Rescaling Social Policies: Towards Multilevel
Transcript
Rescaling Social Policies: Towards Multilevel
Rescaling Social Policies: Towards Multilevel Governance in Europe Public Policy and Social Welfare A Series Edited by the European Centre European Centre Vienna Volume 38 Yuri Kazepov (Ed.) Rescaling Social Policies: Towards Multilevel Governance in Europe © European Centre Vienna, 2010 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise without the prior permission of the publisher. Published by Ashgate Publishing Limited Wey Court East Union Road Farnham Surrey GU9 7PT United Kingdom Ashgate Publishing Company Suite 420 101 Cherry Street Burlington, VT 05401-4405 USA Copy-editing and DTP: Willem Stamatiou European Centre for Social Welfare Policy and Research Berggasse 17, 1090 Vienna, Austria British Library Cataloguing-in-Publication Data. A catalogue record for this book is available from the British Library. ISBN 978-1-4094-1021-8 Printed by Facultas Verlags- und Buchhandels AG, Vienna, Austria To Matti Heikkilä (✝) A good friend and a brilliant researcher Table of Contents Contents List of Figures and Tables............................................................................... 11 The Rescaling Project and This Book ..........................................................15 Yuri Kazepov Setting the Scene: Rescaling and Governance ......................................33 Chapter 1 Rescaling Social Policies towards Multilevel Governance in Europe: Some Reflections on Processes at Stake and Actors Involved .................... 35 Yuri Kazepov Part I: Towards Multilevel Governance in Selected Policy Areas ........................................................73 The Changing Area of Labour Market Activation Policy..................73 Chapter 2 Activation and Rescaling: Interrelated Questions in Social Policy? .......... 75 Stefania Sabatinelli Chapter 3 A Comparative Perspective on Labour Market Changes..........................103 Eduardo Barberis / Beat Baumann Chapter 4 Actors, Scaling and Governance in Activation Policy ............................... 139 Vappu Karjalainen 7 Rescaling Social Policies: Towards Multilevel Governance in Europe The Changing Area of Social Assistance Policy ................................. 175 Chapter 5 Social Assistance Policy Models in Europe: A Comparative Perspective ...........................................................................177 Eduardo Barberis / Stefania Sabatinelli / Annegret Bieri Chapter 6 The Territorial Organisation of Social Assistance Schemes in Europe .... 203 Renate Minas / Einar Øverbye Chapter 7 Actors and Governance Arrangements in the Field of Social Assistance .................................................................... 241 Åke Bergmark / Renate Minas 8 The Changing Area of Long-Term Care for Older People ...............275 Chapter 8 Familiarisation and Regional Variation in Long-Term Care Provision. A Comparative Perspective on Long-Term Care for Older People in Europe ............................................................................277 Rahel Strohmeier Navarro Smith Chapter 9 The Territorial Dimension of Long-Term Care for Older People ............. 313 Signy Irene Vabo Chapter 10 Actors and Governance Arrangements in Long-Term Care for Older People ...........................................................343 Signy Irene Vabo Table of Contents Part II: The Challenges of Multilevel Governance Arrangements within Social Policies ...........................365 Chapter 11 Rescaling Processes in Europe: Convergence and Divergence Patterns towards Multilevel Governance? ............................. 367 Eduardo Barberis / Åke Bergmark / Renate Minas Chapter 12 The Coordination Challenge .........................................................................389 Einar Øverbye / Rahel Strohmeier Navarro Smith / Vappu Karjalainen / Jürgen Stremlow Chapter 13 Rescaling of Social Welfare Policies: Lessons and Future Perspectives .................................................................. 415 Nicolette van Gestel / Jean-Michel Herbillon Appendix......................................................................................................429 Methods and Contexts in the Study of Rescaling.......................................431 Eduardo Barberis References.................................................................................................... 471 The Authors..................................................................................................... 499 9 List of Figures and Tables List of Figures and Tables Figures Chapter 1 Figure 1: Figure 2.a: Figure 2.b: Figure 2.c: Figure 2.d: Figure 2.e: Figure 3: Chapter 5 Figure 1: Chapter 6 Figure 1: Implicit and explicit rescaling ..................................................44 Local autonomy centrally framed: Finland 1990-2005 .......... 56 Centrally framed: France 1990-2005 ........................................58 Regionally framed: Switzerland 1990-2005 ............................58 Regionally framed: Italy 1990-2005..........................................60 Mixed in transition: Poland 1990-2005 ....................................62 The role of redistribution and family ......................................64 Responsibility of public actors for different policy functions in social assistance ......................................193 Sub-national autonomy and central control 1990 and 2005 ............................................................................232 Chapter 11 Figure 1: The role of central authorities, 1990-2005 ............................ 374 Figure 2: The role of central authorities in the three scalar regime types, 1990-2005 ...............................................374 Figure 3: The role of mid-level authorities, 1990-2005 ........................377 Figure 4: The role of mid-level authorities in the three scalar regime types, 1990-2005 ...............................................377 Figure 5: The role of lower-level authorities, 1990-2005 ..................... 379 Figure 6: The role of lower-level authorities in the three scalar regime types, 1990-2005 ...............................................379 11 Rescaling Social Policies: Towards Multilevel Governance in Europe Appendix Figure 1: NUTS-3 level areas where field research has been carried out .................................................................445 Tables Chapter 1 Table 1: Table 2: Table 3: Table 4: Table 5: 12 Chapter 3 Table 1: Table 2: Table 3: Table 4: Table 5: Table 6: Table 7: Table 8: Table 9: Table 10: Table 11: Financial decentralization in selected European countries, 1974-2000 ................................................46 The welfare systems in Europe, 1990-2007 ............................. 54 Third sector actors’ involvement models in European welfare systems ................................................... 65 Dispersion rates of selected indicators at NUTS 2 level, 1999-2007 .......................................................68 Trends in Gini coefficients of income inequality in selected OECD countries, 1985-2005 ................................... 70 Labour market indicators, 1985-2005 ....................................105 Labour market indicators (labour arrangements), 1985-2005 ...................................................................................107 Access to unemployment benefits for a middle-aged redundant and unskilled factory worker (Mr. A) ................ 112 Net replacement rates – initial phase of unemployment: single-earner married couple, no children (profile Mr. A-like). Years 2005, 2006, 2007 ........................... 114 Access to benefits in case of unemployment, 2007 ............. 115 Participant stocks on labour market policies as a percentage of the labour force – Out-of-work income maintenance and support per measure (2000-2005)............ 119 Expenditure on Labour Market Policies / LMP (2005) ....... 121 Public expenditure and participant stocks on LMP per programme (2007) ............................................................. 122 Area of application, financing source and institution responsible for labour market policy training measures, EU (2003) in % ........................................................................... 124 Participation rate in education and training (2007) .............125 Summary of labour markets and labour market policies ...135 List of Figures and Tables Chapter 4 Table 1: Table 2: Table 5: Table 6: Governance types .....................................................................147 Main public actors of activation policy at state, regional, provincial and local level ............................. 150 Responsibilities for various phases of activation policy at different territorial levels .................................................... 153 Vertical coordination in activation policy from the point of view of jurisdiction....................................156 Horizontal coordination of activation policy .......................161 Synoptic view of governance in activation policy ...............170 Chapter 5 Table 1: Table 2: Type of actors and their relevance ......................................... 195 Cash benefits for vignette cases in social assistance............ 202 Table 3: Table 4: Chapter 6 Table 1: Table 2: Legislative and planning power responsibilities ................. 216 Levels of administrative power and changes over time (1990-2005) ...............................................................224 Table 3: Financial power ........................................................................231 Appendix: The dimensions of sub-national autonomy and control ..... 236 Appendix: Sub-national autonomy ........................................................... 238 Appendix: Central control ......................................................................... 239 Chapter 7 Table 1: Table 2: Table 3: Table 4: Chapter 8 Table 1: Table 2: Table 3: Table 4: Table 5: Social assistance: actors at different levels and their pivotal tasks ............................................................. 243 Degrees of targeting, fragmentation and coordination.......250 Purpose of partnerships, who is involved, structure of partnerships .........................................................256 Professional discretion – general overview ..........................266 Type of long-term care delivered to older people: in-kind versus cash benefits ....................................................285 National coverage rates for long-term care for older people: in-kind and cash benefits ................................287 Comparing cash benefits for dependent older people........289 Regional coverage rates for home care and residential care .................................................................. 294 Public expenditure on long-term care for older people (2005) .............................................................297 13 Rescaling Social Policies: Towards Multilevel Governance in Europe Table 6: Table 7: Table 8: Chapter 9 Table 1: Table 2: Table 3: Table 4: 14 Table 5: Table 6: Co-payment by users of long-term care and their relatives ..................................................................... 299 Formal entitlements to long-term care for older people ..... 303 Social rights and professional guidelines of LTC for older people ........................................................................306 Public strategies for care provision, home care and home help to older people...............................................318 Territorial levels with main administrative responsibility for the provision of long-term care services for older people, 2006 ..............................................................320 Regulatory power over long-term care services for older people at different territorial levels, 2006 .............323 Role of, and relationship between, levels of government involved in financing long-term care for older people, 2006 .............................................................................................326 The role of local government in providing long-term care for older people, 2006 ....................................330 Reforms indicating trends of (de)centralisation in long-term care for older people, period 1980-2006 ..............333 Chapter 10 Table 1: Public and non-public providers of in-kind services to older people in six european countries (approximate shares) ...............................................................349 Table 2: Regional variation in actors involved in the delivery of in-kind services to older people......................................... 354 Table 3: The share of single- and multi-purpose agencies delivering long-term care for older people .......................... 356 Table 4: User involvement in the provision of formal long-term care services for older people ..............................360 Chapter 11 Table 1: Convergence trends and patterns and the influence of central, mid- and lower-level authorities .........................380 Appendix Table 1a: Table 1b: Reference areas for field research at the different NUTS levels ....................................................448 Average size of main territorial units in chosen countries .449 The Rescaling Project and This Book The Rescaling Project and This Book Yuri Kazepov This book is the result of the “Rescaling Social Welfare Policies – A comparative study on the path towards multilevel governance in Europe” project. It was a complex comparative project aimed at investigating the changes in both the territorial organisation and the new actors’ organisation of selected social policies in several European countries. The project developed under the auspices of the European Centre for Social Welfare Policy and Research in 2005, following an initial stimulus by Matti Heikkilä to whom this book is dedicated. Matti, a good friend and a brilliant researcher felt that it was important to investigate these issues in a pioneering project. Indeed, when it started this was one of the very few projects on rescaling social policies. No consolidated knowledge base existed to be used as a reference and many methodological questions remained to be addressed even before starting. We approached the issue with quite high ambitions in terms of the policies to be considered, the countries to be involved and the methods to be used. We targeted four policies, twelve countries and multiple methods, but ended up with three policies, eight countries and fewer methods. Among the policies we were able to investigate: a) activation policies in the labour market; b) social assistance; and c) long-term care for the elderly. We unfortunately were obliged to drop migration policies. Having too many policies was not only costly, but also complex in terms of management. Among the countries that we were able to involve in the project are Finland, France, Italy, Norway, Poland, Spain, Sweden and Switzerland. Unfortunately Denmark, Germany, the Netherlands and the United Kingdom did not get funded by those countries’ respective national agencies and were not able to participate in the study. This was a pity, because the territorial dimension of social policies was becoming particularly relevant in their 15 Yuri Kazepov 16 respective political debates. We had originally also planned the use of the EUROMOD tax benefit micro-simulation model for identifying the redistributive impact of territorial reforms. Unfortunately, however, the person assigned to this part of the project left the European Centre and it proved very difficult to replace him. Despite the aforesaid resizing of the project, the countries we investigated, the policies we have considered and the methods used are relevant and interesting in relation to our understanding of the territorial dimension of social policies and of the new governance arrangements. Some additional complications arose, however. The project partners did not know each other before starting the project and the funding was distributed unevenly among them. In fact, the funding mechanism of the project (see below) foresaw that each country had to finance its own part of research. This naturally led to different levels of funding for the different partners, who were first and foremost responsible towards their fund-givers. These challenges contributed to the complexity of the project and required several meetings not only to define the conceptual and methodological frame, but also to ease collaboration among the numerous partners. During our meetings we developed several working hypotheses to frame our interpretation of the ongoing processes. A first hypothesis assumed that different institutional settings diversely frame the way in which rescaling and governance processes are deployed in different countries. This implies that there must be some degree of coherence between the ongoing processes, their direction of change and the organisation of the State and the welfare system/regime these countries belong to. This relation also holds consequences for the interpretation of possible trends of convergence and/ or divergence. Almost all the chapters herein review these aspects against the specificities of the single policy fields. A second hypothesis assumed that the time frame during which specific processes take place influences the way they act. For instance decentralising regulative capacities in a period of economic expansion, when available resources are relatively high, has a completely different meaning than decentralising regulative capacities in times of economic penury and in the presence of neoliberalisation trends. The importance of contextualising trends emerges as a crucial aspect to be considered in interpreting the current changes. The need to adopt a variety of both quantitative and qualitative methods to disentangle the complexity of the processes at stake has also become evident. The analysis of the results stemming from this project was also particularly complex, because we did not opt for country-specific chapters. The Rescaling Project and This Book Rather, we went the more difficult path (but also the more interesting one) of issue-specific chapters, examining these comparatively. To strengthen this aspect, most of the chapters are written by colleagues from different countries and different fields of specialisation, allowing the cross-fertilisation of disciplinary knowledge. As the reader can easily imagine, this process has not been easy. The structure of the book The book is organised into an introduction and two main parts. The introductory chapter presents the main issues at stake and proposes a frame within which to understand the process of rescaling and the path towards multilevel governance arrangements. Some results are also alluded to ahead of time, in more general terms, and a typology of the territorial organisation of social policies in the different countries is presented; in particular this is carried out concerning the changes the different welfare systems are undergoing. One of the main points made is that the joint effect of the processes of territorial re-organisation and the multiplication of actors in policy design, management and implementation is an overall process of subsidiarisation. This process has an ambivalent nature given that it is – as we will see – legitimised by opposing ideologies (i.e. neoliberalism and participatory democracy and grass-roots empowerment). It is also a process characterised by converging rhetorical debates and diverging impacts. The chapter provides some contextual elements to address these differences in social Europe, paving the way for a more in-depth analysis of the policy-area-based chapters of the first part of this book. In the first part of the book the three specific policy areas and – respectively – each individual policy area are addressed in three distinctive chapters. The first provides an overview of the literature in the specific policy fields, which theories are used to understand and interpret them and how they have changed. The second chapter addresses the territorial dimension and how it has changed over the last two-three decades, while the third chapter investigates how the policy-specific actors’ configurations have changed over the same period of time. This structure is more or less replicated for all three policies. Activation policies are addressed in Chapters Two, Three and Four. In the second chapter authored by Stefania Sabatinelli, the literature on activation policies is reviewed and the spatial dimension has been included 17 Yuri Kazepov 18 in the analysis. In particular, the chapter focuses on the implications of the activation paradigm in social assistance and labour market policies, aiming at highlighting some of the main driving forces. The chapter investigates the different meanings of the concept and the narratives developed to accompany its diffusion. From this point of view, activation is a policy area in which the ambivalence of the processes of change is particularly clear: it entails both neoliberal characteristics as well as innovative and empowering policy solutions. The third chapter by Barberis and Baumann compares the effects of scalar configuration changes on labour market policies, providing contextual data and considering the implications of the aforesaid for users. What emerges is clearly that multilevel governance and scale trends provide coordination problems that weaken the effectiveness of active labour market policies. The equity of treatment during activation is also made an open question in all countries, but in particular in countries with complex patterns of multilevel governance and in those undergoing the process of changing their state configuration from a centralist typology to a more or less decentralised one. The fourth chapter by Vappu Karjalainen completes the analysis of activation policies, more specifically taking into consideration the territorial dimension and the emerging governance arrangements. The working hypothesis of the chapter is that activation policies question the governance and distribution of responsibility between territorial levels, due to the contradiction between the macro-nature of the structural causes of unemployment and the local and individual nature of activation services. The governance of activation policy challenges the traditional sector-based administration and service structure of welfare society with bottom-up approaches. The aim is to catch the overall developmental state of activation policy governance in the countries considered in the present study. Social assistance policies are addressed in Chapters Five, Six and Seven. In Chapter Five authors Barberis, Sabatinelli and Bieri provide a general view of the institutional settings within which social assistance policies and the respective welfare systems are embedded. They define the actors’ degree of freedom, which also influences the way in which the interaction and coordination in the policy process takes place at the vertical and horizontal levels. In Chapter Six Minas and Øverbye analyse how, over the last decades, the reorganisation of social assistance policies has entailed shifts of respon- The Rescaling Project and This Book sibility between administrative levels. They address the consequences of the territorial dispersion of authority upwards, downwards or sideways in terms of the search for an optimal organisation of welfare policies. A major trend identified in the chapter is towards the decentralisation of authority. However, the authors show that – rather than representing a single rescaling strategy – decentralisation is a wide umbrella term for diverse and contextualised political, administrative and financial processes which are investigated in different sections of the chapter. The relevance of timing is also highlighted as a crucial perspective to understand current dynamics, i.e. the processes are more or less the same, but different contexts and different timing result in differentiated impacts. To this, the authors add the portrayal of the ambivalent character of the policies stemming out of the tension existing between empowering solutions and new public managementlike solutions which permeate the way in which reforms are legitimised. In Chapter Seven Bergmark and Minas provide an in-depth account of the wide array of actors involved in governing social assistance. They begin with a structured inventory of who is involved at what level and – in that respect – try to capture the relevant political-administrative span with respect to the vertical as well as horizontal division of authority and responsibilities therein. From that they proceed to describe how measures are coordinated between different actors and how that relates to the degree of targeting and fragmentation in the different countries involved in the project. The idea is to identify the extent and character of significant networks and their contextual settings. A particularly relevant section of the chapter addresses the fact that managerial structures and delivery systems are, to a considerable extent, related through the scope of actors and their relations. In order to capture this aspect in a more comprehensive manner the nature of assessments and professional discretion is analysed in-depth, with a special focus on different aspects of street-level decision-making. The third and last policy we investigated is long-term care for older people (LTC), which is addressed in Chapters Eight, Nine and Ten. In Chapter Eight Rahel Strohmeier Navarro Smith provides an overview of trends, driving forces and differentiated impacts of the process of subsidiarisation on the delivery of LTC policies. First, she investigates the changing role of families in general and of women in particular in relation to the different welfare system configurations and the potential “re-familialisation” or “de-familialisation” of long-term care. Second, she analyses the regional variation in long-term care provision. The kind and amount of services and 19 Yuri Kazepov 20 benefits publicly given to dependent older people differ according to the institutionalised territorial frames designed for long-term care provision. The chapter investigates the extent and type of variation, providing an outlook on possible future developments. In Chapters Nine and Ten Signy Vabo addresses, respectively, the territorial dimensions of long-term care policies for the elderly and the emergence of new governance arrangements. In Chapter Nine she clearly shows the complexity of the rescaling processes, which not only entail decentralisation, but also possible re-centralisation trends, according to the countries’ specific conditions. Regulation, management and finances are addressed and analysed as relevant dimensions of these processes, highlighting some critical points: the importance of health policies for specific targets, and the blurring boundaries between formal and informal care. Both assume different meanings in the different institutional arrangements characterising the countries we investigated. Chapter Ten analyses the increasing number of actors from different sectors of society in the field of long-term care policies, combining it with scalar issues. The author also highlights how, in long-term care policies, hierarchy is decreasing in relevance. Meanwhile, market-based and self-organising modes of governance are increasing. The chapter analyses how this takes place considering three different perspectives: the mix of actors, institutional fragmentation and the role of the user. The second part of the book addresses two transversal issues that are crucial in understanding the processes of subsidiarisation and their impacts: a) the convergence-divergence debate; b) the coordination issue. In Chapter Eleven Barberis, Bergmark and Minas address the first of these issues by disentangling the complexities of the convergence-divergence debate. The authors investigate the different phases of convergence studies and the fact that the issue has been approached by two intimately related research concerns: i) the reasons for which convergence has, or has not, occurred, and ii) what exactly has converged. Indeed convergence is a complex and multifaceted phenomenon and general convergence trends are not easy to establish. Our empirical findings tie in with a growing body of data which challenges the idea of increasing similarity among welfare states. Many welfare states have, in fact, proven to be comparatively resilient to external driving forces. The authors make a nuanced analysis of the ongoing processes of change and identify a converging trend in the increased relevance of sub-national authorities and diverging trends as far as the relevance of The Rescaling Project and This Book the nation-state is concerned. In some countries its relevance has declined substantially, while in other countries it has remained stable. Finally, in others it has re-gained relevance. In Chapter Twelve Øverbye, Strohmeier Navarro Smith, Karjalainen and Stremlow address the ways in which the coordination challenge is handled across countries and across policy areas. Indeed the subsidiarisation of social policies brought about a fragmentation of levels and actors involved in policy design, management and implementation. The chapter describes the whole range of possible options adopted to address this problem, highlighting how they are taken up in the different countries to counteract the negative effects of fragmentation. Are these different strategies tied to different welfare systems? Building on previous analyses, the authors look at the situation in the small Nordic countries, contrasted to selected examples from the more complex institutional structures found in Continental Europe. Their analysis confirms the fact that the implementation of converging trends – including coordination measures – has varied depending on the institutional structure of the country’s welfare systems. The last chapter of the book by Van Gestel and Herbillon can be considered a conclusion of the book as it examines the lessons we gained for further developing the theory and practices of the rescaling process of social welfare policies. What is the added value of this study in terms of exploration, evaluation and understanding of macro and micro policies and effects? And how can we describe its specific and original contributions? In answering these questions the chapter returns to the initial research hypotheses and perspectives presented in the first chapter, in order to identify and discuss the issues that lie at the heart of this study: the emergence of new modes of territorial governance in social welfare policies; their implementation and associated risks. Drawing upon the findings of the Rescaling project, the authors discuss the prospects for territorial governance in social welfare policies in two ways. First, they qualify the main trends in the territorial structures of governance at multiple levels as shaped by regulations from different (political) centres of command and as a process of co-production between various actors possessing different levels of status and authority. Second, they discuss the differences between the three policy areas related to their aims and background in the development of social welfare policies. They also reflect on the methodology that has been used for this pilot research project, developing several suggestions for further research and practice in social welfare policies. 21 Yuri Kazepov A substantive appendix by Eduardo Barberis complements the book by providing a short overview on the triangulation method we used in the project. This is not only embedded in the current literature, but also discusses the implications of the use of these methodologies, and in particular of the vignettes, which provided an extremely interesting empirical basis. Thanks and acknowledgments 22 Such a complex project could have never been carried out without the help of many people. First I would like to thank Kai Leichsenring and Mercedes Gonzalez-Quijano from the European Centre for Social Welfare Policy and Research. Kai not only provided efficient and smooth project management but also relevant input throughout the whole project as well as to some of the chapters in their early stages. Mercedes was the organisational backup person, who organised meetings not only with work-related activities but also with important social events. Bernd Marin, the Executive Director of the Centre, not only allowed all this to take place, but also stimulated and supported the whole process of the production of this book. Willem Stamatiou provided the proofreading and typesetting, patiently following my (almost never-ending) requests. Peter Melvyn, also from the European Centre, provided a first comparative draft at the beginning of the project, for which he should be gratefully mentioned. In order to back up my editing work with policy-specific knowledge, I involved colleagues who peer-reviewed each contribution, providing useful comments and suggestions. The review process took place both during the closing seminar of the project held in Lucerne in November 2008 and later on the drafts of the chapters. David Benassi, Rik van Berkel, Giuliano Bonoli, Vando Borghi, Walter Hanesch, Enzo Mingione, August Österle, Joakim Palme, Emmanuele Pavolini, Costanzo Ranci and Barbara da Roit should be gratefully mentioned for their critical comments and help. Tine Rostgaard and CamillaThorgaard should be thanked for providing important insight into the Danish case. Unfortunately the Danish National Centre for Social Research did not get funding, and thus they were only able to provide the WP1, which is an interesting and important background document for Denmark. Marta Plana Serrahima should also be mentioned for her support in integrating the Spanish WP1. The Rescaling Project and This Book Eduardo Barberis helped me throughout the whole project in whatever activity was needed. His flexibility and excellence are a rare quality for which I am very thankful to him. The final quality of the text and the substantial improvements in the language and style herein are due to Catherine Lea Farwell, who did a terrific job. Finally, I would like to thank the partners of the project very much for being so patient with me during the whole period during which we worked together. I am aware that sometimes I might have been tedious and insisting, but I hope it was worth working together. Not all project partners authored a chapter in this book, but they should all be gratefully named, as they all contributed to providing the empirical basis of this book (see list below). The authors of chapters, on the contrary, had the opportunity of receiving my long and never-ending comments on their contributions, so I thank them very much for considering them when revising their respective chapters more than once. They should also be thanked for their patience in waiting for the final product. Last but not least, I owe much to my family, to Simona, Alexander and Lara, who was born and came back to life after a difficult delivery. Without them the book probably would have been published much earlier, but my life would have surely been deprived of the most important experience I have ever had. Urbino, 20 April, 2010 23 Yuri Kazepov Members of the research teams European Centre for Social Welfare Policy and Research Kai Leichsenring (project manager) Mercedes Gonzalez-Quijano (research assistant) Asghar Zaidi (project associate) Manfred Huber (project associate) Bernd Marin (project associate) Erna Riemer (finance and administration) Willem Stamatiou (publications officer) Finnish team 24 Matti Heikkilä (coordination) Vappu Karjalainen (coordination and fieldwork) Taina Rintala (fieldwork) French team Jean-Michel Herbillon (coordination and fieldwork) Agnes Gramain (coordination and fieldwork) Samuel Neuberg (fieldwork) Albane Exertier (fieldwork) Nicolette van Gestel (external expert) Italian team Yuri Kazepov (coordination) Eduardo Barberis (fieldwork and support) Barbara Da Roit (fieldwork) Tatiana Saruis (fieldwork) Marco Arlotti (fieldwork) Matteo Villa (fieldwork) Stefania Sabatinelli (fieldwork) The Rescaling Project and This Book Norwegian team Einar Øverbye (coordination) Signy Irene Vabo (fieldwork) Polish team Joanna Starega-Piasek (responsible) Irena Woycicka (coordination) Marianna Zielenska (fieldwork) Karolina Sztanderska (fieldwork) Spanish team Margarita Bravo Torres (coordination) Maria Luz Cid Ruiz (coordination) Javier del Castillo Pintado (fieldwork) Carmen Diaz (fieldwork) Maria Eugenia Bolaños López (fieldwork) Swedish team Åke Bergmark (coordination) Renate Minas (fieldwork) Dietmar Rauch (fieldwork) Swiss team Beat Baumann (coordination and fieldwork) Jürgen Stremlow (co-coordination) Annegret Bieri (fieldwork) Rahel Strohmeier Navarro Smith (fieldwork) Mathieu Lexton (fieldwork) 25 Yuri Kazepov Sources of funding The project Rescaling Social Policies towards Multilevel Governance followed an experimental funding system within the framework of the European Centre’s governing bodies, representing the governments of the UN-European region. Following an initial outline of the project proposal, seven Member States expressed their interest in participating by funding a national research team and the coordination of the European Centre. Thus each participating country funded its national part of research activities. In addition, the European Centre decided to fund the project’s scientific coordinator, Yuri Kazepov, and thus also the Italian case-study. We gratefully acknowledge the following financial support: 26 Finland STAKES, The National Institute for Welfare and Health, which on 1 January 2009 became the National Institute for Health and Welfare (THL) following the merger of STAKES with the National Public Health Institute (KTL) France Ministère du travail, de la solidarité et de la fonction publique, DARES – Direction de l’Animation de la Recherche et des Études Statistiques (Ministry of Labour, Solidarity and Public Service, DARES) Italy European Centre for Social Welfare Policy and Research, Vienna (Austria) Norway Forsknigsrådet (The Research Council of Norway) Poland Ministerstwo Polityki Społecznej (Ministry of Social Policy), now Ministry of Labour and Social Policy Instytut Rozwoju Służb Społecznych (Institute for Development of Social Services) Instytut Badań nad Gospodarką Rynkową (Gdansk Institute for Market Economics) The Rescaling Project and This Book Spain IMSERSO – Institute for Older Persons and Social Services, International Department Switzerland The Federal Social Insurance Office (FSIO) DORE (DO REsearch) of the Swiss National Science Foundations (SNSF) Lucerne University of Applied Sciences and Arts Sweden Institutet för Framtidsstudier (The Institute for Futures Studies), Stockholm, Sweden 27 Yuri Kazepov Acronyms and abbreviations AFPA ALMP ANPE APEC API APW APA ASSEDIC 28 (Fr) Association nationale pour la formation professionnelle des adultes (National Association for Adult Vocational Training) Active Labour Market Policies (Fr) Agence Nationale pour l’Emploi (National Employment Agency) (Fr) Association pour l’Emploi des Cadres (Job-centre for Executive Managers) (Fr) Allocation Parent Isolé (Allowance for Single Parents) Average Production Worker (Fr) Allocation Personnalisée d’Autonomie (Individualised Allowance for Autonomy) (Fr) Association pour l’emploi dans l’industrie et le commerce (Association for Employment in the Industry and Trade) BSV (FSIO) Bundesamt für Sozialversicherung (Federal Social Insurance Office) CCAS CIS CLIC CMS CNAM CNSA CPI DDASS DIS DRASS (Fr) Centre Communal d’Action Sociale (Municipal Service for Social Action) (Pl) Centrum Integracji Społecznej (Centre for Social Integration) (Fr) Centre Locaux d’Information et de Coordination (Community Information and Coordination Units) (Ch) Centre médico-social (Center for Social and Health Care) (Fr) Conservatoire National des Arts et Metiers (National Academy for Craft and Professions) (Fr) Caisse Nationale de solidarité pour l’autonomie (National Solidarity for Autonomy Fund) (It) Centri per l’impiego (Job-centres) (Fr) Direction Départementale Affaires Sanitaires et Social (Department Directorate for Health and Social Affairs) (Be) Droit à l’Insertion Sociale (Right to Social Insertion) (Fr) Direction Régionale des Affaires Sanitaires et Sociales (Regional Directorate for Health and Social Affairs) The Rescaling Project and This Book EMS ESF EES EU FASILD (Ch) Établissements medico-sociaux (residential care; Older People’s Home) (Eu) European Social Fund (Eu) European Employment Strategy (Eu) European Union FSIO (Fr) Fonds d’aide et de soutien pour l’intégration et la lutte contre les Discriminations (Fund for the Integration of Migrants and the Fight against Discriminations) (It) Fondo Nazionale Politiche Sociali (National Fund for Social Policies) (Ch) Federal Social Insurance Office (see BVS) GDP Gross Domestic Product IAE (Fr) Instituts d’Administration des Entreprises (Institute for Business Administration) (Ch) Interinstitutionelle Zusammenarbeit (Inter-institutional Cooperation) International Labour Organisation International Monetary Fund (Es) Instituto Nacional de Empleo (National Labour Institute) (It) Istituto Nazionale per la Previdenza Sociale (national administration of pension funds and monetary social assistance measures) FNPS IIZ ILO IMF INEM INPS KELA KUTU (FI) Kansaneläkelaitos (National Social Insurance Institution) (FI) Laki kuntouttavasta työtoiminnasta (Act on Rehabilitating Work Experience) LAFOS LMP LTC (FI) Työvoiman palvelukeskus (Labour Force Service Centre) Labour Market Policies Long-Term Care NAP NAV (Eu) National Action Plan (No) NAV Arbeids- og velferdsetaten (The Norwegian Labour and Welfare Administration) Non-Governmental Organisation NGO 29 Yuri Kazepov NPM NUTS New Public Management (Eu) Nomenclature of Territorial Units for Statistics OECD OMC OPS Organisation for Economic Co-operation and Development (Eu) Open Method of Coordination (Pl) Ośrodek pomocy społecznej (Local Social Assistance Centre) MOPS (Pl) Miejski Ośrodek POmocy Społęcznej (Municipal Social Assistance Office) PCPR (Pl) Powiatowe centrum pomocy rodzinie (Family Assistance Centre) Public Employment Services (Fr) Plan locaux pluriannuels pour l’insertion et l’emploi (Local pluri-annual plans for social insertion and employment) Purchasing Power Parities (Fr) Prestation de solidarité dépendance (Dependency Benefit) (Pl) Powiatowy urząd pracy (Local Labour Office) (Pl) Polska Zjednoczona Partia Robotnicza (Polish United Workers’ Party) PES PLIE 30 PPP PSD PUP PUWP RMG RMI RMR ROPS RSA SKOS SPE SPEN SPER (Pt) Rendimento Minimo Garantido (Guaranteed Minimum Income) (Fr) Revenue Minimum d’Insertion (Minimum Income for Insertion) (Ch) Revenue Minimum de Réinsertion (Minimum Income for Re-insertion) (Pl) Regionalny ośrodek polityki społecznej (Regional Social Policy Centre) (Fr) Revenue de Solidarité Active (Active Solidarity Income) (Ch) Schweizerische Konferenz für Sozialhilfe (Swiss Conference for Social Assistance) (Fr) Service Public de l’Emploi (Public Employment Services) (Fr) Service Public de l’Emploi National (National Public Employment Services) (Fr) Service Public de l’Emploi Regional (Regional Public Employment Services) The Rescaling Project and This Book SPED SPEL SVOAM (Fr) Service Public de l’Emploi Departmental (Departmental Public Employment Services) (Fr) Service Public de l’Emploi Local (Local Public Employment Services) (Ch) Schweizerischer Verband der Organisationen für Arbeitsmarktmassnahmen (Swiss Association of Employment Facilication Organisations) UNEDIC (Fr) Union Nationale pour l’Emploi dans l’Industrie et le Commerce (National Union for Employment and Development in Industry and Commerce) VALTAVA (Sf) Sosiaali- ja terveydenhuollon valtionosuuslaki (Act on Planning and Government Grants for Social Welfare and Health Care) WG WP1 Whole-of-Government Work package 1. On Rescaling Social Welfare Policies (in the single countries) Work package 2. Vignettes (on the three policies considered in the project) World Trade Organisation WP2 WTO 31 Rescaling Social Policies towards Multilevel Governance in Europe Setting the Scene: Rescaling and Governance 33 Yuri Kazepov 34 Rescaling Social Policies towards Multilevel Governance in Europe Chapter 1 Rescaling Social Policies towards Multilevel Governance in Europe: Some Reflections on Processes at Stake and Actors Involved Yuri Kazepov Introduction The territorial dimension of social policies has long been a neglected perspective in comparative social policy analysis. Scholars took for granted that social policies were national policies and almost all comparative work which has been done after World War II based comparisons on national data.1 At the same time, the nation-state strengthened its identity and legitimation thanks to the redistributive capacity of welfare policies. Since the end of the 1970s, the deep structural changes that have occurred kicked-off processes of territorial re-organisation of social policies. The structural changes – mainly socio-demographic and socio-economic – are well known and have been investigated in all industrialised countries. They include the ageing of the population, increasing instability of family arrangements, decreasing fertility rates coupled with a deep economic restructuring process, long-term unemployment, the increased participation of women in the labour market, and the spread of services and new technologies which have fostered new and more flexible modes of production identified as the mainstream solution to economic problems. 1 We are, of course, aware of the fact that comparing sub-national levels has always been complicated. A first crucial step for sub-national comparative work has been the introduction of the Nomenclature of Territorial Units for Statistics (NUTS) system by Eurostat almost 30 years ago. The first regionalised statistics were produced accordingly in the second half of the 1970s and solely covered the main demographic indicators. A complex process of negotiating the definition of indicators started in order to provide a single uniform breakdown of territorial units for the production of regional statistics Europe-wide. International agencies (most prominently Eurostat and the OECD) did not systematically organise institutional data in order to consider other territorial levels except the national one up to the 1990s. The NUTS classification has been used in community legislation since 1988, but only after 2003 was an ad hoc regulation approved (1059/03). 35 Yuri Kazepov 36 These changes undermined the functioning of welfare institutions and the effectiveness of social policies by affecting how social risks were produced and the ways in which resources fuelled social policies (Castel, 1997; Taylor-Gooby, 2005; Bonoli, 2006). Postwar policies had developed to address specific risks in a context of economic growth, increasing available resources and relatively stable needs, foreseeing a virtuous positive-sumgame in which redistributive conflicts were kept under control through the availability of more resources. The breaking down of this “system”, often labelled as Keynesian Fordism2, brought about new conflicts and needs (from unemployment to care) which accelerated the emergence of a deep fiscal crisis of the state (O’Connor, 1971), unable to find a new balance between available resources and needs in a context calling for structural reforms (Ferrera, 1998; Armingeon and Bonoli, 2006). The need for structural institutional changes has been met by an intense reform activity which covered most policy areas in the last two decades. The emerging picture is complex and not always easy to disentangle. Reforming welfare states in a context of budgetary constraints is a tricky question and a challenging task. In particular, this is due to the difficulties of intervening in established constituencies and interests, as well as to the path-dependent tendency of expenditure (Boeri, 2005) and access criteria. All of the latter resist the new power balances. In this scenario most reforms have addressed both the territorial dimension of social policies and the actors involved in their design, management, implementation and funding. I have called the combined effect of these two reform processes the subsidiarisation of social policies (Kazepov, 2008). The concept of subsidiarity3 captures well the two processes of change because it addresses both the vertical – i.e. the territorial reorganisation of 2 3 Régulation theory (Boyer and Saillard, 2002) combined the historical analysis of Fordism (Gramsci, 1950) with the analysis of the conditions which allow the permanence and the transformation of the capitalist mode of production. From this point of view Keynesianism played an important and stabilising role. Further developments are provided in Jessop (2002, 2008). The origin of the concept of subsidiarity dates back to the Political Methodice Digesta of Johannes Althusius (1614) who sought protection for the religious associations and the city itself against central power. Pursue of individual liberty and protection from tyranny was also the concern of the American Federalists (e.g. Madison, 1787 quoted in Føllesdal, 1998: 204). The principle of subsidiarity is also connected to the catholic social doctrine which, at the end of the 19th century (with the encyclical Rerum Novarum, 1891 by Pope Leo XIII), attempted to define a middle course between the excesses of laissez-fair capitalism on one hand and the various forms of totalitarianism (Fascism, Communism, etc.), which subordinate the individual to the state, on the other (Waschkuhn, 1995; Føllesdal, 1998). Rescaling Social Policies towards Multilevel Governance in Europe regulatory powers – and the horizontal dimension – i.e. the multiplication of actors – of social policies, pointing to increasingly complex multilevel governance4 solutions to social policy reform needs. The definition of subsidiarity implies that matters ought to be handled by the smallest (or lowest) competent authority, meaning that a central authority should perform only those tasks which cannot be performed effectively at a lower level (Waschkuhn, 1995). This does not necessarily mean that the local level is the best solution for all problems. However, as we shall see, the joint effect of two ongoing processes predominantly has brought about a decentralisation of regulatory powers and an increased role of nongovernmental actors. The relationship between territorial rescaling and the spread of new governance arrangements also becomes evident in the involvement of civil society in the policy-making process, justified with the need of “getting closer” to the citizen (Dahl, 1998; Hirst, 1994; Fung, 2003; Powell, 2007). The concept of subsidiarity has gained momentum since the end of the last century and has now become a rhetorical keystone behind most reforms in Europe.5 Different actors at different scales interact and bargain without necessarily a clearly settled hierarchy (Bache and Flinders, 2004; Le Galès, 2002). The ideological background of the concept and the multitude of interests it serves are usually under-investigated, despite the fact that ideologies are used to interpret and frame the processes themselves (Van Kersbergen and Van Waarden, 2004). The claim to flexibilise policies through market-oriented reforms and reforms targeted to enhance the employability of claimants (Serrano Pascual, 2007) has strongly influenced the direction of social policy change in most European countries. This neoliberalisation process, however, has not been uniform and has varied in time and space (Jessop, 2002; Brenner, 2004). In the field of social policies a different rhetoric coexists, tending to combine the utilisation of market strategies and competitiveness with more innovative forms of solidarity and social cohesion. 4 5 Multilevel governance is a term used to grasp the complexities emerging in the interplay between the two processes of change we are analysing. The term was first used by Marks (1992) to capture developments in EU structural policy following its major reform in 1988. Subsequently, Marks and others developed the concept of multilevel governance to apply it more broadly to the EU decision-making process (Marks et al., 1996). However, soon it became the analysis framework for all kinds of regional issues (Loughlin, 2007: 385). In the European Union, the principle of subsidiarity was established in the Treaty of Maastricht (1992), and is part of the proposed new Treaty establishing a Constitution for Europe (Title 3, Art. I, 11). However, at the local level it was already a key element of the European Charter of Local Self-Government, an instrument of the Council of Europe promulgated in 1985 (see Article 4, Paragraph 3 of the Charter). 37 Yuri Kazepov 38 In this scenario, scholars started to reconsider the territorial dimension in their analytical frameworks taking it no longer for granted, but considering it a privileged perspective from which to understand the ambivalent character of the processes of social policy re-organisation and reform (McEwen and Moreno, 2004; Ferrera, 2005; Zürn and Leibfried, 2005; Clarke, 2005). The spread of the subsidiarity principle as a legitimate basis for both increased effectiveness on one hand and stakeholders’ empowerment on the other influenced the reform process. The aforesaid provided ground for an increasingly complex picture. The aim of this book is to explore this complexity from the point of view of social policies, attempting to highlight their specificity in the rescaling process considering the multiplication of actors as well. In this introductory chapter we will provide some contextual elements to help address the differences existing within social Europe and pave the way for a more in-depth analysis of the policy area-based chapters of the first part of this book. The specific policy areas we selected, namely activation within the labour market, social assistance and long-term care policies for the elderly, are proxies of the welfare systems that will allow us to explore both the existing diversity in social policy arrangements as well as the similarities in the challenges welfare systems have to face. This does not prevent us from starting from the welfare systems/regimes debate to investigate the single policy areas and to see whether or not the degree of intra-system coherence might be challenged for specific policy fields; in particular, considering the rescaling process and the spread of new governance arrangements, two specific perspectives are taken up in the policy area chapters. The policy areas we will consider in this book underwent important reforms from both these perspectives and can be considered particularly promising lenses through which to analyse the processes at stake because they: a) cover claimants and beneficiaries with different positions in the life cycle; b) present a different degree of institutionalisation6 in the overall welfare state organisation and c) present different institutional arrangements and degrees of decentralisation and privatisation. In order to frame this overall objective of the book and considering both dimensions of change, this introductory chapter I starts by relating social policy analysis and the production of scale (§ 1), adopting a middle-range level of theorisation (Merton, 1968: 39-71), i.e. using a theoretically informed 6 Social assistance is the oldest form of institutionalised protection which underwent several paradigm shifts in the last centuries (Alber, 1982), while long-term care for the elderly is the most recent one. Rescaling Social Policies towards Multilevel Governance in Europe pragmatic orientation as the best option to consider the complexities that social policy analysis entails. This implies that I avoid the use of all-encompassing grand theories to explain the changes and present a synthesis of the results of our theoretically informed empirical investigation. Subsequently attention will be devoted to the different forms the rescaling process can take (§ 2) and the multiplication of actors (§ 3). Despite the commonalities these two processes share in all countries, differences between welfare systems in Europe influence the directions these changes may take. In § 4 the different spatial and actor configurations are shortly presented together with the main features of the welfare systems taken into consideration in this book. The nearly prosaic working hypothesis is that the subsidiarisation of social policies is a converging normative rhetoric in most European countries’ social policy reforms, but in most cases not necessarily a convergent practice. Variations in the implementation of this rhetoric into substantive rules occur according to the specificities of the respective institutional frames at national and/or sub-national levels. This assumption – which will be reviewed in depth in the final chapters of the book, in particular in Chapter 11 – will be briefly addressed (§ 5) considering the pros and cons of the processes at stake and the main critical dimensions that emerge as open questions. 1 Social policies and the production of scale In general terms, scale can be defined as “the result of marking territories (...) through boundaries and enclosures, documents and rules, enforcing agents and their authoritative resources” (Marston et al., 2005: 420). It might also be defined as “the nested hierarchy of bounded spaces of different size” (Delaney and Leitner, 1997: 93).7 There is now some consensus that scale is not a fixed category anymore, but rather a socially constructed dimension (Marston, 2000). Conceiving state territoriality as a static background for regulatory processes and running into what Agnew (1994) termed the territorial trap, is not possible anymore: scales are changing in the cultural, economic, political and social dimensions and of these processes evidence 7 The existence of a nested hierarchy does not prevent the existence of multiple scales and hierarchies that might also be mutually disconnected and sometimes follow desynchronised trends (Jessop, 2005: 227). Within social policies this is more common than one might expect, in particular within highly fragmented and categorically structured welfare states like some continental or South-European ones. The de-synchronised reform trend existing between labour market and social assistance policies in Italy is a striking example (Kazepov, 2008). 39 Yuri Kazepov 40 is widely acknowledged. Despite this consensus, however, there is quite a debate on how this construction takes place, which actors, groups and powers are involved and what dynamic is developed out of their interaction. The political economy of scale gave rise to a relevant but contested debate in human and economic geography and other related disciplines sensitive towards “spatial” views of social phenomena.8 This debate has been going on for some decades now; in particular this has been so since the spatial arrangements of economic activities started to change substantially from the 1970s onwards through globalisation and localisation as the result of capitalist forces reshaping their scalar relations. Political-economy explanations argue that rescaling has in general to do with the process of creating conditions that are more favourable to capital accumulation (Somerville, 2004). In particular, it is seen as the attempt to find a new territorial fix to the development of capitalism (Peck and Tickell, 1994; Brenner, 2004), i.e. new scales at which regulatory settings favour the development of a neoliberal frame for economic activities. From this perspective, Cox (2002: 342) sees the main driving force of rescaling in the explicit strategy “of outflanking opposition and resistance (organised mainly at the national level)” to capital accumulation (Somerville, 2004: 142). The debate developed further and the territorial arrangements seem no longer as fixed as they were considered before, because they are increasingly embedded into strategic relational networks (Jessop, 2008) which, in a relatively flexible way, change-over-time both the territorial dimension of regulation and the actors involved in the negotiations concerning the boundaries of those regulations. Despite these developments, the rescaling process remains functional to capitalist accumulation. What happens when we use the concept of scale in addressing social policies? Social policies indeed play an important role in the construction of scale. This has been true since the very beginning when they fuelled the construction of the nation-state, its legitimacy and identity. Social policies structure redistributive communities as the result of the states’ capacity, 8 The literature on this topic is growing fast. See for instance: Swyngedouw (1997, 2005); Brenner (2004, 2009); Somerville (2004); Collinge (2006). For a critical view of scale see Marston et al. (2005) who pursue a flat ontology. For the relation between rescaling and governance see the special issue edited by Gualini (2006). More recently Lobao, Martin and Rodrìguez-Pose edited a special issue of the “Cambridge Journal of Regions, Economy and Society” on rescaling (2009) and Keil and Mahon edited the volume Leviathan Undone? Towards a Political Economy of Scale (2009). Both these publications provide the state-of-the-art on the rescaling processes and also start to engage with the rescaling processes of social policies, even though not as their main focus of analysis. Rescaling Social Policies towards Multilevel Governance in Europe ability and power to define regulative frames, allocating duties and rights and redistributing resources. A scalar perspective on social policy analysis should consider the scale of social policies as the peculiar extension or characteristic of regulative capacity, i.e. as the jurisdiction within which “the rule of law” can exercise its power9 in relation to redistributive issues and services provision. This meso-level institutional approach considers the ability of the state to define and change the boundaries within which to impose its rules and regulations. From this perspective, the welfare state can be seen as the last step of the long-term historical development through which territorially bounded political communities came to introduce redistributive arrangements for their citizens (Ferrera, 2005). The issue at stake in this volume is that the territorial bond of political communities is changing scale, shifting – presumably – also the redistributive capacities states have in different directions as well. In particular, this is true towards to sub-national levels. How can we interpret this change? If we follow the critical political economy approach and apply it to social policy changes, we have to consider rescaling and the multiplication of actors, as an explicit answer of the capitalist system to the collapse10 of the positive-sum-game of economic growth and redistribution. This characterised and legitimised the capitalist system during the trente glorieuses in the postwar period, by allowing most actors to improve their economic and social position. While I share the concerns of scholars highlighting the drawbacks of neoliberalisation in both the political economy of scale and social policy analysis, I also think that the processes taking place are more complex than single ideologies are able to represent (Pickvance, 2008). Neoliberalism critics are able to describe only a portion – even though an important one – of the changes taking place and might underplay (or even neglect) the role of other co-existing processes which are gaining relevance in social policy analysis. For instance, both the spread of individualisation tendencies in social policy design and delivery and the new participatory 9 10 The rule of law developed after the Peace Treaty of Westphalia (1648) where the exclusive right of each state to rule within its own boarders was recognised by all other states. After this “external sovereignty” became established in the international arena, “the rule of law superseded tyranny and the powers of the state were differentiated and separated. (...). With its monopoly of the use of force, the state was then able to consolidate and acquire exclusive rights to these legal powers” (Leibfried and Zürn, 2005: 6). The literature on the reasons why this happened and the relation with the welfare state is quite extensive and varies according to ideology, discipline and aims. See for instance O’Connors (1971), Gough (1979), Offe (1973) for the neo-marxist account, while see Murray (1984) and Mead (1991) for the neoliberal perspective. 41 Yuri Kazepov arrangements necessarily entail a rescaling process downwards to the local level, where disadvantaged people or groups with increasingly differentiated needs can be empowered with tailored – and therefore more effective – interventions, up to their inclusion in the decision-making processes (Craig and Porter, 2006; Powell, 2007). For these reasons, state theorists and social policy analysts tend to embrace a more middle-range theoretical approach, which is not developing a grand theory but attempts to consider the co-existence of different ideologies (e.g. Leibfried and Zürn, 2005). The outcome of their interplay is influenced by a pragmatic orientation of the actors involved in the policy-making process and by the compromises they bargained. In this book, we tie in with this middle-range approach and take some first steps in exploring rescaling processes and emerging governance arrangements for the three policy areas under investigation11 rooting them in their respective theoretical debates. The complexities to be kept under control in this exercise are, however, many and this book provides just a first and empirically grounded exploration. 42 2 Implicit and explicit forms of rescaling social policies Up to the second half of the 1970s most welfare policies were regulated at the national level (McEwen and Moreno, 2005). This was true not only for contributory-based policies, but also for active labour market policies and other standardised welfare services (e.g. job placement services) (Ferrera, 2005). These policies were accompanied by specific forms of spatial Keynesianism through which the state promoted the economic development in peripheral areas such as the Mezzogiorno in Italy or Andalusia in Spain, in the attempt to iron out existing differences at the sub-national levels (Brenner, 2004: 460). In this phase, welfare policies were characterised by a collective management of individual standardised risks, within the frame of a national territorial dimension. The processes of institutional change which were triggered by the sociodemographic and socio-economic transformations followed different directions in Europe, according to the characteristics and internal dynamics of the sphere within which political decisions had to be implemented. Within the economic sphere, both supra-national and sub-national actors become 11 For a short introduction into the project see the preface and the methodological appendix by Barberis in this volume. Rescaling Social Policies towards Multilevel Governance in Europe relevant. On one hand, the European Union established the common market, defined competition rules and introduced the Euro and the European Monetary Union. On the other side, cities and regions acquired a strategic importance in the frame of global competitiveness (Swyngedouw, 1997; Crouch et al., 2001; Salet, 2006; Buck et al., 2005) because the economic performance of their territories increasingly became the basis for their revenues and their ability to guarantee the well-being of their communities. Within the sphere of social policies, the scenario is slightly different. At the supra-national level of the European Union, with a few exceptions mainly related to the rhetoric of the employment strategy (Ashiagbor, 2005), social policies played a rather marginal role.12 Formally important steps have been taken such as the common objectives in the fight against social exclusion and poverty adopted at the Nice European Council (2000), and with the NAPs Inclusion and the NAPs (Hantrais, 2007). These documents, however, did not translate into concrete policy measures at the European level.13 What happened, on the contrary, is that all supra-national market regulations provided – through the Maastricht parameters and the stability pact – the budgetary constraints within which social policies had to develop defining the macro-economic frame for their compatibility. Within this scenario sub-national actors – in particular regions and cities – also increased their role in regulatory terms. This is due to several reasons which reflect the contradictory trends ranging from a neoliberalising rise to the bottom (Ferrera, 2005) to the empowerment of bottom-up initiatives. The resulting landscape – as the single policy area chapters will show – is characterised by the complex coexistence of all policy levels, with differing degrees of freedom in constantly negotiating and renegotiating the new political opportunity structures available (Le Galès, 2002) to them. The rhetoric of subsidiarity legitimising this process highlights the importance 12 13 For an overview up to the mid 1990s see the Green Paper on European Social Policy: Options for the Union (COM (93)551). See also: “The Charter of Fundamental Rights of the European Union” (2000/C364/01). For the developments since then, consult the online database available at: http://ec.europa.eu/social/ This should not underplay the importance of the supra-national level in analysing social policies, which is a rather promising terrain of research. Steps towards transnational social policies include relatively durable arrangements of formal and informal policies, provided by the state or private for-profit or non-for-profit actors, who develop innovative forms of cushioning against targeted emerging social risks. They may range from arrangements addressing post-retirement migration, seasonal migration, medical treatments for specific groups, agreements between health insurance companies, etc. (Faist, 2009). 43 Yuri Kazepov of proximity to citizens, often underplaying the risks that lacking common rights, standards and opportunities might bear. Actually, common rights and standards do not mean standardising policies, but considering equal treatment given equal contextual conditions. This aspect, however, is usually not considered. Within the processes of territorial re-organisation of social policies we can recognise two main trends over the last 30 years: an implicit (A-B in Figure 1) and an explicit form of rescaling (B-C in Figure 1). In the case of implicit rescaling, what changes is the relative weight of specific measures regulated at different territorial levels. For instance, policies regulated at the national level lose resources, claimants and personnel in favour of policies regulated at the sub-national, local level. In the case of explicit rescaling forms of territorial re-organisation of social policies, it is reforms that shift regulatory power to other levels. The latter, for instance, occurred in most European countries for all public employment services, which become increasingly localised (Lundin and Skedinger, 2000; Ferrera, 2006). 44 Figure 1: Implicit and explicit rescaling Europe State Region A B C City -1700 1789 1860-70 1945 1973-79 1990 1995 2000- Rescaling Social Policies towards Multilevel Governance in Europe Both processes took place in almost all European countries and it can be said that in general, since the end of the 1970s up to the mid-1980s, the implicit model of change prevailed. On the other hand explicit reforms become more relevant after the second half of the 1990s. The institutional inertia of Keynesian solutions to specific conditions of need in fact slowed down the process of change, and unemployment had to become a structural phenomenon before the fiscal pressure on the welfare state could trigger deeper reforms. The main reform process started in the second half of the 1990s, in parallel with the attempts to contain the dynamics of unemployment with the spread of both active labour market and activation policies (see Chapters 2-4 herein). Here, we witnessed explicit forms of territorial re-organisation of social policies through reforms aimed at modifying the allocation of regulative capacities at different scales of government, legitimised by the recourse made to the rhetoric of the principle of vertical subsidiarity. In this case as well the existing contextual regulatory conditions influenced the timing of the reforms often in a path-dependent direction.14 In fact, if we take a more nuanced perspective, we see that the timing of changes differed from country to country, for instance Scandinavian countries started the process of explicit decentralisation earlier (even in the 1960s), Italy and Spain in the 1970s, even though consequences started to be felt in the 1980s. Poland started the process of multilevel institutional building incorporating a strong decentralised architecture after joining the European Union in 2004 (Staręga-Piasek et al., 2006). The same differentiation of dynamics took place for the different policy areas according to the characteristics of the respective context of regulation and the overall business cycle. We will consider these aspects more in depth in the single policy area chapters and transversally in Chapter 11.15 An interesting example which clarifies the importance of considering a contextual and multidimensional approach to the study of rescaling processes within social policies comes from the changing degrees of centralisation-decentralisation of the financial relations among territorial levels. In the 1990s most reforms aimed at increasing the overall financial autonomy of sub-national levels of government both in terms of revenues, fostering their own taxation (Stegarescu, 2005: 301) and, in terms of expendi14 15 Path dependency is usually considered intrinsic to institutions as they constrain choice to a limited range of possible alternatives, reducing the probability of path changes and presenting an evolutionary tendency, “a system of nested rules, which are increasingly costly to change” (Goodin, 1996: 23) which produces a self-reinforcing effect over time. On the importance of considering time in social policy analysis see also: Pierson (2000), Fargion (2000) and more recently Bonoli (2007). 45 Yuri Kazepov ture, increasing the costs to be covered by local resources. Both these trends implied a converging reduction of the existing differences of the degree of financial decentralisation16 (see Table 1) (Thieben, 2003; Martinez-Vazquez and Timofeev, 2009) potentially hinting at both a neoliberalisation process implicit in the off-loading process of state financial responsibility (Keating, 1998; Ferrera, 2005) and an empowering process implicit in the development of local control over local resources. Table 1: Financial decentralisation in selected European countries, 1974-2000 !"#$$%&%#'()"$ *+,%+(%"'- !"#$ DK FIN NO SE DE CH FR ES IT UK PL 1974 37,40 0.40 0.33 0.39 0.41 0.38 0.44 0.15 0.15 0.12 0.23 0.212 2000 30,20 0.44 0.33 0.30 0.41 0.39 0.39 0.20 0.27 0.24 0.14 0.27 % 46 &'()* *(*+ *(** &*(*, *(** *(*- &*(*. *(*. *(-) *(-) &*(*, /*(*0 Trend3 2006 +/- –– – –– –– +/ – +/ – + ++ ++ –– ++ + = More decentralisation – = Less decentralisation Source: Source: Own calculations on International Monetary Fund (IMF) Data (1972-2002). 1) The coefficient of variation considers all countries except Poland. 2) Data on Poland refer to 1985 and 2000 respectively. 3) According to the analysis of the reforms considered in the Rescaling Project. The consequences of these trends, despite their apparent convergence, are complex to decipher because a high share of sub-national taxation does not necessarily correspond to a high degree of sub-national autonomy in both defining taxes and expenditure targets. The latter might be strictly regulated by national/federal criteria, which heavily structure and restrict sub-national autonomy (OECD, 2006a; Martinez-Vazquez and Timofeev, 2009). The same is true for sub-national expenditure, which might change 16 The degree of financial decentralisation is usually calculated considering the share of sub-national expenditure in the total public expenditure and the share of sub-national taxes in the total fiscal revenues at the national level. In our case, we have considered four indicators: 1) the share of sub-national expenditure in the total national expenditure; 2) the share of sub-national expenditure in the GDP; 3) the share of sub-national taxes in the total sub-national revenues; 4) the sub-national taxes as a percentage of total taxes. For each country, the total score has been standardised, i.e. 0 = total absence of decentralisation; 1= total decentralisation. Rescaling Social Policies towards Multilevel Governance in Europe according to specific measures within single policy areas as well (on this, the single chapters of the first part offer a more precise insight). In order to understand these changes it is, therefore, necessary to embed them in the regulatory frames which define how specific budgetary items are funded and spent in the different welfare systems. This has to be done in particular considering the territorial dimension of social policy regulatory jurisdictions, identifying “who pays” for what kind of provision at which level. When we have taken that into account, it will no longer be counterintuitive to realise that the most decentralised countries are also the countries with lowest income inequality (see Table 5 below). The institutional meso-level of analysis helps us in that exercise. It also helps us throughout all chapters, to disentangle the ways in which the reforms of the last decades have produced a quite diversified landscape characterised by complex interwoven regulative powers on different scales and among different actors vis-à-vis public-private lines of differentiation in the design, management and implementation of the policies themselves. Chapter 11 provides a precise and transversal account on these aspects. 3 The multiplication of actors and New Governance arrangements in social policies The second process of change that the different welfare systems are facing is represented by the multiplication of actors involved in designing, managing and implementing social policies. We can consider this a horizontal subsidiarisation process in which the number and quality of actors involved increases in numbers and differentiates its tasks. The externalisation of social services (Dixon and Hyde, 2001), the individualisation (Beck, 1998; Paci, 2007) of specific measures and – in some cases – their privatisation (Dixon and Hyde, 2001) have been taking place since the early 1980s. The aforesaid have strongly contributed to a new role for private – both for-profit and non-profit – actors (Powell, 2007). The spread of specific policies, like homecare services for the elderly (see Chapter 8 herein) or activation policies for both unemployed and social assistance claimants (see Chapters 2 and 5 herein), have provided these new actors with new tasks as well. Generally speaking these processes tend to flatten the forms of coordination (see Chapter 12 herein) making the sub-national dimension (either regional or municipal) a crucial territorial level not only for managing and 47 Yuri Kazepov 48 implementing social policies, but increasingly also for the definition of the policy itself or of important parts of it. From this point of view, the vertical and horizontal forms of subsidiarisation reinforce each other and develop frames for action which are relatively coherent with the welfare systems they belong to. The arguments used to legitimise the involvement of new actors are manifold and reflect the ambiguities we already highlighted for the rescaling process with which they are closely intertwined (Powell, 2007). From the economic point of view, budgetary constraints played a major role in supporting both the externalisation and privatisation of social services and provisions. Quasi-markets providing a competitive frame for demand and supply have been seen as a specific way of improving efficiency and effectiveness (McMaster, 2002) almost taking for granted the inevitable direction of change: the neoliberalisation of social policies. This discourse (and actual measures), however, have been paralleled by the spread of a participatory rhetoric which viewed the involvement of civil society – strongly rooted in the full deployment of the subsidiarity principle – as a necessary step to empower the people in finding a solution to their own problems (Ranci, 2006; Fung, 2003; Powell, 2007). A widespread opinion was – and partly still is – that widening the range of actors would increase the legitimacy of the policymaking process (Hirst, 1993). From this perspective, externalisation might support a general reorganisation of the decision-making process involving various stakeholders at different levels, underlining improvements in the empowerment process of civil society broadly conceived (Paci, 2007). Both rhetoric coexist and co-evolve. The involvement of new actors was initially related merely to the implementation of social services and their management. Only in recent years has their role shifted to planning and the co-definition of goals. All that determined a re-definition of the policymaking process, the spread of new models of governance in the management of social policies and a shift from a hierarchical-vertical regulation towards a more cooperative-horizontal one. This re-definition of the policy-making process requires new state capacities aimed at supporting the transition from a government function of a democratically elected body towards a form of governance, in which the public actor aims at facilitating and coordinating the implementation of policies (see Chapter 12 herein). This occurs within a frame in which the discussion about social policy design, its management and implementation takes place in increasingly fragmented and uncertain contexts, characterised by a high degree of negotiation among the actors involved. Rescaling Social Policies towards Multilevel Governance in Europe As a reaction to the crisis of the welfare state, reform processes – in their double meaning of vertical and horizontal subsidiarisation – produced a steady shift from a vertical towards a horizontal coordination of social policies, which finds its ideal level of implementation in the local dimension.17 Despite the fact that these tendencies are common to most European countries, the development and institutionalisation of the new governance arrangements do not converge. On the contrary, the results of these processes of change seem to produce a territorially structured diversification (see Chapter 11 herein). This diversification varies according to socio-economic context and institutional arrangements, with all the specificities this might entail: from a high degree of freedom of the Comunidades Autonomas in Spain, the Länder in Germany or the Kantons in Switzerland, to the relatively low intra-national differentiation in France. DiGaetano and Strom (2003) emphasise the fact that these different institutional contexts favour specific governance arrangements at the local level; that is to say: formal and informal rules actors build and follow, developing practices and networks, attempting to contribute – in one way or another – to the stability and dynamics of the local society and political regime (Le Galès, 2002: 15). Other scholars have gone beyond the local dimension, witnessing how the emerging governance arrangements are quite in line with the respective national welfare systems they belong to, following a path-dependent logic which tends to pre-structure the diversified outcome of the current processes of change (Jessop, 2002; Kazepov, 2005). In this sense, reform processes do not occur in an institutional vacuum, but are deeply rooted in the institutional settings of the individual countries which influence – through context-specific constraints and opportunities – the directions of change. Welfare systems and their functioning are therefore an excellent perspective in which to frame our understanding of the subsidiarisation processes of social policies and their impact. 4 Rescaling and governance in European welfare systems: a comparative perspective Understanding the processes of subsidiarisation of social policies cannot avoid considering what the main features of European welfare systems are. These very features and the underlying regulative principles do translate the common transformation processes into context-specific outcomes and 17 For an in-depth analysis of the differences among single policy areas see the thematic sections of the book and the transversal chapters of the second part. 49 Yuri Kazepov 50 practices which influence the allocation of resources and responsibilities in the different countries and sub-national territories. As Pawson and Tilley (1997) put it, the right equation is: “context+reforms=outcomes”. Following the legacy of Karl Polanyi’s seminal work, The Economy as Instituted Process (1957), in the last few decades the debate on welfare systems has been revitalised by Esping-Andersen (1990 and 1999) who, taking up the contributions by Titmuss (1958), as well connected regulatory principles to specific policies, in particular he related: a) the principle of redistribution to universalistic policies; b) the principle of exchange to liberal market-oriented policies; c) the principle of reciprocity to policies based on family responsibility. We can further specify this responsibility by distinguishing active or passive forms of subsidiarity.18 Both forms share the importance of the family, of third-sector actors, voluntary associations, etc. in the overall welfare system. They differ, however, in relation to the resources that are socially targeted to these actors. Active subsidiarity implies shifts in responsibilities and of “adequate” resources to fulfil these responsibilities (e.g. child care, care of the elderly, ...). Passive subsidiarity just shifts responsibilities without providing and redistributing resources. The recent debate focused both on the number of existing models and the underlying regulatory principles, as well as on the kind of indicators to be considered in the taxonomic exercise. The literature on this issue is quite diversified19 and most authors propose – implicitly or explicitly – a classification based on the balance among different regulatory principles and the way in which the respective institutions socialise risks. In particular, the country-specific equilibrium among the state, the family (plus the third sector) and the market (Esping-Andersen, 1990; Mingione, 1997; Ascoli and Ranci, 2002) contributes to ways in which the allocation of resources is defined. In doing so, this process shapes the constraints and opportunities actors face in defining their life strategies. Within this context, the state (through the rule-of-law) plays a major role, because it also defines – implicitly or explicitly – the role of the other institutions. In fact, by defining access criteria and the generosity of provisions as well as taxation duties, the state not only allocates resources to specific groups, but implicitly (and functionally) also 18 19 I have introduced the distinction between active and passive subsidiarity in order to make the reference to subsidiarity more precise in Garcia and Kazepov (2002). Among the many contributions see: Esping-Andersen (1990, 1999); Ferrera (1998); Mingione (1997); Bambra (2006). For an overview see Art and Glissen (2002). For labour market-related policies see: Gallie and Paugam (2000). For spatially-related implications see: Kazepov (2005). Rescaling Social Policies towards Multilevel Governance in Europe distributes the responsibilities in the production of well-being to the other institutions, be it the market (through commodification) or the family (through familialisation) (Esping-Andersen, 1999). Taking into account these guiding principles and perspectives European scholars identified “four-plus-one” main welfare systems, which allow us to allocate most European countries: the liberal/Anglosaxon welfare system, the social democratic/Scandinavian welfare system, the corporatist/Continental welfare system, the familistic/South European welfare system and the welfare system of transition countries/Central and Eastern European. We are of course aware of the fact that the picture is always more complex than models can reproduce (Crouch, 2005), but the modelling business (Abrahamson, 1999) – if it is not abused – might help to interprete the directions of social policy change and provide a frame of reference beyond the Keynesian-Fordist perspective to a more nuanced analysis of ad hoc measures within specific policy areas. The aforesaid is precisely what this book intends to do. One of the main problems of the conceptualisation of these systems, however, is related to the fact that it does not consider the spatial dimension or specific actors’ configurations. Our attempt is, therefore, to start a line of research which tentatively tries to show the privileged relationship existing between the diverse scale arrangements and actors’ configurations in the different welfare systems. Both dimensions are intimately related to one another, as the analysis of the vertical organisation of powers and the institutional frame within which actors and single policy measures are embedded, define actors’ degrees of freedom. These factors subsequently influence the ways in which they interact and coordinate in the policy process. Agency is – from this point of view – never free-floating: it is always embedded within a specific institutional frame of reference providing both opportunities and constraints (Giddens, 1984; Jessop, 2002 and 2008; Le Galès, 2004; Kazepov, 2005). Considering these aspects we have identified four main spatial configurations, which are taken up in most chapters of the book. These are the result of our attempt to put forward a meso-level analysis by combining macro and micro sources and referring to structural indicators as well as to the results of institutional micro-simulations20 we carried out. In fact, the 20 Information on the macro characteristics refers to both literature review and structural data we collected within the Rescaling project (OECD, Eurostat, ... but also institutional analyses). Information on the micro characteristics refers to targeted interviews we carried out with experts, social workers, policy-makers, etc. to reconstruct the institutional frame for the vignettes (see methodological appendix for details). 51 Yuri Kazepov spatial configurations and the actors’ arrangements we are reporting are the result of an aggregated systematisation of information in which qualitative and quantitative data as well as primary and secondary sources have been considered in order to understand and re-construct the contexts within which rescaling processes and new governance arrangements are taking place and co-evolving. A brief comparative sketch based on data also contained in Table 2 might help in highlighting the existing differences, which will be synthesised in Figure 3 and Table 3 at the end of this section. 1) The Liberal welfare system is characterised by a relatively residual role of the state, which intervenes only when both the market and the family have failed in allocating resources. The market is the prevailing mechanism of regulation and integration in a highly individualised (see position on the y axis in Figure 1) and competitive society. The main example of this model is represented by the US, where the importance of welfare policies is extremely limited and follows, eventually, a form of privatised Keynesianism21 (Crouch, 2008), with all consequences it might hold in terms of the individualisation of risks and responsibilities. Among the European countries, the United Kingdom is the closest one to this model, even though some substantial differences exist, due to the legacy of the Beveridgian welfare state and the relatively developed social services. Poverty and inequality rates are, nevertheless, among the highest in Europe (see no. 23-26, Table 2) and social policies – often means-tested – assumed, after the long period in power of the Conservative government (1979-1997), a rather negative image (Rodney, 2004). Generosity and adequacy of income support levels are at an intermediate position (see the differences between the US and UK position on the x axis in Figure 3), i.e. lower than in Scandinavian countries, but definitely higher than in South-European ones. From the territorial point of view, social policies are centrally framed and differentiation at the sub-national level occurs more in relation to activation policies and to the related local governance arrangements (for training, stages, social insertion,…); furthermore, is mainly market-driven (Lindsay, 2007). This, however, does not translate into high regional dispersion rates for both employment and unemployment. Welfare benefits and other income support meas- 21 Crouch distinguishes Keynesianism – in which it was governments that took on debt to stimulate the economy – from privatised Keynesianism in which it is “individuals, particularly poor ones, [that] took on that role by incurring debt on the market” (Crouch, 2008: 476). 52 Rescaling Social Policies towards Multilevel Governance in Europe ures per se are also more or less managed homogeneously throughout the country, but the joint PES and Benefits agency management of the New Deal is increasingly fragmented and depends very much on the resources available locally. The latter are, in turn, related to the degree of competitiveness that regions and cities can achieve in the market (Buck et al., 2005). The governance arrangements developed within this welfare system tend to be, from this point of view, pluralist and managerial, that means coordinating an increasing number of public-private actors and prioritising efficiency – through forms of new public management (NPM) – over equality. In more recent years a whole-of-government approach has been put forward in order to go beyond NPM and to work across fragmented policy areas in order to achieve a shared goal and an integrated response to particular issues (Christensen and Lægreid, 2007: 1060). Despite successful targeting and efficient poverty reduction, the incidence of poverty and income inequality is higher than in other European countries (see indicator no. 26 in Table 2). Unfortunately our study could not include the UK case.22 2) The Social democratic welfare system is characterised by a pervasive role of the state which takes over responsibilities that other welfare systems delegate to the family or the market. Measures are mostly universalistic, addressed to all citizens according to need. A wide range of in-kind services and monetary transfers is provided, and redistribution is the most important form of allocation, even if in the last decades some second-level insurance-based schemes have been introduced in this system. Market dependency, poverty and inequality are the lowest not only in the EU but also in the world (Ervik and Kuhnle, 1996; Kangas and Palme, 2005; Greve, 2007). Adequacy of income support measures is a fact and local differentiation tends to decrease, except for social assistance and activation measures, which depend on local political and economic contexts. The debate over the consequences of this territorial differentiation, however, kicked-off a trend aiming at homogenising standards in most Scandinavian countries (e.g. Sweden in 1998, Finland in 2001 and Norway in 2005. See also the chapters herein for details on the single policy areas). From the point of view of the territorial organisation of social policies, these countries are, therefore, characterised by what we can describe as a local autonomy 22 On the Rescaling project’s methods, data sources, format and management, see the Appendix by Barberis herein. 53 4 5 6 5 5 6 3 11 10 9 8 22) SA for 1 (PPP) 12 21) Active Labour policies 20) Labour policies 19) Elderly/survivors 10 10 18) Family/children 17) As % on GDP 16) Per capita in PPS Social expenditure 15) Unemployed covered 14) Dispersion NUTS-2 13) Long term (15-64) 12) Youth (15-24) 5 11) Female (15-64) 10) Male (55-64) 7 9) Male (15-64) 5 Unemployment rates 8) Dispersion NUTS-2 7) Youth (15-24) 5 6) Female (15-64) 5) Male (15-64) 5 Employment rates 4) Divorce 3) Births out of wedlock 2) Fertility rate 2 11 1) Child in single parent family Population 1 11,9 0,61 463 964 0,42 1,11 0,43 43,8 6,0 26,4 7410 26,2 24,8 23,8 14,3 5,0 4,1 5,6 5,4 52,9 65,5 77,5 2,6 43,6 1,84 24,0 43,2 8,6 22,4 4585 n.a. 33,9 41,9 17,5 7,8 8,4 12,1 7,5 54,9 60,8 73,9 2,7 27,9 1,83 2007 812 1,68 2,56 36,5 11,7 37,4 6187 n.a. 29,6 15,5 25,6 7,3 1,3 10,7 4,8 37,3 73,1 73,0 2,3 47,0 2,13 n.a. 1990 616! 1,36 2,32 39,3 9,6 30,7 8998 87,0 10,1 13,8 18,6 6,4 4,3 5,8 2,4 42,2 71,8 76,5 2,2 54,7 1,85 n.a. 2007 Sweden 1990 Social-Democratic UK 348! 0,72 2,09 40,7 9,7 27,3 5159 n.a. 24,1 35,0 24,5 12,9 6,0 9,5 7,1 27,6 51,5 67,3 2,1 n.a. n.a. n.a. 1990 362! 0,92 2,32 41,6 8,1 31,1 8200 47,0 35,2 40,2 18,7 8,9 5,3 7,8 6,6 31,5 60,0 69,3 2,5 50,5 2,00 14,0 2007 France Corporative 54 Liberal The welfare systems in Europe, 1990-2007 Welfare systems Table 2: 322 n.a. n.a. 55,1 4,2 24,0 4289 n.a. 68,9 57,7 28,2 13,9 1,6 7,4 17,4 28,3 35,8 69,3 0,5 6,4 1,33 3,3 1990 Italy 375 0,53 1,32 58,3 4,3 26,6 6476 4,4 56,7 47,4 20,3 7,9 2,6 4,9 16,3 24,7 46,6 70,7 0,8 20,7 1,32 6,0 2007 Familistic n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 22,5 n.a. n.a. n.a. 7,5 n.a. 4,8 n.a. n.a. n.a. 1,1 9,4 1,62 n.a. 1990 n.a. 0,45 1,18 59,9 4,3 19,2 2373 n.d. 14,2 51,3 21,7 10,4 7,4 9,0 4,5 25,8 50,6 63,6 1,8 19,4 1,27 9,0 2007 Poland In Transition n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 6,1 n.a. n.a. n.a. n.a. n.a. 1,7 n.a. n.a. n.a. 1990 EU-25 n.a. n.a. n.a. 44,4 7,7 27,0 6630 n.a. n.a. 42,2 15,1 7,9 5,5 6,6 n.a. 38,3 58,6 73,0 2,1 n.a. n.a. 13,0 2007 Yuri Kazepov 14 32,0 37,5 20,0 32,0 33,0 36,7 19,0 30,0 2007 21,0 n.a. n.a. n.a. 1990 23,0 60,7 11,0 28,0 2007 Sweden UK 1990 Social-Democratic Liberal 29,0 42,3 15,0 26,0 1990 23,0 20,0 13,0 33,0 26,0 13,0 50,0 26,0 24,0 32,0 16,7 20,0 n.a. n.a. n.a. n.a. 1990 32,0 37,0 17,0 27,0 2007 Poland 2007 Italy 1990 In Transition Familistic 2007 France Corporative n.a. n.a. n.a. n.a. 1990 EU-25 30,0 38,5 16,0 26,0 2007 n.a. = not available. 1 Last year: 2005. Source: Eurostat in Kazepov (2005, pp.16-17); Eurostat (2006). 2 Last year: 2006. For PL, first year: 1995. For IT, last year: 2005. Source: Eurostat (2009), on-line: http://epp.eurostat.ec.europa.eu/ 3 For PL, first year: 1995. For UK and FR, last year: 2006. Source: Eurostat (2009), on-line: http://epp.eurostat.ec.europa.eu/ 4 For FR, first year: 1995. For all, last year: 2005. Fonte: Eurostat (2009), on-line: http://epp.eurostat.ec.europa.eu/ 5 First year: 1993. Source: Eurostat (2009), on-line: http://epp.eurostat.ec.europa.eu/ 6 Dispersion at NUTS-2 level of total employment rate and unemployment rate. First year: 1999. Source: Eurostat (2009), on-line: http://epp.eurostat.ec.europa.eu/ 7 For PL, first year: 1994. For EU-25, considered: EU-15. Source: OECD (2009), on-line: http://webnet.oecd.org/wbos/index.aspx 8 Unemployed covered by unemployment benefits, 2001. Source: Carbone (2005, p.71). 9 In some cases, provisional data. First year: 1995. Last year: 2006. Source: Eurostat (2009), on-line: http://epp.eurostat.ec.europa.eu/ 10 In some cases, provisional data. For S, first year: 1993. For all, last year: 2006. Source: Eurostat (2009), on-line: http://epp.eurostat.ec.europa.eu/ 11 Incidence (%) of expenditure in labour market policies (both total and active) on GDP. Last year: 2006. Source: OECD (2009), on-line: http://webnet.oecd.org/wbos/index.aspx 12 Social Assistance for one single person per month. PPP = purchasing power parities (Euro = 1). Years considered 1992 and 2001. Source: Cantillon (2006) Minimum Income Protection in Europe.. 13 Incidence (%) of persons with available equivalent income, pre and post social transfers, under the relative poverty line (established at 60% of the equivalent median national income). In some cases, provisional data. First year: 1995. Source: Eurostat (2009), on-line: http://epp.eurostat.ec.europa.eu/ 14 Own calculation taking into account two indicators (23 and 24), according to the following formula: [(23-24)/(23)*100]. 15 In some cases, provisional data. For all, first year: 1995 (except S = 1997). Source: Eurostat (2009), on-line: http://epp.eurostat.ec.europa.eu/. 26) Gini Index 15 25) redistributive capacity 13 13 24) 60% median post-transfers 23) 60% median pre-transfers Poverty and income Welfare systems Rescaling Social Policies towards Multilevel Governance in Europe 55 Yuri Kazepov centrally framed system. This implies that the high autonomy that municipalities retain in managing and funding policies is embedded in a nationally defined regulatory context, which contributes – through the direct provision of many benefits and services – to keeping territorial differentiation under control. Sellers and Lidström call it nationalised local government (2007: 622). Recent trends confirm the importance of the central state (see Figure 2.a for the Finnish case), even though through coordination and steering activities rather than hierarchical control. Income inequality and poverty before taxes and transfers is relatively high, after taxes however it becomes the lowest in Europe (see indicators no. 23-26, Table 2). This reduction is the result of generous transfers and the highest share of the national GDP allocated (see indicators no. 16-17, Table 2) to social expenditure, combined with an effective redistributive capacity of the welfare system. Income transfers like child and family allowances, unemployment benefits (all nationally regulated), and inkind services keep large parts of the population well above the poverty line. As a consequence, this welfare model maintains a high degree of social consensus, as both social and economic results are positive. 56 Figure 2.a: Local autonomy centrally framed: Finland 1990-2005 9 8 7 6 5 4 3 2 1 0 Around 1990 Mid 1990s &'()*$% Around 2000 +,- !"#$% Mid 2000 Rescaling Social Policies towards Multilevel Governance in Europe In this scenario, governance is regulated both by managerial (looking for efficiency) but also participative (also involving claimants) arrangements in which central public authorities, rather than private non- or for-profit actors, play a major role and self-coordinate. Reforms in recent years have been approved in order to keep social expenditure under control, but the overall system configuration is – viewed comparatively – still the most generous and encompassing. 3) The Corporatist welfare system is characterised by social policies conceived in a meritocratic way: contributory schemes reproduce the socioeconomic status that families achieve in the labour market through the position of the breadwinner (Gallie and Paugam, 2000). The family is the prevailing social agency and, coherently with this, it is strongly supported in its caring role by specific targeted in-kind and monetary state provisions (active subsidiarity). Reciprocity is an important form of integration and allocation of resources (see its position on the “y” axis in Figure 3). In fact, even though the state intervenes extensively, this intervention is mediated by the role of the family. Dependency on the market is higher than in the social-democratic model, but lower than in the liberal one. Bismarckian Germany with Austria, France and Belgium are clear examples of this welfare system. From the territorial point of view these countries present quite diversified organisation of models ranging from federal state structures (e.g. Germany, Belgium or Switzerland) – that are regionally framed – to national ones, which are centrally framed (e.g. France). Countries with regionally framed social policies often have the exclusive legislative responsibility allocated to regions, Kantons or Länder and a great potential for differentiation. However, this potential is often kept under control through forms of institutional isomorphism (DiMaggio and Powell, 1983), i.e. through the de facto mutual adaptation of institutional solutions. In countries with centrally framed social policies, on the contrary, benefits and provisions as well as procedures are regulated and funded at the national level. Only the activation of the RMI claimants is fully up to the Départements, but again in a centrally defined frame. Despite this important difference, there is a converging tendency towards forms of intermediate regulation between the national-federal level and municipalities. In Germany it is the Länder, while in Switzerland the Kantons and in France (in particular since 2003) the Départements. In most cases it is national standards which 57 Yuri Kazepov often bind or direct welfare provisions; for this reason the redistributive effects of social policies are relatively high (see indicators no. 23-26 in Table 2). Figure 2.b: Centrally framed: France 1990-2005 9 8 7 6 5 4 3 2 58 1 0 Around 1990 Mid 1990s &'()*$% Around 2000 +,- Mid 2000 !"#$% Figure 2.c: Regionally framed: Switzerland 1990-2005 9 8 7 6 5 4 3 2 1 0 Around 1990 Mid 1990s &'()*$% Around 2000 +,- !"#$% Mid 2000 Rescaling Social Policies towards Multilevel Governance in Europe This downward regulatory convergence does not mean that the central state has lost regulative capacity in France, but that the institutional landscape becomes more articulated (see the level of state intervention which remained high throughout the last 25 years, Figure 2.b). Unlike France, Switzerland (and Belgium) presents a higher level of differentiation and a wider share of options. Figure 2.c shows the basis for this variation very clearly: throughout the last 25 years Kantons have kept their crucial regulative role, coherent with the federal frame of reference. However, in-kind measures may vary more (like activation measures, see Chapter 4 herein), but we have to consider that this variation takes place within a frame of relatively well-guaranteed rights and generous benefits (see Chapter 2 herein). Governance is regulated by corporative arrangements with strong relevance of negotiation among intermediate (often non-profit) bodies and public authorities. Some participative and/ or managerial arrangements might characterise sub-national contexts, varying according to the local political culture. In the scholarly literature, this model is often conflated with the familistic one (Esping-Andersen, 1990, 1999). Indeed, they share the double logic underlying the way in which social risks are managed. Contributory-based benefits on one hand, and the subsidiarity principle regulating the relationship between the state and the family or the subsidiary actors on the other. However, the role of the state is quite relevant in making the difference in terms of generosity (see indicators no. 16-17, Table 2) and services23 provided, defining a relatively high degree of inclusiveness into the respective redistributive communities. 4) The Familistic welfare system, like the corporatist model, conceives social policies in a meritocratic and fragmented way, but it is less generous and very unbalanced in the provision of monetary benefits, which prevail over in-kind services (Naldini and Saraceno, 2008). Fewer resources are targeted to family policies and to other contributory and meanstested schemes. The consequence is that the family is overloaded with social caring responsibilities without, or with very few institutional resources provided by the state (Naldini, 2003) (see Figure 1 and the high scores of the familism index ) bringing about forms of passive sub- 23 The ideological frames which historically influenced the development of social policies in the two models differ as well. The catholic culture in the Mediterranean countries and the monarchical statism rooted in a Bismarckian approach in continental European countries set the stage for the institutionalisation of the deep differences (Alber, 1982; Manow, 2004). 59 Yuri Kazepov sidiarity. These aspects contribute to the segmentation of the labour market in terms of gender and age granting a higher protection to the (often male) breadwinner and slowing down generational turnover. All South-European countries (IT, ES, PT, GR) show a low redistributive capacity (indicators no. 23-25, Table 2) and fewer services, exacerbating – in a context of low incomes – social inequalities. The latter have a strong territorial dimension, in particular in Spain and Italy. In fact, despite strong postwar centralism, since the second half of the 1970s both countries have shifted a great deal of regulatory capacity to their regions. Policies are therefore, at the territorial level, regionally framed to a large extent, highly segmented and targeted at particular categories. Unlike corporatist welfare system countries, regional differentiation does not take place within a frame of relatively well guaranteed rights. On the contrary, it might be even more fragmented, as in the Italian case, i.e. drifting towards very localist variants. Figure 2.d clearly shows the deep changes which have occurred from the territorial point of view in Italy, with the regions becoming the main actors in most of the policies we considered. The dependency on the market resembles the liberal model, with the exception that families extensively reduce this dependency (through low divorce rates, few single households, 60 Figure 2.d: Regionally framed: Italy 1990-2005 9 8 7 6 5 4 3 2 1 0 Around 1990 Mid 1990s &'()*$% Around 2000 +,- !"#$% Mid 2000 Rescaling Social Policies towards Multilevel Governance in Europe etc., see indicators no. 1-4, Table 2). In general, the level of benefits is much lower and local measures and formal entitlements do not always guarantee a payout. Local differentiation and the discretionary power of social workers are much more widespread because of heavy budgetary constraints. Reforms in Portugal and Spain at the end of the 1990s attempted to modernise social assistance schemes, connecting them to national framework laws and RMI-like schemes, but they failed in homogenising access criteria and provision. This is true, in particular, for activation (or insertion) policies and the respective governance arrangements involved, which are in many cases regulated by particularistic and – sometimes clientelistic-oriented by patronage relations – principles with a strong place for private (also non-profit) interests. The typical territorial fragmentation is also reflected in the co-existence of different forms of governance from innovative participatory ones (in progressive regions) to more traditional corporatist forms of interest mediation (in regions with a strong industrial legacy). There is, however, no legally enforceable social right. Income inequalities and poverty rates are higher than in the other European countries. The familistic welfare system is unable to adequately redistribute resources and to prevent poverty. As a matter of fact, poverty rates before and after social transfers in Italy show the lowest capacity to reduce the number of low-income families among the countries considered in Table 2 (see indicators no. 23-26). 5) The welfare system in transition countries is not yet a consolidated model with clear characteristics. Both the conditions producing vulnerability and the institutional frame aimed at contrasting them have experienced a dramatic change since 1989. Most Central European and East European countries belonging to this model underwent a deep structural change in the economy with sharp GDP decreases followed by high increases. The reforms implemented in the last decade to accompany these changes and to contrast their potentially negative impact have had ambivalent consequences. Countries like Poland, for example, give a more important role to market regulation (Staręga-Piasek et al., 2006), while others like Slovenia are investing more in coordinated market and social policies. The starting basis, however, is a quite homogeneous distribution of income with – in most cases – below-average inequalities and strong central control. Yet, the dynamic of change and the impact 61 Yuri Kazepov of the policies adopted in the last decade will bring about consequences in the coming years, most probably causing a hybridisation of models (Hacker, 2009). The first signals are coming from the greater (e.g. Slovenia, Czech Republic) or lesser (e.g. Slovakia) ability of policy transfers to reduce the risk-of-poverty rates significantly (Atkinson et al., 2002). Moreover, becoming part of the European Union was for these countries, even before 2005, an important input for reforming social policies towards institutional designs closer to those of other continental European countries (Cerami, 2006; Fenger, 2007; Aspalter et al., 2009). Changes in the territorial organisation of social policies and in the governance arrangements follow the heterogeneous lines existing in Europe and provide examples of all forms of territorial rescaling dynamics and interaction among actors delineating for these countries a transitional mixed model. From this point of view, Poland is undergoing a heavy territorial re-organisation of social policies, away from central regulation towards more decentralised levels, configuring complex multilevel arrangements which differ from policy to policy (see Figure 2.e). Other countries have kept a stronger central control. Poland also seems more oriented towards forms of competitive pluralism and privatisation (where regional variation occurs vis-à-vis some 62 Figure 2.e: Mixed in transition: Poland 1990-2005 9 8 7 6 5 4 3 2 1 0 Around 1990 Mid 1990s &'()*$% Around 2000 +,- !"#$% Mid 2000 Rescaling Social Policies towards Multilevel Governance in Europe degree of national regulation and responsibility); other countries, like Slovenia and the Czech Republic, have adopted more corporative solutions with concerted decision-making processes. The social model which seems to emerge from these short descriptions sees the “family” and the actors from civil society (the voluntary sector, ...) on one side and the “state” on the other as the main agencies that socialise individual risks, transforming them into social responsibilities. In the first case regulatory principles based on subsidiarity and reciprocity prevail. In the second case redistributive principles prevail. If we consider the “family” (and civil society) and the “state” as two extremes of a continuum, we need to distinguish between passive subsidiarity – which characterises countries in which the state delegates social responsibilities without allocating adequate resources to the actors who take over the burden and responsibility to act – and active subsidiarity in which the state targets specific resources in favour of the actors which take over social responsibilities, intervening in relation to specific needs (e.g. care, training, economic support, …). This distinction intersects with the distinction between vertical and horizontal subsidiarity and allows us to take into account the differences existing between continental welfare systems and south-European ones, providing some indications for interpreting the complex changes taking place in the welfare systems in transition. Figure 3 synthesises what has been briefly sketched out so far using as a proxy two rather rough but helpful indexes: a familism index and a statism index.24 Cross-tabulating the two indexes, the European social models clearly emerge: in Europe it is either the family or the state which take on social responsibilities. This alternative is not without effects on the rescaling processes and the new governance arrangements, as the family lacks the state’s redistributive capacity and provides the ground for increasing inequalities among territories and actors if not adequately backed up by the state’s intervention with ad hoc equalisation funds (as it is the case in 24 The familism index has been calculated considering 5 indicators: 1) Single parent households, as % of all households with children, 2005; 2) Divorce rate, per 1000 inhabitants, 2004; 3) Births out of wedlock, as % of the total live births, 2004; 4) Young males (1834) living with parents, 2003; 5) Young females (18-34) living with parents, 2003. The statism index has been calculated considering 4 indicators: 1) Differential between the at-risk-of-poverty rate before and after social transfers, 2003; 2) Social expenditure, as % of GDP, 2002; 3) Social expenditure in PPS per capita, 2002; 4) Total taxes as % of GDP 2003. Source: Arlotti calculations on Eurostat figures (online statistics http://epp. eurostat.ec.europa.eu); OECD (2005); European Foundation (2005). All variables have been standardised. 63 Yuri Kazepov countries characterised by active subsidiarity) or centrally framed policies. It is indeed the state’s intervention which provides a key element in differentiating both vertical and horizontal subsidiarity in a qualitative and quantitative manner. This holds true for the role of the third sector as well. Figure 3: The role of redistribution and family y Family IT GR ES High Fa amilism!index Reciprrocity!networks 64 PL PT NL HU AT CZ DE FR BE Low SF US UK DK (Market) Low State Redistributive!Institutions High x Statism index Statism!index Analysing elderly care, Ascoli and Ranci (2002) provide some comparative empirical evidence of the role that third sector actors retain in different welfare systems (see Table 3). We can use their results as a useful proxy to frame our understanding of the role of multiple actors in social policy delivery. Of course considering not just delivery, but also design and management might change the balance and the specific policy chapters herein offer more details on these actors’ roles. The scheme, however, is useful because it grounds its classification on how the role of third sector actors intersects with the actual resources allocated and complements well the territorial dimension of social policies. Rescaling Social Policies towards Multilevel Governance in Europe It is a major challenge to try to consider these elements altogether and the analytical distinction between institutional design, the actual management of the specific policies and their delivery on both the vertical and horizontal dimension helps us – in the different chapters – to fine-tune our understanding of the changes that have occurred so far. Table 3: Third sector actors’ involvement models in European welfare systems !./"/0."1)02345"64)78)9:4)#9"94 &214)2B)9:4) $:.5C)#40925 $29"1);<)=>?@ '2D./"/9 ;<)E>?@ (25F25"9.34))*2C41 ,09.34)GH7G.C."5.98 (8*#,%#.#,"501+/*2.' %"59."1);A)=>?@ SI !"D.1.G9.0)D2C41 %"GG.34)GH7G.C."5.98) ( ()*+,-./#01+/*2.' BG RO PL 65 (2DF14D4/9"58 ;A)E>?@ )/.345G"1.G9.0 *2C41)C2D./"94C) 78)9:4)G9"94 ."!"#$%#"&%"' +421.745"1)D"5I9) D2C41 (3#45*6"7*#' Source: Own adaptation from Ranci (2003). 5 The pros and cons of the subsidiarisation process of social policies: some reflections on differences in Europe Considering the fact that different institutional contexts tend to pre-structure both the impact of the rescaling processes and the role of the new actors as well as the directions of change, in this section I would like to provide a synthetic account of the pros and cons emerging from these processes in more general terms. Meanwhile the policy area chapters will carry the burden of addressing these aspects and their implications in a more ad hoc and precise way. Both are potentially present, and the prevalence of one or the other depends on the meaning they acquire within the specific contexts examined. The pros are mainly related to the new role that the local dimension and non-state actors are gaining while increasingly being involved in designing, managing and implementing social policies. In particular, we can highlight the following positive aspects: Yuri Kazepov a) 66 the widening of local experimentation: the options during which new (and old) actors and territories can find (together) new and coordinated solutions to old problems have increased (Mouleart et al., 2007; Drewe et al., 2008: 22) transforming the local level into a social laboratory (Le Galès, 2002; Kazepov, 2005) in which to explore new land. b) More freedom for grass-roots action: more local experimentation also allows the valorisation of existing differences and the development of bottom-linked socially innovative strategies, combining passive and active measures in tailored solutions which could foster social justice and contribute to a better redistribution of resources (Garcia et al., 2008). c) The legitimation of political choice: involving local actors no longer only in the implementation of social policies but increasingly also in the decision-making process, legitimises those political choices which impact on the very same actors and territories, by building trust among them and facilitating mutually adjusting compromises. The cons are related to the ways in which the territorial re-organisation of social policies and the new governance arrangements modify the way in which vulnerability and social risks are produced and institutionally addressed. In particular we can highlight the following critical issues: a) the potential divergence among sub-national territories and the institutionalisation of their increasing disparities: the increasing localisation of the decision-making processes tends not only to consolidate differentiated practices, but also the development of differentiated regulations. This legitimises the development of local models which might further differentiate along socio-economic cleavages, institutionalising uneven treatment (see Chapter 11 herein); b) the territorial coordination of the actors involved in the policy: the multiplication of both private and public actors at the different territorial levels triggers the need for coordination and – at the same time – opens up opportunities for discretion, for potential conflicts and policy implosion (reciprocal vetoes, inability of taking relevant decisions, policy stagnation, ...) (see Chapter 12 herein); c) the spread of blame-avoiding strategies implicit in rescaling processes: often the reduction of financial transfers from the central government forces subnational governments to raise local resources in order to finance their own social programmes. This “decentralisation of penury” (Keat- Rescaling Social Policies towards Multilevel Governance in Europe ing, 1998) has been driven not only by functional pressures but also by political calculations. “Passing the buck” to sub-national governments has the obvious advantage of shifting the blame for unpopular policies of retrenchment onto local governments (Ferrera, 2005: 173-174); d) the accountability of the decision-making process: the multiplication of actors and their territorial fragmentation tends to weaken the democratic control over actors’ responsibilities within the decision-making process, management and implementation (Crouch, 2003; DiGaetano and Strom, 2003; Benz, 2007; Bovens, 2007; Brodkin, 2008). These critical points are related to one another and might be summarised by the risk of falling into the local trap of urban democracy (Purcell, 2006), meaning the uncritical praise of subsidiarity and the tendency to assume a priori that the local scale is preferable to other scales. However, the extent to which this risk occurs in the different European countries and the forms it takes depends on the interplay between the intra-national structural socioeconomic divides and the socio-political specificities and reforms, which grant different degrees of freedom to territories and actors in a given country. Table 4 shows some of the striking differences existing in the European countries considered in this book in terms of intra-national variations taking place at the NUTS 2 level for some selected (mainly) labour market indicators. The total employment rates as well as female employment rates – for instance – are evenly distributed in Scandinavian countries, while very unevenly distributed in Southern Europe: Italy displays dispersion rates that are 10 times higher than in Norway, just to mention the two extremes. We do not go into the reasons for these differences, but we underline their importance in channeling the impact of the processes of change. If we want to assess the impact of rescaling and the multiplication of actors we must, therefore, consider these degrees of variation and the way they interact with the redistributive capacities of social policies and other relevant contextual indicators we have presented in this introduction. The chapters in the first part will do this by considering the three social policy areas under investigation, while the chapters in the second part will address some of the critical issues raised above considering these contextual elements in a cross-country perspective. In synthesis, what we need to understand is the synergic interplay between processes of change and the regulatory contexts within which they take place: similar processes of change filtered by different regulatory frameworks might produce divergent impacts (see 67 Yuri Kazepov 68 synopsis in the appendix to this chapter). As we have seen, within this frame, sub-national bodies (municipalities, but most prominently regions) become increasingly important. The strong accent that recent reforms have put on devolution, decentralisation and active welfare policies has provided them with new regulatory powers which, in a framework of overall fragmentation, have brought about the need for coordination of the increased number of different actors. The degrees of autonomy regions (or other sub-national bodies) have and the resources at their disposal, however, still depend very much on the overall regulation at the national level. In fact, in most countries social policies retain a dual territorial nature which we should keep in mind. They are both national and local: passive contributory-based policies (e.g. unemployment benefits) are still mainly defined at the national level in almost all countries considered, while activation policies and in-kind provisions are increasingly defined at the local level. It is for this very reason that the nation-state’s influence on local policies is still relevant, in particular in relation to redistribution, which still plays an important role in most European countries, despite the differences in the redistributive capacity of the different systems (see indicators no. 23-26 in Table 2). Table 4: Dispersion rates of selected indicators at NUTS 2 level, 1999-2007 SE NO FI 1999 2007 4.8 4.4 2.4 1.7 6.7 5.5 1999 2007 5.6 4.3 3.0 2.3 7.4 6.2 1999 2007 28.3 9.2 n.a. n.a. 30.5 37.3 1999 2007 18.9 20.5 19.0 31.6 9.3 14.9 1990 2006 10.4 13.4 9.2 12.7 15.1 11.6 DE FR IT Total activity rates 5.4 7.1 17.4 6.0 7.1 15.6 Female activity rate 6.9 10.0 30.2 5.8 9.0 25.7 Youth unemployment 36.0 21.7 57.3 40.1 20.1 46.3 Long-term unemployment 8.1 14.1 33.9 9.9 12.8 25.8 Elderly dependency * 10.7 20.7 19.5 7.8 18.3 16.2 ES PL UK 10.8 8.7 4.8 6.4 7.5 6.0 17.6 14.8 6.5 7.6 7.3 6.5 29.2 20.7 32.3 14.4 25.4 18.0 17.1 24.9 16.4 15.7 29.4 22.4 4.5 5.0 16.1 9.5 n.a. 7.8 Source: Own calculations on Eurostat online database (2008). Data for Switzerland not available. Notes: (*) Countries of the Rescaling project. Spain: Ceuta and Melilla excluded; France: Guadalupa, Martinica, Guyana and Reunion excluded; Finland: Åland and Lappland excluded. Rescaling Social Policies towards Multilevel Governance in Europe In this sense, the new forms of multilevel governance that are emerging may well be differentiated and partly fragmented, with relevant differences among policy areas, but as long as relevant resources are regulated and redistributed at the national level – and they are still regulated in most European countries at the national level – the degrees of coherence with national welfare systems are higher than one might expect. Of course the rescaling process significantly affects this state of affairs, in particular when the institutional design of provisions and access criteria are decentralised to sub-national levels. This does not mean that management and implementation are less relevant. The pioneering studies by Lipsky (1980) on the role of street-level bureaucracy in modifying, adapting and influencing the outcome of policies through the discretionary power social workers have, should not be underestimated. Nevertheless, the discretion power of social workers is also embedded in different regulatory contexts which offer them different tools and resources. If a social worker has a series of institutional tools to rely upon, i.e. both passive and active ones as, for instance, a guaranteed minimum income which lifts families above the poverty line, he/she can concentrate on the development of a personalised – and eventually ad hoc designed plan involving the claimant – insertion plan. If he/she cannot do that, then one of his or her main tasks will entail the building of creative (but more often precarious and unevenly distributed) solutions for each case. Alternatively, a social worker must witness the demise of the programme due to budgetary constraints. The comparative data presented in Table 5 provide a rather articulated picture of the development of the Gini index measuring income inequality over the last 25 years. Despite its almost generalised increase, rather pathdependent tendencies persist in the diversity among countries. In particular, the countries belonging to the European social model display a lower degree of income inequality than countries, like the US for instance, where the low level of policy intervention exposes people in need to the increased speed of change of the market and where privatised Keynesianism (Crouch, 2008) has not been able to redistribute – almost by definition – resources. Markets change faster than political redistributive institutions, which are more resilient. This resilience, however, offers sub-national bodies the opportunity to be actors for institutional innovation. Slowing down the pace of change, in fact, extends the time available for exploring new solutions. Sub-national bodies become – once again – laboratories of how social policies construct citizenship, social 69 Yuri Kazepov Table 5: 70 Trends in Gini coefficients of income inequality in selected OECD countries, 1985-2005 Mid-1980s Mid-1990s Mid-2000s ∆ 1985-2005 Denmark Finland * Sweden * Norway * 0,221 0,207 0,198 0,234 0,215 0,228 0,211 0,256 0,232 0,269 0,234 0,276 + 0,011 + 0,062 + 0,034 + 0,042 Austria Belgium France * Germany Netherlands 0,236 0,274 0,300 0,257 0,259 0,238 0,287 0,270 0,272 0,282 0,265 0,271 0,270 0,298 0,271 + 0,029 - 0,003 - 0,030 + 0,041 + 0,012 Greece Italy * Portugal Spain * 0,336 0,309 0,329 0,371 0,336 0,348 0,359 0,343 0,321 0,352 0,385 0,319 - 0,015 + 0,043 + 0,056 - 0,052 Hungary Czech Republic Poland * Slovakia Switzerland 0,273 0,232 n.a. n.a. n.a. 0,294 0,257 n.a. n.a. n.a. 0,291 0,268 0,372 0,268 0,276 + 0,018 + 0,036 n.a. n.a. n.a. Ireland United Kingdom 0,331 0,325 0,324 0,354 0,328 0,335 - 0,003 + 0,010 United States Australia Canada Japan 0,338 n.a. 0,287 0,304 0,361 0,309 0,283 0,323 0,381 0,301 0,317 0,321 + 0,043 n.a. + 0,030 + 0,017 OECD-24 OECD-22 0,293 0,279 0,310 0,293 0,313 0,300 + 0,020 + 0,021 Source: OECD Questionnaire on income distribution (September 2008). Notes: (*) Rescaling countries. Rescaling Social Policies towards Multilevel Governance in Europe inclusion and participation (Le Galès, 2002; Kazepov, 2005; Mouleart et al., 2007; Keating, 2009). The real challenge plays out in the definition of who is included and who is excluded and the way in which rescaling and the new governance arrangements affect the outcome of this renewed struggle. From this point of view, the different European countries under investigation provide interesting perspectives into the potentially positive and negative consequences of the subsidiarisation-of-social-policies process ranging from empowering practices and participatory configurations to a differentiated landscape of rights (and duties) which shape the unequal citizen. The devil is in the detail, and exploring some details is what we are going to do in this book. Thanks I would like to thank Marco Arlotti and Eduardo Barberis very much for organising and calculating the data for all tables in this chapter. They should also be thanked for their extensive and repeated feedback on the content of the chapter and their strong support throughout the whole research project upon which this book and introduction are based. Many thanks also go to Tapio Salonen who invited me to Vaxjö University as a visiting professor where I had access to its wonderful library; those electronic resources served as a basis for this introduction. Useful comments have been provided by Alan Scott, Rahel Strohmeier, David Benassi, Enzo Mingione, Renate Minas, Stefania Sabatinelli and Tatiana Saruis. I am also thankful to Walter Hanesch, who invited me to the Expert Seminar Die Zukunft der Sozialen Stadt in Darmstadt (DE), and Jan Galkowsky who invited me to Lublin (PL) to the European Cities conference, where I had the opportunity to fruitfully discuss and fine-tune parts of the issues included here. The usual disclaimers apply. 71 Low-Medium0 Low0 Medium0 Very high0 Local autonomy centrally framed0 Centrally framed0 Regionally framed0 Regionally framed0 Mixed0 Low-Medium0 Regional variation Territorial frame Southern European) (ES, IT)) Transition mixed) (PL)) Scandinavian ) (SE, FI, NO)) Continental National ) (FR)) Continental Federal ) (CH, DE)) Delivery, management and planning0 Delivery, management and planning0 Highly fragmented (managerial, corporative, clientelistic)0 Highly fragmented (managerial, corporative, clientelistic)0 Low0 Low0 Medium0 Delivery, management and planning0 Corporative0 High0 Users’ involvement Medium0 Delivery0 Governance content Delivery and management0 Corporative and statist managerial0 Managerial participative0 Governance type High (passive subsidiarity)0 High (passive subsidiarity)0 High (active subsidiarity)0 Medium0 Very low0 Intermediate bodies’ involvement 72 0 Mainly central0 Regional and local0 Regional (Kanton, Land)0 Central0 Central0 Bargaining level Non-profit/ for profit0 Non-profit0 Non-profit/ for profit0 Non-profit/ for profit0 For profit0 Main private actors Planning and shared management with limitations0 Planning and shared management with limitations0 Shared in planning and management0 Planning and shared management0 Planning and management0 Role of public actor Yuri Kazepov Appendix. Synopsis of the main characteristics of vertical and horizontal subsidiarity