European Research Reloaded: Cooperation and Integration among Europeanized States (Library of Public Policy and Public Administration)

EUROPEAN RESEARCH RELOADED: COOPERATION AND INTEGRATION AMONG EUROPEANIZED STATES

Library of Public Policy and Public Administration Volume 9

General Editor: DICK W.P. RUITER Faculty of Public Administration and Public Policy, University of Twente, Enschede, the Netherlands

The titles published in this series are listed at the end of this volume.

EUROPEAN RESEARCH RELOADED: COOPERATION AND INTEGRATION AMONG EUROPEANIZED STATES Edited by

Ronald Holzhacker University of Twente, Enschede, The Netherlands and

Markus Haverland Leiden University, Leiden, The Netherlands

A C.I.P. Catalogue record for this book is available from the Library of Congress.

ISBN-10 ISBN-13 ISBN-10 ISBN-13

1-4020-4429-1 (HB) 978-1-4020-4429-8 (HB) 1-4020-4430-5 (e-book) 978-1-4020-4430-4 (e-book)

Published by Springer, P.O. Box 17, 3300 AA Dordrecht, The Netherlands. www.springer.com

Printed on acid-free paper

All Rights Reserved © 2006 Springer No part of this work may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, without written permission from the Publisher, with the exception of any material supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work. Printed in the Netherlands.

TABLE OF CONTENTS About the contributors

vii

Preface

xi

1. Introduction: Cooperation and Integration among Europeanized States - Markus Haverland and Ronald Holzhacker

1

I. Europeanization of the Member States - Beyond Goodness of Fit 2. Beyond the Goodness of Fit: A Preference-based Account of Europeanization - Ellen Mastenbroek and Mendeltje van Keulen

19

3. Framing European Integration in Germany and Italy: Is the EU Used to Justify Pension Reforms? - Sabina Stiller

43

4. Explaining EU Impacts at the Domestic Level – The Europeanization of Regulatory Policy in Denmark - Morten Kallestrup

65

5. Aggregating, Mobilizing and Recruiting: EU Integration and Party Functions - Harmen A. Binnema

89

II. European Integration - Integration and Cooperation among Europeanized States 6. The Europeanization of Central Decision Makers’ Preferences Concerning Europe: a Perpetual Motion? - Femke van Esch

119

7. Domesticated Wolves? Length of Membership, State Size and Preferences at the European Convention - Dirk Leuffen and Sander Luitwieler

151

8. Beyond the Community Method: Why the Open Method of Coordination was Introduced to EU Policy-making - Armin Schäfer

179

v

Contents

vi

III. Conceptual Challenges - Territory, Governance and Changing Notions of Sovereignty 9. European Integration and Unfreezing Territoriality: The Case of the European Health Card - Hans Vollaard

203

10. Governing Mobility: The Externalization of European Migration Policy and the Boundaries of the European Union - Rens van Munster and Steven Sterkx

229

11. Sovereignty Reloaded? A constructivist Perspective on European Research – Tanja E. Aalberts

251

NOTES ON THE EDITORS Dr. Ronald Holzhacker is Assistant Professor for political science at the University of Twente and Fellow at University College Utrecht in the Netherlands. He is broadly interested in the impact of the European Union on national democratic processes in the member states. He is published in such journals as Party Politics, European Union Politics, and the Journal of Legislative Studies. He is most recently editor, with Erik Albaek, of Democratic Governance and European Integration: Linking Societal and State Processes of Democracy (Edward Elgar, forthcoming 2006). In 2005 he is Visiting Professor at the University of Paris 1, Sorbonne and is a 20052006 recipient of the Jean Monnet Fellowship to the European University Institute, Florence. He holds a PhD from the University of Michigan and a J.D from the University of Minnesota Law School. Dr. Markus Haverland is Lecturer in Public Administration at the Faculty of Social Sciences, Leiden University. His research interests include European integration and its effects on the member states, comparative politics and public policy, and the methodology of comparative research. He has published in the Journal of Public Policy, the Journal of Europen Social Policy and West European Politics. He graduated in Public Administration Science at the University of Konstanz and took his doctorate at the University of Utrecht. He has been a Jean Monnet Fellow at the Robert Schuman Center, European University Institute (Florence), and Postdoc and Lecturer at the University of Nijmegen.

viii

About the Contributors

NOTES ON THE CONTRIBUTORS Tanja E. Aalberts is a PhD student at the Department of Political Science, Vrije Universiteit, Amsterdam, The Netherlands and is currently completing her doctoral thesis on sovereignty discourses in the context of EU-Europe and Sub-Saharan Africa. She holds an MA in International Relations and International Law (University of Twente, The Netherlands) and an MScEcon in International Relations Theory (University of Wales, Aberystwyth). She has recently published in the Journal of Common Market Studies 42(1), 2004. Femke A.W.J. van Esch is an Assistant Professor at the Utrecht School of Governance in the Netherlands. She is writing a thesis on the formation of national preferences concerning the establishment of the European Economic and Monetary Union. She has, with others, published ‘Defining National Preferences. The Influence of Inter-national Non-State Actors’ in: B. Arts, M. Noortmann, B. Reinalda (eds.), Non-State Actors in International Relations, Aldershot, Ashgate 2001 and ‘Why States Want EMU. Developing a Theory on National Preferences’ in: A. Verdun (ed.) The Euro. European Integration Theory and Economic and Monetary Union, Lanham: Rowman and Littlefield 2003. Harmen A. Binnema is a Ph.D. candidate at the Department of Political Science, Vrije Universiteit Amsterdam. His dissertation is on the impact of EU integration on the organization and ideology of national political parties. Other research interests include Europeanization, legitimacy, and governance. He recently contributed a chapter on the Netherlands in a volume on the OECD and national welfare states, edited by Klaus Armingeon and Michelle Beyeler and published by Edward Elgar, 2004. Dr. Morten Kallestrup is assistant prof essor in public policy and administration at the University of Aalborg, Denmark, and visiting research fellow at the Danish Institute for International Affairs (DIIS). He has conducted research on how the EU impacts on domestic regulatory policies, in particular on the role of domestic politics in processes of Europeanization. His general research interests include Europeanization of domestic politics and policies, regulatory policy, and the study of ‘politics versus markets’ in Europe. He has published books and articles on Europeanization, regulatory policy-making, and tax policy, as well as co-authored a volume for The Danish Power and Democracy Study in 2004.

About the Contributors

ix

Mendeltje van Keulen is a fellow at the Clingendael European Studies Programme and and PhD student at the Centre for European Studies, University of Twente. She holds Master’s degrees in European public administration from the University of Twente and the College of Europe, Bruges and is completing her dissertation concerning the effectiveness of Dutch EU policies. Her research interests include EU policy-making and co-ordination at the domestic level; the Europeanisation of public administration and EU decision making. Recent publications include: Keulen, M. van (2004), ‘What Happens at Home, Negotiating EU-Policy at the Domestic Level’, in: Meerts, P. and F. Cede (eds.), Negotiating European Union, Basingstoke: Palgrave Macmillan. Sander Luitwieler is a Ph.D. student at the Department of Public Administration, Erasmus University Rotterdam, the Netherlands. His Ph.D. research is on the role of member states and EU institutions during IGC negotiations resulting in EU Treaties, particularly the Treaty of Nice. His research interests concern the European integration process in general and EU Treaty formation in particular. Publication: Luitwieler, Sander and Pijpers, Alfred (2006), ‘The Netherlands: From Principles to Pragmatism’, in: Laursen, Finn (ed.), The Treaty of Nice. Actor Preferences, Bargaining and Institutional Choice, Leiden: Brill Academic Publishers. Dirk Leuffen works as a researcher in the European Politics team of the Center for Comparative and International Studies at the ETH Zürich. In addition, he is a PhD candidate at the University of Mannheim. In his dissertation, he analyses French European policy-making in the context of divided government. His research interests include the analysis of political decision-making, the interactions between domestic politics and foreign policy-making, European Union and French politics. His work is published, among other places, in the British Journal of Political Science. Ellen Mastenbroek graduated with honours in Political Science and Public Administration at the University of Nijmegen, the Netherlands. She is doing PhD research at Leiden University on the transposition of European directives in the Netherlands. Other research interests include quantitative and qualitative methods of political science, Europeanization, international relations, and neo-institutionalism. She has recently published an article in European Union Politics on the transposition of EU directives in the Netherlands. Dr. Rens van Munster studied European Integration and International Relations at the University of Twente, the Netherlands. He holds a PhDdegree from the Department of Political Science, University of Southern Denmark, Odense, where he wrote his doctoral thesis on European security

x

About the Contributors

and immigration. In 2003-2004, he was a Marie Curie Fellow in International Political Community at the Department of International Politics, University of Wales, Aberystwyth. He has published an article in the International Journal for the Semiotics of Law. Dr. Armin Schäfer is researcher at the Max Planck Institute for the Study of Societies, Cologne, Germany. Current research interests include: the history and political economy of European integration, international economic policy coordination, comparative politics, and social policy. Latest publication: ‘A New Form of Governance? Comparing the Open Method of Coordination to Multilateral Surveillance by the IMF and the OECD’, Journal of European Public Policy, forthcoming. Steven Sterkx holds a graduate degree in Political and Social Sciences and a European Master’s degree in Human Rights and Democratization. As a Ph.D. candidate for the Fund for Scientific Research (Flanders), he is currently doing research at the Department of Politics, University of Antwerp, Belgium. His Ph.D. research concerns the asylum and migration policy of the European Union, and in particular its external dimension. A recent publication is ‘The comprehensive approach off balance: externalization of EU asylum and migration policy’, in PSW Paper, 2004/4, Antwerp: University of Antwerp. Sabina Stiller is a junior researcher and PhD candidate at the Department of Political Science, Radboud University of Nijmegen, The Netherlands. She holds a B.A. in European Studies and Spanisch and a M.A. in International Relations. Her research interests include social policy change (both at domestic and EU-level), path-breaking welfare state reform, political leadership and the impact of political ideas. In her Ph.D. project, she looks at explanations for recent structural reforms of the German welfare state. She has written ‘Germany and the Turkish wish to join the EU: Get them in or keep them out?’ Jason Magazine 28 (1), 2003. Hans Vollaard is junior lecturer and co-ordinator of the EU-studies program in the Faculty of Arts, and a Ph.D. candidate in the Department of Political Science at Leiden University, the Netherlands. He studied political science in Leiden between 1995 and 1999. His research project deals with political territoriality and European integration in the policy areas of healthcare and security. In 2005, he co-authored and co-edited a volume on euroscepticism in the Netherlands.

PREFACE The Three Waves of European Research European cooperation and integration has continued to progress forward over the past five decades, with an ever deepening impact on the member states. The first wave of research into these processes concerned European integration, the process of institution building and policy developments at the European Union (EU) level. The second wave, on Europeanization used integration as an explanatory factor in understanding domestic political change and continuity related to the EU. What is now necessary is to link our understanding of these ‘bottom-up’ and ‘top-down’ processes of integration and Europeanization in the EU. This book argues that a third wave of research on the EU is needed to adequately understand the increased interconnectedness between the European and national political levels. We posit that this third wave should be sensitive to the temporal dimension of European integration and Europeanization. In particular, we seek to link the processes of European integration and Europeanization in a new way by asking the question: how has Europeanization affected current modes of integration and cooperation in the EU? Part I. Europeanization of the Member States. Preparing the ground for the third wave, the first part of the book concerns Europeanization. In order to fully understand the feedback of Europeanization on cooperation and integration it is important to analyze how European integration has had an impact on member states in the first place, in particular indirectly, beyond the direct mechanism of compliance with European policies. The research presented here stresses the role which domestic actors and in particular national governments have in utilizing indirect mechanisms to their advantage, hence guiding the Europeanization impact on the member states. Part II. European Integration. The second part of the book concerns integration and cooperation, in line with what we see as the third wave of research. Here we analyze how prior integration effects, that is Europeanization, influences current preferences for integration. We find that earlier integration effects have had a significant influence on those preferences. This has resulted, perhaps somewhat surprisingly, not always in a preference for closer integration, but instead for new forms of looser cooperation between the member states. Part III. Conceptual Challenges. The multi-faceted interrelationships between the EU level and the national level and the increased interconnectedness between them, cast doubt on the appropriateness of traditional readings of central concepts of political science and international relations such as territory, identity and sovereignty. The final section of the book therefore concerns the conceptual challenges faced by the continued development of multi-level governance. These contributions show that a xi

P reface

xii

conceptual reorientation is necessary because up until now these concepts have been almost exclusively linked to the nation state. One of the key findings of the book is the astonishing variation in modes of cooperation and integration in the EU. We suggest that this variation can be explained by taking into account the sources of legitimacy at the national level and at the EU level on which cooperation and integration are based. We argue that whereas economic integration, in particular the creation of a single market, could be sufficiently backed by output legitimacy, deeper integration in other areas requires a degree of input legitimacy that is currently lacking in the EU. Therefore, non-economic integration is often taking forms of looser types of cooperation, such as the open method of coordination and benchmarking, allowing domestic actors more control over the Europeanization of these policies onto the member state. We elaborate on this speculation in the conclusion and believe that it should be part of the future research agenda of the third wave of European research.

About the European Research Colloquium This book emerged from the European Research Colloquium (ERC) of the Netherlands Institute of Government (NIG), which was founded by the editors of this volume in 2002. A small group of researchers from the Netherlands, Germany, Belgium, and Denmark met every six months over the past three years to debate substantive topics, the choice of research design and methodology, and, in particular, the empirical research presented by each author in this book. The ERC offers secondary mentoring to PhD students researching and writing on topics related to the European Union and European comparative politics. During each two-year period, a small group of 14-16 PhD students meet twice a year to discuss their comparative European and EU research with senior scholars from NIG. NIG is a network of eight political science and public administration departments from Universities across the Netherlands. The ERC has the following objectives: • Improve the quality of EU and European comparative PhD dissertations by focusing attention on research design, methodology, and theoretical innovation of the students’ research. • Build a cohort of young researchers stretching across Europe to build the next generation of comparative scholars who will know and cooperate with one another now and in the future.

P reface

xiii

• Create a book length manuscript, consisting of chapters written by each PhD participant, to share the results of the colloquium with the broader academic community. The five 3-day conferences of the group were held at Erasmus University Rotterdam, the University of Twente, University of Nijmegen, University of Utrecht (University College Utrecht and the Utrecht School of Governance), and for the final meeting we returned to Rotterdam. The following senior scholars met with the PhD students in small groups to discuss their research with them during our meetings at the different member institutions of the Netherlands Institute of Government. We extend our gratitude and thanks to them. The students greatly benefited from their wisdom and advice. We thank Tanja Börzel, Peter Geurts, Henk van der Kolk, Andre Krouwel, Bob Lieshout, Sebastiaan Princen, Frans van Waarden, Jaap de Wilde, Bertjan Verbeek, and Kutsal Yesilkagit. We would also like to thank the NIG for their encouragement and guidance in launching the colloquium, including the previous management team at the University of Twente, Jacques Thomassen, Oscar van Heffen, Herman Lelieveldt, Marcia Clifford and Marie-Christine Prédéry, as well as the present one at the Erasmus University Rotterdam, Christopher Pollitt, Sandra van Thiel, and Vicky Balsem.

INTRODUCTION: COOPERATION AND INTEGRATION AMONG EUROPEANIZED STATES Markus Haverland and Ronald Holzhacker European integration has come along way since early visionaries such as Jean Monnet set forth the basic idea of Europe. The three communities formed in the immediate post-war period, the European Coal and Steel Community (ECSC, 1951), the European Atomic Energy Community (Euratom, 1957), and the European Economic Community (EEC, 1957), were limited both in the scope of their supranational decision-making and the resulting impact onto the member states. More recent integration efforts, those memoralized in the Single European Act (1986) and the treaties of Maastricht (1993), Amsterdam (1999), and Nice (2000), established the basis for intensive intergovernmental and supranational decision making in a whole range of policy areas. This evolving process of European integration has had a deep, although varied, impact on the member states. However, the process of integration and Europeanization has not continued uniformly over the past decades. There appears to have been a dramatic change in the relationship between the EU and its member states in the late 1980s and early 1990s. This is a time which corresponds to the fundamental completion of the Single Market and the dramatic developments in Central and Eastern Europe. “The decade from 1985 to 1995 was a watershed in the political development of the EU, for it introduced more intense public scrutiny of European decision-making, more extensive interest group mobilization, and less insulated elite decision-making. The period beginning with the Single European Act and culminating with the decision to establish economic and monetary union created the conditions for politicized-participatory decision making in the EU by increasing the stakes of political conflict, broadening the scope of authoritative decision making, opening new avenues of group influence, and creating incentives for a quantum increase in political mobilization.” (Hooghe and Marks, 2001: 126). This political development has intensified the interconnectedness between the EU and the national level, a phenomenon now widely referred to as multi-level governance (Hooghe and Marks 2001, Kohler-Koch 2003), that has raised – among other things – new concerns about the democratic legitimacy of the European project. It is our conviction that the intensified interaction necessitates bringing the two major strands of research on the European Union together, both the European integration (bottom-up) and the Europeanization of the member states (top-down) perspectives. There have been considerable efforts to explain these processes individually. This book seeks to begin to consider how these two processes can be seen as systematically related processes theoretically and explored through empirical research. The idea is to begin to have a greater appreciation for the development 1

R. Holzhacker and M. Haverland, (eds.), European Research Reloaded: Cooperation and Integration among Europeanized States, 1-17. © 2006 Springer. Printed in the Netherlands.

2

Haverland and Holzhacker

of multi-level governance over time, as a kind of cork-screw rotating continuously with top-down and bottom-up processes of interaction between levels of governance. We see this book as being at the turning point towards a new third wave of research on the EU. The first wave concerned European integration, the process of institution building and policy developments at the EU level. Most of the early thinking in this area was done from an international relations perspective, discussing the interaction of the member states. Later thinking on integration concentrated further on the development of the EU institutions themselves, and involved a wide range of perspectives from institutional thinkers, legal scholars, economists, and policy analysts (Haas 1957, Lindberg and Scheingold 1971, Moravscik 1999, Stone Sweet and Sandholtz 1998). The second wave of research, on the Europeanization of the member states, gained prominence a decade ago and has since exploded with great vigor. This research direction uses EU integration as an explanatory factor in understanding domestic political change and continuity. Here the comparativists came to the fore and sought to compare the impact which the EU and European integration has on the domestic politics of the member states (Cowles, Risse, and Carporaso 2001; Featherstone and Radaelli 2003; Goetz and Hix 2001, Mény, Muller and Quermonne 1996). Though the scholars of the second wave have been able to identify factors and mechanisms that shape the adaptation and non-adaptation of the member states to the European Union, there are still open questions, in particular with regard to mechanisms at work that are more indirect than the direct pressure of institutional compliance enforced by the Commission and the European Court of Justice. Therefore, the first section of the book moves beyond the ‘goodness of fit’ explanations by focussing on indirect mechanisms such as changing opportunity structures and framing. It is likely that member states have not always been aware of these indirect effects of European integration when they decided to delegate competencies to the EU level in the past. This might have led to unexpected Europeanization experiences which now shape member states preferences towards further cooperation and integration. It is therefore important for the focus of this book to analyze these indirect effects in greater detail. Building on the knowledge concerning the significance of the indirect mechanisms, the book then turns to the third wave of research. Here the book draws on the traditional strength of comparative politics and international relations. Some theoretical work has just begun to call for ways to combine bottom-up theories of integration with the top-down theories of Europeanization. Börzel states that “a comprehensive understanding of the relationship between the member states and the European Union requires the systematic integration of the two dimensions,” although she then proceeds to “mainly adopt a top-down perspective” (2005). Börzel and Risse have also discussed the two perspectives and then state “As far as the European Union

Introduction

3

is concerned, we will get a more comprehensive picture if we study the feedback processes among and between the various levels of European, national, and subnational governance,” but then again take a ‘top-down’ perspective (Börzel and Risse, 2003: 57). In the same volume, Wincott states that “In the final analysis research on Europeanization should filter back into our understanding of what the European Union is (on an abstract level) and might even influence the process of European integration itself.” (2003: 282). We try here to move this debate further. There are of course reasons, especially critical in empirical research, to restrict oneself to either a bottom-up or top-down approach to research on the EU and the member states. If both the dependent variable and the independent variable are changing at the same time, any hope of isolating causal factors is lost. But if we truly want to ‘understand the nature of the beast’ (Risse-Kappen 1996), the evolving multi-level political system of the EU and the member states, we must also acknowledge the interactive processes that feed back onto themselves. We argue that it is possible to circumvent the problem by separating Europeanization and European integration chronologically. In other words to take a temporal approach and by explicitly asking – how has the Europeanization experience impacted on (the preferences for) modes of cooperation and integration among the member states? We suggest three mechanisms which drive European integration and cooperation forward: socialisation, path dependency, and learning. Before we elaborate on these mechanisms, we turn to the first section of our book dealing with the Europeanization of member states.

EUROPEANIZATION OF THE MEMBER STATES The focus of the book is on the turning point toward the third wave. This implies that we still need to revisit some issues of the second wave before we are able to move to the third wave. The reason being that in order to analyze the question whether and how earlier Europeanization has affected (national preferences for) current modes of cooperation and integration it is necessary to have a good understanding of the extent to which the EU has impacted on the member states and the underlying mechanisms at work.1 So far, many studies have focused on the goodness of fit between European requirements and the national status quo as an explanation for domestic 1

There are many different usages of the term Europeanisation (for overviews, see Eising 2002, Olsen 2002, Radaelli 2002). Europeanisation is here broadly defined as the effect of European integration on the member states. European integration is characterised by two interrelated processes “the delegation of policy competencies to the supranational level to achieve particular policy outcomes; and the establishment of a new set of political institutions; with executive, legislative and judicial powers” (Hix and Goetz, 2000: 3).

4

Haverland and Holzhacker

policy change. However, empirical research has documented that the degree of Europeanization is not simply a function of the goodness of fit. It is true that at least some misfit is necessary for any adjustment to EU requirements to take place: without misfit, adjustment is not logically possible. But it is not true that misfit is a necessary condition for (any) domestic change caused by the EU. As the case of the French transport policies in the 1990s illustrates, EU induced domestic change is possible even if there is a close fit between EU objectives and national status quo, if this change is divergent from EU objectives (Knill and Lehmkuhl 2002: 261-2). This is a process that has been called ‘retrenchment’ (Börzel and Risse 2003). Apart from this conceptual issue, it has been demonstrated that the explanatory power of the goodness of fit hypotheses is rather weak. The hypotheses that the larger the misfit, the less adjustment observed, has not been sufficiently supported by empirical research. Hence in many instances no adjustments occurred even in case of low misfit, while adjustment has taken place in cases of relatively large misfit (see for instance Haverland 2000, Knill and Lenschow 1998). The contributions in this section therefore go beyond the goodness of fit explanation by focusing on political preferences and indirect mechanism of Europeanization, in particular on changing opportunity structures and the framing of discourses. Changing opportunity structures. The establishment of a new set of institutions at the European level with legislative, executive and judicial powers provides actors with a new layer of access to political decision making. The EU creates new exit, veto, and informational opportunities for domestic actors resulting in a redistribution of powers and resources of public and private actors in the member states (Hix and Goetz 2002, see also Börzel and Risse 2003; Knill and Lehmkuhl 2002). For instance, the single market increases the leverage of export-oriented business at the expense of import-competing firms (see for the general argument Keohane and Milner 1996) and arguably also the power of business interest association vis-à-vis the representatives of diffuse interests, such as the environment (KohlerKoch 1996). Framing. The other indirect mechanism refers to the cognitive impact of European integration and issues of framing. European integration may alter the beliefs and expectations of domestic actors within a given opportunity structure shaping their preferences and strategies. European integration may even lead to cognitive convergence, for instance about the appropriate mode of governance (Börzel and Risse 2003, Kohler-Koch 2002, Knill and Lehmkuhl 2002; Radaelli 2000). We now turn from this general discussion of the possible causal mechanisms for Europeanization to the empirical research in this volume. The first section of the book begins with a contribution by Mastenbroek and Van Keulen that gives a comprehensive criticism of the notion of goodness of fit and than provides an alternative theory arguing that the fate of European policies depends on government preferences rather then the goodness of fit.

Introduction

5

They chose two cases from the Netherlands, the transposition of the 1998 gas directive and the transposition of the Biotech directive, which allow them to test both explanations in a competitive fashion. Next Stiller focuses more explicitly on the two indirect mechanisms elaborated above and analyses whether EU related arguments have been used to justify change in welfare state arrangements. She looks in particular at pension reform and employs a most similar system design, selecting Germany and Italy, two countries with similar pension systems and problems and a comparable number of veto players. In a similar vein, Kallestrup combines the mechanisms of framing and changing opportunity structures and asks to what extent and how Danish politicians and other actors make use of the EU in policy-making processes at the domestic level. Using the technique of process tracing, he looks in particular whether Danish politicians have conceived and assessed EU policies at the domestic level and how they made use of EU policies and pressure to justify reforms to competition law and consumer protections laws in their country. We now turn from our attention on policies to political parties. There is a certain bias in the Europeanization literature to study the effect of the EU on domestic policies, rather than on input processes such as cleavages, parties, and patterns of democratic legitimization (Hix and Goetz 2000: 15). However, in order to assess to what extent the Member States have Europeanized it is also important to evaluate whether and to what extent the EU had an effect beyond policies (political output). Binnema deals with one of the most important mass-elite linkage institutions in democracies, political parties. He analyses whether the EU had an effect on the three most important functions of political parties: aggregation, mobilization and recruitment. He does this for three different periods: around 1970, 1985 and 2000, first of all by comparing the 15 member states on a more general level with other OECD countries, and then by analyzing in greater detail The Netherlands, Denmark and Austria.

EUROPEAN INTEGRATION AND COOPERATION Moving from the second wave to the third wave, the second section of the book deals with the question whether and how Europeanization has in turn an effect on (the preferences for) future cooperation. It is important to note that our approach should not be conflated with neo-functionalist reasoning. To be sure, our attempt to link Europeanization and integration

6

Haverland and Holzhacker

might sound reminiscent of the feedback loops and spill-over effects of the much criticized neo-functionalist approach. Neo-functionalism conceived integration as moving forward from its own dynamic, albeit in fits and starts (Haas 1958; Mitrany 1966, Schmitter 1969). Integration was to occur through two kinds of ‘spillover’, functional and political. Whereas functional spillover referred to the interconnectedness of various economic processes with other societal processes and between issue areas, political spillover referred to how supranational organizations tend to generate a selfreinforcing process of institution building. But here we are not predicting a specific outcome. In particular, we do not posit that there is a process toward an ever closer European Union. Taking a historical perspective on the development of the EU does not automatically lead to a specific expectation about the possible outcome (Schmitter 2002). The scope of governance and the level where decisions are reached in a multi-level system can ‘contract as well as expand’ (Hooghe and Marks 2004: 5). It also does not imply a particular causal mechanism at work. The Europeanization experience might feedback in various ways and with various effects. Below three different mechanisms are discussed drawing from insights from international relations and comparative politics. Socialization. Taking a sociological institutionalist perspective, one can hypothesize that over time member states have been socialized into European norms and have developed a European identity which results in a shared “European” understanding of interests, the problems at stake, and legitimate and workable solutions (Checkel 1999; March and Olsen 1989). In a similar vein Falkner argues “…preference formation is not necessarily exogenous to European integration…we cannot adequately understand EU treaty reform (and, indeed, European integration more generally) if we exclude instances of EU-level socialization and the institutionalization of policy paradigms” (2002, p. 8). This mechanism implies further integration. Path dependency. It is also possible to take an actor-oriented rational choice approach while being sensitive to the temporal dimension of politics. An actor-oriented historical institutionalist perspective would lead us to again hypothesize ever closer integration as policy makers are increasingly constrained by the legacy of consequences of earlier effects of European integration which were often unforeseen and beyond their control (Pierson 1996). Likewise, Kohler-Koch argues that the effect of the deepening of earlier integration has increased the costs of non-decisions (Kohler-Koch 1996: 302). Learning. It may also be the case, however, that the current situation represents a critical juncture, a time at which it is possible to deviate from well-trodden paths. In addition, as attempts of cooperation moves into new areas, the path dependency argument is not likely to be applicable. Governments may desire more room to maneuver in dealing with experiences with European integration. In weighting the costs and benefits of various alternatives of cooperation and integration, member states might

Introduction

7

learn from past experiences of unexpected, indirect Europeanization beyond their control, and thus might now be more cautious when opting for supranational modes of integration. As research on these mechanisms in the context of the EU is still in its infancy, we will not predict which of these mechanisms is dominant. However, the book is based on the assumption that the member states themselves have been transformed in the process of European integration, resulting in a change of preferences toward what they desire at the EU level and how the central institutions of the EU should develop further. This second section of the book begins with a chapter by Van Esch focusing on the preferences of the German and French heads of government and ministers of finance concerning the establishment of a European economic and monetary union. Using cognitive mapping she analyses in particular their world views related to EMU. She analyzes the 1970s and the 1990s which allows her to compare periods with low and high degrees of European integration. She analyses whether and to what extent earlier effects of European integration have become part of their ‘cognitive map’. Turning from the impact of earlier integration effects on the world view of (members of ) governments to its impact on revealed government preferences, Leuffen and Luitwieler study whether the length of membership in the EU impacts on member states’ preferences with regard to five crucial issues during the European Convention. They argue that due to socialisation, older members will favour a more integrationist (or supra-national) institutional design, while new members are more in favour of a design safeguarding their national interests. They test their sociological institutionalist argument against a hypotheses derived from rational choice institutionalism. This hypothesis states that the size of the country explain its preference with regard to these issues. The authors look at the preferences of all 25 members of the convention and test their hypotheses quantitatively. While Leuffen and Luitwieler take length of EU membership as a proxy for the experience of earlier integration effects, Schäfer analyses more concretely a sequence of Europeanization and new modes of cooperation. Taking an actor-oriented historical institutionalist approach that takes into account government preferences and learning effects, he seeks to explain why the open method of coordination (OMC) was introduced to EU policy making. The OMC is a – currently much discussed – alternative to the traditional community method. He analyzes to what extent and how the effects of prior integration, in particular EMU, have limited available choices in the area of employment policy and later in other socio-economic policy areas.

8

Haverland and Holzhacker

CONCEPTUAL CHALLENGES Overall, the chapters of the book document an increased interrelationship between the EU level and the domestic level. This in turn has implications for the reconfiguration of political power and rule internally within, but also externally outside the European Union. In an attempt to address the latter issue, the third wave of European studies calls for a renewed exploration of the way scholars in EU-studies have traditionally imagined the concepts of identity, boundaries and order through the lens of sovereign states. The book therefore concludes with a conceptual reorientation. In a sense, then, the third wave of EU-studies reopens for consideration the concerns of the first wave of theorists, who explicitly dealt with issues of sovereignty, identity, and territoriality. For instance, functionalists like Mitrany (1966) expected territorial orders and identities to fade out in favor of functional polities. Neo-functionalists and federalists foresaw a supranational Euro-state divided into functional and territorial subunits, while realists and liberal intergovernmentalists assumed the nation state would remain. However, this first wave of EU-studies effectively narrowed the question of sovereignty down to a simple yes or no answer, consistent with traditional international relations theory. However, such an approach seems too coarse for analyses that seek to trace the more fine-grained changes in the meaning and significance of sovereignty in arresting the thinking about identity, boundaries and orders. The subsequent studies of Europeanization have often taken boundaries between the EU and its member states and the EU and its environment for granted. Thus, they cannot account for shifts in territorial rule and the changing nature of boundaries. The third wave of EU studies should therefore put forward the conceptual and theoretical devices to address the dynamic interplay between boundaries, identity and order. Thus the final section of this volume on the next generation of EUstudies subsequently deals with the impact of European integration and Europeanization on internal boundaries (Vollaard), external boundaries (van Munster and Sterkx), and the guiding principle of Europe’s territorial order – sovereignty (Aalberts). Vollaard illustrates for the case of the European health card how the unfreezing of Member States’ borders may lead both to the political reconfiguration of member states and the European Union. Van Munster and Sterkx evaluate the ways in which the externalisation of European migration policies exports the Union’s structures of governance beyond its member states. Aalberts concludes the section with a socialconstructivist argument on how European integration and Europeanization result in the changing meaning and significance of sovereignty.

Introduction

9

METHODOLOGICAL CONSIDERATIONS Regardless of whether authors focus on the effect of the EU on the member states or the effect of previous instances of Europeanization on (preferences) for future integration, they all demonstrate both a high level of methodological reflection and rigor in method application to ensure valid and reliable results. This methodological awareness is a response to recent claims that theoretical progress in EU studies should be matched with higher methodological consciousness (Andersen 2003; Anderson 2003; Haverland 2005). The authors of the (comparative) case studies in this volume (Mastenbroeck/Keulen, Stiller, Kallestrup, Binnema, van Esch) are fully aware that the problem of (internal) validity looms large in small ‘n’ designs. Therefore each of them follows a carefully constructed most similar systems design. Cases are selected intentionally in a manner that ensures that potentially theoretically meaningful variables are held ‘constant’ to rule out that they confound the causal effect of the the independent variable in which the author is interested (Frendreis 1983; Lijphart 1971). At the same time the cases exhibit a maximum variation of the independent variable which prevents the problem of selection bias (King et al. 1994). The author of the single historical case study (Schäfer), carefully identifies a number of observable implications from the two theories he studies and then proceeds through pattern-matching, taking into account the timing and sequences of events (Yin 1994). In order to assess the overall impact of the European integration and Europeanization respectively, some case oriented researchers ask the counterfactual question: what would have happened in the domestic area of interest in the absence of the EU (see also Anderson 2003, Haverland 2003, 2005). This additional device to increase the internal validity is necessary as it is difficult to establish the strength of causal effect of the European Union when focusing on indirect mechanisms. Europeanization research is not always aware of this difficulty and generally suffers from a bias towards EU level explanations. Domestic change and continuity is too quickly attributed to the European Union; alternative explanations are not taken into account. Large ‘n’ research has typically less problems with internal validity. The authors that use a quantitative large ‘n’ design (Leuffen and Luitwieler) therefore reflect in particular on concept validity which is often at risk when relatively ‘simple’ quantitative indicators have to be developed to test relatively complex theories. Regardless of the research design chosen all authors pay due notice to the issue of replicability and reliability. They are all explicit about their hypotheses, their operationalization, the source of their data and the methods of data analysis. In order to avoid measurement errors and bias, they typically tap divergent sources in a triangulative fashion (King et al. 1994; Yin 1994).

10

Haverland and Holzhacker

CONCLUSION In lieu of a separate chapter at the end of this volume, we would like to summarize our main results here. First, we will emphasize the results that emerge from the rich empirical research of the chapters contained in the three sections of the book and indicate how this research has begun to answer our main question – how has Europeanization affected current modes of integration and cooperation among the member states? Secondly, we will use our cork screw model to speculate as to why we think the modes of integration and cooperation may have changed over time. This will lay out a fruitful area for future research. In the first section of the book on Europeanization, Mastenbroek and van Keulen have tested the dominant explanation of ‘goodness of fit’ for explaining policy change, against their own explanation based on government preferences. Their results are in line with the preference-based explanation. The Netherlands was in favor of the Gas directive and transposed it successfully, despite a large misfit, while the transposition of the unwanted Biotech directive was a failure, though here the misfit was much smaller or even non-existent. Next, Stiller’s study on the framing of pension reform in Italy and Germany found that domestic actors in Italy used the EU to justify reforms, whereas actors in Germany did not use the EU in this way. She points towards the general popularity of the EU as an important scope condition for framing to occur. The Italian government could tap into a high level of support for the EU, whereas the German government could not. In line with the Italian pension case but contrary to the German case, Kallestrup finds for the case of Danish consumer and competition policy that domestic actors made intensive use of the changed opportunity structures afforded by the EU. The domestic actors pointed to Europe even in the case of national competition policy, even though no direct EU obligations exist in this policy area. The EU was used as an argument for instance to prolong the decision-making process and to delimit the scope of the debate. He also finds that the selective use of the EU as an argument is more important than ‘goodness of fit’ when explaining the reform of regulatory policy. As in the Dutch case (Mastenbroek and van Keulen) the degree of misfit is not correlated with the degree of adaptation. Finally, Binnema finds that the impact of the EU on parties is rather limited, arguing that the pressure for change is weak and that there are no strong mechanisms linking the national level and the EU. Crucially, parties hardly refer to EU issues during national party competition and recruitment is still controlled by the national political parties. European integration might have contributed to the trend of declining membership, however. This Europeanization research allows for the following conclusion: the goodness of fit between EU requirements and the national status quo does

Introduction

11

not account for cross-national variation in adaptation to the EU. Rather, the preferences of governments are a crucial factor. Governments seem to utilize the indirect mechanisms to their own advantage. More concretely: governments get their way by strategically and selectively using the new opportunity to frame a topic in a ‘European’ way. Turning to our next section on European integration and cooperation, van Esch’s chapter using cognitive mapping to compare the worldviews of domestic actors in the late 1960s with the late 1980s regarding EMU, provides support for the hypothesis that as European integration moves forward, the worldviews on which central decision-makers base their preferences have become more Europeanized. Interestingly though, the inclusion of earlier EU integration effects in the cognitive map does not automatically lead to a pro-European stance. The French president Georges Pompidou was strongly against the establishment of (true) EMU but his view is more Europeanized (i.e. more informed by past EU experiences) than that of the pro-European advocate Valery Giscard d’Estaing. This suggests that sometimes learning rather then socialisation mechanisms – as discussed earlier in this introduction – are at work. The results of Leuffen and Luitwieler suggest, however, that generally speaking EU membership does socialize governments into a more proEuropean stance. Their analysis with regard to crucial issues of the European Convention reveals that the length of membership impacts state preferences for further integration. They conclude that the longer a country is a member of the EU, the more it is socialized into the EU and that this is reflected in their preferences for the institutional design for integration. More specifically, they found that long-standing members favor more democratic voting rights in the Council of Ministers based on population, whereas newer members favored institutional designs that emphasize the equality of states. Their alternative explanation drawing from rational choice institutionalism that state-size determines preferences did not receive much empirical support. That Europeanization is not always a question of socialization is however supported by the results of Schäfer’s study on the Open Method of Coordination. He finds that the decision for the OMC is a result both of constraints set by EMU (hence an earlier instance of EU integration) and the political color of governments across the EU. This indicates that not only the policy aims but also the mode of integration in the EU is impacted by prior integration. Thus, while a conservative – liberal coalition at Maastricht created a mode of decision-making for fiscal and monetary policy in the EU to constrain successive national governments, a social democratic majority at Amsterdam relied on soft law to promote its goals in employment and social policy. By implication, this latter mode of integration avoided sovereignty losses for national governments, and helped maintain the control of national actors.

12

Haverland and Holzhacker

The final section of the book turns to re-evaluate central concepts of EU studies and international relations as a result of the third wave of research linking bottom-up and top-down approaches. Vollaard, in his research on healthcare, finds that multi-level governance is beginning to break down conceptions of territoriality in critical components of the social welfare state by giving citizens the option of temporary exit to receive care abroad. He writes that the conception of territoriality is then re-emerging at the EU level. In contrast, Munster and Sterkx find that the EU’s externalization of migration policy, pushing policies out by encouraging countries which border on the EU to tighten border controls, constitutes a deterritorialized response to migration issues. Finally, Aalberts finds that the constructivist perspective toward the mutual construction of structure and agency helps member states maintain the principle of sovereignty within the realities of multi-level governance structures. One of the general research findings of the book is the great importance of national governments for understanding the link between Europeanization and subsequent modes of integration and cooperation. Many contributions have shown that member state governments are capable of controlling the degree and type of Europeanization and are aware of future Europeanization effects when favoring certain EU decision modes over others. More specifically, Mastenbroek and van Keulen find the preferences of national political actors critical to the successful transformation of directives, and both Stiller and Kallestrup find national actors may creatively use arguments about the EU to press for domestic reforms. Van Esch’s study suggest that the world view of governments has become more Europeanized, but that member states can learn different things from earlier instances of EU integration. Schäfer shows how the political color of the majority of governments across the EU influences the mode of decision making chosen for new areas of integration and cooperation. He explains that this is an attempt to steer the Europeanization impacts onto the member states and avoid a loss of sovereign decision-making. Generally speaking, we found an effect of Europeanization on member-states preferences, but there were different mechanisms at work, sometimes in combination, by which member state governments were able to mediate this effect. The result of this decades long process of Europeanization and integration, and the continued development of multi-level governance, is the beginning of the de-territorialization of some decision-making at the national level. This is shown by Vollaard for healthcare policy. In some policy areas, for example in migration policy researched by Munster and Sterkx, decisionmaking has not only re-territorialized at the EU level, but the EU has begun to push policies onto neighboring countries. Finally, this deepening and integration in multi-level governance, continues to occur while the member states continue to view themselves as sovereign, a constructivist turn in the research undertaken by Aalberts.

Introduction

13

Now we turn to the speculative part of our conclusion, that moves beyond the research results presented here to lay out a fruitful area for future research. One of the main results of the empirical research presented here is that there is not a development towards an ever-closer union in the way neofunctionalists originally imagined, but that national actors continue to play a strong role in determining the degree of integration and cooperation and the resultant Europeanization impacts onto the member states. The dual processes of European integration and Europeanization are thus not continuing in a single trajectory. Europeanization does not automatically result in a single type of European integration driven forward primarily by the supranational institutions of the EU, but instead domestic actors continue to play a strong role in deciding on the mode of integration and cooperation. Now we ask, why has there been such a wide variety of modes of integration and cooperation over the decades long development of the European Union? Let us begin here with the model or heuristic device we briefly mentioned at the beginning of the introduction. One way to conceive of the dual processes of Europeanization and European integration is to imagine a rotating cork-screw lying on its side. Each upward rotation of the cork-screw represents a particular aspect of European integration, the subsequent movement downward represents the resulting impact on the member states – Europeanization. Early integration of the EC in the 1950s and 1960s, for example in Coal and Steel entailed minimal Europeanization impacts on the member states in one policy field. The process though quickened and intensified through the 1970s and 1980s with increased integration and increased Europeanization, leading to the development of a multi-level system of governance. Thus when we look at these dual processes over time it appears that although there are periods of faster and slower rotations, there has been a general trajectory since the earliest stages of the European project until the late 1980’s of successively faster rotations. This represents increased activity for European integration and subsequent Europeanization impacts onto the member states. This process was propelled forward by a general consensus, at least by the national actors in the governments in power at the time, that clear economic gains could be achieved by the creation of a single market and peace and stability on the continent could be achieved by a level of political integration. However, after the basic completion of the Single Market in the early 1990s, many of the newer forms of integration and cooperation look quite different than those in the earlier period. It is as if the corkscrew has been profoundly transformed, and subsequent forward progress was splintered in different directions. Some strands of the rotating corkscrew proceeded with a high degree of integration and high Europeanization impacts. That is the case, for example, for the creation of a single currency among 12 of the 15 countries (see chapter van Esch and also Schäfer). But other types of interaction among the member states have been through much looser forms

14

Haverland and Holzhacker

of cooperation, such as bench-marking and other forms of soft-cooperation between member states (see chapter Schäfer). Here one might imagine a thin line spiraling upwards, representing the agreement to discuss and cooperate, but only a dashed lined spiraling downward, representing the limited and voluntary impact back onto the member states of this degree of cooperation. This represents looser types of integration and more minimal impacts of Europeanization. Finally, we also note the increased use of pioneer groups within the EU, where only certain countries agree to cooperate and integrate their policies. The Schengen agreement on border controls is an example of this. Here we might imagine a dashed spiral line upwards toward the EU and a similar one down again onto the member states, because only some EU countries agreed to this kind of cooperation and its subsequent impacts onto the state. We posit that whereas a quickening spiral of integration on the one hand and Europeanization on the other, characterized the period from the 1950’s through the goal of creating a single market in the late 1980’s, the lack of democratic legitimacy for the EU project has resulted in the trajectory of integration and Europeanization in the 1990’s to be quite different than earlier periods.2 The current period of cooperation and integration among Europeanized states is thus characterized by larger variation and generally looser types of cooperation which allow the member states a greater degree of control over the impact of these policies on the member states – thus their Europeanization. Some of these new forms of cooperation began earlier, but their use has now been extended to many more areas than before. We suspect that the distinction between input-oriented authenticity (government by the people) and output-oriented effectiveness (government for the people) is crucial to an understanding of the current modes of cooperation and integration among Europeanized states (Scharpf, 1999). Given clear overall economic gains of the single market program, outputoriented legitimacy was sufficiently high across Europe to propel the European project forward through the 1980’s. Moreover, (economic) “integration could largely be advanced by negative integration pushed by the Commission and the Court, as it were behind the back of politically legitimized actors” (Scharpf 1999: 71). Furthermore, in addition to output legitimacy, the EU likely also benefited from a general sense of input legitimacy nurtured by a feeling of belonging to the Western political community during the period of the Cold War and the provision of peace and stability. However, the situation seems to have changed. Since the creation of the Single Market in the late 1980s and early 1990s, the mode of decision-

2 This is not to say that their was a strictly linear development towards deeper integration and stronger Europeanization during the first four decades of integration. But this was certainly the general trend. There was no large variation in the modes of integration and cooperation. The community method was the prime mode of policy making.

Introduction

15

making in new areas, such as pensions and social security, unemployment, immigration , and security and defense, is quite different than earlier periods. Decision making in the new policy areas is more likely to entail cooperation between the member states, rather than tight integration through the use of the community method. In these policy areas, there are minimal, or at least not immediately obvious economic gains to integration. Moreover, supranational integration in these areas is more visible to the public and demand positive integration, hence explicit approval by the Council of Ministers and, increasingly, the European Parliament. Thus, tacit approval is no longer sufficient for forward progress to occur, instead an “action consensus is needed among a wide range of divergent national and group interests” (Scharpf 1999: 71). In addition, after the fall of the wall and the collapse of communism in Eastern Europe, perhaps general input legitimacy from a sense of belonging to the Western political community has diminished. For both reasons, nationally elected leaders need new sources of input-oriented legitimacy, at least for areas beyond economic and monetary policy where output legitimacy can be more easily secured. Given the lack of input legitimacy, it is likely that further cooperation and integration in non-economic areas can only be achieved through institutional means that allow national elites to better control, monitor, and evaluate the impact of proposals on their country and thus to better steer the process of Europeanization. Thus the insufficient degree of input oriented legitimacy as decision-making moves from the single market project toward non-economic (positive) integration is a reason for the spiral of integration to be splintering into many different modes. New forms of cooperation, rather than integration, are emerging in the EU, such as benchmarking in socioeconomic policy and enhanced cooperation in defense and security, which are more voluntary and less binding because there is a lack of legitimacy to go further. It follows that only if we see an increase in the input legitimacy of the institutions at the EU level, fostered for example by a strengthening of the European Parliament or increasing the transparency of the decision-making within the Council of Ministers, can the kind of European integration pursued during the peak period of the European project be continued. Without this input legitimacy, other types of decision-making involving cooperation in place of the community method will increasingly be used to give more time for national democratic processes to occur.

BIBLIOGRAPHY Anderson, J. 2003, Europeanization in context, in Germany, Europe and the Politics of Constraint. K. Dyson and K. Goetz, Oxford, Oxford UP & British Academy, pp. 37-53. Andersen, S. S. 2003, ‘On a clear day you can see the EU’. Case study methodology in the EU research. ARENA working paper 16/03. Oslo.Börzel, T. A. forthcoming, How the European

16

Haverland and Holzhacker

Union Interacts with its Member States. Member States and the European Union. S. Bulmer and C. Lequesne. Oxford, Oxford UP. Börzel, Tanja A., 2005, How the European Union Interacts with its Member States, in S. Bulmer and C. Lequesne eds, Member States and the European Union, Oxford, Oxford University Press. Börzel, Tanja A., and Thomas Risse. 2003, Conceptualising the Domestic Impact of Europe. In The Politics of Europeanisation, edited by K. Featherstone and C. Radaelli. Oxford, 55-78. Checkel, J. T. 1999, “Social Construction and Integration.” Journal of European Public Policy 64, 545-560. Cowles, Maria, James Caporaso and Thomas Risse, eds. 2001, Transforming Europe: Europeanization and Domestic Change. Ithaca and London, Cornell. Eising, R. 2003, Europäisierung und Integration. Konzepte in der EU-Forschung. Europäische Integration. M. Jachtenfuchs and B. Kohler-Koch. Opladen, Leske + Budrich UTB: 387416. Falkner, G. 2002, “Introduction, EU treaty reform as a three-level process.” Journal of European Public Policy 91, 1-11. Featherstone, Kevin and Claudio Radaelli, eds. 2003, The Politics of Europeanization. Oxford and New York, Oxford, Frendreis, J. P. 1983, “Explanation of Variation and Detection of Covariation. The Purpose and Logic of Comparative Analysis.” Comparative Political Studies 162, 255-272. Goetz, Klaus and Simon Hix, eds. 2001, Europeanized Politics? European integration and National Political Systems. London, Frank Cass. Haas, E. B. 1958, The Uniting of Europe. Political, Social and Economic Forces, 1950-1957. Stanford, Stanford UP. Haverland, M. 2000, “National adaptation to the European Union. The importance of institutional veto points.” Journal of Public Policy 201, 83-103. Haverland, M. 2003, The impact of the European Union on national environmental policies. The Politics of Europeanization. K. Featherstone and C. Radaelli. Oxford, Oxford University Press: 203-221. Haverland, M. 2005, “Does the EU cause domestic developments? The problem of case selection in Europeanization research”, European Integration online Papers (EIoP) 9, 2, http://eiop.or.at/eiop/texte/2005-002a.htm, (2005). Hix, Simon, and Klaus H. Goetz. 2000 Introduction, European Integration and National Political Systems. West European Politics, Special Issue 23 4, 1-26. Holzhacker, Ronald 2002, “National Parliamentary Scrutiny over EU Issues,” European Union Politics 3 4, 459-479. Holzhacker, Ronald and Erik Albaek, forthcoming 2006, Democratic Governance and European Integration: Linking Societal and State Processes of Democracy. Cheltenham, Edward Elgar. Hooghe, L. & G. Marks 2001, Multilevel Governance and European Integration, Oxford, Rowman and Littlefield. Hooghe, L. and G. Marks 2004, The Neo-functionalist were (almost) right: Politicization and European Integration, In The Diversity of Democracy: A Tribute to Philippe C. Schmitter (forthcoming). King, G., R. O. Keohane and S. Verba 1994, Designing Social Inquiry. Scientific Inference in Qualitative Research. Princeton, Princeton UP. Knill, C. and A. Lenschow 1998, “Coping with Europe, the impact of British and German administrations on the implementation of EU environmental policy.” Journal of European Public Policy 54, 595-614. Knill, C. and D. Lehmkuhl 2002, “The national impact of European Union regulatory policy, three Europeanization mechanisms.” European Journal of Political Research 41, 255-280. Kohler-Koch, B. 1996, “Catching up with change, the transformation of governance in the European Union.” Journal of European Public Policy, 33 359-80.

Introduction

17

Kohler-Koch, B. 2002, “European Networks and Ideas, Changing National Policies?” European Integration online Papers 66. Lijphart, A. 1971, “Comparative Politics and Comparative Method.” American Political Science Review 65, 682-693. Kohler-Koch, B., Ed. 2003, Linking EU and National Governance. Oxford, Oxford UP. Lindberg, L. L. and S. A. Scheingold, eds, 1971, Regional Integration. Theory and Research. Cambridge, Harvard UP. March, J. and J. Olsen 1989, Rediscovering Institutions. The Organizational Basis of Politics. London, Macmillan. Mény, Y., P. Muller, and J. L. Quermonne, eds, 1996, Adjusting to Europe. The Impact of the European Union on National Institutions and Policies. London and New York, Routledge. Milner, H. V. and R. O. Keohane 1996, Internationalization and domestic politics, An introduction. Internationalization and Domestic Politics. R. O. Keohane and H. V. Milner eds., Cambridge, UK, Cambridge UP, pp. 3-24. Mitrany, D. 1966, “ The Prospect of Integration, Federal or Functional.” Journal of Common Market Studies 42, 119-49. Moravcsik, A. 1999, The Choice for Europe: Social Purpose and State Power From Rome to Maastricht, Ithaca, Cornell University Press. Olsen, Johan P. 2002, The Many Faces of Europeanization. Journal of Common Market Studies 40 5, 921-950. Pierson, P. 1996, “The Path to European Integration. A Historical Institutionalist Analysis.” Comparative Political Studies 292, 123-163. Radaelli, Claudio 2000, Whither Europeanization? Concept stretching and substantive change, in, European integration online Papers 48, http://eiop.or.at/eiop/texte/2000. Risse-Kappen, Thomas 1996, ‘Exploring the Nature of the Beast; International Relations Theory and Comparative Policy Analysis Meet the European Union,” Journal of Common Market Studies 34 1, pp. 53-80. Scharpf, Fritz W. 1999, Governing Europe. Effective and Legitimate? Oxford. Schmitter, P. 1969, “Three Neofunctional Hypotheses about International Integration.” International Organization 23, 161-166. Schmitter, P. 2002. “Neo-Neo-Functionalism, Déjà-Vu all over again?” available at the website http://www.iue.it/SPS/People/Faculty/CurrentProfessors/bioSchmitter.shtml. Stone Sweet, A., & W. Sandholtz, eds, 1998, Supranational Governance: The Institutionalization of the European Union, Oxford, Oxford University Press. Wincott, D. 2003, The Idea of the European Social Model, Limits and Paradoxes of Europeanization. The Politics of Europeanization. K. Featherstone and C. M. Radaelli, eds., Oxford, Oxford UP, pp. 279-302. Yin, R. K. 1994, Case Study Research and Method. 2nd edition. Thousand Oaks, London, New Dehli, Sage.

I. EUROPEANIZATION OF THE MEMBER STATES – BEYOND GOODNESS OF FIT

BEYOND THE GOODNESS OF FIT

A Preference-based Account of Europeanization Ellen Mastenbroek and Mendeltje van Keulen1 Abstract:

1.

This paper is concerned with formulating and testing a preference-based explanation of EU implementation. The hypothesis is that, rather than the ‘goodness of fit’ with existing policies, the fit with national preferences predicts the ease of implementation of new EU legislation. This hypothesis is tested by a focused comparison between two highly similar recent EUdirectives in the Netherlands. In line with the expectations, implementation of the Biopatent directive, which the Netherlands voted against, turned out to be highly problematic. The Gas Directive, which the Netherlands pushed for vigorously, was implemented timely and correctly. The analysis shows that there is an important scope condition for the hypothesis: as most member states will be internally divided about the preferences regarding new EU policy, the same actor should be involved during the phases of negotiations and implementation. If this is not the case, sound co-ordination should take place between all stakeholders in order to facilitate implementation. Contrary to the rather static goodness of fit argument, this study makes clear that domestic actors’ estimate of the degree of fit with existing national legislation may differ considerably, and secondly, that domestic actors may purposely attempt to change the domestic status quo.

INTRODUCTION

Research on the EU has gone through three analytical phases (Holzhacker and Haverland, this volume). The first wave was focused ‘bottom-up’, explaining processes of European integration and the coming about of the European Union. A next strand of research from the top down, the question being how member states respond to the demands flowing from EU membership. Europeanization research takes integration as an explanatory factor in understanding domestic political change and continuity. A specific

1 Ellen Mastenbroek is assistant professor at the department of public administration, Radboud University Nijmegen. Mendeltje van Keulen is research fellow at the Netherlands Institute for International Relations ‘Clingendael’.

19

R. Holzhacker and M. Haverland, (eds.), European Research Reloaded: Cooperation and Integration among Europeanized States, 19-42. © 2006 Springer. Printed in the Netherlands.

20

Ellen Mastenbroek and Mendeltje van Keulen

question in this ‘second wave’ has been how member states comply with EU policies. Driven by the widespread political attention for the so-called implementation deficit of EU directives, various researchers have contemplated this issue from the late 1980s onwards. Third and more recently, calls have been made to link the first and second stages so as to find out how experiences with Europeanization feed back into consecutive integration processes (Beyers and Trondal, 2003, 2). If one assumes that member states try to upload their policies to the EU level with the aim of laying these down in binding EU legislation (Héritier 1995, 278), it follows that implementation depends on the ‘goodness of fit’ between the EU policy demands and existing national policies (Duina 1997; Knill and Lenschow 1998; Knill 2001; Green Cowles Caporaso and Risse 2000; Börzel and Risse 2003; Börzel 2003a). However, this hypothesis has been rejected on several occasions (Knill and Lenschow 1998, Héritier and Knill 2001; Haverland 2000; Treib 2003). It has been argued that at times member states want to change the status quo, and hence push for EU policies that go against their existing policies. They may even use the EU as a leverage to get this done (Stiller and Kallestrup, this volume). For the same reason, member states may also easily adapt to misfitting EU policies. In assuming that national decision-makers act as guardians of the status quo, the goodness of fit hypothesis thus is overly conservative. In this chapter, we propose a more parsimonious explanation for implementation problems: member states’ preferences. The hypothesis is that member states will ‘balk at complying’ with decisions that do not ‘fit’ their national preferences, but smoothly implement much-wanted negotiating outcomes, even when these run counter to their status quo. In testing this hypothesis, a comparative case study is conducted. The first case is the ‘Biopatent Directive’ (1988/44/EC). Whereas dead against it, the Netherlands, were unable to form a blocking minority in the Council of Ministers. The second case concerns the so-called ‘First Gas Directive’ (98/30/EC), which the Netherlands pushed for vigorously. Apart from the fit with Dutch preferences, these directives are quite similar as regards technical characteristics that influence the ease of implementation, which allows for a focused comparison. The cases are explored with the help of process tracing, dossier analysis and qualitative interviews.

2.

THE IMPLEMENTATION DEFICIT

European integration is claimed to be haunted by an implementation deficit (Weiler 1988, 340). From the late 1980s on, academics and EU officials have shared the view that the member states’ record for complying with EU policies is rather poor. According to Metcalfe (1992, 117), this is one of the most pressing problems for the European Commission in seeking

Beyond the Goodness of Fit

21

to manage the European integration process. Duina (1997, 155) even claims that ‘the inconsistent observance of laws by the member states has stood squarely in the way of unification’. The deficit first came to the fore during the process of internal market building in the late 1980s and early 1990s. The objective of this undertaking, the development of a single European market for goods, services, capital, and people, was to be accomplished by means of an ambitious legislative program, comprising around 300 measures, mostly directives. Whereas progress at the legislative stage was fairly impressive, the project seemed to grind to a halt at the desks of national civil servants and politicians. Member states’ implementation of the new rules turned out to be far from smooth. In 1991, less than a year before the agreed starting date, the transposition rate of the twelve member states averaged 65 %. A mere 24 directives had been transposed by all member states, which equaled one sixth of all the directives in force at that time (Pelkmans 1991, 52). Late and faulty implementation is problematic because it endangers the credibility and the effectiveness of EC law. In the words of Internal Market Commissioner Bolkestein,‘delays in putting directives into effect cause enormous harm to businesses and to citizens’ (European Commission 2004). Moreover, implementation problems put a drain on the Commission’s scarce time and personnel, because investigating instances of alleged late or incorrect implementation is cumbersome and time-consuming. Since the early 1990s, the Commission has therefore taken to battle the deficit. It started out quite cautiously, forming a committee led by former Commissioner Peter Sutherland to study the problem. Yet over time it toughened its stance, shaking off its reputation for being ‘a pussycat when it comes to enforcement’ (Puchala 1975, 513). In doing so, it is aided by the European Court of Justice, which has become a fervent enforcer of EC law. Its threat of legal action has become more and more of a dissuasion for prospective cheaters. It especially gained momentum after its well-known 1991 Frankovich verdict, in which it affirmed the principle of financial liability of a Member State for late or improper transposition. Yet ‘the failure to transpose agreed directives (…) remains a focus for urgent attention’ (European Commission 1999, 13). In line with the Commission, several students of European integration have sought to answer the intriguing question ‘why states fail to implement measures they have previously agreed upon in the context of the Council of Ministers’ (Mendrinou 1996, 4). The overall view is that the implementation deficit is endemic to the EU (Weiler 1988; From and Stava 1993; Knill and Lenschow 1998, 1350; Duina 1997, 155; Lampinen and Uusikylä 1998, 231;

22

Ellen Mastenbroek and Mendeltje van Keulen

Mendrinou 1996, 2; Grant Matthews and Newell, 2000 72; Glachant 2001, 1; Bursens2002, 173) and therefore worth studying.2

3.

EXPLAINING IMPLEMENTATION PROBLEMS

Originally, EU implementation problems were viewed as resulting from legal and administrative difficulties. Starting with the work by Siedentopf and Ziller (1988), the focus was on legal and administrative problems which hampered implementation. An advocate of this view is Demmke (2001, 15), who after the analysis of the EU legal acts in water policy concludes that implementation problems only rarely result from political resistance, but more so from organizational, legal, technical and financial causes. This view is upheld by Falkner et al (2005), who show that the bulk of infringements in the area of EU labor law have their roots in legal-administrative problems. The literature suggests two types of legal-administrative factors that influence the ease of implementation. First, problems are identified at the level of the member states, such as legal perfectionism (Dimitrakopoulos & Richardson 2001b), lack of financial or personnel resources (Ciavarini Azzi 2000, 57; Falkner et al., 2005), and the inefficiency of domestic political institutions (Lampinen and Uusikylä 1998, 240).3 Second, several factors at the level of individual directives play a part, such as the EU decision-making procedure (Ciavarini Azzi 2000, 56); the extent to which it introduces a new topic of EU regulation; the time period allotted for implementation; the legal instrument used for implementation; the complexity of implementation; the existence of Chinese Walls between negotiators and implementers of EU legislation and the number of governmental departments (Bekkers et al., 1993, 197) involved (see Mastenbroek 2003, 376-279). However, in a quantitative study of transposition speed in the Netherlands, Mastenbroek (2003) has shown that such legal-administrative causes are only one side of the coin. Transposition problems sometimes seem to be of a more political nature. Two variants of this argument are present in the literature: the ‘goodness of fit’ hypothesis and the more general ‘domestic politics’ argument.

2

Some authors take a slightly more skeptical view, claiming that the deficit is not too high and not endemic to the EU (Börzel, 2001, 804). Others argue that implementation problems are not special to the EU, but just as prevalent in national political systems (Peters, 1997, 189). 3 For a quantitative study on such country-level variables, see Mbaye (2001).

Beyond the Goodness of Fit

3.1

23

Goodness of fit

The bulk of recent Europeanization literature centers around the ‘goodness of fit’. The idea can be traced back to Héritier (1995, 278), who claims that member states seek ‘to impose their own institutional patterns on EU policymaking because they wish to avoid costs of institutional adaptation and wish want to establish a level playing field for their own industry in European markets’. As follows from this assumption, a member state will have trouble adapting to the decision-making result if it is unsuccessful in uploading its national institutions and policies (Bursens 2002, 181). The hypothesis is that the ease with which European policy demands are implemented depends on the degree to which they ‘fit’ existing national policies and institutions (Duina 1997; Green Cowles et al., 2000; Knill and Lenschow 1998; Knill 2001). A ‘misfit’ results in adaptation costs for national administrations and hence protracted and incorrect implementation. This argument is basically historical institutionalist in nature. However, it has a clearly rationalist undertone, since most authors view the decision to implement as one consciously made by decision-makers seeking to avoid high adaptation costs. In this view, both European integration and Europeanization are governed by the wish of policy makers to uphold the status quo (see figure 1).

European Union

European integration

fit with member state’s status quo

Europeanization

member state

Figure 1. Linking the shaping and taking: the goodness of fit However intuitively appealing, the goodness of fit hypothesis has been rejected on several occasions, for various policy domains. The first to do so were Knill and Lenschow (1998), who confronted the hypothesis with data on the implementation of four environmental directives in the UK and in Germany. The outcomes of their analysis are quite disappointing, since only

24

Ellen Mastenbroek and Mendeltje van Keulen

three of the eight cases are in line with the hypothesis. For instance, two directives requiring major adaptations in the UK were implemented rather smoothly. At the same time, one directive requiring hardly any changes in Germany was transposed with a huge delay. Therefore, Knill & Lenschow (1998, 602) conclude that goodness of fit is in itself insufficient to explain implementation performance. A similar conclusion has been reached by various other scholars. Héritier and Knill (2001, 288), to begin with, tested the goodness of fit hypothesis on several cases in the field of transport policy. They conclude that the goodness of fit is not even a necessary condition for adaptation to occur. Haverland (2000), having analyzed the implementation of the packaging waste directive in several member states, argues that the goodness of fit is not decisive. The UK, for instance, implemented the directive relatively fast and correctly, despite facing a high misfit where Germany, on the other hand, only faced moderate adaptation pressure, but implemented the directive two years late. Finally, Treib (2003) has taken issue with the goodness of fit approach. After studying the implementation of six social policy directives in Germany, Ireland, the Netherlands, and the UK, he finds that the goodness of fit explains only 16 of the 24 cases.

3.2

Policy preferences

All in all, as appears from various empirical studies, the goodness of fit is neither a sufficient, nor a necessary variable for explaining the implementation of EU policies in the member states. The key problem is that it is too static a concept, because it does not allow for a wish among policy makers to change the status quo. In dealing with these disappointing results, broadly, two strategies can be followed. First, one can try to uphold the hypothesis by complementing it with additional or intervening variables despite the heavy empirical weather. For instance, Knill and Lenschow (1998) suggest that the key to their unexpected outcomes is the degree to which national institutions are embedded. The tendency in this literature is to view misfit as a necessary, if not a sufficient condition for domestic change (Green Cowles et al., 2000, Börzel and Risse 2003, 61, Börzel 2003a, 2). The outcomes of the implementation process are determined, besides the goodness of fit, by mediating variables such as the reform stage and dominant belief system (Héritier et al., 2001), political culture (Green Cowles et al., 2000; Börzel and Risse 2003), norm entrepreneurs (Börzel and Risse 2003), learning (Green Cowles et al., 2000), national reform capacity (Héritier 2002), multiple veto points (Green Cowles et al., 2000), the degree of domestic support (Knill and Lenschow 1998), adaptation pressures (Knill 1998; Börzel 2003), the interests of chief executives (Duina and Blithe, 1999), and the differential empowerment of actors (Green Cowles et al., 2000).

Beyond the Goodness of Fit

25

The downside of this approach of adding variables is that the framework becomes less parsimonious. Many of the revised theories are too openended. When new variables are included, like the ones mentioned above, clear hypotheses can no longer be formulated ex ante. The theory can only be used as a heuristic device to interpret outcomes ex post. The second strategy does not suffer from this problem. Rather than making the framework more-encompassing, it is proposed to make it more parsimonious. Rather than considering the fit with existing national policies, the focus is on the preferences of the member states regarding the issue at hand. This strategy allows for the possibility that member states want to change existing domestic policies by means of EU legislation. Thus, by leaving the particular policy preferences of a member state unspecified, rather than making a particular assumption as to their substance, a more parsimonious model can be formulated.4

Figure 2. Linking the shaping and taking: member states’ preferences This rationalist characterization of Europeanization is not new. It ‘fits’ with (liberal) intergovernmentalist accounts of European integration (see Moravcsik 1993) as well as with rationalist accounts of EU implementation. In line with the findings of realist IR scholars, several students of European integration have claimed that implementation of EU policies is political in nature (Snyder 1995, 57; Dimitrakopoulos and Richardson 2001b). Compliance with EU directives is regarded a matter of state choice, based 4 To some extent, we misrepresent the goodness of fit literature, in that it is not meant to be agency-oriented, but structuralist in nature (see for instance Treib 2003, 3). However, most of the goodness of fit literature departs wholly if not partly from an agency-related perspective (Héritier et al., 2001; Knill & Lenschow1998; Börzel, 2003a; Duina 1997; Börzel and Risse 2003; Knill 1998, Green Cowles et al., 2000).

26

Ellen Mastenbroek and Mendeltje van Keulen

upon political calculation (Puchala 1975; Gourevich 1996, 363). Having lost out at the decision-making stage, member states can try to limit their losses by means of ‘opposition through the backdoor’ (Falkner et al., 2004, 3). This can be explained ‘if one assumes rational action by member states who wish to reap the benefits of appearing to be good Europeans yet knowing all about the possibilities for policy erosion away from the high politic venues’ (Richardson 1996, 287). The rationalist solution is also in line with the empirical findings of those who falsified the goodness of fit hypothesis in the first place and did not choose to salvage it. Treib (2003) for instance claims that the key to the variation in implementation performance lies with the political preferences of national governments. Instead of oriented towards the status quo, governments’ reactions are to a great extent determined by their party political preferences (ibid., 24). This means that a misfit may be inconsequential when a national government is willing to change existing policies. The other side of the coin is that political opposition may occur even if the goodness of fit is high. A similar argument is made by Haverland, who claims that the key to his puzzling results is with the domestic political constellation. The more institutional veto points, or stages in the policy process where the agreement of domestic actors is needed, the more problematic implementation will be. However, Haverland focuses not so much on the contents of government preferences, but on the institutional layout of the member state concerning veto points and preferences of political and societal actors such as Parliament and industry (Haverland 2000, 97). The focus on domestic actors with possibly diverging preferences is also present in the approach by Dimitrova and Steunenberg (2000) who present a voting model of EU implementation. Transposition is modeled as a decisionmaking process in which different groups of national actors need to cooperate; each actor can block the implementation process when the directive goes against its interests. A host of actors is identified as veto players: political parties or fractions, parliamentary committees, interest groups and regional and local governments (Dimitrova and Steunenberg 2000, 215). All actors thus dispose of an informal veto power, which brings about a need for government to reach consensus on the implementation of the directive. The strength of these veto player approaches is that they open up the black box of the state. Rather than falling into the realist trap of portraying member states as unitary actors (see Hug 1999), they take a look at their constitutive parts, which allows for a dynamic account of EU implementation. There are, however, two disadvantages to this approach. First, it is unclear from the outset which actors can be classified as veto players - the theory runsthe risk of explanatory openness. Which actor qualifies as a veto player only becomes clear afterwards (Ganghof 2003, 4).

Beyond the Goodness of Fit

27

If predictions are to be formulated, the veto players must be identified beforehand. Second, the veto player models do not allow for testing, because the nature of the implementation ‘game’ beforehand is unclear. There is a need to study the cases in order to be able to select them, which creates a problem of eternal regression. Therefore, it seems wiser to simply focus on one actor: national government, like Treib has done. This allows selection of cases on the independent variable - satisfaction with the bargaining outcome. What is more, this decision can be defended theoretically. It seems farfetched to assume that so many actors have a veto position in implementation processes. From earlier analyses the impression has arisen that, at least in the Netherlands, interest groups are treated very differently when it comes to transposition than when it comes to regular policymaking processes. Government tends to take a rather distant stance towards them, in the case the interest groups do not share their preferences as to transposition. Especially since the stakes are high for government, with an eye on the possibility of financial liability in the face of the 1991 Frankovich case, government is not very likely to give in to interest groups’ demands that it does not share. Something similar holds for national Parliament. It can be argued that in the Netherlands, as well in other parliamentary systems with a strong party discipline, Parliament will rarely go against government’s proposal to transpose a particular directive. For these reasons, the focus is on national government and its preferences, assuming that these remain constant between the stages of decision-making and implementation.

4.

RESEARCH DESIGN

The hypothesis guiding this research is that implementation will be smoother if a national government is presented with a decision-making result that closely matches its policy preferences, than when it is confronted with an undesired decision-making outcome. In other words, the fit with a member state’s policy preferences rather than the goodness of fit determines the ease of implementation (see table 1).

Ease of implementation

Fit with preferences + + Gas Directive -

Biopatent Directive

Table 1: The hypothesized relation between satisfaction and implementation performance

28

Ellen Mastenbroek and Mendeltje van Keulen

In the following, a research design is formulated that allows assessing this hypothesis. In doing so, the focus is on the case of the Netherlands. This strategy allows for the control of potential confounding variables at the national level. As argued above, implementation performance is not solely a consequence of governmental willingness, it is also affected by technicaladministrative variables. In order to control for the effect of such variables at the level of the member state, like legal perfectionism, financial situation, understaffing and the efficiency of domestic political institutions, we propose to focus on one particular member state. Although the single-country focus introduces the risk of low external validity, this can be mitigated by selecting a case that is likely to be crucial (Lijphart 1971, 692). The focus is on the Netherlands, a member state that is claimed to have a strong culture of compliance (Tallberg 2002, 618; Collins and Earnshaw 1992, 217). Dutch governments have a general respect for international law, which makes that acts are generally obeyed once issued, even if the government did not agree with it in the first place. Beyers, Kerremans, and Bursens (2001 77) substantiate this conclusion in their study of transposition in the Benelux: ‘It would be unbearable for Dutch officials not to transpose European law (…).’ And in case deadlines are not met, the Dutch strive to quickly remedy the problem.’ Therefore, if governmental preferences are found to influence implementation patterns in the Netherlands, it is likely to also be the case in other member states.

4.1

Operationalization of satisfaction

In order to test empirically to what extent the ease of implementation is influenced by the fit with governmental preferences, one question is: how to assess ‘satisfaction’? After all, the idea behind EU level bargaining is for all interested parties to get their way and see their positions mirrored in the final text. Because so many interests are involved, it is expected that the outcomes are not often dominated by the interests of one or several parties, but reflect a compromise between, at least, the majority. One intuitively appealing method would be to analytically assess the degree to which the outcome of EU negotiations reflects national preferences with the help of text-analysis. However, EU laws, as national legislation, consists of multiple separate decisions, interconnected in intricate compromises in legal jargon in which the national preferences are interwoven. Therefore, it is not only a complex and therefore time-consuming task, but also only to a limited extent useful to ‘reconstruct’ and subsequently ‘count’ national preferences from the final texts. This paper therefore applies a more straightforward operationalisation of satisfaction: national voting behavior in the Council. As the Council is characterized by a clear preference for unanimity (Hix 1999), in the sense that its provides multiple opportunities for getting an outlying policy

Beyond the Goodness of Fit

29

position in line with the others, it can be assumed that a no-vote constitutes a strong demonstration of the dislike of the decision at stake. This holds even more so in case during the negotiations, a clear majority has emerged in favor of the decision and a no-vote thus only serves ‘for the record’. However, where a no-vote thus serves as a clear indicator for the misfit of preferences and outcomes, the situation is less clear-cut in case of a yesvote. Rather than a ‘one-way street of national interest representation (Christiansen 2001, 152), EU negotiations provide ample opportunities for persuasion, policy-learning and package-deals. Thus, where a national vote in favor of the policy proposal may serve as a strong indication that that particular member state likes the result, it may just as well been bought off by linking the dossier with other agenda items or voted in favor for the sake of consensus. Because of these intervening mechanisms in national voting behavior, it is thus necessary to reconstruct the process of ‘shaping’ national preferences and negotiations, in order to determine their exact nature.

4.2

Operationalization of goodness of fit

Besides testing the core hypothesis of the paper, it is necessary to control for the goodness of fit as a potential explanatory variable. Hence it should be double-checked whether the expected difference in the ease of implementation cannot be attributed to the goodness of fit after all. In the literature, two different distinctions are made when it comes to the goodness of fit. First, a distinction is made between institutional and policy misfit (Börzel and Risse 2003, 59-60, Green Cowles et al., 2000, 6-7; Héritier et al., 2001, 15). The policy dimension relates to the content of the policies, whereas the institutional dimension relates to the regulatory style and structure of a particular policy sector (Knill and Lenschow, 1998, 597). A second distinction concerns the difference between the legal and the practical status quo (Falkner, 2003, 3). It may be the case that certain rules are not laid down in law, but exist informally.

4.3

Operationalization of ease of implementation

Three distinct stages of implementation can be distinguished: transposition, application, and enforcement (Prechal 1995, 5-6). The first stage, that of transposition, refers to the conversion of a directive into provisions of national law by the competent national legislative body or bodies. The second stage is that of application, which refers to the actual carrying out of the national measures transposing the directive. The third stage, labeled enforcement, entails the processes of compelling observance of (the national measures transposing) the directive. For this case study, the focus is on the process of transposition. Application and enforcement,

30

Ellen Mastenbroek and Mendeltje van Keulen

however interesting, can be thought of as totally different games, played at different levels with different players and incentives than transposition. Transposition is subject to two conditions: timeliness and correctness (Duina 1997, 156; Prechal 1995, 20). In the first place, directives need to be transposed in due time, i.e. before the deadline specified by each directive. Secondly, each directive should be transposed correctly. That is, the contents of the measure transposing the directive need to be in line with the contents of the original directive. Both elements of transposition will be considered.

4.4

A focused comparison

The rationale for qualitative case studies lies in the fact that assessing the correctness of transposition for a large number of cases is virtually infeasible. This would involve close reading of the EU and national legal texts, as well as consultation of experts and practitioners in the national setting. Moreover, quantitative analysis would increase the risk of establishing a spurious relationship between governmental willingness and the ease of implementation. Case studies make it possible to explore this link in a much closer way. However, the combination of the complexity of most EU dossiers, the level of detail required in working with methodologies of process tracing imply that only a limited number of dossiers can be thoroughly analyzed. The cases have been selected on the independent variable, so as to avoid inference problems (King et al., 1994, 137). This variable is the fit with government preferences, as reflected by national voting behavior in the Council. A comparison is made between a directive that was very much unwanted by the Dutch and one that was welcomed whole-heartedly. By selecting cases with extreme scores on the explanatory variable, the chances of hitting upon a ‘crucial case’ increase (Lijphart 1971, 692). If the utterly unwanted case is transposed smoothly, or vice versa, the theory is in real trouble. A focused comparison is chosen to prevent the presence of confounding legal-administrative variables of the type discussed in section four. Apart from representing two extremes in the sense of voting behavior, the directives need to be as similar as possible when it comes to these factors. In short, a ‘most similar systems’ design is applied (Przeworski and Teune 1970, 32-34). Implementation problems are expected to occur especially in the realm of qualified majority voting, which involves the risk of being outvoted. As voting analysis has demonstrated that Agriculture and Internal market are the Council formations in which voting occurs most frequently (Mattila and Lane, 2000:42), from these Council formations two case-studies have been selected, representing a no- and a wholehearted yes-vote of the Netherlands. The Biopatent directive (98/44/EC) on the legal protection of biotechnological

Beyond the Goodness of Fit

31

inventions, and the Gas Directive, on the liberalization of the common market for gas (98/44/EC) are very similar in nature. Both are directives aiming at harmonization of national policies, derived from the 1985 White Paper on the Internal Market. Both constituted a first step for the EU into a new policy area and are rather complex in nature. Both were adopted under the co-decision procedure and subject to qualified majority voting. As to the Dutch transposition process, the similarities are striking as well. The directives were dealt with by the same government department, that of Economic Affairs5, and both required the change or adoption of one or more laws. All in all, except for the fit with government’s preferences, the cases are highly comparable. Information on the cases has been drawn from the relevant EU and national official publications, as well as secondary literature and qualitative interviews with representatives from the European Commission, the ministry of Economic Affairs, national and European Parliament and interest groups involved.

5.

THE BIOPATENT DIRECTIVE

5.1

Negotiations

The original 1988 Commission proposal for a directive on the national legal protection of biotechnological inventions6 regulates harmonization of national patent law, to create a level playing field for biotech industry.7 A key feature of the proposal is the principle that plants, animals as well as isolated human elements are patentable, provided the innovation is new, inventive and susceptible of industrial application. The proposal clarifies the scope of legal protection and sets exceptions, amongst which the exclusion from patentability of plant and animal varieties, and those inventions that conflict with public order and morality. At the outset, DG Internal Market of the Commission appeared to have a relatively comfortable position. A majority of member states was actively or passively in favor of the directive, which enjoyed wide support by industry and patient associations. Problems only arose when the conciliation compromise on a final text was unexpectedly rejected by the newly elected 5

One difference is that the biotech directive required cooperation with the Ministry of Agriculture. However, this ministry played only a marginal role, which is why this variable can be excluded as an explanation for any expected difference in the ease of implementation. 6 COM(1988)496final 7 The need for harmonisation as well as the legal base were contested, however, in the course of the negotiations. According to some experts, the Commission was mainly driven by a wish to be involved in the ‘hot’ sector of biotechnology, as European patent law is fully covered by the Munich-based European Patent Convention.

32

Ellen Mastenbroek and Mendeltje van Keulen

European Parliament in spring 1995, after a massive protest campaign by a coalition of interest groups led by Greenpeace. Commission and Council were said to be infuriated about the disloyal behavior of the EP (Paterson 2000, 339). A new proposal, taking the ethical concerns partly into account, was drafted only eight months later. Negotiations on this text started off in a highly politicized environment. The European Parliament was heavily lobbied by non-profit groups highlighting ethical, health and environmental risks associated with exploiting scientific knowledge, which were translated into parliamentary amendments. Interest groups also made their point through national governments, represented in the Council working group on intellectual property, where the text was discussed for two years. In the course of the negotiations, the delegations of the Netherlands, Belgium, Italy, France and Britain voiced various reservations over the patentability of human gene sequences and plants and animals. Although the position of the latter three member states changed to a more willing stance, the Dutch delegation voiced political reservations over a crucial element of the text, relating to the patentability of genetically modified plants and animals. These reservations originated in Parliament, which in 1995 had passed a revision of the national law on patents (Rijksoctrooiwet) in which the patentability of plants and animals had been explicitly excluded. 8 The parliamentarians involved (from the social democrats and the Christian democrats) seemed unaware that, parallel to the revision of national law, the European Commission had been drafting a directive with precisely this controversial issue as a defining element.9 The so-called ‘BNC-fiche’, a paper notifying parliament about the consequences of Commission proposals, did not anticipate on these implementation problems; it stated that ‘no special problems with implementation’ were foreseen.10 The Dutch delegation had agreed with the common position on the draft directive in 1994 and the State Secretary of Economic Affairs claimed not to foresee any problems with the ‘fit’ of a new directive.11 However, parliamentarians made it clear that there could be no question of bypassing national legislation through new and binding EU law. Responsible officials from the Economic Affairs Ministry, all too aware of the political pressures from the Commission and industry, were stuck between a rock and a hard place. Reluctantly, they took political opposition into account in determining the Dutch instruction. As these concerned a principal element in the directive text, the Dutch could or would not be accommodated in the negotiations. Thus, when in December 1997 the 8 Second Chamber: Motie 18(19744), 27 June 1996. This exclusion violated the European Patent Convention (Bostyn et al., 2001, 137). 9 Source: interviews. 10 Lower House Parliamentary Papers, 1995-1996: 22112, nr 60, p. 3. 11 Lower House Parliamentary Papers, 1994-1995, Parliamentary question nr 653 appendix.

Beyond the Goodness of Fit

33

Luxembourg presidency compromise text was decided in the Council, the Dutch delegation voted against the majority, whereas the Belgian and Italian delegations abstained from voting.

5.2

Transposition

The directive being decided upon, the Netherlands was presented with a politically highly controversial result to be transposed into national law. A majority in the Lower House 12 considered the directive an intolerable break with the Rijksoctrooiwet 1995, the central point of contestation being the patentability of plants and animals. The question is whether this can be denoted as a large misfit. As most of the interviewees indicated, the directive itself could be easily accommodated within the Rijksoctrooiwet. What is more, most patents are granted on basis of the European Patent Convention, which does allow for plants and animals to be patented. The directive closely follows the jurisprudence in this field, and hence ‘changes the substantive law of patentability in Europe relatively little’ (Kamstra et al.,, 2002, 45). Yet, the patentability of plants and animals proved highly salient for Parliament. In seeking to evade the directive, Parliament took several unprecedented steps. First, it requested government to lodge legal action with the European Court of Justice (ECJ) to have the directive annulled.13 Though the Ministry of Economic Affairs itself was still in favor of the directive, Cabinet complied with Parliament’s wishes and brought the case before the ECJ.14 The directive was challenged mainly on legal instead of ethical grounds, as it was thought that this would increase the chances for the court case to be successful.15 During Court proceedings, the Ministry of Economic Affairs started administrative preparations for a close transposition of the directive In the discussions with Parliament, the most controversial amendment was number 11, holding that genetically modified plants and animals could not be 12 By and large, the liberals (VVD) and social liberals (D66) were in favor of the directive and transposition, whereas the rest of the parties, forming a majority, were opposed. The main opponents were social democrats (PvdA) and Christian democrats (CDA), supported by smaller left-wing parties (SP, Groen Links) and religious parties on the right (Christen Unie, RPF, SGP). The PvdA, at the time member of the governing coalition, hence showed dualistic behavior. 13 Motion of MP Witteveen-Hevinga, October 15, 1998. Lower House Parliamentary Papers, 19 744, nr. 23. 14 ECJ Case C-377/98; October 19, 1998. 15 For instance, the Netherlands challenged the proposal’s legal base, claiming that the directive should have been adopted using Unanimity instead of Qualified Majority Voting (Kamstra et al., 2002, 67). This issue had been discussed in the initial stages of the Council working group, but had been settled during the negotiations.

34

Ellen Mastenbroek and Mendeltje van Keulen

patented 16 . The Ministry of Economic Affairs, supported by the Dutch Council of State, claimed that this was diametrically opposed to the directive. The ECJ was asked suspension of the transposition, as it became clear that its verdict was not to be delivered before the implementation deadline, but this request was rejected in July 2000. A year later the Court rejected the Dutch claim, making clear once and for all that the Netherlands had to implement the directive. However, Parliament maintained its amendments. In the following interpretation battle between government and Parliament, both sides requested advice from different authoritative institutions, like the European Commission (government), and a legal scholar from the University of Amsterdam (Parliament). Most amendments, including the most controversial one, were nonetheless adopted by Parliament in spring 2002. In the meantime, an infringement procedure started by the Commission in November 2000 was referred to the ECJ in 2003. The Court’s verdict was expected to be in favor of the Commission, and it was not unlikely that the Netherlands would be penalized for late transposition. 17 Moreover, Economic Affairs feared that the Commission would open an infringement procedure for incorrect transposition. Deliberations in the Upper Chamber were delayed during political arousals in 2002-3, until a new Cabinet came to power. It then took another year before the Upper Chamber considered the proposal, posing critical questions about the contents and legal text of the directive.18 Then, however, a remarkable switch occurred in the position of the social democrats. Economic Affairs by this time wanted to implement the directive as fast as possible, though incorrectly, so as to avoid being fined by the ECJ. A majority in the Upper Chamber, composed among others of the social democrats, arguing they could not consent with a faulty bill. Therefore, they proposed that the government came with a novelle, a bill meant to correct a proposal pending in the Upper Chamber. Economic Affairs went along with this. In October 2004, the Upper House finally passed the revised bill, which no longer excluded inventions on plants and animals from patentability, thus avoiding further infringement procedures for incorrect transposition. All in all, the case-study of the Biopatent directive sustains the hypothesis that member states will try to block unwanted directives in the implementation stage. Transposition has proved to be highly problematic for the Netherlands, because of the Dutch reluctance to allow the patentability of plants and animals. The goodness of fit hypothesis cannot explain the late and incorrect transposition, because it is not so much a matter of policy or 16

Lower House Parliamentary Papers, 26 568 (R 1638), nr. 11. Transposition of this case is exceptionally late, i.e. more than two times the standard deviation. See Mastenbroek (2003). 18 Upper House Parliamentary Papers, 26 568 (R 1638), nr. A. 17

Beyond the Goodness of Fit

35

institutional misfit, but of misfit with a highly salient interest of the parliamentary majority, which disposed of a veto. What is more, this case demonstrates that preferences may differ between actors, even within one and the same political party.19

6.

THE GAS DIRECTIVE

6.1

Negotiations

Because of its strategic importance, energy policy and more specifically natural gas supply used to be organized nationally. The 1988 Commission proposal for a single EU energy market had its basis in the 1985 White Paper on the Internal Market. It was expected that more competition would lead to a decrease in energy prices. On the other hand, deregulation implied a loss of policy instruments for central governments to pursue economic objectives. Discussions on the Gas directive, a 1992 proposal by the Commission which was split into a gas and an electricity part, had been ongoing in the Council working group since 1994. The Dutch Ministry of Economic Affairs had a ‘double-hat’ in this discussion. On the one hand, the state was partly owner of Gasunie and a major EU wide gas supplier. Moreover, the Netherlands hosts energy-consuming market sectors (greenhouse horticulture, chemical industry, transport). The Ministry of Economic Affairs was very reluctant regarding liberalization, although there was regular debate with the Foreign Affairs Ministry about this stance. The main Dutch critique was about the risks for public supply security and the split between supply, transport and distribution. On the other hand, the Ministry could not neglect international developments towards more liberalization in almost any other policy sector and slowly realized that its position was untenable in the long run. There also was a strong lobby of the large consumers and the small and medium enterprises for a more liberal gas policy. In 1994, when the liberally oriented Purple coalition entered into power, the Dutch changed their position radically. It was politically decided that the Netherlands should actively push for opening up the EU’s energy markets with the aim to open its national gas market. The Third Energy Report, which was published in December 1995, was the written proof of this move. Liberalization of the gas Market became a top priority of the Cabinet, and specifically of Minister of Economic Affairs Mr. Wijers. At the EU level, after years of deadlock, the directive was ‘revived’ during the Irish Presidency in the second half of 1996. The gas directive was made a Presidency priority in the Netherlands, two months before the start of 19 Alternatively, one could claim that the preferences changed over time. Further research should be done before this claim can be made.

Ellen Mastenbroek and Mendeltje van Keulen

36

its 1997 Council Presidency. The Minister claimed that this goal was supported by a majority of the member states. A junior policy official at the DG Energy was assigned to the Gas Directive dossier and frequented the EU’s Council working group on energy, where the gas directive proposal was discussed in the working group every other week. Although the presidency was very active in the negotiations, The Energy Council in May did not reach agreement on the issue. Failure seems mainly due to the unexpected French elections in June, as a result of which the French were not able to change their position. The Luxembourg Presidency took up the negotiations and issued a new proposal in November. The Council common position reached then in December 1997, was adopted in February 1998. Although majority would have been sufficient, the Presidency aimed at reaching consensus in view of the co-decision procedure. The Gas Directive was rubberstamped in the May 1998 plenary session of the European Parliament, and signed in June 1998.

6.2

Transposition

The resulting directive has as its aim to ensure the free movement of gas and improve the security of supply and industrial competitiveness. As to the goodness of fit, the directive was not at all in line with the Dutch status quo. It provided for liberalization of the Dutch gas market, which ran counter to the existence of the Gasunie as a monopolist in the market. It hence implied a great policy misfit. Moreover, the institutional misfit was large. Since there were no legal rules concerning the transport and delivery of gas, a whole new regulatory framework needed to be erected. 20 However, the directive was clearly in line with the preferences of the government. As said, Minister Wijers was a strong advocate of liberalization, and he wholeheartedly supported the directive, along the lines of the Third Energy Report. The Member states were given two years to implement the directive, the deadline being set at 10 August 2000. However, the transposition process had already been commenced earlier. As early as 1995, the Ministry of Economic Affairs had installed a Taskforce Gas Act (Projectgroep Gaswet), which had the task of drawing up a proposal for a new Gas Act providing for liberalization of the gas market. Rather than initiating this process, the directive served as an input into this process. European developments were only one of the considerations taken into account in the drawing up of the Gas Act (Ministry of Economic Affairs 1997, 3). The proposal for the Gas Act is seen as ‘also implementing directive 98/30/EC’.21 From the early beginning on, the Taskforce kept an eye on the proposal for the first gas directive. By the time the Gas Act was being drawn up, the 20 21

Lower House Parliamentary Papers, 1998-1999, 26 463, nr 3, p. 2. Lower House Parliamentary Papers, 1998-1999, 26 463, nr 3, p. 1.

Beyond the Goodness of Fit

37

Council had already reached agreement on the proposal. For this reason, the EU parameters were considered relatively stable (Ministry of Economic Affairs 1997, 5-6). According to one interviewee at the Ministry, it was clear from the outset that the directive would not be hard to implement, because it aimed at a very limited extent of market opening (somewhere between 33 and 43 % of the market after ten years). The Dutch government wanted to go much further than the required minimum: it aimed at full liberalization in 2007 (Ministry of Economic Affairs 1997). Moreover, the directive seemed to become rather flexible, in that it left countries some choices for transposition. The main option concerned the type of access to the networks: regulated versus negotiated (arts. 15 and 16 of the directive). Another option concerned the use of a reciprocity clause. The Dutch discussion to a great extent focused upon these issues, and the extent of market opening. The bottom line was that Parliament wanted minimal transposition of the directive, whereas government wanted to go much further in its speed of market opening. In its first proposal for a law, dated 30 March 1998, The Ministry of Economic Affairs aimed at full market opening in 2007, and proposed a system of negotiated access. It did not include the reciprocity clause. Most of the Parliamentary proceedings revolved around these three issues. A Parliamentary majority wanted slower market opening, regulated access, and use of the reciprocity clause. The government did not grant the first request, and even sped up the rate of marketing, providing for full opening by the end of 2003. However, it made this conditional to careful evaluation of the experiences with liberalization in the energy market, so as to pacify objections by Parliament. Second, it honored Parliament’s request for tighter oversight of access. In the end, the Gas Act provided for a mixed system of access, combining negotiated access with guidelines set by the Dienst Toezicht Energie (DtE), which was to apply and enforce the directive. Finally, it did not acknowledge Parliament’s request to use the reciprocity clause. The Gas Act entered into force at 10 August 2000, which rendered implementation timely. Furthermore, the directive seems to have been implemented correctly, as the European Commission is very positive about the situation in the Netherlands.22 In an external study of the state of play in the member states, Dutch implementation is denoted as complete.23 This also has to do with the nature of the directive, which leaves the member states considerable choice on the extent of market opening and the regulatory arrangements for the gas market.24 22

http://europa.eu.int/comm/energy/gas/benchmarking/doc/gas98_30_en.pdf, p. 2. Report by Wefa Ltd. for the European Commission, July 2001. See http://europa.eu.int/ comm/energy/gas/publications/doc/finalcor-vol1.pdf. 24 http://europa.eu.int/comm/energy/gas/benchmarking/doc/1/report-amended_en.pdf, p. 8. 23

Ellen Mastenbroek and Mendeltje van Keulen

38

All in all, transposition of the directive has been relatively smooth, despite the huge policy and institutional misfit with the existing practical and legal situation of a monopolized gas sector. Debates in Parliament never called the directive into question, but revolved around the way the directive was implemented, with Parliament adopting a more conservative stance than government, which aimed at fast liberalization. Implementation was smooth because the directive was rather flexible, but mainly because the Netherlands had adopted a clear pro-liberalization stance in 1995. In the words of the interviewee at the Ministry of Economic Affairs: ‘We simply wanted the directive.’

7.

CONCLUSION

This chapter has explored the link between member states preferences and the ease of implementation of EU policy. The guiding hypothesis was that, rather than the goodness of fit with existing policies, the fit with national preferences predicts the ease of implementation of EU policy. In the empirical part of this chapter, this hypothesis has been tested using a focused comparison between the Biopatent and the Gas directive in the Netherlands. The hypothesis predicted that implementation performance would differ, as the Biopatent directive was unwanted by the Netherlands, whereas the Gas directive was vigorously pushed for. The outcomes of both two cases are in line with the hypothesis. Transposition of the Biopatent directive has proved to be highly problematic for the Dutch government. Both the negotiation and implementation of the directive were characterized by a strong reluctance to allow the patentability of plants and animals. The directive eventually was transposed correctly, but after lengthy debates, and more than four years after the deadline. The second case, the First Gas Directive, shows that national preferences may work wonders in overcoming misfit. Whereas on first sight the directive displayed a large misfit with the status quo of a monopolized gas market, transposition was relatively easy because the Dutch Purple Coalition had already planned to change this situation. Liberalization proceeded relatively easy, because of the strong will to do so. The directive was transposed in time and correctly. Even though parliamentary proceedings were rather extensive, they did not question the directive itself, but the maximalist transposition by the Dutch government. At the same time, the analysis shows that the hypothesized link between preferences and the ease of implementation assumes a unitary actor. In the Biopatent case, government and Parliament showed divergent preferences as to key elements of the Directive. However, this condition does not pose insurmountable problems for theory, since in both stages Parliament had the

Beyond the Goodness of Fit

39

final word. During both processes of negotiation and implementation, Parliament’s formal veto position enabled it to get its way. This would have been different, had Government used delegated legislation to transpose the directive and thus bypassed Parliament. Hence, the hypothesis seems subject to the scope condition that, in the face of an internally divided member state, the location of the veto is the same during the stages of shaping and taking. This leads to a second important qualification, which has to do with EU policy co-ordination at the domestic level. If Parliament had not been involved in negotiating the Biopatent directive, the Dutch government would probably have voted in favor. In that case, implementation would have been equally problematic. This situation has arisen in other member states that voted in favor of the directive, but have not succeeded in getting transposing legislation through Parliament (Germany and France, among others). To conclude: there need not only be a formal link between the veto positions, it also needs to materialize, for the hypothesis to hold. The scope condition for the theory to ‘work’ is thus that, in case of an internally divided member state, there must be a formal and material link between the location of the veto during the stages of shaping and taking. Theory assumes that there is effective co-ordination between all domestic actors involved in the ‘shaping’ and ‘taking’ processes. This is consistent with the observation that national governments of the member states have invested much time and resources in mechanisms to link policy-makers and negotiators in national capitals and Brussels with implementers and parliamentarians involved at the domestic level (Kassim 2000). Finally, the analysis of the cases sustains the limited use of the ‘goodness of fit’ approach in analyzing the implementation of EU legislation. First, it shows that the assessment of the actual ‘goodness of fit’ highly depends on the views of the stakeholders. In the Biopatent case, opinions about the viability widely diverged. The Ministry of Economic Affairs and its allies were of the opinion that the directive only necessitated minor legal changes, whereas Parliament and various interest groups found that the directive went against the core of existing legislation. Second, this study reinforces the finding that the goodness of fit is not what drives Europeanization. With the Gas directive, the Dutch clearly sought to change the existing status quo, by radically opening up their national gas market. This is diametrically opposed to the assumption that member states coûte que coûte want to stick to existing policies and institutions. Third, even if member states do upload existing national policies, and are reluctant to change these in the face of a high goodness of fit, it is not so much the goodness of fit that should interest us, but the preferences underlying this. If there is a relationship between fit and Europeanization, this is because policy makers prefer the existing situation to the directive, not because of the mere existence of this status quo. Preference-based accounts for explaining implementation failure provide a promising avenue for further research.

Ellen Mastenbroek and Mendeltje van Keulen

40

REFERENCES Andersen, Sv.S., & Eliassen, K.A. (eds.), 1993, Making Policy in Europe: The Europeif ication of National Policy Making London, Sage, pp. 55-67. Bekkers, V.J.J.M. et al., 1993, Succes- en Faalfactoren bij de Uitvoering van EG-beleid, Bestuurskunde, 2(2): 192-200. Beyers, J. and J. Trondal, 2003, How Nation-States hit’ Europe: Ambiguity and Representation in the EU, European Integration online Papers, 7(5). Beyers, J., Kerremans, B., & Bursens, P., 2000, European Policy Preparation and Implementation in the three Benelux Countries, in: The European Union and the Member States: Cooperation, Coordination, and Compromise, Zeff, E. and Pirro, E. (eds.), Lynne Rienner, Boulder: pp. 59-88. Börzel, T.A., 2001, Non-Compliance in the European Union: Pathology or Statistical Artefact? Journal of European Public Policy, 8(5): 803-24. Börzel, Tanja A., 2003a., Environmental Leaders and Laggards in Europe: Why there is (not) a ‘Southern Problem’. Aldershot, Ashgate, London. Börzel, T.A., 2003b., Shaping and Taking EU policies: Member State Responses to Europeanisation. Queen’s Papers on Europeanisation, 2, 2003. Börzel, T.A., Risse, Th., 2003,Conceptualizing the Domestic Impact of Europe, in: The Politics of Europeanisation, K, Featherstone and C. Radaelli (eds.), Oxford University Press, Oxford, pp. 55-78. Bostyn, S., et al., 2001, Moderne Biotechnologie en Recht: Kluwer, Deventer. Bursens, P., 2002, Why Denmark and Belgium have Different Implementation Records: On Transposition Laggards and Leaders in the EU. Scandinavion Political Studies, 25(2): 173195. Ciavarini Azzi, G., 2000, The Slow March of European Legislation: The Implementation of Directives. Neunreither, Karlheinz, & Wiener, Antje (Eds.), European integration after Amsterdam: Institutional dynamics and prospects for democracy, Oxford: Oxford University Press, pp. 52-67. Collins, K., & Earnshaw, D., 1992, The Implementation and Enforcement of European Community Legislation, Environmental Politics, 1(4): 213-249. Commission of the European Communities, 1999. Communication from the Commission to the European Parliament and the Council: The Strategy for Europe’s Internal Market, COM, 1999) 464, Brussels. Commission of the European Communities, 2004, Internal market: Big disparities Between Member States in Implementing and Aapplying Results. IP/04/33, 12 January 2004. Demmke, Chr., 2001, Towards Effective Environmental Regulation: Innovative Approaches in Implementing and Enforcing European Environmental Law and Policy, Harvard Jean Monnet Working Paper 05/01). Cambridge, MA: Harvard Law School. Devuyst, Y., 1993, De Omzetting van EG-richtlijnen in de Belgische Rechtsorde en de Europeanisering van de Belgische Politiek [The transposition of EC directives in the Belgian legal order and the europeanization of Belgian politics]. Res Publica, 35(1): 39-54. Dimitrakopoulos, D., & Richardson, J., 2001b). Common Law Strife. The Guardian, 23 April 2001. Dimitrova, A. and Steunenberg, B., 2000, The Search for Convergence of National Policies in the European Union: An impossible Quest? European Union Politics, 1(1): 201-226. Duina, F., 1997,E xplaining Legal Implementation in the European Union. International Journal of the Sociology of Law, 25(2): 155-79. ’

Beyond the Goodness of Fit

41

Falkner, G. et al, (ed.), 2004, Non-Compliance with EU Directives in the Member States: Opposition through the Back Door? West European Politics, May 2004, 27(3): 452-73. Falkner, G., 2003, Comparing Europeanisation Effects: From Metaphor to Operationalisation, European Integration online Papers, vol. 7. From, J., & Stava, P., 1993, Implementation of Community Law: The Last Stronghold of National Control. Andersen, Sv. S., & Eliassen, K.A., (Eds.), Making policy in Europe: The Europeification of national policy making Sage Publications London, pp. 55-67. Ganghof, S., 2003, Promises and Pitfalls of Veto Player Analysis. Swiss Political Science Review, 9, 2:1-25. Glachant, M. (ed.), 2001, Implementing European Environmental Policy: The Impacts of Directives in the Member States, New Horizons in Environmental Economics, Edward Elger, Cheltenham. Gourevitch, P.A., 1996, Domestic Sources of International Cooperation International Organization, 50(2): 349-373. Green Cowles, M., Caporaso, J., & Risse, Th. (eds.), 2001, Transforming Europe: Europeanization and domestic change, Cornell University Press, Ithaca and London. Haas, P.M., 1998, Compliance with EU Directives: Insights from International Relations and Comparative Politics, Journal of European Policy, 5(1): 17-37. Haverland, M., 2000, National Adaptation to European Integration: The Importance of Institutional Veto Points, Journal of Public Policy, 20(1): 83-103. Héritier, A., 1995, ‘Leaders’ and ‘Laggards’ in European Clean Air Policy, in: Convergence or Diversity? Internationalization and Economic Policy Response, F. van Waarden & B. Unger (Eds.), Avebury, Aldershot, pp. 278-305. Héritier, A. and Knill, C., 2001, Differential Responses to European Policies: A Comparison. In: Differential Europe: New Opportunities and Restrictions for Policy-making in the Member States, Héritier, Adrienne et al., (eds.), Lanham, MD: Rowman & Littlefield, pp. 257-94. Hix, S., 1999, The Political System of the European Union, Houndmills, Basingstoke, MacMillan. Hosli, M. and B. Soetendorp, 2000, The Hidden Dynamics of EU Council Decision-Making, paper for the 41st Annual Convention of the International Studies Association (ISA), Los Angeles, March 14-17, 2000. Hug, S., 1999, Non-Unitary Actors in Spatial Models: How Far is Far in Foreign Policy? Journal of conflict resolution, 43(4): 479-500. Kamstra, G., et al., 2002, Patents on Biotechnical Inventions: The E.C. Directive, London, Sweet & Maxwell. Kassim, H. et al., 2000, The National Co-ordination of EU Policy: the Domestic Level, in: Eurpean Union, Power and Policy-making, J. Richardson, Oxford: Oxford University Press, pp. 155-78. King, G. et al., 1994, Designing social inquiry: Scientific inference in qualitative research, Princeton, Princeton University Press. Knill, Chr., 2001, The Europeanization of National Administrations: Patterns of Institutional Change and Persistence, Cambridge, Cambridge University Press. Knill, Chr., & Lenschow, A. 1998, Coping with Europe: The Impact of British and German Administrations on the Implementation of EU environmental Policy, Journal of European Public Policy, (5): 595-615. Lampinen, R., & Uusikylä, P., 1998, Implementation Deficit – Why Member States do not Comply with EU Directives? Scandinavian Political Studies, 21(3): 231-51. Lijphart, A., 1971, Comparative Politics and the Comparative Method, The American Political Science Review, 65(3): 682-693. Maas, H.H., & Haersolte, J.C. van, 1994, Tijdige uitvoering van EG-regelgeving in Nederland door middel van wetgeving, SEW: Tijdschrift voor Europees en Economisch Recht, 42(11): 703-26.

42

Ellen Mastenbroek and Mendeltje van Keulen

Mastenbroek, E., 2003, Surviving the Deadline: The Transposition of EU Directives in the Netherlands, European Union Politics, 4(4): 371-96. Mbaye, A.D., 2001, Why National States Comply with Supranational law, European Union Politics, 2(3): 259-281. Mendrinou, M., 1996, Non-Compliance and the European Commission’s Role in Integration, Journal of European Public Policy, 3(1): 1-22. Metcalfe, L., 1992, After 1992: Can the Commission Manage Europe? Australian Journal of Public Administration, 51(1): 117-30. Ministry of Economic Affairs, 1997, Discussienotitie Gasstromen. Moravcsik, A., 1993, Preferences and Power in the European Community: A Liberal Intergovernmentalist Approach, Journal of Common Market Studies, 31(4): 473-524. Paterson, L.A., 2000, Biotechnology Policy: Regulating Risks and Risking Regulation, in: Policy-making in the European Union, Wallace and Wallace (Eds.), Oxford, Oxford University Press [fourth edition], pp. 317-43. Pelkmans, J., 1991, Toward Economic Union, in: Setting European Community Priorities 1991-1992, Brussels, Centre for European Policy Studies, pp. 39-100. Peters, G.B., 1997, The Commission and Implementation in the European Union: Is there an Implementation Deficit and Why?, in: At the heart of the union: Studies of the European Commission, Nugent, Neill (ed.), London, Macmillan, pp. 187-202. Pirro, E., 2000, The European Union and the Member States: Cooperation, Coordination and Compromise, Lynne Riener, Boulder. Prechal, S., 1995, Directives in Europe and Community Law: A Study of Directives and their Enforcement in National Courts, Oxford, Clarendon Press. Przeworski, A. and H. Teune, 1970, The Logic of Comparative Social Inquiry, New York, Wiley. Puchala, D.J., 1975, Domestic Politics and Regional Harmonization in the European Communities, World Politics, 27(4): 496-520. Richardson, J., 1996, Eroding EU Politics: Implementation Gaps, Cheating and Resteering, in: European Union: Power and policy-making, J. Richardson (ed.), London, Routledge, pp. 278-94. Siedentopf, H., & Ziller, J. (eds.), 1988, Making European policies work.: The implementation of community legislation in the member states. London / Newbury Park / New Delhi: Sage. Snyder, F., 1995, The Effectiveness of European Community Law: Institutions, Processes, Tools and Techniques, in: Implementing EC Law in the United Kingdom: Structures for indirect rule Daintith, Terence (ed.), Chichester, John Wiley & Sons, pp. 51-87. Tallberg, J., 2002, Paths to Compliance: Enforcement, Management, and the European Union, in: International Organization, 56(3): 609-643. Treib, O., 2003, EU Governance, Misfit and the Partisan logic of Domestic Adaptation: An Actor-centered Perspective on the Transposition of EU directives. Paper for the EUSA 8th International Biennial Conference, Nashville, March 27-29, 2003. Köln, Max-PlanckInstitute for the Study of Societies. Weiler, J., 1988, The White Paper and the Application of Community Law, in 1992: One European Market, Bierber, R. et al., (eds.), Nomos, Baden-Baden, pp. 337-58.

FRAMING EUROPEAN INTEGRATION IN GERMANY AND ITALY Is the EU used to justify Pension Reforms? Sabina Stiller Abstract:

1.

This chapter looks at indirect mechanisms of Europeanization in the area of social policy, and in particular at EU-related framing of policy-makers in relation to pension reforms. It focuses on the selective ‘use’ of such framing in order to legitimize pension reform in two countries, Italy and Germany. The main conclusion is that ‘Europeanization by framing’, in the realm of social policy, does not occur in a uniform manner in response to comparable pressures, but is likely to be conditional. The analysis of Italy in the mid- to late 1990s shows that Italian politicians readily used Brussels as a scapegoat, while their German counterparts did not resort to an explicit EU-related discourse to help them justify reforms in 2000/2001. The case studies also point to a possible condition for EU-framing: domestic policy-makers may resort to EU-related framing when they can rely on high levels of EUpopularity among the public.

INTRODUCTION

This chapter fits in the line of the previous contribution on biotechnology and energy policy (see contribution by Van Keulen and Mastenbroek) and the subsequent one on consumer protection and competition policy (see contribution by Kallestrup) in the sense that it explores the influence of Europe on domestic policy-making. Following the affirmative findings of Kallestrup in the Danish context, this contribution raises the issue of whether we can also speak of an impact (albeit an indirect one) of European integration in the realm of social policy reform. Social policy, as an important element of national welfare state arrangements, is an issue area that still falls mostly under the jurisdiction of the member states and is not yet heavily regulated by the EU. Thus, it is typically seen as relatively insulated from EU pressures for harmonization. Placing such considerations about the extent of EU involvement in domestic welfare states aside for the moment, it has been observed that pressures for 43

R. Holzhacker and M. Haverland,(eds.), European Research Reloaded: Cooperation and Integration among Europeanized States, 43-64. © 2006 Springer. Printed in the Netherlands.

44

Sabina Stiller

comprehensive reforms of welfare states have been mounting at least since the 1990s. Empirical evidence suggests that at least in some welfare states, such reforms have been initiated despite a background of considerable institutional and electoral resistance to change (Pierson 2001). In this context of debates dominated by domestic reform pressures, it is instructive to see whether external factors, for example in the form of an indirect impact of European integration, played a role in these processes. In the following chapter, I seek to explore whether framing as an Europeanization mechanism, understood as the use of EU-related arguments to justify reform, plays a role in a relatively underdeveloped EU policy area such as social policy, focusing on old-age pensions. Theoretically, I relate to the debate about ‘soft’ Europeanization mechanisms opened up by scholars such as Radaelli (2000), Knill and Lehmkuhl (2002), and Schmidt (Schmidt 2002a, 2002b). According to Knill and Lehmkuhl, EU-related framing of domestic beliefs occurs in policy areas with ‘symbolic’ EU-involvement in domestic matters. Regarding the conditions of European framing, the nature of dominant belief systems and existing institutional opportunities for and constraints on reform are deemed important (Knill and Lehmkuhl 2002, 263). This study introduces a different conception of framing from the one used by Knill and Lehmkuhl. It focuses on the ‘strategic’ dimension of such framing, i.e. whether key political actors use explicit arguments about the EU to legitimize their push for structural reforms. I argue that such framing in domestic reform debates does not always occur; rather, it is likely to be a conditional process. If the Europeanization mechanism of framing is conditional, it follows that the impact of the EU on domestic social policymaking is also likely to vary, as will be exemplified by the case of pensions. Employing a most similar system design (Landman 2000), I compare the experience of Italy (in the mid- to late 1990s) and Germany (2000/2001) in their attempt to modernize their systems of old-age pension provision. Using evidence from previous studies about the case of Italian socio-economic reforms, I show that the domestic debate on pension reform was heavily framed in terms of EU demands in the run-up to European Monetary Union (EMU). In the second case, I use a qualitative content analysis of various sorts of documents to demonstrate that the reform debate in Germany contained at best indirect references to the EU (and EMU) as a justification for reform. Thus, while Italian politicians readily used Brussels as a scapegoat, their German counterparts did not resort to an explicit EU-related discourse to help them justify reforms. Comparing the two cases demonstrates that EU-(and especially EMU) related framing of the need for social policy reform by domestic politicians does not always occur, but is more likely to be conditional. I conclude by pointing towards a condition for EU-related framing in the area of social policy: it relates to the ability of

Framing European Integration in Germany and Italy

45

domestic policy-makers to tap into public opinion that is clearly in favor of the EU (and EMU) at the time of discussing domestic change. The chapter proceeds in three steps. It begins with an overview of recent theorizing on Europeanization and the welfare state, and introduces my conception of EU-related framing inspired by the recent literature on Europeanization mechanisms and discourse. This section also explains how framing as one such mechanism can be linked to processes of social policy change, focusing on policy-makers’ argumentative strategies using the EU as justification for reforms. Second, I will explain the research design and methodology, followed by the discussion of the two cases, Italy and Germany1. The last section reflects on the substantive findings and concludes by pointing towards one condition under which domestic policymakers could make use of EU-related framing, a condition related to the general level of public support for the EU in a given country.

2.

THEORETICAL ISSUES

2.1

Recent theorizing on Europeanization and the welfare state

At first sight, the social policy arrangements of European member states appear to have undergone little change through the dynamics of European integration so far. Or at least, this is a frequently heard assumption when the social policy dimension of the EU is discussed. Does it also correctly reflect the realities of social policy throughout the European Union? Leibfried and Pierson (2000) have characterized Europe’s social dimension as a ‘unique multi-tiered system of social policy’ with three distinctive characteristics: first, it has a propensity towards deadlock and policy immobilism; second, courts are given a prominent role in policy development; and third, it is marked by negative market integration and thus strongly linked to marketmaking processes. A peculiar situation has thus been created where ‘member states have lost more control over national welfare policies, in the face of the pressures of integrated markets, than the EU has gained de facto in transferred authority’ (Leibfried and Pierson 2000). This erosion of the autonomy of nation states to regulate social policy as they please is arguably proceeding via three processes. First, via direct pressures of ‘positive’ integration aimed at developing uniform social standards at the EU level; second, via direct pressures of ‘negative’ policy reforms through market compatibility requirements; and third, via indirect 1

Due to the different character of evidence used, Germany will be discussed at much greater length than Italy.

46

Sabina Stiller

pressures of integration that lead to adaptation of national welfare states (Leibfried and Pierson 2000, 269). It is first and foremost the last category that is relevant for this study since I wish to explore whether pressures stemming from European integration may act as incentives for domestic policy-makers to undertake much needed social policy reform. Such indirect pressures may be of de facto rather than de jure character as they are not systematically related to the development of EU social policy instruments in the form of directives or regulations. Leibfried and Pierson point to possible underlying factors of such pressures such as competitive demands to adapt national economies in the wake of the creation of the single market, and - of a more recent naturethe single currency area, Economic Monetary Union (EMU).

2.2

Pinpointing the Source of Europeanization: Europeanization through EMU

Since the mere reference to European integration dynamics as a source of Europeanization does not point to any area of economic or political integration in particular, it is necessary to define this source more precisely. In this respect, the impact of EMU on the EU member states has received a lot of scholarly attention. To approach the influence of EMU-related Europeanization effects on welfare arrangements implies a search for linkages between the monetary policy dimension of the EU and domestic social policy at the national level. In section 2.3 below I will suggest one possible linkage in the form of politicians resorting to EU-related argumentation, specifically EU-related framing. In what other ways has the link between supranational monetary policy and domestic pension policy been approached? The literature shows considerable disagreement on whether Economic and Monetary Union is an independent explanatory factor for the reform of European welfare states. On one end of the spectrum, some theorists attribute a great deal of importance to the fiscal constraints that were imposed in connection with monetary union (Dyson 2000; Scharpf 2000). Other authors deny that they have an independent effect, arguing that reforms of European welfare states would have occurred anyhow and were, in fact, long overdue. They usually point to changing socio-economic circumstances, such as lower rates of economic growth and increasing population ageing, that would have made these reforms necessary even in the absence of monetary union (Pierson 2001; Rhodes 2002). In their view, although the fiscal constraints at the European level may act as a catalyst for welfare state reforms, they do not cause them in the first place. A recent contribution to this debate, which perhaps suggests an intermediate position, concerns an ‘institutional interference’ mechanism that is illustrated

Framing European Integration in Germany and Italy

47

through the case of German pension policy developments from the mid1990s onwards (Hering 2003). ‘Institutional interference’ refers to the impact of European fiscal institutions, in particular the EU Stability and Growth Pact, on welfare state reform. More specifically, it describes how European fiscal rules played a role in a series of German pension reforms and led to significant benefit cutbacks. The main argument here is that the requirements set by European fiscal institutions concerning austerity are increasingly in conflict with the goals of contribution-based pension arrangements. More drastically even, European fiscal requirements are said to preclude domestic options to adjust pension arrangements in a context of rising social expenditure. As a result, interference effects are arguably opening a window of opportunity for institutional change in pension policy (Hering 2003). Following this brief overview of theoretical and empirical contributions tracing the link between monetary integration and social policy reform, I proceed to present my argument regarding framing as an indirect Europeanization mechanism.

2.3

Europeanization by argumentation? The role of EUrelated framing and discourse

Rather than starting with a broad question such as ‘How does the EU affect social policy reform?’ I focus on the level of domestic policy-makers and ask ‘How are EU-related arguments used by policy-makers in the context of social policy reform?’ This approach implies a different, i.e. actor-centered conceptualization of EU influence on welfare states and pensions in particular (compare to Van Esch, this volume, for a yet again different conceptualization). The aspect I wish to highlight in my analysis of Europeanization as framing in the field of social policy is how key policy-makers legitimize reform proposals through means of argumentation. This is precisely where I expect the influence of Europeanization (and especially EMU) to enter the stage of domestic policy-making. Arguably, policy-makers in need of persuasive arguments to justify unpopular cuts in social provision may find it convenient to point to Brussels in order to spread the blame expected from groups that are affected negatively by reforms. Ahead of the starting date of EMU, Pierson put it this way: ‘If reforms can be presented as legally required or economically imperative because of the single market or moves toward monetary union, national governments may be freed from some blame for welfare-state cutbacks’ (Pierson 1996, 178). More generally, by ‘politically making use of the EMU criteria as weapons in the national political struggles’, reformers are aiming at overriding defense mechanisms against change (van Kersbergen 2000, 27).

48

Sabina Stiller

Therefore, it becomes relevant to ask whether key policy-makers do indeed make use of arguments that explicitly refer to EU-related pressures in order to justify the need for reform. This leads to the question guiding the review of the Italian case and the closer study of the evidence in the German case: To what extent have key policy-makers been referring to EU- (and especially EMU) related pressures to push through structural pension reform and have such arguments been found to dominate the policy debate surrounding reform? This definition (and operationalization) of framing needs to be seen within a more general debate on the role of framing and/or discourse and Europeanization mechanisms. In an article that reflects on the conceptual achievements of the Europeanization literature, Radaelli distinguishes three such mechanisms: positive integration, negative integration and the ‘soft’ mechanisms of framing (Radaelli 2000a). This distinction is inspired by earlier work by Knill and Lehmkuhl (1999) who have more recently drawn up a comprehensive framework to explain the domestic impact of European policy-making. It distinguishes three Europeanization mechanisms: institutional compliance, changing domestic opportunity structures, and, most relevant for the argument here, framing domestic beliefs and expectations (Knill and Lehmkuhl 2002). As to the substance and direction of framing, Schmidt’s work on discourse is important. Her definition of discourse includes both an ideational dimension, stressing the cognitive and normative contents of a given message or argument and an interactive one, referring to the different functions of a discourse (Schmidt 2002b). For the present work, the former dimension is most relevant, as it stresses that policy-makers need to follow the double logics of necessity and appropriateness in order to legitimate their ideas vis-à-vis reform critics and the electorate. With respect to European framing, Schmidt’s analysis of national discourses providing legitimacy to Europe is significant. She argues that the process of adaptation to Europe consists of three dimensions – economic, institutional and discursive – and that the type of discourse used to justify economic and institutional adaptation matters most in terms of legitimization (Schmidt 1997 cited in Radaelli 2000a, 13). In this study, I take such ‘discursive’ adaptation to mean EU-related framing by domestic policy-makers. As to the ways in which framing can make an impact, Knill and Lehmkuhl expect an emergence of domestic policies based on this mechanism in areas with little more than ‘symbolic’ EU involvement or ‘policies that…are designed to increase support for domestic reforms that may facilitate future steps towards further integration’(Knill and Lehmkuhl 2002, 271). Furthermore, they specify that the impact of European framing depends on the dominant national belief systems (where European ideas may possibly act as a ‘focal point’ for domestic problems) and existing

Framing European Integration in Germany and Italy

49

institutional opportunities for and constraints on reform (where European ideas may help to establish a domestic reform consensus). In my view, however, this conception of framing should be restricted to policy areas where European involvement suggests a model for imitation or implementation and remains somewhat vague when it comes to specifying how domestic policy-makers relate to the opportunities for ‘using’ the EU (see also the contribution by Kallestrup, this volume, on the latter point). First, their conception of framing does not allow for the possibility that EU pressures that do not relate to such a model, such as the ones originating from EMU, may equally influence domestic beliefs and expectations concerning other policy areas. Second, it remains vague on how this influence appears and materializes in domestic policy-making. These two points of critique are to be addressed by the definition of framing in this study. Concerning the first point, the Knill and Lehmkuhl framework seems to tie the conditions for framing to the ‘domestic impact of European framing in any particular area’, which can be read as excluding a framing mechanism stemming from European-level developments in a different policy area. But what about instances where a policy area, e.g. social policy, is susceptible to pressures from European-level developments in a different policy area, e.g. monetary policy? As discussed in section 2.2, the literature on EMU and its effects on national welfare states suggests at least some sort of linkage, notwithstanding the disagreements about its precise form. Framing conceived as justifying reform in terms of EU-related arguments is able to accommodate the possibility of ‘cross-sectoral’ framing. Regarding the second point, i.e. making the framing of ‘domestic beliefs and expectations’ more concrete, I draw on Schmidt’s concept of the double logics of discourse to further fill in the link between EU pressures, which provides arguments for the necessity and appropriateness of cutting back extensive pension policies and fostering structural changes in pension systems within domestic reform debates. Rather than tracing the different dimensions of discourse and linking them to successful adjustment of policies, as Schmidt does, I merely concentrate on the legitimizing function of what she calls discourse and look at whether politicians exploit arguments about EU/EMU-pressures for the purpose of legitimizing reform plans.

50

3.

Sabina Stiller

RESEARCH DESIGN AND METHODOLOGY: COMPARING AND CONTRASTING THE EXPERIENCE OF ITALY AND GERMANY

In order to explore the effects of European framing, I employ a most similar systems design (Landman 2000) choosing Italy and Germany as the context of analysis. ‘Most similar’ here refers to both political-institutional as well as welfare-institutional characteristics of these countries2. As to the former, both are parliamentary democracies of the consensual type where the legislature dominates the executive and parliament can withdraw support from the government (Lijphart 1984). Furthermore, in both countries comprehensive welfare and fiscal reforms tend to be difficult, albeit for different reasons: Germany features a rather large number of veto points, not at least because of its bicameral parliament where the upper chamber (Bundesrat) is frequently dominated by a different party majority and thus able to block legislation (Scharpf 1988). On the other hand, in Italy governments tend to be highly fragmented and ideologically polarized coalitions often make the negotiation and implementation of welfare and fiscal policy reforms quite difficult (Sartori 1976). Thus, in both countries policy change becomes a complicated affair since many hurdles have to be taken in the process. In the context of social policy reform, where radical changes to distributive policies potentially inflict losses on beneficiaries and powerful interests, the existence of many veto points tends to be counterproductive to meaningful policy change. Secondly, in terms of old-age pension arrangements, both countries feature Bismarckian welfare state-type institutions that mainly rest on the social insurance principle. This type of welfare state institutions, in turn, is associated with inbuilt adjustment pressures in the face of present socioeconomic change and demographic developments (Esping-Andersen 1999; Scharpf 1997; Esping-Andersen 1996). Especially the area of pensions is considered to be in need of reform in the face of graying populations and its grave implications for the future sustainability of current pension levels. Both Italy and Germany are thus facing similar problem pressures3, but the question is whether policy-makers used similar EU-related framing strategies in order to push for structural changes. 2

I do not claim here that the political institutions of Italy and Germany are identical in all respects, yet I consider that the commonalities are striking enough to justify the label ‘most similar’. 3 Again, I do not claim that the problem of pension insurance schemes in both countries are exactly comparable in terms of magnitude, but rather in the type of problems they face which are common to Bismarckian type welfare regimes (adapting contribution-based schemes to the challenges of low birth-rates, graying, financial strain as result of high unemployment etc.).

Framing European Integration in Germany and Italy

51

The first part of the case discussion summarizes the reform experience of Italy in the 1990s. Due to the fact that several in-depth analyses of the Italian case are available, I draw on these analyses of policy and institutional adjustment for the country’s run-up to EMU membership as a base for my case study. The aim here is to see whether EU-level pressures were actively taken up by Italian politicians in order to legitimize far-reaching reforms, especially in the area of pension reform. The second and empirical part of the case discussion focuses on an important episode of pension reform in Germany in the period 2000/2001. The methodology used here is different, partially because of the recent character of the reform and also for pragmatic reasons. That case study consisted of applying qualitative content analysis to several sorts of documents that reveal the contents of the policy debate. The object of interest in analyzing these documents was whether arguments, which explicitly refer to the EU, also penetrated the German debate about pension reform in order to legitimize it. First, I searched for evidence of EU-related arguments used by policy-makers representing the reformist line. Reformists include the responsible Social Democrat minister Walter Riester as well as Social Democrat parliamentarians, who can be assumed to use similar arguments. Apart from screening the material for EU-related lines of justification, I also took note of other sorts of arguments that were used to justify the reform proposals as they were presented. This procedure was repeated for arguments of the opposition Christian Democrats and interest groups and experts, as will be illustrated in the summary of the results below (see section 4.2).

4.

DISCUSSION

Let us now turn to the Italian experience of pension reforms in the 1990s, drawing on previous studies, before proceeding to the analysis of the German 2000/2001 pension reform.

4.1

EU-related framing and domestic policy-making: Italy and EMU in the 1990s

With respect to the period after the Maastricht treaty (1992) until the late 1990s, the work of Ferrera and Gualmini offers valuable insights into the role of EU-induced exogenous pressures in domestic policy-making. Their study deals with the Italian processes of national socio-economic adaptation which they see grounded in both deep domestic political change, and, relevant for the present argument, the transformation of the international economic environment (Ferrera and Gualmini 2004, 2000a; Ferrera and

52

Sabina Stiller

Gualmini 2000b)4. These two factors in tandem, they argue, ‘forced Italian politicians to modify their strategies substantially and introduce important policy innovations’. While Italy had started to adjust to exogenous challenges when entering the European Monetary System (EMS) in 1979, pressures intensified greatly after the signing of the Maastricht treaty, when ‘the process of ‘restoring to health’(risanamento) public finances (and hence economic and social policy) became more a necessity than a virtue’ (Ferrera and Gualmini 2000b). The Maastricht criteria that laid down the conditions for EMU membership thus came to play a crucial role in social policy reform in Italy, providing an argument to policy-makers that could not simply be ignored: ‘The vincolo esterno (i.e. the constraint imposed directly or indirectly by international regimes and especially by the EU) has in fact come to be an increasingly powerful stimulus for pushing through measures of welfare retrenchment and rationalization on the part of the national executive’ . What was the link between fiscal discipline and social policy, and especially pension reform? If Italy wanted to meet the Maastricht criteria in order to join EMU in the first wave of countries, it had to take serious action against its public debt and deficit. While tax increases were ruled out to achieve this goal, radical spending cuts remained the only realistic option. The new policy of risanamento necessitated strategies to contain both debt and public deficit, and social benefits, among which pension expenditures figured prominently. The relatively large share of total public expenditure, the precarious state of both pension and health care financing, and an already negative image in terms of legality and efficiency explain why social policies drew most attention of Italian reformers (Ferrera and Gualmini 2000b, 378). In the process, EU-related framing was clearly used to justify the sacrifices vis-à-vis the electorate as well as domestic opposition such as interest groups. Several examples will suffice to underline this point. Regarding EU-related framing as part of the domestic discourse, the argumentation surrounding the introduction of the ‘Euro-tax’ (a temporary income surtax) is instructive. The measure was justified by the government as being the ‘price of the last ticket to Europe’, becoming synonymous for the willingness to avoid exclusion ‘from Europe’. Indeed, the tone of the discourse sometimes took on dramatic forms, as Radaelli argues, ‘both the financial crisis of 1992 and the danger of ‘being left out’ in 1996-97 were converted into political opportunities by telling Italians that they were facing a question of ‘life or death’, an apocalyptic message’ (Radaelli 2000b, 226; 4

In their subsequent (and more theoretically grounded) analysis of 2004, the authors draw up a more comprehensive explanation of Italian socio-economic reforms in the 1990s. They reiterate that external developments, including the Maastricht process and internationalization dynamics were important for changing domestic actor constellations and preparing for failureinduced learning dynamics.

Framing European Integration in Germany and Italy

53

Radaelli 2000a). Before concluding the 1995 landmark agreement with the trade unions on pension reform, then Prime Minister Dini had managed to persuade them of the awaited long-term advantages of risanamento and EMU membership in the form of renewed economic growth and reduced interest on public debt (Ferrera and Gualmini 2000a; Radaelli 2000b). Radaelli summarizes the impact of EU-related framing as follows: ‘The risanamento has been presented to Italians as a problem of finance bills and the DPEF (finance and planning document), convergence criteria, and speed of adjustment […] In terms of communicative discourse, the Italian policy makers have been able to craft an effective discourse that secured legitimacy for hard economic choices such as the Euro-tax’ (Radaelli 2000b, 236). The task of creating such a discourse was facilitated by three mediating factors that have supported the double process of financial adaptation and institutional change. First, Italian key policy-makers in charge of reforming public finances found themselves empowered by the Maastricht criteria. Second, Italian elites in general showed a nearly universal support for European integration that made them sympathetic to arguments for austerity framed in the name of ‘Europe’. And finally, a ‘permissive consensus’ on Europe among the Italian public was crucial for those elites, enabling them to overcome opposition at crucial moments in the process of cutting public finances and pension expenditure (Sbragia 2001). A recent comparative study of the politics of pension reform further underlines the importance of the Maastricht treaty and thus European integration as explanatory factor for the Italian changes in pension arrangements (Schludi 2002). The analysis highlights the reasons behind a series of pension reforms that amounted to a turnaround in Italy after 1992. Amongst those reasons, Schludi argues, ‘the dynamics of European integration (in particular the fiscal pressure imposed by the EMU criteria) and its beneficial impact on the political discourse at the national level has been an important driving force behind the reform of the Italian welfare state in general’. More specifically, pressure from EMU enabled several Italian prime ministers including Dini and Prodi to ‘spread an understanding of the costs of non-adjustment, pointing to the necessity of sacrifices in the short run (such as benefit cuts) in exchange for the promise of future benefits’ (Schludi 2002, 30-31). Interestingly, not only the convergence criteria in themselves put on the pressure, but also the threat of several European states to veto Italy’s entry into EMU if it would not tackle pension reform beforehand (Cioccia et al., 2001, cited in Schludi 2002). The evidence presented here clearly shows that Italian policy-makers used EU-related arguments to legitimate pension reforms as part of a larger process of financial adjustments in order to make the country fit for EMU membership. Their efforts eventually proved successful since Italy fulfilled the criteria in time to join EMU in the spring of 1998.

Sabina Stiller

54

Could the Italian experience of pension reform, achieved with the help of European monetary integration, repeat itself in other member states whose welfare state arrangements equally need reform? That is to say, does EMU have Europeanization effects that may trigger the adaptation of domestic social policy in other countries? Or was it rather a singular event tied to a particular confluence of circumstances? To explore this question further, let us now turn to the German case of pension reform in 2000/2001.

4.2

Europeanizing the discourse on social policy reform? The German 2001 pension reforms

4.2.1

Introducing the case: What kind of reforms?

The 2001 pension reforms in Germany were the result of almost two years of frequently changed reform proposals and extensive talks between the government and the opposition on the one hand, and interest groups, especially trade unions, on the other5. The actual parliamentary process took half a year to be completed for the most controversial bill of the three that constituted the whole of the reforms. The two major pieces of legislation that emerged from this period were the Old-Age Provision Act and the Old-Age Provision Extension Act. The first, containing the provisions establishing a new private pension tier, was passed in May 2001 (Bundesgesetzblatt, 29 June 2001). The second, including the legal provisions for benefit cutbacks in the public pension scheme, was passed in January 2001 (Bundesgesetzblatt, 26 March 2001). Yet another change, the reform of reduced capacity and disability pensions was passed as a separate law in December 2000 (Bundesgesetzblatt, 23 December 2000). The following discussion will concentrate on the first two laws. What is the essence of these pension reforms? The main outcome is certainly the creation of a new private pension tier which is voluntary, sizable, and taxsubsidized. In this newly created private scheme, occupational pensions are favored over individual pensions. The introduction of a private pension tier is combined with benefit cutbacks in the public pension scheme which are substantial, but not radical. Moreover, contribution rates are set to be stabilized in the long-term at a level just above the year 2000 level. Yet these provisions reflect a long series of compromises since the original reform proposals went much further. At least four major corrections have been made in the course of the reform debate, either before or during the deliberations in parliament. First, the original draft provided for a mandatory private pension tier. However, 5 This section draws on information compiled in Dunn und Fasshauer (2001) and Hering (2001).

Framing European Integration in Germany and Italy

55

this element had to be withdrawn in reaction to massive protests of parties, lobby groups and the public. Second, the size of the private pension tier turned out to be greater than originally designed. While contributions were first amounting to 2.5 percent of gross income to be gradually phased in until 2007, the law included the provision of 4 percent of gross income to be reached by 2008. The greatest driving force behind this upsizing was the decision of the government to agree to cutbacks of the public pension level, apparently under pressure from the Christian Democrats. These cutbacks had to be compensated, however, by corresponding increases in private provision given the previous commitment of the Social Democrats to hold on to a combined pension level of 70 percent. The third major change concerned the extent to which the state would help individuals to finance the new private pension contributions. During the course of the extra-parliamentary government-opposition talks and several rounds of ‘bidding’, the envisaged modest support for citizens with low incomes was finally extended in the form of tax credits for middle and higher incomes as well, and earmarked as an amount of EURO 10 billion annually. Finally, the role of collectively negotiated occupational pension schemes was strengthened vis-à-vis individual pension plans. As a result of mainly trade union pressure pushing for a contribution of employers to employees private pension savings, a compromise was reached which stipulated that collectively negotiated occupational pensions take precedence over individual pension plans. 4.2.2

Who argued what? Party-political and policy debates on the Riester proposals

Let us now turn to the results of the analyses, starting with speeches and articles of the labor minister Walter Riester. To what extent did he refer to EU-induced pressures in presenting his reform project? One general observation that can be made across all sorts of texts considered is the following: the goals of a structural pension reform were generally given a lot more room in his argumentation than arguments about its necessity, although sometimes the goals were directly linked to such necessity. In a 1999 speech to fellow Social Democrats titled ‘The future of the welfare state’, structural pension reform figured prominently under the reform projects that were to be tackled in the remainder of the cabinet period (Riester 1999). Before discussing the necessity of reform, a list of goals was presented. They included holding on to the ‘generational contract’, a reliable public pension insurance, stable contribution rates at an acceptable level, a stable benefit level and the creation of additional pension provision next to the public provision. The features of the reform package to secure pension provision were presented as a ‘modern and socially balanced solution of the problems of the public pension system’. These problems were then clearly linked to

56

Sabina Stiller

goals that are presented as being more fundamental than the policy-specific goals just named: the need to consolidate public pensions and the federal budget as well as stabilizing contribution rates. All of these can be summarized under the label public spending containment, but a reference to the EU remains absent. In contrast, in another context referring to the opposition’s proposals to finance tax decreases by increasing public debt, the incompatibility of such a policy with the EU Growth and Stability Pact was explicitly pointed out. In an essay intended to take stock of the Social Democrats’ reform achievements halfway through their term in late 2000, two reasons were given for the necessity of structural pension reforms (Riester 2000). While adverse demographic developments were acknowledged, the role of pension contributions in the total of non-wage labor costs was given prominence. Further, Riester made clear that the reform of public pensions cannot be seen independently from general economic policy and, most importantly, should be in line with budgetary policy goals. Here, the climb of the federal taxfinanced contribution to the public pension fund, a growing post in the federal budget, was explicitly referred to. Again, the necessity of reform in connection to budgetary concerns can be found there, albeit without recourse to the underlying constraints on budgetary policy, the EU Maastricht criteria. The goal of budgetary consolidation was again presented separately from the goals directly connected to the characteristics and shortcomings of the public pension scheme: modernization of the current contribution-based system, long-term stabilization of pension provision, achieving a balance between generations, preventing old-age poverty and improving the pension rights of women who have raised children and surviving dependants. Next, let us consider a different sort of texts, namely those documenting the policy debate on reform as it developed during the parliamentary process. Documents used include the bill of the Old-Age Pension Reform Act, a motion by the opposition Christian Democrats urging the government to provide for a ‘sustainable’ pension provision, the parliamentary protocols illustrating the plenary readings of the bill in 2000 and 2001, and summaries of public hearings of experts and interest groups for the parliamentary committee for labor and social affairs. Rather than considering them on a one-by-one basis or in chronological order, I will give a summary of the arguments used to underline the necessity for structural reform by the reformists represented by the government. In addition, I will deal with the reasoning of the opposition and of interest groups and experts. The government position, as formulated in the bill drawn up by the Ministry of Labor and Social Affairs (Deutscher Bundestag 2000a), pointed to one main problem that had to be addressed. Long-term demographic trends were highlighted as the most important reason for and driving force

Framing European Integration in Germany and Italy

57

behind the need for a reform that makes public pension sustainable in the long term. The necessity for creating a private pension tier that enables German citizens to build up additional funds for retirement was directly linked to achieving such sustainability. The goal attached to the measures that should help solve the pension problem was clear: the young generation must not be overburdened with contributions while it should still be entitled to a reasonable pension. The stabilization of contributions linked with this goal should also be an important condition for more growth and employment as well as securing the country’s competitiveness. In the debates during the bill’s readings in parliament, the reformist position reiterated the focus on the demographic trap: in the first reading, minister Riester again put first the need ‘to find a fair balance between generations’. This translated into four goals which he discussed at length: to supplement public pensions with an additional private pension provision, to decrease pension contributions, to support people living on small pensions and to end old-age poverty (Deutscher Bundestag 2000b). At the second reading, the chair of the parliamentary committee made clear that the bill was ‘the answer to the altered and further changing age composition of the population’. Comprehensive reform was the necessary response if the options of increasing contributions, decreasing benefit levels, or increasing eligibility age were not considered acceptable solutions. Minister Riester defended his work pointing to how the goals stipulated earlier were going to be achieved if the bill – by then modified several times – was going to become law (Deutscher Bundestag 2001a). During the last debate to precede the final vote in both Bundestag and Bundesrat in May 2001, arguments about the necessity of the reform were no longer visible. Minister Riester instead focused on insisting that ‘the reform would only produce winners’ and on regretting the opposition’s supposed strategy of blockade and procrastination. Furthermore, party-political argumentation dominated factual argumentation, perhaps not unexpectedly so given the duration of a two-year long discussion process (Deutscher Bundestag 2001c). In any case, what becomes clear is the absence of explicit reference of the government discourse to the EU-level for justification of comprehensive pension reform. Did the opposition, especially the Christian Democrats, make such references in their party-political debate on reform? Their parliamentary motion dating back to June 1999 demanding certain measures to stabilize the institutional framework of public pensions did not look promising in this respect (Deutscher Bundestag 1999). It started from the assumption that the ‘generational contract upon which public pensions rest has been disturbed mostly due to reasons related to demography, the consequence being drastically increased financial pressures bearing on the system. To counter these developments, the government was called upon to present a concept

58

Sabina Stiller

that promotes occupational and private pensions, to improve the independence of women in terms of pension provision, and to illustrate the generational shift by presenting annual balance sheets. The opposition’s contributions to the parliamentary debates centered less on getting across their version of why reform was necessary – it is obvious that their justification does not differ so much from the one given by the government – and more on what they saw as elements of trickery in the bill (Deutscher Bundestag 2001a, Deutscher Bundestag 2001c). For instance, Riester’s ‘compensation factor’, an addition to the pension calculation formula designed to lower the level of pensions over time, was heavily attacked as disadvantaging the younger generation while it was presented as fitting the general goal of sustaining a balance between the generations. Furthermore, many details concerning the standards for private pension schemes generally marked as ‘too bureaucratic and complicated’ and their subsidization remained controversial between the two political camps. In addition, the factual discussion between government and opposition was overshadowed by a contentious advertising campaign of the CDU in their bid to denounce the Riester reform plans. As a result, it can be safely said that the Christian Democrats made no attempt to lay a link between the need for reform and pressures coming from the EU-level. Were EU-related pressures at all part of the overall public debate that ensued in reaction to the Riester reform plans? In order to gain an impression of which issues figured in such debates, I considered a selection of interest group and expert opinions. The latter were presented at the occasion of public hearings held for the parliamentary committee of labor and social affairs (Deutscher Bundestag 2001b)6. As a general observation, it is to be expected that such statements vary widely between organizations representing employers, employees and beneficiaries, but also independent institutions such as the Bundesbank, advisory bodies and individual experts chose to accentuate very different aspects of the proposed changes to the state-run pension scheme. A selection of their opinions will suffice to illustrate the range of issues raised in the hearings. As for the employers’ representatives, the German Association of Industry (BDI) pointed out the ‘compensation factor’ as the main weakness of the bill, burdening the younger generation disproportionally. On the other hand, the German Association of Employers (BDA) doubted the sustainability of the proposed reform measures while criticizing the proposed height of the ceiling of

6

The following discussion draws on final report of the Committee of Labor and Social Affairs which summarizes the contributions to the parliamentary hearings.

Framing European Integration in Germany and Italy

59

contributions at 22 percent in 2030 as economically and socially not justifiable. Trade unions and their umbrella organization, the German Trade Union Federation (DGB), also diverged in their focus: the DGB expressed its worries about a fair balance of the burden between the state, the ones paying contributions and those receiving benefits. Based on the government proposals, only employers would clearly profit from the planned ceiling on the contribution rate while the insured would have to invest additionally into private schemes. Some large unions, such as the employees’ union (DAG) qualified the proposals as continuation of earlier measures aimed to dismantle the welfare state, while the metalworkers’ union (IG Metall) did not even see the necessity of the reform plans since it believed in the continued viability of the current contribution-based system despite demographic developments. On the other hand, the chemical and mineworkers union (IG BCE) greeted the reform plans in principle despite its criticism of the planned pension levels as too low. The Bundesbank and economic advisory bodies were united in their agreement about the plans’ goal of achieving greater intergenerational solidarity. For the Bundesbank, holding on to the status-quo in pension policy would mean to increase unduly the burden on later generations. The IFO Institute for Economic Research pointed to the fact that no one should complain at the same time about lower pension levels and higher contribution rates when ruling out either the increase of the federal subsidy to pensions or the transition to greater capital-cover for the state-run scheme. The opinions of other institutes differed in their focus: the Cologne Institute of German Economy recommended that the contribution rate needs to be stabilized at a level of 20 percent with a view to labor market policy, while the Institute for World Economics welcomed the decision to limit the hike of contribution rates and undertake cuts in pension levels, stressing that balancing the burden between contributors and beneficiaries adequately remains a political decision. Finally, experts from academic institutions tended to criticize different features of the measures contained in the reform plans. The compensation factor was either seen as insufficient ‘in the light of achieving a fair balance between generations at the expense of later generations’, as Professor Rürup argued (Deutscher Bundestag 2001b:18) or, together with the envisaged private contribution, as ‘lacking systematic reasoning, manipulable and arbitrary’, according to Professor Schmähl (Deutscher Bundestag 2001b:19). Others criticized the total absence of measures that regulate the pension entry age or the increase of intransparency that would impair both confidence in and acceptance of the system on the part of beneficiaries. In connection with the latter argument, Professor Bäcker pointed out that the discussion about the extent of the pension burden to be expected by current

60

Sabina Stiller

and future contributors failed to pay attention to the height of the contributions made for private pension plans (Deutscher Bundestag 2001b:18). Summing up the various contributions to the public hearings, only few comments and criticisms of the ones considered were directed at the necessity for a reform that aims to make state pensions financially viable for the future by partially privatizing public pensions. Rather, they took issue with the detailed measures by which inter-generational solidarity, one of the main government goals, is to be achieved. Key features of the proposed reform, such as contribution and benefit levels and the pension calculation formula were singled out for being ‘too high’, ‘too low’ or ‘going not far enough’. What stands out, however, is that the interest group, advisory and expert statements hardly gave prominence to the need for reforms of the current system and its underlying reasons. Possibly, it is not their task to provide the context of justification to the same extent as the government which has to propose and justify the need for reforms. However, that should not stop them from giving their view of why measures that are likely to create both winners and losers and change existing institutional structures, are in place. To be sure, whenever a context of justification was mentioned, it referred to pressures created by the graying of the population, a growing imbalance between generations and the ongoing bleak situation on the labor market. However, nowhere a direct reference to the European context was made. If this spectrum of interest groups and experts cannot be shown to entertain such ideas, though, it is hard to claim that EU-pressures did form part of the public debate. 4.2.3

A debate dominated by domestic reform pressures: whither Europeanization?

To conclude on the findings from the German case, what have we seen in terms of key politicians referring to EU/EMU-related pressures to justify their pension reform proposals? Have such types of argument entered the policy debates surrounding reform? The answer to these questions is fairly obvious. Both the terms ‘European Union’ and ‘Economic and Monetary Union’ remained conspicuously absent from the Social Democrats reformist line of argumentation in favor of pension reform. Neither did the Christian Democrat opposition find it appropriate to bring up the pressures present on the EU-level in its criticism of the government’s plans. The same held true for a selection of the different types of contributors to the public hearings which can be taken as a good indication for the policy debate beyond the party-political context. The arguments that were brought forward to justify the need to reform public pension arrangements were domestic ones. Socio-economic changes

Framing European Integration in Germany and Italy

61

and demographic processes affecting the make-up of German society as well as slow economic growth and high unemployment were found to be responsible for the dire financial situation of the public pension scheme. Such argumentation reflects a belief that the problems affecting pension policy were essentially domestic and internal. Therefore, the analysis of the arguments justifying pension reform suggests that the EU-level did not play a role in the political or the wider public discussion. Coming back to the question whether EU-related framing influences social policy, the conclusion has to be as follows: given the operationalization of the term as reformist policy-makers making their case for social policy reforms by reference to EMU-related pressures, EU-related framing was absent in the case of German pension reform.

5.

CONCLUSION

In sum, the comparison between the Italian and German cases has shown that EMU-induced European framing effects in domestic social policy arrangements do not occur in both cases. The review of the Italian pension reform episode reveals that the domestic reform debate was heavily framed in terms of EU/EMU demands in the run-up to EMU. In contrast, the more substantive analysis of documents in the German case demonstrates the absence of straightforward reference to the EU/EMU as a rationale for reform. As the data suggest, domestic concerns about demographic developments and financial sustainability were evoked as rationale for reform, but not external pressure from the EU. This outcome indicates that the impact of EU-framing in the area of social policy is likely to be conditional and I will sketch one such condition further below. The straightforward outcome of the German case leads me to conclude that policy debates on social policy reform in EU member states are likely to remain dominated by domestic arguments. Yet, in order to corroborate this conclusion, further research needs to be undertaken covering more cases of social policy reform and their accompanying policy debates in a greater number of EU member states. Finally, the comparison of Italy and Germany leads to the observation of a tentative condition for EU/EMU-related framing in the issue area of social policy. As such, it can be seen as complementary to the more general conditions formulated by Knill and Lehmkuhl (see section 2.3). Both the Italian and the German case suggest that the strength of public support towards the EU may play a role in whether policy-makers resort to framing in the presence of EU-level prescriptions. One can think here of the pressures inherent in the EMU convergence criteria (this applied to Italy before 1998) and of the pressures inherent in the EU Stability and Growth

Sabina Stiller

62

Pact (this applied to Germany in 2000/01). As I have shown, while Italian policy-makers were anxious to act in order to meet the EMU criteria and converted this pressure into a persuasive EU-related discourse, German reformists apparently saw no immediate reason to refer to EMU to justify their reform plans. Perhaps the latter was due to the fact that the country still met the prescriptions of the Stability and Growth Pact (in particular the 3 percent budget deficit criterion) at the beginning of the decade. However, whether politicians made use of EU-level arguments in these seemingly comparable contexts in terms of pension reform pressures, may also be linked to the level of public support for the EU which policy-makers arguably take into account before resorting to such a strategy. In this respect, Italian politicians could rely on traditionally high levels of approval for the European Union and especially EMU, while Germans tend to be more skeptical about the EU and the introduction of the Euro (see on Germany Brettschneider, et al., 2003). If one compares the approval rating for EU membership and support of the single currency in these two countries, the difference is indeed notable. In late 1997, the year before joining EMU, 78 percent of Italians approved of the Euro and 69 percent of EU membership (compared to 40 respectively 38 percent of Germans, Eurobarometer 1998). Radaelli points to the fact that at the end of 1999, despite previous years of budgetary stringency, Italy still boasted the highest percentage of supporters of the European currency (European Commission 2000d, cited in Radaelli 2000b, 226). On the other hand, in early 2001, when German pension reform entered its crucial phase, only 45 percent of Germans approved of EU membership and 53 percent supported the Euro (Eurobarometer 2001). Needless to say, these figures only sketch a general picture by lending support to a possible link between EU popularity levels and EU-framing; further systematic research on these two countries and beyond would be needed to substantiate these claims.

APPENDIX Case Study Material: References and documents used for analysis of the German case Publication of legislation Bundesgesetzblatt, “Gesetz zur Reform der gesetzlichen Rentenversicherung und zur Förderung eines kapitalgedeckten Altersvorsorgevermögens (Altersvermögensgesetz - AVmG),” Teil I, 29 June 2001. Bundesgesetzblatt, “Gesetz zur Ergänzung des Gesetzes zur Reform der gesetzlichen Rentenversicherung und zur Förderung eines kapitalgedeckten Altersvorsorgevermögens (Altersvermögensergänzungsgesetz - AVmEG), ” Teil I, 26 March 2001. Bundesgesetzblatt, “Gesetz zur Reform der Renten wegen verminderter Erwerbsfähigkeit,” Teil I, 23 December 2000.

Framing European Integration in Germany and Italy

63

Bills, opposition party motions, parliamentary reform debates and report of expert hearings: Deutscher Bundestag (1999). Drucksache 14/1310, CDU-CSU Fraktion. Berlin, Deutscher Bundestag. Deutscher Bundestag (2000a). Drucksache 14/4595, Gesetzentwurf, Fraktionen SPD und Bündnis 90/Die Grünen. Berlin, Deutscher Bundestag. Deutscher Bundestag (2000b). Plenarprotokoll 14/133. Berlin, Deutscher Bundestag. Deutscher Bundestag (2001a). Plenarprotokoll 14/147. Berlin, Deutscher Bundestag. Deutscher Bundestag (2001b). Drucksache 14/5150, Ausschuss für Arbeit und Sozialordnung. Berlin, Deutscher Bundestag. Deutscher Bundestag (2001c). Plenarprotokoll 14/168. Berlin, Deutscher Bundestag. Speeches, opinion articles and interviews: Riester, W. (1999). ‘Die Zukunft des Sozialstaats’, speech delivered to the annual conference of the Social Democratic Party, 19 November 1999. Riester, W. (2000). ‘Halbzeit, Reform-Zeit: Zeit der Ernte’ Gewerkschaftliche Monatshefte (10). Riester, W. (2001) ‘Rede des Bundesministers für Arbeit und Sozialordnung zur Rentenpolitik in der Debatte der 168. Sitzung im Deutschen Bundestag’, speech delivered to the German parliament, 11 May 2001. Der Spiegel 46/2000, ‘Wir tricksen nicht’, interview with minister Riester, p. 25-26 Süddeutsche Zeitung, 25 January 2001, ‘Ich bin ja kein Masochist’, interview with minister Riester, p. 27 DGB Infodienst Einblick, 23 October 2000, ‘Interview mit Bundesarbeitsminister Walter Riester zur Rentenreform’, interview with minister Riester, http://www.einblick.dgb.de/archiv/0019/tx001901.htm

REFERENCES Brettschneider, F., Maier, M., and Maier, J., 2003, From D-Mark to Euro: The impact of mass media on public opinion in Germany, German Politics 12(2):45-64. Dünn, S., and Fasshauer, S., 2001, Die Rentenreform 2000/2001 - ein Rückblick, Deutsche Rentenversicherung 56(5):266-75. Dyson, K., 2000, The Politics of the Euro-Zone: Stability or Breakdown?, OUP, Oxford. Esping-Andersen, G., 1996, Welfare States in Transition. National Adaptations in Global Economies, Sage, London. Esping-Andersen, G., 1999, Social Foundations of Postindustrial Economies, OUP, Oxford. Eurobarometer, 1998, Survey 48, Autumn 1997, European Commission, Brussels, http://europa.eu.int/comm/dg10/epo. Eurobarometer, 2001, Survey 55, Spring 2001, European Commission, Brussels, http://europa.eu.int/comm/dg10/epo. Ferrera, M., and Gualmini, E., 2000a, Reforms guided by consensus: the welfare state in the Italian transition, West European Politics 23(2):189-208. Ferrera, M., and Gualmini, E., 2000b, Italy - rescue from without?, in: Work and Welfare in the Open Economy Vol II: Diverse Responses to Common Challenges, F. Scharpf and V. A. Schmidt, eds, OUP, Oxford, pp. 351-98. Ferrera, M., and Gualmini, E., 2004, Rescued from Europe? Social and Labor Market Reforms in Italy from Maastricht to Berlusconi, Amsterdam University Press, Amsterdam. Hering, M., 2001, Pension reform under the Schröder government, unpublished.

64

Sabina Stiller

Hering, M., 2003, Institutional Interference in the European Union: The Stability Pact and the Reform of Public Pensions in Germany, paper for European Union Studies Association 8th Biennial International Conference, Nashville, March 27-29, 2003. Knill, C., and Lehmkuhl, D., 2002, The national impact of European Union regulatory policy: three Europeanization mechanisms, European Journal of Political Research 41(2):255-80. Landman, T., 2000, Issues and Methods in Comparative Politics: An Introduction, Routledge, London. Leibfried, S., and Pierson, P., 2000, Social policy - left to courts and markets?, in: PolicyMaking in the EU, H. Wallace and W. Wallace, eds, OUP, Oxford, pp. 267-92. Lijphart, A., 1984, Democracies: Patterns of Majoritarian and Consensus Government in Twenty-one countries, Yale University Press, New Haven. Pierson, P., 1996, The new politics of the welfare state, World Politics 48(1): 141-79. Pierson, P., 2001, Coping with permanent austerity: welfare state restructuring in affluent democracies, in: The New Politics of the W elfare State, P. Pierson, ed., OUP, Oxford, pp. 410-56. Radaelli, C. M., 2000a, Whither Europeanization? Concept stretching and substantive change, European Integration online Papers 4(8), http://eiop.or.at/eiop/texte/2000-008.htm. Radaelli, C. M., 2000b, The Italian state and the Euro: institutions, discourse, and policy regimes, in: European States and the Euro: Europeanization, Variation and Convergence, K. Dyson, ed., OUP, Oxford, pp. 213-37. Rhodes, M., 2002, Why EMU is, or may be, good for European welfare states, in: European States and the Euro: Europeanization, Variation and Convergence, K. Dyson, ed., OUP, Oxford, pp. 305-33. Sartori, G., 1976, Parties and Party Systems: a Framework for Analysis, Cambridge University Press, Cambridge. Sbragia, A., 2001, Italy pays for Europe: political leadership, political choice, and institutional adaptation, in: Transforming Europe: Europeanization and domestic change, M. G. Cowles, J. Caporaso and T. Risse, eds, Cornell University Press, Ithaca. Scharpf, F., 1988, The joint-decision-trap: lessons from German federalism and European integration, Public Administration 61(239-78). Scharpf, F., 1997, Employment and the welfare state: A continental dilemma, Max Planck Institute Working Paper, Cologne. Scharpf, F., 2000, Economic changes, vulnerabilities, and institutional capabilities, in: Welfare and Work in the Open Economy: From Vulnerability to Competitiveness, F. Scharpf and V. A. Schmidt, eds, OUP, Oxford, pp. 21-124. Schludi, M., 2002, The reform of Bismarckian pension systems: a comparison of pension politics in Austria, France, Germany, Italy and Sweden, PhD thesis, Humboldt Universität, Berlin. Schmidt, V. A., 2002a, Does discourse matter in the politics of welfare state adjustment?, Comparative Political Studies 35(2):168-93. Schmidt, V. A., 2002b, The Futures of European Capitalism, OUP, Oxford. van Kersbergen, K., 2000, The declining resistance of welfare states to change?, in: Survival of the European Welfare State, S. Kuhnle, ed., Routledge, London, pp. 19-36.

EXPLAINING EU IMPACTS AT THE DOMESTIC LEVEL The Europeanization of Regulatory Policy in Denmark Morten Kallestrup Abstract:

1.

This chapter questions the value of ‘goodness of fit’ as an explanatory variable in studies of EU impacts on domestic change. Instead it points to the important role of domestic politics. The analysis is based on case studies of domestic regulatory policies of consumer protection and competition policy in Denmark. It is assumed that EU pressures and policies affect domestic policy content not only directly though legal acts, but also indirectly through domestic political actors making use of the EU in policy-making processes at the domestic level. Hence the chapter also explores whether and how domestic actors have been making use of EU policies and pressures in shaping domestic policy content. While the ‘goodness of fit’-hypothesis in itself holds little explanatory value, Europeanization through changing domestic opportunity structures as well as framing beliefs and expectations are important in explaining domestic policy change. The chapter concludes that changed domestic opportunity structures and actors’ beliefs and expectations are much more important than ‘goodness of fit’ when assessing EU impacts on national regulatory policy. An emphasis on actors’ interaction in domestic politics is thus supposed to be a prerequisite for obtaining satisfactory explanatory value in studies of Europeanization.

INTRODUCTION

While the literature on Europeanization has been hastily expanding over the last decade, theoretically founded explanations of when, how, and with what consequences Europeanization occurs are still in their infancy (see e.g. Hix and Goetz 2001; Cowles et al., 2001; Heritíer et al., 2001; Knill and Lemkuhl 2002; Olsen 2002; Featherstone and Radaelli 2003). This chapter ties into the discussion on existing theoretical notions of Europeanization. It employs both the goodness of fit-hypothesis (see chapter by van Keulen and Mastenbroek this volume) as well as the conception of different mechanisms of Europeanization as developed by Knill and Lemkuhl (see chapter by

65

R. Holzhacker and M. Haverland, (eds.), European Research Reloaded: Cooperation and Integration among Europeanized States, 65-87. © 2006 Springer. Printed in the Netherlands.

66

Morten Kallestrup

Stiller this volume). The theoretical frameworks contain some explanatory value as regards direction and content of EU impacts at the national level, yet their value, in particular of the goodness of fit-hypothesis, is neither exclusive nor sufficient when applied to empirical cases. A stronger emphasis on domestic politics is a prerequisite for obtaining satisfactory explanatory value. The central research questions of this chapter are: How does the ‘Acquis Communitaire’ impact on domestic policies; and how are EU policies and pressures made use of by actors in the domestic policy-making process? Based on an analysis of cases of regulatory policy in Denmark, the theoretical frameworks are tested as regards their value in answering these research questions. Consequently EU pressures at the domestic level are also evaluated: what kind of impact did the EU have on domestic policy-makers’ beliefs and on domestic opportunity structures, and what consequences did these effects have on domestic policy content? Based on case analyses of EU consumer protection policy and competition policy, and the impact of these policies onto the corresponding national regulation, this chapter claims that changed domestic actor constellations and actors’ beliefs and expectations are much more important than the goodness of fit when explaining the differential impacts of the EU at the national level. A reservation however is that Knill and Lemkuhl’s coupling of analytical mechanisms with empirical policy archetypes appears rather odd. In the light of empirical cases, all the mechanisms seem to work either simultaneously or sequentially across the different policies. The organization of the chapter is as follows: Second to the introduction, the study’s theoretical framework is explicated as well as is the research design and the methodological propositions. In the fourth section, the empirical based case studies taking issue with the chapter’s research questions are examined and the results summarized. Finally, the case study results are critically assessed and confronted with counterfactual arguments, and the overall conclusions are drawn up.

2.

THEORIES OF EUROPEANIZATION

A prevalent explanatory feature in recent studies of Europeanization is the goodness of fit-hypothesis, which basically implies that the level of compliance between EU policies and national measures determines the extent of domestic change (Cowles et al., 2001; see introduction and chapter by van Keulen and Mastenbroek this volume). A very similar presumption is the emphasis on Europeanization through institutional compliance between European political structures and policies and domestic political structures and policies in explaining differential impacts of the EU onto the domestic

Explaining EU Impacts at the Domestic Level

67

level (Knill and Lemkuhl 2002). Though scholars have pointed to other explanatory and intermediate variables (Cowles et al., 2001; Heritíer et al., 2001), and to the limited explanatory value of the goodness of fit-hypothesis (Radaelli 2003; Knill 2001; Heritier et al., 2001), the literature anyhow sticks to emphasizing the level of structural and policy compliance – alias the goodness of fit – as a core explanatory factor of Europeanization. The importance of the explanatory value may still be overestimated though. Though the misfit-hypothesis has shown its explanatory value in some respects, e.g. in explaining how ‘positive integration’ reshapes existing domestic regulatory provisions, this study questions this prevalent hypothesis by testing it on a variety of regulatory policies. In order to seek explanatory value beyond the goodness of fit-assumption, the study furthermore employs Knill and Lemkuhl’s (2002) framework of three different mechanisms of Europeanization. The goodness of fit-hypothesis is thoroughly introduced in the introduction of this volume as well as in other chapters. Hence it will not be elaborated further here. Knill and Lemkuhl, on the other hand, distinguish between three different mechanisms of Europeanization in the field of EU regulatory policy, respectively by institutional compliance; by changing domestic opportunity structures; and by framing domestic beliefs and expectations. The goodness of fit-explanation is implicitly embedded in the first mechanism. Firstly, according to Knill and Lemkuhl, Europeanization through institutional compliance is related to community policies aimed at replacing existing domestic regulatory arrangements, e.g. positive integration policies such as environmental regulation, consumer protection etc. In these cases the member states must comply with specific institutional requirements prescribed by the EU. Secondly, Europeanization through changing domestic opportunity structures works rather by excluding options from the range of national policy choices. The mechanism is found in market-making policies aimed at abolishing domestic arrangements that distort the functioning of the Common Market, e.g. Internal Marketdirectives and EU competition regulation. Finally, Europeanization through framing domestic beliefs and expectations is prevalent in policy areas in which the level of conflict among member states is too high for common agreements (see also chapter by Stiller this volume for an analysis of ‘framing’ mechanisms). Such policies include e.g. taxation and railway policies that are designed to change the domestic political climate by stimulating and strengthening the overall support for broader European reform objectives (Knill and Lemkuhl 2002: 259). Thus relying on three different explanatory trajectories, the Knill and Lemkuhl-framework is strong and well founded as to its purpose of explaining Europeanization impacts. It must be considered one of the best accounts so far of an Europeanization-oriented explanatory framework. Though both comprehensive

Morten Kallestrup

68

and compelling, the theoretical framework based on the three different Europeanization mechanisms has some weaknesses. While Knill and Lemkuhl hypothesize how the EU impacts directly on domestic policies and actors, it does not go so far as to hypothesize how domestic actors make use of the EU in domestic policy-making processes. This seems to be a missing link in the light of the case studies below. Furthermore, the exemplification of mechanisms through references to concrete policies seems inconsistent as all mechanisms are represented in all policies. Hence the chapter questions the theoretical Europeanization literature focusing on the goodness of fit by scrutinizing the value of institutional compliance as an executive explanatory variable. Furthermore it explores the explanatory value of the differentiation between three different Europeanization mechanisms; and calls into question whether Knill and Lemkuhl’s coupling to policy archetypes is expedient. Accordingly, the study points to domestic politics: and in particular the interaction of actors in domestic decision-making processes, as an underestimated factor of explaining Europeanization.

3.

METHODOLOGY

In accordance with the research questions, the chapter scrutinizes how EU policies and pressures have been absorbed and incorporated into domestic politics and policies. In order to derive inferences on how and with what consequences the EU impacts on domestic regulatory policy-making, the chapter evaluates the EU impacts in cases of Danish consumer protection and competition regulation. The study specifically analyzes how differing degrees of EU pressures impact on the domestic decision-making in the regulatory policy fields of sales of consumer goods and competition policy in Denmark. The independent variable is the ‘acquis communitaire’, i.e. ‘de jure’ pressures (legal acts) and ‘de facto’ pressures (mismatch of system principles, regulation levels etc.) requiring national adaptation. This is measured as the degree of EU adaptation requirement. The dependent variable is domestic policy change. This is measured as the degree of EU adaptation of national policy. Intermediate variables of the framework are domestic actors’ opportunity structures as well as the actors’ beliefs. For the analytical purpose of explaining EU causal effects on domestic politics and policy, process-tracing of how the EU impacts on the domestic decision-making processes prior to the policy output is a crucial element in the analysis. The method of process-tracing holds qualities when the purpose is to obtain a detailed perception of ‘how and what happened’, yet its weakness is its resource-demanding character, which rules out large-n

Explaining EU Impacts at the Domestic Level

69

studies (King et al., 1994, 226-8). In this study then, its applicability appears fairly fruitful. With the methodological aim of maximizing variation in the independent variable, it is evident to compare cases comprising differing degrees of EU pressure for adaptation. But in order to strain off biased explanatory variables, it is also necessary to compare these cases with cases of no or almost no EU pressure for adaptation. In order to ensure variation in the independent variable the selection of observations is thus based on difference in EU adaptation requirements. The continuum of EU primary and secondary legal acts is a useful tool for ensuring difference in the degree of EU-induced ‘de jure’ pressures onto the member-states. Obviously this implies that formal EU policies, i.e. ‘hard law’ (regulations, directives etc.), as well as ‘soft law’ (opinions, the open method of coordination, etc.), and more informal pressures, form part of the independent variable. To ensure variation in the degree of ‘de facto’ pressures, each observation will have to be put in context very thoroughly, and then categorized as to its exposure to ‘de facto’ pressures for change. The selection of sub-cases in the field of regulatory policy enacts a ‘most similar case design’ to keep most other variables constant, though at the same time implicitly seeking to ensure variation in the dependent variable. EU consumer protection policies are thus characteristic by mainly drawing upon the creation of new domestic opportunities and constraints, whereas EU competition policies are much more based on changing domestic norms, ideas, and beliefs. Though the first temporary program for a common EC consumer protection and information policy was adopted in 1975, and followed by a subsequent program in 1981, only few practical provisions were initiated. With the implementation of the EC Common Act in 1987, the new Article 100A of the EC Treaty stated that a high degree of consumer protection had to be taken into account in the carrying through of the Internal Market. In the 1992 Maastricht Treaty it was stated that consumer protective standards were to be considered a means of reaching the goals of the European Union, but it was not until the implementation of the Amsterdam Treaty that the European Commission received the warrant to take steps towards member-states’ consumer protection policies. Hence consumer protection policy is a rather new re-regulatory policy field of the EU. Opposite to this, EU competition policy is a traditional de-regulatory policy field dating back to the creation of the European Economic Community. EU competition policies are designed to break down national barriers and create the foundation for a competitive interaction of businesses at trade markets across Europe, whereas national competition regulations mostly have been aimed at avoiding domestic monopolies and otherwise not disturbing companies too much. However, during the 1990s most western

Morten Kallestrup

70

European nation-states aligned their national competition legislation with EU competition rules and principles. In order to test the goodness of fit-hypothesis as well as to explore the explanatory value of the different mechanisms of Europeanization, it is also of great importance to include observations comprising difference in ‘fit’ with EU-policies, see Table 1 below. The goodness of fit is evaluated by categorizing EU policy as well as national policy as to its regulatory content; and subsequently comparing the categories of the two sets of legislation (i.e. the more correspondence, the lesser goodness of fit). The research design thus establishes the opportunity to derive implications of the theoretical hypotheses. EU requirement (‘de jure’/‘de facto’) and degree of pressure Required model (‘de jure’). HIgh pressure.

Recommended/ Suggested model (‘de facto’). Low pressure.

Observation

Transposing Approximate Expected prime acts at the goodness of mechanism of national level fit before Europeanization transposition

The Sale of Consumer Goods Directive (99/44).

The Sales of Medium. Goods Act from 2002; and the first amendment to the Marketing Practices Act in 1999. The Low. Competition Act from 1997; and the subsequent amendments in 2000 and 2002.

EU competition regulation (prohibition against anticompetitive agreements and abuse of dominant position, merger control, and sanctions).

Goodness of fitexplanation. Europeanization through institutional compliance.

Europeanization through the framing of domestic beliefs and expectations.

Table 1. Observations and theoretical hypotheses In general it has been noted by several that the Europeanization literature notably lacks critical assessments of its own claims (Hix and Goetz 2001; Haverland 2003; Radaelli 2003; Kallestrup 2003). It needs to scrutinize also:

Explaining EU Impacts at the Domestic Level

71

to what extent have the observed changes (and/or continuity) been causally related to the EU? Are there alternative explanations than EU induction and what would have happened in the absence of EU pressure? As Goetz puts is: ‘The relative importance of European integration as a source of executive change can only be assessed if it is systematically compared with rival or complementary explanations. Put differently, having ‘disentangled’ the European effect, Europeanization research has to contextualize it and explore its interactions with other sources of change’ (Goetz 2000, 227). Consequently, the statement that European integration has resulted in a particular outcome implies that if European integration would not have happened, the particular outcome would not have occurred (Haverland 2003, 217). In line herewith, the case study results of this chapter will eventually be confronted with such critical assessments. The data collection is based on the examination of official documents and publications as well as archived files related to the particular law reforms. The archived files have been collected in ministerial archives. The document studies have been supplemented by interviews with the policy-makers involved in the law reforms, e.g. civil servants, representatives of organized interests, committee members, experts in law and economics, politicians, etc. (A more elaborate explication of the data on the particular policy processes can be found in Kallestrup 2004).

4.

CASE STUDIES: SALES OF GOODS POLICY AND COMPETITION POLICY IN DENMARK

As mentioned, the particular case studies of this analysis consist of an investigation of the EU impacts on sales of goods regulation and competition policy in Denmark. National regulatory policies have been exposed to a continuous and increasing influence from EU policies and pressures over the years. Hence certain EU impacts should be expected in domestic consumer protection regulation as well as in domestic competition regulation. Below, each case study is organized into a subsection, comprised of an introduction to the particular policy and an analysis of the Europeanization of the corresponding national regulation. The EU policy and the national status quo have been compared along principle dimensions, as strict comparisons would be far too comprehensive and would not contain further explanatory value. For each case study the issues at stake, the central actors, as well as the actors’ preferences and purposes of action are sought explicated as they are at the midst of the analysis (yet for more elaborate and detailed case studies, see Kallestrup 2004). Following the individual case studies, the overall case study results are summarized.

Morten Kallestrup

72

4.1

Sales of goods policy

Sales of goods regulation operate at the heart of interaction between consumers, traders, and businesses. At the national level these groups are organized around a few large interest organizations: in particular the Confederation of Danish Industries, Danish Commerce and Services, and the Consumer Council, which on their behalf speak their voice. In the lawmaking process the involved authorities, the parliamentary parties, and the ministers are obviously also central actors. At the 2002 reform of the Sales of Goods Act, the issues at stake were so controversial that tactics was a prevalent element in the domestic law-reform process. The EU regulation played a central role in these tactical maneuvers. Following the adoption of the first temporary program for a common EC consumer protection and information policy in 1975, the Council requested the Commissions initiatives on the improvement of guarantees for consumer goods and after-sales services a number of times. Yet the first concrete step towards common EU regulation of the sale of consumer goods was the 1993 green paper on guarantees for consumer goods and after-sales services (COM (93) 509 final). Based on the green paper responses, the Commission worked out a draft directive (COM (95) 520 final), which was put forward in 1996. Contrary to the EU regulation, the Danish sales of goods legislation had been in effect since 1906. Nonetheless a few amendments were enacted in the meantime; it was not until the mid-1990s that issues of the sale of consumer goods were once again raised on the political agenda as left-wing parties made the enhancement of consumer protection a priority. Yet a reform of the Sales of Goods Act did not come through before 2002, partly as a result of the EU sale of consumer goods directive. 4.1.1

The Sales of Consumer Goods Directive and the Danish Sales of Goods Regulation

In 1999, the Council adopted Directive 99/44 of 25 May 1999 on certain aspects of the sale of consumer goods and associated guarantees (the ‘Sale of Consumer Goods Directive’). Due to dispute the legal basis of the Directive was Article 100a (now 95), which referred to the completion of the internal market, rather than Article 129a (1) (now 153), which emphasized the importance of consumer protection. The dispute was settled by including a reference to Article 129a in the first Recital to the Directive. The Directive covered rules on the lack of conformity with the contract as well as rules on manufactures’ and retailers’ guarantees. The overall purpose of the Directive was to initiate rapprochements of member-states’ national laws as regards sales of consumer goods and associated guarantees.

Explaining EU Impacts at the Domestic Level

73

Compared to the 1993 Green Paper on guarantees for consumer goods and after-sales services, the final Sales of Consumer Goods Directive was only modest in its scope and substance, in particular in comparison with the scope of aspects mentioned in the original green paper. Some of the more significant omissions were the exclusion of provisions on after-sales services, network liability, joint liability of seller and manufacturer for nonconformity, horizontal privity, default guarantee, and cross-border enforcement (Twigg-Flesner 1999, 2000). The Commissions original intensions were thus severely enervated during the EU-level negotiations. Both Danish authorities and interest organizations had opposed the Commission’s proposals for directives, but for different reasons. The authorities opposed the draft directives due to implicit legal and technical complications and problems in principle. Business interest organizations referred to the lack of necessity for common EU regulation in the field of sales of goods and associated guarantees, while the consumer protection organizations referred to the draft directives being too weak in respect of consumers’ rights. Despite both Germany and Denmark opposed the adoption of the directive, and the Netherlands did not vote, the Directive was anyhow adopted in 1999. National laws were to comply with the Directive on 1 January 2002. What then appeared as immediate consensus on the opposing of the adoption of the Directive among Danish decision-makers was actually build on extensive and marked conflict among strong domestic interests. Domestic policy-makers were thus forced to behave extremely tactically in order to get on and hence develop their opportunities for reaching particular aims. EU regulation showed to be an important means for that purpose, as it was an effective and convincing argument during domestic negotiations. While the Directive’s provisions on guarantees were transposed into Danish legislation in the 1999 Marketing Practices Act, the rest of the provisions were implemented in the Sales of Goods Act. The minimum requirements to form and substance of consumer sale guarantees of the Sale of Consumer Goods Directive were thus enacted nationally in Denmark in 1999. At the same law reform the Consumer Ombudsman was supplied with increased powers as regards the authority to issue a command. Though the policy-making process leading to the 1999 Marketing Practices Act was characterized by conflict between the cabinet coalition parties and the opposition parties, the issue of conflict was not the implementation of the requirements to consumer sale guarantees, but the powers of the Consumer Ombudsman. Hence the implementation of the Directive provisions was more a matter of form than an issue of conflict. Shortly after the Council’s adoption of the Sale of Consumer Goods Directive in May 1999, the minister of justice set up a working-group, which was to prepare a proposal on how to transpose the rest of the Directive into

74

Morten Kallestrup

national law. The working-group, consisting of high-level jurists, worked from 1999 to 2001, and held meetings with interest organizations regularly. The committee’s final report, published in June 2001, recommended certain changes to the existing sales of goods act: among others that the buyer’s deadline of notification of defects was to be lengthened, so the buyer was obliged to notify defects within two years instead of one year in accordance with the EU Directives provisions. This rule was also to apply to business sales and private sales in order to avoid a notification trap of different time limits for buyers and sellers. Furthermore, the concept of defect was to be specified (Betænkning 1403/2001). All in all the policy effects would be limited, but the issues at stake were considerable for affected interests. As had also been the case during the EU-level negotiations on the Directive, Danish business organizations were displeased with the proposals. Oppositely, the Consumer Council endorsed the draft Bill, though it claimed a distinct need for further improvement and a strengthening of consumers’ rights. During the parliamentary reading, all political parties backed the draft Bill. On 22 April 2002 the Bill was adopted. As national law had had to comply with the Sale of Consumer Goods Directive since 1 January 2002, the 2002 Sales of Goods Act was to be effective retrospectively, which was quite extraordinary. The policy effects of the Bill were manageable though: the seller would be liable for up to two years for any lack of conformity; the consumer was required to inform the seller of the lack of conformity within two months of discovering this; and the concept of defect was specified. Hence the Directive did secure national minimum standards, e.g. as regards the buyer’s deadline of notification of defects, but policy effects were otherwise minimal. 4.1.2

Explaining EU impacts: National actors’ making use of the EU to delay and delimit the domestic process

As it appears, and in spite of the policy effects being manageable, the road towards the new act had been steeped with strongly opposing views of the involved interests, in particular during the preparation of the Sale of Consumer Goods Directive in the 1990s. However, during the national political negotiations leading to the new act the political actors, and in particular the responsible ministers, actively used references to EU legislation to interfere with the national policy process. The ministers used references to EU legislation to control momentum and hence prolong the policy-making process, as well as to delimit the scope of negotiation. For example in 1995 and 1996 the minister of justice used forthcoming EU regulation initiatives in the field of sales of consumer goods as an argument for postponing a revision of the Danish Sales of Goods Act. Consequently,

Explaining EU Impacts at the Domestic Level

75

a revision of the Act did not take place until the enactment of the new act in 2002. Yet the 2002 Act turned out to be not even a result of a general revision, but only an implementation of the Sale of Consumer Goods Directive. Another but similar prolonging incident took place during the reading of the draft Bill in December 2001. The opposition parties demanded a general reform of the act, but the minister of justice referred to forthcoming EU initiatives in the field of EU contract law as an argument for postponing a general revision of the act. Opposite to the former minister’s statements in 1995 and 1996, the purpose of the 2002 reform was changed so as it was only supposed to implement the EU Sale of Consumer Goods Directive. Hence authoritative decision-makers’ references to ‘forthcoming’ EU legislation justified a continual delay of a general revision of the act at several occasions. Also during the political negotiations on the Sales of Goods Act, responsible political spokesmen and ministers referred to the Directive as a means of delimiting the scope of the subject of negotiation. The workinggroup’s commissary from 1999 stated that the working-group members only had to concentrate on provisions initiated by the Directive. Hence a general revision of the Sales of Goods Act was not part of the job. Also during the reading of draft Bill L 26 in October 2001, a number of spokesmen expressed that they did consider the Bill a direct effect of the Directive. Whether or not these statements were meant to substantially delimit the issues open for debate, and / or to dispatch the case more rapidly, is of lesser importance. What is important is that the statements had a deterministic tinge that consequently had a delimiting impact on the general scope of negotiation. The minister of justice also referred to the Directive in February 2002 as a means of delimiting the scope of negotiation during the reading of the Bill. Yet as the Directive was a minimum directive, it did not interfere with the subject of discussion, namely: a higher degree of consumer protection. Though there was a certain misfit between EU provisions and national regulation prior to the reform, the explanatory value of the goodness of fitassumption is marginal. When looking into the domestic reform-making process, many other factors played a much more important role. The interaction among domestic actors, and hence their relative opportunity structures in the light of the EU provisions, proved to be much more important to the speed, direction, and extent of domestic change. At several times national policy-makers thus submitted former and future EU legislation as an argument for delaying the policy-making process as well as for delimiting the scope and substance of negotiation. Hence a number of responsible decision-makers at the national level used references to EU legislation to buy leeway in the national policy-making process, despite this use may have been unjust.

Morten Kallestrup

76

As was the case in the 1999 marketing practices law reform (as well as in subsequent reforms of this act), the allocation of competencies to national authorities was of utmost importance to domestic decision-makers, and even more important than the substantial policy changes. This has also been a prevalent feature of Danish competition policy reforms over the last decade, as will be shown below.

4.2

Competition regulation

While sales of goods regulation regulate the interaction between consumers, traders, and businesses in order to secure a fair amount of consumer protection, competition policies regulate the legal conditions for businesses’ operation and cooperation in order to avoid cartels and dominant positions at the cost of consumers. The central actors in this policy field consist of the private business organizations and the public authorities at various levels. In the subsections below, the general elements of EU competition policy as well as the late developments of Danish competition regulation are drawn up. Following this, a line of argument explaining the extensive EU-harmonization of Danish competition policy is elaborated. At the EU-level, competition regulation has been based on the principle of prohibition and backed by sanctions since 1962. The main elements have been the Articles 81 and 82 on prohibition against anticompetitive agreements and prohibition against abuse of dominant position, as well as the merger regulation and the well-developed legal practice based on rulings by the European Court of Justice and The Court of First Instance. The principle of prohibition inhibits that anti-competitive agreements are illegal from the moment they are carried through. Opposite to this, Danish competition legislation has been based on the principle of control and the principle of publicity from 1937 to 1997. The principle of control means that there is no limit on anti-competitive agreements as long as the competition authorities have not acted upon the particular case, while the principle of publicity inhibits that all information on anti-competitive agreements and monopolies has to be made public through the duty of notification as well as public registration. Hence the misfit between European policy principles and national regulation was remarkable before 1997. It was not until the introduction of the 1997 Competition Act and the subsequent amendments in 2000 and 2002 that Danish competition legislation was changed and aligned with EU competition principles. This considerable change took place even though no EU legal acts ever prescribed the harmonization of national competition laws. There were never legal or official political pressures stemming from the EU onto the member-states as regards the composition of national competition legislation. Below, this rather non-EU-induced

Explaining EU Impacts at the Domestic Level

77

harmonization of Danish competition policy from 1997 to 2002 will be the object of analysis. 4.2.1

The reforming and EU harmonization of Danish competition policy, 1997-2002

As mentioned, business organizations and public authorities have played key roles in this field of policy. Yet in a Danish context the interaction within and among the political parties represented in the parliament has also lead to important consequences for the transformational reforming of Danish competition policy from 1997 to 2002. With the purpose of preparing an introduction of prohibition-based competition legislation, in accordance with EC competition legislation principles, the minister of industry set up the so-called Competition Act Committee in 1993. The committee, consisting f civil servants, representatives of interest organizations, and experts in EC law and economics, was also to take into account how to create an efficient basis for interaction between EC competition legislation and Danish competition legislation. However, during the committee’s work significant disagreements among the participants came apparent. A compromise was obtained though, based on a competition act draft Bill, but only the national authorities and the big business interest organizations backed it. The experts particularly disagreed on the compromise as it was not in conformity with EU competition legislation and thus not in accordance with the purpose of the committee as stated in the commissary (Betænkning 1297/1995). Following this, from 1995 to 1996, the cabinet coalition parties held negotiations, but failed to reach an agreement. Especially the Social Democrats demanded a larger degree of prohibition and EU-alignment of the national competition legislation. In December 1996, a new social-democratic minister of business and industry was appointed. The new minister and the spokesmen of the cabinet coalition parties quickly reached agreement on a new draft Bill, which was partly based on the responses from the former hearing round. Compared to the Competition Act Committees’ draft Bill quite significant changes had been incorporated into the new draft Bill. The general line in the Cabinet’s new proposal was further EU alignment of the national competition legislation. With a few minor adjustments, a large parliamentary majority passed the new Competition Act on 30 May 1997. Among others, the principle of control was totally abandoned and the prohibition against anti-competitive agreements was to cover both horizontal and vertical agreements. The new Competition Act was to be effective from 1 January 1998. However, two years later the Ministry of Business and Industry once again prepared changes of the Competition Act. On 14 October 1999, the minister of business and industry announced a number of amendments to the

78

Morten Kallestrup

existing regulation of competition and consumer protection, including proposals for an amendment to the Competition Act. The most controversial elements included the introduction of merger control, the introduction of fines in case of first-time abuse of dominant position, and a change of the Competition Council’s composition. A number of business interest organizations once again opposed the need for a change of the Competition Act. Anyhow, on 30 March 2000 the minister put forward draft Bill L 242, and in May 2000 the cabinet coalition parties and the right wing opposition parties carried through the act. The 2000 amendment to the Competition Act introduced among others merger control and the option of fining first-time abusers of dominant position. Furthermore the Competition Council obtained the right to enforce Articles 81 and 82 of the EU Treaty, and to require illegal anti-competitive public support to be stopped and if necessary returned. The law was to be effective on 1 October 2000, except the option of fining first-time abusers of dominant position, which was to be effective on 1 July 2002. During the negotiations the minister used references to the need for adaptation to EU legislation to legitimize, with a certain diversionary effect, the carrying through of other domestic and political controversial elements in the act (Boje and Kallestrup 2004). Particularly the role of The Competition Council drew much attention and was reason for much controversy. One year later, in July 2001, the European Commission exposed a cartellike cooperation structure between SAS and Maersk Air. The participants were sanctioned with the biggest fines ever given to Danish companies for anticompetitive arrangements. Executive officers from both companies were fired and the exposure of the case resulted in considerable public attention (Boje and Kallestrup 2004, 266). Combined with the general debate on competition regulation, the minister of business and industry was forced to act on the subject. During September and October 2001, the Ministry worked out a draft Bill on an amendment to the competition act. The minister did not get to put forward the draft Bill as on 31 October the prime minister publicized that general elections were to take place on 20 November 2001. The general elections resulted in a change of cabinet. The Liberals and the Conservatives formed a new coalition cabinet with the liberal Anders Fogh Rasmussen as new prime minister. The leader of the Conservative Party, Bendt Bendtsen, was appointed minister of economic and business affairs, and was hence responsible for the competition regulation. In December 2001, the Ministry of Economic and Business Affairs resumed the preparations for an amendment to the Competition Act. The newly appointed minister declared a critical attitude to the former government’s proposal, but noticed that a rising of the level of fines probably would be necessary. Except of a few changes, the liberal-conservative cabinet’s new draft Bill was similar to the former cabinet’s proposal. On 29 May 2002, the new

Explaining EU Impacts at the Domestic Level

79

Competition Act was adopted in parliament, backed by the cabinet coalition parties, the Social Democratic Party, the Danish Peoples Party, The Social Liberal Party and the Christian Peoples Party. Hence the 2002 amendment to the Competition Act included an increase in the level of fines on anticompetitive agreements. The possibility of lowering the penalty to companies that co-operate during the investigation was introduced, as was a reduction of the lower level for merger control. In less than five years from the adoption of the new Competition Act on 30 May 1997, fundamental transformations through a considerable but voluntary EU harmonization of Danish competition policy had taken place. The most important change was the abandonment of the principle of control and the introduction of the prohibition against anti-competitive agreements and the prohibition against abuse of dominant position. Merger control was introduced in 2000 as well as was the Competition Council’s empowerment to enforce the direct application of the Articles 81 and 82 of the EU Treaty in Denmark. In 2002 the sanctions were tightened and the level of fines increased. Furthermore, the Competition Authority obtained the possibility to carry out more effective control investigations. Some of the changes took place over a stretch of time, e.g. the leveling out of competition regulation of private and public businesses. Yet not only the means, but also the overall aims of competition legislation were changed. Today the purpose of the competition legislation is not only to avoid monopolies on national markets, but also to create the foundation for a sound competition based on European and international competition principles. Hence it is appropriate to characterize the competition regulation changes in Denmark from 1997 to 2002 as a case of paradigmatic change (Hall 1993). Below, the use of EU as a prevalent means for justification throughout all three reforms will be discussed. 4.2.2

Explaining EU impacts: National actors’ making use of the EU as a means of justifying domestic policy change

In the competition policy debate during the last decade, political discourse has been dominated by arguments for change justified by reference to a need for adapting domestic competition legislation to EU competition legislation and principles. A through-going characteristic across all three recent reforms of the Competition Act was domestic actors’ references to a need for adaptation to the EU as an argument for domestic change, though there was no EU ‘de jure’ requirement for adaptation. Paradoxically, the proposed needs were not put to critical scrutiny by an appreciable number of actors during the policy-making processes. Due to the lack of legal requirements, the references to a need for adaptation were based on more practical ‘de facto’ arguments, but this reasoning was also questionable. In

80

Morten Kallestrup

example, during the 1990s a widespread argument was that companies had to apply to two sets of competition legislation, i.e. both EC competition legislation and national competition legislation. Yet this would have been a problem for more than twenty years, just since Denmark entered the EC in 1973. When this was evaluated in the 1980s, it was concluded that this problem could not legitimize a change of the fundamental principles of the legislation. Hence the argument was weak. According to civil servants, the reason for changing the Competition Act in 2000 was also a need for adapting Danish competition legislation to other EU member-states, especially as regards merger control, sanctions, and abuse of dominant position. Furthermore, there was a political wish to support the Commission’s decentralization efforts as regards the appliance of Articles 81 and 82 of the EC Treaty (Levinsen 2001, 23). Since first half of the 1990s, the EU Commission had worked on decentralizing certain authority to national competition authorities, in order to optimize resource allocation and to be able to direct more attention towards large transnational companies. Yet in 2000 there was no major improvement. Hence this argument was also questionable. It seems more plausible that it was budgetmaximizing bureaucrats rather than policy-makers aiming for concrete political results that enforced this argument. Also in the political debate leading to the 2002 amendments, civil servants were continuously referring to the EU as an argument for carrying through changes in the competition legislation, in particular in the rather low level of fines. References to future EU legislation in relation to the decentralization of the application of Articles 81 and 82 were used as arguments for strengthening the Competition Authority’s legal powers. References to the need for bringing Danish competition regulation to the same level as EU competition legislation were used as arguments for strengthening the sanctions and the level of fines. Also in the ministerial note on present Danish competition policy in November 2001, the EU was referred to as an argument for a need for changing the Competition Act. Hence references to a need for EU adaptation as justification for domestic competition policy change were widespread during the processes leading to the three recent reforms of the Competition Act. As mentioned, the change of principles carried through in the 1997 Competition Act was a considerable change. But the important change was not only the change of principles, but also the deliberate change towards an ever-lasting EU conformity of national competition policy. This was of immense importance as it meant that Danish competition legislation consequently must be interpreted and administered in accordance with EU directives and practice: Danish authorities are not supposed to build up and use legal standards that differ from the EU principles. However, the significant direct and indirect harmonization with EU policy did not happen due to EU requirements. No

Explaining EU Impacts at the Domestic Level

81

treaty, directive, or court decision ever required an adaptation of the national competition legislation in order to bring it in accordance with EU principles. Rather it took place as a result of a situation in which EU legislation was a better means of obtaining certain domestic political interests for consumerfriendly political parties as well as for the involved authorities. Especially immensely encumbered ministers furthermore used references to a need for adaptation to the EU tactically to legitimize the carrying through of other domestically controversial issues. Once again the goodness of fit-hypothesis proved to hold only limited explanatory value. On the other hand the EU-induced change of domestic actors’ opportunity structures as well as the impact over time on domestic actors’ beliefs and expectations proved much more valuable in explaining the considerable change and EU-harmonization of domestic competition policy.

4.3

Sum-up of the case study results: Europeanization through changed domestic opportunity structures and framed beliefs and expectations

In each of the above cases, EU policies initiated a process leading to domestic policy change, either directly or indirectly. In all cases domestic actors furthermore made use of EU regulation as a means of reaching certain aims in the domestic policy-making process. The ways and consequences hereof differed from case to case. While the goodness of fit and the pressures for institutional compliance did not contain considerable value with regards to explaining the speed and extent of EU impacts at the domestic level, actors’ making use of references to the EU, whether just or unjust, was a very important factor in either prolonging, shaping, and/or justifying the domestic law-reform and policy change. Hence in all cases the EU impacts on domestic actors’ opportunity structures proved to hold much more explanatory power than did the goodness of fit-hypothesis. In the latter case, the framing of beliefs and expectations also contained explanatory value to a certain extent.

Morten Kallestrup

82 ‘Acquis Communitaire’ (independent variable observation) The Sales of Consumer Goods Directive (99/44).

EU Competition Regulation.

Domestic policy (dependent variable observation)

Theoretically expected prime explanation

Prime explanation derived from case studies

The Sales of Goods Act (2002) and the first amendment to the Marketing Practices Act (1999).

Goodness of fit - Changed domestic opportunity Europeanization structures; national through actors’ made use of institutional the EU with a compliance. temporally prolonging and substantially delimiting effect in the domestic policymaking process. Goodness of fit contained only little explanatory value. Framed domestic Europeanization The Competition Act (1997) and the through the beliefs and expectations and framing of subsequent changed domestic domestic beliefs amendments and expectations. opportunity (2000, 2002). structures. National actors’ made use of the EU as justification for domestic policy change and adaptation.

Table 2. Case study results and derived explanations As regards sale of consumer goods regulation, the EU Sales of Consumer Goods Directive initiated a domestic process of policy change based on requirements for institutional compliance. The determinant however, of the speed and extent of Europeanization impacts, was that domestic actors also used it to prolong the preparation process as well as to delimitate the scope of political negotiation. In the end, and for the benefit of consumers, minimum requirements for the sale of consumer goods were secured under

Explaining EU Impacts at the Domestic Level

83

all circumstances. While EU directives initiated change in national consumer protection in the case of sales of goods regulation, the factor causing change in Danish competition policy was of a much more indefinable kind. EU legislative pressures did not contain much explanatory value in relation to domestic policy change in this case. Oppositely, the beliefs of political decision-makers were of great importance, as a large majority of the parliament opposed business preferences and aligned Danish competition policy with hard-lined EU policy, though such change was neither legally required, nor politically demanded by the EU. Another through-going characteristic was the decision-makers’ continuous references to the EU and to EU competition policy as an argument for carrying through changes of national competition policy. This illustrated the deliberate making-use of changed opportunity structures as a consequence of EU regulation. As explicated, the reference to the EU as an argument for domestic change and/or adaptation, whether valid or not, was used in initiating the new Competition Act in 1997 as well as the subsequent amendments in 2000 and 2002. The result was extensive policy change and harmonization of Danish competition policy in line with EU competition policy principles and legal practice. Hence, while EU directives did only cause limited policy change at the national level in the case of consumer protection, national actors explicitly made use of the EU regulation as an initiator and pawn during the ongoing political processes. In the case of competition policy certain actors used the EU similarly as a role model to enforce extensive policy change of national competition legislation, though no EU legal acts or policy statements required such a change. See Table 2 for an overview of the analytical results as regards the theoretical explanation.

5.

CRITICAL ASSESSMENT AND CONCLUSIONS

A critical evaluation of the analytical results is also a priority of the study’s approach and hence conclusions. Before outlining the overall conclusions, the case study results will be critically assessed in order to establish a proper foundation for deriving inferences. While the direct and causal effect of European integration was obvious in the case of sales of consumer goods regulation, the causal relation was much more indistinct in the case of competition policy. The changes in the Sales of Goods Act may very well have appeared in spite of the Sale of Consumer Goods Directive. As mentioned, domestic initiatives were already underway in 1995 and 1996, but were delayed with reference to forthcoming EU legislation. As the CISG already operated with the measures introduced in the 2002 Sales of Goods Act, an obvious role model for national policy was

84

Morten Kallestrup

already in place. Hence there are strong indications that a similar policy change would have happened anyway, sooner or later. As mentioned, the EU played a prolonging and delimiting role in the case of sales of goods regulation. As regards the changes of the marketing practices regulation in 1999, which was also initiated by the Sales of Consumer Goods Directive, it is quite clear that the particular changes were causally related to the EU. As both consumers and businesses opposed the Sale of Consumer Goods Directive, but for different reasons, the changes in 1999 would not have come through at the national level if it was not for the EU legal act. Yet the policy effects were still manageable. In the case of competition policy, the causal relation between the EU and the changes of national policy was much vaguer. Here alternative explanations to EU-induction are prevalent. Firstly, one may point to ‘de facto’ pressures on the old system of competition control, mainly the lack of convergence between hard-lined EU competition policy and liberal national competition policy. Secondly, the high price level on consumer goods in Denmark was in itself a ‘de facto’ pressure on the system. Thirdly, the revelation of considerable cartels operating on Danish soil affected public opinion on competition regulation (Boje and Kallestrup 2004). Furthermore, since the 1980s the OECD has continuously recommended that national competition legislation be aligned with EU competition policy and principles, in order to counter the relative high price level (OECD 1993, 1997; Wise 2000). Hence the causal effect of the EU onto the changes of national competition regulation is questionable. What anyhow was striking was the prevalent use of references to a need for adaptation to the EU, which prevailed during the political preparation of all three recent changes of the Competition Act. Though there had been no EU legal acts prescribing adaptation of national competition policy, the changes of national policy were politically justified by reference to ‘a need for adaptation’ (Kallestrup 2002). The case of competition policy is thus truly a case in which EU legal pressures and official political pressures have been absent. In despite hereof, the EU was used as a source of political justification for domestic policy change. Concluding theoretically: with the purpose of exploring the value of goodness of fit in explaining domestic policy change in European memberstates, the case studies above strongly indicated that other mediating factors play a much more important role. Changed opportunity structures and hence domestic actors’ use of the EU in domestic policy-making processes is a crucial factor in understanding the speed and extent of domestic policy change. Pressures for compliance cannot explain the speed, extent, and direction of particular policy changes at the national level in a satisfactory manner. The case studies clearly explicate that there is no coherence between the degree of EU-induced pressure put on the member-states and

Explaining EU Impacts at the Domestic Level

85

the extent of domestic policy change. Conversely, domestic actors’ changed opportunity structures, e.g. explicated through the use of references to EU policy, as well as the framing of their beliefs and expectations, seem to contain considerable value in explaining the extent of the particular domestic policy changes. Consequently, it is obvious that the theoretical goodness of fit-assumption must be coupled with the action and interaction of decisionmakers at the domestic level in order to hold explanatory value. Independent from the pressures for adaptation put on the member states, policy change came through domestically, both in the cases of poor goodness of fit (competition policy) as well as in the cases of rather high goodness of fit (consumer protection). Yet another implication from the empirical-based case analyses is that Knill and Lemkuhl’s coupling of the three different mechanisms with empirical policy archetypes is inconsistent. Across the empirical cases, the analytical mechanisms worked either simultaneously or sequentially, and were hence not policy-determined, as implicitly proposed by Knill and Lemkuhl. The explanations deduced from this study’s top-down focus on EU impacts on national policy-making processes seem to hold theoretical value also beyond the immediate implications of the theoretical frameworks involved. In the light of the epistemological discussion of the necessity of a analytical link between bottom-up and top-down processes of Europeanization (see introduction and chapter by van Keulen and Mastenbroek this volume), this study’s results underline that the domestic transposition and implementation processes cannot be bound to pure take-over of the EU outputs. The domestic political transposition and implementation process contains too many explanatory variables to be ignored as a very potential level of analysis in studies of cases in which the EU interferes with national policy. In short: domestic politics matters too much to be ignored! If the purpose of a study is to explain Europeanization impacts at the domestic level, domestic politics cannot be reduced to the measurable extent of ‘correct transposition and implementation of EU directives’. Finally, as explicated in the critical assessment of the case study results, there is an extensive need for defining the role of the EU as one independent variable among others in order to explain domestic political change. The causal relation between the EU and domestic change is in many cases vague. This does not mean that a causal relation does not exist, but the assumed causal relation has to undergo critical scrutiny in order for the researcher to distinguish what is causally related to the EU and what is not. Accordingly there is a pronounced need for applying a counterfactual approach in studies of Europeanization.

Morten Kallestrup

86

REFERENCES Betænkning 1297/1995, 1995, Konkurrencelovgivningen i Danmark – Bind I, Statens Information, Copenhagen. Betænkning 1403/2001, 2001, Gennemførelse af forbrugerkøbsdirektivet, Ministry of Justice, Copenhagen. Boje, P., and Kallestrup, M., 2004, Erhvervsliv, marked og stat. Dansk konkurrencelovgivning og det store erhvervsliv, Århus, Aarhus Universitetsforlag. Commission, 1993, Green Paper on Guarantees for Consumer Goods and after-sales services, COM (93) 509 final, Bruxelles. Commission, 1995, Proposal for a European Parliament and Council Directive on the sale of consumer goods and associated guarantees, COM (95) 520 final, Bruxelles. Cowles, M. G., Caporaso, J., and Risse T., eds, 2001, Transforming Europe: Europeanization and Domestic Change, Cornell University Press, Ithaca and London. Featherstone, K., and Radaelli, C. M., eds, 2003, The Politics of Europeanization, Oxford University Press, Oxford. Goetz, K. H., 2000, European integration and national executives: a cause in search of an effect, West European Politics 23(4):211-231. Hall, P. A., 1993, Policy paradigms, social learning, and the state, Comparative Politics 25(2): 275-296. Haverland, M., 2003, The Impact of the European Union on environmental policies, in The Politics of Europeanization, K. Featherstone and C. M. Radaelli, eds, Oxford University Press, Oxford, pp. 203-224. Heritiér, A., Kerwer, D., Knill, C., Lemkuhl, D., Teusch, M., and Douillet, A. C., 2001, Differential Europe: The European Union Impact on National Policy-Making, Rowman and Littlefield Publichers Inc., Oxford. Hix, S., and Goetz, K. H., eds, 2001, Europeanized politics. European integration and national political systems, Frank Cass, London. Kallestrup, M., 2002, Europeanisation as a discourse: domestic policy legitimisation through the articulation of a ‘need for adaptation’, Public Policy and Administration 17(2):110124. Kallestrup, M., 2003, Anmeldelse af Kevin Featherstone og Claudio M. Radaelli (red.), The Politics of Europeanization, Oxford: Oxford University Press, 2003, Politica 35(4):466468. Kallestrup, M., 2004, Europæisering af nationalstaten. EU’s konsekvenser for national reguleringspolitik og de nationale aktørers undervurderede rolle, PhD Dissertation, Department of Economics, Politics and Public Administration, University of Aalborg. King, G., Keohane, R., and Verba, S., 1994, Designing Social Inquiry. Scientif ic Inference in Qualitative Research, Princeton University Press, Princeton, NJ. Knill, C., 2001, The Europeanisation of National Administrations. Patterns of Institutional Change and Persistence, Cambridge University Press, Cambridge. Knill, C., and Lemkuhl, D., 1999, How Europe matters: different mechanisms of europeanization, European Integration online Papers. 3(7), http://eiop.or.at/eiop/texte/ 1999- 007.htm. Knill, C., and Lemkuhl, D., 2002, The national impact of European Union regulatory policy: three europeanization mechanisms, Journal of European Political Research 41(2):255-280. Levinsen, K., 2001, Konkurrenceloven med kommentarer, Jurist- og Økonomforbundets Forlag, Copenhagen. OECD, 1993, OECD Economic Surveys 1992-1993 – Denmark, OECD, Paris. OECD, 1997, OECD Economic Surveys 1997-1998 – Denmark, OECD, Paris. Olsen, J. P., 2002, The many faces of europeanization, Journal of Common Market Studies 40(5): 921-952.

Explaining EU Impacts at the Domestic Level

87

Radaelli, C. M., 2003, The europeanization of public policy, in The Politics of Europeanization, K. Featherstone and C. M. Radaelli, eds, Oxford University Press, Oxford, pp. 27-56. Twigg-Flesner, C., 1999, The E.C. directive on certain aspects of the sale of consumer goods and associated guarantees, Consumer Law Journal 7:177-192. Twigg-Flesner, C., 2000, New rules for guarantees: for better or worse?, Consumer Policy Review 10(3):100-106. Wise, M., 2000, The role of competition policy in regulatory reform, OECD Journal of Competition law and Policy 2(1):23-77.

AGGREGATING, MOBILIZING AND RECRUITING EU Integration and Party Functions Harmen A. Binnema Abstract:

1.

Europeanization has become a major research focus within EU studies. However, the possible impact of EU integration on political parties has received relatively little attention. This chapter analyses this particular relationship, by considering changes in three main party functions: aggregation, mobilization, and recruitment. It is argued that EU integration adds to existing challenges faced by parties, such as individualization, advances in communication, and the rise of interest groups. The empirical results show some differences in the way in which parties perform their central functions. Concerning aggregation, we find that the salience of the EU increases marginally, whereas the main issues on the party agenda show a remarkable continuity over time. Mobilization has become more difficult for parties, as can be seen in a clear trend of declining membership. Recruitment is still firmly controlled by (national) parties: nearly all members of parliament and government are party members. In addition, there is little interaction between the EU level and the national level when it comes to the career patterns of party politicians. The conclusion is that the impact of EU integration is limited and that one should be cautious in attributing instances of party function change to the EU.

INTRODUCTION

Europeanization has rapidly emerged as a major topic within EU studies. Within this fresh new field, most of the literature has focused on the effects of EU integration on national policy making. Only recently, the impact on political parties has come to the fore as a subject of scientific inquiry (cf. Ladrech 2002). This chapter will also link up with the latter issue, by examining the possible relationship between EU integration and change in party functions, with a particular attention on recruitment. Section 2 introduces the two key elements of this chapter, Europeanization and political parties. After a brief sketch of the emergence

89

R. Holzhacker and M. Haverland, (eds.), European Research Reloaded: Cooperation and Integration among Europeanized States, 89-117. © 2006 Springer. Printed in the Netherlands.

Harmen A. Binnema

90

of this concept in the academic literature, I will argue for a restricted notion of Europeanization. Next to that, the role of parties within a democratic political system will be described, and the functions which parties traditionally are expected to perform. In section 3, I will consider the connection between Europeanization and parties by presenting an overview of the literature on this topic. In particular, it will show that although considerable research has been done in this area, most of it falls outside of the scope of Europeanization. Moreover, the mechanisms of Europeanization involved are different than in the case of policy change, and thus require a specific approach parties and EU integration. In section 4, the link between EU integration and changes in party functions is outlined and then explored empirically. The guiding question is: To what extent does EU integration impact on the functions traditionally ascribed to parties? The three main functions I distinguish are: aggregation of interests, mobilizing the citizenry, and recruitment for office. The last function will be elaborated in greater detail, by focusing on the career patterns of members of national parliaments and European Parliament in Austria, Denmark and the Netherlands. Finally, section 5 contains the conclusion and discussion.

2.

EUROPEANIZATION AND PARTIES

In this section, the two main topics of this chapter, Europeanization and political parties will be discussed. Concerning Europeanization, I will distinguish the main approaches that can be found in the literature and put forward a rather restricted definition, which seems to cover the necessary aspects of Europeanization, whilst avoiding the risk of concept stretching. Second, I will take up the issue of the role of parties within democratic political systems and the functions that are ascribed to parties.

2.1

A brief history of Europeanization

Haverland (2003) argues that Europeanization research can be divided into three phases. The first period, which covers the 1970s, 1980s and early 1990s, saw a number of rather idiosyncratic implementation studies, mostly atheoretical. Moreover, the emphasis was still on the EC as a dependent variable. In the second phase, from the mid-1990s on, Europeanization studies became more comparative and theoretically informed, and the term ‘Europeanization’ became much more widely used (cf. Featherstone 2003). Diverging outcomes in different countries and in different policy areas led to several hypotheses about the effect of Europe. Most prominently among these was the ‘goodness of fit’ hypothesis: the misfit between EU regulation

Aggregating, Mobilizing and Recruiting

91

and national policy would bring about adaptational pressure. Moreover, the importance of veto points or veto players was stressed. Much of this literature displays a neo-institutionalist logic: existing national arrangements explain why the effects of Europe are dissimilar. Currently, the literature is in its transition to a third phase, in which studies are explicitly guided by theoretical assumptions: from induction to deduction.1 The search for logics or mechanisms of domestic change and the testing of these assumptions has become a central issue (see Kallestrup; Stiller; Van Esch, this volume). As Börzel and Risse (2003, 60) succinctly state: ‘The issue is no longer whether Europe matters but how it matters, to what degree, in what direction, at what pace, and at what point of time.’

Unfortunately, the discussion on Europeanization over the last years has not been very fruitful. Instead of moving towards a certain consensus about what Europeanization is and what it is not, the debate tends to be producing more and more developments that apparently should be labeled Europeanization as well.2 The tendency to label a host of different developments, processes and changes ‘Europeanization’, implies that this concept effectively comes to lose its distinctiveness. In fact, this concept stretching (Sartori 1970; cf. Radaelli 2000) makes Europeanization a rather empty notion. Europeanization is to be situated at the domestic level, that is, it is to be seen as a top-down effect. Although I acknowledge that there is a bottom-up process going on as well - which is by no means unimportant to understand the ‘shaping’ and the ‘taking’ - it is rather awkward to call this Europeanization. Analyzing the way in which national preferences collide and coincide at the European level is part of the traditional European integration debate. Therefore, Europeanization should be an intentional isolation of a certain aspect of EU policy making, namely its domestic impact. As a consequence, there is not really much need for new definitions: both Radaelli’s (Radaelli 2000, 4) and Ladrech’s (1994, 69) definition are still valid and useful: ‘… a) construction, b) diffusion and c) institutionalization of formal and informal rules, procedures, policy paradigms, styles, ‘ways of doing things’ and shared beliefs and norms which are first defined and consolidated in the making of EU decisions and then incorporated in the logic of domestic discourse, identities, political structures and public policies’ (Radaelli) 1 Admittedly, the literature linking Europeanization and parties is still in the second, more explorative phase, as will be argued below. No general hypotheses or theoretical frames have been formulated as yet. 2 E.g. the inclusion of a ‘bottom-up’ perspective, reflected in the discussion on the ‘shaping’ and ‘taking’ of EU policy (Börzel 2003; Beyers and Trondal 2002).

Harmen A. Binnema

92

Both authors emphasize that Europeanization has to be seen as a process and they both focus on the domestic impact. I will stick to the definition of Ladrech, whilst embracing the wide scope of Radaelli, who points at various domains of Europeanization.

2.2

Parties and their functions

Some sixty years ago, Schattschneider (1942) argued that ‘democracy is unthinkable save in terms of political parties’. Although this statement may be considered too strong today, the basic idea that parties are vital organizations within society still finds support (cf. Montero and Gunther 2002). Our conception of democracy is intrinsically linked to the existence of political parties and their role in making democracy work. It has been argued that parties are important in two respects: accountability and responsiveness (Keman 1997). The first concept refers to the fact that parties can be held responsible for what they have and have not achieved at the end of an electoral term. To put is simply, it is the opportunity to ‘throw the rascals out’. Responsiveness points at the linkage function of parties, as intermediaries between citizens and the state. It means that parties have to translate citizens’ preferences into policy: there ought to be a close connection between the public agenda and the party agenda. If we accept the idea that parties have a vital role in society and in any political system, the possibility that EU integration would put this role into question is disquieting. But even if we leave aside the normative issue of the value of political parties, it would be worth examining empirically whether EU integration has the assumed effects. It has become common to argue that parties perform certain functions within a political system. Since the classic article of King, ‘Political Parties in Western Democracies’ (1969), in which six essential functions3 were distinguished, many other authors have added their own ‘variations on a theme’. Striving for parsimony, and consciously selecting from the party literature of almost 40 years since King’s contribution, I would suggest to focus on three essential functions:4

3

These functions are: 1) structuring of the vote, 2) integration and mobilisation of the mass public, 3) recruitment of political leaders, 4) organisation of government, 5) formation of public policy, 6) aggregation of interests. 4 These three functions show a certain congruence with the three strategies of parties distinguished by Strøm (Müller and Strøm 1999; 1990): vote-seeking (mobilisation), officeseeking (recruitment), policy-seeking (aggregation).

Aggregating, Mobilizing and Recruiting •

• •

93

articulation and aggregation of interests - the way in which parties bring together a number of ideas about society in a coherent ideology, expressed in party platforms and in every day political debate integrating and mobilizing the citizenry - linking citizens to the state, expressed in party membership and mobilizing people to vote in elections recruitment of political leaders and governance - selection of candidates for representative functions at the various levels of democratic decision making

Thus, the discussion in this chapter on the relation between EU integration and parties fits into the broader topic of party change (cf. Daalder 2002). Dalton and Wattenberg (2000) have pointed to a ‘declining role for political parties in shaping the politics of advanced industrial democracies’. They distinguish between micro-level -, meso-level - and macro-level changes. Under the first heading we find: an increase in educational levels, which leads to a better informed and more critical electorate with political skills of their own, less in need of parties. Moreover, the change towards post-materialist values and the diminution of group-based politics loosen the ties between citizens and parties. Second, the role of other societal actors has increased: the mass media provide an alternative source of information, whereas special-interest groups or lobbyists have taken over part of the representative role of parties. The third category involves advances in technology, allowing for regular opinion polls (e.g. focus groups), new methods of candidate selection and new styles of electoral campaigning. Finally, parties may suffer from the government’s ‘performance deficit’: this reflects the idea that the demands of citizens on governments exceed the latter’s capability. This overview clearly shows that many other factors, next to the possible influence of EU integration, impinge upon parties. In other words, it would be both unconvincing and unsatisfactory to draw up a story about EU and parties only. It is important to understand that much of the party literature starts from (empirical) observations about party changes and then tries to find reasonable explanations, of which EU integration might be one. To put it simple: where does Europe come in? This is quite different from much of the Europeanization literature, which starts from the assumption that Europe produces certain domestic consequences.

94

3.

Harmen A. Binnema

STATE OF THE CURRENT RESEARCH

Up till now, most of the Europeanization literature has focused on policy changes. These studies often invoke the concept of ‘misfit’ to link European integration and national policy. When EU regulations and national laws do not ‘fit’, there is a need for adaptation at the national level: the misfit is seen as a trigger for change. This does not mean that every instance where national and EU regulations do not match, will necessarily lead to adaptation. In certain cases, institutions and actor constellations at the national level may prove to be an impassable hurdle (cf. Knill and Lehmkuhl 2002). Conversely, when it comes to political parties, it is more problematic to speak of a ‘goodness of fit’. It is difficult to find the same logic or mechanisms of change. In other words, it does not seem very fruitful to follow the argument of ‘misfit’ and ‘adaptational pressures’ here. Second, whereas another part of the Europeanization literature already has formulated certain hypotheses which can subsequently be tested, the literature on parties is still in the ‘context of discovery’.

3.1

Europeanization and political parties

Since there is no European standard in which parties can fit or not, the trigger for change has to be found elsewhere. I would suggest that the EU effect can be seen in two respects. First, we have to look at the changes in representation: an additional level of decision making has been created, where parties have their role in the European Parliament, but also (indirectly) through Intergovernmental Conferences. The second element concerns decision making, which is reflected in the decrease of policy autonomy at the national level. Parties have to reconsider their policy proposals and ask themselves which room to maneuver and which instruments they have for acting upon their manifesto promises. EU integration might open up new possibilities for achieving policy ends ‘through the backdoor’, but it may also severely limit parties’ capabilities. One important qualification is therefore that the adaptation of parties is of a much more voluntary nature. We may witness the effect of European integration on parties in two aspects, which, as a shortcut, can be labeled program and organization. First, parties will refer to European policy and institutions in their manifestos, discuss the desirability of transferring decision making to the EU level, and allude to the limits to national policy set by EU regulation. Second, parties will have to adapt their organization to the fact that an EU level of representation has come about. This includes

Aggregating, Mobilizing and Recruiting

95

establishing a procedure for selecting candidates and drawing up a European manifesto. Moreover, MEPs will obtain their place in the national party elite and ties with sister parties in other EU countries become important. Having introduced these two causes and effects, let us now turn to the research that has been done so far on the link between Europeanization and parties. Although there are quite a number of studies dealing with the EU and parties in a broad sense, few of these fall under the heading of Europeanization as it was defined above.5 I will go through the literature, using the five dimensions discerned by Ladrech (2002) in his proposal for a new research agenda: 1. 2. 3. 4.

5.

policy change: an increased mention of EU policy and domestic policy areas where Europe has its influence, together with an emphasis on the possibilities of co-operation with transnational organizations. organizational change: change of party rules and statutes to arrange the role of the European delegation within the national party; besides, the relations with the European federations to which a party belongs party competition: politicization of the European issue; new parties that emerge based largely on their pro- or anti-EU position, and the way in which existing parties use Europe in their (electoral) strategy party-government relations: due to participation of cabinet ministers in EU forums, the relation with the party basis may become strained; parties will ‘push’ for a position that is close to their program and at the same time ‘pull’ to keep policy areas in national hands relations beyond the party system: transnational activities, co-operation with other parties from the same federation may be enhanced; emergence of a form of transnational party organization

As could be expected given the very recent entry of parties in the Europeanization debate, the literature on each of these dimensions is relatively sparse. This holds in particular for the first and the fourth dimension. Mair (2000a) hints at the risk of EU integration ‘hollowing out’ national party competition, but does not empirically study this hypothesis. Moreover, it is argued that that parties tend to depoliticize the EU issue and that they do not often mention the EU influence in various policy areas (Binnema 2003). Second, some work has been done on the relations between

5

To be sure, the authors referred to in this section do not themselves claim their work to be concerned with Europeanization. I merely try to emphasise that Europeanization is a particular focus within the broader literature on EU and parties, which has not often been elaborated.

96

Harmen A. Binnema

national parties and their European delegates: an important conclusion is that most national parties are still in control of the recruitment and nomination of candidates for the European Parliament. Moreover, there are regular meetings between MEPs and the party elite to have access to information and possibly to give voting instructions (Raunio 2000, 1999; Raunio and Hix 2000). Third, quite a number authors deal with the question of how the European issue is incorporated into the national political debate and how parties’ opinions on Europe can be linked to their standing ideologies. One of the findings here is that negative stances can be found at the extremes of the political spectrum, whereas parties in the center tend to advocate further integration or at least endorse existing levels of integration (Marks and Wilson 2000, 1999; Marks, et al., 2002). Others have refined the study of what has become known as ‘Euroscepticism’ by relating it to characteristics of the party system and in distinguishing ‘hard’ and ‘soft’ forms of Euroscepticism, in which the first represents a complete rejection of integration, whereas the latter is a more qualified objection in certain policy areas (Szczerbiak and Taggart 2000; Taggart 1998). The fourth dimension concerns the possible strains between government leaders and their ‘party on the ground’: to my knowledge, there is hardly any research that has taken up this point. Indeed, there seems to be a potential danger for parties, as their ministers are involved in decision-making at EU summits and in other meetings which are hard to control for national parties. In addition, there is the problem of the asymmetry between the party organization and its ministers in terms of timing and access to information (Raunio 1999). But this is merely speculation, since we are lacking empirical research here. Finally, research has been done on the working of party federations and transnational parties. Since the first direct election of the European Parliament in 1979, scholarly attention for this institution has grown. Studies focus on voting behavior of MEPs, (lack of) coherence in the various European party groups and coalition building in the EP (Heidar and Koole 2000; Hix 2001; Scully 1999). As Poguntke et al., (2003) correctly argue, the literature has mainly been oriented to the issue of Euroscepticism and party positions.6 We may question whether this is really the core issue when discussing EU integration and parties. First of all, because many of the scholars involved in this type of research, actually do not use the Europeanization perspective. Second, because simply mentioning Europe or positing anti- or pro-stances does not tell us very much about the extent of Europeanization. The relevant question 6 The authors suggest that this is to a large extent due to media attention, which tends to narrow down the debate to pro- and anti-EU positions.

Aggregating, Mobilizing and Recruiting

97

would be to what extent the EU has been incorporated into the ‘organizational (or ideational) logic’ of a party (cf. Ladrech 1994). One of the unfortunate results of this concentration on one aspect is that other facets of party life have been ignored. Poguntke et al., (2003) go on to suggest various elements of party organization that can be linked to EU integration. The specific focus in their research is on the role of EU specialists within parties, but one could also think of other dimensions of Ladrech’s research agenda that have been neglected so far. The general conclusion that can be drawn from the literature on the Europeanization of parties is that European integration cannot be isolated as a single factor which impacts on parties. Rather, it is argued that the EU adds to existing trends. Two short quotes illustrate this idea: ‘Popular disenchantment with parties (…) has several causes, but it is the contention of this chapter that increased policy competence by the EU aggravates this situation by emphasizing the image of the irrelevance of national parties in various policy sectors, especially economic’ (Ladrech 1999, 95).

‘Evidence indicates that integration has mainly accentuated already existing trends rather than acted as a catalyst for change. Integration has, therefore, reinforced changes that were already taking place’ (Raunio and Hix 2000, 160). It remains to be seen, then, what additional impact EU integration has on parties. Moreover, why would this impact be substantially different from or more pervasive than other challenges parties have faced over the last forty years? The general idea advanced in this chapter is that EU integration adds to a number of factors which have an impact on political parties, as previously mentioned in my discussion of Dalton and Wattenberg (2000). It seems wise to keep this caution about the relative and modest impact in mind.

4.

HOW PARTIES CHANGE

For each of the three functions distinguished above – recruitment, aggregation, mobilization – hypotheses will be formulated concerning potential EU influences. These ideas will be confronted with empirical material: party manifestos, data on party membership, and political recruitment. Data will be presented on all 15 EU member states (prior to enlargement), covering the period between 1970 and 1999. Subsequently, I will focus more in detail on the recruitment function, analyzing the career

Harmen A. Binnema

98

patterns of the members of national parliament and European Parliament in Austria, Denmark and the Netherlands, holding office in 2003.

4.1

Articulation and aggregation

The main challenge for parties concerning the articulation and aggregation function has aptly been described by Mair (2000a). The transfer of decision making authority in many policy fields from the national to the European level has left national parties with limited room to maneuver. Hooghe and Marks (2001, Appendix) argue that in 2000 there were no policy fields left in which decision making was entirely national. What is more, the issues identified as ‘commercial negotiations’ and ‘money/credit’ had become entirely European. We could expect that parties, having less to say on several topics, would have more difficulty to articulate and integrate various interests. National politicians have to acknowledge that the achievement of the pledges they make are not entirely – and sometimes entirely not – in their hands. This may be particularly problematic when policy areas which are vital for the character of a party, such as social policy or immigration policy, are transferred to Brussels. On the other hand, issues related to the EU, such as the democratization of EU decision making, the division of competences between the EU and the national level, or the desirability of enlarging the EU, might also offer new opportunities for parties. These questions might offer new avenues for political competition, if and when parties are able to translate the interests of their potential voters into coherent policy platforms. However, both the lack of clear signals from the general public and the internal divisions within parties over Europe, make this quite a difficult task. As a result, the ideas about Europe which parties put forward in their manifestos tend to be of a rather general nature and the salience of the EU issue is low in many countries. For this reason, it is hypothesized that both the ‘hollowing out’ of party competition and the absence of the EU in many national political debates, lead to an erosion of the ‘articulation and aggregation’ function of parties. This hypothesis will be tested using party manifesto data. Recently, an updated version has been published by the researchers of the Comparative Manifesto Project (Budge, et al., 2001). For those who are not familiar with the manifesto data set, a few introductory comments. The basis is a coding scheme developed in the 1970s, containing 56 policy categories that are

Aggregating, Mobilizing and Recruiting

99

furthermore divided in seven issue groups or domains, like ‘External Relations’ or ‘Welfare’. The coding runs as follows: each manifesto is divided into ‘sentences’ (or quasi-sentences, if there is more than one argument in a sentence), which are the coding units. Each (quasi-) sentence is classified in accordance with the category they best match. Although a few sentences may not fit anywhere, most of the (quasi-) sentences can be categorized under one of the 56 categories. Coders will use the same categories for each country and each party, which enables comparability across countries, across party families and across time. For each category, the frequencies are listed, thus showing the relative emphasis parties put on certain issues. The scores are to be interpreted as follows: if a party program contains 500 sentences, of which 20 are classified as belonging to category 401 Free Enterprise, we find a score of (20/500)*100 = 4.00 for this issue, which depicts its salience. What have been the most salient issues for parties in the 1970s, 1980s and 1990s? The main finding is that there has been an impressive consistency and continuity in terms of the issues prominently articulated by parties. Table 1 shows the five most salient issues for all 15 EU countries in the three different periods. From this table it can be seen that there is only limited variation only between 1970 and 1998. With the exception of Denmark and Portugal, a majority of issues that are prominent in the 1990s, also figure high on the political agenda in the 1970s and 1980s. In Finland, Spain and Sweden this is even the case for all five issues. In other words, parties continue to emphasize more or less the same issues, in spite of the fact that environmental protection or welfare, to name two prominent topics, increasingly have a European dimension too. In addition, if we look at the issues that have disappeared from the political agenda – culture, democracy, political authority – it is hard to see how this would be related to the advance of EU integration. Another interesting finding from table 1 is that the EU issue does not appear in the top-5 in any country, not even in the 1990s. However, it does move upwards in many rankings, especially in recent years. This is illustrated in detail by table 2, which shows the extent the EU issue has come to play a role in national party competition. The scores reported here refer to the relative attention paid to the EC/EU in party manifestos (both in positive and negative terms). It clearly demonstrates that the issue remains rather

Harmen A. Binnema

100

Table 1. Salient issues on the national party agendas 1970-1998 Austria

Belgium

Denmark

Finland

France

1970s 1980s 1. Democracy 1. Environmental 2. Infrastructure Protection 3. Welfare + 2. Government4. Social Justice administrative 5. Education + Efficiency 3. Democracy 4. Free Enterprise 5. Social Justice 1. Decentraliza- 1. tion + 2. Culture 2. 3. Democracy 4. Middle Class 3. 5. Infrastructure 4. 5. 1. Social Justice 1. 2. Economic Orthodoxy 2. 3. 3. Traditional Morality 4. 5. 4. Economic Goals 5. Democracy 1. Welfare + 1. 2. Social Justice 2. 3. Political Authority 3. 4. 4. Traditional Morality + 5. 5. Democracy 1. Democracy 1. 2. Social Justice 2. 3. 3. FreedomHuman Rights 4. 5. 4. Welfare + 5. Agriculture

1990s 1. Political Authority 2. Environmental Protection 3. Social Justice 4. Free Enterprise 5. Governmentadministrative Efficiency Government1. Governmentadministrative administrative Efficiency Efficiency Welfare + 2. Welfare + Environmental 3. Education + 4. Political Protection Authority Infrastructure Education + 5. Democracy 1. Welfare + Economic 2. Political Orthodoxy Authority Social Justice Welfare + 3. Environmental Protection Economic Goals Labor + 4. Labor + 5. Education + Welfare + Environmental Protection Democracy Social Harmony Traditional Morality + Social Justice Welfare + Economic Orthodoxy Democracy Education +

1. Environmental Protection 2. Welfare + 3. Social Justice 4. Social Harmony 5. Political Authority 1. Environmental Protection 2. Social Justice 3. Welfare + 4. Democracy 5. Education +

Aggregating, Mobilizing and Recruiting

101

Germany

Infrastructure Welfare + Education + Political Authority 5. Social Justice

1. Environmental Protection 2. Infrastructure 3. Welfare + 4. Labor + 5. Social Justice

Greece

1. 2.

1. Political Authority 2. Welfare + 3. Democracy 4. Culture 5. Education +

1. 2. 3. 4.

3. 4. 5. Ireland

1. 2. 3. 4. 5.

Italy

1. 2. 3. 4. 5.

1. Environmental Protection 2. Infrastructure 3. Welfare + 4. FreedomHuman Rights 5. Social Justice Democracy 1. Democracy Political 2. Social Harmony Authority 3. Political National Way Authority of Life + 4. Welfare + Social Justice 5. Political AntiCorruption Imperialism Political 1. Welfare + Authority 2. Social Justice 3. Labor + Welfare + 4. Agriculture Productivity Social Justice 5. GovernmentGovernmentadministrative administrative Efficiency Efficiency 1. Political Economic Authority Goals Welfare + 2. GovernmentFreedomadministrative Human Efficiency Rights 3. Economic Goals Government- 4. Freedomadministrative Human Rights Efficiency 5. Welfare + Education +

1. Social Justice 2. Environmental Protection 3. Welfare + 4. Governmentadministrative Efficiency 5. Incentives 1. Political Authority 2. Governmentadministrative Efficiency 3. Environmental Protection 4. Decentralization + 5. Freedom-Human Rights

Harmen A. Binnema

102

1. Welfare + 2. Environmental Protection 3. Infrastructure 4. Education + 5. Governmentadministrative Efficiency Netherlands 1. Welfare + 1. Environmental 1. Environmental 2. Democracy Protection Protection 3. Environment- 2. Government2. Governmentadministrative al Protection administrative Efficiency 4. Social Justice Efficiency 5. Education + 3. Internationaliza- 3. Infrastructure tion + 4. Internationalization + 4. Social Justice 5. Welfare + 5. Welfare + Portugal 1. Democracy 1. Political 1. Welfare + 2. Political 2. Infrastructure Authority 2. Welfare + Authority 3. Government3. Agriculture administrative 3. Democracy 4. Environmental 4. Labor + Efficiency Protection 5. Freedom4. Social Justice Human 5. Culture 5. Labor + Rights Spain 1. Welfare + 1. Welfare + 1. Infrastructure 2. Decentraliza- 2. Decentralization+ 2. Welfare + 3. Decentralization+ tion + 3. Infrastructure 3. Democracy 4. Government4. Government4. Governmentadministrative administrative administrative Efficiency Efficiency Efficiency 5. Democracy 5. Education + 5. FreedomHuman Rights

Luxembourg 1. Economic Planning 2. Welfare + 3. Education + 4. Social Justice 5. Infrastructure

1. Environmental Protection 2. Welfare + 3. Infrastructure 4. Education + 5. Culture

Aggregating, Mobilizing and Recruiting

103

1. Environmental 1. Environmental Protection Protection 2. Welfare + 2. Social Justice 3. Economic 3. Economic Orthodoxy Orthodoxy 4. Welfare + 4. Social Justice 5. Internationaliza- 5. Internationalization + tion + UK 1. Welfare + 1. Welfare + 1. Welfare + 2. Environmental 2. Social Justice 2. FreedomHumanan Rights Protection 3. Democracy 3. Law and Order 3. Education + 4. Incentives 5. Decentraliza- 4. Infrastructure 4. Infrastructure tion + 5. Democracy 5. Law and Order Note: Issues are in italics in the third in fourth column when they have been mentioned in one or both of the previous periods. Sweden

1. Welfare + 2. Environmental Protection 3. Social Justice 4. Democracy 5. Market Regulation

insignificant in most countries. Then again, there is a general increase over the years: from an average of 1.7 in the 1970s to 3.3 in the 1990s. Admittedly, this is by no means spectacular, but it suggests at least a growing awareness among parties of the importance of the EU. In the 1990s, salience is highest in Eurosceptic countries such as Germany, Sweden and the UK. Certain countries diverge from this general pattern of rising salience. In Belgium, the EU issue reached its peak in the 1970s, whereas Luxembourg scores highest in the 1980s. Denmark and the UK, who became member in the 1970s, show a decrease in the 1980s, followed by increasing attention in the 1990s. Would the prospect of joining stimulate attention for the EU issue? In Austria, Finland and Sweden, countries where there used to be hardly any attention for the EU, a rise of salience can be seen in the 1990s (particularly in Sweden, from 0.9 to 5.3). Most likely, this is related to the debates about prospective membership of the EU. A similar argument can be made for Spain and Portugal in the 1980s, if we compare these scores to the previous period. It does not hold for Greece, who also joined in the early 1980s, not even if we assume that most of the debate about accession must have taken place in the 1970s. The clearest increase for Greece can be seen in the 1990s. In other words, even the important topic of joining the EU or not does not automatically lead to increasing significance of the EU in national party competition. Nevertheless, the difference between members and nonmembers are noteworthy.

Harmen A. Binnema

104

Table 2. Salience of the EU issue in national party manifestos 1970-1998 Period Country

1970s

1980s

1990s

Austria

0.8

0.5

1.7

Belgium

3.1

2.2

2.0

Denmark

1.9

1.6

3.0

Finland

0.1

0.0

1.8

France

3.2

2.0

4.2

Germany

4.6

3.6

4.6

Greece

1.5

1.4

4.1

Ireland

0.2

1.0

3.1

Italy

2.0

2.0

2.7

Luxembourg

3.4

4.9

3.6

Netherlands

1.3

2.5

3.5

Portugal

0.7

2.3

4.0

Spain

2.0

3.6

4.2

Sweden

0.3

0.9

5.3

UK

3.4

2.3

4.7

EC/EU

2.4

2.3

3.3

Non EC/EU

0.7

0.5

-

Source: Dataset of the Comparative Manifesto Project, 1945-1998, based on issues per 108 (positive statements about EC/EU) and per 110 (negative statements about EC/EU)

Aggregating, Mobilizing and Recruiting

105

In conclusion, we have to be very cautious and probably a bit skeptical about the hypothesis that the parties are challenged in their function of ‘articulation and aggregation’ through any process of Europeanization. The manifesto data tell us that parties still put forward largely the same set of policy issues as they did before. This can be seen from the considerable overlap both between the three periods and between countries. Even the transfer of certain policy areas to the EU level over the years has not led to a downsizing of the salience of these issues, which contradicts the thesis of the hollowing out of party competition. On the other hand, parties still tend to neglect the EU issue, although there are some notable exceptions. Part of the aggregation of interests could also be for parties to have a coherent vision on EU policy and their institutions: this is an avenue for increased party competition where many parties miss out.

4.2

Integration and mobilization

The changes sketched in connection to aggregation and articulation may also have an impact on the second function, which is mobilization. Examples of this effect can be found in the decreasing membership rates of political on the one hand, and the increasing electoral volatility on the other hand. Also, the decline in turn-out rates at various elections, and political cynicism more in general, point at declining mobilization capacities of parties. All these factors point at an increasingly unsteady and unpredictable relationship between parties and citizens. If, as Mair (2000) argues, EU integration increasingly demonstrates the irrelevance of national politics, this situation is likely only to grow worse. On the other hand, perhaps ironically, EU integration may sometimes turn out to be useful for parties. In particular when nation-wide referenda are held, e.g. about Treaty changes or joining the Euro, anti-EU parties manage to mobilize large numbers of people. In table 3, the levels of party membership in 15 EU countries are reported. Next, data are given for Australia, Canada, New Zealand and Japan to compare with trends outside the EU. Membership rates display a small drop already in the 1980s, but the 1990s show a very strong decline. The times of the mass party, so prominent in the 1950s and 1960s, are clearly over.7 In Austria, Ireland and Sweden, the decline is particularly steep: compared to the 1980s, there is a decrease of respectively 12.4%, 8.8% and

7

It should also be noted that in many countries there has never been such a ‘golden age’ of partisan politics. The mass party was mainly to be found in the Scandinavian countries, whereas e.g. Germany has never had high membership levels.

Harmen A. Binnema

106

16.2%. Other countries, such as Denmark, France, the Netherlands and the UK also show a steady trend downwards. It is interesting to note that ‘new’ democracies Greece and Spain divert from the general pattern. In Greece, membership rates increase consistently and they are above average in the 1990s. Membership rates in Spain are much lower, but they also show a rise from the 1980s to the 1990s. The third newcomer, Portugal, partly fits in this picture as well, but differs through its slight decrease in the 1990s. Turning to the non-EU countries, we find a similar trend of declining membership. In Canada and Australia these are small decreases, since membership has never been high, but in New Zealand this is a rather steep trend downwards. On the contrary, Japan witnesses an increase in party membership, thus belonging to the few countries which move in the opposite direction. What do these results suggest for the mobilization function of parties? First, it is remarkable that the largest decrease of membership takes place in the same period when EU integration intensifies. Moreover, the two most extreme cases of decline, Austria and Sweden joined the EU in the 1990s. On the other hand, this decline is not unlike developments in non-EU countries, like Australia or New Zealand, apart from being somewhat more dramatic. In sum, whereas we can safely conclude that parties have increasing difficulty with their function of mobilization, we have to be cautious again to attribute this to EU integration. In line with the more general idea posited in the literature, we may suggest that EU integration adds to the growing disentanglement of voters and parties (cf. Schmitter 2000), but we cannot claim a strong and direct relationship.

4.3

Recruitment and governance

Mair (2000b) has argued that the procedural functions of parties, especially the recruitment of candidates for parliament and for government, will remain important (and may even gain significance), whereas the representative functions are in decline. As a result, parties are said to change more and more into governing agencies. This argument builds on the ‘cartel thesis’ (Katz and Mair 1995), which asserts that there is a growing interdependence of parties and the state.

Aggregating, Mobilizing and Recruiting

107

Table 3. Party membership 1970-2000 (% of the electorate) Period Country

1970s

1980s

1990s

Austria

29.1

25.9

13.5

Belgium

8.6

8.9

5.9

Denmark

10.1

7.1

5.0

Finland

15.7

13.5

9.2

France

3.2

2.9

1.3

Germany

4.2

4.3

3.1

Greece

3.1

5.8

6.7

Ireland

9.1

11.8

3.0

Italy

11.4

9.6

-

Netherlands

4.6

3.4

2.1

Portugal

2.8

4.9

4.6

Spain

-

1.7

2.8

Sweden

21.5

22.3

6.1

UK

3.7

3.5

2.0

Australia

3.7

2.6

1.5

Canada

-

3.5

2.9

Japan

1.1

2.7

4.2

New Zealand

14.6

7.8

2.1

Source: Krouwel (2004; table 2) for EU countries; Webb et al. (2002) and Dalton/Wattenberg (2000) for other countries

Harmen A. Binnema

108

How are EU integration and the recruitment function of parties related? The nomination of candidates for European Parliaments is controlled by national parties, as long as European elections are nationally based. Second, national parties have their influence through their government representatives in the Council of Ministers. These ministers have also been selected by national parties. The main Treaty changes are negotiated at Intergovernmental Conferences by Prime Ministers, who are also party leaders. Finally, the members of the European Commission are nominated Table 4 Partisan dominance of parliamentary and cabinet seats 1970-2000 (% elected on party list) Period

1970s

1980s

1990s

Parliament

Cabinet

Parliament

Cabinet

Parliament

Cabinet

100.0

100.0

100.0

98.3

100.0

94.2

Belgium

100.0

100.0

100.0

100.0

99.5

100.0

Denmark

100.0

99.0

100.0

100.0

100.0

100.0

Finland

100.0

90.8

100.0

98.2

100.0

97.5

France

NA

84.3

NA

89.4

NA

93.3

Germany

100.0

98.7

100.0

100.0

100.0

98.7

Greece

100.0

100.0

99.8

98.1

99.8

100.0

Ireland

97.9

100.0

97.8

96.7

96.7

100.0

Italy

99.9

98.4

100.0

98.0

99.0

87.6

Luxembourg

99.1

100.0

100.0

100.0

100.0

100.0

Netherlands

100.0

100.0

100.0

100.0

100.0

100.0

Portugal

99.9

41.8

100.0

87.6

100.0

83.6

Spain

99.7

93.9

99.9

97.6

100.0

77.3

Sweden

100.0

98.0

100.0

99.0

100.0

95.7

UK

99.9

100.0

100.0

100.0

99.9

100.0

Country Austria

Source: parliamentary seats based on own calculations from Mackie and Rose (1994) and various websites; cabinet seats are from Siaroff (2000); scores in bold represent cases with more than 10% independent candidates

Aggregating, Mobilizing and Recruiting

109

by national (party) governments. Altogether, parties are quite in control of national and European career patterns, be it directly or indirectly. In particular, the position of national party leaders seems to be strengthened (see Krouwel 2004). The hypothesis to be tested here is that parties are still very strongly performing their recruitment function, and their privileged position both at the national and the European level might even enhance their role. The scores in table 4 refer to the percentage of parliamentary seats (left column) and cabinet seats (right column) which are occupied by partisan candidates. The remaining seats are in the hands of independent candidates, not attached to a particular party or list. This table shows very clearly that parties play an extremely dominant role. In Belgium, Denmark, Germany, Greece, Luxembourg and the Netherlands and the UK, nearly all or all candidates belong to a political party: the share of independent candidates nowhere exceeds 2%. Basically the same can be argued for countries like Austria, Finland, Ireland and Sweden, where partisan candidates always make up more than 90% and often more than 95% of parliamentary and cabinet seats. In contrast to this majority of countries, France and Portugal offer a significantly different picture. In the 1970s, Portugal even had a majority of non-partisan cabinet members. Although the dominance of parties is much larger in the two periods that follow, there remains a share of approximately 15% of independent candidates. In France, independent candidates are also more prominent than in most other countries. As in Portugal, this is particularly the case in the 1970s, whereas partisan dominance increases in the 1980s and is at a considerably high level in the 1990s. Italy and Spain are two deviant cases, since the share of partisan candidates tends to be high, with the exception of the 1990s. In both countries, this is due to one specific cabinet which included a high number of independent candidates (in Italy there was even a 100% independent cabinet). These figures show very clearly that parties are still in charge of recruitment. Looking at the share of independent candidates in parliament, this control has even tightened. Leaving aside the very few exceptions, it seems to be almost impossible to enter parliament without being a party member. When it comes to cabinet composition, the outlook for non-partisan candidates is somewhat more positive. Although it often remains limited to one or two cabinet members, the share of independents is much higher. But the conclusion can only be that recruitment is still very much the prerogative of parties.

Harmen A. Binnema

110

In other words, we can confirm that the recruitment function of parties is not affected. On the contrary, recruitment could turn out to be a vital function for parties to cling to, in order to remain relevant in the long run. The crucial point is that whereas the former two functions might also be performed by other actors, the latter is essentially in the hands of parties. As a matter of fact, one main aspect that distinguishes parties from other societal organizations is their dominant position in the recruitment of candidates for office. This is not without reason: parties have the organizational and financial capacities for recruitment. Parties also allow for ‘trial and error’: candidates have to prove themselves at the local or regional level, to show that they are capable of moving up to more prominent positions.

4.4

Career patterns

The final part of this empirical section will look at the career patterns of MPs and MEPs. It is based on an analysis of the biographies of the members of the Danish Folketing (179), the Dutch Tweede Kamer (150) and the Austrian Bundesrat (183), and those of the Danish (16), Dutch (31), and Austrian (21) members of the European Parliament, all of these in 2003. The aspect of Europeanization which will be explored here is the extent to which parliamentary careers become increasingly intertwined. The hypothesis is that distinctions between the European level and the national level become more and more blurred. Shifts from the regional to the European level and back to the national level are expected to become more normal, rather than being the exception. The question is whether these changes occur or whether these career patterns still very much separated? The reason for choosing these three countries is that they have become member of the EC/EU in different periods. One might expect the ‘interaction’ effect to be strongest in long-standing member the Netherlands, and weakest in Austria, which only joined in 1995. In other words, I expect that the exchange of careers should be highest in the Netherlands, where contacts between party representatives a the EU and the national level has the longest history. At the same time, the context for recruitment of candidates for EU and national posts is quite similar. The electoral system is proportional representation in some form, leading to multiparty systems and rather low electoral thresholds. Austria, Denmark and the Netherlands are also very much alike in the size of both the national parliaments and the EU delegations.

111

Aggregating, Mobilizing and Recruiting

First, we will look at one general characteristic of the members of the national parliaments and EU national delegations: the division according to age. It is a popular view that the European Parliament is a ‘graveyard’ for politicians at the end of their career. In other words, politicians whose role at the national stage is over, find refuge in Strasbourg, never to return to national politics again. In contrast, the interaction hypothesis outlined above expects career moves in both directions. A first hint is provided by table 5: members of the EP are indeed on average several years older than their colleagues in national parliaments. In Denmark and the Netherlands, this difference amounts to approximately 8 years, in Austria it is a bit more than 4 years. This would seem to suggest that membership of the EP comes at a later stage of one’s political career. Of the national parliaments, the Netherlands has the ‘youngest’ representatives (average 44.8), whereas the Danish parliament has the ‘oldest’ (average 48.8).

Table 5. Composition of national parliaments and EP delegations according to age

Austria

< 35

35-50

50-65

> 65

Average

4.4

55.2

39.3

1.1

47.7

EU delegation

0.0

38.1

57.1

4.8

52.0

National

16.8

29.1

50.3

3.9

48.8

EU delegation

0.0

18.8

62.5

18.8

56.3

National

11.3

57.3

30.7

0.7

44.8

67.7

3.2

52.8

National parliament

Denmark

parliament

Netherlands

parliament

EU delegation

6.5

22.6

Looking at the EP, the Austrian delegation has the lowest average age with 52 years, whereas the Danes again score highest with 56.3 years on average. When we specify this further to the different age groups, we find that the first category (below 35 years) is absent in both the Austrian and Danish delegation. In all three EU delegations, the main part of the delegates falls in the age category between 50 and 65. In Denmark, this is also the case for the national parliamentarians, but in Austria and the Netherlands, the plurality can be found in the 35-50 category. This is yet another hint at the fact that membership of national parliament and EP come at different stages, or even that we are dealing with two different groups of party politicians.

Harmen A. Binnema

112

Let us now turn more in detail to the career patterns of our group of parliamentarians. Tables 6 and 7 show which positions MPs and MEPs have occupied in their entire party career prior to their first term in parliament (national or EP). The tables distinguish between three possible other positions within the party: a) representative at the local or regional level,8 b) member of the party executive and c) member of national parliament (for MEPs) and European parliament respectively (for MPs). These three categories have been chosen because they all refer to an ‘internal’ career, i.e. making your way through the ranks of a party. Obviously, these three categories are not mutually exclusive: it is very well possible that a certain politician has been both a member of the party board and a regional representative before going to parliament. Table 6. Previous career of Members of national parliament European

Regional level

Party executive

Parliament Austria

0.0

56.8

10.4

Denmark

1.1

44.7

22.9

Netherlands

1.3

45.3

18.0

Of all 512 national MPs in this dataset, only four had previously held a seat in the European Parliament. As can be seen from table 6, this is a very marginal proportion: no-one in Austria, two in Denmark and two in the Netherlands. Overwhelmingly, parliamentarians have previous experience in regional or local councils. This goes in particular for Austria (56.8%), but Denmark and the Netherlands are also close to 50%. A very common pattern therefore is to climb the ladder within the party hierarchy: first to start and try at the local or regional level, and if one is good enough, he or she gets promoted to national or European politics. In Denmark, a considerable number (22.8%) has previously been member of the party executive, before entering parliament. A comparable situation can be found in the Netherlands (18.0%), whereas in Austria this is a less common path (10.8%). It should

8

This implies that I have discounted previous experience in local committees or local party boards.

Aggregating, Mobilizing and Recruiting

113

also be noted that several MPs have a combination of local experience and membership of the party executive.9 In sum: the career patterns of Austrian, Danish and Dutch MPs show a striking similarity. It is remarkable that practically none of the MPs has previous experience in the European Parliament. In the case of Austria, this may be understandable, since there is only one generation of former MEPs as yet (1996-1999). The last national elections were held when the Fifth EP (1999-2004) was still underway. But it is surprising to find that also in Denmark and the Netherlands, former MEPs play a marginal role in the selection of candidates for national parliament. Table 7. Previous career of Members of European Parliament National

Regional level

Party executive

Parliament Austria

33.3

38.1

14.3

Denmark

31.3

6.3

56.3

Netherlands

12.9

48.4

38.7

Only four MPs had previously been elected to the European Parliament, but as many as 16 MEPs (out of 68) are former MPs. In other words, a move in the other direction is far more common. Compared to the other two countries, only few Dutch MPs switch to the European Parliament (12.9%). Moreover, in contrast to what we saw in table 6, the career patterns of MEPs do show some variation by country. In Denmark, the majority of MEPs has previous experience in the party executive, which again does not hold for Austria and much less for the Netherlands. Whereas previous experience at the local or regional level is again rather dominant in Austria and the Netherlands, this does not apply to Denmark (no more than 6.3%).10

9 As can easily be seen, the percentages do not add up to 100%. Other previous experiences include: membership of interest organisations (such as trade unions or agrarian organisations), party youth organisations, and positions in the private sector or media. 10 Another factor that has to be mentioned here, is that certain parties in Denmark only compete at the EU level, which implies that they do not have regional or national representatives, ruling out two possible career patterns.

Harmen A. Binnema

114

But the importance of the regional level does allow us to expand the ‘climbing of the ladder’ idea to the MEP career as well. Some parliamentarians seem to move to the EP with national representation in between, whereas others go directly to the European level. One explanation may be the growing involvement of the EU with regions, which enhances awareness of the EU among regional representatives. Another hunch is that those regional politicians who do not make on the national list, take a second chance with the EP list. Concluding: European Parliament and national parliaments are still very much two worlds apart. When career moves do take place, they are mostly in one direction: i.e. from the national to the EU level. Only in some rare cases do we find the opposite move. Europeanization understood as increasing interaction between the national and the EU level hardly takes place. Since recruitment is entirely in the hands of national party organization, the influence goes in terms of candidate selection goes in one direction: from the national to the European level.

5.

DISCUSSION AND CONCLUSION

This chapter started, first, with an attempt to come to a useful and sensible definition of Europeanization and second, to introduce party functions as the most sensible way to study party change. Defining Europeanization has meant leaving out several other aspects of EU politics, which, though very interesting and relevant, should probably not be headed under Europeanization, and to deliberately isolate one specific element: the domestic impact of EU integration.. It also became clear that it is difficult to apply the mechanisms of change distinguished in the literature on policy adaptation (based on the logic of ‘misfit’) to political parties. Instead, the main reasons for adaptation - of a more voluntary nature - seem to stem from the additional and growingly influential level of representation and the transfer of decision making. This is not unproblematic, since we have to be very cautious – and explicit at the same time – in linking EU developments and national consequences. The main part of the paper has been the analysis of the way in which EU integration might lead to party function change. Three functions were considered essential: recruitment, aggregation, and mobilization. The analysis has suggested that the representative functions (aggregation and mobilization) are under pressure, but only partially this can be linked to the EU. Whereas one would expect aggregation to become more difficult due to the fact that many decisions are no longer in the hands of national governments and national parties, the manifesto data showed that parties continue to put forward more or less the same policy issues. At the same

Aggregating, Mobilizing and Recruiting

115

time, parties hardly pick up the EU issue in national party competition, which suggests some problems in terms of aggregation. As long as parties do not challenge each other on the EU issue and its domestic impact, they can continue to debate the same issues, whether these are decided at the national level or not. Second, erosion of the mobilization function could be seen in the decline of party membership, and the trend of increasing electoral volatility can be added to this. The link between parties and voters has been weakened considerably; again we can merely suggest that EU integration has added to this trend, but a direct relationship cannot be established. Third, recruitment is far less under pressure; on the contrary, parties continue to be very dominant. The detailed analysis of career patterns of MPs and MEPs, looking at Europeanization as interaction, showed little effect of EU integration. National and European careers are still quite separated, and moves tend to be unidirectional, with the European Parliament as the terminal station. It also hinted once more at the control of parties over recruitment. In particular, the national party organization remains crucial for all kinds of political careers. These findings are quite different from other chapters on Europeanization in this volume. There is not a clear and straightforward effect of EU integration. Instead, the impact of the EU seems be rather limited, as the pressure for change is very low, and there are no strong mechanisms linking the national level to the EU. In other words, changes in politics are simply of a different nature than policy changes. Even if ‘objectively’ the impact of the EU is large, this does not automatically produce party function change. In all three cases, it has been argued that EU integration is not the only factor impacting on these functions; rather, it complements existing trends such as individualization, higher levels of education, advances in technology and communication, and the rise of interest groups - trends that are widely acknowledged in the literature on party change. Considering the balance between the three functions of aggregation, mobilization, and recruitment, the impact seems to be more on more on the latter. In the terms of Mair: parties are becoming governing agencies instead of representatives. As long as the selection of candidates, both at the national and the EU level, is firmly in the hands of parties, they will continue to play an important role. Although this may be good news for parties, it remains to be seen whether this is good news for citizens and democracy as well.

REFERENCES Beyers, J., and Trondal, J., 2002, How nation states hit Europe - ambiguity and representation in the EU, European Integration online Papers 7(5), http://eiop.or.at/eiop/texte/2003005a.htm.

116

Harmen A. Binnema

Binnema, H. A., 2003, How Europe hits parties . . . or not? Europeanisation of party programmes in the Netherlands and the United Kingdom, Working Papers Political Science 3/2003. Börzel, T. A., 2003, Shaping and taking EU policies: member state’s responses to Europeanisation, Queen’s Papers on Europeanisation 2/2003, http://www.qub.ac.uk/ schools/SchoolofPoliticsInternationalStudies/FileStore/Europeanisat ionFiles/Filetoupload, 5270,en.pdf. Börzel, T. A., and Risse, T., 2003, Conceptualising the domestic impact of Europe, in: The Politics of Europeanization, K. Featherstone and C. M. Radaelli, eds, Oxford University Press, Oxford, pp. 57-80. Budge, I., Klingemann, H.-D., Volkens, A., and Bara, J., eds, 2001, Mapping Policy Preferences: Estimates for Parties, Electors, and Governments, 1945-1998, Oxford University Press, Oxford. Daalder, H., 2002, Parties: denied, dismissed, or redundant? A critique, in: Political Parties: Old Concepts and New Challenges, R. Gunther, J. R. Montero and J. J. Linz, eds, Oxford University Press, Oxford, pp. 39-57. Dalton, R. J., and Wattenberg, M. P., 2000, Unthinkable democracy. Political change in advanced industrial democracies, in: Parties without Partisans: Political Change in Advanced Industrial Democracies, R. J. Dalton and M. P. Wattenberg, eds, Oxford University Press, Oxford, pp. 3-16. Featherstone, K., 2003, Introduction: in the name of ‘Europe’, in: The Politics of Europeanization, K. Featherstone and C. M. Radaelli, eds, Oxford University Press, Oxford, pp. 3-26. Haverland, M., 2003, The EU and the nation-state, paper for NIG course State of the Art Political Science, Utrecht, 13-17 January. Heidar, K., and Koole, R., eds, 2000, Parliamentary Party Groups in European Democracies: Political Parties behind Closed Doors, Routledge, London. Hix, S., 2001, Legislative behaviour and party competition in the European Parliament: an application of Nominate to the EU, Journal of Common Market Studies 39(4):663-88. Hooghe, L., and Marks, G., 2001, Types of multi-level governance, European integration Online Papers 5/11, http://eiop.or.at/eiop/texte/2001-011a.htm. Katz, R. S., and Mair, P., 1995, Changing models of party organization and party democracy the emergence of the cartel party, Party Politics 1(1):5-28. Keman, H., 1997, The politics of problem-solving: democratic responsiveness and political accountability, in: The Politics of Problem-Solving in Postwar Democracies, H. Keman, ed., Macmillan, Basingstoke, pp. 162-87. King, A., 1969, Political parties in Western democracies, Polity 2(2):111-41. Knill, C., and Lehmkuhl, D., 2002, The national impact of European Union regulatory policy: three Europeanization mechanisms, European Journal of Political Research 41(2):255-80. Krouwel, A., 2004, All politics is national, but policy is supra-national: a decisive discrepancy, paper for Virtual Conference ‘Changing Party Systems in a Deepening and Widening Europe’, Friedrich Ebert Stiftung, 8 March - 30 April. Ladrech, R., 1994, Europeanization of domestic politics and institutions - the case of France, Journal of Common Market Studies 32(1):69-88. Ladrech, R., 1999, Political parties and the problem of legitimacy in the European Union, in: Legitimacy and the European Union, T. F. Banchoff and M. P. Smith, eds, Routledge, London, pp. 95-112. Ladrech, R., 2002, Europeanisation and political parties: towards a framework for analysis, Party Politics 8(4):389-403. Mair, P., 2000a, The limited impact of Europe on national party systems, West European Politics 23(4):27-51. Mair, P., 2000b, Populist democracy vs. party democracy, paper for ECPR Joint Sessions, Copenhagen, 14-19 April.

Aggregating, Mobilizing and Recruiting

117

Marks, G., and Wilson, C. J., 1999, National parties and the contestation of Europe, in: Legitimacy and the European Union: the contested polity, T. F. Banchoff and M. P. Smith, eds, Routledge, London, pp. 113-33. Marks, G., and Wilson, C. J., 2000, The past in the present: a cleavage theory of party response to European integration, British Journal of Political Science 30(3):433-59. Marks, G., Wilson, C. J., and Ray, L., 2002, National political parties and European integration, American Journal of Political Science 46(3):585-94. Montero, J. R., and Gunther, R., 2002, Introduction: reviewing and reassessing parties, in: Political Parties: Old Concepts and New Challenges, R. Gunther, J. R. Montero and J. J. Linz, eds, Oxford University Press, Oxford, pp. 1-35. Müller, W. C., and Strøm, K., 1999, Policy, Office, or Votes? How Political Parties in Western Europe Make Hard Decisions, Cambridge University Press, Cambridge. Poguntke, T., Aylott, N., Ladrech, R., and Luther, K. R., 2003, The Europeanisation of national political parties: a framework for analysis, paper for 2nd ECPR General Conference, Marburg, 18-21 September. Radaelli, C. M., 2000, Whither Europeanization? Concept stretching and substantive change, European Integration online Papers 4/8, http://eiop.or.at/eiop/texte/2000-008a.htm. Raunio, T., 1999, Always one step behind? National legislatures and the European Union, Government and Opposition 34(2):180-202. Raunio, T., 2000, Losing independence or finally gaining recognition? Contacts between MEPs and national parties, Party Politics 6(2):211-23. Raunio, T., and Hix, S., 2000, Backbenchers learn to fight back: European integration and parliamentary government, West European Politics 23(4):142-68. Sartori, G., 1970, Concept misformation in comparative politics, American Political Science Review 64(4):1033-53. Schattschneider, E. E., 1942, Party Government, Holt Rinehart and Winston, New York. Scully, R. M., 1999, Between nation, party and identity: a study of European Parliamentarians, EPRG Working Paper 5-99, http://www.lse.ac.uk/collections/EPRG/pdf/ Working%20Paper%205.pdf. Strøm, K., 1990, Minority Government and Majority Rule, Cambridge University Press, Cambridge. Szczerbiak, A., and Taggart, P., 2000, Opposing Europe: party systems and opposition to the Union, the Euro and Europeanisation, SEI Working Paper 36, http://www.sussex.ac.uk/ Units/SEI/pdfs/wp36.pdf. Taggart, P., 1998, A touchstone of dissent: Euroscepticism in contemporary Western European party systems, European Journal of Political Research 33(3):363-88.

II. EUROPEAN INTEGRATION – INTEGRATION AND COOPERATION AMONG EUROPEANIZED STATES

THE EUROPEANIZATION OF CENTRAL DECISION MAKERS’ PREFERENCES CONCERNING EUROPE A Perpetual Motion? Femke van Esch Abstract:

1.

The main question addressed in this chapter is how past European integration feeds back into future European integration. The author sets out to answer this question by researching whether European integration affects the worldviews of central decision makers that inform their preferences concerning further (monetary) European integration, and whether there are any indications that the Europeanization of preferences is dependent on the nationality or functional position of decision makers. To do so, the worldviews of the French and German Heads of State and Government, and Ministers of Finance concerning the establishment of a European economic and monetary union are studied using the technique of cognitive mapping. Beliefs held by central decision makers in the early 1970s – hen Europe was still in its infancy– are compared with those held by decision makers in the early 1990s, –when the scope and depth of EU integration had greatly intensified. The chapter shows that –in the 1990s– preferences were more Europeanized than in the 1970s, thus supporting the Europeanization thesis. Furthermore, the study indicates that decision makers functional position may have an intervening effect on the Europeanization of preferences, whereas on the other hand no evidence is found that nationality matters.

INTRODUCTION

As stated in the introduction, one of the central purposes of this book is to go beyond the first and second wave of Europeanization theory and systemically combine and integrate understandings of how European integration comes about (bottom-up), and of how Europe affects national policies and policy-making (top-down). This chapter sets out to contribute to this purpose by researching how past European integration feeds back into

119

R. Holzhacker and M. Haverland, (eds.), European Research Reloaded: Cooperation and Integration among Europeanized States, 119-149. © 2006 Springer. Printed in the Netherlands.

120

Femke van Esch

future European integration via the Europeanization of the worldview on which the preferences of national central decision makers are based. Traditional approaches within IR-theory concerning European integration (bottom-up) have argued that for (further) European co-operation to come into existence a common interest between central European member states needs to exist. In other words, a convergence of national preferences concerning European integration is assumed to be a necessary pre-condition for such integration to be established. As such, the preferences of central decision makers in these member states are pivotal variables in the feedback loop between past and future European integration: they are the crucial link between the ‘shaping’ and the ‘taking’ of Europe. To determine whether European integration may have had an effect on subsequent (non)decisions to engage in further European co-operation it is thus interesting to establish whether a Europeanization of the worldviews on which national preferences are based has actually taken place. This chapter will try to answer this question by comparing the worldviews of central decision makers in France and Germany informing their preferences concerning the establishment of an European economic and monetary union (EMU) in the late 1960s – when Europe was still in its infancy and in the late 1980s – when major instances of European integration like the European Monetary System (EMS) and the signing of the Single European Act (SEA) had taken place. To do this, a technique from the cognitive approach to foreign policy analysis – known as cognitive mapping – will be applied. This chapter will start, however, by defining the central concepts – preferences and Europeanization – and briefly to set out the hypotheses that will be tested in this chapter.

2.

EUROPEANIZATION OF NATIONAL PREFERENCES

In recent years, the meaning of the term (national) preference has become a hotly debated issue in IR-literature (Bueno de Mesquita 2000; Finnemore 1996; Frieden 1999: 41-6; Gourevitch 2002: 313; Hug 2003; Moravcsik 1993, 1997; Princen 2002: 37; Risse 2002: 605, 609; Weldes 1996). While adherents of the Rational Choice approach to international relations narrowly define the concept as referring to some kind of (self-)interest - something that is ‘objectively’ to one’s ‘advantage, good or profit’ (Wall 1975: 489; cf Connolly 1972: 462; Kratochwill 1982: 5-6; Cochran 1973: 751-2),1 Social 1 Most of the time, adherents of the Rational Choice approaches to preferences assume maximalization of power to be an actor’s ultimate (self-)interest. Formally, however, there is

Europeanization of Central Decision Makers

121

Constructivists, on the other hand, reject the notion of objectivity and stress that preferences are of an inter-subjective and normative nature (Checkel 1998: 325-6; Guzzini 2000: 159; Price and Reus-Smit 1998: 266; Weldes 1996: 279). The analysis in this chapter will be based on the assumption that the preferences of central decision makers are informed by their worldviews, their beliefs on ‘how the world works’. These worldviews encompass both decision makers ideas on how events and policy-alternatives affect their ‘interests’ – their ideas of what they perceive is costly and what beneficial – as well as their (shared) ‘principled beliefs’ – their normative ideas of what is right and what is wrong, just or unjust (Goldstein and Keohane 1993: 9). Worldviews are thus assumed to encompass what Rationalist as well as what Social Constructivist perceive to be the dominant origin of preferences. In recent years, the term Europeanization has become as much debated as the term preference is (see the introduction to this volume by Holzhacker & Haverland and the chapter by Binnema). In this chapter, the more a decision maker’s worldview takes into account previous instances of European integration or the existence of the European decision making arena, the more Europeanized its preference is said to be. This means that an actor that is said to have a more Europeanized preference may not be the actor most in favor of European integration, may not be the most ‘pro-European’. The statement merely indicates that its preference is more heavily informed by ‘the experience of Europe’. Such experience may be valued positively, and induce the decision maker to strive to develop European integration even further, however, it is naturally also possible that a decision maker attaches a negative value to the consequences of past European co-operation and has ‘learned’ never to engage in (a particular type of) European integration again.2

2.1

Learning and Socialization

Until now, most Europeanization research has investigated the Europeanization of national policies, legislation, and the national political

nothing that would preclude Rational Choice approaches to making the assumption that actors are driven by idealistic motives rather than material (self) interest (Finnemore and Sikkink 1998:910, 912; Lieshout 1995: 47-50; Risse 2002: 599). 2 In other words, what will not be researched in this chapter is whether (further development of) Europe has led to a change in the value the actor attaches (from positive to negative or vice versa) to the establishment of European economic and monetary union. Furthermore, although this might be an interesting research topic, nor will be researched whether European integration has led actors to become more or less favorable towards European integration.

122

Femke van Esch

system.3 More recently, students have also started to study the effects of Europe on national discourses and national preferences.4 There is however a marked difference between the mechanisms of change underlying the Europeanization of national policies, legislation and the national political system and the Europeanization of national preferences and discourses. In the former case, member states have often made formal commitments to adapt to European legislation and rules and, as such, they are under the legal obligation to Europeanize. Moreover, in these cases European institutions and other advocates of adaptation to the European ‘standard’ may have formal powers to enforce change on the nation states. Because in the case of the Europeanization of national preferences no legal obligation to adapt exists, Europeanization of preferences – by definition – will be ‘voluntary’, and will rely entirely on a change in the worldviews of central decision makers. In other words, Europeanization of central decision makers’ preferences may only occur through voluntary belief change. Previous instances of European integration may cause such ‘voluntary belief change’ by inducing learning, or the socialization or persuasion of central decision makers. In this chapter, learning is defined as the individual5 process of cognitive belief change that is induced by information about the world ‘out-there’ (Finnemore 1996: 12; Keohane and Nye 1987: 749; Levy 1994: 292).6 Such process may either be spurred on by the individual’s realization that the world has changed and its beliefs need updating, or by its acknowledgment that its existing beliefs were faulty (a process which does not necessarily require change in the environment). Learning implies 3 For examples of these kinds of studies in this volume see the chapters by Mastenbroek & van Keulen, Stiller, Kallestrup, and Binnema. 4 For other examples of these kinds of studies in this volume see the chapters by Munster & Sterkx, Leuffen & Luitwieler, and Schaefer. 5 Due to its inherent individual nature, actors do not necessarily ‘learn’ the same lessons from similar experiences (Haverland 1999: 25). However, in the literature, one special form of learning is distinguished that does have the inherent potential to lead to a convergence in policy-preferences: emulation (or imitation). Emulation refers to the process by which an actor becomes aware of the fact that the policies pursued by others lead to a greater successrate in attaining a state-of-affairs that - seeing its values - is more desirable than its own policies, and decides to copy these policies (cf Goldsmith 2000; McNamara 1998: 153-7). Like in the case of learning, this realization is achieved without any social interaction between the imitator and the actor: emulation is caused by observation, not participation, and as such, it is a process of individual cognitive change. Past instances of European integration may facilitate and foster emulation. Previously established European institutions like the Council of Ministers or the European Council form an arena in which the central decision makers from the different member states meet, interact and are confronted with the policy-alternatives chosen by their partner states and the results of these policies (Haverland 1999: 250-6). This may induce them to adopt each others’ worldviews. 6 Due to its basis in methodological individualism, learning is a mechanism of cognitive belief change that is often stressed by adherents of Rational Choice approaches.

Europeanization of Central Decision Makers

123

that after the process has been completed – the individual perceives its newly acquired ideas to be more accurate than those it previously held, and that these ideas will help it to realize its ultimate values in a more effective or efficient manner (Alderson 2001: 424; Etheredge 1981: 77; Haverland 1999: 25).7 However, preferences acquired through a learning-process do not necessarily provide an ‘objectively’ more accurate reflection of the outside world, or help the actor to get closer to the realization of its goals (Keohane and Nye 1987: 749-50; Levy 1994: 292).8 Past European integration may thus influence central decision makers’ preferences concerning further European co-operation by inducing learning. Decisions taken at the European level and the policies that result from these decisions may lead to changes in the political, economic and social environment of actors. The actor may pick up on these changes which may result in a change in its worldview, and in a different evaluation of how certain policy-alternatives may contribute to the realization of its values. Moreover, the implementation of policies decided on at the European level may have an unexpected positive or negative effect on the actor’s values, which may induce the decision maker to change its policy-preferences accordingly (Moravcsik 1998, see also the introduction to this volume by Holzhacker & Haverland). In addition to learning, Europeanization of national preferences may also occur through the socialization of central decision makers. The concept of socialization is based on the ‘sociological’ idea that cognitive beliefs are a property of the social system rather than of the individual’s mind, and that an individual’s worldview and preferences are influenced by the social belief systems that are commonplace in their social environment (their class, or professional, ethnic, or national environment) (Little 1988: 38; Smith 1988: 32). A change in the social environment – like an increased involvement in the European arena – may thus result in a change in an individual’s preferences. Unlike learning, socialization is a social, an interactive process, and - as such - a change in preferences through socialization is not acquired individually but occurs through and during social interaction (Alderson 2001: 423; Checkel 1998: 344; Checkel 2001: 561). Socialization can take two forms. Firstly, it may be the natural and unintended result of belonging

7

As such, learning – by definition - can not result in value change. Levy rightly argues that making the assumption that learning would objectively result in more accurate beliefs would require developing a standard by which we can identify and measure the accuracy of learning. However even in hindsight, it is very hard to categorize beliefs on foreign policy making as accurate (Levy 1994: 292). As such, ‘to insist on an accuracy criterion would either result in research that is less rigorous, more subjective, and more dependent on the analytical and normative biases of the analyst, or it would paralyze analysis because of the lack of measurement standards’ (Levy 1994: 293). 8

124

Femke van Esch

to, and functioning in a certain social group. On the other hand, it may be the result of a third actor’s attempt purposely to convince the other players from the value of its own ideas, a process that is also referred to in literature as ‘persuasion’ (Johnston 2001: 488-9).9 In addition to inducing learning, Europe may also facilitate socializationprocesses. Europe may function as an arena in which close social contacts between decision makers from different states and representatives of the European institutions are facilitated. As such, it may provide central decision makers with a new social environment – a new peer group – and a new social belief system to adopt. Moreover, European institutions may also provide a forum for (mutual) persuasion and argumentation. Finally, international institutions like the European institutions may take on a life on their own, acquire their own autonomous preferences and actively start to persuade its founders of the value of further European integration (Barnett and Finnemore 1999; Van Esch 2001: 116-7). Both the learning and socialization mechanisms lead to the same hypothesis concerning the relation between the level of existing European integration and the level of Europeanization of central decision makers’ preferences. Firstly, a focus on the ‘learning-mechanism’ would lead to the expectation that the more European integration has occurred, the more likely the environment of the individual has been affected by its consequences and the more likely a decision maker has incorporated previous instances of European integration as relevant concepts in the worldview that informs its preferences. Likewise, a focus on the ‘socialization-mechanism’ would expect that more European integration would lead to an increased importance of the European political arena and European institutions in the decision makers’ daily functioning. As a result, the decision maker will participate and interact more frequently and more intensely at the European arena, and it will thus be more likely that the European decision-making arena is an integral part of its worldview. All and all, both the learning and socialization mechanisms lead to the expectation that the more European integration exists, the more Europeanized the preferences of central decision makers will be.

2.2

Alternative hypotheses: Intervened Europeanization

The above hypothesis suggests that the existence of European policies and institutions may alter the preferences of national decision makers. It may be debated whether persuasion – because of its intentional nature – is truly a form of socialization. However, like ‘true socialization’ and in contrast to learning this mechanism does have its effect through and during social interaction.

9

Europeanization of Central Decision Makers

125

However, several authors have hypothesized that Europe may not have the same effect on all actors: Europe may ‘fit’ the existing worldviews of some actors better than those of others, and as such may have weaker, stronger, or a substantively different effect on the preferences of some than on those of others (Risse 2001; Risse, et al., 1999; see also the introduction to this volume by Holzhacker & Haverland). In the literature on preference-formation, factors like the national identity – developed in reaction to specific national historic experiences – and the functional position of a decision maker within the state bureaucracy are generally perceived as the two most important determinants of decision makers’ preferences (Allison and Zelikow 1999: Chapter 3; Krasner 1978: 10; Risse 2001; Risse, et al., 1999). Adherents of the national identity thesis on preference formation posit that the specificity of a national identity is based largely on shared historical experiences, and determines how central decision makers ‘view their instrumental interests and which preferences are regarded as legitimate and appropriate for enacting’ (Risse, et al., 1998: 12). For instance, the collective German aversion to the horrors of Nazi-history is said to have led to the German pro-European stand, and the collective memory of rising inflation would explain their attachment to price stability (Risse, et al., 1998: 28). Adherents of the proposition that the functional position of decision makers in the government bureaucracy decisively influences their preferences, argue that members of an organization may be socialized into adopting the ‘essence of the organization’; ‘the view held by the dominant group in the organization of what the missions and capabilities should be’ (Halperin 1974: 28). This proposition is summed up most eloquently in Allison’s famous quote ‘where you stand depends on where you sit’ (Allison and Zelikow 1999: 307).10 These theoretical notions suggest that central decision makers should not be considered to be ‘tabula rasa’, or be assumed to share a similar worldview when they are confronted with the idea or experience of ‘Europe’. In fact, the above-mentioned authors posit that – depending on their nationality and functional positions – central decision makers’ worldviews may differ significantly, and as such their interpretation and evaluation of European integration may differ also. As such, the ideas underlying European integration may ‘fit’ some national identities or essences of organizations better than others, and the better the ‘fit’, the more easily ‘Europe’ is incorporated in the worldview 10

This tendency may even be reinforced by the existence of the European political arena, for different national functional positions also entail different positions in the European decision making structure: Ministers of Finance meet in the ECOFIN whereas Ministers of Foreign affairs meet in the General Council.

Femke van Esch

126

that informs the central decision makers’ preference. In other words, their national identity and/or functional position may have an intervening effect on the Europeanization of central decision makers’ preferences (see figure 1 & 2)

European

Decision

integration

makers’ preferences National identity/history

Figure 1. The Intervening Effect of Nationality

European

Decision

integration

makers’ preferences

Position in government

Figure 2. The Intervening Effect of Functional Position

3.

RESEARCH-DESIGN

To test the above-mentioned hypothesis that the more European integration exists, the more central decision makers’ preferences will be

Europeanization of Central Decision Makers

127

Europeanized (informed by these instances of European integration), a comparative case-study of the preferences of central decision makers will be conducted. The cases to be studied concern the preferences of the German and French Heads of State and Government, and Ministers of Economics and Finance concerning the establishment of a European economic and monetary union in the late 1960’s and 1980’s. Due to their positions in the national and European political system (seat in the European Council and ECOFIN) these are the pivotal actors in the decision making process on European ‘grand bargains’ in the monetary issue-area. The choice to study central decision makers’ preferences concerning the establishment of an European economic and monetary union was made because both in the 1969 and in 1989, the European member states set out on the road towards the establishment of such a union on the basis of policy-plans (the Werner-plan and the Delors-plan) that were very similar in content (Scholtens, 1990). As such, the topic on which central decision makers in the late 1960s were expected to articulate a preference (the dependent variable) is comparable to that of the 1980s. However, with regards to the independent variable identified by the Europeanization thesis, there is a crucial difference between the first and second attempt to establish a European economic and monetary union. Due to measures like the establishment of the European Council in 1974, the introduction of direct European parliamentary elections, the establishment of the European Monetary System in 1979, and the decision to complete the European single market with the Single European Act of 1986, Europe was integrated to a much larger degree in the late 1980s than in the late 1960s. As such, the research design would allow for the testing of the Europeanization hypothesis introduced above. Moreover, the cases that were chosen to be studied in this chapter also allow for a comparison of the level of Europeanization in the two periods across state and functional position (see table 1 & 2). Werner-report France Head of State/ Government Georges Pompidou Minister of Finance Valery Giscard d’Estaing

Germany Willy Brandt Karl Schiller

Table 1. EMU in the Late 1960s: The Central Decision makers Studied Delors-report Head of State/Government Minister of Finance

France François Mitterrand Pierre Bérégovoy

Germany Helmut Kohl Theo Waigel

Table 2. EMU in the Late 1980s: The Central Decision makers Studied

Femke van Esch

128

Moreover, by choosing to research the degree of Europeanization of central decision makers’ preferences concerning an instance of further European integration (EMU), this chapter will enter the third stage of Europeanization research, and try to combine and integrate understandings of how Europe affects national policies and policy-making (top-down), and of whether European integration comes about (bottom-up). In other words, it will try to investigate how the effects of Europe feed back into the making of Europe through the preferences of central decision makers.

3.1

Measuring Preferences

In studying the preferences of decision makers, one comes across an important methodological problem: preferences are located in the minds of individuals and therefore can not be observed directly. In principle, this problem is not insurmountable. When unable to observe the characteristics of variables directly, a scientist may study its observable consequences which may be located in other aspects of the subject being studied (King, et al., 1994: 41). Following this line of reasoning, one can distinguish two observable consequences that may follow from having preferences; decisions and assertions. In other words, it may reasonably be assumed that when actors have different preferences, these differences are reflected in the decisions they make, and in their assertions. Both of these assumptions have their drawbacks. However, without making either, preferences have no observable consequences and cannot be studied. In my opinion, when trying to answer the question whether preferences of central decision makers are Europeanized studying assertions is to be preferred over studying decisions. For it is doubtful whether studying decisions would even allow for the testing of the Europeanizationhypothesis, for what exactly would constitute a ‘more Europeanized decision’? In my mind, this question could be answered only when Europeanization would be defined as being more pro-European, which would not do justice to the full meaning of the concept. All in all, the research strategy of this chapter will thus be ‘to base what is being measured on what is being asserted’ (Axelrod 1976: 10). More specifically, the three hypotheses introduced will be tested through a technique developed within the context of cognitive foreign policy analysis known as ‘cognitive mapping’. In contrast to content-analysis – which is essentially ‘a counting procedure’- the technique of cognitive mapping is specifically aimed at analyzing the relationships between the concepts that

Europeanization of Central Decision Makers

129

decision makers use (Axelrod 1976: 7). In other words, the basis of analysis is the relationship between concepts, not the concepts themselves. This allows the investigation of whether, and to what degree decision makers have incorporated previous instances of European integration in the worldview that informs their preferences concerning the establishment of European economic and monetary union, and whether they perceive these instances as having led up, or contributed to the establishment of EMU. In general, cognitive maps are derived from existing assertions of policymakers like policy-papers, speeches, and (existing) interviews. This has the advantage that the method is unobtrusive, and a reconstruction of the ideas due to hindsight or the odd workings of memory is avoided (Axelrod 1976: 7). In addition, to increase their reliability, the maps used in this research were composed on the basis of several speeches, writings, or interviews concerning the establishment of a European economic and monetary union. Moreover, when possible, speeches, writing and interviews were chosen that were held in very different settings and before different audiences11. This is to avoid – as much as possible – that the maps would represent rhetoric rather than preferences. From these assertions, the sections that dealt directly with European monetary co-operation were selected and analyzed.12 To transform the assertions into maps ‘the concepts a person uses are represented as points, and the causal links between these concepts are represented as arrows between these points. This gives a pictorial representation of the causal assertions of a person as a graph of points and arrows’ (Axelrod 1976: 5).13

11

For a list of the speeches, writings and interviews used, see appendix 1. The choice for the section was based on content-analysis of the writings, speeches and interviews. Only the sections in which the terms ‘monetary’ ‘co-operation’ and ‘European’ (or any related terms like ‘unification’ ‘union’, the ‘EEG’, ‘EU’, the ‘Community’ or ‘the Six’) where all mentioned, were analyzed. 13 The concepts and relationships represented in the map are thus not pre-determined by the researcher, but represent thoughts and concepts used by the decision maker him/her self (Axelrod 1976: 6-7). 12

Femke van Esch

130

A

+ D

+

+

B

-

+

C

E

Relationships: + Positive - Negative o Non-existent

Figure 3. Example of a Cognitive Map

These arrows are provided with a plus, minus or zero-sign to indicate whether the identified policy-alternative is perceived to contribute to the realization of a goal (value), or not, or whether the decision maker (explicitly) perceived the alternative to have no effect on the goal.14 For instance, Minister Schiller’s assertion ‘wir leben inmitten einer erheblichen internationalen Preis- und Lohnwelle, die uns in unseren isolierten Anstrengungen um mehr Stabilität vor schwere Problemen stellt’ (Bulletin des Presse- und Informationsamt der Bundesregierung: nr. 126, 22 September 1970: S.1301) indicates a perceived negative relationship between ‘inflation abroad’ (Preis- und Lohnwelle) and ‘German price stability’ which will be depicted by an arrow with a negative sign leading from the former to the latter concept (like the relation between B and C in figure 3). Ultimately, what results is a graphical representation of the different policy-alternatives and goals that the decision maker distinguishes in his/her assertions, and the (sign of) the relationship between them (see figure 3). These maps allow for the identification of any previous instances of European integration that are distinguished by the decision maker (see

14

In general, cognitive maps are composed of only the causal relationships. In this chapter – like in many psychological uses of cognitive mapping- the arrows in the maps ‘may be viewed as equivalent, respectively, to the English expressions ‘is usefully or desirably associated with’ and ‘is adversely or undesirably associated with’” (Axelrod 1976: 6-7). As such, these maps encompass relationships indicating the ‘Rationalist’ means-ends relationships between concepts as well as the more ‘Social Constructivist’ normative relationships.

Europeanization of Central Decision Makers

131

table 3, column 1). Moreover, within the literature, techniques have been developed to establish the value the decision maker attaches to each of the concepts and relationships represented in the map. The value of a concept – called ‘centrality’- depends on how many concepts are directly linked to it. In other words, the centrality of a policy-alternative or goal is computed by adding up the number of concepts that are directly linked to it as causes (indegree) and how many as effects (out-degree) (Heradstveit and Narvesen 1978: 81). For instance, in figure 3, the centrality of concept A is two because two arrows depart from concept A and none feed into it. Likewise, due to its two inward-going arrows and one departing arrow, the centrality of concept C is three. By depicting the centrality of each concept as a percentage of the sum of all in-degrees and out-degrees within the map15 the relative value of all concepts within the map may be computed, which makes comparisons of the centrality of concepts across different maps possible (table 3, column 1 & 2). However, a comparison of the relative centrality of concepts referring to previous instances of European integration sheds light only on the general importance of such concepts within the worldview of the decision maker. In order to determine whether, and to what degree decision makers perceive past European integration to contribute positively or negatively to the establishment of a European economic and monetary union or have induced a need for one, a closer analysis of the perceived causes of such a union must be conducted. To do this all so-called ‘antecedent paths’ of the concept European economic and monetary union should be recorded. To derive such ‘antecedent path’ a concept that leads into the goal to be researched (EMU) should be identified, after which a concept must be identified that – in turn – leads to that concept, and to repeat this procedure until no prior concepts can be distinguished (in figure 3, for instance, the sequence A-D-E forms an antecedent path to concept E). This process should be repeated until all possible antecedent paths of a concept within the map are identified (Bonham and Shapiro 1976: 322-4). In other words, until all possible (sequences of) variables that are perceived to lead up to the establishments of a European economic and monetary union are recorded.16 From this list of antecedent paths, it can be derived whether – in the mind of the decision maker – previous instances of European integration have (indirectly) fed into the establishment of European economic and 15

The arrows in a cognitive map are all drawn from one concept to another; none of the arrows begin or end in the void of air. Therefore every arrow represents a cause (in-degree) as well as an effect (out-degree). The total sum of in-degrees and out-degrees is thus – by definition- twice the amount of arrows in the map. 16 Concept E in figure one for instance has five (not three!) antecedent paths: A-D-E, A-C-E, B-C-E, and D-E and C-E.

132

Femke van Esch

monetary union. Moreover, by analyzing the signs of the relationships that link the different concepts in an antecedent path, it may be deduced whether these variables are positively or negatively associated to the establishment of EMU.17 However, the real added value of deriving antecedent paths to merely representing them in the graphical map is that it allows for the perceived relative ‘causal’ power of the different distinguished concepts on the establishment of EMU to be computed. To do this, firstly, the average centrality of each antecedent path should be calculated.18 This is done by adding the centrality of all concepts that are part of the path, and to divide this by the number of relationships that are included in the path (Bonham and Shapiro 1976: 325, n. 8; Heradstveit and Narvesen 1978: 81).19 For instance in figure 3, the centrality of the path A-D-E is calculated as follows: the centrality of A (2) and the centrality of D (2) are added up (4) and divided by the number of arrows (2) which leads to a centrality of two.20 Secondly, the centrality of all the paths leading out from a specific concept into the goal to be researched should be added up. This determines the (absolute) causal value of that concept on the goal that is being researched. In other words, in figure 3 the centrality of the paths A-D-E and that of A-CE should be added up to arrive at the (absolute) causal value of concept A on concept E.21 Finally, the sum of these centralities must be represented as a percentage of the sum of the centrality of all antecedent paths of EMU to 17

An antecedent path between two variables that includes only positive relationships naturally indicates that the variables are positively connected, so does a path that contains (in addition to any amount of positive relationships) an even amount of negative relationships. However, a path that contains an uneven amount of negative relationships indicates a negative relationship between variables. Finally, a path between two variables that contains any number of zero-signs indicates that the causal variable is not connected to the effect (Axelrod 1976: 7). 18 In general, the centrality of paths is used in studies to resolve any imbalances within cognitive maps. An imbalanced map is one in which both a positive and a negative relationship (path) exists between the same two variables, thus contradicting each other. In cognitive theory, it is assumed that the path with the lowest centrality will be suppressed by the decision maker, and its preferences or decisions will be arrived at following the reasoning of the dominant path/relationship (Bonham and Shapiro 1977: 325, n. 8; Heradstveit and Narvesen 1978: 81). 19 This latter step should be made because otherwise, the longer the path between the cause and the effect (EMU), the higher the score. In other words, the more indirect a variable is linked to the value to be researched, the stronger the effect. 20 Because the issue of this chapter is to derive the causal value of certain variables on EMU, the centrality of EMU is not added to that of the others. If it were included, the value that an actor attaches to the establishment of EMU (preference, dependent variable) would become part of the calculation of the values attached to its causes (independent variable). 21 In figure 1, the causal value of A on E is (A-D-E =( 2+2=4/2=) 2 + A-C-E =(2+3=5/2=) 2.5 =) 4.5.

Europeanization of Central Decision Makers

133

allow for comparison across different cognitive maps. All in all, through these calculations it may be established which of the variables (concepts) are perceived by the decision maker to have contributed most usefully or desirably to the establishment of EMU (table 3, column 3), and what the relative salience of the several (or the total of) distinguished instances of past European integration is (table 3, column 4). By comparing the average results calculated for the 1960’s decision makers to those of the 1990s, we may arrive at an answer to the question whether European integration feeds back into the worldviews that inform central decision makers’ preferences concerning further European integration. Due to the small number of cases (8), conclusions drawn from this research will naturally be tentative.22 The ‘many variables, small-N’ problem is even more extensive when drawing conclusions on the ‘intervened-Europeanization’ hypotheses. For although by studying both the German and French Heads of State – and Government and the Ministers of Finance for both periods the research-design allows for an investigation of both these hypotheses, the comparison across nationality and functional position within the two periods that is needed for this analysis will reduce the number of cases per cell to two. As such, the analyses of the ‘intervenedEuropeanization’ hypotheses conducted in this chapter are – at most – ‘plausibility-probes’ of the external validity of the hypotheses. To increase the reliability of the analysis concerning the intervened-Europeanization hypotheses, a qualitative analysis of the whole cognitive map will be added to the quantitative analysis outlined above. Purpose of this qualitative analysis is to see whether – in general – the maps display national or functional characteristics by researching whether there are any obvious similarities in the worldviews (in concepts used or relationships drawn) of central decision makers of the same nationality or functional position.

4.

EUROPEANIZATION OF PREFERENCES

In this section, the results of both the study of the direct and intervened effects of Europe on the worldviews and preferences of central decision makers are reported. However, the section will start by briefly elaborating on the decision makers’ evaluations of the value of the establishment a European economic and monetary union. From the eight decision makers studied, only Pompidou explicitly argued against the establishment of EMU. While in his speeches he refers positively to the establishment of such a 22

Due to the enormous labor-intensity of the technique of cognitive mapping, however, it is unfeasible to include more cases.

Femke van Esch

134

union, he consistently defines EMU as a form of co-operation lacking any supranational institution, and outright opposes the creation of a single currency (Conférence de presse, 2 juillet 1970 & 21 janvier 1971, in; Cousté and Visine 1974: 110, 115). In addition, Schiller, Waigel and Kohl are what I would call ‘reluctant supporters’ of the establishment of EMU. Schiller, for instance, poses serious pre-conditions to its establishment (economic and monetary parallelism, and the ‘Clause de Prudence’, see amongst others; Aussprache zur Großen Anfrage zur Außenpolitik, 29. Januar 1971 in: Bulletin des Presse- und Informationsamt der Bundesregierung: nr. 14, 2 Februar 1971: S.149) and Kohl even stresses that ‘German sacrifices’ have to be made to achieve this goal (Für eine gemeinsame europäische Sicherheitspolitik, 22. Januar 1988 in: Kohl 1989: 157). Waigel – in turn – stresses that only ‘further negotiation’, and ‘long processes’ will lead to the establishment of EMU (Keine Währungs-Union ohne Geldwert-Stabilität in: Rheinischer Merkur, 18 Augustus 1989, S.13). Brandt, Giscard d’Estaing, Mitterrand and Bérégovoy are all ‘convinced supporters’ of the establishment of EMU, and in general perceive a number of (material or normative) benefits to result from its establishment.

4.1

The Direct Effect

The analysis of the cognitive maps of the central decision makers point out that of those in power in the 1960s, the preference of Chancellor Brandt is by far the most Europeanized. In his view, the relative centrality of the previous instances of European integration (the Hague summit and the decision to enlarge the EEG) both make his top-5 of most salient concepts, a result not matched by any of the other 1960s politicians (see table 3, column 2). Moreover, past European integration is not only a salient concept in his general worldview, the Chancellor also identifies both the Hague summit and the widening of the EEG as amongst the pivotal events leading up to the establishment of EMU (table 3, column 3). All in all, this results in a relative causal power of past instances of European integration of 26.2% (see table 3, column 4).

Politician

Previous instances of European integration identified (Centrality in % of total value of concepts)

19601970

Brandt

Hague summit Widening

Pompidou

Hague summit CAP

6.3 6.3

3.6 3.6

Centrality of concepts (in % of total value of concepts)

1) EMU 2) Completion of EEG 3) German policy-making 4) Improved relations with East-E. 5) Hague summit 6) Widening 1) Eur. mon zone 2) Eur. Confederation 3) Intergovernm. decision making 4) Europe as world power Europe as a front

13.5 11.5 10.4 6.3 6.3 6.3 16.7 8.0 6.3 4.5 4.5

Causal power of variables on the establishment of EMU (% of total causal power)

58.6 1) German policy-making 17.4 2) Hague summit 3) Widening 8.7 4) Reinforcement of EEG Inst. 5.8 5) Solidarity 5.1 1) Intergovernm. decision making 31.6 2) Economic harmonisation 13.2 3) Common managed reserves 9.9 7.1 4) Europe as a front 5.7 5) Shared: CAP / Hague

Causal power of past instances of European integration on the establishment of EMU (% of total causal power) Hague summit 17.4 8.7 Widening

Europeanization of Central Decision Makers

Period

Total 26.2 Hague summit 5.7 5.7 CAP

Total

11.3

135

136

19801990

2.7

Schiller

Hague summit

Giscard d’Estaing

Internal market 1.3

Kohl

SEA EMS

9.1 2.3

18.8 1) EMU 2) Inflation in EEG 10.7 4.5 3) Parallelism 4) Benefits for all (U) 4.5 5) Shared: further int / EPU 3.6 10.5 10.5 9.2 5.3 5.3 15.9 11.4 9.1 8.0 5.7

Hague summit 4.7

Total 4.7 Internal market 0.0

Total SEA EMS

0.0 16.5 2.7

Total

19.2

Femke van Esch

1) Further Eur. Integration 2) Eur currency/mon. personality 3) Breton Woods 4) Fr pref. for BW revision US budgetary deficit 1) (Further) Eur. Integration 2) French-German friendship 3) SEA 4) EMU/ECB 5) Price stability

30.5 1) Inflation in EEG 27.3 2) Parallelism 9.4 3) Clause de Prudence 4) French-German cooperation 6.3 5) Shared: o.a Hague Summit 4.7 1) Further Eur. Integration 28.4 2) Serious problems 20.0 3) Breton Woods 18.9 4) Fr. pref. for BW revision 8.4 5) No enforcement of fin. Discipline 6.3 1) Germany’s Nazi history 35.9 2) French-German friendship 27.7 3) SEA 16.5 4.4 4) Price stability 5) Central Bank independence 4.0

2.0 1.0

Waigel

SEA EMS Delors report

9.2 1.3 1.3

Bérégovoy

SEA EMS Delors report

3.1 1.0 1.0

1) (True) Eur. Unification 2) SEA 3) EMU/ECB 4) European inst. Progress 5) Political unification 1) EMU 2) Price stability 3) Economic liberalisation 4) SEA 5) Shared 1) EMU 2) Further Eur. Integration 3) Stable Growth 4) Social Security 5) Loss of sovereignty

18.5 9.8 7.6 7.6 6.5 14.5 14.5 10.5 9.2 5.3 19.4 10.2 6.1 6.1 5.1

1) SEA 2) German re-unification 3) German commitment 4) Delors report

47.4 26.3 21.1 5.3

1) SEA 28.9 2) Economic liberalisation 24.6 3) High governmental deficit 9.6 4) Investments 8.6 8.1 5) Price stability 100,0 1) Delors Report

47.4 SEA E.I. prior to 0.0 SEA Delors report 5.3 Total 52.7 28.9 SEA 0.0 EMS Delors report 5.0 Total 33.9 0.0 SEA 0.0 EMS Delors report 100.0 Total

Europeanization of Central Decision Makers

9.8

SEA E.I. prior to SEA Delors report

Mitterrand

100.0

Table 3. Relative Salience of Previous Instances of European Integration in the Worldviews Informing Central Decision makers’ Preferences Concerning EMU

137

138

Femke van Esch

Despite his aversion of a true (supranational) European monetary union (one which would result in a single currency), the preference of the French President is a good deal more Europeanized than that of his Minister of Finance. For, whereas previous instances of European integration do not take up a pivotal role in his general worldview (relative centrality of The Hague summit and the Common Agricultural Policy is 3.6 %, see table 3, column 1), he does consider these to contribute to the establishment of (what he would call) EMU. Overall, in the view of the former French President, the relative causal power of past European integration amounts to a total of 11.3% (table 3, column 4). The relative causal power of past European integration in the cognitive maps of the 1960s’ Heads of State and Government stands in stark contrast to that in the worldviews of their Ministers of Finance. Both of them identify only one previous instance of European integration in their map: Schiller mentions the The Hague summit, Giscard d’Estaing the internal market. However, whereas Schiller perceives the The Hague summit to make a – admittedly very modest – contribution to the establishment of EMU (relative causal power of 4.7%, see table 3, column 4), the French Minister of Economics and Finance does not draw any relationship between the internal market and the establishment of EMU (see table 3, column 4). All in all, these 1960s Ministers seemed to identify the international monetary system and its institutions to be far more salient than European integration. In the 1980s, the results show a very different image. Firstly, the German Finance Minister, Theo Waigel identifies three instances of past European integration in his map: the Single European Act (SEA), the European Monetary System (EMS), and the Delors-plan (see table 3, column 1). Of these factors, in his view, only the SEA and the Delors-plan are associated (positively) to the establishment of EMU (see table 3, column 4). In fact, he perceives the SEA to be the single most important factor to contribute to the realization of EMU with a causal power of 28.9% (see table 3, column 3). The Single European Act is also the most important instance of past European integration that is distinguished by the German Chancellor, Helmut Kohl (see table 3, column 4). However, in his map, the relative causal power of SEA with regards to the establishment of EMU only amounts to 16.5%. As such, in Kohl’s worldview, the SEA is not only amply outranked as contributing factor by the concepts ‘overcoming Germany’s Nazi history’ (35.9%), and the ‘French-German friendship’ (27.7%), he also awards the Act of 1986 significantly less causal power than his Minister of Finance (see table 3, column 3). Waigel’s preference concerning the establishment of EMU thus seems to be much more Europeanized than Kohl’s. However, the German Finance Minister’s preference is not nearly as Europeanized as that of the French President François Mitterrand. In his

Europeanization of Central Decision Makers

139

map, the causal power of the different previous instances of European integration on the establishment of EMU that he distinguishes amounts to a grand total of 52.7 percent, of which the Single European Act is again the unparalleled most salient event (scoring 47.4 %, see table 3, column 3 & 4). The last central decision maker from the late 1980s whose preference has been studied in this chapter is Pierre Bérégovoy, the French Minister of Economics and Finance. His map indicates that although he distinguishes three previous instances of European integration – the Single European Act, the European Monetary System, and the Delors report – none of these are very salient in his worldview (the relative centrality of these instances of past European integration are respectively 3.1 %, 1% and 1%, see table 3, column 1). Moreover, of the total of 33 concepts distinguished in his cognitive map – of which he perceives 24 to be a direct or indirect result of the establishment of EMU – he only perceives one as leading up to the establishment of EMU: the Delors-report. This leads to the peculiar outcome that – despite the relatively low salience of past European integration in his worldview – the lack of any other concepts perceived to feed into the establishment of EMU results in a relative causal power of 100 percent for the Delors-report.23 Overall, central decision makers in the 1980s have distinguished more instances of previous European integration (an average of 2.75 versus an average of 1.5, see table 4, column 1). Moreover, these instances took up more a salient position in their worldviews than in those of their counterparts (see table 3, column 2). In the 1960s, for instance, only Brandt perceives previous instances of European integration to be salient enough to make top 5, whereas in the 1990s this is the case in three out of four decision makers. Moreover, of the concepts perceived to feed into the establishment of European economic and monetary union, all decision makers from the 1980s perceive past European integration to be an important factor (perceived relative causal powers rank between 19.2 and 52.7 %, see table 3, column 4) whereas in the 1960s, decision makers did not perceive previous instances of European integration to be nearly as salient (perceived relative causal powers rank between 0 and 26.2 %, see table 3, column 4). All in all, the average of relative causal power of past instances of European integration as perceived by central decision makers in the 1980s is more than triple that in the 1960s (see table 4).24

23 This result must be considered as unreliable and has a considerable distorting effect on the collective results. As such, I have chosen to omit these results from the analyses (see below). 24 As indicated in note 23, table 4 reports the results based on only seven cases: the results of Bérégovoy have been omitted from the calculations. However, even when the relative causal power of previous European integration as perceived by Bérégovoy would be assumed to be

Femke van Esch

140 Werner-report 10.6

Delors-report 35.7 (26.5)

Table 4. Europeanization of National Preferences (average of relative causal power of previous instances of European integration based on 7 cases) Overall, the research conducted in this chapter seems to have yielded support for the hypothesis that the more European integration, the more Europeanized the worldviews that inform the preferences of national decision makers.

4.2

Intervened Europeanization Theses

The intervened-Europeanization hypotheses, however, questions whether a direct relationship between the degree of European integration and the level of Europeanization of the preferences of central decision makers really exists. Is this effect not dependent on certain particular characteristics of the decision makers like their nationality or functional position? As regards to the possible intervention of nationality on the Europeanization-effect the analysis conducted in this chapter is inconclusive. On the one hand, the qualitative analysis of the cognitive maps indicates that there are some marked similarities among countrymen and difference across nationality. For instance, all German decision makers identify the relationship with EastGermany as either a goal or a means in their map, a factor clearly pointing back to their specific national historic experience. Moreover, in the worldviews of all German decision makers ‘price stability’ turns up as a very salient concept (combined centralities of Schiller – European, German, and world – price stability-: 16; Waigel: 11; Kohl: 5; Brandt: 3), and in three out of four maps the autonomy of the future European monetary institutions is distinguished as an important concept. In the literature on preference formation, both factors are often carried back to the specific German experience with hyperinflation in the 1920’s. In contrast, only the French decision makers from the 1980s mention the German problem, none of them mentions the autonomy of European monetary institutions, and price-stability is a significantly less salient concept in their worldviews (centralities of Giscard: 3; Pompidou: 1; Mitterrand: 0; Bérégovoy does not mention price stability as such only ‘stable growth’ – growth without inflation – which has a centrality of 6). On zero – the most unfavorable assumption with regards to the Europeanization hypothesis – the average for the 1980s would still amount to 26.5, two and a half times that of the 1960s).

Europeanization of Central Decision Makers

141

the other hand, a typical French concern seems to be the shape and condition of the international monetary system, which all central decision makers but Mitterrand mention in their speeches (from the German decision makers only Schiller mentions this factor); a find that is in line with the French struggle against the might of the US and dollar that is often mentioned in the literature on French foreign (economic) policy. In contrast to the qualitative analysis of the cognitive maps, however, the results of the quantitative analysis do not support the thesis that there is a national dimension to the Europeanization of preferences (see table 5), for whereas the average relative causal power of past instances of European integration was significantly higher in the maps of the German decision makers in the late 1960 than in those of the French, the effect seems to have been reversed in the 1990s.25 In any case, the preliminary analysis conducted in this chapter does not come up with any support in favor of the hypothesis that national characteristics have an intervening effect on the Europeanization of preferences. Germany France

Werner-report 15.5 5.7

Delors-report 26.6 52.7 (26.4)

Table 5. The National Dimension of Europeanization (average of relative causal power of previous instances of European integration based on 7 cases) With regards to the hypothesis that the level of Europeanization of decision makers’ preferences depends on their functional position, both the qualitative and quantitative analysis come up with some tentative support in favor of this thesis. For instance, the rankings of most salient concepts (see table 3, column 2) show that, whereas for only one out of the four Heads of State or Government an ‘economic’ variable is salient enough to make their top 5 (Kohl ranks ‘price stability’ as fifth with a relative centrality of 5.7%, see table 4, column 2), the Ministers of Finance all rank at least two economic factors in their top 5 (relative centralities ranking from 14.5% to

25 Again these results are based on only seven cases. The score in the 4th quadrant of table 5 thus represents the causal value of previous instances of European integration in the perception of Mitterrand. However, even when one would assume the score of Bérégovoy to be zero – the most favorable assumption for the hypothesis – the score of the French decision makers in the 1990s would have been almost the same as that of the Germans (26.4 to 26.6). As such, it may thus be concluded that – at minimum – any effect of nationality on the effects of Europe on decision makers’ preferences that may have been present in the 1960 has disappeared over the years.

Femke van Esch

142

4.5%). In other words, the analysis comes up with some evidence supporting the – somewhat trivial – conclusion that central decision makers seem to be actually fulfilling the different functions they are supposed to be fulfilling: the Ministers of Economics and Finance deal with economy and finance, while the Heads of State and Government are concerned with the more general political means and ends. However trivial the reasoning, the functional effect on the Europeanization of preferences is also supported by the quantitative analysis that was conducted (see table 6). For, although the level of Europeanization between the Ministers of Economics and Finance and the Heads of State and Government differs only slightly in the late 1980s (33.9 versus 36, see table), the result clearly supports the hypothesis. Overall, the general analysis of the cases studied in this chapter seems to indicate that the preferences of the Heads of State and Government are more Europeanized that those of the Ministers of Finance. As such, it may well be that the functional position within government – through the education and experiences associated with such a position – has an intervening effect on the Europeanization of the worldviews that inform decision makers’ preferences concerning further European integration.26 Heads of State/government Min. of Finance

Werner-report 18.6

Delors-report 36.0

2.4

33.9 (17 - 43.3)

Table 6. The Functional Dimension of Europeanization (average of relative causal power or previous instances of European integration based on 7 cases)

26

Again, the score in the 4th quadrant of table 6 represents the score of only one case: that of the German Minister Waigel. By omitting the deviant case of Beregovoy the results are the same as if his score would be the same as Waigel’s: 33.9% (given the low general value Bérégovoy attaches to previous instances of European integration - none of the instances he has distinguished made the centrality top 5, see table 3, column 1 & 2 – this estimate is already high). However, when one would assume the score of Bérégovoy to be 52.7 – the highest score of any of the decision makers, and thus the least favorable assumption for the hypothesis – the score of the 1990’s Ministers of Finance would be 43.3 percent. Under this assumption, the results of the quantitative analysis would no longer support the analysis. A score of zero – the most favorable assumption for the hypothesis – would result in a average score of 17.

Europeanization of Central Decision Makers

5.

143

CONCLUSIONS

Overall, the research conducted in this chapter thus seems to have yielded support for the hypothesis that – in these cases – as European integration advances to a higher level, the more Europeanized the worldviews on which central decision makers base their preferences. The quantitative analysis of this study shows a clear difference in the Europeanization of the decision makers’ preferences in the late 1960’s – which on average scored 10.6% – and the late 1980s which had an average score of 35.7% (or at minimum 26.5%). Seeing the lack of support for the Europeanization-thesis that the different studies in this book seem to come up with, this is a remarkable conclusion. Despite the fact that due to the low number of cases studied in this chapter – a problem that has been significantly aggravated by the outlier status of the scores of Minister Bérégovoy – one should be very careful to make generalizations based on the findings of the study conducted, the tentative evidence that is found in support for the existence of an effect of European integration on central decision makers’ preferences may justify further research into this effect in different issue-areas, and amongst a different set of decision makers. Moreover, the contrast between the results found in this chapter (as well as in the chapter of Leuffen and Luitwieler) and the studies concerning the Europeanization of national policies, legislation and political systems raises the question whether paradoxally the influence of Europe through learning and socialization is not greater than through the exertion of relational power. In addition to the tentative support for the ‘bare’ Europeanization hypothesis that was found, a quantitative and qualitative ‘plausibility probe’ of the intervened Europeanization hypotheses was conducted. With regards to the hypothesis that specific national historic experiences or identities had an intervening effect on the Europeanization of preferences, the cases studied in this chapter supplied no supporting evidence. In contrast, some support was found in favor of the hypothesis that the functional position of decision makers has an intervening effect on the Europeanization of their preferences. Firstly, in the qualitative analysis some obvious differences between the maps of the Minister of Finance and those of the Heads of State – and Government were found. In addition, the quantitative analysis indicated that – in the cases researched – the level of Europeanization of the worldviews of the Heads of State and Government was significantly higher than that of the Ministers of Finance in the late 1960s (18.6% versus 2.4%), and slightly higher in the late 1980s (36% versus 33.9%). Whereas – due to the even lower number of cases per cell – these results must be judged to be very preliminary, the results may warrant further empirical research in the effects of functional position on the Europeanization of preferences. Overall, the research conducted in this chapter thus seems to have yielded some tentative support for the hypothesis that – whether mediated by other

144

Femke van Esch

factors or not – the more European integration, the more Europeanized the preferences of national decision makers. However, all in all this result does not seem to warrant the conclusion that the metaphor ‘perpetual motion’ that was introduced in the title is applicable to European integration (as the (neo)functionalist notion of spill-over once argued). For, a closer view at the results indicates that the level of Europeanization is not, by definition, related to the level of support for Europe. The French President Pompidou, for instance, was strongly against the establishment of a (true) EMU, but his view was more Europeanized than that of the pro-European advocate D’Estaing. Moreover, Waigel and Kohl were both very reluctant supporters of EMU, but Waigel’s preference is significantly more Europeanized that that of his Chancellor. For now, the idea of a European ‘perpetual motion’ seems as unlikely as the physical one.

APPENDIX: LIST OF SPEECHES, PRESSSTATEMENTS, INTERVIEWS AND WRITINGS USED Late 1960s George Pompidou (7) A) Entretien accorde a la revue réalités, 14 avril 1970 (reproduced in: (Pompidou 1975: T.1)) B) Conférence de presse, Paris 2 juillet 1970 C) Conférence de presse, Paris, 21 janvier 1971 (All reproduced in: (Cousté and Visine 1974)) D) Allocution prononcée a l’occasion du dîner offert a l’Elysée en l’honneur de M. Willy Brandt, Chancelier de la République Fédérale D’Allemagne, 25 janvier 1971 (Reproduced in: (Pompidou 1975: T.2)) E) Conférence de presse tenue a l’Elysée – 25 septembre 1971 (Reproduced in: (Pompidou 1975: T.2)) F) Interview télévisée, Paris, 22 décembre 1971 (Reproduced in: (Cousté and Visine 1974)) G) Déclaration a la séance d’ouverture de la Conférence des Chefs d’état ou de Gouvernement des états membres ou adhérents des communautés Européennes a Paris, 19 Octobre 1972 (Reproduced in: (Pompidou 1975: T.2)) Willy Brandt (7) A) Rede vor dem Bundesverband des Deutschen Groß- und Außenhandels am 2. Juni 1970 in Bonn B) Erklärung zur Europapolitik am 6. November 1970 im Bundestag C) Ansprache vor der Vollversammlung des Deutschen Industrie- und Handelstages am 26. Februar 1971 in Bonn D) Erklärung vor dem Bundestag in der 131. Sitzung des Deutschen Bundestages am 24. Juni 1971

Europeanization of Central Decision Makers

145

E) Interview mit Le Monde, Paris, am 6. Juli 1971 zu EWG-Fragen F) “Spiegel”-Gespräch am 27. September 1971 (All reproduced in: (Brandt 1971)) G) Ansprache auf dem 9. Ordentlichen Gewerkschaftstag der IG Druck und Papier in Nürnberg am 24. Oktober 1971 (Reproduced in: (Brandt 1973)) Valéry Giscard d’Estaing (6) A) Discours prononcé le 1er octobre 1963 devant l’assemblée annuelle du Fonds Monétaire International à Washington B) Discours prononcé le 9 Septembre 1964 aux débats annuels du Fonds Monétaire International à Tokyo C) Conférence faite le 11 février 1965 à la Maison du Droit (All reproduced in: (Giscard d’Estaing 1965: 20)) D) Discours prononcé début juin 1966 à Metz. E) Discours prononcé en Septembre 1966 à Saint-Jean-Luz au cours une conférence sur l’Europe. F) Interview par Europe No 1 en Août 1966. (All reproduced in: Le monde, 7 juin 1966, p. 8 ; 10 septembre 1966, p. 5 ; 24 août 1966, p. 14) Karl Schiller (6) A) Bemühen um Geldwertstabilität und leistungsfähige Wirtschaft, Rede zur Eröffnung der Deutschen Industrieausstellung in Berlin am 17. September 1970. B) Voraussetzungen für eine prosperierende Weltwirtschaft, Rede vor der Jahresversammlung der Gouverneure des Internationalen Wahrungsfonds und der Weltbanken in Kopenhagen am 21. September 1970. (All reproduced in: Bulletin des Presse- und Informationsamt der Bundesregierung: nr. 126, S.1300, 22 September 1970; nr. 129, S.1335, 25 September 1970) C) Verklaring voor het Europees Parlement over de economische unie en de vooruitzichten inzake een monetaire unie binnen de Gemeenschappen op 18 november 1970 – repliek. (Reproduced in: Publikatieblad van de Europese Gemeenschappen – Bijlage, november 1970) D) Aussprache zur Großen Anfrage zur Außenpolitik, Beratungen des Deutschen Bundestages am 29. Januar 1971, Rede des Bundeswirtschaftsministers. (Reproduced in: Bulletin des Presse- und Informationsamt der Bundesregierung: nr. 14, S.149, 2 Februar 1971) E) Erklärung des deutschen Bundesminister für Wirtschaft und Finanzen, Karl Schiller, auf der Sondersitzung des Rats der Europäische Gemeinschaften am 8 mai 1971 (Auszuge) (Reproduced in: Europa-Archiv, Folge 18/1971, S. D.417) F) Aktuelle Probleme der Wirtschafts- und Konjunkturpolitik, Rede auf der Öffentlichen Vollversammlung des Deutschen Industrie- und Handelstages in Bonn am 25. Februar 1972. (Reproduced in: Bulletin des Presse- und Informationsamt der Bundesregierung: nr. 29, S.499, 29 Februar 1972) Late 1980s François Mitterrand (4) A) François Mitterrand at the European College in Bruges. Held at the occasion of the opening ceremonial of the academic year of the European College on 13 October 1987 B) Europe and France in François Mitterrand’s View. A long letter to All French Citizens prior to the 1988 presidential elections April 1988. C) President Francois Mitterrand Addresses the European Parliament on 25 October 1989. (All reproduced in: Europe Documents, no. 1479, 31 October 1987; no. 1502, 14 April 1988; no. 1577, 3 November 1989)

146

Femke van Esch

D) Interview accordée par monsieur François Mitterrand, Président de la République a Madame Anne Sinclair dan le Cadre de L’émission 7/7 sur TF1, Dimanche 25 mars 1990 (Reproduced in: Mage, tome VI) Helmut Kohl (7) A) Die Rolle der Bundesrepublik Deutschland in der internationalen Politik. Rede vor dem “Chicago Council on Foreign Relations” am 23. Oktober 1986 in Chicago. (Reproduced in: (Kohl 1989)) B) Das Erbe Ludwig Erhards – Herausforderung an die Wirtschaftspolitik. Rede des Bundeskanzlers auf der Festveranstaltung zum 90. Geburtstag von Professor Dr. Ludwig Erhard am 4. Februar 1987 in Bonn. C) Für eine gemeinsame europäische Sicherheitspolitik. Ansprache bei dem Festakt zum 25. Jahrestag des Deutsch-Französischen Vertrages am 22. Januar 1988 in Paris. (Reproduced in: (Kohl 1989)) D) Europas Zukunft – Vollendung des Binnenmarktes 1992. Ansprache des Bundeskanzlers auf der Gemeinschaftsveranstaltung der deutschen Wirtschaft am 15. März 1988 in Bonn E) Zum 40. Jahrestag der Währungs- und Wirtschaftsreform. Ansprache des Bundeskanzlers in der Feierstunde der Ludwig-Erhard-Stiftung am 12. Juni 1988 in Frankfurt (All reproduced in: (Kohl 1990)) F) 40 Jahre Bundesrepublik Deutschland – Unsere Verantwortung für Deutschlands Zukunft. Rede zur Eröffnung des CDU-Kongresses >>40 Jahre Bundesrepublik Deutschland: Frieden – Freiheit – Soziale Gerechtigkeit<< am 18. Januar 1989 in Bonn. (Reproduced in: (Kohl 1992)) G) Erklärung zum Abschluss des Wirtschaftsgipfels in Paris am 16. Juli 1989 vor der Presse. (Reproduced in: Bulletin des Presse- und Informationsamt der Bundesregierung, nr. 76, S.670, 19 Juli 1989) Pierre Bérégovoy (7) A) Rede des französische Wirtschafts-, Haushalts- und Finanzministers, Pierre Bérégovoy, im Namen der Mitgliedstaaten der Europäischen Gemeinschaften am 26. September 1989 vor der Jahresversammlung des Internationalen Währungsfonds (IWF) und der Weltbank in Washington (Auszüge). (Reproduced in: Europa-Archiv, Folge 21 / 1989) B) Ce n’est pas en creusant les déficits qu’on peut créer des emplois, 8 Octobre 1991 C) Entretien avec M. Bérégovoy, 20 décembre 1989. D) de l’union européenne, à une réunion publique sur le traité de Maastricht organisée à Nevers par le Mouvement européen, 13 juin 1992. E) La ratification du traité aidera à consolider la monnaie, Bérégovoy devant le journal d’Antenne 2, 16 juin 1992. F) Nul n’a l’intention de s’approprier le choix que nous recommanderons aux Français, déclaration Bérégovoy dans la réunion du Congrès à Versailles, 23 juin 1992. G) En matière de sécurité, la France sera plus forte avec ses partenaires que toute seule, déclaration devant les stagiaires civils et militaires de l’institut des hautes études de défense nationale (IHEDN), 3 septembre 1992. (All reproduced in: le monde: 9 octobre 1991, p.30; 20 décembre 1989, p.1&11; 16 juin 1992, p.11; 18 juin 1992, p.9; 24 juin 1992, p.7; 7 septembre 1992, p.12) Theo Waigel (5) A) Wohin steuern die Unionsparteien? Christlich-soziale Grundwerte vor dem aktuellen politischen Hintergrund, März 1988. (In: Politische Studien, März 1988) B) Keine Währungs-Union ohne Geldwert-Stabilität, 18 August 1989. (In: Rheinischer Merkur, 18 Augustus 1989, S. 13.)

Europeanization of Central Decision Makers

147

C) Rede in Berlin anlässlich des Vorstandstreffens der Unternehmen mit Bundesbeteiligungen am 10. Oktober 1989. D) Rede in 613. Sitzung des Bundesrates am 22. Mai 1990 anlässlich der ersten Beratung des Gesetzes zum Staatsvertrag. E) Rede anlässlich des Deutschen Steuerberatertages 1990 am 5. November 1990 in Köln. (All reproduced in: Bulletin des presse- und Informationsamt der Bundesregierung: nr. 104, 12 Oktober 1989; nr. 66, 24 Mai 1990; nr. 129, 8 November 1990)

REFERENCES Alderson, K., 2001, Making sense of state socialization, Review of International Studies 27:415-33. Allison, G. T., and Zelikow, P., 1999, Essence of Decision. Explaining the Cuban Missile Crisis, 2e, Longman, New York. Axelrod, R., 1976, Structure of Decision. The Cognitive Maps of Political Elites, Princeton University Press, Princeton. Barnett, M. N., and Finnemore, M., 1999, The politics, power, and pathologies of international organisations, International Organization 53(4):699-732. Bonham, G. M., and Shapiro, M., 1976, Explanation of the unexpected: the Syrian intervention in Jordan in 1970, in: Structure of Decision. The Cognitive Maps of Political Elites, R. Axelrod, ed., Princeton University Press, Princeton, pp. 113-41. Bonham, G. M., and Shapiro, M., 1977, Foreign policy decision-making in Finland and Austria: the application of a cognitive process model, in: Thought and Action in Foreign Policy. Proceedings of the London Conference on Cognitive Process Models of Foreign Policy, March 1973, G. M. Bonham and M. Shapiro, eds, Birkhäuser Verlag, Basel, pp. 306-55. Brandt, W., 1971, Bundeskanzler Brandt. Reden und Interviews, Hoffmann und Campe, Hamburg. Brandt, W., 1973, Reden und Interviews. Herbst 1971 bis Frühjahr 1973, Hoffmann und Campe, Hamburg. Bueno de Mesquita, B., 2000, Principles of International Politics. People’s Power, Preferences, and Perceptions, CQ Press, Washington DC. Checkel, J. T., 1998, Review article. The constructivist turn in international relations theory, World Politics 50:324-48. Checkel, J. T., 2001, Why comply? Social learning and European identity Change, International Organization 55(3):553-88. Cochran, C. E., 1973, The politics of interest: philosophy and the limitations of the science of politics, American Political Science Review 17:745-66. Connolly, W. E., 1972, On ‘interests’ in politics, Politics and Society 2:459-77. Cousté, P. B., and Visine, F., 1974, Pompidou et l’Europe, Librairies Techniques, Paris. Etheredge, L. S., 1981, Government learning. An overview, in: The Handbook of Political Behavior, S. L. Long, ed., Plenum Press, New York, pp. 73-147. Finnemore, M., 1996, National Interests in International Society, Cornell University Press, Ithaca. Finnemore, M., and Sikkink, K., 1998, International norm dynamics and political change, International Organization 52(4):887-917. Frieden, J. A., 1999, Actors and preferences in international relations, in: Strategic Choice and International Relations, D. A. Lake and R. Powell, eds, Princeton University Press, Princeton, pp. 39-76.

148

Femke van Esch

Giscard d’Estaing, V., 1965, Exposés de M. Valéry Giscard d’Estaing sur les Problèmes Monétaires Internationaux, Paris. Goldsmith, B., 2000, Observational Learning and the Formation of Foreign Policy Preferences, paper for Annual Meeting of the American Political Science Association, Marriott Wardman Park, Washington DC, 31 August - 3 September 2000. Goldstein, J., and Keohane, R. O., 1993, Ideas and Foreign Policy : Beliefs, Institutions, and Political Change, Cornell University Press, Ithaca. Gourevitch, P., 2002, Domestic politics and international relations, in: Handbook of International Politics, W. Carlsnaes, T. Risse and B. Simmons, eds, Sage Publications, London, pp. 309-28. Guzzini, S., 2000, A Reconstruction of Constructivism in International Relations, European Journal of International Relations 6(2):147-82. Halperin, M. H., 1974, Bureaucratic Politics and Foreign Policy, The Brookings Institution, Washington D.C. Haverland, M., 1999, National Autonomy, European Integration and the Politics of Packaging Waste, Netherlands School for Social and Economic Policy Research, University of Utrecht, Utrecht. Heradstveit, D., and Narvesen, O., 1978, Psychological constraints on decision-making, discussion of cognitive approaches: operational code and cognitive map, Cooperation and Conflict. Nordic Journal of International Relations 13:77-92. Hug, S., 2003, Endogenous preferences and delegation in the European Union, Comparative Political Studies 36(1/2):41-74. Johnston, A. I., 2001,Treating international institutions as social environments, International Studies Quarterly 45:487-515. Keohane, R. O., and Nye, J. S., 1987, Power and interdependence revisited, International Organization 41(4):725-53. King, G., Keohane, R. O., and Verba, S., 1994, Designing Social Inquiry. Scientivic Inference in Qualitative Research, Princeton University Press, Princeton. Kohl, H., 1989, Reden zu Fragen der Zukunft, Presse- und Informationsamt der Bundesregierung, Bonn. Kohl, H., 1990, Reden zu Fragen der Sozialen Marktwirtschaft, Presse- und Informationsamt der Bundesregierung, Bonn. Kohl, H., 1992, Die Deutsche Einheit. Reden und Gespräche, Gustav Lübbe Verlag, Bergisch Gladbach. Krasner, S. D., 1978, Defending the National Interest. Raw Materials Investment and U.S. Foreign Policy, Princeton University Press, Princeton. Kratochwill, F., 1982, On the notion of “interest” in international relations, International Organization 36(1):1-30. Levy, J. S., 1994, Learning and foreign policy: sweeping a conceptual minefield, International Organization 48(2):279-312. Lieshout, R. H., 1995, Between Anarchy and Hierarchy: A Theory of International Politics and Foreign Policy, Edward Elgar, Aldershot. Little, R., 1988, Belief systems in the social sciences, in: Belief Systems and International Relations, R. Little and S. Smith, eds, Basil Blackwell, Oxford, pp. 37-56. McNamara, K. R., 1998, The Currency of Ideas: Monetary Politics in the European Union, Cornell University Press, Ithaca, NY. Moravcsik, A., 1993, Preferences and power in the European Community: a liberal intergovernmental approach, Journal of Common Market Studies 31(4):473-524. Moravcsik, A., 1997, Taking preferences seriously: a liberal theory of international politics, International Organization 51(4):513-53. Moravcsik, A., 1998, The Choice for Europe: Social Purpose and State Power from Messina to Maastricht, Cornell University Press, Ithaca, NY.

Europeanization of Central Decision Makers

149

Pompidou, G., 1975, Entretiens et Discours 1968-1974. Tome I et II., Librairie Plon, Paris. Price, R., and Reus-Smit, C., 1998, Dangerous liaisons? Critical international theory and constructivism, European Journal of International Relations 4(3):259-94. Princen, S. B. M., 2002, The California Effect in the Transatlantic Relationship, Universiteit van Utrecht, Utrecht. Risse, T., 2001, A European identity? Europeanization and the evolution of nation-state identities, in: Transforming Europe. Europeanization and Domestic Change, M. Green Cowles, J. Caporaso, T. Risse, ed., Cornell University Press, Ithaca NY, p. 198216. Risse, T., 2002, Constructivism and international institutions, in: Political Science:The State of the Discipline, I. Katznelson and H. V. Milner, eds, W.W. Norton & Company, New York, pp. 597-623. Risse, T., Engelmann-Martin, D., Knopf, H. J., and Roscher, K., 1998, To Euro or not to Euro? The EMU and identity politics in the European Union, EUI Working Papers 98/9, European University Institute, Robert Schuman Centre, Florence. Risse, T., Engelmann-Martin, D., Knopf, H. J., and Roscher, K., 1999, To Euro or not to Euro? The EMU and identity politics in the European Union, European Journal of International Relations 5(2):147-87. Scholtens, B., 1990, On the Viability of an Economic and Monetary Union within the European Community: a Comparison between the Delors Report and th e Werner Report, Department of Economics, University of Amsterdam, Amsterdam. Smith, S., 1988, Belief systems and the study of international relations, in: Belief Systems and International Relations, R. Little and S. Smith, eds, Basil Blackwell, Oxford, pp. 11-36. Van Esch, F. A. W. J., 2001, Defining national preferences: the influence of international nonstate actors, in: Non-State Actors in International Relations, B. Arts, M. Noortmann and B. Reinalda, eds, Ashgate, Aldershot, pp. 109-26. Wall, G., 1975, The concept of interest in politics, Politics and Society (5):487-510. Weldes, J., 1996, Constructing the national interest, European Journal of International Relations 2(3):275-318.

DOMESTICATED WOLVES?

Length of Membership, State Size and Preferences at the European Convention Dirk Leuffen and Sander Luitwieler Abstract:

1.

This chapter analyzes whether sociological or rational institutionalism can better explain governmental preferences in the European Union (EU). Specifically, we are interested in how length of membership impacts on EU member state preferences for future integration. Drawing on literature about the two institutionalisms, we develop two theories of preference formation that we test for five prominent issues at the European Convention: the size of the Commission, the definition and extension of QMV, the extension of codecision, and the organization of the Council presidency. In addition, we construct a variable ‘Citizen Representation’ that combines the individual issues. We present our data using cross tables before estimating linear probability models by OLS. Our quantitative analysis shows that sociological institutionalism is better able to account for preferences at the Convention than rational institutionalism. The dominance of the division between small and large member states, which was so often evoked during the Convention, is not supported by our findings.

INTRODUCTION1

The ‘third wave’ of European integration studies focuses on the interaction between top-down and bottom-up processes of European integration. How does the European Union (EU) impact on its member states and how does that, in turn, change the development of the EU over the longer term? In order to better understand such intertwined processes of

1

The authors are particularly grateful to Jean-Marie Jungblut and Thomas Gschwend, who gave us extremely helpful advice. We also wish to thank Kees Aarts, Harmen Binnema, Matteo Foschi, Bob Lieshout, Francesco Marchi, Frank Schimmelfennig, Arndt Wonka, as well as the members of the European Research Colloquium for their suggestions. Of course, we take on the full responsibility for any mistake that may still be in the paper.

151

R. Holzhacker and M. Haverland, (eds.), European Research Reloaded: Cooperation and Integration among Europeanized States, 151-178. © 2006 Springer. Printed in the Netherlands.

152

Dirk Leuffen and Sander Luitwieler

change, this chapter analyzes whether and how socialization matters for EU member state preference formation concerning questions of future integration. Drawing on literature about sociological and rational institutionalism, we develop two theories of preference formation that we test on data on the European Convention. The Convention, as the most recent instance of preparing for EU Treaty reform, to us seemed a particularly well-suited case to study the impact of socialization on preference formation. This is because of the high number of participating countries and the high variance of length of membership (our initial explanatory variable as to socialization), ranging from 52 years for the founding members of the European Coal and Steel Community (ECSC) to 0 years for the then candidate countries. At the same time, since it was about shaping the future institutional features of the EU of 25 member states, the Convention touched highly sensitive issues such as national sovereignty. Usually, in such cases rational institutionalist approaches are expected to account best for explaining preferences and, indeed, many commentaries, for example, stressed a division between small and large member states at the Convention (e.g. Magnette and Nicolaidis 2004; Allen 2004, 21). Therefore, if sociological institutionalism had something to add in this case, this would certainly underline the importance of this strand of literature. How do we proceed in the chapter? We first develop our two theories: from a sociological institutionalist approach we argue that the longer a country is a member of the EU, the more it will be socialized into certain EU norms and values, and the more these norms and values will be reflected in its preferences for European integration in general and the institutional design of the EU in particular. From a rational institutionalist approach, on the other hand, we argue that state size should best be able to account for member state preferences for future European integration. From these theories we generate hypotheses, which we test on five prominent issues that were debated at the Convention: the size of the Commission, the definition and extension of qualified majority voting (QMV), the extension of codecision, and the organization of the Council presidency. In addition, we construct a variable ‘Citizen Representation’ that combines the individual issues. Concerning the individual issues we present our data using cross tables before estimating linear probability models by OLS applying a variance correction. For the ‘Citizen Representation’ variable we only estimate a linear probability model by OLS.

Domesticated Wolves?

2.

153

NEW INSTITUTIONALIST APPROACHES TO PREFERENCE FORMATION

This volume adopts a top-down perspective to Europeanization, defining it as the effects of European integration on the EU member states. These effects can be traced concerning domestic structures and public policy, but also concerning what Radaelli (2003, 35) calls the ‘cognitive and normative structures’ of the member states. Radaelli subsequently argues that such effects on the norms, values and identities of the member states may in turn ‘transform the interests and preferences upon which [European] negotiations are structured’ (Radaelli 2003, 36; our emphasis). This, of course, would then feed back into future European integration. By analyzing how EU socialization affects member state preferences at the European Convention our chapter precisely takes issue with such considerations. Theoretically, the perspective of sociological institutionalism can capture well the Europeanization of member state identities and preferences. In sociological institutionalism, EU norms and values are considered as ‘institutions’ that affect the identities and behavior of the EU member states. More specifically, ‘institutions’ prescribe which behavior is considered appropriate for members of the institution (the so-called ‘logic of appropriateness’; March and Olsen 1984, 1989, 1998; Peters 1999, 25-42; Aspinwall and Schneider 2001; Kohler-Koch 2002, 4). The agents internalize the institutions to such an extent that they are not able to act independently of them. We thus assume that institutions first impact on member state identities, which then subsequently changes their preferences. The main point is that identities and preferences are not exogenously formed but rather endogenously: the construction of member state identities and preferences cannot be understood without taking into account processes of EU socialization. While we thus theoretically recognize that the socialization of identities precedes the socialization of preferences, empirically our chapter will be confined to the analysis of preferences (cf. Hooghe 2001; KoenigArchibugi 2004). EU socialization is brought about through dense and continuous interaction between member state and EU actors within and across the multiple levels of the EU’s political system. The interaction results in the participating actors internalizing specific norms and values that lie at the basis of and are central to the European integration project. Among these EU norms we consider most important for our analysis the norm of supranational rather than intergovernmental cooperation and decision-making, particularly the supranational community method, and the norm of far-reaching, deeper integration into (ever) more policy areas rather than limited integration or even the re-nationalization of certain policy areas.

154

Dirk Leuffen and Sander Luitwieler

Following a sociological institutionalist approach, we consider length of membership to be the crucial variable explaining member state preferences. An old member state is likely to be socialized into EU norms and values to a greater extent than a new(er) member state, as an old member state has been exposed to the EU norms and values for a longer time period. In general, we therefore hypothesize that the older the member state, the more the EU norms and values mentioned are reflected in its preferences on European integration in general and on the institutional design of the EU in particular.2 Rational institutionalism is generally seen as the natural counterpart of sociological institutionalism (Peters 1999, 43-62; Aspinwall and Schneider 2001) and only recently the possible complementarity of the two approaches has been more prominently put to the fore (Börzel and Risse 2003; KoenigArchibugi 2004). Rather than a theory of institutions, rational institutionalism is a theory about how actors use institutions to achieve their goals. More specifically, it assumes that rational actors use institutions to maximize their utility, i.e. their (economic) self-interest. Institutions not only constrain actors in their strategic behavior, as rational institutionalism would expect, but they can also themselves be understood as the object and outcome of strategic action (Koenig-Archibugi 2004, 142). In contrast with sociological institutionalism, the interests and preferences of actors are exogenously rather than endogenously formed. As Hooghe (2001, 155) puts it, ‘preferences constitute a function of their utility to a priori material goals’. While sociological institutionalism stresses a logic of appropriateness, rational institutionalism depicts a logic of expected consequences (March and Olsen 1998). Applied to the institutional preferences of member states during the Convention, a rational view implies that a member state adopts preferences that in the case of its realization would result in the best formal position, or most formal power within a concrete EU institution or decisionmaking procedure, so as to achieve its (material) goals and interests within the EU. We therefore argue that large member states have an interest in their population being reflected as much as possible in the design of the EU institutions and decision-making procedures, whereas small member states have an interest in being equally treated as much as possible. To operationalize this argument, we expect rational, utility maximizing behavior to express itself so that the principle of citizen equality is reflected

2

As such, ‘length of membership’ reflects a more general ‘scope condition’ for socialization to be likely to happen (cf. Trondal 2004). In taking a macro-level, long-term perspective to EU socialization, we are however not in a position to specify and test several other scope conditions that scholars recently have formulated for and applied in micro-level, more shortterm analyses (e.g. Checkel 1999, 2001; Trondal 2002, 2004).

Domesticated Wolves?

155

in the preferences on institutional design of large member states, whereas the principle of member state equality is reflected in the preferences of small member states. One could expect this rational behavior to be particularly invoked and strengthened by the shadow of future horizontal integration, with the then current member states but also candidate countries anticipating the (effects of) enlargement of the Union up to 25 member states from 1 May 2004. Whereas in a sociological institutionalist approach the explanatory variable is length of membership, here the explanatory variable is state size. In general, we therefore hypothesize that the larger the member state, the more the principle of citizen equality (rather than member state equality) is reflected in its preferences on European integration and on the institutional design of the EU.

3.

THE CONVENTION ON THE FUTURE OF THE EUROPEAN UNION

3.1

Introduction

Though the Treaty of Nice contained institutional reforms to prepare the Union for enlargement, the heads of state and government already in Nice decided to engage another Intergovernmental Conference (IGC) in 2004 in order to reform the Union more radically. In a declaration attached to the Treaty they called for a ‘deeper and wider debate about the future of the European Union’. The Laeken declaration, adopted by the European Council of 14 and 15 December 2001, stated that a Convention on the Future of the European Union would be installed to prepare the next IGC. The declaration formulated a considerable amount of detailed questions the Convention should address and which were related to the following four themes: - The division and definition of competence in the EU. - The simplification of the Union’s instruments. - Increasing democracy, transparency and efficiency concerning the EU institutions and decision-making and the role of national parliaments. - The simplification and reorganization of the Treaties, possibly resulting in the adoption of a ‘constitutional text’ for the Union including the question of the status of the Charter of Fundamental Rights proclaimed in Nice. In addition to chairman Valéry Giscard d’Estaing and vice-chairmen Guilliano Amato and Jean-Luc Dehaene, the Convention was composed of 15 governmental representatives (one per member state), 30 representatives of national parliaments (two per member state), 16 representatives of the European Parliament and two representatives of the European Commission.

Dirk Leuffen and Sander Luitwieler

156

The then candidate countries were in the same way represented as the member states, though they were not attributed the right to veto a developing consensus amidst the EU-15. On 18 July 2003 chairman Giscard finally presented a draft Treaty establishing a Constitution for Europe to the President of the European Council (CONV 850/03), which subsequently formed the basis for the IGC negotiations that started in October 2003 under the Italian presidency. We selected the Convention as our case to test the two institutionalist explanations of preference formation for several reasons. First, it represents the most recent instance of (preparing for) EU Treaty reform. While the greater part of the literature on the subject is descriptive (e.g. Allen 2004), we aim to offer an analytical contribution. Second, the Convention is the first instance in which the preferences of 25 countries can be analyzed, as all of the 10 then candidate countries issued contributions and participated (more or less) fully in the negotiations, whereas they only had an observer status during the Nice IGC. Third, in preparing the IGC and by the very design and idea of this body, the Convention could be expected to display a less intergovernmental and strategic mode of negotiating. We expected this to facilitate the recording of the ‘original’ member state preferences, in comparison with the subsequent IGC. The great openness and transparency of the Convention furthermore facilitated tracking down the preferences of the member states.

3.2

Issues and hypotheses

Though many different issues were examined by the Convention, we decided to select five institutional issues for our analysis following two criteria. First, we decided to analyze those issues that were most prominent in the debate on the draft constitutional Treaty. This qualitative selection is based on an analysis of primary and secondary literature on the Convention.3 Secondly, and related to this, most countries should have expressed a position on an issue. This criterion was necessary because we planned to horizontally compare the different state preferences.4

3 Initially, we had also identified the creation of an EU Foreign Minister or External Representative as an important issue. The crucial questions relating to this issue are, however, closely linked with the other selected issues, especially the organization of the Council presidency, but also the extension of QMV. 4 We are aware of the problem that our data might thus be biased by a pre-selection of issues, which powerful actors in the Convention, such as the Presidium or certain member states might have exercised. If a country, for example, has made an extremely integrationist proposal, which is however supported by nobody else, we cannot account for this in our research design.

Domesticated Wolves?

157

After identifying the most prominent issues we constructed the response items. Although we had at first constructed a list of various items for each issue, we discovered during our data collection that mechanisms of prestructuring must have taken place in the Convention, since only a few options were actually defended by the representatives. We therefore realized that we were able to classify the positions dichotomously without loosing a lot of information and at the same time facilitating the analysis of our data. We coded the data in a way that item 0 always indicates a more state equality oriented position, whereas item 1 is more oriented towards citizen equality. The coding for the individual issues is revealed in the following. Size of the Commission Though a compromise on the size and composition of the European Commission was agreed upon in Nice, the issue came again to the fore in the course of the Convention negotiations. The main options were a reduced Commission composed of one Commissioner per member state (coded as 0) or a reduced Commission below the (future) number of member states, for example to 12 or 15 Commissioners (coded as 1). From a sociological institutionalist perspective, we expect old member states to be in favor of a supranational outlook of the Commission in which not every member state necessarily needs to be represented, and new member states to favor an intergovernmental perspective toward the Commission in which each member state is represented. From a rationalist point of view, we expect small member states to prefer an equal representation with one Commissioner per member state. We expect large member states, on the other hand, to be more inclined to favor a reduced Commission below the number of member states. Definition of QMV The triple majority system defining QMV in the Council, which was decided on in Nice and made Council decision-making more complex, was also renegotiated during the Convention. The main options were maintaining the status quo of Nice (coded as 0) or introducing some kind of double majority system (50% of states + 50% or 60% of population; coded as 1), in which the population criterion weighs more heavily and which represents a more supranational mode of decision-making. From a sociological institutionalist viewpoint, we expect old member states to favor a more supranational decision-making system, which facilitates deeper integration as it is more efficient, and new member states to favor a more intergovernmental decision-making system (the status quo), which limits the possibilities for far-reaching integration into more policy areas. According to a rationalist approach, we expect small member states to favor a solution in which member state equality, instead of the population

158

Dirk Leuffen and Sander Luitwieler

criterion (citizen equality), is mostly reflected and large member states to favor a solution in which the population criterion, instead of member state equality, is mostly reflected. Thus the new and small states should prefer the higher decision-making barriers of Nice and the old and large states should prefer to reduce the veto possibilities. Extension of QMV After QMV was extended in Amsterdam and Nice to only a limited degree and hardly to more sensitive policy areas, another discussion on the issue inevitably popped up during the Convention. The main points at stake were whether to extend QMV to sensitive policy areas, in particular taxation and/or the Common Foreign and Security Policy (CSFP) (coded as 1) or not to extend QMV to these fields (coded as 0). In accordance with a sociological institutionalist approach, we expect old member states to be in favor of extending QMV to taxation and/or CFSP as part of the supranational community method and facilitating deeper integration of these policy areas in the future, and new member states not to be in favor of extending QMV to these fields. From a rational point of view, we expect small member states not to be in favor of the extension of QMV to taxation and/or CFSP, as the status quo of Nice (triple majority) and even more the possible new Council voting rules (a kind of double majority) are to the disadvantage of the small member states. Large member states, on the other hand, should rather be in favor of extending QMV to these policy areas as they profit most from the integration of the population criterion in the current triple majority system, and even more so in a double majority system. Extension of co-decision Related to the extension of QMV is the issue of the extension of the codecision procedure, the main options being no or only a partial extension of the co-decision procedure (coded as 0), or a general extension of co-decision to all legislative acts on which the Council decides by QMV (coded as 1). From a sociological institutional viewpoint, we expect old member states to favor a general extension of the co-decision procedure as it is in line with a supranational mode of decision-making, and new member states to be not in favor of extending co-decision, or only to a limited degree since they are expected to prefer a more intergovernmental mode of decision-making. According to a rational approach, we expect small member states not to be in favor of further extending the co-decision procedure, or only partially, as they have fewer seats in the European Parliament than the large member states and therefore constantly run the risk of being outvoted. For the opposite reason, we expect large countries to be in favor of a general extension of the co-decision procedure.

Domesticated Wolves?

159

Organization of the Council presidency The last issue concerns the organization of the presidency of the European Council and of the Council of the EU. The crucial distinction between the alternatives being proposed during the Convention was whether the element of rotation (coded as 0) or the element of permanency (coded as 1) is dominant in the organization of one or both of the Councils. From a sociological institutionalist point of view, we expect old member states to be in favor of an option in which the element of permanency is dominant, facilitating decision-making as it is more efficient, and new member states to favor an option in which the element of (equal) rotation is dominant. In accordance with a rational approach, we expect small member states to favor a solution of this issue in which the element of (equal) rotation is dominant (in one or both of the Councils), whereas we expect large member states to favor a solution in which the element of permanency is dominant.

4.

DATA AND MEASUREMENT

One of the declared objectives of the Convention was transparency. In order to enable civil society to observe what was being discussed in the Convention, most interventions, contributions and other documents were published on the Internet. That, indeed, proved to be very helpful for our research. But how exactly did we identify the preferences of the now 25 EU member states concerning the institutional future of the Union? As stated earlier, we expected the Convention to be a context with less intergovernmental and strategic negotiating than the subsequent IGC, making it relatively well suited to trace the original preferences of the participating countries. In order to collect the most ‘unspoilt’ preferences, we opted to investigate the positions member states held in as early a stage of the Convention as possible, since the closer the follow-up IGC comes, the more strategically actors can be expected to behave.5 One way to collect this information would have been to question government representatives or other insiders. Since this research was, however, only executed in the spring of 2004 we feared the risk of being trapped by ex-post rationalizations of potential informants. We decided instead to do content analysis of primary sources, such as the contributions to the Convention of the different government representatives, but also speeches of Prime Ministers or Foreign Ministers at other occasions.6 Most of this data was accessible via the Internet on official

5 6

Cf. Schimmelfennig 2003, 178, footnote 25. For a similar approach, see Koenig-Archibugi 2004, 152.

160

Dirk Leuffen and Sander Luitwieler

homepages, such as of the European Convention7 and of national Ministries of Foreign Affairs. We furthermore compared our data with the information provided in the first EUCON report, which has also documented member state preferences for an early stage of the Convention.8 The governmental positions that we report in this article thus refer to late 2002 and early 2003.9 The government representatives did not always clearly list their preferences concerning our issues. If that was the case we had to analyze various contributions and additional documents. Sometimes it remained difficult to find out which solution a government actually preferred or whether it was completely indifferent on an issue. When we were not able to define a position we coded this unit as missing. There are, however, only four missings in our data set. Often government representatives were signatories of quite general, common contributions. Not always, however, did all signatories agree on all the aspects of these common papers. To give an example, the contribution ‘Premises and Principles of EU Institutional Reform’ (CONV 590/03, CONTRIB 263), submitted on 28 February 2003 by several members of the Convention, is signed by the Maltese governmental representative Serracino-Inglott. Serracino-Inglott disagreed, however, concerning the double Council majority proposed on page 3 of this contribution as stated in footnote 1 on the same page. In his intervention during the Convention’s plenary session of 15 and 16 May 2003, SerracinoInglott came again to speak about this issue and underlined the principle of equality between states to be of great importance to Malta.10 On the basis of this information we concluded that Malta should be classified as 0 concerning issue 2.11 The following table reports the state preferences we identified for our five issues. In addition, we supply the length of membership of each country calculated for January 2003. The founder states of the ECSC in 1951, for example, have a length of membership of 52 years, the applicant countries a membership of 0 years. The last column reflects the population size of each country, rounded to 0.5 million.

7

http://european-convention.eu.int/; last accessed on 13 December 2004. http://eucon.europa2004.it/Watch1ed/Questions.htm; last accessed on 13 December 2004. 9 Misfortunately, we could not avoid slight differences concerning the moments of observation in our design. 10 http://www.foreign.gov.mt/pr/docsgov/2003/pr030519c-European_Convention_ Intervention -doi599.htm; last accessed on 22 May 2004. 11 For reasons of space, we cannot cite all the sources we used for our analysis. Therefore, for further information on the data collection please contact the authors. 8

161

Domesticated Wolves? Country AU B CYP CZ D DK ESP EST FIN FR GR HU IRL IT LAT LIT LUX MAL NL POL POR SK SLO SWE UK

Issue 1 0 1 0 0 1 1 0 0 0 1 0 0 0 1 0 0 0 0 0 0 0 0 0 0 1

Issue 2 0 1 1 1 0 0 0 1 1 1 1 0 0 1 0 1 0 0 1 1 0 0

Issue 3 1 1 1 0 1 0 0 1 1 1 1 0 0 0 0 0 1 0 1 0 0 0 0 1 0

Issue 4 1 1 0 1 1 1 1 0 0 1 1 1 0 1 1 1 1 1 1 1 1 1 0 0

Issue 5 0 0 0 0 1 1 1 0 0 1 0 1 0 1 0 0 0 0 0 0 0 0 0 1 1

Membership 8 52 0 0 52 30 17 0 8 52 22 0 30 52 0 0 52 0 52 0 17 0 0 8 30

Population 8 10 0.5 10 82 5.5 39.5 1.5 5 59 10.5 10 3.5 57.5 2.5 3.5 0.5 0.5 15.5 38.5 10 5.5 2 9 59

Table 1. preferences of 25 EU member states on 5 issues at the European Convention, as well as their length of membership and population size (numbers indicate the preferred items; 4 missings reported). Sources: see text

5.

DOMESTICATED WOLVES? EXPLAINING MEMBER STATE PREFERENCES AT THE CONVENTION

How do length of membership and state size relate to the preferences countries held at the Convention on the future of the EU? We will now take a look at how these two variables impacted on the state preferences on the

162

Dirk Leuffen and Sander Luitwieler

different issues. Then we will introduce a variable ‘Citizen Representation’, showing for each country the sum of its preferences divided by the number of issues for which answers were reported. For this continuous variable ‘Citizen Representation’ (possible values lie between 0 and 1) we estimated an OLS regression. To facilitate the interpretation of the individual issues we first present cross tables. The cross tables contribute to better explore the data, but at the same time they introduce a loss of variance since they demand the construction of categories. To compensate for the loss of variance in the cross tables, we finally test our hypotheses on the original data by linear probability models by OLS. Since our dependent variables for the individual issues are dichotomous, we applied a variance correction.12 For the cross tables we had to construct categories of the variables length of membership and state size. Concerning length of membership we decided to construct 5 categories according to the different EU enlargement rounds. Coded 0 are the ten new member states – Cyprus, Czech Republic, Estonia, Hungary, Latvia, Lithuania, Malta, Poland, Slovakia and Slovenia; coded 1 are Austria, Finland and Sweden; coded 2 are Greece, Portugal and Spain; coded 3 are Denmark, Ireland and the United Kingdom; coded 4 are the six founding states Belgium, France, Germany, Italy, Luxembourg and the Netherlands. This variable is thus scaled ordinally. Underlying this categorization is the expectation that the longer a country adheres to the EU, the more it should be socialized into EU norms and values. Building categories for population size was more complicated. Are Poland and Spain, for example, large or small countries? When we take the population of the Union’s largest country, Germany, and divide that by two, Poland and Spain should be classified with the small countries. When drawing the line by dividing the countries into two camps, Sweden becomes the median and Belgium, the Czech Republic, Greece, Hungary, the Netherlands and Portugal are classified with the large member states. The population differences, however, remain very high amongst the countries in this potential category. Because of these difficulties we decided to construct four categories. Coded 0 are the eight countries that have a population of below five million inhabitants (‘small’) – Cyprus, Estonia, Ireland, Latvia, Lithuania, Luxembourg, Malta and Slovenia; coded 1 are the eleven countries with a population between five and 15.5 million inhabitants

12

Cf. Aldrich and Nelson 1984, 9-22. Since the number of cases was at most 25 (for issue 2 we had only 22 cases and for issue 4 only 24), we could not rely on logistic regressions. We controlled our findings, however, by estimating logits and probits. The results clearly reflect our findings, but are not reported in this chapter. The small n, of course, always remains a handicap, which we, however, could not circumnavigate.

163

Domesticated Wolves?

(‘medium small’) – Austria, Belgium, the Czech Republic, Denmark, Finland, Greece, Hungary, the Netherlands, Portugal, Slovakia and Sweden; coded 2 are Poland with 38.5 million and Spain with 39.5 million inhabitants (‘medium large’); coded 3 are the four largest countries in the EU (‘large’) – France, Germany, Italy and the United Kingdom. Issue 1 Issue 1 concerns the size of the European Commission. The two main options were a reduced Commission to the number of member states (0) or a Commission reduced below the number of member states (1). Our hypotheses expected larger and older member states to prefer solution 1 and smaller and newer member states to prefer solution 0. Let us first take a look at how the preferences on the Commission issue are related to population size. We find that none of the smallest states desires the Commission to be reduced below the number of member states. This is, on the other hand, exactly what all the four largest member states seem to prefer. Whereas the two medium large states, Poland and Spain, tend towards the one Commissioner per member state solution, two of the eleven medium small states, Belgium and Denmark, prefer a more reduced Commission provided that a system of equal rotation is introduced.13

Population 0 1 2 3 Total

Issue 1 0 8 9 2 0 19

1 0 2 0 4 6

Total 8 11 2 4 25

Table 2. population size and Commission size preference Pearson chi2(3) = 16.0287 Pr = 0.001 Cramer’s V = 0.8007 The chi2 associated with our table has 3 degrees of freedom and is 16.03. The observed differences are quite significant. Cramer’s V mounts to 0.8007

13

See Prime Minister Verhofstadt’s speech ‘Montesquieu and the European Union’, held at the College of Europe on 18 November 2002 and Prime Minister Rasmussen’s speech ‘Danish EU Policy after the Presidency’, held at the Danish Institute for International Studies on 15 January 2003. Links to both speeches can be found at: http://europa.eu.int/futurum/ congov_en.htm; last accessed on 13 December 2004.

Dirk Leuffen and Sander Luitwieler

164

and thus suggests an almost perfect correlation.14 State size indeed seems to matter, but we clearly have to see that drawing a line between our categories two and three seems to be important. How is length of membership related to the Commission issue? Concerning our five categories, we find that only countries belonging to categories three and four, i.e. the elder member states, seem to prefer a Commission size reduced below the number of member states. Of the founding member states, only the Netherlands and Luxembourg seem opposed to this idea. Concerning the three countries of the first enlargement group, only Ireland is not ready to temporarily loose its Commissioner. Denmark as well as the UK prefer a Commission below the number of member states.

Membership 0 1 2 3 4 Total

Issue 1 0 10 3 3 1 2 19

1 0 0 0 2 4 6

Total 10 3 3 3 6 25

Table 3. length of membership and Commission size preference Pearson chi2(4) = 14.0351 Pr = 0.007 Cramer’s V = 0.7493 The chi2 associated with this table has 4 degrees of freedom and is 14.04. Again the differences are significant. A Cramer’s V of 0.7493 suggests a strong correlation between length of membership and member state preferences for the Commission’s size. In order to more clearly compare the impact of our two independent variables, we decided to estimate a linear probability model by OLS with variance correction on our original data set, with population size and length of membership being treated as continuous variables. When only applying this technique to state size, we indeed find that population is significant. When we, however, control for length of membership, this variable turns out to have a stronger impact on the Commission size issue. Population size remains significant, but when adding length of membership to the model it turns out to be slightly less important.

14

Cramer’s V is a correlation coefficient ranging from 0.0 (no correlation) to 1.0 (perfect correlation).

Domesticated Wolves?

165

Dependent Variable Commission size preference (1) (2) Ind. Var. Coef. Std. Err. p Coef. Std. Err. Population .0120589 .0007362 0.000 .0037735 .0015831 Membership .0135012 .002373 Constant .0129762 .0508389 0.801 -.016226 .0100855 R-squared 0.9210 0.9864 N 25 25

p 0.026 0.000 0.122

Table 4. linear probability model by OLS with variance correction for issue 1. First only for population size (1), then for population size and length of membership (2) Our data thus confirm both hypotheses for issue 1. Length of membership as well as population size impact on the preferences countries held towards the size of the Commission. The coefficients are positive: thus, the elder and larger a member state, the more likely it prefers a reduced Commission below the number of member states. Issue 2 Issue 2 is about the definition of QMV. In our hypotheses we expected larger and elder states to be in favor of departing from the Nice solution and of reducing the necessary majorities. The smaller and newer member states, on the other hand, were expected to rather tend towards the compromise elaborated at Nice. In the following tables the status quo of Nice is coded as 0, while 1 stands for a dual majority option. Table five shows the results for issue 2 relating to population size:

Population 0 1 2 3 Total

Issue 2 0 3 4 2 2 11

1 2 7 0 2 11

Total 5 11 2 4 22

Table 5. population size and definition of QMV preference (three missings reported in category 1) Pearson chi2(3) = 3.0182 Pr = 0.389 Cramer’s V = 0.3734

Dirk Leuffen and Sander Luitwieler

166

Our findings concerning issue 2 are less clear. The only clear category is category 2, which however suffers from its small n. Both Poland and Spain prefer the Nice solution. The large member states are divided: whereas the UK and Italy prefer to stick to the Nice Treaty rules, Germany and France are in favor of changing it. Three of the five small states for which we have identified a position on this issue prefer the Nice position (Estonia, Latvia and Malta), standing in opposition is Slovenia and Luxembourg. A majority of the medium small countries seems to be in favor of departing from Nice, as only four countries of this block (Austria, Denmark, Portugal and Sweden) prefer the status quo. For this table we get a chi2 of 3.02, but our results are not significant. Cramer’s V takes a value of 0.3734, thereby suggesting only a slight correlation. Thus there seems to be no strong link between population size and the definition of QMV. When looking at length of membership and the definition of QMV, it is interesting to see that five of the six founding states of the ECSC are in favor of reducing the necessary majorities. Only Italy is of a different opinion on this issue and is joined by the UK and Denmark, the 1973 adherents. We could not identify the Irish preference on this issue. Spain and Portugal stick to the Nice provisions, whereas Greece is willing to change the decisionmaking rules in the Council. The same result is reported for the northern enlargement member states: Austria and Sweden are in favor of the Nice position, but Finland tends towards a double majority system. The new member states are divided on this issue.

Membership 0 1 2 3 4 Total

Issue 2 0 4 2 2 2 1 11

1 4 1 1 0 5 11

Total 8 3 3 2 6 22

Table 6. length of membership and definition of QMV preference (three missings reported) Pearson chi2(4) = 5.3333 Pr = 0.255 Cramer’s V = 0.4924 The chi2 associated with this table has 4 degrees of freedom and is 5.33, but the differences are not significant. The Cramer’s V, however, suggests

167

Domesticated Wolves?

that there is a correlation between length of membership and the definition of QMV. When estimating a linear probability model by OLS with variance correction for issue 2, our second hypothesis is confirmed: length of membership is positively correlated with a preference for a double majority system. Our model does not, however, fit very well, since we can only predict about 23% correctly. With this technique we find a negative correlation between state size and a preference for a double majority system, but this finding is not statistically significant.

Dependent Variable Definition of QMV preference (1) (2) Ind. Var. Coef. Std. Err. p Coef. Std. Err. Population -.0008558 .0047404 0.859 -.0064524 .0047102 Membership .0107295 .0046679 Constant .5173406 .146893 0.002 .4149874 .1553721 R-squared 0.0016 0.2320 N 22 22

p 0.187 0.033 0.015

Table 7. linear probability model by OLS with variance correction for issue 2 The differences between the results of Pearson’s chi2 test and this model might be due to the loss of information in the cross tables, since we lose variance because of the categorization of the data. In the linear probability model population size and length of membership are treated as continuous variables, whereas this is not the case in the cross tables. This observation, of course, underlines that our findings remain less robust because of the small number of cases. Cramer’s V, however, points in the same direction as the findings of the linear probability model. Issue 3 Concerning issue 3, the extension of QMV, our rationalist hypothesis expects the larger states to be more likely to prefer an extension of QMV to taxation and/or CFSP. From a sociological institutionalist point of view, we expect elder member states to be more inclined towards the extension of QMV to these fields than newer ones. If countries prefer not to extend QMV

Dirk Leuffen and Sander Luitwieler

168

to taxation and/or CFSP this is coded as 0, when they are in favor of including taxation and/or CFSP into the range of QMV this is coded as 1. What appears when looking at the relation between population size and the extension of QMV?

Population 0 1 2 3 Total

Issue 3 0 5 5 2 2 14

1 3 6 0 2 11

Total 8 11 2 4 25

Table 8. population size and extension of QMV preference Pearson chi2(3) = 2.2635 Pr = 0.520 Cramer’s V = 0.3009 Again the Pearson’s chi2 test reveals that the differences are not significant and Cramer’s V is low. Five of the eight small states prefer not to extend QMV to taxation and CFSP; however, Cyprus, Estonia and Luxembourg are ready to extend QMV to at least one of these sectors. In category 1 the tendency slightly points in a different direction: 6 of the 11 states are in favor of a general extension of QMV. The two states of category 2, Poland and Spain, are again on the conservative side, while the four large states are divided on the issue. The UK and Italy are now more hesitant than France and Germany. In total 14 of 25 countries are hesitant to further extend QMV to taxation and/or CFSP. What does length of membership tell us about the preference on the extension of QMV?

Membership 0 1 2 3 4 Total

Issue 3 0 8 0 2 3 1 14

1 2 3 1 0 5 11

Total 10 3 3 3 6 25

Table 9. length of membership and extension of QMV preference Pearson chi2(4) = 12.4188 Pr = 0.014 Cramer’s V = 0.7048

169

Domesticated Wolves?

The chi2 associated with this table has 4 degrees of freedom and is 12.42. The observed differences are significant (<0.05). This is supported by a Cramer’s V of 0.7048. Length of membership thus seems to matter. Eight of the ten new member states are against a general extension of QMV including taxation and/or CFSP. Only Cyprus and Estonia take different positions. The opposite direction appears in category four: five of the six founding members are in favor of extending QMV to taxation and/or CFSP. As stated above, only Italy remains hesitant. What is interesting is that categories 1 and 3 show opposite patterns. The three Nordic enlargement countries stick less to unanimity than the three countries of the first enlargement wave, Ireland, Denmark and the UK. Our linear probability model by OLS with variance correction for issue 3 confirms these findings. There does not seem to be a clear relation between population size and the extension of QMV. The state size hypothesis is clearly falsified (see model 1 and 2 in table 10). Length of membership, on the other hand, again proves to be statistically significant. Our sociological institutionalist hypothesis thus remains valid, since the elder member states indeed seem more likely to approve an extension of QMV to taxation and/or CFSP. Dependent Variable Ind. Var. Population Membership Constant R-squared N

Extension of QMV

(1) Coef. Std. Err. p .0002919 .0045504 0.949 .434776 0.0002 25

.1316746 0.003

Coef. -.006522 .0132219 .298754 0.3331 25

preference (2) Std. Err. p .004268 0.141 .0040592 0.004 .1293833 0.031

Table 10. linear probability model by OLS with variance correction for issue 3 Issue 4 Concerning co-decision we again defined two items. Code 0 means that a country prefers no or only a partial extension of co-decision, while 1 means that a country prefers a general extension of co-decision to all legislative acts on which the Council decides by QMV. From a rationalist point of view, we expected larger states to tend towards a general extension of co-decision, whereas smaller states should be more hesitant. From a sociological

Dirk Leuffen and Sander Luitwieler

170

institutionalist point of view, we expected elder rather than newer member states to stronger prefer a general extension of co-decision. When looking at the cross table on the population size variable, we actually realize that our expectation concerning state size does not at all seem to apply.

Population 0 1 2 3 Total

Issue 4 0 1 2 0 3 6

1 7 9 1 1 18

Total 8 11 1 4 24

Table 11. population size and extension of co-decision preference (one missing reported) Pearson chi2(4) = 6.6061 Pr = 0.086 Cramer’s V = 0.5246 Seven of the eight smallest member states seem to prefer a general extension of co-decision. Only Cyprus takes a different position. The opposite is the case for the four largest countries. Here only Germany is in favor of the more integrationist perspective. The eleven medium small countries reflect the pattern of the small states. Only Finland and Sweden take a more restrictive position. Concerning category 2 only the Spanish preference for a general extension of co-decision is reported. Our rationalist hypothesis cannot be confirmed. When looking at the relation between length of membership and the extension of co-decision, we find that in all categories but category 1 a majority of countries prefers a general extension of co-decision. In total only six countries prefer not to generally extend the co-decision procedure: Cyprus, Finland, France, Italy, Sweden and the UK.

Membership 0 1 2 3 4 Total

Issue 4 0 1 2 0 1 2 6

1 9 1 2 2 4 18

Total 10 3 2 3 6 24

Table 12. length of membership and extension of co-decision preference (one missing reported) Pearson chi2(4) = 4.9778 Pr = 0.290 Cramer’s V = 0.4554

171

Domesticated Wolves?

The observed differences for neither state size nor length of membership are significant. Though Cramer’s V suggests a correlation, this measure cannot account for the contrasting directions of the findings and our rationalist expectation. Thus, neither the rationalist nor the sociological institutionalist hypothesis can be confirmed. Our linear probability model by OLS with variance correction replicates this finding. No correlation between state size, length of membership and extension of co-decision seems to exist. Since this issue cannot be accounted for by these variables, other variables, such as, for example, parliamentary traditions or political culture could be explored in the future.

Dependent Variable Extension of co-decision preference

(1) Std. Err. p

Coef.

(2) Std. Err. p

Ind. Var.

Coef.

Population Membership Constant R-squared N

-.0060655 .0042907 0.171 -.0077012 .0046786 0.115 .0029994 .004248 0.488 .8590029 .1003579 0.000 .8321774 .1150787 0.000 0.0833 0.1143 24 24

Table 13. linear probability model by OLS with variance correction for issue 4

Issue 5 Concerning the organization of Council presidencies, we introduced two items. Item 0 reflects a rotating European Council presidency, possibly combined with a permanent Council of the EU presidency. In contrast, item 1 reflects a more permanent organization of both the European Council and Council of the EU presidencies. As always, our hypotheses expect newer and smaller states to prefer solution 0. Elder and larger states should, on the other hand, have a preference for solution 1.

Dirk Leuffen and Sander Luitwieler

172

Generally, there seems to be a majority of 17 member states opposing a more permanent organization of both Council presidencies. Of the eight states preferring the opposite solution, none belongs to the group of the smallest states. Of the eleven medium small states only three, Denmark, Hungary and Sweden, seem to prefer solution 1.

Population 0 1 2 3 Total

Issue 5 0 8 8 1 0 17

1 0 3 1 4 8

Total 8 11 2 4 25

Table 14. population size and Council presidency preference Pearson chi2(3) = 12.6755 Pr = 0.005 Cramer’s V = 0.7121 Of the six largest states only Poland is not in favor of permanency for both Council presidencies. The chi2 associated with this table is 12.68 and the observed differences are quite significant. This is supported by a strong Cramer’s V. Concerning issue 5, population size, indeed, seems to matter. This confirms our rationalist hypothesis. Concerning length of membership, we cannot detect any statistical significance and Cramer’s V is rather small. Although a massive majority of the new member states opposes the idea of generally changing the organization of the Council presidencies (only Hungary seems ready to accept a new solution), the tendencies in the other categories are less clear and no general trend is discernable. In particular, the European founding states are divided and only the three largest states are in favor of establishing more permanent Council presidencies.

Membership 0 1 2 3 4 Total

Issue 5 0 9 2 2 1 3 17

1 1 1 1 2 3 8

Total 10 3 3 3 6 25

Table 15. length of membership and Council presidency preference Pearson chi2(4) = 4.7794 Pr = 0.311 Cramer’s V = 0.4372

173

Domesticated Wolves?

Our linear probability model by OLS with variance correction supports these findings. Dependent Variable Council presidency preference Ind. Var. Population Membership Constant R-squared N

(1) Coef. Std. Err. p .0112816 .0010204 0.000 .080303 0.8416 25

.0770409 0.308

(2) Coef. Std. Err. p .0110065 .0022139 0.000 .0005422 .003836 0.889 .0746708 .0880303 0.405 0.8419 25

Table 16. linear probability model by OLS with variance correction for issue 5 State size, for the first time, clearly proves to have a stronger impact than length of membership. We can thus confirm the rationalist hypothesis: basically all the larger states tend towards the introduction of more stable European Council and possibly Council of the EU presidencies. Generally the fit of our model is excellent: we can predict 84% of the cases correctly. Our socialization hypothesis, on the other hand, is clearly falsified. Concerning this issue socialization effects do not seem to point into the same direction: Luxembourg, Belgium and the Netherlands disagree with the larger founding countries on this issue. Citizen Representation After estimating the impact of our two explanatory variables on single issues, we decided to construct a variable combining the member state preferences on all our issues, in order to get a more general impression of their position towards citizen representation in the EU. For each country we added all preferences for the single issues and divided this sum by the number of issues for which preferences were reported (generally five, but for four cases four).15 A higher score on this variable reflects a more citizen equality oriented approach, a lower score tends more towards the status quo and member state equality. Our rationalist hypothesis expected that larger

15

Thus we did not account for the possibly different saliency of the five issues, which would have been very difficult to measure.

Dirk Leuffen and Sander Luitwieler

174

states should generally favor citizen representation, thus should score higher, whereas smaller member states should tend towards the smaller side of the scale. When testing this hypothesis on the basis of a linear probability model by OLS only for population size – we did not need to include a variance correction this time because the dependent variable was not dichotomous anymore – we indeed find a significant positive correlation between state size and ‘Citizen Representation’ as model 1 in table 17 shows.

Dependent Variable Citizen representation Ind. Var. Population Membership Constant R-squared N

(1) Coef. Std. Err. p .0038408 .0017341 0.037 .3710191 .0503226 0.1758 25

0.000

(2) Coef. Std. Err. .0007353 .0016778 .0063972 .0018328 .3034569 .0455916 0.4695 25

p 0.665 0.002 0.000

Table 17. linear probability model by OLS for ‘Citizen Representation’ When controlling for length of membership, however, the significance of population size is wiped out and only length of membership remains significant. Concerning the sum of all five issues we can thus falsify the state size hypothesis and confirm the socialization hypothesis: the longer a state has been a member in the EU, the more likely it will favor a higher citizen representation in the EU. On our scale Germany, as the largest state and a founder, scores highest of all with 1.0. For the then candidate countries, on the other hand, only Hungary with 0.6 can be classified amongst the more EU citizen oriented countries. Of the founding countries Italy gets the lowest score in the sum. Spain (0.25) and Portugal (0.2) end up in the lower range, while Denmark (0.6) and the UK (0.4) end up with medium results. Figure 1 shows the relation between length of membership and EU citizen representation.16

16

This graph is designed with Stata 8.0.

175

Domesticated Wolves? Fitted values

Sum/issues D

1

Citizen Representation Index

FR

B

DK

NL AU

EST SK CZ SLO

LUX

GR

HU

FIN

UK SWE

IT

IRL

MAL

.2

LIT LAT CYP POL

ESP POR

0

Membership

52

Figure 1. Length of Membership and Citizen Representation To sum up, our findings show that many commentators on the EU Convention might have overestimated the division between small and large member states. Of the issues that we analyzed, only the Commission size and the organization of the Council presidency clearly confirm the state size hypothesis, and for the Commission size our models show that it might even be trumped by the length of membership variable. Our findings so far, of course, lack a certain robustness since they are based on a relatively small number of observations. We think, however, that our analysis shows that length of membership is an important element for explaining member state preferences in the EU.

6.

CONCLUSION

In this chapter we compared a sociological with a rational institutionalist explanation of preference formation. On the basis of new institutionalist literature, we developed two general hypotheses explicating how length of membership and state size impact on EU member state preferences. We then generated a set of specific hypotheses geared to five institutional issues of the Convention, which we subsequently tested with the help of cross tables and linear probability models by OLS. Finally, we combined the different issues into a new variable that we called ‘Citizen Representation’ and on which we also tested our hypotheses.

176

Dirk Leuffen and Sander Luitwieler

We generally find that both approaches are able to account for member state preferences at the Convention, but not for all our issues in the same way. In particular the issue of the extension of co-decision is not well captured by any of our theories. Potentially other variables, such as parliamentary traditions, might better account for the issue. The rationalist explanation on the basis of state size scores particularly well concerning the Council presidency issue. It also quite well predicts the preferences concerning the Commission size. When we, however, control for this issue for the sociologically oriented variable length of membership, the importance of the division between large and small member states is clearly reduced. Since this effect even more applies to our combined ‘Citizen Representation’ variable, our analysis strongly highlights the relevance of the sociological institutionalist approach. Although our statistical findings still suffer from a rather small n, they at least question the division between small and large member states that was perceived by several media commentators and scholars alike during the Convention. Future research could engage in constructing more analytical categories in order to better explain preference formation. This equally applies to the rational and the sociological institutionalist explanation. As to the latter, we applied ‘length of membership’ as a more general ‘scope condition’ for socialization to be likely to happen. In addition, it could be particularly interesting to investigate the scope condition ‘degree of membership’ (or, the membership–non-membership dichotomy): a full member of the EU is likely to be socialized into EU norms and values to a greater extent than a pre- or non-member, as the ‘density of interaction’ between full members of the in-group is higher (Checkel 1999, 549; Trondal 2002, 470-71). An objection that could be raised against our findings is the problem of a spurious relationship. According to this logic, detecting more supranational preferences amidst the elder member states does not necessarily mean that these countries were more strongly ‘Europeanized’ or socialized into EU norms and values. Instead, their original integrative preferences might explain both their decision to join the Communities in the first place and their current integration preferences. Thus the ex ante preferences would be the explanatory variable. These preferences, however, remain difficult to measure and therefore to control for, especially since not all countries always had the opportunity to join the EU, which of course particularly applies to the Central and Eastern European states. One could also criticize the specific concept of socialization we used in our research design. We assumed that actors are socialized into EU norms and values making them more pro-integrative. They might, however, also learn to act more strategically and thus adopt preferences that are more in line with a rational institutionalist approach. To increase our knowledge of such questions, future qualitative research is needed. Applying process

Domesticated Wolves?

177

tracing, for example, seems a suitable way to better understand why certain countries at the Convention acted as they did. In addition, future research could consider how to further increase the number of observations, in order to better understand whether and to what extent elder member states are actually more ‘Europeanized’. How does our chapter relate to this book project? In fact, we tried to reflect the third wave approach of European integration studies as closely as possible. In investigating how length of membership impacts on state preferences for future integration we clearly combined top-down and bottom-up processes of European integration. Our research thus again points towards the dynamic interactions within the EU multilevel system of governance.

REFERENCES Aldrich, J. H., and Nelson, F. D., 1984, Linear Probability, Logit, and Probit Models, Sage, Newbury Park. Allen, D., 2004, The Convention and the draft constitutional treaty, in: The Future of Europe. Integration and Enlargement, F. Cameron, ed., Routledge, London, pp. 18-34. Aspinwall, M., and Schneider, G., 2001, Institutional research on the European Union: mapping the field, in: The Rules of Integration. Institutionalist Approaches to the Study of Europe, G. Schneider and M. Aspinwall, eds, Manchester University Press, Manchester, pp. 1-18. Börzel, T. A., and Risse, T., 2003, Conceptualizing the domestic impact of Europe, in: The Politics of Europeanization, K. Featherstone and C. M. Radaelli, eds, Oxford University Press, Oxford, pp. 57-80. Checkel, J. T., 1999, Social construction and integration, Journal of European Public Policy 6(4):545-60. Checkel, J. T., 2001, Why comply? Social learning and European identity change, International Organization 55(3):553-88. Hooghe, L., 2001, Top Commission officials on capitalism: an institutionalist understanding of preferences, in: The Rules of Integration. Institutionalist Approaches to the Study of Europe, G. Schneider and M. Aspinwall, eds, Manchester University Press, Manchester, pp. 152-73. Koenig-Archibugi, M., 2004, Explaining government preferences for institutional change in EU Foreign and Security Policy, International Organization 58(1):137-74. Kohler-Koch, B., 2002, European networks and ideas: changing national policies?, European Integration online Papers 6(6), http://eiop.or.at/eiop/texte/2002-006a.htm. Magnette, P., and Nicolaidis, K., 2004, Coping with the lilliput syndrom. Large vs. small member states in the European Convention, Politique Européenne 13(1):68-92. March, J. G., and Olsen, J. P., 1984, The new institutionalism: organizational factors in political life, American Political Science Review 78(3):734-49. March, J. G., and Olsen, J. P., 1989, Rediscovering Institutions: The Organizational Basis of Politics, Free Press, New York. March, J. G., and Olsen, J. P., 1998, The institutional dynamics of international political orders, International Organization 52(4):943-69. Peters, B. G., 1999, Institutional Theory in Political Science. The ‘New Institutionalism’, Pinter, London.

178

Dirk Leuffen and Sander Luitwieler

Radaelli, C. M., 2003, The Europeanization of public policy, in: The Politics of Europeanization, K. Featherstone and C. M. Radaelli, eds, Oxford University Press, Oxford, pp. 27-56. Schimmelfennig, F., 2003, The EU, NATO and the Integration of Europe. Rules and Rhetoric, Cambridge University Press, Cambridge. Trondal, J., 2002, Beyond the EU membership–non-membership dichotomy? Supranational identities among national EU decision-makers, Journal of European Public Policy 9(3):468-87. Trondal, J., 2004, Re-socializing civil servants: the transformative powers of EU institutions, Acta Politica 39(1):4-30.

BEYOND THE COMMUNITY METHOD

Why the Open Method of Coordination was introduced to EU Policy-making Armin Schäfer Abstract:

1.

This chapter looks at the introduction of the Open Method of Coordination (OMC) to EU policy-making. This new mode of governance has been developed over the last decade and has received considerable attention in the literature. However, much of this writing fails to put the OMC into the broader context of European Monetary Union (EMU); in contrast, this chapter links the Amsterdam employment title to the prior Maastricht decision to form a monetary union. It seeks to contribute to the literature on European integration in two ways: First, this chapter offers three refinements to Pierson’s historical institutionalist account of European integration. Second, it provides an alternative to functional explanations of the OMC. In brief the argument is that a conservative-liberal coalition created hard law in fiscal and monetary policy to constrain its successors at Maastricht, while the social democratic majority relied on soft law to promote its goals in employment and social policy at Amsterdam. While the former effectively limited later policy-choices, the latter largely avoids sovereignty losses for national governments.

INTRODUCTION

Over the last decade, a new policy-instrument has emerged within the European Union. Since the European Council summit at Lisbon, this new soft law procedure has been called the ‘Open Method of Coordination’ (OMC). It differs from the ‘Community Method’ as it is more intergovernmentalist and neo-voluntaristic. What accounts for the European Union selecting less integrationist soft law in employment and social policy while relying on hard law in monetary and fiscal policy? Why has integration been uneven? This chapter traces the paradox of the EU becoming a state in monetary policy while remaining an international

179

R. Holzhacker and M. Haverland, (eds.), European Research Reloaded: Cooperation and Integration among Europeanized States, 179-202. © 2006 Springer. Printed in the Netherlands.

180

Armin Schäfer

organization in economic policy at large. More narrowly, it seeks to answer the question of why the OMC was introduced to EU policy-making. In line with the chapters of van Esch and Leuffen/Luitwieler, I will analyze how prior European integration influences the preferences and policy-choices of national actors. To do so, I will rely on the historical institutionalist framework developed by Pierson (1996). He has argued that even though member states (in fact, governments) design European institutions in their own interests, these will thereafter not always conform to their expectations. Instead, ‘gaps in control’ emerge due to the partial autonomy of supranational institutions, restricted time horizons of political decision makers, unanticipated consequences, and shifts in governments’ preferences. Supranational actors use these gaps in control to further their own interest and advance integration. In sum, there will be a ratchet-up effect of integration despite the efforts of national decision-makers to avoid sovereignty losses. While this account offers valuable insights to European integration, I nonetheless suggest a number of refinements that allow for a broader range of outcomes. First, governments can choose between different policyinstruments for cooperation. Not all of them lead to gaps in control. Instead, some of them – most notably the newly created Open Method of Coordination – strengthen the influence of governments at the detriment of Community actors. Second, governments learn that some modes of governance are more likely to produce ‘agency losses’ than others. In particular, they learn to restrict the autonomy of supranational actors as a result of past experiences. Third, changing political majorities in the (European) Council can influence the choice of policies and policyinstruments. Hence, we have to look at how national governments align along party political lines. Overall, these amendments seem suited to explain how the decision for European Monetary Union influenced later policychoices. To illustrate this argument, this chapter provides a case study that reconstructs the sequence from Maastricht to Lisbon to demonstrate how prior integration, changing actors, and learning led to the choice of a soft law approach in employment policy – subsequently dubbed the Open Method of Coordination. In the existing literature, the politics of choosing policy-instruments is oftentimes neglected. Most of the writing on the OMC focuses instead on its presumed effects, that is, they adopt a top-down perspective. It is argued that soft law was selected as a problem-solving device that facilitates mutual learning and a diffusion of best practices (de la Porte and Pochet 2002, 12; Radaelli 2003, 52). Implicitly, it is assumed that policy-instruments are chosen on a functional logic. A historical institutionalist approach, in contrast, stresses how decisions are mediated and constrained by the already

Beyond the Community Method

181

existing institutional setting (cf. Thelen and Steinmo 1992, 2). Prior decisions have an impact on later ones. Accordingly, as Holzhacker and Haverland point out in the Introduction, ‘third wave’ studies of EU policymaking have to systematically integrate bottom-up and top-down perspectives. Therefore, this chapter seeks to show how institution building limits available policy-choices and feeds back into the creation of new institutions. It thus contributes to the literature on the origins and the development of the OMC, questioning functional explanations and offering a historical institutionalist account instead.

2.

CONCEPTUALIZING EUROPEANIZATION

INTEGRATION

AND

From a historical institutionalist view, there is a sequential relationship between integration and Europeanization where they will impact on each other in the following way: At t1 a favorable actor constellation decides to push for integration, that is, to create ‘at the European level distinct structures of governance’ (Risse, et al., 2001, 3). The kind of institutions created and the mode of governance selected will depend on the compromises struck between the actors involved. Once created, institutions are sticky because altering them depends on unanimous consent. Subsequently, at t2 the mode of governance will impact on national policymaking and restrict certain policies while enabling others (i.e., Europeanization). At t3 a new actor constellation might have emerged; however, in most cases, it cannot reverse the steps taken before and the room for maneuver decreases. Accordingly, the range of policy choices or available instruments becomes more limited at the end of the sequence. This three-step model was spelled out most forcefully by Pierson (1996).1 He sought to provide more realistic a framework for analyzing European integration than either intergovernmentalism or neo-functionalism. Taking into account the intergovernmentalist premise that in negotiations governments will not accept any compromise running counter to their interests, he then moved on to show why, over time, gaps between their institutional preferences and the actual functioning of institutions emerge nonetheless (ibid, 131). There are, according to Pierson, four reasons why they do: autonomous actions by supranational actors, restricted time horizons of decision makers, unintended consequences, and actors with different preferences taking office. 1

Cf. Hix’s (1998, 47) summary.

182

Armin Schäfer

Equally intriguing is the question why these control gaps cannot be undone by governments. If actors face unwanted consequences to prior decisions, why are they not able to alter them? There are three main reasons: the resistance of supranational actors, institutional obstacles to reform, and sunk costs (Pierson 1996, 142-5). European integration in particular empowers supranational actors. Over the years, the Commission has used its right of initiative to increase Community competencies.2 The European Court of Justice (ECJ) has also interpreted its mandate extensively. Supranational actors can make use of their powers to further expand their competencies. This phenomenon is known as ‘agency loss’ in the literature on delegation: The principal always has difficulties in fully controlling the agent’s actions (Pollack 1997). In brief, this analysis suggests a ratchet-up effect of European integration. Once decisions are taken, moves towards integration cannot be easily reversed. Consequently, at a later point in time, the options available become more limited. Pierson (1996, 147-8) believes that this account offers a better explanation for ‘task expansion’ (Pollack 1994, 96) and ongoing integration than neo-functionalism because it offers sound micro-foundations. His approach to historical institutionalism is both actor-centered and rationalist.3 While I accept the thrust of Pierson’s argument, I nevertheless propose some amendments to it. In particular three points deserve attention: (1) not all policy-choices lead to a ratchet-up effect of integration; (2) governments learn to choose forms of cooperation that limit agency losses; (3) there is an oft-neglected political dimension to the choice of policy-instruments. While the next few paragraphs elaborate on these points, section three will offer an empirical demonstration of the extended model in a case study. Three amendments First, the consequences – unintended or not – that follow from a decision for integration depend on the policy-instrument chosen. There is no need to assume they all have the same effects. Especially since different policyinstruments in varying degrees transfer power to supranational actors and can be more or less precise as well as binding. Abbott et al. (2000, 401) use delegation, precision, and obligation to analyze differences between soft and hard law. The loss of control for national governments depends on the

2

Rhodes (1995: 99-104) refers to this as the ‘treaty-base game’. Usually, three institutionalisms are juxtaposed: sociological, rational choice, and historical institutionalism (Hall and Taylor 1996). However, the first two can both lead to the lock-in that the last usually stresses; there can well be either a sociological historical institutionalism are a rationalist historical institutionalism. Thelen (1999) provides a discussion of the relationship of the three variants. 3

Beyond the Community Method

183

specification of these three dimensions. Notably, delegation can trigger agency losses: “Legalized delegation, especially in its harder forms, introduces new actors and new forms of politics into interstate relations. […] actors with delegated legal authority have their own interests, the pursuit of which may be more or less successfully constrained by conditions on the grant of authority and concomitant surveillance by member states. Transnational coalitions of nonstate actors also pursue their interests through influence or direct participation at the supranational level, often producing greater divergence from member state concerns. Deciding disputes, adapting or developing new rules, implementing agreed norms, and responding to rule violations all engender their own type of politics, which helps to restructure traditional interstate politics” (Abbott, et al., 2000, 417). Obviously, the Community Method delegates considerable power to the Commission and the ECJ and avails ample opportunity to act independently of their principals. This mode of governance is characterized by the Commission monopoly of the right of initiative; a widespread use of qualified majority voting in the Council; an active role for the European Parliament; and the uniform interpretation of Community law by the European Court of Justice. However, already the second and third pillars of the EU-Treaties circumscribe supranational actors’ competencies. These pillars are based on an intergovernmental logic of cooperation in which the Commission’s right of initiative is shared with the member states or limited to specific areas of activity; the Council generally acts unanimously; the European Parliament is confined to a purely consultative role; and the Court of Justice, if at all, plays only a minor role. Furthermore, and more importantly for this chapter, the Open Method of Coordination also refrains from delegating authority. Its low degree of legalization makes it a soft law approach (Kenner 1999). Hence, there are alternatives to the Community Method that limit agency losses. More generally, governments can choose between a range of different modes of coordination that differ with regard to delegation and, accordingly, the degree of legalization. Table 1 gives a number of examples of how modes of governance differ in terms of delegation and legalization. The choice of a ‘softer’ policy-instrument limits losses of control for national governments. The two extremes are pure competition on the one hand and complete integration (or supra-nationalization) on the other.

Armin Schäfer

184 Delegation

Legalization

low

soft law

high

hard law

Mode governance

of

Example

wage policy; tax policy; OMC Social cooperation Inclusion coordination EES (Art. 125without sanctions 130); BEPG (Art. 98-100) (soft coordination) coordination with fiscal policy (Art. 104, and SGP) political sanctions coordination with competition policy (Art. 81-89) delegated sanctions (Community Method) competition

integration (complete delegation)

monetary policy (Art. 105-112)

Table 1. Menu of EU Modes of Governance A second point of departure from Pierson’s model concerns the impact of unintended consequences on future decisions about policy-instruments. While it is true that the reversal of past decisions can be very difficult under unanimity rule, governments learn that certain instruments make unwanted consequences more likely than others. Once they have realized this, they might well opt for more intergovernmental forms of cooperation. Later sections of this chapter will show that this has happened in the field of employment policy. More abstractly, this means that a sequence can be much longer than merely three steps which, in turn, necessitates studying longer periods of time – thus calling for an extended historical institutionalist account. If we prolong the time period under observation, we might realize that Pierson’s three step model in fact is but a subsection of a longer sequence. Hence, there will be various lock-in, feedback, and learning

Beyond the Community Method

185

effects that need examination.4 Not all of them will work in the same direction; in fact, they can conflict with and contradict each other. Third, the kind of policy-instrument chosen depends on the beliefs of salient actors and, therefore, on actor constellations. That is to say, the choice of policy-instruments depends on politics. Even though Pierson refers to the British Conservative government’s opt-out of the Social Protocol at Maastricht (for more details Lange 1993) and goes on to stress that a single Labour victory could reverse this decision, he fails to more generally consider the impact of the partisan composition of the European Council on policy-choices. However, had more governments shared the Conservative’s position on social policy the Social Agreement would not have been possible. Instead, Christian democratic parties strongly influenced the IGC that strove for flanking market-making with some – limited – social policy (Johansson 2002). Similarly, had the 1980s been dominated by center-left governments rather than the New Right, the Single European Act would either have not occurred or looked differently. In general, it seems important to take the party affiliation of negotiating governments into account. This last point is particularly important if Hix and Lord’s (1997, 4) assertion holds that “key players rush to form agreements while there is a favorable conjuncture in the domestic politics of member states. They may also seek to neutralize future uncertainties in national party politics by locking successor governments into EU-level agreements.” In this case, the partisan composition of the European Council will matter a great deal and deserves attention. The next section will show that a center-right coalition opted for hard law in monetary and fiscal policy at Maastricht, while a center-left coalition relied on soft law for employment policy at Amsterdam. Both timing and politics matter in the choice of policy-instruments.

3.

FROM EMU TO OMC: A HISTORICAL INSTITUTIONALIST ACCOUNT

This section provides a case study that uses the amended Pierson model to explain a specific feature of the EU’s political economy: Why Maastricht created hard law for monetary and fiscal policy, while Amsterdam relied on soft law for employment policy? I would argue that we cannot understand 4

Historical institutionalism and especially arguments about path-dependency always face the difficulty to delimit the beginning and the end of a path. The very time segment studied can influence the estimation of the causal effect. Unless we have a theory about the forces that sustain (and end) a path, it is hard to repel the claim that any particular path actually started even earlier than maintained. While this is an important problem that deserves attention, this brief chapter can by no means do justice to it.

Armin Schäfer

186

these different degrees of integration and Europeanization without looking at changing actor constellations over time, learning, and politics. To show how actor constellations vary, this chapter uses a data-set on the party composition of government in all member states. We observe a large variation in the number of center-right and center-left governments between 1991 and 2003. Based on this observation, the case study will reveal this sequence: A market enhancing actor constellation (t1) fostered integration following a macroeconomic design demanding ‘hard coordination’ for monetary and fiscal policy (t2), which on the one hand has had strong effects on national policy-making (t3) and on the other limited subsequent policy choices. Accordingly, between 1997 and 2001, the social democratic majority in place merely created a subsidiary employment policy, supplementing market integration. Today, the main focus of social policy and social inclusion is ‘commodification,’ i.e., including as many people as possible into the labor market.5 As an illustration, we can depict the relation between integration and Europeanization in the following way: actor constellation t1 monetary and fiscal policy employment and social policy

conservativeliberal coalition t4 social democratic coalition

integration t2

Europeanizati on t3

strong

strong

t5

t6

weak

weak

resulting regulation EMU: hard law

OMC: soft law

Table 2. Sequence of Integration and Europeanization

3.1

European Monetary Union

This section deals with the first part of the sequence shown in Table 2. It seeks to show how the conservative-liberal coalition facilitated an agreement on EMU. Governments in Maastricht agreed upon ‘integration’ as the mode of governance for monetary policy. By transferring the decision-making power to the European Central Bank, they chose to even surpass the 5 “While the term ‘Social Europe’ accentuates the social goals behind the strategy, these goals are almost exclusively approached from the direction of employment. The underlying message is that the social value of an individual is primarily determined by his/her potential contribution as a worker. The overarching operational aim is to get those able to work working and those unable to work enabled” (Hvinden, et al., 2001, 174).

Beyond the Community Method

187

Community Method. Therefore, not only the impact on national policymaking has been strong but also the room for maneuver for later actors has severely been limited. This chapter does not claim, however, that party politics was the only factor bringing about EMU. The Maastricht Treaty came at a time when the political landscape of Europe was rapidly changing due to the end of the Cold War. It could also build on a relatively high degree of economic convergence between the member states and pre-established forms of monetary cooperation. Yet, the conclusions actors draw from such developments depend on who these actors are. The dominance of market-oriented governments clearly had an impact first on the integration project chosen and second on its institutional design. Therefore, to understand the peculiar political economy of the European Union, we have to begin with the Maastricht decision to form a Monetary Union. It entails the creation of an independent central bank and a set of rules for fiscal policy. Owing to this, EMU is strong on delegation, obligation, and precision and has a strong impact on national policies. Why was it possible after more than 20 years of discussions to agree on EMU? Two aspects in particular seem important. First, an expert consensus about the design of a monetary union existed at the time of the negotiations. Second, a favorable actor constellation built up supporting these experts’ ideas. Ideational convergence met with political power to create EMU.6 An expert consensus was possible because a small circle of Ministers of Finance and Central Bankers negotiated the framework for EMU. This took place in the Committee of Central Bankers, the Monetary Committee and the Delors’ Committee. The participants formed an ‘epistemic community’ (Haas 1992) that concentrated mainly on the problem of how to secure price stability in a monetary union. This narrow focus also led to a narrow definition of the institutional requirements. Members of the Committees came to share the view that an independent central bank built on the German example would ensure sound money. As the German Bundesbank was involved in these discussions, its later approval was facilitated. This, in turn, helped the government to domestically justify giving up the D-Mark (Moravcsik 1998, 395). The evolving monetarist consensus meant abandoning earlier proposals for a monetary union. Both the Werner Plan (1970) and the MacDougall Report (1977) had stressed the need for centralizing decisions on fiscal policy and increasing the Community budget (Tsoukalis 1993, 275); centralizing monetary policy without centralizing economic policy decisions 6

Dyson and Featherstone (1999) provide a detailed account of the developments leading to the Maastricht Treaty.

188

Armin Schäfer

was inconceivable at that time. Once Keynesian thinking had been abandoned, stabilizing aggregate demand for the entire Community became less salient. Instead, fiscal policy came to be seen mainly as a means supporting price stability. Accordingly, the Delors’ Committee suggested defining a limit to national budgets. EMU could only work if national deficits and debts were controlled by a clear set of binding rules. In contrast, the overall policy-mix was to be achieved by a soft coordination of national policies. Gouvernement économique did not seem necessary any longer. Limiting transfers of sovereignty exclusively to monetary policy not only made for an institutionally lean concept of EMU but also facilitated agreement. During the 1980s, due to the European Monetary System most governments had already lost their monetary autonomy as they followed the low inflation policy of the Bundesbank. Thus, they could even hope to gain influence by transferring monetary authority to a joint central bank (Sandholtz 1993, 37-8). Germany, on the other hand, was the one state that had to abdicate power. This strengthened its bargaining position. All other governments had to agree to a version of EMU that could find the support of the German government as well as the Bundesbank. Modeling the new European central bank on its German predecessor eased that problem. As Radaelli (2000, 30) points out, the provisions on EMU are a case in point of ‘institutional isomorphism.’ Exactly how important German unification in this process was is disputed. Moravcsik (1998, 428) denies that the fall of the Berlin Wall made any difference since the German negotiation position did not change between 1988 and the Maastricht summit. Dyson and Featherstone (1999, 757), in contrast, argue that German unification gave a sense of urgency primarily to French policy-makers to conclude the negotiations on EMU. It seems fair to say that the changed geopolitical situation speeded up negotiations that otherwise might have been inconclusive (see Sandholtz 1993, 31-4).7 In sum, an expert consensus on monetary union, the demand for only limited transfers of sovereignty along with German unification gave impetus to creating EMU. These were necessary conditions, though; by themselves, they did not suffice. Only the overwhelming majority of center-right 7

To mollify France’s concerns, the German government agreed to a precise schedule towards monetary union. Earlier, any definite time-table had been rejected with the argument that prior economic convergence would have to occur. This so-called ‘coronation theory’ had traditionally been opposed by the French view that monetary cooperation would trigger economic convergence (‘locomotive theory’) (Kruse 1980, 62-70; Tsoukalis 1993, 178). The will to come to an agreement in Maastricht bridged these seemingly contradictory strategies: While the start of EMU was fixed for 1999, only those states that until then would have achieved a sufficient degree of convergence should partake. Hence, the Maastricht Treaty spelled out a number of convergence criteria as a precondition for participation.

Beyond the Community Method

189

governments paved the way to EMU. In 1991, only three out of twelve member states were governed by the left. One of them was France that actively supported further steps in European integration. In the other countries, center-right parties were in government that followed a program of liberalization and fiscal austerity. For them, EMU was a way to permanently enshrine price stability in a European framework. They also saw the opportunity to discipline their successors (much in Hix and Lord’s sense) and to add legitimacy to their national course of action. In short, at t1 of the sequence, center-right governments seized the opportunity to adopt hard law to bolster EMU and created an autonomous central bank committed to price stability. They turned their policy-preference into hard law that subsequently could not easily be undone. What does this mean for the concept of integration and Europeanization presented earlier? Monetary policy is quite exceptional to EU policy-making in precluding any influence of the member states. In contrast, even the Community Method does not work without the cooperation of national authorities. Moreover, with the ECB an independent actor was created whose mandate can only be changed by unanimous consent. Even if governments wanted to pursue a different course of action, they would be inhibited by the institutional framework created in Maastricht.8 The European Central Bank has the power to retaliate against any attempt to instigate growth by a less restrictive fiscal policy. Due to the priority given to price stability, monetary policy cannot be employed to strengthen growth. In fact, insolating the ECB from demands for a growth-inducing policy-mix signals that it pursues a ‘dominant strategy’ (Bernhard, et al., 2002, 706). There is no need for an exante coordination of policies because it is assumed that all other actors will have to adapt to the central bank’s strategy. Hence, Issing (2002, 34) points out that the Maastricht Treaty provides an ‘efficient initial assignment’ rendering other forms of coordination superfluous. In contrast, von Hagen and Mundschenk (2001) are skeptical of whether this institutional setting in itself will produce positive outcomes. Be that as it may, the critical point is that in monetary policy there is strong legalization and accordingly economic policy is locked-in to a monetarist path. The resulting impact (i.e., Europeanization) of EMU on national policymaking is also strong. It forestalls the use of exchange rate changes or fiscal stimuli during crises and, in combination with the single market, considerably increases competition; member states have few instruments at their disposal to react to asymmetrical shocks or low growth; as labor 8

This poses serious questions for the democratic legitimacy of ECB’s decisions. Cf. Scharpf (1999, 155).

190

Armin Schäfer

mobility across borders remains marginal, the need for domestic labor market flexibility mounts; price competition is heightened and there is a strong pressure to reduce ‘slack,’ i.e., subsidized or sheltered unproductive jobs. As a result, the options for a European employment policy have been markedly reduced by EMU and the SGP’s rules for fiscal policy. As Scharpf (2002, 655) stresses, a number of traditionally available strategies are beyond the scope of such an initiative: “Thus if unemployment rises in the Euro Zone generally, Luxembourg EES guidelines could not recommend lower ECB interest rates; if unemployment rates rises nationally, EES recommendations could neither relax the deficit rules of the Stability Pact nor the competition rules on state aids to depressed regions or industries. Similarly if expenditures on health care are rising, OMC could not recommend controls or “positive lists” for pharmaceuticals; and if social services are being eroded by fiscal constraints, there is no chance for guidelines promoting either a concerted increase of taxes on capital incomes of failing that, the re-introduction of effective capital exchange controls.” Prior integration explains why the European employment strategy is limited in its contents. It was both a supplement and subordinated to EMU. Any more ambitious, employment-creating initiatives were circumscribed by the importance attached to sound money and the credibility of maintaining price stability. Therefore, employment policy was residual in character. On this account, Pierson’s three-step historical institutionalist model holds for the first part of the sequence of table 1. Yet, employment was not only limited to supply-side measures but also confined to soft law. Why was a non-binding, soft mode of coordination chosen? The next section addresses this question. It deals with the second part of the sequence in table 2.

3.2

The Origins of the Open Method of Coordination

In the aftermath of the Maastricht summit, European integration faced a legitimacy crisis. Danish voters in 1992 rejected the Treaty amendments in a referendum while the French accepted them only by a tiny majority. At the same time, the European economy turned sore and unemployment rose again. There was a widespread notion that integration was too exclusively focused on market integration, disregarding unemployment as the most pressing problem. The permissive consensus supporting integration was at risk. In this moment of crisis action had to be taken to prove that governments cared for unemployment without, however, granting more power to distant ‘Brussels bureaucrats.’ At the Copenhagen summit in 1993, the European Council instructed the Commission to produce a White Paper. It was asked to spell out a strategy

Beyond the Community Method

191

for higher growth, competitiveness and employment. The Commission’s report became known as the ‘Delors’ White Book.’ However, the first real step towards a European employment initiative came with the Essen summit in 1994. Heads of state and government agreed on a number of objectives to fight unemployment (European Council 1994).9 These included investment in human capital (up-skilling of the workforce), increasing the employmentintensiveness of growth, reducing non-wage labor costs, improving the effectiveness of employment policy by moving from passive to active labor market policy as well as supporting groups particularly hard hit by unemployment. An important role was also conferred to social partner dialogue.10 Crucially, however, none of these objectives was legally binding or enforceable. Those governments most skeptical of a European employment policy knew that a binding agreement would make for undue Commission interference in domestic policy-making. Accordingly, they fought for a soft law approach. As a result, member states were merely urged ‘to transpose these recommendations in their individual policies into a multi-annual program having regard to the specific features of their economic and social situation’ (European Council 1994). The European Commission and the Labor and Social Affairs as well as Economic Financial Affairs Council were asked to monitor national developments and report annually to the European Council about their progress. Hence, the core of the Open Method of Coordination – common objectives, national implementation and surveillance by the Commission and member states – were in place already in 1994. Content and form followed from a disagreement in substance and a consensus to limit obligation. A soft coordination padded with mutual monitoring served both. Another parent of the OMC was the convergence process introduced with the Maastricht Treaty (Pochet, et al., 2001, 295). Member states drew up a set of criteria defining eligibility for entering monetary union. Again, it was up to each individual government by which means it wanted to meet these goals. There was a yearly assessment of the achievements in so-called convergence reports. The convergence process did not depend on any (negative) sanctions because compliance promised club membership. It was a soft coordination mechanism in that it was not legally binding, though it entailed strong incentives to comply. In the end, the decision which member states in fact were eligible for monetary union was a political one, too, but 9

As a matter fact, the Essen Employment procedures were meant to fend-off more ambitious proposals of the ascending social democratic majority in the European Council (Hix/Lord 1997, 194). 10 These objectives until today are at the heart of the European Employment Strategy.

192

Armin Schäfer

no one doubted that several states had taken great pains to meet the Maastricht criteria and that a fair level of convergence had been achieved. Thus, during the Intergovernmental Conference leading to the Amsterdam Treaty those governments supporting a chapter on employment could point towards already agreed upon principles for employment policy as well as a successful method. Especially those countries that had joined the Union only in 1995 – Sweden, Finland, Austria – supported a higher EU profile in tackling unemployment. Still, until early 1997 chances were slim that the negotiations would lead to an agreement as conservative governments in Great Britain, France, and Germany opposed even this limited approach. Only after the elections in the first two countries, bringing New Labour and the French socialists to power, did an agreement became possible (Szyszczak 2000, 204). Finally not even the Kohl government was keen to block treaty reform for an issue considered of minor relevance. It wanted to make sure, though, that any transfer of power to the European Commission would be strictly limited. In the end, employment became part of the Amsterdam Treaty. Without the leftward electoral swing in Europe this would not have happened (Jenson and Pochet 2002, 8). In an article on the genesis and the development of the European Employment Strategy, Goetschy (1999, 125) offers a similar conclusion: “Consensus in the [Amsterdam] IGC over employment was only possible on two conditions: national employment policies should continue to play the principal role, and major and costly programmes at EU level should be avoided. Some parties (ETUC, the EP and some national delegations) would have preferred the Treaty to have formally specified the relationship between economic and employment policies. Instead the desire to bring the two into balance was declared in a formal Council resolution (albeit without binding effect) on growth and employment. The inequality of status between the mandatory Stability and Growth Pact agreed at the December 1996 Dublin summit and the resolution on employment illustrates the continuing imbalance between monetary and political integration.” Since the amended Treaty would only be ratified in 1999, governments decided to have a special summit on employment in 1997. It took place in Luxembourg, baptizing the European Employment Strategy ‘Luxembourg process.’ Consciously modeled on the Maastricht convergence process, it introduced the following cycle (European Council 1997): (1) every year, members states agree on ‘Employment Guidelines’ specifying common objectives; (2) each governments draws up a ‘National Employment Action Plan’ (NAP) detailing the strategy of how it seeks to achieve these goals;

Beyond the Community Method

193

(3) based on a Commission draft, the Council and the Commission publish the ‘Joint Employment Report’ which assesses and evaluates member states’ NAPs as well as their policies; (4) the European Council annually reviews this reports and modifies, if appropriate, the Guidelines. While this procedure mimicked the convergence program it refrained from introducing sanctions. The Luxembourg process tried to compensate for its limited legal weight by putting a stronger emphasis on the commitment made by governments. They were asked to draw up a NAP to detail their approach to fighting unemployment. This document thereafter could be used to compare word and action. This Commission hoped to gain influence on national policies by reminding reluctant governments on their prior commitment. At the Luxembourg summit no consensus could be reached on defining a target for unemployment as a counterpart to those concerning public deficits, inflation, and debts (Rodrigues 2001, 4). The Commission proposed a more rigorous set of employment guidelines and urged the member states to accept clearly specified quantitative targets but could not overcome their resistance (Biagi 2000, 57). That they affected a compromise at all was largely due to the negotiation skills and personality of Jean-Claude Juncker, Luxembourg’s prime minister.11 Juncker, a Christian Democrat, convinced his colleagues that some action was necessary. Reducing employment policies to a voluntary opportunity for learning without obligatory targets facilitated support for it (Trubek and Mosher 2003, 38). In the end, such diverse governments as those led by Jospin, Blair, Aznar and Kohl were able to accede to the thus conceived European Employment Strategy (EES). There are two reasons why the EES thereafter did not turn out to be a stillborn child. First, the European Commission and the newly created Employment Committee used their leverage in due course to upgrade the process by pushing for quantitative indicators, rigorous benchmarking and made increasing use of ‘naming and shaming’ by implicitly ranking member states’ performances. During 1999 and 2000 two cycles of the Luxembourg Process were completed. National governments and administrations got used to drawing up Action Plans and even learned to accept yearly recommendations. Moreover, in 2000 the economic outlook was favorable. The Commission’s forecast predicted GDP growth of about three per cent for 2001 and 2002 and unemployment declining to a twenty-year low of under eight per cent in 2002 (European Commission 2000, 33-7). Hence, the EES had gained credit for being a smoothly working policy tool. 11

The crucial role Juncker had played was confirmed frequently in interviews the author conducted with Commission and national officials as well as ETUC representatives.

194

Armin Schäfer

Second, and more importantly, since Maastricht the actor constellation had changed. Europe was now governed by social democrats (see Graph 1 and 2). In 1997, for the first time in the 1990s a majority of governments were led by the left, a development that peaked in 1999 when eleven out of fifteen governments were headed by social democrats. Since then, this number has declined considerably to currently only four out of fifteen of EU-15. Thus, between 1998 and 2001 a supportive coalition for the EES existed, not only solidifying but even strengthening it. These governments were – at least principally – committed to welfare state reform, turning ‘vices into virtues’ (Levy 1999). The European Employment Strategy gave them a common vocabulary as well as a European complement to their national endeavors. Yet, the “magical return of Social Democracy” (Cuperus and Kandel 1998) brought the New Left to power, most prominently, New Labour in Britain. These modernized social democrats were far less interventionist than their predecessors and had come to accept the market. Hence, the EES of the Amsterdam Treaty “most clearly reflects the new centre-left programme of Tony Blair, which seems likely to dominate the Union’s social agenda in the years to come” (Pollack 2000: 269). When the Left faced the opportunity – though, within limits – to transform the EUagenda, it had itself been transformed.

Figure 1. Governments dominated by the left in percentage12 12

Based on a data-set of Schmidt (2003). 2003: own data. For 1997 and 2000 the data refer to the time of the Amsterdam and Lisbon summit respectively. Numbers in graph 1 refer to the

Beyond the Community Method

195

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Left

Right

Independent

Figure 2. Governments led by prime minister of the left in percentage Under the Portuguese presidency the soft method of coordination used in employment became generally accepted as combining Community action with subsidiarity, thus reconciling effectiveness and legitimacy. The Portuguese government led by the Social Democrat Antonio Guterres advocated transforming the procedures probed with the Employment Strategy – European guidelines, NAPs, peer review, and peer pressure – into a policy-making tool of its own right (Hodson and Maher 2001, 724). At the Lisbon summit in March 2000, the EES-approach was dubbed the Open Method of Coordination, defined by: ƒ “fixing guidelines for the Union combined with specific timetables for achieving the goals which they set in the short, medium and long terms; ƒ establishing, where appropriate, quantitative and qualitative indicators and benchmarks against the best in the world and tailored to the needs of different Member States and sectors as a means of comparing best practice;

percentage of ministers of the left (Communists, Socialists, Social Democrats and Greens) and the right (Christian Democrats, Centrists, Conservatives, Liberals and the non-regionalist Right). Others refers to ministers or prime ministers without party affiliation; in graph 2 these mainly are technocratic governments in Italy.

196

Armin Schäfer

ƒ translating these European guidelines into national and regional policies by setting specific targets and adopting measures, taking into account national and regional differences; ƒ periodic monitoring, evaluation and peer review organized as mutual learning processes” (European Council 2000, § 37). At the same summit, heads of state and government invented the ‘Lisbon strategy.’ They agreed to a new strategic goal for the next decade: “to become the most competitive and dynamic knowledge-based economy in the world, capable of sustainable economic growth with more and better jobs and greater social cohesion” (European Council 2000, §5). The strategy aimed at full employment and thereby strengthening social cohesion. The Council suggested that the overall goal should be raising the EU employment rate to 70% and to increase the number of women in employment to more than 60% by 2010. Today, after fire years the European Union falls far short of the Lisbon goals and is in danger of missing them altogether. At its summit, the European Council concluded that “the pace of reform needs to be significantly stepped up if the 2010 goals targets are to be achieved” (European Council 2004, §7). Somewhat schizophrenically, the member states demanded improved implementation and a speedier translation of the goals into concrete measures (ibid, §10) – as if someone else was responsible for their own inaction. This gap between solemn declarations and real actions points towards an explanation of OMC’s attractiveness for governments. A neo-voluntarist, soft mode of governance:13 ƒ allows for flexibility (policy initiatives can be adapted to the diverse institutional arrangements, legal regimes and national circumstances in the member states); ƒ minimizes sovereignty losses (member-state compliance remains essentially voluntary. Compliance is not enforceable); ƒ helps superseding political resistance (flexible agreements may overcome member states’ opposition to EU mandates considered too disruptive of or too alien to national arrangements); ƒ permits shifting blame (it enables domestic actors to blame the EU for unpopular decisions); ƒ offers symbolic politics (new initiatives foster the impression of determination. European governments jointly take action against unemployment, social exclusion, and for economic growth);

13

The following draws on Abbott and Snidal (2000, 436-43). See also Schäfer (2004).

Beyond the Community Method

197

ƒ avoids agency losses (member states stay firmly in control of politics. Rather than enabling supranational actors to follow their own agenda [principal-agent problem], member states use the Open Method to shield them from outside interference – unless it turns out to be useful for their own domestic aims. An internationalization of domestic decisions strengthens executive actors [Wolf 1999]). Whereas EMU and prior integration largely explain the limits in substance of the Open Method of Coordination, its form owes to these general features of soft law. Although developed incrementally, it has proved a useful tool for intergovernmental cooperation beyond the Community Method. Governments have willingly applied it to new fields.14 The OMC has not replaced the Community Method but it has widened the menu member states can choose from. Governments rely on it when they favor common action though remain anxious not to lose control (Dehousse 2002). Arguably, governments have learned to avoid instruments which threaten to incur lasting agency losses. If we look at the origins of the OMC, four points deserve attention. First, neither EES nor OMC call into question the predominance of the European Monetary Union. They are not meant to challenge the logic put forward in the Maastricht Treaty but instead to supplement it. Lisbon cannot be considered ”Europe’s Maastricht for welfare” (Rhodes 2000) as the institutionalization remains much weaker. Both delegation and obligation are extremely low for various OMC-procedures. Second, since it was inaugurated in 1994, employment policy has relied on a voluntarist and intergovernmental coordination procedure. There has never been as strong a coalition in favor of employment policy as there was in Maastricht for monetary policy. Whereas center-right governments accepted to impose strict rules for fiscal policy also on themselves to lock-in their successors, social democratic governments were reluctant to transfer power and, thus, to create equally binding rules for employment policy. They instead favored more intergovernmental modes of cooperation. As demonstrated elsewhere, the level of support for European integration – enhanced supranational competencies – systematically co-varies with shifting political majorities. While the Christian democrat-liberal coalition at Maastricht scored high on integration-mindedness, support for further transfers of power subsequently plummeted with the ascending social democratic majority (Manow et al., 2004). Moreover, the left was afraid that the credit for fighting unemployment – an electoral asset for center-left parties – would then go to the EU (Ladrech 2003, 119). 14

For an overview see Hodson and Maher (2001: 726).

Armin Schäfer

198

Third, the European Employment Strategy was created during the run-up to EMU. Any radical departure from the Maastricht framework would have jeopardized successfully launching monetary union. Moreover, the institutions earlier created, most notably the European Central Bank and the SGP, enshrined a macroeconomic logic that precluded utilizing monetary policy or fiscal policy for fostering growth. The ‘initial assignment’ puts pressure on labor market liberalization and wage policy as the way towards accelerated growth. Monetary and fiscal policies, on the other hand, respectively have to secure price stability and signal a clear commitment to responsible government action. Fourth, across sectors, integration does not automatically lead towards further integration. Quite contrary to the neo-functionalist logic of spillover, the far-reaching integration in monetary policy might have prevented a stronger form of economic coordination or a more meaningful employment policy. We only detect these ‘spillbacks’ by looking at the longer sequence since Maastricht. While the preponderance of EMU explains the contents of the EES, governments’ reluctance to transferring power accounts for its form, that is, for why soft law was chosen. The OMC was selected to limit Europeanization; it was the attempt of social democratic governments – who inherited institutions built by their center-right predecessors – to craft a social policy onto the exiting framework without unduly restricting their own room for maneuver.

4.

CONCLUSION

What factors account for selecting either hard or soft law in EU economic policy? Converging economic ideas and a supportive actor constellation rendered possible an agreement on EMU in Maastricht. Restricting transfers of sovereignty to monetary policy facilitated this agreement. Moreover, the decision to model the ECB on the Bundesbank secured German consent. The Maastricht Treaty led to far-reaching albeit narrow integration built on hard low. Whereas monetary policy was centralized and a set of rules devised for fiscal policy, all other fields of economic policy-making rested with the national level (Cameron 1998, 213). However, multilateral surveillance – a soft coordination procedure – was introduced for aligning national decisions. As EMU was based on the monetarist paradigm, the institutional framework created with Maastricht firmly entrenched supply-side policies. From the mid-1990s onwards, center-left governments came to office pushing to advance employment policy at the European level. However, due to prior integration available options were limited. Without putting EMU

Beyond the Community Method

199

into question, employment policy could only focus on labor market reform. Furthermore, the employment chapter of the Amsterdam Treaty was only acceptable to all governments because it refrained from delegating decisionmaking competencies. Reluctant governments made sure that there would not be any strong outside interference with national decisions. Soft law was a means to reconcile a common approach with the national prerogative for action. Yet, contrary to harder – legally binding – modes of governance, the OMC does not create the kind of lock-in that Pierson detected as a mechanism that propels integration. At Maastricht, the British Conservatives had to opt-out of the Social Protocol to avoid its consequences. After Amsterdam, this is no longer necessary since soft law does not bind reluctant governments in the first place. In contrast to the existing literature, I would argue that with the OMC governments have found a way to prevent further integration. They have learned avoid unwanted gaps in control. Combining European guidelines, national decisions and multilateral discussions became a new policy instrument called the Open Method of Coordination at Lisbon. It has been extended to more than ten other policy areas since then. A nonbinding form of cooperation makes it easier to find compromises and seems particularly advantageous to governments as it widens their strategic room for maneuver. Accordingly, the OMC became a way of cooperating beyond the traditional Community Method. What does the analysis presented in this chapter have to say about the impact of the OMC? Will Europeanization be strong or weak? First, soft law is not necessarily ineffective. However, its impact will more strongly depend on the will of national governments than harder forms of policy-making do. Second, not only integration depends on supportive actors but so does Europeanization, especially in the case of nonbinding procedures. Between 1997 and 2001 – when many of the OMC-processes were initiated – most governments were led by social democrats. One might well assume than many of them domestically supported the measures of the EES. Yet, this situation by now has changed. Presently, in eleven out of fifteen OLD member states there are center-right parties in office. That might well mean that the momentum for combining equity and efficiency in welfare state reform will be lost. If governments opt for further retrenchment, the Open Method of Coordination will be unable to stem the tide. While it is still too early to tell exactly what effects OMC will have, we can nevertheless conclude that the Maastricht center-right coalition was far more successful in binding its successors’ hands than the Amsterdam center-left coalition.

200

Armin Schäfer

REFERENCES Abbott, K. W., Keohane, R. O., Moravcsik, A., Slaughter, A.-M. and Snidal, D., 2000, The concept of legalization, International Organization 54(3):401-19. Abbott, K. W. and Snidal, D., 2000, Hard and soft law in international governance, International Organization 54(3):421-56. Bernhard, W., Broz, J. L. and Clark, W. R., 2002, The political economy of monetary Institutions, International Organization 56(4):693-723. Biagi, M., 2000, The impact of the European Employment Strategy on the role of labour law and Industrial Relations, The International Journal of Comparative Labour Law and Industrial Relations 16(2):155-73. Cameron, D., 1998, Creating supranational authority in monetary and exchange-rate policy: The sources and effects of EMU, in: European Integration and Supranational Governance, W. Sandholtz and A. Stone Sweet, eds, Oxford, Oxford University Press, pp. 188-216. Cuperus, R. and Kandel, J., 1998, The magical return of social democracy: an introduction, in: Transformation in Progress. European Social Democracy, R. Cuperus and J. Kandel, eds, Amsterdam, Friedrich Ebert Stiftung / Wiardi Beckman Stichting, pp. 11-28. de la Porte, C. and Pochet, P., 2002, Introduction, in: Building Social Europe through the Open Method of Co-ordination, C. d. l. Porte and P. Pochet, eds, Brussels et al., PIE Lang, pp. 11-26. Dehousse, R., 2002, The Open Method of Coordination: A New Policy Paradigm?, Paper presented at the First Pan-European Conference on European Union Politics: “The Politics of European Integration: Academic Acquis and Future Challenges”, Bordeaux, 26-28 September 2002. Dyson, K. and Featherstone, K., 1999, The Road to Maastricht. Negotiating Economic and Monetary Union, Oxford, Oxford University Press. European Commission, 2000, The EU Economy: 2000 Review, European Economy 71. European Council, 1994, Presidency Conclusions, Essen, 9 and 10 December 1994. European Council, 1997, Presidency Conclusions. Luxembourg extraordinary European Council meeting on employment, Luxembourg, 20 and 21 November 1997. European Council, 2000, Presidency Conclusions, Lisbon, 23 and 24 March 2000. Goetschy, J., 1999, The European Employment Strategy: genesis and development, European Journal of Industrial Relations 5(2):117-37. Haas, P. M., 1992, Introduction: epistemic communities and international policy Coordi Coordination, International Organization 46(1):1-35. Hall, P. A. and Taylor, R. C. R., 1996, Political science and the three new institutionalisms, Political Studies 44:936-57. Hix, S., 1998, The study of the European Union II: the ‘new governance’ agenda and its rival, Journal of European Public Policy 5(1):38-65. Hix, S. and Lord, C., 1997, Political Parties in the European Union, Houndmills, Macmillan Press. Hodson, D. and Maher, I., 2001, The Open Method as a new mode of governance: The case of soft economic policy coordination, Journal of Common Market Studies 39(4):719-46. Hvinden, B., Heikkilä, M. and Kankare, I., 2001, Towards activation? The changing relationship between social protection and employment in Western Europe, in: Nordic Welfare States in the European Context, M. Kautto, J. Fritzell, B. Hvinden, J. Kvist and H. Uusitalo, eds, London/New York, Routledge, pp. 168-97. Issing, O., 2002, On macroeconomic policy co-ordination in EMU, Journal of Common Market Studies 40(2):345-58. Jenson, J. and Pochet, P., 2002, Employment and social policy since Maastricht: Standing up to the European Monetary Union, Prepared for The Year of the Euro, Nanovic Institute for

Beyond the Community Method

201

European Studies, University of Notre Dame, December 5-8, 2002: Canada Research Chair in Citizenship and Governance. Johansson, K. M., 2002, Another road to Maastricht: The christian democrat coalition and the quest for European Union, Journal of Common Market Studies 40(3):871-93. Kenner, J., 1999, The EC Employment Title and the ‘Third Way’: Making soft law work?, The International Journal of Comparative Labour Law and Industrial Relations 15(1): 33-60. Kruse, D. C., 1980, Monetary Integration in Western Europe: EMU, EMS and Beyond, London et al., Butterworths. Ladrech, R., 2003, The left and the European Union, Parliamentary Affairs 56:112-24. Lange, P., 1993, Maastricht and the social protocol: Why did they do It?, Politics and Society 21(1):5-36. Levy, J. D., 1999, Vice into virtue? Progressive politics and welfare reform in continental Europe, Politics & Society 27(2):239-73. Manow, P., Schäfer, A. and Zorn, H., 2004, European social policy and Europe’s partypolitical center of gravity, 1957-2003, MPIf G Discussion Paper 04/6:Cologne. Moravcsik, A., 1998, The Choice for Europe. Social Purpose and State power from Messina to Maastricht, London, UCL Press. Pierson, P., 1996, The path to European integration: A historical institutionalist analysis, Comparative Political Studies 29(2):123-63. Pochet, P., de la Porte, C. and Room, G., 2001, Social benchmarking, policy making and new governance in the EU, Journal of European Social Policy 11(4):291-307. Pollack, M. A., 1994, Creeping competence: The expanding agenda of the European Community, Journal of Public Policy 14(2):95-145. Pollack, M. A., 1997, Delegation, agency, and agenda setting in the European Community, International Organization 51(1):99-134. Pollack, M. A., 2000, A Blairite treaty: Neo-liberalism and regulated capitalism in the Treaty of Amsterdam, in: European Integration after Amsterdam. Institutional Dynamics and Prospects for Democracy, K. Neunreither and A. Wiener, eds, Oxford, Oxford University Press, pp. 266-89. Radaelli, C. M., 2000, Policy transfer in the European Union: Institutional isomorphism as a source of legitimacy, Governance 13(1):25-43. Radaelli, C. M., 2003, The Open Method of Coordination: A new governance architecture of the European Union?, Stockholm: Swedish Institute for European Policy Studies (SIEPS). Rhodes, M., 1995, A regulatory conundrum: Industrial relations and the social dimension, in: European Social Policy. Between Fragmentation and Integration, S. Leibfried and P. Pierson, eds, Washington, D.C., The Brookings Institution, pp. 78-122. Rhodes, M., 2000, Lisbon: Europe’s ‘Maastricht for welfare?’ ECSA Review 13(3). Risse, T., Cowles, M. G. and Caporaso, J., 2001, Europeanization and domestic change: Introduction, in: Transforming Europe. Europeanization and Domestic Change, M. G. Cowles, J. Caporaso and T. Risse, eds, Ithaca/London, Cornell University Press, pp. 1-20. Rodrigues, M. J., 2001, The Open Method of Coordination as a new governance tool, in: L’evoluzione della governance europea, special issue of Europa/Europe, M. Telò, ed., Rome, pp. 96-107. Sandholtz, W., 1993, Choosing union: Monetary politics and Maastricht, International Organization 47(1):1-39. Schäfer, A., 2004, A new form of governance? Comparing the Open Method of Coordination to multilateral surveillance by the IMF and the OECD, MPIfG Working Paper 04/5:Cologne. Scharpf, F. W., 1999, Governing in Europe: Effective and Democratic?, Oxford, Oxford University Press.

202

Armin Schäfer

Scharpf, F. W., 2002, The European social model: Coping with the challenges of diversity, Journal of Common Market Studies 40(4):645-70. Schmidt, M. G., 2003, Die parteipolitische Zusammensetzung von Regierungen in 23 OECD Demokratien 1945-2002, Universität Bremen, Zentrum für Sozialpolitik, Unpublished data set. Szyszczak, E., 2000, The evolving European Employment Strategy, in: Social Law and Policy in an Evolving European Union, J. Shaw, ed., Oxford, Portland, Hart, pp. 197-220. Thelen, K., 1999, Historical Institutionalism in Comparative Politics, Annual Review of Political Science 2:369-404. Thelen, K. and Steinmo, S., 1992, Historical institutionalism in comparative politics, in: Structuring Politics: Historical Institutionalism in Comparative Analysis, S. Steinmo, K. Thelen and F. Longstreth, eds, Cambridge/New York, Cambridge University Press, pp. 1-32. Trubek, D. M. and Mosher, J. S., 2003, New governance, employment policy, and the European social model, in: Governing Work and Welfare in a New Economy: European and American Experiments, J. Zeitlin and D. Trubek, eds, Oxford, Oxford University Press, pp. 33-58. Tsoukalis, L., 1993, The New European Economy. The Politics and Economics of Integration. Second revised edition, Oxford, Oxford University Press. von Hagen, J. and Mundschenk, S., 2001, The functioning of economic policy coordination, ZEI Working Paper B 08, Bonn: Zentrum für Europäische Integrationsforschung. Wolf, K. D., 1999, Defending state autonomy. Intergovernmental governance in the European Union, in: The Transformation of Governance in the European Union, B. Kohler-Koch and R. Eising, eds, London, Routledge, pp. 231-48.

EUROPEAN INTEGRATION AND UNFREEZING TERRITORIALITY The Case of the European Health Card Hans Vollaard Abstract:

1.

In 2004, a European health card has been introduced to cover the costs of cross-border emergency care in the European Union (EU). Facilitating EU citizens to seek healthcare outside the domestic welfare systems, the health card challenges the territorial organization of healthcare in the member states. This chapter seeks to explain when and how territory would matter more, less or differently for the organization of healthcare in the EU-area. The first and second waves of EU-studies lack explanations of changing territoriality – briefly defined as politics organized according to territory. This negligence stems from the territorial divide between theories on politics within the territorial states, and theories on politics among territorial states. The European health card is used as an illustration of how Stein Rokkan’s notions on polity (re)formation helps to transcend this territorial divide in explaining changing territoriality. Reasoning along Rokkan-Hirschmanian lines, the European health card seems to be a de-territorializing means to sustain the legitimacy of national health authorities. This rescue of the national healthcare systems coincides with a re-emergence of territoriality at a European level.

THE EUROPEAN CHALLENGE TO TERRITORIALITY

After 25 years of discussion, the Council of the European Union and the European Parliament agreed on the introduction of a European health insurance card from 1 June 2004. This card for socially insured citizens of the European Union (EU) Member States simplifies administrative procedures to acquire healthcare within the territory of the European Economic Area (EEA). Facilitating most EU citizens to seek healthcare outside their domestic welfare systems, the adoption of the health card raises two fundamental questions about the territorial organization of the healthcare states in the EU/EEA-area. Facilitating national citizens to obtain cross-border

203

R. Holzhacker and M. Haverland,(eds.), European Research Reloaded: Cooperation and Integration among Europeanized States, 203-227. © 2006 Springer. Printed in the Netherlands.

Hans Vollaard

204

healthcare, questions the territorial sovereignty and control of national health authorities. Furthermore, citizens’ increasing opportunities of exiting and entry to healthcare systems may undermine the national solidarity and financial fundaments of the territorially confined healthcare states. In the eyes of Klaus Offe (1998, 102, 111), this ‘Entgrenzung’ (de-bordering) and the consequent decline of ‘Anwesenheitzwang’ (coercion to stay) will lead to the ‘Entplfichtung’ (de-obligationing) in the membership of healthcare states. Thus, European integration seems to pose a serious challenge to one of the fundamental sources of the legitimacy of states and their governments: a satisfactory ‘output’ of healthcare. Uneven patterns of patient mobility within the EU/EEA area have also evoked voices for controlling the healthcare burden at a European level. As a consequence, the unfreezing of state borders may entail a far-reaching (territorial) reconfiguration of healthcare resulting in healthcare organized in a multi-level Europe. Although territory is often mentioned as an important element in the political reconfiguration of Europe, both first-wave and second-wave EUstudies suffer from a lack in theorizing the relation between European integration and territoriality. Section 2 expounds how these EU-studies did not and could not address the implications of European integration for political territoriality – shortly defined as politics organized according to territory. The section subsequently presents an explanation to explore the relation between European integration and political territoriality. The explanation is based on the works by Stein Rokkan, one of the few political scientists who theorized changing territoriality. The case of the European health card is used as an illustration of the European challenge to political territoriality, and in section 3 Rokkan’s theory is applied to the case of the European health card. Section 4 concludes on the significance of territoriality in the organization of healthcare in a multi-level Europe.

2.

THEORIZING CHANGING TERRITORIALITY

2.1

Defining territoriality

Theorizing changing territoriality follows a recent call by Miles Kahler. In a recent state of the art overview of state theory, he argued: ‘Model[ling] the institutions of territoriality, which are central in defining state and unit variation, should become a central part of the institutional research agenda on the state’ (Kahler 2002, 79). In his view, the oversimplification in debates on whether politics is going beyond ‘the’ state due to globalization and European integration will be overcome by more refined studies of one of the states’ ‘institutional variables’, territoriality. A refined approach of territoriality

European Integration and Unfreezing Territoriality

205

will help to provide the necessary precision of the avalanche of concepts varying in solidity and clarity to describe the authority amalgam in the EU. Robert Sack is one of the few scholars dealing with territoriality in depth. He defines political territoriality as ‘the attempt by an individual or group to affect, influence, or control people, phenomena, and relationships, by delimiting and asserting control over a geographic area’ (Sack 1986, 19). Territoriality is an effective and efficient means of exclusionary control, classification and communication: ‘Territoriality classifies, at least in part, by area rather than by type. (…) We need not stipulate the kinds of things in place that are ours or not yours. Thus territoriality avoids, to varying degrees, the need for enumeration and classification by kind and may be the only means of asserting control if we cannot enumerate all of the significant factors and relationships to which we have access. This effect is especially useful in the political arena, where a part of the political is its concern with novel conditions and relationships. Territoriality can be easy to communicate because it requires only one kind of marker or sign – the boundary.’ (Sack 1986, 32). A territorial logic of organization is an effective disciplinary instrument of surveillance from a hierarchic centre, because geographically centralized supervision requires only a few guards within a clearly demarcated area to keep outsiders out, and insiders in. This hierarchical territorial circumscription of responsibility also facilitates a clear system of impersonal accountability for everything and everyone within the area. Classification by territory also simplifies efficient planning. Territorial planning can separate the area of planning from the actual things, persons, dynamics, and events within that area. Enumeration by kind would make surveillance, accountability and planning much more complex and less visible. Defining territoriality in this way means that a process of de-territorialization does not only involve people and governments caring less about territory in their political behavior. It also refers to the territorial logic of exclusionary control and planning losing its significance in organizing policies, polities and politics.

2.2

The territorial divide

The histories of the European states have been closely intertwined with political territoriality. The Peace Treaties of Westphalia (1648) are usually seen as starting point for a Europe increasingly ordered by state territoriality. The template of the Westphalian states’ territoriality has become a mutually exclusive, hierarchic, centralized, impersonal and sovereign entities having clearly demarcated, contiguous and fixed territories (cf. Weber 1956, 27; Ruggie, 1993).

206

Hans Vollaard

European integration seems to challenge the Westphalian organization of politics. Unfortunately, theorizing when and how territory matters more, less or differently for political organization is a somewhat neglected issue in political science. Although territoriality is often perceived politically significant, it has been barely theorized. That holds for International Relations, Regionalism and Federalism studies, Comparative Politics as well as Political Geography (Forsberg 1996; Ruggie 1993; Keating 1998, 1; Newman 2001; Rokkan and Urwin 1983, 1). Studies of European integration and Europeanization also lack refined explanations of changing territoriality. ‘First wave treatises’ on political transformation in Europe basically assume a certain prospect of territoriality. Functionalists expects that crossborder technocratic co-operation in policy-areas will eventually dissolve territorial states, whereas the enduring significance of the principle of territoriality is key to realism. Furthermore, intergovernmentalists expect territorial states to survive due to European integration, while neofunctionalists and federalists expected the outcome of European integration to be a European super-state divided in respectively functional or territorial subunits (Risse 1996, 57). Although a territorial outcome of European integration might eventually be proven to be empirically correct, it should not be assumed before the evidence is given. Explanations of the formation of Euro-polity and transformation of the EU Member States should transcend the territorial state bias, but should not exclude the territorial state as a potential outcome. Moreover, the significance of territoriality is not just to be narrowed down to a yes/no question on territorialization or deterritorialization. The implications of European integration may instead lead to a re-territorialization of political structures deviating from the Westphalian template of state territoriality (cf. Forsberg 1996). ‘Second wave studies’ of Europeanization have touched upon the issue of politics and territory, such as in contributions on the impact of European integration on intergovernmental relations in (semi-)federal Germany and Spain (Börzel 2001), and the position of the regions in a multi-level Europe (e.g., Le Galès and Lequesne 1998). Both studies of multi-level governance and Europeanization, however, focuses more on daily decision-making and substantial problem-solving than on the when and how of shifts in structures of power and rule in the EU-area: ‘Europeanisation tends to be discussed in substantive terms, while the procedural characteristics, which can only be adequately grasped from a longer-term perspective, are often neglected.’ (Hix and Goetz 2000, 16; see also Jachtenfuchs 2001). The fairly detailed empirical studies of Europeanized (parts of) policy areas may show bits and pieces of the new political configurations that emerge, but do not show an overarching view, and the fundamental polity-forming mechanisms of reterritorialization in the EU area.

European Integration and Unfreezing Territoriality

207

The lack of a thorough-going exploration of changing territoriality may be excused by the fact that studies of European integration and Europeanization did not seek explanations of the changing nature of (state) borders and the mechanisms of re-territorialization. However, this negligence of the issue of territoriality stems from a more fundamental conception of political life in political science. Since state borders are often used as the dividing line between theories explaining politics within territorial states and theories explaining politics among territorial states, (state) territoriality is overlooked too easily. This territorial divide hampers a full understanding of polity formation in the EU-area, and in particular the relation between European integration and territoriality. As Hans Daalder (1993, 29) put it: ‘The long-standing assumption of a natural division of labor between the study of international relations engaged in analyzing the interactions of states, and comparative politics concerned with the study of processes within states, always rested on somewhat dubious ground. It left unclear what scholars were to handle the formation of (new) states; it glossed over the importance of domestic political processes on the making of foreign policies; it belittled what became known in the international relations literature as ‘transnational’ politics; and it postulated a degree of political independence for ‘sovereign’ states which never completely fitted the realities of an interdependent world.’ This division of labor is also visible in the distinction between studies of Europeanization and European integration. This might be for good reasons, but goes at certain costs: ‘While it is necessary to disentangle European integration as an explanatory variable and the European effect as a dependent variable, both will need to be “re-entangled” if we wish to do justice to the real-life interdependency of forces of economic, political and cultural change.’ (Hix and Goetz 2000, 23). In similar vein Johan Olsen (2001, 21) argues in an overview of Europeanization literature: ‘A focus on uni-causal relations and the language and logic of fixed dependent and independent variables, can become a strait jacket preventing an adequate theoretical and empirical analysis of European dynamics of change. However, no coherent empirical research programme is possible if everything is seen as endogenous and in flux.’ Be that as it may, an explanation of changing territoriality should encompass instead of assume ‘the territorial divide’ between politics within territorial states affected by the EU (comparative politics; Europeanization studies), and politics among territorial states within the EU (International Relations; European integration studies). Taking states and their borders (implicitly) for granted prevents the analysis of changing borders and the territorial underpinnings of polities. Criticizing ‘the “statecentered” account of spatiality of power’, John Agnew (1998, 49) describes the ‘territorial trap’ in analyses of polity formation, be it at national, global

208

Hans Vollaard

or regional level: the geographical assumptions of territorial state sovereignty, the separation of domestic and foreign realms of politics in terms of analyses and rules, and the designation of distinct societies by state borders precludes analyses of polity-formation that question these static assumptions. The territorial divide in analyzing politics in the European Union easily allures to perceive European integration and its impact (Europeanization) as hollowing out territorial states, while (re)nationalization is seen as an automatic loss of European influence in politics (Clark 1998, 480). Domestic and foreign politics are, however, not necessarily zero-sum games as the reasoning of functionalism and realism/intergovernmentalism imply. The loss of state authority over economic, cultural, social or information systems may lead to simultaneous shrinkage and expansion of the foreign and domestic realms, or to a differentiated impact depending on place or policy area. Similar to the debates on the relation between globalization and ‘the’ state, European integration and Europeanization are fully intertwined with the reconfiguration of the states, and cannot be studied from a fixed image of a territorially divided world: ‘It is futile to locate sources of explanation and outcomes of political processes at specific substructures in world politics, for the very reason that these substructures are too interdependent to allow for ontological differentiation.’ (De Wilde forthcoming, 4; see also Clark 1998, 489). A third wave of EU-studies should therefore overcome the (territorial) fragmentation in theory and research practice to explore the European challenge to (Westphalian) territoriality. Some scholars have already attempted to overcome the territorial divide in analyses of political reconfiguration. They have aimed at finding scientific tools to avoid thinking about political reformation based on the template of state territoriality. In International Relations theories scholars like John Ruggie (1993), James Rosenau (1997) and Yosef Lapid et al., (2001) have posed the question how to deal theoretically with transformative change beyond the ‘Idealtypical’ state. Ruggie and Rosenau were the ones who raised the issues of unbundling territoriality and governance without government, thus offering terms and wordings to grope in the dark ages after the modern state. Concepts as neo-medievalism and empire were issued to ‘free our minds from the confines of modern state system, whose assumptions prejudice us to the degree that they obscure new developments’ (Wæver 1995, 424; see further Bull 1977; Anderson 1996). Lapid et al., introduced the analytical triad of ‘Identities, Borders, Orders’ (IBO) as an ‘antidote to reification and essentialism’ known in statebased Realism and Liberal-Institutionalism (Lapid et al., 2001, 11). This triad gives way to an awareness of the continuous efforts to (re)set the boundaries of political identities in the processes of political transformation. This IBO-approach is a ‘heuristic tool’ to get rid of static assumptions of ‘the’

European Integration and Unfreezing Territoriality

209

territorial state, focusing on the processual, mobile, fluid, relational, dynamic, and discursive nature of politics instead. Although these concepts and approaches are helpful for not taking borders for granted, they lack a fine-grained explanation of when and how political territoriality is changing in significance. In effect, the same argument largely accounts for the multilevel governance approaches (MLG). This ‘compelling metaphor’ (Rosamond 2000, 110) has helped to understand that the making and implementation of EU-policies is running through less territorialized, non-hierarchic, decentralized, and flexible networks, it gives no explanation as to when and how political territoriality is of different significance at regional, state or EU-level. Theoretical connections between the domestic and foreign political realms have been put forward by Keohane and Milner (1996) following the work by Peter Gourevitch (1978) on the international impact on domestic politics (the so-called ‘second image reversed’ approach). They scrutinize the mutual interdependence of the international structure and the domestic configuration of power. Although they spend considerable effort to show this linkage, they do not question the nature of the divide between domestic and foreign politics itself. They rather analyze the second-image reversed, than the third and second image merged. Nevertheless, Keohane and Milner (1996, 19, 244, 250) hint at the work by Albert Hirschman, which may provide the way out of territorially divided analyses of political formation. Stein Rokkan has done so.

2.3

Theorizing changing territoriality: exit, voice and loyalty

Inspired by famous political geographers like Friedrich Ratzel, Richard Hartshorne and Derwent Whittlesey, Stein Rokkan provides the building blocks for an explanation of changing territoriality. His theoretical notions regarding mechanisms of state formation are of considerable help to explore present-day processes of the formation of the Euro-polity and reformation of (healthcare) states (see Peter Flora in Rokkan 1999). In particular, Stefano Bartolini did a magnificent job in applying Rokkan’s notions to polity(re)formation in the EU-area (Bartolini 1998). Rokkan perceived polity-formation as a continuous tension between forces of boundary-transcendence versus boundary control. He described the ensuing interdependence of polities’ internal structuring and external consolidation in terms of Hirschman’s ‘exit, voice and loyalty’ (Hirschman 1970). Hirschman’s taxonomy intends to analyze the behavior of dissatisfied members in any social organization. Hirschman’s basic position is that (aggregated) actions from dissatisfied individual members eventually generate pressure on the management, rulers or authorities of an organization

210

Hans Vollaard

to reform and thereby to recuperate the entire system. Polity (re)formation and fluctuations in political authority thus start with dissatisfaction. Two mechanisms by which members express dissatisfaction can start the processes that mould individual actions into systemic reforms: exit and voice. Members could choose to leave an organization if they are dissatisfied with its values. This escape is called “exit”. Such an exit can be physical by moving (e.g. voting by feet), and institutional (e.g. canceling membership). Rokkan further distinguishes ‘primary exits’ from ‘secondary exits’. A primary exit comprises the institutional and physical opportunities to escape. A secondary exit is the actual escape from the organization. Voice is ‘any attempt at all to change, rather than to escape from, an objectionable state of affairs, whether through individual or collective petition to the management directly in charge, through appeal to a higher authority with the intention of forcing a change in management, or through various types of actions and protests, including those that are meant to mobilize public opinion.’ (Hirschman 1970, 30) “Voicers” try to change the organization by articulating and aggregating their desires and grieves, desiring a satisfying improvement of the system and its products, or an expansion of their say into the organization’s affairs. These voices can be expressed through formal and informal channels for complaints, representation, and protestation. Members may voice their dissatisfaction, because they have no option to leave the system. Even if members do have such an option, they may be too loyal to the system and choose for voicing instead of leaving the deteriorating system. Loyalty (attachment to the organization) intervenes with considerations of exit and voice. Loyalty may even restrain members from voicing, as their commitment to the organization prevents them from doing so. In choosing between exit and voice, members also consider the possibilities for improvement, and the costs of letting their voice be heard and effective: ‘…the conditions implying the resort to voice when exit is available depend upon: 1) the extent to which members are willing to trade off the certainty of exit against the uncertainties of an improvement in the deteriorated product; 2) the estimate members have of their ability to influence the organization.’ (Hirschman 1970, 77) Another consideration for members is the uncertainty of an exit. The costs of exiting to an unknown organization may be higher than voice in a known organization. The more a member knows about another organization, the easier exiting becomes. (Hirschman 1970, 81) A member’s calculation of costs and benefits further depends on the infrastructure to exit and voice. When institutions exist to voice dissatisfaction, the costs are quite different than in a system without or with weak voice institutions. Institutionalization of voice is much more expected in externally closed polities, since fellow-voicers have few opportunities to defect from the

European Integration and Unfreezing Territoriality

211

collective efforts to express dissatisfaction. Moreover, the costs of collectively organizing voice are perceived to be lower as voice institutions can be used again. Because voice behavior is repeated, uncertainty and even distrust gradually disappear in favor of mutual and standardized expectations. This stabilization of behavior into institutionalized and standardized patterns would not occur if members could easily choose to exit. (Bartolini 1998, 36-37) This routinized behavior within a closed polity entails mutual loyalty and the socialization of a political identity, making voice an even more secure instrument of expressing dissatisfaction in members’ “own” polity. (Dowding et al., 2000, 492) Blocking entrance of outsiders also prevents the tempting voices to select exit. Based on this Hirschmanian reasoning, Rokkan’s basic contention on polity-formation is the mutual interdependence of a polity’s external consolidation (boundary demarcation) and its internal structuring(institutionalizing voice and loyalty). Boundary demarcation is a continuous struggle between ‘boundary transcendence’ (exit) and ‘boundary maintenance’. The maintenance of boundaries can be pursued by strengthening the boundaries, but also by attaching members to the polity more closely. External consolidation enables authorities to extend their central grip and the setup of an effective and efficient internal hierarchical organization, since authorities know better to whom and what value satisfaction should be directed. Thus, planning to allocate values is easier for a fixed group within an externally consolidated polity. To summarize Rokkan’s main argument somewhat crudely regarding state formation, the territorial closure of states eventually enhanced their internal structuring in the establishment of a secure, liberal and democratic welfare state. State rulers established borders to block members’ exits and non-members’ entry, and fostered loyalty to them and the state. This external consolidation allowed for internal institutional developments of participation by the state’s residents through the judiciary and mass democracy (voice), and internal solidarity through redistribution within a territorially closed community. Depending on the geographical and social opportunities for members to escape from states’ territories, the internal structuring of states became tighter. (Hirschman 1974, 16) The territorial closure of states entailed the centralization of organization, control and planning, enduring and impersonal structures of accountability, and exclusivity of membership – the territorial logic of organization. (Mann 1986) In the EU, the dynamic between internal structuring and external consolidation seems to work in the reverse. EU-policies are to open states’ territories and to establish cross-border institutions such as the European Commission. This would consequently imply a weakening of the internal structuring of the EU-Member States (as long as they do not anticipate or respond to this potential weakening). Nowadays, members may choose to

Hans Vollaard

212

exit the state territory instead of voicing within a state’s democracy in case of dissatisfaction. However, Member States may foster national loyalty to prevent members from leaving: ‘whereas today the general talk is about the decline of the nation State, Rokkan’s perspective also leaves open the possibility that the process of European unification may lead to a revivification of national thinking and acting’. (Flora 2000) From a Rokkanian perspective the internal structuring of the Euro-polity might be difficult, since its boundaries are unconsolidated due to the various geographic enlargements and the increase of co-operation in many policy domains. The constitutionalization of the Euro-polity through the recent series of Intergovernmental Conferences may keep the internal structuring strong enough. Rokkan’s theory is a worthwhile one to explore the territorial reconfiguration within the EU-area, since it meaningfully links legitimacy (dissatisfaction and exit), democracy and legal participation (voice), solidarity (loyalty), and territoriality (borders). Moreover, Hirschman’s taxonomy has been developed to analyze responses to the (declining) performance of an organization in the delivery and allocation of values among its members. Thus dealing with value (dis)satisfaction, it might be perfectly suitable in analyzing contemporary polities in which legitimacy is increasingly based on output and performance in welfare instead of warfare. The Rokkan-Hirschmanian framework also follows the dominance of neoinstitutionalism in Europeanization research, because it perceives the trajectory of the territorial (re)formation of both the EU Member States and the Euro-polity dependent on the institutional paths taken by them and their subunits, as well as the present settings and environment they are in. Comparing the EU and its Member States to older processes of state formation does not imply that the EU would become a state or that the Member States will remain Westphalian states. Like scholars before argued that it does not make sense to develop a theory of Sweden or England, here theoretical notions on the mechanisms of polity formation are applied to a comparative case of the formation of the Euro-polity. The case of the European health card provides an insight into how the Rokkanian internalexternal dynamic may re-territorialize the organization of healthcare in the EU area.

3.

THE CASE OF THE EUROPEAN HEALTH CARD

3.1

Healthcare states in Europe

Healthcare systems in Europe have basically been developed within the territorial confines of the national states. Since the Second World War they

European Integration and Unfreezing Territoriality

213

have been gradually extended towards almost universal, obligatory insurance or service coverage of citizens’ basic health needs. They take up a considerable share of states’ public expenditures, labor force, and gross domestic product. Although the European systems are all (in)directly regulated by states, the organization, financing and delivery of healthcare differ from country to country. Two families can be distinguished among the European healthcare states (Moran 1999; European Parliament 1998). The ‘command-and-control healthcare state’ is characterized by a state-guaranteed universal health insurance covering citizens’ basic health needs, the planning and provision of mainly publicly owned national health service, the funding of healthcare through taxation, and political decision-making at national, regional or local level. Costs of supplementary health could be covered by private voluntary insurance or direct payments. This model can be found in the United Kingdom and Scandinavia. Within a ‘corporatist healthcare state’ insurance and provision of healthcare is largely in the hands of public law bodies – health insurance funds (or sickness funds or mutualities) and hospitals - in which health professionals’ associations, and social partners have a large say. This type of healthcare state is largely financed through a social insurance system of income-related social security contributions. The state operates as a director of this corporatist amalgam, only showing its hierarchy in times of (financial) urgency. Countries like France, Belgium, Germany, and to some extent The Netherlands have corporatist healthcare systems. A majority of European respondents express themselves against the dismantling of their healthcare systems and cuts to their basic health package, and adhere to the principle of solidarity subscribing to the statement that health care rights of the lower incomes should not be diminished (European Commission 1998). Since citizens perceive governments both in commandand-control and corporatist responsible for affordable and timely access to healthcare, the issue of waiting lists have been a hot topic in recent elections in the United Kingdom and the Netherlands. Healthcare was the most important issue in the Dutch elections of 2002 and 2003, as 57% and 52% of respondents, respectively, mentioned healthcare as the most important issue in the Dutch Elections Studies and held the government responsible for solving the problems of the waiting lists (personal communication with Dr. J.J.M. van Holsteyn, director Dutch Election Studies 2002, 2003; Centrum voor Verzekeringsstatistiek 2002). Rising assertiveness and expectations among patients and the ageing population further heighten demand for healthcare. Since European governments have to curtail budgets in accordance with the EMU-norms (but sometimes don’t) and limit the burden of premiums and taxes to remain internationally competitive, the fulfillment of citizens’ healthcare demands is under pressure.

Hans Vollaard

214

3.2

Regulation 1408/71

Command-and-control centralization and corporatist entanglements have an inherent tendency to close off healthcare systems within state borders. Only some patients have made use of healthcare across these borders, because of the dissatisfactory state of healthcare at home or the insufficient availability of advanced top-clinical care. The cross-border consumption of healthcare took a fairly negligible share of the total health expenditures within the EU-area until the late 1990s (Palm et al., 2000). In principle, no barriers exist to access healthcare across borders. The real issue is the coverage of costs for healthcare obtained outside the healthcare states. Since 1958 regulations have existed to cover cross-border healthcare costs for socially insured workers by coordinating the participating social security systems based on the 1957 Treaty of Rome article to facilitate the freedom of workers. The present regulation 1408/71 provides an extensive procedure to determine the competent healthcare state to cover certain cross-border healthcare costs. When an employee plans to seek treatment in another Member State, he/she should request prior authorization from the competent health insurance institution based on an E-112 form. This authorization cannot be refused if two conditions are fulfilled, i.e. ‘1. the desired treatment is part of the employee’s healthcare package, and 2. the treatment cannot be given within the period that is normally necessary, in view of his current state of health and the probable course of his disease.’ Costs of immediately necessary care during a temporary stay abroad for professional or private purposes alike can be covered based on a so-called E-111 form. This is to be obtained from the health insurance institution at home in advance. In case of acute, unplanned health provision abroad, this latter institution will then cover the health costs. In certain cases, the foreign health insurance institution should also approve the E-111 form in advance to obtain reimbursement. The E-111 procedure is quite an administrative burden to assure only a potential treatment abroad. The right of coverage of healthcare abroad has been gradually extended from employees and their relatives to virtually all EU-citizens like selfemployed and their families, students (E-128 form), transport workers (E-110), pensioners, posted workers (E-128), unemployed persons looking for a job in another Member State (E-119) and civil servants, and in due time to all legally resident socially insured third-country nationals. Through agreements between the EU and the European Free Trade Area (Liechtenstein, Norway, Iceland) these co-ordination scheme are applicable within the European Economic Area (EEA). Since June 2002 the E-111 form is also applicable in Switzerland.

European Integration and Unfreezing Territoriality

3.3

215

Overruling the principle of territoriality

Regulation 1408/71 overrules the principle of territoriality of national healthcare systems, since the Member States in certain cases could no longer territorially delimit the coverage of healthcare costs, the consumption of health (even if a EU/EEA-citizen did not contribute), and the provision of healthcare: ‘the co-ordination system deterritorialises the national systems in order to ensure that migrants are entitled to benefits on the basis of their own insurance record.’(Van der Mei 2001, 75). The healthcare systems in the EU-area are no longer ‘closed shops’ within state borders (Leibfried and Pierson 1995, 50ff; 2000: 283). Insurance rights follow the worker wherever in the area of the Member States, meaning a ‘personalisation’ of previously territorially restricted rights (Watson 1980). While regulation 1408/71 originally aimed at the freedom of movement of workers, it has been expanded to the effect that ‘under certain circumstances, the fact that a person has never worked or resided in another Member State is not, as such, an obstacle to entitlement to medical care in another Member State.’ (Cornelissen 1996, 463) Membership of a healthcare system is less bound by the territories of the healthcare states within the EU/EEA area. If crossborder patient mobility would grow (in uneven patterns), territorial planning of the healthcare infrastructure will be hampered by under- or overutilization. Thus, by unfreezing territoriality European integration impacts on citizenship and planning of healthcare states. As long as cross-border patient mobility remained low, this encroachment on territorial sovereignty was not of great concern to the Member States. The creation of a Single European Market in the late 1980s gradually changed that. For instance, Dutch market-oriented reforms were accompanied by warnings that European legislation on competition and free movement would constrain this reform of the healthcare state. After a Conference on Health under the Dutch presidency in 1991 the EU-health ministers consequently ‘recognize that Member States need to make allowances for the effects that the completion of the internal market may have on the operation of health-care services and their nature and extent’. They therefore agreed to exchange information and initiate research on the impact of the internal market, leading to the introduction of experiments with cross-border healthcare in border regions (Council of Ministers 1994; Vollaard 2004). The European Commission, EU and EFTA Member States also initiated research on using information and communication technology in healthcare. They aimed at creating a coherent European infrastructure for healthcare information, standardization of terminology and software in a ‘European e-Health Area’ under the subsequent labels of AIM (Advanced Informatics in Medicine) and eEurope (European Commission 2004).

216

Hans Vollaard

The Member States’ health ministers insisted on their national prerogatives, as they stated in 1991: ‘it is a matter for the Member States to determine the organization and funding of health-care systems and to make fundamental health policy choices’ (Council of Ministers 1991). Although the 1991 Maastricht Treaty acknowledged Community action to ensure a “high level of human health protection”, the Member States jealously guarded their healthcare competencies in the European Commission’s public health programs: ‘The Council and the Ministers for Health point out that, public health policy as such, except in cases where the Treaties provide otherwise is the responsibility of the Member States’ (Council of Ministers 1993). However, the aftermath of the 1996-1997 BSE-crisis resulted into an extension of involvement of European Commission to improve public health in all Community policies and activities. Nevertheless, the Member States emphatically stipulate their exclusive competences in the Amsterdam Treaty (1997), as is repeated in the Draft Constitutional Treaty: ‘Union action in the field of public health shall fully respect the responsibilities of the Member States for the organization and delivery of health services and medical care.’ (European Convention 2003, Art. III-179, sub 7). Judgments by the European Court of Justice (ECJ) in the 1998 shattered Member States’ image of exclusive prerogatives in healthcare (ECJ 1998). According to the Court’s interpretation of European law, patients are allowed to obtain reimbursement of certain cross-border healthcare since in principle the freedom of services and goods also accounts for medical treatments and devices. This interpretation caused much upheaval among national health authorities because it implied that – in the words of a Dutch health official – the ECJ’s interpretation ran counter to the principle of sovereignty and potentially open the gates of national healthcare systems without any restraint. In later cases, the Court further explained its position: Member States are allowed to restrict the freedom of services and goods for intramural care provided that it is motivated upon objective, nondiscriminatory criteria of public health, financial equilibrium, planning and accessibility of the intramural healthcare facilities. Furthermore, the Court ruled in May 2003 that patients are free to purchase extramural care across borders while keeping the right of reimbursement (ECJ 2003). Since most Member States considered healthcare as their exclusive competence, talks on the single market’s impact on their healthcare systems were initially strictly informal. The Belgian minister for Health Vandenbroucke pushed for a discussion on cross-border patient mobility and the consequences of European legislation for national healthcare systems, confronted with an increasing number of Dutch and British patients visiting Belgian hospitals. Only the Dutch and British minister (concerned with waiting lists) the Spanish (confronted with hibernating foreign pensioners using Spanish healthcare), and to some extent the German minister of health

European Integration and Unfreezing Territoriality

217

were eventually willing to join him. In an informal meeting under Spanish EU-presidency, the health ministers did nevertheless acknowledge that the legislation on the internal market and the interpretations by the European Court of Justice need them to discuss an issue like patient mobility. In July 2002 an ‘informal process of high level reflection on patient mobility and healthcare developments in the EU’ subsequently started (Health ministers 2002). Meanwhile, Member States’ concerns on unemployment, sluggish economic growth and social exclusion led them to decide to create the most competitive economy of the world, while preserving a Social Europe and a healthy ageing population (the so-called Lisbon strategy). Intergovernmental taskforces and the European Commission subsequently argued among other things for accessibility, quality and financial viability of healthcare systems in the era of ageing, and for geographic mobility in the internal market (European Commission 2001; 2002). One of the proposals that emerged from the discussions both on geographic mobility and patient mobility was the introduction of a European health card. The European Council of Barcelona, March 2002, requested the European Commission to present a proposal to introduce this card. Since June 2004 the health card has been issued.

3.4

The European health card

The coincidence of advancing information and communication technology, the political momentum of the Lisbon-strategy, and the growing attention to the issue of patient mobility set the stage for the European health card. Long before in 1978, the ministers of Health suggested that such a card would be helpful in particular in case of emergency care. The European Parliament subsequently adopted in 1981 a resolution concerning a European health card to replace the E-111 form, because of its complexity for patients and insurance institutions (European Parliament 1981) Another advantage to patients would be the correct transmission of health information in case of emergency care to a doctor, whom most patients have never consulted before (European Commission 1995). In 1986, the Council of Health Ministers adopted a proposal by the European Commission to introduce a voluntary health card to be issued by the Member States willing to. It conceived a uniform emergency card as a means to enhance freedom of movement of ‘European citizens’ helping to create a ‘People’s Europe’ (Council of Ministers 1986) It appeared in 1989 that most Member States did not take any action, and the Council requested the European Commission to conduct research to establish ‘a harmonized European social insurance card’ (Council of Ministers 1989).

218

Hans Vollaard

During the 1990s the technological feasibilities of a (European) health card were explored through AIM-research (in close co-operation with the G-8), and through the issuing of millions of national health cards in Germany and France. The standardization of medical terminology and health registration, harmonization of use of data cards in healthcare, and technical interoperability of telematic networks by AIM-research and the EU/EEA European Committee for Standardisation (CEN) prepared the European health card’s technological ground. The adoption of a data protection directive (95/46) in 1995 has been the legal preparation to tackle the issue of privacy. Despite this progress and another call for a health card from the European Parliament in 1996 (European Parliament 1996), the Member States were fairly reluctant. They still foresaw ‘a number of legal, ethical, economic and technical difficulties’ to be addressed before any idea of introduction, awarding it no priority (Council of Ministers 1996). The discussions on the Lisbon-strategy and patient mobility led the Member States to change their minds. While the Health ministers in early March 2002 emphasized the need to study the advantages of a European health card to pensioners and workers staying abroad, in late March 2002 the executive leaders of the Member States’ governments adopted the Commission’s proposal for a common health card in June 2004. That proposal says ‘an EU-wide health card should be introduced, aimed at transforming the relevant European paper forms into an electronic card. Card holders will be able to claim access to immediately necessary health care in a Member State other than the one where they are insured, the latter being nevertheless responsible for the costs.’ (European Commission 2002) The European Council emphasized that ‘such a card will simplify procedures, but will not change existing rights and obligations’ (European Council 2003). In February 2003, the Commission presented its plans to introduce the European health card that ‘will facilitate temporary stays abroad, initially holidays, the E-111 form being the first to be replaced; and, later, employees posted to another country (E128), international road transport (E110), study (E128) and job seeking (E119)’ (European Commission 2003a). Despite all research efforts, a common telematic infrastructure to read electronic health card is still lacking. Therefore, an eye-readable plastic card is gradually introduced within a transitory period from June 2004 till December 2005 by the national health insurance authorities. The card would initially only mention the name and health identification number of the cardholder and information on the competent health insurance authority. In a later stage, an electronic card containing information on the cardholder’s health history may follow. In a press release European Commissioner for Social Affairs Anna Diamantopolou contended that the card ‘will also have a powerful symbolic value: after the euro, the European health card is another piece of Europe in your pocket.’ (cited in European Commission 2003b) In

European Integration and Unfreezing Territoriality

219

similar vein, the European Parliament stated that the introduction of a European health card would ‘contribute significantly to the promotion of the free movement and the European Citizenship’ (European Parliament 2003). The Charter of Fundamental Rights of the Union seems to underline this European citizenship by effectively defining a European right of healthcare: ‘Everyone has the right of access to preventive health care and the right to benefit from medical treatment under the conditions established by national laws and practices. A high level of human protection shall be ensured in the definition and implementation of all Union policies and activities.’ (European Convention 2003, Art. II-35). Thus, the health card symbolizes the European challenge to national conceptions of citizenship, and the territorial control of health consumption. While the membership of healthcare systems has been heavily entrenched into the states’ territories, the European health card further unfreezes territoriality. The expected extension of regulation 1408/71 (plus the health card) to third-country nationals would entail an increase of cross-border healthcare consumption, since they are much more mobile than EU-citizens. Next to pensioners in Spain, and the export of British and Dutch patients on waiting lists to Belgium, this third-country nationals’ mobility may lead Member States to urge the sharing of the health burden at the EU-level. This rescue of the national healthcare systems entails the sacrifice of states’ principle of territoriality. Will this rescue imply the demise of the fundamental organizational principle of territoriality, or is the formation of a Healthy Europe (as an equivalent to Social Europe) guided again by the principle of territoriality?

3.5

The European health card: a primary exit and its implications

Member States collectively dominate the EU because they use it as a vehicle to maintain the acceptance from their respective state members. The dissatisfaction of Member States’ members is the driving motor of integration and reformation in the EU-area. Integration has predominantly taken place in areas in which state authorities need co-operation to maintain value provision for their state members (agriculture, trade), and not so much on issues in which state authorities can still individually gain prestige in their members’ eyes (foreign affairs and defense). When resources are available at EU-level to buy loyalties, the governments still have a decisive or exclusive say (e.g., agricultural and structural funds). The pace and content of European integration is predominantly set by the interests of the Member States to satisfy their respective state members. The growing number of Member States and the ensuing diversification decrease the possibility that Member States would find a solution to satisfy them all. As a consequence,

220

Hans Vollaard

Member States have opted for partial exits to keep their national say within the policy area like foreign affairs and defense, and monetary co-operation. Nevertheless, the EU Member States stick together as ‘[t]he costs of exit, even selective, are high. This disciplining mechanism stands behind the institutionalization of Europe.’ (Caporaso and Jupille 2004). Moreover, the 1990s policies to enhance the social and territorial cohesion of Europe – internal structuring – may also keep (partial) exit at bay. Political reconfiguration in the EU-area starts from a combination of dissatisfaction within the Member States and a primary exit. The European health card offers the primary exit for dissatisfied state citizens to express their discontent. The card makes citizens aware of and facilitates access to healthcare across borders, and enhances the resemblance of the EU/EEAMember States. Thus, it lowers the uncertainty and costs for a citizen to seek cross-border healthcare. Indeed, both dissatisfaction (on waiting lists and treatment), and familiarity with foreign healthcare systems were the main motives of citizens to seek cross-border healthcare, as was shown in the experiments in Dutch-German and Dutch-Belgian border regions (Grunwald and Smit 1999). At present, the use of cross-border healthcare is still marginal both in numbers and amounts within the EU-area. The hurdles of complicated administrative procedures (the E-111 and E-112 forms), problems with aftercare and language prevent many from going abroad. The unfamiliarity with other doctors, hospitals, insurers or foreign healthcare systems and loyalty restrains them to escape their present health providers and insurers to which they are loyal now (Brouwer et al., 2003). Even within the Netherlands the mobility is fairly low, since most patients seek healthcare within their own region (Brouwer 1999, 25). Nevertheless, cross-border healthcare rises in case of a deteriorating situation at home. Italians have sought healthcare in France and Belgium dissatisfied about domestic healthcare (France 1997). Similarly in the Dutch Zealand Flanders region, the demand for healthcare in Belgium increased (which could be easily obtained in a Euregional experiment) after negative stories about the regional Terneuzen hospital (CVZ 2003). As also an analysis of rather similar cases of healthcare consumption across Swiss canton borders suggests, the fewer obstacles to cross-border care, the more chance of crossborder healthcare (Crivelli 1998). In case of dissatisfaction with healthcare, an alternative to exit and voice is not to be expected soon. Apart from elections, the voice of patients is relatively difficult to organize as they form a highly fragmented and volatile group, with the possible exception of chronic patients. A combination of dissatisfaction about the performance of a healthcare systems and European exit options could therefore lead to the immediate use of the escape route of cross-border healthcare. However, the choice between exit and voice is a

European Integration and Unfreezing Territoriality

221

complicated one in a healthcare system. Health users have different exit opportunities, ranging from the possibility to switch from one healthcare provider or health insurance fund to another, to seek private healthcare provision and insurance instead of public, or escape from one national healthcare system to another (Freeman 1999, 109-10). The introduction of exits within citizens’ own healthcare system may therefore be introduced to siphon off dissatisfaction: ‘Nowhere did user dissatisfaction with health care states imply a demand for the quasi-marketization of health care, though that is the way in which, in part, governments have sought to meet it’ (Freeman 1999, 116). For instance, Dutch health authorities adopted exit opportunities among health insurance funds, ending their regional monopolies in 1992 and allowing patients to join the fund of their choice. The Dutch health authorities further introduced opportunities to voice introducing grievance procedures in hospitals and consultative patient platforms in the 1990s. Through offering more internal voice and exit opportunities, the authorities channel dissatisfaction which prevents health users to leave the system. Since the Dutch strict budgeting policy in the healthcare has deteriorated the broadly demanded delivery of timely healthcare, these internal exit and voice options can be interpreted as a means to spread the responsibility and blame of the dissatisfactory functioning of the healthcare system. Thus far, Dutch citizens still consider the national government and not so much the health insurance funds as main responsible for the healthcare system (Centrum voor Verzekeringsstatistiek 2002). European integration holds incentives for governments to perform better in providing public benefits such as healthcare, because they are confronted with better-informed and exit-threatening (affluent) patients. As long as patient mobility is marginal, it is a valve for temporary inconveniences of waiting lists, siphoning off the dissatisfaction in border areas, and saving money by a more efficient use of healthcare facilities just across border. An improvement in health provision may however attract foreign patients with the effect that the system becomes overloaded. Investment and planning in medical education and health facilities would become increasingly difficult. Anti-discriminatory legislation in Europe prevents governments from selecting benefit recipients to keep the system closed for ‘health tourism’. The necessary rise in taxes or premiums to build extra facilities may stimulate affluent citizens to leave at the expense of maintaining solidarity. Governments must therefore rely on different instruments to maintain their solidarity systems. ‘Receiving’ healthcare states like Belgium has tried to control the flows of patients, as Belgian parliamentarians and the Minister of Health Frank Vandenbroucke feared the export of Dutch and English waiting lists would make Belgians wait too long for treatment in Belgian hospitals, resulting in dissatisfaction among Belgian health consumer-citizens. The Belgian and

222

Hans Vollaard

British governments concluded a contract on treatments of NHS patients in Belgian hospitals in 2002. The financing of Belgian hospitals was further changed in the same year to the effect that it was no longer attractive for Belgian hospitals to treat as many patients, including foreign ones. The introduction of the European health card has also been motivated as a means to obtain information on patient flows, allowing for better control. ‘Sending’ Member States have also an interest in keeping the health consumption at home at a certain level to maintain the healthcare infrastructure, and uphold legitimacy based on health performance. That explains the limited coverage of the European health card – restricted to emergency care. The interests of both receiving and sending Member States reach further: questions on burden-sharing (i.e. solidarity) and planning of health facilities at EU-level have been raised. The rising costs of advanced medical technologies demanded by the well-informed public will be shared by the establishment of ‘centres of reference’ planned at EU-level. National governments try collectively to prevent dissatisfaction by unfulfilled expectations of their respective citizens.

4.

ORGANIZING HEALTHCARE: TERRITORIALITY IN A MULTI-LEVEL EUROPE

Borders have never sealed states completely off. They rather function as one of the filters for access to public benefits, and to keep public benefits systems ‘plannable’ and affordable. The European health card will force state governments to consult neighboring states to control flows of patients partly by constructing an EU-health area. This undermines the territorial exclusivity of state membership, and the centrality of decision-making and planning within the previously territorial state. Even if cross-border patient mobility would be limited, this de-territorialization of the healthcare states will be brought about. In the pre-health card period, the small numbers of cross-border healthcare consumption ‘provided more than the critical mass necessary to generate continuous and increasing litigation at the ECJ level’ (Leibfried and Pierson 2000, 278). The court’s decisions have had a creeping and lasting effect on one of the strongholds of national legitimacy, healthcare. The European health card will only stimulate cross-border healthcare, since the card is an easy reminder and facilitator of the EUcitizens’ right of reimbursement of emergency care. Furthermore, court cases on why the European health card does not cover ‘E-112/planned care’ may follow soon. Patients from Central European EU-Member States may press ahead for coverage of E-112 care, since the healthcare provision within

European Integration and Unfreezing Territoriality

223

Western Europe is better. Meanwhile, cross-border workers and hibernating pensioners from Western European EU-Member States may claim the same from their respective governments. Therefore, the de-territorialization of the healthcare states by the 1408/71 Regulation is expected to continue with the introduction of the European health card; but to a certain extent. Because performance in healthcare is one of the cornerstones of governments’ legitimacy – in particular in command-and-control states the legitimacy link between voters/health users and their governments is quite direct – the healthcare systems are expected to remain largely within national confines. A loyalty-fostering reaction to cross-border migration would strengthen that. The territorial infrastructure of particularly command-andcontrol states will therefore remain largely intact, although the boundaries of healthcare states are more fluid due to the de-territorialization of healthcare states. National governments have started co-operation at EU-level to make their healthcare systems perform well for their citizens. Thus, the deterritorializing European rescue of the healthcare state is accompanied with consociationalist practices at the EU-level by the national governments. The national compounds try to keep as much autonomy within their segments to keep Europeanization – penetration and standardization from “Brussels” – as low as possible, while collectively aim at sustaining their respective authority in their healthcare nation. A common understanding within the European society of healthcare nations to maintain a Social Europe helps the national governments to co-operate. Similar to European citizenship, the European healthcare provisions such as the European health card is rather a facilitative addition to national citizenship than the creation of a new European layer of identity. Although a somewhat fictitious image, the Rokkanian internal-external dynamic of polity-formation has thus entailed a further weakening of Westphalian state territoriality but not necessarily of national solidarity. The principle of territoriality seems to emerge in the Euro-polity. Eligibility to Regulation 1408/71 (including the European health card) is soon to be based on legal residence within the territories of the EU-area, instead of national citizenship. Thus, membership space coincides with geographic space, indicating territorial exclusivity. Central planning in the Euro-polity’s centre(s) is initiated by allocating centers of reference throughout the EU-area. The Euro-polity is still in construction since its borders are uncertain yet due to (future) enlargements and an increase of internal patient mobility. Therefore, refreezing territoriality by consolidating external borders and internal territorial cleavages in a European health area is not expected to occur too soon. Rokkan’s theory combined both the European and domestic political reams to explain when and how territory is significant for, among other

Hans Vollaard

224

things, the definition of formal competences of European and national authorities, as well as the actual range of action of individual patients. The new exit opportunities offered by the creation of the internal market have set going both Europeanization and renewed European integration. The Rokkanian reasoning also provides Claus Offe a reassuring answer to his concerns that de-territorialization by European integration may weaken the welfare states (Offe 1998, 102). The continuous strength of national loyalties may keep both exit and entry at bay, which would keep national solidarity and plannability of healthcare provision largely intact. Satisfaction of national citizens by providing a European extension of health rights might even be the necessary argument for citizens to remain loyal to their national governments and fellow-citizens. In this light, the introduction of the European health card is a slightly de-territorializing European rescue of national legitimacy and solidarity.

REFERENCES Agnew, J. 1998, Geopolitics: Revisioning World Politics, London, Routledge Anderson, J. 1996, The Shifting Stage of Politics: New Mediaeval and Postmodern Territorialities, Environment and Planning D: Society and Space 14: 133-153. Bartolini, S. 1998, Exit Options, Boundary Building, Political Structuring, EUI working paper. Florence, EUI. Bull, H. 1977, The Anarchical Society: A Study of Order in World Politics, London, MacMillan. Börzel, T.A. 2001, Europeanization and Territorial Institutional Change: Toward Cooperative Regionalism?, in: Transforming Europe: Europeanization and Domestic Change, Green Cowles, M., Caporaso, J. and Risse, K., eds, Ithaca/ London, Cornell University Press, pp.137-158. Brouwer, W.B.F., Exel, J. van, Hermans, H.E.G.M., and Stoop, A., 2003, Should I stay or should I go? Waiting Lists and Cross-Border Care in the Netherlands, Health Policy, 63(3): 289-298. Brouwer, W.B.F. 1999, Het Nederlandse Gezondheidszorgstelsel in Europa: Een Economische Verkenning. Zoetermeer: Council for Health and Social Services (RVZ). Caporaso, J.A., and Jupille, J. 2004, Sovereignty and Territoriality in the European Union: Transforming the UK Institutional Order, in: Restructuring Territoriality: Europe and the United States compared, Ansell, C. and Palma, G. di, eds., Cambridge, Cambridge University Press, pp. 67-89. Centrum voor Verzekeringsstatistiek 2002, De Consument aan het Woord: Onderzoek naar de Mening van de Consument over de Gezondheidszorg en de Ziektekostenverzekering, Den Haag, Centrum voor Verzekeringsstatistiek. Clark, I., 1998, Beyond the Great Divide: Globalization and the Theory of International Relations, Review of International Studies, 24: 479-498 . CVZ, 2003, Grensoverschrijdende Zorg (internal report). Amstelveen, College Voor Zorgverzekeringen. Cornelissen, R. 1996, The Principle of Territoriality and the Community Regulations on Social Security (Regulations 1408/71 and 574/72), Common Market Law Review, 33: 439-471.

European Integration and Unfreezing Territoriality

225

Council of Ministers, 1986, Resolution of the Council and of the Representatives of the Governments of the Member States, meeting within the Council, of 29 May 1986, concerning the Adoption of a European Emergency Health Card’, OJ 1986 C 184, 4-7. Council of Ministers, 1989, Council Conclusions of 29 September 1989 on a European Card for the Provision of Immediate Care’, O J 1989 C 277/03. Council of Ministers, 1991, Resolution of the Council and the Ministers for Health, meeting within the Council of 11 November 1991 concerning Fundamental Health-policy Choices’, O J 1991 C 304/05. Council of Ministers, 1993, Resolution of the Council and the Ministers for Health, meeting within the Council of 27 May 1993, on Future Action in the Field of Public Health, O J C 174, 25 June 1993: 1-3. Council of Ministers, 1994, Council Resolution of 2 June 1994 on the Framework for Community Action in the Field of Public Health, O J C 165, 17 June 1994: 1-2. Council of Ministers, 1996, European Health Card - Presidency Conclusions. Press release 314, no. 11437/96, 12 November 1996. Crivelli, L.A., 1998, Cross-Border Care between Swiss Cantons: A Testing Lab for the Single European Market, in: Health Care and its Financing in the Single European Market, Leidl, R., ed, Amsterdam, IOS Press, pp. 285-305. Daalder, H., 1993, The Development of the Study of Comparative Politics, in: Comparative Politics, Keman, H., ed, Amsterdam, Free University Press, pp. 11-30. Dowding, K., John, P., Mergoupis, T. and Vugt, T. van 2000, Exit, Voice and Loyalty: Analytic and Empirical Developments, European Journal of Political Research, 37: 469-495. European Commission, 1995, Final Report AIM DG XIII “Eurocards” concerted action, http://www.cix.co.uk/~cic/euhci.htm. European Commission, 1998, Eurobarometer Survey 49, http://europa.eu.int/comm/public_ opinion/archives/eb/eb49/eb49_en.htm. European Commission, 2001, Communication from the Commission: The Future of Health Care and Care for the Elderly: Guaranteeing Accessibility, Quality and Financial Viability, COM(2001) 723 final. International Public Policy and Management: Policy Learning beyond Regional, Cultural, and Political Boundaries, 2002, Communication from the European Commission: Commission•s the action plan for skills and mobility, COM(2002) 0071 final. European Commission, 2003a, Communication from the Commission concerning the Introduction of a European Health Card, COM(2003) 73 final. European Commission, 2003b, European Health Card for 1 June 2004. Press Release IP/03/271. European Commission, 2004, Communication on E-Health: Making Healthcare better for European Citizens: An Action Plan for a European e-Health Area, COM (2004) 356. European Convention, 2003, Draft Constitutional Treaty, The Policies and Functioning of the Union, Luxembourg, Office for Official Publications of the European Communities. European Council, March 2002, Promoting Skills and Mobility in the European Union – Presidency Conclusion no 34, http://europa.eu.int/abc/doc/off/bull/en/200203/i1029.htm. European Court of Justice, 28 April 1998, Case C-120/95 (Nicolas Decker vs. Caisse de Maladie des Employes Prives) and Case C-158/96 (Raymond Kohll vs. Union des Caisses de Maladie). European Court of Justice, 13 May 2003, Case C-385/99 (V.G. Müller-Fauré vs. Onderlinge Waarborgmaatschappij OZ Zorgverzekeringen UA and E.E.M. van Riet vs. Onderlinge Waarborgmaatschappij ZAO Zorgverzekeringen). European Parliament, 1981, Resolution on a European Health Card. OJ 1981 C 287: 32. European Parliament, 1996, Report on the European Health Card. A4-0091/96. European Parliament, 1998, Health Care Systems in the EU: A Comparative Study, Luxembourg: EP DG for Research.

226

Hans Vollaard

European Parliament/Committee on Employment and Social Affairs, 2003, Draft Report on Proposal for European Parliament and Council regulation on amending Council Regulations 1408/71 and 574/72, Provisional 2003/0138(COD). Flora, P., 2000, Formation of External Frontiers and Internal Structuring: Europe and its Nations. A Research Perspective on Rokkan’s Lines (translation I.L. Fraser), EUI-RSC seminar paper; published as Flora, P. 2000, Externe Grenzbildung und Interne Strukturierung: Europa und seinen Nationen: Eine Rokkanische Forschungperspektive, Berliner Journal für Soziologie, 10: 157-166. Forsberg, T., 1996, Beyond Sovereignty, Within Territoriality: Mapping the Space of LateModern (Geo)Politics, Cooperation and Conflict, 31(4): 355-386. France, G., 1997, Cross-border Flows of Italian Patients within the European Union: An International Trade Approach, European Journal of Public Health, 7(3 supplement):19-32. Freeman, R., 1999, The Politics of Health in Europe, Manchester, Manchester University Press. Le Galès, P. and Lequesne, C., eds, 1998, Regions in Europe, London, Routledge. Hix, S. and Goetz, K.H., 2000, Introduction: European Integration and National Political Systems, West European Politics, 23(4): 1-26. Gourevitch, P.A., 1978, The Second Image Reversed, International Organization, 32(4): 881-912. Grunwald, C.A. and Smit, R.L.C., 1999, Grensoverschrijdende Zorg: Zorg op Maat in de Euregio Maas-Rijn: Evaluatie van een Experiment (report at the request of the Health Insurance Board, no. 816). Utrecht, NZi. Health Ministers, 2002, Results of the Meeting of Health Ministers, Málaga 8 February 2002, http://www.ue2002.es. Hirschman, A.O., 1970, Exit, Voice, and Loyalty : Responses to Decline in Firms, Organizations, and States, Cambridge, Harvard University Press. Hirschman, A.O., 1974, “Exit, Voice, and Loyalty”: Further Reflections and a Survey of Recent Contributions, Social Science Information, 13(1): 7-26. Jachtenfuchs, M., 2001, The Governance Approach to European Integration, Journal of Common Market Studies, 9: 1-24. Kahler, M,. 2002, The of the State in World Politics, in: Political Science: State of the Discipline, Katznelson, I., and Milner, H.V. eds, New York, Norton/Washington, APSA. Keating, M., 1998, The New Regionalism in Western Europe: Territorial Restructuring and Political Change, Cheltenham, Edward Elgar. Keohane, R.O., Milner, H.V., eds, 1996, Internationalization and Domestic Politics. Cambridge: Cambridge University Press. Lapid, Y., Albert, M. and Jacobson, D., eds, 2001, Identities, Borders, Orders: Rethinking International Relations Theory, Minneapolis, University of Minnesota. Leibfried, S. and Pierson, P., 1995, Semisovereign Welfare States, in: European Social Policy: Between Fragmentation and Integration, Leibfried, S. and Pierson, P. eds, Washington, Brookings Institution, pp. 43-77. Leibfried, S., Pierson, P., 2000, Social Policy, in: Policy-Making in the European Union, Wallace, H. and Wallace, W., eds., 4 th ed. Oxford, oxford University Press, pp. 267-292. Mann, M., 1986, The Autonomous Power of the State: Its Origins, Mechanisms and Results, in: States in History. Hall, J.H., ed., Oxford, Basil Blackwell, pp.109-136. Mei, A.P. van der, 2001, Free Movement of Persons within the European Community: CrossBorder Access to Public Benefits (dissertation), Maastricht, Maastricht University. Moran, M., 1999, Death or Transfiguration? The Changing Government of the Health Care State, EUI Working Paper 99/15, Florence: EUI Newman, D., 2001, Boundaries, Borders and Barriers: Changing Geographic Perspectives on Territorial Lines, in: Identities, Borders, Orders: Rethinking International Relations Theory, Lapid, Y., Albert, M. and Jacobson, D., eds., Identities, Borders, Orders:

European Integration and Unfreezing Territoriality

227

Rethinking International Relations Theory, Minneapolis, University of Minnesota, pp. 137-151. Offe, C., 1998, Demokratie und Wohlfahrtsstaat: Eine Europäische Regimeform unter dem Streß der Europäischen Integration, in: Internationale Wirtschaft, Nationale Demokratie: Herausforderungen für die Demokratietheorie, Streeck, W., ed., Frankfurt, Campus Verlag, pp. 99-136. Olsen, J. (2001), The Many Faces of Europeanization, ARENA Working Paper 01/2, http://www.arena.uio.no/publications/wp02_2.htm. Palm, W., Nickless, J., Lewalle, H. and Coheur, A., 2000, Implications of Recent Jurisprudence on the Co-ordination of Health Care Protection Systems (Report at the request of the DG Employment and Social Affairs), Brussels, AIM. Risse, T., 1996, Exploring the Nature of the Beast: International Relations Theory and Comparative Policy Analysis Meet the European Union, Journal of Common Market Studies, 34(1): 53-80. Rokkan, S. and Urwin, D. 1983, Economy, Territory, Identity: Politics of Western European Peripheries, London, Sage. Rokkan, S., 1999, State Formation, Nation-Building, and Mass Politics in Europe: The Theory of Stein Rokkan (based on his collected works; edited by P. Flora, S. Kuhnle, D.Urwin), Oxford, Oxford University Press. Rosamond, B., 2000, Theories of European Integration, Basingstoke, MacMillan. Rosenau, J.N., 1997, Along the Domestic-Foreign Frontier: Exploring Governance in a Turbulent World. Cambridge, Cambridge University Press. Ruggie, J.G., 1993, Territoriality and beyond, International Organization, 47(1): 139-174. Sack, R., 1986, Human Territoriality: Its Theory and History, Cambridge, Cambridge University Press. Vollaard, J.P., 2004, Solidarity, Territoriality and Healthcare: Cross-National Policy Learning in Europe, in: International Public Policy and Management: Policy Learning beyond Regional, Cultural, and Political Boundaries, Vigoda-Gadot, E., Levi-Faur, D., eds., International Public Policy and Management: Policy Learning beyond Regional, Cultural, and Political Boundaries, Marcel Dekker, pp. 267-296. Watson, Ph., 1980, Social Security Law of the European Communities, London, Mansell Publishing. Weber, M., 1956, Staatssoziologie, Berlin, Duncker and Humblot. Wæver, O., 1995, Identity, Integration and Security: Solving the Sovereignty Puzzle in the E.U., Journal of International Affairs, 48(2): 389-431.

GOVERNING MOBILITY

The Externalization of European Migration Policy and the Boundaries of the European Union Rens van Munster and Steven Sterkx Abstract:

1.

The central question to be explored in this chapter is in which ways the externalization of migration policies inscribes the boundaries of the European Union. This question poses a significant challenge to current theories of governance and Europeanization which take as point of departure that the EU is a territorial unit in which there is a clear distinction between the EU-inside and EU-outside. As such, these theories cannot account for the fact that the boundaries of inclusion and exclusion do not solely correlate with territorial borders. Therefore, moving beyond the conceptualization of governance and Europeanization within European studies, this chapter proposes to conceptualize governance as the ‘conduct of conduct’. By looking at policy instruments that have been deployed to govern mobility between the EU and third countries, it argues that the externalization of migration policies can be understood as a set of deterritorialized practices to govern the proper conduct of freedom within the EU by rendering forms of improper conduct increasingly immobile.

INTRODUCTION

The institutionalization of a space in which individuals can conduct their social and economic freedom has been a central objective of EU governance. To realize this objective practically, the EU has adopted facilitating measures such as the creation of the internal market, the establishment of the EU as an area of freedom, security and justice (AFSJ) and the introduction of a European citizenship that enables individuals to be recognized as legal subjects across national borders. The institutionalization of a space in which individuals can conduct their freedom has, however, not just been effected by facilitating measures. Since the introduction of the Justice and Home Affairs pillar (JHA) in Maastricht in 1992, the more security and controloriented aspects of governing the conduct of freedom have become increasingly salient in EU policy-making. In official EU discourse these

229

R. Holzhacker and M. Haverland, (eds.), European Research Reloaded: Cooperation and Integration among Europeanized States, 229-250. © 2006 Springer. Printed in the Netherlands.

230

Rens van Munster and Steven Sterkx

control measures have been described as necessary flanking measures to compensate and supplement the abolishment of internal borders. As a result, the topic of third country immigration has become increasingly described as something that needs to be controlled in order to constitute and maintain the EU as an area of freedom, security and justice. This chapter will examine a relatively new part of these so-called flanking measures, namely the externalization of European migration and asylum policies. Partly because attempts to settle internal differences regarding the management of migration constrained effective policy-making and partly in response to electoral demands to boost the restriction of migration, the European Council decided at the 1999 Tampere summit that the management of migration also required the input of competences and instruments in the field of external relations. For the success of a harmonized asylum and migration policy, it was argued, would not just depend upon internal governance but also upon the Union’s capacity to address political and economic issues in countries of origin and transit itself. The externalization of migration policies poses a significant challenge to current theories of governance and Europeanization because these are still bounded by a territorial imaginary of the EU in which there is a clear demarcation between ‘EU-inside’ and ‘EU-outside’ (cf. Walker 2000). The externalization of migration policies, however, constitutes a non-territorial response to the issue of migration. Moreover, because current theories of Europeanization take the Euro-polity as the independent variable with which to explain change in the domestic settings of member states, these theories cannot account for the ways in which the boundaries of inclusion and exclusion no longer solely correlate with the territorial border of the EU. That is, more specifically, the externalization of the management of migration flows between the EU and the external environment has a constitutive bearing upon the EU by redrawing its very boundaries through the inscription of what is to be included and what is to be excluded from the area of freedom, security and justice. The central question to be explored in this chapter, therefore, is in which ways the externalization of migration policies (re)draws the boundaries of the European Union as an area of freedom, security and justice. In doing so, this chapter will propose an alternative way of looking at governance. The next section, thereto, critically examines the literature on governance and Europeanization. It considers the ‘governance turn’ in EUstudies and concludes that the shift towards top-down theorizing this entailed has generally bypassed the question of how European governance inscribes, modifies, strengthens, transforms or multiplies the boundaries between the EU and its external environment. Section 3, therefore, proposes an alternative way of conceptualizing governance as the ‘conduct of conduct’. It will be argued that the externalization of the governance of migration

Governing Mobility

231

between the EU and its external environment can be conceptualized as the governance of the excessive conduct of freedom. A conceptualization of governance as the conduct of conduct also brings out the constitutive relationship between freedom and control. Rather than viewing control and mobility in terms of a trade-off, it will be argued that the two form a constitutive relationship: the proper exercise of freedom within the EU depends on controlling inappropriate forms of freedom and mobility. Subsequently, section 4 will empirically discuss four of the main types of governance that fall under the wider label of externalization, arguing that these are not merely policy tools for the management of migration but also founding practices for the EU-polity as such.

2.

A CRITICAL EXAMINATION OF GOVERNANCE APPROACHES IN EU-STUDIES

In recent years EU-studies have witnessed an increasing use of the term governance in research on the European Union (see, among others, Schmitter 1996; Jachtenfuchs 1997; Kohler-Koch and Eising 1999; Hooghe and Marks 2001; for an overview see Hix 1998; Jachtenfuchs 2001). Constituting a powerful challenge to the more traditional theories of intergovernmentalism and neofunctionalism, governance approaches have contributed significantly to our empirical understanding of EU-policymaking across a great variety of sectors (see also chapters 1-5 in this volume). More importantly, though, the governance turn has also had a profound theoretical impact on EU-studies. Arguably, the most significant contribution of the governance approach to European integration is its unpacking of the Westphalian assumptions of state sovereignty that underpin the more established approaches of intergovernmentalism and neofunctionalism (see e.g. Jachtenfuchs 1995; Caporaso 1996; Wind 1997; Aalberts 2004). The governance approach challenges the traditional division between hierarchy/anarchy, inside/outside and the domestic/international by pointing out that order is also possible in the absence of a sovereign enforcer (see Marks et al., 1996).1 For instance, Hooghe and Marks argue that decisionmaking competencies in the EU are shared by multiple actors across different levels, which implies that the state does no longer play the central gatekeeper role between the domestic and the international: ‘the separation between domestic and international politics, which lies at the heart of the 1

A similar argument is made in the constructivist literature within the discipline of International Relations. This similarity between the governance approach and constructivist theories of IR is explicitly dealt with by Risse-Kappen (1996). See also the chapter by Aalberts in this volume.

232

Rens van Munster and Steven Sterkx

state-centric model, is rejected by the multi-level governance approach’ (Hooghe and Marks 2001, 4). Thus, although the EU does not take the form of a state, it is nevertheless characterized by a structure of political authority in which governmental agents at different levels of decision-making are embedded. The insight that the EU constitutes a multi-level polity has brought about a significant shift in the focus of European Studies which is best described as a shift from a bottom-up view to a top-down view on European integration (Jachtenfuchs 2001). This shift towards a more top-down approach has in turn spurred much research on the impact of Europeanization which until recently was a much forgotten dimension of European integration (Jachtenfuchs and Kohler-Koch 2004, 109). Although the concept of Europeanization is essentially contested among EU-scholars (see e.g. Radaelli 2000; Olsen 2001; Eising 2003), much of the self-styled Europeanization literature seems concerned with the top-down question of how European governance influences ‘domestic policies, politics, and polities of the member states’ (Börzel and Risse 2003, 57). The analytical move towards top-down theorizing is not without tradeoffs, however. First, top-down theorizing risks turning analytical choices about what to take as an independent or dependent variable into tacit ontological claims about the nature of the EU (cf. Fearon and Wendt 2002). For turning the EU into an independent variable also means that the latter is turned into something fixed or static. A clear division between independent variables (EU-polity) and dependent variables (domestic adjustment, enforcement) cannot capture the feedback effects of governance processes on the EU-polity because the latter is treated as a static and fixed structure. This is of course not to say that the EU should never be treated as an independent variable which may have real causal effects. The point is rather that analytical choices should be considered as just that, i.e., as practical and methodological choices made by the researcher to delimit a research subject. They should not be interpreted, however, as reflections upon an unchanging reality ‘out there’.2 A second and not wholly unrelated problem of top-down theorizing concerns the fact that it posits a clear divide between EU-inside and EUoutside. This is a somewhat paradoxical trait of a tradition built around the assumption that the inside/outside divide can be transcended. In fact, it is the very assumption of overlapping authorities that makes top-down theorizing possible. For only if it is accepted that the EU is more than institutionalized 2 Consider for instance Caporaso’s (1996) claim that the EU has transformed into a ‘postontological entity’, which is bewildering both from a philosophical stance (as the notion of post-ontological seems to imply that the EU has moved beyond being) and from a more empirical point of view as it pays no attention to how processes of governance shape and reshape the ‘nature of the beast’ (Risse-Kappen 1996).

Governing Mobility

233

intergovernmental bargaining, the top-down effects of European governance can be analyzed and measured in any meaningful way. But if governance approaches transcend the distinction on one level, they also reify it on another level by taking as a point of departure the EU as a well-established and fixed structure with a clear inside (the EU member states) and outside (the external environment). As a result, the literature on Europeanization is largely inward-looking and pays little attention to the externalities of EU governance and the role of the latter in shaping the EU (see Cederman 2001; cf. de Wilde forthcoming). Friis and Murphy are therefore right to observe that ‘the focus of the governance school on territorial or political space within the Union creates the impression that the extent of EU governance is confined to its membership, i.e. membership and governance are implicitly linked. This neglects the development of governance “beyond the EU”’ (Friis and Murphy 1999, 213). Although the literature on Europeanization has largely considered the effects of the EU-polity on its member states, it could be argued to its defense that this is not so much due to theoretical deficiencies as empirical priorities of individual scholars. In principal, the effects of Europeanization on non-members can be measured as easily as the effects on member-states.3 However, and notwithstanding the possible benefits of adding an external component to the Europeanization literature, a more outwards-looking focus nevertheless replicates the error of taking the EU as a given entity with a fixed inside and outside. It would still take the EU, the independent variable, as something out there and the external environment as the dependent variable in which change can be detected. But it would not be possible to view how this interaction impacts upon the Union itself. As Friis and Murphy argue: ‘It is ironic that the application of a concept [governance] which assumes a blurring of boundaries between inside and outside…has largely ignored or downplayed the significance of the ‘outside’ in shaping the ‘inside’ (Friis and Murphy 1999, 213). In the present context of the externalization of migration policy, then, it would prove impossible to consider the ways in which the externalization of migration policies has a constitutive feedback effect upon the EU’s boundaries. To conclude, then, it can be argued that while the governance turn in European Studies has contributed significantly to our insight into the dynamics of Europeanization, it is necessary, if we aim to obtain a firmer grip on the nature of the EU, also to inquire into how ‘its boundaries emerge out of specific interactions with the EU’s external environment’ (Cederman 2001, 2). In order to be able to do so, however, it is first necessary to provide 3 In this context, it could for instance be interesting to explore some theoretical and methodological links between the Europeanization literature on the one hand and the literature on the EU as an external actor on the other (see e.g. Hill 1993; Bretherton and Vogler 1999).

234

Rens van Munster and Steven Sterkx

a conceptualization of governance that does not start from top-down theorizations of the EU-polity.

3.

TOWARDS A CONCEPTUALIZATION OF GOVERNANCE AS THE CONDUCT OF CONDUCT

In this section, then, we seek to move away from the top-down view of governance as the patterning of social relations towards of view of governance as the ‘conduct of conduct’ (Gordon 1991, 2; Dean 1999, 10; Rose 1999, 3). Such a conceptualization does not take as a point of departure EU governance as an independent variable but, instead, conceptualizes governance as an activity that seeks to shape the conduct of people towards certain objectives. Defined as such, governance in liberal societies can be described as the conscious attempt to influence, mould, guide and direct the ways in which people exercise their freedom (Dean 1999; Rose 1999). Contrary to what is commonly assumed, this assumes that freedom is not something to be achieved by minimizing the governance of individuals. Rather, it views freedom as an effect or artifact of governance. Governance in liberal societies seeks to make people free in different domains such as the market, society and the family. As the neo-liberal philosopher Hayek argues: ‘Man has not developed in freedom…Freedom was made possible by the gradual evolution of the discipline of civilization which is at the same time the discipline of freedom’ (quoted in Rose 1999, 67). This in turn draws attention to the types of governance that can be used to bring about freedom. Broadly, a distinction can be made between two categories: facilitating and control-oriented types of governance (Dean 2002). In an interesting account of political and social transformations in Europe, Andrew Barry has pointed out some of the more facilitating types of governance. According to him, the project of European integration aims to encourage individuals to make autonomous social and economic choices on the basis of their individual interests (Barry 1993, 1994). As he points out, the economic rationale behind harmonization is to correct excessive state interference in the economy in order to allow individuals to exercise their own economic freedom. A second, more social rationale for the creation of a European space refers to the idea that the EU should become a kind of ‘animator state’ within which agents would actively seek to implement appropriate strategies for social problems themselves (Barry 1993). As such, techniques of harmonization (e.g. the introduction of a European citizenship, the introduction of a European health card, the abolishment of internal

Governing Mobility

235

frontiers) are crucial in carving out a political and social space on the European level (see also the chapter by Vollaard in this volume). However, the optimal exercise of freedom within the EU does not only depend upon facilitating measures that seek to establish the optimal conditions under which individuals actively start governing their own conduct of freedom (cf. Dean 2002). The proper conduct of freedom within the EU also depends on the governance of what is considered to be improper or perverse exercise of freedom. As article 2 of the TEU argues, the objective of the EU is ‘to maintain and develop the Union as an area of freedom, security and justice, in which the free movement of persons is assured in conjunction with appropriate measures with respect to external border controls, asylum, immigration and the prevention and combating of crime.’ Hence the conduct of freedom within the EU depends not only on the creation of the optimal conditions for the exercise of freedom but also on the control of what are considered to be excessive forms of social and economic freedom (Dean 2002; Huysmans forthcoming, chapter 5). Within the EU, migration has been defined as a danger to the free exercise of social and economic freedom within the EU. As a result, it has become increasingly subjected to control-oriented forms of governance. The specific way in which the EU has sought to govern the conduct of freedom led them to expect that ‘the market would not only improve free movement of law-abiding agents, but would also facilitate illegal and criminal activities by terrorists, international criminal organizations, asylum-seekers and immigrants’ (Huysmans 1999, 760). It is striking in this context that the source of external danger to the EU is not so much attributed to a territorial unity such as a state but to transversal and transnational activities. As Walters observes: ‘It is not the collective defense of the EU from Russia, Turkey, or Morocco as military powers, or national movements which is at stake. Rather, the security threat takes the form of a host of transnational, social threats, and is often personified in the racialized figure of Islamic and nonwhite people’ (Walters 2002, 570). It is important to note, though, that there is nothing inherent to migration that qualifies it as a security issue that needs to be managed through techniques of control. Migration can be framed in many different ways (as a humanitarian, an economic, or cultural issue for example) and it was not until it was rendered as a negative side effect to the functioning of the internal market and the conduct of freedom that it became a security issue. Because of the emphasis on transnational phenomena, immigration has become inscribed as a form of excessive conduct of freedom which should be controlled and separated from more legitimate forms of the exercise of freedom (see also article 29 TEU). Whereas sovereign states have sought to control excessive migration flows through the management of the territorial

236

Rens van Munster and Steven Sterkx

border, territorial concerns are less important for the EU as a multilevel polity. As Barry argues: ‘For societies governed by a principle of sovereignty the territoriality of the state was of overriding importance. The territory was a space whose borders needed to be defended, and within which sovereign rule could be exercised. By contrast, liberal government depended not primarily on the exercise of sovereignty, but on the development of institutions and procedures concerned with regulating and improving the health, wealth, and security of the population’ (Barry 1993, 318). While the control of outer territorial borders by no means has disappeared, it would be incomplete to view these controls simply as a reterritorialization of national borders on a European level. More specifically, such a view would ignore the changing meaning that is ascribed to territorial borders. As the above quote by Barry suggests, territoriality is not the crucial, all-defining mechanism that dictates how the illegitimate conduct of freedom is to be governed between the EU and the external environment. In fact, territorial controls – while still important – have become one among many elements that seek to condition the conduct of freedom of the population within the EU. Indeed, as the next section will show, the institutionalization of freedom through more control-oriented forms of governance is increasingly made dependent upon deterritorialized forms of management, i.e. the conduct of freedom through control-oriented forms of governance increasingly takes place beyond the territorial borders of the EU.

4.

GOVERNING FREEDOM: THE EXTERNALIZATION OF EU ASYLUM AND MIGRATION POLICY

This section empirically discusses the ways in which the more controloriented forms of the governance of freedom have been deterritorialized. To this end, subsection 4.1 will discuss the increased importance of the externalization of migration control in general terms. Subsections 4.2 to 4.5, subsequently, will briefly discuss four of the main policy instruments that have been deployed to govern the ‘improper’ conduct of freedom between the EU and its external environment. These are, respectively, (1) the action plans of the High-Level Working Group on Asylum and Migration, (2) the conclusion of readmission agreements, (3) carrier responsibility and (4) the recent proposals to process asylum-seekers in transit camps at the outer border of the EU in combination with increased reception in the region. While this list is not exhaustive, it provides a good oversight of some of the

Governing Mobility

237

more important policy instruments aimed at the control of migration beyond the territorial borders of the EU.

4.1

The Deterritorialization of Control: The Externalization of Migration and Asylum Policies

At the 1999 Tampere summit, the European Council decided in favor of a comprehensive approach to migration that included the management and control of (illegal) migratory flows as well as the prevention of these flows through targeting the root causes of migration. In practice, however, policymaking under the umbrella of the comprehensive approach has been strongly biased towards restrictive policies (van Selm 2002; Boswell 2003a, 2003b). The financial budgets reflect this priority. In 2001 a new budget line (B7667) of EUR 10 million was created to support ‘[c]ooperation with third countries in the area of migration’.4 Since then the budget has been doubled to approximately EUR 20 million in 2003. The cooperation with third countries shares the overall objective of trying to influence migratory movements. It can be divided into three strands: migration management, the fight against illegal immigration, and voluntary return and efficient execution of readmission obligations. Specific measures include, among others, legislative and institutional development of asylum systems in target countries, development of information campaigns on illegal and legal immigration, capacity-building to implement readmission obligations, support to border controls, training and transfer of knowledge, provision of equipment and other materials, development of airport and immigration liaison officers’ networks, and supporting capabilities to detect falsified documents. Moreover, in March 2004 a multi-annual financial framework has been put into operation for the years 2004-2008 which aim is to provide financial and technical aid to third countries in the area of migration and asylum. A total amount of EUR 250 million is made available ‘to provide a specific additional response to the needs of third countries in their efforts to manage more effectively all aspects of migratory flows, and in particular to stimulate third countries’ readiness to conclude readmission agreements, and to assist them in coping with the consequences of such agreements.’5 In general, this financial instrument acknowledges the priority attached to the external 4

EUROPEAN COMMISSION, 28 September 2001, Communication from Commissioner Vitorino to the Commission on Framework for preparatory actions, Budget line ‘Cooperation with third countries in the area of migration (B7-667)’. 5 COUNCIL OF THE EUROPEAN UNION, 10 March 2004, Regulation establishing a programme for financial and technical assistance to third countries in the areas of migration and asylum (AENEAS), 491/2004, published in Official Journal of the European Union on 18 March 2004, L 80/1, par. (7).

238

Rens van Munster and Steven Sterkx

dimension of asylum and migration and will probably make future cooperation with third countries and the implementation of measures less problematic. The EUR 250 million will be spent on information campaigns, development of legislation, exchange of information and know-how, staff training, capacity-building in the areas of border controls and security of travel documents and visas, technical and operational assistance, purchase of equipment and supplies, reintegration of returnees and resettlement programs, and assistance in the negotiation by the third countries concerned of their own readmission agreements with relevant countries. The externalization of migration and asylum policy is a deliberate process to export the Union’s asylum and migration policy through the mobilization of third countries. Hence, Guiraudon speaks in this context of ‘finding new allies’ and ‘co-opting sending and transit countries’ (Guiraudon 2000, 266). As such, the externalization of control nicely displays the decreased meaning of territorial borders for the management of migration flows. With the geographical site of the territorial border no longer the sole area where migrants are admitted or refused, the governance of freedom already start in the countries of origin and transit. Below some of the most salient policy initiatives will be considered in more detail.

4.2

The Action Plans of the High-Level Working Group

At the end of 1998, in line with the Austrian Presidency’s infamous strategy paper, the Dutch delegation to the Council put forward a proposal to create a horizontal task force on asylum and migration in order to establish a common, integrated, cross-pillar approach targeted at the situation in the most important countries of origin of asylum-seekers and migrants.6 The JHA Council of 3-4 December 1998 endorsed the idea and the General Affairs Council of 6-7 December 1998 officially established this cross-pillar task force under the name ‘High-Level Working Group on Asylum and Migration’ (HLWG). In accordance with the tasks proposed by the Dutch delegation in its note, the HLWG drew up a list of the most important countries of origin and transit for which cross-pillar actions plans would be drafted, and prepared its terms of reference during its first two meetings of 17 December 1998 and 11 January 1999. The following countries and regions were selected: Afghanistan/Pakistan, Albania and its neighboring region, Morocco, Somalia and Sri Lanka. In addition, the HLWG would also

6 DUTCH DELEGATION TO THE COUNCIL, 23 November 1998, Note from the Dutch delegation: Task Force on Asylum and Migration, 13344/98 JAI 37 AG 15.

Governing Mobility

239

take up and assess the already existing action plan on the influx of migrants from Iraq.7 According to the terms of reference that were endorsed by the General Affairs Council of 25 January 1999, the action plans should comprise elements such as a joint analysis of the causes of influx, suggestions aimed at strengthening the common strategy for development with the country concerned, identification of humanitarian needs and proposals to this end, proposals for intensification of political and diplomatic dialogue with the selected countries, indications on readmission clauses and agreements, on possibilities of reception and protection in the region, safe return, repatriation, as well as on the cooperation with intergovernmental, governmental, nongovernmental organizations and the United Nations Office of the HighCommissioner for Refugees (UNHCR).8 Although the action plans were intended to establish a cross-pillar approach with respect to the situation in a number of important countries of origin or transit, they received severe criticism from non-governmental organizations, the European Parliament and Morocco (one of the target countries).9 The main points of critique concerned, firstly, the lack of dialogue with the target countries and the consequent lack of a genuine partnership and, secondly, the manifest emphasis of the HLWG on securityrelated measures. This second critical assessment points to the imbalanced nature of the plans where a strong weight is given to measures aimed at controlling and preventing migration into EU territory. The plans’ main focus was on the conclusion and implementation of readmission agreements, but it also included proposals with regard to sending liaison officers to airports in order to improve gate checks, support for the adoption of visa requirements by the target countries for third-country nationals, technical and financial support to the target countries for the identification and detection of criminal trafficking networks, strategies for early detection of false documents, the organization of information campaigns in the target countries to advise on migration options and to warn against the 7

Influx of Migrants from Iraq and the Neighbouring Region: EU Action Plan, 5573/98 ASIM 13 EUROPOL 12 PESC 27 COMEM 4 COSEE 4, adopted by the General Affairs Council on 26 January 1998. The United Kingdom had taken the initiative for this plan. 8 COREPER, 22 January 1999, Terms of reference of the High Level Working Group on Asylum and Migration; preparation of action plans for the most important countries of origin and transit of asylum-seekers and migrants, 5264/2/99 REV 2 LIMITE JAI 1 AG 1. 9 AMNESTY INTERNATIONAL, EUROPEAN UNION ASSOCIATION, 23 December 1999, Comments by Amnesty International on the implementation of the action plans adopted by the High Level Working Group on Asylum and Migration. EUROPEAN PARLIAMENT, 29 February 2000, Report on asylum-seekers and migrants – action plans for countries of origin and transit, Rapporteur: Jorge Salvador Hernandez Mollar, FINAL A5-0057/2000. EUROPEAN PARLIAMENT, 12 October 2000, Report on the Draft Action Plan for Albania and neighbouring regions, Rapporteur: Anna Karamanou, FINAL A5-0287/2000.

Rens van Munster and Steven Sterkx

240

consequences of illegally entering EU member states, assistance and training of asylum and migration officials and border authorities, and increasing reception and protection capacities in the region of origin. As such, drafting the action plans, the security of the EU member states and not the development of the target countries would have been the HLWG’s major concern.

4.3

EU Readmission Policy

The entry into force of the Amsterdam Treaty marked the beginning of EU competency to regulate repatriation of illegal residents. Article 63(3) of the EC Treaty (as amended by the Treaty of Amsterdam) establishes the Union’s right to negotiate and conclude readmission agreements with third countries. Readmission can be defined as the decision by a receiving state on the re-entry of an individual. The first EU readmission agreement ever – between the European Community and Hong Kong – was signed on 27 November 2002 and formally adopted by the Council on 17 December 2003.10 In the meantime, Macao and Sri Lanka have also entered into a readmission agreement with the EU. Readmission negotiations with Albania have been completed, while agreements with Russia, Morocco and Ukraine are under way. Moreover, negotiations are soon to open with Algeria, China, Pakistan and Turkey.11 Community readmission agreements with third countries are considered a primary tool in the management of migration flows and the fight against illegal immigration. Their aim is to regulate the mutual taking-back of certain categories of persons without any formalities other than those specified in the agreement. Basically, it concerns the following two categories: first, the party’s own nationals, and secondly nationals of a third country or stateless persons who have unlawfully entered the territory of one of the contracting parties coming (directly) from the territory of the other party. A second element of EU readmission policy is the insertion of readmission clauses into agreements with third countries.12 These clauses intend to commit the contracting parties to readmit their own nationals, third-country nationals and stateless persons. How this will 10

COUNCIL OF THE EUROPEAN UNION, 17 December 2003, Decision concerning the conclusion of the Agreement between the European Community and the Government of the Hong Kong Special Administrative Region of the People’s Republic of China on the readmission of persons residing without authorisation, 2004/80/EC. The Agreement was published in the Official Journal of the European Union on 24 January 2004, L 17/25-39. 11 EUROPEAN COMMISSION, 30 December 2003, Biannual update of the scoreboard to review progress on the creation of an area of “Freedom, Security and Justice” in the European Union (second half of 2003), COM(2003) 812 final, p. 36. 12 COUNCIL OF THE EUROPEAN UNION, 2 December 1999, Decision on readmission clauses in Community agreements or mixed agreements, 13461/99. These clauses are based on art. 13 of the Cotonou Agreement with ACP-countries.

Governing Mobility

241

actually proceed is left to implementing agreements. A novelty is that these clauses need to be inserted in all future Community or mixed agreements.13 Central to readmission agreements is that the responsibility for migration management is shifted to third countries through the introduction of remote control. While this implies a certain amount of cooperation and partnership with third countries, it predominantly alludes to the EU’s policy priority of containing migration flows. As such, readmission policy constitutes exclusion as it reinforces the image of the EU as a ‘fortress’. The fact that readmission agreements are a politics of exclusion becomes apparent from the fact that ‘leverage’ is needed in order to persuade third countries to enter into readmission agreements. These agreements seem to asymmetrically ‘benefit’ the European Union, since third countries have little to gain from them. The Commission has pointed out that the Community should provide incentives to obtain the co-operation of the third countries involved (Reerman 2002: 23), although the (financial) margin of maneuver is relatively limited.14 The negotiation and conclusion of readmission agreements are only likely to succeed if they are part of a broader co-operation agenda. On a country-by-country basis, specific accompanying support should be granted. For instance, in the case of negotiations with Albania, the leverage needed is the parallel negotiation of a Stabilization and Association Agreement. As such, political pressure of the European Union and Albania’s aspiration for future membership provide the necessary incentives. In other cases, the amount of leverage required can be far more substantial. Negotiations with Russia will – as the Kaliningrad issue has shown – centre on compromises in the field of visa policy, while negotiations with Morocco might entail economic or financial concessions and accords on legal immigration. In the case of Sri Lanka, the readmission negotiations should be seen against the background of the Tamil-fight: the government of Sri Lanka did not ask for visa exemption, but the deal was that the EU would help with combating terrorism and the financing of it. In general, and as a corollary of the JHAchapter negotiations with the candidate countries, the EU can be said to pursue a strategy of functional inclusion under the heading of partnership: the functional involvement of third countries in the establishment of an area of freedom, security and justice.

13 This decision was taken by the JHA Council of 3 December 1998. Meanwhile, readmission clauses have already been inserted in agreements with Algeria, Armenia, Azerbaijan, Croatia, Egypt, Georgia, Lebanon, Macedonia, Uzbekistan and the Cotonou Agreement between the EU and the ACP-countries. See EUROPEAN COMMISSION, 10 April 2002, Green Paper on a Community Return Policy on Illegal Residents, COM(2002) 175 final, p. 24. 14 Ibid., par. 4.1.1. See also EUROPEAN COMMISSION, 3 December 2002, Communication on Integrating migration issues in the European Union’s relations with third countries, COM(2002) 703 final, p. 25.

Rens van Munster and Steven Sterkx

242

4.4

Carrier Liability

Another instance of remote control is the shifting of immigration control liability to (mainly airline) carriers, in combination with the sending of liaison officers to support the control tasks ascribed to these carriers. The idea of carrier liability was first elaborated in the context of Schengencooperation. Under the heading of accompanying measures for the abolition of internal border checks, article 26 of the Schengen Implementing Agreement states that ‘if aliens are refused entry into the territory of one of the Contracting Parties, the carrier which brought them to the external border by air, sea, or land shall be obliged to assume responsibility for them again.’ The carrier in question will be sanctioned in two ways. First, the carrier will have to bear the costs of return (or the costs of stay and return if immediate onward transportation is not possible). Secondly, the Contracting Parties will impose penalties upon the carrier. In order to avoid these sanctions, carriers have taken measures to make sure third-country nationals are in possession of valid travel documents upon entering the Schengen-territory. These measures mainly consist of tightened security and control checks and involve a shifting out of traditional public tasks such as border control and immigration management to private actors. Building upon the Schengen acquis and in accordance with the Amsterdam Treaty Protocol on integrating Schengen into the EU framework, the Council of the European Union has meanwhile supplemented article 26 of the Schengen Implementing Agreement by means of a directive, with the overall objective of curbing migratory flows and combating illegal immigration.15 Article 2 provides that the provisions of the Schengen Convention shall also apply to third-country nationals in transit. Articles 3 and 4 commit EU member states to impose respectively the costs of repatriation and financial penalties upon the carriers. Although the directive contains safeguards acknowledging the obligations on refugee protection resulting from the 1951 Geneva Convention, a British Airways manager has already stated that airlines, as a corollary to avoiding carrier sanctions, are denying persons the opportunity to claim asylum by preventing individuals from flying to a country. Furthermore, the effect of these sanctions could very well be the encouragement of high quality document fraud and of people-smuggling, as such watering down the goal of fighting illegal immigration (Forster 2002, 27).

15

COUNCIL OF THE EUROPEAN UNION, 28 June 2001, Directive 2001/51/EC supplementing the provisions of Article 26 of the Convention implementing the Schengen Agreement of 14 June 1985, published in the Official Journal of the European Communities on 10 July 2001, L 187/45-6.

Governing Mobility

243

Complementary to the control function ascribed to carriers, the EU member states are sending and appointing liaison officers and document experts in third countries of origin and transit which are the source of illegal immigration, i.e. ‘immigration hot spots’. Airline security personnel will be trained by these specialists in order to detect document forgeries before actual travel has started.16

4.5

Reception in the Region and Transit Camps

The latest evolution under the heading of a comprehensive approach entails the ideas on external processing of asylum applications, along with an elaboration of regional reception facilities (Loescher and Milner 2003; Noll 2003). Both strategies reflect the growing attention for remote protection and governance export. Following up on earlier developments on migration management partnerships, capacity-building in countries of origin and transit, and return and readmission of illegal residents, the UK government drafted a plan to establish a global network of safe havens which was leaked to the public. The ‘new vision of refugees’ plan defines these safe havens as ‘particular camps whose prime purpose is to provide a place of safety and process claims.’17 Asylum-seekers would be able to go directly to one of the safe havens. In addition, and most importantly, any asylum-seeker that would arrive in the UK or elsewhere would be relocated to a safe haven. A later and final version of this draft in early March 2003 dropped the notion of safe havens and introduced the concept of ‘regional protection areas’. A paper attached to a letter by Tony Blair to his Greek colleague Costas Simitis – Greece held the EU Presidency during the first half of 2003 – elaborated on improved regional management of migration flows and launched the notion of ‘transit processing centers’.18 These processing centers would be on transit routes to Europe, just outside the EU territory. Asylum-seekers arriving in the UK and other EU member states would be transferred to the transit processing centers, where their asylum claims would be assessed. The centers could be managed by the International Organization for Migration (IOM), with the screening of asylum requests approved by UNHCR. In response to the UK plans, the UNHCR proposed a ‘three-pronged approach’, which was presented at the informal meeting of the JHA Council 16

GENERAL SECRETARIAT OF THE COUNCIL OF THE EUROPEAN UNION, DG H, February 2002, EU Schengen Catalogue: External borders, removal and readmission – recommendations and best practices, p.12-3. 17 UK HOME OFFICE, 5 February 2003, A new vision for refugees. Final report. Draft, p. 10. It was leaked to The Guardian and The British Daily. 18 TONY BLAIR, 10 March 2003, Letter to His Excellency Mr Costas Simitis: New international approaches to asylum processing and protection, http://www. statewatch.org/news/2003/apr/blair-simitis-asile.pdf.

244

Rens van Munster and Steven Sterkx

in Veria, Greece.19 It consists of a regional prong aimed at enhancing protection in the region of origin; an EU prong with the possibility for closed reception centers and processing by EU teams to address caseloads composed primarily of economic migrants; and a national prong being the effective functioning of individual member state asylum systems. The UK proposals were supported by Denmark, Spain, the Netherlands and Italy and it was agreed that the UK could present an elaborated proposal at the June Thessaloniki European Council. Before the Thessaloniki Summit, the European Commission – responding to the request made by the Spring European Council to explore the UK ideas further20 – presented its communication on more accessible, equitable and managed asylum systems.21 The Commission clearly stated that new approaches should not undermine the process of establishing a common asylum system, nor shift the burden away from the European Union territory, thereby strongly opposing some of the core content of the UK proposals.22 The final discussions at the Thessaloniki Summit revealed that there was not sufficient support for the controversial UK plans. Germany, Sweden and the Commission reacted fiercely against the idea of transit processing centers, which was dropped in the end. The European Council Conclusions only mention the need for enhanced protection capacity in the region of origin.23 Nevertheless, British officials have already said that they will continue to work on their idea with other like-minded EU countries: pilot protection zones for refugees could be set up outside Europe by ‘a coalition of the willing’.24

19 UNHCR, 28 March 2003, Statement by Mr. Ruud Lubbers, United Nations High Commissioner for Refugees, at an informal meeting of the European Union Justice and Home Affairs Council, http://www.unhcr.ch. 20 BRUSSELS EUROPEAN COUNCIL, 20 and 21 March 2003, Presidency Conclusions, 8410/03 POLGEN 29, par. 63. 21 EUROPEAN COMMISSION, 3 June 2003, Communication from the Commission to the Council and the European Parliament. Towards more accessible, equitable and managed asylum systems, COM(2003) 315 final. 22 Ibid., p. 22. Instead, the EU member states should consider the introduction of Protected Entry Procedures, i.e. asylum applications at EU embassies in third countries, and of resettlement schemes. 23 THESSALONIKI EUROPEAN COUNCIL, 19 and 20 June 2003, Presidency Conclusions, 11638/03, par. 26. 24 EURACTIV, 20 June 2003, Thessaloniki on immigration: more money but no external protection zones, http://www.euractiv.com. Media have already reported on a pilot scheme for transit camps in Croatia; see EUOBSERVER, 16 June 2003, Asylum seekers to wait in Croatia, http://www.euobserver.com.

Governing Mobility

4.6

245

Concluding Remarks

The institutionalization of freedom within the EU has been a key dynamic through which the externalization of restrictive migration policies has taken place. On the basis of the policies described above, the externalization of migration policies can be subdivided into three categories. The first one is the notion of remote control which refers to the transfer of border controls to third countries (Zolberg 2003; see also Guiraudon 2003). Through screening at the source, the EU and its member states have erected new boundaries in third countries in an attempt to deny immigrants and asylum-seekers territorial access to their systems of legal protection and asylum. Secondly and parallel to the idea of remote control, Lavenex has coined the term remote protection to describe the increased emphasis on the extra-territorial dimension of the protection of refugees in either regions of origin or in transit centers close to the EU border (Lavenex 2003; Loescher and Milner 2003; see also Noll and Vedsted-Hansen 1999). Thirdly, it is possible to single out capacity-building as a means by which the EU seeks to ensure migration control in third countries. Through the transference of know-how, technologies, equipment and legislation, third countries are actively being involved in EU efforts to manage migration. The EU is providing technical and financial assistance to non-member states in order to supply the necessary leverage to implement control at the ground. Although there are considerable differences between the different types of governance in terms of their respective origins and contexts in which they first emerged and the kind of actors (e.g. public or private) and interventions (e.g. capacity-building or protection), they are all structured by a discourse that defines immigration as a threat to the proper conduct of freedom within the EU. As a result, these heterogeneous types of governance are all purposefully directed towards the objective of restricting the improper and dangerous conduct of freedom. And they do so in a deterritorialized manner. Although the physical location of the geographical border still possesses important symbolic value, the control of migration flows is increasingly pushed outwards. In contradistinction to more territorial forms of governance, the decisions on inclusion and exclusion are not just made at the geographical intersection between EU-inside and EU-outside. Understood as the conduct of conduct, governance in the EU can be considered as the regulation of mobility in which the mobility of some is made dependent on the immobility of others. As Verstraete argues: ‘The freedom of mobility for some (citizens, tourists, business people) could only be made possible through the organised exclusion of others forced to move around as illegal “aliens”, migrants, or refugees’ (Verstraete 2001, 29). Facilitating the proper conduct of freedom for some, in other words, cannot go without defining what is to be considered as the improper or even

Rens van Munster and Steven Sterkx

246

dangerous forms of conduct (cf. Foucault 2003).25 As a consequence, the boundaries between the EU and its outside are increasingly inscribed through migration policies that extend well beyond the territorial border of membership and which seek to define, at a distance, what constitutes legitimate and what constitutes illegitimate forms of the exercise of freedom and mobility. The externalization of control is not just a technical process of managing threats to the EU; rather, it is also a way of imagining and ordering the European space by defining and constructing migration as a form of illegitimate movement from which the EU-order should be protected.

5.

CONCLUSION

In this chapter, we have argued that current theories of governance and Europeanization are unable to consider how the boundaries of the Union are shaped through its relations with the external environment. Taking the EU as an independent variable, the Europeanization literature works with a static notion of the nature of the Union (see introduction to this volume; chapter by Mastenbroek and van Keulen in this volume; chapter by Aalberts in this volume). Therefore, this chapter has argued for a different conceptualization of governance as the conduct of conduct, arguing that the project of European integration can be understood as a project aimed at establishing the optimal economic and social conditions under which individuals can exercise their freedom. The establishment of a European space in which individuals can conduct their freedom depends on facilitating measures (European citizenship, the abolishment of internal borders) as well as on more security and control-oriented types of governance. This chapter argued that the externalization of migration policies to third countries (remote control, remote protection and capacity-building) can all be understood as relatively recent and innovative attempts to govern the proper conduct of freedom within the EU through governance aimed at the management of the improper and dangerous exercise of freedom. More specifically, the externalization of migration policies are deterritorialized forms of control that seek to render those forms of improper conduct of freedom increasingly immobile. It should be noted here that the governance of mobility and freedom is by no means the logical next step or functional spillover from more facilitating measures. The management of migration is not just an instrumental response 25

Foucault has made a similar statement about racism, which first function is to fragment the population in worthy and less worthy individuals and to subject each category to different forms of governance (Foucault 2003, 255). In this view, inclusive facilitating measures on the one hand and exclusive control-oriented measures on the other are not two opposites but two sides of the same coin.

Governing Mobility

247

to an objective problem; to the contrary, the association of migration with terrorism, organized crime, human smuggling and the drugs trade constructs migration as a security issue (rather than, say, an issue of economic or humanitarian concern) (Huysmans 1995; Doty 1996; Bigo 2000). The conduct of freedom, therefore, is not without social costs as it risks rendering immobile the most vulnerable groups that seek entrance to the EU. Moreover, control policies that seek to curb the improper conduct of freedom paradoxically also support these forms of conduct. For the more difficult it becomes to enter the European Union legally, the more immigrants are forced to rely on illegal forms of mobility such as human smuggling to entry the European space (see Andreas 2000). Also, it is not unlikely that the representation of migrants as a danger has repercussions for the position of migrants that already live within the European Union (Boswell 2003a). For these reasons, it could be argued that the externalization of migration policies should be focused less on restrictive policies than on preventive policies that attack the root causes of migration such as economic deprivation and political oppression. This would mean that governance does not target the conduct of freedom by migrants but the conditions under which immigrants are likely to migrate. Migration policies, then, would turn into development policy. But, while the targeting of root causes seems more humane than control-oriented types of governance, this is not necessarily the case. If a more preventive approach is still structured by a security discourse, it might only lead to a blurring of the security and development field and a distortion of development policies (e.g. by making security concerns a precondition for development) (see Boswell 2003b; cf. Duffield 2001). By way of conclusion, then, it might be argued that the increased emphasis on security and control in the governance of freedom might well be indicative of a broader shift in the EU in which the project of European integration is driven by the distribution of danger and the political principle of fear (Bigo 2002). If this is true, the conceptualization of governance as the conduct of conduct could generate interesting and important new insights into the political and social transformations in the EU by examining in which ways the governance of freedom and definitions of the proper conduct of freedom have changed over time. Defining governance as the conduct of conduct, then, would not only be able to theorize more systematically the ways in which the EU seeks to mould the conduct of freedom between itself and the external environment but could provide important new insights into the dynamics, norms and discourses that underpin representations of the EU and which determine the way it is governed and ordered.

Rens van Munster and Steven Sterkx

248

REFERENCES Aalberts, T. E., 2004, The future of sovereignty in multilevel Europe – A constructivist reading, Journal of Common Market Studies 42(1):23-46. Andreas, P., 2000, Border Games: Policing the U.S.-Mexico Divide, Cornell University Press, Ithaca. Barry, A., 1993, The European Community and European government: harmonization, mobility and space, Economy and Society 22(3):314-26. Barry, A., 1994, Harmonization and the art of European government’, in: A New Europe? Social Change and Political Transformation, C. Rootes and H. Davis, eds, UCL Press, London, pp. 39-54. Bigo, D., 2000, When two become one: internal and external securitisations in Europe, in: International relations theory and the politics of European integration, M. Kelstrup and M. C. Williams, eds, Routledge, Londen, pp.171-204. Bigo, D., 2002, Security and immigration: toward a critique of the governmentality of unease, Alternatives 27(special issue):63-92. Börzel, T. A., and Risse, T., 2003, Conceptualizing the domestic impact of Europe’, in: The Politics of Europeanisation, K. Featherstone and C. M. Radaelli, eds, Oxford University Press, Oxford, pp. 55-78. Boswell, C., 2003a, European migration policies in flux. Changing patterns of inclusion and exclusion, Chatham House Papers, The Royal Institute of International Affairs, Blackwell, London. Boswell, C., 2003b, The ‘external dimension’ of EU immigration and asylum policy, International Affairs 79(3):619-38. Bretherton, C., and Vogler, J., 1999, The EU as a Global Actor, Routledge, London. Caporaso, J. A., 1996, The European Union and forms of state: Westphalian, regulatory or post-modern?, Journal of Common Market Studies 34(1):29-52 .. Cederman, L-E., 2001, Political boundaries and identity trade-offs, in: Constructing Europe’s Identity: The External Dimension, L-E. Cederman, ed., Rienner, London, pp. 1-34. Dean, M., 1999, Governmentality. Power and Rule in Modern Society, Sage, London. Dean, M., 2002, Liberal government and authoritarianism, Economy and Society 31(1):37-61. Doty, R. L., 1998/99, Immigration and the politics of security, Security Studies 8(2/3):76-93. Duffield, M. R., 2001, Global Governance and the New Wars: The Merging of Development and Security, Zed Books, London. Eising, R., 2003, Europäisierung und Integration. Konzepte in der EU-Forschung, in: Europäische Integration, 2nd ed., M. Jachtenfuchs and B. Kohler-Koch, eds, Leske und Budrich, Opladen, pp. 379-426. Fearon, J., and Wendt, A., 2002, Rationalism v. constructivism: a skeptical view, in: Handbook of International Relations,W. Carlsnaes, T. Risse and B. A. Simmons, eds, Sage, London, pp. 52-72. Forster, J., 2002, Sanctions imposed on carriers: impact on migration flows and asylum requests, in: Managing Migration Flows and Preventing Illegal Immigration. Schengen Justice and Home Affairs Colloquium, C. Faria, ed., EIPA, Maastricht, pp. 25-8. Foucault, M., 2003, Society Must Be Defended, Allen Lane, London. Friis, L., and Murphy, A., 1999, The European Union and Central and Eastern Europe: governance and boundaries, Journal of Common Market Studies 37(2):211-32. Gordon, C., 1991, Governmental rationality: an introduction’, in: The Foucault Effect. Studies in Governmentality, G. Burchell, C. Gordon and P. Miller, eds, Chicago University Press, Chicago, pp. 1-52. Guiraudon, V., 2000, European integration and migration policy: vertical policy-making as venue shopping, Journal of Common Market Studies 38(2):251-71..

Governing Mobility

249

Guiraudon, V., 2003, Before the EU border: remote control of the ‘huddled masses’, in: In Search of Europe’s Borders, K. Groenendijk, E. Guild and P. Minderhoud, eds, Kluwer Law International, The Hague, pp. 191-214. Hill, C., 1993, The capability-expectations gap, or conceptualizing Europe’s international role, Journal of Common Market Studies 31(3):305-28. Hix, S., 1998, The study of the European Union II. The ‘new governance’ agenda and its rival, Journal of European Public Policy 5(1):38-65. Hooghe, L., and Marks, G., 2001, Multi-Level Governance and European Integration, Rowman and Littlefield, Lanham. Huysmans, J., 1995, Migrants as a security problem: dangers of ‘securitizing’ societal issues, in: Migration and European Integration: The Dynamics of Inclusion and Exclusion, R. Miles and D. Thränhardt, eds, Pinter, London, pp. 53-72. Huysmans, J., 1999, The European Union and the securitization of migration, Journal of Common Market Studies 38(5):751-77. Huysmans, J., forthcoming, Insecurity and the Political. On the Political Significance of Securitising Migration. Jachtenfuchs, M., 1995, Theoretical perspectives on European governance, European Law Journal 1(2):115-33. Jachtenfuchs, M., 1997, Conceptualizing European governance, in: Reflective Approaches to European Governance, K. E. Jørgensen, ed., MacMillan, London, pp. 39-50. Jachtenfuchs, M., 2001, The governance approach to European integration, Journal of Common Market Studies 39(2):245-64. Jachtenfuchs, M., and Kohler-Koch, B., 2004, Governance and institutional development’, in: European Integration Theory, A. Wiener and T. Diez, eds, Oxford University Press, Oxford, pp. 97-116. Kohler-Koch, B., and Eising, R., eds, 1999, The Transformation of Governance in the European Union, Routledge, London. Lavenex, S., 2003, Europeanization and asylum policy, paper for 2nd ECPR Conference, Marburg, 18-21 September 2003. Loescher, G., and Milner, J., 2003, The missing link: the need for comprehensive engagement in regions of refugee origin, International Affairs 79(3):595-617. Marks, G., Hooghe, L., and Blank, K., 1996, European integration from the 1980s: statecentric v. multi-level governance, Journal of Common Market Studies 34(3):341-78. Noll, G., 2003, Visions of the exceptional: legal and theoretical issues raised by transit processing centres and protection zones, European Journal of Migration and Law 5(3):303-42. Noll, G., and Vedsted-Hansen, J., 1999, Non-Communitarians: refugee and asylum policies, in: The EU and Human Rights, P. Alston, ed., Oxford University Press, Oxford, pp. 360-410. Olsen, J. P., 2001, The many faces of Europeanization, ARENA Working Papers 01/2, www.arena.uio.no/publications/wp02_2.htm. Radaelli, C. M., 2000, Whither Europeanization? Concept stretching and substantive change, European Integration Online Papers 4(8), http://eiop.or.at/eiop/texte/2000-008.htm. Reermann, O., 2002, Cooperation with countries of origin and transit: readmission agreements, in: Managing Migration Flows and Preventing Illegal Immigration. Schengen Justice and Home Affairs Colloquium, F. Cláudia, ed., EIPA, Maastricht, pp. 13-24. Risse-Kappen, T., 1996, Exploring the nature of the beast: international relations theory and comparative policy analysis meet the European Union, Journal of Common Market Studies 34(1):53-80. Rose, N., 1999, Powers of Freedom. Reframing Political Thought, Cambridge University Press, Cambridge.

250

Rens van Munster and Steven Sterkx

Schmitter, P. C., 1996, Imagining the future of the Euro-polity with the help of new concepts, in: Governance in the European Union, G. Marks, F. W. Scharpf, P. C. Schmitter and W. Streeck, eds, Sage, London, pp. 121-50. van Selm, J., 2002, Immigration and asylum or foreign policy: the EU’s approach to migrants and their countries of origin, in: Migration and the Externalities of European Integration, S. Lavenex and E. M. Uçarer, eds, Lexington Books, Lanham, chapter eight. Verstraete, G., 2001, Technological frontiers and the politics of mobility, New Formations 43(special issue):26-43. Walker, R. B. J., 2000, Europe is not where it is supposed to be, in: International Relations Theory and the Politics of European Integration, M. Kelstrup and M. C. Williams, eds, Routledge, London, pp. 14-32. Walters, W., 2002, Mapping Schengenland: denaturalizing the border, Society and Space 20(5):561-80. de Wilde, J. H., forthcoming, European Studies beyond the Union. The case of multilevel governance’, in: The Changing Focus in European Studies, D. Hansen, G. Kristensen and A. Vanags, eds. Wind, M., 1997, Rediscovering institutions: a reflectivist critique of rational institutionalism’, in: Reflective Approaches to European Governance, K. E. Jørgensen, MacMillan, Basingstoke, pp. 15-38. Zolberg, A. R., 2003, The archeology of ‘remote control’, in: Migration Control in the North Atlantic World: The Evolution of State Practices in Europe and the United States from the French Revolution to the Inter-War Period, A. Fahrmeir, ed., Berghahn, New York.

SOVEREIGNTY RELOADED? A Constructivist Perspective on European Research Tanja E. Aalberts∗ Abstract:

1.

This chapter addresses three issues, beginning with the sovereignty puzzle that emerges from multilevel governance analyses (in terms of the endurance of sovereignty within structures of overlapping authorities), it suggests supplementing the static view of multilevel governance with the dynamic perspective of Europeanization literature as an important step forward for the next generation of EU studies. In addition, it calls for a ‘constructivist turn’ in order to elaborate the dynamics identified by Europeanization approaches. It is argued that this provides the key to the sovereignty puzzle by analysing the link between interaction and identity. Finally, the constructivist perspective of the mutual constitution of structure and agency is advocated as a fruitful lane for the third wave of EU research as a way to overcome its struggles with unidirectional, causality notions of ‘bottom-up’ and ‘top-down’ relationships within multilevel governance structures.

INTRODUCTION

In the history of European studies there has always been a love-hate relationship between European integration and the (sovereign) memberstates. The dynamics of integration are often portrayed as balancing along a continuum between the extremes of survival of the nation-state versus the Union as (developing into) a new ‘suprastate’. With the emergence of multilevel governance analyses in the 1990s, attention has shifted from the process of integration to the emerging political system, leading to a description of the European Union (EU) as a complex and dynamic institution, with shifting patterns of authority amongst multiple actors across several levels, combining integration and cooperation. However, this does

251

R. Holzhacker and M. Haverland, (eds.), European Research Reloaded: Cooperation and Integration among Europeanized States, 251-272. © 2006 Springer. Printed in the Netherlands.

252

Tanja E. Aalberts

not settle the sovereignty issue. To the contrary, ‘multilevel governance’ renders it even more relevant and puzzling. After all, it directly challenges the traditional view of a clear-cut separation between inside and outside (Walker 1993). In the conventional reading sovereignty balances on this distinction and is Janus-faced: looking both inwards (in terms of governmental supremacy) and outwards (connoting the absence of an overarching authority) at the same time. Sovereignty is conceived as a legal, absolute and unitary condition (James 1999), and is, ultimately, indivisible. It hardly needs elucidation that such a depiction fits uneasily with reality in the EU-context, as described by multilevel governance (see also Vollaard’s discussion on notions of territoriality, this volume). Multilevel governance in some form or another has been widely accepted as a description of policy-making within the EUropean polity. It also forms the context in which a large number of chapters in this volume have been written, as it is the intensified interconnectedness and concurrent new modes of interaction between different levels of governance which necessitates the proposed ‘third wave’ of research on the EU (Haverland and Holzhacker, this volume). That the sovereignty issue transpires almost automatically (if indirectly) at a more concrete level too, becomes clear when considering the specific issue areas as discussed in the previous chapters. Generally, it is at the heart of the discussion on qualified majority voting, which was one of the key issues of the 2003 European Convention (see Leuffen and Luitwieler, this volume).1 Moreover, multilevel governance is not limited to so-called ‘low politics’, but includes areas such as internal security, defence and immigration (see Van Munster and Sterkx, this volume), as well as monetary and fiscal policies (see chapters by Van Esch and Schäfer, this volume) – issues which traditionally are considered to be at the very heart of sovereignty (see for instance Spruyt 1994). In addition, three chapters focus on directives (Kallestrup, Versluis, and Mastenbroek and Van Keulen, this volume), which are a tangible outcome of the supremacy and direct legal effect of community law in the legal order of the member-states (even if they are not directly applicable, see Craig and De Búrca 1998), as the European Court has claimed in the Van Gend and Loos Case.2 Whereas these developments at first hand appear to connote a zero-sum game between sovereignty and multilevel governance, member-states still play a key role in these governance structures (see notably Marks, et al., 1995). Apparently, sovereign statehood is not ‘obsolete’ or dead. After all, 1

In addition, the progress with regard to the integration of a Charter of Rights and the development of a European Constitution irrevocably change the formal set up. As this would include written changes to the founding documents, member states would lose their status as Herren der Verträge (Wind 2001). 2 Case 26/62, NV. Algemene Transporten Expeditie Onderneming van Gend en Loos v.Nederlandse Administratie der Belastingen [1963] ECR 1, [1963] CMLR 105.

Sovereignty Reloaded

253

when it comes to the crunch, member-states can still appeal to their status as sovereigns. From a governance perspective there is ‘irritating evidence’ of the ‘unrestricted vitality of national governments’ (Kohler-Koch 1996b, 364). Combined with the recognition that the EU is far from a (supra)state itself, this leaves us with a rather ambiguous picture of sovereign statehood. What comes to the fore is a patchwork or amalgam of differing and overlapping political arrangements (see chapter by Vollaard, this volume). At the same time, a pure legal reading of sovereignty as a juridical principle seems hardly satisfactory, as it does not capture the dynamics of the alleged postmodern context, and the changing institutions as described by multilevel governance. Unfortunately, the multilevel governance literature leaves the sovereignty issues by and large unaddressed. While it indeed provides a ‘compelling metaphor’ (Rosamond 2000) of the dispersal of authority within the contemporary polity of the EU, and appears to be a step forward for European Studies as it moves beyond the discussion of sovereignty as something that is ‘ever present or about to disappear’ (Walker 1990, 168), multilevel governance fails to discuss how sharing of authority amongst a variety of actors relates to the sovereign(ty of) member-states as key actors in this structure. This chapter sets out to analyse the puzzle of the resilience of sovereign statehood within a multilevel governance context. In order to try and make sense of this paradoxical state of affairs, there is a need for conceptual and theoretical clarification. A persistent objection against multilevel governance is its focus on description, at the cost of theorization. In this chapter, I propose to combine the ‘governance turn’ in EU studies with both the notion of Europeanization as the rejoinder to the classical, unilinear integration theories (see Haverland and Holzhacker, this volume) and the ‘constructivist turn’ in IR theory (Checkel 1998). As it focuses on the dynamics behind the emergent European political order, and the multifaceted interrelationships between the different levels within this order, Europeanization can serve as frame of reference for the analysis of the continuation and change of the institution of sovereignty within the alleged postmodern EU-context. Constructivism adds to this a theoretical notion on the ‘construction of social reality’, and a conceptualization of identity formation (see Wendt 1999), as well as, crucially, the mutual constitution of structure and agency (see notably Wendt 1987; Gould 1998; Adler 2002). Together this will help us to better understand the ‘nature of the beast’ (Risse-Kappen 1996) in terms of an evolving multi-level polity beyond the member-states. Moreover, it will enable us to account for the alleged ‘schizophrenic’ character of sovereignty, i.e. the changing but prolonged status of sovereign statehood within a multilevel governance context. In

Tanja E. Aalberts

254

addition it is argued that the structure-agency discussion might turn out to be beneficial to the Europeanization debate in its struggles with causality notions of ‘bottom-up’ and ‘top-down’ relationships. As set out in the introductory chapter, so far this literature has had a hard time in dealing with the interactive processes of mutual change and influence between the member-states and European institutions. In my view, a constructivist perspective has a lot to offer in terms of conceptual clarification in this broader context too and as such can provide helpful insights at this turning point towards the so-called third wave of research on the EU.3 This chapter proceeds as follows. The next section will give an overview of the multilevel governance approach (section 2.1) and discusses its remnant statism (section 2.2). It will be argued that multilevel governance is unable to deal with the endurance of sovereignty in contemporary EUEurope because of a static view. Section 2.3 will briefly discuss the lack of ‘cross-fertilization’ between multilevel governance and Europeanization approaches, and maintains that the added value of the latter literature consists of a dynamic dimension. In order to develop this tentative ‘processual view’(see Van Munster and Sterkx, this volume) and to overcome the struggles with unidirectional biases, it is argued that the Europeanization approaches could benefit from the structure/agency debate that prevails in IR theory. Constructivism has elaborated this in terms of ‘mutual constitution’ within a framework of intersubjective understandings (section 3.1) and analyses the link between interaction and identity (section 3.2). Section 3.3 then discusses how these elements together enable us to clarify the ‘continuity in change’ of sovereign identities in a postmodern, multilevel governance context. To conclude, section 4 links the discussion back to the broader framework of the book and suggests how it can advance the proposed ‘third wave’ in EU studies.

2.

MULTILEVEL GOVERNANCE

2.1

Key points

Whereas originally European Studies consisted of theories of integration, since the nineties there is a move away from seeking to understand the process of integration and to predict where it is heading, towards analysing the ‘nature of the beast’ that had emerged in the meantime (Risse-Kappen 3

For other (calls for) constructivist approaches to EU studies see amongst others Christiansen et al. (2001), Checkel (1999; 2001b), Jupille et al. (2003), Jørgensen (1997b), Risse (2004), Shaw and Wiener (2000), Wiener and Diez (2004), as well as special issuses of the Journal of European Public Policy 6(4), 1999, and Comparative Political Studies 36(1/2), 2003.

Sovereignty Reloaded

255

1996; cf. Puchala 1972). In the process of theorising integration, the emerging ‘Euro-polity’ had previously been left underdeveloped. However, the ambiguous picture that has evolved from the integration process so far has prompted discussion about the European order as ‘the first truly postmodern international political form’ and ‘multiperspectival polity’ (Ruggie 1993, 140, 72; Anderson 1996; Caporaso 1996), in which states ‘…operate within a much more complex, cross-cutting network of governance, based upon the breakdown of the distinction between domestic and foreign affairs, on mutual interference in each other’s domestic affairs, on increasing mutual transparency, and on the emergence of a sufficiently strong sense of community to guarantee mutual security’ (Wallace 1999, 519). The multilevel governance approach captures the mood of these (postmodern) times well. Instigated by insights from domestic and comparative politics, multilevel governance sets out to study the EU as a polity. As such the EU as a political system has moved from explanans to explanandum (Jachtenfuchs 2001; Caporaso 1996); and hence the criticism that multilevel governance does not address the ‘causal motor of integration’ (Jordan 2001, 201) seems off the mark as that is precisely what it did not set out to do (see George 2004). Besides this move beyond the intergovernmentalism vs supranationalism/neofunctionalism debate, multilevel governance also tries to overcome the disciplinary cleavages between Comparative Politics and IR theory. Indeed the notion of governance indicates a transgressing of the boundaries between the inside and the outside, between domestic politics and foreign affairs, and between public and private spheres. The additive multilevel in turn indicates the attempt to encompass the paradoxical yet parallel developments of increasing centralization (towards the European level) and regionalization (to subnational as well as private institutions) (Jørgensen 1997a).4 Drawing from several strands of multilevel governance, Hooghe and Marks (2001a, 3-4) summarize the approach into three main characteristics. First, leaving behind the intergovernmentalist notion of the decision-making monopoly of national governments, multilevel governance is characterized by a sharing of competencies amongst a variety of actors at a variety of levels. Much in line with supranationalism, the focus is on the autonomous role played by the European institutions (notably the Commission) in the policy-making process. Rather than acting as agents of national governments, these have become actors in their own right. Moreover, they increasingly have an independent impact within the jurisdiction of memberstates; this is illustrated most clearly in the case of the European Court of 4

Rosenau has suggested referring to these combined developments fragmentation and integration as ‘fragmeration’. For a most recent elaboration, in which he links the governance approach in IR theory to the ‘governance turn’ in European Studies, see Rosenau (2004).

256

Tanja E. Aalberts

Justice. Besides these supranational actors, both subnational and nongovernmental actors have gained access to, and an independent impact on the decision-making process. This has resulted in a second characteristic, namely the undermining of the conventional separation between domestic and international politics. Whereas state governments used to be important gatekeepers when it came to representation of interests at the supranational level, this role has been circumvented by new lines of communication and representation due to multiple points of access for subnational and transnational/non-governmental actors. The once clear-cut separation between internal and external affairs has been blurred, and the domestic versus international arenas have become ‘almost seamless’ within the EU context (Hooghe and Marks 2001a, 28, 78). Overall, one can distinguish a tripartite move away from national governments: i) upwards, as a most direct consequence of European integration; ii) downwards, because of regionalization and subsequent empowerment of subnational actors; and iii) sideways, to non-governmental actors (e.g. public-private partnerships). This latter shift is foremost elaborated in the analysis of EU politics in terms of ‘network governance’. In this reading, ‘[p]olitical reality is held to be depicted far more accurately in terms of a network that can trace the tight, compact patterns of interaction between public and private actors of the most varied nature and at the same time able to make clear that we are not, in fact, dealing with a set of pre- or subordinate relationships, but instead with a bargaining process between strategies of action being pursued by mutually dependent, but at the same time autonomous, actors’ (Kohler-Koch 1996b, 369-70).5 The third element of multilevel governance seems to follow logically ; it states that a new mode of decision-making has emerged, which has institutionalized the loss of control for national governments. It can be characterized as a ‘multi-tier negotiating system’, in which issues are being dealt with at several levels concomitantly, throughout all phases of the process (Kohler-Koch 1996a). Still, this does not imply a complete loss of any say national governments might have. To the contrary, multilevel governance does not argue the obsolescence of state actors, but continues to appoint them a key role in European politics (Marks, et al., 1995). Hence while multilevel governance appears to share some of the characteristics of both intergovernmentalism (centrality of member-states) and supranationalism (European institutions as autonomous actors), it clearly distances itself from 5 See also Kohler-Koch and Eising (1999) and Jachtenfuchs (1995; 2001). Hooghe and Marks have only recently added the move sideways to their analysis (Hooghe and Marks 2001b, 2003; Marks and Hooghe 2004). Hence Rosenau’s (2004) criticism of multilevel governance, that it implies hierarchy, and lacks to address the horizontal dimension, only counts for the early multilevel governance literature by Liesbet Hooghe and Gary Marks.

Sovereignty Reloaded

257

that debate. It does so by emphasising on the one hand that a member-state is but one actor amongst many others. And on the other hand by interpreting governance not as something ‘above’ the state (like a suprastate structure), but as governance beyond the state (Jachtenfuchs 1997). This connotes governance with governments. And national governments still are primus inter pares as well (Jessop 2004, 57). As such, multilevel governance seems to present a compromising ‘in-between’; it rejects the ‘either-or’ discussion with interstate bargaining and transnational coalition building as mutually exclusive options (Risse-Kappen 1996), and regards the current structures as an intermediate arrangement in its own right (Anderson 1996; Kohler-Koch 1996a).6

2.2

Remnant statism

Linking this to the central issue of this chapter, the burning question is how this intermediate arrangement relates to the sovereign status of memberstates. Crucial in light of our discussion, multilevel governance combines the transformation of the nation-state with its erosion (Bache and Flinders 2004). Yet, multilevel governance can still be regarded as a statist approach. While governance approaches are often juxtaposed to statist (or state-centred) approaches (Marks, et al., 1996; Jessop 2004), multilevel governance does not fully escape the ‘practice of locating the state at the centre of the scheme’ (Rosenau 2004, 40). To be clear, to some extent it does stay safe from state centric hazards. For one thing, the (member-)state is not examined in isolation, and treated consequently as an independent variable (Jessop 2004, 51). Moreover, multilevel governance distinguishes between the state as institution and state executives, who can pursue their own interests. Hence, differing from intergovernmentalism, the state is no longer treated as a unitary actor (see f.i. Kohler-Koch 1996a; Hooghe and Marks 2001a). Finally, while both intergovernmentalism and supranationalism can be criticized for ‘fetishising formal constitutional and juridical features’ and ignoring de facto state capacities (Jessop 2004, 55), modalities of power are indeed central to multilevel governance analyses. With a focus on the ‘locus of control’ (Hooghe and Marks 2001a, 2) multilevel governance does indeed explicitly address the latter element.7 6

For a criticism of this possibility, see Neyer (2003). For a discussion of multilevel governance as a substitute for neofunctionalism, see George (2004). He argues that ‘… multilevel governance does not escape the dichotomy, but is simply a more sophisticated restatement of one side of it … although minus functional spillover, which of course is why it needs a new name.’ (George 2004, 108, 12). Jordan (2001) also identifies neofunctionalism as the intellectual roots of multilevel governance. 7 However, it can be questioned whether this is anymore ‘essential’ to sovereignty than formal constitutional features, as will be argued below.

258

Tanja E. Aalberts

However, while multilevel governance maintains the transformation of the member-state, rather than its robustness (intergovernmentalism) or its withering away and rescaling of statehood on a higher level (supranationalism/ neofunctionalism), there is some remnant statism in this approach, too. Apart from the obvious fact that states are still addressed as the key actors in governance arena, it also serves as a point of reference. Theoretically the member-state is still the referent from which the other actors are conceptualized, be it upwards, downwards or sideways. For instance, the role of transnational actors is interpreted as a move sideways, away from the state. In this context Shaw and Wiener (2000) speak of the paradox of the European polity, which appears a ‘near-state’ and is antithetical to stateness at the same time: ‘The risk of studying European governance then lies in the continuous revival of the idea of stateness, whether that takes the form of resistance against or reform towards the establishment of statelike patterns.’ (Shaw and Wiener 2000, 65, italics in the original). Multilevel governance approaches fail to address this paradox and to elaborate on how these changes are compatible with the resilience of the principle of sovereignty. If we put the above profile of multilevel governance next to the Westphalian framework of reference, an ambiguous picture transpires indeed. With the emerging governance structures outside the state on the one hand, and the undermining of governmental representation on the other, multilevel governance ‘escapes our conventional understanding of statehood’ (Hooghe 1996, 15) and seems to connote the ‘world turned inside out and outside in’ (Anderson 1996, 135). Indeed, at face value, multilevel governance appears a direct impingement of sovereign statehood, from three directions – both bottom-up (loss of gate-keeping); top-down (direct effect of EU law); and sideways (increasing role of non-state actors). As such, the developments described by multilevel governance challenge both the notions of international anarchy (external sovereignty) and governmental supremacy (internal sovereignty). In this context it is remarkable that most multilevel governance theorists address sovereignty at most anecdotally (with the exception of Marks, et al., 1995; for a critique of the readings of sovereignty within multilevel governance literature, see Aalberts 2004). In line with ongoing debates in IR theory, there is yet another way in which multilevel governance can be conceived as a statist approach. ‘Statism’ then not only connotes an exclusive focus on states as actors in the (inter)national arena, but more so to taking the state for a ‘given’, i.e. an immutable and so-called brute fact (see Bartelson 1998). This is also referred to as reification, which can be defined as: ‘…the apprehension of the products of human activity as if they were something else than human products – such as facts of nature, results of cosmic laws, or manifestations of divine will. … The reified world is … experienced by man as a

Sovereignty Reloaded

259

strange facticity, an opus alienum over which he has no control rather than as the opus proprium of his own productive activity’ (Berger and Luckmann 1991

[1966], 106, italics in the original).

Indeed, two prominent authors of multilevel governance, Liesbet Hooghe and Gary Marks, briefly address the flaw of reification, and accept the reading of states as social institutions. In their explication this results to a varying degree in differentiation from one’s environment as well as the possibility to act coherently. Along this line, they suggest that ‘[i]f states are viewed as sets of commonly accepted rules that specify a particular authoritative order, then one should ask how such rules may change over time, and whether and how they will be defended.’ (Hooghe and Marks 2001a, 74). This mirrors a notion of the relationship between member-states and the EU in regime-like terms. Still, this hardly seems an adequate picture, which does not really do justice to the institutional features that have advanced since the beginning of the integration process. If, according to a common definition, regimes are described as a set of ‘implicit or explicit principles, norms, rules and decision-making procedures around which actors’ expectations converge in a given area of international relations’ (Krasner 1983, 2), then the EU at a minimum connotes a ‘saturated regime’, founded on the core institution of the ‘embedded acquis communautaire’ (Christiansen, et al. 1999, 539). The EU connotes more than a regulative agreement between independent states.8 This is a gap in multilevel governance where Europeanization could provide a clarifying perspective, given its more explicit focus on dynamics of adaptation (see also Van Munster and Sterkx, this volume).

2.3

Multilevel governance vs Europeanization

In contrast to multilevel governance the focus of the Europeanization literature is (again) on process. However, rather than contemplating the imminent end product of European integration (as was the focus of the classical debate), attention is turned to the impact of the process as it is, i.e. the dynamics of the evolving European polity, and notably the changes in and among institutions involved. The relationship between Europeanization studies and multilevel governance approaches is a bit ambiguous. Generally speaking, these seem parallel but isolated branches of literature, with little cross-referencing. Alternatively the relationship is addressed only parenthetically. Just to pick a few examples as a (far from representative) illustration: handbooks like George/Bache (2001) and Rosamond (2000) focus on multilevel governance, 8

Moreover, in my reading regime theory is not particularly suited to analyse the progressing transformation of states within a multilevel governance context, as it starts from the realist assumption that states are self-interest maximizers, whose identities and interests remain constant throughout the process (cf. Krasner 1983). I will take up on this issue below.

260

Tanja E. Aalberts

with ‘Europeanization’ not even being listed in the index. The contributions to Cowles and Smith (2000), to mention another often quoted volume, discuss both multi-level governance and Europeanization, but not in connection to each other. Nor do the editors in their introductory chapter combine the two concepts.Cowles et al. (2001) obviously discuss Europeanization; and ‘multilevel governance’ is mentioned only once throughout the book. See also Featherstone and Radealli’s (2003b) most recent volume on Europeanization, with merely four hits for multilevel governance, three of which appear in the same chapter. Olsen in his discussion of the many faces of Europeanization, is most explicit in the sense that he describes one reading in terms of multilevel governance. Europeanization then ‘involves the division of responsibilities and powers between different levels of governance. All multilevel systems of governance need to work out a balance between unity and diversity, central co-ordination and local autonomy. Europeanization … implies adapting national and sub-national systems of governance to a European political centre and European-wide norms’ (Olsen 2002, 923-4). A possible exception to the practice of neglect is Jachtenfuchs and Kohler-Koch (2004), who dedicate a paragraph to Europeanization in their chapter on governance approaches. While they do call for an extension of the Europeanization agenda beyond policy-making to more institutional issues (such as shifting boundaries between public and private spheres – i.e. multilevel governance), they fail to explicitly address the link between the approaches. In my reading, Europeanization adds to multilevel governance as it focuses on the dynamics behind the emergent European political order, whereas multilevel governance is first and foremost a description of interlinkage between the levels within the European polity, in terms of the ‘locus of political control’. This descriptive focus leads to the often-launched criticism that multilevel governance lacks a proper theoretical framework (but see George 2004). The Europeanization literature, conversely, analyses the (continuing) processes of interaction and mutual adaptation. In this context it sets out to explore explanations for institutional change and continuity and explores the impact of the changing environment on the member-states. As such Europeanization could serve as a theoretical supplement to multilevel governance in order to substantiate this latter approach. However, Europeanization research is far from unproblematic itself. For one thing, if suffers from an overkill of definitions and applications (see Olsen 2002; Radaelli 2000, and Binnema, this volume). A popular definition is offered by Radaelli (2000, 3), who describes Europeanization as: ‘[p]rocesses of (a) construction (b) diffusion and (c) institutionalization of formal and informal rules, procedures, policy paradigms, styles, ‘ways of doing things’ and shared beliefs and norms which are first defined and consolidated in the

Sovereignty Reloaded

261

making of EU decisions and then incorporated in the logic of domestic discourse, identities, political structures and public policies’

This definition seems most helpful for the purposes of this chapter as it addresses a wider scale of dimensions than most other definitions, some of which limit their focus to changes in policy-making. Moreover, it hints at structure/agency issues as it includes both notions of construction and institutionalization and consolidation, which are then linked to identity issues. As will be shown in section 3, this is an important issue with regard to our sovereignty puzzle. Still, the above definition could suffer from a second and more critical flaw of Europeanization research so far: a unidirectional perspective. This has resulted in either a bottom-up or a topdown perspective, which has impeded abilities to account for ‘feedback loops’, as has been convincingly argued on several occasions. In order to overcome these biases, this book set out to introduce a third wave in European Studies, which considers both processes as systematically related processes as a kind of rotating cork-screw (see Haverland/Holzhacker, this volume).9 For the further theoretical exploration of these interaction processes between different levels and actors, constructivism provides some interesting insights in terms of the structure/agency debate in IR theory. Whereas IR theory has often been pushed aside as irrelevant for the study of the European Union given the moribund dichotomy between intergovernmentalism and supranationalism that emerged from that discipline, and Comparative Politics is proposed as a better perspective to study the emerging governance structures (cf. Jachtenfuchs 2001), it will be argued that the constructivist turn can illuminate the dynamics of mutual adaptation. In addition, and particularly relevant to our analysis of the sovereignty puzzle, is the constructivist notion of identity formation.10 This will allow us to explore both change and continuity, for it is a misreading of constructivism that it can only account for change. As such it adds a theoretical framework to Europeanization, with which we can explore both change and continuity without falling into the trap of reification.

9

Radaelli’s definition fits this cork-screw model insofar as it does include a time dimension (‘first …. then’), which Holzhacker and Haverland (this volume) advocate as an important element of the third wave, too. A constructivist perspective adds to this a notion of the mutual constitution of structure and agency, in which case it might not always be so clear what the time sequence is. 10 This is the crucial difference between constructivism and (sociological) institutionalism. Whereas these labels are sometimes used interchangeably (see for instance Börzel and Risse 2003), institutionalism limits its focus to the analysis of change in behaviour (and maybe preferences), whereas constructivism takes also interests and identities to be endogenous to interaction.

Tanja E. Aalberts

262

3.

A CONSTRUCTIVIST PERSPECTIVE

3.1

Key elements

The above quote about ‘reification of the state’ can be considered one of the axioms of constructivism. Crucial in this approach is the distinction between the natural and the social world, between physical, ‘brute’ facts and ‘institutional facts’, i.e. facts that depend on an intersubjective frame of reference (Searle 1995). In other words: ‘even our most enduring institutions are based on collective understandings … they are reified structures that were once upon a time conceived ex nihilo by human consciousness … [which] were subsequently diffused and consolidated until they were taken for granted’ (Adler 1997, 322). This focus on intersubjectivity as basis for social reality, and on the dynamics between structure and agency, can help us to unravel the sovereignty puzzle in multilevel governance Europe. In order to do so, I will make use of Wendt’s analysis. He is considered one of the key figures in constructivism and his book (Wendt 1999) is one of the most extensive elaborations of social constructivism. More specifically, his analysis of identity formation on the basis of intersubjective understandings will prove to be clarifying for our analysis of sovereignty in a multilevel governance context.11 Crucial for the understanding of identities is the mutual constitution of structure and agency. In a nutshell: the identity of social entities (read: member-states) are not intrinsic, but relational and emerge from structures. These consist of (i) shared knowledge; (ii) material resources; and (iii) practices (Wendt 1995).

11

For a more critical analysis of Wendt's discussion of sovereignty, see Bartelson (1998) and Aalberts (2002).

Sovereignty Reloaded

263

interaction

among agents

mutual understandings structures (institutional facts)

‘what states do and what they are’

regulation behaviour + constitution identity

Figure 1. Identity construction a la Wendt

It follows that structures are not exogenously given either, but emerge through interaction amongst agents. Interaction and practice result in the constitution of shared meanings, which develop into structures that in turn affect behaviour and constitute identities (see figure 1). Thus, key structures are intersubjective, i.e. social instead of material, and exist by grace of interaction. As such they are continuously in process. Consequently, structure has no meaning independent of (state) practice to accept certain institutions as a basic rule in international politics (i.e. agency). Hence the focus is on ‘distribution of knowledge’ and intersubjective understandings. In that sense, the ‘real world’ of international politics consists of brute, material facts and institutional facts alike (Wendt 1999, 140, 10; Wendt and Friedheim 1995). As identities in turn depend on these intersubjective structures, they are inherently relational and, consequently, changeable. Still this does not mean that they are fluid and instable. Not at all, as structures tend to support certain behaviours, while discouraging others and as such can be considered to have a conservative impact. In contrast to an individualist ontology – in which case all the constitutive power lies with the agents – constructivism emphasizes the constitutive feedback of structures on identities, too. Intersubjective constructions confront agents as obdurate institutional facts (Wendt 1994, 389), which is nicely captured by the notion of ‘structure’.12 Hence, while they are intersubjective in their nature, this does not mean institutional facts can easily (be) change(d), as for instance Vollaard (this volume) analyses with regard to the territoriality principle. 12 Actually, Wendt refers to ‘social facts’. However, in my opinion the label institutional fact is more accurate. For a distinction between social vs institutional facts, see Searle (1995).

264

Tanja E. Aalberts

Once constituted, institutional facts ‘behave as’ and are ‘acted upon’ as brute or hard facts. Nevertheless, they are intersubjective and exist on the basis of interaction. Interaction as such forms the foundation of social reality. In this sense, both structures and (identities of) agents are dependent variables at the same time – they are mutually constitutive. One author aptly summarized the crux of the debate as follows: structure is both the outcome and the medium of (inter)action, and ‘[a]n actor can act socially only because there exists a social structure to draw on, and it is only through the actions of agents that structure is reproduced (and, potentially, transformed)’ (Dessler 1989, 452).

3.2

Constitution of (sovereign) identity: two dynamics of interaction

The quote on the link between structure and agency conceals that this dual dynamic of reproduction and transformation counts for structure and (identity of) agents alike. Wendt discusses two ways in which interaction can affect identity. The basic relationship is the connection between what actors (here: states) do and what they are. Interaction brings about and sustains identity. On the one hand, interaction can serve to reinforce the status quo identity of agents. When agents are communicating, they are not only pursuing their (self-interested) goals, but ‘are also instantiating and reproducing a particular conception of who they are’ (Wendt 1999, 341). This is the constitutive effect of interaction. In the process of communication, agents participate in the joint constitution of their identities and counteridentities. Hence it should be noted, that even when identities (and interests) remain relatively stable, this constancy is still endogenous to interaction.13 The second dynamic of interaction builds upon this endogenity of identity. As they are always in process, and results of practice, the boundaries of the agent’s identity (here the member-state) are, in principle, always ‘at stake’. Precisely because identity is a ‘dependent variable’, this allows for the emergence of collective identities – as is the case in the EU context. This means that as a result of interaction and the shared meanings evolving from that, a sort of ‘super-ordinate identity’ can develop, above and beyond the state.14 Hence, state identity is both constituted, reinforced and 13

This in contrast to rational choice and regime theory, which considers interests and identity to be given and as such exogenous to interaction. Arguably multilevel governance shares this individualist ontology in combination with an instrumental logic of action (Checkel 2001a; Risse 2004; George 2004). 14 This general idea can be traced back to Karl Deutsch. In the 1950s he pointed out the likelihood of the formation of new (security) communities on the basis of a ‘shared identity’ as a consequence of increased social interactions within the EC-context (Deutsch 1953; Deutsch, et al., 1957; see also Adler and Barnett 1998).

Sovereignty Reloaded

265

can be transformed through interaction. And it is this understanding of the concomitant dynamics of identity construction and change that can be useful for our discussion about multilevel governance and sovereignty. Sovereignty is first and foremost part of what Wendt has labelled a state’s social or role identity. Role identities are not based on intrinsic properties an agent might have, but only exists in a social context, in relation to other agents (who vice versa possess relevant counter-identities). As such, role identities cannot be enacted unilaterally, but exist on the basis of shared expectations and collective understandings. Sovereignty is part of such a role identity, as it can be conceived as a status granted by fellow-states by means of recognition. It is recognition that renders sovereignty an institution shared by many. Consequently, sovereignty as an institutional fact exists by grace of a specific audience, ‘an international society which [still] recognizes the distinction between internal and external as valid and acts upon the belief in the existence of that distinction’ (Werner and de Wilde 2001, 288). Such an understanding of the contingent feature of state identity and the institution of sovereignty is crucial for our attempt to account for the continuity and change of sovereignty within multilevel governance Europe. The crucial element lies in the understanding that this institution is an ‘institutional fact’: it consists of intersubjective meanings that emerge from interaction. These institutional facts in turn not only affect behaviour, but concomitantly constitute identity, including sovereign identities.

3.3

Constructivist reading of sovereignty in multilevel governance structures

By conceiving sovereign statehood as the historical innovation it is, shifting the focus to the context of complex political practices of modern states, and the ‘minute rituals through which states are constantly made and remade’ (Walker 1991, 452), constructivism can help to clarify ‘what’s going on in Europe’ in terms of paradoxical developments of multilevel governance on the one hand, and the ‘irritating evidence of the unrestricted vitality’ of the institution of sovereignty on the other.’ How does multilevel governance fit into this picture? From the above discussion it follows that it can be seen as an emerging structure (i.e. multilevel polity) and process or practice, at the same time. To put it more accurately: multilevel governance consists of an (evolving) intersubjective structure emerging from, and therefore endogenous to, process and interaction. From the interaction amongst member-states mutual understandings transpire, which in turn lead to the development of intersubjective structures and institutional arrangements, i.e. multilevel governance structures. As aforementioned, multilevel governance academics reject a unitary view of the state. Hooghe and Marks (2001a) are most

266

Tanja E. Aalberts

explicit in this, and argue that the focus should be on state executives. Whereas this shows sensitivity to at least one aspect of the analytical problem of ‘statism’, it is only a first step, for such a perspective entails the risk of ending up with an individualist ontology (cf. neo-realism). Taking individuals as starting point for explaining collective patterns of action begets a bottom-up perspective towards the agency/structure relationship. Consequently, actors (agents) have a constitutive effect on structures, but such a perspective cannot account for the constitutive feedback of structures on agents. This is where the added value of the Europeanization approaches lies. Within this literature the attention is shifted to the impact of institutional arrangements (structure) on the member-states (agents). These intersubjective structures constitute the context that sets the boundaries in which the member-states (and, as one of the key features of multilevel governance, other actors) operate. This not only refers to constraints and opportunity structures in terms of their preferences and behaviour, but reaches further to the boundaries of ‘Self’, i.e. identity.15 As set out in the introduction to this volume, ‘if we truly want to [understand] the evolving multi-level political system of the EU and the member states, we must also acknowledge the interactive processes that feed back onto themselves’ (Haverland and Holzhacker, this volume), i.e. onto the agents. Europeanization hence addresses and potentially can account for the change in behaviour, and, crucially, identity of agents. Indeed, multilevel governance underlines the need to shift the focus from the disappearance of their identity as such (supranationalism) versus the givenness and constancy of identity (intergovernmentalism), to the transformation of member-state identities. However, the sovereignty puzzle does not merely entails change as such, but rather consists of the endurance of sovereignty within an alleged ‘nonsovereign’ context (consisting of overlapping authorities and the curtailment of state gatekeeping) of multilevel governance. Here constructivism provides helpful insights, by analysing sovereignty as an institutional fact that emerges from interaction which affects identity. The reality of sovereignty hence ‘consists in its use and acceptance’(Werner and de Wilde 2001, 304; Ruggie 1998). As long as states mutually accept each other as sovereigns, and tune their behaviour accordingly, they are. In essence, this is what their (sovereign) individuality entails. Hence, while Europeanization approaches are very correct in pointing out that interaction in terms of multilevel 15

Again, this is where constructivism adds a dimension to regime theory and institutionalism, which analyse institutions as ‘persistent sets of rules that constrain activity, shape expectations, and prescribe [behavioural] roles’ (Keohane 1988, 384). In this regard regimes connote specific institutions, as they are focused on particular issues.

Sovereignty Reloaded

267

governance influences the (identities of) member-states, this does not trounce upon sovereign individualities as such, because of a simultaneous sovereignty discourse. This is the ‘conservative’ dynamic of interaction. When states (or rather, state executives) are communicating in the European arena, they are constantly constituting and reproducing their mutual quality as sovereign states (irrespective of their particular vision for the future of Europe). In this sense, states can be considered self-fulfilling prophecies and sovereignty as an institutional fact can be said to entail a ‘sedimented discourse’ – a discourse that as a result of political and social practice has become relatively permanent and durable (Howarth 1995, 127-8, 32). As straightforward as it may appear, this dynamic of interaction ensures the continuation of sovereign identity within ostensibly incompatible and divergent multilevel governance structures. In sum, multilevel governance and sovereignty are not a zero-sum game. While the multilevel governance structure impacts on state identities, and statehood to an increasing degree is defined by EU-membership (cf. Risse 2004), these ‘Europeanized states’ retain their sovereign identity due to a simultaneous sovereignty discourse.16 The degree of Europeanization is not inversely proportional to sovereignty in terms of ‘more Europeanized = less sovereign’. Such a ‘formulae’ is flawed insofar as it disregards that sovereignty is an institutional fact that exists on the basis of a dual logic of interaction which lies at the basis of the mutual constitution of structure and agency.

4.

CONCLUSION

This chapter set out to address three issues that might have seemed rather distant at first sight. However, they are related and interlinked through the understanding of social reality and the constitution of structures and agents. The first issue followed from the lack of theoretical backbone within multilevel governance approaches, and in particular their failure to address the sovereignty puzzle that emerges from their analyses. In this context it was suggested to supplement multilevel governance with bottom-up and topdown dynamics in the Europeanization literature. Whereas Europeanists so far have not discussed this sovereignty puzzle either, they add a processual view to the static approach of multilevel governance, by focusing on the impact of the changing environment on member-states. This is an important, 16

Risse (2004) discusses the construction of European identity too, but focuses in this context on citizen’s identities, i.e. their ‘sense of belonging’ and (supra)national attachments. However, he reaches a similar conclusion: ‘studying questions of European identity highlights the importance of analysing the discursive construction of meanings’ (2004, 171; see also Wæver 2004).

Tanja E. Aalberts

268

if only first, step for the next generation in EU studies. In order to elaborate these dynamics in terms of ‘continuity in change’ with regard to sovereign identity, this chapter secondly advocated a constructivist perspective. Such a perspective not only clarifies our discussion through analysing the link between interaction and the continuation and/or transformation of (sovereign) identity, but also provides helpful insights in terms of mutual constitution of structure and agency. As a third issue, the structure/agency debate in turn is argued to be particular relevant to the general call for a third wave in EU research that provides the background for this book. So far Europeanization literature has suffered from a unidirectional perspective, emphasizing either bottom-up or top-down processes, without being able to account for ‘feedback loops’. It has been argued here that as soon as one acknowledges and aims to analyze these double dynamics, one is dealing with ‘mutual constitution’ and the structure/agency debate comes into the picture. Rather than isolating either the top-down or the bottom-up perspective with regard to causal impact, a constructivist perspective emphasizes that actors play an active part in the constitution of social structures, which in turn constitute their identities and relationships and influences their behaviour and preferences – which then again feed back into the institutional arrangements (structure).17 As such this would be a clarifying perspective and fruitful line of research for the third wave of EU research in its aim to tackle the dynamics of European integration and Europeanization in tandem.

REFERENCES Aalberts, T. E., 2002, Multilevel governance and the future of sovereignty: A constructivist perspective, Working Papers Political Science 04/2002, Amsterdam, Vrije Universiteit, http://politicologie.scw.vu.nl/wpps/web042002.pdf. Aalberts, T. E., 2004, The future of sovereignty in multilevel governance Europe – a constructivist reading, Journal of Common Market Studies 42(1):23-46. Aalberts, T. E. and Munster, R. V., 2003, The Third Debate revisited: The unfulfilled promise of social constructivism, Working Paper 14/2003, Centre for European Studies, University of Southern Denmark. Adler, E., 1997, Seizing the middle ground: Constructivism in world politics, European Journal of International Relations 3(3):319-63. 17 While Featherstone and Radaelli (2003) acknowledge the importance of notions of structure and agency for the Europeanization research, they still apply causality language, which does not fit a notion of mutual constitution (see Aalberts and Van Munster 2003). Risse (2004) addresses the usefulness of the structure/agency-debate too, but in his elaboration he tends to limit the focus to March and Olsen’s (1998) Logic of Appropriateness (as opposed to its rational, instrumental counterpart - the Logic of Consequences) and the impact on ‘behavioural effects’ and preferences, rather than the deeper and more abstract notion of ‘identity’.

Sovereignty Reloaded

269

Adler, E., 2002, Constructivism and international relations, in: Handbook of International Relations, W. Carlsnaes, T. Risse and B. A. Simmons, eds, London, Sage, pp. 95-118. Adler, E. and Barnett, M., eds, 1998, Security Communities, Cambridge, Cambridge University Press. Anderson, J., 1996, The shifting stage of politics: new medieval and postmodern territorialities?, Environment and Planning D: Society and Space 14(2):133-53. Bache, I. and Flinders, M., 2004, Themes and issues in multi-level governance, in: MultiLevel Governance, I. Bache and M. Flinders, eds, Oxford, Oxford University Press, pp. 1-15. Bartelson, J., 1998, Second natures: is the state identical with itself ?, European Journal of International Relations 4(3):295-326. Berger, P. L. and Luckmann, T., 1991 [1966], The Social Construction of Reality. A Treatise in the Sociology of Knowledge, London, Penguin. Börzel, T. A. and Risse, T., 2003, Conceptualising the domestic impact of Europe, in: The Politics of Europeanisation, K. Featherstone and C. M. Radaelli, eds, Oxford, Oxford University Press, pp. 55-78. Caporaso, J. A., 1996, The European Union and forms of state: Westphalian, regulatory or post-modern?, Journal of Common Market Studies 34(1):29-52. Checkel, J. T., 1998, The constructivist turn in International Relations theory, World Politics 50(2):324-48. Checkel, J. T., 1999, Social construction and integration, Journal of European Public Policy 6(4):545-60. Checkel, J. T., 2001a, Constructing European institutions, in: The Rules of Integration: Institutionalist Approaches to the Study of Europe, M. Aspinwall and G. Schneider, eds, Manchester, Manchester University Press. Checkel, J. T., 2001b, Why comply? Social learning and European identity change, International Organization 55(3):553-88. Christiansen, T., Jørgensen, K. E. and Wiener, A., 1999, The social construction of Europe, Journal of European Public Policy 6(4):528-44. Christiansen, T., Jørgensen, K. E. and Wiener, A., eds, 2001, The Social Construction of Europe, London, Sage. Cowles, M. G., Caporaso, J. A. and Risse, T., eds, 2001, Transforming Europe: Europeanization and domestic change, Ithaca, Cornell University Press. Cowles, M. G. and Smith, M., eds, 2000, The State of the European Union. Volume 5: Risks, Reform, Resistance and Revival, Oxford, Oxford University Press. Craig, P. and De Búrca, G., 1998, EU Law. Text, Cases, and Materials, 2nd edition, Oxford, Oxford University Press. Dessler, D., 1989, What is at stake in the agent-structure debate?, International Organization 43(3):441-73. Deutsch, K. W., 1953, Nationalism and Social Communication: An Inquiry into the Foundations of Nationality, Cambridge, MIT Press. Deutsch, K. W., Burrell, S. A., Kann, R. A. and Wagenen, R. W. v., 1957, Political Community and the North Atlantic Area: International Organization in the Light of Historical Experience, Princeton, Princeton University Press. Featherstone, K. and Radaelli, C. M., 2003, A conversant research agenda, in: The Politics of Europeanisation, K. Featherstone and C. M. Radaelli, eds, Oxford, Oxford University Press, pp. 331-41. George, S. A., 2004, Multi-level governance and the European Union, in: Multi-Level Governance, I. Bache and M. Flinders, eds, Oxford, Oxford University Press, pp. 107-26. George, S. A. and Bache, I., 2001, Politics in the European Union, Oxford, Oxford University Press.

270

Tanja E. Aalberts

Gould, H. D., 1998, What is at stake in the agent-structure debate?, in: International Relations in a Constructed World, V. Kubálková, N. Onuf and P. Kowert, eds, London, M.E. Sharpe, pp. 79-98. Hooghe, L., 1996, Introduction: Reconciling EU-wide policy and national diversity, in: Cohesion Policy and European Integration: Building Multi-Level Governance, L. Hooghe, ed., Oxford, Oxford University Press, pp. 1-26. Hooghe, L. and Marks, G., 2001a, Multi-level governance and European Integration, Lanham, Rowman & Littlefield. Hooghe, L. and Marks, G., 2001b, Types of multi-level governance, European Integration online Papers 5(11), http://eiop.or.at/eiop/texte/2001-011.htm. Hooghe, L. and Marks, G., 2003, Unraveling the central state, but how? Types of multilevel governance, American Political Science Review 97(2):233-43. Howarth, D., 1995, Discourse theory, in: Theory and Methods in Political Science, D. Marsh and G. Stoker, eds, London, MacMillan, pp. 115-33. Jachtenfuchs, M., 1995, Theoretical perspectives on European governance, European Law Journal 1(2):115-33. Jachtenfuchs, M., 1997, Conceptualizing European governance, in: Reflective Approaches to European Governance, K. E. Jørgensen, ed., London, MacMillan, pp. 39-50. Jachtenfuchs, M., 2001, The governance approach to European integration, Journal of Common Market Studies 39(2):245-64. Jachtenfuchs, M. and Kohler-Koch, B., 2004, Governance and institutional development, in: European Integration Theory, A. Wiener and T. Diez, eds, Oxford, Oxford University Press, pp. 97-115. James, A., 1999, The practice of sovereign statehood in contemporary international society, Political Studies 47(3):457-73. Jessop, B., 2004, Multi-level governance and multi-level metagovernance. Changes in the European Union as integral moments in the transformation and reorientation of contemporary statehood, in: Multi-Level Governance, I. Bache and M. Flinders, eds, Oxford, Oxford University Press, pp. 49-74. Jordan, A., 2001, The European Union - an evolving system of multi-level governance or government, Policy and Politics 29(2):193-208. Jørgensen, K. E., 1997a, Introduction: Approaching European governance, in: Reflective Approaches to European Governance, K. E. Jørgensen, ed., London, MacMillan, pp. 1-12. Jørgensen, K. E., ed. 1997b, Reflective Approaches to European Governance, London, MacMillan. Jupille, J., Caporaso, J. A. and Checkel, J. T., 2003, Integrating institutions. Theory, method, and the study of the European Union, Comparative Political Studies 36(1):7-41. Keohane, R. O., 1988, International institutions: Two approaches, International Studies Quarterly 32(4):379-96. Kohler-Koch, B., 1996a, The strength of weakness: The transformation of governance in the EU, in: The Future of the Nation State: Essays on Cultural Pluralism and Political Integration, S. Gustavsson and L. Lewin, eds, London, Routledge, pp. 169-210. Kohler-Koch, B., 1996b, Catching up with change: The transformation of governance in the European Union, Journal of European Public Policy 3(3):359-81. Kohler-Koch, B. and Eising, R., eds, 1999, The Transformation of Governance in the European Union, London, Routledge. Krasner, S. D., ed. 1983, International Regimes, Ithaca, Cornell University Press. March, J. G. and Olsen, J. P., 1998, The institutional dynamics of international political orders, International Organization 52:943-69. Marks, G. and Hooghe, L., 2004, Contrasting visions of multi-level governance, in: Multi-Level Governance, I. Bache and M. Flinders, eds, Oxford, Oxford University Press, pp. 15-30.

Sovereignty Reloaded

271

Marks, G., Hooghe, L. and Blank, K., 1995, European integration and the state, EUI working paper RSC no. 95/7, Badia Fiesolana, San Domenico (FI), European University Institute Florence, Robert Schuman Centre. Marks, G., Hooghe, L. and Blank, K., 1996, European integration from the 1980s: Statecentric versus multi-level governance, Journal of Common Market Studies 34(3):341-78. Neyer, J., 2003, Discourse and order in the EU: A deliberative approach to multi-level governance, Journal of Common Market Studies 41(4):687-706. Olsen, J. P., 2002, The many faces of Europeanization, Journal of Common Market Studies 40(5):921-52. Puchala, D. J., 1972, Of blind men, elephants and international integration, Journal of Common Market Studies 10(2):267-84. Radaelli, C. M., 2000, Whither Europeanization? Concept stretching and substantive change, European Integration online Papers 4(8), http://eiop.or.at/eiop/texte/2000-008.htm. Risse, T., 2004, Social constructivism and European integration, in: European Integration Theory, A. Wiener and T. Diez, eds, Oxford, Oxford University Press, pp. 159-75. Risse-Kappen, T., 1996, Exploring the nature of the beast; International Relations theory and Comparative Policy analysis Meet the European Union, Journal of Common Market Studies 34(1):53-80. Rosamond, B., 2000, Theories of European Integration, 2nd edition, Houndmills, Palgrave. Rosenau, J. N., 2004, Strong demand, huge supply, in: Multi-Level Governance, I. Bache and M. Flinders, eds, Oxford, Oxford University Press. Ruggie, J. G., 1993, Territoriality and beyond: Problematizing modernity in international relations, International Organization 47(1):139-74. Ruggie, J. G., 1998, What makes the world hang together? Neo-utilitarianism and the social constructivist challenge, International Organization 52(4):855-85. Searle, J. R., 1995, The Construction of Social Reality, New York, Free Press. Shaw, J. and Wiener, A., 2000, The paradox of the European polity’, in: The State of the European Union. Volume 5: Risks, Reform, Resistance and Revival, M. Green Cowles and M. Smith, eds, Oxford, Oxford University Press, pp. 64-88. Spruyt, H., 1994, The Sovereign State and its Competitors: an Analysis of Systems Change, Princeton, Princeton University Press. Wæver, O., 2004, Discursive approaches, in: European Integration Theory, A. Wiener and T. Diez, eds, Oxford, Oxford University Press, pp. 197-217. Walker, R. B. J., 1990, Sovereignty, identity, community. Reflections on the horizons of contemporary political practice, in: Contending Sovereignties. Rethinking Political Community, R. B. J. Walker and S. H. Mendlovitz, eds, Boulder, Lynne Rienner, pp. 159-85. Walker, R. B. J., 1991, State sovereignty and the articulation of political space/time, Millennium 20(3):445-61. Walker, R. B. J., 1993, Inside/Outside: International relations as political theory, Cambridge, Cambridge University Press. Wallace, W., 1999, The sharing of sovereignty: The European paradox, Political Studies 47(3):503-21. Wendt, A., 1987, The agent-structure problem in International Relations theory, International Organization 41(3):335-70. Wendt, A., 1994, Collective identity formation and the international state, American Political Science Review 88(2):384-96. Wendt, A., 1995, Constructing international politics, International Security 20(1):71-81. Wendt, A., 1999, A Social Theory of International Politics, Cambridge, Cambridge University Press. Wendt, A. and Friedheim, D., 1995, Hierarchy under anarchy: Informal empire and the East German state, International Organization 49(4):689-721. ’

272

Tanja E. Aalberts

Werner, W. G. and de Wilde, J. H., 2001, The endurance of sovereignty, European Journal of International Relations 7(3):283-313. Wiener, A. and Diez, T., eds, 2004, European Integration Theory, Oxford, Oxford University Press. Wind, M., 2001, Sovereignty and European Integration. Towards a Post-Hobbesian Order, Houndmills, Palgrave.

Library of Public Policy and Public Administration 1.

D. Rahm, J. Kirkland and B. Bozeman: University-Industry R&D Collaboration in the United States, the United Kingdom, and Japan. 2000 ISBN 0-7923-6073-7

2.

R. Torenvlied: Political Decisions and Agency Performance. 2000 ISBN 0-7923-6092-3

3.

O. Elgstr¨om: Images and Strategies for Autonomy. Explaining Swedish Security Policy Strategies in the 19th Century. 2000 ISBN 0-7923-6571-2

4.

O. van Heffen, W.J.M. Kickert and J.J.A. Thomassen (eds.): Governance in Modern Society. Effects, Change and Formation of Government Institutions. 2000 ISBN 0-7923-6653-0

5.

H. Wagenaar (ed.): Government Institutions: Effects, Changes and Normative ISBN 0-7923-6720-0 Foundations. 2000

6.

M.W. Bauer: A Creeping Transformation? The European Commission and the ISBN 1-4020-0006-5 Management of EU Structural Funds in Germany. 2001 Volumes 1–6 were published by Kluwer Academic Publishers.

7.

C. Agocs (ed.): Workplace Equality. International Perspectives on Legislation, Policy and Practice. 2002 ISBN 90-411-1843-8

8.

B. Denters, O. van Heffen, J. Huisman and P.-J. Klok (eds.): The Rise of Interactive Governance and Quasi-Markets. 2003 ISBN 1-4020-1742-1

9.

R. Holzhacker and M. Haverland (eds.): European Research Reloaded: Cooperation and Integration among Europeanized States. 2006 ISBN 1-4020-4429-1

www.springer.com