 per cent) of New Deal and Employment Zone participants moving into employment immediately after the programme, and remaining in a job for six months on average (NAO, 2007b, p. 16). The same data tracked programme participants in the 2002–3 cohort and found that participants in the New Deal for Lone Parents and the New Deal for Young People spent, on average, about one quarter of the four years after they took part in the programme in employment – the same proportion of time on benefits and approximately 5–6 per cent of their time back on the programme (NAO, 2007b). Results were worse for the New Deal 25 Plus cohort, who spent only 10 per cent of the four years following programme participation in employment (ibid.). Even the more successful Employment Zone evaluations showed that almost half of mandatory participants left without

Tomáš Sirovátka

102

Steering with Sticks, Rowing for Rewards

securing paid work (Griffiths and Durkin, 2007), and increasing proportions of disabled people are reported to have stopped claiming benefits, without finding employment (Stafford, 2007). This reinforces longstanding concerns over the third of young people who left the New Deal for Young People to ‘destinations unknown’ (Hasluck, 2000). Intractable problems of low pay and poor job quality continue to present barriers even for the most enthusiastic providers (Griffiths and Durkin, 2007), especially in ‘demand-deficit’ local labour markets (Hasluck and Green, 2007). There has been a lack of employer engagement in employment programmes (Dewson et al., 2007). Sustainability of employment was a major issue, with criticism of early job placement into entry-level jobs, which may be unsuitable and unsustainable (usually not lasting into the longer term) without any additional employability, development of skills or advancement (Griffiths and Durkin, 2007; Hasluck and Green, 2007; NAO, 2007b; Yeo, 2007). This was found to lead to ‘churning’ between low-paid unsustainable employment, benefits and employment programmes (Hasluck and Green, 2007). Indeed, around two-thirds of the annual 2.4 million new JSA claims were repeat claims and around 40 per cent of JSA claimants who gain employment will make a repeat claim within six months (NAO, 2007b). The work-first design of programmes and contracts was problematic. Outcome payments rewarding jobs lasting 13 weeks were found to be ‘too short, and falls short of the point at which most employment programmes deliver a positive return on investment’ (NAO, 2007b, p. 10). Recommendations were made for longer-duration contracts and targets that could take into account individual jobseeker needs; provide for disadvantaged participants who want better-quality jobs; and avoid creaming and parking, perhaps by measuring aggregate employment (Griffiths and Durkin, 2007; NAO, 2007b). Recommendations were also made for more effective in-work support and initiatives to increase skills and qualifications, that is, more of a holistic human capital development approach than a short-term work-first approach (ibid.). However, there are contradictions, since Employment Zones appeared to achieve better results, except for lone parents (when measured according to immediate job entry) than the New Deal Programmes but offered much worse value for money (Griffiths and Durkin, 2007). This reinforces the message that quick fixes are expensive and do little to address the very real, acute and multiple difficulties faced by the small proportion of long-term unemployed people in the UK (Hasluck and Green, 2007).

Tomáš Sirovátka

Sharon Wright

103

Outputs Several studies have highlighted the sensitivity of providers to financial incentives and contract requirements, leading to efforts being directed at those who are most profitable rather than those with greatest need (Griffiths and Durkin, 2007; Stafford, 2007). Perhaps the clearest evidence that the goal of cost cutting was paramount (over, for example, the desire to raise the employment rate or eradicate child poverty) was from contract negotiations for the New Deal for Disabled People, where successful providers had their numbers of participants capped by the DWP because of over-performance (Stafford, 2007). There has also been disquiet from a range of providers, since the unprofitability of some programmes led to contract renegotiation (in some cases withdrawal from provision) – a process that benefitted larger private providers who could cross-subsidize from alternative internal or external sources (Stafford, 2007). This tension between cost effectiveness (or even cost cutting) and performance seems set to continue, and Employment Zone evaluations also showed that steps to increase value for money could endanger the success of the initiative (Griffiths and Durkin, 2007). Several evaluations have pointed towards the quality improvements made possible by flexibility (for example, to offer users longer and more frequent interviews at times to suit them); discretion and autonomy; and individually tailored practical and financial assistance (Dewson et al., 2007; Griffiths and Durkin, 2007; Hasluck and Green, 2007; NAO, 2007a). However, this combination of flexibility comes at the price of transparency and accountability within the implementation black box (Griffiths and Durkin, 2007). There is little evidence that these improvements could not have been made through Jobcentre Plus provision. However, there has been a general lack of innovation (Dewson et al., 2007), and any that has developed has failed to benefit the most disadvantaged (Griffiths and Durkin, 2007; Hasluck and Green, 2007). Processes Over the last decade, there has been an observable simplification of the problem that social security policies are designed to tackle: ‘worklessness’ has come to be seen as the functional equivalent of unemployment. The policy instruments that form the solution have been blunted. The principles and practices that were originally developed (the provision of work-first employment information and advice to stimulate jobseeking activities towards early job entry (see Hasluck and Green,

Tomáš Sirovátka

104

Steering with Sticks, Rowing for Rewards

2007)) for one client group, the unemployed, have been generalized and applied to others, principally lone parents and those who are ill or disabled. The impacts of reforms in the governance of activation are potentially significant and far reaching for a range of different social actors: primarily the citizens who are subject to these reforms, but also front-line advisers, agencies (of all types, including those newly subcontracted as well as Jobcentre Plus) and even whole sectors (for example, the voluntary sector). It is certainly conceivable, for example, that the voluntary sector, in its unique position of representing the interests of vulnerable and disadvantaged people and organizationally susceptible to changes in funding, may be affected to a large degree by becoming dependent on DWP contracts to provide work-first job search support. Evaluation of several marketized programmes (such as Employment Zones, Working Neighbourhoods Pilots and New Deal programmes) showed disappointing results in relation to the promised development of participant choice (Dewson et al., 2007; Griffiths and Durkin, 2007; Hasluck and Green, 2007; Stafford, 2007). Choice of provider was often not available at all for mandatory randomly allocated referrals from Jobcentre Plus and, where available (more commonly for voluntary participants), it was either restricted by availability of locality or limited to initial selection, after which point changing provider was impossible (Griffiths and Durkin, 2007). Mandatory enrolment in employment programmes has also created difficulties, since motivation plays a key role in effectiveness, with good outcomes in voluntary programmes (Stafford, 2007). If benefit recipients ‘feel coerced into taking provision, where they feel such provision is inappropriate, where they feel that the cost to them of overcoming barriers is greater than the benefits of entering employment, their chances of moving from benefit into employment are much less’ (Hasluck and Green, 2007, p. 46). This undermines the functioning of a quasi-market, a necessary precondition of which is user choice (see Wright, 2008). Another inherent flaw in the quasi-market model is that ‘policies designed to develop the market and demonstrate the commercial viability of delivery appear to run counter to objectives designed to increase provider competition and achieve better value for money’ (Griffiths and Durkin, 2007, p. 73). One benefit of contracted-out support appears to be better customer satisfaction, perhaps due to a more customer-orientated culture and greater autonomy and flexibility to respond to users needs (Griffiths and Durkin, 2007), or due to greater resource allocation and more intensive case-loading. Better-off calculations have been identified as

Tomáš Sirovátka

Sharon Wright

105

an effective tool for advisers (NAO, 2007b). However, even in the best performing contracted-out provision, the Employment Zones: ‘there is no evidence that private sector providers necessarily perform better than the voluntary sector, or indeed the public sector’ (Griffiths and Durkin, 2007, p. 72). For Working Neighbourhoods Pilots, marketized to different degrees in some areas, there was also no evidence that contractors performed better than Jobcentre Plus (Dewson et al., 2007).

Conclusion The UK has been remarkably robust in retaining a substantial role for the civil service in providing centralized administration of social security benefits and employment services. Early signs of reforms began in the late 1980s, with activation through stricter conditionality, mandatory interviews and tougher sanctions. The governance of this new approach to stimulating job entry was to outsource training services. Major reforms were undertaken in the 1990s, starting with the creation of the Employment Service as an executive agency, working at arm’s length from central government to deliver job-matching assistance to unemployed people. This organizational transformation also involved the application of new managerialist working practices, cuts in staffing levels and a simplified management hierarchy. During the 1990s, interagency cooperation was evident through partnership working arrangements in the delivery of training and back-to-work programmes. Efforts both to activate (including a widening of the target group to include previously inactive groups) and to reform governance (particularly through marketization and the development of a quasi-market in employment services) were intensified in the late 1990s and continue into the 2010s. The governance of activation in the UK presents an example of a hybrid model (Wiggan, 2009; Considine and Lewis, 2010). Strong aspects of procedural governance still exist, although the overarching governance logic has shifted firmly towards corporate and market governance. Aspects of network governance also continue to be important, but are overshadowed by corporate and procedural governance. Overall, a highly centralized mandate (largely excluding alternative voices, for example social partners, campaigners and policy makers at devolved (Scotland, Wales and, to a lesser extent, Northern Ireland) and municipal levels) exists for decentralized delivery, which is closely controlled through contracts and new managerialist incentive structures. Whilst policy-making processes are closed-off and hierarchical, this contrasts with the rhetoric of inclusiveness and empowerment that

Tomáš Sirovátka

106

Steering with Sticks, Rowing for Rewards

surrounded the introduction and development of partnership working and marketized delivery arrangements. Since the late 1980s, developments in work-first activation and changing governance structures and practices have been driven by a cost-cutting rationale, backed up by ideologically based calls to end welfare dependency. A high level of continuity and cross-party political consensus regarding these policy drivers has allowed for the widespread application of market values (despite a lack of evidence that competition can deliver the intended outcomes, or outcomes that are beneficial to citizens) and seemingly conflicting developments, such as simultaneous decentralization and re-centralization, which may not otherwise have been possible.

Note 1. 1990–2001 social security was within the remit of Department of Social Security; Employment Services were the responsibility of the Department of Employment (1990–4); then the Department of Education and Employment (1994–2001); social security and employment were merged from 2001 onwards within the Department for Work and Pensions.

References Barlow, J. and M. Röber (1996) ‘Steering not rowing: co-ordination and control in the management of public services in Britain and Germany’, International Journal of Public Sector Management, 9/5–6, 73–89. Bredgaard, T. and F. Larsen (2008) ‘Quasi-markets in employment policy in Australia, the Netherlands and Denmark: Do they deliver on promises?’, Social Policy and Society, 7/3, 341–52. Clarke, J. and J. Newman (1997) The Managerial State: Power, Politics and Ideology in the Remaking of Social Welfare (London: Sage). Clasen, J., G. Duncan, T. Eardley, M. Evans, P. Ughetto, W. van Oorschot and S. Wright (2001) ‘Towards “Single Gateways”? A cross-national review of the changing roles of Employment Offices in seven countries’, Zeitschrift fuer Internationales und Auslaendisches Sozialrecht, 1, 43–63. Considine, M. (2001) Enterprising States: The Public Management of Welfare-to-Work (Cambridge: Cambridge University Press). Considine, M. and J. Lewis (2010) Taking the Herd to Market: Governance, Steering and Service Delivery in Australia, the Netherlands and the UK, presented at ‘The global finanical crisis and new governance of labour market policies’, RESQ International research conference, Copenhagen, 24–25 June. Davies, S. (2008) ‘Contracting out employment services to the third and private sectors: A critique’, Critical Social Policy, 28/2, 136–64. Department for Work and Pensions (2003) The UK Employment Action Plan (London: The Stationery Office).

Tomáš Sirovátka

Sharon Wright

107

Department for Work and Pensions (2008a) DWP Commissioning Strategy (London: The Stationery Office). Department for Work and Pensions (2008b) Raising Expectations and Increasing Support: Reforming Welfare for the Future (London: The Stationery Office). Department for Work and Pensions (2009) ‘Supplying DWP: Flexible New Deal’, http://www.dwp.gov.uk/supplying-dwp/what-we-buy/welfare-towork-services/flexible-new-deal/ (accessed 2 November 2010). Dewson, S., J. Casebourne, A. Darlow, T. Bickerstaff, D.R. Fletcher, T. Gore and S. Krishnan (2007) Evaluation of the Working Neighbourhoods Pilot: Final Report (London: The Stationery Office). Finn, D. (2009) ‘The “welfare market”: The role of the private sector in the delivery of benefits and employment services’, in J. Millar (ed.) Understanding Social Security (Bristol: The Policy Press). Finn, D., M. Knuth, O. Schweer and W. Somerville (2005) Reinventing the Public Employment Service: The Changing Role of Employment Assistance in Britain and Germany (London: Anglo-German Foundation). Finn, D. and D. Taylor (1990) The Future of Jobcentres: Labour Market Policy and The Employment Service (London: IPPR). Fletcher, D.R. (1997) ‘Evaluating special measures for the unemployed: Some reflections on the UK experience’, Policy and Politics, 25/2, 173–84. Foster, D. and P. Hoggett (1999) ‘Change in the Benefits Agency: Empowering the exhausted worker?’, Work, Employment & Society, 13/1, 19–39. Freud, D. (2007) Reducing Dependency, Increasing Opportunity: Options for the Future of Welfare to Work. An Independent Report to the Department for Work and Pensions (London: Corporate Document Services). Gregg, P., S. Harkness and L. Macmillan (2006) Welfare to Work Policies and Child Poverty: A Review of Issues Relating to the Labour Market and Economy (York: Joseph Rowntree Foundation). Griffiths, R. and S. Durkin (2007) Synthesising the Evidence on Employment Zones (London: The Stationery Office). Grover, C. (2009) ‘Privatizing employment services in Britain’, Critical Social Policy, 29/3, 487–509. Handle, J. (2003) ‘Social citizenship and workfare in the US and Western Europe: From status to contract’, Journal of European Social Policy, 13/3, 229–43. Harker, L. (2006) Delivering on Child Poverty: What Would it Take? A Report for the Department for Work and Pensions (London: The Stationery Office). Hasluck, C. (2000) ‘Early lessons from the evaluation of New Deal Programmes’, Labour Market Trends, 108/8, 369–77. Hasluck, C. and A.E. Green (2007) What Works for Whom? A Review of Evidence and Meta-Analysis for the Department for Work and Pensions (London: The Stationery Office). Horton, S. and J. Jones (1996) ‘Who are the new public managers? An initial analysis of “Next Steps” Chief Executives and their management role’, Public Policy and Administration, 11/4, 18–44. Lindsay, C. and M. Mailand (2004) ‘Different routes, common directions? Activation policies for young people in Denmark and the UK’, International Journal of Social Welfare, 13/3, 195–207.

Tomáš Sirovátka

108

Steering with Sticks, Rowing for Rewards

Lindsay, C., R.W McQuaid and M. Dutton (2007) ‘New approaches to employability in the UK: Combining “Human Capital Development” and “Work First” strategies?’, Journal of Social Policy, 36/4, 539–60. Millar, J. (2002) ‘Adjusting welfare policies to stimulate job entry: The example of the United Kingdom’, in H. Sarfati and G. Bonoli (eds) Labour Market and Social Protection Reforms in International Perspective: Parallel or Converging Tracks? (Aldershot: Ashgate/ISSA). Millar, J. (ed.) (2009) Understanding Social Security: Issues for Policy and Practice (Bristol: The Policy Press). National Audit Office (2007a) Helping People from Workless Households into Work (London: The Stationery Office). National Audit Office (2007b) Sustainable Employment: Supporting People to Stay in Work and Advance (London: The Stationery Office). National Statistics (2008) ‘Employment’,http://www.statistics.gov.uk/CCI/ nugget.asp?ID=12 (accessed 16 April 2008). Newman, J. (2001) Modernising Governance: New Labour, Policy and Society (London: Sage). Newman, J. (2004) ‘Modernising the State: A new style of governance’, in J. Lewis and R. Surender (eds.) Welfare State Change: Towards a Third Way? (Oxford: Oxford University Press). Newman, J. (2005) Remaking Governance: Peoples, Politics and the Public Sphere (Bristol: The Policy Press). Nickell, S. (1999) ‘Unemployment in Britain’, in P. Gregg and J. Wadsworth (eds.) The State of Working Britain (Manchester: Manchester University Press). Piggott, L. and C. Grover (2009) ‘Retrenching Incapacity Benefit: Employment Support Allowance and paid work’, Social Policy and Society, 8/2, 157–70. Price, D. (2000) Office of Hope: A History of the Employment Service (London: Policy Studies Institute). Showler, B. (1976) The Public Employment Service (London: Longman). Stafford, B. (2007) New Deal for Disabled People: Third Synthesis Report – Key Findings from the Evaluation (London: The Stationery Office). Theodore, N. and J. Peck (1999) ‘Welfare-to-work: National problems, local solutions?’, Critical Social Policy, 19/4, 485–510. Tonge, J. (1999) ‘New packaging, old deal?: New Labour and employment policy innovation’, Critical Social Policy, 19/2, 217–32. Van Berkel, R. (2010) ‘The provision of income protection and activation services for the unemployed in “active” welfare states. An international comparison’, Journal of Social Policy, 39/1, 17–34. Van Berkel, R. and V. Borghi (2008) ‘The governance of activation: A literature review’, Social Policy and Society, 7/3, 393–402. Van der Veen, R. and W. Trommel (1999) ‘Managed liberalization of the Dutch welfare state: A review and analysis of the reform of the Dutch social security system, 1985–1998’, Governance: An International Journal of Policy and Administration, 12/3, 289–310. White, M. (2000) ‘The New Deal for Young People: Towards an ethical employment policy?’, Policy Studies, 21/4, 281–95. Wiggan, J. (2009) ‘Mapping the governance reform of welfare to work in Britain under New Labour’, International Journal of Public Administration, 32, 1026–47.

Tomáš Sirovátka

Sharon Wright

109

Wright, S. (2006) ‘The administration of transformation: A case study of implementing welfare reform in the UK’, in P. Henman and M. Fenger (eds) Administering Welfare Reform: International Transformations in Welfare Governance (Bristol: The Policy Press). Wright, S. (2008) Contracting Out Employment Services: Lessons from Australia, Denmark, Germany and the Netherlands (London: Child Poverty Action Group). Yeo, A. (2007) Experience of Work and Job Retention among Lone Parents: An Evidence Review (Leeds: Department for Work and Pensions).

Tomáš Sirovátka

6 The Governance of Activation Policies in Italy: from Centralized and Hierarchical to a Multi-Level Open System Model? Paolo R. Graziano and Annelies Raué

Introduction The Italian welfare state has traditionally been characterized by extensive protection of the workers and insiders, and very limited provisions in terms of security or employment opportunities for the outsiders. Low participation rates and rising unemployment – already surfacing in the 1970s – were partially consequences of this traditional organization of the welfare state. The early warnings of the unsustainability of the system could, however, still be ignored in times of economic prosperity. Yet the oil shocks and the subsequent economic downward trends signalled the beginning of a new period of more troubled times. Although the first, cautious steps to deal with the situation were taken in the 1980s, significant reforms found enough political and social support for policy adoption and implementation only in the 1990s. Also, due to external European constraints (Ferrera and Gualmini, 2004; Graziano, 2004), reforms became unavoidable and the social policy governance structure was also affected. The main aim of this chapter will be to answer, with regard to the Italian case, the common research question raised in this book: Have new forms of governance emerged from the most recent reforms? Emphasis lies on the reforms in the Italian system of unemployment benefits, employment services and social assistance during the past decade. Although it is not yet possible to draw sound conclusions regarding the effects of the reforms, we will present some evidence and focus more on the challenges that the new governance setting poses to the Italian institutional and non-institutional actors. From a more normative standpoint, in order to deliver properly (that 110

Tomáš Sirovátka

Paolo R. Graziano and Annelies Raué

111

is, contributing to lower unemployment and poverty rates, and raising the overall employment rate), the key processes detected in the Italian case (decentralization and marketization) need to be fully coordinated and supplemented by central institutions in order to avoid further exacerbation of the traditional North–South divide, which has been one of the main political concerns from the late nineteenth century. The chapter will be structured as follows. The first section will present and discuss the analytical tools that will be used in order to trace governance changes. The following section will provide an overview of the situation on the Italian labour market since the arguments used in order to support both policy and governance changes in the 1990s were strongly connected to the peculiar substantive difficulties (high unemployment and poverty rate, low employment rate and high territorial differentiation) faced by the Italian government. The third section will deal with the main features of employment and social assistance policies in comparative perspective. The fourth section focuses on the most relevant policy reforms and their consequences on the overall social policy governance system. The fifth section discusses the (limited) evidence regarding the effects of the reform and the new emerging challenges. The final section will deal with the implications of these reforms on the governance regime and will point out further areas for research in order to draw conclusions on the success of these policies in shaping an employment-friendly welfare system.

The governance of employment and social policies: in search of a typology As stated in Chapter 1 of this book, in a noteworthy contribution Newman (2001) distinguished four models of governance: the hierarchy model, the rational goal model, the open system model and the self-governance model. The first model is characterized by control, predictability and accountability and is based on formal authority; the rational goal model is oriented at the maximization of output and outcome; the open system model focuses on flexibility, expansion and adaptation, and it is based on flows of power within networks; finally, the self-governance model is oriented towards devolution, participation and sustainability, and is based on citizen and community power. This distinction is quite interesting, since it focuses on the possible main features of different empirical social policy governance systems in developed countries. Nevertheless, it may be problematic since it focuses on two different analytical criteria: the first is the main goal of

Tomáš Sirovátka

112

The Governance of Activation Policies in Italy

the system, whereas the second concerns the power relationship nature within the system. If we consider the first distinctive criterion we may be somewhat puzzled in trying to operationalize the main goal of the system concept, since it could be hard to find solid indicators signalling the phenomenon we want to investigate. The second criterion is a more classic concept in social science and in the governance literature (Mayntz, 1999), and it does not pose severe difficulties in its operationalization. To be sure, the distinction suggested by Newman has another fundamental merit: it focuses – rather explicitly – on the overall openness of the governance system. This criterion needs to be considered seriously, since several reforms of the 1990s have been inflated by the partnership rhetoric (Boerzel and Risse, 2005; Vesan and Graziano, 2008). Nevertheless, looking at the partnership and governance literature, we also realize that often the two criteria have been considered synonymous, and therefore we may allow ourselves to merge the two into one more comprehensive and consolidated label: policy style (Richardson, 1982). Chapter 1 also discusses Considine’s contribution (2001) which leads – using the sources of rationality, forms of control, primary virtues and service delivery as selective criteria – to four governance models: procedural governance, corporate governance, market governance and network governance. These models are also very promising analytically, but again not easy to operationalize. Therefore, for the purpose of our case study we will simplify the analytical framework by focusing on the policy (or policy-making) style – which may be hierarchical or nonhierarchical – together with another, echoed in Considine’s volume and strongly connected to the content of recent social policy reforms, especially in Italy (Graziano, 2007): the territorial organization of the governance system. In this respect we distinguish between a state-centric model, where the central government is the key (if not the only) institutional actor involved, and a multi-level model where sub-national actors play a relevant role in the overall governance system. Table 6.1 illustrates the typology. As will be shown in this chapter, the Italian case over the years has increasingly moved away from a state-centric and hierarchical model Table 6.1

Employment and social policy governance models: a typology

Policy style/territorial organization

Hierarchical

State-centric Multi-level

Tomáš Sirovátka

Non-hierarchical

Paolo R. Graziano and Annelies Raué

113

to a multi-level and non-hierarchical one. But, more specifically, how will we proceed in the Italian case in order to answer our basic research question: has – and, if so, how – the Italian social policy governance system changed, especially during the past decade? In order to pursue properly such an endeavour we need to clarify our research design and method. First, we will follow a two-step research design. The first step consists of an illustration of the main features of the governance model prior to the reforms that were adopted and implemented from the late 1990s. We will consider policy-making style and the centralized/decentralized nature of the system in order better to classify the Italian case. We will then focus on the policy reforms and investigate the governance changes. From a methodological standpoint, our research is grounded in historical neo-institutionalism and process tracing (Pierson, 2000) and the sources used consist of policy documents and secondary literature. Before we focus on our main research question, a country characterization with respect to the specificities of the employment and social assistance problems faced by Italy in past years is required.

The main features of Italian employment and social assistance in comparative perspective In order to provide a snapshot of the long-lasting Italian employment and social assistance challenges, an overview of the development of the unemployment, employment and poverty rates over the past decade will be presented here. Data will highlight the prevailing dualism that marks the country’s North–South divide. Moreover, a comparison between countries in terms of social policy expenditure will provide greater insight into the challenges the Italian welfare system is currently facing. As Figure 6.1 clearly demonstrates, especially during the late 1990s total unemployment in Italy was comparatively high. Only the early years of the new millennium saw a more positive trend, with Italy’s unemployment rate dropping to single-digit numbers and even below the EU27 and EU15 averages.1 However, this positive picture changes somewhat when the regional disparities are taken into consideration. Although in all regions the unemployment rate has dropped during the past few years, a large divergence between the northern and the southern regions persists. The South (and the Islands) is characterized by an extremely high unemployment rate compared with both the national average and the EU27 average. Most striking, however, are the regional imbalances: in 1999 the unemployment range was between 4.6 per cent and 23.5 per cent.

Tomáš Sirovátka

114

The Governance of Activation Policies in Italy

12 10 8 6 4 2 0

1996 1997 1998 1999 2000 2001 2002 0 0 0 8,6 8,5 8,9 EU27 0 EU15 10.1 9.8 9.3 8.5 7.7 7.2 7,6 Italy 11.2 11.3 11.3 10.9 10.1 9.1 8.6

2003 2004 2005 8,9 8,9 9 7.9 8 8.1 8.4 8 7.7

2006 2007 8,1 7,1 7.7 7 6.8 6.1

Years EU27 Figure 6.1

EU15

Italy

Total unemployment rate (percentages)

In the late 2000s, however, the rate continued to be several times lower in the northern regions than in the South. Furthermore, when we unpack the data along a territorial differentiation, another case of regional disparity becomes evident. The southern regions are characterized by low employment rates (although somewhat increasing) while the other three areas (north-east, north-west, centre) hover around the 60 to 65 per cent mark. Finally, in comparative perspective the overall public expenditure on employment policy is very low, and particularly low until the mid-1990s was the expenditure on Active Labour Market Policies (Table 6.2). Along with a long-standing high level of unemployment, the Italian poverty rate has been above the EU average for the past decade or so. As Table 6.3 clearly shows, the share of persons with an equalized disposable income below the poverty threshold is higher than in other Table 6.2 Passive and active labour-market policies (ALMP) expenditure in selected EU countries (percentage of GDP, 2005)

Sweden Germany UK Italy

ALMP

Passive policies

Total

1.3 1.1 0.5 0.6

1.2 1.3 0.3 0.5

2.5 2.4 0.8 1.1

Source: OECD, Social Indicators Database.

Tomáš Sirovátka

Paolo R. Graziano and Annelies Raué

115

Table 6.3 Share of persons with an equalized disposable income below the poverty threshold in the EU

Sweden UK Germany Italy EU15 EU27 ∗ 2005

1999

2004

2008

8 18 11 19 15 −

11.3 18∗ 12.2∗ 19.1 17 −

12.2 18.8 15.2 18.7 16.4 16.5

data.

EU countries (with the exception of the UK) and the EU15 average – although lower than the EU27 average. What is also particularly interesting is that if we adopt a diachronic perspective, the challenge for Italian policy makers became initially more stringent and then levelled out over the past decade since the relative poverty rate was 8.3 in 1980 and 10.2 in 1994 (Commissione d’indagine sulla povertà, 1996), and it remained fairly stable from 1999 to 2008 (Table 6.3). Furthermore, as in the case of low employment and high unemployment rates, the territorial distribution of relative poverty is particularly relevant in the Italian case (Table 6.4). Poverty is concentrated in the southern part of the country and this constitutes a specific challenge for the governance of social policy, as we will see in the following sections. Finally, if we consider the data regarding social services expenditure and if we adopt a comparative perspective, it emerges quite clearly how traditionally underdeveloped the public provision of social services and benefits has been in Italy (Figure 6.2). Table 6.4 Relative household consumption-based poverty in Italy, regional distribution 2006–9

North Centre South Total

2006

2009

5.2 6.9 22.6 11.1

4.9 5.9 22.7 10.8

Source: ISTAT (2010).

Tomáš Sirovátka

116

The Governance of Activation Policies in Italy

5.00 4.50 4.00 3.50 3.00 2.50 2.00 1.50 1.00 0.50 0.00 1980

1985

1990

1995

Eu-15 average

Italy

France

Uk

2000

SW; FIN; DK average

Figure 6.2 Expenditure for services targeted to families, elderly people and the disabled (1980–2000) Note: Health-care services not included. Source: OECD, Social Indicators Database.

In sum, in many regards the Italian situation can be labelled as quite peculiar compared with European averages; often, unemployment rates rose above EU averages and employment rates lagged behind; the same applies to the poverty rate figures. Although recent years have seen promising trends in employment and unemployment figures, this is combined with a persisting territorial dualism that characterizes the Italian employment situation and the endurance of a high level of poverty rate. Unemployment is concentrated in the southern regions where rates are generally three times higher than those in the North. Employment, on the other hand, is clustered in the North. This is especially true for certain disadvantaged groups in society, such as women, as well as for both younger and older workers. The same applies to poverty distribution, since this phenomenon is particularly concentrated in the southern part of the country. All of these general social policy features have historically determined specific challenges to the governance of social policies. More specifically, in the overall context of Italian low public expenditure on employment policy, the internal balance between active and passive expenditure has traditionally been in favour of the passive option. Therefore, given the overall lack of activation procedures, following a path-dependent

Tomáš Sirovátka

Paolo R. Graziano and Annelies Raué

117

hypothesis (North, 1991) we should expect that the activation reforms of the 1990s have proved particularly difficult to implement due to the limited administrative activation experience at the national level, but also, and especially, at the regional level. Before we discuss the main features of the recent governance reforms and their effects, we need to further characterize the traditional Italian state-centric and hierarchical model of employment and social policy governance.

Governing unemployment and social assistance benefits in Italy: limited social protection in a state-centric and hierarchical governance model (1948–97) Unemployment benefits are provided at the national level, with the main actors being the Ministry of Labour and the Italian National Social Security Institute (INPS). Social assistance for the unemployed in the form of minimum resources is mostly arranged at the sub-national level (mainly regional and municipal). As illustrated elsewhere (Graziano, 2007; Jessoula et al., 2010), activation policies have not been at the heart of the traditional Italian employment and social assistance policy, and the governance system tradition mirrored such limited development. In order to provide a detailed account of the governance system prior to the reforms of the late 1990s, we will focus separately on employment and social assistance policies. Unemployment benefits, employment services and vocational training The provision of unemployment benefits in Italy has traditionally been very limited. The following three are amongst the main instruments governed by the Labour Ministry’s Directorate General of the Labour Market and by the Istituto Nazionale per la Protezione Sociale (INPS): (1) the ordinary unemployment benefits, or indennità ordinaria di disoccupazione; (2) short-term unemployment benefits, either ordinary or special: la Cassa integrazione gaudagni ordinaria e Cassa Integrazione guadagni speciale (CIG and CIGS); and (3) the mobility benefit, or indennita di mobilità. The ordinary unemployment benefit is a compulsory social insurance scheme financed through employers’ contributions. Benefits are earnings related: to qualify a worker needs at least two years of coverage with a minimum 52 weeks of contributions, and the replacement rate varies as a function of the average salary received over the reference gross salary. Currently this rate stands at 60 per cent for the first six months,

Tomáš Sirovátka

118

The Governance of Activation Policies in Italy

50 per cent for a further two months and 40 per cent for the remaining four months, if applicable (that is, the worker is over 50 years of age). In addition, there are special provisions for workers in specific sectors such as construction and agriculture. INPS manages both the collection of contributions and the benefit allocation. The case of short-term unemployment benefits, CIG and CIGS presents some interesting peculiarities. These benefits consist of earnings supplements for workers who are temporarily unemployed, due to either impermanent reduction, suspension of activity (CIG) or sector- or area-specific firm restructuring (CIGS). These benefits are funded by both employers’ contributions (CIG) and general taxation (CIGS). It should also be borne in mind that these workers are not considered unemployed, since they are considered to be temporarily suspended from work. The replacement rates do not differ between CIG and CIGS (80 per cent of the previous salary with a ceiling, in both cases), yet the duration does: a maximum of 12 months for CIG and 36 for CIGS. The provision of such benefits is decided by the Labour Minister in coordination with the trade unions and business associations, which allows for some interesting bargaining manoeuvres and influence for these actors in the adoption phase of benefit allocation. Also, in this case contributions’ collection and benefit allocation are guaranteed by INPS. The third and last form of unemployment benefit is the mobility allowance introduced in 1991 by Law 223, with the aim of facilitating the re-inclusion of dismissed workers who were covered by CIGS but following the end of the maximum benefit duration could not be re-employed. The mobility allowance is mainly funded via general taxation, although there is a residual contribution paid by the employer concerned. The beneficiary needs at least one year of coverage and six months of employment in order to be paid 100 per cent of the CIGS benefit during the first year and 80 per cent for the following months. Its duration varies according to the age of the worker and the region where the firm is based yet should, at maximum, last 48 months. However, in some cases workers might be entitled first to CIG(S) and then to the mobility benefit, resulting in a maximum benefit period of eight years. Again, the benefit allocation is managed by INPS offices. The Italian system of public employment services (PES) dates back to 1949, when the state introduced national unemployment lists and centralized governance of both unemployment benefits (INPS) and PES (uffici di collocamento – job placement offices). Until 1997, the PES was state-centric and organized in a standardized and hierarchical way throughout the Italian territories. This system was known to be highly

Tomáš Sirovátka

Paolo R. Graziano and Annelies Raué

119

inefficient, facilitating little matching of labour demand and supply and creating few actual job placements (Ichino, 1982). The services provided were very limited, since the main task was to register the unemployed and allocate the unemployed status, and to pass on the list of unemployed to the recruiting companies, which often found other ways to hire personnel (Gualmini, 1998). Finally, from the late 1940s until the 1970s, education and vocational training was primarily offered by the state vocational institutes (Istituti di formazione professionale) and state technical institutes (Istituti tecnici statali) aimed primarily at education training, whereas from the late 1970s regional institutes started to offer initial vocational training (basic and post-compulsory training and post-diploma and post-degree training; ISFOL, 2000). Although fully public, like the PES the education and vocational training system was mainly state-centric and hierarchically organized – although after 1978, with respect to the above-mentioned initial vocational training measures, some regional differentiation started to emerge. Social assistance benefits Social assistance aimed at ensuring a minimum level of income to citizens who have either a low disposable income or limited overall wealth is even less developed than the system of unemployment benefits. Usually, the social assistance benefits can be provided either to people who are no longer entitled to receiving unemployment benefits or to those who do not meet the conditions of such benefit in the first place and are below a certain income or wealth threshold. First, after the period of unemployment benefits ends, the unemployed are not automatically directed into the social assistance provisions, because their organization is in the hands of different institutions. Secondly, until the late 1990s no universal basic income scheme was present at the national level. In fact, due to the (historical) lack of a national social safety net, systems of social assistance were traditionally arranged and managed at the regional and municipal levels (Bifulco et al., 2008), More specifically, after a long period of nationalization between 1920 and 1970, the establishment of regional governments in the 1970s determined a new governance configuration to social assistance based primarily on the role of the regions. Furthermore, the municipalities increased their role in social assistance policies provision, leaving the state as the main (residual) funding function – via the National Social Assistance Fund (Fondo Nazionale per le politiche sociali) – and coordinating role, until recent times. This means that conditions for social assistance policies ranging

Tomáš Sirovátka

120

The Governance of Activation Policies in Italy

from minimum income schemes to housing and social care policies vary significantly with respect to benefits or services granted, and eligibility and accessibility show great regional and municipal variation, depending on the availability of financial resources (INPS, 2008; CIES, 2010). Nevertheless, these policies could be considered as residual and underdeveloped with respect to the other main pillars of the Italian welfare state. Moreover, until the most recent reform the system was formally open to the (non-institutionalized) involvement of third-sector actors, voluntary organizations and private players, but these actors played only a residual role (Bifulco and Vitale, 2006). In sum, before the wave of reforms that started towards the end of the 1990s, the Italian governance regime of employment and social policies can be labelled as a hierarchical model based on state-centric control, predictability and accountability. This does not entail that local institutions were completely absent, but rather that – when present as in the case of social assistance policies – their role was residual with respect to the overall design of the governance of employment and social policies. With respect to societal (that is, non-institutional) actors, trade unions played a role in important implementation decisions concerning some unemployment benefits (namely the CIGS), but their overall role in the governance of both employment and social assistance policies was very limited. The state and its agencies were controlling the overall functioning of the system (Gualmini, 1998), and also societal interests – if ever relevant – were organized hierarchically and in a centralized way (Graziano, 2004). Furthermore, in the employment policy network there was no room for interest representation of the unemployed (Baglioni et al., 2008), rendering the governance system virtually closed to other interests. Therefore, we can label the 1948–97 Italian employment and social policy configuration as a state-centric and hierarchical governance model.

Changing the social policy governance model: a decade of reforms, 1997–2010 From the second half of the 1990s, the Italian welfare state system has gone through very relevant reforms (Ferrera and Gualmini, 2004; Graziano, 2007, 2009). Activation procedures following flexibility principles have gained increasing ground in the Italian context, in reorientating public expenditure towards active rather than passive policies (see Table 6.2), in decentralizing PES in order to overcome the traditionally limited effectiveness of national employment services (Ichino,

Tomáš Sirovátka

Paolo R. Graziano and Annelies Raué

121

1982), and introducing private actors in the implementation of both employment and social policies. More specifically, the main reforms concerned the extension of coverage for ‘mid-siders’, the expansion of non-standard contracts, the decentralization of employment services and social assistance, and marketization of vocational training and, to a certain extent, social assistance (Jessoula et al., 2010).

Unemployment benefits, employment services and vocational training reforms Reforms started taking place during the 1990s to address the challenges illustrated in the second section of this chapter that were also becoming increasingly under scrutiny due to the launch of the European Employment Strategy (Graziano and Jessoula, 2011). If we look at the governance features, unemployment benefits remained virtually unchanged after the reforms. Among the main actors involved in the provision of these benefits are the National Institute for Social Security (INPS) and its branch offices. Nevertheless, the most recent policy adopted (Law 2/2009) was aimed at contrasting the impact of the crisis by expanding the beneficiaries of ordinary unemployment benefits and providing a new role for regions and regional social partners in the provision of specific unemployment benefits (ammortizzatori sociali in deroga; see also ISFOL, 2008). With respect to PES and education/vocational training, in 1997 a comprehensive reform was adopted at the national level aimed at setting the stage for a process of decentralization of administrative functions to regional and local levels. Among the decentralized policy areas were PES and education and vocational training services. The design and aim of this reform was to expand public-level cooperation to include not only national actors, but also to give greater responsibilities to the provincial and regional levels in terms of employment services. Subsequently, legislative decree 469/97 – aiming at the creation of a single, integrated instrument for coordination of local employment policies – further implemented decentralization and defined specifically the roles and functions of institutions in job placement and employability policies. The 1997 reform provided increasingly new functions to both the regional level and the provincial one (ISFOL, 2004), since the main aim of the reform was to boost regional functions in coping with unemployment protection. The reform implied the transfer of personnel (about 80 per cent of formerly governmental personnel was transferred to the regions and the

Tomáš Sirovátka

122

The Governance of Activation Policies in Italy

provinces; Gandiglio, 2004) and functions such as ‘a) collecting information on labour supply and labour demand in the local labour market; b) identifying priority target groups (long term unemployed, unskilled, women, disabled, immigrant citizens); c) providing individual services and placement programmes; d) supporting job search and participation in professional training courses, easing the access to the labour market; e) providing counselling to companies, information and support on existing specific incentives (collective dismissals policies, tax reductions, assistance on outplacements, etc.); f) promoting self-employment (job creation schemes)’ (Naticchioni and Loriga, 2010). On the other hand, flexibility in combination with facilitated (re-)entry of the labour market became an additional focal point of the reforms.2 The increased range of labour contracts, such as temporary work and fixed-term contracts, is illustrative for this. Furthermore, the expansion of labour contracts offered another opportunity to the private sector to enter the active policy sector: the 1997 law allowed for the opening of (private) temporary work agencies for the matching between this specific kind of labour demand and supply. These temporary work agencies managed to open the market of employment services to private actors. Finally, and in connection with the reforms of vocational training discussed in the previous section, they gained extended responsibilities in the field of education and vocational training (for which they received state funding). Thus, the reforms introduced mainly two new processes: decentralization (with respect to PES, vocational training and specific unemployment benefits) and marketization (with respect to the emergence of new private actors such as temporary agencies and private companies offering vocational training). It is certainly the case for employment policies that the central government (via the Ministry of Labour) remains the key actor. Nevertheless, in all the various stages of policy implementation, other sub-national and private actors have entered the scene. Although central institutions via specific directories generally carry the main responsibility with respect to the other two tasks (LSU/LPU and PIR) that fall under its sub-section ‘Social shock-absorbers and incentives for occupation’ (Ammortizzatori sociali e incentivi all’occupazione), from the mid-1990s there have been clear signs of the creation of a new, open and multi-level governance model. First to be introduced were Lavori Socialmente Utili (LSU) and Lavori di Pubblica Utilità (LPU) – Socially Useful Jobs, which were initially meant as an active measure yet developed into social assistance programmes (Graziano, 2007). Over the years, regional and municipal governments took up some of these

Tomáš Sirovátka

Paolo R. Graziano and Annelies Raué

123

functions and currently the (few) LSU/LPU left are often managed at the sub-national level. Secondly, the progetti e programmi di incentivazione al reinserimento o inserimento lavorativo (PIR – projects or programmes to promote the inclusion or re-inclusion of workers) are projects or programmes aiming at specific groups, are managed centrally with fixed budgets and are of limited duration. Regional levels of government including provinces, regions and municipalities gained increased influence and responsibility in PES. Table 6.5 indicates their reformed responsibilities and tasks, amongst which fall education and vocational training as well as job placement. In addition, education and vocational training – due also to the increased use of the European Social Fund, which is a financial instrument provided by EU regional policy – became increasingly regionalized and open to private actors (both NGOs and associations connected with the trade unions), which has increased their role in providing publicly funded vocational training courses in the various Italian regions (ISFOL, 2000). In sum, with the exception of unemployment benefits, in all the employment policy relevant sub-fields under scrutiny (education and vocational training and PES), the evidence shows clearly how much more decentralized and privatized (or marketized) the Italian governance model is becoming. Social assistance policy reforms Social assistance has traditionally been covered primarily by local administrations. The reforms adopted in the late 1990s tried to introduce a national scheme in order to close the gap with other EU countries where minimum income schemes have been developed. In fact, the 1998–2001 period did witness the experiment of a nationally managed social assistance scheme in the shape of minimum insertion income, the reddito minimo de inserimento (RMI). This provision was meant for unemployed citizens living on an income below a certain threshold. The guidelines were set at the national level, to be further determined and implemented by the regions (a very good example of decentralization), in cooperation with the municipalities and local health centres. The beneficiary was supposed to participate in activities in activation measures, such as vocational training services promoted by the regions or other sub-national institutions, in order to remain eligible for benefits. The aim of such a scheme was to reduce the number of people living below the poverty line, and to introduce a form of conditionality in order to promote vocational training and employability.

Tomáš Sirovátka

124 Table 6.5

Distribution of tasks before and after the 1997 reforms

Ministry of labour

Regions (n = 20)

Provinces (n = 103)

Private sector

PES: orientation (for jobseekers), information and counselling (for jobseekers and firms); coordination of matching supply and demand; management, information supply to SIL

Job placement services’ agencies

Before the 1997 reforms Political guidelines

School counselling Professional training

Planning and coordination of labour policies Government management of PES After the 1997 reforms General political and administrative guidelines

Planning

SIL (Labour Information System): planning, management and development

School counselling

Temporary work agencies

Agencies providing services (orientation, skills balance for jobseekers) to PES

Professional training

Coordination of actions aimed at matching labour supply and demand Source: Adapted from Borghi and Van Berkel (2007, p. 92).

Tomáš Sirovátka

Paolo R. Graziano and Annelies Raué

125

During the implementation of the pilot scheme large regional divergence in success rates emerged. For example, the percentage of RMI beneficiaries involved in social insertion programmes ranged between 8 per cent in certain regions and 92 per cent in others (Borghi and Van Berkel, 2007). This is illustrative for the difficulties the project was facing, especially with respect to the sub-national administrative capacities needed to effectively implement the policy. In fact, the RMI project failed due to lower administrative capacities within the regions, which could not handle the provisions that were first managed at the ministerial level (Madama, 2010). A second reason for its cancellation lies more in political circumstances. When the experiment was cancelled, arrangements were once again left to the regions – some continued with universality and conditionality (Emilia Romagna), others initially did not (Campania). In 2000, Italian social assistance policy seemed to take another big step towards decentralization and better coordination with the approval of a national framework law (Law 328/00), according to which an integrated system of social services and benefits, characterized by strong cooperation between the different levels of government and public–private actors, was to be developed. More specifically, the reform [addressed] two main and connected objectives: 1. to promote the well-being of all citizens according to a principle of ‘selective universalism’; 2. to assign precise powers to the state while setting conditions for a decentralized and negotiated public action. With reference to the first objective, the reform introduces LIVEAS, namely basic standards of social services, which fix the typologies of supply that has to be guaranteed across the entire nation. To target the second objective, the reform promotes a new governance architecture that allocates political-administrative responsibilities to state, regions and municipalities on a principle of vertical subsidiarity. This recalls a model of social citizenship in which the state is to guarantee a principle of (selective) universalism while the local governments are called to take into account the specificities of people and places. In this respect, the reform aims at overcoming the limitations of excessive localism by positively combining local autonomy and national regulation. Bifulco et al. (2008, p. 145) Furthermore, in the formulation of the so-called Local Programming Plans (Piani di zona), the reform called local non-institutional actors

Tomáš Sirovátka

126

The Governance of Activation Policies in Italy

(such as NGOs, trade unions and individual citizens) to participate in local programming (and implementation) activities. This was an open door to subsidized private social assistance services that were offered by NGOs or cooperatives in the various fields of social assistance policy. More specifically, private actors have been involved in local welfare via public administration contracting-out, accreditation and partnerships in co-formulation of local welfare plans. Within the social assistance policy field, the main private actors were non-profit actors, unlike those that became increasingly relevant in employment policy. Put in a different way, the reforms were marked by a clear decentralization and (some) marketization that followed very closely the overall trends inaugurated by employment policy developments. In sum, after the wave of reforms which started towards the end of the 1990s, the Italian governance regime of employment and social policies can be labelled as a multi-level and non-hierarchical model of governance characterized primarily by flexibility principles, expansion of and (partial) adaptation to new services and institutional and noninstitutional actors that have increasingly become part of new policy networks. More specifically, over the years new players – both subnational institutions and private actors – have entered into the social policy game, while marketization procedures have become more relevant in employment and vocational training services, and also in social assistance policies. Beyond trade unions, NGOs, non-profit actors and sub-national governments (at the regional, provincial and municipal level) started to play a relevant role in various decision-making phases (formulation, adoption and implementation) of employment and social policies.

The effects of reforms and new governance challenges All of the above illustrated reforms shared the common features of decentralization and marketization. In particular, the employment services, vocational training and social assistance reforms required a greater involvement from regional and private actors, in both the formulation and implementation phases. Furthermore, with respect to the PES, although territorially differentiated there has been an expansion of employment services (ISFOL, 2008). Finally, vocational training is better targeted, therefore facilitating the inclusion of the unemployed (ISFOL, 2008), and social assistance local plans now take more into account the specific social needs of the territory. Both reforms, however, were differently implemented in the regions: with respect to the PES in general terms, it can be stated – although more

Tomáš Sirovátka

Paolo R. Graziano and Annelies Raué

127

research needs to be done at the regional level – that those regions where the institutional capabilities were already fairly well developed (that is, northern and central regions),3 the overall implementation of the reform was satisfactory, whereas in other regions (which, furthermore, happened to be the ones where unemployment problems are concentrated) significant implementation gaps have been detected – especially in the southern part of the country PES are not as well structured as they are in several centre–north regions (ISFOL, 2008). Also, with respect to the reform of social assistance, regional variation is emerging during the implementation phase. As the Osservatorio per l’attuazione della 328/2000 (trade union-based observatory aimed at evaluating the implementation of the policy) has clearly stated (2009), the main concern regards the capacity to maintain the same effectiveness of the policy in every territory characterized not only by different levels of employment and social challenges but also by significantly varied administrative capacities. In fact, some social actors (such as the main Italian trade union, CGIL) are starting to ask for more coordination and central monitoring on the implementation of Law 328/2008 (Osservatorio per l’attuazione della 328/2000, 2009). Especially after the approval of the Law 42/2009, which further decentralizes the overall Italian administrative system, the fragmentation effects of the reforms are increasingly feared by trade unions and NGOs. In addition, institutional bodies such as the National Anti-Poverty Committee (CIES, Commissione nazionale di indagine sull’esclusione sociale), hosted by the Ministry of Labour, Health and Social Policies, has expressed worries due to the strong signs of regional divergence in policy implementation that have been seen in 2009 (CIES, 2009) and in 2010 (CIES, 2010). In the latest report of the anti-poverty commission, the administrative capacity problem is put forward very clearly and, implicitly, the risk that even if innovative measures are introduced – such as a universal minimum income scheme – the impact will vary significantly if the administrative capacities of several Italian regions are not enhanced.

Conclusion: Italian social policy from state-centric and hierarchical to a multi-level and open governance model Although more research is needed in order further to understand the effects of the decentralization and marketization trends of governance reforms, it emerges quite clearly that two main challenges are currently faced by Italian institutions with respect to employment and social policy governance. The first challenge is to ensure a cohesive decentralization process, avoiding expansion of the already territorial social

Tomáš Sirovátka

128

The Governance of Activation Policies in Italy

differences arising from the new policy design and implementation, as analysed earlier in this chapter. This is particularly difficult in a context where cost-containment strategies are guiding both central and local governments and where the new overall strategy of the government is further supporting a scenario where territorial solidarity will become residual and rudderless with respect to basic, national shared goals. Administrative capacities need to be spread out among sub-national institutions in order to ensure that the decentralization process does not turn into a greater divergence in policy outcomes, that is, employment and social assistance benefits and services provision. The second main challenge regards the lack of opportunity in the Italian employment and social assistance governance systems to hear the voice of the excluded. On the one hand, the substantive effect of the flexibilization of employment relationships has created a new group of workers who are not insiders and – although workers – share many features of the outsiders. Jessoula et al. (2010) suggest that these new, flexible – if not ‘precarious’ – workers may be labelled ‘mid-siders’, since they are not fully excluded from employment relations but at the same time do not share the same level of protection as the insiders, who typically have regular life-long contracts and considerable employment protection legislation. Furthermore, the flexible or precarious workers are becoming an important segment of the Italian workforce: about 12 per cent of Italian workers (about three million people) have temporary contracts (Jessoula et al., 2010). And these workers, in a context where the unemployed are not able to organize collectively at the national level (Baglioni et al., 2008) and have therefore not been included in the creation of the reforms, have limited opportunity to have their voice heard within the traditional trade unions. Against the overall Italian union density of about 33 per cent (Visser, 2006), even if we consider there to be two million temporary workers in the Italian case, the overall union density of temporary workers is around 5 per cent. The data in Table 6.6 regarding the trade union membership of pensioners, standard and non-standard workers confirm the marginal role played by atypical workers’ interests within the trade unions. In principle this consideration will be less relevant for the socially excluded, since in the process of decentralization they may be able not only to be better targeted by social policies but also more included in the decision-making process. But if we take into account the low administrative (and decision-making) capacities of some Italian local administrations, especially in the South (Putnam, 1993; Milio, 2007), and we also add the fact that the cost-containment imperative remains

Tomáš Sirovátka

Paolo R. Graziano and Annelies Raué Table 6.6

129

Main Italian trade unions’ internal composition (2009)

Members (standard contract workers)

Members (pensioners)

Members (fixed-term contract workers)

Total

CGIL

2,705,336 (47.1%)

41,628 (0.8%)

5,746,167

CISL

2,301,702 (50.8%)

28,233 (0.6%)

4,531,085

UIL

1,579,066 (72.6%)

2,994,003 (52.1%) 2,201,150 (48.6%) 574,985 (26.5%)

20,100 (0.9%)

2,174,151

Source: Websites of the trade unions (last accessed: 29 October 2010).

the main goal of both central and local institutions, we may fear that the advantages of being close to decision making may not provide strong support towards the political inclusion of the (already) socially excluded. Finally, even if the reforms have triggered a more efficient and effective system of governance, many social actors underline the fundamental lack of resources allocated in particular to social assistance programmes. Since the late 1990s, the only signs of policy success have been associated with the employment rate increase, but as the CIES (2009, p. 14) has observed, recent years have seen a rise in the working poor. These social problems cannot be solved unless the total social envelope increases. In sum, before the 2008–9 world economic crisis the Italian social policy fostered improvements in unemployment and employment rates, but not in the poverty rate. As suggested previously, if we adopt a governance perspective it is possible to register a rather interesting change from a state-centric and hierarchical model to a multi-level and nonhierarchical system of governance. Decentralization and marketization led to new relations between different levels of government, and between actors from the public and private sectors. More specifically, the employment policy reforms had an impact on the overall governance setting by overcoming the traditional hierarchical model and providing new opportunities for the development of a non-hierarchical (or at least less so) system of governance characterized by some flexibility and greater inclusion of social and institutional actors in a multi-level setting. This does not imply that all the dimensions analysed fall similarly under the same multi-level and non-hierarchical governance model. As stated previously, almost all unemployment benefits are still strongly in the hierarchical hands of the central government. However, if we take into consideration the most recent expansion of unemployment

Tomáš Sirovátka

130

The Governance of Activation Policies in Italy

benefits connected with the crisis, in this field too the traditional hierarchical model is challenged since regional governments are increasingly responsible for the allocation of portions of unemployment benefits.

Notes 1. All data was retrieved from Eurostat unless otherwise stated. 2. The introduction of the mobility allowance as an unemployment benefit is another sign of the increased emphasis on (re-)entry into the labour market. 3. For a classic analysis of patterns of institutional capabilities in Italy, see Putnam (1993).

References Baglioni, S., D. della Porta and P. Graziano (2008) ‘The contentious politics of unemployment: The Italian case in comparative perspective’, European Journal of Political Research, 47/ 6, 827–51. Bifulco, L. and T. Vitale (2006) ‘Contracting for welfare services in Italy’, Journal of Social Policy, 35/3, 495–513. Bifulco, L., M. Bricocoli and R. Monteleone (2008) ‘Activation and local welfare in Italy’, Social Policy & Administration, 42/2, 143–59. Boerzel, T. and T. Risse (2005) ‘Public-private partnerships: Effective and legitimate tools of international governance?’, in E. Grande and L.W. Pauly (eds) Complex Sovereignty: On the Reconstruction of Political Authority in the 21st Century (Toronto: University of Toronto Press). Borghi, V. and R. van Berkel (2007) ‘New modes of governance in Italy and the Netherlands: The case of activation policies’, Public Administration, 85/1, 83–101. CIES (2009) Rapporto 2009 (Rome: CIES). CIES (2010) Rapporto 2010 (Rome: CIES). Commissione d’indagine sulla povertà e sull’emarginazione (1996) La povertà in Italia 1980–1994 (Rome). Considine, M. (2001) Enterprising States: The Public Management of Welfare-to-Work (Cambridge: Cambridge university press). Eurostat, Social Protection in the European Union – Various Years (Luxembourg: Eurostat). Ferrera, M. and E. Gualmini (2004) Rescued by Europe? Social and Labour Reforms from Maastricht to Berlusconi (Amsterdam: Amsterdam University Press). Gandiglio, A. (2004) ‘La riforma dei servizi per l’impiego’, Secondo rapporto annuale sull’attuazione del federalismo (Rome: ISAE), pp. 21–38. Graziano, P. (2004) From a Guaranteed Labour Market to a Flexible Worker Model?, Jean Monnet Series, Working Paper, Forum Européen, Fondation Nationale de Sciences Politiques, 4/2004. Graziano, P. (2007) ‘Adapting to the European employment strategy? Recent developments in Italian employment policy’, International Journal of Labour Law and Industrial Relations, 23/4, 543–65.

Tomáš Sirovátka

Paolo R. Graziano and Annelies Raué

131

Graziano, P.R. (2009) ‘Choosing welfare or losing social citizenship? Citizens’ free choice in recent Italian welfare state reforms’, Social Policy & Administration, 43/6, 601–16. Graziano, P. and M. Jessoula (2011) ‘Eppur si muoveva: The Italian trajectory of welfare state reform from “rescued by Europe” to Euroscepticism’, in P. Graziano, S. Jacquot and B. Palier (eds) Europa, Europae: The EU and the Domestic Politics of Welfare State Reform (Basingstoke: Palgrave). Gualmini, E. (1998) La politica del lavoro (Bologna: Il Mulino). Ichino, P. (1982) Il collocamento impossibile (Bari: De Donato). INPS (2008) L’indagine censuaria sugli interventi e sui servizi sociali dei comuni (Rome: INPS). ISFOL (2000) The Vocational Training System in Italy (Rome: ISFOL). ISFOL (2004) Sistemi di welfare e gestione del rischio economico di disoccupazione (Milan: Franco Angeli). ISFOL (2008) Le procedure di accertamento dello stato di disoccupazione e di attivazione dei disoccupati nei Centri per l’impiego (Rome: ISFOL). ISTAT (2010) I consumi delle famiglie (Rome: ISTAT). Jessoula, M., P. Graziano and I. Madama (2010) ‘ “Selective” flexicurity in segmented labour markets. The emergence of mid-siders in the Italian case’, Journal of Social Policy, 39/4, 561–83. Madama, I. (2010) Le politiche assistenziali (Bologna: Il Mulino). Mayntz, R. (1999) ‘La teoria della governance: sfide e prospettive’, Rivista Italiana di scienza politica, 29/1, 3–21. Milio, S. (2007) ‘Can administrative capacities explain differences in regional performances? Evidence from Structural Funds implementation in Southern Italy’, Regional Studies, 41/4, 429–42. Naticchioni, P. and S. Loriga (2010) Short and Long Term Evaluations of Public Employment Services in Italy, Dipartimento di Economia – Sapienza di Roma and ISFOL, Discussion Paper No. 10. Newman, J. (2001) Modernising Governance: New Labour, Policy and Society (London: Sage). North, D. (1991) Institutions, Institutional Change and Economic Performance (Cambridge: Cambridge University Press). Osservatorio sull’applicazione della 328/2000 (2009) Contributo al libro verde [sul welfare], Rome. Pierson, P. (2000) ‘Path dependence, increasing returns and the study of politics’, American Political Science Review, 94/2, 251–67. Putnam, R. (1993) Making Democracy Work: Civic Traditions in Modern Italy (Princeton: Princeton University Press). Richardson, J. (ed.) (1982) Policy Styles in Western Europe (London: George Allen and Unwin). Vesan, P. and P. Graziano (2008) ‘New modes of governance: Explaining local partnership consolidation’, NewGov Policy Brief, 18. Visser, J. (2006) ‘Union membership statistics in 24 countries’, Monthly Labor Review, 129/1, 38–49.

Tomáš Sirovátka

7 The Liberal Governance of a Non-Liberal Welfare State? The Case of the Netherlands Rik van Berkel and Willibrord de Graaf

Introduction In the literature on welfare state regimes, the Netherlands has been characterized as a hybrid welfare regime. According to Esping-Andersen (1999), this is due to its mixture of social democratic and conservative welfare regime characteristics. Esping-Andersen argues that when looking at income protection systems, the Netherlands are social-democratic as its systems are universalist, provide comprehensive coverage and are generous and de-commodifying. However, these characteristics are mixed with Catholic familialism, which becomes manifest in female labour-market participation and the level of services provided to encourage participation in paid work for women. To these conservative characteristics could be added policies aimed at reducing the labour-market participation of older workers, which were especially popular in the 1980s and 1990s. One could rightfully raise the issue whether this characterization of the Dutch welfare state is still adequate after a series of welfare state retrenchments, reforms based on activation and a serious rise of the labour-market participation of women (even when often in part-time jobs) and, though more recently, of older workers. These developments have made the Dutch welfare state both less socially democratic (less universalist and generous) and less conservative (increasing labourmarket participation of women and older people, increasing childcare services). At the same time, liberal characteristics have entered Dutch welfare state arrangements (increasing selectivity and conditionality, re-commodification as a consequence of increasing responsibilities of unemployed people in the context of activation). An alternative 132

Tomáš Sirovátka

Rik van Berkel and Willibrord de Graaf

133

interpretation could be that Esping-Andersen’s characterization of the Dutch welfare state as hybrid is still valid, but that the mix of welfare state regime characteristics looks different today than, say, 20 years ago. When we focus on governance rather than substantive reforms, a rather similar development can be observed. If we relate (in a simplified way) social democratic governance regimes with a strong role of state institutions in the administration and provision of social services, conservative governance regimes with a strong role of civil society and particularly social partners, and liberal governance regimes with a strong role of the market, we may observe a decreasing role of conservative (especially corporatist) arrangements, an increasing role of the market and a rather ambiguous development where the role of the state is concerned. In this chapter, we will elaborate this argument as follows. We will start by briefly presenting the main actors and agencies involved in administering income protection schemes and activation services for the unemployed at the end of the 1980s (the start of active welfare state reforms in the Netherlands) and today (2010). This overview already reveals some of the major governance reforms that have been taking place in the 1990s and 2000s. The third section will analyse these reforms in more depth, followed, in the fourth section, by an interpretation of these reforms in terms of shifts in governance regimes. In the fifth section, we will discuss some of the effects of these reforms. The chapter ends with some conclusions.

Governance reforms: a brief overview The 1980s was a decade of significant welfare state reforms in the Netherlands. Although these reforms mainly focused on welfare state retrenchments, the first experiments with what we know now as activation programmes took place. These were particularly focused on young unemployed people dependent on social assistance, in a context of youth unemployment rates above 10 per cent throughout the 1980s (Spies and Van Berkel, 2001). Activation gained momentum at the end of the 1980s, when an influential report by the Scientific Council of Government Policy (WRR, 1990) was published, which identified the low labour-market participation rate of various groups in Dutch society rather than high unemployment as the main threat to the sustainability of the Dutch welfare state. At that time, which can be denoted as the starting point of active welfare state reforms in the Netherlands, the implementation structure for

Tomáš Sirovátka

134

The Liberal Governance of a Non-Liberal Welfare State?

Table 7.1 Implementation structure for income protection and activation in the Netherlands (late 1980s and 2010) Late 1980s

2010

Unemployment benefit administration

Social Insurance Bodies

UWV

Social assistance administration

Municipal welfare agencies

Municipal welfare agencies

Activation insured

PES

UWV WERKbedrijf + private providers

Activation uninsured

PES

Municipalities + private and public providers

income protection and employment services for the unemployed looked as shown in Table 7.1. Unemployment benefits were administered by various sectorally organized Social Insurance Bodies (bedrijfsverenigingen). The Social Insurance Bodies – which were also responsible for the administration of Disability and Sickness benefits – were governed by the organizations of employers and employees. The second tier of the Dutch income protection system, social assistance for the unemployed who are not or no longer entitled to unemployment benefits, was administered by municipal welfare agencies, though fully funded by national government. These agencies operate under the responsibility of local authorities. Employment services for the unemployed (in those days mainly consisting of job mediation, a vacancy database and training programmes) were provided by the PES, which was then organized as a so-called directorate-general of the Ministry of Social Affairs and Employment. Twenty years on, this implementation structure has changed significantly. Of the institutions responsible for administering income protection programmes and employment services in the late 1980s, only the municipal welfare agencies still exist. These agencies are now responsible not only for the administration of social assistance, but also for the activation of social assistance recipients. Local welfare agencies can either provide activation services in-house or subcontract them to public or private organizations operating in a quasi-market. The Social Insurance Bodies were abolished in the 1990s as the result of a process of marginalizing the role of social partners in the administration of social insurance schemes. Administering social insurance now is the responsibility of one single benefit agency, an independent administrative body operating under supervision of the Ministry. This agency is also

Tomáš Sirovátka

Rik van Berkel and Willibrord de Graaf

135

responsible for activating unemployment benefit recipients, for which they are obliged to purchase services from private providers. The PES no longer exist as a separate organization. During the 1990s, the responsibilities of the PES in the area of activation were gradually transferred to the benefit and social assistance agencies, giving these agencies more control over service provision by the PES. The department of the PES originally responsible for activation was eventually privatized when the market for activation services was established in 2002. The remaining parts of the PES were reorganized into Centres for Work and Income: one-stop agencies for unemployed people irrespective of the nature of income support they receive. In 2009, these Centres were merged with the benefit agency UWV, and are now known as UWV WERKbedrijf (UWV Work Company). This short overview already highlights two important governance reform trends in the Netherlands: a strong trend towards marketized provision of activation services and a decline in corporatist governance, which has effectively abolished the role of social partners. We will now look at some of the governance reforms briefly sketched above in more detail.

Main trends in governance reforms An important trigger for many of the governance reforms summarized in Table 7.1 were the outcomes of various parliamentary inquiries and evaluation studies that took place in the first half of the 1990s and involved the functioning of the Social Insurance Bodies, the municipal welfare agencies and the PES. In a recent study of these reforms (Van Gestel et al., 2009), the authors characterize the avalanche of governance reforms during the 1990s and 2000s as a reform swamp: no master plan was guiding the reforms, they took place in rapid succession, often in an attempt to correct disappointing results of earlier reforms, and reveal a whimsical pattern when it comes to the principles underlying the reforms. Exit the social partners The early 1990s were characterized by contradictory developments where the role of social partners in the governance structures of social insurance schemes and employment services was concerned, a first example of the whimsical reform pattern mentioned above. In 1991, the role of social partners in the PES was strengthened which, among others, was intended to promote the commitment of employers in

Tomáš Sirovátka

136

The Liberal Governance of a Non-Liberal Welfare State?

the labour-market integration of the unemployed and, more specifically, the long-term unemployed. At the same time, social partners were accused of using their position in the social insurance system to frustrate national government’s attempts to activate the Dutch welfare state. This criticism particularly focused on the ways in which social partners abused – as it was called in the political debate – their position in the social insurance system by using the Disability Benefit system as a labour-market exit route for redundant (and often older) workers. As we saw before, this eventually resulted in the exclusion of social partners from the implementation structure of social insurance: the Social Insurance Bodies were replaced by a new agency (the UWV) in which social partners had no role whatsoever. One of the consequences of this was that the Ministry strengthened its control over benefit administration processes, and especially over the procedures for testing eligibility for Disability Benefits. With social partners no longer involved in the management of the benefit agency, the government hoped to be able to reduce the numbers of Disability Benefit recipients. The increasing involvement of social partners in the PES during the 1990s lasted only for several years: with the transfer of responsibilities for activating the unemployed to the benefit agency and the local welfare agencies, the privatization of the PES department responsible for activation and the marketization of activation services, their involvement came to an end less than a decade after it was installed. Currently, the role of social partners is limited to their participation in a national advisory body concerning social security and activation policies, in which they participate together with representatives from the municipalities. Decentralization and (re-)centralization Reforms of the benefit agency and the PES are a complex mix of both centralization and decentralization processes. This complex mix seems to reveal that urgencies arising in specific administrative contexts rather than a master plan guided the design of the reforms. The establishment of the new benefit agency itself can be interpreted as a process of centralization, as it replaced the social insurance bodies with one national agency, marginalized the role of social partners in benefit administration and strengthened political control over the agency’s management. At the same time, administrative decentralization took place in the benefit agency and the PES (merged in 2009, which in itself can be seen as another example of a process of centralization) by giving these agencies the status of an independent administrative body (Zelfstandig

Tomáš Sirovátka

Rik van Berkel and Willibrord de Graaf

137

BestuursOrgaan or ZBO). This is not an exceptional case: many public agencies in the Netherlands have a similar status nowadays. Decentralization trends are most clearly visible in the area of activation. Evidently, the introduction of a quasi-market for activation services is a form of what in Chapter 1 was called competitive decentralization. Political decentralization and deregulation affected the activation of both unemployment benefit and social assistance recipients. In the 1990s, activation used to be regulated through national activation programmes for more or less clearly demarcated target groups of unemployed people. This changed rather drastically during the first decade of the twenty-first century. Nowadays, national activation programmes are completely non-existent: the benefit agency and local welfare agencies have full policy autonomy concerning the nature of activation services they want to provide or to purchase, as well as concerning the activation offers made to specific target groups of unemployed people. An interesting exception in this context concerns a recently introduced (2009) Act on the integration of young (under 27) unemployed social assistance recipients. This Act obliges municipalities to make young unemployed people an activation or training offer. Here we see another centralization trend in an overall context of decentralization and deregulation: the Act reduces room for local policy making where young unemployed social assistance recipients are concerned. Another example of decentralization concerns the fact that UWV has decentralized the process of purchasing activation services from external providers. This used to take place at the national level, but has been regionalized. One of the triggers of this decentralization process was the introduction of so-called Individual Reintegration Agreements (IRO), which give unemployed people a voice in the design of their activation plan and the selection of the provider of their activation services (Sol and Westerveld, 2005).1 The introduction of these IROs also resulted in a considerable increase of the numbers of external providers contracted by the regional offices of the benefit agency (Schrijvershof et al., 2008). For local welfare agencies, decentralization and deregulation not only involved the nature of activation, but also the organization of service provision. When the market for activation services was established in 2002, both the UWV and the local welfare agencies were obliged to outsource activation services for the unemployed.2 However, soon after the introduction of a new Social Assistance Act in 2004, the obligation for local welfare agencies to outsource activation services was abolished. Nowadays, local welfare agencies are free to decide whether they provide activation in-house or purchase it from external (public or private)

Tomáš Sirovátka

138

The Liberal Governance of a Non-Liberal Welfare State?

providers. UWV is still obliged to outsource the provision of activation services. A final development that stimulates decentralization is related to the importance attached to regional/local cooperation between municipal local welfare agencies on the one hand and UWV (or, after the merger with the PES, UWV WERKbedrijf) offices on the other. We will return to this issue later in this chapter, but in terms of decentralization this implies that within the UWV organization a tension exists between central steering and management and regional/local flexibility to organize forms of inter-agency cooperation tailored to the variety of local or regional contexts in which this cooperation has to take shape. Marketization The Netherlands is well known for the radical marketization in the area of activation and employment services that was introduced in the early 2000s. The delegation of service provision to private so-called reintegration companies and the abolishment of publicly provided activation (with some minor exceptions, see below) followed a period in which various attempts were made to intensify and improve publicly provided activation. Initially, the PES was responsible for activating the unemployed, but in the late 1990s the role of principal in activation was transferred to the benefit agency for unemployment benefit recipients, and to the local welfare agencies for people dependent on social assistance, thus giving the PES less room for manoeuvre in providing services. So when marketization was introduced, the benefit and welfare agencies became purchasers of services on the market. The public remains of the PES, from then on termed Centres for Work and Income, were by law made responsible for the provision of some specific employment and activation services targeted at the unemployed (unemployment benefit as well as social assistance recipients) during their first six months of unemployment, such as direct job mediation or support in job applications. Outsourcing of all other activation services was made obligatory, which implied that the benefit agency and the local welfare agency were not allowed to provide activation services in-house. Of course, this regulation, which is quite unique in a European perspective, implied a significant boost for the development of the service market. According to an evaluation study of the marketization of activation services (Mallee et al., 2006), there were about 1800 private providers contracted by the public purchasers in 2006. One of the characteristics of the Dutch activation market is that it is minimally regulated by the state. Unlike other countries (such

Tomáš Sirovátka

Rik van Berkel and Willibrord de Graaf

139

as Denmark (Bredgaard and Larsen, 2008); see also Chapter 8 on Switzerland in this volume), successive Dutch governments have relied on the self-regulation powers of the activation service market: national regulation is limited to the rules concerning the process of putting work out to tender. One of the initiatives that private providers have taken is the establishment of a branch association that, among others, developed a quality mark. At the moment, about 165 reintegration companies have joined this association (www.boaborea.nl). Although this is a relatively small proportion of all reintegration companies that are active in the market, one should bear in mind that the size of the companies and the volume of services provided by them vary enormously. Over the years, the functioning of the activation market has changed significantly. Initially, the benefit agency and local welfare agencies generally referred unemployed people to private providers, delegating most decision making concerning individual activation interventions to the providers. During the early years of marketization, the numbers of unemployed participating in activation programmes increased significantly (see below), but the results were generally considered disappointing (Mallee et al., 2006). The public purchasers were quick to blame the private providers for this, but as De Koning (2009) rightfully argued, this analysis is too simple, as it ignores the role of the purchasers in the failure of the market (see also Groot et al., 2006). According to De Koning, this role was characterized by a lack of expertise and professionalism. Purchasers not only had little knowledge of the market (which in itself is understandable, of course, in the context of a completely new market), but also had little insight into their clients’ needs, which is a necessary condition to contract providers on the grounds of the nature and quality of services they offer. This is one of the reasons why market competition focused on price rather than quality, which made private providers reluctant to innovate, invest in quality and develop adequate services for those most difficult to employ (Mallee et al., 2006). Over the years, the benefit and welfare agencies have changed the ways in which they organize service provision. The disappointing results of marketization stimulated this, but changes in regulation and legislation did so as well. The introduction of the Individual Reintegration Agreement forced the benefit agency to deal with private providers in a different way, and the local welfare agencies saw their autonomy in organizing service provision increase as a consequence of the abolishment of obligatory outsourcing. In addition, the agencies gradually increased their expertise in the area of activation, at both the management and street level (Buurman, 2008; Van Berkel

Tomáš Sirovátka

140

The Liberal Governance of a Non-Liberal Welfare State?

et al., 2010). Although under different conditions (for example, the IRO and obligatory outsourcing for the benefit agency, autonomy in organizing service provision for the local welfare agencies), we can observe some clear trends in the ways in which the agencies tried to strengthen control over, and their role in, activation processes. First, less use is made of performance contracts such as no-cure no-pay or no-cure lesspay contracts (RWI, 2008). Instead, the benefit agency and local welfare agencies have intensified monitoring of clients who are referred to private providers, and interfere more strongly in the content of individual activation services. In addition, they buy more diversified, individualized and modular – that is, specific components of activation trajectories rather than full trajectories – services, and their street-level workers are more involved in determining the nature of individual activation processes, instead of leaving these decisions to external providers. In the benefit agency, this is also stimulated by the IRO: unemployment benefit recipients have to discuss their reintegration plans with their coach at the benefit agency, who needs to approve of the plan. The proportion of activated unemployment benefit recipients making use of the IRO is considerable: in 2007 this was close to two-thirds (Schrijvershof et al., 2008). In local welfare agencies, street-level workers may also play a role in deciding whether social assistance recipients are serviced in-house or referred to external providers. A mix of in-house and externally provided activation has become more common. In 2005, 56 per cent of the municipal funds spent on activation services went to private providers (Divosa, 2007). In 2008, this was 22 per cent: 28 per cent was spent on in-house services, and for the remaining 50 per cent other municipal agencies were contracted (RWI, 2008). So, especially in the area of the activation of social assistance recipients, a process of demarketization has taken place in recent years, increasing the importance of the public provision of activation. New public management The introduction of a market for publicly funded activation services is a clear example of how new public management instruments have been introduced in the Netherlands. The introduction of strict dividing lines between the responsibilities of principals and agents, and the implementation of the steering-not-rowing principle where the role of public agencies in activation is concerned, were the cornerstones of the introduction of the market. Nevertheless, as shown above, currently a more modest form of these principles is being practised that has made the dividing lines between purchasers and providers less clear. Other forms

Tomáš Sirovátka

Rik van Berkel and Willibrord de Graaf

141

of contracting private providers rather than performance contracts have become more common, and in the area of municipal activation we can even see a process of demarketization. New public management, however, lies not only at the basis of how relationships between public and private actors are structured, it has also come to shape relationships between national government on the one hand and public agencies (the UWV) and local authorities and their local welfare agencies on the other. In the context of municipal activation, one of the first examples of new public management was the introduction – before the marketization of activation – of performance agreements that were concluded between the national state and the municipalities (Van Berkel, 2006). These agreements made funding for activation services dependent on the numbers of social assistance recipients participating in activation programmes, and the proportion of participants finding a job after participation in activation. A more far-reaching form of new public management steering was introduced with a new Social Assistance Act in 2004. This Act not only brought the decentralization and deregulation discussed before, but also a new regime for financing local social assistance expenses. Until then, all local social assistance expenses were reimbursed by national government. In other words, the funding system did not provide financial incentives for municipalities to decrease the numbers of social assistance recipients which, with the growing popularity and dominance of the incentive paradigm (Van der Veen and Trommel, 1999) in public administration, was interpreted as highly problematic. The new financing regime is a clear example of this incentive paradigm, as it considerably strengthened municipal financial responsibilities. Nowadays, municipalities receive a fixed yearly budget for financing social assistance expenses: if they spend less, they can add the surplus to municipal resources; if they spend more, they will have to find additional financial resources in the municipal budget. The new financing regime functions as an implicit centralization process, in the sense that it strongly influences autonomous local decisionmaking processes by focusing policy choices and decisions increasingly on a reduction of social assistance dependency – which is exactly what national government intended to realize with the new financing regime. In the case of the merged PES and benefit agency, new public management has taken different forms. Each year, performance agreements are concluded between the Ministry and UWV WERKbedrijf. These agreements include a variety of performance areas, such as preventing benefit entry, correct and timely provision of benefits, client orientation,

Tomáš Sirovátka

142

The Liberal Governance of a Non-Liberal Welfare State?

integrated service provision and efficiency. The consequences for not reaching the performance targets are not as clear and direct as is the case for the municipalities. One of the reasons for this is that whereas social assistance expenses are financed through taxation, unemployment benefits are paid from premiums paid by employers and employees, which cannot be used as a steering instrument by the state. Funds for the activation of unemployment benefit recipients are paid by the state, however, and thus offer national government more steering (and sanctioning) opportunities (De Koning, 2009). Inter-agency cooperation The introduction of marketization was part of a larger reform programme (the so-called SUWI-reform, Structure for the administrative organization for work and income) which also aimed at promoting inter-agency cooperation between the public agencies involved: the benefit agency, the PES (then named the Centre for Work and Income) and the municipal welfare agencies. The Centres for Work and Income were supposed to act as a one-stop shop agency, especially for newly unemployed people. CWI found itself in a complex position: on the one hand, it had to cooperate with the nationally managed benefit agency; on the other, with the over 400 Dutch municipalities and their local welfare agencies. In the cooperation process, the local welfare agencies demanded flexibility from the CWI to establish decentralized, tailor-made forms of collaboration which, for the CWI agencies, turned out to be difficult to realize (Blommesteijn et al., 2005; Fermin et al., 2006; PWC, 2006). Not surprisingly, against this background the municipalities were unsatisfied with the quality of services provided by the CWI, which involved – collecting the information from newly unemployed people on the basis of which decisions on income protection entitlements should be made; – determining newly unemployed people’s labour-market distance, which should support the benefit agency and local welfare agencies in making decisions about activation offers; – offering newly unemployed people (irrespective of whether they were entitled to unemployment benefits or to social assistance) some basic employment and activation services during the first six months of unemployment. However, lack of resources often implied that CWI did not offer much support to the newly unemployed.

Tomáš Sirovátka

Rik van Berkel and Willibrord de Graaf

143

What made municipalities especially critical about the functioning of CWI was the fact that they – after the introduction of the new financing regime for social assistance – were dependent on the quality of services provided by CWI without having strong steering instruments to influence CWI service provision (Blommesteijn et al., 2005). For example, if CWI did not provide adequate services to newly unemployed social assistance recipients (which, according to municipalities was often the case) and if this would result in longer periods of social assistance dependency, not the CWI but the municipalities would bear the financial consequences of prolonged (and thus higher) social assistance payments. Despite this criticism, a group of municipalities started experimenting with further-reaching forms of local cooperation between UWV, CWI and local welfare agencies in so-called showroom experiments (AKO, 2007). These experiments were based on several principles, such as one contact person for individual unemployed people as well as for employers, joint purchasing of activation services and one IT system. In some cases, these experiments also implied the integration of tasks of some frontline workers in the public agencies,3 so that they could serve newly unemployed people irrespective of the kind of income support to which the unemployed were entitled. Since the merger of UWV and CWI, these experiments have inspired a new wave of developments in inter-agency cooperation. UWV/CWI (UWV WERKbedrijf) and local welfare agencies have to co-locate frontline services in so-called Work Plazas, where forms of cooperation based on the showroom principles should be implemented. Milestones have been defined for inter-agency cooperation that include the showroom principles mentioned above, as well as some additional ones, such as one management team for each Work Plaza, one joint yearly action plan and one counter for all clients. Progress in realizing these milestones is monitored regularly, showing a quite diverse state of affairs (see http://landkaart.samenvoordeklant.nl). A related form of inter-agency cooperation concerns the creation of (inter-)municipal Youth Counters, specifically aimed at young people. In most municipalities, these Counters join the services of UWV WERKbedrijf, local welfare agencies and schools for professional education, but they may also include the services of, for example, youth care, agencies providing assistance with debts, or the registration centre for school drop-outs. It is too early to reach conclusions about the successfulness of these new initiatives in inter-agency cooperation. Problems in promoting

Tomáš Sirovátka

144

The Liberal Governance of a Non-Liberal Welfare State?

cooperation between nationally managed agencies and agencies that operate under the responsibility of local authorities are likely to persist. Occasionally, proposals have been launched to make the municipalities fully responsible for the activation of all unemployed. This would imply the adoption of the Danish model: Denmark introduced a full municipalization of activation in 2009 (Bredgaard and Larsen, 2009). Until now, however, these proposals have not been followed – and one could argue that the merger of UWV and CWI has not made the full municipalization of activation more likely.

A shift towards a new governance regime? When we interpret the governance reforms analysed above in terms of governance regimes, some conclusions can be drawn. First of all, corporatist structures have been eroded in the Dutch policy areas of social insurance and employment/activation services. In the area of employment and activation services, employers are now seen as clients (besides the unemployed) of the local or regional Work Plazas. Secondly, the successors of the traditional corporatist structures are different in the areas of social insurance and activation. In social insurance, a mix of (in Considine’s words; see Chapter 1) procedural and corporate governance has been introduced. UWV WERKbedrijf now functions as a formally independent agency under the responsibility of the Minister of Social Affairs and Employment, which was seen as a necessary step to re-install the primacy of politics. The agency is steered through yearly action plans that specify critical performance targets. In the provision of activation and employment services, the market governance model has become dominant. This was especially the case in the early years of marketization, when outsourcing was obligatory for both the benefit agency and the local welfare agencies. Thirdly, several trends could be observed that point to a development from a radical market governance model towards a more moderate market governance model. It should be emphasized, however, that this is not the result of explicit government regulation aimed at reducing the role of the market,4 but rather of the ways in which the benefit and local welfare agencies act as purchasers on the market. A demarketization trend is most clearly visible where the activation of social assistance recipients is concerned, as local welfare agencies have strengthened the role of in-house production of activation services. But public responsibilities have increased in activating unemployment benefit recipients as well, in the sense that the public purchasers strengthened their roles in individual activation processes.

Tomáš Sirovátka

Rik van Berkel and Willibrord de Graaf

145

Against the background of these developments, one could argue that the borderlines between market governance and network governance have become blurred. In some cases, relations between purchasers and providers are not merely based on strict competition principles, but involve forms of cooperation as well. This seems most clearly to be the case in the provision of services for groups of unemployed most difficult to employ, and is probably related to the complexity of needs that should be met on the one hand and the variety of actors involved in meeting these needs on the other. All in all, the early 2000s saw a clear shift away from corporatist towards market and corporate governance. The fate of procedural governance was more ambiguous: it was strengthened in some respects, but weakened in others. In the years following this neo-liberal dominance, a gradual mitigation seems to have occurred, especially where marketization is concerned. On the one hand, one could interpret this as a learning process: problems of (specifically long-term) unemployment and social inclusion are more complex, wicked and unmanageable than the ‘naive thought that ( . . . ) “the market would do the job” ’ (De Koning, 2009, p. i) presupposed. In his evaluation of Dutch governance reforms, De Koning (2009, p. i) concluded: ‘Although a mixed system may be seen as neither fish nor flesh, it might work better than a system developed from one leading principle’. On the other hand, not only pragmatic but also ideological considerations seem to play a role in an erosion of the support for neo-liberal thinking in managing public issues. In the Dutch public and political debate, trust in the virtues of the market has come under increased criticism – evidently, the financial and economic crisis has been an important catalyst in this process. It is too early to say whether this will lead to a sustainable move away from a neo-liberal governance regime, but compared with the early 2000s the reputation of neo-liberal governance reforms has certainly been damaged.

Effects of governance reforms Making statements about the effects of governance reforms is a rather precarious enterprise, not only because a large variety of effects can be distinguished which in itself may be an object of political controversy, but also because the quality of available research data leaves a lot to be desired. This is not simply (and maybe not even primarily) a problem of the quality of evaluation research, but rather of the conditions under which research takes place that make it hard to reach reliable and

Tomáš Sirovátka

146

The Liberal Governance of a Non-Liberal Welfare State?

robust conclusions. In this section, we shall discuss some of the effects of governance reforms. In terms of the outcome effects mentioned in Chapter 1, most evaluation studies conclude that it is impossible to say anything reliable about the impact of marketization on labour-market entry. One of the main reasons for this is that no comparisons can be made with the period prior to marketization (De Koning, 2009). Of course, nowadays, comparisons could be made between public and private activation in the local context, but as far as we know, no such studies are available. Data are available concerning the percentage of people finding a job after participation in activation. For example, Buurman (2008) refers to a study that found that 67 per cent of unemployment benefit recipients who started activation in 2004 had found a job two years later. For social assistance recipients, this percentage was 19 per cent.5 There is also evidence that these figures have improved in the 2002–4 period. During the same period, the sustainability of labour-market reintegration has slightly improved for former recipients of unemployment benefits (Ministerie SZW, 2008). However, these figures refer to gross effectiveness, whereas most evaluation researchers would prefer data on net effectiveness. The (far more scarce) studies focusing on net effectiveness found only modest effects of activation. For example, Groot et al. (2008) found that the probability that social assistance recipients will find a job within 1.5 years increases with 2.5 percentage points when they are offered activation services in the first year of social assistance dependency; when an activation offer is made after one year, this figure drops to 1.5 percentage points. According to this study, net effectiveness is somewhat lower for people receiving unemployment benefits. But again, all these figures tell us nothing about the effectiveness of marketized versus publicly provided activation. In the context of a discussion of outcome effects of marketized activation, the impact on the effectiveness of activation of the IROs for unemployment benefit recipients is interesting. The IRO evaluation study compared placement rates after IRO trajectories with placement rates after regular (that is, non-IRO) trajectories. These data showed that placement rates of IRO trajectories are 20–40 per cent higher than the percentages realized with regular trajectories (Schrijvershof et al., 2008), but of course these are gross, not net effectiveness, figures. In addition, costs of IRO trajectories are on average 30 per cent higher than regular trajectories. At the same time, one could argue that savings on benefits are likely to be higher than the extra costs of activation, since the probability that IRO trajectories result in a labour contract for 12 months

Tomáš Sirovátka

Rik van Berkel and Willibrord de Graaf

147

or more is higher than is the case with regular trajectories (Schrijvershof et al., 2008). Anyhow, at first sight one may conclude that in the context of marketized activation, increasing voice and choice of unemployed people seems to be effective. An evaluation study of the showroom experiments focused on promoting local inter-agency cooperation showed some (though not very robust) indications that these experiments reduced benefit dependence periods and reintegrated the unemployed into the labour market more quickly (Ministerie SZW, 2008). An outcome effect not mentioned in Chapter 1, but relevant in the Dutch debates on the effectiveness of governance reforms, is the reduction of benefit or welfare dependency. This indicator of effectiveness does not coincide with labour-market entry, as people may have reasons for not or no longer claiming income support other than labour-market entry. Kok et al. (2007) carried out an econometric study in which they tried to assess the impact of the new financing system in social assistance on the volume of social assistance recipients. Their conclusion is that, in the period 2004–6, the new financing system caused a reduction in the numbers of social assistance recipients of 4 per cent, which amounts to 13,300 people. They expected that this percentage will rise to about 14 per cent (compared with 2003) in 2010. The researchers argue that this reduction is the consequence of both a lower number of new social assistance entries and an increased number of social assistance exits. More specifically, they found that women and young people in the age group 25–34 years of age were less likely to become dependent on social assistance. As far as social assistance exits are concerned, the effect of the new financing regime was strongest for social assistance recipients in the age group 35–44 years of age. Of course, it is not easy to interpret these figures in a straightforward way. On the one hand, one could argue that local activation services are more successful for younger social assistance recipients; on the other, it may reveal that local welfare agencies mainly prioritize younger people in providing services to the unemployed. Below, we will shed some more light on this issue. Focusing on output effects, various studies have pointed out that marketization has contributed to an increase in the numbers of unemployed people participating in activation (Mallee et al., 2006), even though the exact impact of marketization on the volume of participation in activation is hard to isolate. Table 7.2 shows that during the period 1999–2005, during which marketization and several other governance reforms were introduced, the numbers of unemployed participants in activation increased considerably.

Tomáš Sirovátka

148

The Liberal Governance of a Non-Liberal Welfare State?

Table 7.2

Volume of started activation trajectories (1999–2005)

Year

Unemployment benefit recipients

Social assistance recipients

1999 2000 2001 2002 2003 2004 2005

12,306 18,834 23,510 n/a 28,925 52,150 46,573

n/a n/a 11,000∗ 64,000 70,000 109,000 98,000

∗ Second

half-year period. Sources: Van den Berg et al. (2006); RWI (2008).

Another output effect concerns the content of activation services. Governance reforms have not been without consequences for the content of activation services. Many local welfare agencies now focus on activation services that promote quick labour-market reintegration: according to Divosa (2009), municipalities invest less in education and social activation (activation programmes that aim at promoting social participation in unpaid work, mainly focused at the groups of clients most difficult to reintegrate) and more in job mediation and guidance activities, although there is evidence that some municipalities started to consider education of more importance for those dependent on social assistance for less than one year (RWI, 2008). In addition, so-called Work First programmes seem to become increasingly popular among municipalities, especially for the activation of new social assistance claimants. About 88 per cent of municipalities will have implemented Work First projects by now (Divosa, 2009). The number of municipalities where Work First is used has increased, and as Table 7.3 shows, new Table 7.3 Target groups of Work First (percentage of municipalities in which the named target group is the main target group of Work First programmes, 2005–6) Target group

2005

2006

17 31 20 33

12 46 27 15

Young recipients New recipients Other None Source: Divosa (2006).

Tomáš Sirovátka

Rik van Berkel and Willibrord de Graaf

149

recipients as well as young recipients (of whom most will be new recipients) are the main target groups of these programmes. Work First approaches imply that people are placed in work activities immediately when claiming social assistance. For those already receiving social assistance, the diversity of activation programmes has increased. Two remarks should be made in this context. First, programmes are focused more exclusively on participation in the regular labour market for these groups as well. Other outcomes (participation in subsidized work or in unpaid forms of work) are seen nowadays as stepping stones towards regular labour-market participation, whereas in the past they could also – for certain subgroups of social assistance recipients – be ends in themselves. Secondly, the profile of programmes has changed in the sense that we see more behaviourdependent approaches: many municipalities nowadays develop specific programmes for what they see as unwilling or unmotivated client groups. Another output effect of governance reforms concerns the activation of the most vulnerable groups of the unemployed, many of whom are long-term unemployed receiving social assistance. Improving services for these groups was one of the motives of many of the reforms discussed before: marketization, decentralization and inter-agency cooperation should benefit these groups by making services more integrated, tailor made and responsive to their needs. Nevertheless, there are indications that the reforms have done little to improve the situation for the most vulnerable unemployed: their situation in terms of access to effective activation services seems to have worsened. Evaluations of the marketization process show that private service providers find it hard to develop adequate services for this group (for example, De Koning, 2009). The new funding regime for social assistance made municipalities inclined to focus on the easiest to reintegrate: 63 per cent of local welfare agencies participating in a monitor study reported in 2004 that they prioritized the easiest to reintegrate as well as new social assistance claimants; only 1 per cent prioritized the difficult to employ and longterm social assistance recipients (Divosa, 2005). The new funding regime also affected the use of more or less permanent subsidized jobs as an activation instrument, which had been created during the 1990s for the hard to employ. When national funding of these jobs was abolished with the 2004 Social Assistance Act, municipalities started to reduce the numbers of permanent subsidized jobs, because they heavily burden local activation budgets. In 2004, they spent 55 per cent of their activation budgets on these jobs (Divosa, 2006) and in 2008 they spent 34 per cent

Tomáš Sirovátka

150

The Liberal Governance of a Non-Liberal Welfare State?

(Divosa, 2009). Although detailed figures are not available, it is likely that some of the people who found employment in subsidized jobs have again become unemployed. In other words, the unemployed people that are most in need of support in order to improve their employability, labour-market or social participation, seem to have profited least from the governance reforms that have been introduced in the Netherlands.

Conclusion The Netherlands have witnessed far-reaching governance reforms in the policy areas of income protection and activation for the unemployed. As the effect studies we discussed illustrated, they have strengthened the transformation towards an active Dutch welfare state in some respects, but not in others: the number of unemployed people participating in activation has increased impressively, but in particular the activation of long-term unemployed people is under pressure from the strong emphasis in activation on (quick) labour-market reintegration. How this will develop in the future is hard to predict. On the one hand, further initiatives in strengthening local inter-agency cooperation may have a positive impact on the quality of services for the long-term unemployed. Although the new government that entered office in the autumn of 2010 has not announced new governance reforms, the Conservative– Liberal Party (VVD, the largest party in parliament) wants to abolish the UWV WERKbedrijf and make activation a full municipal responsibility. In other words, the debate about municipalization of Dutch activation has not yet ended, probably. In addition, the new government – that in general does not have a very positive attitude towards activation services – wants to use activation services selectively, as it is called in the coalition agreement, for vulnerable people only. On the other hand, the new government intends to introduce considerable cuts in the budget available for activation services. Furthermore, municipalities have been confronted recently with a significant increase in the numbers of social assistance recipients: during the first six months of 2010, the number of social assistance recipients increased by 7.5 per cent (the number of unemployment benefit recipients decreased slightly in this period). There is no doubt that these developments will affect the services provided by UWV WERKbedrijf and local welfare agencies, the prioritization of groups in activation and possibly the inter-agency cooperation at local level. But probably more importantly, the Netherlands may experience a shift towards a more liberal approach in activation, which puts less emphasis on activation services to support people’s return to

Tomáš Sirovátka

Rik van Berkel and Willibrord de Graaf

151

the labour market, and a stronger emphasis on financial incentives (that is, less generous income support).

Notes 1. Interestingly, a similar regulation to the Individual Reintegration Agreement does not exist for social assistance recipients, although local welfare agencies may decide to introduce similar instruments for (groups of) social assistance recipients. However, given the resources spent on this instrument (1.4 million ¤ in the 73 municipalities participating in a monitoring study carried out in 2008 (Divosa, 2009)), its quantitative significance seems to be modest. 2. For municipalities, this involved 70 per cent of the budget available for activation services. 3. Until the merger of UWV and CWI, this would involve frontline workers employed by UWV, CWI and local welfare agencies. 4. Abolishing the obligatory outsourcing for local welfare agencies was intended to decentralize decision making regarding the organization of service provision, not to push back the role of the market in service provision, even though that was one of its consequences. 5. Buurman (2008) underlines that, for administrative reasons, these figures are not fully comparable.

References AKO (2007) Eindevaluatie toonkamers. Eén loket voor werk en inkomen. Samenvatting (Amsterdam: AKO). Blommesteijn, M., M. van Gent, C. Groot, J. Soethout and R. van Waveren (2005) Gemeenten en de WWB I. Geprikkeld tot werken (Amsterdam: Regioplan). Bredgaard, T. and F. Larsen (2008) ‘Quasi-markets in employment policy: Do they deliver on promises?’, Social Policy and Society, 7/3, 341–52. Bredgaard, T. and F. Larsen (2009) ‘Redesigning the governance of employment policies – Decentralised centralisation in municipal jobcentres’, in F. Larsen and R. van Berkel (eds) The New Governance and Implementation of Labour Market Policies (Copenhagen: DJOeF). Buurman, M. (2008) ‘Re-integratiebeleid: wat zijn de resultaten en wat zijn ze waard?’, Tijdschrift voor openbare financiën, 2, 40–8. De Koning, J. (2009) Reforms in Dutch Active Labour Policy during the Last 20 Years: An Evaluation (Rotterdam: SEOR). Divosa (2005) WWB monitor: Een jaar Wet Werk en Bijstand (Utrecht: Divosa). Divosa (2006) WWB monitor 2006: Meer perspectief voor mensen. Twee jaar Wet Werk en Bijstand (Utrecht: Divosa). Divosa (2007) Divosa monitor 2007: Verschil maken. Drie jaar Wet Werk en Bijstand (Utrecht: Divosa). Divosa (2009) Divosa monitor 2009. Meer dan ooit. Sociale diensten en participatiebevordering (Utrecht: Divosa). Esping-Andersen, G. (1999) Social Foundations of Postindustrial Economies (Oxford: Oxford University Press).

Tomáš Sirovátka

152

The Liberal Governance of a Non-Liberal Welfare State?

Fermin, B., C. Wevers, J. Sanders and J. van der Wolk (2006) SUWI: Structuur in samenwerking? Onderzoek naar de SUWI-samenwerking rond bemiddeling en reïntegratie (The Hague: RWI). Groot I., M. de Graaf-Zijl, P. Hop, L. Kok, B. Fermin, D. Ooms and W. Zwinkels (2008) De lange weg naar werk: Beleid voor langdurig uitkeringsgerechtigden in de WW en de WWB (Amsterdam: SEO). Groot, I., D. Hollanders, J. Hop and S. Onderstal (2006) Werkt de reïntegratiemarkt? Onderzoek naar de marktwerking op de reïntegratiemarkt (Amsterdam: SEO). Kok, L., I. Groot and D. Güler (2007) Kwantitatief Effect WWB (Amsterdam: SEO). Mallee, L., J. Mevissen, J. Soethout and H. Weening (2006) De re-integratiemarkt geijkt. Evaluatie introductie marktwerking bij re-integratie van niet-werkenden (Amsterdam: Regioplan). Ministerie SZW (2008) Beleidsdoorlichting re-integratie (The Hague: SZW). PWC (2006) SUWI-evaluatie 2006. Een evaluatie van de Wet structuur Uitvoeringsorganisatie Werk en Inkomen (Wet SUWI) (Rotterdam: PriceWaterhouseCoopers). RWI (2008) Reintegratiemarkt-analyse 2008 (The Hague: RWI). Schrijvershof, C., C. Thijs, K. Visscher and L. Aarts (2008) Eindevaluatie IRO (The Hague: APE). Sol, E. and M. Westerveld (eds) (2005) Contractualism in Employment Services: A New Form of Welfare State Governance (The Hague: Kluwer). Spies, H. and R. van Berkel (2001) ‘Workfare in The Netherlands. Young unemployed people and the Jobseekers Employment Act’, in I. Lødemel and H. Trickey (eds) An Offer You Can’t Refuse: Workfare in International Perspective (Bristol: The Policy Press). Van Berkel, R. (2006) ‘The decentralization of Social Assistance in the Netherlands’, International Journal of Sociology and Social Policy, 26/1–2, 20–32. Van Berkel, R., P. van der Aa and N. van Gestel (2010) ‘Professionals without a profession? Redesigning case management in Dutch local welfare agencies’, European Journal of Social Work, 13/4. Van den Berg, M., H. Bosselaar, M. den Hoedt (2006) Terugbladeren in de Agenda voor de Toekomst (The Hague: Ministry of Social Affairs and Employment). Van der Veen, R. and W. Trommel (1999) ‘Managed liberalization of the Dutch welfare state: A review and analysis of the reform of the Dutch social security system, 1985–1998’, Governance: An International Journal of Policy and Administration, 12/3: 289–310. Van Gestel, N., P. de Beer and M. van der Meer (2009) Het hervormingsmoeras van de verzorgingsstaat. Veranderingen in de organisatie van de sociale zekerheid (Amsterdam: Amsterdam University Press). WRR (1990) Werkend perspectief. Arbeidsparticipatie in de jaren ‘90 (The Hague: Sdu).

Tomáš Sirovátka

8 Marketization in a Federal System: New Modes of Governance in Unemployment Insurance and Social Assistance in Switzerland Franziska Ehrler and Fritz Sager

Introduction In international terms the Swiss welfare state has developed relatively recently. The late establishment of national programmes does not mean, however, that there were no previously existing social security systems in place, but that these operated at the cantonal and local levels (Armingeon, 2001). The same applies to the two income insurance systems for the unemployed: first, unemployment insurance, and secondly, social assistance. Both systems are built on structures developed over time, which were originally created as a result of private initiative (Obinger, 1998). Unlike unemployment insurance, social assistance is means tested and is still managed by the cantons (SKOS, 2007). Unemployment in Switzerland was below 1 per cent until into the 1990s, and it was not until 1994 that the unemployment rate increased significantly (4.7 per cent). This in effect signalled the end of full employment in Switzerland, and the income protection systems faced their first major test. It became apparent that the preventative measures and the implementation structures relating to unemployment insurance were inadequate in the new economic context and that change was needed (Engler, 2005). Different reforms led to a shift in outlook from passive income protection to a stronger emphasis on activation (Engler, 2005). This chapter will look closely at the reforms aimed at adapting governance structures to active labour-market policies. The nature of these governance reforms in Switzerland will be analysed in terms of the various guiding principles of the new modes of governance outlined in Chapter 1. The three reforms analysed in detail concern primarily 153

Tomáš Sirovátka

154

Marketization in a Federal System

unemployment insurance, because there are no general reforms of governance in social assistance due to its highly decentralized organization. The effects of these reforms and trends will be assessed in terms of employability as the principal aim of active labour-market measures, but also in terms of service provision and treatment of clients. Our focus will therefore be on implementation structures. It should be remembered that in Switzerland, with its historically developed federal structures and powerful cantons, substantial differences in implementation may arise and federal laws are not always implemented satisfactorily (Linder, 2004; Sager and Rüefli, 2005). In order to examine these issues, the first section will describe in more detail Swiss social welfare and unemployment insurance and the current players and governance structures will be outlined. In the second part, the three lines of reform referred to above will be examined in depth with regard to processes such as centralization, marketization, cooperation, individualization, contractualization and the involvement of social partners. The third part will look at the effects of these new modes of governance.

Income protection, activation and players Switzerland has two systems of income maintenance for the unemployed: unemployment insurance that is part of the Swiss system of social insurances and means-tested social assistance. Both systems have been developed over time and there are few links between them. Accordingly, each institution has its own criteria for inclusion, target group definitions and measures. The objectives of subsistence protection and integration into employment are weighted differently, and there is no common basic political agreement on the objectives of the Swiss social system. For the unemployed themselves, the overlap between the two systems is low and there is no automatic transition between unemployment insurance and social assistance (Handbuch IIZ, 2004; Fluder et al., 2009). Unemployment insurance is organized at the national, cantonal and regional levels, while social assistance is regulated primarily in the cantons and locally. This makes cooperation rare on legislation and implementation structure, creating major problems. Due to this lack of linkage the two systems will be described separately.

Unemployment insurance Swiss unemployment insurance is part of the Swiss social insurance system and governed by federal law. It is financed mainly by employer

Tomáš Sirovátka

Franziska Ehrler and Fritz Sager

155

and employee contributions. The provisions are implemented at the cantonal and the regional levels. Apart from the self-employed, all individuals in employment are covered by obligatory unemployment insurance. The unemployed in Switzerland are entitled to 400 days’ allowance over a two-year period. Young adults newly entered the labour market and persons with long absences from the labour market because of childcare, for example, are also entitled to benefits but under different conditions (for example, only 260 days’ allowances for young adults entering the labour market). The insured parties receive 70–80 per cent of their insured salary, depending on their socio-economic status (Gerfin et al., 2005). Activation measures were introduced by the Federal Law of 1984, but as the unemployment insurance fund covered only partial amounts and the level of unemployment was low, the cantons made little use of providing activation schemes. Not until the 1990s and, specifically, as a result of the comprehensive reform of unemployment insurance in 1995, did active labour-market measures increase in importance (Engler, 2005). In 2008 around 25 per cent of all registered unemployed participated in an activation programme (SECO, 2008a). Due to its design and content, the system of Swiss unemployment insurance is categorized by the typology of unemployment regimes by Gallie and Paugam (2000) as an unemployment-centred regime with high coverage, level and duration of cover dependent on previous employment experience, and developed active labour-market policies. Income protection and activation are dealt with separately in Swiss unemployment insurance. Both responsibility for policy and administration are shared between national and cantonal institutions. Whereas income protection is mainly organized within public agencies, the provision of activation programmes and services is shared between public agencies and private partners. The following paragraph describes the national institutions before a further paragraph outlines the cantonal institutions and their missions. Financing is concentrated at the national level. The Equalization Fund acts as a collecting pool into which, in addition to employee and employer contributions, all other income of the unemployment insurance fund flows. The Equalization Fund covers all expenses of unemployment insurance (Leu, 2005). It is managed by the federal administration service for Economic affairs and supervised by a Supervisory Board made up of representatives of employers, employees, federal and cantonal institutions and science. The employee and employer contributions represent by far the largest proportion of finance for the Swiss unemployment insurance system. Unemployment benefits

Tomáš Sirovátka

156

Marketization in a Federal System

constitute the largest expenses item, while around 15 per cent of overall expenses are for administration and implementation of activation measures. In 2007, 3.2 billion ¤ (0.83 per cent of GDP) were spent on unemployment insurance (BFS, 2009). National institutions are not only responsible for administrating finance but also for implementing active labour-market policies. Whereas the service for economic affairs occupies a steering function on active labour-market mechanisms, the Supervisory board has an advisory mission, issuing guidelines on the implementation of activation measures (SECO, 2003; Leu, 2005). The distribution of benefits and the implementation of activation programmes and services are carried out by cantonal administrations. However, different institutions are responsible for benefit distribution and the implementation of active labour-market policies. The distribution of benefits is assured currently by 39 administrative organizations that have unemployment insurance funds. In addition to 26 public unemployment insurance funds, employee and employer associations with national, regional or cantonal influence hold private funds. The unemployed are free to choose between public and private insurance funds (SECO, 2008b). The Regional employment centres (REC) make up the core of the institutions created in 1995 for the task of implementing active labour-market policies. Key activities are advising jobseekers, checking and clarifying entitlement, imposing sanctions, allocating available jobs, checking employability, provision and allocation of activation measures, as well as acquiring new jobs (Egger et al., 2001). The RECs cooperate with private employment agencies on job allocation. For the provision of activation measures the RECs work in conjunction with the Logistics Offices of active labour-market policies (ALMP). Each canton has a Logistics Office of ALMP, which is responsible for the negotiation of contracts with public and private suppliers of activation measures (Engler, 2005).

Social assistance Public social assistance is organized at the cantonal and local levels. Each canton has its own social welfare legislation, but there are huge differences in organization and level of service. Unlike social security, social assistance is financed entirely from general taxation. As with legislation, the financing of social assistance is incumbent on the cantons (SKOS, 2007). In some cantons, social welfare is financed entirely by the canton, while in others the municipalities bear the entire costs. There are also various intermediate mixed forms of financing (SKOS, 2006).

Tomáš Sirovátka

Franziska Ehrler and Fritz Sager

157

Overall, around 2.5 billion ¤ (0.65 per cent of GDP) were spent in 2007 on social assistance (BFS, 2009). In order to guarantee a certain level of uniformity, the Schweizerische Konferenz für Sozialhilfe (SKOS) issues guidelines for the drawing up and calculation of social assistance payments. The SKOS is made up of representatives from municipalities, cantons and the federal authority, as well as private organizations in the social sector. While not legally binding, the SKOS guidelines have become a central benchmark in Swiss welfare assistance, and are adhered to by the federal court (Piller, 2006; SKOS, 2007). In 1998 a revision of the SKOS guidelines introduced the dual principle of income protection and activation (Hohn, 2005). This move towards activation was underlined by the 2005 revision, which introduced benefits with incentives for taking part in activation schemes (SKOS, 2005b). Social assistance, as the final social security safety net in Switzerland, is basically a needs-based system, but it also contains performance-related elements. It is needs based as a recipient of social assistance is entitled to have housing and healthcare costs paid for, as well as basic means of subsistence. In addition to these needs-based services, social welfare provides performance-related or so-called incentive-based services. These are, first, integration bonuses for welfare recipients who make a particular effort to integrate in both social and employment terms, for example by participating in a social or economic integration programme. Secondly, there are tax-exempt payments (claimants who work have part of their salary exempted from benefit calculation and so effectively have greater entitlement), which leave employed welfare recipients better off in comparison with totally unemployed welfare recipients, thereby giving them an incentive to maintain or increase their workload. Because of these supplementary benefits, social assistance is often higher than the minimum subsistence level, but for only a small number of social assistance recipients is a swift return to the labour market possible. For this reason, the social assistance activation measures also include measures intended to encourage the social and professional integration of recipients via various qualifications (SKOS, 2005a). The implementation of social assistance differs greatly between cantons. Public social assistance is controlled by the cantonal administration, but in the majority of cantons this responsibility is delegated to the municipalities. The organizational structure of social assistance depends to a large extent on the size of the municipality. In small municipalities local government decides the entitlement to social assistance benefits and a part-time civil servant is responsible for

Tomáš Sirovátka

158

Marketization in a Federal System

implementation. In this case social assistance remains limited to benefit payments (Fluder and Stremlow, 1999). In most municipalities a special commission is responsible for entitlement decisions and most large municipalities, in particular towns, have a well-developed social services department with integrated social services (advisory service, activation, and so on) for implementing social assistance (Fluder and Stremlow, 1999). Private institutions such as relief services, churches and charitable foundations also play a key role in Swiss social assistance. They are active in the area of advice, individual financial assistance and measures for economic and social integration (Pfister, 2007). However, as they act independently from public social assistance, the following discussion of governance of social assistance will concentrate on public social assistance. In the last few years a trend towards regionalization and professionalization has become apparent in social assistance (both for social services and implementation players), with the former closely linked with regionalization in other areas of social security (SKOS, 2006; Stremlow, 2008). In the Bern canton, for example, since 2002 all municipalities have been required to provide access to professional social services (Bütler Liesch, 2006). The Vaud canton provides a different example: it regionalized social services as long ago as 1997, by creating 11 regional social assistance centres (Piotet, 2006, p. 12). An evaluation of activation measures showed that in most social assistance agencies the social worker is responsible for the assignment of a client to an activation scheme or measure. Some agencies introduced a special service within the agency that is responsible for activation, and one canton introduced a separate office responsible for the activation of clients of unemployment insurance, social assistance and disability insurance (Pfister, 2009). Private institutions again play a key role in the provision of activation measures (Pfister, 2007). Gough et al. (1997) and Gough (2001) identify the Swiss social assistance regime of the early 1990s as a decentralized, discretionary relief system, this regime type being described as localized and discretionary. It is discretionary in the sense that there is no explicit right to benefits. It is also a relief system with high benefit levels linked to social work and wider kin obligations. In this system relatively few people claim assistance (Gough et al., 1997). In the meantime, a debate on activation and disincentives to work emerged in Switzerland and led to changes. People who work or attend an integration programme receive higher benefits than other claimants, but the discretionary component of the allocation of these benefits is considerable and the application varies even between the different social workers within a service (Pfister, 2009).

Tomáš Sirovátka

Franziska Ehrler and Fritz Sager

159

Social assistance in Switzerland still has no national legislation and the recipient has no legal claim to social assistance benefits. Social assistance continues to be financed mainly by the municipalities. But it can be supposed that the regionalization and professionalization trend in social assistance has reduced the scope for this discretionary aspect, as the two trends aim for a higher equality of treatment of recipients within the cantonal legislation (SKOS, 2006). The number of recipients has risen considerably since the 1990s, but with only 3.1 per cent of the Swiss population depending on social assistance benefits in 2007 it is still at a low level (BFS, 2009). The Swiss social assistance scheme remains a decentralized, discretionary relief system but with a trend towards a regionalization and a stronger emphasis on activation.

Reforms There have been three reforms to unemployment insurance and social assistance that have had an effect on governance structures. These relate primarily to unemployment insurance, due to the fact that reforms in social assistance occur first and foremost at the cantonal level and thus affect only the governance structures of individual cantons. Some national developments that resulted from this have been mentioned above. Table 8.1 provides an overview of the main governance reforms. Table 8.1

Main governance reforms in Switzerland

Year of reform

Policy field

Main objective of reform

Main reform aspects

1995

Unemployment insurance

To adjust the system of unemployment insurance to meet the demands of the new economic situation with higher unemployment rates.

(Re)centralization; professionalization; marketization

2000/ 2002

Unemployment insurance

Efficient control of the implementation of the Unemployment Insurance Act

Contractualization; marketization

2002

Social insurance and social assistance

To improve inter-institutional cooperation between unemployment insurance, social assistance, disability insurance and other related institutions

Inter-agency cooperation; individualization

Tomáš Sirovátka

160

Marketization in a Federal System

The three reforms – and, in particular the 1995 reform – can be seen as key in the change of outlook from mere income protection to activation that also took place in other fields of social security (for example, disability insurance). The reforms reflect the tendencies towards New Public Management that emerged in the Swiss Public Administration system in the mid-1990s (Lienhard et al., 2005). Subsequently, these three reforms will be looked at in more detail and the individual aspects of each reform will be discussed.

Reform of unemployment insurance 1995 – systematic change and new strategies The sharp increase in the unemployment rate from 1.1 per cent in 1991 to 4.7 per cent in 1994, with a rise in the level of the long-term unemployed (more than one year of unemployment) from 4 per cent to nearly 30 per cent of all unemployed in 1994, made reform essential (BFS, 2010). The way in which the unemployment insurance system was organized and structured meant that it was not able to meet the demands of the new economic context. While initially, the Federal Council wanted simply to reform the rules on daily allowances, the parliament decided on a fundamental change in the system. As a result, a new daily allowance regime was introduced: the cantons were required to operate RECs and to provide a minimum number of activation programmes (Engler, 2005). The reform was decided in 1995 and implemented in 1996–7. The system change was also an expression of a change in strategy from the purely passive payment of daily allowances to a greater emphasis on activation (Engler, 2005). In order to be able to implement the change in the system it was necessary to create new institutions and to reform existing ones. The structures agreed on in 1995 correspond with those described in the previous section, whereby the three institutions of Regional Employment Centres (REC), Logistics Office of ALMP and the Tripartite Committee were created. The RECs formed the core of the new structures, and replaced more than 3000 local employment agencies (Ferro Luzzi et al., 2004). They took over job placement, advising jobseekers and supervising the recipients of unemployment allowances (Engler, 2005). All cantons are obliged to set up one REC per 1000 unemployed persons, or at least one per canton (Egger et al., 2001). The RECs are financed entirely by the Unemployment Insurance Fund, whereas the local employment offices were funded by the municipalities (Engler, 2005).

Tomáš Sirovátka

Franziska Ehrler and Fritz Sager

161

A two-way re-centralization of public employment placement in the highly decentralized system of unemployment insurance resulted. First, the managerial power of discretion was transferred from the local to the regional level through the introduction of new institutions. This recentralization can be seen in the light of a process of building up capacity. Secondly, the policy responsibilities were partially shifted as a result of the greater centralization of taxation and financing at the national level. The local employment offices often did not have the resources and expert staff required to effectively fulfil their tasks (Engler, 2005). While the local employment offices, particularly in smaller municipalities, were often run on a part-time basis by a civil servant, with the introduction of the RECs it was possible to appoint 2500 members of staff within a year (Egger et al., 2001; Engler, 2005). The weekly stamp checks for the unemployed were replaced by monthly advisory sessions and, in contrast to the local employment offices, all RECs are connected to the nationally run job databank (Egger et al., 2001; Engler, 2005). The regionalization of public job placement therefore also led to its professionalization. In addition to the RECs, each canton was provided with financing for a Logistics Office of ALMP, to ensure the provision of activation programmes. Following the 1995 reforms, the cantons are required to provide a total of at least 25,000 places on activation programmes (Engler, 2005). The Logistics Office of ALMP determine, in consultation with the RECs, requirements relating to activation programmes, plan them, award contracts to service providers and evaluate performance (Engler, 2005). In this way, a quasi-market was created in which the Logistics Offices of ALMP, as public bodies representing the unemployed, purchase the activation programmes. The suppliers are public and private organizations, with the majority being in competition with one another, but which, depending on the nature of the programme, are not allowed to make any profit (Egger et al., 2000). On the basis of this separation between the service providers and the customer and the resultant competition, Swiss unemployment insurance was marketized. Tripartite commissions were created as the third new institution. In keeping with the legislation, these advise the REC and its members work within their organizations towards the provision of a sufficient range of activation measures. In this way the social partners, in addition to their already existing involvement, were also provided with a central role in the new structures.

Tomáš Sirovátka

162

Marketization in a Federal System

Service agreements between the federal authority and the cantons: New Public Management and the emphasis on outcome control The new governance structures, repositioning activation measures at the cantonal or regional level, created a classic principal-agent situation. The state secretary for economic affairs (SECO), responsible for the strategic management of public employment placement, takes the role of the principal. The cantonal employment centres, the RECs and Logistics Offices of ALMP that answer to them, being responsible for the operations, can be described as agents. The SECO directs the cantons in achieving objectives. This is difficult to monitor, since SECO has more specialist expertise of the macroeconomic situation of the labour market, while the cantons (or the RECs) specialize in available employment and demand for jobs (Egger et al., 2001). From 1997 to 1999 attempts were made to control cantonal implementation, and the RECs and Logistics Offices of ALMP, via output factors. The cantons, as explained above, were required to provide a specific number of placements on activation programmes. If a programme could not be provided for a job-seeker, the canton had to contribute more by way of daily allowances than would otherwise be the case. The newly created RECs were encouraged to place as many people in employment as possible. Both control factors led to inefficient behaviour on the part of the cantons and RECs. The first factor led to misplacements in activation programmes, so that recipients often did not take up the designated placements or terminated them after only a short time (Egger et al., 2001). A first evaluation showed the inefficiency of output control and as a consequence the SECO, on the basis of elements of New Public Management, agreed a new, goal-driven control mechanism with the cantons. The following objectives were set: reintegration as soon and on as permanent a basis as possible; avoidance of long-term unemployment; and setting limits to benefit entitlement periods. In order to guarantee the efficient working of and better results from the REC, they were to compete with the other RECs. To this end, a penalty/reward system was introduced based on four indicators (average number of days claimed; proportion of long-term jobseekers; proportion of unemployed reaching the end of their benefit entitlement period; and proportion of re-registrations within four months) (Behncke et al., 2006). If, at the end of the year, a canton was above average, not only were all of its costs covered, but in addition a bonus payment was made, which could be up to 3 per cent. If, however, a canton was more than 5 per cent below the average for three years running, it would itself have to bear an additional

Tomáš Sirovátka

Franziska Ehrler and Fritz Sager

163

share in the costs of 3 per cent. The results were published in order to increase effectiveness (Egger et al., 2001; Engler, 2005). However, the cantons were no longer required to provide a minimum range of activation measures (Engler, 2005). The system was implemented by way of a service agreement between the federal authority and the cantons at the beginning of 2000, and the contract could theoretically be rescinded by either side. To measure the indicators two different sets of data were used: the placement and labour-market statistics generated by the RECs and the statistics of the payment system of the cantonal unemployment funds. The second data set could not be influenced by the RECs (Egger et al., 2001). However, in 2002, before the initial three-year period had elapsed and the first RECs would have been liable to sanctions, the reward/penalty system was scrapped. The performance measure continued to be applied and the results published, but this did not involve any reward or penalty payments, first and foremost due to the inadequacies of the effects measurement. One-line budgets were introduced, leading to the allocation of cost ceilings for each canton which applied to the REC, the Logistics Offices of ALMP and the cantonal employment offices. Many powers held by the SECO could now be delegated to the cantons, increasing their room for manoeuvre and responsibility (Engler, 2005). This service agreement was of a limited duration and was renewed for another three years in 2006. A similar development took place in the financing and control of activation measures. Until 2003, the Logistics Offices of ALMP had to submit to the SECO a budgeted outline of planned measures to obtain funds. Since 2003 cantons can authorize activation measures for up to 3.8 million ¤ without SECO approval. In 2006 the system of oneline budgets was also introduced in the area of activation measures. This change gave cantons more liberty, as they were free to use oneline budgets that are not dependent on the number of participants or performance-related flat rates in contracts with suppliers of activation measures (Engler, 2005). These developments, as with the post-1995 reform developments of governance structures, show a trend towards New Public Management and the marketization this brings with it. There is, also, a contractualization on varying levels: first, between the federal authority and the cantons; and secondly, between the Logistics Offices of ALMP and the suppliers of activation measures. However, in both kinds of contract, the contracting parties are not of equal power, as one party (in the first case the cantons, in the second the providers of activation measures) has no get-out clause, because each is dependent on the financial

Tomáš Sirovátka

164

Marketization in a Federal System

support of the other (Bonvin and Moachon, 2007). In both instances, the restriction of the powers of the latter players is the objective, in order to ensure that they act in accordance with the specifications of the federal authority (Bonvin and Moachon, 2007).

Inter-institutional cooperation as a project to improve inter-agency cooperation Cooperation between the various social security institutions in Switzerland is difficult due to the various structures, values, organization cultures and processes that have developed over time in these systems (Handbuch IIZ, 2004). The lack of cooperation between Disability Insurance, Social Assistance and Unemployment Insurance leads to issues of institutional responsibility being seen as more important than the best possible integration of the unemployed. This is an issue particularly for the increasing number of people with specific problems, who therefore find integration difficult. In order to resolve this issue, in 2001 the cantonal governments introduced the inter-institutional cooperation project (Handbuch IIZ, 2004). The objective of inter-institutional cooperation is to improve cooperation between Disability Insurance, Social Assistance, REC and careers advisory services with the common goal of integrating the unemployed into the labour market (Handbuch IIZ, 2004). The project consists of a national programme as well as programmes originating from individual cantons (Handbuch IIZ, 2004). The implementing bodies for inter-institutional cooperation are the RECs and Logistics Offices of ALMP, the cantonal and local social assistance offices, the regional careers advisory services and the Disability Insurance Offices (Zürcher, 2002). The list of actors shows this to be an inter-institutional cooperation between different existing public agencies of the major welfare state institutions (unemployment insurance, social assistance, disability insurance) involved in activation policies. In pilot projects, the cantons developed various modes of cooperation. These range from inter-agency coordinating bodies, to joint assessment centres and re-integration programmes, through to interinstitutional offices, bringing together different services under one roof, as a point of contact for those seeking help. The institutions involved in a project vary between cantons (Handbuch IIZ, 2004). Lastly, at the end of 2005 a project for the joint implementation of medical employment market assessments in the course of case-managements was set up by the Unemployment Insurance Fund, Disability Insurance and Social Assistance with the aim of the swiftest possible reintegration into the

Tomáš Sirovátka

Franziska Ehrler and Fritz Sager

165

regular labour market of persons with multiple problems, via crossinstitutional processes (IIZ-MAMAC, 2008). The project is a continuation of the inter-institutional cooperation, which is made more binding (EDI, 2006). In 2010 the government announced the introduction of a new national committee to advance inter-institutional cooperation (EDI, 2010). This suggests that a process of greater inter-agency cooperation is underway in Switzerland, but is far from being complete. The cooperation works well at the level of individual case management, but does not yet fulfil the institutional expectations of cooperation between the different systems of the welfare state (Lindenmeyer and Walker, 2010). Only in a small number of cases did the increased cooperation lead to a local merger of offices (for example, Social Assistance in Zurich). New bodies were created (assessment teams, inter-institutional platforms) within the existing institutions, to which the already established offices can refer multi-problem cases (Handbuch IIZ, 2004). For partial or even complete mergers of institutions, massive changes to legal status and political conditions would be required, which is why such models are not widely discussed (Handbuch IIZ, 2004). One of the aims of the cooperation of assessments is an individualization of the relationship between the client and the system, since the objective of inter-institutional assessments is a binding, individual integration plan. This integration plan includes objectives and indicators and determines measures and financing of programmes and the subsistence level (IIZ-MAMAC, 2006). The analysis of the three major reforms related to new modes of governance shows that changes have taken place. The twofold tendency of re-centralization within the unemployment insurance, the implementation of quasi-markets for activation measures as well as the introduction of a new steering mechanism between national and cantonal levels for financing the RECs and the provision of activation measures are the most important, along with consolidated changes in governance structures. The trend of inter-institutional cooperation and individualization is an ongoing project, which could be related to the institutional level having not yet improved cooperation between the different social security systems to the desired extent.

Effects of altered governance structures Mixed effects on labour-market integration (outcome effects) Several studies on outcome levels have assessed the effects of active labour-market measures. In concentrating on the measures, the relation

Tomáš Sirovátka

166

Marketization in a Federal System

to governance structures has been left unexamined. Further research is needed to establish, for example, whether success or failure of a measure is due to the deployment process of participants by the RECs or whether competition between suppliers of active labour-market measures leads to the development of measures adequately answering labour-market demands. All evaluations emphasize the fastest possible economic integration and do not deal at all with sustainability (Ragni, 2007). The evaluation results for the different activation measures for unemployment insurance clients are mixed. The most positive effects are found for temporary subsidized employment (Gerfin and Lechner, 2000; Lalive-d’Epinay et al., 2008). The employment programmes are only successful for specific subgroups. It has been seen that the programmes are more effective for the less well qualified and the long-term unemployed (Gerfin et al., 2005). The success of educational measures is found to be low, but with better results for women (Gerfin and Lechner, 2000; Lalive d’Epinay and Zweimüller, 2000). The results of the evaluation of activation measures for social assistance recipients and their employability are also relatively negative (for an overview, see Pfister, 2009). It should, however, be remembered that the activation measures for social assistance recipients are only partially aimed at integration into the regular labour market. Local differences in implementation (output effects) After their introduction in 1995, RECs were put in charge of advisory, placement and activation activities (Bieri et al., 2006). Perret et al. (2007) show that some cantons have only partially fulfilled the activation part of their mission, whereas other cantons have set up structures for activation measures. In Switzerland as a whole a broad variety of programmes is offered (for an overview, see Gerfin and Lechner, 2000; Pfister, 2009), but the local offer for social assistance recipients can be too narrow for a real choice (Pfister, 2009, p. 43). Nationally, over 25 per cent of all unemployed persons registered in 2008 by the unemployment insurance participated in an activation measure, the largest proportion having a temporary subsidized job. Participation rates of unemployed persons in activation measures have increased continuously over recent years (SECO, 2008a). The coverage is much lower for social assistance. Only 2.7 per cent of the social assistance clientele over the age of 15 participated in 2008 in an activation measure, and the majority of these took part in programmes organized by the unemployment insurance. It should be remembered, however, that around 30 per cent of the social assistance recipients already have a job and another third are unable

Tomáš Sirovátka

Franziska Ehrler and Fritz Sager

167

to enter the labour market because of care responsibilities, education, non-employability and so on (Lindenmeyer and Walker, 2010). Activation measures on offer have not only been comprehensively evaluated but have also developed considerably. As a result, active labour-market measures today are designed with greater sensitivity to the labour market, and qualifications are given higher priority than just employment. The educational measures have been expanded at the cost of employment programmes and the participants in employment programmes are also provided with reports or certificates of qualification, to improve their chances in the labour market (Bieri et al., 2004). Another development is the diversification and individualization of measures. This is the cantons’ reaction to local circumstances. While in one canton executive unemployment may demand special measures, in another measures for the young were expanded (Bieri et al., 2004). Overall, the varying requirements of the jobseekers led to an increase in the measures on offer, and the RECs pay more attention to matching personal requirements to the objectives of a measure and to ensuring, through flexible entry options, rapid placement on a measure. Participants in programmes are also guaranteed individual support and advice (Bieri et al., 2004). How these developments are directly or indirectly linked to altered governance structures is difficult to assess. The introduction of the RECs initiated a capacity-building process that led to more differentiated demands on activation measures. But to assess how this process directly affected the field of programme suppliers, further research would be needed. Effects that are clearly linked to changed governance structures are deployment strategies. As a result of the performance indicators introduced in REC, a creaming-off effect occurred. As the swiftest possible reintegration into the labour market is by far the most important indicator, strong incentives are created to concentrate resources on those jobseekers who are easiest to place (Ragni, 2007). However, the shift from output to outcome control not only occurred in unemployment insurance. Within the system of social assistance integration (primarily economic integration) and New Public Management, strategies were discussed and led to changes. Classification systems based on targets for the different groups of recipients have been discussed or introduced in several social assistance agencies. One criterion is therefore a client’s potential for economic integration. For individuals with a high potential and those with low prospects of economic integration, social assistance emphasis is on income protection, because it is believed that the first group does not need an activation measure to reintegrate into the labour

Tomáš Sirovátka

168

Marketization in a Federal System

market and that the second can not be reintegrated anyway. Only people with reasonable prospects but moderate potential for economic integration qualify for the participation in an activation measure. However, as such classification systems are relatively new, their impact has not yet been evaluated (Heinzmann, 2009). Kutzner (2009) shows that two groups of social assistance clientele – the working poor and lone parents – are not anticipated to take part in activation measures. As their status already proves activity (through employment or care work), social assistance is limited to income protection for these recipients. Individualization with limited choice (process effects) The RECs’ dual role after the 1995 reforms of service providers and supervisory bodies led to a conflict of interests for REC advisors during face-to-face sessions with clients, as they have to act as both their controllers and advisers (Magnin, 2004). Depending on the advisors’ strategy, activation measures are adjusted to meet individual requirements, either to be as rapid and as targeted as possible or emphasize motivation and qualification (Frölich et al., 2007). The advisors do not advise all jobseekers in the same way, but apply different targets, procedures and measures depending on the characteristics and employability of the jobseeker. Qualifications, experience, motivation, health and age are seen as decisive factors. The question of whether the qualified jobseekers are from an industry with good or restricted career prospects is also taken into account. The former tend to be allocated to temporary employment while, for the latter, employment programmes and education measures are given priority (Frölich et al., 2007). Pfister (2009), in her evaluation of activation in social assistance, states that directors of social assistance agencies prioritize their clients’ opinion on the choice of an activation measure, but a real choice is often not possible due to a lack of offer. However, most clients participating in a measure are satisfied. There are clients who are forced to participate in an activation measure against their will, but this is a rare exception in all social assistance agencies. Different cantons have introduced individual contracts between institution and client. In the Basel canton activation for unemployed persons in unemployment insurance or social assistance is organized by contracts between the client and the case manager. The contract contains details of the clients’ resources, goals, control indicators, responsibilities of the two contract partners and the activation measure. The target is usually the integration into the regular labour market (Pfister, 2009). Another example of individualization/contractualization is an institutional cooperation project between RECs and Social Assistance,

Tomáš Sirovátka

Franziska Ehrler and Fritz Sager

169

which was started in the Vaud canton. With the aim of basic material security and reintegration, a contract is concluded between the institution and the client, by which the individual agrees to undertake active measures to reintegrate both socially and economically (Handbuch IIZ, 2004).

Conclusion The analysis of governance structures in the Swiss social security system for the unemployed has shown that both unemployment insurance and social assistance are organized on a decentralized basis in the federal system and that they act for the most part independently of one another. The reforms to Swiss unemployment insurance have led to fundamental changes in the governance structure over the last decade. The reforms implemented since the mid-1990s, however, show a tendency towards recentralization as part of a capacity-building process, where the regional level within the cantons has been strengthened at the cost of the local units. This development has already been implemented in unemployment insurance, and in social assistance numerous projects within various cantons are taking the same route. Social assistance recentralization has been aimed at a greater equality of treatment of the recipients. In addition to the trend of recentralization, the implementation of active labour-market policies has led to a marketization. To supply the activation measures, quasi-markets were created where private suppliers of activation measures compete with each other for public funding via service agreements. This tendency towards contractualization has been seen in public structures. These new modes of governance show a development towards a market governance regime in Switzerland. At the same time, elements of the corporate governance regime were introduced. That the system has to be goal driven became a priority leading to the change from output to outcome control (for the different governance regimes, see Chapter 1 and Considine and Lewis, 1999). The Swiss system can be seen as a mix of the market and corporate governance regime, with a clear emphasis on capacity building. The effects of these new modes of governance have been various. On the one hand, activation measures became better suited to meeting individual requirements and allocation is carried out on a more individualized basis. On the other hand, the introduction of the quasi-market and the weighting of certain performance indicators have led to false incentives and to a creaming-off of the unemployed. Another tendency in the change of governance modes is a process of inter-institutional cooperation, which is far from complete. It can be

Tomáš Sirovátka

170

Marketization in a Federal System

assumed that further changes will occur in the governance structures in Switzerland, but whether these will be as far reaching as the changes to date and to what extent these will affect the trend towards new modes of governance remain to be seen.

References Armingeon, K. (2001) ‘Institutionalising the Swiss welfare state’, West European Politics, 24/2, 145–68. Behncke, S., M. Frölich and M. Lechner (2006) Aktive Arbeitsmarktpolitik in Deutschland und der Schweiz (Discussion Paper no. 2006–13, Universität St. Gallen). BFS – Bundesamt für Statistik (2009) Sozialhilfestatistik. Standardauswertungen zur Sozialhilfestatistik 2007 (Bern). BFS – Bundesamt für Statistik (2010) Statistisches Lexikon, http://www.bfs.admin. ch (accessed 5 April 2010). Bieri, O., F. Müller and A. Balthasar (2004) Übersicht über die Professionalisierung der arbeitsmarktlichen Massnahmen (AMM) seit 1997, http://www.treffpunkt-arbeit. ch/downloads/studien/seco (accessed 24 April 2008). Bieri, O., R. Bachmann, D. Bodenmüller and A. Balthasar (2006) RAVStrategien zur Arbeitsmarktintegration und deren Wirksamkeit (Seco Publication, Arbeitsmarktpolitik N◦ 17). Bonvin, J.-M. and E. Moachon (2007) ‘The impact of contractualism in social policies: The case of active labour market policies in Switzerland’, International Journal of Sociology and Social Policy, 27/9–10, 401–12. Bütler Liesch, D. (2006) ‘Regionalisierung im Kanton Bern: Neues Sozialhilfegesetz, Qualität als oberste Priorität’, ZeSo Zeitschrift für Sozialhilfe, 103/1, 12. Considine, M. and J.M. Lewis (1999) ‘Governance at ground level: The frontline bureaucrat in the age of markets and networks’, Public Administration Review, 59/6, 467–80. EDI – Eidgenössisches Departement des Innern (2006) Projektdossier IIZ-MAMAC, http://www.iiz.ch (accessed 12 April 2008). EDI – Eidgenössisches Departement des Innern EDI (2010) Gesamtschweizerische Strategie zur Armutsbekämpfung. Egger, M., U. Birchler, T. Dreher, C. Lenz, V. Merckx and T. Studer (2000) Evaluation des Einsatzes arbeitsmarktlicher Massnahmen (Bern: Seco). Egger, M., C. Lenz and B.A. Zürcher (2001) Die Vereinbarung zwischen Bund und Kantonen über den Vollzug des Arbeitslosenversicherungsgesetzes in der Schweiz (Bern: Seco). Engler, H. (2005) . . . und weiter auf Erfolgskurs. 10 Jahre regionale Arbeitsvermittlungszentren (RAV). http://www.treffpunkt-arbeit.ch/downloads/studien/ seco (accessed 18 April 2008). Ferro Luzzi, G., Y. Flückiger, J.V. Ramirez and A. Vassiliev (2004) ‘Swiss unemployment policy, an evaluation of the Public Employment Service’, Swiss Journal of Sociology, 30/3, 319–37. Fluder, R. and J. Stremlow (1999) Armut und Bedürftigkeit. Herausforderungen für das kommunale Sozialwesen (Bern: Haupt).

Tomáš Sirovátka

Franziska Ehrler and Fritz Sager

171

Fluder, R., T. Graf, R. Ruder and R. Salzgeber (2009) Quantifizierung der Übergänge zwischen Systemen der Sozialen Sicherheit (IV, ALV und Sozialhilfe) (Bern: Bundesamt für Sozialversicherungen). Frölich, M., M. Lechner, S. Behncke, S. Hammer, N. Schmidt, S. Menegale, A. Lehmann and R. Iten (2007) Einfluss der RAV auf die Wiedereingliederung von Stellensuchenden (Bern: Seco). Gallie, D. and S. Paugam (2000) ‘The experience of unemployment in Europe: The debate’, in D. Gallie and S. Paugam (eds) Welfare Regimes and the Experience of Unemployment in Europe (Oxford: Oxford University Press). Gerfin, M. and M. Lechner (2000) Evaluationsprogramm Arbeitsmarktpolitik. Ökonometrische Evaluation der arbeitsmarktlichen Massnahmen in der Schweiz (Bern: Seco). Gerfin, M., M. Lechner and H. Steiger (2005) ‘Does subsidised temporary employment get the unemployed back to work? An econometric analysis of two different schemes’, Labour Economics, 12, 807–35. Gough, I. (2001) ‘Social Assistance regimes: A cluster analysis’, Journal of European Social Policy, 11/2, 165–70. Gough, I., J. Bradshaw, J. Ditch, T. Eardley and P. Whiteford (1997) ‘Social assistance in Oecd countries’, Journal of European Social Policy, 7/1, 17–43. Handbuch IIZ (2004) http:// www.treffpunkt-arbeit.ch (accessed 20 April 2008). Heinzmann, C. (2009) ‘Klassifizieren in der Sozialhilfe: zwischen individueller Fallabklärung und standardisierten Modellen – Entwicklungen in der Schweiz nach dem Zweiten Weltkrieg’, in S. Kutzner, U. Mäder, C. Knöpfel, C. Heinzmann and D. Pakoci (eds) Sozialhilfe in der Schweiz. Klassifikation, Integration und Ausschluss von Klienten (Zürich/Chur: Rüegger Verlag). Hohn, M. (2005) ‘Die SKOS-Richtlinien zur Unterstützungsbemessung’, in SKOS (ed.) Von der Armenpflege zur Sozialhilfe. Ein Jahrhundert SKOS & ZeSo – Ein Lesebuch (Bern: SKOS). IIZ-MAMAC (2006) Project dossier IIZ-MAMAC, http://www.iiz.ch (accessed 20 April 2008). IIZ-MAMAC (2008) Weisung/Richtlinie zur Interinstitutionellen Zusammenarbeit in der Projektphase IIZ-MAMAC, http://www.skos.ch (accessed 20 April 2008). Kutzner, S. (2009) ‘Sozialhilfe und Ausschluss’, in S. Kutzner, U. Mäder, C. Knöpfel, C. Heinzmann and D. Pakoci (eds) Sozialhilfe in der Schweiz. Klassifikation, Integration und Ausschluss von Klienten (Zürich/Chur: Rüegger Verlag). Lalive d’Epinay, R. and J. Zweimüller (2000) Arbeitsmarktliche Massnahmen. Anspruch auf Arbeitslosenentschädigung und die Dauer der Arbeitslosigkeit (Bern: Seco). Lalive d’Epinay, R., J.C. van Ours and J. Zweimüller (2008) ‘The impact of active labour market programmes on the duration of unemployment in Switzerland’, The Economic Journal, 118, 235–57. Leu, A. (2005) Die arbeitsmarktlichen Massnahmen im Rahmen der Arbeitslosenversicherung in der Schweiz, Dissertation (Zurich: Schulthess Juristische Medien AG). Lienhard, A., A. Ritz, R. Steiner and A. Ladner (2005) 10 Jahre New Public Management in der Schweiz. Bilanz, Irrtümer und Erfolgsfaktoren (Bern: Haupt). Lindenmeyer, H. and K. Walker (2010) Arbeitslosenversicherung und Sozialhilfe: Zusammenarbeit bei der Arbeitsvermittlung Studie zur Zusammenarbeit RAV – Sozialhilfe (Bern: SECO).

Tomáš Sirovátka

172

Marketization in a Federal System

Linder, W. (2004) Schweizerische Demokratie. Institutionen – Prozesse – Perspektiven (Bern: Haupt). Magnin, C. (2004) ‘Ein für den aktivierenden Sozialstaat typisches Handlungsdilemma’, Swiss Journal of Sociology, 30/3, 339–61. Obinger, H. (1998) Politische Institutionen und Sozialpolitik in der Schweiz. Der Einfluss von Nebenregierungen auf die Struktur und Entwicklungsdynamik des schweizerischen Sozialstaats (Frankfurt am Main: Peter Lang). Perret, V., O. Giraud, M. Helbling and M. Battaglini (2007) Les cantons Suisses face au chômage. Fédéralisme et politiques de l’emploi (Paris: L’Harmattan). Pfister, N. (2007) ‘Sie springen ein und fangen auf’, ZeSo Zeitschrift for Sozialhilfe, 104/2, 12–3. Pfister, N. (2009) Integrationsauftrag der Sozialhilfe in der Praxis (Bern: SKOS). Piller, O. (2006) Die soziale Schweiz. Die schweizerischen Sozialwerke im Überblick (Bern: Haupt). Piotet, G. (2006) ‘Regionalisierung im Kanton Waadt. Eine Anlaufstelle, viele Dienstleistungen’, ZeSo Zeitschrift for Sozialhilfe, 103/1, 12. Ragni, T. (2007) Die Wirksamkeit der öffentlichen Arbeitsvermittlung in der Schweiz. Übersicht über die Hauptergebnisse des ‘Follow up’ der Evaluation der aktiven Abeitsmarktpolitik und Beurteilung der sechs Abschlussberichte, Discussion paper of the Department for Economic Policy of SECO, http://www.seco.admin.ch (accessed 25 April 2008). Sager, F. and C. Rüefli (2005) ‘Die Evaluation öffentlicher Politiken mit föderalistischen Vollzugsarrangements. Eine konzeptionelle Erweiterung des Stufenmodells und eine praktische Anwendung’, Swiss Political Science Review, 11/2, 101–29. SECO (2003) Hauptsache Arbeit, die schweizerische Arbeitsmarktbehörde, http:// www.treffpunkt-arbeit.ch (accessed 15 April 2008). SECO (2008a) Die Lage auf dem Arbeitsmarkt im März 2008, http://www.seco. admin.ch (accessed 10 Mai 2008). SECO (2008b) Über uns, http://www.treffpunkt-arbeit.ch (accessed 10 Mai 2008). SKOS (2005a) Richtlinien für die Ausgestaltung und Bemessung der Sozialhilfe (Bern: Schweizerische Konferenz für Sozialhilfe). SKOS (2005b) SKOS-Richtlinien-Revision 2005, Grundzüge, http://www.skos.ch/ Dokumente (accessed 20 April 2008). SKOS (2006) Regionalisierung und Professionalisierung sozialer Dienste. Auf dem Weg zu einer tragfähigen und wirksamen Sozialhilfe, http://www.skos.ch (accessed 20 April 2008). SKOS (2007) Häufig gestellte Fragen zur Sozialhilfe, http://www.skos.ch (accessed 20 April 2008). Stremlow, J. (2008) ‘Wie die Behörden organisiert sind: Ein Überblick’, ZeSo Zeitschrift für Sozialhilfe, 105/2, 4–5. Zürcher, E. (2002) ‘Bildung einer Nationalen Koordinationsgruppe zur interinstitutionellen Kooperation (KG IIZ)’, Soziale Sicherheit CHSS, 4, 210–1.

Tomáš Sirovátka

9 Governance of Activation Policies in the Czech Republic: Uncoordinated Transformation Tomáš Sirovátka and Jiˇrí Winkler

Introduction Activation policies represent one of the key common trends in the development of the welfare states in Europe (compare Van Berkel and Møller, 2002; Serrano Pascual, 2004; Zeitlin and Trubek, 2005), no matter that individual national activation strategies are far from being uniform. Policies aiming to increase labour-market participation seem to be appropriate for achieving both the goal of sustainable public finance and that of eradication of poverty and social exclusion. In postcommunist countries the policies of activation are being implemented in a specific societal and institutional context: certain failures of governance and implementation conditions have already been identified (Winkler and Žižlavský, 2004; Sirovátka, 2007, 2008; Sirovátka et al., 2007). In this chapter we examine the question as to what specific kind of governance of activation policies have emerged in the Czech Republic and, secondly, the various effects of the specific mode of governance of activation policies. The chapter is structured as follows: after the introductory part we characterize the institutional framework of activation. In the third section we explain the key trends of governance of activation policies since 2004 in the Czech Republic. In the fourth part we assess the effects of governance of activation policies, and in the last and concluding section we discuss the main findings.

Institutional framework of activation The formation of labour-market and employment policy institutions in the Czech Republic goes back to the beginning of the 1990s when the 173

Tomáš Sirovátka

174

Governance of Activation Policies in the Czech Republic

governance structure of the Public Employment Services (PES) was being shaped. This structure survived with no important changes until 2004. During this period, the establishment of local (district) Employment Offices (EOs), with relatively tough powers to implement active policies, took place within a model of a two-tier PES: this included the PES section at the Ministry of Labour and Social Affairs (that is, Employment Services Administration (ESA)) in the centre (strategic and coordination level) and 77 EOs in the districts, originally District Employment Offices, at the second level (executive level). The ESA as an integral section of the Ministry of Labour and Social Affairs is responsible for the coordination and delivery of active employment policy and unemployment benefits – these tasks are performed by local Employment Offices. After the split of the Czech/Slovak federation in 1993, the power to create employment policy was conferred on the national ministry (more specifically on the ESA), but without transfer of the necessary personnel and financial capacity that previously existed at the federal ministry. Since its establishment in 1991 and during its whole existence, the PES has thus been understaffed and underfinanced when compared with other European Union (EU) countries. This fact is mirrored first of all in the poor managerial capacity of the centre (ESA), as well as in the work overload at local/district EOs. The abovementioned deficits have translated themselves into a limited scope and quality of active labour-market policies (ALMPs) and job mediation. For example, if we compare the International Labour Organization (ILO) figures (2003) for the number of the unemployed with the total number of PES staff, we see that the workload (in numbers) is about three times as high in the Czech Republic when compared with Germany and Sweden, and even higher when compared with the UK. The EOs at the local level are responsible for delivery of unemployment benefits as well as for the implementation of ALMPs. They are guided by legislation (Employment Act) and by decisions of the PES governing body – ESA. This central body outlines the strategy of labour-market policy (embedded since 1999 in the National Action Plans of Employment (NAPE)). The decisions of ESA are directed mainly towards the distribution of financial resources allocated for ALMP measures to the EOs, based on specified criteria (such as the unemployment rate in the region, the proportion of long-term unemployment) and towards recommendations (methodical notes) on ALMP measures delivery. Secondly, ESA exerts influence on the formation of the Employment Act. Notwithstanding the conditions given by legislation and the financial resources allocated by the Ministry of Labour and Social Affairs

Tomáš Sirovátka

Tomáš Sirovátka and Jiˇrí Winkler

175

(ESA), the local/district EOs are largely independent in making decisions about the scope and kind of measures to be realized, about the groups of the unemployed to be involved in the measures, depending on the local labour-market conditions, as well as about the specific approach to the unemployed and the job-mediation procedures. The high degree of autonomy enjoyed by the EOs is not, however, a result of a purposeful delegation of power but rather an unintentional outcome of the given institutional frame. While the central ESA possesses only a limited capacity for conceptual decision making, the EOs usually have a considerable knowledge of the local labour market. The control and coordination functions of the PES centre (ESA) are therefore significantly weakened. It does not possess much authority and influence, either at the stage of planning and formation of activation policy or during monitoring and overseeing implementation. On the other hand local EOs do not influence the formation of national employment policy in any way, in contrast to Germany, Austria and other countries. When the Regional Self-Government Units were established in the Czech Republic in 2000, under the Public Administration Reform initiated due to the country’s accession to the EU, 14 of the local (district) EOs were entrusted with extended competences as partners to the new Regional Self-Administration Units. Nevertheless, these entrusted EOs were granted very limited decision-making powers: their task is mainly to ensure administrative procedures necessary for implementation of the European Social Fund (ESF) projects/measures. They also fulfil some methodological and coordination activities; however, they are not provided with decision-making discretion related to the national ALMPs. Other actors, such as the National Tripartite Council, the Parliamentary Commission for Health Care and Social Issues or the interministerial commission for the NAPE (or later, the National Reform Programme), fulfil only an advisory role in the field of ALMP. Since the early 1990s, the directors of local EOs have consulted with the so-called Counselling Committees (advisory bodies established at local level) where local social partners, NGOs, municipalities and educational institutions are represented. These, however, are not endowed with any decision-making powers. At the local level, EOs represent the principals in the field of activation. Municipalities may be considered passive actors: they deliver only social assistance benefits that are fully refunded from the central government budget. They do not receive any funds for activation, since the only body possessing resources and competences concerning activation and implementation of ALMP is the PES. The EOs can

Tomáš Sirovátka

176

Governance of Activation Policies in the Czech Republic

provide from their budget wage subsidies for public works organized by municipalities, but municipalities themselves have no economic incentives available for any activation measures. Cross-sectional coordination between EOs (subordinated to the ESA as a section of the Ministry of Labour and Social Affairs) and municipalities (subordinated to the Ministry of Regional Development) in the field of activation is limited to the organization of public works and unsystematic exchange of information. In spite of the increase in unemployment and in the numbers of welfare state clients during the late 1990s, the government did not allow an increase in public administration staff, with no exemption for PES. Instead, the government established following elections in 2006 promised to decrease public administration staff by 9 per cent, and this objective was also applied to PES that are considered to be part of public administration. A similar practice was adopted following the crisis in 2008–9, when further reductions in public servant numbers were announced in spite of the fact that unemployment exploded during 2009 and that Individual Action Plans (IAPs) that had been used to a limited extent (see below) since 2004 became obligatory, which significantly increased the workload for EOs. Employment Offices had to reduce staff, and evidence shows that there are now only around 1800 staff working directly with the clients as mediators or counsellors (Kalužná, 2008), while there were 540,000 unemployed people at the end of 2009 (that is, a ratio of about 300 clients to one frontline staff).

Trends in governance of activation since accession to the EU The Czech labour market was not hit by unemployment increases until the end of the 1990s (registered unemployment level came close to 10 per cent), which were, however, marked by a high proportion of longterm unemployment (more than 50 per cent of unemployment in 2004) and sharp differences in specific unemployment rates according to education, health status, age and gender. Activation became a challenge. This period overlaps with the country’s accession to the EU: minor institutional reorganization took place in order that the ESF projects could be implemented. In principle, the governance framework did not change. The lack of top-down governance reforms complicated – as we will argue – effective implementation of activation policies in several respects. Nevertheless, this fact does not imply that no changes in governance appeared. We have evidence of several transformations that

Tomáš Sirovátka

Tomáš Sirovátka and Jiˇrí Winkler

177

occurred as spontaneous, uncoordinated adaptations to the demands raised by a more intensive implementation of activation measures within the regime of the ESF schemes/projects. The co-existence of weak coordination by the centre and spontaneous adaptations of governance at the local and regional levels may be considered to be a specific feature of Czech governance of activation policies. When assessing the trends in governance of activation policies in the Czech Republic since accession to the EU (2004) we focus primarily on the key dimensions of governance of activation as outlined in Chapter 1: decentralization, marketization, inter-agency cooperation, new public management and coordination by objectives, with the most important aspects being decentralization and marketization (for their characteristics, see Table 9.1). Decentralization and inter-agency cooperation Decentralization represents the first and most significant (spontaneous) trend of the governance changes, although the system has remained highly centralized in terms of financing the ALMP measures. While administrative reforms of PES in many European countries (Austria, Germany, Denmark and Ireland) involved a formally governed functional and territorial decentralization and devolution of some services, decentralization in the system of Czech employment services takes place as an unexpected consequence of governance failure. We may label the trend as a spontaneous transformation of governance. The transformation of governance was necessitated by the appearance of new measures financed through the ESF. However, this reform was little coordinated by the centre (ESA) as a strategic change, except for formal steps such as the establishment of an administrative unit at the central level responsible for administration of the implementation of the ESF projects (see below). The spontaneous transformation at the local and regional levels followed as a consequence of the complex process of social learning, consisting of several parallel processes, such as rational problem solving, normative (legislative) transformation, imitation and ideational transfer (March and Olsen, 1989; Obinger et al., 2005; Taylor-Gooby and Zinn, 2005). At the same time, the availability of new resources of the ESF and opportunities for participation of various actors confronted central authorities of PES/ESA with a serious challenge concerning their coordinating ability, and made their managerial and capacity deficits more apparent. This was a catalyst to further decentralization (and devolution) rather than centralization.

Tomáš Sirovátka

178

Governance of Activation Policies in the Czech Republic

Table 9.1

Trends in governance of activation after accession to the EU (2004)

Decentralization

Spontaneous decentralization

Political– administrative

Poor capacities of the centre to coordinate the policies (legal instruments and central budget being the main tools) Employment offices continue to maintain a high level of discretion (political decision-making discretion) while being restricted by central budget constraints Municipalities continue to be responsible for administration of social assistance (administrative decentralization)

Non-competitive

Purchase of training, individual diagnostics, counselling, private mediation – increasing in scope due to availability of ESF resources and implementation of agency work in 2004 (new Employment Act), low numbers of competitors.

Internal

Decision-making power given to local EOs (implicitly due to outsourcing in conditions of weak governance, most decision power flows to the providers) – devolution.

De-concentration

Entrusted EOs are responsible for coordination of methodical functions, control and assessment of activation and, more recently, for implementation of regional projects under the ESF

Marketization

Unregulated market failures

Purchasers

Public EOs (with no use of vouchers for services, no role played by the unemployed themselves)

Providers

Public (educational institutions), private (increasing role), NGOs

Contracts

Classic but weak regulation: soft conditions – lack of incentives in terms of performance targets due to the weak role of service purchasers

Cooperation among local EOs through informal networks

No duty to outsource (however, due to a lack of PES capacities, outsourcing implicitly becomes a dominant strategy)

Due to spontaneous decentralization even prior to 2004, EOs had spontaneously developed an adaptation mechanism for the governance of problems emerging from the deficits on the part of ESA (Sirovátka, 2003; Winkler and Žižlavský, 2004). In conditions of weak central coordination, directors of EOs had begun to create informal regional support networks among themselves, coordinating procedures to resolve new

Tomáš Sirovátka

Tomáš Sirovátka and Jiˇrí Winkler

179

problems, issues concerning competence or vaguely defined tasks. These clusters of directors and clusters of other staff of EOs occupying similar positions now cooperate during preparation of calls under the ESF project schemes. They also offer consultations to potential recipients/coordinators of the ESF projects concerning preparation of their individual project proposals. In a number of regions, EOs became informal coordinators of collaboration between EOs, employers, NGOs and municipalities during the formation of local programmes of social and economic development. This bottom-up activity was later institutionalized by the Employment Act from 2004 as a specific labour-market programme within what PES called Purpose-built Programmes in the Local Labour Markets (Winkler et al., 2005). Consequently, and especially after 2004 (in connection with implementation of the ESF projects), public–private partnerships (PPPs) in the regional labour markets have developed, emerging in two forms: first, partnerships associated with local and regional programmes, and secondly, those associated with national programmes and serving to stabilize the national strategy in regional conditions. The central role of local EOs (and the entrusted local EOs) as principals is typical of the latter type of PPPs. Their role is to organize independent tenders announced ad hoc under specific calls for projects, and subsequently to contract the services providers. The PPPs later merged with the first type of the ESF projects, called the Regional Integrated Projects, which are created from the bottom up through cooperation among local EOs coordinated by the EOs entrusted with methodological guidance. Due to this spontaneous and institutionally supported decentralization and networking, the performance of PES is characterized by a high degree of decentralization, even fragmentation. It is often the case that the EOs do not follow the government policy guidelines given by the ESA, with the centre thus losing control over implementation of activation programmes and their outcomes. Secondly, the formation of the state employment policy and the labour-market policy is strongly politicized. In such a situation the ESA faces difficulties in defining any widely acceptable government policy objectives. Lastly, the incorporation of other private participants (each having their own specific interests) in the implementation structure increases an already high dispersal and heterogeneity of decision making and causes chaos in policy making at the local level, and misunderstandings in communication between the centre and the regions/districts. Similarly, the cooperation between state and non-state

Tomáš Sirovátka

180

Governance of Activation Policies in the Czech Republic

participants does not proceed on the basis of consensus over activation policy objectives in the labour market. The fulfilment of the principal role of the EOs in this process of implementation of activation is an uneasy task. The staff of EOs often settle for less than completely satisfactory results in cooperation with external service providers who do not provide better measures/services (regarding their scope and quality) than the EOs, although the cost of the outsourced services is considerably higher. At the same time, they emphasize the limited capacity of local EOs to coordinate and control implementation of activation by private agencies and NGOs, as well as the poor communication with the responsible coordination units for the ESF projects at the entrusted EOs (Winkler et al., 2005; Winkler, 2009). On the other hand, innovative bottom-up initiatives and cooperation sometimes emerge to substitute national programmes, the capacity of which appears insufficient to solve local problems. These are usually ad hoc regional networks of actors with a principal role played by the local EO. Subjects involved in such networks have a strong social capital (trust, in the first place) built during cooperation on projects at the regional level. Their cooperation is associated with the possibility of identifying a common regional interest that inspires joint action by different subjects. Cooperation among EOs and municipalities on public works projects in underdeveloped regions (border regions, inner peripheries) is one example. Existing research shows a high degree of satisfaction of the participants in such cooperation with their ability to find a common platform. This finding is even more relevant as the participants report that controversies were more common ten years earlier – when they started cooperation on these projects (Winkler, 2009). Marketization Operation of private employment agencies has been possible in the Czech Republic since the early 1990s. Employment Offices have also contracted out retraining programmes to educational institutions, including private ones. In addition, the new Employment Act (2004) made agency employment possible, and subsequently the number of private actors increased rapidly. Their activities further expanded as a result of the ESF projects enabling EOs to subcontract a range of various activities, such as job mediation, individual diagnostics, counselling, vocational training, job placement of vulnerable groups of the unemployed, and individual assistance. The ESA as a central body has been outsourcing activities such as analytical and conceptual work, special educational programmes for PES staff, marketing, tenders and

Tomáš Sirovátka

Tomáš Sirovátka and Jiˇrí Winkler

181

so on. The range and scope of activities being outsourced has increased dramatically. There are two kinds of the ESF project schemes: (1) ‘national’ projects, where the entrusted EOs play the role of project coordinator and services purchaser, and private and public agencies as well as NGOs as services providers; and (2) ‘individual grant’ projects, where the entrusted EOs take on the role of tender managers, and private and public agencies and NGOs act as project coordinators and may purchase those services that they are unable to provide by themselves from other providers. Between 2004 and 2008 there occurred a marked increase in the numbers of non-governmental (profit and non-profit-making) organizations providing educational, counselling and mediation services in the labour market. Strong, long-standing agencies operating in this market have gained a dominant position in several economically significant regions. This is well documented in the attitudes of EOs’ managers who evaluate best those agencies with which they have cooperated for a long time (Winkler and Zelenková, 2010). Surprisingly, such a development has not caused any increase in competition between service providers. The given circumstances prevent the emergence of competition. According to EOs’ staff (interviews led by the authors), the established leading agencies misuse their market position by working together in order to divide the market amongst themselves and do not allow competition to develop with other educational and mediating agencies. In the case of the ESF projects, the performance of service providers is supervised by PES on the basis of obligatory financial and output indicators. However, these output indicators seem to be rather soft and vague and thus do provide the services providers with a lot of space for manoeuvre (see below). For example, indicators of quality are missing from evaluations of programmes provided to vulnerable groups of the unemployed. Similarly, the selection criteria for evaluating project proposals do not include indicators of quality. Failed attempts at new public management and lack of coordination The Czech Republic has witnessed two processes in connection with new methods of governance since 2004. First, an attempt to move towards a management-by-objectives approach in ALMP has taken place, as expressed in the continued specification of the NAPE guidelines and targets. In 2004, the ESA formulated ‘A plan of institutional, merit and time-related preconditions for the realization of the NAPE measures in 2005–2006’. This plan introduced

Tomáš Sirovátka

182

Governance of Activation Policies in the Czech Republic

quantified targets: it specified the number of Individual Action Plans to be realized (for both young and long-term unemployed people) and the number of long-term unemployed and other disadvantaged people qualified to participate in the ALMP measures. Additionally, since 2005 PES have obliged EOs to meet certain quantified targets in the form of ALMP targeting, and have also specified the targets concerning the application of selected tools (vocational training and work rehabilitation for disabled people). However, the targets set for 2006 and 2007 lacked ambition, since the required proportion of participants from the target groups did not reach their designated proportion in the unemployment stock: for example, participation of 30 per cent of people unemployed for more than six months in ALMP measures was required (while there were twice as many of those in the registers), as well as 20 per cent of unemployed people above 50 years of age (30 per cent in the registers) and 25 per cent of unemployed people under the age of 25 years of age (which slightly exceeded their proportion among the unemployed). So it comes as no surprise that these goals had already been fulfilled at the majority of the EOs. Another issue is that EOs have never been evaluated with regard to the achievement of the set objectives (see MPSV, 2008). We can characterize these developments as unfinished, incremental reforms of the governance of activation. The intended shift towards the managerial approach has not, as yet, significantly affected the prevailing bureaucratic-administrative style of policy making, given the weak conceptual and control capacity of the centre: consequently, the set goals have been neither ambitious nor strictly required and evaluated. Secondly, the ESF projects also induced organizational changes within PES. In a relatively short time a plural implementation structure was created. The administrative agenda of the ESF schemes was handed over to the newly established ministry sections. A multi-track administration emerged, one track being the ‘ESF line’ and another being the ‘national ALMP line’, with little coordination between them. This lack of coordination is apparent in particular at the level of local EOs that are responsible for the ‘national’ ALMP measures and are also involved in the roles of tender managers, purchasers or project coordinators in the ESF projects. The current stage of development of Czech activation policy and its governance is seen as a result of historically developed national rules and routine procedures: a low level of internal feedback/self-reflection and an inability of the system of employment policy to reform itself on the one hand, and the influence of the European Employment Strategy,

Tomáš Sirovátka

Tomáš Sirovátka and Jiˇrí Winkler

183

mostly implemented by means of the Open Method of Coordination (OMC), as well as programming and funding from the ESF, on the other hand. Owing to mutual interaction of the factors mentioned above, the present governance system shows a number of deficiencies and compromises (institutional and organizational gaps) that may affect the efficiency of Czech activation and labour market policy (March, 1999). Given this situation, the central authorities have attempted to reform this malfunctioning system, although they prefer bureaucratic methods to a formal centralization of powers. The emerging mode of governance Considine (2001) distinguished between procedural, market, corporate and network types of governance. Activation processes in practice always includes a mixture of various procedures and means of organizing activation policy represented by these models (see Chapter 1). The modes of governance are rather the ideal type, that is, tools enabling empirical measurement and descriptions. Hence it is no surprise that the current Czech mode of activation policies delivery has hardly been converging towards any of the ideal types. It is not fully compliant with the rules of either the market or corporate governance mode, since the ESA has not been successful in coordinating employment services by using operationalized objectives. In addition, as yet there exists no competitive market of non-state services, and rational client demand for such services is still lacking. Neither is the mode of coordination close to the rules of network governance, which would be underpinned by mutual dialogue among the various interested parties (entrepreneurs, regional political representatives, employees and jobseekers, and others) and by the citizens’ and service providers’ trust of the coordinative role of the state – for example, in the case of cross-sectional partnerships. Such a role of the principle is not appropriately fulfilled by PES. Finally, the current situation is far from the procedural governance mode where providers of PES perform under a hierarchical control by the policy makers. We assess Czech governance of activation as a hybrid mode, positioned somewhere between the procedural mode (which is strongly embedded due to path dependency, but is characterized by the insufficient managerial and implementation capacity of the ESA, and therefore spontaneously decentralized) and the market mode of governance (based on widespread contractualization and outsourcing, and characterized by weak regulation of the market and a lack of incentives – vague contracts are typical). This mode of governance is, however, enriched

Tomáš Sirovátka

184

Governance of Activation Policies in the Czech Republic

with fragmented, unfinished and inconsistent innovations evoked primarily by external stimuli after the accession to the EU (the European Employment Strategy (EES) and ESF projects). It is also characterized by a low level of professional capacity, a strong influence of political actors on the process of policy making, and a high degree of decentralization in real policy making/implementation. Considering this specific institutional set-up of activation policy, the question follows as to how effective activation policies can be.

The effects of the governance of activation It is difficult to assess the effects of the governance changes in the case of the Czech Republic. First, hardly any governance reforms were implemented from the top down, except for new activation measures such as Individual Action Plans, increased benefit conditionality, making-work-pay measures and an increased scope for ALMP measures. In spite of this lack of top-down reforms, spontaneous processes of marketization/privatization and decentralization emerged, in particular in connection with the spread of the ESF schemes/projects. The absence of top-down governance reforms hampered effective implementation of activation measures. Nevertheless, we observed that the spontaneous reforms from the bottom (local level) up enabled, to some extent, implementation of activation policies at the local level. Although they could not effectively countervail all the governance deficits, they have facilitated timely implementation of the ESF projects. High expectations were associated with the implementation of measures under the ESF, concerning an increased scope of activation measures and greater involvement of the vulnerable groups (Horáková et al., 2010). More individualized treatment and more intensive job search facilities were expected from the IAPs. For these reasons we will attempt to compare here the effects of the new measures financed through the ESF and IAPs (where the spontaneous changes of governance such as privatization/marketization and decentralization were most significant) with the other (traditional, national) ALMP measures. We are interested to discover whether and how the spontaneous changes of governance enable more effective implementation of activation – where these are more strongly associated with the newly implemented ESF projects rather than with the old ALMP measures. In line with the classification of the effects of governance of activation policies as outlined in Chapter 1, we will focus here on the

Tomáš Sirovátka

Tomáš Sirovátka and Jiˇrí Winkler

185

outcomes (in our case, indicated by the indirect measure of employment effects) and the outputs of activation measures (in our case, the scope and targeting of the policies/measures and their content), and finally on procedural effects (the degree of individualization of activation). Outcomes: employment effects of activation measures Using administrative data, measurement was made of how often the participants in ALMPs and the ESF projects appeared in the registers during the 12 months following completion of their participation in the national ALMP and ESF projects in 2006 (that is, an indirect measure of the employment effect) (Kulhavý and Sirovátka, 2008). We found evidence that there were differences in the employment effects of various national ALMP programmes: in the case of training and public jobs the occurrence of unemployment immediately after participation in the programme was over 60 per cent, dropping to approximately 30 per cent after 12 months. The participants in more selective and less scaled national measures of job creation in the private sector exhibited lower rates of unemployment when compared with the participants in training activities and public jobs (only about 10 per cent of the former appear in the registers shortly after completion of the programme, and also after 12 months), and even lower rates in the case of self-employment support schemes. As regards the employment effects of the ESF measures, these were apparently better than those of the national requalification programmes in 2007 (after 12 months only about 10 per cent of the participants appear in the registers). These encouraging results, however, show only gross employment effects and include a considerable dead weight, mainly due to creaming-off effects. The findings indicate higher employment effects for the ESF projects when compared with labour-market training and public works under national ALMPs. Nevertheless, analysis of the implemented measures one year later, in 2007 (Hora et al., 2009), documented a converging trend in effectiveness: after 12 months 27 per cent of the participants in the ESF measures were on the registers, as compared with 25 per cent in national requalification programmes and only 16 per cent in the national public works programme. The decreasing relative effectiveness of the ESF measures and the increasing effectiveness of national ALMP measures are due to the increasing scope of the ESF measures (see below), which have reduced creaming-off in selecting participants for these measures. To summarize, it seems that the new ESF measures only temporarily improved the placement effects, which lasted only until the scope of these measures was increased.

Tomáš Sirovátka

186

Governance of Activation Policies in the Czech Republic

The outputs: the scope, creaming-off/targeting and the contents of activation measures The scope Basically, labour-market measures are among the least developed types of public policies in the Czech Republic. The expenditure on these measures amounted to only about 0.12 per cent of GDP in recent years, which is five or even seven times less than in countries with a comparable unemployment rate (see OECD, 2007). Nevertheless, since 2004 the ESF has made it possible to finance specific projects of ALMP. This simple fact has brought new resources and more opportunities for participation that have been used by actors from both the profit and non-profit sectors (as services providers, see above). Thanks to financing from the ESF, the scope of ALMP increased, with the proportion of active labour-market participants in the unemployment stock rising from 19 per cent to 39 per cent between 2005 and 2007. In 2008, the absolute number of participants in ALMP measures (85,000) dropped below the levels of 2006 and 2007 (141,000 and 137,000, respectively), and the proportion of active labour-market participants in the unemployment stock was 24 per cent, with participants in the ESF projects representing 50 per cent of all participants. Data from 2009 show that the scope of ALMPs had not increased much in spite of the crisis (102,000 participants, although unemployment increased from 352,000 to 539,000); the share of all labour-market measures participants in the unemployment stock was only 19 per cent, with two-thirds of these being financed through the ESF projects (MPSV, 2010). With regard to the last five years, the ESF measures have clearly replaced a substantial part of previously extant national ALMP measures without any added value in terms of the scope of measures. This must be attributed primarily to a lack of managerial capacity at the EOs in administering an increased number of ALMP measures, even where these are being outsourced to other agents. The government shows no commitment to changing the situation, since it does not consider the modest scope of ALMPs to be a policy deficit. Instead, it prefers to improve incentives to work (see NRP, 2005, 2008) and continues its restrictive staffing policy. Targeting It is interesting to note that during the development of activation policies no positive change has appeared in relation to targeting and

Tomáš Sirovátka

Tomáš Sirovátka and Jiˇrí Winkler

187

eliminating the creaming-off effects. The national ALMPs have been analysed (Kulhavý and Sirovátka, 2008; Hora et al., 2009) as being somewhat biased towards selection of those unemployed people at lower risk of exclusion (that is, unemployed persons with an intermediate–higher level of education are more likely to participate in some kind of ALMP than those with only elementary education, even though the unemployment risks are reversed between the two groups). The reason for this is that the EOs select those unemployed people whom they expect to represent the least demanding workload. For example, while in 2007 around 32 per cent of the registered unemployed were people with only basic education and 31 per cent were over 50 years, their proportion among requalification/training participants represented 15 per cent and 17 per cent, respectively (Kulhavý and Sirovátka, 2008). Although the implementation of the ESF projects is based on management by indicators and objectives, which should improve targeting of the unemployed at higher risk of unemployment (in the Czech case, these are poorly educated, the disabled, the unemployed over 50 years of age and the long-term unemployed), the creaming-off effects are most evident in the programmes of vocational training for the unemployed with basic education, the unemployed over 50 years of age and the unemployed with (partial) invalidity, who are under-represented both in the national ALMP measures and in the ESF projects (Kulhavý and Sirovátka, 2008; Hora et al., 2009). For example, a logistic regression analysis shows that in 2007 the probability of participation in the national ALMP measures was twice as high for those having completed secondary education when compared with those with only a basic/elementary education. In the case of requalification/training it was three times as high, while in the case of the ESF programmes it was 2.3 times as high. The other above-mentioned target groups are also somewhat under-represented, in both the national ALMP measures and the ESF schemes. We can see, when looking at the actual specifications of the target groups in the ESF projects, that these definitions are often rather broad, allowing easy selection between or within groups (the latter not being apparent in data, but still representing a typical strategy to circumvent the demands set by the conditions of the ESF projects), in favour of the better-equipped unemployed. In this creaming-off process, a crucial role is played by specific employers (for whom the unemployed are typically being retrained) and their negotiations concerning the final selection of participants with the private agencies (services providers). Despite all this, targeting is still slightly better under the ESF

Tomáš Sirovátka

188

Governance of Activation Policies in the Czech Republic

programmes in comparison with the national schemes, due to the introduction of management by indicators. Even though the target groups are rather broadly/generally defined and the target indicators within the ESF projects are rather soft, the objectives to be achieved are still somewhat more explicit and more binding than those that regulate the targeting of measures of the national ALMP. Contents of activation The content of activation policies has, to a large degree, been influenced by the institutional framework and governance of the policies. As we have already explained, the institutional framework has not been open to much innovation, especially in terms of more demanding forms of activation and labour-market policies like vocational training and individual counselling. The Employment Act from 2004 and the National Reform Programme 2005–8 showed a strong preference for measures improving work incentives, both positive incentives (such as improvements in the minimum wage and tax bonuses, back-to-work benefits and in-work benefits/tax credits) and – especially since 2006 – negative incentives (such as a delayed revaluation of the living minimum, implementation of the existence minimum – set at a lower level than the living minimum), with the latter being provided to the unemployed assessed as passive or non-cooperative, and since 2009 to all social assistance claimants who (after six months) do not participate in public service jobs. Furthermore, greater conditionality of benefit entitlements, a stricter definition of suitable job and, finally, to some extent individual casework with the unemployed are principally aimed at increasing the administrative pressures on the unemployed. Procedural effects The lack of top-down governance reforms and the deficits in institutional capacity imply a rather narrow approach to activation. Individual action plans, which represented a policy innovation, have not been effectively implemented. In most countries, the process of individualizing services is a significant feature of the current governance of activation. Individualization is associated with voice being given to the unemployed (their voice is strong when they have rights in terms of co-determining the content of their individual contract and when they are provided with a choice of instruments/providers, for example in the form of vouchers). Such empowering forms have not been used in the Czech Republic. Since individual contracts, which had been strongly recommended in guideline 1 of the EES (prevention and early

Tomáš Sirovátka

Tomáš Sirovátka and Jiˇrí Winkler

189

activation), appeared to the ESA to be a convenient and inexpensive instrument for eliminating passivity and benefit dependency or abuse, they succeeded in embedding this instrument in the new Employment Act No. 435 from September 2004, thus obliging EOs to offer the IAP to every unemployed person under 25 years after a certain period of unemployment (participation was voluntary). While the original intention was to invite large numbers of the unemployed to participate in IAPs, EOs in locations with high unemployment, where this goal was beyond the capacity of available staff, immediately deviated from this target. Not only did the planned impacts of the programme not materialize due to the limited numbers of participants who agreed to contract the IAP, but also a formal approach in most EOs narrowed the measure to a mere signing of the contract and a rather standard job-mediation process, while simultaneously focusing the IAPs on more cooperative (more motivated) clients. While the National Reform Programme 2005–8 promised 25,000 individual contracts to be concluded in 2006, the realized number was only 7000 (MPSV, 2008). Nevertheless, since 2009, the centre-right government has established IAPs as obligatory for all unemployed people after five months of unemployment, in spite of the lack of personnel capacity at EOs. Thus the IAPs have largely become a formal step, possibly increasing the administrative pressure on the unemployed. Since the start of 2009 the ESA has been negotiating with the government regarding the possible cancellation of the IAP as an obligatory measure, on the grounds of increased numbers of unemployed and the lack of capacity for this task (this statement is based on the authors’ interview with a representative of the ESA). These ambiguous effects have resulted from the development of the IAPs: at present, IAPs are understood to be a specific tool of enforcement policy and as stimulation for the unemployed to become active in the labour market (Sirovátka et al., 2007). This basic activation tool is simply a contract that exerts a specific form of pressure on the applicant: breaking the contract may be a reason for removal of the individual from the register of unemployed people and for withdrawal of unemployment benefits. There are also other reasons for ineffectiveness in terms of individual treatment. Privatization and outsourcing of employment services creates an expectation of a more individual approach and a higher quality of service provision due to the expected increase in interest to satisfy individual demands under conditions of competition among various service providers. However, studies from other countries show that implementation of purely quasi-market relationships (formulation

Tomáš Sirovátka

190

Governance of Activation Policies in the Czech Republic

of impersonal supply and demand, contracting services) as the decisive coordination mechanism is not a sufficient precondition for improving the quality of public services. Competition among providers under approximately identical conditions is not the only important factor (Hood, 1991) – services providers’ responsibility is also significant (Goodin, 2003). Some authors emphasize the necessity of the coexistence of various coordinative mechanisms in order that services improve. According to Gray and Jenkins (2007), in the case of educational services, for instance, the combination of a strictly defined legislative framework, privatization and the contracting of services from private agencies, as well as the development of professional communities, are all essential. In the case of the Czech activation policy, the contracting-out of services has so far been accompanied by neither fair competition among service providers nor the development of professional education and the formation of professional communities as a coordinative mechanism for cross-sector collaboration. The present state of unfinished privatization (which lacks the principles of fair competition) and the ungoverned level of professionalism in human resources reduce any effort at individualization of services to a rather formal contract concluded between clients, service providers and the state. To summarize, the governance of activation has not brought any significant procedural effects. With regard to the overall effects, it seems that insufficient personnel resources, low professional competence (implying a rather procedural governance in the implementation phase of activation), unfinished and inconsistent innovations, as well as mostly instrumental forms of cooperation between PES on the one hand and employers and NGOs on the other, have made it possible to use the ESF resources without paying attention to the establishment of an adequate governance framework. This has now led to a situation where governance changes have only unimportant and ambiguous effects on activation.

Conclusions: the role of governance and implementation conditions We have shown that in the Czech Republic the governance of activation policies is traditionally a weak element of public policy making characterized by deficits in the management of activation, unfinished reforms, spontaneously emerging mechanisms and informal support networks filling the gaps in governance, and a neglected system of PES in terms of their managerial and personnel capacity.

Tomáš Sirovátka

Tomáš Sirovátka and Jiˇrí Winkler

191

In 2004, external factors of a governance change gained more strength with the country’s accession to the EU: the programming of the ESF projects associated with new formal procedures in governance, an increase in resources available for activation measures and the transmission of new ideas from the EES conveyed through the NAPE. Nevertheless, no important top-down governance reforms have been implemented despite the fact that the focus on activation has increased. In these circumstances, the policy makers deliberately focus largely on simple measures of activation that do not require a very advanced coordination of policies: these are typically work incentives implemented into the benefit schemes. The introduction of the ESF projects into activation policy has brought some change: at least an initial extension of the scope of active labour-market measures, marketization and outsourcing, and a pluralization of actors. The coordination and managerial capacity of the centre is still weak: coordination by objectives and performance measures at the national and local levels as well as the rules set to govern the involvement of other actors are insufficient in terms of promotion of the strategic objectives of the centre or EOs, such as better targeting (at specific groups) and better quality of certain aspects of measures. Similarly, the implementation capacity of EOs needed to meet more demanding tasks, such as a more individualized and comprehensive service to clients, is also inadequate. Consequently, an important factor is the governmental and managerial deficit concerning the preparation of an institutional environment and adaptation of the ESF programmes and facilitating their implementation in the Czech context. The key government objectives have been the making of savings in this area and a reduction in public administration staff. For this reason the ESF project measures have been aimed simply at substituting the national ALMP measures. At the same time, some functioning mechanisms of local/regional cooperation have been weakened since the National Integrated Projects that replaced the national policies have been prioritized. Despite the foregoing, the governance of activation policies has become more complex with the implementation of new measures under the ESF. This has happened without any institutional reform pursuing the objective of better coordination, and so the previously existing institutional constraints have worsened. On the one hand, the spontaneous adaptation mechanisms of the system of PES make the functioning of the new and more complex governance structure possible, while on the other the effectiveness of the ESF projects is not very good in

Tomáš Sirovátka

192

Governance of Activation Policies in the Czech Republic

several respects. Initially, the ESF projects brought an increased scope of labour-market policy measures, but this effect did not last due to the insufficient managerial capacity of the EOs. Besides, the added value of these measures is doubtful since they have barely improved the targeting of activation policies towards vulnerable groups, and are sometimes associated with increased selection and creaming-off processes during implementation. The development of effective implementation of activation measures (with regards to their scope, targeting, contents and quality) is blocked by a weak coordination capacity, a poorly coordinated process of outsourcing that does not promote competition among private agencies, and deficient professional competences among PES staff. Although local actors (directors of EOs) try to respond to some of these deficits by forming local networks of mutual support, this effort cannot eliminate the substantial system deficits and the fragmented governance of the PES. The spontaneous decentralization evoked by the new challenges represents a specific feature of the Czech governance of activation, making it possible to implement the ESF projects and adapt them to the existing mode of governance. Nevertheless, neither decentralization nor marketization, which are the key trends, have brought, by themselves (that is, without effective management by the centre), any significant effects on activation policies.

Acknowledgements This study was written with the support of the Ministry of Education of the Czech Republic (NPVII2D06009).

References Considine, M. (2001) Enterprising States: The Public Management of Welfare-to-Work (Cambridge: Cambridge University Press). Goodin, R.E. (2003) ‘Democratic accountability: Distinctiveness of the Third Sector’, European Journal of Sociology, 44, 359–96. Gray, A. and B. Jenkins (2007) ‘Profession and bureaucracy’, in J. Baldock (ed.) Social Policy (Oxford: Oxford University Press). Hood, Ch. (1991) ‘A public management for all seasons?’, Public Administration, 69/1, 3–19. Hora, O., T. Sirovátka, H. Tomešová Bartáková and J. Vyhlídal (2009) Hodnocení programu˚ aktivní politiky zamˇestnanosti realizovaných v roce 2007 se zamˇeˇrením na rekvalifikace (Prague: VÚPSV). Horáková, M., O. Hora and J. Vyhlídal (2010) Hodnocení pˇrínosu projektu˚ Operaˇcního programu Rozvoj lidských zdroju˚ (Prague: VÚPSV).

Tomáš Sirovátka

Tomáš Sirovátka and Jiˇrí Winkler

193

ILO (2003) Summary of International Public Employment Services (PES) Administrative Statistics (Updated September 2003), http://www.sedi.oas.org/ddse/english/ documentos/cancun/012003PESAdminStats-%20Ellen%20Hansen-ILO.doc (accessed 28 September 2010). Kalužná, D. (2008) Main Features of the Public Employment Services in the Czech Republic (Paris: OECD). Kulhavý, V. and T. Sirovátka (2008) Hodnocení efektivity programu˚ APZ a doprovodných nástroju˚ a projektu˚ politiky zamˇestnanosti v roce 2007 (Prague: VÚPSV). March, J.G. (1999) The Pursuit of Organizational Intelligence (London: Blackwell). March, J.G. and P. Olsen (1989) Rediscovering Institutions: The Organizational Basis of Politics (New York: Free Press). MPSV (Ministry of Labour and Social Affairs) (2008) Analýza situace na trhu práce v roce 2007 (Prague: MLSA). MPSV (Ministry of Labour and Social Affairs (2010) Data available through web portal http://portal.mpsv.cz/sz/politikazamest/trh_prace/, own computations (accessed 5 October 2010). NRP (2005) National Lisbon Programme 2005–2008 (National Reform Programme of the Czech Republic). Available on http://ec.europa.eu/archives/ growthandjobs/national-dimension/member-states-2005-2008-reports/index_ en.htm (accessed 30 October 2010). NRP (2008) National Reform Programme of the Czech Republic 2008–2010. Available on http://ec.europa.eu/archives/growthandjobs_2009/documentation/index_ en.htm#national (accessed 30 October 2010). Obinger, H., S. Leibfried and F.G. Castles (2005) ‘Bypasses to a social Europe? Lessons from federal experience’, Journal of European Public Policy, 12/3, 1–27. OECD (2007) Employment Outlook (Paris: OECD). Serrano Pascual, A. (ed.) (2004) Are Activation Policies Converging in Europe? The European Employment Strategy for Young People (Brussels: ETUI). Sirovátka, T. (2003) Problémy trhu práce a politiky zamˇestnanosti, Research report (Brno: Masaryk University). Sirovátka, T. (2007) ‘The individual approach in activation policy in the Czech Republic’, in R. van Berkel and B. Valkenburg (eds) Making It Personal. Individualising Activation Services in the EU (Bristol: The Policy Press). Sirovátka, T. (2008) ‘Activation policies under conditions of weak governance: Czech and Slovak cases compared’, Central European Journal of Public Policy, 2/1, 4–29. Sirovátka, T., P. Horák and M. Horáková (2007) ‘Emergence of new modes of governance in activation policies: Czech experience’, International Journal of Sociology and Social Policy, 27/7–8, 311–23. Taylor Gooby, P. and J. Zinn (eds) (2005) Risk in Social Science (Oxford: Oxford University Press). Van Berkel, R. and I.H. Møller (eds) (2002) Active Social Policies in the EU. Inclusion through Participation? (Bristol: The Policy Press). Winkler, J. M. Žižlavský and P. Horák (2003) Institucionální pˇredpoklady ˇceské politiky zamˇestnanosti, Research report (Brno: Masaryk University). ˇ Winkler, J. and M. Žižlavský (eds) (2004) Ceský trh práce a Evropská strategie zamˇestnanosti (Brno: Masaryk University). Winkler, J., L. Klimplová and M. Žižlavský (2005) Úˇcelové programy na lokálním trhu práce (Brno: Masaryk University).

Tomáš Sirovátka

194

Governance of Activation Policies in the Czech Republic

Winkler, J. (2009) ‘Institucionální a organizaˇcní limity individualizace cˇeské politiky zamˇestnanosti v podmínkách duálního trhu práce’, in T. Sirovátka, J. Winkler and M. Žižlavský (eds) Nejistoty na trhu práce (Brno: Masaryk University). Winkler, J. and I. Zelenková (2010) ‘Služby zamˇestnanosti a formy spolupráce s obˇcanským sektorem na regionálním trhu práce’, in J. Winkler and L. Klimplová (eds) Nová sociální rizika na trhu práce a potˇreby reformy ˇceské veˇrejné politiky (Brno: Masaryk University). Zeitlin, J. and D. Trubek (eds) (2005) Governing Work and Welfare in a New Economy (Oxford: Oxford University Press).

Tomáš Sirovátka

10 Decentralization and Back to Centralization: the Swedish Case Renate Minas

Governing activation in Sweden: features and trends Labour-market policy reform has a long tradition in Sweden. With Sweden as a pioneer, Nordic governments introduced active labourmarket policies early in the twentieth century and relied on these to facilitate structural changes in the economy. A significant feature of Swedish labour-market policy since then has been extensive state intervention. With the deep economic recession and massive increase in unemployment in the early 1990s, the conditions for labour-market policy were, however, fundamentally altered. To counter fears of a longterm increase in working-age people dependent on benefit payments, as well as growing costs for social assistance and unemployment compensation, the government introduced a wide array of reforms aimed at creating and strengthening the link between work and welfare systems. To understand recent reform processes concerning the governance of activation policies in Sweden, several specific conditions have to be taken into account. One is the fact that local government and local responsibility for welfare provision have a long history, and municipal self-governance is enshrined in the constitution. Nevertheless, Sweden is a traditional unitary state where legislative power is held entirely at the national level, and the constitution also gives parliament the right to define the municipal scope of authority and to establish the range of local political influence. In this respect, national government has the ultimate power to decide how far decentralization should be carried through (Bergmark and Minas, 2007). In practice, this means that those parts of the Swedish welfare sector that are administered by the municipalities are under national government control to varying degrees. This control is exercised by means of a number of steering instruments that have a more or less imperative character. The other 195

Tomáš Sirovátka

196

Decentralization and Back to Centralization

aspect is that even if active labour-market policy is a central government policy domain, local governments also have a key role. Local governments organized public relief work throughout the twentieth century and have worked in close cooperation with the Public Employment Services (PES) in carrying out activation measures. Thus, the analysis here will concentrate mainly on local government’s involvement in activation policies and its relation to central government agencies. Changes in modes of governance in activation policies need to be understood as consequences of national contingencies, institutional choices, traditions and values of the particular country, but also of course as consequences of other influences such as New Public Management ideas pushing processes such as decentralization, privatization or building of partnerships (Table 10.1).

Table 10.1

Main governance reforms in Sweden

Year of reform

Policy field

Main objective

Main reform aspect

New local government act, more local autonomy in how to organize the internal organization

Decentralization

Social assistance and labour-market policy

Shift of responsibility for unemployed youth to the local level

Decentralization

1996

Labour-market policy

Municipality got majority on employment service committee

Decentralization

1998

Social assistance

Introduction of national norm

Re-centralization

Introduction of workfare in social assistance act

Conditionality

1991

1995/96 1997

Restricted right to appeal municipal decision 2004

Labour-market Policy

Supervising of employment relief funds by Unemployment Insurance Board (IAF), before this control was performed by the National Labour Market Board.

Tomáš Sirovátka

Differentiation

Renate Minas

197

2005

Social Insurance

Establishment of the new Swedish Social Insurance Agency

Centralization, increased governmental steering

2007

Labour-market Policy

Reorganization of the National Labour Market Administration

Centralization, increased governmental steering

2007

Social assistance and labour-market policy

Shift of responsibility for unemployed youth to the national level

Re-centralization

2008

Labour-market Policy

Swedish Public Employment Service was commissioned by the Government to purchase private employment services

Multiplication of actors Privatization

Changes in modes of governance have often been discussed as fostering multi-level governance, a concept at the centre of lively debates in European policymaking (Stoker, 1998; Scharpf, 2001; Hooghe and Marks, 2003; Bach and Flinders, 2004). Multi-level governance implies that important actors and centres of authority are dispersed across several administrative levels and that local, regional, national and supranational authorities interact with each other across different levels of government (vertical dimension) and with other relevant actors at the same level (horizontal dimension) (Paraskevopoulos, 2002). In the Swedish context, a dispersion of actors and authorities across various tiers of governance certainly occurred during the 1990s, but this has been replaced more recently by a concentration of activation responsibilities at the central level, implying a reduction in the number of actors involved in the policy area. The chapter will start with a short overview of trends of unemployment and social assistance recipiency in Sweden, followed by an outline of the current governance structure for the administration of labourmarket policy and the social assistance scheme, with the main focus on steering responsibilities and actors involved. In a subsequent section, the main trends of reforms concerning governance of activation and income protection for the unemployed over the previous 20 years are identified. Finally, the effects of these reforms are considered, and the case of Sweden will be discussed in the light of the governance regime types outlined in Chapter 1.

Tomáš Sirovátka

198

Decentralization and Back to Centralization

Unemployment and social assistance in Sweden The changes in governance that have taken place must be understood in relation to the dramatic changes that took place in the labour market. The Swedish labour market was for a long time characterized by very low unemployment rates and high levels of labour-force participation. However, mass unemployment hit Sweden in the 1990s, and in the course of that decade a considerable part of the population was affected by unemployment at some point in time (Korpi and Stenberg, 2001). The downturn also led to a rise in the number of social assistance recipients. The situation for people dependent on either the social assistance scheme or other social security programmes can be characterized by a tightening of eligibility requirements and less generous compensation levels (Bergmark 2000, 2003). Even if expenditures for social assistance during the twentieth century in general have accounted for a relatively marginal part of total social expenditures in Sweden, the increase in social assistance expenditures during the 1990s has often been characterized as dramatic. The share of social assistance recipients in the population rose from 6 per cent in 1990 to 8.5 per cent in 1997, and expenditures almost doubled (Bergmark, 2000). Since social assistance expenditures are financed by the municipalities, the development of social assistance recipiency during the 1990s represented an extraordinary burden on the municipal economy. From 1998 the situation improved for a few years, but worsened again with the financial crisis of the late 2000s.

The governance structure Since the elections in 2006, Sweden has been ruled by a centre-right majority government. Prior to this change Sweden had been ruled by the social democrats since the early 1980s, with only a short interruption from 1991 to 1994. The territorial level of government in Sweden has a three-tiered structure (national, regional and local levels), with a clear division of responsibility regarding the regulation, administration and financing of various public policy areas. Regulation and administration of labour market policy Labour-market policy is a national issue, with the central government as the principal actor. The government regulates the policy area by formulating policies and goals and by distributing resources. The Ministry of Employment is responsible for labour-market policies and their

Tomáš Sirovátka

Renate Minas

199

implementation, the employment offices, the adaptation of work and rehabilitation focusing on working life, as well as other labour-market issues relating, among other things, to people with disabilities and unemployment benefit. Several national acts and ordinances regulate and control activation and labour-market policies. The policy area is considerably regulated (top-down), regarding both the aims and the means of achieving these aims. General guidelines for PES activities are laid down in the Labour Market Policy Activities Ordinance, which includes annual targets for labour-market policy established in resolutions on the government’s budget bill. More specific objectives or quantifiable operational targets are given in the government’s appropriation warrant and in other directives. Thus, in order to achieve the overall goal of an effective national labour market with full employment and good economic growth, management by objectives is the principal instrument for governmental control over the agency responsible for organizing labour market policy – the Swedish PES. The Swedish PES is a public authority responsible for organizing and coordinating labour-market policy. It passes resources to geographical labour-market areas, but also has the autonomy to formulate its own aims and guidelines. The Swedish PES is subdivided into four supra-regional areas that, in turn, are subdivided into sub-regional labour-market areas with a total of 320 local PES. The PES implement the central government decisions at the local level. Rules, laws and guidelines regulate the work of the PES, but the local officers have considerable latitude in deciding on the appropriate action in individual cases. The Swedish parliament also legislates on unemployment insurance and provides the main financing. However, administration of unemployment insurance in Sweden is organized somewhat differently to that in many other countries. It is not administered by the state or local social insurance offices, but by independent unemployment insurance funds normally affiliated to a trade union. The unemployment funds play a significant role in administration of unemployment insurance benefits by making decisions on compensation and handling of payments. Both the unemployment funds and the PES are supervised by a special government agency, the Swedish Unemployment Insurance Board. Actors at the sub-national level are the 68 geographical supra-regional labour-market councils. These councils have a purely advisory function by mapping local labour requirements and discussing questions that are of relevance to the local labour market. An additional important task

Tomáš Sirovátka

200

Decentralization and Back to Centralization

is improving cooperation between businesses, municipalities and other organizations with respect to the geographical labour market. Private actors are part of the employment service market, predominantly with respect to delivery of services. In 2008 the Swedish PES was required by the government to ensure that so-called complementing actors (private actors) are part of the agencies’ range of services. The Swedish PES is required to offer purchased services or services performed by private actors for one-third of the unemployed participating in the Job and Development Guarantee or the Job Guarantee for Youth. In 2009 approximately 20 per cent of the target group had contact with a private actor at some point during the year (Bennmarker et al., 2009). In absolute numbers, 25,554 unemployed participated in the Job and Development Guarantee and 15,642 in the Job Guarantee for Youth (Arbetsförmedlingen, 2010a).

Regulation and administration of social assistance With regard to social assistance, one key distinction in relation to the unemployment compensation scheme is the different rationale upon which social assistance is built. Social assistance is a selective benefit involving a test of the person’s financial resources or means. There is therefore no general right to social assistance; instead, eligibility is connected to individual circumstances. In contrast to the mainly national governance of labour-market policy, social assistance is a municipal task. Regarding activation of social assistance recipients one can therefore see references to a dual system of activation policies where the municipalities are responsible for the uninsured unemployed and the PES for the insured unemployed. Although the benefit is regulated by national law (Social Service Act 2001, No. 453), the specific construction of the act gives municipalities and individual social workers extensive discretion in deciding over benefits levels, duration of benefit receipt and demands regarding participation in activation measures in individual cases. The Act is often described as a framework law, and thus determines only in a general manner the eligibility for social assistance. Implementation of the Act is instead handed over to local municipalities that have the autonomy to formulate their own local policies and, furthermore, are responsible for financing social assistance. Thus, with respect to decision making, Swedish municipalities enjoy considerable autonomy and local politicians are comparatively independent from national government (Page and Goldsmith, 1987). Other Nordic countries have similar features, and there is sometimes talk of a ‘Nordic

Tomáš Sirovátka

Renate Minas

201

model’ for the municipal system, including municipal taxation rights, municipal self-governance established within the constitution and a general mandate that, within certain limits, allows the municipalities to define the scope of municipal concerns (Bergmark and Minas, 2008). In spite of that, central government regulates municipal activities through various steering mechanisms (constitutional regulation, more specific legislation such as the social service act, supervision and so on) and has the power to either enlarge or reduce the municipal scope of authority. Besides parliament and government (which each year sets the national benefit standard for social assistance), another important central actor in this policy area is the National Board of Health and Welfare. Its role is to interpret the legal practice and give support and recommendations regarding the assessment of individuals’ right to social assistance. A further task is the supervision of the social services. To summarize, the current governance structures of the two policy areas are quite different and separated from each other. This is true for the level of governance responsible for regulating, administrating and financing the policy area, as well as the type and number of actors involved. Furthermore, steering mechanisms differ between a policy area where regulations, management by objectives and monitoring specify the client treatment versus a policy area that is regulated by a less detailed, so-called framework law giving municipalities and case workers a considerable degree of discretion. Nevertheless, the reforms that took place during recent decades changed the balance of power between the central and the local levels and resulted in the emergence of new actors. In the following section these reforms will be placed in a broader context, pointing to more general trends in governance reforms.

Governance reforms: principles, actors and capacities What, then, are the main trends of the reforms in Sweden concerning governance of activation, and did these reforms change the nature of governance? The concept of multi-level governance draws attention to the fact that influence over a certain area is normally dispersed, negotiated and shared between actors at different levels. Furthermore, institutional changes are often stated to have essentially decreased the influence of central state, increased regional/local discretion and institutionalized different modes of vertical and horizontal coordination. When it comes to reforms in the governance of unemployment and social assistance schemes that have taken place in Sweden, the picture is partly consistent with that description but there are also discrepancies.

Tomáš Sirovátka

202

Decentralization and Back to Centralization

Key changes in governance structures point to decentralization trends and the appearance of new actors carrying out activation programmes. In the field of national activation policies private actors emerged performing national programmes, whereas at the local level municipalities took the role of organizing a large variety of local measures. However, there have also been cases of re-centralization. These developments will be elaborated upon in more detail below, starting with modifications along the vertical dimension, for example shifts that imply decentralization and re-centralization processes. Secondly, modifications at the horizontal level will be highlighted. Vertical governance: territorial changes regarding authority and areas of responsibility Shifts of responsibility falling between these levels are not novel or specific to the previous 20 years. By the second half of the twentieth century decisions at the national level to transfer authority from the national to the local level or to assign new responsibilities for the implementation of reforms to the local level resulted in growth of the municipal sector and in its significance for welfare provision (Bergmark and Minas, 2007). A decentralization of responsibility for activation policies for the insured unemployed as well as for social assistance recipients in the early 1990s continued this trend. None the less, recent shifts in territorial responsibility point to a trend away from decentralized to centralized governance of the policy area. We will start with the first trend, decentralization. Trends of decentralization In the 1990s a transfer of political autonomy from the central state to the municipalities implied an increased municipal responsibility for activation policies directed towards social assistance recipients. However, the new political responsibility followed a practice that already existed. Confronted with an increasing number of social assistance recipients and escalating social assistance costs, many municipalities had already started to organize vocational guidance, placement services and activation measures targeted at these recipients. This development then gathered further momentum with legal changes that shifted the political responsibility for unemployed individuals from the state to the municipalities. The ‘Act on Municipal Responsibility for Young People between 20–24 Years’ (1997, No 1268) stated that municipalities may offer employment to unemployed youths who had not obtained jobs, regular training or a place in a labour-market policy programme within

Tomáš Sirovátka

Renate Minas

203

90 days. The job should be full-time, and the programme could comprise both job experience and training. In addition, the revision of the Swedish Social Services Act in 1998 gave the municipalities the option of making participation in activation programmes mandatory for social assistance recipients between 20 and 24 years of age (and for other longterm unemployed young adults as well). The municipalities were also empowered to either refuse or lower the assistance given if the individual did not participate in the programme assigned without due reason. However, the municipalities were not obliged to offer programmes, and this is hence not a right for the unemployed. Although the PES for a long time has been the main provider of active measures to unemployed citizens, during the 1990s municipalities became providers of such measures to an increasing extent. One reason for this development is that the PES had been less successful in helping social assistance recipients find jobs. The explanation provided by the PES was that many unemployed social assistance recipients were too far from the labour market and therefore unattractive to employers. As a result of the rising number of local activation programmes that appeared – besides the national ones – a dual system of activation policies emerged. Those with a stable position in the labour market could rely on the first national tier of activation policies and higher levels of social security, whereas those with an unstable position in the labour market were forced to rely on social assistance and be exposed to broadly based municipal discretion with respect to quality and content of activation programmes (Salonen and Ulmestig, 2004). Other decentralization measures aimed at strengthening the role of the municipalities in local labour-market policies included the reorganization of the local employment service committees (today’s labourmarket councils). These committees were cooperative bodies at the local level with representatives of local business sectors, trade unions, employment offices and municipalities, and were organizationally a part of the previous National Labour Market Administration. A reorganization in 1996 was intended to increase municipal influence over labour-market policy and improve prospects for local adaptation of policy by giving the municipalities the majority within the committees (Ds, 1999, p. 54). Thus it became increasingly common for municipalities to establish special units with responsibility for labour-market policy. However, in the subsequent reform of the Swedish PES, these committees were replaced by the labour-market councils, and the role of the municipalities in local labour-market policies was reduced (see below).

Tomáš Sirovátka

204

Decentralization and Back to Centralization

Trends of (re-)centralization In the period from 1995, however, decentralization processes have been less frequent and shifts of power in the opposite direction have taken place. In part, these new policies can be understood as a reaction to large local variations in service availability and welfare standards. Thus, a number of reforms and national government actions were carried out that have mainly served to redirect the initiative to the state. An example not mentioned before but related to the design of social assistance is the introduction in 1998 of a monetary benefit standard for social assistance, replacing the monetary benefit guidelines introduced by the National Board for Health and Welfare in 1985. This standard was intended to guarantee a minimum standard, including a certain number of budgetary items identical countrywide, thus limiting the municipalities’ opportunity to introduce its own levels. Re-centralization was, in general, a consequence of the poor performance of local activation programmes and the difficulties in gaining an overall picture of the content and quality of the unnumerable existing local projects. One example of a reduction in municipal responsibility for activation policies was undoubtedly the elimination of the municipal youth programmes in 2007 that implied a transfer of responsibility back to the national PES. In its place, a national job guarantee for young people (16–24 years old) was introduced for youths who had been registered as unemployed for over three months. The reason for termination of the municipal programmes was evaluation studies showing the negative effects of programme participation on young unemployed peoples’ opportunities to establish themselves in the labour market (Sibbmark and Forslund, 2005; Forslund and Nordström Skans, 2006). It was argued that the young unemployed had lost contact with the national PES and that it would be too difficult for the government to follow up, steer and control municipal activities for unemployed youth. Activation of unemployed youth in the first 90 days of their unemployment and of older social assistance recipients is, by contrast, still performed at the local level. Following abolishment of the municipal youth programmes, the opportunities for the municipalities to organize activities for unemployed youth have become limited. The ordinance on the Swedish PES cooperation with the municipalities regarding labour-market programmes (SFS 2010, No. 83) regulates that only when the PES are unable to purchase a more suitable actor will it be permissible to sign a contract with the municipalities.

Tomáš Sirovátka

Renate Minas

205

The trend to concentrate responsibility for activation of the unemployed with the PES has continued. In 2010 the government passed legislation transferring the responsibility for coordinating measures for the integration of immigrants from the municipalities to the PES (Lagen om etableringsinsatser för vissa nyanlända invandrare 2010, No. 197). The PES are in charge of initiating a so-called establishment plan that will help the immigrant to integrate into the labour market. A strengthening of the central level was also implicitly implied in the merger of the National Labour Market Board and county labourmarket boards into the Swedish PES (2008), at the local and regional levels’ expense. As mentioned earlier, the municipalities lost their influence over local labour-market policies and the county labour boards were closed down. Even if the intention was to guarantee more efficient government of labour-market policy and to increase the quality of the organization, the new organization faced a whole range of problems and coordination with the local level becoming more problematical. To summarize, whereas the early 1990s was a period of increasing decentralization, we can now see a trend towards a stronger steering of activation policies by the central level, linked to a weakening of sub-national tiers of governance.

Horizontal governance: changing balance of power between relevant stakeholders Changes in governance structures also occur by altering the balance of power between relevant stakeholders within one level. An important part of the Swedish model has been the agreement of the social partners to work towards full employment. The Saltjöbaden agreement of 1938 symbolizes the cooperative will of the social partners, evident also in subsequent tripartite agreements. Thus, the administration of labour-market policy was for a long time based on cooperation between the Swedish PES (then the National Labour Market Board directorate), labour market authorities at the regional and local levels and the social partners (Svenningsson, 2003). Nevertheless, in 1990 the employers withdrew from all formal tripartite cooperation; they decided to withdraw from most of the institutions based on social partnership, such as the National Labour Market Board and the local employment committees (Svenningsson, 2003). They found it more beneficial to stay outside these arrangements, arguing that the old solutions do not fit the requirements of a new, flexible labour market.

Tomáš Sirovátka

206

Decentralization and Back to Centralization

The position of the trade unions as an important partner in labourmarket policies was recently further challenged when the newly elected right-wing government decided to increase the unemployment insurance funds’ membership fees. The government’s main motive for this change was to alter the conditions for wage setting by weakening those trade unions having a high number of unemployed members. The increase resulted in a substantial drop in membership in the unemployment insurance funds: within three years 260,000 people left the unions and even more (430,000) left the funds (http://www.lotidningen.se/? id_item=26588). Some of the trade unions’ unemployment insurance funds have now been merged to meet the new development. After heavy criticism and the fact that many unemployed that left the unemployment funds were without unemployment insurance and were obliged to apply for social assistance, in 2009 the Swedish Parliament opted for reduced membership fees. Introduction of New Public Management instruments In the debate on modes of governance, it is argued that the emergence of New Public Management has had a significant influence in promoting institutional changes with regard to performance, quality and case management. Performance management in the public sphere is often carried out by an authority setting the overall policy aims for a particular policy field. The aims are then operationalized into quantifiable national targets and indicators and implemented by, for instance, national PES. Here sub-national targets are subsequently negotiated with regional and local (labour-market) authorities, who are monitored and regularly reviewed (Weishaupt, 2010). In this context, Management by Objectives, that is quantitative output targets concerned with the effectiveness and efficiency of particular instruments, has become increasingly popular. Within Europe, Sweden and Norway have the longest experience with Management by Objectives systems, which in the field of labour-market policy were introduced in the mid-1980s. None the less, in the early years it had only limited impact because there were neither quantified target levels nor performance incentives. This situation changed in 1997 when control was improved and, most importantly, target levels were introduced. The more effective implementation of the Management by Objectives approach was first the consequence of the introduction of a computerized management information system (LEAD), which made it easier both to detect whether target levels had been reached and to implement a real feedback process. LEAD presents statistical information at the national, regional and local levels and makes it possible to update

Tomáš Sirovátka

Renate Minas

207

results regularly. Monthly follow-ups of the national labour-market policy objectives have been conducted, as well as the provision of weekly information on numbers of registered jobseekers. A stricter implementation of the results of the county performance in some cases even led to the replacement of county public employment services’ directors for not implementing the Management by Objectives approach sufficiently rigorously (Mosley et al., 2001). Private actors In the early 1990s private actors played only a minor role in the delivery of labour-market services. Although a purchaser–provider split model for labour-market training was introduced in Sweden in the mid-1980s, ten years later a government agency (AMU-gruppen) still bore the main responsibility for vocational training and carried out more than 40 per cent of all educational programmes. This changed over time, as the number of activation measures rose and policies altered. When massive unemployment hit Sweden at the beginning of the 1990s there were four activation programmes available, and by the end of the decade the total range of options was considerably extended. This significant increase in activation measures coincided with the growing importance of private actors in carrying out certain programmes. Consequently, by 1999 the agency’s share of vocational training programmes had dropped to 22 per cent (Ds, 2000, p. 38). By this time, the agency had also been transformed into a governmental company. Another factor behind the increased number of private actors was the Public Procurement Act introduced in 1994 (SFS 1992, No. 1528). The proportion of participants in training measures offered by private companies increased from 15 per cent in 1992 to 55 per cent in 1999. The opening of labour-market policies for private solutions accelerated in 2007, with an experiment in three counties in which employment services were organized by private actors (so-called complementing actors) during a trial period of six months. The services were purchased by the PES and included job coaching for 700 unemployed. The complementing actors were paid in three stages: when they accepted the client, when the client obtained a job and when the client had held the job for three months (Sjögren Lindquist, 2007). With the reorganization of the Swedish PES in 2008 this cooperation with private actors was expanded considerably, and the PES was commissioned by the government to purchase private employment services. The government’s assignment for 2008 stated that the employment service shall guarantee external actors the opportunity to function as complements

Tomáš Sirovátka

208

Decentralization and Back to Centralization

in the activities of the PES. It was emphasized that the other actors’ knowledge and competence would be important in helping the unemployed find jobs, in particular with respect to unemployed youth and the hard to employ unemployed. At least one-third of all participants within the Job and Development Guarantee and the Youth Guarantee schemes should be offered measures purchased from external actors. During the first half of 2010, 970 private actors supplied employment services to almost 120,000 unemployed for roughly 715 million SEK (approximately 70 million ¤) (Arbetsförmedling, 2010b). There are clear signs of an increased level of privatization in the delivery of national labour-market policy, but more systematic evidence regarding market influences on local labour-market policies is difficult to find. In bigger cities there exists well-established cooperation between public and private actors (for example, Jobbtorg Stockholm), but it is difficult to obtain a comprehensive picture for the country as a whole.

Inter-agency cooperation Cooperation between different welfare agencies across different tiers of government and within one level is a common phenomenon in Sweden, and has been so for many years. However, the form and content of cooperation vary extensively, especially at the local level within and among municipalities. The Social Service Act obliges local social welfare boards to cooperate with other relevant actors, and in many municipalities the boards have developed a close relationship with a large variety of different actors (for example, the local PES, the social insurance offices, voluntary organizations, health centres and local enterprises). These various cooperative initiatives can, however, be more or less institutionalized, concern development of measures for special groups of recipients, handle joint cases or develop common policies, and so on (Minas, 2008). Furthermore, several studies have pointed to obstacles related to organizational and cultural differences within the agencies, divided loyalties of the officials and limited resources available to deal with the complex needs of the clients (see, for example, Norman and Axelsson 2007). At the national level, too, a number of collaborations between various welfare institutions can be found. In the field of vocational rehabilitation, for example, new legislation in 2003 – the Act on Financial Coordination of Rehabilitation Measures (2003, No. 1210) – made it possible for various institutions (PES, the regional health authority, the

Tomáš Sirovátka

Renate Minas

209

municipal social service and the national social insurance administration) to build local associations for financial coordination (Ahgren et al., 2009). All the institutions involved contribute towards financial coordination. The services provided can be summarized as preventive and promotional activities aiming to improve health and to prevent sickness absence and social exclusion; and socio-medical activities, including a treatment programme for early and coordinated rehabilitation and occupational activities aiming to get people back into work, or into a rehabilitation programme, as soon as possible. These activities are carried out by multidisciplinary teams consisting of professionals from the different sectors and institutions involved, for example physicians, nurses, physiotherapists, psychologists, economists, lawyers and social workers. The teams are supervised by coordinators appointed by the association. In general, one can say that there is a desire for more institutionalized cooperation from the viewpoint of the municipalities that, however, is not shared by the governmental agencies. As an example, the Swedish Association of Local Authorities and Regions works towards a one-door policy for unemployed people, an initiative that is supported by 90 per cent of the municipalities but has received no support from neither the government, Swedish PES nor the Swedish Social Insurance Agency. In contrast, the more recent reorganizations of these latter two state agencies have resulted in more complicated relations between them and the municipalities. These governmental agencies point out that the existing level of cooperation is sufficient and that it is difficult to cooperate in a more institutionalized way with the municipalities. To date, no plans for integrated activation services directed towards the unemployed far from the labour market are visible. A more comprehensive reform of the social insurance system is, however, the task of a parliamentary commission that started its work in September 2010.

Changed modes of governance? Effects and governance regime types This chapter aimed to identify the main trends over the last 20 years in Swedish reforms concerning governance of activation and income protection for the unemployed, and to question whether one may observe a closer interlinking of activation policies and income protection for the unemployed. On both counts, it seems necessary to point to differences with respect to the regulation of the policy area and the delivery of the services.

Tomáš Sirovátka

210

Decentralization and Back to Centralization

The regulation of activation policies at both national and municipal levels ties benefit receipt to participation in activation programmes in a stricter way than at the beginning of the 1990s and, in addition, tightened the eligibility criteria for income protection. However, if we consider the division of work between the unemployment insurance funds and the PES, the situation has not changed. The PES are responsible for providing information to jobseekers and check that they comply with the conditions for receiving unemployment insurance, and the unemployment funds decide on and pay unemployment insurance. The situation starts to change on the delivery side, however, since the PES are no longer exclusively responsible for the delivery of service since private providers made their appearance. The situation is somewhat different at the municipal level. With the development of municipal activation programmes, a closer interlinking of the organization of activation policies and social assistance benefit has certainly taken place. A stricter conditionality has emerged that binds activation and benefit receipt more tightly, also in the sense that one actor – the municipal social services – is (or has been) responsible for both aspects. What the new regulations, the concentration of activation measures within the Swedish PES, imply for the connection between activation policies and social assistance receipt is too early to foretell. The closer interlinking of work and welfare can be seen as a result of various governance processes. Two processes stand apart. One is a stricter state regulation of national and municipal activation policies, the second a stronger influence from market mechanisms that primarily influences the administration and delivery of national activation programmes. Stronger state regulation is manifested, for example, by the introduction of the national benefit standards, the changes in the Social Service Act that regulated the municipalities’ responsibility for unemployed youth, the close-down of municipal activation programmes and the concentration of tasks within the PES. The basis for this state regulation is traditional legal instruments. Behind these processes of re-centralization was increasing local variation in service availability and welfare standards, but it was also a reaction against a lack of efficiency in, especially, local activation programmes. Thus, the initial argument for decentralization, a concern with efficiency where some matters simply are handled better in smaller units or closer to the citizens and service users, was no longer relevant. Instead, the aspect of the overall responsibility of the government for welfare within the country was stressed in the sense that the government has to make sure that certain

Tomáš Sirovátka

Renate Minas

211

minimum levels of welfare services are upheld. In more recent years an ambition to strengthen the control of work within the social insurance sphere and a more uniform application of laws and regulations strengthened re-centralization processes. This is particularly true with respect to the large emphasis put on the work principle by the current government. It is also interesting that shifts towards re-centralization have been performed by both social democratic and centre-right governments. Looking at the driving forces behind the shifting responsibility for activation policies upwards or downwards, these are actually legitimized by the same argument: increasing efficiency. It has been observed elsewhere that the same claims are sometimes made in favour of decentralization and re-centralization processes. In addition, in ‘practice shifts between centralization, decentralization and re-centralization, defined as shifting the power and responsibility over policies towards the national and local level respectively and back again, seem to be ongoing cycles in which trends and taking sides in the discussion succeed one another continuously’ (de Vries, 2000, p. 194). The more recent transfer of power from the local to the national level in Sweden was largely a consequence of bad performance of local activation programmes. However, evidence supporting that statement is somewhat difficult to find. There are few studies that looked into the efficiency of local labour-market programmes for social assistance recipients. This low number is partly due to problems with systematic documentation and following-up of those programmes (Milton and Bergström, 1998; Salonen and Ulmstig, 2004; Thorén, 2005; Giertz, 2007; Lundin, 2008). Furthermore, the results of these studies are difficult to utilize since they often concern small projects that faced difficulties in finding control groups. Nevertheless, the general impression of these studies is that municipal programmes directed towards this group do not have positive effects (Giertz, 2007). With respect to youth programmes that were once part of the national labour-market programmes but organized by the municipalities, two studies exist that examined the efficiency of that particular measure. Carling and Larsson (2005) studied the effect of programme participation on the subsequent labour-market attachment. They found a slightly increased probability of employment during the first 120 days of unemployment, suggesting that the guarantee works more as a threat than a promise. However, the impact disappears over time. Another finding was that only less than one-third of the target group was assigned to a programme within the promised 100 days. The authors surmised that the reluctance to put the

Tomáš Sirovátka

212

Decentralization and Back to Centralization

programme into effect at the local employment offices may have been due to either the offices mistrusting the municipal authorities or economic disincentives. Forslund and Nordström Skans (2006) also studied the municipal youth programmes with respect to various labour-market outcomes and came to a similar conclusion, in that the municipal youth programmes showed few positive effects. However, some positive effects for the participants’ likelihood of studying (at a level lower than university) could be found. The second governance process is the increased influence of market mechanisms by the introduction of management by objectives and the increasing participation of private actors in the delivery of services. Market mechanisms are more concerned with administration and delivery of national programmes. In terms of governance, market mechanisms are most obvious at a horizontal level when combining governmental interference with market-oriented actors. Here, the argument of efficiency is brought forward by stating that market-oriented actors increase efficiency and effectiveness. Wide dissatisfaction with the costs and effectiveness of government programmes were important driving forces for introducing these new instruments. The introduction of private actors was strongly promoted by the conservative governments that launched the Public Procurement Act in 1994 and reinforced the development towards more competition after the election in 2006. However, the social democratic government in the mid-1990s did not reverse the change from 1994. The emergence of private or complementing actors is also a very recent phenomenon in Sweden, and only very few studies have evaluated its effects. One study carried out by Bennmarker et al. (2009) focused on the above-mentioned experiment in three counties in which employment services were organized by private actors. Their results showed that the unemployed using private services were more satisfied with the help they received and had a closer contact with their case workers than the unemployed who had contact with a PES. However, the authors found no differences between the two forms of employment service regarding the unemployed’s chances of finding a job. Private actors were somewhat better at helping those unemployed with immigrant backgrounds, whereas PES were better at helping unemployed youth to find a job. What does this development in Sweden mean for the discussion about governance regime types? Considine and Lewis (2003) point to the declining popularity of the procedural model of bureaucracy in the 1970s, with a corporate model emerging in the 1980s and the market

Tomáš Sirovátka

Renate Minas

213

governance model of the 1990s. However, the development in Sweden indicates a high reliance on legislation and state regulation in a traditional bureaucratic way. Irrespective of the challenges of the previous 20 years, a hierarchical order of decision-making processes in the area of activation policies can be found. The initial transfer of autonomy to the local level and then back to the central level strengthens the picture of a national government that has the ultimate power to decide how far decentralization should be carried through: a national steering and control of policy making through legislation and top-down steering. The re-centralization measures that have been carried out during recent years have served to secure the political ambitions of national government at the expense of local self-regulation. However, governance of activation policy in Sweden cannot be placed exclusively into one of the governance types mentioned in Chapter 1. Characteristics of other governance types can also be identified, resulting in a certain mixture of governance principles regarding activation policies. The setting of quantifiable national targets and indicators that are monitored and regularly reviewed is thus a typical feature of corporate governance that emphasizes goal-oriented planning and management improvement. None the less, market mechanisms are undoubtedly an important element in today’s labour-market administration and delivery of primarily national activation policy, where contracting-out and competitive tenders have become general practice. To summarize, we can identify a hybrid form of governance regarding activation policies.

References Ahgren, B., S. Bihari Axelsson and R. Axelsson (2009) ‘Evaluating intersectoral collaboration: A model for assessment by service users’, International Journal of Integrated Care (IJIC), 9/1, e03. Arbetsförmedling (2010a) Resultatrapport om kompletterande aktörer samt intern coachning, Rapport 2010 (Stockholm: Arbetsförmedling). Arbetsförmedling (2010b) Arbetsförmedlingens återrapportering. Fördjupade analyser (Stockholm: Arbetsförmedling). Bach, I. and M. Flinders (2004) Multi-Level Governance (Oxford: Oxford University Press). Bennmarker, H., E. Grönqvist and B. Öckert (2009) Betalt after resultat – utvärdering av försöksverksamhet med privata arbetsförmedlingar, IFAU Rapport 2009:23 (Uppsala: Institutet för arbetsmarknadspolitiska utvärderingar). Bergmark, Å. (2000) ‘Socialbidragen under 1990-talet’, in Å. Bergmark (ed.) Välfärd och försörjning, SOU 2000:40 (Stockholm: Fritzes). Bergmark, Å. (2003) ‘Activated to work? Activation policies in Sweden in the 1990s’, RFAS, 4, 291–306.

Tomáš Sirovátka

214

Decentralization and Back to Centralization

Bergmark, Å. and R. Minas (2007) Decentralized Welfare or Universal Standards? The Transformation of Territorial Authority and Local Discretion in Sweden, Presentation at the Annual Meeting of the American Sociological Association 2007. Bergmark, Å. and R. Minas (2008) ‘Ska makten över välfärden decentraliseras?’, Framtiden, 4, 27–9. Carling, K. and L. Larsson (2005) ‘Does early intervention help the unemployed youth?’, Labour Economics, 12/3, 301–19. Considine, M. and J. M. Lewis (2003) ‘Bureaucracy, network, or enterprise? Comparing models of governance in Australia, Britain, the Netherlands, and New Zealand’, Public Administration Review, 63/2, 131–40. Departementsserien (Ds) (1999) Samverkan för färre bidrag och fler jobb, DS1999:54 (Stockholm: Socialdepartementet). Departementsserien (Ds) (2000) Från kartläggning av behov till genomförande av arbetsmarknadsutbildning, DS 2000:38 (Stockholm: Socialdepartementet). De Vries, M. (2000) ‘The rise and fall of decentralization: A comparative analysis of arguments and practices in European countries’, European Journal of Political Research, 38, 193–224. Forslund, A. and O. Nordström Skans (2006) ‘Swedish youth labor market policies revisited’ Working Paper 2006:6 (Uppsala: IFAU Institutet för arbetsmarknadspolitiska utvärderingar). Giertz, A. (2004) Making the Poor Work: Social Assistance and Activation Programs in Sweden (Lund: Lund University). Hooghe, L. and G. Marks (2003) ‘Unraveling the central state, but how? Types of multi-level governance’, American Political Science Review, 97, 233–43. Korpi, T. and S.-Å. Stenberg (2001) ‘Massarbetslöshetens Sverige: Arbetslöshetens karaktär och effekter på individers levnadsförhållanden’, in J. Fritzell, M. Gähler and O. Lundberg (eds) Välfärd och arbete i arbetslöshetens årtionde SOU 2001:53 (Stockholm: Fritzes). Lundin, M. (2008) Kommunerna och arbetsmarknadspolitiken IFAU Rapport 2008:13 (Uppsala: Institutet för arbetsmarknadspolitiska utvärderingar). Milton, P. and R. Bergström (1998) Uppsalamodellen och socialbidragstagarna – En effektutvärdering CUS Skrift 1998:1 (Stockholm: Centrum för utvärdering av socialt arbete, Socialstyrelsen). Minas, R. (2008) ‘Arbete med socialbidrag. Organisation och metoder, kontinuitet och förändring’, in Å. Bergmark, R. Minas, T. Lundström and S. Wiklund (eds) Socialtjänsten i blickfånget – Organisation, resurser och insatser. Exempel från arbete med barn och ungdom, försörjningsstöd, missbruk (Stockholm: Natur och Kultur). Mosley, H., H. Schütz and N. Breyer (2001) Management by Objectives in European Public Employment Services Discussion Paper FS I 01-203 (Berlin: Wissenschaftszentrum Berlin für Sozialforschung). Norman, C. and R. Axelsson (2007) ‘Cooperation as a strategy for provision of welfare services – A study of a rehabilitation project in Sweden’, The European Journal of Public Health, 17/5, 532–36. Page, E.C. and M.J. Goldsmith (1987) (eds) Central and Local Government Relations (London: Sage). Paraskevopoulos, C.J. (2002) ‘New institutionalism and EU regional policy: Multi-level governance by “thin” or “thick” institutions?’, Current Politics and Economics of Europe, 11/2, 59–79.

Tomáš Sirovátka

Renate Minas

215

Salonen, T. and R. Ulmestig (2004) Nedersta trappsteget. En studie om kommunal aktivering Rapportserie i socialt arbete, 1:2004 (IVOSA, Växjö University). Scharpf, F.W. (2001) European Governance: Common Concerns vs. the Challenge of Diversity, MPIfG Working Paper 01/6. Sibbmark, K. and A. Forslund (2005) Kommunala arbetsmarknadsinsatser riktade till ungdomar mellan 18 och 24 år IFAU rapport 2005:9 (Uppsala: IFAU Institutet för arbetsmarknadspolitiska utvärderingar). Sjögren Lindquist, G. (2007) Unemployment Insurance, Social Assistance and Activation Policy in Sweden. Peer Review: Implementierung the news basic allowance for job seekers, Berlin. Stoker, G. (1998) ‘Governance as theory: Five propositions’, International Social Science Journal, 50/1, 17–28. Svenningsson, L. (2003) ‘Sweden, new pathways to labour market policy’, in OECD, Managing Decentralisation: A New Role for Labour Market Policy (Paris: OECD). Thorén, K. (2005) Kommunal aktiveringspolitik: en fallstudie av det praktiska arbetet med arbetslösa socialbidragstagare IFAU rapport 2005:11 (Uppsala: Institutet för arbetsmarknadspolitiska utvärderingar). Weishaupt, T. (2010) ‘A silent revolution? New management ideas and the reinvention of European Public Employment Services’, Socio-Economic Review, 8/3, 461–86.

Tomáš Sirovátka

11 Governance of Integrated Activation Policy in Finland Vappu Karjalainen and Peppi Saikku

Towards an integrated approach in activation Implementation of activation policy in Finland has been a question of coordination between the state (Public Employment Services (PES); Social Insurance Institution (KELA)), the municipalities (social welfare, health care, rehabilitation) and other actors (service producers, employers, third sector, education). In particular, the collaboration between the state (PES) and municipal social welfare has been crucial in activation because the main target groups of activation policy have been the long-term unemployed, who are often clients of both the employment office and the social welfare office. The collaboration towards integrated services has continued for two decades. Activation policy arrived in Finland mainly via the European Union (EU) and the EES (European Employment Strategy), and was prepared within EU institutions. The first phase of activation in Finland started in the early 1990s, with a series of activation projects funded by the European Social Fund (ESF). The target populations of the projects were the long-term unemployed and jobseekers with disabilities. The aim was to develop rehabilitation and activation measures, create good practices and support the long-term unemployed in re-entering the labour market. The results of the ESF projects were reasonable but not sustainable. However, many good activation practices were developed and documented, although reflections on the dissemination of practices were seldom an objective of the developmental work (Mannila et al., 2001). On the whole, activation projects were important in networking various local actors and in combining activation resources within local environments; nevertheless, the project boom did not bring about any major structural changes in governmental organization of public services. 216

Tomáš Sirovátka

Vappu Karjalainen and Peppi Saikku 217

Activation policy was implemented in the existing administrative structures of employment policy (state), social insurance (state) and social policy (municipalities), with the emphasis on closer cooperation between these sectors and on networking with the service producers and the third sector. Cooperation between local-level public actors presupposed a certain degree of freedom in arranging coordinated services and, indeed, the autonomous position of municipalities (social welfare, social assistance, health care) since 1992 has facilitated this process. Employment services had been state driven, organized by the state and decentralized regionally and locally during the 1980s. The second phase of activation took place around the year 2000. According to the new regulations set out in the Social Assistance Act (1998), a person could be subject to a maximum 40 per cent reduction in social assistance for refusing the offer of work or training. A similar conditionality in the form of an interruption of employment benefit payment had already existed in employment services since 1994. The Rehabilitative Work Experience Act (2001) imposed additional obligations on the long-term unemployed. On receiving labour market support (benefit) or social assistance for an extended period (maximum 500 days/180 days), the recipient was obliged to participate in drawing up an activation plan. Regulations were binding on unemployed persons aged under 25 years. The Act also imposed an obligation on employment officials and social workers to cooperate in drawing up an activation plan with the client. The social rights of clients were weakened and mechanisms for controlling people who were longterm unemployed got stronger. Alongside the reforms, minor structural changes were made within related sector administration: the coordination of employment and social services in individual activation required collaborative arrangements (such as working pairs, teams) on a permanent basis between the sectors. The early signs of an integrated service structure could be identified between the sectors. The third phase of the Finnish activation policy was implemented between 2002 and 2007. In 2002, a pilot period started for interagency collaboration between employment services, social welfare and health services, and social insurance services. Joint service units for the long-term unemployed were established in 18 municipalities. Now the challenge for authorities was to coordinate public and outsourced services in individual cases in such a way that the available local resources would provide maximum support for an activation client. The evaluation results (Arnkil et al., 2004) of the pilot phase of the joint service units were positive, and from the beginning of 2004 Labour Force

Tomáš Sirovátka

218

Governance of Integrated Activation Policy in Finland

Service Centres (LAFOS units) became a permanent integrated service structure for the long-term unemployed.1 The establishment of LAFOS was based on the Government Employment Policy Programme and on a contract between state-level authorities. No new legislation was established. A kind of contractual phase in the development of horizontal coordination within the public sector began. The conditionality of labour market support (benefit) for the longtermed unemployed was further tightened in 2006. According to the new regulations the state and municipalities shared the passive costs of the last-resort benefits2 for the long-term unemployed in cases where the unemployed person does not participate in the activation measure offered after 500 days of unemployment. The reform worked as a carrot for municipalities: if municipalities facilitate participation of the longterm unemployed in activation, they are released from passive benefit costs because the state is primarily responsible for activation and the costs related to that. Co-financing motivated municipal actors to participate in the LAFOS collaboration with employment authorities and other parties of activation. The co-financing provided a crucial base for permanent coordination, thereby forming a structural basis for setting up joint service units. The ongoing fourth phase can be characterized as a period of developing intermediate labour markets3 and further strengthening of the obligation towards activation. For example, rehabilitative work became mandatory for the unemployed aged over 25 years from the beginning of 2010. LAFOS units are taking further steps in horizontal coordination, with an emphasis on more efficient activation and better employment results. The Ministry of Labour has merged with the Ministry of Trade and Industry,4 which as such illustrates the emerging direction in activation policy: better performance results in activation are expected. While the objectives of employment policy are getting more result oriented in activation, those of social welfare policy are hidden, that is, they are merging with the objectives of the employment policy. The Finnish activation paradigm has emerged during the past two decades. Step by step, the conditionality of activation has been strengthened and there are elements of a new paternalism in the Finnish activation policy (Keskitalo, 2008). The governance of activation has emerged as a policy area with its own implementation procedures within the public sector. Its development can be characterized as a shift from a single-sector (employment) approach to a cross-ministerial and multisector approach, with horizontal and coordination structures of its own. The path towards an integrated activation policy has proceeded in

Tomáš Sirovátka

Vappu Karjalainen and Peppi Saikku 219

stages, culminating in the establishment of permanent LAFOS units for the long-term unemployed. Activation policy has transformed the service paradigm of the welfare state from a purely sector-based silo to a multi-sector, joined-up service delivery with respective governance. This process makes fundamental demands on the administration and management of the public sector. In the following, we analyse the integration development from the point of view of governance. We argue that an integrated approach – or a network structure – for the delivery of activation services is more compatible with the needs of the long-term unemployed than a traditional sector-based service, but it involves many problems related to the arrangements of multi-sector governance.

LAFOS – a network approach As Chapter 1 depicts, the governance of activation policy is closely connected to a general, ongoing reorganization of welfare governance and administration. There are some key principles underlying public administration and management reforms that are generally discussed in many countries. Decentralization, marketization, partnerships and new public management are examples of the themes of this discussion. Implementation of activation policy is thus part of a wider governance development in each country, rooted in country-specific administration and governance traditions. Hence, the models of activation governance should be addressed in relation to different governance regimes and to governance models adopted in various countries. Mark Considine and Jenny Lewis (2003) have formulated four ideal types of governance in relation to synergies between the public and private sectors, with the starting point being that the state is a partner in creating new strategic collaboration and governance. The types are: procedural, corporate, market and network governance (see Chapter 1). While the public interest in classifying the types has been the starting point, attention has also been paid to the potential dynamism of the markets in the theoretical underpinnings of these governance types. This makes them useful and relevant when analysing the governance arrangements of activation. Each country certainly emphasizes different aspects in the governance of activation, and the priority of these aspects is highly dependent on the developmental stage of performance management (NPM) and the ethos of activation policy in that country. In Finland, there has been a tendency to develop integrated approaches in activation as a

Tomáš Sirovátka

220

Governance of Integrated Activation Policy in Finland

means of tackling structural unemployment. The integration process of sector policies and the establishment of LAFOS units have emerged without new legislation related to governance. Thus, instead of procedural governance arrangements, the driving force for these reforms has been the strengthening of the partnership within the public sector and with other relevant actors. This approach comes nearest to the network governance type. Network governance is an emerging newcomer to public governance. It provides an answer to coordination dilemmas posed by public multisector systems and aims to rely on outside agencies in the form of a stronger strategic partnership. The network approach is less motivated by rules and less defined by supervision, and more focused on building joint action and trust between collaborative actors. Fixed organizational boundaries and roles are not relevant in the same way as previously, and a new rationality based on the creation of shared aims and culture is emphasized (Considine and Lewis, 2003). Eva Soerensen and Jacob Torfing (2007) have been working on the concept governance network and define it as (1) a relatively stable horizontal articulation of interdependent, though operationally autonomous, actors (2) who interact through negotiations (3) that take place within a regulative, normative and cognitive framework (4) that is self-regulating within a limit set by external actors and (5) which contributes to the production of public services. To some degree, these features are recognizable in the integration development of Finnish activation policy and especially in joint service units (LAFOS). Governance network research is a research area that ‘has not yet become sedimented into a new paradigm with its own clear-cut definitions, taxonomies and methods’ (Soerensen and Torfing, 2007, p. 6). The research strategy is often based on retroduction rather than purely empirical induction or theoretical deduction. Thus, the focus of network governance research is usually on the backward mapping of the key policy actors and on the policy outputs produced by the collaboration and coordination of key actors. This strategy has guided the evaluation of LAFOS units in Finland. In the following sections the key findings of our evaluation research (Karjalainen and Saikku, 2008) are presented.

Multi-sector implementation of activation and joint services The Government Employment Policy Programme (2003–7) has been a major tool in creating the political consensus for shared responsibilities

Tomáš Sirovátka

Vappu Karjalainen and Peppi Saikku 221

and integrated policy arrangements. The Programme worked as a policymaking framework, being prepared inter-ministerially and politically approved with the aim of tackling structural unemployment.5 The Ministry of Labour was the leading agency in carrying out the budgeting and administrative planning for the Programme. The Policy Programme provided the political approval necessary to initiate the piloting of joint service for the long-term unemployed (2002–3) and the establishment of LAFOS units in 2004. The Ministry of Labour was responsible for the preparation of the reform in close cooperation with the Ministry of Social Affairs and Health, the Social Insurance Institution and the representatives of the association of municipalities. Services for the long-term unemployed relating to employment, social welfare, health, rehabilitation and social insurance were to be provided under the same roof. The establishment of LAFOS was voluntary for local employment authorities and municipalities. Although no new legislation was enacted, the Ministry of Labour reserved extra financing for local actors in the employment sector to start up a centre. The municipalities were responsible for funding half of the running costs. They did not receive any extra money from the state, only the ordinary annual state grants for municipal services. The administrative solution for the establishment of LAFOS was exceptional. The representatives of the central government, that is, the Ministry of Labour, the Ministry of Social Affairs and Health, the Social Insurance Institution and the Association of Finnish Local and Regional Authorities, entered into a contract and recommended that local-level actors establish LAFOS units. A contractual procedure seemed to be the only way to set up a new network organization that could operate between the sectors. Each sector had its own legislation and administration, and the enactment of new legislation for an integrated service structure was not supported by any party. LAFOS and the related policy making marked the launch of joined-up government (Ling, 2002) – or whole of government (WG) as it was later called (Christensen and Laegreid, 2007) – in the Finnish public administration. According to Pollitt (2003, p. 35) joined-up government has at least four underlying goals. At first, it can eliminate tensions between different policies and therefore aims to increase the effectiveness of these policies. Secondly, joined-up government makes better use of available resources through elimination of possible duplication between different sector programmes. Thirdly, it can improve the flow of good ideas and strengthen cooperation between various policy stakeholders, thus producing synergies. The fourth goal is related to providing a more

Tomáš Sirovátka

222

Governance of Integrated Activation Policy in Finland

integrated set of services from the point of view of the citizen. Since the efforts to tackle structural unemployment within one sector had proved ineffective, joined-up government was a natural path to finding progress. A general feature of joined-up government is that different public sector bodies coordinate their actions, while in joined-up governance public sector bodies cooperate not only with each other but also with service producers and third-sector and other actors (Pollitt, 2003, p. 38). Joined-up governance was becoming a framework for the further development of LAFOS. The public objectives of LAFOS emphasized the lowering of structural unemployment and expenditures on last-resort benefits.6 The task of LAFOS was to coordinate rehabilitation, activation and employment services individually with the expertise of a multi-professional team. The team consisted of experts from the employment office, social welfare office, health care, rehabilitation and social insurance. Compared with services provided in parent offices, the LAFOS service represented a specialized service unit for the long-term unemployed. For example, NAV offices – which are the equivalent joint service units in Norway – are frontline units charged with the task of providing general guidance and basic services (Andreassen et al., 2007). In recruiting personnel for LAFOS, it was beneficial to have expertise and former experience in guidance, rehabilitation and questions of employability. From the perspective of activation policy, LAFOS was a type of experiment: the hypotheses to be tested concerned whether a proper and individually organized service delivery would provide better activation policy results. From the point of view of an unemployed person, LAFOS represented a new opportunity to receive individually relevant services and long-lasting support in the efforts to re-enter into work. The LAFOS organization was a newcomer within the sector-based service system, from the point of view of both administration and service delivery (see Figure 11.1). Furthermore, as a network organization it represented a major managerial challenge: who are qualified to make decisions concerning strategy and the operational management of LAFOS activities at state and local level? The nature of a network organization is collaborative and connective; it does not have one centre but is multi-nodal and has many centres (Castells and Cardoso, 2006). Each network partner is concurrently an autonomous actor and a collaborative partner within the network. The LAFOS approach raised a new question with respect to network management. It opened up the challenge of the horizontal dimension of governance in the public sector as opposed to the former, mainly vertical, orientation.

Tomáš Sirovátka

Vappu Karjalainen and Peppi Saikku 223 State level

Ministry of employment and the economy

Ministry of social affairs and health

Social insurance institution

Assoc. of finnish local and regional authorities

Steering group

Regi onal level Local level

Regional office

TE-centre Municipal social work Local steering group Health services

PES-office

Local office

Management group Education

Lafos

Third sector

Service producers

Employers

Figure 11.1 Position of Labour Force Service Centres (LAFOS) within the administrative structure for activation policy

The Finnish welfare state has been developed within the framework of a Nordic model, with an emphasis on universalism and equality in social security and service delivery. This principle did not fully fit in with the establishment of a network organization. The legitimization of LAFOS is based on contractualism, not on legislation. LAFOS (39) covers about 40 per cent (161) of the total number of municipalities in Finland. Thus, unemployed citizens are in an unequal position with regard to receipt of LAFOS-services. In particular, the small and remote municipalities are outside the LAFOS network.

State-level steering mechanisms Integrated activation policy implied a completely new challenge for traditional sector-based public government. Model and tools for crosssector governance had to be created in state and local settings because no relevant models were available. State-level tools were needed to manage the LAFOS implementation. The Ministry of Labour set up a

Tomáš Sirovátka

224

Governance of Integrated Activation Policy in Finland

steering group for the purpose with representatives from the Ministry of Social Affairs and Health, the Social Insurance Institution, the Association of Finnish Local and Regional Authorities and regional government involved in labour administration (Figure 11.1). This was the body intended to oversee the implementation and follow-up of LAFOS. Nevertheless, the steering group’s mandate was not quite clear; it remained an intergovernmental coordination structure with the task of planning state-wide implementation. The steering group worked primarily as a negotiation forum and it had a mandate to provide recommendations to local-level actors concerning the organization of LAFOS. In addition, the steering group approved the establishment of LAFOS units although the formal decisions were made by the Ministry of Labour. In principle, the steering group could play a central role in strategic leadership for the integrated policy and implementation of LAFOS, but it never fully assumed this role. According to the evaluation (Karjalainen and Saikku, 2008), the main reason for the weak role of the steering group was the unclear jurisdiction of the group and its position as a decision maker. As a tool for network governance, the steering group consisted of a highly diverse array of state-level actors. The Ministry of Labour and PES at regional and local levels are state authorities with a relatively strong central and regional government, while the Ministry of Social Affairs and Health has little power over the municipalities, which are themselves responsible for providing social welfare and health services. Municipalities have an autonomous position in relation to central government, and the role of the Ministry of Social Affairs and Health is rather one of steering through information and recommendations than by regulations. The Social Insurance Institution is also a state organization with strong central government and centralized administration. The Association of Finnish Local and Regional Authorities is a nonprofit-making organization for municipalities, putting pressure on state authorities in order to keep the costs of municipalities at an acceptable level. In these circumstances, the steering group’s own management tools comprised state-level contracts between the parties involved and recommendations to local-level actors. In this sense the group proved to be a pure network organization. From the local perspective the statelevel steering group was not profiled as a decision maker, although the LAFOS-units had high expectations for that. A recommendation introduced by the steering group concerns the criteria for referring clients to LAFOS. While the goals of LAFOS were approved by the steering group, no shared performance targets or operational objectives were set up jointly. The issue of vertical and

Tomáš Sirovátka

Vappu Karjalainen and Peppi Saikku 225

horizontal accountability in a network structure is a difficult one, and no sustainable solution to this issue has so far been identified. The Ministry of Labour – later the Ministry of Employment and the Economy – has worked as a lead agency in the steering group, with the consequence that the other parties have regarded it as domineering. As a tool for network governance, the steering group has met all of the typical problems and challenges involved in the whole of government approach. Legitimacy, accountability, power relations and mutual trust between authorities are crucial issues when developing network governance within the public sphere (Agranoff, 2006; Christensen and Laegreid, 2007). At state level, one lesson to be learned is that top-down government and sector-based steering systems as such do not work in a network context. The vertical dimension of governance becomes secondary to the horizontal. Jurisdiction and practices for horizontal decision making are of importance.

Local-level management The regional level plays only a minor role in the implementation of activation policy in Finland. There are regional Employment and Economic Development Centres (T & E Centres) within the vertical employment administration that supervise employment offices and provide financing to local offices. The T & E Centres do not have a regional counterpart in the field of social and health care administration. Thus, from the point of view of LAFOS governance, the regional level is not pivotal. Contracts between employment authorities and municipalities at local level have been the basis for the establishment of LAFOS units. According to the locally formulated contract, both parties provide the unit with the same number of workers and share half of the expenditure required for the running of services (for example, the costs of localities and equipment). The management of the LAFOS units is similar to that of central-level management (see Figure 11.1). There is a steering group at the local level charged with the task of implementing LAFOS. The steering group is responsible for long-term planning, budgeting and partnership recruitment for LAFOS. Representatives from the employment sector, municipality (social welfare and health care), social insurance institution and other actors (for example, third sector) are members of the group. For the more operational management of LAFOS, a management group has been set up with directors from parent offices as members. Its task is to support the operational management of

Tomáš Sirovátka

226

Governance of Integrated Activation Policy in Finland

LAFOS that is in the hands of one or two leaders. The two-sector leadership is a type of partnership management, with one person in charge from both an employment office and a social welfare office. At first, the two-leader model was a solution but, over time, most units have transferred to a one-leader model. In the respective NAV units in Norway, unitary management – as they call it – has also been adopted (Askim et al., 2010). The evaluation of LAFOS revealed that problems existed in both strategic and operational management. The roles of the steering group and management group were unclear and overlapping. Both were weak in decision making and the mandate for that was limited. Since the groups had no legitimate role in an administrative sense, they were largely confined to a negotiative and consultative position. As network organizations the groups did not make decisions, they were more reaching agreements on certain important principles (see also Agranoff, 2006). Negotiations within the network context are always seeking knowledge, and they involve many learning aspects and the building of a new interaction culture (Senge, 1990). Stakeholders involved in the steering group and management group were working in a new administrative frame that challenged them to reflect on the legitimacy of their own decision-making positions. The operational leaders of LAFOS were nominated from within staff. According to the two-leader model, decisions made by one of the leaders were relevant to only the sector (s)he was representing. So, if the leader came from the municipality (s)he could not make decisions concerning workers from the employment office. The legislation and rules of each sector were also valid in the network context. In this way, procedural governance was involved in LAFOS governance. The need for clarification on the administrative position of LAFOS in relation to parent offices was evident. The one-leader model with specified rules was introduced in most LAFOS units. The autonomy of LAFOS in planning, setting up performance targets and managing everyday activities turned out to be a crucial issue. Accountability is a major challenge in network governance (Pollitt, 2003; Bryson et al., 2006). Within LAFOS, accountability was a problem from the very beginning. Each party also had sector-based follow-up and data information systems that were utilized in LAFOS units. Employment officials had a data-processing connection to the information system of the employment government, social workers to that of social welfare government, and so on. A shared data system or the possibility to link information systems across organizational boundaries was

Tomáš Sirovátka

Vappu Karjalainen and Peppi Saikku 227

needed. Linking separate systems proved to be difficult, and subsequently a state-wide joint data system for LAFOS has been set up. According to the evaluation, the most positive experiences of LAFOS were found at the operational level of service delivery. Workers recognized the limited nature of a one-sector approach in complex client situations and were eager to utilize the flexibility and wider resources of the network structure. Similar evidence comes from the USA, where it has been found that networks involving administrative-level actors are prone to conflict, whereas those coordinating service delivery among line staff experience greater collaboration (Bryson et al., 2006). The network approach works more effectively the lower one moves on the hierarchic levels of government. To a great extent network governance is about lower-level politics, with the aim of coordinating activities across sectoral and organizational boundaries without removing the boundaries themselves (Christensen and Laegreid, 2007). Network governance is a bottom-up policy and cannot easily be imposed from the top down (Pollitt, 2003). It challenges traditional, vertical and hierarchical power relations and calls for models of horizontal decision making.

Multi-professional and multi-sectoral service structure Each LAFOS unit is a separate local service unit with personnel, partnerships and working methods of its own. They are profiled as holistic and multi-professional service providers for the long-term unemployed. Professionals from the employment office, social welfare office, municipal health care unit and local Social Insurance Institution moved to work in LAFOS. They brought with them the traditions, regulations and administrative cultures of their parent offices. A LAFOS culture and way of working had to be created around a diverse expertise and the former sector-based administrative practices. In the beginning most of the LAFOS teams drifted into conflict. All four phases of team development were met during the pilot phase of the units (Tuchman and Jensen, 1977). At first there was enthusiasm and high hopes in relation to joint services within the new service structure (forming). While working together, differences in expertise, working cultures and values between sectors were recognized and conflicts started to arise in everyday work (storming). Little by little, the origins of conflicts were identified and discussed collectively (norming). After that the team was usually more realistic and ready for the developmental work of joint services (performing). Some teams progressed fairly smoothly from the

Tomáš Sirovátka

228

Governance of Integrated Activation Policy in Finland

very beginning, while some disbanded during the pilot phase and a new team was set up to initiate LAFOS. In LAFOS a working-pair method was developed, which meant that each client is met by both an employment official and a social worker. An individual activation plan is drawn up in this tripartite meeting. The driving force for joint services lies in bringing diverse expertise together, where the advantages are obtained by combining differences, not similarities. For the professionals the new service structure was demanding, however, because it made each individual’s expertise transparent to other professionals. Some social workers became disappointed at the expertise level of employment officials, and vice versa. Professionals may have unrealistic expectations concerning the competence of other professionals in cases that approach the limits of their own competence. LAFOS, as a network structure, opened up a set of resources with the challenge of forming reasonable service combinations in order to offer more individual support for a single client. The joint service threw up new challenges for professionals: they had to reflect on their expertise in relation to other expertise. The network structure as such turned out to be an effective learning community for them in questions related to activation.

Clients in integrated services There were 26,000 clients in LAFOS in 2007, which represented a minor portion of all the hard-to-employ citizens.7 Hence clients that had been referred to LAFOS had multiple problems in their employability: longterm unemployment, low occupational skills, little work experience, limited social skills, problems with their physical or mental health or abuse of alcohol or drugs. The need for multi-sector and individualized services in improving employability was apparent. One of the main objectives in developing integrated services was improving the employability of the clients (McQuaid and Lindsay, 2005). The broad concept of employability consists of three inter-related components that influence a person’s employability: (1) individual factors (2) personal circumstances and (3) external factors. Individual factors involve a person’s employability skills and attributes, such as personal competencies, basic social skills, qualifications, work experience, health and well-being. Personal circumstances include a range of socioeconomic factors related to an individual’s social and household circumstances, such as caring responsibilities and access to resources, such as transportation. External factors include demand factors, such

Tomáš Sirovátka

Vappu Karjalainen and Peppi Saikku 229

as the level of local labour demand and employers’ recruitment procedures, as well as enabling support factors, such as measures to match labour demand and supply. All three components should be considered when evaluating a person’s employability. Client insights on integrated services have been compiled since the pilot phase of the joint service units. According to the customer satisfaction surveys in 2002 and 2003, over 80 per cent of clients were satisfied with the services they had received in the unit. Clients viewed the best aspects of the services as being individuality, versatility, receiving different services under the same roof and the friendly and approachable nature of the staff (Lahti, 2004). Clients’ feedback on the quality of the integrated services has also been very positive during the LAFOS phase (Arnkil et al., 2007; Karjalainen and Saikku, 2008). Individual employability factors have been taken into consideration and handled with much greater scope and intensity in LAFOS than in PES or standard social services. Clients were offered personalized health services, combinations of services in individual situations, group activities for gaining basic social and everyday life skills, education and practical training. Due to the integrated social services, the client’s social situation (family relations, housing and so on) was made visible and was considered in relation to labour-market activities such as supported work or education. LAFOS clients experienced a better understanding of their own situation, possibilities and future prospects. The ultimate goal of LAFOS was to reduce structural unemployment by improving and supporting the employability of the long-term unemployed. Participation rates in activation measures of people on labour-market support (benefit) rose in Finland by nearly 10 per cent from 2005 to 2007, being 35 per cent in 2007. However, one problem was that activation measures did not open up new paths to the open labour market. The gap between the demands of local labour markets and the abilities of the clients remained. Some clients found themselves in a cycle of continuous activation measures that did not evolve or lead to any permanent solution. The labour market entry results following LAFOS activation have not been very high. Approximately 10 per cent of the unemployment periods end due to transfer to the open labour market and about 20 per cent enter into pay-subsidized work.8 As can be seen from Table 11.1, the situation has remained very stable from the beginning of LAFOS activity. The labour-market entry results and the insights of clients suggest that insufficient attention has been paid to the external factors of

Tomáš Sirovátka

230

Governance of Integrated Activation Policy in Finland

Table 11.1

Ended unemployment periods in LAFOS (2004–7)

Ended unemployment periods

2004

2005

2006

2007

To

%

n

%

n

%

n

%

n

Open labour market Pay-subsidized work Education Outside the labour market Other reason/no knowledge Unemployment pension

12 21 16 22 29

876 1571 1231 1699 2172

10 22 12 25 30

2139 4555 2575 5092 6278

11 21 11 26 31

3158 6252 3260 7653 9431

11 21 10 26 32

3950 7507 3695 9187 11621

0.2

16

0.1

27

0.1

39

0.1

41

Source: Statistics from the Ministry of Labour/the Ministry of Employment and the Economy.

employability. Because of the heterogeneous and demanding clientele, a lot of resources and effort have been put into improving their individual skills and attributes, as well as their personal circumstances. Although LAFOS workers are widely networking with various third-sector and service-providing organizations, private companies and employers have proved to be much harder to reach and involve. In the views of the clients, this, however, seems to be the crucial issue in obtaining sustainable employment results. From the client perspective, activation measures need to provide real progress and lead to a result in order to retain their legitimacy. The integrated approach of activation policy and services has produced contradictory results from the client perspective. On the one hand, the quality of the service has improved. Individual situations are better recognized and taken into account in planning and delivering services. Multi-sector services mean wider possibilities and better reconciliation of service processes. Clients feel their insights are valued and they are treated in a respectful manner by staff. On the other hand, the activation policy has tightened its grip on clients. New sanctions and obligations have been introduced. In integrated services, activation may go beyond the commitment to work-related activities. Clients are presumed to reveal their entire life situation and be willing to improve it. Individualization of the services means also an emphasis on individual responsibility. The legitimacy of activation is questioned by citizens if their increased efforts and commitment do not lead to satisfactory results. From the client perspective, the network approach has overcome many obstacles in complex, cross-sector service production. The clients’

Tomáš Sirovátka

Vappu Karjalainen and Peppi Saikku 231

voice is better heard than before. However, the involvement of the wider network, including private companies and third-sector organizations such as citizen organizations, is low. The external factors of employability should be identified more closely. Although the employability of the clients has improved, employment results have been modest. Service processes in LAFOS tend to last a long time, approximately two years. Thus the efficiency of LAFOS has been questioned, especially by the Ministry of Employment and the Economy. LAFOS seems to have more of an effect on the well-being of clients than on their employment (Valtakari et al., 2008). For the governance of an integrated activation policy, this will impose challenges in negotiating targets that also take into account the well-being of the citizens.

Mixture of governance models LAFOS is a result of integrated policy development in Finnish activation. Actors from public sectors have engaged in collaborative strategy planning and in multi-professional service delivery. A network approach is very different from sector-based silo government, meaning that new structures are needed for interaction and collaboration, in both administration/management and service delivery. The art of contracting, negotiating and dialogue are qualifications that are of value in integrated policy implementation. The network model is an administrative response to the complexity of the life situations of the long-term unemployed. In this sense the network approach represents progress, that is, the transfer to a more sophisticated model of service delivery and governance. Evaluation provided evidence on both the advantages and problems involved in network governance. These are listed in Table 11.2. The advantages are those of flexibility, better utilization of diverse resources and client-centred encounters in service delivery. The Table 11.2

Network governance model in activation policy

Advantages

Problems

Client-centred approach Flexible, low hierarchy Coordination of separate resources Challenge for expertise

Jurisdiction of management Unclear discretion Accountability, separate data systems Sector-based traditions

Tomáš Sirovátka

232

Governance of Integrated Activation Policy in Finland

problems are founded on principles of legitimacy, discretion and accountability in the context of multi-sectoral administration. Network governance is in a developmental phase in Finnish activation policy. LAFOS has worked as a platform for experiments in joint planning, management and service delivery. A future challenge concerns the clarification of network principles and maintaining the commitment of various parties to the demanding developmental process. Since no former models are available, practices have to be created within each network context and by the stakeholders involved. A network approach deals strongly with interaction and communication, which places an emphasis on cultural aspects. Our evaluation results revealed that changes in cultural conventions are slow, and a decade is a short period in this respect. Although network governance is currently profiled in Finnish activation policy, there are, of course, elements of procedural, corporate and market governance (see Chapter 1). Procedural governance based on sector-based legislation and universal regulation forms an essential background for the network approach, especially in relation to benefit administration. From the perspective of clients, procedural governance secures a certain level of equality and rights – even obligations – within social security and services. However, in the context of joint services it represents a stagnant approach. Procedural governance is slow and limited when encountering multifaceted phenomena and trying to identify administrative solutions in cross-sector situations, whether these concern a citizen or service delivery structures. The features of corporate governance in activation imply elements of performance management (NPM) that have spread throughout the Finnish public sector during the last decade. Public and private organizations have focused on results and on meeting performance targets. Usually, performance targets in activation policy are set up separately within the employment administration, the social welfare administration and so on, and the follow-up of reaching the targets is carried out via sector-based data systems. In a joint service structure, such as LAFOS, a major challenge lies in setting up cross-cutting targets and giving them equal status with the sector targets. Market governance with competition, contracts and cost control is also featured in the implementation of Finnish activation policy. The public sector has introduced outsourcing and the marketization of certain services on a permanent basis in the provision of individual activation. Questions concerning the producer–purchaser relationship, subcontracting regulations, quality control and performance targets for

Tomáš Sirovátka

Vappu Karjalainen and Peppi Saikku 233

outsourced services have to be solved. In the context of network governance, outsourcing of services is one basic component of service delivery. Tendering of services, quality standards for service products and the case management of outsourced services are crucial issues in the service provision of LAFOS units. Market governance in Finnish activation policy is at an emergent stage compared with the respective development in, for example, the Netherlands (see Chapter 7). In all likelihood, marketization will characterize more strongly the next developmental phase of activation governance in Finland.

Conclusions The trend in Finnish activation policy over the last decade has been the strengthening of activation and the tightening of obligations imposed on the long-term unemployed. A shared responsibility between the central actors (state/employment government, social insurance institution and municipalities/social welfare and health government) has gradually developed, being nowadays a necessary prerequisite in the implementation of activation policy. Several reforms have facilitated the development of an integrated approach and the most concrete result of the integration process is LAFOS units that are state-wide, covering the urban regions with specialized and holistic services for the long-term unemployed. A political consensus has been reached in favour of this kind of policy making. In the long term, the question is whether a joint service and whole-of-government approach will be mere political attainments that will fade away to be replaced by a new reform, or whether this kind of integration reflects a deeper change in the welfare society. One might ask whether we have reached the point where individualization is embedding itself in the structures of the welfare services on a permanent basis and transforming them to be more compatible with citizens’ life situations. According to our results at least, it remains too early to talk about a fundamental and sustainable change in this respect. The integration process has revealed that new problems are arising alongside network governance in the environment of the sector-based public service system. Models for network governance have to be created in each network context by means of contracts, negotiations, dialogue and trust. The bottom-up approach of the network approach gives rise to many new questions concerning legislation and regulation, discretion in the network context and territorial equity in service delivery. The traditional welfare system faces a shift in the service paradigm with

Tomáš Sirovátka

234

Governance of Integrated Activation Policy in Finland

controversial aspects. On the one hand, the network approach represents (post-)modern service development in the sense of taking account of the multifaceted conditions in which people live and by providing a client-centred, holistic service. On the other hand, network governance questions the competency of decision making and management of the public sector within a multi-actor environment and challenges public actors to engage in new reflections. LAFOS – and the preceding integration development – have been an unusual intervention in the public sector administration in Finland. The efforts of integrated approaches have aimed at producing a deeper coordination of administration and service delivery than can normally be achieved in public services. Arrangements of network governance are increasingly seen as a suitable response to the question of how to tackle multifaceted, uncertain and conflict-prone policy problems (Sorensen and Torfing, 2007). The advantages of LAFOS are experienced by the long-term unemployed themselves. They are encountered more individually and appreciated as clients of welfare services. However, the hypothesis as to whether proper and coordinated service delivery would provide better results in activation does not generate unambiguous evidence. Network structures are flexible in connecting actors from different environments and bridging the potential gap between the long-term unemployed and working life. The coordination of resources provides an opportunity for joint utilization of the scant or diminishing resources of one actor, but the network structure gives rise to difficulties in administrative arrangements. The self-regulating nature of network governance is problematic in relation to public government: Who are the actual decision makers in a network context? How are the shared issues regulated in and through negotiations? A kind of regulation of selfregulation of networks is needed. Soerensen and Torfing (2007) refer to the development of meta-governance in network management with the purpose of making agreements regarding the governance issues and processes in networks. Guiding principles are necessary in the management of network organizations. The future challenge of network governance could be in attempts to produce relevant technologies and tools of meta-governance.

Notes 1. There are 39 LAFOS units in Finland. In 2007 these covered 52 out of a total of 114 PES offices, and 161 municipalities out of 416. The staff resource in

Tomáš Sirovátka

Vappu Karjalainen and Peppi Saikku 235

2. 3.

4. 5.

6. 7. 8.

LAFOS was 650 man-years in 2007: 320 contributed by the state (PES) and 330 by the municipalities (social work and health care). The contribution of the National Insurance Institution consisted mainly of consultancy. In 2007 there were 26,000 clients in LAFOS service, in 2009 about 25,000. Last-resort benefits for the long-term unemployed are labour-market support (PES) and social assistance (social welfare, municipality). The concept of an intermediate labour market refers to a bridge between longterm unemployment and the open labour market, consisting of a variety of measures and facilitators (work practice, supported work, rehabilitative work and so on). From 1 January 2008 the Ministry of Employment and the Economy. Other main objectives of the Programme were ensuring the availability of a skilled labour force, encouraging workers to stay in the labour market and increasing work productivity. Other objectives were raising the activation rate and improving capacity and work ability of the long-term unemployed. In 2007, an average of 125,000 jobseekers were classified as hard to employ and, of these, an average of 52,000 were long-term unemployed. In PES offices in 2007 overall, 44 per cent of employment periods ended in open labour market entry and around 11 per cent in pay-subsidized work (Ministry of Employment and the Economy, 2008, p. 2).

References Agranoff, R. (2006) ‘Inside collaborative networks: Ten lessons for public management’, Public Administration Review, 66/s1, 56–65. Andreassen, T.A., I. Drange, T. Thune and L.C. Monkerud (2007) På vei mot en integrert velferdsforvaltning? Erfaringer fra pilotprosessen iden nye arbeids- og velferdsforvaltningen, AFI-rapport 4/2007 (Oslo: Arbeidsforskningsinstituttet AS). Arnkil, R., V. Karjalainen, S. Aho, T. Lahti, S-M. Lyytinen and T. Spangar (2004) Yhteispalvelusta palvelukeskuskonseptin kehittämiseen. Yhteispalvelukokeilun arvioinnin loppuraportti. Työhallinnon julkaisu 339 (Helsinki: Työministeriö). Arnkil, R., V. Karjalainen, P. Saikku, T. Spangar and S. Pitkänen (2007) Työvoimatoimistot ja työvoiman palvelukeskukset tänään ja huomenna. Väliraportti Suomen työvoimapalvelujen uudistuksesta. Työhallinnon julkaisu 373 (Helsinki: Työministeriö). Askim, J., A.L. Fimreite, A. Moseley and L. Holm Pedersen (2010) ‘One stop shops: An emerging instrument for joining up the 21st century welfare state’, in T. Christensen and A.L. Fimreite (eds) NAV in an International Context (Bergen: Stein Rokkan centre for social studies). Bryson, J.M., B.C. Crosby and M. Middleton Stone (2006) ‘The design and implementation of cross-sector collaborations: Propositions from literature’, Public Administration Review, 66/s1, 44–55. Castells, M. and G. Cardoso (eds) (2006) The Network Society: From Knowledge to Policy (Washington, DC: John Hopkins for Transatlantic Relations). Christensen, T. and P. Laegreid (2007) ‘The whole-of-government approach to public sector reform’, Public Administration Review, 67/6, 1059–66.

Tomáš Sirovátka

236

Governance of Integrated Activation Policy in Finland

Considine, M. and J.M. Lewis (2003) ‘Bureaucracy, network, or enterprise? Comparing models of governance in Australia, Britain, the Netherlands, and New Zealand’, Public Administration Review, 63/2, 131–40. Karjalainen V. and P. Saikku (2008) ‘Työvoiman palvelukeskukset: erityispalvelua vaikeasti työllistyville’, in R. Arnkil, V. Karjalainen, P. Saikku, T. Spangar and S. Pitkänen (eds) Kohti työelämälähtöisiä integroivia palveluja. Työvoimatoimistojen ja työvoiman palvelukeskusten arviointitutkimus (Helsinki: Työ- ja elinkeinoministeriö). Keskitalo, E. (2008) Balancing Social Citizenship and New Paternalism. Finnish Activation Policy and Street-Level Practice in a Comparative Perspective (Jyväskylä: National Research and Development Centre for Welfare and Health). Lahti, T. (2004) ‘Asiakastyytyväisyys ja palvelukokemukset’, in R. Arnkil, V. Karjalainen, S. Aho, T. Lahti, S-M. Lyytinen and T. Spangar (eds) Yhteispalvelusta palvelukeskuskonseptin kehittämiseen. Yhteispalvelukokeilun arvioinnin loppuraportti (Helsinki: Työministeriö). Ling, T. (2002) ‘Delivering joined-up government in the UK: Dimensions, issues and problems’, Public Administration, 80/4, 615–42. Mannila, S., M. Ala-Kauhaluoma and S. Valjakka (eds) (2001) Good Practice in Finding Good Practice: International Workshop in Evaluation (Helsinki: Rehabilitation Foundation). McQuaid, R. and C. Lindsay (2005) ‘The concept of employability’, Urban Studies, 42/2, 197–219. Ministry of Employment and the Economy (2008) Employment Report 2007, http://www.tem.fi/files/19650/TEM17_2008_aktiivinen_2.pdf (accessed 19 December 2008). Pollitt, C. (2003) ‘Joined-up government: A survey’, Political Studies Review, 1/1, 34–49. Senge, P. (1990) The Fifth Discipline: The Art and Practice of the Learning Organizations (New York: Doubleday). Soerensen, E. and J. Torfing (2007) Theories of Democratic Network Governance (London: Palgrave Macmillan). Tuchman, B.W. and M.A. Jensen (1977) ‘Stages of small-group development revisited’, Group & Organizational Studies, 2/4, 419–27. Valtakari, M., H. Syrjä and P. Kiuru (2008) Julkisen työvoimapalvelun palvelurakenteen uudistamisen vaikuttavuus (Helsinki: Ministry of Employment and the Economy).

Tomáš Sirovátka

12 The Governance of Active Welfare States in Europe in a Comparative Perspective Rik van Berkel, Willibrord de Graaf and Tomáš Sirovátka

Introduction Throughout this book, two research questions guided our analysis of reforms of governance structures for the administration of income protection systems and the provision of activation or employment services for unemployed people. First of all, we wanted to know what governance reforms European countries have introduced or are introducing. As was elaborated upon in the first chapter, we were specifically interested in the following governance reform characteristics or principles: processes of decentralization and (re-)centralization; the introduction of markets and the principle of competition in service provision processes; inter-agency cooperation and partnerships; and the introduction of governance instruments derived from the New Public Management in the steering of public agencies involved in social policy implementation. Interpreting governance reforms at a more abstract level, we also raised the question whether these reforms are producing a shift of governance regimes in the social policy areas on which this book focused, using the typology of governance regimes developed in Mark Considine’s (2001) study of the public management of welfare-to-work programmes as a starting point. In terms of the traditional divisions of work in the social-scientific (sub-)disciplines, these questions belong to the domains of public administration and public management studies. However, we have argued that they are of relevance from a social policy studies perspective as well. Not only are governance reforms not neutral in terms of the influence they exercise on substantial aspects of social policy programmes and social services, these reforms also often explicitly intend to have an impact on programmes and services. This is where the second 237

Tomáš Sirovátka

238

The Governance of Active Welfare States in Europe

question that this book seeks to answer entered: what do we know of the intended and unintended effects of governance reforms? The preceding chapters have answered these questions for nine European countries representing a variety of welfare state regimes: social-democratic welfare states (Finland and Sweden), conservative welfare states (Germany, France, Switzerland), a liberal welfare state (UK), a hybrid welfare state (Netherlands), a Southern European welfare state (Italy) and a post-communist welfare state (Czech Republic). In this concluding chapter, we will sum up the findings presented in the preceding chapters by presenting answers to our leading questions from a comparative perspective. But before doing that, some general remarks concerning the governance reform processes we studied can be made. All countries in our study have introduced governance reforms as part of broader reform programmes aimed at making welfare states more activating, which include both governance reforms and reforms of social policy programmes. Although national contexts are relevant in determining the nature and timing of reforms, some overarching objectives are pursued in all or most countries: promoting the coordination between various income protection systems as well as between income protection systems and employment/activation services; and strengthening the capacities for an efficient and effective implementation of activation programmes and provision of activation services. Nevertheless, considerable differences between the countries exist as well, and not only regarding the characteristics of the reforms introduced to realize the objectives just mentioned. Some countries started governance reforms considerably earlier than others: in this respect, the UK can be considered an early reformer, starting reforms in the 1980s, whereas Germany introduced its main reforms in the 2000s. Furthermore, some countries have been quite radical in their reforms, whereas others seem to have taken a more modest reform course. The Swedish reforms, for example, are quite moderate and incremental compared with the radical introduction of marketization in the Netherlands or the introduction of New Public Management instruments in Switzerland. In addition, countries are different in the degree of central orchestration of governance reform processes. Whereas the role of central government is traditionally strong in the UK, the case of the Czech Republic is a clear example of reforms taking place without strong coordination by the national state, at least as far as governance reforms are concerned (national reforms of the content of activation did take place). This is why the Czech governance reforms were qualified as spontaneous reforms. To a certain extent something similar applies to Italy, but here significant welfare state

Tomáš Sirovátka

Rik van Berkel et al.

239

functions (social assistance, employment services) have been decentralized. This reveals that differences in reform processes may also point to different capacities of national states to introduce and implement these reforms, as well as at different levels of commitment to deal with reform urgencies as national urgencies. The structure of this chapter will follow the two research questions mentioned at the start of this introduction. First, the various governance reforms will be compared, followed by a section that deals with the question of whether these reforms can be interpreted as shifts of governance regimes. In the third section, we will look at the effects of the governance reforms. Based on this analysis, the final section will address the issue of the ways in which governance reforms and substantial social policy characteristics are related.

Governance reforms In this section, a comparative analysis of governance reforms will be presented, structured according to the reform characteristics or principles that guided our analyses throughout this book. Decentralization Of the various types of governance reform processes taking place, decentralization is probably the most complex to analyse and compare. Even when we limit our attention to activation policies, programmes and services – which, as we showed in Chapter 1, are often considered as requiring decentralization in order to increase their effectiveness – it is quite difficult to unravel what actors and agencies have authority in taking what types of decisions concerning activation. More detailed and systematic research than we have been able to do is required to come up with reliable insights into these questions. Because of that, the focus in this chapter will be different: we will analyse some of the complexities involved in comparing decentralization reforms. First of all, the provision of activation services involves a variety of policy choices and decisions. For example, it involves choices regarding the nature of activation programmes and services; regarding the definition of target groups of programmes and services; regarding the procedures to determine which unemployed persons belong to what target groups and should be referred to what type of programme or service; regarding the organization of service provision processes; and regarding the organization of primary processes at the frontlines of the agencies involved in service provision. Each of these types of decision may

Tomáš Sirovátka

240

The Governance of Active Welfare States in Europe

be more or less decentralized, as can be illustrated using the German case as an example: in Germany, there is a national profiling system to categorize the unemployed in various groups reflecting various levels of employability, purchasing services from external providers takes place at regional levels, and local actors have gained more decisionmaking room in spending budgets for activation services. Comparing the countries, activation in the UK is most centralized. The most obvious manifestation of this is the fact that the activation of recipients of the UK safety net provision is the responsibility of Jobcentre Plus, rather than the municipalities. This means that the activation of recipients of both the contribution-based and means-tested Jobseeker’s Allowance is subject to the same national and central regulation. In the UK, most types of policy decisions in the area of activation are subject to central regulation: for example, the content of activation (although more flexibility has been introduced recently), the involvement of external providers, target groups of activation programmes as well as the organization of the work of frontline workers. At the other extreme there is the Dutch case of Social Assistance. The activation of Dutch social assistance recipients is highly decentralized, although work and activation obligations of the unemployed are regulated nationally: local actors have decision-making authority regarding the content of activation, the organization of service provision, as well as the procedures for profiling social assistance recipients. Another complicating factor in this context concerns the fact that different territorial agents may be responsible for various services relevant in the context of activation. In France, for example, the regions are responsible for Vocational Training services (as is the case in Italy), the departments provide job-search support for social assistance recipients and municipalities are primary actors in programmes aimed at promoting social inclusion. This points to a second complexity in analysing decentralization: the fact that in most countries (in our study, the UK and the Czech Republic are the exceptions) the activation of the insured on the one hand and social assistance recipients on the other is governed in different ways, involving different actors and agencies when it comes to primary responsibilities for activating both groups of unemployed, and involving different forms of decentralization. In many countries we see processes of political decentralization (see Chapter 1) in the area of activating social assistance recipients, involving not only municipalities but sometimes also regions (in federal states (Germany, Switzerland) as well as in countries such as France and Italy) in decision making

Tomáš Sirovátka

Rik van Berkel et al.

241

concerning activation. Administrative decentralization is more common in governing the activation of recipients of unemployment insurance schemes, as the activation of this group of unemployed is often the responsibility of national agencies and their regional or local offices: the Public Employment Services (PES) agency (for example, Sweden and Finland) or the joint benefit and PES agencies (France, Germany, the Netherlands). In this context, Italy is an interesting case, as PES functions have been subject to political decentralization, delegating policymaking authorities to regions and provinces. Overall, we can conclude that the activation of social assistance recipients is more decentralized than the activation of recipients of unemployment insurance. The role of municipalities in administering (and often also funding) Social Assistance schemes plays an obvious role in explaining this. Another explanation might be found in the different characteristics of the two populations (social assistance and unemployment insurance recipients): the heterogeneity of the social assistance population, together with the multiple social and individual problems characterizing the situations of many social assistance recipients, is seen as creating a greater urgency for tailor-made and individualized (and, thus, decentralized) services than is the case for the insured unemployed. Thirdly, in assessing degrees of (de-)centralization, not only the absence or presence of national and central regulation is relevant, but also the nature of regulation. The UK provides a clear (though certainly not unique) example of this. The UK used to have a range of New Deal programmes, each programme being targeted at a specific target group (for example, young unemployed people or single parents). More recently, the programmes were integrated into a more flexible programme, providing more decision-making room concerning the nature of activation in individual cases to agencies involved in activating the unemployed. In other words, more ‘open’ national programmes can be interpreted as a form of decentralizing concrete decisions about activation to lower-level agents. Fourthly, room for decentralized policy making is not merely a function of the explicit delegation of power from national or central to lower-level actors. The Czech case shows some interesting examples of this: decentralization can be the consequence of insufficient capacities (or legitimacy) of central actors to enforce regulation. Thus, decentralization may be unintended or, as it was called in the chapter on the Czech case, spontaneous rather than intended. In addition, the Czech case made clear that the availability of funding from sources other than the central government – in this case, funds from the European Social

Tomáš Sirovátka

242

The Governance of Active Welfare States in Europe

Fund – may strengthen decentralization processes as well. Something similar seems to have laid the foundation for the introduction of municipal activation programmes for social assistance recipients in countries such as Sweden and Germany, where dissatisfaction with the activation services provided by the PES on the one hand and the availability of local financial resources on the other created the urgency and the feasibility of local initiatives in this area. The Czech case shows that there may be more decentralized policymaking room than officially granted to regional or local actors. But the reverse may be the case as well – and this is the fifth issue making analysis of decentralization complex: officially granted room for policy making may be considerably reduced by other measures or regulations that central governments or other actors introduce, or fail to introduce. A well-known and notorious example in this context is Italy, where the delegation of policy-making authority for social assistance and activation services to regional, provincial and local actors is not accompanied by any redistribution of financial resources. This implies that poor regions with low financial resources and high levels of need have little room for manoeuvre, even though in a formal sense their policy-making room is considerable. But in other countries as well, formal decision-making room may in practice be reduced as a consequence of the context in which decision making takes place. As we saw in the preceding chapters, New Public Management instruments may seriously affect the ways in which regional or local actors use their policy-making authority. Funding regimes, contracts containing specific performance indicators, monitoring and benchmarking mechanisms: these and other instruments that steer outcomes rather than outputs or processes of service provision make certain policy choices more likely than others, even though, formally, regional or local actors are not obliged to make these choices. Because of that, we qualified these processes as implicit centralization. Finally, processes of decentralization and centralization are not aims in themselves but embedded in strategies aimed at realizing complex sets of objectives that do not always unambiguously point in the direction of either decentralization or centralization. That is one of the reasons why processes of decentralization and centralization often occur simultaneously or subsequently. The Swedish case provided an example of this: in the 1990s, municipalities acquired the authority to start developing programmes aimed at the young unemployed. Decentralized programmes were seen as creating room for tailor-made services that were expected to be more effective than standardized

Tomáš Sirovátka

Rik van Berkel et al.

243

programmes. Recently, however, the responsibility of activating the young unemployed in Sweden was transferred to the PES, partly using the same argument that was used before to support a decentralized approach, namely to improve the effectiveness of services. Another argument used in favour of recentralization was to promote equality of treatment and reduce diversity – issues in the context of decentralized policy making that indeed deserve serious attention. Marketization and competition When we compare the ways in which the countries in our study organize the provision of income support and activation for the unemployed, it is clear that the reforms that were discussed in the preceding chapters not only point in the direction of multi-level governance, but also of multi-actor governance. One of the ways in which the shift towards multi-actor governance becomes manifest is the increasing role of quasi-markets and competition in the provision of activation services. At the same time, we should be careful to equate the rise of service markets with the introduction of multi-actor governance. Some countries, notably Germany and Switzerland, traditionally involved other actors in the provision of social services, as is, for example, illustrated by the role of the Church in providing services to the poor in Switzerland. Thus, marketization may have the effect of multiplying the range of actors involved in service provision, but it may also imply that actors traditionally involved in service provision now find themselves in different relationships with each other (competitors) or with the main agency (agents versus principal). In addition, marketization may have consequences for the traditional main agency (a benefit, welfare or PES agency), which itself may be subjected to processes of competition, as in the case of the UK and Germany. Probably the most striking conclusion that we can draw from our national analyses of reform processes is that marketization has been introduced in all countries. Finland and Sweden have been least eager to introduce this governance reform, but recently the Swedish government obliged the PES to purchase a specified proportion of services on the market, thus strengthening the role of marketization in the provision of activation. As discussed in Chapter 1, the common trend towards marketized service provision does not diminish the fact that service markets differ in several respects. Some of the main dimensions of variation can be outlined as follows.

Tomáš Sirovátka

244

The Governance of Active Welfare States in Europe

First of all, the proportion of activation services provided by market actors differs between countries. Although we do not have data that could provide more detailed information on this issue, the role of market actors seems to be most modest in Finland, whereas in the Netherlands their role is strong, especially because, with the introduction of marketization in the Netherlands, both the local welfare agencies and the benefit agency were obliged to outsource most activation services. In more recent years, the Netherlands witnessed a process of demarketization of the activation of social assistance recipients: the abolishment of the obligation to outsource services for local welfare agencies reduced the proportion of the budget available for activation services that is spent on the market. Maybe surprisingly, the UK has seen a rather incremental growth of the role of the market in service provision, even though it was one of the first countries to introduce this governance reform. More recently, the Freud report recommended a stronger role for the market in the provision of services. Secondly, the nature of providers differs between the countries. In the Netherlands, private for-profit companies play an important role, while in Germany non-profit organizations are important players. In the UK, a mix of for-profit and voluntary organizations can be found. In several countries, temp agencies play a role in providing job placement services, for example in Italy, France and Germany. In Italy, temp agencies also have a role in the provision of education and vocational training. Outsourcing training programmes is also common practice in the Czech Republic and Sweden, where the first steps towards marketization in the 1990s involved mainly training programmes. In most countries, the purchasers – the benefit agency, the PES, local welfare agencies – also act as providers. Thus, the split between the roles of purchaser and provider is not complete. Marketization involves the purchasers in their role as providers in competition processes in various degrees. For example, in Germany the purchasers do in certain cases compete with market actors in the provision of services (see below). In the UK, Jobcentre Plus finds itself competing with market actors as well. In the Netherlands, local welfare agencies act as both purchaser and provider, but they themselves decide what services are being outsourced, so that they do not compete with market actors. Competition is also absent in cases where purchasers are obliged to outsource (a specific proportion of, or all) services, as in Sweden or in the case of the Dutch benefit agency. In several countries – but not all, as the UK case illustrates – service provision for unemployment benefit recipients is more marketized than service provision for social assistance recipients. In France, for example,

Tomáš Sirovátka

Rik van Berkel et al.

245

municipalities seem to prefer negotiating and cooperating with external providers rather than subjecting them to competition. Something similar applies to Sweden, where the role of marketized service provision in municipal activation programmes used to be modest. Interestingly, in the context of the recent attempts to strengthen the role of market actors in Swedish activation, it was mentioned that the involvement of private actors is considered of particular importance in activating the young unemployed and the hard to employ – which used to be the target groups of municipal activation programmes. Why marketization is more common in the case of unemployment benefit recipients is unclear. It may be that municipalities believe that services for social assistance recipients are difficult to marketize, given the sometimes complex and multiple needs of this group of unemployed. But more mundane motives may play a role as well: for example, avoiding a reduction in the numbers of municipal civil servants. Fourthly, market competition seems to be fiercer in some countries than in others. Regarding the Czech Republic it was noted that, despite the increasing importance of marketized service provision, which is partly a consequence of activation programmes funded by the European Social Fund (ESF), competition remains weak as the market is dominated by a small number of providers. Something similar seems to be the case in the UK, where some prime contractors manage to dominate the market and squeeze out competitors. In the Dutch case, on the contrary, competition among providers is strong which, especially in the first years of marketization, made it hard for private providers to survive. A fifth dimension of diversity concerns the issue of decentralization of purchasing processes. Centralized purchasing, at least for part of the services, can be witnessed in several countries. Examples are the UK, where the Department for Work and Pensions sometimes acts as purchaser of services; Sweden, where purchasing by the PES used to take place at country level but now takes place at national level; Germany, where regions acquired a stronger role in purchasing services. In Switzerland, the establishment of Regional Employment Centres and Logistics Offices implied a centralization process as well, as they replaced local employment offices – though marketization was introduced only after this regionalization process took place. Purchasing is decentralized in the Czech Republic, where local (or regional) Employment Offices purchase services for unemployment benefit and social assistance recipients; and in the Netherlands, where local welfare agencies act as purchasers of activation services for social assistance recipients. Countries that decided to centralize purchasing did so for various reasons. In the UK, for

Tomáš Sirovátka

246

The Governance of Active Welfare States in Europe

example, centralization was an instrument to subject Jobcentre Plus to competition processes: in these cases, Jobcentre Plus was no longer purchaser, but one of the potential providers. In Sweden, centralization aimed at reducing diversity and at strengthening central control over activation policies. In Switzerland, the decision to organize purchasing processes at the regional level was intended to professionalize outsourcing of services. The issues of expertise and capacities needed in outsourcing activation services also play a role in the Netherlands, where a regionalization of purchasing processes took place in various areas: several so-called inter-municipal welfare agencies (voluntary cooperation among groups of local welfare agencies) were established to cope with the increasing number of local tasks and responsibilities, including purchasing services. A final issue we want to mention concerns the role of the unemployed themselves in the market of activation services. In most countries, the unemployed have no role whatsoever in purchasing services: benefit agencies, welfare agencies or PES agencies decide what services are outsourced and what providers are contracted. Two countries provide exceptions to this rule, however. In Germany, a voucher system was introduced that empowered the unemployed in the sense of providing them with purchasing power on the service market. Interestingly, this voucher system also strengthened the competition between public and private providers. In the Netherlands, unemployment benefit recipients are allowed to develop their own reintegration plan and select a provider of their own choice. Even though they need the approval of the benefit agency, this does give the unemployed some voice and choice in activation processes. Inter-agency cooperation and partnerships Welfare state reforms aimed at promoting activation and labour-market participation have created an urgency for promoting various forms of coordination: coordinating activities of agencies responsible for income provision (benefit/welfare agencies) and PES; coordinating activities of benefit agencies and welfare agencies; and coordinating activities of providers of activation services and providers of other social services that are of relevance in promoting the labour-market participation of the unemployed. All countries have introduced reforms to tackle one or more of these coordination issues. Noteworthy are reforms aimed at coordinating the activities of benefit agencies and the PES. As is the case with all coordination challenges, existing structures for administrating income protection systems and

Tomáš Sirovátka

Rik van Berkel et al.

247

providing activation services play an important role in determining what coordination problems arise in the context of implementing the activation paradigm. For example, in Germany and the Czech Republic the PES are responsible for activation as well as for administering unemployment benefits. Something similar applies to the UK, where Jobcentre Plus implements both functions. Coordinating the administration of unemployment benefits and activation services will be more urgent in countries where these functions are organized separately. To promote coordination, France and the Netherlands recently decided to merge the benefit agency and PES. In both countries, these mergers were preceded with less far-reaching forms of coordination: experimental Employment Houses in France and co-located services of the local offices of the benefit agency and the PES in the Netherlands. In Switzerland, coordination of benefit administration and activation takes place in the cantonal Regional Employment Centres, which are responsible for benefit administration as well as activation. In the other countries, coordination initiatives to date have been less far-reaching. In Sweden, for example, no one-stop shop agencies exist: coordination is dependent on contacts between the staff of the PES agencies and the insurance funds. In Italy, coordination issues are mainly situated at regional and local levels. Far more difficult is the issue of coordinating the activities of the (often nationally managed) benefit and PES agencies on the one hand, and the municipalities and their local welfare agencies (and other local social services agencies) on the other. Generally speaking, two types of coordination issues are important in this context: coordination of the activities of local welfare agencies and the PES in the context of activating social assistance recipients; and the coordination of the activities of unemployment benefits and local welfare agencies. The reason for promoting the latter type of coordination is to prevent unnecessary transfers of unemployed people from one income protection system to the other, which is a frequent practice in several countries in order to reduce the costs of one agency/system at the expense of the other. Both national coordination problems in this area and national solutions are very diverse. As a consequence of the features of the UK social assistance scheme, the UK is a special case, as municipalities do not play a role in the administration of this scheme. In the other countries, municipalities (or regions) do play a role in the administration of social assistance, but not always in the activation of social assistance recipients. In the Czech Republic, for example, the functions of social

Tomáš Sirovátka

248

The Governance of Active Welfare States in Europe

assistance administration and activating social assistance recipients are divided between the municipalities and local employment offices. Coordination between the municipalities and the employment offices is no major issue in the Czech Republic, although one could argue that the conditionality of social assistance entitlements makes exchange of information between local welfare agencies and local employment offices necessary. Sweden is an interesting case in this context. As we saw before, the responsibility for the activation of important groups of social assistance recipients was transferred from the municipalities to the PES. Thus, whereas social assistance administration and activating social assistance recipients used to be integrated, now these functions have been separated, at least for some groups of the unemployed. One might expect that this reform is likely to increase, rather than decrease coordination problems, especially since Swedish Social Assistance is funded by local taxes. Swedish municipalities are in favour of improving cooperation with national agencies, but the national agencies are not keen to take initiatives in this area. Coordination is also relatively low profile between French municipalities and the local agencies of the merged benefit and PES agency, although the departments (responsible for organizing activation services for social assistance recipients) may decide to enter into partnerships with the local agencies of the merged institution. Finland, Germany, Switzerland and the Netherlands have taken more far-reaching, though again quite diverse, initiatives to promote coordination. Germany introduced various coordination models. In the dominant ARGE model, a new agency jointly managed by the municipalities and the PES should provide integrated services for the long-term unemployed (UB-II recipients). In the second model, the municipal model, municipalities are fully responsible for providing income support and activation to the long-term unemployed. In the third model, in which responsibilities remain divided, coordination is barely realized, as the municipalities and the PES provide their services separately. This variety of models makes the German provision of services for the long-term unemployed fragmented, despite the original intentions of the Hartz reforms. In the Netherlands, coordination involves the local welfare agencies and the offices of the merged benefit and PES agency. Co-location at the local level of front offices of the agencies in so-called Work Squares is the main initiative taken to improve coordination. In terms of the partners involved, these agencies resemble the German ARGE agencies, but the tasks of the partners as well

Tomáš Sirovátka

Rik van Berkel et al.

249

as the target groups of the agencies differ. The level of integration and coordination within the Dutch Work Squares differs significantly between municipalities. Coordination initiatives in Switzerland are fairly recent, and very heterogeneous, as it is up to the cantons and municipalities to decide about forms of coordination. Even though a national coordination group exists that monitors the coordination initiatives at cantonal levels, these initiatives are more bottom-up compared with, for example, Germany and the Netherlands. One of the reasons for this is that Switzerland does not have a national/federal Social Assistance system, making the introduction of a national model for cantonal coordination impossible. Like Germany, Finland introduced special agencies responsible for service provision for the longterm unemployed as well (LAFOS). However, the Finnish answer to coordination issues did not take the form of a national model that is centrally regulated and rolled out over all Finnish municipalities. Instead, a network model of coordination was introduced in which national and local agencies voluntarily decided to join forces in providing activation services for the long-term unemployed. This approach also carried the implication that local LAFOS units are not present in all Finnish municipalities – which, as mentioned before, raised issues about equality and uniformity. The aims of the Finnish coordination initiatives were to bridge the gaps between employment and social policies, to avoid a silo approach in service provision and to provide a more integrated service package responding to the needs of the long-term unemployed. The Finnish voluntary network initiative is rather singular compared with the coordination solutions we find in other countries. Although not discussed extensively in this book, we may expect network approaches to be used in other countries as well, especially at local or regional levels aimed at providing more integrated services for the long-term unemployed. We have already mentioned the French case where municipalities seem to prefer the network and partnership approach rather than marketization in dealing with coordination issues in service provision; another example is Italy. Although the fragmented nature of the Italian system makes it impossible to provide a systematic overview of governance structures in Italy, it is likely that a variety of network and partnership constructions exists to coordinate local services and, among others, functions of the PES (which are divided among national, regional and provincial levels). Nevertheless, the scale of the Finnish network initiative, as well as the fact that this type of coordination is used to solve the

Tomáš Sirovátka

250

The Governance of Active Welfare States in Europe

issue of coordinating activities of local social services agencies, the national PES and the social insurance agency, make the Finnish case exceptional. New Public Management In one way or another, all the core public institutions involved in providing income protection and activation services to the unemployed (local welfare agencies, benefit agencies and PES agencies) have been confronted with New Public Management instruments. New Public Management affected relationships between the central state and its relevant Ministries on the one hand and the agencies on the other, but is often also introduced to manage the agencies internally. As we discussed in Chapter 1, crucial issues in introducing the New Public Management style of steering include the use of performance indicators, the selection of these indicators, as well as the nature of the financial incentives and disincentives that accompany performance steering. When we compare the countries involved in our study, some have introduced New Public Management more vigorously than others. On the one hand, we find countries such as Finland, France and the Czech Republic where a rather soft version of New Public Management has been introduced. For example, forms of contractualism were introduced to manage the French PES, but these contracts were formal rather than seriously affecting the functioning of the agency. In the Czech case, output indicators (such as the number of Individual Action Plans and the number of long-term unemployed participating in activation measures) defined to steer the PES were considered to be weak: first, because they had already been realized in various regions of the country; and secondly, because they lacked ambition in the sense that they did not accurately reflect the seriousness of the problem of long-term unemployment. The Finnish case revealed that initiatives aimed at coordinating the activities of public agencies should not leave unaffected the performance indicators to which these agencies are subjected. The fact that each of the agencies involved in the LAFOS initiative had to realize its own performance objectives makes coordination and the setting of joint objectives difficult. In other countries, similar problems arose when coordination initiatives did leave an agency-based management by objectives system in place: for example, the functioning of the German ARGE is strongly influenced by the performance indicators set for the Federal Employment Agency.

Tomáš Sirovátka

Rik van Berkel et al.

251

On the other hand, Switzerland and the Netherlands are examples of countries where New Public Management has been embraced quite enthusiastically. Both countries introduced output indicators in the late 1990s, specifying the numbers of unemployed that should participate in activation and employment programmes. In Switzerland, these indicators were used in managing the Regional Employment Centres and Logistics Offices; in the Netherlands, these indicators were defined in contracts between the Ministry and the municipalities, thus affecting the activation of social assistance recipients. Both countries also introduced systems of financial incentives related to the realization of the specified output. In Switzerland, not realizing the objectives would have resulted in a larger cantonal contribution to unemployment benefit payments; in the Netherlands, the realization of objectives affected the budget made available to municipalities for activation services. Since then, both countries have refined their New Public Management instruments. Switzerland shifted from output indicators towards outcome indicators. It also introduced a system of competition between the Regional Employment Centres using bonuses for good and penalties for bad performers. However, before the first Centres would be confronted with sanctions, the bonus/penalty system was abolished. In the Netherlands, the benefit agency and the PES agency (now merged) conclude a contract with the Ministry every year. These contracts contain various performance objectives that include output as well as outcome indicators. As far as the local welfare agencies are concerned, the budget system that was introduced to finance local social assistance expenses now serves as the main mechanism for steering local service provision. Of course, New Public Management instruments do not necessarily replace more traditional forms of regulation. The cases of the UK and Germany illustrate this, as these countries steer their benefit and PES agencies (Jobcentre Plus in the UK, the Federal Employment Agency in Germany) through a combination of traditional regulation and New Public Management instruments. In Germany, for example, a nationwide, standardized system for profiling the unemployed exists. In addition, the Federal Employment Agency is subject to management by objectives, linking inputs (budgets) to outcomes (results in terms of labour-market integration). In the UK, we find a combination of, for example, central regulations aimed at standardizing frontline work and performance management instruments, such as a system through which Jobcentre Plus offices earn points by placing unemployed people in the labour market. As was the case with marketization in the UK, New

Tomáš Sirovátka

252

The Governance of Active Welfare States in Europe

Public Management rhetoric may have been stronger than New Public Management practices.

Governance regimes When we look at the governance reforms compared in the former section from the perspective of governance regimes, it seems quite clear that shifts are taking place in countries’ regimes for governing income protection schemes and activation services. The transformation from passive into active welfare state arrangements is one of the important developments that may explain these regime shifts, as they imply strengthening of the focus on service provision (especially activation and employment services), a greater emphasis on the conditionality of income provision and the rise of a variety of coordination issues (between various income protection schemes; between social and employment policies; and between activation and other social services). Following Considine’s typology of governance regimes and their characteristics, the following observations can be made. The first governance regime, procedural governance, is, according to Considine, characterized by its emphasis on law, rules, reliability and universal treatment. Here, we may understand procedural governance as governance through national, central regulation, as we do not have detailed information on how regional or local actors govern. As a consequence, decentralization is interpreted as a decrease in procedural governance, which is of course questionable, since national rules and regulations may simply be replaced by regional or local rules and regulations. Generally speaking, we can conclude that procedural governance has become both less dominant and – though more difficult to assess – less important in the context of activation-focused welfare states. Less dominant refers to the fact that all countries witnessed a hybridization of governance regimes: governance reforms have introduced (or strengthened) elements of other governance regimes, making it increasingly difficult to classify national governance regimes in terms of a single regime type. Claiming that procedural governance has become less important means that countries rely less heavily than formerly on national rules and regulations in governing specific policy fields. Shifts in the importance of procedural governance are most clearly visible in the area of activation programmes and services: when we look at income protection schemes the role of national rules and regulations has changed far less. Indicators for procedural governance in activation

Tomáš Sirovátka

Rik van Berkel et al.

253

are, among others, the existence of national standardized activation programmes, the degree to which details of these programmes are laid down in national regulation or are open for (local) discretion, the existence of national standardized profiling instruments, and the existence of national regulations concerning frontline work. Against this background, Sweden is a country where procedural governance is still strong, several recent reforms even showing trends of a strengthening of procedural governance. In the UK, procedural governance is still relatively strong as well, though recent reforms of the New Deal programmes imply less emphasis on programme standardization and regulate activation in less detail. The Netherlands is an example of a country where the importance of procedural governance in the governance of activation has significantly declined. One could argue something similar in the context of the Finnish LAFOS agencies: the content of services provided to the long-term unemployed is not simply determined by national regulations, but is partly the product of processes of negotiation and cooperation between the actors involved in the network. In several countries, procedural governance plays a more important role in governing the activation of unemployment benefit recipients compared with the activation of social assistance recipients. This seems to be the case, for example, in Germany, France and – evidently – countries without national Social Assistance systems (Switzerland, Italy). It also used to be the case in Sweden although, as we saw, this changed recently. Of course, procedural governance requires that states dispose of the capacities to effectively implement national rules and regulations. In this context we can point at the Czech case where a lack of capacities of the state to rule effectively through procedural governance has contributed to a situation in which nationally initiated governance reforms have been largely absent, whereas spontaneous governance reforms took place in attempts to cope with regional or local coordination problems, or to adapt to regulations in the context of activation projects funded by the European Social Fund (ESF). Besides procedural governance, other governance regimes have become more prominent in all countries. The rise of the market and corporate governance regimes are clearly visible in all countries, irrespective of the welfare state regime to which they belong. In a sense this is surprising, as the marketization of public services and New Public Management, which implies the introduction of private sector principles into the public sector, is seen as a neo-liberal strategy and one might expect that market and corporate governance fit better with liberal welfare states than with other types of welfare state. Nevertheless,

Tomáš Sirovátka

254

The Governance of Active Welfare States in Europe

even though it is true that the liberal welfare state included in our study, the UK, was the first country to start introducing service markets and New Public Management, several continental countries – notably Switzerland and the Netherlands and, more recently, Germany – match the UK’s enthusiasm for these governance regimes. As far as market governance is concerned, one might even argue that Switzerland and the Netherlands have surpassed the UK, at least until recently. Compared with the other countries, the social-democratic welfare states in our study, Finland and Sweden, have been most reserved in introducing these governance regimes; as a matter of fact, Sweden adhered to the traditional procedural governance model most strongly, although in recent years market governance has started to become more important. In Italy, France and the Czech Republic, the market governance regime gained ground as well. Especially in France and the Czech Republic, to date corporate governance has played a relatively minor role. One could argue that these discrepancies between welfare state and governance regimes are an artificial construct. Most welfare state regime typologies – and especially Esping-Andersen’s typology, which is still widely used in social policy studies – were developed prior to the welfare state reforms aimed at activating welfare state arrangements. So, in a sense, we could be criticized for comparing apples and oranges: welfare state regimes reflecting the pre-activation period and governance regimes in the context of modernized welfare states. In our view, this is a fair point. The introduction of the activation paradigm into social policies has been interpreted by many scholars as a shift from solidaristic and collective towards liberal and individualist arrangements, which has affected all welfare states, not only liberal welfare states. Against this background, we may conclude that we are witnessing similar developments in welfare state and governance regimes, rather than discrepancies. Network governance is the fourth governance regime distinguished by Considine. The first variety of network governance is corporatist governance, in which social partners – that is, organized interests of employers and employees – are strongly involved in policy-making and policy-implementation processes. Corporatist governance was traditionally weak in the UK and the Czech Republic, and far more important in the other countries, especially in the policy areas of social insurance, PES and labour-market policy. Governance reforms have had various effects on corporatist governance. In one group of countries (Italy, Switzerland, Sweden, Finland) no major change in the involvement of social partners took place. In the second group of countries (France, Germany,

Tomáš Sirovátka

Rik van Berkel et al.

255

the Netherlands) corporatist governance clearly weakened, although in various degrees: in the Netherlands, social partners were completely removed from social insurance and PES institutions, in France their position in the policy fields of activation and training was reduced, and in Germany social partners experienced a decline of their roles in the entire income protection system (UB-I and UB-II), as well as the PES. The second variety of the network governance regime, where clients and service providers are involved in joint service production processes, was found mainly in Finland in the context of activating the long-term unemployed. However, as mentioned before, our findings concerning this form of network governance are biased, as our analyses focused on national developments, whereas we expect a stronger proliferation of forms of network governance at regional and local levels (the French, UK, Italian and German cases provided some examples of this).

Effects of governance reforms In this book we have looked at three types of governance reform effects: outcome effects, output effects and process effects. In all countries, research of some sort is available that provides us with insights into the various types of reform effects. At the same time, the robustness of these insights leaves a lot to be desired: first, because much research does not really allow us to make reliable statements about the effects of specific governance reforms. Reforms often come as reform packages, which makes it hard to unravel the effects of governance reforms and those of social policy reforms, or to disentangle the effects of various governance reforms. Although in most countries available research data provide a general picture of the effects of activation, it is difficult if not impossible to find out how specific policy or governance (reform) characteristics contribute to (in-)effectiveness. Secondly, systematic comparative research which either compares pre-reform and post-reform effects or effects of various governance models existing at the same time, is scarce. This type of research is especially important when we want to gain insight into outcome effects. International comparative research into effects of governance reforms is even harder to find. Thirdly, some types of governance reforms are studied in more detail than others. For example, relatively speaking marketization has received considerable attention, whereas New Public Management or decentralization reforms have been studies less frequently. All this

Tomáš Sirovátka

256

The Governance of Active Welfare States in Europe

implies that our findings concerning the effects of governance reforms are rather tentative. Outcome effects Most available studies on the outcome effects of governance reforms focus on labour-market reintegration. Other potential outcomes receive less or no attention, for example: sustainable labour-market participation, employability, social inclusion or benefit independence. The Swiss case provided an illustrative example of this: Swiss research showed that the activation of social assistance recipients was not very effective in bringing people back to the labour market; at the same time, Swiss activation of social assistance recipients is only partially aimed at realizing this objective. As far as the impact of marketization of activation services is concerned, the results of available studies are not very clear. In Germany, no evidence was found that private providers produce better results than public providers, and several British studies reached a similar conclusion. In France public providers were found to be more successful than private providers, whereas in Sweden the effectiveness of public and private service provision turned out to be different for different target groups. However, as the Dutch case showed, these results are difficult to interpret: they may tell us something about marketized service provision per se, but positive or negative effects may also be due to specific characteristics of service markets. In the context of marketization, both the German and the Dutch case showed that increasing the voice and choice of unemployed people in activation may have a positive impact on the effects of marketized services. The outcome effects of promoting inter-agency cooperation are also inconclusive. Studies of the Finnish LAFOS agencies revealed that the services provided by these agencies contributed to the employability of the unemployed, but that the effects of the services in terms of reintegration into work were modest. These two effects were also investigated in German studies of the various models for servicing the recipients of Unemployment Benefit II. These studies found that the ARGE model was more successful in promoting labour-market reintegration, whereas the municipal model was more successful in promoting the employability of the unemployed. Experiments with improved local cooperation in the Netherlands (the showroom experiments) also showed some positive outcome effects, in terms of shortening benefit periods and quicker labour-market reintegration of unemployed people.

Tomáš Sirovátka

Rik van Berkel et al.

257

Studies of the outcome effects of decentralization are scarce, which is surprising given the fact that decentralization is often advocated in order to promote tailor-made activation services. Interestingly, disappointing results of municipal activation programmes in Sweden were used as an argument to centralize the activation of some groups of unemployed that used to be target groups of municipal activation. In the Czech Republic, some comparative data are available on the effects of central, national programmes and more decentralized ESFfunded activation projects. These data showed that, although ESF projects turned out to be more effective initially, differences between the results of national and ESF projects diminished over the years. In the Netherlands, where a far-reaching process of decentralizing the activation of social assistance recipients took place, outcome effect studies focused mainly on the new system for financing local social assistance expenses, showing that the new financing system contributed to lower volumes of social assistance recipients – which, however, does not necessarily coincide with improved effectiveness in terms of labour-market reintegration. Output effects The chapters in this book have demonstrated that governance reforms have had various output effects, although here again, methodological issues make it difficult to provide robust statements on how specific governance reforms produce specific effects. There are some indications that governance reforms have contributed to the participation of larger groups of unemployed people in general, or specifically of long-term unemployed people, in activation. In France and the Netherlands, for example, the involvement of other actors in the provision of activation services has had an impact in this respect; the German ARGE as well as the Finnish LAFOS agencies strengthened the efforts aimed at activating long-term unemployed people. In the Czech Republic, ESF projects improved the coverage of activation (temporarily), although it is difficult to establish whether this is mainly the effect of the availability of extra funds, or of the ‘spontaneous’ governance reforms that the ESF projects triggered. In Switzerland, where national governance reforms affected mainly the activation of unemployment benefit recipients, the proportion of benefit recipients involved in activation is considerably higher than the proportion of activated social assistance recipients. Of course, promoting the activation of unemployed people (and treating increasing numbers of benefit recipients as if they were ‘normal’ unemployed people – see, for example, the UK

Tomáš Sirovátka

258

The Governance of Active Welfare States in Europe

case) has also been an explicit objective of substantial or formal policy reforms in all countries. In other words, insofar as governance reforms have increased the coverage of activation services – in terms of the proportion of unemployed people participating in activation – governance and social policy reforms have reinforced each other. The relation between social policy and governance reforms may be more problematic where activating the hard-to-employ is concerned. Various chapters pointed at ambivalent effects of governance reforms on activation services for the hard-to-employ, and showed that governance reforms may weaken rather than strengthen activation efforts aimed at this group of unemployed. This is especially the case when efficiency and effectiveness (in terms of (quick) labour-market reintegration) considerations are emphasized as a consequence of governance reforms. Both in the context of marketization and the introduction of New Public Management principles, the processes of creaming-off and prioritizing groups of easy-to-employ unemployed occur. Creaming-off processes in the context of marketization were observed, for example, in France, Germany and the Netherlands. New Public Management has an impact on the activation of the hard-to-employ in a variety of ways. Performance indicators may create an incentive for public service purchasers or providers to prioritize short-term job placements and the easier-toreintegrate unemployed, as the UK and Swiss cases show. A similar effect may result from the ways in which public organizations are financed, as was shown in the cases of the German UB-I recipients and Dutch social assistance. Against this background, the results of the study comparing the German models for activating UB-II recipients is interesting: the ARGE services (where the impact of NPM instruments is more clearly felt than in the municipal model) are more strongly focused on direct labour-market integration, which may make creaming-off more likely than in the case of the municipal model, where direct labour-market integration is less dominant. However, prioritizing the easier-to-reintegrate is not only a consequence of efficiency or effectiveness considerations: as the Czech case shows, where management by objectives was characterized as weak and lacking ambition so that its impact on activation practices is probably limited, there may also be the effect of a lack of resources and capacities to deal with more complex cases and needs. Several countries reported that governance reforms may affect the content of services and again, these effects are ambivalent, as we have seen examples of both service standardization and service differentiation, as well as of improved and reduced service quality. Service

Tomáš Sirovátka

Rik van Berkel et al.

259

standardization may be the consequence of policy centralization, as is traditionally the case in the UK (although a more flexible activation policy was introduced recently); and as took place in Sweden with respect to the activation of some traditional municipal target groups. Under certain conditions, marketization may promote standardization (rather than differentiation, which is often one of the intended effects of marketization) as well. In Germany, standardization as well as reduced quality of services was related to the stricter requirements that private providers have to meet, which triggered risk minimization strategies among private providers. Something similar was the case in the early years of marketization in the Netherlands, when price competition was strong, contracts covered relatively short time periods and private providers were not eager to innovate. More recently, however, new ways of involving market actors seem to have contributed to service diversification. Interestingly, the Czech case seems to suggest that weak competition may provide a disincentive for innovation and diversification as well. Marketization may also contribute to a reduction of diversity in activation when competition pushes specific types of providers from the market (as was reported in the UK and is considered a risk in the French cases), which may lead to a more standardized offer of services. Finally, New Public Management may contribute to standardization as well, as we saw: it may contribute to a situation in which public organizations concentrate on providing or purchasing those services that are considered to promote short-term labour-market integration. Governance reforms may also promote service de-standardization and diversification. As the above-mentioned case of marketization showed, conditions under which market providers operate and relationships between purchasers and providers both play a role in determining whether markets promote standardization or diversification. Evidently, decentralization is most often cited as contributing to service diversification. Service diversification and service individualization were mentioned as effects of decentralization in, among others, Finland, Sweden, the Netherlands, Germany (especially the municipal model) and Italy. However, service diversification is not always valued in a positive way, as the case of Sweden showed where diversity was not considered to contribute to the effectiveness of activation services and programmes. Furthermore, in several countries the issue of how service diversification, decentralization and individualization relate to equal treatment, equal access and transparency is still unresolved. In addition, the Czech and Italian cases make clear that the potential positive

Tomáš Sirovátka

260

The Governance of Active Welfare States in Europe

effects of decentralization can only be realized under certain conditions, of which the availability of resources and capacities at regional and local levels are probably the most important. Process effects As becomes clear from the above, the impact of governance reforms on more individualized service provision – an objective that seems to be more strongly pursued where the activation of the long-term unemployed is concerned – is rather ambiguous: some reforms provide more room for it, whereas others counteract it. Furthermore, implementation conditions play an important role in making service provision processes more or less individualized and tailor-made: such as caseloads, which had a positive impact on the services provided by the German Federal Employment Agency but in the Czech case turned service individualization into a rather formal and administrative process; the diversity of available services, which limited the attempts to individualize services in the Swiss case; or a range of organizational concerns resulting from, for example, mergers or performance management that place organizational rather than client needs in the centre (see, among others, the German and French cases). In the context of governance reforms, and especially marketization, attempts to promote the voice and choice of the unemployed in activation processes by providing them with institutionalized instruments to exercise influence are scarce, and were found only in Germany and the Netherlands. In the German case, only a very small proportion of the unemployed make use of the voucher system, which is partly due to the lack of market transparency for users of vouchers. In the Dutch case, the use of the Individual Reintegration Agreement for the recipients of unemployment benefits is much more widespread. Apart from these institutionalized instruments for increasing the voice and choice of the unemployed, frontline workers may give the unemployed more or less voice and choice in individual activation processes. This seems to have been the case in the Finnish LAFOS agencies: one of the reasons for the high client satisfaction with LAFOS services was that the voice of the unemployed was better heard. However, our study was not designed to investigate the impact of governance reforms on frontline practices and the nature of individualized activation. Finally, several UK and Swedish studies found that clients of private providers were more satisfied than those of public providers, although it is not clear whether this is related to the fact that private providers give the unemployed more room for voice and choice.

Tomáš Sirovátka

Rik van Berkel et al.

261

Governance reforms and social policy reforms Social policy reforms and governance reforms are interrelated in complex ways. As was elaborated upon in Chapter 1, social policy reforms may create an urgency for governance reforms, which is clearly the case in the context of the increasing importance of activation and the increasing conditionality of income support in active welfare states. But as the former section showed, the opposite may also be the case: governance reforms may also have an impact on social policies, or rather on social policy practices, that is, the ways in which the rights and obligations of the unemployed are put into practice, the nature of the services they are entitled to, the accessibility of services and so on. In addition, governance reforms may have effects that are in line with official social policy objectives, but may also counteract them. The first is the case when governance reforms involve more unemployed people in activation, contribute to more diversified and individualized services (in as far as social policies intend to provide less standardized and more tailor-made services) or improve the effectiveness of services. The latter is the case when governance reforms exclude specific groups from activation (where social policies aim to be inclusive), promote service standardization (where social policies aim to diversify services) or narrow the scope of activation services to quick labour-market reintegration. Of course, the match or mismatch between social policy objectives and governance reform outcomes is not always easy to establish: social policy objectives are often multifarious, and so are the objectives and actual outcomes of governance reforms. In addition, as we saw, outcomes of governance reforms are difficult to investigate. Nevertheless, tentative and inconclusive as the findings on effects of governance reforms presented in this book may be, we think that the following conclusions can be drawn. First, available research confirms our expectation that governance reforms are not neutral in terms of their impact on social policy practices and, thus, on social policy outcomes. Secondly, governance reforms may be conducive to, or counteract, social policy objectives and intentions. Thirdly, the complex interactions between social policies and the governance structures through which they are implemented, delivered and provided make governance regimes and governance reforms an interesting topic for social policy researchers who are interested not only in ‘official’ social policies but also in (the conditions of) the actual ‘production’ of these policies in the organizations involved in implementation processes.

Tomáš Sirovátka

262

The Governance of Active Welfare States in Europe

Another conclusion with respect to the interrelations between social policy and governance reforms relates to the fact that throughout this book we have seen that governance reforms come in many shapes and varieties. Because of that, it seems quite worthless to talk in a generalized way about the activation market, the activation service network, the New Public Management and so on. Instead, there are various types of markets and networks (and combinations of the two), various types of performance management (NPM) strategies, various mixes of centralized and decentralized policy making and implementation; and in assessing the impact of governance reforms on social policies and social policy practices, it is virtually impossible to generalize and abstract from the more concrete characteristics of governance structures and governance regimes. This book has made only a start in unravelling the various dimensions that might be relevant in recognizing variety within specific forms of governance. More theoretical as well as empirical work will be needed to elaborate upon this – in this context, we believe that comparative research (at both the national and international level) can be very useful. Finally, a comment on the capacities that actors need to make governance reforms actually work: markets require more than assigning the roles of purchasers and providers; networks require more than co-locating service providers; decentralization requires more than delegating decision-making powers. The ‘care for institutions’ (and, we might add, for the people working in these institutions) mentioned in Kooiman’s and Bavinck’s definition, cited in Chapter 1, seems to be a vital element of governance reforms. Too often, this element seems to be neglected, bringing policy makers to the conclusion that governance reforms have failed whereas what actually went wrong is that institutions were not cared for. A research agenda, focusing on the ways in which social policy and governance reforms are intertwined, should not overlook the ways in which institutions involved in policy implementation and service provision are enabled and equipped to perform the functions and tasks expected of them. The theme of a shift in forms of citizenship should be part of this research agenda. Since activation and its governance affect, as we have seen, the rights and duties of the unemployed, it is important to analyse the direction of this change. Traditionally, the form of deserving citizenship implied the protection of the worker against risks such as unemployment by offering a substitute income, but now a transformation towards a different form of deservedness is developing. This form considers the productive citizen as a new ideal: a citizen who contributes

Tomáš Sirovátka

Rik van Berkel et al.

263

to society, preferably through work, and takes responsibility for protecting themselves against the risks and for grabbing the opportunities of the current society. The governance reforms we have described try to deliver the instruments for bringing about this shift and, as such, a study of the relation between governance reforms and citizenship would be instructive to look at the precise implications of this relationship.

Tomáš Sirovátka

Index

Note: The letters ‘f’ and ‘t’ following the locators refer to figures and tables respectively. activation, organization and provision of, 27–36 actors and agencies, 35–36 Czech Republic: democratic labour-market and employment institutions, 29; ESF, 29; post-communist phenomenon, 28; Public Employment Services, 29 Finland: co-funding, 32; external providers, 33; LAFOS, 32 France: AFPA, 34; ANPE, 34; DIRECCTE, 34; external service providers, 34 Germany: Hartz reforms, 31; Unemployment Benefits I and II (UBI and II), 31 issues, 28 Italy: flexibilization of labour market, 35 municipalities, role of, 30–31 national (PES-run) or local programmes, 31 Netherlands: benefit agency (UWV), 31–32; outsourcing activation, 32; Work Plazas, 32 Sweden: ALMP, 28; PES, 30 Switzerland: absence of national social assistance legislation, 33; Logistics Offices, 33; private social assistance, role of, 34; REC, 33 UK: Jobcentre Plus, 29; role of municipalities, 29 Active Labour Market Policies (ALMP), 1, 27, 114, 114t, 153, 155–156, 160–164, 169, 174–175, 177, 181–182, 184–188, 191

Active Solidarity Income (RSA), 25, 34, 44, 48, 56 active welfare states in Europe, comparative perspective governance reforms: decentralization, 239–243; inter-agency cooperation and partnerships, 246–250; marketization and competition, 243–246; new public management, 250–252; outcome effects, 256–257; output effects, 257–260; process effects, 260 governance regimes: corporatist governance, 254; Esping-Andersen’s typology, 254; joint service production, 255; neo-liberal strategy, 253; network governance, 254; procedural governance, 252–253 social policy reforms: capacities needs of actors, 262; effects of, 261; multifarious objectives, 261; shift in forms of citizenship, 262–263; urgency for, 261; varieties in, 262 see also individual countries administration of income provision and activation activating welfare state arrangements, 36 activation of social assistance recipients, 37 emphasis on activation, 36 income support, see income support systems, administration of

264

Tomáš Sirovátka

Index organization of activation, see activation, organization and provision of Agencies for Work (AfW, Germany), 65t, 67–68, 70 ALMP, see Active Labour Market Policies (ALMP) ARGE consortia (Germany), 65t, 68, 72 Association for Adult Vocational Training (AFPA), 34, 35t, 41, 43t, 46, 48, 53 Association for Managers’ Employment (APEC), 43t, 44, 46, 52–53 Catholic familialism, 132 Centre for Work and Income (CWI), 142 Christian Democratic–Liberal coalition, 63 competitive decentralization, 7–8, 137 contracts, nature of classic and trust, 9 contractualism, 4, 12, 40, 47, 49–50, 55, 223, 250 contractualization, 86, 92, 95, 97–98, 154, 159t, 163, 168–169, 183 cooperation between public/ private/voluntary sector, 98 Course Towards Enterprises (France), 57 CWI, see Centre for Work and Income (CWI) Czech Republic activation, institutional framework, 173–176: ALMP, 174; autonomy of EOs, 174–175; Counselling Committees, 175; EOs at local level, 175–176; ESA, capacity of, 174; IAPs, 176; Ministry of Labour and Social Affairs, 174; NAPE, 174; National Tripartite Council, 175; Regional Self-Government Units, 175; two-tier PES, 174 decentralization and inter-agency cooperation, 177–180: bottom-up initiatives and

Tomáš Sirovátka

265

cooperation, 180; EOs, difficulties, 180; EOs, informal regional support networks, 178; imitation and ideational transfer, 177; incorporation of private participants, 179; normative (legislative) transformation, 177; politicized formation of policies, 179; purpose-built Programmes, 178–179; rational problem solving, 177; Regional Integrated Projects, 179; role of local EOs, 179; spontaneous decentralization, 177; transformation at the local and regional levels, 177 emerging mode of governance, 183–184: hybrid mode, characteristics, 183–184 ESF measures, expectations, 184 marketization, 180–181: ESF projects, national/‘individual grant,’ 181; increase in NGOs, 181; misuse of market position, 181; new Employment Act, 180; private employment agencies, 180; subcontracting activities, 180 new public management and lack of coordination, 181–183: ESF line/national ALMP line, 182; management-by-objectives approach in ALMP, 181–182; OMC, 183; organizational changes within PES, 182 outcome effect, 185: employment effects of the ESF measures, 185 output effect, 185; contents of activation, 188; Employment Act from 2004, 188; incentives to work, improvisation of, 186; logistic regression analysis, 187; National Reform Programme 2005–8, 188; scope, 185; targeting, 186–188

266

Index

Czech Republic – continued procedural effects, 188–190: IAPs, development of, 189; individualization, 188; privatization and outsourcing of employment services, 189–190 top-down reforms, absence of, 184 trends in, 176–184, 178t decentralization capacity building, 7 centralization, 7–8 comparative perspective, 239–243: administrative decentralization, 240–241; centralization, 241–243; complexity in analysing, 240; degrees of (de-)centralization, 240–241; policy choices and decisions, 240; spontaneous than intended, 241 competitive vs. non-competitive, 6–8 Czech Republic, 177–180: bottom-up initiatives and cooperation, 180; EOs, difficulties, 180; EOs, informal regional support networks, 178; imitation and ideational transfer, 177; incorporation of other private participants, 179; normative (legislative) transformation, 177; politicized formation of policies, 179; purpose-built programmes, 178–179; rational problem solving, 177; Regional Integrated Projects, 179; role of local EOs, 179; spontaneous decentralization, 177; transformation at local/regional levels, 177 Finland: LAFOS, 219–220 France, 47–49: call-for-tender procedures, 56; impact of the mid-2000s, 56; law No 2003-276, 47; law No 2004-258, 47; levels, regional/ departmental/local, 48;

procedural and institutional reforms, 56; RMA, 48; RSA, 48; vocational training policies, 48 Germany, 71–73: contracting-out of services, 72; labour-market, centralization/standardization, 72; regional purchasing centres, creation of, 72; SGB III, 71–72 incentive paradigm, 8 internal vs. external, 6–7 Italy, 121–122: effects of reforms, 127; social assistance policy, 125 Netherlands, 136–138: Act on integration of young (under 27) unemployed, 137; competitive decentralization, 137; Individual Reintegration Agreements (IRO), 137; new benefit agency, 136; Social Assistance Act in 2004, 137; UWV, decentralization of, 137 political vs. administrative, 6–7 reasons to opt, 6–7 strategic choices, 6 Sweden, 202–203: ‘Act on Municipal Responsibility for Young People between 20–24 Years,’ 202; local employment service committees, 203; National Labour Market Administration, 203; Swedish Social Services Act in 1998, 203 UK, 94–96: area-based Employment Zones, 95; contracting-out, 94; Department for Work and Pensions, 96; ‘District Delivery Plans,’ 95; Flexible New Deal, 95–96; New Deal programmes, 94; New Deal Task Force, 95; new public management practices, 94; NPM practices, 94; target groups, 95; Working Neighbourhoods Pilots, 95 welfare states, governance of, 6–8 Department for Work and Pensions (DWP), 85–86, 89, 92–93, 96–98, 103–104, 245

Tomáš Sirovátka

Index Directorate General for Labour, Employment and Vocational training (DGTEFP), 41, 43t Disability Benefit system (Netherlands), 136 DWP, see Department for Work and Pensions (DWP) employment policy/assistance, see under individual countries ESF, see European Social Fund (ESF) European Employment Strategy, 1, 40, 121, 182, 184, 216 European Foundation for the Improvement of Living and Working Conditions, 11 European Social Fund (ESF), 29, 123, 175, 216, 245, 253 Finland activation policy: fourth phase, 218–219; labour market support in 2006, 218; second phase, 217; third phase, 217–218; via EU and EES, 216–217 clients in integrated services, 228–231: contradictory results of LAFOS, 230; employability, components, 228; employment results, 231; ended unemployment periods in LAFOS, 230t; external factors, 228–229; hard-to-employ citizens, 228; insights/feedback, 229; problems, 229 governance models, 231–233: network governance model, 231–232, 231t; procedural/corporate/market, 232–233 integrated approach in activation, 216–219: KELA, 216; Rehabilitative Work Experience Act, 217; Social Assistance Act, 217 joint services, multi-sector implementation, 220–223: Employment Policy Programme (2003–7), 220; joined-up

267

government, goals of, 221–222; LAFOS organization and objectives, 221–222; Ministry of Labour, function of, 221; Nordic model, 223 LAFOS, network approach, 219–220 local-level management, 225–227: accountability, 226–227; evaluation of LAFOS, 226; operational leaders of LAFOS, 226; steering group, role of, 225–226; T & E Centres, 225; two-sector leadership, 226 multi-professional/sectoral service structure, 227–228: conflicts, LAFOS teams, 227; working-pair method, 228 state-level steering mechanisms, 223–225: Association of Finnish Local and Regional Authorities, 224; Ministry of Labour and PES, 224; referring clients to LAFOS, 224–225; steering group’s mandate, 224 Flexible New Deal (UK), 90, 95–96 France changes in activation strategies, see France, activation strategies changes in governance, 54–55: intermediaries, 54; law and tripartite negotiations, 54; unions’ regulations, 54–55 contracting/outsourcing, 50–53: co-contractors/subcontractors, 52; conflicts in employment institutions, 51; equal access to PES, 52–53; 2005 Law on Social Cohesion, circles, 43t, 52; mid-2000s, 51; PARE/SMP/PAP, 51 decentralization, 47–49: Constitutional by law No 2004-258, 47; Constitutional Law No 2003-276, 47; levels, regional/departmental/local, 48; RSA/RMA, 48; vocational training policies, 48

Tomáš Sirovátka

268

Index

France – continued devolution: creation of Local Missions (1980), 49; downward/upward logic, 48 impacts of governance reforms, 55–58: call-for-tender procedures, 56; contracting-out, 56–57; decentralization in practice, 55–56; impact of mid-2000s decentralization, 56; inter-agency cooperation, 57–58; OPP-CVE experimentation, 57; procedural and institutional reforms, 56 inter-agency cooperation and merging services, 53–54: Employment Houses, 53; joint location, tool, 53; Law of 13 February 2008, objectives, 53–54; unemployment insurance agencies and ANPE, 53 local benefit agencies, 39 national insurance fund, 39 new public management, 49–50 NPM and state contractualism, 49–50: modernization process, 49; performance contracts (progress contracts), 50; policy evaluation and accountability, 50; RGPP, 50 unemployment benefits and income support, 49 France, activation strategies, 39–47 corporatist regulations, 46–47: AFPA, 46; APEC, 46; PARE, 47; ‘reduced activity,’ 46; tri-partite convention of 13 June 2001, 47; Unédic, 46 income protection schemes, 44–46: API, 44; CAF, 46; National Family Fund (CNAF), 46; National Unemployment Insurance Fund, Unédic, 44; RMI, 44; RSA, 44; social assistance, 45t, 46; unemployment compensation

(insurance and solidarity schemes), 44–46, 45t Local Missions, 44 NGOs, 44 PES, institutional governance, 40–44: AFPA, 41; agencies at the regional/departmental/local level, 41; ANPE, 41; centralized and corporatist, 40; co-contractors, Local Missions, 44; decentralized services, 41; DGTEFP, 41; DIRECCTE, 41; labour-market institutions, 41; Ministry of Labour, 41; PES for placement and unemployment compensation, 42t; second circle, contributing to PES, 41; Social Cohesion, law on, 41; Subcontractors, NGOs, 44; third circle, may participate in PES, 41; three circles of public employment services, 43t shifting activation policies, 40: economic rationality, arguments, 40; frontline workers, 40; income support recipients, 40; institutional change, 40; institutions and frontline workers, 40 General Revision of Public Policies (RGPP), 50 Germany Agencies for Work (AfW), 67 ARGE consortia, 68 combination of formal and operational elements, 64 decentralization and (re-)centralization, 71–73: contracting-out of placement services, 72; creation of regional purchasing centres, 72; purchase of labour-market, 72; SGB III, 71–72 employability, goal of labour-market policy, 63 FEO: Federal Employment Agency, 67; means-tested benefit, municipalities, 66;

Tomáš Sirovátka

Index Unemployment Insurance Benefit/Assistance, 66 Hartz commission (Modern Services in the Labour Market), 64 Hartz reforms, 63–68: decentralization and (re-)centralization, 71–73; inter-agency cooperation, 69–71; main trends, 68; marketization, 74–75; new public management, 73–74; social partners, role of, 68–69 implementation structure for income protection and activation, 64, 66t insurance-based with income regime, comparison, 79–80 inter-agency cooperation, 69–71: division of benefit administration, 70; fragmented financial/organizational structure, 70–71; incentives, 71; Jobcentres, FEO turned, 69; local AfW and municipalities, division of, 70; twofold motivation, organizational reform, 69 labour-market policy, 67 marketization, 74–75: contracting-out of placement services, 74; 1-Euro-job, 75; individual placement budget in 2009, 75; market share and effectiveness, 74; Personnel Service Agencies (PSAs), set up of, 75; public monopoly for job placement, 74; 1990s, job placement conciliation, 74 Modern Services on the Labour Market (2002 and 2005), 64 new public management, 73–74: action programmes and resource allocation, 73; clients for counselling/ activation/qualification, 73; cooperation of person seeking assistance, 73–74; cost-effectiveness, 73; costs of input with output, 73;

269

incentive assistance clients, 73; market clients, 73; service centres, 73 opted-out communes, 68 outcome effects, 76: ARGE/municipal providers, performance of, 76; long-term unemployment, comparison, 76 output effects, 77–78: incentive assistance clients, 77; PSAs, importance of, 77; quality of activation, 77; training voucher, criticism of, 78 pathbreaking change or paradigm shift, 79 placement scandal in 2002, 63 process effects, 78–79; specialized services, 79 self-governance via tripartite corporatism, 79 Social Code Book (SGB), I/II, 67 social partners, role of, 68–69: indirect marginalization, 68; representation through trade unions, 69; three-person Executive Boards, 68 three-tier system until 1960s, 62 trends in governance reforms, 65t, 66t Unemployment Benefit II (UB II), 67 universal working-age benefit, 79 governance, 13–16 definitions, Kooiman and Bavinck, 2–3 empowerment, concept of, 14 ideal-type, models: hierarchical model, 14–15; open-system model, 14–15; rational goal model, 14–15; self-governance model, 14–15 procedural governance, 15 uses of concept, 2 hard-to-employ citizens, 228 Hartz reforms, see Germany

Tomáš Sirovátka

270

Index

income provision and activation, administration of activating welfare state arrangements, 36 activation of social assistance recipients, 37 emphasis on activation, 36 income support, see income support systems, administration of organization of activation, see activation, organization and provision of income support systems, administration of, 23–27 actors and agencies, UB and SA schemes, 28 Czech Republic: PES, 27; social assistance, 25; unemployment benefit systems, 24 differences between, 23–24 Finland: labour market support, 24; local responsibilities, strengthening, 25; social assistance, 25; voluntary unemployment insurance system, 23 France: ANPE, 27; dual unemployment compensation system, 24; local Employment Counters, 27; social assistance system, 25; solidarity scheme, 24; unemployment compensation systems, 26 Germany: Agency’s Service Centres, 26; Hartz reforms, 25; income-related system of unemployment benefits, 24; unemployment benefit I, 26 Italy: mobility benefit, 24; National Institute for Social Security (INPS), 27; social assistance provision, 26; unemployment benefit arrangements, 24 Netherlands: income-related system of unemployment benefits, 24; UWV (Agency for the Administration of Employees’ Insurance), 27

reforms of the administrative structures, 24–25 Sweden: KELA, 26; social assistance by local taxes, 25; Social Service Act of the early 1980s, 25; Unemployment Insurance Board (IAF), 26; voluntary unemployment insurance system, 23 Switzerland: Swiss Welfare Conference (SKOS), 26; unemployment benefit systems, 24 UK: JSA, contribution/income-based, 23 see also individual countries Individual Action Plans (IAPs), 176, 182, 184, 188–189, 250 inter-agency cooperation, 3 comparative perspective, 246–250: coordinating of benefit agencies and PES, 246; coordinating of municipalities and their local welfare agencies, 247; coordination models, 248; national coordination problems, 247–248; Work Squares, 248–249 Czech Republic, 177–180: bottom-up initiatives and cooperation, 180; EOs, difficulties, 180; EOs, informal regional support networks, 178; EOs, role of local, 179; imitation and ideational transfer, 177; incorporation of other private participants, 179; normative (legislative) transformation, 177; politicized formation of policies, 179; purpose-built Programmes, 178–179; rational problem solving, 177; Regional Integrated Projects, 179; spontaneous decentralization, 177; transformation at the local and regional levels, 177 forms of: one-stop agencies, 10 France, 53–54: Employment Houses, 53; joint location, tool, 53; Law

Tomáš Sirovátka

Index of 13 February 2008, objectives, 53–54; unemployment insurance agencies and ANPE, 53 Germany, 69–71: division of benefit administration, 70; fragmented financial and organizational structure, 70–71; incentives, 71; Jobcentres, FEO turned, 69; local AfW and municipalities, division of, 70; twofold motivation, organizational reform, 69 Netherlands, 142–144: (inter-)municipal Youth Counters, creation of, 143; services provided by CWI, 142–143; showroom experiments (UWV/CWI/local welfare agencies), 143; SUWI-reform, 142; Work Plazas, 143 partnership: concept of, 11; factors contributing to, 11 reasons for promoting, 10 Sweden, 208–209: Act on Financial Coordination of Rehabilitation Measures, 208; collaborations between welfare institutions, 208; institutionalized cooperation, desire for, 209; Social Service Act, 208; socio-medical activities, 209; vocational rehabilitation, 208 Switzerland, 164–165: cross-institutional processes (IIZ-MAMAC), 165; inter-institutional cooperation project, 164; lack of cooperation, 164; medical employment market assessments, 164; multi-problem cases, 165 ISFOL, 119, 121, 123, 126–127 Italian National Social Security Institute (INPS), 27, 117–118, 120–121

271

Italy effects and challenges, 126–127: implementation of Law 328/2008, 127; National Anti-Poverty Committee, 127 employment and social policies, 111–113: Considine’s contribution, 112; hierarchy model, 111; open system model, 111; rational goal model, 111; self-governance model, 111; typology, 112t hierarchical governance model (1948–97), 117 reforms (1997–2010), 120–126 social assistance and employment, comparison, 113–117: ALMP, 114t; EU27 and EU15 averages, 113; expenditure for services, families/elderly people/ disabled, 116t; relative household consumption-based poverty, 115t; share of persons with equalized disposable income, 115t; total unemployment rate, 114f social assistance benefits, 119–120: housing and social care policies, 119–120; minimum income schemes, 119–120; minimum level of income to citizens, 119; National Social Assistance Fund, 119 social assistance policy reforms (1997–2010), 123–126: decentralization, 125; local non-institutional actors, 125–126; Local Programming Plans, 125; social insertion programmes, 125 state-centric/hierarchical to multi-level/open model, 127–130 tasks before/after 1997 reforms, 124, 124t trade unions’ internal composition, 129t

Tomáš Sirovátka

272

Index

Italy – continued unemployment benefits/ employment/vocational training, 117–119: education and vocational training, 119; mobility allowance/benefit, 118; ordinary unemployment benefits, 117–118; short-term unemployment benefits, 118 unemployment benefits/ employment/vocational training (1997–2010), 121–123: decentralization, 121–122; INPS, role of, 121; Lavori di Pubblica Utilità (LPU), 122; Lavori Socialmente Utili (LSU), 122; marketization, 122; PIR – projects or programmes to promote the inclusion or re-inclusion of workers, 123; Socially Useful Jobs, 122; ‘Social shock-absorbers and incentives for occupation,’ 122; transfer of personnel and functions, 121–122 Job and Development Guarantee (Sweden), 200, 208 Job Guarantee for Youth (Sweden), 200 Jobseeker’s Allowance (JSA), 23–24, 27–29, 88–91, 94, 96, 102, 240 JSA, see Jobseeker’s Allowance (JSA) Labour Force Service Centres (LAFOS), 32, 35, 37, 218–234, 249–250, 253, 256–257, 260 Labour Market Policy Activities Ordinance (Sweden), 199 LAFOS, see Labour Force Service Centres (LAFOS) LAFOS, network approach administrative solution for LAFOS, 221 conflicts among teams, 227 contradictory results of, 230 developmental stage of performance management (NPM), 219 ended unemployment periods, 230t

evaluation of, 226 flexible network structures, 234 ideal types of governance: procedural, corporate, market and network, 219 legitimization of, 223 network governance: definition, 220; research, 220 objectives of, 222 organization and objectives, 222 position within administrative structure, 223t referring clients to, 224–225 unusual intervention, 234 see also Finland Local Family Funds (CAF), 45t, 46 Lone Parent Allowance (API), 44 marketization activation services, aspects of, 9: competition, nature of, 9; contracts, nature of, 9; market, nature of, 9; providers, nature of, 9; purchasers, 9 comparative perspective, 243–246: decentralization of purchasing processes, 245–246; market competition, 245; nature of providers, 244; proportion of activation services in countries, 244; role of quasi-markets, 243; role of the unemployed themselves, 246 competitive decentralization, 8 Czech Republic, 180–181: ESF projects, national/‘individual grant,’ 181; misuse of market position, 181; new Employment Act, 180; NGOs, increase in, 181; private employment agencies, 180; subcontracting activities, 180 Finland, 119, 232–233 France, 40, 51, 54, 57 Germany, 74–75: contracting-out of placement services, 74; 1-Euro-job, 75; individual placement budget in 2009, 75; market share and effectiveness,

Tomáš Sirovátka

Index 74; PSAs, set up of, 75; public monopoly for job placement, 74; 1990s, job placement conciliation, 74 Italy, 111, 122, 126–127, 129; vocational training marketization, 121 Netherlands, 138–140: Individual Reintegration Agreement, 139; mix of in-house/externally provided activation, 140; national regulation, 138; no-cure no-pay or no-cure less-pay contracts, 140; purchasers, 139; radical marketization, 138 quasi-markets, 8 Switzerland, marketization in a federal system, see Switzerland UK, 97–98: conservative–Liberal Democrat coalition government, 97; development of the quasi-market, 97; market-based reforms, 97; private providers/privatization, 98; profit-making multinationals, 98 marketizers, 4 Minimum Activity Income (RMA), 48 Minimum Integration Income (RMI), 44 Monthly Personalized Support (SMP), 51 NAPE, see National Action Plans of Employment (NAPE) National Action Plans of Employment (NAPE), 174 National Anti-Poverty Committee (Italy), 127 National Board of Health and Welfare (Sweden), 201 National Employment Agency (ANPE), 38, 41, 49–50, 53–54, 57 National Family Fund (France), 46 National Labour Market Administration, Sweden, 203 national social assistance scheme, 24, 29

273

National Social Insurance Institution (KELA), 26, 28, 216 Netherlands Catholic familialism, 132 Centres for Work and Income, one-stop agencies, 135 decentralization and (re-)centralization, 136–138: Act on integration of young (under 27) unemployed, 137; competitive decentralization, 137; Individual Reintegration Agreements (IRO), 137; new benefit agency, 136; Social Assistance Act in 2004, 137; UWV, decentralization of, 137 Dutch welfare state, characterization of, 132–133 effects of reforms, 145–150: activation trajectories, volume of, 148t; content of activation services, 148; diversity of activation programmes, 149; IRO evaluation, 146; labour-market reintegration, sustainability of, 146; outcome effect, 147; output effects, 147–148; showroom experiments, 147; vulnerable groups of unemployed, 149–150; Work First programmes, target groups, 148–149, 148t income protection/activation, implementation structure, 134t inter-agency cooperation, 142–144: (inter-)municipal Youth Counters, creation of, 143; services provided by CWI, 142–143; showroom experiments (UWV/CWI/local welfare agencies), 143; SUWI-reform, 142; Work Plazas, 143 marketization, 138–140: Individual Reintegration Agreement, 139; mix of in-house/externally provided activation, 140; national regulation, 138;

Tomáš Sirovátka

274

Index

no-cure no-pay or no-cure less-pay contracts, 140; purchasers, 139; radical marketization, 138 new governance regime, shift towards, 144–145: from corporatist to market/corporate governance, 144–145; UWV WERKbedrijf, 144 new public management, 140–142: incentive paradigm, 141; merged PES and benefit agency, 141–142; municipal activation, 141; Social Assistance Act in 2004, 141 overview of reforms, 133–135; 1980s, state reforms, 133 social democratic/conservative characteristics, 132 Social Insurance Bodies, 134 social partners, exit of, 135–136: Disability Benefit system, 136; Social Insurance Bodies, replacement of, 136 trends in governance reforms, 135–144: decentralization and (re-)centralization, 136–138; inter-agency cooperation, 142–144; marketization, 138–140; new public management, 140–142; social partners, exit of, 135–136 new public management, 11–13 comparative perspective, 250–252: combination of traditional regulation, 251–252; vigorous and soft version of, 250–251 Czech Republic, 181–183: ESF line/national ALMP line, 182; management-by-objectives approach in ALMP, 181–182; OMC, 183; organizational changes within PES, 182 definition/characteristics, 11–12 Finland, 219, see also LAFOS, network approach France, 49–50

Germany, 73–74: action programmes and resource allocation, 73; Clients for counselling and activation, 73; Clients for counselling and qualification, 73; cooperation of person seeking assistance, 73–74; cost-effectiveness, 73; costs of input with output, 73; Incentive assistance clients, 73; market clients, 73; service centres (Kundenzentrum), 73 incentive paradigm, concept of, 12 management by objectives: internal public contracts, 12 Netherlands, 140–142: incentive paradigm, 141; merged PES and benefit agency, 141–142; municipal activation, 141; Social Assistance Act in 2004, 141 performance indicators, 12–13 Sweden, 206–207: computerized management information system (LEAD), 206; cooperation between public and private actors, 208 Switzerland, 162–164: cantonal implementation, control of (1997 to 99), 162–163; financing and control of measures, 163; state secretary for economic affairs (SECO), 162 UK, 94, see also decentralization OMC, see Open Method of Coordination (OMC) Open Method of Coordination (OMC), 183 outcome effect comparative perspective, 256–257: decentralization, 257; impact of marketization of activation services, 256; promoting inter-agency cooperation, 256 Czech Republic: employment effects of the ESF measures, 185

Tomáš Sirovátka

Index Germany: ARGE/municipal providers, performance of, 76; long-term unemployment, comparison, 76 Netherlands, 147 Switzerland, 165–166: temporary subsidized employment, 166 UK, 100–102: increased benefit recipients, 101; longer-term employment outcomes, 101; New Deal for Lone Parents/Young People, 101; New Deal 25 Plus cohort, 101–102; sustainability of employment, 102 output effect comparative perspective, 257–260: content of services, 258–259; NPM, impact of, 258; participation of larger groups of unemployed people, 257; relation between social policy and governance reforms, 258; service de-standardization and diversification, 259; service diversification and individualization, 259 Czech Republic, 185: contents of activation, 188; Employment Act from 2004, 188; incentives to work, improvisation of, 186; logistic regression analysis, 187; National Reform Programme 2005–8, 188; scope, 185; targeting, 186–188 Germany, 77–78: incentive assistance clients, 77; PSAs, importance of, 77; quality of activation, 77; training voucher, criticism of, 78 Netherlands, 147–148 Switzerland, 165–168: classification systems, 167–168; creaming-off effect, REC, 167; diversification and individualization, 167; educational measures, 166–167 UK, 103: combination of flexibility, 103; contract for New Deal for Disabled People, 103

275

PAP, see Personalized Action Plan (PAP) PARE, see Plan to Support Return to Employment (PARE) Personalized Action Plan (PAP), 51 Personalized Project for Returning to Employment (PPAE), 51 Personnel Service Agencies (PSAs), 75, 77 Plan to Support Return to Employment (PARE), 47, 51 PPAE, see Personalized Project for Returning to Employment (PPAE) private providers for job placement (OPP), 57 process effect comparative perspective, 260: implementation conditions, 260; marketization, context of, 260 Germany: specialized services, 79 Switzerland, 168–169: contracts, institution and client, 168–169; decisive factors, jobseekers, 168 UK, 103–105: benefit of contracted-out support, 104–105; mandatory enrolment, 104; marketized programmes, evaluation, 104 PSA, see Personnel Service Agencies (PSAs) Public expenditure on labour-market policies in 2007, 19t quasi-markets, 7–10, 16, 38, 50, 56, 100, 165, 169, 243 re-centralization Germany, 71–73: centralization/standardization, purchase of labour-market, 72; contracting-out of placement services, 72; creation of regional purchasing centres, 72; Hartz reforms, 71–73; SGB III, 71–72 Netherlands, 136–138: Act on integration of young (under 27) unemployed, 137; competitive decentralization, 137; Individual Reintegration

Tomáš Sirovátka

276

Index

Agreements (IRO), 137; new benefit agency, 136; Social Assistance Act in 2004, 137; UWV, decentralization of, 137 Sweden, 204–205: monetary benefit standard, 204; municipal youth programmes, abolishment, 204; National Board for Health and Welfare in 1985, 204; PES, role of, 205 UK, 96–97: Benefits Agency, 96; Jobcentre Plus, 96–97; use of contracts, 96; work-first and work-for-all approach, 96 Reconciling Work and Welfare (RECWOWE), 4 RECWOWE, see Reconciling Work and Welfare (RECWOWE) reforms of operational policies, 2 Regional Direction of Firms, Competition, Consumption, Work and Employment (DIRECCTE), 34–35, 41, 43 Regional Employment Centres (REC), 33, 35, 156, 160–164, 167–168, 245, 247, 251 Rehabilitative Work Experience Act (Finland), 217 RGPP, see General Revision of Public Policies (RGPP) RMA, see Minimum Activity Income (RMA) showroom experiments (Netherlands), 143, 147, 256 SKOS, see Swiss Welfare Conference (SKOS) SMP, see Monthly Personalized Support (SMP) Social Assistance Act, 137, 149, 196, 217 Social Code Book (SGB), I/II, 67 Social Insurance Institution (KELA), 26, 28f, 216, 221, 223f, 224–225, 227, 233 Social Service Act (Sweden), 25, 200–201, 208, 210

state secretary for economic affairs (SECO), 162 Structure for the administrative organization for work and income (SUWI), 142 Sweden activation, features/trends, 195–197: extensive state intervention, 195; local government and local responsibility, 195; multi-level governance, 197; PES, 196 computerized management information system (LEAD), 206 decentralization, trends of, 202–203: ‘Act on Municipal Responsibility for Young People between 20–24 Years,’ 202; local employment service committees, 203; National Labour Market Administration, 203; Swedish Social Services Act in 1998, 203 governance reforms, 196t, 197t, 201–205 governance regime types and effects, 209–213: dissatisfaction with costs/market, 212; increased influence of market mechanisms, 212; interlinking of work and welfare, 210; procedural model of bureaucracy, 212–213; re-centralization measures, 213; regulation of activation policies, benefit of, 210; state regulation, strengthening of, 210–211; transfer of power, local to the national level, 211; youth programmes, 211 governance structure, 198–201 horizontal governance, 205 inter-agency cooperation, 208–209: Act on Financial Coordination of Rehabilitation Measures, 208; collaborations between welfare institutions, 208; institutionalized cooperation, desire for, 209; Social Service Act, 208; socio-medical

Tomáš Sirovátka

Index activities, 209; vocational rehabilitation, 208 labour market policy, 198–200: actors, supra-regional labour-market councils, 199–200; guidelines for PES activities, 199; Job and Development Guarantee, 200; Job Guarantee for Youth, 200; Labour Market Policy Activities Ordinance, 199; private actors, employment service market, 200; Swedish parliament, 199; Swedish PES, subdivisions, 199; unemployment funds, 199 multi-level governance, concept of, 201 new public management instruments, 206–207: computerized management information system (LEAD), 206; cooperation between public and private actors, 208 private actors, 207–208: Job and Development Guarantee, 208; Public Procurement Act, 207; Youth Guarantee, 208 (re-)centralization, trends of, 204–205: monetary benefit standard, 204; municipal youth programmes, abolishment, 204; National Board for Health and Welfare in 1985, 204; PES, role of, 205 Saltjöbaden agreement of 1938, 205 social assistance, 200–201: dual system, municipalities/PES, 200; National Board of Health and Welfare, 201; ‘Nordic model’ for the municipal system, 200; Social Service Act 2001, 200; steering mechanisms, central government, 201 trade unions, 206 unemployment and social assistance, 198: financial crisis, late 2000s, 198; labour-force participation, 198; social assistance recipients, 198

277

vertical governance, 202–205 Swedish Social Services Act in 1998, 203 Swiss Welfare Conference (SKOS), 26, 153, 156–159 Switzerland activation, income protection, 154: means-tested social assistance, 154; unemployment insurance, 154 inter-agency cooperation, 164–165: cross-institutional processes (IIZ-MAMAC), 165; inter-institutional cooperation project, 164; lack of cooperation, 164; medical employment market assessments, 164; multi-problem cases, 165 new public management, 162–164: cantonal implementation, control of (1997 to 99), 162–163; financing and control of measures, 163; state secretary for economic affairs (SECO), 162 outcome effects, 165–166: temporary subsidized employment, 166 output effects, 166–168: classification systems, 167–168; creaming-off effect, REC, 167; diversification and individualization, 167; educational measures, 166–167 process effects, 168–169: contracts, institution and client, 168–169; decisive factors, jobseekers, 168 reforms, 159–160, 159t social assistance, 156–159: cantonal and local levels, 156; as decentralized, discretionary relief system, 158; evaluation of measures, 158; implementation of, 157; incentive-based services, 157; needs-based system, 157; no legal claim for recipient, 159; organizational

Tomáš Sirovátka

278

Index

structure of, 157–158; private institutions, 158; regionalization and professionalization, 158; Schweizerische Konferenz für Sozialhilfe (SKOS), 157; tax-exempt payments, 157 unemployment insurance, 154–156: cantonal institutions, role of, 156; Equalization Fund, 155; Federal Law of 1984, measures, 155; financing, 155; national institutions, role of, 155–156; Regional employment centres (REC), 156; Swiss social insurance system, 154 unemployment insurance (1995), 160–161: decentralized system, 161; local employment offices, 161; Logistics Office of ALMP, 161; RECs, 160–161; tripartite Committee, creation of, 160 UK centralized control, 93–94: cross-party consensus, 94; Department for Work and Pensions, 93 contractualization, 98: cooperation between public/private/ voluntary sector, 98 decentralization, 94–96: area-based Employment Zones, 95; contracting-out, 94; Department for Work and Pensions, 96; ‘District Delivery Plans,’ 95; Flexible New Deal, 95–96; New Deal programmes, 94; New Deal Task Force, 95; new public management practices, 94; target groups, 95; Working Neighbourhoods Pilots, 95 incentive-based financial reward systems, 85–86 income protection and activation, implementation structure, 91t

Tomáš Sirovátka

marketization, 97–98: conservative–Liberal Democrat coalition government, 97; development of the quasi-market, 97; market-based reforms, 97; private providers/privatization, 98; profit-making multinationals, 98 ‘moving people from welfare into work,’ 85 New Deal programmes, 89 new hybrid regime, shift towards, 98–100: contracting of private providers, 99; corporate governance, forms of, 99–100; incentive paradigm, 99; market values to the procedural governance, 99 outcome effects, 100–102: increased benefit recipients, 101; longer-term employment outcomes, 101; New Deal for Lone Parents/Young People, 101; New Deal 25 Plus cohort, 101–102; sustainability of employment, 102 output effects, 103: combination of flexibility, 103; contract for New Deal for Disabled People, 103 overview, 87–91: 1990 and 2002, local offices and government departments, 88; cost cutting, 90; developments, period of, 90; DWP Commissioning Strategy, 89–90; Flexible New Deal 1, 90; Job preparation/training, Blair’s New Labour party, 89; Jobseeker’s Allowance (JSA) Act, 88–89; Labour governments (1997–2010), 89; 1990–2002, local offices/government departments, 88; mass unemployment in the 1980s, 87–88; New Deal programmes, 89; new Employment Service, 88; public employment service in 1909, 87; quasi-privatization,

Index 88; Work-first policy proposals, 91 process effects, 103–105: benefit of contracted-out support, 104–105; mandatory enrolment, 104; marketized programmes, evaluation, 104 re-centralization, 96–97: Benefits Agency, 96; Jobcentre Plus, 96–97; use of contracts, 96; work-first and work-for-all approach, 96 trends in governance reforms, 92–98 work-first type, 86–87: Child Tax Credit, 86; client-centred approaches, 86; National Childcare Strategy, 86; National Minimum Wage, 86; Working Tax Credit, 86 Unemployment Insurance Board (IAF), 26, 196t, 199 universalism, 40, 125, 223 welfare states, governance of decentralization, 6–8 governance of social policies, 2 governance regimes, 13–16 income provision and activation, 3

279

inter-agency cooperation and partnerships, 10–11 marketization and competition, 8–10 new public management, 11–13 potential effects, 16–18: advantages of quasi-markets, 16; dependent variable, 17; forms of governance, 16 public expenditure on labour-market policies in 2007, 17 RECWOWE, 4 reforms, unemployed people, 3 research questions, 4–6: effects of reforms, 5–6; nature of reforms, 4–5 Work First programmes, target groups, 148, 148t work-first type, UK, 86–87 Child Tax Credit, 86 client-centred approaches, 86 National Childcare Strategy, 86 National Minimum Wage, 86 Working Tax Credit, 86 Working Neighbourhoods Pilots (UK), 95

Tomáš Sirovátka