The State of the European Union: Building a European Polity? 9781685857943

Table of contents :
Contents
Tables and Figures
Preface
1 Introduction: Integration in Theoretical Perspective
PART 1 REFLECTIONS ON EUROPEAN INTEGRATION
2 The European Union and Regional Integration Theory
3 The Justiciability of Subsidiarity
4 Common, Collective, or Combined? Theories of Defense Integration in the European Union
5 Economic Uncertainty and European Solidarity Revisited: Trends in Public Support for European Integration
6 Political Group Cohesion in the European Parliament, 1989-1994
7 European Monetary Diplomacy and the Rolling Crisis of 1992-1993
8 Integration Theory and the Enlargement of the European Union
PART 2 EUROPEANIZATION OF NATIONAL POLITICS
9 Germany's Länder and the Federalization of the European Union
10 The Franco-German Relationship in the Post-Maastricht Era
11 Institutions and Leadership: Germany, Maastricht, and the ERM Crisis
12 The Importance of Being Independent: Central Bank Independence and the European System of Central Banks
13 National Interest and Convergence of Preferences: A Changing Role for Spain in the EU?
14 The European Union and the Nordic Countries: Impacts on the Integration Process
PART 3 EUROPEAN PUBLIC-POLICY MAKING: INTERNAL AND EXTERNAL DIMENSIONS
15 Promiscuous Policymaking: The European Policy Style?
16 Regional Actors in an Intergovernmental Play: The Making and Implementation of EC Structural Policy
17 EU Research Policy: The Politics of Expertise
18 The European Commission as Corporate Actor? European Telecommunications Policy After Maastricht
19 Integrating the Environment into the European Union: The History of the Controversial Carbon Tax
20 The European Community and the Conclusion of the Uruguay Round
21 The Lomé Convention: An Aging Dinosaur in the European Union's Foreign Policy Enterprise?
Acronyms
Contributors
Index
About the Book

Citation preview

THE STATE OF THE

EUROPEAN UNION

A biennial project of the

European Community Studies Association

THE STATE OF THE

EUROPEAN UNION Vol.3

Building a European Polity?

edited by

Carolyn Rhodes Sonia Mazey

L Y N N E R I F N ftFR FUBLISHIM

LONGMAN

Published in the United States of America in 1995 by Lynne Rienner Publishers, Inc. 1800 30th Street, Boulder, Colorado 80301 Published in the United Kingdom in 1995 by Longman Group Limited Longman House, Burnt Mill, Harlow Essex CM20 2JE England and associated companies throughout the world © 1995 by Lynne Rienner Publishers, Inc. All rights reserved

Library of Congress Cataloging-in-Publication Data The state of the European Union : building a European polity? / edited by Carolyn Rhodes, Sonia Mazey. Includes bibliographical references and index. ISBN 1-55587-605-6 (alk. paper) 1. European Union. 2. European Union countries—Politics and government. 3. European Union countries—Economic policy. I. Rhodes, Carolyn, 1953. II. Mazey, Sonia. JN30.S73 1995 341.24'2—dc20 95-9034 CIP British Cataloguing in Publication Data A Cataloguing in Publication record for this book is available from the British Library. ISBN 0-582-27813-9

Printed and bound in the United States of America

@

The paper used in this publication meets the requirements of the American National Standard for Permanence of Paper for Printed Library Materials Z39.48-1984. 5 4 3 2 1

Contents List of Tables and Figures Preface 1

vii ix

Introduction: Integration in Theoretical Perspective Carolyn Rhodes & Sonia Mazey

1

PART 1 REFLECTIONS ON EUROPEAN INTEGRATION 2 3 4

5

6

7

8

The European Union and Regional Integration Theory James A. Caporaso & John T. S. Keeler The "Justiciability" of Subsidiarity Daniel G. Parian Common, Collective, or Combined? Theories of Defense Integration in the European Union Patricia Chilton Economic Uncertainty and European Solidarity Revisited: Trends in Public Support for European Integration Christopher Anderson Political Group Cohesion in the European Parliament, 1989-1994 Joanne Bay Brzinski European Monetary Diplomacy and the Rolling Crisis of 1992-1993 David M. Andrews Integration Theory and the Enlargement of the European Union Lee Miles, John Redmond & René Schwok

PART 2 9

29 63

81

111

135

159

177

EUROPEANIZATION OF NATIONAL POLITICS

Germany's Länder and the Federalization of the European Union Richard E. Deeg

v

197

vi 10

11

12

13

14

Contents

The Franco-German Relationship in the Post-Maastricht Era Pia Christina Wood Institutions and Leadership: Germany, Maastricht, and the ERM Crisis Michael E. Smith & Wayne Sandholtz The Importance of Being Independent: Central Bank Independence and the European System of Central Banks Hugo M. Kaufmann National Interest and Convergence of Preferences: A Changing Role for Spain in the EU? Carlos Closa The European Union and the Nordic Countries: Impacts on the Integration Process Lee Miles

221

245

267

293

317

PART 3 EUROPEAN PUBLIC-POLICY MAKING: INTERNAL AND EXTERNAL DIMENSIONS 15 16

17 18

19

20

21

Promiscuous Policymaking: The European Policy Style? Sonia Maze y & Jeremy Richardson Regional Actors in an Intergovernmental Play: The Making and Implementation of EC Structural Policy Mark A. Pollack EU Research Policy: The Politics of Expertise John Peterson The European Commission as Corporate Actor? European Telecommunications Policy After Maastricht Gerhard Fuchs Integrating the Environment into the European Union: The History of the Controversial Carbon Tax Anthony R. Zito The European Community and the Conclusion of the Uruguay Round Youri Devuyst The Lomé Convention: An Aging Dinosaur in the European Union's Foreign Policy Enterprise? Olufemi A. Babarinde

Acronyms List of Contributors Index About the Book

337

361 391

413

431

449

469 497 501 505 520

Tables and Figures

Tables 5.1 5.2 5.3 5.4 6.1 6.2

Support for European Integration: 1973-1992 Support for Membership in the European Community: 1973-1992 Pearson Correlation Coefficients of Support for European Integration and GDP Growth: 1973-1992 Support for the Maastricht Treaty: 1992-1993

118 121 125 129

Political Groups in the European Parliament: 1989-1994 Group Cohesion in the European Parliament: Average Index of Agreement Correlations Between Index of Agreement Scores and National Party Influence, Ideology, and Size Correlations Between Size Variables and the Index of Agreement and Group Unanimity Voting Participation by Nationality and Political Group: Average Percentage of MEPs Absent and Share of Eligible Votes Not Cast Because of Absences

141

13.1

Economic Profile

299

13.2

EC Expenditure in Spain

301

14.1

Statistical Profile of the Applicant States

328

16.1 16.2

Major Redistributive Bargains Regional Fund Budget

364 371

17.1 17.2

EC/EU Spending on R&D: 1987-1993 Defining Characteristics of Different Policy Fields

393 406

6.3 6.4 6.5

vii

145 148 149

152

viii

Tables and Figures

Figures 2.1 2.2 5.1 5.2

European Community Articles in the New York Times Integration Scholarship and European Community News Public Support for European Integration Linear Trends in Public Support for European Integration

38 38 120 123

Preface

This volume is the third in a biennial series of "yearbooks" on the state of the European Community and covers the years 1993 and 1994, during which the European Union was created and enlarged to include fifteen member states. In particular it considers integration and enlargement in the post-Maastricht era in a critical and reflective context. It examines a range of analytical and substantive issues that emerged during this period, evidencing a concerted effort to place recent events regarding European integration in theoretical perspective. This is the first volume in the series to confront the institutional changes created by the Treaty on European Union (TEU), as well as the somewhat complicated terminology it has introduced. Having consulted officials in the Commission (as well as Chapter 3 of the TEU), we have adopted the following guidelines for references to the European Community and European Union: All references to the European Economic Community as per Article G of the Treaty on European Union will now refer to the European Community. Thus, the European Community has become one of the three European Communities (along with the Coal and Steel Community and Euratom). Issues involving the single market, competition policy, and external trade relations still fall under the European Community. Also, we have chosen to adopt the guideline that all references prior to ratification of the TEU in 1993 to the collective community of member states will utilize "European Community." References to the European Union are appropriate when referring to the collective union of member states after ratification of the TEU in 1993. This is the case when authors refer to the European Union generally. In addition, references to the Common Foreign and Security Policy and to cooperation in the fields of justice and internal affairs will utilize European Union. The Commission remains the Commission of the European Communities while the Council is now the Council of the European Union. Thus, readers will encounter different usages regarding the EC and EU depending on the context under discussion. ix

X

Preface

This biennial series is one of several projects sponsored by the European Community Studies Association ( E C S A ) . E C S A exists to develop a community of scholars, teachers, and policy practitioners interested in the E U and to raise the level of knowledge regarding European integration. Although there are a number of associations and organizations in the United States that focus on Europe or on individual European countries or groups of countries, E C S A is the only organization devoted solely to the EU. It welcomes members from all academic disciplines, as well as the business and public sectors. E C S A ' s objectives are to promote study and awareness of the E U by facilitating research projects and other collaborative works among the association's members in order to increase the published literature on the EU; encourage links between the association and European counterpart associations and individuals; publish an informational newsletter; hold conferences and workshops; organize panels on the E U at conferences and workshops sponsored by other organizations; seek and disburse funds to support the association's activities; draw on membership to speak publicly about E U issues; and undertake other programs, activities, and projects deemed appropriate. The association is committed to promoting links with interested individuals and organizations in other parts of the world. In coediting this volume, we have been fortunate to receive assistance and support from many individuals and institutions. We are particularly grateful to the executive committee of E C S A and to William Burros, E C S A ' s administrative director. The following institutions generously provided financial and/or administrative support: Utah State University Milton R. Merrill Endowment, Department of Political Science and Women and Gender Studies Institute; Churchill College, University of Cambridge; and The Ford Foundation. We would also like to thank Stephanie White, Kimberly Riggs, and especially Denise Brooks of Utah State University for helping to prepare the final manuscript. Carolyn Rhodes Sonia Mazey

1 Introduction: Integration in Theoretical Perspective Carolyn Rhodes & Sonia Mazey

In preparing this third biennial volume we have been guided by two goals: to update readers on important developments and ongoing issues in European integration and to provide a theoretical context within which to place these events. Focusing on the question of whether a European polity is emerging after nearly fifty years of integrative cooperation, we have collected a very interesting set of studies that will help readers make their own judgments. Not surprisingly, depending upon the issue area or institutions involved, conclusions are mixed. Many view the European Union as an ongoing struggle of give and take between member states, where no real European polity has emerged. Others, however, see real movement in the direction of a polity, as individuals, corporations, and government actors increasingly identify with, and act according to, European-level institutions and processes. We have urged our contributors to be as analytical as possible in treating their topics and to utilize theoretical perspectives whenever such frameworks help to clarify and explain the particular issue under study. A concerted effort has been made to cover a range of topic areas and issues; however, because contributions were selected for their quality and analytical interest rather than merely for their topicality, coverage is by no means universal. In writing this introduction, we have been influenced equally by a desire to alert readers to important recent events and a desire to emphasize the value of different theoretical and methodological approaches to the study of EU affairs. Our introductory chapter is divided into four main sections. First, we offer a brief review of the events of 1993-1994 that we consider most important to the question of integration, focusing primarily on the implications of the new Treaty on European Union (or Maastricht Treaty). Second, we explain the organization of this volume. Third, we discuss the major analytical approaches that guide EU studies, differentiating

1

2

The State of the EU

the analytical approaches that treat the E U as an international organization f r o m those that treat the E U as a state-like entity. T h i s section c o n s u m e s the bulk o f our introduction, as we link the work o f our contributors to the theoretical and analytical perspectives identified, suggesting to students o f E U integration that a wealth o f analytical perspectives exists to guide research in m e a n i n g f u l and productive w a y s . A l s o , we argue that e x i s t i n g approaches need to b e m o r e precisely applied and differentiated, and that all o f us who work in this e x c i t i n g area o f research should c h a l l e n g e and refine our concepts to hone our understanding o f the decisions and p o l i c i e s involved in the integration process. Finally, we c o n c l u d e our discussion by h i g h l i g h t i n g the a n a l y t i c a l t h e m e s that e m e r g e f r o m the twenty chapters that follow, reinforcing the view that there is a very useful partnership b e tween theory and c a s e study analysis.

Recent Events and EU Integration T h e years 1 9 9 3 and 1 9 9 4 were difficult, yet interesting, f o r the European integration p r o c e s s . T h e T r e a t y on E u r o p e a n U n i o n , w h i c h c a m e into e f f e c t in N o v e m b e r 1 9 9 3 , pledged m e m b e r states to an ambitious new set o f g o a l s : a c o m m o n foreign and security policy, monetary union, and s o c i a l and environmental policy cooperation. T h e Maastricht Treaty f o l l o w e d the e a r l i e r l o g i c o f the 1 9 8 6 S i n g l e E u r o p e a n A c t ( S E A ) . H o w e v e r , whereas the latter had b e e n primarily c o n c e r n e d with " n e g a t i v e i n t e g r a t i o n " (the r e m o v a l o f barriers to the free m o v e m e n t o f g o o d s , s e r v i c e s , capital, and l a b o r ) , the M a a s t r i c h t T r e a t y sought to strengthen " p o s i t i v e " integration measures (the coordination o f social and environmental politics and m o n etary p o l i c y ) as well. T h i s marked an important d e v e l o p m e n t in the integration p r o c e s s , not least b e c a u s e it implied a stronger c o m m i t m e n t to building a E u r o p e a n polity. Not surprisingly, t h e r e f o r e , the M a a s t r i c h t Treaty prompted widespread debate. To counter concerns among some m e m b e r governments (and some subnational governments) that such a project would result in e x c e s s i v e supranationalism, the Maastricht Treaty also introduced the subsidiarity principle. In a c c o r d a n c e with this principle, responsibility for public p o l i c y m a k i n g should rest with the lowest (or as Daniel Partan explains in Chapter 3, the most efficient) governmental level possible. In short, the E U should be permitted to assume responsibility only for those policies that cannot satisfactorily be carried out by national (or subnational) governments. N o t w i t h s t a n d i n g this safeguard f o r those worried about loss o f national sovereignty, ratification o f the M a a s t r i c h t T r e a t y was not a s i m p l e matter; public d e b a t e o v e r the p r o v i s i o n s and i m p l i c a t i o n s o f the treaty during the ratification process was heated, and ratification itself was only n a r r o w l y a c h i e v e d . In F r a n c e , the l o n g - s t a n d i n g political c o n s e n s u s in

Introduction

3

favor of European integration was seriously challenged for the first time, and the r e f e r e n d u m held on the Maastricht Treaty produced only a tiny majority in favor of ratification. In Denmark, ratification was achieved only after the introduction of special provisions for the Danes. In the United Kingdom, parliamentary opposition to the treaty forced the prime minister to make ratification the subject of a parliamentary vote of confidence, even though Britain had arranged to "opt out" of key treaty provisions relating to social policy and monetary union. Finally, the accession debates in the Nordic countries revolved around deep-seated concerns over the loss of sovereignty, ultimately resulting in N o r w a y ' s rejection of EU membership. Several recent developments within and beyond the European Union help to explain this ambivalence toward further European integration, which ratification of the Maastricht Treaty has done little to dispel. Within the Union, high u n e m p l o y m e n t levels and economic recession have provided fertile conditions for the rise of populist, extreme right-wing parties such as the French Front Nationale and the German Republikaner party, which have sought to link European integration with increased immigration, high unemployment, and a loss of national identity. In general, public opinion regarding deeper integration (as Christopher Anderson notes in Chapter 5) has always been affected to some degree by economic circumstances, and recent difficulties have shaken public confidence. Political opposition to Euro-federalism and ideological objections to the "interventionist" economic and industrial policies of the Commission have prompted widespread public uneasiness about the implications of the Maastricht agreement. Meanwhile, the 1992-1993 European monetary crisis, tensions in the traditional Franco-German alliance, and the European ramifications of German reunification have tempered enthusiasm for the Union. The difficulties experienced by the Union in reaching agreement—both internally and with the United States—in the Uruguay Round of GATT negotiations, demands f r o m former Soviet bloc countries for closer ties with the Union, and the uncertainty that surrounds the future development of a European defense policy (especially in the midst of the Yugoslavian disaster) have called further into question the unity and purpose of the Union. T h e s e issues were evident in the Nordic accession debates where politicians and the public alike expressed ambivalence over the benefits of membership. Even as the twelve member states embrace Austria, Sweden, and Finland and contemplate further enlargement eastward, there is already in the U n i o n ' s infancy a critical reassessment of the state of European integration. Yet, despite these specific misgivings, there continues to be considerable public appreciation for the general value and increasing importance of collective European arrangements. The degree of sovereign prerogative that member states and their electorates are willing to forfeit to European level institutions, and the ability

4

The State of the EU

of European level actors to set the policy agenda across a number of issue areas, have become measures of the Union's integrative achievement. Thus, as our contributors assess the events of 1993 and 1994 and explicate the ways in which decisions are made in the European Union, they also tell us a great deal about the evolving character and limits of European integration and whether a European polity is emerging from the architecture of past cooperation.

Organization of This Volume There is a serious effort underway among scholars to place recent events and their relationship to economic and political integration into a broader theoretical and historical perspective. This in turn has been influenced by the ongoing debate among policymakers both within member states and at the E U level about the appropriate direction of future European integration. This reflective and somewhat critical mood is evident in this volume as contributors review E U developments and debates and analyze them within the wider historical and theoretical contexts of European integration. Reflections on European

Integration

Part 1, "Reflections on European Integration," is specifically devoted to assessing European integration, both in theoretical terms and in the playing out of integrative efforts. Thus, Chapter 2 by James Caporaso and John Keeler takes readers through European Union and regional integration theory as it has evolved since the 1950s, establishing the historical backdrop for much of the theoretical analysis that follows. The next six chapters examine key dimensions of European integration in the context o f recent developments. Daniel Partan's legal analysis of the subsidiarity principle in Chapter 3 helps predict how this pivotal concept will influence future integrative directions. As one of the most controversial aspects of the Treaty on European Union, it is a subject of considerable interest. In Chapter 4 Patricia Chilton assesses the theories of defense integration, applying them specifically to recent cooperative defense arrangements as well as to the Common Foreign and Security Policy component of the T E U . Christopher Anderson's quantitative analysis in Chapter 5 of the correlation between economic uncertainty and public support for European integration provides readers with some perspective on the ups and downs in public support. Joanne B a y Brzinski's more specific study in Chapter 6 of political groups in the European Parliament is predicated on the crucial theoretical question of whether a "European" identity is emerging in place of national allegiances, offering a very interesting analysis of polity development within this democratic European-level institution.

5

Introduction

David Andrews's analysis of the European monetary crisis of 1 9 9 2 1993 has been included in Part 1 because this subject lies at the heart of recent reassessments of the goals of the Single European Act and the T E U . It demonstrates that, despite the obvious rift between France and Germany and the collapse of the exchange rate mechanism, subsequent events evidence an ongoing commitment to the goals of economic and monetary union. Finally, Part 1 concludes with an analysis of integration theory and the enlargements (including the most recent enlargement) of the European Union, written by L e e Miles, John Redmond, and René Schwok. This final reflection on European integration clarifies how acceding member states influence the character and direction of European integration, and it offers new insights about the relative merits of functionalist and intergovernmental explanations in this context. Europeanization

of National

Politics

Part 2, which is titled the "Europeanization of National Politics," examines the ways in which governing principles, institutions, and rules are being shaped and implemented at the European level. Focusing specifically on how member states (or subnational groups) are influencing and adapting (or reacting) to European Union initiatives and institutional structures, these chapters directly assess whether, under what circumstances, and in what form a European polity is being built. In some instances domestic institutions and actors have proven crucial in shaping the particular direction of E U integration. A striking example is Richard Deeg's Chapter 9, which demonstrates that the future character of the E U polity will likely evidence the influence o f the highly active German Länder. Not surprisingly, the general consensus among contributors is that member states recently have been resisting "Europeanization" when it directly challenges domestic policies. Pia Christina Wood's assessment of the Franco-German relationship in the post-Maastricht period reveals the strains in this pivotal partnership, reaffirming long-standing national positions on European issues as well as demonstrating the emergence of new cooperative difficulties. In Chapter 11 Michael Smith and Wayne Sandholtz examine how powerful domestic institutions influence the integrative effort at the European level, citing as their example the German Bundesbank's behavior in influencing German positions at Maastricht and in contributing to the E R M crisis of 1 9 9 2 - 1 9 9 3 . Carlos Closa offers, in Chapter 13, a reassessment of Spain's position toward European integrative efforts and its effort to influence European-level arrangements in recent years, arguing similarly that shifting domestic circumstances, as well as changing Europeanlevel opportunities, are altering Spain's relationship to the EU. At the same time, however, it is clear that in certain realms Europeanlevel principles and rules of behavior are being shaped that will influence the future character of the European polity. Institutions such as a European

6

The State of the EU

independent central bank (as explained by Hugo Kaufmann in Chapter 12) will be crucial in affecting member state financial policy, if and when it is fully created. Finally, this section ends with a chapter on the 1994 enlargement, which helps reveal broader questions about the level of "Europeanization" expected, desired, or feared by new member countries. Lee Miles in Chapter 14 examines Nordic accession arrangements with the EU and the subsequent ratification debates, identifying the issues that prompted both membership interest in the first place and later shaped the elite and public debates that followed. His use of domestic and international factors offers a range of insights about how much "Europeanization" these countries really want, implying too that this and prospective enlargements could have a significant effect on the Europeanization of national politics in the future. European Public-Policy

Making: Internal and External

Dimensions

Part 3 of this volume deals directly with the process and outcomes of public-policy making at the European level, both domestically within the EU and in terms of its foreign policy. Launching this section is a chapter by Sonia Mazey and Jeremy Richardson, which characterizes public-policy making in the EU as a multilayered process where European-level institutions are quite open to influence, thus somewhat "promiscuous." This general chapter on European policy style lays the foundation for the subsequent case studies that examine specific areas of European public policy. Leading the case studies is a strong analytical chapter by Mark Pollack, which examines the ways in which EC structural policy gets made, how budgets are negotiated and implemented, and how member states wield their power in this area. Following this examination of structural policy is an excellent analysis of EU research policy by John Peterson, who places most of the credit for policy formulation and effectiveness with the experts, unlike Pollack who emphasizes intergovernmental bargains. Chapter 18 offers an innovative analysis of telecommunications policy, where Gerhard Fuchs characterizes the Commission as a corporate actor that can influence policy directions. Anthony Zito's analysis of the EU's carbon tax effort is embedded in a strong theoretical framework that illustrates the processes involved in shaping EU environmental policy. Two chapters deal with the external dimension of EU public policy, both focusing on trade relations. The first, Chapter 20 by Youri Devuyst, examines the multilevel negotiations that were necessary to achieve EC agreement in the GATT Uruguay Round. Taking into account both Commission and Council involvement, this study focuses on the successes and failures of the bargains struck between member states and between the EC and the United States to achieve agreement. Finally, in Chapter 21 Olufemi Babarinde offers a comprehensive assessment and explanation of the status

7

Introduction

of the Lomé Convention in the priorities of the EU. Identifying the domestic and international factors that have undermined the EU's commitment to Lomé, he presents a grim prediction for its future. Thus, the policy case studies offer a mixture of conclusions about the future directions of internal and external public policy. In some cases national actors remain highly influential. In other cases, technocrats and private interest lobbyists are instrumental in shaping policy. However, all of these chapters contain useful analytical explanations about the making and implementing of EU policy. They refine our understanding of the integration process, identify where power and influence reside, and aid us in deriving new paradigms and insights from their rich variety of approaches.

Approaching Integration Analysis: What International Relations and Comparative Politics Have to Offer Guided by the analytical themes that emerged from our contributors as well as by existing theoretical paradigms, our purpose in this section is to introduce and demonstrate the relevance of a wealth of analytical perspectives from international relations theory, comparative politics, and public policy analysis to the study of European integration. Broadly speaking, academic interest in the European Union has focused upon two related sets of questions: explaining and assessing the process of European integration; and evaluating the policymaking capacity, structures, and processes of the EU. Whereas those researchers interested primarily in the former tend to view the European Union as an international organization, those interested primarily in the EU policy process tend increasingly to view the Union as a system of governance (albeit a peculiar and incomplete one). At the risk of oversimplification, we suggest that these two approaches to European Community studies are usually associated with integration theory and comparative politics/public policy making theory respectively. This discussion is illustrated by references to the ways in which contributors to this book utilize these various theoretical approaches to analyze their respective subjects. In offering this blend of theory and substance we hope to clarify how scholars examine events and to offer our readers an analytical compass by which to guide their exploration of recent developments in European integration. Integration Theory and International Relations of European Union: Intergovernmentalism and

Explanations Neofunctionalism

Analysis of European integration has been dominated by two general interpretive approaches: intergovernmentalism and neofunctionalism. The intergovernmentalist perspective assumes that the member states of the

8

The State of the EU

European U n i o n remain the key actors in determining o u t c o m e s . R e g a r d l e s s o f h o w m u c h institutionalization o c c u r s at the E u r o p e a n l e v e l , this p e r s p e c t i v e m a i n t a i n s that it can only e x i s t at the b e h e s t o f the m e m b e r states, and crucial decisions remain in their hands. T h u s , political bargains struck b e t w e e n m e m b e r states determine the character o f European c o o p eration and the scope and limits o f European institutions. 1 Accordingly, analysts seek to explain European policy, d e c i s i o n m a k i n g , and m e m b e r state c o n f l i c t or cooperation by f o c u s i n g their attention on m e m b e r state capabilities, state-to-state negotiations, and resulting bargains, downplaying or ignoring alternative explanations. A n important limitation o f this model o f E U d e c i s i o n m a k i n g is its tendency to assume the e x i s t e n c e o f a single national interest. R e c e n t l y , Andrew M o r a v c s i k has attempted to modify this assumption by i d e n t i f y i n g what he c a l l s " l i b e r a l i n t e r g o v e r n m e n t a l i s m , " w h i c h takes into a c c o u n t the pressures o f d o m e s t i c i n t e r e s t s . 2 I f o n e a c cepts this m o d i f i c a t i o n , intergovernmentalist analysis has b e e n well suited to characterize a significant portion o f the political and e c o n o m i c arrangem e n t s a m o n g m e m b e r states and to d e s c r i b e to a large d e g r e e how integration has occurred. T h e f u n c t i o n a l i s t tradition a s s u m e s that c o o p e r a t i v e o u t c o m e s are products o f private and public interest forming incremental, transnational linkages, which spill from issue to issue, causing cooperation to breed c o operation and integration to spur integration, with supranational institutional development a logical c o n s e q u e n c e . L i n k a g e s between political and e c o n o m i c e l i t e s f o r g e further institutional d e v e l o p m e n t , the p r o c e s s e x panding progressively as elites r e c o g n i z e the benefits o f institutionalizing the linkages. H e r e the f o c u s is not on the m e m b e r states as the key actors but on political and e c o n o m i c elites and interest groups and the Europeanlevel institutions they create (and within which they participate). N e o f u n c t i o n a l i s t s m o d i f i e d this view by r e c o g n i z i n g that constraints on integration e x i s t — t h a t integration is not necessarily a progressive, linear process. T h e ups and downs, they state, are due in part to systemic f a c tors that can shape i f not i m p e d e the integrative p r o c e s s . W h e r e a s f u n c tionalists a s s u m e d the e x i s t e n c e o f a p o l i t i c a l c o n s e n s u s in f a v o r

of

transnational c o o p e r a t i o n , n e o f u n c t i o n a l i s t s a c k n o w l e d g e the e x i s t e n c e and variable impact o f political c o n f l i c t and bargaining between European policymakers. T h u s n e o f u n c t i o n a l i s t s a c k n o w l e d g e that intergovernmental relations do have an i m p a c t , admitting that " p e r i o d s o f i n c r e m e n t a l a d v a n c e were l i k e l y to be p r e c e d e d by a political b a r g a i n . " 3 E v e n so, the f u n c t i o n a l i s t neofunctionalist perspective remains f o c u s e d on the linkages between poli c y m a k i n g elites across boundaries and tends to emphasize the importance o f E u r o p e a n - l e v e l institutions f o r further e n c o u r a g i n g t h o s e

linkages

across issue areas. A s such, neofunctionalism falls within the pluralist tradition, f o c u s i n g upon the activities o f societal groups rather than intergovernmental bargaining.

Introduction

9

Generally, then, international relations scholars tend to either explain phenomena as resulting from intergovernmental bargaining between member states or as resulting from the influence of European-level institutions and transnational interests expanding incrementally across issues as political and economic elites become accustomed to their benefits and demand further integration. While the past two decades have produced scholarship suggesting that other, more normative, and institutionally oriented theories might also inform our understanding of European integration, these two general approaches—intergovernmentalism and functionalism—guide most of the analyses that have emerged. Analysts tend to adopt the functionalist perspective when integration and institutionalization are proceeding without significant member state interference and to utilize intergovernmentalism either when major new debates or initiatives emerge or when acrimony among member states is obstructive to the integration effort. This pattern is evident in the historical review of integration theory presented by Caporaso and Keeler in Chapter 2, and it probably reflects valid analytical shifts, depending upon how integration is proceeding. Not surprisingly, most of the contributors to this volume follow this same pattern. However, in addition, several authors offer new observations regarding integration, modifying and refining our application of intergovernmentalism and functionalism, as well as offering new propositions about cooperation, conflict, and institution building and maintenance in the European Union. Interestingly, the majority of the authors in this volume utilize intergovernmentalist interpretations. It is clear from these studies, however, that merely categorizing all events involving member state bargaining as intergovernmentalism and assuming that this informs us about cause and effect is inadequate. While this may focus our attention at the point where the crucial activity is occurring, it offers little in the way of explanation. The authors utilizing the intergovernmentalist perspective demonstrate that it is also important to discern not only that international bargaining created the situation at hand but why and how particular bargains are struck. What factors made this bargain possible? Under what conditions might this situation be repeated? What does this tell us about future member state agreements and disputes? How does this affect our view of the integration process? A number of discrete approaches within intergovernmentalism can differentiate and characterize outcomes of member state relationships and interactions. Differentiating Intergovernmentalism: Neorealism, Bargaining Theory, Domestic Sources of Foreign Policy, and Game Theory Neorealism. The dominant theory underlying intergovernmentalism is neorealism. Yet the theoretical classification of neorealism is often carelessly applied in European integration studies, with no clear distinction

10

The State of the EU

made among realism, neorealism, bargaining theory, domestic sources of foreign policy, and game theory. All of these perspectives assume that interactions among rational nation-states are what produce significant outcomes in international affairs (including cooperative arrangements like the Europe Union), but each has its own characteristics. Realism, it is generally agreed, was never a systematic theory of behavior and outcomes in international affairs, but rather a descriptive set of assumptions about the character of nation-states in an anarchical world system. 4 Neorealism is considered the theoretical heir to realism because it attempts to systematize the assumptions of realism into a theory about national behavior in international affairs. It has been the dominant paradigm in international relations for thirty years. 5 Neorealism assumes that the only significant actors in international affairs are nation-states, discounting and discrediting all transnational organizations and linkages as pawns or appendages of the individual nation-states. Neorealists assume that nationstates are rational, self-interested actors in constant pursuit of power and influence. They view all forms of cooperation as temporary and tenuous— linked entirely to whether they serve the self-interest of the nation-states involved. This self-interest is defined in terms of survival and power, with military power the orthodox measure of relative capability. This theoretical perspective helps us to understand the international environment within which EU member states function, and security analysts consider it vital to providing insights into the constraints and opportunities facing EU member states in pursuing a common foreign and security policy. Patricia Chilton's examination of security and military integration in Chapter 4 illustrates this point. Despite some integrative successes in tactical military operations, security integration overall has remained elusive. Functionalist explanations fall short of explaining the difficulties encountered by member states in integrating security policy. Neorealism, however, which accounts for the presence of the United States military in the European structure of power and notes its asymmetrical capabilities within NATO, offers a more convincing explanation. Another realm where neorealism offers insights into the European integrative process involves the external security relations of member states. Analysts have long argued that the Cold War fear of Soviet aggression created the impetus for Western European integration and helped maintain it until 1990. 6 Especially if one considers the French view that Western Europe could trust neither the Soviet Union nor the United States, this supplement to the Atlantic Alliance made sense in realist terms, even though neorealism has not been well equipped to account for economic cooperation. However, the neorealist view also suggests that with the end of the Cold War the cement that held the EC together as a defensive alliance (as well as the Community with the Atlantic Alliance) disappeared, and this may account for some of the current cooperative difficulties within the Union.

Introduction

11

The disappearance of the geopolitical divisions in Europe that previously cemented the European Community opened opportunities for such countries as Finland, Poland, Hungary, and others to become members. This is noted by Lee Miles in Chapter 14 on the Nordic accessions. Neorealism directs our attention toward the structure of the international system and how it affects member state (and prospective member state) calculations of interests. This systemic structural theory does explain the international context at a given point in time, even if it is inadequate for explaining how changes in context occur or how subnational factors and linkages between nations may affect member interests. 7 Most commonly, neorealist assumptions are utilized by intergovermentalists to describe and characterize in general terms how bargains between member states are made and how those bargains produce new cooperative arrangements and institutions. The stress in neorealism on relative national power capabilities and national interests is particularly useful in this regard. Thus, intergovernmenalists adopt realist assumptions when explaining interactions between member states whether military power is at issue or not. This approach is especially relevant when significant new initiatives are proposed. Thus, as Smith and Sandholtz explain in Chapter 11, for example, the position taken by the economically powerful German government during the Maastricht negotiations regarding economic and monetary union was instrumental to the kind of agreement reached. Similarly, Carlos Closa's explanation of Spain's national interests and bargaining efforts, Mark Pollack's analysis of structural policy, and Pia C. Wood's review of the Franco-German relationship highlight the importance of focusing on member state interests and capabilities in order to understand policy outcomes at the European level. As the dominant paradigm in international relations theory, neorealism has undergone a great deal of criticism and modification over the past thirty years. Much of that criticism centers on its simple assumptions about cause and effect and the role of nation-states as unitary actors. Critics argue that the world is a complicated place, where definitions and loci of power are not straightforward and where interests often transcend national boundaries and identities. Correspondingly, other international environmental structures besides military power may be important in creating opportunities and constraints for individual nation-state choices. The structure of economic relationships—trade flows and financial arrangements— may be more deterministic in a world so characterized by economic interdependence. The existence and success of the European Community offers a case in point. Therefore, there are those who accept neorealism's general premise that nation-states behave in self-interested ways but who also recognize that the structure and character of domestic political and economic institutions and interests affect the position of the national government in

12

The State of the EU

international affairs. 8 Again, Closa, Pollack, Wood, and Smith and Sandholtz utilize this modified definition of neorealism, as they take into account the domestic forces that influence national interests and bargaining positions. Acknowledging the validity of the utilitarian assumptions of neorealism, both interdependence theory and bargaining theory claim that multiple sources of influence exist within and between nations. Both analytical approaches, though their respective focus is somewhat different, attempt to identify under what circumstances some forms of influence are more effective than others and how and why institutional structures and transnational relationships affect the bargaining process. This is particularly useful in examining member state relationships and negotiations within the European Union. Interdependence theory. Interdependence theory examines how the complex communication, commercial, financial, environmental, and military arrangements among nations (and their domestic interests) create vulnerabilities and sensitivities to each other's influence. 9 Interdependence theory sees the linkages between domestic interests across national boundaries as significant factors in determining and/or limiting national foreign policy choices. It also offers explanations of why bargaining leverage may result from factors other than military might and why a more discerning view of international bargaining may be necessary. Within the European Union, where an array of complex linkages exists across member states, interdependence theory should be particularly relevant and deserves broader application. However, most analysts do not make formal use of this theory; instead they more loosely apply the concept of intergovernmentalist bargaining without differentiating sources of influence across issues. Bargaining theory. Like interdependence theory, bargaining theory focuses on how nations influence each other, but it looks more specifically at how disputes are resolved and how the actors, interests, and strategies affect outcomes. 1 0 Some bargaining theory focuses on how the international interaction itself affects outcomes, how strategies as well as interests produce certain results, and how the bargaining context affects choices and the kinds of leverage available to nation-states. Recent work in this area shows that the particular institution that provides the milieu for bargaining may be crucial to its results. 11 The institutional arrangements in the Atlantic defense community provide a case in point. Patricia Chilton's comparison of defense and security cooperation within the EU notes the significance of institutional framework for explaining differences between the NATO and W E U arrangements. Also, in his explanation of telecommunications policy, Gerhard Fuchs's characterization of Commission influence depends to a large degree on the sector's institutional setting. Finally,

Introduction

13

the future character of European multilevel governance, the subsidiarity principle, the interpretive role of the Court of Justice in this area, and the impact of federal-type arrangements point to a new institutional setting that may affect bargaining between member states and between subnational actors and member states. This is foreshadowed by Daniel Partan's analysis of the subsidiarity issue in Chapter 3, as well as more directly addressed by Richard Deeg in his examination of the German Länder and European Union federalization in Chapter 9. Clearly, as these authors argue, we must look carefully at the institutional setting if we are to better understand member state bargaining outcomes. Game theory. As a complement to bargaining theory, game theory has been useful in providing descriptions of such interactions. This has helped reduce complex sets of encounters to simple abstractions of real bargaining situations and from these put forward some general propositions about how outcomes are determined. 12 Except for the application of Robert Putnam's notion of two-level games discussed below, game theory is not utilized by the authors in this volume. However, we note it here, since there are a number of European-level bargaining situations that lend themselves to such formal, mathematical description. Comparative foreign policy analysis. Similarly, comparative foreign policy analysis has recognized for years that, while the structure of the international system may present individual nations with constraints and opportunities in the pursuit of their interests, it is often the character and pattern of domestic institutional arrangements that determine national foreign policy choices. 1 3 Some analysts, such as Robert Putnam, explain that national governments are involved in two sets of bargaining interactions— one at the domestic level and one at the international level. 1 4 Informed by Putnam's characterization of international bargaining as a two-level game, Youri Devuyst's analysis of the Uruguay Round negotiations in Chapter 20 builds on this framework and explains outcomes as resulting from multilevel games involving domestic national interests, national governments, and European-level actors. Though less formalized, a number of the other chapters implicitly characterize member state behavior as resulting from national governments being caught between European-level pressures on the one hand and domestic pressures on the other. A nation-state's failure to cooperate or strong initiatives to shape key policies may be traced to the strength and vociferousness of domestic interests and the institutional structures that protect them. Richard Deeg's analysis of German federalism and European Union, Carlos Closa's analysis of Spanish interests in the European Union, Lee Miles's examination of Nordic accession, Mark Pollack's insights regarding regional policy, Wayne Sandholtz and Michael Smith's institutional

14

The State of the EU

analysis of German monetary policy, Pia Wood's reexamination of the Franco-German relationship, Hugo Kaufmann's analysis of the importance of an independent European central bank, and Olufemi Babarinde's assessment of the Lomé Convention demonstrate why it is important to take into account both international and domestic forces. From a more theoretical perspective, these case studies may also identify the domestic factors and international circumstances that influence member state behavior across issue areas. Regime theory. Regime theory developed at the fulcrum point between intergovernmentalism and functionalism and draws from many of the international relations approaches already discussed. While, like neorealism, it assumes that nation-states are rational actors that make deals with one another for mutual benefit, it goes further in considering the internal actors of nation-states and the decisionmaking processes involved. It assumes that nations engage in patterns of cooperation and purposefully create sets of rules, norms, and procedures around which their expectations converge. 1 5 For many regime theorists, national self-interest may be the result of compromises between competing domestic interests. Cross-national linkages may provide nations with strong incentives to cooperate and to regularize their interactions. This view has been adopted by many analysts of European integration. That is, the demand for further integration is fueled by influential national interests that benefit from transnational arrangements. 1 6 Regime theory is thus more pluralistic in its view of the nation-state than is neorealism and has been commonly associated with liberal interpretations of international cooperative arrangement. Furthermore, regime theory assumes that, as expectations converge around certain regimes, national behavior is in part shaped by those expectations. In this regard, regime theory assumes that nations choose to cooperate over time on a regular basis depending upon the issues and interests involved, but that once this cooperation is institutionalized, the regime itself has influence over national behavior and may at times actually govern it. In this regard, regime theory helps us to understand the process toward European integration, and, in turn, integration theory has been important in influencing regime theory. 17 Probably the best implicit example of the use of regime theory in this volume is the reflective analysis of enlargement in Chapter 8 by Lee Miles, John Redmond, and René Schwok. Because it examines motives for membership as well as impacts of membership and the effects of enlargement, this analysis fits well within the regime theory tradition. Also, one could categorize Patricia Chilton's analysis of European security or Richard Deeg's analysis of the federalization of Europe Union within this perspective, because both chapters deal with "regime creation," the effort of member states to create cooperative arrangements and institutions that will influence future interactions.

Introduction

15

However, once we analytically recognize that regimes are intervening variables between international interactions and actual outcomes, we may be moving away from the intergovernmentalist category of analysis toward neofunctionalism. In neofunctionalist theory the character of elite linkages and institutional development becomes the crucial factor in explaining the integrative process. As noted above, this neofunctionalist perspective assumes that progressive integration (though subject to fits and starts caused by intergovernmental bargaining) will proceed from the emergence of transnational interests. Research focusing on the formation, scope, and influence of these transnational interests helps explain how European-level bureaucracies in turn influence the integrative process and the character of European-level policies across issues. 18 Yet neofunctionalism, like intergovernmentalism, is too broad an analysis to totally explain European integration effectively. This is because, even though it describes in general spillover and incrementalist cooperation and institution building, it also encompasses a range of phenomena that demands further differentiation. Therefore, it is necessary that we move beyond the international relations perspective where European cooperation, norms, and governance are explained as the creation and performance of an international organization. We must also consider the analytical approaches that are more commonly applied to nation-states and the comparisons of different national polities.

Comparative Politics Approaches to Studying the European Union As the European Union assumes more attributes of a sovereign nationstate, political scientists have increasingly turned toward comparative (or national) political theory to explain the nature of EU policymaking and the behavior of European policy actors. The application of comparative politics paradigms to the European level is now widespread among political scientists seeking to understand the practice of European "governance" and to characterize the European "polity." The distinction between international relations and comparative political science approaches to European integration should not, however, be overstated. Both disciplines acknowledge the importance of national contexts in shaping European policy outcomes. Similarly, comparative political scientists are also interested in the impact of European policy actors and policymaking structures on European integration. Thus, they also tend to analyze phenomena within the wider theoretical parameters of intergovernmentalism and neofunctionalism. Political science approaches to the study of th E U are well represented in this volume. Several chapters, notably the case studies of E U

16

The State of the EU

policymaking in Part 3, utilize theoretical concepts and frameworks developed in the context of national politics and policymaking to analyze developments at the European level. There are two basic types of comparative political science approaches to European Union politics: those that focus primarily on the activities of individuals, groups, and governments and those that stress the importance of institutions. 19 Within the former category, rational choice and pluralist approaches represent the comparative political science counterparts to the neorealist and regime theory approaches associated with international relations. Rational

Choice

Perspectives

Rational choice theory assumes that individuals are essentially self-interested and have a rank order of preferences. Rational choice theorists also assume that individual and institutional policy actors try to maximize their objectives. Having made these assumptions, rational choice analysts then seek to explain individual behavior and policy outcomes by means of models such as game theory. The application of rational choice models to EU policymaking is a fairly recent, but potentially significant, development. Moravcsik's explanation of the 1986 Single European Act (SEA), for instance, prompted widespread debate and interest. In this study, Moravcsik argued that the SEA was not the result of neofunctionalist pressures for further integration but the product of calculated self-interest on the part of member states. 20 Of particular relevance to EU decisionmaking is George Tsebelis's model of "nested games." 2 1 Tsebelis argues that even when actors appear to make suboptimal choices, they are in fact behaving rationally. The explanation for apparently irrational behavior has to do with the fact that the observer's view of the game is incomplete. "The observer focuses attention on only one game, but the actor is involved in a whole network of games—what I call nested games." 2 2 Tsebelis identifies two types of nested games: games in multiple arenas and institutional design. In games in multiple arenas, choices made in one arena have consequences in all arenas. An individual may, therefore, choose a suboptimal strategy in one arena if this maximizes her overall position. In the case of institutional design, a rational individual seeks to increase the number of choices available by trying to redefine the rules of the game. The notion of nested games is thus helpful to rational choice explanations of EU decisionmaking because it accounts for member-state governments being involved in multilevel, multifaceted negotiations. This point is also made by Mark Pollack in his analysis of Community Structural Funds in this book. Other observers, including Gary Marks, have claimed that the European Commission and regional governments have become increasingly

17

Introduction

influential in determining the distribution of EC Structural Funds. 2 3 However, Pollack argues in Chapter 16 that all aspects of Community structural policy, including the redistributive decisions, can be explained in terms of side-payments in larger, intergovernmental bargains between national governments. Within this "rational actor" framework, the apparently "irrational" willingness of wealthier member states to redistribute EU resources to their poorer counterparts is explained by (1) the former states more intense preferences for market integration and monetary union and (2) the ability of the poorer EU member states to veto progress toward these wider objectives. Institutional design is analytically important in the sense that it highlights the link between individual actions and institutions. The latter are not merely "givens"; they may be shaped to suit individual interests. As demonstrated by the SEA, the Maastricht Treaty, and the forthcoming Intergoverment Conference in 1996, the EU rules of the game—or more loosely, the Community's Constitution—are still in a state of flux. The role of the European Court, the powers of the European Parliament, the role of regional governments, the future status of the European Central Bank, and even the geographical dimensions of the Community have not been precisely defined. These issues will undoubtedly be hotly debated by the member governments at the 1996 Intergovernmental Conference. However, rational choice approaches to the EU suffer important limitations. In particular, like neorealist approaches, they tend to assume that member states are unitary actors with a single set of rank-order preferences. Interministerial conflicts, partisan cleavages, and bureaucratic interests tend to be ignored. Rational choice explanations also take the concept of member state interests as given and rarely ask how and why some policy goals are included in the national interest while others are excluded. Thus, by focusing on the actions of a small set of state actors, they tend to underestimate the less obvious (but no less important) policymaking significance of nongovernmental actors, organized interests, and political ideology. Pluralist Approaches: Organized Interests and Policy

Networks

In sharp contrast, pluralist approaches to the European Community stress the political importance of social and economic groups in shaping policy outcomes. Interest group formation, strategies, and influence therefore constitute the focus of study. This paradigm now dominates research on the EU policymaking process. Comparative politics analysts agree that the decisionmaking process in the European Union is characterized by institutional fragmentation and openness. As such, it is more akin to the U.S. political system than several West European governmental systems. As

18

The State of the EU

Sonia Mazey and Jeremy Richardson point out in Chapter 15, there are many access points for groups (subnational, national, and transnational) to enter the policymaking process. In consequence, the European policymaking process is characterized by uncertainty and competitive agenda setting. As Guy Peters has argued, the institutional features of the European Community make it to some extent a "prospective agenda-setter's paradise." 2 4 This point is highlighted in Chapter 19 by Anthony Zito's analysis of the EU policy "agenda-setting" role played by environmental groups, though, as he acknowledges, such groups find it more difficult to secure their favored policy outcome. This is because the institutional complexity and openness of the EU policymaking process means that any "entrepreneurial coalition" is likely to meet with opposition. The undeniable proliferation of EU lobbyists in recent years has been matched by a similar increase in the number of interest group-based analyses of EU politics (demonstrating once again the link between political developments and intellectual fashion). 2 5 Since the focal point for EU lobbyists is the European Commission, much academic attention has focused upon highlighting linkages between groups and EU Commission officials. Such studies confirm that—just as is the case in national politics— groups are not equally influential at the European level. 26 Thus, it seems accurate to describe the European political system as one of limited pluralism. The policymaking influence of groups also depends upon the nature of the policy sector under review. In short, so-called "high politics" issues such as European security and foreign policy and EU institutional reform tend to be processed primarily by the member governments with little direct input from organized interests. However, more technical, scientific, or "low politics" policies (such as telecommunications policy) and the day-to-day management of established policies offer greater opportunities for groups to influence policies. Thus, the pace of European integration is variable depending upon the nature of the policy under consideration and the principal actors involved. The highly visible, much-reported business of European summits is understandably dominated by intergovernmental bargaining. Meanwhile, a number of contributions suggest that the development of EU sectoral policies (for example, in the fields of telecommunications and research) is more consistent with neofunctionalist theories of integration. One reason why groups are more influential in sectoral policymaking is because Commission officials are particularly dependent upon the scientific knowledge and expertise of the groups affected by proposals in these sectors. A good example of this phenomenon is provided here by John Peterson's analysis in Chapter 17 of the negotiations surrounding the fourth EU research framework program. Once the broad parameters of the program were set by national governments, Commission technocrats and outside experts filled in the details. A similar pattern of shifts in the

Introduction

19

institutional location of policymaking influence emerges from Smith and Sandholtz's analysis of Germany's role in the Exchange Rate Mechanism (ERM) crisis in Chapter 11. While the German government was influential during the Maastricht Treaty negotiations on Economic and Monetary Union (EMU), thereafter monetary policymaking power within Germany shifted from the government to the Bundesbank. Thus, both contributions remind us that the same policy issue may be processed at different stages by different actors in different institutional settings. Satisfactory explanations of the EU policy process may, therefore, require researchers to embrace more than one conceptual framework. In an attempt to impose some kind of order upon this apparently chaotic, pluralistic image of the EU political process, a number of scholars have invoked the "policy networks" model of group-state relations to describe linkages between groups and EU policymakers. The concept of a policy network was originally used by Hugh Heclo to describe the increasing importance of organized interests in the U.S. policy process. 2 7 Heclo's concern was with the dynamics of the policy process. His argument was that the entry of new groups into the policy process had created new "webs of influence" or "issue networks," which formed the basis of a loosely structured, open, and somewhat disorderly policy process. "Policy network" is in fact a generic term. In reality, there may be several different types of networks. Richardson and Jordan, for instance, developed the concept of "policy communities" to highlight the informal relationships between organized interests and national policymakers in Britain. The term policy community was intended to convey a very close and stable exchange relationship between policy actors based on a shared language and understanding of problems. 2 8 More recently, Rhodes and Marsh have argued that policy communities and the much looser, less consensual, issue networks represent the two ends of a continuum along which other types of networks (e.g., professional networks) are located. 29 The multinational character and often highly technical subject matter of much EU policymaking places a high premium on knowledge, expertise, and ideas. Thus, in Chapter 15 Mazey and Richardson use the term "epistemic communities," associated with the work of R Haas, to describe the policymaking importance of transnational networks of scientific and professional experts. 3 0 Similarly, they suggest that Sabatier's concept of "advocacy coalitions" comprising prominent individuals with shared ideas and beliefs may change intellectual fashion and policy. 3 1 In Chapter 19 Zito suggests that the existence and activities of such communities may help to explain the development of EU environmental policy. As critics have pointed out, the policy networks approach is essentially a descriptive model, which in itself does little to explain policy outcomes or the existence of different sorts of policy networks. Nevertheless, it is a valuable analytical tool for categorizing group-EU relations, which

20

The State of the EU

may also help us to understand the manner in which policies are processed within the European Commission. Moreover, the mapping of group activity on any particular policy issue further confirms the degree to which E U policymaking is a multilevel game requiring groups, governments, and EU actors to participate in several overlapping policy-networks with an interest in a policy sector. 3 2 To the extent that this approach reveals the mobilization of groups, it also highlights the development and significance of sociopolitical cleavages at the European level (e.g., between left and right over social policy and between industrialists and environmentalists). The existence of such cleavages is one attribute of a liberal democratic polity and thus a significant development in terms of European integration. Finally, of course, the focus on the positive attitude of many socioeconomic groups toward some E C policies (e.g., the Single Market program and monetary integration) has done much to revive interest in the importance accorded to organized interests by neofunctionalist theories of integration. 3 3

Institutional

and Constitutional

Perspectives

Institutional and constitutional/legal analyses of the Community have flourished in recent years, complementary to agency-based approaches to E U politics. Moreover, whereas earlier studies of European Community institutions tended to focus on the formal powers and organizational characteristics of these institutions, recent analyses have been more ambitious and more sociological in approach. Key issues addressed by "new institutionalist" approaches to the E U include the impact of institutional environments and institutional reform upon the E U policy process, the contribution made by institutions to the integrative process, and the relevance of concepts such as federalism and consociationalism to European Union governance. 3 4 A key assumption of new institutionalism is that institutions (and institutional membership) play an important role in shaping "rational" actor strategies. This approach also suggests that institutions tend to impose a "path dependency" on the policy process and a bias against radical policy change. 3 5 Support for this view is provided by Olufemi Babarinde's analysis of the development of L o m é in Chapter 2 1 , which concludes that, notwithstanding the current problems surrounding the E U ' s policy toward less-developed countries, institutionalization of the policy within the Commission and the African, Caribbean, and Pacific states makes its demise unlikely. The impact of E U institutions upon the decisionmaking process has attracted widespread interest. The role of the Council of Ministers in the European policymaking process has attracted considerable attention, especially from analysts keen to demonstrate the intergovernmental nature of the decisionmaking process. 3 6 Wessels, for instance, has argued that the

Introduction

21

need to satisfy national political demands may 3 7 impede the effectiveness of European policymaking within the Council and produce suboptimal European policy outcomes. The European Parliament—hitherto a rather neglected area of study—has also received more attention following the introduction of the "cooperation procedure" and "codecisionmaking power" in the SEA and the Maastricht Treaty respectively. Discussion of Parliament's codecisionmaking powers with respect to EU research policy and its budgetary powers with regard to environmental policy in Chapters 17 and 19 respectively indicate that the Parliament is likely to become an increasingly important policy actor. A recent evaluation of the Parliament's policymaking role concluded that its capacity to influence policy outcomes varies considerably among issues. More specifically, the same study argued that Parliament is more influential on issues of high political saliency (e.g., the environment) and on issues that do not require unanimous support within the Council of Ministers. The extent of interinstitutional discussion of policy proposals (between the Parliament, Commission, and Council) and the institutional resources (staffing, knowledge, and so on) available to members of the European Parliament (MEPs) also has an effect. 3 8 Though the formal powers of the European Parliament remain limited, several studies have highlighted the incremental development of the Parliament's political importance. 39 A significant aspect of this development, namely the emergence of transnational cooperation within the political groupings, is examined in Chapter 6 by Joanne Bay Brzinski. Her analysis of MEPs' voting behavior concludes that group cohesion, though still fragile, is increasing. Considerable attention has focused upon the "neofunctionalist" contribution of European institutions to the European integration process. There seems to be widespread support for the view that the European Commission, in its capacity as policy initiator, has generally promoted integration by extending its sphere of policy competence, by actively encouraging the creation and development of transnational policy networks in support of its initiatives, and by pursuing integration in a tactical, political manner. 4 0 In Chapter 18 Fuchs singles out the European Commission's "pro-active" role as the most important factor in explaining the development of European telecommunications policy. Community law specialists have also drawn attention to the significant contribution made to the integration process by the European Court of Justice. Such studies highlight the political significance of the legal doctrines of "direct effect" and "supremacy" of EU law over national law. In short, these principles, together with the Court's power of judicial review, provide the basis of a supranational legal system and European constitution. 41 As one commentator reminds us, these developments were not inevitable; they are the result of a conscious strategy on the part of the Court. 4 2 In a similar vein, Daniel Partan argues in Chapter 3 that the way in which the

22

The State of the EU

Court perceives its role in interpreting the subsidiarity principle will be important in determining the scope of European legislation. Other institutional analysts, rather than focusing on one institution, have examined the relationships between different EU institutions and the institutional linkages between regional, national, and European-level governments. The unique institutional structure of the European Union means that it is neither a sovereign state nor an international organization. Attempts to conceptualize this system have drawn heavily upon the federal principle. In particular, the model of "cooperative federalism" based upon concurrent powers has been suggested as an appropriate characterization of the European Community's decisionmaking structure, especially since the introduction of the subsidiarity principle. 4 3 This theme is pursued in Deeg's analysis (in Chapter 9) of the influence wielded by the German Länder during the Maastricht Treaty negotiations. Deeg argues that implementation of the treaty has brought about a new balance of power among the Länder, the federal government, and the EU.

Conclusion For EU member states the period 1993-1994 was dominated by the ratification of and adaptation to the Maastricht Treaty. As highlighted in this volume, few national governments found this a simple task. In part, their difficulties can be explained by the ambitious nature of the Maastricht Treaty, in part by the worsening economic conditions within the Union, and in part by the dramatic changes in the wider geopolitical situation. These difficulties gave rise to fresh doubts about the direction in which the Union should proceed. Thus, at one level, this period may be characterized as a time of reflection and reassessment. Yet, as highlighted by several contributions to this volume, European integration has continued, as reflected in the accession of Sweden, Finland, and Austria to the Union, in the development and institutionalization of key EU policies, and in the incremental Europeanization of national politics. The potential significance of these developments for European integration should not be underestimated. This ambiguity underlines the complex and uncertain nature of the European integration process. Viewed in historical context, the Maastricht Treaty represents only the latest step in a longer, piecemeal process of integration. As the European Community has evolved into a European Union, analysts have begun to refer to the Union as a "European polity" and "system of governance." Such terms are not intended to imply that the European Union is a sovereign state but that it constitutes an important political system with supranational legislative authority. As such, the EU is a unique phenomenon for which there exists no constitutional blueprint or institutional design. In the dynamic content of the objectives of European

Introduction

23

integration, there can be n o s i n g l e approach to s t u d y i n g the E u r o p e a n U n i o n . Our aim in this introductory chapter has b e e n to demonstrate the need for theoretical and m e t h o d o l o g i c a l pluralism in the f i e l d o f European studies. A s highlighted in the preceding d i s c u s s i o n , international relations theories concentrate primarily on the impact of s y s t e m i c and d o m e s t i c factors u p o n the attitudes of national g o v e r n m e n t s and other k e y actors toward integration. M e a n w h i l e , c o m p a r a t i v e p o l i t i c s approaches to the E U f o c u s o n characterizing and e x p l a i n i n g the internal p o l i t i c s and institutional balance of p o w e r within the Union. The chapters that f o l l o w illustrate the utility o f both approaches, demonstrating that a range o f explanatory m o d e l s broadens our understanding of the d y n a m i c s o f European U n i o n , leading us to a more comprehensive analysis of European integration.

Notes 1. The most prominent intergovernmentalist approach to the European Community is represented by Paul Taylor, "Intergovernmentalism in the European Communities in the 1970s: Patterns and Perspectives," International Organization 36 (Autumn 1982): 741-766; and Paul Taylor, The Limits of European Integration (New York: Columbia University Press, 1983); and more recently by Andrew Moravcsik, "Negotiating the Single European Act: National Interests and Conventional Statecraft in the European Community," International Organization 45 (Winter 1991): 19-56. 2. Andrew Moravcsik, "Preferences and Power in the European Community: A Liberal Intergovernmentalist Approach," Journal of Common Market Studies 31 (1993): 473-524. 3. James Caporaso and John Keeler, "The European Union and Regional Integration Theory," The State of the European Union Vol. 3: Building a European Polity? (Boulder: Lynne Rienner, 1995), 56. 4. Descriptive realism has been most associated with Hans Morgenthau, George Kennan, and Hedley Bull. However, Kenneth Waltz made the greatest contribution toward systematizing realism as a theoretical paradigm. See the discussion below of neorealism. 5. See Kenneth Waltz, Theory of International Politics (Reading, Massachusetts: Addison-Wesley, 1979); Morton Kaplan, System and Process in International Politics (New York: Wiley, 1957); Hedley Bull, The Anarchical Society (New York: Columbia University Press, 1977); and Richard Rosecrance, Action and Reaction in World Politics: International Systems in Perspective (Boston: Little, Brown, 1963). 6. See, in particular, Karl Deutsch et al., Political Community and the North Atlantic Area (Princeton: Princeton University Press, 1957); and Ernest Haas, The Uniting of Europe: Political, Economic and Social Forces, 1950-1957 (Stanford: Stanford University Press, 1958). 7. Robert Keohane, "Realism, Neorealism, and the Study of World Politics," Robert Keohane, ed., Neorealism and its Critics (New York: Columbia University Press, 1986), 17-18. 8. Andrew Moravcsik, "Preferences and Power in the European Community," 473. Thus, for many, "liberal intergovernmentalism" is a modification of neorealism

24

The State of the EU

that assumes that member states remain the crucial actors but also takes into account the domestic pressures that influence state policies. 9. Robert Keohane and Joseph Nye, Power and Interdependence: World Politics in Transition (Boston: Little, Brown, 1977). See also Richard Cooper, The Economics of Interdependence (New York: McGraw Hill, 1968). 10. See, for example, John Odell, "Latin American Trade Negotiations with the United States," International Organization 34 (Spring 1981): 207-228; John Odell, "International Trade Conflicts: U.S. and South Korea," International Studies Quarterly 29 (1985): 263-286; Carolyn Rhodes, "Reciprocity in Trade: The Utility of a Bargaining Strategy," International Organization 43 (Spring 1989): 2 7 3 - 2 9 9 ; and Klaus Stegemann, "Policy Rivalry Among Industrial States: What Can We Learn from Models of Strategic Trade Policy?" International Organization 43 (Winter 1989): 73-100. 11. Among others, Robert Keohane, Beth Yarbrough and Robert Yarbrough, and Kenneth Oye focus our attention on the value of organizational form at the international level, while Robert Keohane and Stanley Hoffman, Geoffrey Garrett, Wayne Sandholtz, and Gary Marks make similar arguments for focusing on institutional frameworks at the community level. See Robert Keohane, After Hegemony: Cooperation and Discord in the World Political Economy (Princeton: Princeton University Press, 1984); Beth and Robert Yarbrough, Cooperation and Governance in International Trade (Princeton: Princeton University Press, 1992); Kenneth Oye, Economic Discrimination and Political Exchange: World Political Economy in the 1930s and 1980s (Princeton: Princeton University Press, 1992); Robert Keohane and Stanley H o f f m a n , "Institutional Change in Europe in the 1980s," Robert Keohane and Stanley Hoffman, eds., The New European Community: Decisionmaking and Institutional Change (Boulder: Westview Press, 1991); Geoffrey Garrett, "International Cooperation and Institutional Choice: The European Community's Internal Market," International Organization 46 (Spring 1992): 5 3 3 - 5 6 0 ; Wayne Sandholtz, "Institutions and Collective Action: The New Telecommunications in Western Europe," World Politics 45 (January 1993): 2 4 2 - 2 7 0 ; and Gary Marks, "Structural Policy and Multilevel Governance in the EC," Alan Cafruny and Glenda Rosenthal, eds., The State of the European Community Vol. 2: The Maastricht Debates and Beyond (Boulder: Lynne Rienner, 1993). 12. See Thomas Schelling, The Strategy of Conflict (New York: Oxford University Press, 1960); Glenn Snyder and Paul Diesing, Conflict Among Nations: Bargaining, Decision Making and System Structure in International Crises (Princeton: Princeton University Press, 1977); Robert Axelrod, The Evolution of Cooperation (New York: Basic Books, 1984); Kenneth Oye, ed., Cooperation Under Anarchy (Princeton: Princeton University Press, 1986). For an overview of the use of public choice approaches to international relations theory and analysis, see Bruno Frey, "Public Choice and Global Politics," International Organization 38 (Winter 1984): 199-223; and Duncan Snidal, "The Game Theory of International Politics," in Oye, ed., Cooperation Under Anarchy, 25-57. 13. See, for example, J. David Singer, "The Level-of-Analysis Problem in International Relations," in Phil Williams, Donald Goldstein, and Jay Shafritz, eds., Classic Readings of International Relations (Belmont, California: Wadsworth Publishing, 1994), 86-97; B. P. White, "Decision-Making Analysis," in Williams et al., Classic Readings, 107-126; and Graham Allison, "Conceptual Models and the Cuban Missile Crisis," in Williams et al., Classic Readings, 131-155. See also Joe Hagan, "Regimes, Political Oppositions and the Comparative Analysis of Foreign Policy," in Charles Hermann, Charles Kegley, Jr., and James Rosenau, eds., New

Introduction

25

Directions in the Study of Foreign Policy (Boston: Allen and Unwin, 1987), 339-365; and David Skidmore and Valerie Hudson, The Limits of State Autonomy: Societal Groups and Foreign Policy Formulation (Boulder: Westview Press, 1993). 14. Robert Putnam, "Diplomacy and Domestic Politics: The Logic of TwoLevel Games," International Organization 42 (Summer 1988): 427-460. 15. Stephen Krasner, "Structural Causes and Regime Consequences: Regimes as Intervening Variables," Stephen Krasner, ed., International Regimes (Ithaca: Cornell University Press, 1983), 1-22. 16. A recent example of this perspective is Wayne Sandholtz and John Zysman, "1992: Recasting the European Bargain," World Politics 42 (October 1989): 95-128. 17. Ernst Haas has been particularly influential in the cross-fertilization between integration theory and regime theory. See "Turbulent Fields and the Theory of Regional Integration," International Organization 30 (1976): 173-212. See also his more general contribution to understanding regimes as part of processes of cooperation in "Words Can Hurt You; or, Who Said What about Regimes," in Stephen Krasner, ed., International Regimes, 23-59. 18. For two useful overviews of integration theory and the role of the f u n c tionalist process, see Michael Hodges, "Integration Theory," in Trevor Taylor, ed., Approaches and Theory in International Relations (London: Longman, 1986), 237-256; and A. J. R. Groom and Alexis Heraclides, "Integration and Disintegration," in Margot Light and A. J. R. Groom, eds., International Relations: A Handbook of Current Theory (London: Pinter, 1985), 174-193. 19. Simon Hix, "The Study of the European Community: The Challenge to Comparative Politics, West European Politics 17 (January 1994): 1-30. 20. Moravcsik, "Negotiating the Single European Act," 19-56. 21. George Tsebelis, Nested Games: Rational Choice in Comparative Politics (Berkeley: University of California Press, 1990). 22. Ibid., 7. 23. Gary Marks, "Structural Policy in the European Community," in Alberta Sbragia, ed., Euro-Politics: Institutions and Policymaking in the "New" European Community (Washington, D.C.: Brookings Institution, 1992), 191-224. 24. Guy Peters, "Agenda-setting in the European Community," European Journal of Public Policy 1 (1994): 3 - 2 6 . 25. See, for example, Sonia Mazey and Jeremy Richardson, eds., Lobbying in the European Community (Oxford: Oxford University Press, 1993); M. Van Schendelen, ed., National Public and Private EC Lobbying (Aldershot: Dartmouth Press, 1993); J. Greenwood, J. Grote, and K. Rönit, eds., Organized Interests and the European Community (London: Sage Publications, 1993); S. S. Andersen and K. A. Eliassen, "European Community Lobbying," European Journal of Political Research 20 (1991): 173-187; and G. Jordan, A. M. McLaughlin, and W. A. Malone, "Corporate Lobbying in the European Community," Journal of Common Market Studies 31 (1993): 191-212. 26. Sonia Mazey and Jeremy Richardson, "Environmental Groups and the EC: Challenges and Opportunities," in D. Judge, ed., A Green Dimension for the European Community: Political Issues and Processes (London: Frank Cass Ltd., 1993), 109-128. 27. Hugh Heclo, "Issue Networks and the Executive Establishment," in Anthony King, ed., The New American Political System (Washington, D.C.: American Enterprise Institute, 1978). 28. Jeremy Richardson and Grant Jordan, Governing Under Pressure: The Policy Process in a Post-Parliamentary Democracy (Oxford: Martin Robertson,

26

The State of the EU

1979), 7 3 - 7 4 . For a recent discussion of the utility of "policy networks" approaches to the European Union, see Jeremy Richardson, "Actor Based Models of National and EU Policy-Making: Policy Communities, Issue Networks and Epistemic Communities," in A. Menon and H. Kassim, eds., The EU and National Industrial Policy (London: Routledge, 1995). 29. See R. A. W. Rhodes and D. Marsh, "New Directions in the Study of Policy Networks," European Journal of Political Research 21 (1992): 181-205; R. A. W. Rhodes and D. Marsh "Policy Networks in British Politics," in D. Marsh and R. A. W. Rhodes, eds., Policy Networks in British Government (Oxford: Clarendon Press, 1992), 1-26. 30. Peter Haas, "Introduction: Epistemic Communities and International Policy Co-ordination," International Organization 49 (1992): 1-35. 31. Paul Sabatier, "An Advocacy Coalition Framework of Policy Change and the Role of Policy-Orientated Learning Therein," Policy Sciences 21 (1988): 128-168. 32. Sonia Mazey and Jeremy Richardson, "Policy Co-ordination in Brussels: Environmental and Regional Policy," in S. Baker, K. Milton, and S. Yearly, eds., Protecting the Periphery: Environmental Policy in the Peripheral Regions of the European Union (London: Frank Cass, 1994), 22-44. 33. Hix, "The Study of the European Community," 1-30. 34. Keohane and Hoffmann, eds., The New European Community; and Sbragia, ed., Euro-Politics. 35. S. Bulmer, "The Governance of the European Union: A New Institutionalist Approach," Journal of Public Policy 13 (1994): 351-380. 36. E. J. Kirchner, Decision Making in the European Community: The Council Presidency and European Integration (Manchester: Manchester University Press, 1992). 37. W. Wessels, "The EC Council: The Community's Decision Making Center," in Keohane and Hoffman, eds., The New European Community, 133-164. 38. David Judge, David Earnshaw, and Ngair Cowan, "Ripples or Waves: The European Parliament in the European Community Policy Process," Journal of European Public Policy 1 (1994): 27-52. 39. R. Wittke-Bourguignon et al., "Five Years of the Directly Elected Parliament: Performance and Prospects," Journal of Common Market Studies 24 (1986): 3 9 - 5 9 ; J. Lodge, "EC Policy-Making: Institutional Dynamics," in J. Lodge, ed., The EC and the Challenge of the Future, 2d ed. (London: Pinter, 1993); and F. Jacobs, R. Corbett, and M. Shackleton, The European Parliament, 2d ed. (London: Longman, 1992). 40. Laura Cram, "Calling the Tune without Paying the Piper? Social Policy Regulation: The Role of the Commission in European Community Social Policy," Policy and Politics 21 (1993): 135-146; and Sonia Mazey, "Bureaucratic Expansion on Behalf of Women? An Evaluation of EC Equal Opportunities Policies," Public Administration (forthcoming, 1995). 41. G. F. Mancini, "The Making of a Constitution for Europe," in Keohane and Hoffmann, eds., The New European Community, and M. Shapiro, "The European Court of Justice," in Sbragia, ed., Euro-Politics, 177-194. 42. Daniel Wincott, "The Role of Law or the Rule of the Court of Justice," Journal of European Public Policy 2 (forthcoming). 43. Fritz Scharpf, "The Joint Decision-Trap: Lessons from German Federalism and European Integration," Public Administration 66 (1988): 239-278; and W. Wessels, Alternative Strategies for Institution Reforms (working paper, no. 85, European University Institute, 1984).

PARTI REFLECTIONS ON EUROPEAN INTEGRATION

2 The European Union and Regional Integration Theory James A. Caporaso & John T. S. Keeler

This chapter traces the development of theories designed to illuminate the process of regional integration in Europe from the era of the European Coal and Steel Community ( E C S C ) to the launching of the European Union (EU). It should be stressed at the outset that our chapter makes no pretense of covering all the important European Community (EC) literature published over the last forty years. Many insightful historical works, case studies of developments within particular issue areas, and typological efforts will be neglected here, for we focus exclusively on the works we consider to be of the greatest theoretical significance. Moreover, our treatment of the literature may strike some readers as excessively "U.S.-centric" or perhaps "Anglo-American-centric." We readily acknowledge that we are most familiar with publications in English and may thus have slighted some significant works in other languages. However, it is our conviction that the U.S.-centric nature of our piece also reflects, more defensibly, the fact that U.S. scholars have manifested a disproportionate interest in developing social science theories related to the phenomenon of European integration. We approach our subject matter historically, examining in turn the major stages in the development of regional integration theory. The first section deals with the theory of functionalism elaborated during the 1940s and 1950s. The second section focuses on neofunctionalism, the predominant theory developed from the late 1950s through the early 1970s. Ideally this section would also include an extensive discussion of the communitarian theory pioneered by Karl Deutsch; we readily acknowledge that this approach—grounded in cybernetics and focusing on transactions, messages, and the emergence of shared values among core groups—was important and had a sizable following during the 1960s and early 1970s. 1 However, the communitarian approach faded more quickly than neofunctionalism and has been all but absent in the current debates; in light of that

29

30

Reflections on European Integration

fact and our spatial constraints, we thus deal only in passing with the Deutschian perspective. 2 The third section deals with the contributions of what has been termed the "doldrums era," an oft-neglected period stretching roughly from Ernst Haas's The Obsolescence of Regional Integration Theory of 1975 to the mid-1980s. While U.S. scholars turned away from the study of the E C and integration theory in this era, those scholars on both sides of the Atlantic who continued working in the field produced a variety of insightful studies and built much of the empirical foundation on which the latest wave of debates over integration theory rests. The fourth section examines the rich "new wave" of literature that has emerged since the mid-1980s. The conclusion offers some summary judgments on the development of integration theory and some reflections on the future of theory related to the European Union.

The First Wave: Functionalism The theory of functionalism has historical roots that reach into the nineteenth century. While functionalism preceded the drive for European integration, it nevertheless became prominent as the theoretical backdrop for the movement toward formation of the E C S C . The primary spokesperson for functionalism in the postwar period was David Mitrany, whose A Working Peace System became a bible for functionalist theory and practice. 3 While functionalist theory has never been codified, its logic is best seen in terms of a threefold distinction between background conditions, process conditions, and the conditions necessary for task expansion (or spillover) to take place. But prior to these three categories, and cutting across them, is a belief about the relationship between problem solving and territory in the modern world. Mitrany argued that traditional views of problem solving—as tied to authoritative territorial structures—are not the only way to deal with practical problems among countries. The control of navigation on the Rhine or Danube, the regulation of pollution in the Mediterranean, and the management of air traffic across several countries do not necessarily require a territorial authority with national representation, voting rights, and state control. The principle of "rule" could also be functional, with experts solving problems without primary regard to national jurisdictions. 4 Concerning background conditions, functionalism accepted many of the structural and ideological conditions later characterized as part of "the New Europe." 5 These included a muting of ideological conflict among key social groups (not only workers and capitalists but also religious and ethnic-national groups). Society was reoriented away from its cleavages toward the collective gains resulting from cooperation. The "economic problem" was defined as productivity rather than distribution.

The EU and Regional Integration Theory

31

The waning of ideological conflict and the differentiated structure of advanced capitalism provided the backdrop for the process conditions of integration. Europe was an area of dense economic exchange, of trade, capital f l o w s , tourism, communication, and so on. These international processes, undramatic in one sense, provided a fertile environment for further cooperative ventures. A basic infrastructure for cooperation was already in place. Instituting a Coal and Steel Community, and later a free trade area and customs union, could be seen as extensions of what already was in place rather than as radical departures. Finally, functionalism thought in terms of the conditions favoring task expansion. Once initial integrative processes had started, how could they be sustained or expanded? The key mechanism sustaining these secondary (but crucial) integrative processes was spillover. Spillover referred to two different but related processes. The first, sectoral spillover, involved the expansion of integrative activities from one sector to another, e.g., from coal and steel to agriculture or from customs union to monetary policy. A second type of spillover involved increasing politicization of sectoral activity as, for example, when the coordination of monetary policies was replaced by a more centralized system of governance. Task expansion in this case required a greater delegation of political authority by member states to international institutions. A powerful attraction of functionalism was that its appeal cut several ways at once. It simultaneously provided a theory of the social structure of advanced liberal capitalism and a theory of action. In other words, functionalism had a distinctive praxis. In addition, functionalism offered a positive theory of how the world worked and a set of prescriptions for how to improve it. While meliorist and incremental in means, the functionalist program offered far-reaching—not to say revolutionary—long-term results. Given functionalism's preoccupation with the details of production and commerce, we often forget the long-term normative charge that it carried. In the eyes of many scholars and practitioners, the federalist project was tied excessively to nation-states—to nationalism, militarism, bilateral deals, balance of power, and so forth. The opportunity to think and speak in terms of transnational economic ties, mutual cooperation, the taming of nationalism, spillover, and community spirit seemed both analytically sensible and morally uplifting. It was part of Jean Monnet's genius that he created a new way of talking about what people were doing transnationally, and this vocabulary was remarkably nonstatist. According to Monnet's doctrine, the most mundane facts took on major significance. The smallest acts of everyday l i f e — o f production, commerce, and voluntary associations—could lay the groundwork for new ways of organizing relations among countries. While Adam Smith showed how individual self-seeking—without cooperative motivation—could lead to the common good, Monnet argued that self-interested

32

Reflections on European Integration

transnational interactions of the most practical kind could transform the interstate system. In this environment, the collection of data (the mundane facts referred to above) became a religious experience. This preoccupation with the technical and noncontroversial, and the forestalling of the political, reflects well the functionalist logic. Avoid comprehensive plans for integration and minimize the political, at least initially. Take advantage of social differentiation and carve society into sectors, the more specialized and autonomous the better. Beginning with piecemeal cooperation, allow the chief dynamic mechanism—spillover—to do its work. This, in highly simplified form, was the functionalist program. Purely as a sociological theory of modern industrial capitalism, functionalism seemed constructed on a contradiction. The differentiation of society into subsystems, especially sectoral ones, allowed a greater role for the expert and hence greater autonomy. To the extent that distance could be put between various economic sectors and the state, experts had greater room to maneuver and greater freedom to solve problems. Functionalists never theorized the space for autonomous expert action nor provided much of an argument about the conditions under which political agents would challenge economic integration. Spillover, on which so much depended theoretically, could go many ways, as the terms spillover, spillback, and spillaround suggest. Thus, the major variable in terms of explaining the dynamics of integration was indeterminate, adaptable to any outcome but lacking a distinctive predictive component. Functionalism did not have a well-developed theory of either politics or political choice. Its central message, a valuable one, focused on changing constraints and preferences due to exogenously specified social and economic cooperation. But let us anticipate a rational choice argument; the incentive structure alone cannot explain outcomes. In a world of strategic indeterminacy, even with preferences specified, many outcomes are usually possible. A theory of choice and a theory of politics (i.e., power and bargaining) are needed, but functionalism falls short of this.

The Second Wave: Neofunctionalism As a theory of regional integration, functionalism was largely superseded by neofunctionalism. Many of the crucial concerns of functionalism were carried forward: the importance of social and economic actors (in contrast to the state-centric paradigm), the role of issue-linkage and spillover, and the way in which initial cooperative probes altered the incentives facing political elites. What neofunctionalism changed were three things. First, there was a methodological reorientation, which can best be described as an effort to systematize and express in social science concepts the subject matter of regional integration. Neofunctionalism relied on a mix of

The EU and Regional Integration Theory

33

languages drawn from interest group theory, systems theory, and, to a lesser extent, economics. Second, neofunctionalism attempted to bring together into a unified model theoretical fragments relating to background conditions, conditions at the time of union, process conditions, and outcomes. A s a theory, it sought to present a comprehensive account of how the parts fit together. Third, neofunctionalism recognized the importance of supranational institutions and tried to provide a theory of politics. As theories, functionalism and neofunctionalism recapitulated the substantive process they were trying to explain. Functionalism presented the initial challenge. It downplayed the state, emphasized social actors, and introduced the language of interest and incremental problem solving. It elevated the mundane to the platform of world history. But it did not provide the political and institutional details necessary to explain how social and economic processes resulted in integrative (or disintegrative) outcomes. Functionalism went straight from structure to process to outcome, with spillover functioning as a touchstone in regard to intervening factors but not as a theoretical compass. Neofunctionalism made several refinements in regional integration theory. First of all, due largely to the work of Haas, Lindberg, Nye, Puchala, and Schmitter, neofunctionalism clarified the dependent variable—regional integration—and set it apart from related terms such as regional cooperation, regional organization, and regional systems. Haas defined the task as one of "explaining how and why they [states] voluntarily mingle, merge, and mix with their neighbors so as to lose the factual attributes of sovereignty while acquiring new techniques for resolving conflict between themselves." 6 In contrast to Deutsch, who saw the flow of goods, services, people, and messages across countries as central, Haas placed emphasis on institutions and attitudes. With regard to institutions, decisionmaking patterns were critical. With regard to attitudes, the learning of integrative habits was important. Institutional loyalties, orientations, and lobbying activities should refocus themselves around the European Commission, Council of Ministers, Court of Justice, and European Parliament as practical problems were addressed, and perhaps solved, by E C institutions. B y identifying authority-legitimacy transfers as the central concern, Haas provided a dependent variable that was highly general, continuously variable, and not confined to discrete, holistic outcomes (confederation, federation). 7 Second, neofunctionalism contributed importantly to our understanding of spillover. As mentioned, spillover shouldered most of the burden of explaining change. If the concept of spillover could be successfully employed, neofunctionalism had some claim to being a dynamic theory. Without this, it was at best reduced to comparative statics. Within the E C , spillover had a mixed record across different sectors. The progressive implementation of the customs union was closely linked to progress in both agriculture and transport. However, agriculture was a sector of much

34

Reflections on European

Integration

greater accomplishment than transport. 8 Similarly, the full integrative effect of the free trade area and common external tariff could not be felt without some coordination o f e x c h a n g e rates, and this in turn implied broader cooperation regarding monetary policy. T h e positive linkages between agriculture and the customs union and between customs union and monetary policy are examples o f spillover. In other areas, for example, transport, integration did not proceed very far and was self-contained. Functionalism's treatment o f spillover relied on a kind o f technical self-determination. Given the right background conditions, the right initial cooperation, and favorable attitudes on the part o f elites, cooperation was bound to expand. At best, this formulation suffers from underdetermination (not enough information to yield precise predictions). At worst, it b e c o m e s tautological, if "the right" background conditions and "favora b l e " attitudes cannot be specified ex ante. Haas and others (especially Schmitter) 9 recognized the problem and sought to fill the gap between structure and outcome with a theory of politics. At a minimum, neofunctionalism provided an extensive descriptive account o f the poorly understood region lying between initial cooperation and integrative or disintegrative outcomes. B y doing so, it pointed toward what needed to be better understood and accented the need for a theory of politics. Third, neofunctionalism attempted to develop a theory of politics on many fronts, not just as applied to spillover. Functionalism self-consciously downplayed the importance o f politics and the politician. B y focusing on social and economic sectors, functionalism seemed to be saying that political actors could be " f i n e s s e d . " If concrete habits o f cooperation could be put into place among countries by private actors, governments would have to adjust policies and institutional frameworks. Neofunctionalists disagreed. They argued that a delegation of political power (however circumscribed) to a central authority was critical. O n c e in place, integration could proceed within this institutional framework; indeed, the process o f integration might outgrow the original framework and require institutional changes. Thus, for neofunctionalists supranational political elites were important catalysts to change. T h e work of Ernst Haas represented the leading edge of neofunctionalism, and its political scope was striking. He recognized that D e Gaulle's resistance to European integration was not idiosyncratic but rather was a metaphor for a more widespread phenomenon that corresponded somewhat (at least) with nationalism and the institutional interests o f the state. D e Gaulle, as c h i e f executive o f a m a j o r member state, could bring the process of integration to a halt, despite the enormous reservoir o f e c o nomic benefit tied up in the E C . F a c e d with this challenge, neofunctionalism had two options: ( 1 ) it could ignore De Gaulle, label the general's actions idiosyncratic, and wait for the long-term economic forces to assert themselves; or ( 2 ) it could try

The EU and Regional Integration

Theory

35

to incorporate what De Gaulle represented. To H a a s ' credit, he took the second path and tried to theorize the limits of a style of decisionmaking that was rooted in fragmented, issue-oriented, incremental problem solving. Haas noted: T h e c h i e f i t e m in this l e s s o n is the r e c o g n i t i o n that pragmatic-interest p o l i t i c s , c o n c e r n e d with e c o n o m i c w e l f a r e , has its o w n built-in limits. . . . P r a g m a t i c interests, b e c a u s e they are pragmatic and not r e i n f o r c e d with d e e p i d e o l o g i c a l or philosophical c o m m i t m e n t s , are ephemeral. Just b e c a u s e they are w e a k l y held they can be readily scrapped. A n d a political p r o c e s s that is built and projected from pragmatic interests, therefore, is bound to be a frail process susceptible to r e v e r s a l . 1 0

This recognition of the limits of functional integration was not just ad hoc, a descriptive cop-out. It drew a lesson, or perhaps two: First, that integration was not a machine that could go of its own motion; second, that politics and political bargaining could not be ignored for long. Quiet periods of incremental advance were likely to be preceded by a political bargain. Once the limits set by the political bargain were met, politics entered the fray. The absence of day-to-day involvement by political authorities should not be misinterpreted. Neofunctionalism's fourth, and final, refinement was to recognize the importance of external environment. If Western Europe (or six of its m e m bers) was a distinct area, it was nevertheless embedded within a larger international system and one that impinged on its members in different ways. Stanley H o f f m a n n had been saying this for quite some time and in the p r e f a c e to the second edition of The Uniting of Europe (1968), Haas gave H o f f m a n n his due. H o f f m a n n ' s point was quite simple, that each member state was a distinct entity with its own culture, interests, "national situation," and capabilities. Members were also situated differently within the broader global system, and the tugs and pulls were not the same for all. In crucial m o m e n t s — t h e oil crisis, expansion of membership, monetary policy—they went in different directions. 1 1 Not much new theory came from this, but, as with spillover, there was at least a descriptive recognition that something was lacking. This recognition provided a stimulus that set realists to thinking about the global context of regional integration. The heyday of neofunctionalism corresponded to the early period of integration in the EC, f r o m the entry into force of the Treaty of R o m e in 1958 to the completion of the customs union in 1968. The years 1961 through 1965 witnessed implementation of the c o m m o n agricultural policy, including measures controlling the price of agricultural commodities as well as structural policy. To be sure, the neofunctional "period" did not include only integrative success. The constitutional crisis of 1 9 6 5 - 1 9 6 6 also occurred during this time. Nevertheless, neofunctionalism continued to be influential well into the 1970s.

36

Reflections on European Integration

Despite neofunctionalism's advances, problems remained. The first derived from the fact that the European Community (or rather the EC conflated with its precursor, the ECSC, which was the focus of Haas' seminal book) represented a statistical population of one. 1 2 Findings did not travel well. The limitation was not statistical, springing from the difficulties of generalizing from one case. The problem ran deeper. What passed for theory was often simply empirical generalization. 13 Such empirical generalizations, while based on observed regularities, do not meet the standard of theory in two senses. First, they are not part of an intellectual structure that explains the regularities. Second, they do not support counterfactual claims about what "would happen if. . . . " This second limitation is serious because it implies explanatory difficulties. In short, the evidence from observation on the EC cannot do double duty, it cannot simultaneously generate a hypothesis inductively and serve as evidence for it. With few if any other regional units to use as independent sites, and with cross-issue comparisons suspect because of Galton's problem, research on the EC is in a bind. A second criticism was that neofunctionalism had shortcomings relating to a theory of politics. It focused our attention on political elites, and it argued the significance of parties, interest groups, domestic politics, and the bargaining that takes place in Brussels and between Brussels and national capitals. But it did not provide either a theory of bargaining or of political choice. Given the close attention paid to decisionmaking in the EC, this may seem an odd criticism. But this attention was almost completely of a descriptive nature with the "presence or absence of political will" often invoked to "account for" the outcomes.

Theoretical Darkness—and Light—in the Doldrums Era From the perspective of Ernst Haas and other neofunctionalist scholars who dominated the regional integration field through the 1960s, the European Economic Community represented an exciting phenomenon whose development seemed explicable in terms of a powerful theory. As Haas summed up the situation in early 1965, the Community's activities seemed to have "come close to voiding the power of the national state in all realms other than defense, education and foreign policy" and "the functional logic which may lead, more or less automatically, from a common market to political unification, seemed to be neatly illustrated by the history of the E E C . " 1 4 But the ink was barely dry on such assessments of Euro-reality and theory when the 1 9 6 5 - 1 9 6 6 Empty Chair crisis provided the rudest of jolts. B y 1967 Haas was already attempting to cope with the possibility that De Gaulle had "killed the Common Market" by revising his theory to

The EU and Regional Integration

Theory

37

account for the prospect of "disintegration," and by 1975 he was announcing the "obsolescence of regional integration theory." 15 Although the EC was not in fact "killed" by De Gaulle, the phenomenon of European integration did appear less dramatic in the wake of the Empty Chair crisis and the economic crises that followed. The period from the early 1970s to the early 1980s has often been characterized as the doldrums era or the "Dark Ages" for the Community. 1 6 Moreover, it has frequently been portrayed—at least by American scholars—as a Dark Age for the development of integration theory. From the perspective of the mid1990s, to what extent does the "Dark Age" image do justice to both the reality of the EC and theory related to it as they evolved from the early 1970s to the mid-1980s? The EC's development stagnated in some respects during this period, and, as Figure 2.1 illustrates, the perceived newsworthiness of the Community—as measured by coverage in the New York Times—certainly diminished. However, it is important to note that the Community did not suffer the sort of precipitous disintegration that Haas and others once thought possible. 17 Many scholars, especially in Europe, have persuasively argued that the EC's survival "with so little damage to its basic structure" in the face of the adverse environment of the 1970s should be viewed as a considerable achievement. 1 8 Such an assessment seems even more cogent when one acknowledges that the Community managed to widen its membership substantially and make gains in other areas during that era. 1 9 From 1970 to 1980 the perceived power of "Brussels" was sufficient to generate an increase in the number of Euro-lobbying groups from approximately 300 to 439. 2 0 And as Ginsberg demonstrates, foreign policy activities of the EC, if not in a growth mode, certainly held their own. 2 1 The resilience of the EC during these troubled times looks even more impressive from the post-1989 perspective; after all, the Community has now outlived a good number of nation-states created long before its inception! Whether or not it was justified by the reality of the EC, it is indisputable that the "doldrums era" featured a precipitous decline of scholarly interest in the Community on the part of those (mainly U.S.) scholars who had earlier viewed it as a topic of vital theoretical concern. Figure 2.2 shows the extent to which studies of the EC and integration theory diminished in the United States after Haas issued his obsolescence declaration and news of the Community began to disappear from the headlines. Two measures of theory-driven scholarship on integration and the EC are plotted in this figure: (1) the number of articles on regional integration and the EC in International Organization (the average number of such articles for each three-year period is listed at the mid-point year, i.e., the 1990 figure includes 1989-1991 2 2 ); and (2) the average number of citations of Haas' The Uniting of Europe in the Social Science Citation Index for each three-year

38

Reflections on European Integration

Figure 2.1 European Community Articles in the New York Times •

Year

The EU and Regional Integration Theory

39

period. As Smith and Ray note in their recent study, the 10 data provide a useful but obviously imperfect measure of "the rise and fall of integration studies." We thus felt that it would be illuminating to complement their measure by using references to Haas' classic, which is cited in virtually all articles (especially by U.S. scholars) on the EC and integration that have any theoretical component; the Haas measure is also useful given that his book's original publication date ( 1 9 5 8 ) made it available for citation throughout the years on which our data focus. As Figure 2.2 shows, the 10 data and the Haas data follow roughly parallel trajectories; the correlation between the two data sets is 0.77. Figure 2.2 also plots the New York Times data from Figure 2.1 (now presented in a slightly different form: the average annual number of articles for each three-year interval) to illustrate the extent of covariance between news coverage of and scholarly attention to the EC. William Wallace has argued that Americans became overly pessimistic about the development of European integration in the 1970s in part as a reaction to their excessive optimism in the 1950s and 1960s. 2 3 Europeans (Britons included) may have had a more realistic view of E C developments; at the least they had a more immediate concern with the progress of the Community. For whatever reason, European scholars produced more work on the E C than their U.S. counterparts from the mid-1970s through the mid-1980s. How important was the theoretical content of the EC scholarship produced during this era? A recent essay (from the United States) discussing the evolution of scholarship on European integration reflects the prevalent U.S. perspective in this regard. After discussing the evidence of decline in U.S. scholarship on integration, Smith and Ray assert that integration studies "did continue, mainly in Europe, but the focus was most often on microlevel case studies rather than the theory-building orientation that was a large part of earlier work." 2 4 In the same vein, two other authors began a recent article by asserting that the European integration field has been deemed "academically moribund for much of the past two decades." 2 5 Is this a fair assessment, or should the major studies of the "doldrums era" be viewed as part of an on-going effort at integration theory-building? Our contention is that the scholarship of that period merits more attention and respect than it has been accorded. It is true that a good deal of this work was self-consciously atheoretical and indeed sometimes mocked Haas-style efforts at grand theory with alleged predictive capacity. However, it is also true that the best "Dark Age" research presented cogent theoretical insights or produced "microlevel case studies" that implicitly made significant contributions to the enterprise of theory building. First, this literature effectively challenged Haas' argument that the institutions of the Community would not endure in a state of messy equilibrium. In his "Turbulent Fields" piece, Haas had disagreed strongly with

40

Reflections on European

Integration

those (e.g., Puchala, Taylor, Lindberg, Scheingold, and Wallace) 26 who argued " t h a t . . . the present mixed institutional structure is likely to continue indefinitely." The "halfway house" could not last, he insisted, because its lack of a clear central authority would make it ineffective in "coping with complexly linked and highly controversial issues on the European agenda" and would ultimately lead to a collapse of legitimacy. 27 Two of the responses to this position are worth noting. William Wallace acknowledged that the EC "may not survive the slow erosion of its authority," but he also stressed that the "halfway house," which seemed inherently untenable within the bounds of Haas' neofunctionalism, might appear far more promising if viewed in terms of an evolutionary federalist theoretical perspective. Wallace chastised U.S. scholars for forgetting that in domestic political systems like their own (or that of West Germany), federalism entailed a messy, constantly shifting sharing of authority. 28 He also stressed that federations or confederations "do not follow unilinear paths from unity to disunity. 29 Authority crises might thus represent temporary setbacks followed, under changed environmental conditions, by greater federal/EC authority and capability. Puchala's (implicit) response to Haas focused on a different issue. In his "Domestic Politics" piece of 1975, Puchala stressed that his case study on regional harmonization showed little evidence that the EC Commission was very "supranationally authoritative" even in the era before the Empty Chair crisis. "In contrast to notions about how regional systems should supersede enfeebled nation-states," wrote Puchala, "the European experience demonstrates that successful integration demands strong national governments for purposes of carrying regional programs into effect." The power of the EC's member states had thus hardly been "voided" (Haas' term), and the Commission often acted more like a "pussycat" than a commanding authority, yet "European" solutions could still be achieved. 30 Second, the "doldrums era" literature both elaborated useful models of the EC's policymaking process and specified the dynamics that allowed for integrative outcomes. In terms of the general model to guide analysis of policymaking, a consensus emerged that earlier work on the EC had been far too Brussels-centric, for "the Community institutions themselves represent but the tip of an iceberg." 31 Precise formulations of a more realistic model varied from Puchala's "Concordance System" to Webb's "dense network" to Wallace's "confederation" or "partial political system." All of these models converged, however, on the central point that the Community needed to be studied as a complex, multilevel system encompassing numerous actors at the EC, national, and even subnational levels. In terms of the dynamics of integration, proponents of these models were unable to offer a parsimonious alternative to Haas' "spillover" (indeed, Puchala made a point of noting sardonically that "nothing spilled anywhere" in a case study he assessed), but they did offer a realistic and

The EU and Regional Integration

Theory

41

useful emphasis on complex bargaining, side-payments, and log rolling. 32 As will be noted in the next section, exemplars of the latest wave of more self-consciously theoretical integration analyses often pick up where these "Dark Ages" analysts left off. A crucial point made repeatedly in the work of Puchala and others who carried out (oft-maligned) case-studies of policymaking was that the integration/bargaining process often makes more continuous—but painfully slow—progress in particular sectors than might be apparent to generalists and casual observers. Through what Puchala deemed "political reverberation," policies were seen as shuttling back and forth several times between the EC level and the national (or even subnational) level before a "European" solution could be achieved. Slow and tedious though the process is (one might note that Haas described the integration process as occurring in small steps, but he seemed to expect larger steps than were observed), and absent though it might be from headlines, it often leads to a successful outcome. It has taken as long as fourteen years to achieve harmonized policy involving matters ranging from Value Added Tax (VAT) to jam. 3 3 Third, the "Dark Ages" literature clearly established that the "EC system" developed in an asymmetrical fashion, with certain policy sectors maintaining or steadily acquiring relatively high levels of policy integration while little progress was made elsewhere. Agriculture was widely recognized as the only economic sector in which the Community approached "the full powers of a federal government." 3 4 Doldrums-era scholarship by Feld and Pearce on the Common Agricultural Policy greatly clarified the scope—and perhaps more importantly, the limits—of Brussels' power on this issue. Their analyses found that agriculture is largely governed by the member states, yields policies much less "common" (in terms of both uniformity and administrative control) than often assumed, and has been increasingly "renationalized" over time. 35 On a different issue, work by Stein and Weiler demonstrated the extent to which—during the allegedly Dark Age—the European Court of Justice (ECJ) managed to make the legal sector the one most closely approximating the dynamic of a federal system. 3 6 During the EEC's first decade the ECJ was not yet prepared, in either jurisprudential or (using the term broadly) political terms, to make its presence felt beyond the narrow bounds of the legal community. However, the ECJ set the stage for a more visible future during those years by elaborating the principal doctrines (especially "supremacy" and "direct effect") and establishing the contacts with national courts that would steadily increase its case load and enable it to make a significant impact. By 1980 the Court was delivering more than six times as many rulings per year as it had in 1968; and the rate and range of ECJ decisional output—as well as the constraints posed by its jurisprudence—would continue to increase greatly throughout the 1980s. We will

42

Reflections on European

Integration

return in section 4 to issues posed by the E C J ' s bold development of "a constitutional framework for a federal type structure in Europe." 3 7 The persistent asymmetry of Community development and the precise nature of policymaking dynamics in sectors such as agriculture and law fit poorly with both initial neofunctionalist predictions and the later, more pessimistic projections of Haas. Although no doldrums-era grand theory emerged to explain these developments, a variety of authors did view them as grounds for accepting what Sbragia would later term a "segmented federalist" perspective on the E C ' s integration process. 3 8 Fourth, those who worked on the issue of mass attitudes toward the EC during the "doldrums era" produced findings that are quite relevant now in the post-Danish referendum era: Studies undertaken in the 1970s found no evidence of a "socialization-pushed movement" for integration, 3 9 and a case study of German attitudes showed that EC affairs seemed "grossly boring" to most citizens. 4 0 In the wake of the no-vote in Denmark and the near-no-vote in France on the Maastricht Treaty, these studies stand as a reminder that it may be time for integration theorists to direct some research once again, Deutsch-style, to the issue of the extent to which Euro-progress will necessitate the cultivation of a community with genuine "we-feeling," and to the issue of how the Community and member states have or have not successfully engaged in furthering such a spirit. 4 1

The New Wave: Current Theory and Research While the "doldrums era" thus produced more significant contributions than many (especially U.S.) scholars have recognized, it is true that it was marked by a decline of interest in both the EC and integration theory. As Figure 2.2 illustrates, the 1992 program and the agreement on the SEA in the mid-1980s generated a marked reversal of this trend. In this section, we will focus on four developments from the new wave of theory and research: (1) the reemergence of grand theory; (2) attention to the microfoundations of regional integration; (3) focus on institutions and multilevel governance; and (4) recognition of the increasing importance of the European Court of Justice and law for the integration process. Reemergence

of Grand

Theory

Without diminishing the importance of the prior empirical and theoretical efforts, the post-1987 period reintroduced theoretical debates on a grand scale. As the real-world integration enterprise in Europe regained momentum, academics rediscovered Haas' neofunctionalism and rekindled old debates about the process of integration. Functionalist and neofunctionalist concerns with the importance of civil society, socioeconomic interests, and

The Ell and Regional Integration

Theory

43

technocracy are carried forward with different emphases by Sandholtz and Zysman, Burley and Mattli, and Green. In addition, some of realism's theoretical concerns—state interests, power, and bargaining—are reflected in the work of Moravcsik (an intergovernmental institutionalist) and Grieco, a realist. 4 2 Other scholars open new lines of inquiry: Sbragia develops a model of federalism in novel ways; Marks speaks of an emerging system of multilevel governance; and Majone elaborates a model of the regulatory state. To a striking extent, however, the new debates parallel the old. Realism and neofunctionalism remain important and provide analytic categories that still organize much of the research. Neofunctionalism remains for the most part a theory of the transformative potential of civil society. It provides a way of understanding how opportunities for private gain can be surreptitiously exploited in alliance with supportive international institutions in such a way as to influence integrative outcomes. Realism, however, has subtly changed the terms of discourse. It no longer emphasizes, as it once did, anarchy, sovereignty, conflict, and power politics. Contemporary versions of realism focus more on domestic interests as they are influenced and defined by domestic political processes rather than by the purely positional derivation of interest suggested by classical realism. Starting with domestic interests so defined, realists draw out the consequences for bargaining behavior and policy outcomes. In the United States, Sandholtz and Zysman got the debate off the ground with their article "1992: Recasting the European Bargain." They argued that changes in international economic structure "triggered the 1992 process." 4 3 These changes had to do principally with the economic successes of Japan and the newly industrializing countries and with the relative economic decline of the United States. However, Sandholtz and Zysman do not provide an explanation that is purely structural. They acknowledge the importance of the Commission, which, along with business groups, "was able to mobilize a coalition of government elites that favored the overall objective of market unification." 4 4 Sandholtz and Zysman point to the importance of many factors, including domestic politics, international institutions, and changes in the international economy. But one comes away feeling that changes in the international economy are crucial, not in the sense that they determine outcomes (they do not), but in the sense that they provide the source of changing opportunities and constraints. Once new opportunities and constraints are in place, actors at the domestic level are pressured to find appropriate responses. While variation in domestic preferences and institutions may lead to different outcomes, the overall pattern of explanation is systemic. Andrew Moravcsik's "Negotiating the Single European Act" is an important response to the Sandholtz and Zysman piece. Setting his sights on

44

Reflections on European

Integration

the negotiations leading up to the SEA, Moravcsik challenges the view "that institutional reform resulted from an elite alliance between EC officials and pan-European interest groups." 4 5 Instead, he presents a model based on intergovernmentalism, lowest common denominator bargaining, and strict limits on sovereignty transfer. 4 6 This model mirrors the central claims of realism 4 7 with the important exception that interests are not structurally derived, i.e., position in the international power distribution does not determine interests. For this, Moravcsik argues, we must turn to domestic politics. 48 By referring to domestic politics as the source of state interests, he provides an explanation partly based on second-image views of international politics. Moravcsik's story is partly told in terms of domestic actors and domestic politics. Critical events include the advent to power of conservative governments in Britain and Germany, the French turnaround, the clash between different wings of the socialist party in France, and the failure of Keynesian policies. The convergence of interests among leaders of France, Germany, and Britain is an important precursor of agreement on the Single European Act. A central component of Moravcsik's argument is that interests are treated as exogenous. These interests are described, and he relates how the preferences of leaders of France, Germany, and Great Britain converged. Our point is simply that these preferences are themselves theoretically unexplained. Once interests are given, the negotiating history is supportive of the intergovernmental model. But taking interests as given is a powerful move that allows realism (modified to be sure) to claim much theoretical ground. We should recall that neither classical realism nor neorealism says much about state interests beyond the ways in which changes in systemic structure imply changes in perceptions of security. Moravcsik's argument is persuasive to the extent that he can show that state leaders not only made the final decision, but that they made them independently of the causal influences that are captured by neofunctional theory. If neofunctionalism can explain state interests, through a logic of spillover or elite socialization, the intergovernmental model loses much of its theoretical bite. The response to the intergovernmental paradigm can logically follow two paths. First, one can return to neofunctionalism's preoccupation with economic forces and international actors and argue that the full extent of their influence has not been appreciated. Second, one can try to provide an account, consistent with neofunctionalism, in which preferences are determined. Maria Green's paper, "Setting the Agenda for A New Europe: The Politics of Big Business in EC 1992," strives to do the former. 49 Wayne Sandholtz's "Choosing Union: Monetary Politics and Maastricht" 50 and David Cameron's "British Exit, German Voice, French Loyalty: Defection, Domination, and Cooperation in the 1992-93 ERM Crisis" 51 try to do the latter.

The EU and Regional Integration Theory

45

Maria Green provides an argument about the importance of business groups and multinational corporations. She intensively examines the activities of the European Roundtable of Industrialists ( E R T ) and finds an important role in structuring the E C agenda. She does not argue that the E R T determines the outcome, only that key business leaders helped to formulate the issues and shape the options. Thus, by the time state leaders came to the bargaining table, a substantial amount of work had been accomplished. Green reasserts the importance of neofunctionalism by arguing that the negotiation process responded to options shaped by economic forces and economic actors. In "Choosing U n i o n , " Sandholtz confronts the difficult task of explaining state preferences. On the one hand, systemic perspectives claim too much by arguing that changes in the international distribution of power and wealth force specific domestic responses. Unless systemic constraints are tied to "single-exit outcomes," (the legendary fire in the theater with one exit), the systemic level is necessarily incomplete. On the other hand, sometimes domestic politics approaches claim too little, providing a locus within which potentially important domestic factors transpire but substituting descriptive claims for theory-based explanation. Sandholtz critiques the intergovernmental model for implying that preferences are formed by some "hermetic process" and then transported to Brussels. He agrees with Moravcsik that "Community decisions are bargains that reflect state interests, but those interests are shaped in part by membership in the E C . " Sandholtz's analysis of the effects of the E C on state interests provides a theoretical bridge between international institutions and domestic preferences. B y doing so, he succeeds in recovering some of the ground that neofunctionalism lost to realism. Sandholtz argues that, in trying to understand the choice of monetary union, spillover may be a necessary but not sufficient part of the explanation, e.g., spillover from the 1992 project to Maastricht. However, an additional part of the explanation centers on within-EMS politics. One element of preference formation simply has to do with constraints. As countries integrate into E M S , the costs of exiting increase. Also, France, Belgium, the Netherlands, and Italy felt that they had more leverage inside a new, fully integrated monetary union than in a Bundesbank-dominated EMS. David Cameron's recent paper on the 1 9 9 2 - 1 9 9 3 E R M crisis takes further Sandholtz's stress on the importance of E C membership for state preference formation. Cameron argues that a structural realist approach to this case "underpredicts" the outcome while a domestic politics approach "seriously mispredicts" the actual outcome, because it "leads one to expect that it would have been Britain, not France, that would remain in the E R M and France, not Britain, that would e x i t . " 5 2 For Cameron, the "British Exit, French L o y a l t y " outcome is explicable—and even predictable—if

46

Reflections on European

Integration

and only if one takes into account the impact on preference formation of the ERM regime "viewed as a polity." In his account, France's position is attributable to the fact that the policy preferences of its officials were "transformed" (i.e., pushed toward convergence with German preferences) by long interaction within the ERM, and Franco-German cooperation was further facilitated by the "internalization of norms." Lacking such a "history of interaction, shared norms, convergent preferences and cooperation," Cameron argues, British officials were almost inevitably led to the "exit" option. 53 Compelling though the Sandholtz and Cameron arguments on preference formation may seem, it must be noted that they are based largely on inference rather than hard evidence. It is important and illuminating that such views are being presented now with increasing frequency, but clearly they need to be refined and probed through further research. How can we identify compelling evidence for the "shared norms" hypothesis? We should be able to identify such normative patterns independently of the length of interaction (an imperfect proxy at best). How can the normative hypothesis become more persuasive than one arguing that monetary stability was crucial for France or the hypothesis that France wanted to cement Germany (especially a unified Germany) to the EU? If we step back from the debates, two points stand out as crucial for understanding the process of European integration. The first point is that the interests of governments, specifically the interests of the executive branch, as well as power and bargaining among states, remain important. This should not be surprising. The traditional concerns of domestic and international politics—power, interest, and conflict resolution—do not melt away in the rosy environment of Brussels, Strasbourg, and Luxembourg. The second point is that the interests of governmental leaders are not inherent, as if they were the product of fixed national characters or derived from slow changes in the international distribution of power. Quite the contrary, these interests are constantly in the process of being "produced." Theories of learning, of acquisition of norms, and of socialization stress how preferences (and the learning of norms and rules) come into being. Theories of bargaining and strategic interaction stress how states act on the basis of these preferences. While we cannot develop this point here, the synthesis of preferences and power-bargaining is the central concern of a new (1993) article by Moravcsik, "Preferences and Power in the European Community: A Liberal Intergovernmentalist Approach." Moravcsik's dual emphasis "seeks to account for major decisions in the history of the EC by positing a two-stage approach. In the first stage, national preferences are primarily determined by the constraints and opportunities imposed by economic interdependence. In the second stage, the outcomes of intergovernmental negotiations are determined by the relative bargaining power of

47

The EU and Regional Integration Theory

governments and the functional incentives for institutionalization created by high transaction costs and the desire to control domestic agendas." 54 While the new wave of grand theory may not be so "grand" as the earlier wave—few are talking about eclipsing the nation-state or creating an autonomous supranational organization—it has focused on the importance of state and non-state actors, the influence of international institutions, and the role of distinctive domestic processes in animating national political agendas. In addition, Sandholtz's concern with the formation of preferences is not unconnected to the issue of identity politics. 55 Not everyone greets debates in grand theory with enthusiasm. At their worst, they tend to be overly abstract and detached from both middle-level propositions and empirical evidence. One danger of grand theory is empirical remoteness and involution, i.e., theory turned in upon itself and concerned only with internal matters of consistency and deduction. Sophisticated grand theory is of course general, but it serves to consolidate isolated empirical generalizations and provides vertical integration between abstract concepts and empirical observations. Grand theory, in this sense, provides an encompassing structure to help us make sense of "findings" and identifies gaps and inconsistencies in the collective research enterprise. Microfoundations:

Rational Choice

Theory

Both realism and neofunctionalism are macrostructural theories. They deal with economic, social, and political aggregates. For realism, the first-order facts—anarchy and the international distribution of power—are considered macro. Interests and bargaining serve as second-order variables and are closer to the micro level. Neofunctionalism deals with background variables such as regime-type, social complexity, bureaucratic expertise, and process-level variables such as size and scope of transactions and spillover. While the important facts are different for each theory, they are both structural. Both provide only a weak account of agency. 56 This is a problem if there is no straight line from structure to outcome. Collective action problems provide good examples of gaps between individual preferences and collective outcomes. Since integration implies coordinated action among several states to overcome social dilemmas, a sound microtheory might be deemed relevant. Given the importance of rational choice theory within the social sciences today, there is a tendency to equate microtheory with rational choice. But there are numerous possible ways to supply microfoundations, only one of which is rational choice theory. Even the choice theoretic umbrella covers not only rational choice but also psychological models of choice such as prospect theory 57 and models of human problem solving and decisionmaking. 58 In addition, there are social psychological theories

48

Reflections on European

Integration

stressing the learning of norms and roles. 5 9 However, it is the rational choice approach, emphasizing self-interest and constrained maximization, that has informed most of the work on microfoundations. There are numerous people working in the rational choice approach to integration, broadly defined. 6 0 For illustrative purposes, let us comment briefly on Garrett and Weingast's "Ideas, Interests, and Institutions: Constructing the E C ' s Internal Market" (1991). Garrett and Weingast take institutional outcomes, in particular the organizing principle of the internal market, as the dependent variable. Very simply, they treat these outcomes as results of rational instrumental calculations. In much the same way that policies are chosen, so too are institutions. But there is immediately a problem, quickly recognized by the authors—there is no unique institutional solution to a problem facing an integrative union. As the authors put it, "There are many stable paths to cooperation that cannot readily be differentiated in terms of their consequences for aggregate welfare." 6 1 What is at stake is neither rationality nor efficiency; these are assumed. At issue is how choices are made from a number of efficient alternatives. Accident, history (path-dependence), clear precedents, and force (though this violates efficiency) might help to decide the outcome. The factor singled out by Garrett and Weingast is ideas, i.e., the influence of constructed focal solutions centering on some prominent idea, in this case the principle of mutual recognition. The basic logic of the approach is functional. The European Court of Justice's decision in the Cassis de Dijon case is " g o o d " for the member states. The set of principles decided by the E C J are those (or are among those) that states might decide for themselves. While rational choice theory opens up some interesting lines of inquiry, there are problems of integrating this line of research into the acquis of integration theory. The major reason is that the terms of debate are shifted. Rational choice theory provides analytic categories that are not only different from, but also incommensurate with, neofunctionalism and realism. Indeed, rational choice does not offer a theory so much as an elaboration of the proximal mechanisms through which other theories work. The rational choice approach allows us to see more clearly the incentives implied by changes in international power structures and socioeconomic conditions. It is not so much an alternative to neorealism and neofunctionalism as a specification of the intermediate mechanisms through which these theories work. This does not give us much leverage in terms of adjudicating the major debates of regional integration theory.

Political Institutions

and Multilevel

Governance

While the study of institutions is a recent addition to mainstream economic theory, it has been around for a long time in political science. Yet, ironically,

The EU and Regional Integration Theory

49

the study of institutions had been pushed into the background during the postwar period and is only recently being revived. 6 2 On the domestic front, pluralism encouraged a focus on the activities of social and economic groups, particularly as they became organized as interest groups. The job of the political analyst was to expose the economic and social foundations of political institutions and beliefs. 6 3 In the international arena, realism similarly downgraded the importance of institutional phenomena by emphasizing the primacy of power. Institutions per se had little autonomous importance. Neofunctionalism, for its part, allowed a distinctive place for institutions but treated them as dependent variables (form follows function) that varied along an intergovernmental-supranational continuum. Furthermore, neofunctionalism offered explanations of institutional outcomes on the basis of an explanatory structure that was itself lacking in institutions. This is theoretically incomplete. The E C is an institution, or set of institutions, as well as a market and free trade area. It is a complex of rules, procedures, norms, and common understandings that constitutes its members, structures their interactions, and mediates between power and interests to influence policy outcomes. Despite the seeming importance of the E C ' s institutional components, with f e w exceptions institutions have played a scant role theoretically in accounts of European integration. 64 In broad terms, there are three ways that institutions could be important for European integration. First, there is simply conceptualizing the E C in institutional terms. This is a pretheoretical yet conceptual task. Such work might include identifying the type of political system of which the E C is an instance (intergovernmental organization, regulatory regime, or concordance system), which institutions perform legislative, executive, and judicial functions, the democratic versus nondemocratic character of E C institutions, and so on. Second, work is needed in responding to the question of how institutions make a difference. How will the shift to majority voting change policy outcomes? How about the new cooperation procedure? Much has been made of the "democratic deficit." How would increasing the power of the European Parliament alter the decisionmaking procedures in the E C ? 6 5 Third, how do institutions themselves change? What are the factors that are responsible for the growth, decline, differentiation, and change along other dimensions of the institutional structures of the E C ? This question is the most difficult of the three. It asks us to make institutions themselves the object of explanation, i.e., to develop a theory of institutions. In the sections below we comment briefly on points one and two. To our knowledge, there is almost no work on theories of institutional change in the E C . How should we conceptualize the overall institutional structure of the E C ? Many are not satisfied with treating institutional outcomes as somewhere on the intergovernmental-supranational continuum. 66 It is increasingly

50

Reflections on European

Integration

clear to many scholars that models useful for understanding the formation of nation states may not be useful for understanding how states organize their affairs between one another. In short, EC developments cannot always be shoehorned into our preconceived categories. Dissatisfaction with the current state of affairs does not mean that there is agreement on alternatives. In the remainder of this section, we explore two conceptions of multilevel governance. In his article, "Structural Policy and Multilevel Governance in the EC," Gary Marks raises a big question: "What kind of order is emerging in Europe?" 6 7 Echoing sentiments expressed by Puchala years ago, Marks argues that the debate over EC institutions has been misguided because it is based on a conception of the relative power of the EC compared to that of the Community's member states. There is a fixed pie (of power? of performance?), and EC institutions compete for scarce resources at the expense of other institutions. In his examination of structural policy, Marks comes to this conclusion: Structural p o l i c y in the EC d o e s not fit along a continuum running from continued nation state predominance to the e m e r g e n c e of a Euro-state. Instead, it appears to be a t w o - s i d e d process, i n v o l v i n g decentralization of d e c i s i o n - m a k i n g to subnational l e v e l s of g o v e r n m e n t as well as centralization of n e w p o w e r s at the supranational level. If w e e n c o m p a s s the experience of structural p o l i c y in our notion of the future European policy, it can be v i e w e d as the leading e d g e of a system of multilevel governance in w h i c h supranational, national, regional, and local g o v e r n m e n t s are enm e s h e d in territorially overarching p o l i c y n e t w o r k s . 6 8

For Marks, then, institutional changes should not be viewed as taking place on a continuum from decentralization to centralization. Instead, there is a simultaneous devolution and centralization. Moreover, compared with the situation Puchala described almost two decades ago, the current system features both more empowered Commission officials—armed with a larger budget and room for autonomy in the implementation process—and an unprecedented "mobilization and empowerment of subnational governments." 69 The focus of Fritz Scharpf is quite different. In "The Joint Decision Trap," he explores the pathologies of decisionmaking that emerge when policies are jointly decided at two levels, in this case, presumably the Council of Ministers and member governments. Scharpf sees problems when two conditions exist: (1) when central governments depend on the agreement of constituent units, and (2) when this agreement must be unanimous. 7 0 The problem with this institutional structure has to do with the conflict between what is optimal with respect to policy and with respect to institutional interests per se. Regarding policy, there are often political economies of scale associated with larger units of government. As the problems dealt with assume broader scope—because of changes in division

The EU and Regional Integration

Theory

51

of labor, interdependence, and externalities—the case for more comprehensive policymaking is strengthened. However, these more comprehensive arrangements do not necessarily (or even normally) apply to institutional interests per se, especially to those occupying institutional roles. As a result, Scharpf argues that Politikverflechtung (joint decisionmaking) produces substantive deficiencies in outcomes that are not explainable by the goals of the actors, their resources, or the behavior of interest groups. The problem is with political technology—in short, with institutions. S c h a r p f ' s analysis may be less salient today when the requirement of unanimity has been somewhat relaxed. However, his central point is relevant. The interplay of private and public forces, of interest groups, transnational corporations, and nation states will not ensure optimal outcomes (optimal does not mean perfect, just the best given resources and preferences). These factors tell only part of the story. It is still necessary to pay attention to institutional design. The EC} and Integration

Theory

The dramatic emergence of the ECJ f r o m its previous obscurity has generated one of the more interesting debates within the new wave literature. The central question in this debate is the following: How can one explain the extraordinary progress of legal integration, i.e., the fact that, of all the C o m m u n i t y institutions, the Court has been willing and able to go "farthest in limiting national autonomy." 7 1 Within the literature produced since Stein and Weiler's seminal articles, the most provocative and contradictory positions on this debate have been staked out by Garrett and by Burley and Mattli. Garrett's rational choice approach explains the growth in the power of the ECJ by focusing, not on the Court, but on the preferences of the member states. His basic argument is that the member states have enabled and encouraged the ECJ to limit national autonomy in certain respects because doing so serves, in general terms, their collective interests. Only by delegating substantial power to the ECJ can m e m b e r states be certain that compliance with EC regulations will be effectively monitored and that the "incomplete contracting p r o b l e m " will be overcome through the Court's elaboration of vaguely defined rules. 7 2 At this point in his argument, Garrett adds two important caveats. First, while member states have been willing to allow the ECJ to constrain their autonomy, they have done so at least in part because of the limits to the Court's power and autonomy. Various safeguards serve to curb any threat potentially posed by the Court, he argues; the ECJ must be " f e a r f u l " that an excessively intrusive ruling could yield a challenge to its authority through noncompliance, a more politicized (antiactivist) policy of appointment of justices by member states, or even a treaty revision diminishing the

52

Reflections on European

Integration

Court's powers or jurisdiction. Second, the most powerful EC states have the least to fear from the ECJ, because "the principles governing decisions of the European c o u r t . . . are consistent with the preferences of France and Germany." Certainly on matters related to the internal market, he contends, "the EC legal system serves the purposes of the French and Germans extremely well." 7 3 Burley and Mattli sharply criticize certain elements of Garrett's statecentric theory of EC. First, they argue that Garrett "is simply wrong" in claiming that the ECJ's rulings "are consistent with the preferences of France and Germany." As evidence for this contention, they note that both Germany and France have "argued explicitly and strongly against the Court's ultimate position" in a good number of cases. 7 4 It is true that Garrett does not directly address this issue at all; indeed, he discusses very few cases, and his account of the crucial Cassis de Dijon case neglects to acknowledge that it produced a ruling staunchly opposed by Germany. 7 5 Garrett merits criticism on this point, quite clearly; as will be discussed below, however, this does not mean that his general line of reasoning regarding the limits to Court autonomy should be ignored. Second, Burley and Mattli argue that "there is absolutely no evidence that the Court actually attempts . . . to track the positions of the member states." 76 It is true that Garrett merely infers this and provides no concrete evidence to support his position. However, Burley and Mattli seemingly undermine their own critique in this regard by noting in their piece that "the ECJ uses the EC Commission as a political bellwether," taking the Commission's "position as an indicator of political acceptability to the member states." 77 Surely the Commission "tracks" member state interests in a sense, and although the Commission argues against the positions of states in particular cases, 78 one would assume that the Commission makes calculated assumptions about—in Burley and Mattli's words—"how far the member states can be pushed toward the Court and the Commission's vision of maximum integration." 79 Third, Burley and Mattli note that the "Maastricht Treaty on European Union reflects a determination on the part of the member states to limit the ECJ." Not only is the Court "entirely excluded from two of the three 'pillars' of the treaty: foreign and security policy and cooperation in the spheres of justice and home affairs," but also "a number of specific articles are very tightly drafted to prevent judicial manipulation." At first glance, they concede, these Maastricht provisions "appear to confirm the GarrettWeingast theory" regarding the limits on the autonomy of the Court. However, they argue, the Maastricht provisions actually undermine the Garrett thesis: "if indeed the Court ensures the protection of its authority and legitimacy by assiduous fidelity to state interests rather than to law, then why worry?" That is, why should the member states worry enough about potential ECJ action to exclude it from the new pillars? 8 0 The response

The EU and Regional Integration Theory

53

from a neorationalist perspective seems obvious: while states have been willing to risk potential negative rulings by the Court on particular cases in exchange for the collective gains provided through the Court's role in the economic sphere, they have not viewed the risk-gain equation as equally acceptable in the "high politics" areas covered by the Maastricht pillars. A s Corbett has written of the Maastricht drafting process, the E C J and other "federal characteristics" of the Community "were sufficiently important to make some Member States wary about giving the Community full competence in the sensitive areas of foreign policy and, especially, security." 8 1 Burley and Mattli's critique of Garrett's thesis thus seems problematic in some respects, but they do present a fascinating countertheory. Whereas Garrett explains the broad political limits to the E C J ' s power, Burley and Mattli explain how and why the E C J has been willing and able to test and incrementally stretch those limits over the past thirty years. They categorize their theory as neofunctionalist because "the drivers" of the process of legal integration, like those of economic integration in the Haas scheme, "are supranational and subnational actors pursuing their own self-interests within a politically insulated sphere." 8 2 A s they document, the E C J has worked to develop close relations with the member states' lower courts as well as community law professors, and this coalition of legal community actors has steadily advanced the process of legal integration in a manner that has entailed "reciprocal empowerment." 8 3 The E C J has had "the power to pursue its own agenda" with considerable autonomy mainly because the technical and seemingly nonpolitical nature of its legal behavior has enabled it, in neofunctionalist fashion, to be "shielded from the interplay of direct political interests." 8 4 With law functioning "both as mask and shield," Burley and Mattli contend, the "pursuit of individual interests" by the European justices has allowed for the extension of the E C J ' s jurisdiction through a dynamic reminiscent of functional spillover and resulted in "the accretion of power" on the part of the Court. 85 Proof of the Court's power vis-à-vis member states, they argue, may be found in the fact that "different governments strongly disagree" with particular rulings of the court, but, over time, "they tend to accept the Court's position and regard the path chosen as inevitable." Neorationalism à la Garrett, they insist, "is at a loss" to explain such an outcome. 86 The Burley-Mattli portrayal of E C J development is compelling in certain respects. It provides an important corrective to Garrett's depiction of a " f e a r f u l " Court more conscious of its limits than its potential reach; on the basis of Garrett's account alone, one would have a hard time understanding the recent appearance of works depicting the E C J as extremely or even excessively activist. 87 It also gives a more politically realistic explanation for the E C J ' s steady integrative progress than one finds in traditional "legalist" studies. 88 However, on the crucial question of the Court's

54

Reflections on European

Integration

power and autonomy, it is hard to accept fully the Burley-Mattli assessment for two reasons. First, neorationalist perspectives are hardly "at a loss" to explain the grudging tolerance member states have frequently manifested for negative rulings by the Court. As noted in the Garrett section, such cases may seem acceptable in light of the collective benefits secured through Court action. Moreover, however "ferociously" certain states may have criticized some decisions, 8 9 the practical losses inflicted by those decisions should not be exaggerated. The "sheep meat cases," for example, could hardly have been expected to generate a judicial-political confrontation on a par with that in France when the ECJ's domestic counterpart, the Constitutional Council, invalidated the Mitterrand government's nationalization law. 9 0 During the 1960s and 1970s the Court often "spoke the brave language of teleological interpretation," but its legal doctrine was "relatively modest and set few limits, either substantive or procedural, on the organs of national . . . government." 9 1 Even the f a m o u s Cassis de Dijon case, often depicted as an exemplar of the Court's heroic assertiveness, has been shown to have gained its reputation largely through a Commission communication that exaggerated the boldness of the decision. 9 2 An additional crucial point regarding the states' tolerance of negative Court rulings has been developed at length by Alter and Meunier-Aitsahalia. Even the relatively bold rulings of the ECJ do not normally impose final policy outcomes; instead, they simply "succeed in jarring the political process" in a fashion that ultimately yields compromises "reflecting the concerns of mobilized interest groups and of the different member states." In the Cassis case, for example, the member states used their power in the Council to dull the effects of the ECJ ruling and later "undermined the jurisprudence of the Court" by attaching a restrictive Article 100b to the Single European Act. 9 3 Second, one must ask just how effective the Court's legal "mask and shield" have been. Burley and Mattli portray the ECJ as effectively "preserving its ability to c a m o u f l a g e controversial political decisions," but they would appear to be at a loss to explain such developments as the member states' restriction of the Court's jurisdiction in the SEA, exclusion of the Court f r o m two of Maastricht's three pillars, and consideration of a proposal to restrict lower-level courts at the national level from forwarding to the ECJ preliminary ruling questions. 9 4

Conclusion As the recent literature on issues ranging f r o m the ECJ to the Maastricht Treaty makes clear, neofunctionalism has proven to be a far more resilient theoretical approach to integration than most commentators assumed two decades ago. The new wave analysts no longer see neofunctionalism as a simple theoretical key that can unlock the secrets of Europe's future, and

The EU and Regional Integration

Theory

55

they no longer share the assumption that integration is a zero-sum game destined to be won by the forces of supranationalism at the expense of antiquated nation states. Nonetheless, many of them do stress, quite cogently, that the process of integration cannot be properly understood without attention to the dynamic on which Haas focused in the 1950s. As Sandholtz noted in a recent paper, for example, spillover may no longer be viewed as automatic or unidirectional, but as the density of common policies in the EU has become greater, the likelihood of substantive issue linkages (the "driving force behind spillovers") occurring has become greater as well. 95 By the same token, as analysts such as Burley and Mattli, Marks, and Dehousse and Majone have argued, it makes less sense now than it did during the heyday of neofunctionalism to portray the dynamics of integration solely in terms of member state bargaining. Euro-institutions such as the ECJ and the Commission are now far more influential actors or "policy entrepreneurs" than they were several decades ago. 9 6 As the intergovernmentalists remind us, however, Haas was hardly wrong to deduce from the jolts of the Gaullist era that one had to back off from expansive claims for neofunctionalism. From the ECSC to the EC to the EU, the machinery of "Europe" has assured that national governments continue to play a decisive role in the integration process; this role is no doubt most decisive, and most visible, in the case of the biggest leaps of integration such as those embodied in the SEA and Maastricht. 97 Much of the most interesting literature of the new wave consists of efforts to trace the interactions between the states and the nonstate actors, to underscore the "mutual influence of the Commission and the member states," and to untangle the increasingly complex and sometimes impenetrable "power play" that shapes EU outcomes. 9 8 Looking toward the future of integration theory, two points should be made. First, there is a need for a better marriage of theoretical and empirical work on the European Union. Even if one is sympathetic to the U.S. argument that European scholarship on the EU has often been too atheoretical, too descriptive, and too focused on technical details within narrow issue areas, the opposite lament should not be ignored. Our European colleagues have been quite right to stress that much of the theory-driven U.S. literature has seemed wrongheaded precisely because it has been oblivious to what is now an extraordinarily rich accumulation of information about the complex EU machinery. Building parsimonious theories requires simplification and precludes attention to many details; on the other hand, such theories cannot be made credible to technical experts if they are supported by flimsy evidence or if the insistence on parsimony appears to be an excuse for ignoring data that inconveniently contradict one's assumptions. Second, it is important to acknowledge that longer the only theoretical game in town for those behavior in the European Union. In the era of started his work, or the formative era of the EC,

integration theory is no concerned with political the ECSC, when Haas when neofunctionalism

56

Reflections on European

Integration

seemed most compelling, "Europe" represented a unique experiment in international relations and what seemed to justify theoretical attention was the question of how its "would-be polity" might develop at the expense of the nation-state. As an increasing number of scholars have argued of late, however, the very success of this venture seems to require a shift of theoretical focus. It is not that integration theory is outmoded; it will continue to have an important place as long as EU actors continue to struggle over institutional development and underlying authority-legitimacy transfers. However, alongside integration theory a new body of theoretical literature is developing that deals with what might be termed the "normal" politics of the EU. Sbragia, for example, has argued that a comparative federalism framework should be applied to the political games being played within the EU's current structure. Along similar lines, Hix has encouraged comparativists to apply theoretical insights (those related to consociational democracy as well as federalism, among others) to the EU's complex "internal political arena." Majone has shown how useful it can be to examine the EU's regulatory policy from a comparative perspective with the United States as the principal contrast model. In a new edited volume, Bueno de Mesquita and Stokman have employed an expected utility model and an exchange or logrolling model to explain and predict the decisions of the EU's key institution, the Council of Ministers. From a different angle, Sandholtz has argued that bargaining within the EU might be better understood if viewed comparatively with theories developed to explain multilateral games in other international institutions such as the GATT, the IMF, and even the U.N. Security C o u n c i l . " The above suggestions, particularly those of Sbragia and Hix, reflect an important development. The EU is no longer an incipient organization struggling to come into existence. It was natural for earlier studies to define the central explanatory task as one of accounting for the emergence and development of the Community's institutional structure. But scholars increasingly assume that some institutional structure is in place and examine what goes on inside these structures. Politics and policymaking within institutions have assumed an analytical place alongside the politics of institutional change. For the future, a wide variety of theoretical options are open. Given the number of talented scholars attracted to the EU field by the political developments of the last decade, it seems safe to assume that one can expect a rich outpouring of theoretical literature through the remainder of the 1990s.

Notes This chapter w a s originally prepared for delivery as a plenary address at the Third Biennial International Conference of the European Community Studies Association,

The EU and Regional Integration

Theory

57

2 7 - 2 9 May 1993, in Washington, D.C. We would like to thank Mark Schneider and Anne Teissier for their assistance with the research and Stephanie MacLachlan for her editorial assistance. Willem Dicke of the European Community's Information Section in Washington, D.C. also merits our thanks for providing data unavailable elsewhere. We owe special thanks to the following individuals for providing valuable comments and criticisms on previous drafts of the manuscript: Karen Alter, Simon Bulmer, Christine Ingebritsen, Peter Katzenstein, Sophie Meunier-Aitsahalia, Andrew Moravcsik, Wayne Sandholtz, Bengt Sundelius, Anne-Marie Slaughter, and Alec Stone. 1. Karl Deutsch, Political Community in the North Atlantic Area (Princeton: Princeton University Press, 1957); and Karl Deutsch et al., "Political Community and the North Atlantic Area," International Political Communities: An Anthology (New York: Doubleday and Co., 1966). 2. Just before sending this chapter off for publication, we learned of a conference held at the University of Wisconsin-Madison on the topic of Pluralistic Security Communities: Past, Present and Future, 30 April-1 May 1994. The conference was organized by Emanuel Adler and Michael Barnett. One broad goal of the conference was to reassess the role of Deutschian analysis of security communities in the contemporary world. See their paper, "Pluralistic Security Communities: Past, Present and Future," (unpublished, University of Wisconsin-Madison, 1994). 3. David Mitrany, A Working Peace System (Chicago: Quadrangle Books, 1966). 4. David Mitrany, "The Prospect of Integration: Federal or Functional?" in Joseph Nye, ed., International Regionalism: Readings (Boston: Little, Brown, 1968), 58-61. 5. There are many influential essays in this volume, the most important of which is that by Ernst Haas, "Technocracy, Pluralism and the New Europe," in Stephen Graubard, ed., A New Europe? (Boston: Houghton Mifflin, 1964), 6 2 - 8 8 . 6. Ernst Haas, "The Study of Regional Theorizing: Reflections on the Joy and Anguish of Pretheorizing," International Organization 24 (1970): 610. 7. Ibid., 33. 8. Leon Lindberg and Stuart Scheingold, Europe's Would-Be Polity: Patterns of Change in the European Community (Englewood Cliffs: Prentice-Hall, 1970), 141-143. 9. See Philippe Schmitter, "A Revised Theory of Regional Integration," International Organization 24 (1970): 836-868. 10. Ernst Haas, The Uniting of Europe, 2d ed. (Stanford: Stanford University Press, 1968), xxiii. 11. Stanley Hoffmann, "Obstinate or Obsolete? The Fate of the Nation-State and the Case of Western Europe," Daedalus 95 (1966): 3. 12. Of course, this need not be so. The classification of particular cases is a conceptual—not empirical—issue. The EC could be, and sometimes is, considered an instance of a more general phenomenon. For example, see Joseph Nye, "Comparing Common Markets: A Revised Neofunctional Model," International Organization 24 (1970): 796-835; Ernst Haas, "International Integration: The European and Universal Process," in M. Hodges, ed., European Integration (Harmondsworth: Penguin, 1972), 9 1 - 1 0 7 ; and Walter Mattli, "The Logic of Regional Integration: Europe and Beyond" (working paper, Program in International Politics, Economics, and Security, University of Chicago, Chicago, Illinois, 1993), 1 - 6 7 . The tricky part is that the EC is in some ways historically novel, its structures and processes different from anything found in the European Free Trade Association, the Arab League, the East African Common Market, or the Latin American Free Trade Association. Whether these differences represent variation in common variables or

58

Reflections on European Integration

differences in defining properties is the key issue. A nomothetic approach would opt for the former. 13. Ernst Haas, "Turbulent Fields and the Theory of Regional Integration," International Organization 30 (1976): 183. 14. Ernst Haas, " T h e Uniting of Europe and the Uniting of Latin America," Journal of Common Market Studies 5 (1967): 3 2 4 - 3 2 5 . 15. Haas, " T h e Uniting of Europe," 316; see also, Ernst Haas, The Obsolescence of Regional Integration Theory (Berkeley: Institute o f International Studies, 1975); Haas, "Turbulent Fields," 1 7 7 - 1 7 8 . 16. Robert Keohane and Stanley Hoffmann, "Institutional Change in Europe in the 1980s," in Robert Keohane and Stanley Hoffmann, eds., The New European Community: Decisionmaking and Institutional Change (Boulder: Westview Press, 1991), 8. 17. See Norma Heathcote, "The Crisis of European Supranationality," Journal of Common Market Studies 5 (1966): 171. 18. William Wallace, "Europe as a Confederation: The Community and the Nation-State," Journal of Common Market Studies 20 (1982): 63. 19. Carole Webb, "Theoretical Perspectives and Problems," in Helen Wallace, William Wallace, and Carole Webb, eds., Policy-Making in the European Community, 2d ed. (Chichester: Wiley, 1983), 2; Dale Smith and lames Ray, "European Integration: Gloomy Theory Versus Rosy Reality," in Dale Smith and James Ray, eds., The 1992 Project and the Future of Integration in Europe (Armonk, New York: M. E. Sharpe, 1993), 20ff; Alberta Sbragia, "Asymmetrical Integration in the European Community: The Single European Act and Institutional Development," in Dale Smith and James Ray, eds., The 1992 Project, 96; and Joseph Weiler, "Community, Member States and European Integration: Is the Law Relevant?" Journal of Common Market Studies 20 (1982): 3 9 - 5 5 . 20. Svein Andersen and Kjell Eliassen, "European Community Lobbying," European Journal of Political Research 2 0 (1991): 1 7 3 - 1 8 7 . 21. Roy Ginsberg, Foreign Policy Actions of the European Community: The Politics of Scale (Boulder: Lynne Rienner, 1989), 5 5 - 1 1 6 . 22. See Smith and Ray, "European Integration, 3 2 - 3 3 ; their data are updated to include data for 1 9 8 9 - 1 9 9 1 and the first ten of twelve issues for 1 9 9 2 - 1 9 9 4 . 23. W. Wallace, "Europe as a Confederation," 58. 24. Smith and Ray, "European Integration," 33. 25. Anne-Marie Burley and Walter Mattli, "Europe Before the Court: A Political Theory of Legal Integration," International Organization 47 (1993): 41. 26. See especially Leon Lindberg and Stuart Scheingold, Europe's Would-Be Polity (Englewood Cliffs: Prentice-Hall, 1970). 27. Haas, "Turbulent Fields," 2 0 3 - 2 0 4 . 28. William Wallace, "Less than a Federation, More than a Regime: The Community as a Political System," in Wallace, Wallace, and Webb, eds., Policy-Making in the European Community, 407; see also Helen Wallace, "Negotiation, Conflict and Compromise: The Elusive Pursuit of Common Policies," in Wallace, Wallace, and Webb, eds., Policy-Making in the European Community, 46. 29. W. Wallace, "Europe as a Confederation," 63. 30. Donald Puchala, "Domestic Politics and Regional Harmonization in the European Communities," World Politics 27 (July 1975): 5 1 0 - 5 1 5 ; see also Helen Wallace, "The Impact of the European Communities on National Policy-Making," in M. Hodges, ed., European Integration, 2 8 5 - 3 0 3 ; and Paul Taylor, "The Politics of the European Communities: The Confederal Phase," World Politics 28 (1975): 336-366.

The EU and Regional Integration

Theory

59

31. Webb, "Theoretical Perspectives and Problems," 4. 32. Donald Puchala, "Worm Cans and Worth Taxes: Fiscal Harmonization and the European Policy Process," in Wallace, Wallace, and Webb, eds., Policy-Making in the European Community, 243; W. Wallace, "Europe as a Confederation"; and Simon Bulmer, "Domestic Politics and European Community Policy-Making," Journal of Common Market Studies 21 (June 1983): 349-363. 33. Donald Puchala, "European Fiscal Harmonization: Politics During the Dutch Interlude," in Leon Hurwitz, ed., Contemporary Perspectives on European Integration (Westport, Conn.: Greenwood, 1980), 2 0 9 - 2 2 2 ; and Diana Welch, "From 'Euro Beer' to 'Newcastle Brown': A Review of European Community Action to Dismantle Divergent 'Food' Laws," Journal of Common Market Studies 22 (September 1983): 56. 34. Fritz Scharpf, "The Joint-Decision Trap: Lessons from German Federalism and European Integration," Public Administration 66 (1988): 251. 35. Werner Feld, "Two-Tier Policy Making in the EC: The Common Agricultural Policy," in Leon Hurwitz, ed., Contemporary Perspectives, 123-149; and Joan Pearce, "The Common Agricultural Policy: The Accumulation of Special Interests," in Wallace, Wallace, and Webb, eds., Policy-Making in the European Community, 143-175. 36. Eric Stein, "Lawyers, Judges, and the Making of a Transnational Constitution," American Journal of International Law 75 (1981): 1 - 2 7 ; Joseph Weiler, "The Community System: The Dual Character of Supranationalism," Yearbook of European Law 1 (1981); and Joseph Weiler, "Community, Member States and European Integration: Is the Law Relevant?" Journal of Common Market Studies 20 (1982): 39-55. 37. Stein, "Lawyers, Judges," 1. 38. Alberta Sbragia, "Thinking About the European Future: The Uses of Comparison," in Alberta Sbragia, ed., Euro-Politics: Institutions and Policymaking in the "New" European Community (Washington, D.C.: Brookings Institution, 1992), 262. 39. David Handley, "Public Opinion and European Integration: The Crisis of the 1970s," European Journal of Political Research 9 (1981): 335-362; see also Ronald Inglehart and Jacque Rene Rabier, "Europe Elects a Parliament: Cognitive Mobilization, Political Mobilization, and Pro-European Attitudes as Influences on Voter Turnout," in Leon Hurwitz, ed., Contemporary Perspectives, 27-51. 40. Elizabeth Noelle-Neumann, "Phantom Europe: Thirty Years of Survey Research on German Attitudes Toward European Integration," in Leon Hurwitz, ed., Contemporary Perspectives, 53-74. 41. See Richard Eichenberg and Russell Dalton, "Europeans and the European Community: The Dynamics of Public Support for European Integration," International Organization 47 (Autumn 1993): 507-534. 42. See Joseph M. Grieco, "State Interests and International Rule Trajectories: The Politics of European Economic and Monetary Union" (unpublished manuscript, Duke University, 1992), 1-64. 43. Wayne Sandholtz and John Zysman, "1992: Recasting the European Bargain," World Politics 42 (1989): 95. 44. Ibid., 96. 45. Andrew Moravcsik, "Negotiating the Single European Act," International Organization 45 (Winter 1991): 20. 46. Ibid., 25. 47. While we do not wish to belabor the labeling issue, we point out that most theoretical approaches are described by numerous factors, not all of which are

60

Reflections on European

Integration

found in a particular work. Realism can be described by its focus on states-as-actors, by anarchy, by state or national interests, by the preservation of sovereignty, by the highlighting of power as opposed to voluntary exchange, by bargaining style, and by the manner in which interests are derived from changes in international power structures. Our goal is less to assign a label to a person's work than to identify the key concepts used and their intellectual lineage. In our view, Moravcsik's approach owes much to realism. It differs in important ways too: Interests are not derived from power structure, domestic politics are important in determining preferences, and interstate relations are not described in purely conflictual terms. 48. Moravcsik, Negotiating the Single European Act, 27. 49. Maria Green, "Setting the Agenda for a New Europe: The Politics of Big Business in EC 1992" (working paper, study group on European integration and domestic policymaking, Center for European Studies, Harvard University, 1993). 50. Wayne Sandholtz, "Choosing Union: Monetary Politics and Maastricht," International Organization 47 (1993): 3. 51. David Cameron, "British Exit, German Voice, and French Loyalty: Defection, Domination, and Cooperation in the 1992-93 ERM Crisis" (working paper, Workshop on International Political Economy, Columbia University, March 1993). 52. Ibid., emphasis in original, 4. 53. Ibid., 4, 58, 63. 54. Andrew Moravcsik, "Preferences and Power in the European Community: A Liberal Intergovernmentalist Approach," Journal of Common Market Studies 31 (December 1993): 517. 55. Alexander Wendt, "Anarchy Is What States Make of It: The Social Construction of Power Politics," International Organization 46 (1992): 391-425; and Henry Nau, "Identity and International Politics: An Alternative to Neorealism" (working paper, Annual Meeting of the American Political Science Association, Washington, D.C., 1993). 56. We say "weak" rather than "absent" because both theories could claim an implied microfoundation. Realist theory requires a state representative to "execute" the changed incentives caused by changing international variables. Neofunctionalism was always concerned with the instrumental calculations of self-interested actors. See Ernst Haas, "The Study of Regional Integration: Reflections on the Joy and Anguish of Pretheorizing," International Organization 24 (1970): 627. 57. George Quattrone and Amos Tversky, "Contrasting Rational and Psychological Analyses of Political Choice," American Political Science Review 82 (1988): 719-736. 58. Herbert Simon, "Rationality in Psychology and Economics," in Robin Hogarth and Melvin Reder, eds., Rational Choice: The Contrast Between Economics and Psychology (Chicago: University of Chicago Press, 1987), 2 5 ^ - 0 . 59. Wendt, "Anarchy Is What States Make of It," 3 9 1 ^ 2 5 . 60. Geoffrey Garrett, "International Cooperation and Institutional Choice: The European Community's Internal Market," International Organization 46 (1992): 533-560; Geoffrey Garrett and Barry Weingast, "Ideas, Interests and Institutions" (working paper, NBER Conference on Political Economics, Washington, D.C., 1991); Peter Lange, "Maastricht and the Social Protocol: Why Did They Do it?" (working paper, Duke University, 1993); and Scharpf, "The Joint Decision Trap," 239-278. 61. Garrett and Weingast, "Ideas, Interests and Institutions," 3. 62. James March and Johan Olsen, "The New Institutionalism: Organizational Factors in Political Life," American Political Science Review 78 (1984): 734-749.

The EU and Regional Integration

Theory

61

63. Sheldon Wolin, Politics and Vision (Princeton: Princeton University Press, 1960), 287. 64. Scharpf, "The Joint Decision Trap"; Peter Ludlow, "The European Commission, in Keohane and Hoffmann, eds., The New European Community, 85-132; Sbragia, "Thinking About the European Future, 257-291; and Wolfgang Wessels, "The EC Council: The Community's Decisionmaking Center," in Keohane and Hoffmann, eds., The New European Community, 133-154. 65. Some progress has been made regarding the importance of the E U ' s changing institutional and decisionmaking structure. Gerald Schneider, "Getting Closer at Different Speeds: Strategic Interaction in Widening European Integration," in Pierre Allan and Christian Schmidt, eds., Qualitative Games in International Relations (Chiltenham, England: Edgar Elgar, forthcoming). See also Madelaine Hosli, "Admission of Free Trade Association States to the European Community: Effects of Voting Power in the European Community Council of Ministers," International Organization 47 (Autumn 1993), 629-643. 66. Ludlow, "The European Commission," in Keohane and H o f f m a n n , eds., The New European Community, 85-132; Gary Marks, "Structural Policy and Multilevel Governance in the European Community," in Alan Cafruny and Glenda Rosenthal, eds., The State of the European Community (Boulder: Lynne Rienner, 1993), 391-410; M. Huelshoff, "Domestic Politics and Dynamic Issue Linkage: A Reformulation of Regional Integration Theory," International Studies Quarterly 38 (1994); and Philippe Schmitter, "Interests, Powers and Functions: Emergent Properties and Unintended Consequences in the European Polity" (unpublished manuscript, Stanford University, Center for Advanced Study in the Behavioral Sciences, April 1992). 67. Marks, "Structural Policy," 391. 68. Ibid., 401-402. 69. Ibid., 407. 70. Scharpf, "The Joint Decision Trap," 254. 71. Keohane and Hoffmann, "Institutional Change in Europe," 11. 72. Garrett, "International Cooperation and Institutional Choice," 556-558; for a related argument, see Joseph Weiler, "A Quiet Revolution: The European Court of Justice and Its Interlocutors," Comparative Political Studies 26 (January 1994): 527. 73. Garrett, "International Cooperation and Institutional Choice," 5 5 6 - 5 5 9 ; see also Garrett and Weingast, "Ideas, Interests, and Institutions." 74. Burley and Mattli, "Europe Before the Court," 51. 75. See Karen Alter and Sophie Meunier-Aitsahalia, "Judicial Politics in the European Community: European Integration and the Pathbreaking Cassis de Dijon Decision," Comparative Political Studies 26 (January 1994): 535-561. 76. Burley and Mattli, "Europe before the Court," 51. 77. Ibid., 71; see also, 51. 78. See Stein, "Lawyers, Judges." 79. Burley and Mattli, "Europe Before the Court," 51, emphasis in original. 80. Ibid., 74. 81. Richard Corbett, "The Intergovernmental Conference on Political Union," Journal of Common Market Studies 30 (1992): 278. 82. Burley and Mattli, "Europe Before the Court," 43. 83. Ibid., 64. 84. Ibid., 75 and 57. 85. Ibid., 65, 72, and 73. 86. Ibid., 51.

62

Reflections on European

Integration

87. H. Rasmussen, On Law and Policy in the European Court of Justice (Boston: Nihoff, 1986); see also Mary Volcansek, "The European Court of Justice: Supranational Policy-Making," West European Politics (July 1992): 109-121. 88. See Burley and Mattli, Europe Before the Court, 4 5 - 4 6 . 89. Ibid., 51. 90. John Keeler and Alec Stone, "Judicial-Political Confrontation in Mitterrand's France," in George Ross, Stanley H o f f m a n n , and Sylvia Malzacher, eds., The Mitterrand Experiment (New York: Oxford University Press, 1987), 161-181. 91. Martin Shapiro, "The European Court of Justice," in Sbragia, ed., EuroPolitics, 127. 92. Alter and Meunier-Aitsahalia, Judical Politics in the European Community, 26, 535-561. 93. Ibid., 552-557; see also Shapiro, "The European Court of Justice," 127. 94. Burley and Mattli, "Europe Before the Court," 70; and Alter and MeunierAitsahalia, "Judicial Politics and the European Community," 558. 95. Wayne Sandholtz, "Membership Matters: The European Union and State Preferences" (unpublished manuscript, University of California-Irvine, 1994). 96. Renaud Dehousse and Giandomenico Majone, "The Dynamics of European Integration" (working paper, European Community Studies Association's Third Biennial Conference, Washington, D.C., 2 7 - 2 9 May 1993), 3. 97. Alan Milward, The European Rescue of the Nation-State (Berkeley: University of California Press, 1992). 98. Dehousse and Majone, "The Dynamics of European Integration," 36. 99. Alberta Sbragia, "The European Community: A Balancing Act," Publius 23 (Summer 1993): 23-38; Simon Hix, "The Study of the European Community: The Challenge to Comparative Politics," West European Politics 17 (January 1994): 1-30; Giandomenico Majone, "The European Community Between Social Policy and Social Regulation," Journal of Common Market Studies 31 (1993): 153-170; Bruce Bueno de Mesquita and Frans Stokman, eds., European Community Decision Making (New Haven: Yale University Press, 1994); and Wayne Sandholtz, "Membership Matters."

3 The Justiciability of Subsidiarity Daniel G. Partan

This chapter 1 examines the subsidiarity principle contained in the Maastricht Treaty on European Union 2 and asks how that principle might apply in litigation before the European Court of Justice. Following a brief sketch of subsidiarity as applied to EC environmental action by the Single European Act, 3 this chapter examines the subsidiarity provisions adopted at Maastricht 4 and the impact that those provisions may have on European Court practice. Subsidiarity will undoubtedly function as a guide for actions by political organs of the European Union. 5 My inquiry examines subsidiary's legal dimension. I ask in what way and to what extent the European Court will apply the subsidiarity principle as a rule of law in its review of Community legal action. I use the term justiciability to define my inquiry. By this I mean the capacity of a court (the European Court) to adjudicate a particular type of question (subsidiarity). In my use of the term, issues are justiciable when by their nature they are capable of being decided through the judicial process. Justiciability might also refer more broadly to a court's legal competence or its subject-matter jurisdiction, that is, to the court's statutory or constitutional authority to entertain and decide particular issues. When justiciability is used in this latter sense, what is meant is that the matter at issue must fall within the court's jurisdiction or legal competence to be justiciable. But I do not address questions of legal competence or jurisdiction. 6 My inquiry and my use of the term justiciability refer only to the judicial capacity of the European Court to interpret and apply the Maastricht Treaty's subsidiarity standard as a rule of European Union constitutional law. When I speak of the European Court's capacity to adjudicate questions of subsidiarity, I recognize that there is no uniform cross-cultural concept of capacity to adjudicate. Capacity to adjudicate is culture- and courtspecific. Judicial systems that share the ideals of an independent judiciary

63

64

Reflections on European

Integration

and the rule of law may well differ in the line drawn between matters for courts and matters for legislatures. Policies and values that are appropriately determined by judges in one system may be reserved for political organs in another. Indeed, even within a single judicial system, issues that are justiciable at one level or in one type of court may not be justiciable in another. Hence the question of justiciability of subsidiarity in the European Court cannot be resolved by abstract principle. History and analogy provide the only effective guide. I will endeavor to assess the justiciability of subsidiarity by examining European Court practice on analogous questions. As explained in the discussion that follows, the Maastricht Treaty makes subsidiarity a basic requirement of Community law. Since the European Court's role under the EC Treaty is to ensure that "the law is observed" and "the law" includes treaty law, 7 compliance with subsidiarity lies within the Court's formal legal competence. The justiciability question therefore is not whether the Court should apply subsidiarity, but rather in what way and to what extent the Court will or should do so. I resist the suggestion that the Court's use of subsidiarity should be confined to procedural or process concerns. 8 No doubt the procedural approach would ensure that consideration is given to subsidiarity in the Community legislative process. However, it was contemplated at Maastricht that subsidiarity, and hence the Court's supervisory role, must also reach into the substance of Community lawmaking. The defining difference between subsidiarity as a judicial issue and subsidiarity as a political issue is the consequent locus of substantive lawmaking authority. Operationally, the justiciability question becomes whether the substantive role of subsidiarity will be left to the political process or whether it will be shaped and guaranteed by the European Court. Treating subsidiarity as nonjusticiable, or as subject only to procedural requirements, would in effect entrust the line between member state and Community lawmaking authority to political decisions made through the Community legislative process. Treating subsidiarity as justiciable in the substantive sense would ensure a continuing role for the European Court in shaping and defining the contours of Community lawmaking authority. The question of justiciability of subsidiarity thus affects core concerns about the distribution and separation of powers in the European Union. Indeed, the justiciability of subsidiarity may be a watershed constitutional issue. If subsidiarity is applied by the Court to define the substantive reach of Community law, the Court is positioned to continue the trend toward enhanced Community lawmaking that was so vigorously begun by the Court's direct effect and supremacy decisions. 9 By contrast, if the Maastricht Treaty's subsidiarity provisions give rise at most to procedural requirements in litigation before the Court, subsidiarity will leave the substance and reach of Community lawmaking to the political process. This

The Justiciability

of

Subsidiarity

65

would mean that politics and not law will define the line between national and Community competence. 1 0 The Court's attitude toward subsidiarity thus can have profound consequences, but it is by no means clear which route is preferable. Cogent arguments can be advanced for both political and judicial decisionmaking. Since political solutions are presumed responsive to the political will, the capacity for political choice implicit in nonjusticiability may be praised for flexibility and democratic accountability. But the political model may also be criticized because political solutions lack stability and fail to provide a principled basis for extrapolation to new situations; constitutions should not be left to the winds of political change. In other words, justiciability may enhance the stability of the law since the Court will develop a subsidiarity doctrine that clarifies the line between Community and member state legislative power. My purpose here is not to argue for one route over the other or to urge judicialization of matters that some may see as more appropriately resolved by the political process. Rather I intend simply to define the issue and to ask how the justiciability of subsidiarity is likely to be resolved by the European Court.

An Overview of Subsidiarity in Community Law Although I recognize that the roots of subsidiarity lie in theology and in political theory, my inquiry is confined to the potential role of subsidiarity in European Court practice under the European Community Treaty. 11 Nevertheless, it is well to remember the sharp language used by Pope Pius XI responding to the expanding powers of Italy's fascist government, language that finds an echo in the views of some contemporary political leaders: It is an injustice, a grave evil and a disturbance of the right order, for a larger and higher association to arrogate to itself functions which can be performed efficiently by smaller lower societies. This is a fundamental principle of social philosophy, unshaken and unchangeable. 12

Invocations of subsidiarity in European Community usage are typically more muted. According to a 1990 EC Task Force on the Environment, subsidiarity means that "primary responsibility and decision-making competence should rest with the lowest possible level of authority of the political hierarchy." 13 Further, "Member States should be given the maximum flexibility in choosing how to meet [Community] environmental standards, and should be free to impose environmental standards higher than the Community norm." 1 4 Though the Commission similarly uses less majestic language than that of the Pope, it is equally categorical in its

66

Reflections on European

Integration

judgments. In comments prepared for Parliament and the Council in 1992, the Commission articulated the following attitude toward subsidiarity: Subsidiarity means that, when exercising its powers, the Community must, where various equally effective options are available, choose the form of action or measure which leaves [to] the Member States, individuals or businesses concerned the greatest degree of freedom. 15

Although the term was not used, the basic idea of subsidiarity was inserted in the text of the EC Treaty in 1986 by the Single European Act. Article 130r(4), made applicable solely to the new chapter on the environment, authorized Community action "to the extent to which" EC environmental objectives "can be attained better at Community level than at the level of the individual Member States." 16 The reference in Article 130r(4) to Community environmental objectives that "can be attained better at Community level" looks in two directions. Under this provision, subsidiarity operates both to empower and to restrict; the provision both unleashes Community action and limits that action. And, centrally important to the justiciability inquiry, the limits of Community action are defined in terms of comparative efficiency. Article 130r(4) authorizes Community action only where the Community can attain the treaty's environmental objectives "better" than can be done by action at the level of the individual member states. Since "better" normally includes with greater efficiency, the provision seems to require a comparison between the efficiency of Community and member state action. For Article 130r(4) subsidiarity to be enforced in the European Court, the Court will need to compare the efficiency of the Community measure with that potentially achieved through member state implementation. 1 7 Since environmental protection is embraced by the celebrated subsidiarity provisions of EC Treaty Article 3b, the Maastricht Treaty eliminated the subsidiarity provisions of Article 130r(4). Although the new article applies only in "areas which do not fall within [the Community's] exclusive competence" 1 8 and restricts rather than authorizes action, Article 3b added by the Maastricht Treaty 19 now mandates subsidiarity throughout the range of Community action. Article 3b provides that "the Community shall take action, in accordance with the principle of subsidiarity, only if and in so far as the objectives of the proposed action cannot be sufficiently achieved by the Member States and can therefore, by reason of the scale or effects of the proposed action, be better achieved by the Community." 2 0 Article 3b is not a model of transparency. Since Community action is authorized "only if" Community objectives "cannot be sufficiently achieved" by the member states, the article appears to exclude Community action altogether when the Community's objectives can be "sufficiently achieved by the Member States." This may mean that Community action is

The Justiciability

of

Subsidiarity

67

foreclosed even where inefficient member state action will be effective, that is, where inefficient member state action will "sufficiently" achieve the objectives of the proposed action. 21 If so, comparative efficiency may no longer be a feature of subsidiarity. But the article also seems to invoke comparative efficiency by referring to "the scale or effects of the proposed action" as the reason that the Community's objectives "cannot be sufficiently achieved by the Member States." It is for reasons of "scale or effects" that the Community's objectives can "be better achieved by the Community." At least where scale or effects implicate efficiency, efficiency becomes a Community objective, and the Article 3b formula seems to mean that comparative efficiency will justify Community action. Comparative efficiency may also be implicit in the overall structure of Article 3b. In place of the more straightforward comparative efficiency test of the Single European Act (Community objectives "can be attained better at Community level"), the language of Article 3b seems to envision two conditions. The article requires both that the Community's objectives "cannot be sufficiently achieved" by states and that they "can . . . be better achieved by the Community." When the two conditions are linked together, the article can be seen to retain the focus on comparative efficiency. This is because the Community will first define the "objectives of the proposed action" and then will act, but "only if and in so far as" the two conditions are satisfied. Where an objective can be "better achieved" by Community action, it would normally be no stretch to say that it cannot be "sufficiently achieved" by state action. Since the Community defines the objective, the efficiency of its attainment can be made an integral part of that objective. In such a case inefficient member state action will always be insufficient. 2 2 Hence Maastricht's Article 3b, like the SEA's Article 130r(4), seems to frame subsidiarity in terms of comparative efficiency. The subsidiarity question thus becomes: Can the Community's objective be "better achieved" by Community as compared to national action? My task is to examine whether this question can be answered through judicial review in the European Court.

The Central Role of "Necessity" Thus far we have addressed only the subsidiarity provisions of the second paragraph of Maastricht's Article 3b, quoted in the text above. The article contains two other restrictions on Community action. The initial paragraph incorporates the delegated powers principle, 2 3 and the third paragraph restricts Community action to measures that do not "go beyond what is necessary to achieve the objectives" stated in the treaty. 24 This is somewhat confusing. Although Article 3b begins by confining Community action to the objectives stated in the treaty, which are to be attained by use of the

68

Reflections on European

Integration

powers conferred in the treaty (the delegated powers principle), the article ends by repeating that Community authority does not "go beyond what is necessary to achieve" the treaty's objectives. 2 5 This apparent redundancy is perhaps best explained by a desire to put into the treaty text the Community law principle of proportionality. As developed by the European Court of Justice, proportionality is a "general principle of law" that has its source in the constitutional traditions common to the member states. The principle applies to actions taken by the Community as well as to actions taken by member states in implementing Community law. 26 In both instances the proportionality principle is said to have three fundamental aspects. First, it requires that the Community or national action under review be reasonably related to the Community's objective; that is, the action must be "suitable for the purpose of achieving the desired objective." Second, the principle also requires that the action must not be out of proportion in comparison with the objective in the sense that the action must not "go beyond what is necessary" to achieve the Community's objective. 27 This element requires that the Community choose the least restrictive regulatory alternative, that is, the alternative that is the least invasive of member state authority or of affected private interests. The third element that appears in the European Court's case law calls for a weighing or balancing of the measure's costs against its benefits. This factor is discussed in the next section. Proportionality's first two elements seem to be embraced by the delegated powers principle articulated in the first paragraph of Article 3b. It is implicit in the delegated powers principle that Community action that either is not reasonably related to the Community's objective, or goes beyond what is necessary to achieve that objective, will for that reason fall outside of the Community's delegated powers. To paraphrase the language favored by the Court, both types of Community action are manifestly illsuited to attaining the desired Community objectives. The restatement of these limits through the formula used in the final paragraph, which stipulates that Community measures "shall not go beyond what is necessary to achieve" treaty objectives, would seem to have no purpose other than to make clear that Article 3b incorporates the European Court's proportionality jurisprudence. 2 8 A recent Commission paper seems basically to agree with this approach. 2 9 After characterizing subsidiarity as a "commonsense principle" and "essentially [a] political principle," the Commission paper notes that subsidiarity satisfies the two basic requirements of proportionality, which are "the need for Community action and the need to ensure that the means employed are commensurate with the objectives pursued." 3 0 The three elements of Article 3b (delegated powers, subsidiarity, and proportionality) thus share the idea of necessity.

The Justiciability of Subsidiarity

69

The link between subsidiarity and necessity is also shown in the Commission's approach, which relates subsidiarity to the intensity of Community action. According to the Commission paper quoted above, proportionality requires that Community action always take the least invasive form capable of accomplishing the Community's objectives. This means that each element of a Community measure must be shown to be "necessary" to achieve the Community's objectives. The Commission distinguishes three forms of Community action on an increasing scale of intensity: "priority should be given to support measures rather than regulations, to mutual recognition rather than harmonization, [and] to framework directives rather than detailed rules." In the Commission's view, subsidiarity implies that "the most binding instruments should be used [only] as a last resort." 3 1 In other words, any increase in intensity must be justified by necessity. The relationship between the "scale or effects" of Community action and the proportionality element of subsidiarity presents some conceptual difficulty. This is because the subsidiarity provisions of Article 3b rely upon the "scale or effects" of Community action to establish the need for Community action; the "scale or effects" show that the objectives sought can be better achieved by Community action. Indeed, at the Edinburgh summit in December 1992, the European Council affirmed that to reflect subsidiarity, "the Council must be satisfied that action at the Community level would produce clear benefits by reason of its scale or effects compared with action at the level of the Member States." 3 2 Even so, the Commission considered that for Community legislation to be necessary to accomplish the Community's objectives, "the subsidiarity principle dictates that Community legislation and national measures [are each to be given] its own respective role: Community legislation forms the framework into which national action must be fitted." 3 3 Hence two conditions are needed for Community action to satisfy both the intensity and the proportionality standards of Article 3b. First, the "scale or effects" of Community action must be substantial enough to justify Community action by showing that its objectives "cannot be sufficiently achieved by the Member States." Second, the "scale or effects" of Community action must not be so substantial as to exceed what is required to accomplish the Community objective. This in effect restates what I have said about the basic elements of the proportionality principle. 34

Judicial Review of Necessity and Proportionality As has been shown in European Court proportionality jurisprudence, 35 the element of a reasonable relation to the objective sought is without doubt justiciable. This "reasonable relation" element confines implementing

70

Reflections on European Integration

action to measures that are shown by a logical reasoning process to be suitable for, or capable of, achieving the desired objective. This use of logical reasoning is the quintessential role of a court. The remaining two proportionality elements are intended to ensure a proper proportion between the measure and its objectives. One requires that implementing measures do not "go beyond what is necessary" to achieve their objectives. 3 6 The other seeks to ensure that the costs of an implementing measure do not manifestly outweigh its benefits. 37 The necessity element of the Court's proportionality assessment presents no serious issue of justiciability. To ask whether a measure is necessary to achieve the desired objective is simply to ask whether the objective can be achieved by some less radical or less intrusive means. Such questions are the normal grist for the judicial mill. Given the benefit of presumptions and of burdens of proof and persuasion, evidence submitted in an adversarial process will enable a court to decide whether a measure goes beyond that which is necessary to achieve the desired objective. As with the "reasonable relation" element, logical analysis and judicial reasoning are sufficient to provide the answer. In contrast, the final proportionality element that requires balancing costs against benefits presents justiciability concerns. Reviewing courts can and do require decisionmakers to identify and to consider or balance the parties' interests and to balance costs against benefits. Where the requisite fact of such consideration or balancing is not established, reviewing courts also can and do void or vacate the decision and remand the matter for reconsideration. For a court to do more, that is, for the court itself to compare or weigh the competing interests, or to balance costs against benefits, would displace the decisionmaker and put the court squarely in that role. Where a court engages in interest or cost-benefit balancing, the court in effect takes over the substance or the merits of the decision and substitutes its own views for those of the agency whose decision is under review. The question of cost-benefit or interest balancing thus presents both issues of the court's authority or legal competence and issues of judicial capacity. For purposes of this discussion we can assume that the European Court has the legal competence to engage in interest balancing. 38 My inquiry asks whether adjudication in the European Court can strike a balance between costs and benefits, or between the interests of the Community and those of the member states, or of the private parties whose interests are affected by the Community's acts. I approach that inquiry through analysis of cases in which the European Court has been asked to undertake the balancing task. The European Court's apparent willingness to balance costs against benefits appears in the early proportionality ruling in the Internationale Handelsgesellschaft case. 3 9 That case dealt with a 1967 Council regulation that made a member state's grant of export licenses for certain grains

The Justiciability of Subsidiarity

71

conditional upon prior payment of a deposit that was forfeited if the export was not made. The Community objective was to ensure effective control of the grain market, which required precise forecasts of future imports and exports. In the light of this objective the Court ruled that: The requirement of import and export licenses involving for the licensees an undertaking to effect the proposed transactions under the guarantee of a deposit constitutes a method which is both necessary and appropriate to enable the competent authorities to determine in the most effective manner their interventions on the market in cereals. . . . [Furthermore], the costs involved in the deposit do not constitute an amount disproportionate to the total value of the goods in question and of the other trading costs. It appears therefore that the burdens resulting from the system on deposits are not excessive and are the normal consequences of a system of organization of the markets conceived to meet the requirements of the general interest. 4 0

The Court's explanation of its ruling uses cost-benefit balancing language: The deposit system's "burdens" are "not excessive" because "the costs involved in the deposit [are not] disproportionate to the total value of the goods." This approach seems to balance costs against benefits, but it fails to set forth a logical framework for the conclusion that the costs are "not excessive." The Court would know how to value the costs, but how are the benefits to be valued? Anomalously the Court seemed to measure the benefits (grain market stability) by the total value of the goods for which a deposit had been paid. The Court did this by comparing the costs in forfeited deposits to the total value of the goods. Does the Court mean that the costs (the forfeited deposits) do not exceed the benefits (grain market stability) unless those costs approach (or exceed) the value of the goods? Moreover, does it matter that the benefits are conferred on society at large whereas the costs are borne by identified litigants? The Court did not address these questions. Instead it said only that the costs were the "normal consequences" of market regulation. The Court's conclusion seems to bypass cost-benefit balancing. Since the Community measure had been taken to achieve a legitimate objective, presumably costs that are the "normal consequences" of that "necessary" measure must be borne regardless of the balance. In proportionality terms this assumes a finding that the measures taken (the forfeit of deposits) are the least invasive means of achieving the goal of grain market stability. To hold otherwise would be to rule that, for reasons of excessive cost, the Community could not institute market control measures that lie within its treaty authority. And it would be implicit in such a ruling that the Community's market control objective (the benefit) has less value than the entrepreneur's forfeited deposits (the cost). In a more recent case 4 1 the Court addressed a Council directive that banned the sale of certain growth hormones used for livestock. The objectives

72

Reflections on European Integration

of the measure were to remove distortions of competition and barriers to trade caused by differences in national legislation on growth hormones and to alleviate consumer anxiety about the human health effects of growth hormones. 4 2 The directive was challenged in a British court on proportionality grounds, which led to a reference to the European Court for a preliminary ruling. Three arguments were advanced. First, the outright prohibition of the sale of hormones is impossible to apply in practice. Second, since there was no showing of an actual danger to human health, less restrictive measures, such as dissemination of information and labeling requirements, would have been sufficient to allay the anxieties of consumers. Third, "the prohibition in question entails excessive disadvantages, in particular considerable financial losses on the part of the traders concerned, in relation to the alleged benefits accruing to the general interest." In his assessment of the proportionality arguments, the advocate general gave an extended cost-benefit analysis: Not only does the measure adopted have no positive effect but, on the contrary, it causes a number of disadvantages. It causes economic loss which is sustained first of all by the pharmaceutical companies which are no longer able to sell certain preparations, then by farmers who are denied the savings in costs and gains in productivity made possible by the use of these substances, and finally by veterinarians who are denied clinical freedom to administer the hormones and the income they would have earned from doing so. The effect of the prohibition, they say, is that the beef produced is higher in fat content than that treated with hormones, with all the ill consequences which that entails for public health. It has also caused a considerable expansion of a black market in dangerous substitute substances, which previously existed only in Member States where the prohibition was already in force. 4 3

Nevertheless the challenge failed. The Court found that the growth hormone ban was both reasonably related to the Community's objectives and necessary to attain those objectives. The Court considered that only a total prohibition of growth hormones was capable of allaying consumer anxieties. Moreover, "any system of partial authorization would require costly control measures whose effectiveness would not be guaranteed. It follows that the prohibition at issue cannot be regarded as a manifestly inappropriate measure." Taking account of the requirements of health protection, the Council was entitled to take the view that "the removal of barriers to trade and distortions of competition could not be achieved by means of less onerous measures such as the dissemination of information to consumers and the labeling of meat." With regard to costs and benefits, the Court said only that "the importance of the objectives pursued is such as to justify even substantial negative financial consequences for certain traders." The Court's ruling does not undertake a detailed balance of costs against benefits. Instead it simply

The Justiciability

of

Subsidiarity

73

allows the achievement of the Community's stated health objectives to take precedence over costs to affected individuals. The advocate general took the same perspective but explained the result in cost-benefit terms: When weighing "damage caused to individual rights against the benefits accruing to the general interest, it should be stated that the maintenance of public health must take precedence over any other consideration." But the effect of this perspective is to externalize the costs of the Community's acts, which is the antithesis of cost-benefit balancing. Since "the maintenance of public health must take precedence over any other consideration," the cost to individuals is not actually a factor, and there is no role for judicial cost-benefit balancing. Another example of the externalization of costs can be drawn from the field of consumer protection. A member state's perception of its consumer protection needs may lead it to adopt product or other standards that can be met by local industry but impose expensive or insuperable problems for producers located in other member states. The European Court indicated in its landmark Cassis de Dijon ruling, 4 4 decided prior to the SEA amendments, that the logic and law of the common market embodied in EC Treaty Article 30 45 require that a product that is lawfully produced and put into circulation in one member state must be freely marketed in every other member state. In Cassis de Dijon, Germany had prohibited the marketing of a French-produced fruit liqueur, Cassis de Dijon, since that liqueur had less than the minimum alcoholic content required by German law. As consumer protection is not among the treaty exceptions to free movement of goods, 4 6 the Cassis de Dijon ruling allowed for implied exceptions to the otherwise categorical prohibition of quantitative restrictions on imports. The Court recognized that consumer protection and other "mandatory requirements" of national law could exclude imported goods, but it declined to allow the German rule to prevail because it was not "necessary" for consumer protection. The Court ruled that Germany's interest in consumer protection could have been satisfied in other ways. In an unprecedented formal response to the Court's Cassis de Dijon ruling, a Commission statement acknowledged that "a product lawfully produced and marketed in one Member State must, in principle, be admitted to the market of any other Member State." 4 7 Nevertheless the Commission considered that both consumer protection and the protection of the environment 4 8 were exceptions to the Cassis de Dijon market access rule. States may restrict the free flow of goods for environmental and consumer protection reasons so long as such measures are both "necessary, that is appropriate and not excessive," and "essential for the purpose to be attained," that is, they utilize "means which are the most appropriate and at the same time least hinder trade." 4 9 The Commission's formulation seems totally deferential to the member state's definition of its environmental and consumer protection objectives.

74

Reflections on European

Integration

The Commission will accept any "necessary" and "proportionate" national protective measure. The measure must be "necessary" in the sense of being essential to achieving the state's objective and of being least invasive of the Community interest in the free flow of goods. It must also be "proportionate" in the less precise sense of not being out of proportion to the objective pursued. Under the Cassis de Dijon concept, both the Commission and the Court have a role in applying the standards of "necessity" and "proportionality." The Commission acts through oversight of member state applications of Community law and through enforcement actions under EC Article 169. The Court's rulings can come both in Article 169 actions and through preliminary rulings on questions submitted by national courts under EC Article 177. The framework for the Court's review distinguishes between necessity and proportionality. As the advocate general said in a recent Sunday trading case, the concepts of necessity and proportionality are not co-extensive. T h e requirement of necessity has t w o aspects: first of all that the national legislation in question is in fact relevant for the attainment of the objective p u r s u e d — i n other words, that there is at least p o t e n t i a l l y a causal c o n n e c t i o n b e t w e e n the two; secondly, that there is no alternative to such legislation w h i c h is equally e f f e c t i v e but less restrictive of intra-Community trade (the criterion of the least restrictive alternative). The criterion of proportionality, o n the other hand, is to be understood as m e a n i n g that e v e n if legislation is relevant and is the least restrictive of trade, it is n e v ertheless i n c o m p a t i b l e w i t h A r t i c l e 3 0 . . . w h e r e the o b s t a c l e w h i c h it creates to i n t r a - C o m m u n i t y trade is out of proportion to the o b j e c t i v e pursued.50

The Court has used the language of interest balancing in its assessment of necessity and proportionality in the Cassis de Dijon context, but the cases have not actually required the Court to balance the m e m b e r state's environmental objectives against the C o m m u n i t y ' s interest in the f r e e flow of goods. Indeed, it is not clear f r o m the Court's jurisprudence that the Court would actually weigh or balance the C o m m u n i t y interest against the interests of member states. An example is the so-called Danish Bottle Case.51 D e n m a r k adopted a system that required all containers for beer and soft drinks to be returnable; these beverages could be marketed only in reusable containers approved by the Danish National Agency for the Protection of the Environment. 5 2 Citing its earlier jurisprudence, the Court ruled that "measures adopted to protect the environment must not 'go beyond the inevitable restrictions which are justified by the pursuit of the objective of environmental p r o t e c t i o n . ' " 5 3 The Court considered the use of deposit-and-return containers "an indispensable element of a system intended to ensure the re-use of containers." Such a system was therefore

The Justiciability

of

Subsidiarity

75

"necessary" to achieve the state's environmental protection objective. The Court also ruled that "the system for returning non-approved containers is capable of protecting the environment." 5 4 Hence the Danish returnable container system was approved despite its impact on the free movement of goods. Denmark's further requirement that beverages be marketed only in containers approved by the Danish National Agency for the Protection of the Environment was rejected by the Court. The approved-container requirement, said the Court, would impose substantial additional costs for foreign producers and make the importation of goods very difficult. But, because the returned containers could be removed by the foreign producer, it was also not necessary for Denmark to require the use of approved containers. Hence the Court held that the approved-container aspect of the Danish rule violated Article 30 and could not be applied against foreign producers. The approved-container aspect of the Court's ruling is not entirely clear but seems consistent with a pattern in which the Court first defines the "essential element" in a state's objective and then confines the state's action to measures that the Court considers "necessary" to achieve that objective. So viewed, the Court exercises judicial control, but it does not balance interests. Instead, the Court accepts the state's articulation of its objectives and allows the state to take measures to achieve those objectives, so long as such measures satisfy the criteria of "rational relation" and of "necessity." The state's measures must be rationally related to the state's objectives, and the measures must be "necessary" in the sense that they are the least invasive of the Community's interest in the free movement of goods. The Court's opinion may go further and speak of a need to ensure that the burdens of implementation measures are not out of proportion to the benefits of the objectives sought, but the Court has yet to examine with any care just how such a balance is to be struck. The Court has had no occasion as yet to address the subsidiarity principle. In the Danish Bottles Case, the Court had only to assess environmental action taken in the Danish national interest against the challenge that such action conflicted with the treaty objective of free movement of goods. The Court's approach was to ask only whether the Danish rule exceeded "inevitable restrictions" justified by the national environmental protection objective, not whether the costs of the Danish rule exceeded its benefits.

Conclusion The approach taken by the Court in the Danish Bottles Case seems the likely mode of analysis for the subsidiarity principle. Where the Community's objectives are legitimate, that is, where they fall within powers delegated to

76

Reflections on European

Integration

the Community, those objectives are in the first instance defined by the Community and then accepted by the Court. The Court's task would then be to satisfy itself that the proposed Community action is reasonably related to the Community's objectives. It must also determine that the action does not go beyond what is necessary to achieve those objectives. Finally the Court would determine whether the Community-defined objectives could be accomplished at least as well by national action. Presumptions and burdens of proof and persuasion would assist the Court in making this judgment. For example, the burden of persuasion might be placed upon the proponent of subsidiarity, that is, the party that has asked the Court to rule in favor of member state action. Where national action is not shown to be equivalent or preferable, the Community would be free to act. In this scenario the Court would not be called upon to assess the costs of Community action or to balance those costs against the costs of member state action. With subsidiarity no bar, the Court could consider that the costs of Community action reflect only "inevitable restrictions" justified by the need to attain the essential elements of the Community's objective. There would be no occasion for the Court to balance costs against benefits or to balance the Community's interests against those of the member states. The difficult balancing question would be presented where Community objectives could be attained both by member state and by Community action. In such a case the Court is certain to be asked whether Community action is to be preferred on efficiency grounds. The argument would be that member state action, though effective, would be inefficient, and therefore that the Community's objectives "can . . . be better achieved" by Community action. 55 If the Court were to apply the comparative efficiency standard, the Court's task would be to determine whether Community implementation action would be less costly than member state action. The monetary costs of implementation could be assessed by the Court, but the costs of denial of subsidiarity would also be a factor. Balancing the costs of Community action against the costs of member state action would require the Court to place a value on subsidiarity from which to infer the cost of a failure to apply subsidiarity. This would involve a political choice for the Court, which it may consider best left to the Community's political organs. This is especially true once the Community hat chosen to move forward; the Court may well be reluctant to substitute its view for that of the Community's political organs. This conclusion is supported by the Court's past jurisprudence in which, while appearing to affirm its authority to balance costs against benefits, the Court has refrained from such an exercise, preferring to explain its decision on other grounds. The pattern of decisionmaking just described leaves the choice of objectives to the Community political authorities, with the Court deciding only whether the measures taken are rationally related to their objectives and whether they are "necessary" in the sense of being the least restrictive or invasive alternative. As applied to subsidiarity, these inquiries go beyond

The Justiciability

of

Subsidiarity

77

procedure into substance (the procedural issue). T h e Court w o u l d d e c i d e not only whether the political organs have addressed subsidiarity but a l s o whether C o m m u n i t y action taken to a c h i e v e legitimate C o m m u n i t y object i v e s is c o n f i n e d to m e a s u r e s n e c e s s a r y to a c h i e v e the d e f i n e d o b j e c t i v e s (the substantive issue). But judicial reasoning will be s u f f i c i e n t to answer these inquiries. There will be n o n e e d to substitute the Court's j u d g m e n t o n substantive p o l i c y i s s u e s for that of the C o m m u n i t y ' s political organs. W h i l e the Court's precedents d o not e x c l u d e the m o r e p o l i c y - o r i e n t e d comparative e f f i c i e n c y approach, neither do they s u g g e s t that the Court is l i k e l y to e n g a g e a c t i v e l y in a b a l a n c i n g o f C o m m u n i t y against national interests.

Notes 1. I am indebted to Robert P. McHale of the Class of 1995 at Boston University School of Law, who worked as my research assistant on this project. 2. Footnote citations to the basic treaty documents of the European Union distinguish between the original provisions of the 1957 "Treaty Establishing the European Economic Community" ("Treaty of Rome") and provisions amended or added by the 1986 "Single European Act" ("SEA"), or by the 1992 "Treaty on European Union" ( " T E U " or "Maastricht"). The EEC is now called the European Community ("EC"); the Treaty of Rome as amended is referred to as the EC Treaty or the Community Treaty. 3. EC Treaty Article 130r(4), added by the SEA. 4. EC Treaty Article 3b, added by the TEU, quoted in the text at note 20. 5. For an early insightful work, see Renaud Dehousse, "The Legacy of Maastricht: Emerging Institutional Issues," Collected Courses of the Academy of European Law for 1992, vol. 3, part 1 (the Hague: Kluwer, 1994), 181. 6. Issues of subsidiarity can be expected to be presented in the European Court in actions for annulment under EC Treaty Article 173 and references for preliminary rulings under Article 177. 7. EC Treaty Article 164. 8. See, for example, George A. Bermann, "Taking Subsidiarity Seriously: Federalism in the European Community and the United States," Columbia Law Review 94 (1994): 331. 9. The seminal cases are Van Gend en Loos, Case 26/62, (1963) ECR 1; and Costa v ENEL, Case 6/64, (1964) ECR 585. 10. Restraint exercised in the Community political process could check the expansion of Community law whatever role is played by the European Court. Even so, matters referred to the Court for preliminary rulings pursuant to EC Treaty Article 177 will have potential for expanding the domain of Community lawmaking. 11. By adding Article 3b to the EC Treaty, TEU Article G, para. B(4), applies subsidiarity to all "areas which do not fall within [EC] exclusive competence." TEU Article B, final paragraph, also calls for respect for the principle of subsidiarity as defined in EC Article 3b in achieving the new objectives set out in the TEU. This provision may give subsidiarity the same status with respect to the TEU as it has with respect to the amended EC Treaty, but this essay is confined to the role of subsidiarity in the EC. There is no reference to subsidiarity in the treaties establishing EURATOM or the European Coal and Steel Community.

78

Reflections on European

Integration

12. Catholic Church, Quadragesimo Anno (London: Catholic Truth Society, 1936), 31, para. 79. 13. "1992" The Environmental Dimension (EC Task Force Report on the Environment and the Internal Market, 1990), 16, para. 3.2. 14. Ibid., 17, para. 3.2. The quoted passage reflects the provisions of EC Treaty Article 130t, added by the Single European Act. 15. EC Commission, "The Principle of Subsidiarity: Communication of the Commission to the Council and the European Parliament," (Doc. SEC[92] 1990 final, 27 Oct. 1992) sect. IV, 13. 16. EEC Treaty Article 130r(4), added by the Single European Act. EC environmental objectives are stated in paragraph 1 of the article: (1) To preserve, protect, and improve the quality of the environment (2) To contribute towards protecting human health (3) To ensure a prudent and rational utilization of natural resources The Treaty on European Union rewords these objectives slightly and adds a fourth objective: "promoting measures at international level to deal with regional or worldwide environmental problems." 17. The Court has had no occasion to pass upon the comparative efficiency aspect of Article 130r(4). 18. The EC Treaty does not specify areas of the Community's "exclusive competence," wherein Community action is presumably not subject to the restraints of subsidiarity. For the present discussion I assume that such matters as environmental, consumer, and health protection fall within areas of concurrent competence and hence are subject to Article 3b subsidiarity. 19. The term "subsidiarity" appears in three provisions of the Treaty on European Union: (1) in the preamble, penultimate paragraph; (2) in Article B, final paragraph; and (3) in Article G, para. B(5). The latter paragraph adds Article 3b to the Treaty of Rome. The idea of subsidiarity also appears in TEU Article A, which refers to the "process of creating an ever closer union . . . in which decisions are taken as closely as possible to the citizen." 20. The quoted provision is EC Article 3b, para. 2. Article 3b, para. 1, limits Community action to "the limits of the powers conferred upon it by this Treaty," and para. 3 provides that Community action "shall not go beyond what is necessary to achieve the objectives of this Treaty." 21. Koen Lenaerts, "The Principle of Subsidiarity and the Environment in the European Union: Keeping the Balance of Federalism," Fordham International Law Journal 17 (1994): 846, 876-877. 22. The new text also explains the utility of Community action to attain the Community's objectives by reference to the "scale or effects of the proposed action." One implication is that the "scale or effects of the proposed action" will show whether the objectives of that action "cannot be sufficiently achieved by the Member States." See text at notes 30-34. 23. "The Community shall act within the limits of the powers conferred upon it by this Treaty and of the objectives assigned to it therein." EC Article 3b, para. 1. 24. EC Article 3b, para. 3. 25. Note that the objectives referred to in the first and third paragraphs of Article 3b are the objectives of the Treaty of Rome as amended by the Treaty on European Union, whereas the objectives referred to in the second paragraph are the objectives of the proposed Community action. The objectives of EC action must of course fall within the objectives of the EC Treaty.

The Justiciability

of

Subsidiarity

79

26. In Internationale Handelsgesellschaft, the European Court of Justice recognized "respect for fundamental rights [as] an integral part of the general principles of law protected by the Court of Justice." Proportionality is one such fundamental right "inspired by the constitutional traditions common to the Member States." Case 11/70, (1970) ECR 1125, para. 4. 27. Walter Rau Lebensmittelwerke v EEC, Cases 279, 280, 285, and 286/84, (1987) ECR 1069, para. 34. 28. See the discussion of judicial review of necessity and proportionality in the next section. 29. EC Commission, "The Principle of Subsidiarity: Communication of the Commission to the Council and the European Parliament" (Doc. SEC[92] 1990 final, 27 October 1992). 30. Ibid., Annex, 1, paras. 1 - 3 . 31. Ibid., Annex, 5, para. 2(b). 32. EC Council, "Overall Approach to the Application by the Council of the Subsidiarity Principle and Article 3b of the Treaty on European Union," Guidelines to the second paragraph of Article 3b, para, ii (Council DOC/92/8, 13 Dec. 1992). 33. EC Commission Communication (Doc. SEC[92] 1990 final, 27 Oct. 1992) 15, para. 2. 34. Moving down the Commission's scale of intensity of Community action may drop below the point at which the "scale or effects" of Community action would warrant the conclusion that "the objectives of the proposed action cannot be sufficiently achieved by the Member States." As adopted by the Edinburgh European Council, the Commission's "Guidelines" to implement the subsidiarity principle suggest that the Community "should legislate only to the extent necessary" and that "directives should be preferred to regulations and framework directives to detailed measures." This hierarchical order, including the preference for framework directives, is surely unexceptional. But the Guidelines move even further down the scale of intensity. They call for giving preference to "non-binding measures such as recommendations" and for giving consideration to "voluntary codes of conduct." Low intensity measures such as these drop off the subsidiarity scale. Since such measures contemplate voluntary action by member states, they are not subject to the subsidiarity idea of deference to the member states. See EC Council, "Guidelines to the third paragraph of Article 3b" (DOC/92/8, 13 Dec. 1992), para. v. 35. Rather than attempt an exhaustive survey of the European Court's proportionality jurisprudence, I have selected cases that seem to me illustrative of the approach taken by the Court. 36. Walter Rau Lebensmittelwerke v EEC, Cases 279, 280, 285, and 286/84, (1987) ECR 1069, para. 34. 37. See, e.g., Internationale Handelsgesellschaft, Case 11/70, (1970) ECR 1125. The Court's use of cost-benefit analysis is addressed below. 38. The reasoning is somewhat circular. If cost-benefit or interest balancing is a judicial task, it presents legal questions that the Court is competent to address because EC Treaty Article 164 provides: "The Court of Justice shall ensure that in the interpretation and application of this Treaty the law is observed." 39. Internationale Handelsgesellschaft v Einfuhr und Vorratsselle fiir Getreide und Futtermittel, Case 11/70, (1970) ECR 1125. 40. Ibid., 1158. 41. Regina v Minister of Agriculture, Fisheries and Food, Case C-331/88, (1991) 1 CMLR 507. 42. Council Directive 88/146 states in its preamble that differing member state assessments of the human health effects of growth hormones have led to varying

80

Reflections on European

Integration

regulations governing their use and that "this divergence distorts the conditions of competition . . . and is a serious barrier to intra-Community trade. 43. (1991) 1 CMLR 507. The advocate general was Mischo. 44. Rewe-Zentral AG v Bundesmonopolverwaltung fur Branntwein (Cassis de Dijon), Case 120/78, (1979) ECR 649. 45. EC Treaty Article 30 guarantees the free flow of goods by prohibiting quantitative restrictions on imports in intra-Community trade subject to exceptions stated in Article 36. Article 30 also prohibits "all measures having equivalent effect." 46. Under EC Treaty Article 36 the provisions of Article 30 "shall not preclude prohibitions or restrictions on imports . . . justified on grounds of public morality, public policy or public security [or] the protection of health and life of humans, animals or plants." 47. EC Commission, "Communication from the Commission concerning the Consequences of the Judgment given by the Court of Justice on 20 February 1979 in Case 120/78 (Cassis de Dijon)," Official Journal (10 March 1980) C 256/2. 48. See Procureur de la Republique v ADBHU, Case 240/83, (1985) ECR 531. 49. Commission Cassis de Dijon Communication, note 47. 50. Union Departmentale des Syndicats CGT de L'Aisne v Société Internationale de Distribution d'Equipements Familiaux, Case 312/89, (1991) ECR 1. The advocate general was Van Gerven. 51. Commission v Denmark (beverage containers), Case 302/86, (1988) ECR 4607. The claim was that a national law requiring a 24-hour weekly rest period on Sundays could not prevent SIDEF Conforama from opening its shops on Sunday as such a prohibition would have "equivalent effect" to quantitative restrictions and thus conflict with EC Treaty Article 30. See note 45 above. 52. The sole statutory exception allowed very limited quantities of imported beverages to be sold by foreign producers in order to test the market, but even then the foreign producers had to use deposit-and-return containers, which were in fact returned to the foreign producer. 53. Commission v Denmark (beverage containers), Case 302/86, (1988) ECR 4607, para. 11, quoting from Procureur de la Republique v ADBHU, Case 240/83, (1985) ECR 531. 54. Commission v Denmark (beverage containers), Case 302/86, para. 21 (1988) ECR 4607. 55. EC Treaty Article 3b, para. 2, quoted in full in the text at note 20.

4 Common, Collective, or Combined? Theories of Defense Integration in the European Union Patricia Chilton The present state of European Union studies in the defense and security field reflects the dichotomy of the Cold War period—a dichotomy between the study of the European Community/Union as a "civilian power" conveniently bracketed off from security considerations, and the study of European security in an international relations framework, which still fights shy of integration theory and largely disregards the European Union (EU) as an international actor. The theoretical dimensions of the Cold War dichotomy were clearly set out in the long-lived debate between François Duchêne and Hedley Bull. 1 According to one point of view, the new European Economic Community arrangements could be regarded as having solved the security problem in Europe by binding France and Germany into a permanent "civilian" relationship. Thus, Duchêne contended, a transnational "civilian power" was in the process of replacing military power and reversing the traditional power relationships in Europe. In that case, military competition would become impossible between partners within the Community, while external relations would be controlled by, on the one hand, the lack of military threat posed by the Community to its neighbors and, on the other hand, the economic power of the Community, which would make military aggression from the outside unlikely. According to the other point of view, the new pattern of civilian relationships altered nothing in the international anarchy of relations between sovereign states. Hedley Bull argued that the only power was military power. In that case, although it was possible that at some time in the future the European Community would become a sovereign state in its own right, the modern security state would remain the basic unit of international relations, and the only question was what sort of alliances such states would form. The course of European history, from the late 1950s until 1989, produced a curious effect. It became possible for both visions to be held without apparent contradiction. In accordance with Bull's framework, the

81

82

Reflections on European

Integration

Western European states aligned themselves more or less unanimously with the United States under the security alliance of the North Atlantic Treaty Organization (NATO). Meanwhile, Duchene's framework flourished at the economic level. It did so in a positive security vacuum, unruffled by any higher-level concern that the Community's security dimension was ultimately provided by NATO. This provision was assumed to be permanent. It was relatively unremarked, during the forty-year standoff with the Eastern bloc, that NATO provided something more tangible in terms of West European integration. 2 NATO's function as an integrator of armed forces on the European continent cannot be overestimated. This function has appeared more valuable in retrospect, since the demise of the Warsaw Pact has given way to the eruption of national and subnational armed conflict on the European periphery. In the 1970s, the development of interdependence and regime theories added considerable weight to the functionalist model of European integration, which was closely allied to Duchene's premises. Haas, Keohane and Nye, and Krasner all provided more sophisticated backing to Duchene's contention that economic and organizational power and the skillful use of international organizations could, in the modern world of global interdependence, replace or be superior to the military power traditionally wielded by unitary states. 3 The standard view of European integration, with its functionalist model of the way limited transnational economic regimes (such as the European Coal and Steel Community) would, by their spillover, automatically create more complex and increasingly political interdependencies, fits this mold perfectly. But the problem of military power was only momentarily removed from theoretical consideration. The dissolution of the blocs in 1989 and the collapse of Warsaw Pact cohesion shattered illusions that the European Community was immune from the need to take decisions about security. In some quarters this produced extreme reactions, such as the ultrarealist response that predicted rather old-fashioned power struggles between European states and a scenario in which Germany would want its own independent nuclear weapons. 4 In general, the analyses of international relations theorists and defense experts followed less outrageously in the same direction. There were three schools of post-Cold War strategic studies. The first was a realist retrenchment, which clung to the Cold War containment paradigm and betrayed an ideological attachment to its hard categories. 5 The second was subscribed to by both liberal and neorealist traditions, which recognized that a paradigm was lost, yet urged intellectuals to "wait and see" what institutional patterns would emerge before attempting prophecy. 6 The third, critical of realist and neorealist world views, argued that it was time to change the paradigm proactively. 7 Of these three schools, the third was

Theories of Defense

Integration

83

small indeed. Yet it was this train of thought that might have found some commonality with integrationist tradition. At the same time, however, European integration theory was in trouble. The classic route for integration, proceeding first to economic harmonization and cooperation, then to the development of the necessary political institutions, and finally to full integration of all the nation-state's functions, including foreign policy, security, and defense cooperation, was not the one followed by events. Defense cooperation, and other areas of security policy, received a surprising boost post-1989, while economic integration suffered unexpected setbacks, and political integration progressed fitfully in the face of popular, and populist, dissent. A rapidly developing defense union in the shape of the Western European Union (WEU) and an embryonic European army in the shape of the Franco-German Eurocorps marched alongside a new wave of "Europe-of-nation-states" rhetoric and glorification of narrow national interests. In this climate, theorists were cautious. The long-standing parallel approaches to Europe-building on the one hand and to adapting to the new strategic situation on the other maintained their hold, though their products jarred increasingly, not only with each other but with reality. Events, particularly the conflict in the former Yugoslavia, continued to force security onto the agenda. The disillusion about Europe's positive security vacuum had begun a few years earlier (most noticeably in response to the U.S. Strategic Defense Initiative of 1983) and had already led to the reactivation of the Western European Union as a focus for the emerging European defense identity (EDI). This shift of perspective was formalized in the Hague Platform of 1987. 8 The Single European Act of 1986 adumbrated Title V of the Maastricht Treaty. But it was with Maastricht that defense and security unequivocally entered the post-Cold War European integration framework. 9 The time is now ripe for theorizing defense and security integration in a way that impacts on integration theory in general. In the past five years, bringing defense and security studies into the European integration framework has been largely a matter of one of the following: (1) descriptive study of institution-building like the WEU and Eurocorps; (2) analysis of transatlantic relations, mainly reverting to the traditional realist categories of international relations; (3) defense analysis consisting of threat assessment and review of defense capabilities at the European level; and (4) a consumer-choice model of security alternatives, otherwise known as the European security architecture debate. Theoretical responses to the change in the relation between Europe-building and European security have lagged behind. 1 0 For the first two or three of the post-Cold War years, the defense structures of the major powers and organizations were themselves subject to upheavals of an uncoordinated kind, lacking strategic innovation or

84

Reflections on European

Integration

direction. The early stages of the NATO force structure review, resulting in the flaccid 1991 strategic concept, are a good example. 1 1 Similarly, the French-German versus British-Italian defense debate up to Maastricht had little bearing on the shape of Europe's future security concerns. More recently, however, new kinds of cooperation have been emerging, and current developments in European security and defense cooperation have policy implications that are particularly interesting. These need to become part of the integration picture. When these developments are seen as part of the full picture, we need to rethink some of the old questions. We want to know what theoretical perspectives are relevant to the emergence of a European defense identity that is nonetheless destined to remain locked into much broader security structures. Three immediate questions suggest themselves: (1) Is security integration necessary for economic or political union? (2) Can security integration follow a separate course? (3) Will security integration determine the parameters of "European integration"? A further question follows closely: (4) Is sovereignty (whose emblem is security) indivisible? In order to answer such questions, it is necessary to consider the nature of the new developments in European security and defense cooperation. Defense is now a constitutional function of the European Union. Maastricht's Title V allows for the creation of a "common defense" issuing from the practices of common foreign and security policymaking and, in due course, common defense policymaking. At present, the "common defense" mentioned in the Maastricht Treaty is a long way off. Also, the content of Maastricht's "common defense" has deliberately not been too closely defined. Nevertheless, Maastricht charts a course toward the goal of common European defense, and one of the specific aims of the Inter Governmental Conference review in 1996 will be to see how far this goal is being realized and how far it is realizable. Maastricht also formally links the Western European Union with the defense function of the EU. 1 2 Ten of the EU states are full members of the WEU, while Denmark and Ireland (together with the three new EU states, Austria, Finland, and Sweden) are WEU observers. Thus the Petersberg Declaration of 19 June 1992, refers to the WEU as "the defense component of the European Union." But Western European states already have an operative defense organization in NATO—albeit a "collective defense" organization. NATO's "collective defense" framework is established in the guarantee of mutual assistance set down in Article 5 of the North Atlantic Treaty. 13 This framework ensures that "an armed attack against one or more of them . . . shall be considered an attack against them all" and invokes Article 51 of the United Nations Charter, which recognizes the right to "collective self-defense." Of the twelve pre-1995 EU states, only Ireland is outside the North Atlantic Alliance (France and Spain are members of the alliance, although they are outside the integrated military command of NATO), and even Ireland is arguably covered by the NATO security

Theories of Defense

Integration

85

umbrella. 1 4 Furthermore, the trend is toward "combined" rather than common, or even collective, alliance structures in the future. The most significant indicators of this trend are the proposals made at the NATO summit in January 1994 to provide a framework for the combined operation of NATO and WEU in Combined Joint Task Forces (CJTF). 1 * The Partnership for Peace (PfP) proposals, which had a higher profile at the summit conference, further illustrate this trend by providing a framework for combined peacekeeping operations between NATO and the former Warsaw Pact countries. The CJTF initiative is used here as a case study to test the following theoretical claims: (1) that security integration follows economic and political integration (functionalist models), and (2) that sovereignty is indivisible (realist models). The first claim returns to our original three questions. The second claim, being the substance of our fourth question, touches the core of both integration and security theory. New empirical work in this area examines CJTFs as an example of the asset-sharing arrangements that are now being made between the WEU and NATO. 1 6 This provides a key to reassessments of both integration and international relations theories. In this study, we first follow the different dimensions in which security integration is taking place—that is, in new multilateral peacekeeping arrangements, in the sharing of military assets, multinationality of forces, command and control arrangements, use of intelligence, and provisions for political control. Then we examine the implications of the CJTF mechanism both for European security structures, and for European policymaking. Finally, we discuss the theoretical implications of these new developments. The issues they raise will become central to the European debate as the EU approaches new security and geopolitical configurations.

Dimensions of Security NATO and the WEU are two different methods of defense integration. CJTF, as a mechanism for combining the assets and command structures of both, represents a new departure. For NATO, the strategic implications are clearer now than they were in 1991. NATO strategy has embraced international peacekeeping as its "number two" mission (after collective selfdefense). 1 7 Through peace support operations, offered on a case-by-case basis to the United Nations, NATO will be able to act out-of-area (as it has done in Bosnia). 1 8 Thus the global context of the military alliance is gaining in importance by comparison with the territorial defense of Western Europe. This evolution is consistent with U.S. foreign policy, which has made "assertive multilateralism" a plank of its policy toward international organizations and has used this to resolve burden-sharing issues, in particular with Western Europe. 1 9 For the WEU, the strategic and policy

86

Reflections on European Integration

implications are not clear. C J T F is an offer the W E U cannot refuse. But the EU has not formulated a consistent policy with regard to this new form of military integration and may lack the political machinery to do so. Thus military integration may precede political integration in this case. Analysis of C J T F shows features related (theoretically) to other levels of European integration. Shared assets, multinationality, unified command and control, and shared intelligence count among those features. Also implied are questions of political control and accountability. The recurrent theoretical question is how far political and military integration can proceed along different paths, and whether one will be determined by the other.

NATO and the WEU: The Effect of Peacekeeping In the early days of W E U reactivation, the W E U was seen as a rival to NATO. William Taft, U.S. permanent representative on the North Atlantic Council, spoke bluntly to a series of European audiences in February 1991: "I would like to say just a word about the American public and its view of [NATO]. . . . To get specific . . . the U.S. public would not understand what was going on if the Europeans stopped using NATO or began replacing it with other structures to perform its historic tasks." 2 0 There was some U.S. interest at that time in exploiting the potential of the W E U as a mechanism whereby NATO members could act out-of-area if the case arose. While NATO itself might be prevented under Article 6 of its treaty from going out-of-area, there was nothing to prevent the W E U from doing so, if necessary with Atlantic partners. Such potential arrangements accord with the principle of "double hatting." This means having forces ready to act as either NATO units or W E U units, depending on the tasks and on the geography, as well as on the international politics of a given security crisis. As the "new world order" idea has taken shape, however, NATO has formulated distinctive plans for out-of-area action. Based on the post-Cold War explosion of demand for peacekeeping, these plans constitute a new strategic concept that embraces the WEU. They offer peace support operations both to the United Nations and to what is now a UN regional organization, the Conference on Security and Cooperation in Europe (CSCE). 2 1 "Peace support" covers a range of activities from diplomatic initiatives to full-scale military operations. Coming under the auspices of the UN, none of these will be subject to either Article 5 or Article 6 of the North Atlantic Treaty. NATO and the W E U are now collaborating on peace support operations, though they were in competition during 1992, as events in former Yugoslavia forced peacekeeping onto both of their agendas. NATO first offered support for C S C E peacekeeping operations at a North Atlantic Council meeting in Oslo on 4 June 1992. The W E U followed closely with an

Theories of Defense

Integration

87

offer of support for both CSCE and UN peacekeeping on 19 June 1992, at its Bonn-Petersberg meeting. The wording used was broadly the same as NATO's, but the W E U ' s Petersberg Declaration added that WEU forces might also be deployed for "tasks of combat forces in crisis management including peacemaking." NATO closed the gap in December 1992 by stating that: "We are ready to respond positively to initiatives that the UN Secretary General might take to seek Alliance assistance in the implementation of UN Security Council Resolutions." 2 2 Practical experience in former Yugoslavia also shaped the outcome of the institutional competition between NATO and the WEU. The EU's action in this area has not been notable for its success except in one instance. The Adriatic embargo, which the EU initiated through the WEU in July 1992 to implement the UN's resolutions, almost failed. It had the necessary naval assets but lacked the intelligence resources to carry out its mandate. In order to overcome these difficulties, it merged during its first year with a parallel operation under the command of NATO, which had access to superior surveillance systems. Enforcement of the embargo followed from November 1992, command and control arrangements were set up and tested, and the experiment was deemed a success in terms of military cooperation. 2 3 In June 1993 it became Operation Sharp Guard and has since served as a model for a more permanent procedure for combined NATO and W E U operations. This evolved into the Combined Joint Task Force concept in the latter half of 1993. At the NATO Summit in January 1994, the Heads of State and Government endorsed the CJTF concept. It was presented as an arrangement that would strengthen the European security and defense identity (ESDI), and that also offered immediate operational advantages to the WEU. Still within the global peacekeeping framework, CJTF proposed the setting up of specific task forces for "peacekeeping and other contingency" operations. Since January 1994 NATO's military commanders developed the details of the CJTF concept within the framework of NATO's ongoing force structure review. CJTF sets the tone for the future relationship between NATO and WEU activities. It is seen as part of a review of the balance of responsibilities between the European and the North American allies. 2 4 The equipping, staffing, and financing of future operations, along with new command and control arrangements, are to be decided in the context of the emerging European security and defense identity (ESDI) represented by the WEU. 2 5 The CJTF concept is designed to provide a concrete basis for the maintenance of alliance and European forces which will be "separable but not separate." 2 6 Avoiding the duplication of effort and resources that would result from distinct and independent military structures is cited as the rationale. The new concept has at its core the establishment of mobile CJTF headquarters for which specific groups of officers, from existing NATO regional headquarters, are the nucleus.

88

Reflections on European Integration

Militarily, the concept is based on the U.S. Joint Task Force Concept introduced in the late 1970s with multinationality added. The idea is to develop flexible, contingency-dependent force packages for different types of military mission (task forces). These task force packages could be drawn from any of the different armed services, land army, airforce, navy, and others (joint). They could also be drawn from a wide range of national and multinational contributors (combined). Thus C J T F caters for participation from either a narrower or a wider group of nations, rather than relying on all and only NATO members to contribute in every contingency.

Shared Military

Assets

The C J T F concept gives maximum flexibility for assigning and using assets, in that there is a case-by-case commitment only from participating states. It is important to note that NATO itself has no military assets to speak of. Assets belong to contributor states and are assigned to the organization. The proposal to set up C J T F s will enable the alliance to utilize the military assets thus assigned in peacekeeping and other out-of-area activities. Because of the peace support arrangement with the UN, there is no need to amend the North Atlantic Treaty. C J T F provides a mechanism whereby the alliance can utilize its military assets (1) for operations outside the NATO Treaty area as defined in Article 6; (2) in respect of nonArticle 5 action (not arising from the treaty's collective defense guarantees); and (3) in integrated operations with non-NATO partners. " [ C J T F ] . . . would allow new flexibility for organizing peacekeeping and other tasks. It would enable NATO to take effective action in contingencies that do not evoke Article 5 of the North Atlantic Treaty." 2 7 Specifically, C J T F is aimed at W E U participation, allowing in principle for W E U command of agreed C J T F operations. This means, for example, that French forces, which are not part of NATO's integrated command structure (but which represent one of Europe's vital assets), could participate fully in C J T F operations. It could also overcome French objections to the inevitable U . S . command of NATO operations through S A C E U R (Supreme Allied Commander Europe), if W E U command became a reality. Spanish forces would also be captured by this mechanism, and the E U enlargement process promises to bring in other assets shortly. Force-structure planning in the NATO area has been moving for some time toward the capability required by a C J T F concept. Several Multi-National Divisions (MNDs) have been created so far. O f these, the most elaborate is the A R R C ( A C E — A l l i e d Command Europe—Rapid Reaction Corps), which is under British command, and allows for a distinctive British input. French input is also distinctive and embodied in a parallel (and sometimes seen as rival) enterprise, the Eurocorps. Major advantages of the Eurocorps project in the French view are ( 1 ) legitimization of the

Theories of Defense

Integration

89

stationing of French troops on German soil; (2) integration of German troops into a European army, however embryonic, prior to an anticipated massive withdrawal of U.S. troops from Europe in the next few years; and (3) standardization of equipment between participating European armies, independent of NATO standardization. Included in the French rationale for the Eurocorps is a defense-industrial strategy. However, as will be demonstrated below, Eurocorps troops will not have operational credibility without NATO command, control, and intelligence (C 2 I) infrastructure. The NATO-Eurocorps relationship is fully within the framework of the MND, and CJTF, concepts. While the force retains its independent command for normal purposes, it accepts the operational command of NATO's integrated structure for military operations. The operational capability this gives to the fledgling European army is considerable, and the Eurocorps may thereby achieve some of the more important command roles in multinational operations where the allies do not deem it essential to have direct SACEUR command. However, NATO's senior military planners think it unlikely that any significant roles will go to the Eurocorps, even in its new NATO-friendly guise. Officials indicate that the ARRC has been earmarked as a provider of potential mobile headquarters for the CJTFs and that, unless a particular contingency makes it politically desirable to use the Eurocorps, the ARRC will be the command structure of choice for NATO's key nations. British national policy has been oriented in this direction, as it has made the ARRC its instrument for the creation of a special British role in NATO, and national policy goals have so far been met. Nevertheless, Britain will be at least as much constrained as France (which bitterly opposed British policy on the ARRC) by the outcomes of this policy process. That is, European responses to security questions will depend increasingly on U.S. willingness to support operations through the use of alliance assets, and the major European states are committing themselves to this decisionmaking framework.

Multinationality

of Forces

The task force approach gives a considerable amount of flexibility for either NATO or the WEU to choose from the military forces offered by contributing nations. This flexibility will be used partly to ensure military effectiveness and partly for political acceptability, depending on the type and location of the conflict. An obvious advantage of CJTF is that it will overcome some of the concerns military planners have in dealing with multinational task forces set up for specific contingencies. They have stressed the need for such forces to become more effective in unity of command and control, as well as common doctrine, planning, training, and the interoperability of their equipment. Nevertheless, the CJTF arrangement raises a number of questions. Using NATO-linked headquarters implies a highly

90

Reflections on European

Integration

U.S.-dominated structure. The U.S. agreement to leave its staff officers in place, even if the United States is not participating in a specific CJTF operation, may solve the problem of restaffing headquarters at short notice, but it may also cause political problems. The United States might be seen as still seeking significant political-military control, even when it is not contributing forces to an operation. In addition, there is likely to be a lack of influential staff positions in these headquarters for staff officers from outside NATO's integrated military structure (as in the case of France and Spain). There are no indications that countries that contribute a large proportion of forces to CJTF operations (as France might do) will be represented proportionately within the headquarters. Particular concern arises about the role of U.S. permanent officers in CJTF headquarters, if that headquarters should be called upon to run a WEU-commanded operation. Would these officers have too much influence over the outcome of the WEU operation? NATO military planning for peace support operations begins by "taking into account the principle of case by case decisions of the Alliance . . . and recognizing that national participation in peace support operations will remain subject to national decision." 2 8 The alliance has taken care to ensure member states' autonomy in deciding whether to sign up for specific peace support operations, and this concern for autonomy is carried over into the CJTF concept. At the same time NATO and the WEU have developed force structures, in particular the Immediate and Rapid Reaction Forces (RRF) that are largely based on the principle of multinationality, down to the divisional level and below. Since these forces are to be deployed early during a crisis, this structure will make autonomous national decisionmaking difficult. The speed of reaction of these forces, together with the degree of integration, is bound to increase pressures on the smaller states to conform with majority or large-power decisions on deployment. Member states may feel obliged to participate at the earlier stages, but may be powerless to influence the later stages of decisionmaking. While this structure is intended to signal the alliance's cohesion once the decision to deploy forces has been made, it may have negative effects on overall political cohesion, given the scope for policy divergence. Another effect may be the gap between policymaking and implementation. Member states could as a group achieve consensus on a deployment policy, for instance on a containment policy in former Yugoslavia requiring the deployment of 10,000 troops. But they could subsequently fail to produce the necessary assets to implement it in their case-by-case commitments. Such a scenario is not unknown in the UN for example. Decisionmaking procedures about peace support operations and force-package deployments do not foster cohesion. Indeed, given the number of allies and their very different geographic, economic, and political priorities, there is a high probability of dissent. In addition, decisions about crisis management may be made under crisis-typical time constraints. In such circumstances,

Theories of Defense Integration

91

it may not be possible to avoid damaging the national interests of some members. Member states may even perceive that they are being blackmailed in some way by their alliance partners. Germany, for instance, is particularly sensitive to the possibility of being pressured by alliance partners to participate in military operations that would meet with fierce domestic resistance. Perhaps the most damaging effect of multinational integration, if these decisionmaking processes do not function satisfactorily, will be a lack of decisive action. If the combined forces of NATO and the W E U cannot be used for peacekeeping on behalf of the UN because the legitimate concerns of member states lead to immobilism, both organizations will lose credibility. Unified Command and Control Under the C J T F arrangement, the W E U is integrating into structures over which it has very little control in policy terms. The PfP countries have even less. Command and control (C 2 ) arrangements are far more than a technical means to ensure the effectiveness of military operations. In a multinational environment, they also illustrate who has political and military control. Thus analyzing the C 2 arrangements envisioned for future peace support operations gives insight into national interests and political intentions. The C 2 arrangements also illuminate the relationships between the organizations that mandate multilateral military activities, those who conduct them, and the auspices under which such activities take place. The C J T F concept supplements NATO's traditional integrated military structure. While day-to-day operational planning, training, and military contacts continue to be conducted through the normal NATO command structure, the C J T F headquarters will operate in a specific contingency. Depending on the contingency, a task force from the nations committed to the task will be selected and subordinated. As long as the operation continues, both the task force and the headquarters will remain under the command of whichever body has been chosen. This will depend on the political decisions taken. C J T F is supposed to allow for the W E U to have command and control of some operations. However, as described above, the chances of C 2 being in the WEU's rather than NATO's hands are negligible. Also, the differences between the C 2 arrangements envisioned for the C J T F combinations on the one hand and the PfP combinations on the other are considerable. Command and control of the potential task forces is not even in principle extended to Central and Eastern European partners, as it is to W E U countries. If PfP countries are involved in the peacekeeping operations, the UN will be in command. Also, PfP countries are not expected to be involved in peace enforcement operations (falling mainly under Chapter Seven of the UN Charter) but only in the lower levels of peacekeeping, such as lightly armed operations to maintain existing ceasefires (under Chapter Six of the UN Charter and under UN command).

92

Reflections on European

Integration

Policy guidelines on c o m m a n d arrangements within NATO require that wherever NATO assets are deployed directly as NATO assets, NATO will be in command. If NATO assets are specifically assigned to the U N , however, then the U N is in c o m m a n d . Similarly, if NATO assets are assigned to the W E U under the C J T F arrangement, then the W E U will be in command. But this is only likely to occur if there is a political need to designate an operation a " W E U operation." In this case, U.S. troops are unlikely to be involved. U.S. troop deployment is a special problem in this context, because the U.S. policy preference for keeping U.S. troops strictly under U.S. command has not altered in the age of multilateralism. On the contrary, it has become more insistent. All of the policies of assertive multilateralism (UN peacekeeping and CJTFs included) are concerned with ensuring smooth mechanisms of cooperation with able partners, while maintaining U.S. command of U.S. troops. U.S. command through NATO is acceptable, but the UN c o m m a n d of combat operations involving U.S. troops is not. The mechanisms sought in response are either those of the peace support concept, which allow the United States, through NATO, to opt for the UN mandates corresponding to its command and control preferences (the Gulf War being the model of choice), or those of the C J T F and its P f P derivative, which preserve the option for the United States to leave selected peace operations in the hands of the W E U or the UN, while retaining significant political control and influence over the agenda. Because C 2 arrangements not only involve technical aspects of efficiency and coordination but confer political authority on those in charge of operations, it is important to note these details of the C 2 environment being created by CJTF. Those European states that are accustomed to some influence through NATO, or to being among one of the military heavyweights in the European Union, may find their positions on security matters weaker in the future. While decisionmaking in the W E U may become easier, those decisions will count for less. The W E U forum proposes, NATO disposes. Meanwhile decisionmaking in NATO may become less accessible both to the W E U and to individual European states, as the pool of countries with whom key NATO states could combine becomes larger. "Political" decisionmaking in the NATO context also may count for less. What will matter will be the ad hoc decisions on participation in particular operations, and who can most successfully control those decisionmaking processes.

Shared

Intelligence

Once again, it is important to note that just as in the case of other military assets the intelligence assets that NATO has at its disposal are mainly national assets and that the most important of them belong to the United States. Thus N A T O ' s sharing of intelligence assets depends directly on U.S. national decisions. Such decisions could veto or undermine a combined operation. This is indeed what happened with the U.S. change of

Theories of Defense

Integration

93

policy on Operation Sharp Guard described below. CJTFs are meant to operate on behalf of either NATO or the WEU. However, the WEU is at a serious disadvantage, not only with regard to command and control infrastructure but also with regard to intelligence capabilities, by comparison with U.S.-supported NATO operations. A WEU Assembly report emphasizes that in such possible WEU operations, a CJTF headquarters can be effective only "if essential collective Alliance assets such as satellite intelligence and AWACS (Airborne Warning and Control System) are available at the same time." According to the report, the WEU Planning Cell "needs to have access to NATO and national intelligence including secret material if its work is to be taken seriously. An intelligence agreement between the WEU and NATO is urgently needed." 2 9 Europe's post-Gulf War analyses, particularly French analyses, highlighted the lack of independent satellite photographic, signals, and communications capability. As a first step, the WEU set up a satellite information analysis station in Spain. Unfortunately the analysts there have been using only Landsat photos (which are technically inferior and can in any case be bought elsewhere) and, when budgets permit, the technically better French SPOT (Système Probatoire d'Observation de la Terre) photos. Nothing matching the U.S. system is yet available through the WEU. For the long term, the French government has increased the budget for the Direction du Renseignements Militaires, its newest intelligence service, from 39 million francs in 1993 to 243 million in 1994. These sums are being used to upgrade European satellite capability. In May 1994 the Federal Security Council of Germany agreed to pay up to 20 percent of the cost of the HELIOS 2, part of a new French-designed satellite system consisting of two photo-reconnaissance satellites, two radar satellites, and a released relay satellite. On a national basis, France is also buying signals intelligence satellites. Proposals for a multinational intelligence agency set up and operated outside NATO structures were made in 1994. Pressure is also being exerted to reassess the European satellite station and the capacity of the WEU to operate its own independent intelligence-gathering and communications system. However, there is little to suggest that a purely European organization could compete with what NATO has to offer. This may not appear problematic as long as a high degree of coincidence between U.S. and European interests is assumed. Yet there are many areas of significant policy divergence between Europe and the United States, even where their general interests and political intentions remain similar. Peacekeeping is precisely such an area. At present, neither individual states within the European Union nor the WEU can conduct their own operations without NATO's C 2 I infrastructure and logistic support. Thus, at a time when the EU is assuming a political identity and developing a Common Foreign and Security Policy (CFSP), the EU lacks the capacity for independent action. U.S. Secretary of Defense Les Aspin left no doubt where decisionmaking responsibility

94

Reflections

on European

Integration

would lie when he explained how CJTF is expected to function: "In the case of WEU operations, the WEU commander would have full authority even if drawing on NATO collective assets. The SACEUR would train, package and provide the assets to the WEU. NATO would make CJTF assets available to the WEU or other groups, provided that NATO remains the central forum for decision making about common security issues." 30 NATO's C 2 I is one of the critical areas that the United States has in its power to deny, not by a formal veto but a de facto one, to European partners. The late NATO Secretary General Manfred Worner was optimistic that "common WEU stances will be increasingly introduced into Alliance consultations." 3 1 But politically acceptable solutions may be hard to achieve where one partner always holds the trump card. This will undoubtedly lead to political tensions between NATO and the European core, if interests diverge, as they did in the 1980s over the Strategic Defense Initiative (SDI).

Political Control and

Accountability

Politically, the CJTF concept is ambiguous. On the face of it, the concept is intended to allow the WEU to draw on NATO assets once the decision has been made that the WEU, rather than NATO, should take action. This is presented as part of the effort to strengthen the development of a European security and defense identity and could, according to French Foreign Minister Alain Juppé, imply that NATO would accept CJTFs under WEU operational command, instead of under SACEUR. 3 2 Indeed, French foreign policy might be seen as one of the main targets of the CJTF strategy. Since French reluctance to commit its forces to NATO's integrated military command structure has remained firm, alternative mechanisms for overcoming that reluctance have had to be found, if only because France's military assets are too important to be left outside the NATO-WEU coalition framework. At the same time, CJTF is intended to be used as the format for cooperation with Central and Eastern European countries in the context of PfP. It would provide sufficient flexibility for future participation by non-NATO (and non-WEU) member countries, for example, in the framework of the North Atlantic Cooperation Council (NACC), in peacekeeping, and other similar military operations. Thus it reflects the increased demand on NATO to commit itself more strongly to pan-European security (supporting the CSCE) as well as to supporting the UN. But Central and Eastern European partners, while seeking military cooperation with the West, and accepting PfP as the best offer, are far from convinced of NATO's commitment to pan-European security. In the proceedings of their NACC working group on cooperation in peacekeeping, they have expressed disquiet on several counts, including the orientation of NATO's peacekeeping doctrine and the procedures envisioned for the negotiation of mandates and C 2 arrangements with the UN. 3 3 In some of these concerns they may find common cause with WEU partners.

Theories of Defense Integration

95

The major advantage claimed for C J T F is that of offering the WEU an operational capability much earlier than would otherwise be possible, thus contributing to E U autonomy and to the evolution of an independent CFSP. However, serious questions may also be raised in this context. Firstly, the matter of political control is not settled by declaratory statements, which place overall political control in the hands of the UN, or alternatively in the hands of individual nation-states. In defense and security policy areas in particular, political control is often exercised most effectively by technical means. How command arrangements are decided, and how intelligence is shared, are two of the key indicators. Secondly, the fundamental question of accountability is blurred in the pursuit of national interests through multinational organizations. Who is to be held accountable for an operation that is mandated by the UN, commanded by the WEU, staffed by the French and British militaries, equipped from national assets assigned to NATO, and politically controlled by the supply of military information gathered by the United States and processed by NATO? Neither the institutions of the E U , nor those of the Atlantic Alliance, nor those of the UN, are fully attuned to such eventualities. In cases where there is consent about the WEU taking action and the United States not participating in that action, the C J T F concept should be militarily sufficient. However, serious controversy could arise in cases where the WEU countries want to take action, but the United States does not. And even greater damage would result from a situation in which the United States initially agreed to a WEU operation using NATO assets and C J T F headquarters, but later withdrew from this position. W E U parliamentarians have voiced concern that, as long as "it remains to be seen to what extent NATO and the Americans will agree to waive their right of refusal," and "in the absence of procedures automatically ensuring that the assets of the Alliance will be made available to it, W E U must maintain its autonomous military planning capability and develop its own operational capability in order to act independently or at the request of the European Union." 3 4 In other words, as long as there is no commitment by NATO not to block W E U decisions by denying necessary assets, the C J T F concept will ensure overall U.S. leadership and control, while enabling the United States to carry less of the burden. In fact, the concept might be seen as a step toward maintaining NATO as the main forum of security consultation, as well as a maneuver to maintain European reliance on NATO and U.S. military assets.

Implications of C J T F for European Security Structures and Policy How does this departure in defense organization relate to European security policymaking? Analysis of decisionmaking in the European Union has

96

Reflections on European

Integration

focused on the polarized categories of the supranational and the intergovernmental. Typically, defense is viewed as an area of tension between efforts to form a European superstate, in which decisions on defense and security would emanate from supranational bodies, and the opposing efforts of nation-state governments to retain sovereignty over the institutions of their own national defense. NATO is usually seen as an exemplar of the intergovernmental, while the W E U arouses both the hopes and fears of supranationalism. Where, then, does the C J T F combination fit in? Neither EU autonomy nor member-state autonomy shows net benefit (though both get payoffs) f r o m the new arrangement. CJTF, giving as it does undue weight to an intervening variable (U.S. asset-sharing willingness), overrides the tension between supranational and intergovernmental modes of decisionmaking. It brings fully into play "the soft power of using institutions to co-opt others to share the burden." 3 5 Thus, the European security and defense identity could only be embraced officially by NATO (in January 1994) once C J T F had been institutionalized. This is because C J T F transforms ESDI, by changing the relations between NATO and W E U , and in so doing gives the United States a powerful institutional tool for co-opting European powers to share peacekeeping burdens. It is in this way that we shall attempt to analyze security structures, not as competing "architectures" but as a multilayered set of power relations between states and alliances. Multilateralism and defense cooperation are the name of the game, and the trick is organizing multilateralism in one's own interest. 3 6 There are significant strategies emerging f r o m within the EU that show that this is clearly understood at some levels. The 1993 Balladur plan for a European Stability Pact is just such a strategy. It offers Eastern Europeans alternative routes to both security and economic integration by making acceptance of the Pact a "necessary but not sufficient" condition for E U membership. The Balladur plan is competing in the field of soft power (considerably softer, some would add, than its NATO-PfP rival), but it is competing. Official awareness of the problems EU independent policymaking faces is expressed more openly than hitherto. 3 7 Institutional power is now being tested. Soft power turns into hard power under pressure of events, however, as the Sharp Guard experience demonstrated in November 1994. The effects of C J T F arrangements in crisis situations, such as those which have arisen over policy divergence on former Yugoslavia, are revealing, and will be addressed below.

European Security

Structures

There were possibilities for fundamental restructuring of NATO from January 1994, utilizing European defense identity aspirations to shape the core of a politically more balanced organization. A f o r m u l a might have been found to allow the W E U to occupy the central geopolitical space in a new NATO, which would continue to bridge the Atlantic effectively and

Theories of Defense Integration

97

which would also hammer out a new set of security arrangements with the C S C E - w i d e range of states to the East. In fact, the January summit moved toward greater caution with regard to Eastern Europe and less openness with regard to fundamental restructuring of the alliance. The emphasis remains on ad hoc coalitions—neither common nor collective security but an advanced machinery for coalition-building. Individual states may decide to participate in NATO's "peace support" operations, on a case-by-case basis. Les Aspin has pointed out how "other nations such as France—as well as peace partners over time—could establish a presence at a C J T F headquarters and commit forces for C J T F contingencies while remaining separate from NATO's integrated military structure." 38 NATO's ability to function as a broker of ad hoc coalitions, drawing together the most appropriate mix from its sixteen members for any specific contingency, not only follows logically from the force restructuring exercise but also points the way to an extension of this function via the C J T F concept. On the face of it, C J T F is simply a method of combining elements from different national forces in order to carry out specific missions. The C J T F formula adopts the pattern already established in NATO for peace support operations. States participate on a case-by-case basis. Task forces are set up in specific contingencies and last only as long as the mission they are expected to fulfill. States may choose the U N missions they want to support and those they turn down. The only difference is that C J T F is not limited to the sixteen NATO nations. This is where reconfiguration starts. Even if C J T F is limited, as it will be in the first instance, to full W E U members, all of which are also NATO members (except for Ireland, which now functions as an observer of the W E U but is unlikely to join NATO), the relationship of nation-states in Europe to each other and to existing security organizations will be altered. France (and Spain) may begin to participate fully in N A T O ' s military operations while remaining outside the integrated military structure. Theoretically Austria, Finland, Sweden, and even Ireland (the W E U observers that are unlikely to join NATO) could do so in the future. Some of the central European states may participate, via the W E U route, as their association with the E U progresses. Indeed, as the Balladur initiative suggests, military integration may come first in future or even be a prerequisite to economic and political integration. Organizing at the W E U level, before combination with NATO, relieves NATO's nonEuropean members of administrative and, most important, financial responsibility, facilitating a new approach to the problem of burden-sharing. The possibility of the W E U organizing, and even commanding, forces in combined operations, opens the field wide in politically sensitive areas. NATO assets could be used without U.S. troops being involved in missions or even present in large numbers on the European continent. This also represents a substantial reconfiguration with regard to C S C E . NATO has already assumed many of the military tasks of the C S C E

98

Reflections on European Integration

through the creation of the North Atlantic Cooperation Council. There will be more fundamental reconfiguration through the new Partnership for Peace program. The C J T F formula has been adapted to serve PfP. While the framework agreement signed by all the Central European and most of the ex-Soviet states provides for a permanent consultation, training, and transparency arrangement between NATO and the PfP signatories, there is also the possibility of undertaking peacekeeping operations at the request of the UN or CSCE. This would be done on the same basis as the CJTFs, with Central and Eastern European states participating on a case-by-case basis, and with decisions on the particular Task Force and its command structure being dependent on the contingency. Like the W E U countries faced with the C J T F offer, the Central and Eastern European countries are finding that NATO's offer is one they cannot refuse. Although it falls short of their requirements for security guarantees and membership of NATO, it may enable them to carry out peacekeeping operations that they would not be able to do themselves. Conversely, it might be very useful to NATO countries to support such operations technically and logistically without having to risk their own troops on the ground. The C J T F concept therefore implies structural changes in the decisionmaking processes affecting European security policy. In effect, these innovations amount to a reconfiguration of the major European security organizations. Recent predictions had envisioned structural changes of this order. A more European NATO in which the United States would play a smaller role was one scenario; another was a more cooperative security arrangement between the former Warsaw Pact and NATO states; yet another was a three-tier NATO in which the W E U would occupy the first tier at the European core, NATO would represent the second sixteen-nation tier, and the C S C E would take care of the vast (mainly East European) periphery or third tier. 39 In reality, the present structure is NATO at the center of a web of international peacekeeping forces. But given that the W E U is firmly woven into this web, providing potential military resources to a degree unlikely to be matched by its control over decisionmaking, this reconfiguration is particularly important for European security policy.

Policy

Implications

The reconfiguring of international security organizations that we have described raises important questions for European policymaking. European states, whether individually or collectively as the EU, are bound to be concerned about (1) their capacity to engage in independent action, (2) their capacity to disengage from multinational actions, and (3) the problem of policy divergence. This problem is most likely to occur, for geopolitical reasons, between European and transatlantic allies. But while EU members move toward the common policy positions of CFSP, none of the E S D I

Theories of Defense Integration

99

institutions, captured as they are now in the logic of the C J T F concept, can guarantee a capacity for independent action by the EU states. On the one hand, the power to engage in independent action is rendered negligible by the dependence on NATO intelligence gathering and communications. "The ARRC requires a multinational intelligence and targeting group . . . that will link intelligence assets, fuse and process their data, and ensure timely and relevant products. This raises the problems of shared intelligence and data fusion. . . . Conceptually, therefore, the 'pendulum of control' must move to favor NATO commanders in this new era of multinationality." 40 The A R R C is not unique in this respect. The same would be true of other multinational forces. The Eurocorps, despite its professed "European" profile, is operationally just as dependent on the intelligence provisions that only NATO can provide. Neither individual states nor the W E U have the capacity for conducting their own peacekeeping operations. Whether acting alone (for example, Britain in the Falklands or France in North Africa), or in an EU grouping (as in the former Yugoslavia), their military operations are increasingly dependent on NATO's C 2 I infrastructure and logistic support. These are assets that the United States has in its power to deny. But there is no European alternative, because creating such an alternative would be more expensive, and less reliable, than tapping into U.S. resources, and might even, it is argued, precipitate the large-scale U.S. troop withdrawal that Europeans fear. On the other hand, the power to disengage from multinational actions is severely reduced by the level of multinationality in NATO's new Rapid Reaction Force structure and command. Not only is it understood that the multinational force commander must have a greater say in what has hitherto been the prerogative of individual nations, but the capacity of W E U member states to remain outside NATO's ad hoc peace support operations is in doubt. NATO's military planning stipulates that no such operations will be undertaken, except in accordance with the principle of case-bycase decisionmaking by the alliance, and also stipulates that national participation in peace support operations will always remain subject to a national decision. Nevertheless, the new force and command structures taking shape around NATO's R R F concept introduce a new element into decisionmaking: the integration of forces at ever lower levels, these being the earliest deployment levels in a conflict, will make it difficult for any one partner to disengage. Multinationality now extends down to division level, in the various multinational divisions that are most likely to be engaged at the beginning of an operation. It will be difficult for individual member states to refuse participation in, say, noncombat roles in a preventive deployment or monitoring situation, if by doing so they risk endangering the enterprise. By the same token, it will become more difficult for individual members to disengage once conflict has begun, however great the escalation.

100

Reflections on European

Integration

Current prospects are for fewer U.S. forces in Europe, but for the availability of all multinational forces (many of which will of necessity be forces answerable to the WEU or Eurocorps forces) to take part in out-ofarea operations under NATO command. The successful use of NATO infrastructure, logistics, and training in the Gulf War is the template on which NATO's new peace support function is being built. This does not take account of two political issues. Who decides when to intervene and when not? And how long will European populations be willing to host the machinery of U.S. foreign policy? These political issues are the essence of the European defense identity debate. The likelihood of policy divergence raises further questions about the wisdom of counting on U.S. cooperation in cases in which there may be fundamental differences of approach. Peacekeeping doctrine, for example, is an area where wide divergences have been revealed in the course of coordinating policy on Bosnia. There is a "European" view of peacekeeping (shared by the Canadians) that makes impartiality paramount, emphasizes consent-promoting techniques, and insists on the conceptual separation of traditional peacekeeping (under Chapter Six of the UN Charter) from the strictly military peace enforcement missions that may also be mandated by the UN (under Chapter Seven). This view is well informed by the experience of European ground troops (primarily French and British) in the former Yugoslavia. The U.S. view of peacekeeping, in contrast, prioritizes "overwhelming force," conceptualizes the range of peace support operations as a continuum of "levels of violence," and posits a "middle ground" between peacekeeping and peace enforcement, which is anathema to European practitioners. Predictably, the view projected in NATO's military planning and doctrine for peace support operations (that is, the environment within which the CJTF mechanism is supposed to work) reflects U.S. rather than European doctrine. 41 This is a particularly interesting example as it provides the background both to a significant event in our CJTF case study and to an important piece of evidence about the effect of operational military integration on higher-level policymaking. As noted earlier, a successful prototype CJTF was the Adriatic embargo, translated into Operation Sharp Guard in June 1993. Technically, this operation proved impeccable. In policy coordination terms, however, it came perilously close to shattering alliance cohesion, because of the different understandings of impartiality and the mixing of peace operations. On 11 November 1994, the United States announced its intention to withdraw from Operation Sharp Guard. Unable to persuade either the UN Security Council or its European allies to lift the arms embargo on the Bosnian government (the U.S. policy preference), the United States proposed to implement a Congressional decision of 5 October 1994 which, though it would not provide arms to Bosnia, would stop the funding of the operation to enforce the arms embargo. U.S. ships would

Theories of Defense Integration

101

no longer participate fully in the Adriatic embargo. More significantly (for the French and British ships operating as forces answerable to the WEU), the United States would limit the sharing of intelligence on Adriatic shipments with its European allies, thereby rendering them deaf and blind. Such a denial of U.S. intelligence to a NATO-WEU operation, set up to enforce a UN resolution, would have reversed EU policy on Bosnia overnight. Sharp Guard illustrated graphically the exercise of political control by technical means, which is the essence of the C J T F strategy. The outcome of the Sharp Guard crisis, however, illustrated something else, which might be termed the "spillover effects of military integration." Although a U.S. domestic decision almost put an end to the participation of the European partners in humanitarian intervention in Bosnia (the EU policy preference), the resulting crisis was so severe that the U.S. administration succeeded in finessing that decision so that the arms embargo would remain unaffected for the time being. Thus the peacekeepers on the ground prevailed. EU policy gained a respite, and French and British troops did not have to scramble ignominiously out of Bosnia under NATO air cover, as they might have done had the impartial UNPROFOR mission been abandoned. In the Bosnian case, the independent European position was slow to materialize politically. Yet independent policy was operationalized through the engagement of European troops in the UNPROFOR mission. This operational strength has in turn affected U.S. policy, preventing or delaying the implementation of policies (such as extensive use of air power, or the lifting of the Bosnian arms embargo) that could undermine the UNPROFOR mission. There is some evidence from the above example that, whereas defense integration does not necessarily follow economic and political integration, political integration does follow operational military integration. For example, new levels of policymaking have ensued from Franco-British cooperation in Bosnia. Practical coordination was needed on the rules of engagement in the Bosnian field, followed by doctrinal coordination on the theory and practice of impartiality. Militaries conferred, creating a community of expert knowledge that impinged on policymaking. 4 2 Without this experience, there would have been no strong European foreign policy position, of the kind revealed in November 1994, capable of convincing a U.S. administration to tailor its Bosnian policy to European needs. There are three possible paths of development with regard to European defense integration. First, the EU may be seen as a subcomponent of the Atlantic Alliance (with the W E U as an integral pillar of NATO). This shows some signs of being developed along the lines of the "new Atlantic Alliance" theory, in which the security dimension would fall into place as just one dimension of a wider community shared by Europe and the United States. This would maintain the congruity of security and economic-political areas, though the area would not be that of Europe. Second, pressure

102

Reflections on European

Integration

from NATO or the transatlantic partnership may provoke the elusive "closer union" between European states (as did SDI in 1983 and the Reykjavik Summit in October 1986 4 3 ) in order to secure those interests that they have in common and that are different from the interests of the United States. Divergent U.S. and European policies in Bosnia have had the same effect. This would efficiently achieve a new security union congruent with the economic-political union of Europe, though it would reverse expectations of functionalist integration theory in making security the driver of political integration. Third, a path between these two alternatives would have most interesting implications for integration theory, because it would mean a successful separation of sovereignty and security.

Conclusion: Theoretical Implications In the countries of the European Union, military integration is proceeding ahead of political integration. This is clearly the case for European members of the Atlantic Alliance, for whom new layers of European military cooperation are being created through existing NATO structures. But it also applies to some of the future enlargement countries (from the former Warsaw Pact) for whom military cooperation is now on offer as a prerequisite to economic cooperation. This could turn conventional neofunctionalist integration theory on its head. Realists, on the other hand, might find it a convenient illustration of the primacy of power relations. More important, however, military integration is occurring on a different track from political and economic integration. Instead of taking place in the context of Maastricht and the European Union, military integration is taking place primarily through the C J T F mechanism and the management of the NATO-WEU partnership by the United States, which holds the ultimate decisionmaking power over the use of essential resources. In this form, military integration is unlikely to enhance the European defense identity or to transform common foreign and security policy into a common defense policy as envisioned in the Maastricht framework. Here we consider the forms military integration is taking in the EU. Neither common defense nor collective defense fit the developments initiated by CJTF, and we have therefore characterized them as "combined defense." Common

Defense

There are strong factors preventing the achievement of a "common defense" for Europe in the full sense of the Maastricht Treaty. 44 The national interests of the European nation-states are not by any means the only, or even the strongest, factor. What the present study shows is that the international security organizations and the manipulation of these organizations by dominant states, both inside and outside Europe, appear to be more

Theories of Defense Integration

103

important factors. "Common defense" implies both the common ownership of major defense assets and a common political process for decisionmaking on security. These are features usually associated with national (or supranational) sovereignty and enshrined in constitution or treaty. Without a unified political authority (similar to that, say, in the post-Civil War United States), a common defense is unimaginable. Should this combination be achieved in Europe's case, it would bear out classic integration theory in which security integration would follow political integration, which would first follow economic integration. This is the path mapped out for the European Union in Maastricht's Title V (J.4.1), the political finality of which is a larger state of Europe. It is the only model that ensures congruity between political and security institutions. But there is too much competition from the other defense integration track, as outlined above, for this outcome to be predictable. Collective

Defense

There is still a strong trend toward a European "collective defense" (based on the W E U ) taking over the primary NATO role of collective self-defense in Europe. This is a trend encouraged in a limited way by the United States as it moves toward new goals of assertive (and pluralist) multilateralism in which NATO is a more fluid concept than it appeared up to now. However, the distinction between collective defense and collective security should be borne in mind. 45 Collective security is the kind of security framework offered by the UN and C S C E . Sometimes associated with cooperative security (and its emphasis on confidence-building measures), collective security does not include security guarantees. A collective security regime aims to prevent or manage conflict within the group of signatories, rather than to defend the group against outsiders. In the civilian power sense (echoing Duchene), the E U is already a collective security regime and a successful one at that. "Collective defense" is a more traditional alliance form. It implies a territorial alliance of sovereign states in which an attack on one is deemed an attack on all (as in NATO's Article 5). It is thus based explicitly on security guarantees and (in the post-1945 world) is usually associated with integrated command structures and extensive asset-sharing, enshrined in international treaties and agreements. The W E U could assume this function easily as a subcomponent or "European pillar" of NATO. But it would continue to depend on the availability of infrastructural support that is presently not guaranteed to the W E U by NATO. Combined

Defense

As our case study shows, this collective defense function is undermined by the forcing of the new "combined" options on an underdeveloped W E U ,

104

Reflections on European

Integration

which cannot but choose the new style of cooperation offered by NATO through CJTFs. A new security configuration of combined defense arrangements looks likely to prevail in Europe in the short to medium term. This combined security configuration covers the European Union, wider Europe (to the East), and transatlantic partnerships. It separates sovereignty, defined in terms of the security-state unit, from the restructuring of political and economic sovereignty as debated in the integration of the European continent. "Combined defense" along the lines sketched out for CJTF allows for a somewhat weaker collective defense relationship, or even for several of these within the larger entity. With decisionmaking on a case-by-case basis, there would not be the same watertight security guarantees as could be presumed in Cold War NATO. Yet there could be subsets, such as the WEU states or the Visegrad Group, 4 6 which might offer each other more permanent security guarantees. Combined defense appears to maximize choice, but its force structures are constraining. It allows for extensive asset-sharing, but on an ad hoc alliance basis where the largest asset-holders call the tune. In sum, military integration is proceeding, but it is following a different track from economic and political integration. While, under the impact of the Maastricht Treaty, the WEU is beginning to provide some kind of EU-wide framework for the coordination of security policies within the Union, the military assets of the WEU are being locked into the broader structure of NATO with its dominant transatlantic dimension. Even given the strains of the post-Cold War realignments, this relationship is unlikely to be prised apart. Its disintegration is not in the interests of either European or transatlantic partners. However, this does not preclude significant policy divergence. The theoretical implications have become clear. In answering the questions we put earlier, we can at least point the way to further work. 1. Is security integration necessary for economic and political integration? The answer must be no, insofar as the Union has progressed to Maastricht and beyond with only the most slender of military-organizational ties at EU level. What might be regarded as a necessary prerequisite, however, is the existence of a security environment within which the participating states are already in some sense "integrated." NATO has clearly furnished this environment in the past. 2. If this suggests that security integration can follow a separate course, however, we have to think again. Europe cannot both follow the U.S. lead (as it did in the Gulf) and forge a common European foreign policy (as it tried to do in Bosnia) that is, for geopolitical reasons, different from that of its U.S. ally. The answer to whether security integration can follow a separate course is therefore, yes—but only so far. It seems from

Theories of Defense

Integration

105

our examples that, whereas defense integration does not necessarily follow economic and political integration, political integration may follow operational military integration (which is a variation on the functionalist paradigm). 3. Will security integration determine the parameters of European integration? Interestingly, there has been some effect observed along these lines, for example, on the processes of EU enlargement. Balladur-type security pacts have now been invented as alternative-to-NATO and prior-toEU membership requirements. In this paradigm, security integration arrangements have deliberately been made a prerequisite to economic and political integration. 4. Finally, is sovereignty indivisible? The trend toward CJTF-type multilateralism can be read either as the maintenance of nation-state sovereignty through the case-by-case decisionmaking process (which will ensure that no state is bound by the security interests of an ally state) or as the fragmentation of security interests such that allies will tend to cluster around issues in different combinations and with different levels of commitment. The second reading suggests a high degree of divisibility. A state's need to be part of a larger entity, and yet reluctant to do anything above the minimum, may paralyze policymaking in this context. It may, however, have the effect of driving policymaking toward more cooperative security structures rather than the collective defense structures of the Atlantic Alliance. A state may be sovereign to limit its involvement (e.g., no U.S. ground troops) but not to withdraw infrastructural assets once committed (e.g., U.S. intelligence in Operation Sharp Guard). Dividing sovereignty, in international peacekeeping, for example, may make nolonger-sovereign states more safe. In further study, functionalists might do well to consider the spillover effects of operational military cooperation. How should they be regarded? Like those of economic cooperation? Under what circumstances does military cooperation create new institutions for political decisionmaking? Meanwhile, the realist question about what sort of alliances the modern security state would form is one that should still exercise us. In separating sovereignty and security, are these states now forming alliances that undermine the modern security state? If this separation is occurring, European integration theory will have to take it into account and ask whether it is generalizable. If interests are now so ineffectually aggregated by the sovereign state, in particular with respect to the security of national populations, perhaps it is appropriate that the defense and promotion of security interests should be separated from the nation-state and that defense structures and policies should cease to be regarded as a necessary level of the state's apparatus.

106

Reflections on European

Integration

Notes The author wishes to acknowledge the support received in primary research assistance and resources form the British American Security Information Council, the Berlin Information Centre for Transatlantic Security, and the Centre for European Security and Disarmament in Brussels, as well as the time and expertise given by staff at NATO headquarters, SHAPE, and the North Atlantic Assembly in the background research for this chapter. Responsibility for the interpretation for this material rests solely with the author. 1. See, for example, François Duchêne, "Europe and Changing Superpower Relations," The Round Table 61 (October 1971): 577-584; François Duchêne, Europe's Role in World Peace," Richard Mayne, ed., Europe Tomorrow (London: Fontana, 1972); and Hedley Bull, "Civilian Power Europe: A Contradiction in Terms?" Journal of Common Market Studies 21 (September-December 1982): 149-170. 2. A useful perspective is provided in Mary Kaldor, The Imaginary War: Understanding the East-West Conflict (Oxford: Basil Blackwell, 1990). 3. Key texts of the debate that flourished in the 1970s are: Ernst Haas, Beyond the Nation-State'. Functionalism and International Organization (Stanford: Stanford University Press, 1964); Robert Keohane and Joseph Nye, Power and Interdependence (Boston: Little, Brown, 1977); and Stephen Krasner, ed., International Regimes (Ithaca: Cornell University Press, 1983). 4. John Mearsheimer, "Back to the Future: Instability in Europe after the Cold War," International Security 15 (Summer 1990): 5 - 5 7 . 5. Joseph Grieco, "Understanding the Problem of International Cooperation: The Limits of Neoliberal Institutionalism and the Future of Realist Theory," in David Baldwin, ed., Neorealism and Neoliberalism: The Contemporary Debate (New York: Columbia University Press, 1993), 301-338; Henry Kissinger, "What Kind of New World Order?" The Washington Post, 3 December 1991; and Richard Pipes, "The Soviet Union Adrift," Foreign Affairs 70 (1991): 70-87. 6. Stephen van Evera, "Primed for Peace: Europe after the Cold War," International Security 15 (Winter 1990/91): 7 - 5 7 ; John Lewis Gaddis, "Toward the Post-Cold War World," Foreign Affairs 70 (Spring 1991): 102-122; and Robert Keohane, "Institutionalist Theory and the Realist Challenge after the Cold War" (working paper, no. 92-97, Harvard University Center for International Affairs, 1992). 7. Jiri Dienstbier, "Central Europe's Security," Foreign Policy 83 (Summer 1991): 119-127; Jack Snyder, "Averting Anarchy in the New Europe," International Security 14 (Spring 1990): 5 - 4 1 ; and Richard Ullman, "Enlarging the Zone of Peace," Foreign Policy 80 (Fall 1990): 102-120. 8. "Western European Union: Platform on European Security Interests," adopted by the Foreign and Defense Ministers of the (then) seven W E U states at The Hague on 27 October 1987. WEU Assembly Document 1122, 27 October 1987. 9. "The common foreign and security policy shall include all questions related to the security of the Union, including the eventual framing of a common defense policy, which might in time lead to a common defense." Treaty on European Union, Title V, Article J.4.1, (Luxembourg: Office for Official Publications of the European Communities, 1991), 126. 10. Among the exceptions to this charge are: Reinhardt Rummel, "Integration, Disintegration, and Security in Europe: Preparing the Community for a Multi-Institutional Response," International Journal 47 (Winter 1991-1992): 64-92; and Mathias Jopp, "The Strategic Implications of European Integration" (Adelphi Paper 290, London: Brassey's for IISS, 1994).

Theories of Defense

Integration

107

11. "The Alliance's New Strategic Concept," Heads of State and Government participating in the meeting of the North Atlantic Council in Rome, 7 - 8 November 1991, reprinted in NATO Handbook, (Brussels: NATO Office of Information and Press, 1992), 147-160. 12. "The Union requests the Western European Union (WEU), which is an integral part of the development of the European Union, to elaborate and implement decisions and actions of the Union which have defense implications. The Council shall, in agreement with the institutions of the WEU, adopt the necessary practical arrangements." Treaty on European Union, Title V, Article J.4.2, 129. 13. "The North Atlantic Treaty," Article 5, (Washington, D.C.: U.S. GPO, 1949) 4 April 1949. 14. Austria, Finland, and Sweden, the three states that joined the EU on 1 January 1995, are not full members of the WEU. Nor are they NATO members. All assumed observer status of the W E U along with EU membership. All are active members of the CSCE/OSCE. All are signatories of PfP, and Finland acts as an occasional observer of the NACC. 15. NATO Press Communiqué M-1(94)3 (Declaration of Heads of State and Government Participating in the Meeting of the North Atlantic Council held 11 January 1994 at NATO Headquarters, Brussels). 16. This work, on which I draw for much of the case study, is detailed in Patricia Chilton Otfried Nassauer, Dan Plesch, and Jamie Patten, NATO, Peacekeeping, and the United Nations, BASIC Report 94.1 (Washington, D.C.: BASIC, 1994). 17. MC 327 "NATO Military Planning for Peace Support Operations," a decision of NATO's Military Committee taken 5 August 1993. 18. Article 6 of the North Atlantic Treaty defines specific territorial limitations to the application of Article 5. Thus NATO action outside "the territory of any of the Parties" or "the North Atlantic area" is problematic. 19. Presidential Review Directive 13 and Presidential Decision Directive 25, the published version of which is "The Clinton Administration's Policy on Reforming Multilateral Peace Operations" (Washington, D.C.: U.S. Department of State Publication 10161, May 1994). 20. William Taft, "Taft talks about European Security in New World Order," Official Text (United States Information Service, 12 February 1991). 21. From 1 January 1995 on, the CSCE has become the Organization for Security and Cooperation in Europe (OSCE). 22. NATO Press Service: M-NAC-2(92)106, Ministerial Meeting of the North Atlantic Council, NATO Headquarters, Brussels, 17 December 1992, Final Communique, Part 2, "NATO's Role in Peacekeeping." 23. "Joint Session of the North Atlantic Council and the Council of the Western European Union held in Brussels on 8 June 1993," NATO Review, June 1993, 21.

24. For background on this theme see John Oneal, "The Theory of Collective Action and Burden Sharing in NATO," International Organization 44 (Summer 1990): 379-402; and Steve Weber, "Shaping the Postwar Balance of Power: Multilateralism in NATO," International Organization 46 (Summer 1992): 633-680. 25. For discussion of the EDI/ESDI debate see Catherine Guicherd, "A European Defense Identity: Challenge and Opportunity for NATO," Congressional Research Service Report 9 1 ^ 7 8 RCO, 12 June 1991; and Alexander Moens, "Behind Complementarity and Transparency: The Politics of the European Security and Defense Identity," Journal of European Integration 16 (1992): 29^18. 26. NATO Press Service, "NATO and the W E U are Two Viable Building Blocks in Our Broader Effort to Create a New Euro-Atlantic Security Order"

108

Reflections on European

Integration

(speech by the late NATO Secretary General Manfred Wörner to the WEU Assembly, Paris, 29 November 1993). 27. Warren Christopher, U.S. Secretary of State (speech at NATO Headquarters, Brussels, 2 December 1993). 28. "NATO Military Planning for Peace Support Operations," MC 327. 29. "Assembly of the WEU: The WEU Planning Cell" (WEU-Document 1421, Paris, May 1994). 30. Les Aspin, U.S. Secretary of Defense (speech at NATO Headquarters, Brussels, 8 - 9 December 1993). 31. NATO Press Service, "NATO and the WEU are Two Viable Building Blocks in Our Broader Effort to Create a New Euro-Atlantic Security Order," (Paris speech by the late NATO secretary general, Manfred Wörner to the WEU Assembly, 29 November 1993). 32. Alain Juppé, French foreign affairs minister (speech at the Institut des Hautes Etudes de la Défense Nationale, Paris, 21 January 1994). 33. See for example NATO Press Service, Meeting of the North Atlantic Cooperation Council in Athens, Greece, Report to Ministers by the NACC Ad Hoc Group on Cooperation in Peacekeeping, M-NACC 1(93)40, 11 June 1993; and NATO Press Service, NACC Informal Working Group on Cooperation in Peacekeeping Planning: "Draft NACC Planning Principles and Guidelines for Combined Peacekeeping Operations," Brussels, 17 March 1994. 34. Mr. Baumel, rapporteur, "The Evolution of NATO and Its Consequences for the WEU," WEU-Document 1410, 23 March 1994, part 3. 35. Joseph Nye, "What New World Order?" Foreign Affairs 71 (Spring 1992):83-96. 36. Michael Brenner, "Multilateralism and European Security," Survival 35 (Summer 1993): 138-155; and Trevor Taylor, "Western European Security and Defense Cooperation: Maastricht and Beyond," International Affairs 70 (January 1994): 1—16. 37. See Willem van Eekelen, secretary general of the Western European Union, "WEU after Two Brussels Summits: A New Approach to Common Tasks," (Brussels, 27 January 1994); Karel de Gucht, rapporteur, "Report of the Committee on Institutional Affairs on the Future Relations Between the European Union, WEU and the Atlantic Alliance" (European Parliament Session Documents A30041/94, 27 January 1994); "Western European Union: Kirchberg Declaration," WEU Council of Ministers, 9 May 1994; and "Western European Union: Noordwijk Declaration," WEU Council of Ministers, 14 November 1994. 38. Summary of Statement by Les Aspin, U.S. Secretary of Defense, to Fall 1993 Meeting of the NATO Defense Planning Committee, Brussels, 8 - 9 December 1993. 39. Examples can be found in Malcolm Chalmers, "Beyond the Alliance System: The Case for a European Security Organization," World Policy Journal (Spring 1990): 215-250; Jonathan Clarke, "Replacing NATO," Foreign Policy, no. 93 (Winter 1993/94): 22-40; and Douglas Stuart, "NATO's Future as a Pan-European Security Institution," NATO Review, August 1993, 15-19. 40. Lieutenant General Sir Jeremy Mackenzie, Commander of the ARRC, "The ACE Rapid Reaction Corps—Making It Work," RUSI (Royal United Services Institute) Journal (February 1993): 16-20. 41. Primary sources for the debate on peacekeeping doctrine are: Lieutenant Colonel Charles Dobbie, British Army Field Manual, Wider Peacekeeping, 5th draft, 1994; Rapport au premier ministre, "Participation de la France aux Opérations de Maintien de la Paix," François Trucy, Sénateur du Var, Maire de Toulon, Parlementaire

Theories of Defense

Integration

109

en Mission 4 Août 1993-4 Février 1994 (Paris: Documentation Française, 1994); U.S. Army Field Manual 100-123 (Draft Version 6), January 1994; and NATO Doctrine for Peace Support Operations Draft, Change 1, document prepared by the Peacekeeping Section (SHOPP), OPS/LOG DIV, of Supreme Headquarter's Allied Powers Europe, 28 February 1994). 42. The study of this phenomenon as an epistemic community has yet to be done. See Peter M. Haas (ed.), Knowledge, Power, and International Policy Coordination, Special Issue of International Organization, 46, (Winter 1992). 43. At Reykjavik, the private agreement reached by Reagan and Gorbachev on dismantling Intermediate Nuclear Forces, without consultation with the European allies on whose territories those weapons were deployed, left Western European leaders with a sharp sense of their lack of leverage on U.S. policymaking and led directly to the drafting of The Hague Platform on European Security Interests by the W E U states in 1987. 44. Emil Kirchner, "Has the Single European Act Opened the Door for a European Security Policy?" Journal of European Integration 13 (1989): 1 - 1 4 ; Jacques Delors, "European Integration and Security," Survival 33 (March/April 1991): 9 9 - 1 0 9 ; and Anand Menon, Anthony Forster, and William Wallace, "A Common European Defense?" Survival 34 (Autumn 1992): 98-118. 45. Stuart Croft, "Cooperative Security in Europe," Brassey's Defense Yearbook 1993 (London: Brassey's, 1993), 101-116; Gregory Flynn and David Scheffer, "Limited Collective Security," Foreign Policy 80 (Fall 1990): 7 7 - 1 0 1 ; and Josef Joffe, "Collective Security and the Future of Europe: Failed Dreams and Dead Ends," Survival 34 (Spring 1992): 36-50. 46. The Visegrad Group (composed of the Czech and Slovak Republics, Hungary, and Poland) dates from 1990 and takes its name from a joint declaration on trilateral security cooperation issued after a meeting in Visegrad, Hungary, February 1991.

5 Economic Uncertainty and European Solidarity Revisited: Trends in Public Support for European Integration Christopher Anderson This chapter examines recent public opinion trends regarding the European Community in the context of existing analysis and looks in particular at the relationship between economic factors and public opinion shifts. 1 Over the past several years, the European public has become reluctant to endorse the European integration project as envisioned by policymakers in Brussels and at home. Moreover, the citizens of the European Community (EC) have repeatedly demonstrated that any further integration will be difficult without their support. Polls that track public support for European integration demonstrate that approval is waning: Support declined from a high of 81 percent among EC citizens in favor of efforts to unify Europe in 1991 to 73 percent in the fall of 1993; from a high of 72 percent who thought their country's membership in the EC was a good thing in 1991 to only 54 percent who agreed in the spring of 1994; and from a high of 59 percent in 1991 who would regret the dissolution of the EC to a mere 4 0 percent voicing regret in late 1993. 2 To make matters worse, the Danes voted against the ratification of the Maastricht Treaty in a referendum in June of 1992, and in France—one of the founding members of the European Community—less than 51 percent of the voters cast their ballot in favor of ratifying the Maastricht agreement in October of 1992. 3 It is safe to say that the European public has become increasingly aware and critical of the consequences resulting from the European integration process. 4 New initiatives in areas such as social, educational, environmental, and economic policy have led to greater levels of involvement by the EC in the way national governments formulate policy. Decisions made in Brussels, Strasbourg, and Luxembourg increasingly and more directly affect the lives of ordinary citizens. As European integration continues, the E C has become an important and salient issue on national agendas, given that popular support has been crucial for the implementation of the far-reaching initiatives embodied in the Single European Act 111

112

Reflections on European

Integration

(SEA) signed in 1985 and the Maastricht Treaty signed in 1991. In fact, as the referenda held in Ireland, Denmark, and France on the SEA and on Maastricht show, citizens have the ability and willingness to constrain, modify, or eventually forestall the integration process.

Public Opinion and European Integration It is generally acknowledged that the European Community has long suffered from a "democracy-deficit," that is, a shortage of formal and/or institutionalized opportunities for the citizens of the member states to influence the decisionmaking process within the Community. Moreover, it is commonly assumed that the early efforts to unify Europe during the 1950s and 1960s were largely a political process dominated by elites that embodied few apparent costs for the average citizen. However, the most recent stages of the integration process, such as the SEA and the signing of the Maastricht Treaty, have made citizens more aware of the direct and significant consequences that EC policies have on their daily lives. This increased awareness is partially a function of the efforts of the European Commission, under the leadership of Jacques Delors, to rally the citizens of the member states behind its proposals. The term "1992" was as much a public relations slogan as a political program. Such mundane things as the sale of beers, wines, and cheeses, and more significant issues such as the nature of welfare delivery systems and the future of Europe after the Cold War have become subjects of intense public debate since the mid-1980s. As a result, the "permissive consensus" described by Lindberg and Scheingold that seemed to characterize the interplay between European public opinion and elite action aimed at greater integration has become a more conscientious consensus and at times seems to have turned into a chorus of disharmony. 5 Early functionalist theory aimed at explaining and predicting the European integration process had little room for public opinion. The integration process was seen as an elite-driven venture, and popular opinion played only a minor role, as the following quote by Haas demonstrates: It is as impracticable as it is unnecessary to have recourse to general public opinion surveys, or e v e n to surveys of s p e c i f i c a l l y interested groups. . . . It s u f f i c e s to s i n g l e out and d e f i n e the political elites in the participating countries, to study their reactions to integration and to a s s e s s c h a n g e s in attitude on their part. 6

While it may indeed be accurate to characterize the early phase of the integration process as an elite-driven phenomenon for which public opinion was largely inconsequential, it is not clear that public opinion and citizen

Trends in Public Support for

Integration

113

involvement in the integration process in general has always been unimportant; in fact, it seems to have become increasingly important over time. A number of students of the integration process did not regard the integration process exclusively as an elite-driven phenomenon for which citizens' involvement was inconsequential. The work of Deutsch and Inglehart, for example, stressed the importance of public opinion for the integration process. Others emphasized the importance of developing a popular consensus toward the institutions of the European Community as well as an overall "sense of community." 7 According to Inglehart, public opinion plays an important role for the study of European integration because it exerts considerable influence on the decisions taken by national governments and because policies, in turn, have measurable effects on mass attitudes themselves. Put simply, European integration, pushed forward at the elite level, cannot progress without public support. Such a complex social, political, and economic process requires the active tolerance, understanding, and support of the publics. 8

Research on Public Support for European Integration Compared to studies of public opinion on national politics and policy issues, research by political scientists on public support for European integration has been less extensive. Public opinion research on attitudes toward the European integration process was for a long time considered secondary to the more important decision processes that take place at the elite level. Given the recent decline in support and the considerable difficulty elites have had trying to sell the Maastricht Treaty to skeptic populations, however, researchers have started to pay greater and more systematic attention to what Europe's citizens think about integration. Postmaterialism

and the

Economy

The literature on mass support for European integration in general, and the European Community in particular, has—until recently—been fairly onesided in its theoretical perspective. Studies of public support for European integration typically emphasized the investigation of the microlevel foundations (i.e., individual behavior), and explanations of individuals' support for the integration process were for a long time dominated by Inglehart's notions of postmaterialism and cognitive mobilization. 9 Inglehart argues that the European public has become more "postmaterialist" as a result of the economic and social changes that have taken place in advanced industrialized societies since World War II. Moreover, he contends that the European public has become cognitively mobilized. The cognitive mobilization thesis argues that "rising educational levels of recent decades, coupled

114

Reflections on European Integration

with the growing availability o f information about things happening in distant p l a c e s , is c o n d u c i v e to an i n c r e a s i n g l y c o s m o p o l i t a n o u t l o o k on the part o f Western p u b l i c s . " 1 0 T a k e n together, I n g l e h a r t ' s argument suggests that support f o r E u r o p e a n i n t e g r a t i o n — a postmaterialist g o a l — i n c r e a s e s as individuals' l e v e l s o f education and s o c i o e c o n o m i c standing i n c r e a s e . B e c a u s e individuals born after W o r l d W a r II have also b e e n found to b e m o r e likely to support European integration, Inglehart portrays the typical supporter o f European integration as young, highly educated, c o s m o p o l i tan, and postmaterialist. R e c e n t l y , however, the postmaterialist approach has c o m e under serious attack by the e c o n o m i c approach. 1 1 In f a c t , the e c o n o m i c approach has rivalled, i f not replaced, postmaterialism as the dominant approach for e x plaining public opinion toward E u r o p e a n integration. A c c o r d i n g to the e c o n o m i c explanations, support for European integration varies with evaluations o f national and personal e c o n o m i c conditions, as well as national and personal e c o n o m i c benefits derived f r o m C o m m u n i t y m e m b e r s h i p . 1 2 Support for integration goes up, so these authors argue, when the national e c o n o m y p e r f o r m s well, and when a nation b e n e f i t s e c o n o m i c a l l y f r o m Community membership. L e g i t i m a t e doubts have also b e e n raised as to whether e c o n o m i c c o n ditions during an individual's youth or current e c o n o m i c conditions c a u s e individuals to support postmaterialist goals in the first p l a c e . 1 3 Critics have charged that it is predominantly the latter. F o r the literature on support for E u r o p e a n integration this would m e a n that the r e l a t i o n s h i p b e t w e e n a c o u n t r y ' s l e v e l o f postmaterialism and its level o f support for integration is largely spurious, given that e c o n o m i c conditions drive both postmaterialism and support for integration. 1 4 W h i l e this chapter is in general agreement with the e c o n o m i c approach, it cannot, and is not intended to, resolve whether it is postmaterialism or e c o n o m i c conditions that drive individua l s ' support f o r E u r o p e a n integration. G i v e n the r e c e n t c r i t i c i s m s o f the postmaterialist thesis, however, I find it plausible to assume that e c o n o m i c c o n d i t i o n s — a t least p a r t i a l l y — d r i v e support for integration. In that c a s e , the relationship b e t w e e n postmaterialism and support f o r integration i s — at least p a r t i a l l y — a s s u m e d to b e spurious. Although almost all o f these studies have generally assumed that attitudes toward E u r o p e a n integration are m e a n i n g f u l , a contrasting view o f public opinion has questioned whether the public has meaningful attitudes toward the E C . C o n v e r s e and R o s e n a u , f o r e x a m p l e , have argued that citizens generally do not have consistent b e l i e f s on policy issues, in particular on foreign p o l i c y . 1 5 With regard to public opinion toward European int e g r a t i o n , this view would imply that the E C populations do not have consistent attitudes on E C m a t t e r s . 1 6 A n d even i f attitudes toward the E C can be measured reliably and have inherent m e a n i n g , the e v i d e n c e for key explanatory approaches for integration support, such as the postmaterialist

Trends in Public Support for

Integration

115

thesis, is mixed at best. This is also not the place to pass final judgment on these debates; suffice it to say that we are willing to assume here that public support for European integration can be measured reliably and is not simply a "nonattitude," even though public opinion toward European integration may be qualitatively different than opinions on domestic policy issues. Micro Versus

Macro

As indicated above, most studies of support for European integration have focused on the micro/individual level, that is, on explanations of individual behavior, even though there are considerable differences across countries in the aggregate levels and dynamics of support. 1 7 Yet, studies that explicitly focus on aggregate dynamics of support for European integration are rare, and the aggregate (national) differences in support as well as changes in support over time are not well understood. This scholarly neglect is unfortunate, given that aggregate dynamics in support matter greatly to politicians: 1 8 U s e f u l , and therefore consequential, opinion is aggregate. Politicians care about the v i e w s of states, districts, areas, cities, w h a t - h a v e - y o u . Individual o p i n i o n is u s e f u l o n l y as an indicator of the aggregate. For a politician to pay attention to individual v i e w s is to m i s s the main g a m e . . . . The politician must, as a matter of i m a g e , appear to be c o n c e r n e d about individuals, but aggregate opinion is what matters. 1 9

Support for European integration is a dynamic phenomenon that not only involves ups and downs within countries over time, but it also encompasses significant variation across countries. And the dynamics matter to the integration process because they matter to elites engaged in the formulation and implementation of further efforts at integration. It is of little use to the decisionmakers in Bonn, Paris, London, or Brussels—and thus less consequential to the integration process—to know that postmaterialists are more likely to support European integration. But it matters greatly to them to know whether there are more people for integration than against it or that levels of support have been increasing or decreasing systematically and in predictable ways over a specified period of time. If mass opinion matters to the integration process, it matters in the aggregate. This chapter therefore seeks to trace the macrolevel differences in support for European integration and offers empirical evidence for the importance of national economic conditions to explain changes in support for European integration over time. The approach taken in this chapter is of a descriptive and an explanatory type. The descriptive part of this chapter traces the dynamics of public support for integration, and the explanatory part systematically relates indicators of economic performance to evidence

116

Reflections on European Integration

from mass opinion surveys. The following section describes the long-run (aggregate) trends and cross-national differences in public opinion about the European integration process. We then consider the impact of national economic conditions on the ups and downs in aggregate support to examine whether the economy has a systematic impact on support levels. Finally, we consider the implications of the empirical findings and examine the role that public opinion may play in the integration process as a whole.

Determinants of Public Support for Integration As Dalton and Eichenberg point out, the dynamics of aggregate support for European integration vary across space and time. 20 In their seminal work on public opinion support for European integration, Inglehart and Rabier found that support for integration was higher among the populations of the original six members states of the Community (France, Germany, Italy, and the Benelux countries) than among the citizens of countries that joined later (particularly Britain and Denmark). 21 Based on the notion of a socialization process that begins when a country joins the EC, they argued that individuals were more likely to support the integration process the longer their countries had been members of the EC. As more citizens witness the workings of the Community and are familiarized with the benefits of EC membership, the average levels of support are expected to increase. In the 1970s, Inglehart and Rabier could explain the differences in support between the original six and the newer members (Denmark, Ireland, and Britain), along these lines. Moreover, this kind of reasoning could explain the upward changes in support among the public of the (then) nine member states. Following a somewhat different logic, Inglehart and Rabier and others have also argued that citizens at least partially blame the EC for economic performance, in particular when a country joined the EC during hard economic times. Hence, support is lower in countries that joined during the height of the economic crisis of the 1970s. 2 2 Differences across countries could thus be explained with the help of the dual factors of length of membership in the EC and economic conditions. The same factors that are used to explain differences in the levels of support for integration across countries—length of membership and economic conditions—have also been used to explain changes in support for integration over time. As mentioned above, the best known empirical regularity is the overall upward trend in support for integration that was established by (among others) Inglehart and Rabier in an analysis of data from 1950 through 1975. On the other hand, Handley finds that support declined toward the end of the 1970s and the beginning of the 1980s as

Trends in Public Support for

Integration

117

economic conditions worsened. 2 3 Almost uniformly, however, support for integration is higher today than it was during the early days of the EC. 2 4 Both length of E C membership and economic conditions are hence thought to be associated with differences in levels in support across countries and changes in support within countries over time. Consequently, length of E C membership and economic performance are thought to be associated with both levels of support and trends in support for integration. The rest of this chapter takes a new look at the earlier findings of Inglehart and Rabier. In particular, I will focus on the notions that support drifts upward over time and that economic hardship can halt or even reverse this upward trend.

Patterns of Public Support for European Integration 25 The public opinion data on which this chapter is based are f r o m public opinion surveys conducted by the Commission of the European C o m m u nities on matters dealing with the EC and European integration. These surveys have been conducted on a twice-yearly basis since the early 1970s and are based on random national samples in each of the m e m b e r states. These so-called Eurobarometer polls include a set of identical questions in each survey. Because questions concerning European integration have been included in each survey since 1973, we can track aggregate levels of support over time and across countries in a systematic and comprehensive fashion. 2 6 How, then, has public support for European integration developed over the past two decades? Mass support for European integration can be measured in a number of ways. Fortunately, the battery of questions that has been consistently asked since the inception of the Eurobarometer polls has included items that seek to tap different dimensions of support. The most important ones are reviewed and analyzed here. Earlier research has identified two dimensions of support for European integration: diffuse/affective and evaluative/utilitarian. 2 7 The diffuse/affective dimension of support is expected to measure "the contemporary vision of the 'idea of Europe,' without tying answers to any specific political or economic institutions." 2 8 The evaluative/utilitarian conceptualization of support, on the other hand, involves the "calculated appraisal of the immediate costs and benefits of membership in the Community." 2 9 Three questions that tap both the affective and utilitarian dimensions of support for European integration were utilized in this chapter: 3 0 1. In general, are you for or against efforts being made to unify Western Europe?

118

Reflections on European

Integration

2. Generally speaking, do you think that (your country's) membership of the Common Market is a good thing, a bad thing, or neither good nor bad? 3. If you were to be told tomorrow that the European Community (Common Market) had been scrapped, would you be . . . very sorry, indifferent, relieved? Table 5.1 provides the average levels of integration support in the EC member states over the 1973-1992 period as well as the rank of the individual countries in terms of support (1 means the country has the highest level of support, 12 means it displays the lowest average level of support). Overall, mean levels of support for the ideal of European integration are uniformly high, ranging from an average of 83 percent in Italy to 47 percent in Denmark, with most countries lying somewhere in the 65-75 percent range. Generally, support for the (least utilitarian) European integration ideal is highest, followed by support for a country's EC membership, where the range spans from 77 percent support for EC membership in the

Table 5.1 Country

Belgium Netherlands Luxembourg Germany France Italy Britain Denmark Ireland Greece Spain Portugal

Support for European Integration: 1973-1992 (average support in percentages) European Integration 3

European Community''

Sorry if EC was scrapped 0

Avg.

Rank

Avg.

Rank

Avg.

Rank

72.2 76.9 79.8 77.0 78.4 83.0 62.9 47.3 64.4 68.6 71.2 62.2

6 5 2 4 3 1 10 12 9 8 7 11

62.8 77.6 76.7 58.8 61.1 73.1 37.8 38.8 57.2 53.9 59.4 48.9

4 1 2 7 5 3 12 11 8 9 6 10

37.6 51.8 59.5 49.4 46.8 49.6 25.9 29.6 43.6 37.3 43.9 38.8

9 2 1 4 5 3 12 11 7 10 6 8

Average Rank

7 3 1 5 4 2 11 12 8 9 6 10

Source: Compiled by author from Eurobarometer : Public Opinion in the European Community (Commission of the European Communities, Directorate-General Information, Communication, Culture, Brussels). Notes: a. "In general, are you for or against efforts being made to unify Europe?" Percentage of respondents answering "for." b. "Generally speaking, do you think that (your country's) membership of the European Community (Common Market) is a good thing, a bad thing, or neither good nor bad?" Percentage of respondents answering "good thing." c. "If you were to be told tomorrow that the European Community had been scrapped, would you be very sorry, indifferent, relieved?" Percentage of respondents answering "very sorry."

Trends in Public Support for

Integration

119

Netherlands to 38 percent in Britain. 31 Scores of regret if the Community were dissolved are lowest across the board; here Luxembourg has the highest average support with 60 percent, and Britain is again lowest, with a mere 26 percent saying (on average) that they would be very sorry if the Community were dissolved. The pattern that emerges from these numbers is similar, but not identical, to the one observed by Inglehart and Rabier during the 1970s. With regard to integration in general and EC membership more specifically, the populations of the original six members of the EC: France, Germany, Italy, and the Benelux countries, have higher support scores than the rest of the Community. However, Spain is one notable exception, ranking ahead of Germany on support for EC membership. The overall pattern also holds on the most specific measure of EC membership and EC cooperation: regret if the EC were dissolved. Here, the only exception among the original six is the low rank of Belgium, which ranks 9 out of 12 on this measure. The public in Luxembourg, the Netherlands, Italy, Germany, and France would most regret the dissolution of the EC, while the Belgian public on average expresses less regret than Spain, Ireland, and Portugal. The results of average levels reported here are mostly in line with the conventional wisdom that the populations of the original six display higher levels of support for Europe. The only deviations from the pattern are the generally high level of Spanish support and the low level of regret of the Belgians if the EC were to be dissolved. Conversely, the citizens of Britain and Denmark are consistently the lowest supporters of integration. Ireland (which joined the EC with Denmark and Britain in 1973) ranks consistently higher than Denmark and Britain, as do the newer members Portugal and Greece. The average support scores for European integration between 1973 and 1992 show that there are three distinct groupings with regard to the levels of support for European integration: first, the original six plus Spain; second, Ireland, Greece, and Portugal; and third, Denmark and Britain. The picture is thus a bit more differentiated than the simple "original six versus the rest" dichotomy employed by Inglehart and Rabier and commonly used in the description of support levels for European integration. However, averages can only tell part of the story of public support for an integrated Europe, especially when these averages are calculated over a time span that extends as far as twenty years. While averages are good summary indicators, they do not describe the ups and downs of public support. What long term trends—if any—can be observed? Figure 5.1 graphs responses to the three questions averaged across the Union as a whole between 1972 and 1992. The one feature that stands out most prominently from the graphs is the clear difference between the levels of the three series. This indicates that respondents are less likely to support integration the more a question

120

Reflections on European Integration

Figure 5.1 Public Support for European Integration (all EU member states) 90

1

1

r

n

I

r

80

.2 o

S 3

60 i\ I\

50

/

CO

40 30

Integration

72

74

—I 76

1

78

1

80

1

82

'

84

'

86

Membership

I

Scrapped 90

92

94

Year

requires them to calculate the trade-offs of integration and E C membership. When citizens are prompted to reflect on the costs and benefits of E C membership, they are more critical of the integration project (in the aggregate). Another pronounced property that stands out is the fluctuations in support, although the three series do appear to move together most of the time (and—not shown here—in most of the countries as well). The average Pearson correlation between the three series across the twelve member states is r = 0.68. There appears to be somewhat of an upward trend in support that is dampened in 1990, after which support declines. The overall upward trend since 1970 is interrupted by a low point in the early 1980s. In order to see more clearly whether there are significant bumps in the series within individual countries, I calculated support for a country's membership in the E C between 1973 and 1992 at three-year intervals. 3 2 The results are shown in Table 5.2. The table shows upward trends in support for E C membership in the E C member states but also significant country-specific irregularities. Especially in France, Britain, and Ireland, support for EC membership drops significantly in the late 1970s/early 1980s after which it rises again. Countries that have a more linear upward trend in support for E C membership include Belgium and the Netherlands, while support in Germany, Luxembourg, and Italy is more stationary. The upward trend in public support for integration is important. Yet there is a serious problem with thinking that public support for integration

Trends in Public Support for Integration

Table 5.2

Country

Belgium Netherlands Luxembourg Germany France Italy Great Britain Denmark Ireland Greece Spain Portugal

121

Support for Membership in the European Community: 1973-1992 (in percent) 3 1973-1976 1977-1980 1981-1984 1985-1988

61.3 68.4 72.1 58.0 61.7 71.7 38.6 34.7 53.0

60.6 78.4 77.4 61.4 55.9 72.1 30.4 34.8 55.5

—

—

—

—

—

—

56.3 76.9 76.0 55.0 56.0 69.6 29.3 32.9 45.4 41.3 50.9 26.4

1989-1992

66.9 80.3 80.8 57.3 68.9 76.3 41.0 39.3 59.9 50.3 62.3 57.2

69.9 84.0 76.9 63.0 63.3 75.9 51.4 53.9 73.9 72.7 66.9 69.0

Change 1973/19761989/1992 +8.6 +15.6 +4.8 +5.0 + 1.6 +4.2 + 12.8 + 19.2 +20.9 +31.4 +16.0 +42.6

Source: Compiled by author from Eurobarometer: Public Opinion in the European Community (Commission of the European Communities, Directorate-General Information, Communication, Culture, Brussels). Note: a. Respondents answering "good thing" when asked "Generally speaking, do you think that (your country's) membership of the Common Market is a good thing, a bad thing, or neither good nor bad?"

is increasing with time and is higher in "older" EC member states. There is a ceiling of 100 percent support for integration. If—all else being equal— support for integration is higher in the older member states and if there is an upward trend in support, some states will reach this ceiling faster than others, unless the upward trend is much steeper in the newer member states. The fact that there is a ceiling may consequently have an impact on the nature of the trend in support observed in the EC member states. A straightforward way to establish the nature of the upward trend in public support for European integration in the EC member states is to calculate the linear trend in support between 1973 and 1992. It can be calculated using OLS regression techniques with the help of the following model: Support^ = a + B * Time + e

t

where a and 13 are constants and e t denotes a random error with zero mean. The trend is given by 6, indicating the change in the mean level per unit time, in this case the periods between two consecutive Eurobarometers. 33 Two pieces of information can be gleaned from this calculation. The trend provides a measure of (1) whether the trend in support has indeed been upward, and, if so, (2) how steep it has been over the 1973-1992 period. We can thus calculate in which countries the movement toward greater levels of support has been strongest.

122

Reflections on European

Integration

Figure 5.2 plots the trends—expressed by the coefficient 13—of the three series for each of the EC member states. 34 The plots display a very consistent and illuminating configuration with regard to the rate of increase in support for integration, which are steeper the more recently a country has become a member of the Community. Conversely, the longer a country has been a member of the EC, the less steep the increases in integration support. The results point to a leveling off phenomenon with regard to support for European integration. In fact, increase in support appears to have come to a standstill or have been reversed in two countries of the original six. Luxembourg and Germany both have slightly negative trends in support for integration in general (Germany also displays a negative trend in regret if the EC were dissolved). Furthermore, the rates of increase over time are smallest in the other member states of the original six. On the other hand, the rates of increase in support are highest in the three newest members of the Community, Greece, Spain, and Portugal with Ireland, Denmark, and Britain somewhere in the middle. This finding does not necessarily come as a surprise. Longer membership in the EC may lead to higher levels of support for integration as Inglehart and Rabier had predicted. However, it appears that some of the older EC states have reached the ceiling of support for integration. This ceiling effect may have important consequences for elite efforts aimed at deepening European integration. Overall, the results of average support scores and trends in support show mostly consistent patterns across time and space. The public of the original six member states and Spain are most favorably disposed toward European integration, while the British and Danish public consistently show the lowest level of support. Moreover, support for integration in general (the ideal) is almost uniformly higher than more specific measures of support for EC membership. There is an upward trend in support for integration in most countries, but this trend seems to have leveled out in some of the oldest member states (Germany and Luxembourg). With regard to the cross-temporal fluctuations, increases in support during the 1970s slowed and even slightly reversed in several countries during the early 1980s, after which support increased again. The next section of this chapter constitutes an attempt to explain the cross-temporal differences in support for European integration.

Economic Conditions As mentioned above, Inglehart and Rabier and others have argued that support for European integration varies with the business cycle. Support is expected to be lower during bad economic times and to increase when the economy performs well. In order to test whether this is indeed the case, we

123

Trends in Public Support for Integration

Figure 5.2 Linear Trends in Public Support (or European Integration a. Support (or Integration

GB DK I R L

B

OR ESP

P

NL GB DR IRL ESP GR

P

Country

b. Support (or EC

D

I

L

F

B

Country

c. Regret i( EC Were Dissolved

.s a

GB NL

B

F

DR

Country

I

IRL f

124

Reflections on European Integration

can correlate measures of economic performance with support for integration. The economic variable utilized here is the annual change in GDP growth in each member state prior to the date of the individual Eurobarometer survey. Table 5.3 shows the simple bivariate Pearson correlation coefficients of GDP growth and support for the period between 1973 and 1992 in each member state. There are basically two ways to read Table 5.3. First, we are interested in the direction of the correlation between support and GDP growth. A positive correlation indicates that support is higher when the economy performs well, and, conversely, that support declines when economic performance slumps. A negative coefficient means that positive economic growth translates into lower levels of support for integration. Second, we are also interested in the statistical significance of the relationships. These are denoted by the double asterisks in Table 5.3. The correlation analysis shows that national economic performance is positively related to integration support measures in the overwhelming number of cases. The better a country's economy performs, the generally higher the public support for integration and EC membership and the greater the likelihood that citizens would regret the dissolution of the EC. Put simply, when the economy performs well, people support European integration. However, this relationship is not uniformly statistically significant and is not positive for all countries. General support for European integration is statistically related to economic performance in Germany, Belgium, the Netherlands, Luxembourg, Ireland, Portugal, and Spain. The relationship is positive and statistically significant in all these countries except for Luxembourg and positive in all E C member states except for Luxembourg and Denmark. The strongest positive correlations can be found in Spain and Portugal. Thus the populations in the most recent member states react most strongly to changes in economic conditions when asked about the Community. The negative coefficient for Luxembourg case can be explained with regard to the particular position that the country occupies in Europe and the Community. Its economy is so heavily dependent on those of its neighbors (Germany, France, Belgium, and the Netherlands) that a downturn in economic performance would actually cause a rise in pro-European attitudes. When the Luxembourg economy goes sour, Luxembourgers see the EC as an avenue for help. The significance of the relationship of economic performance and public support for the EC is reduced for the EC membership and regret questions. Only Ireland, Portugal, and Spain display statistically significant positive relationships on the E C membership question, and Ireland and Belgium exhibit a strong positive relationship between economic conditions and regret if the EC was scrapped. The countries that display the most consistent correlations of economic performance indicators and public support for integration are thus

Trends in Public Support for

Table 5.3

125

Integration

Pearson Correlation C o e f f i c i e n t s of Support for European Integration and G D P Growth: 1973-1992

Country

Germany France Italy Belgium Netherlands Luxembourg United Kingdom Denmark Ireland Greece 3 Portugal 3 Spain 2

European Integration 0.36** 0.11 0.20 0.37** 0.39** -0.35** 0.01 -0.24 0.41** 0.05 0.55** 0.56**

(31) (31) (31) (31) (31) (32) (31) (31) (31) (24) (20) (21)

European Community 0.22 0.15 0.04 0.26 0.12 0.11 -0.04 -0.15 0.41** 0.10 0.57** 0.60**

(38) (38) (38) (38) (38) (38) (38) (38) (38) (23) (24) (25)

Sorry If EC Was Scrapped -0.09 0.22 0.90 0.41** 0.15 0.02 -0.19 -0.25 0.49** 0.02 -0.41 -0.35

(29) (29) (29) (29) (29) (29) (27) (29) (29) (23) (14) (14)

Source: Compiled by author from Eurobarometer: Public Opinion in the European Community (Commission of the European Communities, Directorate-General Information, Communication, Culture, Brussels). Notes: a. Data for Greece, Portugal, and Spain from 1980 to 1992. **: Significant at the 0.05 level

Ireland (all three questions), Portugal and Spain (both on integration and EC membership), and Belgium (integration and regret if the EC was scrapped). The fact that the populations in the oldest member states do not appear to apply much of a cost/benefit calculus when evaluating Community membership is noteworthy in this regard. This may point to a split among the member states; the public of late joiners may be more critical of the integration project in times of economic hardship than the public of the original six. Overall, the relationships between national economic performance and support for integration are overwhelmingly positive but by no means breathtaking. What they do indicate, however, is that the relationship between economic uncertainty and support for integration is not necessarily straightforward or omnipresent. It should not come as a surprise that there are probably other factors such as the domestic political environment or subjective perceptions of economic conditions that mediate the relationship between the economy and public opinion. The analysis of such factors is an avenue for research that has hardly been pursued but would be worth considering in future research with the help of a multivariate model.

Economic Uncertainty and European Solidarity Revisited: Some Conclusions Public support for European integration varies widely across countries and over time. Moreover, public opinion has come to play an increasingly

126

Reflections on European

Integration

important role in the European integration process. This chapter has attempted to describe and explain such differences and aggregate movements in support for European integration across countries and across time. Thus, it sought to supply a systematic and broad explanation for general processes that underlie the fluctuations and differences in levels in support for integration. The chapter does not—and cannot—account for "individual-country" stories of variations in support. Instead, it focuses on the general phenomena of time and economic performance that presumably impact populations in general. This chapter also constitutes an attempt to revisit key propositions about the dynamics of public support for integration as they were formulated by Inglehart and Rabier. They found that while support for integration increased with time (length of EC membership), levels of support also varied as a result of domestic economic developments. There is considerable evidence to suggest that length of membership and economic performance have an impact on what the peoples of the member states think about the Community. 35 On the one hand, there has been an upward trend in support for integration efforts; on the other hand, this upward trend has at times been tempered, especially during times of economic hardship. I found that support for European integration increases with favorable economic performance and decreases during times of economic hardship. Moreover, support is consistently lower in some countries than in others. The finding that levels of support for integration are higher the longer a country has been a member of the EC explains differences in integration support within countries over time and differences in the levels of support across countries. The mechanism by which time has an influence on integration support is less clear than the fact that time matters. We can only speculate that there is a general socialization process associated with EC membership; more research is needed in this area. However, Inglehart and Rabier's findings are also modified in this paper. The use of the length of EC membership to explain the dynamics of public support for European integration involves the following problem: Support can only rise so high, and upward trends in support for European integration are slowed over time or even reversed as support levels approach 90 percent of the population. Support thus cannot rise much further; instead it levels off after a while and may even decline temporarily with downturns in economic performance. The analysis also found that the relationship between economic performance and support for integration is not equally strong across all member states. Countries such as Portugal, Spain, Ireland, and Belgium display greater levels of significance with regard to the impact of economic conditions on support for integration. The relationship between support for integration and national economic performance is reversed in the case of Luxembourg.

Trends in Public Support for

Integration

127

Some Implications What do these results mean for the connection between the European public and the integration process? T h e significant impact o f national e c o nomic performance on public support for integration indicates that the public's approval of the E C project is subject to national political and economic circumstances. We agree with "almost all of the major theorists who . . . concur that favorable economic payoffs are conducive to—and perhaps even essential t o — t h e processes of national and supra-national integrat i o n . " 3 6 Favorable national e c o n o m i c and political circumstances are needed for the integration project to proceed with the support of the European public. Similarly, we have known for a long time that national politics has played an important role in referenda on E C measures. T h e recent referenda on the Maastricht Treaty are but the latest examples. 3 7 This part o f our analysis echoes the perspective o f researchers who emphasize the impact o f national politics on the integration p r o c e s s . 3 8 When e c o n o m i c times are good, citizens are more supportive o f the integration project; when economic times are bad—as they have been for several years, especially when measured in unemployment figures—citizens are reluctant to endorse a deepening o f integration. I f the populations o f newer m e m b e r states employ more o f a cost-benefit calculus when it comes to evaluating E C membership, we can expect the populations of the states that are set to join the E C to be more influenced by national economic conditions as well. But public opinion and public involvement in the integration process need not be obstacles to the completion o f a project that was initially based on elite action and consensus. 3 9 In fact, the results of our analysis would suggest that the European public may b e c o m e more supportive of the integration process the more familiar they are with the workings of European institutions and the consequences o f E C membership. Naturally, the immediate prospects for the integration process are unfavorable if European e c o n o m i e s continue to slump. T h e most recently available data from the Eurobarometer polls taken in 1993 and the first half of 1994 would support this contention. Support for E U membership has dropped further during 1 9 9 3 and early 1 9 9 4 in light o f only sluggish e c o n o m i c recovery across Europe, coupled with persistently high and increasing unemployment levels. However, hopes and fears about the Single Market also seem to have reached some sort of equilibrium in 1993 and 1994. A majority of the E U ' s citizens ( 5 4 percent) say that they feel very or rather hopeful about the Single Market, over 35 percent who say that they are very or rather fearful o f it. As o f m i d - 1 9 9 5 , these numbers have held steady for a year and a half. 4 0 It is important to note, though, that the impact o f economic conditions on support for integration is relatively unrelated to citizens' experiences

128

Reflections on European Integration

with the EC. The analysis of the length of EC membership demonstrates this point. This means that national politics and national economic performance are not the only factors affecting public support for integration. In fact, our analyses demonstrate that experience in terms of EC membership plays an important and independent role. Thus, it is sensible to conclude that both national and supranational politics matter to the European public. The fact that public support for integration may hit or has already hit a ceiling would indicate, furthermore, that we could expect some convergence of public opinion toward the EC across member states or differences in public opinion on EC matters that do not distinguish countries by length of membership but by some other criterion. One way to test this proposition is to examine public opinion about the Maastricht Treaty. Table 5.4 shows public support for the Maastricht Treaty, averaged across the two Eurobarometer surveys conducted after the signing of the treaty by the member states. The table is noteworthy for several reasons. It shows that there was considerable variation in support for Maastricht across the member states and that, on average, the original six member states were more supportive of the treaty than the later joiners. Thus 85 percent of the Italian public supported the treaty while only 41.5 percent of the British public did, with most countries in the 6 0 - 7 5 percent range. However, the table also shows that these differences in support are relatively small. Even more important, the member states display considerable heterogeneity in support. In fact, the original six versus the rest dichotomy appears to have been broken on the Maastricht issue. Thus, the Irish, Greek, and Portuguese populations are as supportive of the Maastricht Treaty as the Belgian, Dutch, and Luxembourgish populations. What is more, French and German public opinion is among the least supportive of the Maastricht Treaty—a finding that is stunning, given the importance of these two countries for the integration project. It is entirely possible, of course, that the French and German people were aware, or believed, that their countries would have to underwrite Maastricht and/or lose sovereignty, while Greek, Irish, and Portuguese citizens felt that their countries would benefit from further integration. However, this would not contradict the notion that the ceiling in support has been reached and that future integration will deal to a much greater extent with distributive issues. Suggestions for Further

Research

There are several issues that remain unresolved with regard to the analysis of public opinion dynamics regarding European integration. First, the way in which the economy influences public support for European integration has not been sufficiently studied, given that attitudes toward integration may be qualitatively different than attitudes toward national

Trends in Public Support for

Table 5.4

129

Integration

Support for the Maastricht Treaty: 1992-1993 (in percent) 3 (average support of two surveys taken in late 1992 and early 1993)

Country

Support in percent

Rank

Belgium Netherlands Luxembourg Germany France Italy Great Britain Denmark Ireland Greece Spain Portugal

78.5 79.0 70.5 60.0 54.5 85.0 41.5 55.5 74.5 76.5 68.5 73.5

3 2 7 9 11 1 12 10 5 4 8 6

Average Original Six Average Later Six

71.3 65.0

5.5 7.5

Source: Compiled by author from Eurobarometer: Public Opinion in the European Community (Commission of the European Communities, Directorate-General Information, Communication, Culture, Brussels). Note: a. Respondents answering " f o r " when asked "If there were a referendum on whether to agree to the Maastricht Treaty or not, would you vote for or against?"

governments. Moreover, the performance of the national economy may be only one of several economic variables that should be taken into account when investigating the dynamics of integration support. Certainly, membership in the Community brings with it economic benefits in the form of subsidies and net returns from the Community's budget. Along with indicators of national economic performance, future studies need to consider such economic benefits as well. Second, aside from understanding the role of time and economic conditions and benefits, we need to systematically incorporate "political" variables into explanations of public support for European integration. These political variables will almost certainly include phenomena at the national as well as the international levels, such as national elections, referenda, the ideological preferences of governments in member states, and the extent of a country's dependence on the success of EC institutions and policymaking. Some promising efforts have already been made in this direction. 4 1 Third, it is not clear by what mechanism length of EC membership translates into increased levels of support for European integration. Fourth, subsequent work needs to establish more systematically the role that public opinion plays in the integration process as a whole. Finally, and most fundamentally, the issue of whether attitudes toward the European Community are meaningful remains unanswered. It is possible, if not likely, that the extent to which the European public is able

130

Reflections on European

Integration

to make sense of the integration process has changed over time. This would mean that there has been a trend toward greater awareness and attitude consistency over time aside from the upward trend in levels of public support. As support for integration reaches its ceiling in a greater number of states, elites will face the task of managing and maintaining the support of populations that are ever more familiar with the EC. Their support is indispensable to the success of an integrated Europe.

Notes 1 . 1 w i s h to t h a n k t h e C e n t e r f o r A d v a n c e d S t u d y in t h e B e h a v i o r a l S c i e n c e s , S t a n f o r d U n i v e r s i t y , f o r its h o s p i t a l i t y w h i l e w r i t i n g t h i s c h a p t e r . T h a n k s a l s o to Cary Funk and Karl Kaltenthaler for useful discussions and Carolyn R h o d e s and S o n i a M a z e y f o r h e l p f u l c o m m e n t s o n an e a r l i e r d r a f t . 2. E u r o p e a n C o m m i s s i o n , Eurobarometer: Public Opinion in the European Community 4 0 ( D e c e m b e r 1993): 1 2 - 1 3 ; a n d E u r o p e a n C o m m i s s i o n , Eurobarometer: Public Opinion in the European Community 4 1 (July 1994): 8 - 9 . 3. I r e l a n d a l s o h e l d a r e f e r e n d u m o n t h e M a a s t r i c h t T r e a t y . It e a s i l y r a t i f i e d t h e treaty in J u n e 1 9 9 2 b y a v o t e of 6 4 to 3 6 p e r c e n t . D e n m a r k v o t e d a s e c o n d t i m e in t h e s p r i n g of 1 9 9 3 a n d a l s o f i n a l l y r a t i f i e d t h e treaty. 4. Joseph Janssen, "Postmaterialism, Cognitive Mobilization, and Public S u p p o r t f o r E u r o p e a n I n t e g r a t i o n , " British Journal of Political Science 21 ( 1 9 9 1 ) : 443-468. 5. L e o n L i n d b e r g a n d S t u a r t S c h e i n g o l d , Europe's Would-Be Polity ( E n g l e w o o d C l i f f s : P r e n t i c e H a l l , 1970). 6. E r n s t H a a s , The Uniting of Europe ( S t a n f o r d : S t a n f o r d U n i v e r s i t y P r e s s , 1968), 17. 7. S e e K a r l D e u t s c h et al., Political Community and the North Atlantic Area: International Organization in the Light of Historical Experience (Princeton: P r i n c e t o n U n i v e r s i t y P r e s s , 1957); K a r l D e u t s c h et al., France, Germany, and the Western Alliance: A Study of Elite Attitudes on European Integration and World Politics ( N e w York: Scribner, 1967); R o n a l d Inglehart, " C o g n i t i v e M o b i l i z a t i o n a n d E u r o p e a n I d e n t i t y , " Comparative Politics 3 ( 1 9 7 0 ) : 4 5 - 7 1 ; R o n a l d I n g l e h a r t , " C h a n g i n g V a l u e P r i o r i t i e s a n d E u r o p e a n I n t e g r a t i o n , " Journal of Common Market Studies 10 ( 1 9 7 1 ) : 1 - 3 6 ; a n d R o n a l d I n g l e h a r t , The Silent Revolution (Princeton: P r i n c e t o n U n i v e r s i t y P r e s s , 1977). 8. N o t e , h o w e v e r , that p a y i n g attention to t h e role of p u b l i c o p i n i o n in t h e int e g r a t i o n p r o c e s s is n o t e q u i v a l e n t to s t r e s s i n g t h e n e e d to a l l e v i a t e t h e d e m o c r a c y deficit. Public opinion toward, and public perceptions of, the integration process m a y p l a y an i m p o r t a n t r o l e i r r e s p e c t i v e of w h e t h e r e f f o r t s a r e b e i n g m a d e to f o r m a l l y i n c l u d e t h e c i t i z e n s of t h e m e m b e r s t a t e s in C o m m u n i t y d e c i s i o n m a k i n g to a greater extent than has previously been the case. 9. R o n a l d I n g l e h a r t , " A n E n d to E u r o p e a n I n t e g r a t i o n ? " American Political Science Review 61 ( 1 9 6 7 ) : 9 1 - 1 0 5 ; I n g l e h a r t , " C o g n i t i v e M o b i l i z a t i o n a n d E u r o pean Identity," 4 5 - 7 1 ; Inglehart, " C h a n g i n g Value Priorities and European Integration," 1-36; Janssen, "Postmaterialism, Cognitive Mobilization," 4 4 3 ^ 6 8 . 10. R o n a l d I n g l e h a r t a n d J a c q u e s - R e n e R a b i e r , " E c o n o m i c U n c e r t a i n t y a n d E u r o p e a n S o l i d a r i t y : P u b l i c O p i n i o n T r e n d s , " Annals of the American Academy of Political and Social Science 4 4 0 ( 1 9 7 8 ) : 86.

Trends in Public Support for

Integration

131

11. Harold Clarke and Nitish Dutt, "Measuring Value Change in Western Industrialized Societies: The Impact of Unemployment," American Political Science Review 79 (1991): 10-27; Thomas Trump, "Value Formation and Postmaterialism: Inglehart's Theory of Value Change Reconsidered," Comparative Political Studies 24 (1991): 365-390; Raymond Duch and Michael Taylor, "Postmaterialism and the Economic Condition," American Journal of Political Science 37 (1993): 747-779; Paul Abramson and Ronald Inglehart, "Education, Security, and Postmaterialism: A Comment on Duch and Taylor's 'Postmaterialism and the Economic Condition,'" American Journal of Political Science 38 (1994): 797-814; and Raymond Duch and Michael Taylor, "A Reply to Abramson and Inglehart's 'Education, Security, and Postmaterialism,'" American Journal of Political Science 38 (1994): 815-824. 12. See Russell Dalton and Richard Eichenberg, "Economic Evaluations and Citizen Support for European Integration" (working paper, Annual Meeting of the American Political Science Association, Washington, D.C., 1991); Russell Dalton and Richard Eichenberg, "A People's Europe: Citizen Support for the 1992 Project and Beyond," in Dale Smith and James Ray, eds., The 1992 Project and the Future of Integration in Europe (Armonk, N.Y.: M. E. Sharpe, 1992), 7 3 - 9 1 ; Richard Eichenberg and Russell Dalton, "Europeans and the European Community: The Dynamics of Public Support for European Integration," International Organization 47 (1993): 507-534; Christopher Anderson and Karl Kaltenthaler, "Time, Economics, and Public Support for European Integration, 1973-1992" (working paper, Annual Meeting of the Southern Political Science Association, Savannah, Georgia, 1993); Matthew Gabel and Harvey Palmer, "What's Behind All the Commotion? Explanations for Cross-National and Cross-Sectional Differences in Public Support for European Integration" (working paper, Annual Meeting of the Midwest Political Science Association, Chicago, Illinois, 15-17 April 1993); Matthew Gabel and Harvey Palmer, "Understanding Variation in Public Support for European Integration," European Journal of Political Research 27 (forthcoming, 1995); Russell Dalton and Richard Eichenberg, "Economic Perceptions and Citizen Support for European Integration" (unpublished manuscript, University of California, Irvine, 1994); Christopher Anderson, "Public Support for Membership in the European Community, 1980-1992" (working paper, Ninth Annual Conference of Europeanists, Chicago, Illinois, 31 March-2 April 1994); and Matthew Gabel, "All Workers Are Not Created Equal: Examining Occupation-Based Differences in Public Support for European Integration" (working paper, Annual Meeting of the American Political Science Association, New York, 1 - 4 September 1994). 13. If, indeed, European integration is a postmaterialist goal. 14. See also Janssen, "Postmaterialism, Cognitive Mobilization," 443-468. 15. See Philip Converse, "The Nature of Belief Systems in Mass Publics," in David Apter, ed., Ideology and Discontent (New York: Free Press, 1964), 206-261; and James Rosenau, Public Opinion and Foreign Policy (New York: Random House, 1960). 16. Martin Slater, "Political Elites, Popular Indifference, and Community Building," in Loukas Tsoukalis, ed., The European Community: Past, Present, and Future (Oxford: Basil Blackwell, 1983), 6 9 - 8 6 . 17. Dalton and Eichenberg, "A People's Europe, 73-91. 18. Notable exceptions are the studies by Inglehart and Rabier, "Economic Uncertainty and European Solidarity, 86; Eichenberg and Dalton, "Europeans and the European Community," 507-534; and David Handley, "Public Opinion and European Integration: The Crisis of the 1970s," European Journal of Political Research 9 (1981): 335-364.

132

Reflections on European

Integration

19. James Stimson, Public Opinion in America: Moods, Cycles, and Swings (Boulder: Westview Press, 1991), 12. 20. Dalton and Eichenberg, "A People's Europe," 73-91. 21. Inglehart and Rabier, "Economic Uncertainty and European Solidarity," 86. 22. See Janssen, "Postmaterialism, Cognitive Mobilization," 443-468. 23. See Handley, "Public Opinion and European Integration," 335-364. 24. Inglehart and Rabier take note of the nexus between long-term membership, economic conditions, and support for integration in their analysis and conclude that The results of these analyses clearly demonstrate the preponderance of long-term effects over those of the current economic environment, at least as measured by these standard economic indicators. The economic context is important. Nation-specific effects, apparently linked with longterm membership, are even more important. (Inglehart and Rabier, "Economic Uncertainty and European Solidarity," 77-78) Thus, they resolve the differences across time and space in favor of national differences and long-term membership effects. 25. The analysis presented in this section is similar in spirit to that presented in Anderson and Kaltenthaler, "Time, Economics, and Public Support." 26. At the time of writing, the public opinion data are available up to the first half of 1994. Since the economic data are not available consistently for all member states after 1992, the analysis is restricted to the end of 1992. Note, however, that this analysis covers the longest time-span of any since Inglehart and Rabier's paper. 27. See Lindberg and Scheingold, Europe's Would-Be Polity, Handley, "Public Opinion and European Integration," 335-364; Werner Feld and John Wildgen, Domestic Political Realities and European Unification (Boulder: Westview Press, 1976). 28. Miles Hewstone, Understanding Attitudes Toward the European Community (New York: Cambridge University Press, 1986), 24. 29. Ronald Inglehart and Karl-Heinz Reif, "Analyzing Trends in West European Opinion: The Role of the Eurobarometer Surveys," in Karl-Heinz Reif and Ronald Inglehart, eds., Eurobarometer: The Dynamics of European Public Opinion (New York: St. Martin's Press, 1991), 7. 30. Note that these are also the questions most frequently analyzed by others and for which the most complete data series are available. 31. The only exception is the Netherlands where the scores for integration in general and EC membership are approximately equal. 32. Because the EC membership question series is the most complete series of the three, it is utilized here. 33. Compare Christopher Chatfield, The Analysis of Time Series: An Introduction, 4th ed. (New York: Chapman and Hall, 1989). 34. The actual numbers for this figure are given in Appendix 5.1 on page 134. 35. See, however, Harvey Palmer and Matthew Gabel, "National Interest, Public Opinion, and European Integration: A Political-Economy Model of NationalLevel Support" (unpublished manuscript, George Mason University, 1994), for a critical assessment of "length of membership" as an explanatory variable. 36. Inglehart and Reif, "Analyzing Trends in West European Opinion," 7. 37. See Mark Franklin, Michael Marsh, and Lauren McLaren, "The European Question: Opposition to European Unification in the Wake of Maastricht" (working

Trends in Public Support for

Integration

133

paper, Annual Meeting of the Midwest Political Science Association, Chicago, Illinois, 15-17 April 1993). 38. Compare Feld and Wildgen, Domestic Political Realities; and James Caporaso, The Structure and Function of European Integration (Pacific Palisades: Goodyear, 1974). 39. See also Haas, The Uniting of Europe; Wayne Sandholtz and John Zysman, "1992: Recasting the European Bargain," World Politics 42 (1989): 9 5 - 1 2 8 . 40. European Commission, Eurobarometer: Public Opinion in the European Community. 41. Gabel and Palmer, "Understanding Variation in Public Support"; Dalton and Eichenberg, "Economic Evaluations and Citizen Support"; Gabel, "All Workers Are Not Created Equal"; and Christopher Anderson, "Public Support for Membership in the European Community: Explaining the Cross-National Differences" (unpublished manuscript, Rice University, 1994).

134 A p p e n d i x 5.1

Country

Germany Netherlands Luxembourg Belgium France Italy Britain Denmark Ireland Greece Spain Portugal

Reflections

on European

Integration

L i n e a r T r e n d s in S u p p o r t for E u r o p e a n I n t e g r a t i o n : 1 9 7 3 - 1 9 9 2 ( b a s e d on: S u p p o r t t = « + P, + s t ( t = t i m e ) , v a l u e s s h o w n a r e for the s l o p e p.) European Integration 1

European Community 2

Sorry if E C was scrapped^

Trend

Trend

Trend

-0.02 0.12 -0.20 0.60 0.26 0.30 0.55 0.55 0.58 1.12 1.30 2.65

0.07 0.41 0.17 0.29 0.20 0.15 0.42 0.55 0.56 1.88 0.79 2.57

-0.18 0.09 0.03 0.11 0.13 0.23 0.06 0.20 0.36 1.66 0.85 1.18

Source: Eurobarometer : Public Opinion in the European Community (Commission of the European Communities; Directorate-General Information, Communication, Culture; Brussels). Notes: 1. "In general, are you for or against efforts being made to unify Europe?" Percentage o f respondents answering "for." 2. "Generally speaking, do you think that (your country's) membership o f the European Community (Common Market) is a good thing, or neither good nor bad?" Percentage of respondents answering "good thing." 3. " I f you were to be told tomorrow that the European Community had been scrapped, would you be very sorry, indifferent, relieved?" Percentage of respondents answering "very sorry."

6 Political Group Cohesion in the European Parliament, 1989-1994 Joanne Bay Brzinski

One of the notable political changes in Europe since World War II has been the growing integration and cooperation in the European Community. The Community has begun to take on powers and responsibilities formerly characteristic of nation-states. The extent of this integration within institutions of the Community and the prospect of further integration have been a central concern among those studying Europe. Evaluating the extent of transnational cooperation in the European Union and overall integration of member states is a difficult if not impossible task. This chapter has a limited goal, to look at cooperation within political groups or factions in the European Parliament between 1989 and 1994. The European Parliament is an interesting setting in which to look at cooperation among elites from different nations because it was designed specifically around partisan and ideological rather than national interests. Its structure is organized to facilitate transnational cooperation. The Parliament is not organized by nationality, and indeed all the delegates to Parliament from one single nation rarely if ever meet as a national group. 1 Instead the Parliament is organized by partisanship in parliamentary political groups. Members of Parliament are elected, rather than appointed, in regular European elections (held every five years). In contrast, other institutions in the European Union are either appointed by national governments (e.g., the Commission) or represent the governments themselves (e.g., the Council of Ministers). This organization was meant to emphasize the common interests and perspectives that cross national boundaries. It also was designed to represent the full range of viewpoints and ideological perspectives in the Community. More than in any other European institution, the Parliament can claim to reflect and represent the citizens in the Union. Parliamentary political groups or factions have been given a large and vital role within the European Parliament. Political groups are primarily

135

136

Reflections on European

Integration

responsible for selection of Parliamentary leaders, setting the Parliamentary agenda, and selecting committee delegations. Political groups that include members from many member states benefit under Parliamentary rules. Groups with members from a single state must have twenty-three members to form, while those including more member states have lower thresholds (with two member states, 18 members; and with three or more member states, 12 members). The transnational membership of these political groups makes them an ideal setting in which to examine cooperation between national elites. Nonetheless, there are limits to the role of political groups. Those limits are due to the role of national parties in European elections and to the role of the parliament in the European Union (these limits are discussed in more detail below). The limited sanctions and rewards controlled by the political groups impede their ability to enforce group unity and discipline. Have political groups managed to achieve unity and cohesion despite those barriers? What is the reason for that unity? These questions are addressed in this chapter. I have chosen to look at parliamentary political groups rather than the transnational parties. Transnational parties are extraparliamentary organizations of partisan interests at the European level that originated in European party federations in the 1970s. They reformed into political parties following the Maastricht Treaty's strong statement of support for European political parties. Their function is both to serve as a forum for exchanging ideas among like-minded parties and to promote common action among national parties. In the 1994 European elections, for instance, the three transnational parties (Socialists, European People's Party [the Christian Democrats or PPE], and the Liberals) each wrote a manifesto used to some extent by member parties. However, the transnational political parties are still underdeveloped compared to the parliamentary political groups, and their role in the European Union is unclear. Political groups already have a defined role (though of limited power) in the policymaking process. They also have ongoing and relatively secure funding because of their role in the European Parliament. The political groups reflect a larger number and more varied ideological positions as well. The transnational parties, in contrast, have an uncertain role and depend heavily on the political groups and national parties for funding. 2 In addition, only the largest political groups have a corresponding transnational party. Generally, we may conclude that the role of extraparliamentary political parties in European integration is growing but remains uncertain. In addition, the success of political groups in achieving cohesion can be gauged in ways that transnational parties cannot. Because the political groups have been active and involved in the Parliament, the potential of cooperation between elites is more readily and credibly evaluated by looking at the actions of group members in the Parliament. Analyses of national parliamentary parties frequently focus on their ability to achieve

Political Cohesion in the European

Parliament

137

voting majorities. Parliaments and assemblies make authoritative statements by voting on bills and resolutions. This is true in the European Parliament as elsewhere. The ability of political groups to achieve internal party cohesion and discipline in parliamentary votes is an indication of their success or failure. I use this criterion to judge whether political groups have developed unity and cohesion as a group. Explaining the variation in voting cohesion across political groups will illuminate the conditions and factors that facilitate integration.

Data and Measures The data for this analysis relies primarily on a randomly selected set of roll call votes in the European Parliament between 1989 and 1994. Most votes in the European Parliament involve a simple show of hands. The Parliament only takes roll call votes when the outcome of the vote is unsure, when an absolute majority of members is needed, 3 or when twentythree members of a group request a roll call vote in advance. Only votes that are formally requested by groups are recorded and published in the Official Journal of the European Communities. The population of votes used, therefore, is only a portion of all votes taken in the Parliament. However, roll call votes frequently are called by groups on votes they consider important. Groups request a roll call vote when they want an official record of the vote, when they want to check their own members' votes, or when they want to publicize the position of another group. The European Right has been particularly assiduous in requesting roll call votes. The limited number of votes recorded does introduce some bias into the votes sampled. 4 One of the problems with this sampling strategy is that all votes are included in the sample, even relatively routine or broadly accepted votes. Also, political groups emphasize cohesion primarily on votes that they care about. The approach taken in this chapter does not indicate the importance of those votes. Using all roll call votes reported in the Official Journal of the European Communities as the universe of cases, 73 votes from the 1989-1994 Parliament were randomly selected. 5 In analyzing group cohesion, I used only votes when more than one third of the members (173) of Parliament were present and when at least 10 percent of those present did not vote with the majority. Fifty-five votes met those criteria. I used those votes to calculate an index of agreement for each group. This index is a measure of voting cohesion in the political groups. It is adapted from a measure widely used to evaluate cohesion in national legislatures, the Rice party cohesion measure. 6 That measure takes the percentage of party members supporting the majority position (for or against a measure) and subtracts the percentage in the minority, producing a score

138

Reflections on European

Integration

from 0 to 100. This measure is, however, inappropriate for use in the European Parliament. There, abstentions are fairly frequent and seem to be meaningful but would be ignored using the Rice measure. Members may vote for or against a measure but may also officially abstain from voting (and those abstentions are recorded with votes for and against the measure). Interviews revealed that abstentions frequently (but not always) were used by members who disagree with the group position but did not wish to vote against it. 7 Excluding abstentions from the analysis would, therefore, underestimate internal disagreement in the groups. Fulvio Attina uses a variant of the Rice measure to evaluate political group cohesion in the European Parliament, which takes abstentions into account. 8 Three voting options are considered in his analysis: voting for a measure, against a measure, and abstaining. His index of agreement "is the percentage measure of the relation between (a) the difference between the highest numbering modality and the sum of the other two modalities in a vote by the MEPs of a Group, and (b) the total number of votes cast by the Group" 9 :

IA=

highest modality - sum of the other two modalities total number of votes cast by the group

x 100

As a result, the index of agreement used in this analysis ranges from - 3 3 to 100. 10 A score of 100 indicates perfect unanimity among voting members in the group. Positive scores indicate that a majority of the group voted together. A zero score indicates that exactly half of the group members did not vote for the modal position. Negative scores indicate that most group members did not vote for the modal position. A score of - 3 3 indicates that the group was perfectly divided between the three options. It should also be noted that measures of group cohesion measure unity within the group, independent of the votes by other groups. In the analysis, the modal vote of the group is considered the group position. This is open to question. The fact that most of the group voted in one way does not mean that the group actually instructed its members to vote in that way (though in all cases where I had any information on group positions, the modal position and the group instructions corresponded). It would be helpful to have copies of the group whip lists (i.e., the voting instruction) to indicate both the actual group position and the intensity of group instructions. In the absence of such information, the sampling strategy and measurement strategy were the best options available. From among the 55 valid votes, for each group only votes with at least half of the group members voting were used to measure cohesion. The number of votes that qualified varied by group, ranging from 5 to 26. For each vote with sufficient group participation, an index of agreement was calculated. Those indices were averaged by group to determine an average index of agreement. These are the values reported in this analysis. On

Political Cohesion in the European

Parliament

139

average 14 votes were used for each group to calculate the index of agreement. Because the number of votes used to determine the index of agreement is small, the findings are somewhat tentative. The sample of roll call votes is also used to look at the problem of attendance at plenary sessions. In that analysis all 73 votes are used. Most votes lost in calculating the index of agreement were characterized by high levels of absence in the Parliament and within the groups. This alone suggests the need to look at absences. Absences are also interesting because groups spend time and resources to mobilize their memberships. The high level of absences indicates a failure by political groups to enforce discipline among their members. In addition to the roll call votes, members of the full-time staff of the two largest groups, the Socialists and the PPE, were interviewed during the last two weeks of June 1994. The interviews included the senior members from each group. I spoke with four members of the PPE and six in the Socialist group. The staff of the political groups are chosen from among the Parliamentary secretariat, i.e., civil servants attached to the Parliament. A small number of Parliamentary staff (about 500) are attached to the Parliamentary groups. 1 1 The members of the secretariat have generally been present in the Parliament and/or the European Community for much longer than the Members of the European Parliament (MEPs) and are responsible to the political group, rather than to national parties. 1 2 These factors give them a unique long-term and transnational perspective about parliamentary groups and on the conflicts between national and transnational interests. Though the interviews were open-ended and unstructured, all were asked about six things: 1. The group's process for deciding on voting positions and for resolving conflicts on votes within the group 2. The conflicts (if any) between national interests and group positions 3. Ideological ties and differences within groups, and their relationship to national interests 4. The effects of absences at plenary sessions and turnover in the group on party cohesion 5. Changes in the Parliament and political groups since Maastricht 6. The relationship between political groups and transnational parties In this chapter, the interviews are used primarily to confirm the findings of the roll call analysis and for anecdotal evidence.

The Role of Political Groups in the Parliament The organization of the Parliament on the basis of partisanship rather than nationality is meant to encourage and facilitate cross-national cooperation.

140

Reflections on European

Integration

It organizes members based on shared principles and interests rather than national divisions. The Maastricht Treaty considered the partisan organization of Parliament to be a crucial model in the further integration in Europe. "Political parties at the European-level are important as a factor for integration within the Union. They contribute to forming a European awareness and to expressing the political will of the citizens of the Union." 1 3 Because the MEPs are directly elected, the membership of the Parliament reflects the ideological diversity of European political parties and European citizens. The political groups during the 1989-1994 session are listed in Table 6.1 (roughly by size). 14 In most groups, a common position is chosen on all votes, and group members are given instructions for votes during the plenary session. It is appropriate, therefore, to evaluate adherence to group positions. The Rainbow Group is a notable exception to this rule. It was originally organized after the 1979 elections to give a role to independent MEPs and to factions too small to qualify as a political group. Its formal title is, in fact, the Group for the Technical Co-ordination and Defense of Independent Groups and Members. Because much of Parliament's business is conducted through political groups and because political groups have access to extra staff and offices, the Rainbow group offered advantages to members who would otherwise be nonaligned. The largest faction in the group is made up of regional parties, but it also includes members without national partisan ties and single members from national parties. Consequently, the group is extremely heterogeneous, and it does not attempt to develop common positions in most instances. Two major shifts in group membership occurred in 1992 and early 1993. The members of the European Democratic Group (EDG) and the British and Danish Conservatives joined the PPE in 1992, and the EDG ceased to exist. The Italian Communists, the major members of the Group of the European Left (EUL), joined the Socialist Group in January 1993. Other members of that group dispersed, some joining Left Unity, others becoming nonaligned, still others following the Italians into the Socialist Group. These political groups are included in the analysis, using votes until the time the groups dissolved. In addition to these major shifts, there are other isolated shifts in membership. Where possible, 1 have tried to reflect those shifts in the analysis. Why might one expect cohesion, either among parties or in political groups? There are two basic reasons that parties are thought to be cohesive. 1 5 First, party members generally share ideological positions and a basic political philosophy. Their shared world view leads them in most cases to hold similar positions on issues. Second, the organization of political system affects party cohesion. When rules and processes reward parties for acting cohesively, cohesion tends to be higher. The costs of

Political Cohesion in the European Parliament

Table 6.1

141

Political Groups in the European Parliament: 1989-1994 3

Political Group Names

Abbreviation

Nature of Member of Parties

Socialist Group

PSE

Socialist and Social Democratic Parties

The European People's Party

PPE

Christian Democratic and (in 1992) Conservative Parties

Liberal, Democratic, and Reformist Group

LDR

Liberal and Centrist Parties

European Democratic Group

EDG

Conservative Parties

The Greens

—

Green and Alternative List Parties

Group of the European Left

EUL

Eurocommunists

Group of the European Democratic Alliance

EDA

Gaullist and other Right-Wing Parties

Technical Group of the European Right

ER

Nationalist Parties

The Left Unity Group

LU

Orthodox Communist Parties

ARC

Regionalist and Antimarket Parties

The Rainbow Group

Source: Compiled by the author from information available in the Official Journal of the European Communities (1989-1994) and the European Parliament, List of Members (Luxembourg: Office for Official Publications of the European Communities, 1992 and 1994). Note: a. The bulk of the E D G ' s members joined the PPE in 1992. EUL ceased to exist in January 1993 when its largest member, the Italian Communist Party, joined the Socialist Group.

defection from the party position vary depending on whether the system is parliamentary or presidential, on the electoral system, on legislative rules and organization, and on internal party rules and practices. Epstein notes that the reasons for cohesion can vary: "In order that a party can fulfill a collective governing responsibility, its elected office-holders plainly have to agree regularly on common policies, either from conviction or from expediency." 16 Certainly parties differ in the importance placed on both of those elements. Some parties have fewer shared principles. Some political systems give parties more sanctions and rewards to ensure discipline. These differences are generally thought to be related to differences in cohesion.

142

Reflections on European

Integration

The European Parliament and its political groups mirror in many ways national parliaments and national political parties. Like parties in a party system, political groups reflect many ideological positions and include all of the major party families in European politics. 1 7 The political groups play a prominent role in the work of the European Parliament in much the same way that parties do in national parliaments in Europe. 1 8 The political groups are involved in selecting Parliamentary leaders, in setting the parliamentary agenda, and in allocating speaking time and committee assignments. They have their own staffs and offices, providing support and information for their members. The schedule of the Parliament includes a group week, three or four days before the plenary sessions, in which groups discuss votes and issues on the plenary agenda. The group weeks are used to decide voting positions for the group on all plenary issues. During the plenary session itself, time is allocated for further group meetings. Groups are seated together in the Parliament as well. The role given to political groups and their prominence in Parliamentary business is seen by many as the first step toward a federal and sovereign European Union. These features of political groups make it likely that they will be cohesive to some extent. Both ideological ties within groups and the importance of political groups in parliamentary organization suggest that some level of group cohesion is likely. The parallels with national parliaments and parties are limited, and there is good reason to expect that political groups will be less cohesive than parties. Although the political groups play a large role in the Parliament, the Parliament and the political groups do not have many of the resources and powers held by national parliaments and political parties. They lack many of the organizational mechanisms that enforce party cohesion in national legislatures. 19 The European Parliament is not a legislative body but a consultative one. The Council of Ministers holds the legislative power in the Union. As a consequence, Parliament's decisions are advisory and have been disregarded by the Council and the Commission in many instances. The European Union also lacks a government requiring the support of Parliament, though the Parliament does approve the Commission president. The process by which member states chose the president of the Commission in 1994 (without formal participation and against the preferences of the Parliament) demonstrated that the choice is driven by national interests rather than Parliamentary concerns. The fact that there is no government to support eliminates one of the motivations for cohesion found in many parliamentary assemblies. When there is a government to support, party members may follow the party position because failing to do so may lead to dissolution of the parliament. Groups in the European parliament have a fixed term in office unrelated to voting with the group. The Single Europe Act and the Maastricht Treaty both expanded the role of the Parliament,

Political Cohesion in the European

Parliament

143

introducing cooperation and codecision procedures to force other European institutions to take Parliamentary views more seriously. 20 Nonetheless, the Parliament continues to have a smaller role in the European Union than most national parliaments. Political groups also have fewer powers than national political parties and few ways to discipline their members in comparison with national parties. Part of this stems from the fact that there is no government to support. As mentioned above, the penalties for ignoring group positions are less than in a national parliament, where lack of party discipline can lead to new elections. The existence of national parties, and their role in European elections, also limits the abilities of political groups to control their members. Party lists and district candidates are chosen by national parties, and elections have generally revolved around national issues and priorities rather than MEPs' actions in the Parliament. For ambitious politicians, participation in the European Parliament can be a handicap rather than an asset because of this emphasis on national politics in selection of party lists. The weakness of political groups in this regard is best illustrated by an example from the Socialist group during the most recent election. National political parties, rather than the political groups, decide on party lists for the European elections. Their decisions about the list, and the ranking of members on the list, are frequently influenced by national rather than transnational interests. The result is high turnover during election periods. In the 1994 election, six of the eight Dutch representatives in the Socialist group were excluded from the top of the list. One Socialist noted: That is a sign to members: Don't bother about working your socks off in Strasbourg and Brussels, but just keep in with the party nationally. And for the French, it was even worse. It is appalling that only three or four of the existing members go back on the list. Particularly, if it had been a punishment for bad behavior, absenteeism, but it wasn't. It was a punishment for presence. We have a pretty hefty turnover in the group as a whole, and in the Parliament it must be over 5 0 percent. It is not an encouragement. 2 1

The decisions on electoral lists do not reflect or contribute to the needs of the political group but only those of the national party. Literature on parties and candidates has generally emphasized reelection as the most important motivation for individual members of parliaments. Though parliamentarians may also have more substantive policy goals, fulfillment of those goals also requires reelection. 2 2 In the U.S. context, where parties have little or no control over nominations, this has meant relatively low levels of party cohesion most of the time. Elsewhere, when parties do control nominations, party loyalty is rated more highly by members of parliament, and voting discipline is also much stronger. The fact that political groups do not control whether members of the European

144

Reflections on European

Integration

Parliament are renominated nor their list position should lead to weaker group cohesion and voting discipline. There is also more diversity within political groups than within national parties. Though political parties in different nations may share similar names (e.g., the French Socialists and the Spanish Socialists), unique historical and national experiences have made these parties distinct from one another. Therefore, political groups include a heterogeneous set of parties. 23 Members from two ideologically different Socialist parties may not share common views about issues, also impairing group cohesion. The European Parliament uses political groups as a primary means of organization. It gives political groups a great deal of power to set agenda and appoint leaders in the Parliament. Groups do, therefore, have a great deal of control over positions in the European Parliament and over some "perks" of office, and groups do have the means to exert social pressure on members through regular and close proximity. Group meetings and internal resources encourage participation in political groups and cooperation among their members. However, these resources do not give political groups the ability to sanction members by denying them reelection or the right to run in European elections independent of national political parties. The ability of groups to sanction or reward members is limited and excludes the most potent means of compelling discipline. These conflicting elements of Parliamentary organization lead one to expect moderate levels of cohesion within the groups. Despite the expectations of only moderate levels of cohesion, the political groups in the Parliament generally show very high levels of agreement in voting, much higher than one would expect given the absence of sanctions and paucity of rewards they can provide to members who follow the group position (see Table 6.2). All except three groups have scores of more than 80 on measures of cohesion. This means that on average, more than 90 percent of the members of these groups vote for the group position. The Rice measure of party cohesion provides a point of reference for comparison, though it is not identical. In the United States, with notoriously low levels of cohesion, party cohesion measures have ranged between 55 and 89, generally hovering around 70. 2 4 Parliamentary parties evidenced higher levels of party cohesion, generally achieving scores of over 90. 25 Though the European Parliament does not have a government to support, and despite the fact that political groups have little ability to sanction defection, the level of voting cohesion of groups in the European Parliament is much closer to that found in European parliamentary systems than in the U.S. Congress. The notable exception to the generally high level of cohesion is the Rainbow group (ARC). Its score is only 25, significantly below the others. The Rainbow group has more diversity in its membership than any other

Political Cohesion in the European Parliament

Table 6.2

145

Group Cohesion in the European Parliament: Average Index of Agreement Political Groups

Index of Agreement Score

LU EUL ER PSE PPE Green EDG EDA LDR ARC

98.6 91.5 91.2 90.4 85.3 85.0 84.5 74.9 70.8 25.9

group in the Parliament. It also deliberately does not enforce cohesion. The heterogeneity of the group and the absence of a national partisan base among some members make it unlikely that this group would be cohesive. In general, the groups with low levels of cohesion are not dissimilar in most ways from those with higher levels. The three lowest vary by size. The Liberal, Democratic, and Reformist Group (LDR) is the third largest group, averaging 47 members, and includes 10 countries and between 16 and 17 parties during the parliamentary term. 2 6 The Rainbow Group is among the smallest groups, averaging 14.5 members, 8 or 9 member states, and 10 or 11 parties during the parliamentary term. The Group of the European Democratic Alliance (EDA) is somewhere in between the two with 21 members but with only 4 nations and 4 parties included among its membership. The three groups with the most cohesion are small, but their rates of cohesion are virtually the same as the largest group, the Socialists. There is also diversity among those with the lowest levels of cohesion in party dominance and ideology. The EDA is dominated by the French Gaullists, a condition that would lead one to expect higher levels of cohesion because most group members share membership in one national party. The LDR and Rainbow Group, in contrast, have no clearly dominant party or national group. Ideologically, there is also no clear pattern. Parties of the right are found both among the most and least cohesive groups. Though EDA represents parties of the right, the Rainbow Group has no particular ideological position, and the LDR is generally thought to be center right. The European Right is unexpectedly among the top three groups in cohesion. It could be expected that levels of cohesion differ across groups in the Parliament, but those patterns of cohesion bear further investigation. To understand how and why groups have come to cooperate, it is important to understand the varying levels of cohesion among the political groups.

146

Reflections on European

Integration

Explaining Differences Across Groups What factors are likely to influence cohesion of political groups in the European Parliament? I investigated three possible influences: the nationality of group members, ideology, and the overall size o f the group. National interests dominate in most institutions in the European Community. Though the Parliament is not organized along national lines, the role o f national parties is still great. National parties select candidates running in European elections, and those campaigns generally focus on national priorities. In most cases, parties rather than individual M E P s j o i n Parliamentary groups. Most M E P s have national partisan experience and continue participating in party activities while in the European Parliament. Political ambitions o f M E P s generally involve participation in national politics and parties rather than in the European parliament and the political groups. T h e s e factors give national parties influence over M E P s . In some cases, therefore, though the political group cannot enforce group cohesion, the national party can encourage disciplined voting. When there are few national parties in a political group or when there is one party with the largest share of seats in a political group, group cohesion is likely to be higher. Ideology may be a factor in group cohesion. Parties o f the Left share a clearly articulated political philosophy and a great deal o f practice in cross-national cooperation. This was noted, for instance, in Converse and Pierce's study of France. They found that Socialists and Marxists had considerably higher levels of discipline than other parties. 2 7 Marxism provides a coherent and relatively well-defined ideological starting point. Both Marxist and Green ideologies emphasize international cooperation and international consequences o f actions. Those in the Socialist Group emphasized the ideological connections in the group, developed both through the written canon and long contact with one another. " W e must remember that all the parties that belong to the Socialist Group belong also to the International Socialist Party. In some way the ideology is the same for everybody. You have some very important points that are used as ideals in every country. After that you can make a model for the country where you live. B u t more or less there are three or four points that are present in every country. This is the connection." 2 8 Though this may understate the differences among Socialists in the Parliament, leftist parties do rely on a clearly articulated and internationalist ideology. 2 9 In contrast, there is much more disagreement within the right about what constitutes generally accepted principles. T h e r e are at least three m a j o r political families in the European center right: the Liberals, the Christian Democrats, and the Conservatives. Fascist and nationalist movements also compete with these movements on the far right. This reflects the absence o f the philosophical touch point that Marxism provides the

Political Cohesion in the European

Parliament

147

left. Even within those three major families, there is not much agreement about general principles. 3 0 Political movements on the right tend to be more heterogeneous in motivation and less defined by particular philosophical bases. The need to clarify those common values has been one function of the parliamentary PPE. Practical experiences of parties of the right and left differ as well. Parties of the left have long organized and worked in international assemblies. As early as the late nineteenth century, international labor and socialist organizations existed in Europe. As noted earlier, the Socialist International serves an important connection within the Socialist Group. There is in fact a formal relationship between the group and the Socialist International. The right does not have as long a tradition of working as a group in international assemblies nor are their formal connections between the PPE political group and the PPE transnational party—some members of the group, e.g., the British Conservatives, are not members of the party. The final influence reflects an organizational element that encourages cooperation among groups. The overall size of a group should make a difference in group cohesion. Reaching a consensus among a small group of people is generally easier than among a large group. There is more opportunity for division within a large group than within a small group. Large groups are also likely to be more heterogeneous, with more national delegations and more parties. This should contribute to disunity. Correlations between political group scores on the index of agreement and on three measures of these variables test more systematically the impact of national party influence, ideology, and size on party cohesion (see Table 6.3). National party influence is gauged by the share of seats in the political group held by its largest member. Among the European Democrats, for instance, the British Conservatives held thirty-two of thirty-four seats, or 94.1 percent of the seats. Ideology is a dummy variable (Left = 1, Right = 0), with the Left Unity Group (LU), EUL, the Greens, and the Socialists classified as "left." Size is the average number of members a group had during the parliamentary term. 31 The first two correlations show the expected relationships, but their magnitude is modest. The larger proportion of seats controlled by a single party, the more cohesion in the group. Leftist parties were also more cohesive. The correlation between group size and group cohesion is positive, indicating that larger groups are more cohesive. This is unexpected. To reevaluate the role of size in parliamentary voting, I compared a variety of size variables: the overall number of group members, the dummy size variable used in the regression equation, the number of member states represented in each group, and the number of parties affiliated with each group. All of these gauge the number of interests and people encompassed by a group. The first two variables define size by the number of group members. The last two reflect the heterogeneity of the group, and the number of

148 Table 6.3

Reflections on European

Integration

Correlations Between Index of Agreement Scores and National Party Influence, Ideology, and Size

Variables

Party Share of Seats Ideology: Left Parties Number of Group Members

Correlations with Index of Agreement Score .29 .48 .22

interests (parties or nationalities) reflected in the group. It may not be the size of the group that inhibits voting agreement but the diversity of that membership. When the four size variables are correlated with the index of agreement score, the direction of the relationship varies. Both the overall group size and the dummy variable are positively correlated with the index, while the number of member states and the number of affiliated parties show a negative correlation. These results suggest that diversity rather than size inhibits group cohesion. More heterogeneous (rather than larger) groups are less likely to agree on voting positions. However, the sizes of the coefficients for all these variables are small, indicating weak relationships with the index of agreement. Size is not irrelevant, however, in voting outcomes in the Parliament. Large groups are able to achieve a great deal of voting agreement but have fewer unanimous votes (see Table 6.4). The groups most likely to achieve unanimity 3 2 were among the smaller groups: Left Unity (13-14 members, 91.7 percent votes unanimous); the European Right (14 members, 90 percent votes unanimous); EUL (28 members, 75 percent votes unanimous); and the Greens (27-29 members, 53.9 percent votes unanimous). Though the EDA did have a low index of agreement score, 40 percent of its votes were unanimous. The Rainbow Group is the exception among small groups. It had no unanimous votes. When the proportion of votes with unanimity within the political group is substituted for the index of agreement, size variables are much more important and consistent. The same four variables show a more consistent and stronger relationship with group unanimity. Both size and diversity significantly decrease the proportion of unanimous votes. These coefficients are much more robust than those associated with the index of agreement, and the diversity measures are statistically significant. The levels of cohesion within political groups is much higher than one would expect, given the institutional structure of the Parliament. In talking with members of the political groups' secretariats it is even more striking how well established the norm of group cohesion and group discipline is. One Christian Democrat said:

149

Political Cohesion in the European Parliament

Table 6.4

Correlations Between Size Variables and the Index of Agreement and Group Unanimity

Size Variables

Index of Agreement

Percentage of Votes with Group Unanimity

Number of Group Members

.22

-.41

Dummy Variable (large = 1)

.08

-.35

Number of Member States Represented in the Group

-.25

-.72*

Number of Parties Affiliated with the Group

-.20

-.71*

Note: *p