Justice Contained: Law and Politics in the European Union 9781501722646

In this probing analysis of the European Union's transnational legal system, Lisa Conant explores the interaction b

144 102 27MB

English Pages 272 [271] Year 2018

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
Contents
Figures
Tables
Preface
Abbreviations
1. Introduction: Justice Contained
2. The Variable Reach of the Law: The European Court of Justice and the Politics of Legal Integration
3. Containing Justice: Institutional Constraints on Law in the European Union
4. From Jurisprudence to Policy: The Liberalization of European Telecommunications
5. European Law and What States Made of It: Intergovernmental Reform of Electricity
6. From Law to Policy and Practice? Nationality and Access to Public-Sector Employment
7. Justice Contained and Reversed: Nationality, Territory, and Access to Social Benefits
8. The European Court of Justice and Judicial Authority: Comparisons and Conclusions
Index
Recommend Papers

Justice Contained: Law and Politics in the European Union
 9781501722646

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

JUSTICE CONTAINED

JUSTICE CONTAINED LAW AND POLITICS IN THE EUROPEAN UNION

Lisa Conant

Cornell University Press

ITHAcA AND LoNDoN

Copyright© 2002 by Cornell University All rights reserved. Except for brief quotations in a review, this book, or parts thereof, must not be reproduced in any form without permission in writing from the publisher. For information, address Cornell University Press, Sage House, 512 East State Street, Ithaca, New York 14850. First published 2002 by Cornell University Press Printed in the United States of America

Library Of Congress Cataloging-in-Publication Data Conant, Lisa]. Justice contained : law and politics in the European Union I Lisa Conant. p. em. Includes bibliographical references and index. ISBN o-8014-3910-8 (cloth: acid-free paper) 1. Court ofJustice of the European Communities. 2. Political questions and judicial power-European Union countries. 3· Law-European Union countries-International unification 4· Law and politics. I. Title. K]E 54 61 .C66 2002 2001005606

341.5'52'094--dC21

Cornell University Press strives to use environmentally responsible suppliers and materials to the fullest extent possible in the publishing of its books. Such materials include vegetable-based, low-VOC inks and acid-free papers that are recycled, totally chlorine-free, or partly composed of nonwood fibers. For further information, visit our website at www.cornellpress.cornell.edu. Cloth printing

10

9

8

7

6

5

4

3

2

FOR DICCON

[Although] indi\idual oppression may now and then proceed from the courts of justice, the general liberty of the people can never be endangered from that quarter. ALEXANDER HAMILTON,

78th Federalist Paper

Contents

List of Figures

ix

List of Tables

xi

Preface

xiii

List of Abbreviations

xvii

1.

Introduction: Justice Contained

2. The Variable Reach of the Law: The European Court of Justice and the Politics of Legal Integration 3. 4. 5. 6.

15

Containing Justice: Institutional Constraints on Law in the European Union

50

From Jurisprudence to Policy: The Liberalization of European Telecommunications

95

European Law and What States Made of It: Intergovernmental Reform of Electricity

122

From Law to Policy and Practice? Nationality and Access to Public-Sector Employment

151

7. Justice Contained and Reversed: Nationality, Territory, and Access to Social Benefits

177

8. The European Court of Justice and Judicial Authority: Comparisons and Conclusions

213

Index

243

vii

Figures

2.1

Organizational and Institutional Determinants of Policy Responses

32

4.1

Competing Interests and Policy Outcomes in Telecommunications

117

5.1

Competing Interests and Policy Outcomes in Electricity

148

6.1

Incidence of Established Infringements in Five Fields

157

6.2 Competing Interests and Policy Outcomes in Public-Sector Employment 7.1

ix

Competing Interests and Policy Outcomes in Social Benefits

172 208

Tables

2.1

Interest Aggregation and Incentives to Mobilize Legally and Politically

2.2 Variation on the Dependent Variable: Policy Responses to Innovative ECJ Interpretation 3.1

Incidence of Delayed Compliance with ECJ Judgments, 1983-1999

27 33 71

3.2 Aggregate Legal Inquiries in the European Union, 1983-1997

76

3.3 Aggregate Legal Inquiries by Member State, 1983-1997

82

3.4 French, German, and U.K. Courts' Citations of the ECJ and European Law

83

3.5 National Court Interaction with the ECJ up to 1999

85

3.6 Distribution of ECJ Citations in French, German, and U.K. Courts

88

4.1

Chronology of the Enforcement of Article 86 (90 EEC)

100

4.2 European Regulatory Framework for Telecommunications

108

5.1

127

Evolution of Legal Obligations under Articles 86 and 31 (90 and 37 EEC)

5.2 Evolution of Reform Proposals for Electricity Liberalization 6.1

Incidence of Established Infringements: Employment and Social Benefits

135 156

6.2 Legal Challenges to Public-Sector Employment Restrictions

159

6.3 Foreign Workers in France, Germany, and the United Kingdom

167

6.4 Labor Costs in France, Germany, and the United Kingdom

167

xi

xii

Tables

6.5

Unemployment Rates in France, Germany, and the United Kingdom

167

7.1

Evolution of Migrants' Legal Entitlement to Social Benefits

186

7.2

Legal Challenges to Social Benefits Restrictions for EU Nationals

190

7.3

Legal Challenges to Discrimination Prohibited by Association Agreements

198

7.4

Distribution of Algerian, Moroccan, Tunisian, and Turkish Migrants, 1997

200

7.5

Distribution of Migrant EU Nationals

203

Preface

My first inquiries into the subject of this book began when I became puzzled by accounts of an "international" court exercising tremendous influence over the Member States of the European Union. I was initially intrigued by the prospects for resolving conflicts in the absence of any mechanisms of coercion and curious about how a court might have contributed to taming states that had been combatants in the deadliest wars of the twentieth century. As I learned more about the European Court ofJustice (ECJ) and traditional domestic courts, my curiosity about the enforcement issues dominating the international relations puzzle shifted to a concern about the actual scope of the ECJ's authority in European politics and policy-making. The "hero" of integration seemed a bit overbearing in its apparent "rule" over liberal democratic states and their societies. This book tries to explain how these two puzzles meet: the ECJ, like all other courts in liberal democracies, has absolutely no means to impose its decisions and rarely exercises any power by having its decisions enforced by the coercive mechanisms of executive state agents. Instead, it exercises its influence beyond the courtroom largely through the efforts of other state and societal actors who come to support its decisions. The completion of this project was possible only with support from many sources. I appreciate the encouragement and feedback of many mentors, colleagues, and friends. Jim Caporaso, John T. S. Keeler, and Christine Ingebritsen all provided invaluable guidance as this work progressed through the dissertation stage and beyond. I am indebted to Chuck Epp and Jonathan Golub for very constructive feedback on the entire manuscript. Special thanks also go to Karen Alter for many helpful exchanges on the xiii

xiv

Preface

ECJ and detailed comments on sections of this book. I thank Roger Haydon of Cornell University Press for his support of the project and the speed with which he shepherded it through. For comments on individual chapters and discussions that helped clarify many points, I thank Karen Anderson, Miriam Aziz, lain Begg, Cecilia Chessa, Bruno de Witte, Clea Finkle, John Gilliom, Kenny Lawson, Andrew Moravcsik, Diana Pallais, Lee Ann Patterson, Milada Vachudova, and Carolyn Warner. For their careful editing of the entire manuscript, I thank Karen Hwa and Jack Rummel. I thank Stuart Scheingold and Alec Stone Sweet for taking time to share their work and discuss how literature in U.S. and comparative public law could inform my arguments. I also appreciate the support of my work by Maria Green Cowles, Michael McCann, Joel Migdal, Thomas Risse, Patricia Weitsman, and Julie White. I thank the European lawyers, officials, and representatives from firms and interest associations I interviewed for sharing their insights with me. I am extremely grateful for the time, documentation, and encouragement they offered me throughout a year of initial field research. Special thanks go to Bernd Lange heine who introduced me to legal study of the European Union while he was an EU fellow at the University ofWashington, provided valuable feedback as I initially defined my project, and helped me make my first contacts in Brussels and Luxembourg. I am also extremely grateful for assistance from the law and EU documentation librarians at the European University Institute (EUI). Machteld Nijsten spent long hours helping me to gather evidence on national case law from four data bases in three languages. Emir Lawless uncovered a wealth of rather obscure EU documents for me and resolved my difficulties with Celex searches on public service. Karin Stromberg and Carole Davison in Seattle; Ingeborg Mehser, Sylvia Fingerle, and Sandra Unbehaun in Berlin; and Federica Sergori, Filipa De Sousa, and Eija Heikkinen in Florence facilitated the logistics of my research at three universities and, in Berlin and Florence, helped with housing, residence permits, and travel. I never understood the need for "technical support" when I used a Mac, but with an involuntary transition to the PC, I am now deeply grateful for all the cheerful and patient assistance provided by Eric Gallo and my much more PC-proficient office mates at the EUI. Like litigation, research requires financial support, and my research benefited from the generous funding of institutions on both sides of the Atlantic. Dissertation research that forms the core of the book's evidence was supported with fellowships granted by the Berlin Program for Advanced German and European Studies of the Freie Universitat Berlin and the Social Science Research Council, the European Community Studies Association and the Ford Foundation (Grant number 940-0466 to ECSA), the

Preface

xv

Foreign Language and Area Studies Fellowship of the U.S. Department of Education, and james Caporaso as Bloedel Chair of Political Science at the University of Washington. I am grateful for a teaching reduction from the Department of Political Science at Ohio University, which allowed me to make revisions that were necessary to transform the project into a book. I thank my new colleagues and dean at the University of Denver for granting me leave during the very first year of my appointment so I could take a Jean Monnet fellowship at the Robert Schuman Centre of the EUI. The unparalleled comparative law resources and unscheduled time I had at my disposal at the EUI enabled me to revise the entire book and substantially expand the scope of evidence to support its central arguments. The conclusions, opinions, and other statements in this book are mine and not necessarily those of the sponsoring institutions. Finally, I appreciate all the support that my husband, Diccon Conant, and our families extended to my pursuit ofthis project. I appreciate the encouragement from my parents, Paul and Ramona Frey; sister, Carol Frey; and Diccon's parents, Gretchen andjonathan Conant, and I thank them for their indispensable help in keeping all of the logistical details of our U.S. life in order while we lived in Florence for ten months. Our Cavalier King Charles Spaniel, Dudley, who curled up on my lap throughout the winter, contributed also by keeping me warm enough to write in my lovely, but frequently freezing, EUI office at Villa La Fonte. I reserve my biggest thanks for Diccon, who provided the support that made this project both possible and rewarding. Above all, his love and humor enabled me to sustain my work on this research over many years. LisA CoNANT

Denver, Colorado

Abbreviations

DG

Directorate General (of the European Commission)

EC

European Community

ECJ

Court of Justice of the European Communities or European Court of Justice

EDF

Electricite de France

EEC

European Economic Community

EOC

Equal Opportunities Commission

EU

European Union

GDP

Gross Domestic Product

GIST!

Groupe d'lnformation et de Soutien des Travailleurs lmmigres

IGC

Intergovernmental Conference

MEP

Member of the European Parliament

MNC

Multi-National Corporation

ONP

Open Network Provision

SEA

Single European Act

SGCI

Secretariat General de Comite lnterministeriel pour les Questions de Cooperation Economique Europeenne

TEU

Treaty on European Union (Maastricht Treaty)

TPA

Third-Party Access

xvii

JUSTICE CONTAINED

1 Introduction: Justice Contained

In the midst ofthe recession and protectionism that plagued Europe in the 197os, a German importer challenged national law that prohibited the sale of the French liqueur Cassis de Dijon in Germany. German law prohibited the import of cassis on the grounds that it did not have a sufficient alcohol content to be marketed as a liqueur. The frustrated importer, Rewe Zentrale AG, claimed that German law posed an illegitimate obstacle to the free movement of goods under European Community (EC) law. The German government defended its law by arguing that national regulation was necessary to protect legitimate values such as consumer health and fair competition in the absence of common European regulations. The absence of common European regulations was a frequent problem because unanimous voting procedures complicated the negotiation of all new EC legislation prior to the adoption of the Single European Act (SEA) of 1g86. The Court of Justice of the European Communities (ECJ) resolved the dispute over cassis by rejecting the claim that a minimal alcohol content protects legitimate values and ruling that there is therefore no valid reason why goods that are lawfully produced and marketed in one of the Member States should not be introduced into any other Member State. 1 In principle, this reasoning eliminates the need for precise standardization efforts at the European level; states accept others' regulations as equivalent to their own instead of establishing common standards. Such mutual recognition of regulation simplifies market integration by obliterating the majority of non tariff 1. Rewe Zentrale AG v Bundesmonopolverwaltungfilr Branntwein C-120/78, (1979) European Court Reports (E. C.R.), 649.

1

2

Justice Contained

barriers to trade. Yet, it also undercuts national sovereignty and democratic processes by allowing foreign rules to regulate imports. If national rules continue to govern domestic production, mutual recognition also leads to reverse discrimination, burdening domestic producers more than foreign competitors. Eliminating this problem by easing requirements for domestic producers threatens to generate a "race to the bottom," where regulatory standards sink to those of the state with the lowest level of protection. The ECJ's acceptance of the right to restrict imports when legitimate values are at stake alleviates the sting of the mutual recognition approach only if everyone either agrees on legitimate values or trusts a court to determine these values. Regulatory differences originate in the distinctive values communities prioritize in the first place, and their dismissal by a foreign court is likely to inspire resentment. In spite of all these dilemmas, however, the mutual recognition approach of the cassis case did transform the European regulatory process and reinvigorated the momentum to advance integration through the internal market program of the SEA. Mutual recognition became a central mechanism to remove barriers to trade in both goods and services despite its problematic implications for sovereignty and democracy. Why did the Member States accept a court ruling that could undermine their control over regulatory regimes? How did such a court ruling reorient the approach to market integration? Why did the principle of mutual recognition become enshrined in the SEA of 1g86? What are the sources and conditions ofjudicial power over policy-making in the European Union (EU)?

Sources and Conditions of Judicial Power in the European Union The power of the supranational European Court of Justice, ruling over a membership of national states, has been attributed to the compelling legal and functional logic of its decisions, 2 its strategic accommodation of national interests, 3 and the support of national courts.' Conventional expla2. Ulrich Everling, 'The Member States of the European Community before Their Court of Justice," European Law Review 9 (1984): 215-41; G. Federico Mancini, 'The Making of a Constitution for Europe," in The New European Community: Decisionmaking and Institutional Change, ed. Robert Keohane and Stanley Hoffman (Boulder, Colo.: Westview Press, 1991); Joseph Weiler, 'The Transformation of Europe," Yale Law]ournal1oo (June 1991): 2403-83. 3· Geoffrey Garrett, "International Cooperation and Institutional Choice: The European Community's Internal Market," International Organization 46 (1992), 533-60; Geoffrey Garrett, R. Daniel Kelemen, and Heiner Schulz, 'The European Court of Justice, National Governments, and Legal Integration in the European Union," International Organization 52 (winter 1998): 149-76. 4· Anne-Marie Burley and Walter Mattli, "Europe before the Court: A Political Theory of Legal Integration," International Organization 47 (winter 1993): 41-76; Walter Mattli and

Introduction

3

nations place particular emphasis on the enforcement of many ECJ decisions by national courts and the difficulty that unanimous voting poses for overturning ECJ decisions through treaty revision. 5 These factors certainly contribute to the ECJ's authority and the binding European legal system it pioneered. In this book I argue, however, that judicial influence over major processes of reform relies on a much broader mobilization of pressure for policy change. Pressures for policy responses to innovative judicial interpretations consist of strategic litigation campaigns of copycat cases, the mass filing of parallel claims before bureaucracies, the systematic prosecution of parallel cases by enforcement agencies, and the lobbying of officials and elected representatives. This legal and political mobilization of pressure derives from the activism of organized societal actors and the responsiveness of public institutions that support legal claims. The mobilization of pressure is critical to the ECJ's influence in policy processes because evasive responses to its case law usually contain the justice that is available through the European legal system. European '1ustice" encompasses the rights and obligations that emerge from ECJ interpretation of the treaties and legislation of the EC (and later the EU). EU Member States typically limit ECJ interference with the content of European rights and obligations by containing their compliance: national administrations obey individual ECJ judgments and simultaneously ignore the implications that unwelcome judicial interpretation has for the universe of parallel situations. As a result, European legal authority fragments as individual national courts apply ECJ principles in single shots while national administrations maintain incompatible policies. Concerted litigation and political mobilization on the local or national level can encourage a patchwork of change, encompassing only some administrative units and some Member States. Sustained pressures across all levels can encourage more general processes of European reform. Yet the mobilization of competing interests in these reform efforts will also influence policy outcomes, diluting, conditioning, and transforming ECJ interpretation. The ultimate policy outcome usually restricts the application of judicial principles by specifYing the

Anne-Marie Slaughter, "The Role of National Courts in the Process of European Integration: Accounting for Judicial Preferences and Constraints," in The European Court and the National Courts-Doctrine and jurisprudence: Legal Change in its Social Context, ed. Anne-Marie Slaughter, Alec Stone Sweet, and Joseph Weiler (Oxford: Hart 1998); Karen Alter, "The European Court's Political Power," West European Politics Quly 1996): 458-87. 5· Karen Alter, "Who Are the 'Masters of the Treaty?': European Governments and the European Court of Justice," International Organization 52 (winter 1998): 121-47; Alec Stone Sweet and Thomas Brunell, "Constructing a Supranational Constitution: Dispute Resolution and Governance in the European Community," American Political Science Review 92 (March 1998): 63-81.

4

Justice Contained

scope and conditions of particular rights and obligations. In this way, justice may again be contained. Others have argued that pressures from organized groups and institutions contribute to judicial policy impact in both the EU and United States.'; I build on the foundations of this research. Yet my analysis is distinctive in developing a general theoretical framework to explain the origins and consequences of variable patterns of legal and political mobilization. Most research concentrates on a single issue and a particular legal or political strategy. My argument contributes to understanding judicial power by assessing the type of mobilization that is likely to occur across issues and by tracing the ways in which the mobilization of pressure mediates policy outcomes.

Implications of the Contained-Justice Explanation The "contained-justice" explanation of judicial power that I advance is significant for debates in the fields of European integration and constitutional law and politics. First, the contained-justice explanation speaks to debates about the institutional autonomy of the European Court of Justice relative to national states. Second, it speaks to debates about tensions between democratic processes and the protection of fundamental rights by courts. The debates address a common question: can courts impose their decisions against the will of states and governing majorities? In the EU context, the capacity of the ECJ to impose its will seems to reflect the end of sovereignty and the domestication of an international system. More broadly, the capacity of courts to impose their will appears to sacrifice majority decisions for the protection of fundamental rights. Judicial power in the European Union is central to the common question behind both debates. The resolution of disputes by the ECJ frequently 6. Carol Harlow and Richard Rawlings, Pressure through Law (New York: Routledge, 1992); Karen Alter and Sophie Meunier-Aitsahalia, 'Judicial Politics in tlle European Community: European Integration and the Patllbreaking Cassis de Dijon Decision," Comparative Political Studies 26 (January 1994): 535-61; Karen Alter and Jeannette Vargas, "Explaining Variation in the Use of European Litigation Strategies: European Community Law and British Gender Equality Policy," Comparative Political Studies 32 (May zooo): 452-82; Karen Alter, 'The European Union's Legal System and Domestic Policy: Spillover or Backlash?" International Organization 54 (summer 2ooo): 4S9-518; Marc Galanter, "Why tlle 'Haves' Come Out Ahead: Speculations on the Limits of Legal Change," Law and Society Review 9 ( 1 97 4): 95-160; Joel Handler, Social Movements and the Legal System (New York: Academic Press, ILJ7S); Susan Olson, "Interestgroup Litigation in Federal District Court: Beyond the Political Disadvantage Theory," Journal of Politics !)2 ( Il)lJO): 8:,4-S2; Michael McCann, Rights at Win-k: Pay Equity Reform and the Politics of Legal Mobilization (Chicago: University of Chicago Press, 1994); Charles Epp, The Rights Revolution: Lawyers, Activists, and Supreme Courts in Comparative Perspective (Chicago: University of Chicago Press, 1998), among otl1ers.

Introduction

5

challenges the national laws and practices ofEU Member States. European treaties and legislation function as a higher law against which national law is judged for its compatibility. Analogous to constitutional review, ECJ interpretation can require the democratic choices of national polities to be set aside for the sake of European integration. European rights trump national rights whenever a conflict arises between the two. The challenge to democratic processes occurs on two levels. First, similar to constitutional interpretation within states, ECJ interpretation creates and expands European rights far beyond the original intentions of the drafters. Second, the European texts that the ECJ uses to grant rights derive from a less inclusive and transparent democratic process than the national texts that face denunciation because executive leaders dominate the European legislative process through closed proceedings in the Council of Ministers. Debates in European integration typically overlook the implications of judicial power for democratic processes because they focus almost exclusively on the prospects for supranational control over states. Competing accounts insist that either the ECJ or the Member States ultimately determine outcomes in disputes over European legal obligations. Conventional explanations defend the ECJ's capacity to subject states to a rule of law; how democratically responsive that rule of law is does not receive much attention.7 By contrast, explanations grounded in realist international relations theory emphasize the enduring primacy of states and deny that European law and institutions exert any real authority over policy processes. In these accounts, the apparent obedience of Member States is contingent on their prior determination of outcomes: ECJ rulings respect the preferences of Member States and are therefore virtually epiphenomenal." Here, the tension between democracy and fundamental rights does not emerge since judicial power does not really exist. Meanwhile, debates that concentrate on constitutional law and politics within countries devote much attention to the tension between democracy and the interpretation of fundamental rights by courts. Europeans are particularly suspicious ofjudicial influence in politics and have traditionally been more likely to prioritize democratic processes over judicial authority. My thesis on contained justice helps resolve both the puzzle of supranational judicial power and the democratic dilemma of fundamental rights protection by courts. By interpreting European law, the ECJ has certainly created rights and obligations that national governments never intended to honor. But, if this unwelcome European justice is contained by its exclusive application to individual disputes, neither states nor democratic majorities face a gross encroachment of their prerogatives. And, if a broad mobiliza7. Burley and Mattli 1993; Alter 1996 and 1gg8; Stone Sweet and Brunell 1gg8. 8. Garrett 199~; Garrett, Kelemen, and Schulz 1gg8.

6

Justice Contained

tion of legal and political pressure is necessary to expand the practical application of controversial interpretations, the policy process reflects a broader consensus than the dictates of a few judges. In this respect, my analysis is parallel to Charles Epp's argument that constitutional rights revolutions within countries develop out of broad patterns of support and organized collective action in civil society. 9 Moreover, if the mobilization of pressure contains justice by conditioning the broader application of judicial principles, the ECJ ceases to impose its will against states or majorities. Instead, those affected by legal interpretation compromise over the precise scope of rights and obligations under European law. Indeed, instances in which Member States react by restricting, or even overturning, ECJ decisions through legislation or treaty amendment constitute democratic responses that clearly communicate the extent of rights available under European law. A crucial problem that remains, however, is the actual exclusion of many individuals from their entitlement to equal treatment under the law. Most existing research presumes a pluralism of litigating interests, which, because of resource constraints, does not actually exist. Bringing a single claim before a court is beyond the capacity of many individuals, but this is exactly what is required to access European justice in the event of pervasive evasion through contained compliance. Few associations or institutions exist to help disadvantaged actors realize their European rights. As long as access to European litigation is far from universal, equal protection before the law remains elusive. While a long trail of ECJ rulings suggests that individuals enjoy particular rights, few individuals enjoy these rights in practice. The hypocrisy of this situation threatens the legitimacy of the European legal system on two levels: frustration prevails both for those who resent an unaccountable judicial expansion of rights as well as for those who continue to face treatment that is inconsistent with existing case law. Yet, until access to European legal recourse is expanded, the hidden dismissal of individual rights through contained compliance will endure, and reform efforts inspired by ECJ interpretation will reflect the concerns of the more organized groups in society that can mobilize pressure through collective action in legal and political arenas.

Origins of the Contained-Justice Explanation The standard emphasis in existing research on the content of ECJ decisions, national judicial enforcement, and institutional barriers to overturng. Epp 1gg8.

Introduction

7

ing the EGJ reflect a preoccupation with distinctions between international and domestic politics. Focused on the supranational character of the ECJ, most analysts have treated the ECJ as an international organization whose legitimacy and authority must be explained. Solid legal reasoning and respect for broad collective interests certainly contribute to the EGJ's legitimacy, and the enforcement ofEGJ decisions by national courts also bolsters the EGJ's authority. Yet looking beyond problems of international organization to the dynamics of constitutional law and politics within countries uncovers challenges that all courts face as institutions. For instance, national judicial enforcement becomes problematic; national courts rely on executive cooperation to enforce decisions against governments just as the ECJ does. Domestic courts typically have few means to allocate resources and rely on others to implement their decisions to any meaningful degree. And, the prevalence of high thresholds for constitutional amendment in domestic systems also means that domestic court decisions are not easily reversed, but few scholars of domestic courts would consider constitutional amendment to be the only means to curtail unwelcome judicial activism. Conventional accounts of the EGJ largely overlook these problems because they never follow legal interpretation out of the courtroom. Most research on the EGJ remains plagued by what Stuart Scheingold labels the "myth of rights" in the U.S. context: the tendency to forge an unproblematic linkage between judicial stimulus and affirmative political responses. Those faithful to the myth of rights assume that politics and policy consistently reflect patterns of rights and obligations under law. 10 Indeed research on the U.S. case, widely considered to possess one of the most powerful judicial systems in the world, chronicles a myriad of obstacles associated with judicial influence in policy disputes." These findings inspired me to look beyond supranational judicial declarations and the compliance problems of international organizations to the broader political dynamics that surround the interpretation of fundamental rights by courts. Parallels between the dominant challenges faced by the EGJ and by domestic courts suggest that the ECJ experiences the standard limitations of courts rather than the typical weaknesses of international organizations. Courts everywhere ultimately depend on parties to bring disputes before them. Unlike other branches of government, which may act independently of the particular demands placed on them, courts make decisions only in reaction to specific claims. Variations in the ability of interested parties to 10. Stuart Scheingold, The Politics of Rights: Lawyers, Public Polil)'. and Political Change (New Haven: Yale University Press, 197 4). 11. For a prominent example, see Gerald Rosenberg, The Hollow Hope: Can Courts Bring about Social Change? (Chicago: University of Chicago Press, 1991).

8

Justice Contained

initiate legal challenges and mobilize pressure to broaden the scope of legal victories, therefore, will be crucial to judicial impact. As a result, I selected four issue areas for comparative analysis: 1) regulation of the telecommunications sector, 2) regulation of the electricity sector, 3) national discrimination in public-sector employment, and 4) national discrimination in access to social benefits. Each area became the target of innovative ECJ interpretation, but they differ in several important respects, particularly in their potential to affect the interests of actors with variable capacities to mobilize pressure in support of their goals. In the first two areas, legal obligations provoked responses from state and societal actors with significant financial and organizational resources. By contrast, legal developments in the last two areas pitted the state against a diffuse class of individuals who would be unlikely to develop the capacity to mobilize legal and political pressure on their own. These four case studies reflect the central focus of European justice and are particularly useful to assess judicial impact on policy processes. European rights and obligations are overwhelmingly economic in nature because European integration has largely concentrated on the creation of a transnational European market. As a result, the European treaty provisions and legislation that the ECJ interprets primarily address the regulation of competition in the EU. The scope of this regulation is broad, reaching beyond simple efforts to encourage trade to balancing environmental protection against economic interests and prohibiting gender and national discrimination in the labor market. Indeed, European integration has been distinctive in its efforts to combine international economic liberalization with a modicum of social justice. Promises related to workers' rights date from the early days of integration while commitments to develop poorer regions and protect consumers and the environment expanded as market liberalization accelerated in the mid-1g8os. Enforcement of these commitments can lead to innovative legal interpretation whenever the ECJ either identifies violations of "forgotten" or "dormant" obligations or creatively adjusts the meaning of provisions. In the cases of telecommunications and electricity regulation, ECJ interpretation reminded Member States that European competition rules extended even to traditionally monopolistic sectors. In the cases on national discrimination in access to public-sector employment and social benefits, ECJ interpretation creatively expanded rights to enhance the economic prospects of EU migrants 12 and their families. ECJ rulings in each area clashed with national law and policy. Historically, 12. I refer to all EU nationals and associated third-country nationals who enjoy European legal rights as "migrants" if these individuals have ever worked, provided services, established businesses, or lived outside their country of citizenship within the EU.

Introduction

9

Member States regulated the telecommunications and electricity industries as public utilities, and they defined eligibility for employment in the public sector and entitlement to social benefits. Therefore, the ECJ's identification and expansion of European rights and obligations in these fields implied major reform of public policy and not merely adjustment by private sector actors. The actual emergence of judicial power in a process of reform develops from a sustained set of legal and political interactions. The process follows three primary steps: ( 1) the ECJ innovates by identifying or creating rights and obligations in its case law; ( 2) public and private actors mobilize to support or resist the broader application of these rights and obligations; and (3) Member States and EU institutions accommodate the new rights and obligations by adjusting their law and policy. This book traces these stages in each issue. Each stage of the process is necessary for judicial policy impact, and justice is usually contained along the way. During the initial step, Member States contain their compliance to individual disputes before court~. If the second stage develops, the mobilization of competing pressures usually compels policymakers to contain justice by restricting the scope of new rights and obligations within broad policy guidelines. I trace the development of European rights and obligations in the body of ECJ case law relevant to each issue. I gauge affected interests and mobilization by tracking litigation, complaints to EU institutions, infringement (enforcement) proceedings, communications and reports by the European Commission, parliamentary debates, discussions of major disputes in trade publications and media coverage, and official statistics. Data from ninetysix interviews in Bonn, Brussels, London, Luxembourg, and Paris provides information about the positions and practices of officials at the European Commission, ECJ, national ministries, lawyers for telecommunications and electricity providers, legal aid associations, regulation specialists in telecommunications firms, national and EU industry associations, and EU organizations that represent the interests of employers, workers, migrants, and citizens. Finally, I measure the impact of judicial interpretation on policy with reference to streams of EC'J rulings and national court litigation that enforce parallel points of law against states, annual reports on the implementation ofEC law by the European Commission, national administrative orders, interview data, and legislation and treaty changes. Policy response is a complex issue, but variations in the extent of judicial control can be assessed by tracing discrepancies between rights and obligations in ECJ case law and ongoing practices and follow-up legislation. Policy responses can occur on botl1 the EU and domestic level. EU policy responses include the adoption of European legislation or treaty provisions to accommodate ECJ interpretation. Domestic policy responses can occur

10

Justice Contained

independently of EU policy responses when individual Member States feel pressures to adapt their law and policy according to the requirements of evolving ECJ interpretation. To examine domestic policy responses in the four case studies and the general institutional reception of European law in Member States, my evidence primarily addresses France, Germany, and the United Kingdom. Data on the interests and mobilization behind EU-level policy responses also concentrates on these three Member States. Given the complex nature of tracing interests, mobilization, and policy processes, a comprehensive study of all EU Member States is not feasible. Focusing on France, Germany, and the United Kingdom is justifiable for a number of reasons. These three states are roughly equivalent in terms of their overall demographic, economic, and political clout in the EU. Indeed, as the most powerful EU Member States, their responses to European law are important to the efficacy of the entire integration project, and their support is critical to the adoption of new European legislation and treaties. These three states are also home to telecommunications and electricity industries of comparable importance, and they host comparably sizeable migrant populations. As a result, actors from these Member States can be expected to have some stake in innovative ECJ case law in the four issue areas that I explore. Finally, France and Germany hold average records of compliance with European law, while compliance in the United Kingdom ranks above average. These Member States' compliance patterns rule out both systemic hostility to European law and a fundamental incapacity to implement any law effectively. No other EU Member State compares with France, Germany, and the United Kingdom on all of these factors. Italy comes closest on most criteria, but it has by far the worst compliance record of any Member State. The political instability and lack of administrative capacity that has led to pervasive noncompliance in Italy precludes investigation of more indirect forms of judicial influence. If Italy cannot regularly obey direct court orders, it can hardly be expected to align broader policy with evolving ECJ case law.

Significance of Judicial Interventions in Policy-Making Given the importance my explanation places on the mobilization of legal and political reactions to innovative legal interpretation, it is important to consider how significant the initial step ofjudicial intervention is for policymaking. Do ECJ rulings matter? Or do judicial contributions merely reflect a broader pattern of demands that would eventually find their way into policy reforms? And does the restriction and dilution of ECJ principles in legislation or treaties fundamentally undermine judicial interference in policy

Introduction

11

processes? I suggest that while the ECJ cannot unilaterally impose major policy changes, there are a number of reasons why judicial interventions in policy-making are nonetheless significant. ECJ rulings matter because they can create solutions that did not exist in prior debates on policy problems. Whether particular ECJ innovations would have eventually entered the policy debate is a counter-factual that cannot be evaluated, but there is often little to suggest that policy reforms would have developed in a parallel fashion to judicial solutions. Courts often provoke political responses by surprising other actors and recasting debates with decisions that no one anticipated. The political response may try to contain the implications of judicial solutions by restricting and diluting innovative legal interpretation. Even if this effort succeeds, the original judicial intervention matters if its content survives to any extent in the policy outcome. Furthermore, the judicial interference matters whenever it motivates definitive policy action. Problems frequently arise before courts precisely because the normal channels of policy-making are deadlocked. Prior to judicial interventions, the stalemate and consequent continuation of the status quo may serve the interests of many actors. Once courts begin toresolve disputes with unwelcome solutions, formerly intransigent actors may seek compromises to limit the extent to which courts continue making decisions in the absence of broadly accepted guidelines. Even if the original judicial intervention is overturned in such a case, it was nonetheless critical to the achievement of a policy change. Individual court rulings can also be significant as bargaining chips for actors engaged in a wide variety of negotiations. Innovative legal interpretation may embolden the European Commission in negotiations over legislation or infringement proceedings with Member States because it feels confident that the ECJ will support it in any disputes. Similarly, sympathetic national officials may use ECJ case law to bolster arguments for changes in national law and policy. Firms, groups, and individuals may also invoke ECJ interpretation in attempts to gain leverage relative to other private or public actors. The scope of the changes compelled by such legalized bargaining depends on the extent of legal and political mobilization by contending actors, but the shadow of judicial interference frames the context of negotiations nonetheless.

An Illustration: Contained Justice and Mutual Recognition The importance of legal and political mobilization to judicial authority over policy-making is evident in the process that incorporated mutual recognition into the internal market program of the SEA. The ECJ's cassis

12

Justice Contained

decision articulated a new approach to economic integration, but it did not generate any automatic or immediate policy effects. Indeed the innovative reasoning of cassis attracted virtually no attention until the European Commission announced its intention to use the ruling to recast its approach to European regulation. Appearing a year after the ECJ issued cassis, a communication of the European Commission coined the term "mutual recognition" and highlighted the potential of the ECJ's reasoning. 13 The European Commission sought to apply the mutual recognition approach broadly as the new means to eliminate non tariff barriers to the free movement of goods and services. According to the European Commission, common EC regulations would be necessary only in those areas of legitimate exceptions to free movement listed under the treaty and the "mandatory requirements" defined by the ECJ in its case law. 14 Member States challenged the European Commission's claims about cassis, directing the Council of Ministers' legal service to issue a counterinterpretation of the ruling. 15 Business associations, environmental organizations, and consumer protection groups also mobilized to support or challenge the mutual recognition approach. 16 As a political debate on mutual recognition began, a series of parallel legal challenges reached the Court ofJustice. The ECJ administered its mutual recognition principle on a case-by-case basis, evaluating the balance between the protection of legitimate values and excessive restrictions on trade. Specific outcomes ranged widely and ultimately affected the marketing of 1 15 individual products. 17 This litigation compelled only piecemeal changes, but it did shift the policy-making context for Member States. Prior to the mutual recognition litigation, Member States were free to protect "legitimate" values through national legislation as long as no common European regulation existed, even if this impeded trade. To the extent that Member States prioritized their specific regulatory protections over any benefits from trade, common Eu13. "Communication from the Commission Concerning the Consequences of the Judgement Given by the Court ofJustice on 20 February 1979 in Case 120/78 ("Cassis di Dijon'');' Official journal of the European Communities C-2 56 I 2 ( 1o March 1980): 2-3. 14. Renaud Dehousse, The European Court ofJustice (New York: St. Martin's Press, 1998): 87-88. Daniel Wincott, "Institutional Interaction and European Integration: Towards an Everyday Critique of Intergovernmentalism," Journal of Common Market Studies 33 (1995), 604-5· 15. Dehousse 1998, 87. 16. Alter and Meunier-Aitsahalia 1994, 551-55. 17. Martin Shapiro, ''The European Court ofJustice," in Eunrpolitics: Institutions and Policymaking in the "New" European Community, ed. Alberta Sbragia (Washington, D.C.: Brookings, 1991), 133-34; Alter and Meunier-Aitsahalia 1994, 54 7-48; Jeanne-Mey Sun and Jacques Pelkmans, "Regulatory Competition in the Single Market," journal of Common Market Studies 33 (1995): 6g-70.

Introduction

13

ropean regulations floundered and economic integration stagnated. After cassis individual national provisions became vulnerable to legal challenge, subjecting Member States to ECJ decisions over which they had no control. Suddenly, the collective failure of Member States to agree to common European regulations did not preserve their right to determine the range of legitimate regulatory objectives. Deadlock simply gave the ECJ the capacity to chip away at particular regulatory protections. 18 Anyone that could benefit from trade had an incentive to initiate legal challenges, and actors engaged in trade are generally likely to possess the financial resources to wage legal battles. This prospect of a continuing stream of legal challenges in turn motivated Member States to agree to common regulations. The policy process inspired by cassis ultimately culminated in a compromise to apply mutual recognition on top of a base of common, minimum standards. 19 The consensus achieved in the Single European Act simultaneously expanded and restricted the mutual recognition approach: it turned mutual recognition into a policy that affected all regulation but it also allowed Member States to define the level of regulatory protection that must be achieved prior to its application. 20 Member States regained control over the European regulatory process by specifying the degree of standardization and mutual recognition they could tolerate. 21 Throughout the policy process, justice was contained. Initially justice was confined to the individual disputes that arose before the ECJ. Ultimately, justice was contained by its limited application to partially harmonized standards. Judicial intervention was certainly an impetus to major policy reform in this case. But the ECJ and cassis were important as the catalysts of a complex policy-making process rather than the definitive authority on the policy outcome. The ECJ created a means to circumvent a stumbling block that had plagued trade for the initial decades of European integration. However, the stumbling block was effectively removed through the consensus of Member States, who in turn reacted to the legal and political pressures mobilized in the wake of the European Commission's use of cassis as a bargaining chip to promote major reform. Without ECJ intervention, the process of achieving common regulations would have continued to be painstaking, accelerated only by the adoption of qualified majority voting in the SEA. Without promotion of the new mutual recognition approach by the European Commission, the ECJ's ruling might have remained obscure and failed to instigate a stream of parallel legal challenges. Without the 18. Dehousse Igg8, C)l-C)2. 19. Alter and Meunier-Ait,ahalia 1994, 551-55· 20. Sun and Pelkmans 1995, 6g-7o. 21. K Lenaerts, "Some Thoughts about the Interaction between Judges and Politicians in the European Community," Yearbook ofEuropean Law 12 ( 1992): 15.

14

Justice Contained

mobilization of organized interests, Member States could have safely ignored the European Commission's efforts to expand the mutual recognition approach beyond the individual cases that resulted in litigation. The achievement of the internal market through mutual recognition rested on far more than careful reasoning, enforcement of cases by national courts, and institutional barriers to overturning ECJ decisions. All of these factors are important foundations for judicial power, but none are sufficient to explain how innovative legal interpretation ultimately relaunched European integration and eliminated the vast majority of non tariff barriers to trade in the contemporary EU. Furthermore, none of these factors account for the manner in which the mutual recognition approach was contained in order to protect a minimal set of common standards. Instead, the new approach to integration grew out of the collective efforts of EU institutions, national governments, businesses, and associations representing a variety of societal interests. And, the practical ramifications of this approach have varied across the EU. Some producers, such as German beer makers, have embraced reverse discrimination by invoking more stringent domestic standards as a means to promote the superiority of their products. By contrast, domestic producers of Italian pasta won the right to circumvent national standards after the Italian Constitutional Court denounced reverse discrimination as unconstitutionaP 2 European consumers, meanwhile, will all have to read their labels more carefully. 22. Commission, "Sixteenth Annual Report on Monitoring the Application of Community Law," COM (gg) 301 final (gJuly 1999), 268-6g.

2 The Variable Reach of the Law: The European Court of Justice and the Politics of Legal Integration

The European Court of Justice is heralded for constructing a binding

supranational legal system that consolidates past EU agreements and propels regional cooperation forward. Why national governments obediently conform to ECJ judgments has inspired the most interest within political science. In existing accounts, prudent judgment, national judicial enforcement, and institutional barriers to overturning judgments figure prominently in the emergence of a supranational legal system that imposes its decisions as effectively as a domestic system. These factors provide a compelling account of a dramatic institutional development, but they fail to explain how this new legal system exerts any influence over the broader politics of integration and the policies that structure interactions within the EU. Most accounts implicitly or explicitly assume that ECJ rulings are automatic catalysts for policy change and that innovative legal interpretation prompts wide-ranging reforms. This assumption vanished long ago in public law scholarship on the U.S. case. A large body of research documents the frequently limited impact of Supreme Court judgments. 1 Alexander 1. Gordon Patrie, 'The Impact of a Court Decision: Aftermath of the McCollum Case," journal of Public Law6 ( lldstein and Robert Keohane (Ithaca, N.Y.: Cornell University Press, 1993); Geoffrey Garrett, 'The Politics of Legal Integration in the European Union," International Organization 49 (1995b): 171-81; Geoffrey Garret, R. Daniel Kelemen, and Heiner Schulz, 'The European Court ofJustice, National Governments, and Legal Integration in the European Union," International Organization 52 (winter 1998): 149-76; Andrew Moravcsik, "Liberal Intergovernmentalism and Integration: A Rejoinder," Journal of Common Market Studies 33 (December 1995): 612, 622-23; Stone Sweet and Brunell 1998a. 55· Weiler 1982, 1991, 1994; Everling 19R4; Mancini 1991.

40

Justice Contained

ical attacks. Appeals to common professional norms also help to generate a loyal following among national judges, lawyers, and academics. 56 Finally, prudent political judgment contributes to compliance. Here, the ECJ deploys a classic judicial tactic of incrementally developing new legal principles in cases with minimal material impact. The ECJ then gradually expands or limits their scope and meaning, depending on initial reactions. If the expansive application of a principle eventually leads to controversy, the legal consistency reflected in the parallel resolution of past disputes bolsters the ECJ's legitimacy, even if periodic noncompliance results. 57 A variety of actors can be the target of the ECJ's political prudence. In arguments grounded in realist international relations theory, the ECJ builds its case law strategically, balancing its desire to construct a supranational legal system with its need to accommodate the interests of the more powerful Member States. 58 Here, national executives are the important actors to appease. Meanwhile, conventional accounts stress that the ECJ created an effective legal system by promoting the interests of most national judges, private lawyers, academics, and societal actors who benefit from integration. Most importantly, the ECJ encouraged national courts to enforce EC law by empowering them with the ability to overturn national law on the basis of its incompatibility with the "higher" European law. Since only specialized constitutional courts had ever exercised this type of judicial review in those Member States with constitutional courts, most national judges gained a significant new competence through the European legal system. 59 Co-opting constitutional and other supreme courts has remained a distinct challenge. Karen Alter argues that competition for influence within national judiciaries gradually forged a cooperative relationship between national high courts and the ECJ. As many national courts seized the opportunity to exercise "European" judicial review, high courts found it necessary to engage the European legal system in order to maintain any control over legal developments within their jurisdictions. 60 The dominant trend is one of national judicial cooperation with the ECJ, but legal debates continue to surface that reflect the need for the ECJ to reach mutually accommodating 56. Burley and Mattli 1993; Walter Mattli and Anne-Marie Slaughter, "Law and Politics in the European Union: A Reply to Garrett," International Organization 49 ( 1995): 183-go. 57· T. C. Hartley, The Foundations of Eumpean Community Law, 3d ed. (Oxford: Clarendon Press, 1994), 87-88; Garrett 1992; Garrett, Kelemen, and Schulz 1gg8; Karen Alter, "Who Are the 'Masters of tiie Treaty'?: European Governments and the European Court of Justice," International Organization 52, 1 (winter 1998): 130-32. 58. Garrett 1992; Garrett, Kelemen, and Schulz 1998. 59· Burley and Mattli 1993; Mattli and Slaughter 1gg8; Karen Alter, 'The European Court's Political Power," West European Politics (July 1996): 458-87. 6o. Karen Alter, Establishing European Law Supremacy: The Making of an International Rule of Law in Europe (Oxford: Oxford University Press, 2001).

The Variable Reach of the Law

41

arrangements with national judges who consider themselves partners rather than underlings in a hierarchy. 61 Conventional accounts also emphasize how the ECJ elicited the support of a variety of societal actors, including lawyers, litigants, and academics. The opportunity to replace national law with European law offered private lawyers and their clients a new avenue of recourse to pursue their claims. Any societal actors whose interests could be served by European law gained a new ally in disputes with public authorities or other societal actors. Likewise, the relationship between ECJ justices and academics has tended to be mutually empowering. Law professors analyzing the European legal system increased demand for their work by raising awareness and emphasizing the importance of European law. Respected as authoritative, their scholarship then facilitated the reception of ECJ case law in national legal communities. This mutually beneficial interaction has been directly evident in the cases in which individuals rotated between academic and judicial appointments. 52 Prudent judgment has indisputably contributed to the creation of an effective supranational legal system in the EU. The ECJ has managed torecruit the support of a wide variety of state and societal actors by designing a legal structure that simultaneously serves many distinct interests. Martin Shapiro considers the ECJ's achievements to be "arguably the most significant judicial instrument of political development in the twentieth century."63 Indeed the ECJ is one of possibly two "international courts" (the other being the European Court of Human Rights in Strasbourg) whose authority is on par with national judiciaries in democracies that respect the rule of law. This book contests neither the critical role that prudent judgment has played in the construction of the European legal system nor the virtually "domestic" performance of the ECJ as a court. However, prudent judgment arguments do not explain how ECJ interpretation might affect the politics of policy-making within the EU. First, it is problematic to account for any variations in policy impact in terms of the functional utility of ECJ judgments. Attributing policy changes to the more functional ECJ rulings and contained compliance to the less functional rulings results in a circular tautology unless there is some commonly accepted 61. Mancini 1991; K. Meessen, "Hedging European Integration: The Maastricht:Judgment of the Federal Constitutional Court of Germany," Fordham International raw journal1 7 ( 1994): 511; Steve Boom, "The European Union after the Maastricht Decision: Will Germany Be the 'Virginia" ofEurope?" AmericanjournalojComparativeLaw43 (spring 1995): 177-226. 62. Burley and Mattli 1993. 59. 65. 63. Martin Shapiro, "Public Law and Judicial Politics," in Political Scieuce: The State of the Discipline II, ed. Ada W. Finifter (Washington, D.C.: American Political Science Association, 1993), 368.

42

Justice Contained

means to measure functional utility. But specifying functionality independently from the relative success of outcomes is difficult. Second, arguments that convincing legal reasoning can camouflage the damaging implications of rulings and shield the ECJ from political confrontation suggest that legal interpretation fails to have any broad impact. Legal masks would become transparent once judgments produced important economic, political, or social consequences. The absence of political controversy is likely to result from the narrow application of legal principles rather than the capacity of the law to shield judges from any and all criticism. Finally, attributing policy impact to political prudence is also suspect. The logic of most arguments about political prudence explains only compliance with individual rulings. The argument about the incremental development of legal principles presumes that ECJ decisions apply exclusively to particular disputes before courts. Arguments about national judges, private lawyers, societal actors, and academics share this limitation. The ECJ may have created incentives for European litigation, but litigation results only in the resolution of concrete disputes for the parties before courts. Legal scholarship and the development of national judicial support for ECJ case law may help to convince national judges to decide cases consistently, but this remains a very narrow application of law as policy. National judicial Enforcement and Respect for the Rule of Law

The relative success of the ECJ in motivating national courts to participate in the European legal system has meant that the majority of ECJ decisions are actually enforced directly by national courts. National judges "refer" questions about European law to the ECJ, who provides interpretations of relevant measures in a "preliminary ruling," which the national court then applies to the dispute at hand. Most of the literature on integration stresses the importance of this mechanism of national judicial enforcement, which distinguishes the European legal system from most systems of international law. The preliminary ruling procedure enlists the preestablished domestic mechanisms of enforcement in the service of European law. Respect for judicial independence and the rule of law normally operate to delegitimize overt political assaults against domestic courts. Joseph Weiler makes this point most starkly by observing that "a state, in our Western democracies, cannot disobey its own courts. "64 Once again, I do not dispute that the enforcement of much of European law by national courts has contributed enormously to the effectiveness of the supranational legal system forged by the ECJ. However, as I noted 64. Weiler 1991, 2421.

The Variable Reach of the Law

43

above, national courts apply European legal principles in single doses as concrete cases come before them. As important as this process is to the establishment of a functioning legal system, it tells us nothing about the role of this legal system in the politics of policy-making in the EU. Existing studies of cross-national and cross-sectoral variation in national court references for ECJ preliminary rulings65 inform us about the extent to which national judicial enforcement holds different Member States accountable to their European obligations but do not explain variation in policy impact due to a focus on individual court cases. Institutional Barriers to Overturning the ECJ

Finally, institutional barriers to overturning ECJ decisions frequently appear in accounts of supranational judicial authority. The ECJ typically justifies its rulings on the basis of the EC treaties, interpreted essentially as a higher law or "constitution" against which all secondary forms of European or domestic legislation must be compatible. As a result, the only guaranteed means to ensure the reversal of ECJ interpretation is to change the treaty. Treaty revision requires unanimous approval and domestic ratification from all EU Member States. Domestic ratification involves at least the approval of domestic legislatures and in some cases popular approval through referenda. If a single Member State rejects a proposed treaty amendment, it will fail to come into effect. Therefore, the barriers associated with overturning ECJ case law are formidable. Member States will have to achieve consensus and be willing to expend considerable effort to reverse the ECJ definitively. 66 Obstacles associated with unanimous treaty revision certainly contribute to the ECJ's autonomy and help explain why very few of its rulings have been formally overturned. Yet individual Member States should rarely feel 65. Jonathan Golub, "Modeling Judicial Dialogue in the European Community: The Quantitative Basis of Preliminary References to the ECJ," EUI Working Paper RSC 96/58, 1996a; Alec Stone Sweet and Thomas Brunell, 'The European Court and the National Courts: A Statistical Analysis of Preliminary References, 1961-;-95," journal ofEuropean Public Policy 5 (March 1998b): 66-g7; lisa Conant, "Europeanization and the Courts: Variable Patterns of Adaptation among National Judiciaries," in Transforming Europe: Europeanization and Domestic Change, ed. Maria Green Cowles, James Caporaso, and Thomas Risse (Ithaca, N.Y.: Cornell University Press, 2001). 66. Fritz Scharpf, 'The Joint-Decision Trap: Lessons From German Federalism and European Integration," Public Administration 66 (autumn 1988) 239-78; Andrew Moravcsik, "Preferences and Power in the European Community: A Liberal lntergovemmentalist Approach," in Economic and Political Integration in Europe: Internal Dynamics and Global Context, ed. Simon Bulmer and Andrew Scott (Oxford: Blackwell, 1994), 6g-7o; Mark Pollack, "Delegation, Agency, and Agenda Setting in the EC," International Organization 51, 1 (winter 1997): 118-19; Alter 1998, 136-42; Stone Sweet and Brunell1998a, 66.

44

Justice Contained

compelled to revise the treaties in response to ECJ decisions. The only instances in which particular Member States would perceive the need to reverse ECJ case law is when it threatened to have widespread negative effects. The question of why individual rulings would have these broad effects still needs to be explained. The difficulty of treaty revision, then, does not explain why treaty revision would be essential in the first place. Most existing scholarship on the ECJ did not set out to explain its impact on policy, but concentrated on the emergence of a binding legal system. A few studies consciously recognize that policy impact is a separate question,"7 but others inflate judicial policy impact by assuming it necessarily accompanies compliance with individual cases. 68 Alec Stone Sweet and Thomas Brunell assert the boldest claims about the power of the ECJ over policy outcomes. Drawing attention to a series of apparent ECJ victories in policy disputes, they argue that national governments have been "forced" to amend legislation to conform to the ECJ's evolving case law and that the ECJ has used preliminary rulings to "legislate, by judicial fiat," provisions vetoed in the Council of Ministers. 69 In their account, all policy processes engaged by the ECJ appear identical: the Member States enact case law as policy. Yet the mechanisms that "force" states to ratify individual rulings as legislation remain unspecified. Stone Sweet and Brunell rely exclusively on conventional factors, which convincingly explain compliance only with individual rulings. Indeed, trying to extend conventional explanations to account for policy impact leads to exaggerated expectations of judicial power. If prudent judgment, national judicial enforcement, and institutional barriers to overturning ECJ decisions are sufficient to determine policy effects, we should see EC and national legislation, as well as administrative practices, promptly adjusting to the demands of innovative ECJ legal principles. Policy responses should be almost uniformly positive and virtually immediate since conventional accounts specify no mechanisms that lead to variation or delay. By contrast, if the mobilization of pressure is critical to policy impact as the contained-justice argument holds, we should expect delays between innovative legal principles and legislative or administrative changes. The lag in policy response might persist indefinitely. If policies eventually shift in 67. Moravcsik 1995, 624; Alter 1998, 136; Mattli and Slaughter 1998, 184. 68. Alec Stone, "Governing with Judges: The New Constitutionalism," in Governing the New Europe, ed.Jack Hayward and Edward Page (Durham, N.C.: Duke University Press, 1995); Carlos Ball, ''The Making of a Transnational Capitalist Society: The Court ofJustice, Social Policy, and Individual Rights Under the European Community's Legal Order," Harvard International Law journal37 (spring 1996): 353-67; Garrett, Kelemen, and Schulz 1998; Stone Sweet and Brunell 1998a. 69. Stone Sweet and Brunell 1998a, 76-77.

The Variable Reach of the Law

45

response to ECJ rulings, delays in policy impact should frequently be accompanied by some of the following: complaints about "violations" of innovative legal principles, streams of copycat cases before national courts and the ECJ, interest group and firm lobbying, and debates among elected officials. The mobilization of legal or political pressure should be associated with the activism of organizational and institutional actors, including interest groups, firms, and public enforcement or legal aid agencies. Policy responses can be expected to vary widely, but most can be expected to dilute the ECJ's principles to some extent.

Structure of the Study and Overview of the Findings The empirical investigation begins with a cross-national assessment of official practices that contain justice across the European Union, with a focus on administrative and judicial practices in France, Germany, and the United Kingdom. However, my analysis emphasizes a cross-sectoral comparison of the role of law in the politics of policy-making. Efforts to promote market liberalization in the telecommunications and electricity sectors and attempts to eliminate national discrimination in access to publicsector employment and various social benefits met with a variable degree of success and affected actors with a diverse array of interests and organizational capabilities. The strong legal basis for reforms in the sectors I examine could be expected to produce automatic policy changes according to conventional accounts. The "direct effect" ofEC law in the cases on market liberalization and national discrimination put proponents of reform in an ideal legal position to challenge prevailing policy: directly effective European legal measures can be enforced before national courts even if no domestic implementing measures ever "transposed" the EC law into national law. However, as the following preview of my findings indicates, purely legal obligations were insufficient to motivate major policy change. With respect to market liberalization, Article 86 (go EEC) of the treaties subjects public enterprises to EC competition rules and empowers the supranational European Commission to issue competition directives (secondary legislation) without the consent of the Council of Ministers or European Parliament. 70 Long dormant and forgotten, these treaty obligations came to the attention of the European Commission through ECJ case law. The application of competition law through Article 86 (go EEC) simultaneously expands the legislative competence of the European Commission 70. Amsterdam Treaty (2 October 1997); Treaty Establishing the European Economic Community (25 March HJ57).

46

Justice Contained

and directly challenges the exclusively national regulation of multiple economic sectors. Both the telecommunications and electricity sectors became targets for liberalization under Article 86 (go EEC) as the European Commission prepared its own directives to open these fields to competition. 71 The ECJ's endorsement of the European Commission's competence to legislate unilaterally meant that monopolistic public enterprises and enterprises granted exclusive rights by national law became vulnerable to a program of liberalization directed by supranational institutions. The direct effect of Article 86 (go EEC) enabled the European Commission and third parties to challenge the legality of exclusive provision that was sanctioned by national regulation. 72 The legal basis for prohibitions against national discrimination is also strong. Treaty provisions prohibit discrimination on the grounds of nationality, generally under Article 12 ( 7 EEC) for individuals engaged in economic activity: as a worker under Article 3g (48 EEC), as an entrepreneur under Article 43 (52 EEC), and as a service provider under Article 4g (sg EEC). Secondary legislation has extended rights of movement and residence to those with sufficient financial means, retirees, and students. 73 Originally designed to open European labor markets, these provisions began targeting national law once the ECJ applied free-movement obligations to public-service sectors traditionally reserved for nationals. All of these provisions have direct effecf• and represent "fundamental freedoms" in the EC legal order, which require that any exceptions be interpreted restrictively.75 EC law prohibits both direct and indirect forms of discrimination: Member States must eliminate policies which may appear neutral in letter, but have either a greater impact on the nationals of other Member States or hinder the free movement of persons. 76 Because distinctions 71. I selected these two fields because the European Commission initiated liberalization efforts with Article 86 (go EEC) in both instances, unlike some other cases of liberalization in service monopolies, for instance, air transport liberalization, which the European Commission initiated under Article 9.5 ( 1ooa EEC) before it "rediscovered" Article 86 (go EEC). 72. Giuseppe Sacchi v Italian RepublicC-1.5.5/73, ( 1974) European Court Reports (E.C.R.), 409; Merci convenzionali porto di Genova SpA v Siderurgica Gabrielli SpA Cr179/9o, (1991) E.C.R., I-.588g; Procedure penak v Paul Corbeau C-320/91, ( 1993) E.C.R., I-2.533. 73· Council Directive go/ 364 EEC, Official journal of the European Communities L-180 ( 13 July 1990): 26; Council Directive go/36.5/EEC, Official journal of the European Communities L-180 (13]uly 1990): 28; Council Directive 93/96 EC, Official journal of the European Communities L-317 ( 18 December 1993): .59· 74· Commission v France C-167/73, (1974) E.C.R., 3.59; VanDuyn v Home Office C-41/74, ( 1974) E.C.R., 1337; jean Reyners v Kingdom of Belgium C-2/74, (1974) E.C.R., 631; johannes Van Binsbergen v Bedrijfmereniging voor de Metaalnijverheid C-33/7 4, ( 197 4) E.C.R., 1299. 7.5· D. Levin v Kingdom of the Netherlands C-.53/81, (1982) E.C.R., 103.5; VanDuyn C-41/74, 1975; Reyners C-2/74, 1974; Van Binsbergen C-33/74, 1974. 76. Catherine Barnard, EC Employment Law (Chicester: John Wiley & Sons, 199.5), 107-8.

The Variable Reach of the Law

47

among Articles 39, 43, and 49 (48, 52, 59 EEC) are frequently blurred, the ECJ has declared that they must be read as a whole, such that principles applicable in one area extend to the others. 77 In order to facilitate labor mobility, EC law extends rights beyond employment and residence to include the equal provision of "social and tax advantages" for EU migrants and their families and equal access to social security schemes. 78 Association agreements between the EU and Algeria, Morocco, Tunisia, and Turkey also provide for the equal treatment of "third-country-national" migrant workers with respect to employment, education, and social security schemes. 79 Judicial interpretation of these measures, which I refer to collectively as "social benefits," directly affects Member States by interfering in the operation of national regimes for social welfare, educational and social services, residence, and taxation. ECJ legal interpretation has been important in all of these areas because the general application of judicial principles could have significant policy consequences for the targeted industries, fields of employment, or state budgets. Legal attempts to liberalize the nationally regulated telecommunications and electricity sectors represent massive extensions of EC competition law, which traditionally applied only to private enterprise. Member States never anticipated that European institutions would enforce competition law against public enterprises and industries to which they granted exclusive privileges. 80 The ECJ's rejection of nationality-based discrimination interferes with public employment policies and programs for welfare and taxation, traditionally the preserve of the national state and generally the site of distributional conflict. Therefore, ECJ legal interpretation in each area threatened to interfere with national policy. The crucial distinguishing features between these four case studies reside in the characteristics of the rulings' potential beneficiaries and losers. The legal and political resources that affected parties could bring to bear in litigative and policy-making arenas varied across these sectors. Telecommunications and electricity liberalization both engaged the intense interests of relatively concentrated actors: large commercial enterprises had intense interests both as producers and consumers of telecommunications and elec77· Jean Rnyerv Kingdom of Belgium C-48/75, ( 1976) E.C.R., 497· 78. Council Regulation 1612/68/EEC of 15 October 1968, Official journal of the European Communities 2 ( 1968): 475; Council Regulation 1408/71/EEC of 14]une 1971, Official journal oftheEurojJean Communities2 (1971): 416. 79· See chapter 7 for detailed discussion of the distinct provisions in different association agreements. 8o. David Gerber, 'The Transformation of European Community Competition Law," Harvard International Law journal35 ( 1994): 136--41; Stephen Wilks and Lee McGowan, "Disarming the Commission: The Debate Over a European Cartel Office," Journal of Common Market Studies 32 (1995): 261,264.

48

Justice Contained

tricity. Telecommunications and electricity also represent fields that are of central interest to state agencies in their pursuit of broader economic and political goals. However, the constellations of support and opposition to liberalization differed across the two industries, leading to variable pressures for adaptation in corresponding regulatory regimes. By contrast, national discrimination prohibitions have engaged the intense interests of a diffuse class of beneficiaries. Both the size and distribution ofthe potential beneficiary populations and the adjustment costs associated with policy change varied between public-sector employment and social benefits. Differences in the availability of institutional support for reform also distinguished the policy process in these two areas. The book's findings illustrate that the mobilization of legal and political pressure mediate the role oflaw in the politics of policy-making in the EU, usually acting to contain justice. In chapter 3, I explore why ECJ decisions usually have a very narrow impact across the EU. The institutional structure of the European legal system confines the scope of most ECJ judgments: incremental and particularistic legal reasoning, the lack of binding precedents and class actions, and discretionary enforcement by national courts and the European Commission all limit the scope of judicial authority over policy-making. Evidence on the treatment of ECJ case law by national administrations, national courts, and the European Commission documents a process by which innovative forms of European justice are carefully contained. Yet some innovative legal interpretation does instigate significant changes in policy, a phenomenon explored iri the case studies on market liberalization and national discrimination. In chapter 4, I analyze the contribution ECJ interpretation made to the liberalization of telecommunications and the legislative competence of the European Commission in EC competition policy. Contrary to common perception, the supranational ECJ and European Commission did not force a fundamentally unwelcome policy reform on intransigent Member States. Instead, the evolution of economic interests in this field, combined with the European Commission's manipulation of a decade worth of innovative legal interpretation, ultimately mobilized a relatively broad base of support for liberalization. This foundation of political support enabled Member States to tolerate the opening of markets through directives unilaterally legislated by the European Commission. Furthermore, Member States retained control over the nature of competition in telecommunications by collectively agreeing to the new EU-wide regulatory framework. In chapter 5, I extend this analysis to a parallel effort to liberalize electricity markets. This chapter demonstrates that the mobilization of resistance prevented the European Commission from using a legitimate legal

The Variable Reach of the Law

49

basis to open markets unilaterally. The lack of a political consensus to liberalize the electricity sector and threats to stall the progress of all EC legislation convinced the European Commission to proceed with a process of intergovernmental reform. A prolonged debate, operating under the shadow of ongoing ECJ litigation, ultimately produced a restricted framework for competition. Legal pressures contributed to policy changes, but Member States rather than the ECJ controlled the content of these changes. In chapters 6 and 7, I trace the impact of ECJ decisions that denounce national discrimination in access to public-sector employment and social benefits. These chapters illustrate the challenges faced by individuals who seek their rights under EC law. EU migrants, who are largely unorganized for collective action, generate only sporadic litigation. A~ a result, Member States can usually confine EC rights to litigants in particular disputes and avoid broader policy change. The decisive role of institutional support in promoting reform is evident in the case of discriminatory access to publicsector employment, which the European Commission prosecuted vigorously after 1991. Relentless enforcement proceedings compelled multiple Member States to abandon their evasion of EC legal obligations in this field and adopt necessary reforms of national law. By contrast, a lack of comparable institutional commitment in the case of social benefits has facilitated persistent discrepancies between national policy and legal obligations articulated in EC] case law. Furthermore, Member States have even unanimously overruled unwelcome ECJ decisions on the transnational mobility of specific social benefits. Finally, in chapter 8 I conclude with a comparative analysis of judicial power in policy-making, situating the ECJ as a court with greater institutional similarities to domestic courts than international organizations. Comparisons emphasize the difficulties associated with judicial influence over policy in the U.S. case, where federal authority has long been established.

3 Containing Justice: Institutional Constraints on Law in the European Union

The European Court of Justice and its case law command respect in the

EU. Yet the operation of an effective supranational legal system will not automatically lead to judicial control over policy outcomes. Compliance with individual ECJ judgments cannot be equated with policy change. Existing accounts often treat innovative rulings as binding policy prescriptions, conflating the application of legal interpretation in a particular case with the adoption of general reforms. Understanding the role of courts and law in European politics requires greater awareness of the institutional rules and practices that contain justice by enabling a narrow application of ECJ decisions. In this chapter I demonstrate that the institutional structure of the European legal system usually confines the impact ofECJ case law. In contrast to conventional accounts of the ECJ, this analysis explores limits on judicial influence in the EU. A realist international-relations approach could not explain these limits merely with reference to the lack of supranational mechanisms of coercion, because the institutional constraints that create these limits originate within the state and afflict domestic courts as well as the ECJ. First, I show that judicial sources oflaw exert relatively weak pressures for change because the incremental and particularized nature of case law dilutes the policy implications it generates. Second, I analyze how formal European rules and prevailing practices typically insulate policy processes from judicial interference. Third, I evaluate the discretionary enforcement of European law by the European Commission and national courts. Together, judicial sources of law, formal rules, standard practices, and discretionary enforcement generate variable opportunities for individ-

50

Containing Justice

51

uals to invoke European law and hold Member States differentially accountable to their legal obligations.

Courts as a Source of Law Courts contribute to the body of law that governs society as they apply constitutional provisions and statutes in cases before them. Judges necessarily modify rules as they specify vague or competing legal provisions in order to resolve concrete disputes. 1 Many Western European legal systems operate on the pretense that all law is embodied in the codes enacted by legislatures.2 Yet, most contemporary legal scholars agree that all courts engage in supplementary and interstitial lawmaking, which fills in the details of statutory or customary law.' Judgments include the broadest policy implications when they address grievances about the operation of public programs or vindicate constitutional and statutory rights. Abram Chayes conceptualizes these legal contests as public law litigation: the judicial interpretation of economic and social legislation impinges on regulatory policy because interpretation given to one set oflitigants is often relevant to many others in similar situations. 4 Achieving Law through Courts? Courts also take a distinctive approach to decision-making that confines the policy ramifications of their rulings. Public law scholars of domestic systems stress the incremental nature of judicial contributions to policy debates. Courts generally take exploratory steps, linking decisions to the particular facts of a case and eschewing comprehensive policy guidelines. Decisions frequently involve important elements of compromise that produce intermediate resolutions for the affected parties. As they confront political, social, and economic responses to their rulings, courts alternatively extend, restrict, or qualify their original principles. This gradual approach to deci1. Robert Dahl, "Decision-making in a Democracy: The Supreme Court as a National Policy-Maker," journal of Public Law 6 ( 195 7): 279-95; Louis Fisher, Constitutional Dialogues: Interpretation as Political Process (Princeton, NJ.: Princeton University Press, 1988), 37-38; Martin Shapiro, Courts: A Comparative and Political Analysis (Chicago: University of Chicago Press, 1981),20. 2. Shapiro 1981, 2H; Carol Harlow and Richard Rawlings, Pressure through Law (New York: Routledge, 1992), 315. 3· Shapiro 1981, 35· 4· Abram Chayes, "The Role of the Judge in Public Law Litigation," Harvard Law Review 89 (1976): 1294-95, 1304; Abram Chayes, "Foreward: Public Law Litigation and the Burger Court," HarvardLawReview96 (1982): 4-5.

52

Justice Contained

sion-making stems from the desire to encourage consent and avoid further conflict. 5 This feature of judicial decision-making has important implications for policy. Cass Sunstein argues that courts ordinarily produce incompletely theorized agreements on particular outcomes. 6 According to Sunstein, "If judges disavow large-scale theories [by linking agreements and their justifications to specific events], then losers in particular cases lose much less. They lose a decision, but not the world. They may win on another occasion."7 Because courts do not put the world at stake, the application of their decisions remains open to debate. Uncertainty about the extension of legal principles across cases precludes rapid policy reform. Instead, the parameters of a court's solution slowly develop as remaining questions are settled through ongoing litigation, interpretations by executive leaders and bureaucrats, and legislators' efforts to accommodate gaps left by judicial silences. 8 Like other courts, the ECJ incrementally develops its principles in reference to particular situations. Ambiguities associated with the application of new rules across cases encourage contained compliance, while any efforts to apply innovative legal interpretation widely remain vulnerable to challenge on contextual grounds. Pressures for reform may slowly build if case law reinforces a rule and expands its scope. However, this pressure typically advances at a glacial pace and can involve the restriction of rules as well as their expansion. As a result, any exclusively judicial development of European legal obligations generates a relatively weak pressure for change. The uncertainties involved in applying judicial interpretation broadly can affect rulings that prescribe changes in substantive policies or institutional structures. The following discussion demonstrates that the ECJ case law responsible for constructing a supranational legal system includes a myriad of limitations that obfuscate the responsibilities and opportunities states and individuals face under European law. These institutional ambiguities, products of judicial decision-making themselves, then compound the difficulties of applying rulings with implications for substantive policy. Achieving Europe through Courts?

ECJ decisions on direct effect, supremacy, and state liability attract attention as powerful mechanisms to align national policy with European obli5· Chayes 1976, 1316; Fisher 1988, 10, 276; Shapiro 1981, viii, 15, 36; Cass Sunstein, Legal Reasoning and Political Conflict (New York: Oxford University Press, 1996), viii, 37. 6. Sunstein 1996, viii. 7· Ibid., 41. 8. Fisher 1988, 243, 256; Charles Johnson and Bradley Canon, judicial Politics: Implementation and Impact (Washington, D.C.: Congressional Quarterly, 1984), 32.

Containing Justice

53

gations and accord individuals their European rights. Yet the judicial origin of these celebrated institutional developments results in gaps and escape clauses that complicate the uniform application of the law and access to supranational legal recourse. The issue of "direct effect" arises in European law because the treaties that constituted the European Communities and one major form of European secondary legislation, directives, do not apply directly. Laws that apply directly confer rights and obligations without requiring any further implementing measures. The Treaty of Rome only provided for one type of European secondary legislation, regulations, to apply directly. As a result, individuals and firms rely on European regulations as if they were domestic legislation, enforceable before national courts. By contrast, international treaties typically bind only states and cannot serve as the basis for legal claims made by other actors against those states. Like other treaties, the Treaty of Rome did not provide for its own direct effect. Furthermore, European directives, which establish common rules in many policy areas, required transposition into national law because they bind Member States to achieve a particular result but leave the choice of "form and methods" to national authorities. Directives originally constituted frame-work legislation without any direct effect. The absence of direct effect for treaties and directives limited opportunities to challenge violations of EC legal obligations. The European Commission and Member States could initiate proceedings against infractions, but affected individuals and firms within Europe would have no recourse against national governments that disregarded provisions of the treaties and directives. In order to promote the application of European law, the ECJ granted direct effect to treaty articles in 1963 and directives in 1974. By granting supremacy to European law as well in 1964, the ECJ is widely recognized as having "constitutionalized" the treaties. 9 Constitutions confer rights and obligations on citizens as well as governments, and they serve as the higher law in disputes about competing legal norms. Finally, by holding states liable for damages individuals suffered as a result of the g. lvV. Algemene Transpo-rten Expeditie Onderneming van Gend en Loos v Nederlandse Administratie der Belastingen C-26/62, (1963) European Court Rep&ts (E.C.R.), 1; Ramino Costa v ENEL C6/64, (1964) E.C.R., 585; VanDuyn v Home Office C-41/74, (1974) E.C.R., 1337; Eric Stein, "Lawyers,Judges, and the Making of a Transnational Constitution," American Journal of International Law 75 (1981 ): 1-27; G. Federico Mancini, "The Making of a Constitution for Europe," in The New European Community: Decisionmakin{!; and Institutional Change, ed. Robert Keohane and Stanley Hoffman (Boulder, Colo. :Westview, 1991), 177-194; Joseph H. H. Weiler, The Constitution ofEurope: "Do the New Clothes Have an Emperor?" and Other &says on European Integration (Cambridge: Cambridge University Press, 1999). The constitutional nature of the treatybased legal system is contested by a number oflegal scholars who argue that true constitutions arise out of deliberate democratic decisions, rather than judicial fiat, and provide for comprehensive judicial protection of individual rights, rather than the patchy and indirect nature of legal protection available in the EU.

54

Justice Contained

nonimplementation of European law in 1991, the ECJ created a new mechanism to enforce European rights. 10 These rulings transformed interstate agreements into a supranational legal order binding states, firms, and individuals. However, the resulting legal order reflects the typical limitations of judicial sources of rules and generates much less pressure for the convergence of law and policy than is commonly assumed. While the supremacy doctrine confers an unambiguous precedence to European law, ECJ rulings on direct effect and state liability are classic examples of judicial decision-making in incrementally expanding and limiting the nature of rights and obligations. As the ECJ extended direct effect to treaty provisions and directives, it also rejected the direct effect of measures whose terms are insufficiently precise, conditional in their obligations, or grant discretion to Member States. 11 As a result, the material scope of direct effect expanded gradually, determined through ongoing litigation on discrete treaty articles and directives. Although the ECJ liberally extended direct effect, 12 some legal provisions have failed the test and others remain open to dispute. More significantly, restrictions the ECJ has attached to the nature of obligations imposed by direct effect eliminate many actors' need to conform with European law. Specifically, EC provisions vary in the obligations they impose on private parties. Regulations, whose direct effect was established by treaty, have a general application that imposes obligations on both public and private parties. 13 In this case, direct effect operates both "vertically" between governments and their subjects and "horizontally" between private actors. By contrast, while granting vertical direct effect to clear and precise treaty articles and directives, ECJ interpretation has limited the horizontal direct effect for these forms of law. The ECJ has applied direct effect horizontally between private parties to a few treaty articles, for instance, Article 141 (119 EEC), which guarantees equal pay for men and women, and Articles 81 and 82 (85 and 86 EEC), which prohibit agreements that restrain trade and abuses of dominant market positions. 14 Yet for directives, the ECJ rejected the principle of horizontal direct effect 10. Francovich and Bonafaci v Italy Joined G-6, 9/90, (1991) E.C.R., I-5357· 11. VanGendC-26/62, 1963; VanDuynC-41/74, 197+ 12. Lynne Watson and Bermann v Italian Republic C-rrS/75, (1976) E.C.R., 1185; Damian Chalmers, judicial Preferences and the Community Legal Order," Modern Law Review 6o (March 1997): 183. 13. Article 189 EEC; George Bermann, Roger Goebel, William Davey, and Eleanor Fox, Cases and Materials on European Community Law (St. Paul, Minn.: West Publishing, 1993), r8o. 14. Gabrielle Defrenne v Sabena (II) C-43/75, ( 1976) E.C.R., 455; Bermann eta!. 1993, r8o; Amsterdam Treaty (2 October 1997); Treaty Establishing the European Economic Community (25 March 1957).

Containing Justice

55

in decisions throughout the 1g8os and 1990s. 15 Because Article 249 (189 EEC) binds only Member States to the objectives of directives, the ECJ holds that directives can only confer rights on individuals against states. The ECJ reasoned that states must be subject to direct enforcement of directives because they cannot rely on their own failure to implement a directive properly and on time in order to escape obligations under EC law. 16 Legal certainty, meanwhile, provides a legitimate justification for rejecting horizontal direct effect: individuals following clear provisions of national law have a legitimate expectation that these national measures apply to their situation. 17 However, for a program of integration designed to foster exchange relations, the absence of horizontal direct effect for directives constitutes a critical gap in legal protection. Because directives cannot impose obligations on private parties, directives cannot be relied on to enforce EC rights in disputes between private parties.18 Therefore, directives can take precedence over conflicting national law to govern relations between the state and individuals, but not relations among individuals. The ECJ developed an indirect route to the enforcement of European rights against private parties by obligating national courts to interpret national law in conformity with EC law, including directives. The 1984 Von Colson decision obligates national courts to interpret any national legislation that was adopted to implement directives in conformity with the requirements of EC law. For six years it remained unclear whether national courts should interpret national law that was not designed to implement directives in conformity with EC directives. The 1990 Marleasingruling settled this ambiguity by obligating national courts to interpret national law adopted either before or after the adoption of a directive to be in conformity with EC law. 19 However, the distinction between this indirect obligation to interpret and the obligation to apply European law directly is ambiguous. Reading national law to be consistent with EC directives is functionally equivalent to direct effect. Some scholars hail ECJ case law on interpretive obligations, or "indirect effect," as a means to the enforcement of directives between private parties. Yet, ECJ case law on interpretive obligations and direct effect limits the practical utility of this mechanism of enforcement. First, the 15. M. Helen MarshaU v Southampton and South West Hampshire Area Health Authority C152/84, (1986) E.C.R, 723; PaoloFacciniDori v Recreb SrlC-91/92, (1994) E.C.R., l-3325; El Corte Ingles SA v Cristina Bltizquez Rivero C-192/94, ( 1996) E.C.R., l-1281. 16. Pubblico Ministero v Tullio Ratti C-q8/78, ( 1979) E.C.R., 1629. 17. SalumiC-212-17/80, (1981) E.C.R., 2735· 18. Marshall C-152/84, 1986; Kolpinghuis Nijmegen v Kingdom of the Netherlands C-8o/86, (1987) E.C.R., 3969; and El Corte InglesC-192/94• 1996. 19. Von Colson and Kamann v Land Nordrhein-Westfalen C-14/83, (1984) E.C.R., 1891 and Marleasing SA v La Comercial lnternacinnal de Alimentacion SA C-106/89, ( 1990) E.C.R., l-4135; Salvatore C'rrimaldi v Fonds des Maladies Professionnelles Cr3 2 2/88, ( 1989) E. C.R., 4407.

56

Justice Contained

ECJ allows national courts to escape the interpretive obligation with its order to interpret domestic legislation "in so far as it is given discretion to do so under nationallaw" 20 and "in so far as possible in light of the wording and purpose of the directive. "21 In any direct conflict between national law and an EC directive, national courts can decide that formally overriding national law is "impossible" on the basis of national requirements. This judicial escape clause is necessary for the ECJ to avoid overstepping its jurisdiction since interpretation of national law is the exclusive preserve of national courts. 22 This judicial escape clause also complicates efforts to achieve legal certainty. Gniinne de Biirca argues that "it would be extremely difficult for those potentially affected by a measure to know whether to act according to domestic law as currently interpreted, or in accordance with an EC Directive in conformity with which the domestic law would be read. "23 Subsequent ECJ case law maintained the escape clause by continuing to recognize that national courts may not be able to interpret national law to be in conformity with European law.~• As a result, the potential to enforce rights through the "indirect effect" of the interpretive obligation remains precarious. Second, the ECJ frustrates the application of EC law to private disputes every time it explicitly rules that directives cannot be enforced against private parties, as it did after the articulation of its interpretive obligation. 25 To add to the general state of confusion, the ECJ also found in favor of plaintiffs appealing to EC directives in instances that had the implication of horizontal direct effect, although they were formally litigated by public authorities or pursued indirectly against a decision of a public authority. 26 This fundamentally incompatible case law on the direct and indirect enforcement of EC directives between private parties27 generates an inconsistent pressure for the alignment of practices with legal obligations. Reliance on EC directives in private disputes succeeds if national courts interpret national law in conformity with European law, and it fails if national courts do 20. Formulated in Von Colson C-14/83, 1984 and cited in MarleasingC-w6!89, 1990. 21. MarleasingC-w6!89, 1990. 22. T. C. Hartley, The Foundations of European Community Law, 3d edition (Oxford: Clarendon Press, 1994), 223-2423. Gn'i.inne de Burca, "Giving Effect to European Community Directives," Modern Law RPr vif!lv 55 (March 1992): 219. 24. Wagner Miret v P'ondo de Garantia Salarial C-334/92, (1993) E.C.R., I-6911; Chalmers 1997, 190. 25. FacciniDori Crg1/92, 1994 and El Corte Inglis C-192/94, 1996. 26. Rnfael Ruiz Bernal.dez C-129/94, (1996) Common Market Law Reports 76, 2: 889 and Panagis Pafitis and Others v Trapew Kentrikis Ellados AE and Others Cr441/93, (1996) E.C.R., 11347· 27. Hartley 1994, 223, 227; Bermann eta!. 1993, 192; De Burca 1992.

Containing Justice

57

not directly apply EC directives. Both options are equally legitimate under ECJ case law, but one outcome relies on European law while the other rests exclusively on national law. De Burca observes, "There is a conflict between [the ECJ's] desire to ensure the effectiveness and enforcement of EC Directives and its unwillingness to declare them directly enforceable .... [This] leaves the legal position of those who would benefit from or incur responsibilities under Directives in a state of considerable confusion.''28 These ambiguities manifest themselves in the inconsistent application of the ECJ's interpretation across and within national court systems. For example, the Netherlands' Hoge Raad (Supreme Court) explicitly invoked the ECJ's escape clauses to avoid interpreting national legislation in line with a European directive whose transposal was overdue. The Dutch national court claimed that national provisions precluded an interpretation in line with the directive, and that any effort to stretch the interpretation would violate legal certainty since it would indirectly place obligations on individuals, something directives cannot do according to the ECJ. Recognizing indirect effect as equivalent to the forbidden horizontal direct effect, the Dutch court refused to apply European law over the conflicting national measure. 29 Meanwhile, Italy's Corte di Cassazione (Supreme Court) acknowledged that provisions of a directive whose transposal was overdue could not be applied as rules of law in litigation between individuals, but nonetheless found it possible to allow the directive to be interpreted as the legal framework to decide a judgment between individuals."" Here, willingness to invoke indirect effect led to the application of European law. Such discrepancies even plague decisions within the same national court. Spain's Tribunal Supremo (Supreme Court) decided that one directive had horizontal direct effect in three cases, but also expressly denied the horizontal direct effect of this same directive in another case, and finally articulated doubt that this directive could have the horizontal direct effect it had previously attributed to it.'1 The prospects for redress in this situation are far from optimal. EGJ decisions on state liability potentially rectify this problem by providing an alternate form of recourse for situations in which individuals are unable to invoke directives against other private parties. The ECJ first declared that governments could be liable for damages caused to individuals by a failure to transpose directives in 1991. By 1996, subsequent 28. De Btirca 1992, 229. 29. Commission, "Fourteenth Annual Report on Monitoring the Application of Community Law-1996," Official journal of the European Communities C-332 (3 November 1997): 20430. Commission 1997, C-332, 203. 31. Commission, "Sixteenth Annual Report on Monitoring the Application of Community Law-1gg8," COM (gg) 301 final (gjuly 1999), 261-62.

58

Justice Contained

cases established that states could also be liable for the inadequate implementation of European law. 32 Prior to this case law, if a government did not transpose a directive promptly and properly, private parties could rely on existing national law to escape European obligations. By holding states responsible for the losses suffered in such a case, however, the ECJ created a remedy for individuals who could benefit from European provisions. 33 State liability also provides governments with an incentive to transpose directives appropriately in order to avoid paying financial compensation to those suffering under national rules that are incompatible with European obligations. Correctly transposed directives are essentially European law reconstituted in the form of national law. Since individuals can invoke national law against private parties, proper transposal of directives alleviates the problem that arises from the inability of individuals to enforce directives horizontally against other private parties. However, establishing state liability remains difficult in practice. Tests to determine liability rely on the discretion of national courts and the prevailing state of national procedures. The ECJ has articulated conditions to determine state liability that usually allow national courts to rule on questions of liability themselves and require national courts to find or invent the procedures necessary for individuals to claim compensation. National courts have started to find states liable for damages on the basis of the ECJ s criteria, but preliminary evidence suggests that national courts have awarded few damages and dismiss more claims for compensation than they grant. National judges appear to have no trouble justifying their denial of claims on the basis of ECJ interpretation and on the basis of inadequate procedures under national law. First, the following ECJ guidelines delegate most responsibility concerning the identification of state liability to national courts. For states to incur liability, the relevant directive must confer rights on individuals, and a causal link must exist between the failure to implement a directive and the loss suffered. 34 States are strictly liable for damages caused in circumstances where they have no discretion, such as the timely transposal of a directive or the failure to comply with an interim order of the ECJ. 35 For situations 32. Francovich C-6, 9/90, 1991; Brasserie duPecheurSA v Germany and R v Transport Secretary ex p. Factortame Ill joined C-46, 48/93, ( 1996) E.C.R., I-1029; Dillenkofer v Germany Joined C178, 179, 188, 190/94, ( 1996) E.C.R., I-4845; Hedley Lomas v Agriculture and Fisheries (MAFF) Cr5/94, (1996) E.C.R., l-253; R v Treasury ex p. British Telecommunications C-392/93, ( 1996) E.C.R., I-1631. 33· Chalmers 1997, 191. 34· FrancovichC-6, 9/90, 1991. 35· Brasserie and Factortame III C-46, 48/93, 1996; Dillenkofer C-178, 179, 188, 190/94, 1996.

Containing Justice

59

where states have wider discretion, such as precisely how to transpose a directive into national legislation, breaches will be sufficiently serious to incur liability if national measures violate a provision of European law that is clear and precise, or if they constitute infringements according to prior ECJ case law.'" Since these guidelines fall under the circumstances that obviate the need to make references for preliminary rulings, Damian Chalmers argues that the ECJ has created a system of enforcement that will operate principally through national courts and local interpretive culturesY Indeed, national courts appear willing to determine whether suspected breaches are causally related to damages suffered, sufficiently serious, or flagrantly violating clear provisions of EC law. National courts have dismissed claims on the basis of their application of these guidelines ten times, and established instances of state liability only eight times, according to European Commission reports that track the application of ECJ case law on state liability. 38 The European Commission does not claim to present a comprehensive account, but seems determined to indicate progress in the application of European law. Findings of liability relevant to individual breaches find their way into multiple reports, usually because governments appeal against these decisions until they reach the highest court. Reports that follow such cases indicate that upper courts have upheld initial findings of liability, but they also suggest that the European Commission has found no abundance of instances in which national courts are actually prepared to award damages against the state. Findings of liability are sure to increase over time as national courts face more claims and become more familiar with ECJ interpretation. Yet, the remaining gaps in ECJ guidelines on state liability invite divergent application by national courts and create uncertainties that are likely to discourage many actors from attempting to seek redress through state liability in the first place. The second difficulty associated with state liability is that the ECJ subjects this remedy to national rules on damages. 39 Questions to be settled by national procedures include which state agency is liable to pay damages when governments fail to transpose directives properly, which national courts are competent to decide on liability, what constitutes an adequate level of compensation, and what time limits are appropriate to impose on

36. Brasserie and Factortame III C-46, 48/93, 1996; Lomas C-5/ 94, ( 1996); British Telecommunications C-392/ 93, 1996. 37· Chalmers 1997, 195-96. 3R. Commission 1997, C-332, 205-6; Commission, "Fifteenth Annual Report on Monitoring the Application of Community Law-1997," Official Journal of the European Communities C250 (10 August 1998): 203-4; Commission 1999, COM (99) 301 final, 271-75; Commission, "Seventeenth Annual Report on Monitoring the Application of Community Law-1999," COM (2ooo) 92 final (23june 2000), Annex VI, 31-34. 39· Francovich C-6, 9/go, 1991 and BrasserieC-46, 48/g3, 1gg6.

60

Justice Contained

the acceptance of claims. These are not trivial gaps when national procedures vary and national courts decline the opportunity to invent procedures that do not already exist under national law. For instance, five years after the ECJ's Frankovich ruling held the Italian state liable for damages caused by the failure to transpose a directive, the Italian Corte di Cassazione annulled a lower court's decision to grant compensation for a parallel claim, arguing that no provisions of Italian law enabled the state to be held liable for action taken in the exercise of its legislative powers.'0 This court repeated its position that rights to compensation do not flow from unlawful acts by the state until June tgg8, when it decided that compensation could flow from the civil liability of the state under the Italian Civil Code. 41 This particular reversal opens the way to recourse in Italy, but a judge from this court nonetheless finds state liability very difficult to implement in practice. When neither the ECJ nor national legislation specifies the precise means to grant compensation, national judges remain at a loss as to how they are supposed to proceed lawfully. 42 This is not a particularly inspiring avenue for potential claimants, given the time and resources absorbed by litigation. In an analysis of the ECJ's state liability case law, Carol Harlow concludes that reliance on national systems of liability may be creating an illusion of remedy where few remedies are available in practice.'' More generally, ECJ case law on the provision of remedies obscures the nature and extent oflegal recourse available under European law. The ECJ has sustained a basic conflict in its case law on remedies for over a decade. Initially, the ECJ ruled that national courts use all available national remedies to aid in applying EC law,'1 but need not create new remedies. 45 The ECJ originally sought nondiscrimination: it has required that EC rights be enforced to a degree parallel to national rights, and that it not be impossible or excessively difficult to seek EC rights. However, the ECJ left questions of judicial procedure and sanctions to domestic rules. Subsequently, the ECJ contradicted its earlier rulings by holding that national courts are 40. Commission 1997, C-332, 205-6. 41. Commission 1998a, C-250, 203-4; Commission 1999, COM (gg) 301 final, 273. 42. Giovanni Giacalone, 'The Community Judge and the National judge: The Francovich Case," in Towards a Europeanised judiciary? Practitioners' Experiences of National judges with the Europeanisation of Private Law, ed. Sonja Feiden and Kristina Riedl, EUI Working Paper LAW no. 2000/3. 2000,46-55· 43· Carol Harlow, "'Francovich' and the Problem of the Disobedient State," EUI Working Paper RSC no. g6/62, 1gg6, 32. 44- Rewe v Landwirtschaftskammer Saarland C-33/76, (1976) E.C.R., 1g8g; Comet v Produktschap voor Siergewassen C-45/76, ( 1976) E.C.R., ~043; Carmin Russo v Azienda di Stato per Glllnterventi sul Mercato Agricolo (AlMA) C-6o/75, (1976) E.C.R., 45; Amsterdam Bulb Bv v ProducktschapvoorSiergewassenC-50/76, (1977) E.C.R., 137. 45· RewevHauptzollamtKielC-158/So, (1981) E.C.R., 1805.

Containing Justice

61

obligated to provide effective, adequate remedies. 46 If national remedies do not provide effective or adequate remedies, national courts would appear to need to create new remedies. The implications of both positions surfaced in case law throughout the 1g8os and 1ggos: in most instances the ECJ recommended reliance on national rules, 47 but in others it required national courts to ignore domestic limitations and effectively create new remedies.'8 Advocate General Jacobs has argued that the ECJ assumes national remedies normally protect EC rights sufficiently, concluding that the need to intervene is exceptional. 49 Yet, he has also acknowledged the inequities generated by the existing framework: "If Community law is to be uniformly applied, if undertakings are to benefit from comparable levels of judicial protection in different Member States, and if Member States themselves are to be subject to comparable burdens, then it is desirable, perhaps even essential, that there should be a more uniform approach to remedies and procedural rules governing the enforcement of Community rights. "50 Given the contradictory elements in the ECJ's case law, national courts have tremendous discretion over remedies and individuals face unpredictable procedural requirements. National courts can easily justify the application of national procedures that restrict the ability to invoke European rights. For example, the German Bundesfinanzhof cited more restrictive ECJ case law to justify the use of national procedures that would eliminate the opportunity to adjust a claimant's tax assessment in line with an EC directive. The Bundesfinanzhof insisted on the validity of a deadline for ap46. Von ColsonC-14/83, 1984. 47· Mireco SAS v Amministrazione delle Finanze dello Stato C-826/79, (198o) E.C.R., 2559; Deutsche Milchkontor v Germany C-205-15/82, (1983) E.C.R., 2633; San Giurgio SPA v Amminstrazione delle Finanze dello Stato C-199/82, ( 1983) E.C.R., 3595; H. Steenhorst-Neerings v Bedrijfsvereniging voor Detailhande~ Ambachten en Huisvrouwen C-338/ 91, ( 1gg3) E.C.R., I-5475;]ohnson v Chief Adjudication OjficerC-410/g2, (1gg4) E.C.R., I-5483; ColorollPensions Trustees Ltd. v Russell C-22/91, (1gg4) E.C.R., I-2043; Peterlnveck, Van Campenhout SDS and Cie SCS v Belgium C312/ g3, ( 1gg5) E.C.R., I-45gg; Van Schijndl and van Veen v Stichting Pensioenfonds voor Fysiotherapcuten C-43o-1 I 93, ( 1gg6) E.C.R., I-4 705; Francovich C-6,g/ go, 19g1; Brasserie and Factortame ///C-46, 48/g3, 1996. 48. UNECTEFvHeylensC-222/86, (1987) E.C.R., 4097; R v Secretary of State for Transport, ex parte Factortame (no. 2) C-2 13/89, ( 1990) E.C.R., I-2433; Emmott v Minister Jpr Social Welfare and the Attorney General C-208/ go, ( 1gg1) E.C.R., I-4269; and M. Helen Marshall v Southampton and South West Hampshire Area Health Authority (no. 2) C-271/g1, (1993) E.C.R., I-4367; Erika Szyszczak, "Making Europe More Relevant to Its Citizens: Effective Judicial Process," European LawReview21 (19g6): 351-64. 49· Opinion in Van SchijndlC-43D-1/93, 1996, E.C.R., l-4705, paragraphs 28-30. 50. Christopher Himsworth, "Things Fall Apart: The Harmonisation of Community Judicial Procedural Protection Revisited," European Law Review 22 (1997): 300, quoting F. G. Jacobs, "Remedies in National Courts for the Enforcement of Community Rights," in Hacia un Nuevo Orden Internacional y Europeo: Estudios en Hemenaje al Professor Don Manuel Diez de Velasco

(1gg3).

62

Justice Contained

peals even though it recognized that national rules had not been properly interpreted by courts or correctly implemented by the government. 51 The Italian Corte di Cassazione and French Cour de Cassation (a Supreme Court) also imposed national time limits on applications for the reimbursement of taxes whose collection had been incompatible with EC directives. Like the Bundesfinanzhof, the Cour de Cassation legitimated its restrictive decisions with reference to ECJ case law. 52 Appeals on the "exceptional" chance that a situation requires intervention to ensure effective legal protection is apparently not a very promising option when therecovery of inappropriate taxes is at issue. Both EU officials and legal scholars recognize the shortcomings of a purely judicial approach and advocate the legislation of a common European system for legal recourse. Advocate General Jacobs acknowledges that disparity in the application of EC law is inevitable in the absence of harmonized rules on remedies, procedure, and time limits. 53 ECJ justices express regret over the absence of provisions to harmonize rules, but reject the notion that they can impose a comprehensive system on their own. 54 The European Commission also realizes that divergent means of national redress threaten the effective functioning of EC law and contribute to an illusory system of legal protection. 55 Christopher Himsworth argues that a systematic harmonization oflegislation would provide superior legal protection to the "incremental and piecemeal approach adopted by the Court of Justice so far." 56 De Biirca blames much of the national confusion over remedies on the lack of a legislated system of harmonized remedies and procedural rules to deal with breaches of European rights. 57 Meanwhile, Member States rejected the European Commission's proposal to harmonize rules for remedies and state liability. 58 In summary, the ECJ's limited development of European provisions for state liability and other remedies compromise the practical utility of invok51. Commission 1997. C-332, 200. .~2. Commission 1997, C-332, 204; Commission 2000, COM ('woo) 92, Annex VI, 31-2. 53· Van Schijndl Cr43o-1 I 93, 1996, E.C.R., I-4 705, paragraph 45· 54· Express Dairy Foods LTD. v Intervention Board for Agricultural Produce C-130l79, (1g8o) E.C.R., 1887 and Van Schijndel C-43o-1 I 93. 1996. 55· Commission, "Tenth Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law-1992," Official Journal of the European Communities C-233 (August 1993): 7; Commission, "Thirteenth Annual Report on Monitoring the Application of Community Law-1995," COM (96) 6oo final (29 May 1996), 14; Kieran Bradley and Alastair Sutton, "European Union and the Rule of Law," in Maastricht and Beyond: Building the European Union, ed. Andrew Duff, John Pinder, and Roy Pryce (London: Routledge, 1994), 251-52· 56. Himsworth 1997, 292, 307-11. 57· De Biirca 1992, 239· s8. Harlow 1996, 30.

Containing Justice

63

ing European law through direct or indirect effect. ECJ case law provides no comprehensive guidelines to national courts that are prepared to invoke European law. The potential to seek compensation through damages and access to other remedies continues to vary substantially across jurisdictions, with the result that parties in virtually identical situations can face very different legal outcomes. 59 As striking as the concepts of direct effect and state liability appear, therefore, they must be recognized as typical products ofjudicial decision-making that constrain the very obligations and rights that they create. Enduring restrictions and ambiguities in the EU's legal architecture generate a much weaker source of pressure to reform policy and a riskier, more unpredictable system of legal recourse than is typically assumed.

The Limits of the Law in Europe Ambiguity and incrementalism are not the only characteristics of judicial decisions that insulate policy processes from ECJ interference. The formal obligations to apply ECJ case law are also very narrow. The European system of judicial review operating through Article 2 34 ( 1 77 EEC) references only compels compliance with individual judgments because ECJ preliminary rulings do not constitute binding "precedents." The absence of generally binding legal obligations enables national officials to disregard principles that are articulated in ECJ case law. National administrations capitalize on this restriction to contain justice. Moreover, restrictive standing requirements for individuals and organizations impede the capacity of societal actors to broaden the impact of individual rulings.

Precedent and the Lack of Generally Binding Legal Effects

Formal European legal institutions do not oblige Member States to ascribe any general effects to most ECJ judgments. Originating in a community of civil law jurisdictions, the European legal system lacks the doctrine of stare decisis that establishes the institution of precedent. Article 234 ( 177 EEC) preliminary rulings have no erga omnes (generally binding) or ultra partes (beyond the parties) effects. With the exception of judgments declaring an act of an EU institution void in a direct action for annulment under Articles 2 30 and 231 ( 17 3 and 1 7 4 EEC), ECJ rulings have only inter partes effects, a

59· Chalmers 1997. 186.

64

Justice Contained

relative effect binding only between the parties in the particular case. 60 Once the ECJ clarifies the meaning of a European measure in a preliminary ruling, therefore, the referring national court is bound to apply the interpretation only to the particular facts of the case before it. The scope of the decision is formally limited to this single lawsuit. The ECJ decision has no direct effect for parties who are not in court. 61 Similarly situated individuals must bring their own proceedings before national courts to ensure direct enforcement of their EC rights. The subsequent case-by-case application of legal interpretation is not a particularly potent force to transform national law and policy. Restrictions on erga omnes effects, which formally govern interactions between courts, derive from the continental desire to limit the power of the judiciary relative to legislative and executive institutions. The obligation of a court to base decisions on prior judicial decisions, as required under the doctrine of stare decisis, has the effect of creating a rule for general application. Generally binding rules constitute law, which only legislators make in civil law systems. Judicial decisions lack the status of law if no court is bound by the decision of any other court. 6" The formal absence of binding precedents need not result in judicial practices that diverge substantially from those in common-law systems. Just as common-law judges distinguish cases from precedents and overturn precedents, civil-law judges take prior judgments into consideration and do not overrule previous cases without grounds for doing so. 63 John Henry Merryman observes that a civil-court judge may refer to previous decisions "because he is impressed by the authority of the prior court, because he is persuaded by its reasoning, because he is too lazy to think the problem through himself, because he does not want to risk reversal on appeal, or for a variety of other reasons. "64 The ECJ's consideration of its previous decisions reflects a convergence of practices. Prior to 1973, the ECJ generally just repeated its decisions without any indication it had ever ruled on the issue before. It also ruled in 1963 that national courts could treat the interpretation in a prior preliminary ruling as authoritative if materially identical questions were raised be6o. A. G. Toth, The Oxford Encyrlopaedia of European Community Law (Oxford: Clarendon Press, 1990), 1:202, 321, 465;Jeffrey Cohen, 'The European Preliminary Reference and U.S. Supreme Court Review of State Court Judgments: A Study in Comparative Judicial Federalism," American Journal of Comparative Law 44 ( 1996): 435-3 7. 61. SnupatvECSCHighAutlwrityCr32, 33/58, (1959) E.C.R., 275· 62. Henry Schermers and Denis Waelbroeck,judicial Protection in the European Communities, 4th ed. (Deventer: Kluwer, 1987), 83-84;John Henry Merryman, The Civil Law Tradition, 2d ed. (Palo Alto: Stanford University Press, 1985), 22,36-43, 45· 63. Schermers and Waelbroeck 1987, 83-84; Merryman 1985, 47; Hartley 1994, 83. 64. Merryman 1985, 47·

Containing Justice

65

fore them. 65 After 1973, when common-law judges from the United Kingdom and Ireland joined the Court of justice, the ECJ started to cite its previous judgments by name more frequently. 56 In a 1982 decision, the ECJ also decided that national courts could rely on interpretations from prior preliminary rulings even if the questions under consideration were not strictly identical. 67 The ECJ even began to send previous judgments to national courts that requested preliminary rulings, asking if they could find sufficient clarification without further references. For example, before deciding El Corte Inglis in 1996, 68 the ECJ sent the national court a copy of the 1994 Faccini Dori ruling69 and asked if the national judge still needed further guidance. This pattern of behavior mirrors common-law practice, and it invites national courts to treat ECJ rulings as de facto precedents. Indeed, Paul Craig and de Burca conclude that the ECJ developed a system of precedent by allowing national courts to rely on previous preliminary rulings. 70 Yet, the ECJ still looks more like a civil-law court when it diverges from previous case law, usually ignoring the discrepancy and avoiding any explicit overturning of earlier decisions. 71 Moreover, the system betrays its civil-law origins precisely in its refusal to confer any genuinely binding effects to prior judicial decisions. National courts may choose to apply legal principles articulated in ECJ case law across related situations, or they may ask for a new preliminary ruling if they prefer that the ECJ reconsider an issue.72 However, neither treaty obligations, secondary legislation, nor ECJ case law bind national courts to principles decided in prior preliminary rulings. A critical obligation is missing from this situation: national courts are not explicitly obliged to follow previous rulings until they resubmit the question and receive a different decision from the ECJ. 73 The lack of explicit precedential authority permits divergent concep65. Da Costa en Schaake Nv, Jacob Meijer NV and Hoechst-Holland NV v Nederlandse BelastingadministratieC-28-3o/62, (1963) E.C.R., 31. 66. Schermers and Waelbroeck 1987, 86. 67. Srl CILFIT and Lanificio di Gavardo SpA v Ministry of Health C-283/81, (1982) E.C.R., 34 1 5· 68. C-192/94, 1996. 69. C-91/92, 1994· 70. Paul Craig and Griinne de Burca, EU Law: Text, Cases, and Materials, 2d ed. (Oxford: Oxford University Press, 1998), 417,423. 71. Hartley 1994, 84. 72. Da CostaC-28/62, 1963; CILFITC-238/81, 1982; Schermersand Waelbroeck 1987,85, 399; Hartley 1994, 301; Alberto Trabucchi, "L' effet 'erga omnes' des decisions prejudicielles rendues par Ia Cour de justice des Communautes europeennes," Revue trimestrielle de droit europeen 10 (1974): 64-65, 7o--71, 81. 73· Toili 1990, 321; Hartley 1994, 301; Cohen 1996,428, 435·

66

Justice Contained

tions of the responsibilities that derive from EQJ case law. Former ECJ Advocate General Alberto Trabucchi and A. G. Toth argue that the relatively abstract formulation of preliminary rulings implies that they can be generally applicable: the ECJ interprets the meaning of EC law in preliminary rulings, and the national court then applies the interpretation to the facts of a specific case. 71 Since the meaning of the law should not change even if its application to particular situations may vary, giving general effects to interpretive rulings appears logically necessary. If this were not the case, the implication would be that EC provisions do not have unique interpretations.75 Toth argues that a national court confronted with questions already decided in preliminary rulings "will either apply that interpretation or refer the question to the European Court for a further reading. Such a [national] court may not adopt any different interpretation without misapplying Community law. "76 T. C. Hartley concurs that it "would be improper for [a national court] simply to depart from the ruling because it thought that it was wrong."77 Yet Trabucchi also stresses that the possibility to invoke previous case law as if it were a source of law is distinct from an obligation to apply past judicial decisions. According to Trabucchi, opportunities to invoke existing case law or request new preliminary rulings facilitate an efficient and flexible system ofjustice, while obligations to give binding force to past judgments crystallize ECJ case law into a rigid, immutable system of law. 7" Trabucchi's concern with rigidity and immutability reflect his misunderstanding of the use of stare decisis in common-law systems, where judges avoid such detrimental inflexibility by distinguishing and overturning precedent. Nonetheless, the distinction Trabucchi makes between the possibility and the obligation to invoke "precedent" is important because it leaves considerable room for national judicial discretion. National courts have accorded ECJ rulings a variable degree of precedential authority in doctrine and practice. National courts accepted the notion that preliminary rulings might apply across parallel cases at different points in time. And, national courts continue to fail to accord prior preliminary rulings with precedential effects in cases where they make decisions in direct contradiction to existing ECJ case law. National courts that attributed broader effects to preliminary rulings early on include the House of Lords, French Cour de Cassation, and German Bundesverwaltungsgericht (Federal Administrative 74· 75· 76. 77· 78.

Trabucchi 1974, 78-8o; Toth 1990,419-20. Cohen 1996,441. Toth 1990, 419. Hartley 1994, 301. Trabucchi 1974,64-65, 70-71, 81.

Containing Justice

67

Court). 79 Lord Denning of the United Kingdom even instructed national judges to apply existing preliminary rulings to related cases and avoid further references unless judges specifically sought to challenge previous EGJ case law. 8° Consistent with the common law, this instruction effectively treats preliminary rulings as precedents. By contrast, Italian courts initially held that they were free to decide whether they would apply prior rulings of the EGJ in cases before them. 81 A decade after the articulation of this cavalier, but legally accurate, attitude toward "precedent," the question that an Italian court referred to the ECJ reflects how foreign the notion of binding precedent was to the national judge. The Italian court wondered whether an earlier ECJ case that had declared a particular EC regulation to be invalid was effective in subsequent litigation, or whether such a finding was relevant only to the court that had initially received the EQJ's ruling." 2 The question perhaps respects that an EGJ decision has generally binding effects only if it derives from a direct action for annulment under Articles 230 and 231 (173 and 174 EEC). Yet, declaring a European regulation to be null and void would seem to imply that it is always null and void, regardless of the legal basis of the case in which invalidity was established. In response, the EGJ refrained from imposing any binding precedent, and instead merely informed the Italian court that it could rely on the previous finding of invalidity to regard the act as void in disputes between other parties. The EGJ acknowledged that the previous finding of invalidity was directly addressed to the national court that had originally referred the question and indicated that other national courts maintained the power to refer the matter once more to the ECJ. It was not until1997 that the Italian Corte di Cassazione held that the ECJ's preliminary rulings exerted effects in the domestic legal order that are analogous to a binding precedent: the Italian court ruled that ECJ decisions are immediately applicable, with the effect that national courts should disapply any national statutes found to be incompatible with the ECJ's interpretation of European law. 8' In practice, national courts base many decisions on prior EGJ rulings but 79· Schermers and Waelbroeck 1987, 399· 8o. P. P. Craig, "Report on the United Kingdom," in The European Court and the National Courts-Doctrine and jurisprudence: Legal Change in Its Social Context, ed. Anne-Marie Slaughter, Alec Stone Sweet, andJoseph Weiler (Oxford: Hart, 1998), 205-6. 81. Sorifla Anonima Fablnica Accumulatori (SAFA) v Amministrazione delle Finanze dello Stato, Corte di Appello di Milano, 1972, Common Market Law &ports 66 (1973), 155-s6; Schermers and Waelbroeck 1987, 398-99. 82. International Chemical Corporation v Amministrazione delle Finanze della Stato C-66/8o, (1gH 1) E.C.R., 1983; Craig and de Burca Igg8, 417-1 H. H3. Commission 1998, C-250, 202.

68

Justice Contained

they also continue to make decisions that directly conflict with existing ECJ case law. The Dutch Centrale Raad van Beroep refused to apply the ECJ's principle of nondiscrimination in a case on retirement pensions to a parallel case on invalidity benefits."' In a system with binding precedents, referring the case to the ECJ would have been more appropriate than simply refusing to respect ECJ interpretation in similar cases. But this Dutch court was simply following the practice of two supreme jurisdictions in the Netherlands. The Raad van State (Council of State) made three decisions in 1998 in direct conflict with ECJ rulings, 85 and the Hoge Raad ruled in direct contradiction to existing ECJ case law by overturning a lower court's decision that was compatible with a previous preliminary ruling."6 In Greece, the Symvoulio tis Epikrateias (Council of State) refused to recognize diplomas where some instruction was given in Greece by the staff of universities from other Member States. Two judgments on this issue in 1997 and 1998 clearly conflict with multiple treaty articles, a directive on the mutual recognition of diplomas, and ECJ rulings on EU nationals' access to employment as university instructors. 87 The Spanish Tribunal Supremo only decided in 1997 to reverse its refusal to recognize retirement contributions earned in other Member States, a position that had been in direct conflict with ECJ rulings for a decade. 88 As national judges become more familiar with European law, the incidence of direct conflict between national and ECJ case law may decline. Yet, inviting national courts to rely on prior preliminary rulings as guidance in the ongoing resolution of disputes is not equivalent to requiring national courts to apply ECJ "precedent." Only national judges have ever instructed national courts to treat prior ECJ rulings as binding in related cases. The formal absence of binding precedents in the European legal system, by contrast, constitutes a permissive condition that can always be relied on by national courts to legitimate a restrictive approach to ECJ interpretation. Evasion of Policy Implications by State Administrations

The narrow scope of ECJ rulings generates a particularly permissive environment for national governments and their administrations, which face no obligation to treat ECJ interpretation as a source of generally binding rules. The lack of erga omnes effects for ECJ judgments dilutes legal requirements, freeing national officials to claim that individuals must sue to prove 84. 85. 86. 87. 88.

Ibid., 196. Commission Commission Commission Commission

2000, COM 1999, COM 2000, COM 1999, COM

(2000) 92 final, Annex VI, 19. (gg) 301 final, 26g. (2ooo) 92 final, Annex VI, 17-18. (gg) 301 final, 260.

Containing Justice

69

that their specific cases merit the application ofEC law as interpreted in existing ECJ case law. Formally justifiable, such evasion contains judicial impact over policy processes. Even if national courts regularly granted precedent-setting authority to ECJ preliminary rulings, their decisions would still only apply to individual cases. To give EC law a meaningful general effect, national governments must often adapt legislation or initiate changes in administrative practices. Yet, initiating changes in law and practices is generally much more onerous than simply ignoring the universe of situations that may be parallel to those of past litigants. If a national measure is the subject of a preliminary ruling, therefore, compliance with an individual judgment is typically followed by business as usual. The national administration is unlikely to introduce broader changes in policy. Moreover, other Member States that have similar measures, but are not directly targeted by the preliminary ruling, face no obligation to accommodate the new interpretation of European law. Evidence I gathered from interviews with officials in France, Germany, and the United Kingdom indicates that this response of contained compliance constitutes standard administrative practice. Member State officials do not routinely treat preliminary rulings as binding policy prescriptions. Officials in ministries that apply law in specific policy areas were particularly likely, relative to their colleagues in central coordinating ministries, to stress difficulties in applying judicial principles as general policy. They distinguished directives, which would give them guidelines they must address, from case law, whose guidelines could be read restrictively. 89 Responses of contained compliance restrict the policy impact of ECJ decisions, but do not represent the defiant will of outlaws. France, Germany, and the United Kingdom maintain average to good records of compliance with individual ECJ rulings. The incidence of noncompliance with different ECJ judgments is near the mean (41.3) in France, well under one standard deviation above the mean (79) in Germany, and below the mean and median (27) in the United Kingdom. Outliers include Belgium, which is 89. Interviews at Foreign Ministry of France, Paris, t:i October 199s; Secretariat general de comite interministkriel pour les questions de cooperation economique europeenne (SGCI), Paris, 11 October 1995; Treasury Solicitors Office, London, 1 1 July 1995; Cabinet Office, London, 18july 1995; Bundesministerium fiirWirtschaft, Bonn, 18 and 19]anuary 1996; British Home Office, London, 24]uly 1995; Department of Social Security, London, 25 July 1995; Center for Social Security for Migrant Workers, Paris, 29 February 1996; Service for Women's Rights, Paris, 29 February 1996; Commission Directorate General V-A, Brussels, 14 September 1995; Commission Legal Services, Brussels, 18 September 1995· For a detailed case study on a long-standing instance of administrative evasion in the United Kingdom, and a scathing normative critique of the practice, see Mike Radford and Allan Kerr, "Acquiring Rights-Losing Power: A Case Study in Ministerial Resistance to the Impact of European Community Law," ModernJ.awReview6o (January 1997): 2')-43·

70

Justice Contained

over one standard deviation above the mean and Italy, which is over two standard deviations (116.8) above the mean (see table 3.1). Furthermore, given the narrowly binding scope of most ECJ rulings, the British, French, and German administrative practices are formally legitimate. The contained compliance engaged in by the administrations also does not necessarily reflect any deliberate attempts to evade legal obligations. National administrations may genuinely distinguish the context of a particular judgment from the characteristics of a potentially related national policy. They may also be patently unaware of the exact steps they need to take in order to fulfill a new legal obligation. The ambiguities associated with EQJ case law generate few obvious policy guidelines for Member State officials. Uncertainties that arise from indeterminate judicial principles dissuade general policy responses. As long as the application of legal principles remains open to debate, Member State officials generally maintain preferred policies. 90 Interpretations of case law by the European Commission usually do not persuade skeptical national administrations to change their policies. One national legal advisor dismissed the European Commission's communiques on ECJ case law: 'The Commission is the Commission; it is not objective. "91 An administrator at the European Commission conceded that most Member States pay virtually no attention to the European Commission's statements about the meaning of ECJ case law."" Instead, Member States usually await explicit legal challenges. If a lawsuit materializes, the government can attempt to persuade the ECJ of its position. If no claims emerge, governments remain unperturbed in their administration of domestic policy. Even a loss before the ECJ has an advantage: Member State officials get a judgment that addresses the particulars of their policies, which may be considerably more accommodating than the legal implications that the European Commission reads into ECJ case law. By waiting for a directly targeted ruling to confirm their obligations, Member States avoid "overimplementation," which entails policy adaptation that the ECJ may never have required. Officials from central coordinating ministries who must defend national actions before the ECJ lament the tendency for ministries engaged in the go. Interviews at Foreign Ministry of France, Paris, 6 October 1995; SGCI, Paris, 11 October 1995; Treasury Solicitors Office, London, 11 July 1995; Cabinet Office, London, 18 July 1995; Bundesministerium fiir Wirtschaft, Bonn, 18 and 1 9 January 1gg6; Commission Legal Service, Brussels, I .June 1995; British Home Offict>, London, 24]uly 1995; Center for Social Security for Migrant Workers, Paris, 29 February 1996; Service for Womens' Right,, Paris, 29 February 1996; Ministry for Women and Youth, Bonn, 17 January 1996; Commission Directorate General V-A, Brussels, 14 September 1995; Department of Social Security, London, 25 July 1995· 91. SGCI, Paris, 1 1 October 1995. 92. Interview at Commission Directorate General V-A, Brussels, 14 September 1995.

Table 3.1. Incidence of Delayed Compliance with ECJ Judgments, 1983-1999 Member State Italy Belgium Germany Greece France Spain Ireland Luxembourg Netherlands United Kingdom Denmark Portugal Austria Finland Sweden

Incidents of Delayed Compliance with Different Judgments

Average Number of Years Delayed

Longest Incident of Delay in Years

130 97 52 50 46 29 25 19 17 16 8 6 1 0 0

2.7 2.3 2.4 2.3 2.8 2.0 1.8 1.7 2.4 3.0 1.5 1.7 1.0 not applicable not applicable

12 13 7 9 10 6 5 4 6

9

4 2 1 not applicable not applicable

Sources: Commission, "First Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law-1983," COM (84) 181 final (11 April1984), 27-30; Commission, "Second Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law-1984," COM (85) 149 (6 May 1985), 28-31; Commission, "Third Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law-1985," Official Journal of the European Communities C-220 (1 September 1986): 22-26; Commission, "Fourth Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law-1986," Official Journal of the European Communities C-338 (16 December 1987): 26--32; Commission, "Fifth Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law-1 987," Official Journal ofthe European Communities C-31 0 (5 December 1988): 34-42; Commission, "Sixth Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law-1988," Official Journal of the European Communities C-330 (30 December 1989): 41-52; Commission, "Seventh Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law-1989," Official Journal of the European Communities C-232 (17 September 1990): 43-53; Commission, "Eighth Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law-1990," Official Journal of the European Communities C-338 (31 December 1991): 62--138; Commission, "Ninth Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law-1 991 ," Official Journal of the European Communities C-250 (28 September 1992): 55--£9; Commission, "Tenth Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law-1992," Official Journal of the European Communities C-233 (August 1993): 207-211; Commission, "Eleventh Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law-1 993," Official Journal of the European Communities C-154 (6 June 1994): 169-173; Commission, "Twelfth Annual Report on Monitoring the Application of Community Law-1994," Official Journal of the European Communities C-254 (29 September 1995): 16D-163; Commission, "Thirteenth Annual Report on Monitoring the Application of Community Law-1995," COM (96) 600 final (29 May 1996), 399-412; Commission, "Fourteenth Annual Report on Monitoring the Application of Community Law-1 996," Official Journal of the European Communities C-332 (3 November 1997): 193-1 97; Commission, "Fifteenth Annual Report on Monitoring the Application of Community Law-1997," Official Journal otthe European Communities C-250 (10 August 1998): 19Q-194; Commission, "Sixteenth Annual Report on Monitoring the Application of Community Law-1998," COM (99) 301final (9 July 1999), 23Q-240; Commission, "Seventeenth Annual Report on Monitoring the Application of Community Law-1 999," COM (2000) 92 final (23 June 2000), Annex V, 2-11.

72

Justice Contained

administration of policy to wait for a specific judgment in instances in which previous ECJ case law virtually guarantees they will lose. 93 Once a challenge arises, the defense is likely to seem futile in principle and practice. Indeed, the need to exercise consistent control over administrative action historically led to the relaxation of restrictions on erga omnes effects for particular courts within civil-law countries. Merryman explains the functional logic: "A court that was not bound by its own or other courts' prior decisions, and that consequently could make a decision binding only on the parties to the case before it, was inadequate to the task of keeping administrative action within acceptable bounds.''94 Such functional utility did not lead to the institutionalization of erga omnes effects for ECJ preliminary rulings, however, and as a result, national officials remain free to decide how they will treat existing ECJ case law. Despite the chagrin of those who represent recalcitrant administrations, the most common practice remains contained compliance. In exceptional cases, national administrators voluntarily choose to apply judicial decisions beyond litigants. A clear principle of EC law, combined with an unambiguous target population, improves the chances that national administrations will adjust their actions on a general level. For example, the U.K administration applied an innovative legal principle universally in the case of Antonissen, where the ECJ decided that those seeking work may reasonably reside and collect social benefits for six months. Because it would be extremely difficult to justifY why any one individual or another would need less time than this, the United Kingdom accepted six months as a standard for anyone applying for social benefits. 95 Yet, such a response is not automatic as patterns of contained compliance in similarly "obvious" cases demonstrate in chapter 7. Indeed, national officials who deal exclusively with EC law in particular policy areas had difficulties thinking of examples in which they might consider initiating a general policy response to an ECJ ruling. In those they could cite (rarely more than one or two when reflecting back on the last decade), national officials emphasized that they interpret judicial principles narrowly. 96 93· Interviews at Foreign Ministry of France, Paris, 6 October 1995; SGCI, Paris, 11 October 1995; Treasury Solicitors Office, London, 11 July 1995; Cabinet Office, London, 18 July 1995; Bundesministerium fur Wirtschaft, Bonn, 18 and 19January 1996; British Home Office, London, 24 July 1995; Center for Social Security for Migrant Workers, Paris, 29 February 1996; Department of Social Security, London, 25July 1995; Service for Womens' Rights, Paris, 29 February 1996; Ministry for Women and Youth, Bonn, 17 January 1996. 94· Merryman 1985, 133-34. 95· Gustaf!Desiderius Antonissen (ex parte) v United Kingdom C-2g2/8g, ( 1991) E.C.R., I-745; Interview at British Home Office, London, 24july 1995. 96. Interviews at British Home Office, London, 24July 1995; Center for Social Security for Migrant Workers, Paris, 29 February 1996; Department of Social Security, London, 25 July 1995·

Containing Justice

73

Restrictive Standing Another institutional restriction that limits the scope of Eq decisions is the lack of class actions in the European legal system. The Court of justice rejected the notion that groups, as representatives of a class, could be individually concerned with general measures, blocking the way for class actions.97 The ECJ has generally denied standing to individuals or groups who seek to represent collective interests. This contrasts with relaxed standing requirements in the United States, where judges decided that the collective or general quality of an interest should not exclude it from judicial protection. Restricted opportunities for standing in the EU have blocked access to European legal recourse for many individuals and small businesses. 98 As a result, actors who might have difficulty organizing collective action in political arenas lose a critical source of legal leverage as well. The aggregation of claims in a class action requires a minimum of coordination, but presents a defendant with a deluge of simultaneous demands for change. Case-by-case application of European law typically proceeds, therefore, in the most incremental and individualized manner possible. The increased pressures that class actions generate will only be accessible when national courts apply European law and national procedures allow for class actions. European legal institutions and the common practices of national officials together confine the impact of ECJ decisions in policy processes. The narrow scope of most ECJ rulings, responses of contained compliance within national administrations, and prohibitions against class actions all act to insulate policy-making from judicial interference.

Discretionary Enforcement by the European Commission and National Courts Two mechanisms of enforcement can be engaged to broaden the impact of Eq interpretation: prosecution by the European Commission and the application of European law by national courts. Centralized prosecution by the European Commission through Article 226 (169 EEC) infringement proceedings was originally intended to be the primary means to enforce European law. But the ECj's rulings on supremacy, direct and indirect effect, and state liability also created opportunities for private actors to engage national courts in a decentralized system of enforcement. Individuals 97. Confederation National£ des Fruits et Producteurs des Fruits et Legumes v Council C-16, 17/62, (1962) E.C.R., 937· g8. Carol Harlow, "Towards a Theory of Access for the European Court of justice," in Yearbook of Eumpean Law, ed. A. Barav, D. A. Wyatt, and Joan Wyatt (Oxford: Clarendon Press, 1993), 12:233. 237·

74

Justice Contained

can enforce their European rights if a national court directly applies European law to their case. Yet both legal procedures constitute highly discretionary means of enforcement that vary in the extent to which they pressure Member States to align policy with European legal obligations. Prosecution under centralized infringement proceedings is contingent on the prevailing resources and goals of the European Commission and remains largely closed to outside scrutiny and challenge. The benefit of this mechanism of enforcement, however, lies in its typical requirement that Member States reform legislation to accommodate European obligations. As a result, evasion through contained compliance is not generally viable. Decentralized enforcement though national courts, meanwhile, depends on the variable ability and willingness of national courts to deploy the European legal system to the advantage of claimants. The benefit of this mode of enforcement is that affected actors enjoy recourse as parties to the proceedings and may appeal adverse decisions to higher jurisdictions. Yet recourse through this channel ultimately relies on national courts oflast instance, which are not the more subservient allies of the ECJ and its legal interpretation. More generally, the disadvantage of this enforcement mechanism is its limited scope. The only way for national judicial enforcement to combat contained compliance is through persistent litigation that repeatedly holds administrations accountable to their European obligations. Evidence on national judicial behavior suggests that opportunities to generate pressure through persistent litigation remain limited. The European Commission and Infringement Proceedings

Under Article 226 (169 EEC), the European Commission holds the power to bring infringement proceedings against Member States that violate their EC legal obligations. Proceedings begin with a "letter of formal notice" drafted by the European Commission, which identifies a suspect Member State law or practice and aims to establish a dialogue on the issue. If the European Commission is not satisfied with the Member State's response, it issues a "reasoned opinion," which represents the European Commission's finding of an infringement. The European Commission resolves over So percent of its infringement proceedings at these two prelitigation stages of the process. 99 If a Member State fails to accommodate changes required at the reasoned opinion stage, this should trigger prosecution in a "direct action" before the European Court of Justice; that is, no national courts are involved. 100 If Member States then delay compliance with the ECJ' s ruling, the European Commission may initiate proceedings to sanction the nongg. Commission 1997, C-322, g. Commission 1993, C-233, 6-7.

100.

Containing Justice

75

compliance with penalty payments under Article 228 ( 171 TEU) .101 Intended to have a deterrent effect, these fines accrue on a daily basis and are calibrated according to the offending Member State's Gross Domestic Product (GDP) and voting weight in the EU. According to the European Commission, its threats to fine states and its initiation of proceedings for penalty payments have induced rapid compliance in the targeted Member States. Since 1997, the European Commission has pursued sanctions for eighteen cases of noncompliance with ECJ judgments but had imposed only one fine as of July 200o."" Successful infringement proceedings should generally produce an actual policy change because compliance usually entails legislative reform that establishes new rules for general application. The main dilemma associated with this centralized system of enforcement is the lack of procedures to hold the European Commission accountable to its obligation to enforce European law. Suspected infringements come to the attention of the European Commission overwhelmingly through the complaints of firms and individuals who are frustrated by national policies. Table 3.2 indicates that inquiries by the European Commission constitute a small percentage of the cases that result in letters of formal notice. 103 However, the decision to initiate and sustain infringement proceedings rests entirely with the European Commission. The European Commission does not possess the resources to pursue all potential violations of European law/ 04 so it strategically selects infringement cases that are promising on legal grounds and serve the European Commission's broader political and institutional interests. 105 Engaged in a continual process of negotiation to approve new legislation and implement existing law uniformly, the European Commission will not choose lightly to antagonize Member States with infringement proceedings. In reacting to legal obligations that arise out of innovative ECJ interpretation, therefore, the European Commission rarely acts on the basis of a few rulings, but usually waits to see that a legal principle develops consistently through a series of cases. 106 Such caution improves its chances of sue101. Treaty on European Union ( 1993), the "Maastricht Treaty." 102. Commission 1997, C-332, 15-16; Commission 1998, C-250, 8-9; Commission 1999, COM (99) 301 final, introduction; Commission 2ooo, COM (2ooo) 92 final, 11; Commission v GreeceC-387197, (2ooo) E.C.R., not yet reported. 103. Commission 1996, COM (96) 6oo final, 6, 14. 104. Bradley and Sutton 1994, 249-50 and Harlow and Rawlings 1992, 276, 278. 105. Bradley and Sutton 1994, 257; Harlow 1993, 223. Interviews at Commission Legal Service, Brussels, 1 and 2 June 1 995 and 18 and 21 September 1995; Commission Directorate General V-D, Brussels, 31 May 1995 and 18 and 27 September 1995; Commission Directorate General V-A, Brussels, 14 September 1995; Commission Directorate General IV-B, Brussels, 30 May 199.~; Commission Directorate General IV-A, Brussels, 2june 199.5· 106. Interviews at Commission Legal Service, Brussels. 2 June 199'); Commission Directorate General V-D, Brussels, 31 May 1995 and 18, 22, and 27 September 1995.

76

Justice Contained Table 3.2. Aggregate Legal Inquiries in the European Union, 1983-1997

Legal Inquiry Letters of formal notice European Commission inquiries Complaints References for preliminary rulings National court decisions on European law

1983-1997 12,428 4,078 13,838 2,548 14,671'

Sources Court of Justice of the European Communities, Synopsis of the Work of the Court of Justice and Annual Report of the Court of Justice (Luxembourg: Office for Official Publications of the European Communities, 1986, 1988, 1990, 1991, 1992, 1995, 1996, 1997, 1998), 83,136, 125, 137,136 and 163, 30Q-301, 234,202, 228; Commission, "Fifth Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law-1987," Official Journal ofthe European Communities C-310 (5 December 1988): 23; Commission, "Sixth Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law-1988," Official Journal of the European Communities C-330 (30 December 1989): 53; Commission, "Eighth Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law-1990," Official Journal of the European Communities C-338 (31 December 1991 ): 77, 83; Commission, "Tenth Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law1992," Official Journal of the European Communities C-233 (August 1993): 23, 212; Commission, "Twelfth Annual Report on Monitoring the Application of Community Law-1994," Official Journal of the European Communities C-254 (29 September 1995): 56, 63, 164; Commission, "Thirteenth Annual Report on Monitoring the Application of Community Law-1995," COM (96) 600 final (29 May 1996), 91-95,103,111, 414-415; Commission, "Fourteenth Annual Report on Monitoring the Application of Community Law1996," Official Journal otthe European Communities C-332 (3 November 1997): 198; Commission, "Fifteenth Annual Report on Monitoring the Application of Community Law-1997," Official Journal of the European Communities C-250 (1 0 August 1998): 195. 'This figure includes cases decided on points of European law, regardless of whether the national court made a reference or not. Due to a change in reporting procedures in the ECJ's Annual Report, cases decided between July 1, 1995, and December 31, 1995, are not included in this total.

cess before the ECJ, but may delay prosecution designed to align general policies with evolving legal obligations. The failure to initiate infringement proceedings or the choice to abandon these proceedings is virtually never open to challenge. Individuals who claim that the European Commission neglects its obligations to enforce EC

Containing Justice

77

law by failing to take action against a Member State normally have no standing to protest the European Commission's "failure to act" under Article 232 (175 EEC). Under this treaty provision, individuals gain standing only when they can prove that the European Commission has failed to initiate infringement proceedings against a violation of an EC decision that is addressed to them personally. This prevents most individuals from challenging lax prosecution efforts by the European Commission. The ECJ, meanwhile, has taken no steps to tame administrative discretion by expanding standing to allow individuals to challenge the European Commission. 107 As a result, the European Commission is free to abandon most complaints, whether they are well founded or not, without facing the prospect of judicial review. 108 Even the European Ombudsman, who provides redress for citizens who are denied their rights or face "maladministration" in EU institutions, has not yet challenged the discretion of the European Commission in infringement proceedings. So far the European Ombudsman has adopted a conservative and formal approach that reserves maladministration for situations in which an institution acts outside the limits of its legal authority, thus respecting the broad discretion enjoyed by the European Commission according to ECJ case law. 109 Complainants have not even possessed rights to be informed about the status of their complaints or the reasons why the European Commission has decided to drop a proceeding at any given stage. Historically, infringement proceedings have been handled in a discretionary, cooperative, and closed process, with the goal of inducing compliance through quiet accommodation. The European Commission and Member States justifY the need for secrecy on the logic that it promotes open dialogue, mutual confidence, and effective compromises. 110 These rationales may hold at the first stage, involving letters of formal notice. However, demands for greater transparency and procedural consistency at the later stages have compelled the European Parliament and European Ombudsman to criticize the infringement process.m The European Commission has responded to these criticisms by speeding up the processing of suspected infringements, issuing press releases when proceedings reach the stage of reasoned opinions or 107. Lord Bethell v Commission C-246/81, (1982) E.C.R., 2277; Harlow 1993, 227-31; Harlow and Rawlings 1992, 276, 279-81; Carol Harlow, "A Community oflnterests? Making the Most of European Law," Modern Law Review 55 (May 1992): 348-49; Bradley and Sutton 1994, 24 7-48, 256-6o; Christopher Harding, "Who Goes to Court in Europe? An Analysis of Litigation Against the European Community," European Law Review 17 (April 1992): 107; Hartley 1994, 362-67, 411-14; Richard Rawlings, "Engaged Elites: Citizen Action and Institutional Attitudes in Commission Enforcement," European Law journal6 (March 2000): 1 2. 108. Bradley and Sutton 1994, 248. 109. Rawlings 2000, 1.~. 18. 11 o. Ibid., 9, 13. 111. Commission 1997, C-332, g-w; Commission 1998, C-250, 9-10.

78

Justice Contained

referrals to the ECJ, and notifying claimants of decisions to close cases. Proceedings at the initial stage remain confidential, but the European Commission may now ask for claimants' opinions before closing cases and provide reasons for why it decides to close cases. 112 Despite these improvements in efficiency and transparency, complainants remain fundamentally dependent on the discretion of the European Commission. Because complainants are not parties to the proceedings at any stage of the infringement process, they also rely entirely on the European Commission's representation when infringements reach the ECJ. Individuals have no right to present arguments or evidence in interventions before the ECJ in cases between or among Member States and EU institutions. 113 Procedures at the Court of Justice derive predominantly from the French model of civil law, where an independent court official (the Advocate-General) fulfills the role of amicus curiae that includes societal interventions in the U.S. system.u• Therefore, individuals and firms, who are usually the ones to suffer the immediate consequences of Member States' disregard of European law, exert no direct influence in infringement proceedings. The relatively slow process of resolving infringements constitutes the final frustration in enforcing European law through this mechanism. Cases where Member States failed to communicate how they would transpose European law typically last two to three years, and cases of incorrect transposal or application average five years.u 5 Delayed implementation of judgments against Member States (see table 3.1) and the necessity of starting the whole process over again to pursue fines then further extends the enforcement process. Recognizing the limitations of infringement proceedings, the European Commission advocates an expansion in the decentralized system of enforcement available through national courts: it encourages complainants whose claims it does not have the resources or interest to pursue to bring actions before national courts. 116 This position reflects little sensitivity to the practical barriers to litigation that many individuals face, particularly in those areas of law where interests are weak or diffuse. Environmental pro112. Commission 1gg7, C-332, g-1g; Commission 1gg8, C-250, 8-10; Rawlings 2000, 15-22. 113. Article 37 of the Protocol on the Statute of the Court ofJustice of the European Economic Community, 17 April1g57. 114. Harlow and Rawlings 1gg2, 281. ll5. Rawlings 2000, 22. 116. Bradley and Sutton 1g94, 247; interviews at Commission Directorate General V-D, Brussels, 31 May 19g5 and 18 and 27 September 19g5; Commission Directorate General V-A, Brussels, 14 September 1995; Commission Legal Services, Brussels, 18 September 1995.

Containing Justice

79

tection constitutes one of the few such areas where the European Commission has consistently recognized these challenges and responded proactively by facilitating complaints and even funding environmental nongovernmental organizations (NGO) .117 More generally, the European Commission's call to delegate enforcement tasks is problematic in light of its explicit recognition that national means of redress for infractions of European law are far from adequate. 118 Variable Enforcement before National Courts

Litigation before national courts constitutes the second means to enforce European legal obligations. Actors with a stake in European rights bring their disputes about the application of European law before national courts. Any national judge may then refer questions about the interpretation of "traditional" EC provisions from the EU's "first pillar" to the ECJ, or national judges may choose to apply ECJ interpretation from prior preliminary rulings on related points. National judges from lower and intermediate appellate courts can also simply disregard European measures and decide the case on the basis of national law alone. In this event, parties seeking to enforce European rights must appeal to higher jurisdictions, potentially until they reach the court of last instance, which is formally obligated to refer questions of EC law to the ECJ for a preliminary ruling. Exceptions to this procedure exist for two areas of the EU's "third pillar" that became legally binding only in the aftermath of the Amsterdam Treaty. First, the Amsterdam Treaty brought measures on visas, asylum, immigration, and other policies concerning the free movement of persons under the new Title IV of the EC, where only national courts oflast instance may refer questions to the ECJ for preliminary rulings. Second, the Amsterdam Treaty allows Member States to declare whether or not they accept the ECJ's jurisdiction over measures from the remaining third pillar issues of police and judicial cooperation in criminal matters. Ten Member States decided to allow any of their national courts to make references to the ECJ on measures adopted under this pillar. Spain chose to restrict reference rights to courts of last instance, while Denmark, France, Ireland, and the United Kingdom have not yet declared that they accept the ECJ' sjurisdiction at all. 119 Finally, all national judges may apply European law on their own, without any ECJ guidance. According to the ECJ, national courts should only apply European law independently if its correct interpretation is so obvious as to leave no scope 117. II

Rawlings 2000, 14. Commission 1996, COM (96) 6oo final, 7,

14 and Commission 1993, C-233, 7· g. Official journal of the European Communities L-1 14 (1 May 1999): 56.

118.

80

Justice Contained

for any reasonable doubt. Assessing the meaning of European law entails a comparison of the different language versions of European measures, all of which are equally authentic, and an understanding of the state of evolution in European law as a whole. 120 The ECJ is clearly best situated to perform these assessments, which suggests that it would prefer that national courts make references rather than apply European law without any of its interpretive guidance. This decentralized system provides many points of access to recourse under European law, enabling the pursuit of far more claims than could ever be processed directly through EU institutions alone. Individuals seeking to enforce European rights before national courts are also parties to the proceedings, with direct representation of their interests. However, the impact of this decentralized mode of enforcement will often be very narrow, tailored to the particular problems of the individuals who file suit. National courts apply EC provisions over conflicting national measures for parties before them, generally producing case-by-case application of European measures rather than major changes in legislation or policy. As a result, overcoming practices of contained compliance requires either persistent litigation or credible threats of unrelenting legal challenge. Efforts to contain compliance only become futile in the face of a deluge of copycat cases. Orchestrating sustained legal pressure through national courts, however, remains difficult for a number of reasons. Access to recourse and remedies is complicated by uncertainties about the concrete application of European legal doctrines such as direct effect and state liability, as the first section of this chapter demonstrated. Patterns of national judicial behavior, which I turn to next, also suggest that opportunities to invoke European rights vary across jurisdictions within and between Member States. Insufficient knowledge about European law, delays for preliminary rulings, and resistance to ECJ case law all contribute to incidents in which national courts fail to enforce European rights. Forum shopping and appeals are generally the only means to circumvent these problems because neither the European Commission nor the European Ombudsman are willing to sanction national courts for inappropriate behavior. As a result, actors who lack the resources to appeal and forum-shop will generate only sporadic litigation, which allows Member States to contain compliance. And, access to recourse effectively disappears if national courts of last instance pose obstacles to the application of European law. The following evidence shows that national courts handle a large volume of "European" litigation, but often on terms independent of ECJ guidance and unlikely to generate systematic pressures for broader policy changes. 120. Srl Clll1T and Lanificio di Gavardo SpA v Ministry of Health C-283/81, (1982) E.C.R, 3415-

Containing Justice

81

National courts do resolve a significant number of disputes that turn on questions of European law. Their decisions on EC law slightly outnumber infringement proceedings that begin with the European Commission's "letters of formal notice" in the EU as a whole. Only a few small Member States (Greece, Ireland, Luxembourg, and Portugal) face substantially fewer European legal inquiries through their national courts than through the European Commission's letters of formal notice and reasoned opinions. The overwhelming majority of national court decisions take place outside the context of a reference to the ECJ for a preliminary ruling (with the exception of those in Austria, Finland, and Sweden, who joined the EU in 1995; see tables 3.2 and 3.3). National courts are most often applying European law, therefore, without any formal interaction with the ECJ. The uniformity of this legal recourse, its conformity to ECJ interpretation, and its likelihood of generating major pressures for policy change, however, are all questionable. Unfamiliarity with European law remains an important initial obstacle to effective legal protection before national courts. In 1997 the European Commission found it necessary to launch an action program, the Robert Schuman Project, to improve awareness of European law within the legal professions, including national judges, prosecutors, and lawyers. Results from eighty local projects supported by this campaign in 1997 and 1998 showed that judges and lawyers in EU Member States still need a great deal of training in European law. 121 National judges also personally report that it is difficult to gain access to EC law and to keep up with ECJ case law. 122 The potential magnitude of basic knowledge gaps is evident from a European Commission data base that represents one of the few comprehensive efforts to analyze all existing national case law in a particular area of law, in this case a directive on unfair terms in consumer contracts: only 4·4 percent of the judgments delivered by national courts in the field covered by the directive referred to the EC text. What is particularly striking is that the European Commission saw this minuscule percentage as a reflection of the "progressive impact" of EC law on nationallaw. 123 This would leave 96.6 percent of the litigants utterly dependent on the correct and timely trans121. Commission 1998, C-250, 1o-11, 13-14; Commission 1999, COM (99) 301 final, 2. 122. Claire Favre, "De l'Efficacite de Ia Mise en Oeuvre du Droit Communautaire de Ia Concurrence par le Juge National aIa Difficile Connaissance du Droit Materiel Europeen," in Towards a Europeanised .Judiciary? Practitioners' Experiences of National .Judges with the Europeanisation of Private Law, ed. Sonja Feiden and Kristina Riedl, EUI Working Paper lAW no. 2000/3 (2ooo), 26-35; G. 0. Zacharias Sundstrom and Matti R. S. Kauppi, eds., Access to justice: A Record of the Thoughts and Ideas Dealing with the Interrelationship between National Law and Courts and Community Law and Courts (Helsinki: Finnish Association for European Law, 1999), 50. 123. Cosimo Monda, "Monitoring the Enforcement of the Unfair Contract Terms Directive: The European Commission Database on Case Law About Unfair Contractual Terms (CLAB)," in Towards aEu1opeanisedjudiciary? 58-59·

82

Justice Contained

Table 3.3. Aggregate Legal Inquiries by Member State, 1983-1997

Member State Belgium Denmark Germany Greece Spain France Ireland Italy Luxembourg Netherlands Austria Portugal Finland Sweden United Kingdom

References for Preliminary Rulings

National Court Decisions on European LaW'

Letters of Formal Notice

Reasoned Opinions

264 61 657 48 66 414 25 407 29 286 43 24 9 17 198

1,251 152 3,388 198 709 2,525 165 2,039 96 2,835 55 65 19 24 1,150

1,054 574 995 1,309 895 1,277 818 1,448 780 766 245 943 370 146 826

450 60 361 539 260 440 278 724 265 187 40 330 8 6 173

Sources: Court of Justice of the European Communities, Synopsis of the Work of the Court of Justice and Annual Report of the Court of Justice (Luxembourg: Office for Official Publications of the European Communities, 1986, 1988, 1990, 1991, 1992, 1995, 1996, 1997, 1998), 83, 136, 125, 137, 136 and 163, 300-301, 234, 202, 228; Commission, "Fifth Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law-1987," Official Journal ofthe European Communities C-310 (5 December 1988) 23; Commission, "Sixth Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law-1988," Official Journal of the European Communities C-330 (30 December 1989): 53; Commission, "Eighth Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law-1990," Official Journal of the European Communities C-338 (31 December 1991 ): 77; Commission, "Tenth Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law1992," Official Journal of the European Communities C-233 (August 1993): 23, 63, 212; Commission, "Twelfth Annual Report on Monitoring the Application of Community Law-1994," Official Journal of the European Communities C-254 (29 September 1995): 164; Commission, "Thirteenth Annual Report on Monitoring the Application of Community Law-1995," COM (96) 600 final (29 May 1996), 111, 415; Commission, "Fourteenth Annual Report on Monitoring the Application of Community Law-1996," Official Journal of the European Communities C332 (3 November 1997): 198; Commission, "Fifteenth Annual Report on Monitoring the Application of Community Law-1997," Official Journal of the European Communities C-250 (10 August 1998): 195; Commission, "Sixteenth Annual Report on Monitoring the Application of Community Law-1998," COM (99) 301 final (9 July 1999), 245; Commission, "Seventeenth Annual Report on Monitoring the Application of Community Law-1999," COM (2000) 92 final (23 June 2000), Annex VI, 14; Annex II, 13. a As noted in Table 3.2, the figures for court decisions include all cases decided on points of European law, with the exception of the period between July 1, 1995, and December 31, 1995.

posal of this directive into national law, because only a precious few national courts were applying the directive directly, indirectly, vertically, or horizontally. National judges who do invoke European provisions in their decisions are likely to interpret the treaties, regulations, and directives without any explicit reliance on ECJ case law. Table 3·4 indicates that the independent application of EC measures outpaces citations of ECJ case law in French,

Containing Justice

83

Table 3.4. French, German, and U.K. Courts' Citations of the ECJ and European Law

French Gaur de Cassation and its courts French Conseil d'Etat and its courts German courts U.K. courts

ECJ Rulings

Treaties

Directives

Regulations

1,347 227 1,171

1,967 821 2,529

1,051

768 682 3,273

176

901

892 1,538 over 1,000

968

Sources: Jurifrance searches for French case law on 17, 20, and 23 November 2000; Juris searches for German case law on 17 November and 3 December 2000; Lexis search for U.K. case law on 23 November 2000. Note: Comparisons should be made exclusively among court systems within the same country due to variations in the coverage of databases. The greater rate of French citation to ECJ decisions is primarily due to the need to do free text searches on judgments in Jurifrance, which yield higher numbers than Juris and Lexis searches that isolate cases formally grounded on ECJ case law. Juris and Jurifrance also both cover more case law from lower courts than Lexis, which makes direct cross-national comparisons untenable.

German, and U.K. courts. Only the normal judiciary courts in the French Cour de Cassation system are more likely to cite ECJ case law than directives or regulations. And, the judges in this branch of the French judiciary are still more prone to apply the treaties than cite ECJ rulings. Because of differences in the coverage of databases on each country's courts, cross-national comparisons of these citations are not justifiable. The important trend is that within each system, covered by the same database, heavy reliance on ECJ "precedent" is not apparent. Instead, national judges who decide to apply European law appear to consider themselves capable of determining its meaning on their own. This practice of independent, decentralized judicial review is very likely to lead to divergent interpretations of European law across jurisdictions. Most national judges are obviously not going to compare the different language versions of EC texts, so linguistic distinctions alone will produce some variation in interpretation across countries. More important, even national judges who follow ECJ case law carefully are hardly going to follow much of the national case law on European law developing within fifteen other Member States. Moreover, national judges who have little experience with the European legal system have nonetheless demonstrated their willingness to apply EC law independently. Supreme administrative courts in Finland and Sweden declined to make references to the ECJ on questions of interpretation despite a disagreement on the application of EC provisions within the Finnish court and despite the need to apply six different treaty articles and two directives without the guidance of prior ECJ case law in the case of the Swedish court. 1 ~ 4 And, it is clear that national judges who

124.

Commission

2000,

COM (2ooo)

92

f1nal, Annex VI,

20.

84

Justice Contained

apply EC provisions in areas lacking ECJ case law can come up with different legal interpretations than the ECJ. For example, the German Bundesverwaltungsgericht decided that a particular German measure restricting women's employment options in the military was in accordance with a directive on equal treatment between men and women. Despite its formal obligation to refer questions of EC law to the ECJ, this court of last instance did not bother to make a reference for an interpretation of the relevant EC provision. About six months later, the Hannover Verwaltungsgericht received a preliminary ruling from the ECJ on the same provision, which clearly indicated that the blanket prohibition in the German law would be incompatible with the directive. 125 Variations in interpretation are inevitable features of court systems and need not pose undue threats to the protection of rights if they remain open to correction. Appeals normally provide high courts with the opportunity to reconcile conflicting lines of legal interpretation from courts beneath them. But both the ECJ and parties seeking to enforce European rights face the obstacle that references for preliminary rulings are not "appeals" under the control of the ECJ. Even those parties who can afford appeals have no right to appeal to the ECJ, but instead depend on national courts to refer their cases to the ECJ. The ECJ, in turn, will only be able to consider discrepancies in national courts' case law on European law when national courts refer questions that address such conflicts. As a result, national courts are critical gatekeepers, determining the nature and extent of recourse under European law. National judicial preferences appear to influence the propensity for national judges to refer cases for preliminary rulings or explicitly cite prior ECJ case law. The propensity to send references for preliminary rulings is broadly consistent with conventional accounts of national judicial empowerment: national judges from lower and intermediate appellate courts, which have the most influence to gain through the opportunity to exercise "European judicial review" refer questions to the ECJ most frequently. 126 In the EU as a whole, national courts of first instance referred the greatest number of questions to the ECJ from 1961 to 1980, while intermediate ap125. Ibid., 15. 126. Anne-Marie Burley and Walter Mattli, "Europe before the Court: A Political Theory of Legal Integration," International Organization 47 (winter 1993): 41-76; Joseph Weiler, "A Quiet Revolution: The European Court of Justice and Its Interlocutors," Comparative Political Studies 26 (January 1994): 510-34; Karen Alter, 'The European Court's Political Power," West European Politics (July 1996): 458-87; Karen Alter, Establishing European Law Supremacy: The Making of an International Rule of Law in Europe (Oxford: Oxford University Press, 2001); Anne-Marie Slaughter, Alec Stone Sweet, and Joseph Weiler, eds. The European Court and the National Courts-Doctrine and ]urisprudrnce: Legal Change in Its Social Context (Oxford: Hart, 1998).

Containing Justice

85

Table 3.5. National Court Interaction with the ECJ up to 1999 Member State Germany Italy France Netherlands Belgium United Kingdom Spain Austria Denmark Greece Luxembourg Ireland Portugal Sweden Finland

References for Preliminary Rulings from Lower Courts

References from Courts of Last Instance

Total References for Preliminary Rulings

812 532 534 214 333 255 86 66 66 48 21 13 16 15 11

350 93 77 302 77 36 39 47 15

1162 625 611 516 410 291 125 113 81 56 46 39 38 22 15

8

25 26 22 7 4

Sources: Court of Justice of the European Communities, Report of Proceedings 1992-1994: Synopsis of/he Work of the Court of Justice and of the Court of First Instance of the European Communities (Luxembourg: Office for Official Publications of the European Communities, 1995), 261; Commission, "Thirteenth Annual Report on Monitoring the Application of Community law-1995," COM (96) 600 final (29 May 1996), 415; Commission, "Fourteenth Annual Report on Monitoring the Application of Community Law-1996," Official Journal of the European Communities C-332 (3 November 1997): 198; Commission, "Fifteenth Annual Report on Monitoring the Application of Community Law-1997," Official Journal of the European Communities C-250 (10 August1998): 195; Commission, "Sixteenth Annual Report on Monitoring the Application of Community Law-1998," COM (99) 301 final (9 July 1999), 245; Commission, "Seventeenth Annual Report on Monitoring the Application of Community Law1999," COM (2000) 92 final (23 June 2000), Annex VI, 14.

pellate courts gained a slight edge over courts of first instance as of 1981. 127 Courts of first instance continue to possess the most active referral rates in Belgium, Germany, Greece, and Portugal, while intermediate appellate courts refer more questions in Denmark, France, Ireland, Italy, the Netherlands, Spain, and the United Kingdom. 12K Historically, courts of last instance have referred a minority of cases in most Member States, with the exception oflreland, Luxembourg, the Netherlands, and Portugal (see table 3-s). Meanwhile, the constitutional courts of Belgium, France, Germany, and Italy, none of which have anything to gain by interacting with the ECJ, have not made any references to the ECJ.' 29 The concentration of lower-court references has positive implications for 127. Alec Stone Sweet and Thomas Brunell, 'The European Court and the National Courts: A Statistical Analysis of Preliminary References, Ig61-1995·" Journal ofEuropean Public Policy (March 1gg8a): figure 7. 128. Stone Sweet and Brunell 1gg8a, figure 8. 129. Alter 1gg6, 46,1).

86

Justice Contained

the application of EC law: prompt referrals to the ECJ obviate the need to appeal cases through multiple levels of the national judiciary. Conversely, the concentration of references from courts of last instance in some Member States indicates that efforts to invoke EC law must percolate to the top level of the national judiciary. A Dutch Member of the European Parliament has criticized concentrations in upper court references, true of the Netherlands, as illegitimate attempts to restrict the reference powers of lower courts, which burdens litigants with the need to appeal. 1' 0 This is the precise constraint that all Member States ensured litigants would face for Title IV provisions on visas, asylum, immigration, and the free movement of persons. Limiting reference rights to courts of last instance is particularly likely to stifle claims in this field since these actors are among the least likely to be able to sustain a case through multiple appeals. The scarcity of references from most national courts of last instance, including their complete absence from constitutional courts, reflects yet another problem with access. Low rates of referral from most courts of last instance, despite their obligation to refer all relevant cases to the ECJ, suggests that many courts of last instance are not the most receptive forum to pursue European rights. A national court's refusal to refer at this stage eliminates individuals' final (and in Title IV areas first) opportunity to access European justice. In addition to institutional empowerment, preferences about the substantive content of legal interpretation also appear to influence national judicial behavior. Karen Alter has demonstrated that lower courts initially referred questions to the ECJ when they expected favorable interpretations and sought to avoid the possibility of reversal on appeal within the national system. 131 Jonathan Golub has shown that British judges withheld references to the ECJ and interpreted EC law independently when they wanted to shield domestic environmental policies from unfavorable ECJ case law. 132 In a study of all reported U.K. court rulings on questions of EC law, Chalmers found that efforts to invoke EC law before British courts tend to be successful when EC law extends state powers, providing further means to control and criminalize behavior. Conversely, he found that efforts to invoke EC law were much less successful when EC law might prevent the state from promoting "collective ties" or applying criminal sanctions on the grounds that they breach EC law. The willingness to make references followed a similar pattern, where U.K. judges were more receptive to requests 130. "Debates of the European Parliament," Official journal of the European Communities, Annex no. 3-434 (14 September 1993): 50. 131. Alter 1 996, 466--68. 132. Jonathan Golub, ''The Politics ofJudicial Discretion: Rethinking the Interaction Between National Courts and the European Court of Justice," West European Politics 19 (April 1 996): 376, 379-81.

Containing Justice

87

for references in civil proceedings and were relatively resistant to referring cases at the behest of defendants in criminal proceedings. 133 U.K. courts were also more receptive to applying EC law when it would reinforce private legal relationships and more resistant when EC law might disrupt stable private relationships, thwart private initiatives, or impede the exercise of property rights. In sectoral terms, these judicial preferences have meant that it has been easier to enforce EC rights related to sex discrimination, which consolidate employment contracts, and harder to enforce EC rights related to the free movement of persons, which challenge collective national ties. It has also been more difficult to enforce European competition and environmental law, which often interfere with private contractsY4 Furthermore, the hesitance of U.K. courts to invoke EC law if it would favor a criminal defendant complicates recourse for economic actors who assume they can rely on more liberal EC provisions when they act "illegally" under national law. These findings cast doubt on any assumption that national courts automatically serve as conduits in the enforcement of European rights. Indeed, characteristics of the parties claiming EC rights appear particularly important in determining the propensity of national courts to cite prior ECJ case law or refer questions for preliminary rulings. Both explicit citations of ECJ case law and references are concentrated in those national courts that overwhelmingly serve commercial interests. Table 3.6 demonstrates that the incidence of citing ECJ case law is highest in the civil and commercial jurisdictions of French, German, and U.K. courts. Referrals for preliminary rulings are also most prevalent in this set of national courtsY" These national courts resolve disputes related to market interactions and economic regulations, with commercial enterprises the most frequent parties to the proceedings. Many of the cases before administrative courts and the House of Lords, a court of last instance with a general jurisdiction, can also involve enterprises' challenges to national economic regulation. This pattern of citations and references is only partially an artifact of the economic orientation of European law. Alec Stone Sweet and Thomas Brunell have posited a simple functional relationship between economic 133. Damian Chalmers, 'The Much Ado About judicial Politics in the United Kingdom: A Statistical Analysis of Reported Decisions of United Kingdom Courts Invoking EU Law 1973-1998," Harvard jean Monnet Working Paper 1/oo (zooo): 32. 134. Ibid., 27-28. 135. Lisa Conant, "Europeanization and the Courts: Variable Patterns of Adaptation among National Judiciaries," in Transforming Europe: Europeanization and Domestic Change, ed. Maria Green Cowles, James Caporaso, and Thomas Risse (Ithaca, N.Y.: Cornell University Press, 2001), table 6.5, 109.

88

Justice Contained

Table 3.6. Distribution of ECJ Citations in French, German, and U.K. Courts

French Courts

German Courts

Court of Appeal Civil Division

55

74

High Court of Justice, Queens Bench Division

41

79

High Court at Justice, Chancery Division

10

Administrative courts

44

House of Lords

40

104

Social courts

41

41

466

Labor courts

20

Employment appeal tribunal Others

Gaur de Cassation Commercial courts

746

Tax courts

Gaur de Cassation Civil courts

106

Cour de Cassation Tribunal de Grande Instance

158

Bundesgerechtshol (civil jurisdiclions) Oberlandesgerichte (mostly civil)

Conseil d'Etat Administrative court system

227

Cour de Cassation Social courts Cour de Cassation Criminal courts Others

U.K Courts 913

9

38

Sources Jurilrance searches lor French case law on 17 and 20 November 2000; Juris searches lor German case law on 17 December 2000; Lexis search for U.K. case law on 23 November 2000. Not!!. As noted in Table 3.4, comparisons should be made exclusively among court systems within the same country due to variations in the coverage of databases.

rights and societal demand for them, where they find a high correlation between referral rates and levels of intra-EU trade. 136 As my previous research demonstrates, however, this "transactionalist" account fails to explain the majority of references that bear no clear relationship to trade. 137 For example, disputes over the application of European provisions on migrant workers and the equal treatment of men and women address economic rights that have little to do with trade. Social and labor courts would handle most of these cases, although administrative courts would deal with some questions related to migrants. These courts' references and citations to ECJ case law are much less extensive than their civil and commercial counterparts who are more likely to resolve disputes affecting commercial enterprises. This discrepancy relates to the variable capacity of affected parties to engage in litigation effectively, rather than simply variations in interest in European rights. Commercial enterprises typically possess more resources to pursue litigation than most actual "individuals" with intense economic 136. Alec Stone Sweet and Thoma~ Brunell, "Constructing a Supranational Constitution: Dispute Resolution and Governance in the European Community," American Political Science Reviewg2 (March 1gg8b): 63-81; Stone Sweet and Brunell1gg8a. 137. Conant 2001, 114-15.

Containing Justice

89

interests in European rights, such as the employees and retirees seeking the application of European provisions on national or gender discrimination. Enterprises are more likely to be able to "purchase" knowledge ofECJ case law, which enables them to request its application to their disputes and contributes to more prevalent citation ofECJ "precedent." My past cross-sectoral research on referral rates indicates that the participation of ( 1) actors with the means to finance litigation, such as commercial enterprises, (2) societal interest organizations, or (3) public agencies engaged in enforcement and advocacy contribute to higher rates of references for preliminary rulings. 138 These parties are all likely to possess the resources necessary to understand their prospects under EC law and to appeal and forum-shop within the national judicial system to improve their chances of facing a receptive national judge. Likewise, in a case study on the concentration of gender discrimination references in U.K. courts, Karen Alter and Jeannette Vargas found that fonun shopping and appeals were critical to achieving references and national case law favorable to EC rights. Institutional support in the form of the British Equal Opportunities Commission (EOC) was particularly important in sponsoring litigation in this field. 139 Research on other areas by Damian Chalmers, Carol Harlow, and Richard Rawlings has also found that references overwhelmingly originate in cases where businesses are engaged in regulatory bargaining with the state or where other organizational and institutional actors explicitly seek to challenge domestic provisions through European judicial review. 140 These three categories of actors are also more likely to be willing to wait during the long delay between a reference and the ECJ's preliminary ruling. Indeed, Chalmers's research on U.K. courts found that only businesses and interest groups intent on challenging the content of domestic law were interested in waiting for preliminary rulings. 141 The delay is bad enough to have inspired a challenge at the European Court of Human Rights (ECHR), on the grounds that the delay lengthened domestic civil proceedings so long as to violate the concept of a "fair and public hearing." Although the ECHR agreed that prolonging the proceedings by over two years and seven months seemed quite long, it rejected the challenge on the 138. Ibid., 98-99, 108-14. 139. Karen Alter and jeannette Vargas, "Explaining Variation in the Use of European Litigation Strategies: European Community Law and British Gender Equality Policy," Comparative Political Studies 33 (May 2000): 452-82. 140. Chalmers 2000, 34-6; Harlow 1992, 344-48; Harlow 1993, 333; Richard Rawlings, "The Eurolaw Game: Some Deductions from a Saga," journal of Law and Society 20 (autumn 1993): 309-40. 141. Chalmers 2000, 34-35.

90

Justice Contained

basis that it would work against the aim of the reference procedure. 142 National judges are sensitive to the delay caused by references to the ECJ, and will avoid references in cases where prompt recourse is important even when they acknowledge that the interpretation of a disputed measure is far from obvious. 143 As a result, the practical enforcement of European rights for many parties will be based on the interpretation of national courts, rather than the ECJ. Whether parties seek strictly personal recourse or a general shift in domestic law, national judicial enforcement does not necessarily lead to changes in the domestic status quo for anyone beyond the parties to the litigation. Chalmers's research uncovers a clear case ofwhat I conceptualize as contained compliance: U.K. courts typically apply EC provisions without directly addressing the status of incompatible national provisions. Indeed, up through 1999, only two decisions before U.K. courts actually engaged the question of whether domestic legislation might be suspended. 144 Otherwise, contained compliance prevailed, where national courts applied EC provisions in a case-by-case manner and ignored the broader implications for national law. It is hardly surprising that Chalmers concludes that U.K. administrations face little judicial pressure to accommodate obligations under European law. They are not alone, as the failure of other national courts to challenge evasive administrative practices suggests. For instance, the French Conseil d'Etat holds that it cannot review nonlegislative administrative decisions for compatibility with EC law. 145 Essentially the French administration can act as it pleases, as long as it does not codifY its choice in subordinate legislation. Hence, legal challenges in administrative courts under the Conseil d'Etat will not exert pressures to align much administrative action with European legal obligations. Such variable patterns of judicial participation in the European legal system can pose serious obstacles to individual attempts to invoke EC rights. Outright national judicial resistance to ECJ case law can also reduce the potential to challenge evasion of European obligations. The European Commission regularly uncovers explicit instances of resistance by national high courts. In annual reports on the application of EC law, the European Com142. Commission 2000, COM (2ooo) 92 final, 35· 143. Commission 1999, COM (99) 301 final, 2.')3; Commission 1997, C-332, 203; H.P. Bulmer Ltd. v. ]. Bolinger SA, Court of Appeal, 1974, Common Market Law Reporter 2 ( 1974): 91; Chalmers 2000, 22, 32-33. 144· Chalmers 2000, 10. 145. Commission, "Eleventh Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law-1993," Official journal of the E.uropean Communities(', 154 (6 June 1994): 176; Commission, ''Twelfth Annual Report on Monitoring the Application of Community Law-1994." Official journal of the European Communities Cr2 54 ( 29 September 1995): 166.

Containing Justice

91

mission documents ninety-four incidents of inappropriate action taken by high courts between 1985 and 1999. Courts from Member States with high rates of references for preliminary rulings were among the worst culprits: the European Commission targeted seventeen instances of inappropriate judicial action in both France and Germany, eighteen in the Netherlands, sixteen in Spain, seven in Italy, six in Belgium, three each in Austria and Sweden, one in Finland, and two each in Greece, Portugal, and the United Kingdom. The high courts cited either failed to refer when obligated to do so, interpreted EC law in direct contradiction to existing EQJ case law, or issued rulings that impeded the application of an EC measure. The rate of these incidents is underreported because the European Commission only reviews a subset of high court actions. 146 National judicial intransigence escapes prosecution by the European Commission, which seeks to promote an entirely voluntary cooperation between national courts and the ECJ. As a result, national courts face little to no pressure from EU institutions to accommodate the requirements of the European legal system. The European Commission takes no active measures to enforce the obligation of national courts to refer cases/ 17 to encourage the uniform application of ECJ case law as a body of precedents, or to oblige national courts to respect the preliminary rulings they request. The European Commission refrains from using infringement proceedings in cases of national judicial resistance because it considers that such a procedure potentially exposes independent judiciaries to political pressures 146. Commission, "Third Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law-198 5," Official journal of the European Communities G-220 ( 1 September 1986): 27-28; Commission, "Fourth Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law-1986," Official journal of the European Communities C-338 ( 16 December 1987): 33-34; Commission, "Fifth Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law-1987," Officialfoumal of the European Communities G-31 o (5 December 1988): 43-45; Commission, "Sixth Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law-1988," Official Journal of the European Communities C-330 (30 December 1989): 53-54; Commission, "Seventh Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law1989," Official journal of the European Communities G-232 ( 17 September 1990): 232; Commission, "Eighth Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law-1990," Official journal of the European Communities C-338 (31 December 1991): 78; Commission, "Ninth Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law-1991," Official journal of the European Communities C-250 (28 September 1992): 71; Commission 1993, G-233, 213; Commission 1994, G-154, 174-75: Commission 1995, G-254, 164-65; Commission 1996, COM (96) 6oo final, 417-21. 147. The Commission made an exception in 1989, when it initiated infringement proceedings against France due to the incorrect application of EC law by the Cour de Cassation. Commission 1989, G-330, 53-54·

92

Justice Contained

and endangers future prospects for judicial cooperation. 148 The European Ombudsman has taken the same approach to complaints concerning national courts. 149 Voluntary judicial cooperation has been forthcoming from national courts around the EU. However, cooperation has always coexisted with resistance in a contested process of interaction that continues to the present. The enduring independence of national courts in their relationship to the ECJ is not purely a function of the supranational status of the European legal system. Competition for influence and efforts to preserve entrenched legal norms characterize interactions among domestic courts as well. 150 What is important about the contested relationship between the ECJ and national courts is not that disputes exist, but that various points of contention mediate the application of EC law within the Member States. Accounts of the reception of major ECJ innovations, such as the supremacy and direct effect of EC law, uncover how national courts respond to ECJ decisions on their own terms. For example, the French Conseil d'Etat resisted recognizing the supremacy of EC law until 1g8g. This represents one of the longest-standing incidents of national judicial intransigence and contrasts sharply with the 1975 acceptance of European legal supremacy by the French Cour de Cassation."' Resistance from the Conseil d'Etat long impeded the application of EC provisions in France because a large proportion of European law targets activities of the state. However, the supremacy debate is far from over in the EU because national courts define the parameters of their cooperation even as they make concessions to the ECJ. Karen Alter argues that in "basing the supremacy of EC law on national constitutions and not tying themselves to the ECJ's legal reasoning regarding EC law supremacy, high courts left open legal avenues through which they could refuse the authority of the ECJ in the future with148. Commission 1986, C-220, 27; Commission 1987, C-338, 33; Commission 1988, C310, 43; Commission 1989, C-330, 53-54. 149· Rawlings 2000, 10. 150. Fisher Ig88, 7-8, 222-24;Johnson and Canon 1984, 32, 43-45· 51, s6, 71; Martin Shapiro, Law and Politics in the Supreme Court: New Approaches to Political jurisprudence (New York: Free Press, 1964), 25; Shapiro 1981, 40; Alter 1996;jens Plotner, "Report on France," in The European Court and the National Courts-Doctrine and jurisprudence: Legal Change in Its Social Context, ed. Anne-Marie Slaughter, Alec Stone Sweet, and Joseph Weiler (Oxford: Hart, 1998); Rawlings 1993. 151. Alec Stone Sweet, "Constitutional Dialogues in the European Community," in The European Court and the National Courts-Doctrine and jurisprudence: Legal Change in its Social Context, ed. Anne-Marie Slaughter, Alec Stone Sweet, and joseph Weiler (Oxford: Hart, 1998); Alter 200 l.

Containing Justice

93

out contradicting their jurisprudence on EC law supremacy. "152 And anumber of constitutional courts have indeed problematized supremacy once again. In its controversial Maastricht decision of 1994, the German Bundesverfassungsgericht (Federal Constitutional Court) made the primacy of EC law contingent on national constitutional stipulations. 153 The Bundesfinanzhof and Frankfurt Verwaltungsgericht followed this lead, explicitly detying ECJ case law by questioning the validity of EC acts on the basis of incompatibility with the German Basic Law (constitution) .154 Italian and Greek courts have also indicated that they will review the constitutional acceptability of EC obligations. 155 Conversely, Austrian and Belgian courts have actually accepted the idea that European law is supreme even over their national constitutions. 156 The Conseil d'Etat's long-standing refusal to accept the ECJ's supremacy doctrine and the more recent assertions of some constitutional courts of the "supremacy" of national constitutions constitute direct clashes with the ECJ's clear statement on the precedence of European law. These cases illustrate that even unambiguous ECJ interpretation does not automatically elicit uniform, deferential responses. Other notable cases of resistance include the rejection of direct effect for directives by the French Conseil d'Etat and German Bundesfinanzhoffrom the 1970s to the mid-198os, 157 and U.K courts' restrictive stance toward indirect effect. While the German and French courts rebelled against the idea that framework legislation requiring transposal could have any direct effects, U.K courts have narrowly construed their obligation to interpret all national laws to be in conformity with such framework legislation. De Burca finds that U.K courts have not applied the presumption that Parliament intends to comply with EC law to measures that were not specifically designed to implement EC law. 158 This response to the interpretive obligation, or indirect effect, inhibits the ability to invoke directives in disputes between private parties. Variable behavior among national courts suggests that opportunities to invoke EC law as a means to challenge national policies vary within and across Member States. The variable reception of ECJ case law in national 152. Alter 1996,470. 153. Steve Boom, "The European Union Mter the Maastricht Decision: Will Germany Be the 'Virginia' of Europe?" American journal of Comparative Law 43 (spring 1995): 177-226. 154· Commission 1997, C-332, 201; Commission 1998, C-250, 198. 155. Commission 1999, f:OM (99) 301 final, 266-68; Commission 2000, COM (2000) 92 final, Annex VI, 17. 156. Commission 2000, COM (2ooo) 92 final, Annex VI, 24; Commission 1997, C-332, 202. 157. Hartley 1994, 243-47. 158. De Burca 1992, 221.

94

Justice Contained

courts complicates individual legal recourse and mediates the ECJ's impact over policy processes. National courts that respond favorably to European developments will encourage litigation that demands the enforcement of new legal principles. Conversely, national courts that resist ECJ innovations will contain their application and discourage suits by those who seek to challenge the domestic order. In conclusion, this chapter demonstrates that the ECJ constrains the same novel forms oflegal recourse that it creates. Judicial developments in EC law, characterized by incrementalism and particularized decisions, generate considerable uncertainty. This uncertainty, in addition to rules that limit the binding scope of individual ECJ decisions, narrowly confine rights and obligations defined in case law. Governments and their administrations capitalize on these constraints, responding to legal interpretation by applying it in single shots as discrete disputes come before national courts. This response produces a fragmentation in European judicial authority. National judges' variable willingness to engage the European legal system provides another source of fragmentation. Taken together, these features contain the policy implications of ECJ decisions. Long focused on the centralizing force of ECJ case law, research on judicial authority in the EU must take account of these limitations. Declarations that courts act as policymakers exaggerate the force that the ECJ exerts in politics. By exploring how legal institutions generate fragmented patterns of authority, this chapter identifies gaps between ECJ case law and policy processes in the EU and its Member States. Case studies in the following chapters will demonstrate that the institutional constraints that characterize the European legal system contained justice until institutional and organizational actors mobilized pressure in response to innovative ECJ legal interpretation.

4 From Jurisprudence to Policy: The Liberalization of European Telecommunications Public enterprises historically monopolized the telecommunications sector throughout the European Union. Similar to many other networkbased services, exclusive provision traditionally seemed justifiable as both efficient and necessary to achieve universal service. Yet the legal status of such exclusive provision was inherently ambiguous in light of competing provisions in the Treaty of Rome and its successors. Starting in 1974, ECJ interpretation of Article 86 (go EEC) challenged national restrictions on competition in areas dominated by public undertakings and other undertakings granted exclusive rights to provide particular services. In 1982, ECJ interpretation of Article 86 (go EEC) also confirmed the European Commission's competence to issue directives in order to ensure the application of competition rules to public undertakings and undertakings granted exclusive rights. The European Commission capitalized on this legal interpretation to launch a comprehensive liberalization of European telecommunications between 1988 and 1gg8. The process of reform was striking: the European Commission legislated unilaterally to open telecommunications markets under Article 86 (go EEC), 1 abandoning the traditional approach of proposing directives for approval by national representatives in the Council of Ministers under Article 95 (10oa EC). 2 ECJ decisions upheld the European Commission's ap1. Amsterdam Treaty ( 2 October 1997); Treaty Establishing the European Economic Community (25 March 1957). 2. Single European Act, Article 18 amends Article 100 EEC Qanuary 1987).

95

96

Justice Contained

proach as Member States challenged its competence to issue competition directives independently. And, even as Member States contested exclusive supranational control over the opening of particular telecommunications markets to competition, their representatives in the Council of Ministers responded cooperatively by harmonizing a progressively liberal regulatory framework for the operation of a telecommunications market at the EU level. This pattern of events is surprising because both the European Commission and Member States originally left traditional public-service sectors such as telecommunications out of the 1992 internal market program, intending to preserve national provision of services. European proposals in telecommunications initially focused on the harmonization of technical standards and the development of a common infrastructure to facilitate the use of new information technologies. When the European Commission then decided to advocate the introduction of competition as a means to improve the performance of the European telecommunications industry, it could utilize what has been dubbed a "crowbar approach to judicial review" to expand the scope of the internal market program in services.' Telecommunications liberalization under Article 86 (go EEC) ultimately involved Member States' acceptance of both a significant increase in the legislative competence of the European Commission and a major policy reform. In this chapter I show that the evolution of interests and mobilization of pressure associated with telecommunications liberalization contributed to this dramatic instance of reform. Furthermore, I will demonstrate that Member States retained a substantial degree of control over the process of regulatory policy-making. The competing interests associated with liberalization foreclosed the capacity of national governments to either actively evade or passively accept legal obligations. Failing to block the introduction of competition by supranational institutions, Member States have been able to exercise influence over the pace and content of policy reform. The chapter proceeds in three sections. First, I discuss the legal basis that justifies both the liberalization of telecommunications markets and the European Commission's authority to legislate liberalization unilaterally. Second, I trace the development of European competition policy in this field. Third, I relate this process of reform to the mobilization of political pressure by specifying the pattern of interests affected by liberalization. 3· Jacques Pelkmans, 'The Significance ofEC-1992," Annals of the American Academy ofPolitical and Social Science 531 (January 1994): 101.

From Jurisprudence to Policy

97

Legal Innovations: Article 86 (90 EEC) Brings Competition Law to Public Services The legal obligation to apply European competition law to public-service sectors such as telecommunications is actually embedded within the 1g57 Treaty of Rome. Yet competing treaty provisions condition this legal obligation. In its first paragraph, Article 86 (go EEC) prohibits national measures that are contrary to treaty rules, particularly the competition rules, with respect to public undertakings and undertakings granted exclusive rights. But the second paragraph of Article 86 (go EEC) allows for exemptions from European competition rules in cases where their application would inhibit an undertaking's ability to provide "services of general economic interest," which typically include universal service obligations that are uneconomic due to the costs of providing service to remote locations. The second paragraph of Article 86 (go EEC) requires these exceptions to be balanced against their impact on EU trade. Finally, the third paragraph of Article 86 (go EEC) grants the European Commission competence to address appropriate directives and decisions to Member States. Meanwhile, Article 2g5 (222 EEC) leaves Member States free to determine regimes of property ownership, allowing public and private enterprises to perform various economic activities. And no treaty provision prohibits monopolies per se. However, Article 15 7 ( 130 EEC) obligates the EC and Member States to ensure that the conditions necessary for the competitiveness of European industry exist. Given the historic prevalence of exclusive provision for many services (postal delivery and telecommunications, air and rail transportation, energy and water distribution), the only operational treaty provisions were the exemption for services of general economic interest and the freedom of states to sanction public and private monopolies. The potential to apply competition rules to traditional public-service sectors attracted little attention prior to the entry into force of the Single European Act (SEA) in 1987. Yet since the early 1g7os, both public and private actors periodically invoked the competition provisions of Article 86 (go EEC). In 1971 the European Parliament requested that the European Commission prepare directives and decisions to eliminate distortions in competition between public and private enterprises.• This elicited no immediate response from the European Commission, but in 1g7 4, the ECJ ruled that exemptions from competition rules under Article 86, 2 (go, 2 EEC) were permissible only if their application would be incompatible with the performance of tasks associated with the general economic interest. 4· Official journal of the European Communities G66 ( 1 July 1971): 11-13.

98

Justice Contained

Member States maintaining exemptions would need to prove to national courts that the application of competition rules would impede the accomplishment of these tasks. 5 This decision asserted the exceptional nature of exemptions to competition rules, emphasizing the obligation to allow competition. This case attracted virtually no attention, but in 1g8o the European Commission responded to the European Parliament's inquiries by issuing a directive to increase transparency in the financial relations between Member States and their public enterprises. This "transparency directive" required Member States to reveal financial information necessary for the European Commission to investigate potential state aids abuses. 6 France, Italy, and the United Kingdom contested the European Commission's competence to issue this directive unilaterally, claiming that the Council of Ministers needed to adopt any directives whose impact extended beyond specific situations. But in 1g82 the ECJ upheld the European Commission's right to adopt even such a broadly targeted directive, arguing that this directive fell within the scope of the European Commission's duties to enforce the competition provisions of Article 86 (go EEC). 7 Vindicated by its victory before the ECJ, the European Commission extended the application of this directive in 1g8 5 8 and began to adopt individual decisions under Article 86 (go EEC) as well. 9 Decisions of the European Commission are legally binding measures that target particular abuses of EC law by individual firms or public agencies. As such, decisions do not possess the general regulatory character of directives, which address entire classes of behaviors and actors. Yet, even these narrowly targeted decisions by the European Commission inspired Member State opposition. The Netherlands led a legal challenge against one of these decisions, but the ECJ again supported the rights of the European Commission to take decisions related to European competition law and public enterprises under Article 86 (go EEC). 10 5· GiuseppeSacchiC-155/73• (1974) European Court Reports (E.C.R.), 409. 6. Commission Directive 80/723/EEC, Officialfournal of the European Communities L-195 (zgJuly 1g8o): 35· 7· France, Italy, and the United Kingdom v the Commission C-188/8o, (1982) E.C.R., 2545· 8. Commission Directive 8 5/41 3/EEC, Official journal of the European Communities L-2 29 (28 August 1985): 20. g. Commission Decision 85/276/EEC of 24 April 1985, Official journal of the European Communities L-152 (11 June 1985): 25; Commission Decision 87/359 EEC of 22June 1987, Official Journal of the HurofJean Communities L-194 ( 15 July 1987): 28; Commission Decision go /I 6 EEC of 20 December 1989, Official Journal of the European Communities L-10 ( 12 January 1990): 47; Commission Decision go/ 456/ EEC of 1 August 1990, Official journal of the European Communities L-233 (26 August 1990): 19; Commission Decision 94/119/EEC of 21 December 1993, Official journal of the European Communities L-55 (26 February 1994): 52. 10. Commission Decision go/16 EEC; Kingdom of the Netherlands and Others v Commission of' the European CommunitiesC-48, 66/go, (1992) E.C.R., I-s6s.

From Jurisprudence to Policy

99

Meanwhile, ECJ decisions in 1g85 identified abuses of dominant positions by public monopolies, providing further legal justification to extend European competition rules to public enterprises and enterprises granted exclusive rights under nationallaw. 11 The British Telecom case specifically addressed the application of competition rules to a particular telecommunications service. The ECJ rejected the Article 86,2 (go,2 EEC) exemption in this instance, finding that Italy failed to establish that the application of competition rules would create problems for British Telecom in performing any tasks of general interest assigned to it. 12 This case law, as well as complaints about exclusive rights and the momentum to create an internal market through the SEA, contributed to the European Commission's decision to prepare directives to open particular telecommunications markets to competition as of 1g88. This use of Article 86 (go EEC) allowed the European Commission to exercise broad regulatory powers, demanding the end of exclusive rights conferred at the nationalleveJ.1 3 Member States opposed the unilateral exercise of regulatory authority by the European Commission and challenged the first two directives issued under Article 86 (go EEC). The ECJ again upheld the competence of the European Commission to adopt directives in order to enforce competition rules/ 4 and the ECJ continued to limit the capacity of public enterprises to restrict trade and competition in preliminary rulings from 1g8g to 1gg3. 15 11. &British Telecommunications: Italy v CommissionC-41/83, (1985) E.C.R., 873; CentreBelge dEtudes de Marchi- Tete-marketing SA v Compagnie Luxembourgeoise de Tetediffusion SA (CBEM-Telemarketing v CLT) C-311/84, (1985) E.C.R., 3261; George Bermann et al., Cases and Materials on European Community Law (St. Paul: West, 1993), 881-82; Paul Craig and Grainne de Burca, EU Law: Texts, Cases, and Materials, 2d ed. (Oxford: Oxford University Press, 1998), 1971. 12. British TelecomC-41/83, 198!). 13. Commission Directive 88 I 30 1/EEC of 16 May 1988, Official journal of the European CommunitiesL-131 (27 May 1988): 73; Commission Directive go/388/EEC of28June 1ggo, Official Journal of the European Communities L-192 (24 July 1ggo): 10; Commission Directive 94/ 46/EC of 13 October 1994, Official Journal of the European Communities L-268 ( 19 October 1994): 15; Commission Directive gs/ 51 /EC of 18 October 1995, Officialjournal of the European Communities L-256 (26 October 1995): 49; Commission Directive g6/2/EC of 16 January 1gg6, Official.Journal of the European Communities L-20 ( 16 January 1996): 59; Commission Directive g6/1g/EC of 13 March 1gg6, Official journal of the European Communities L-74 (22 March 1gg6): 13. 14. French Republic v Commission of the European Communities C-202/88, ( 1991) E.C.R., 11223 and Kingdom of Spain and Others v Commission of the European Communities C-271, 281, 28g/go, (1992) E.C.R., l-5833. 15. Ahmed Saeed Flugreisen & Silver Line Reisebiiro GmbH v Zentrale zur Bekiimpfung Unlauteren Wettbewerbs eVC-66/86, ( 1g8g) E.C.R., 8o3; Regie des ttitegrajJhes et des tetephones v GB-lnno-BM SA C-18/88, (1991) E.C.R., I-5941; Elliniki Radiophonia Tileorassi AE v Dimotiki Etairia PlirojiJrissis and Sotirios Kouvelas C-260/Sg, (1991) E.C.R., I-2925; Merci convenzionali porto di Genova SpA v Siderurgica Gabrielli SpA C-179/ go, ( 1991) E.C.R., l-588g; Klaus Hafner & Fritz Elser v Macrotron GmbH C-41 /go, (1991) E.C.R., l-1979; Christian Poucet v Assurances generales de France (AGF) et Caisse mutuelle regionale du Languedoc-Roussillon (Camulrw:) and Daniel Pistre v Caisse autonome na-

100

Justice Contained

Table 4.1. Chronology of the Enforcement of Article 86 (90 EEC) Meaning of Article 86 (90 EEC)

Member States must respect EC competition rules and refrain from engaging in national discrimination in the case of public undertakings and undertakings granted exclusive rights. Undertakings providing for services of general economic interest are subject to treaty rules, including competition rules, insofar as these rules do not obstruct the provision of these services. The development of trade must not be affected to such an extent as would be contrary to the interests of the EC. The European Commission shall ensure the application of Article 86 (90 EEC) and address appropriate directives or decisions to Member States.

Enforcement of Article 86 (90 EEC)

Treaty of Rome, 1957

None

European Parliament requests directives and decisions, 1971.

None

ECJ Case law, 1974-1993

Case by case

European Commission issues Article 86 (90 EEC) transparency directive, 1980.

All Member States must reveal financial information about public enterprises.

ECJ upholds Article 86 (90 EEC) directive alter Member States challenge its legality, 1982. European Commission adopts individual decisions under Article 86 (90 EEC), 1985-1994.

Case by case

ECJ upholds Article 86 (90 EEC) decision challenged by the Netherlands, 1992. Realization of Article 86 (90 EEC) in Telecoms Policy

European Commission issues Article 86 (90 EEC) directives to liberalize telecommunications, 19881999.

Telecommunications markets open progressively across the EU.

ECJ upholds two Article 86 (90 EEC) directives alter Member States challenge their legality, 1991-1992.

Table 4.1 represents the evolution in the use of Article 86 (go EEC) over time. The European Commission's adoption of Article 86 (go EEC) directives and the ECJ's legitimation of this independent regulatory power constituted an important new direction in European competition policy. The European Commission traditionally focused on private infringements of comtionale de compensation de /'assurance vieillesse des artisans (Cancava) C-159-160/91, (1993) E.C.R., l-637·

From Jurisprudence to Policy

101

petition rules and concentrated its enforcement actions against private enterprises.16 Changing its priorities, the European Commission argued in 1990 that the greatest market barriers were those created by state interventionY In its 1994 Competition Report, the European Commission associated systems of exclusive rights under national regulation with higher prices, insufficient technological innovation, and slow employment growth. 18 According to David Gerber, this reorientation represents a dramatic substantive shift away from concerns with private conduct toward the problem of government interference in the competitive process. 19 This extension of competition law also constitutes a major shift in relations between Member States and supranational institutions. The European competition regime always gave the European Commission exceptional autonomy: Articles 81-89 (85-g4 EEC) and Regulation 17 of 1g62 have enabled the European Commission to investigate, codifY, exempt, and fine. The only restraints include the need to achieve European Commission endorsement and to defend challenges before the ECJ. "' Member States sanctioned this supranational autonomy because they realized each national government would be prone to coddle domestic industries. But Member States intended the European Commission to wield its power against private parties. The ECJ's unexpected enforcement of the linkage between competition law and public enterprises under Article 86 (go EEC) suddenly subjected national governments to autonomous supranational control. The application of competition law under Article 86 (go EEC) can oblige Member States to eliminate all exclusive rights that directly or indirectly inhibit trade. Once the European Commission prohibits exclusive rights in a field, all interested parties can enter this market confident that any national efforts to enforce restrictive government regulations will be considered a treaty violation. Therefore Article 86 (go EEC), subject exclusively to the authority of the European Commission and ECJ, can immediately open markets for competition. Unanticipated and unwelcome for Member States, liberalization under Article 86 (go EEC) nonetheless rapidly ceased to be contested as a lawful

16. Stephen Wilks and Lee McGowan, "Disarming the Commission: The Debate over a European Cartel Office," Journal of Common Market Studies 32 ( 1995): 261; David Gerber, 'The Transformation of European Community Competition Law," Harvard International Law Journal 35 (1994): 137-38. 17. Commission, Twentieth Report on Competition Policy (Luxembourg: Office for Official Publications of the European Communities, 1990), 50. 18. Commission, Twenty-third Report on Competition Policy (Luxembourg: Office for Official Publications of the European Communities, 1994a), 31. 19. Gerber 1994, 137. 20. Wilks and McGowan 1995, 261, 264.

102

Justice Contained

procedure. Mter two bouts before the ECJ over the first two telecommunications directives, Member States acknowledged the legal status of supranational regulatory authority under Article 86 (go EEC). The European Commission utilized Article 86 (go EEC) to issue a series of further telecommunications directives and has consistently insisted on its legal capacity to use this tool across sectors traditionally dominated by publicservice provision. 21 However, legal recognition did not coincide with political acceptance. In response to the second telecommunications directive issued under Article 86 (go EEC), a member of the European Parliament (MEP) exclaimed to the European Commission, "Vous avez juridiquement raison, mais politiquement tort. "22 Member States and MEPs alike were uniformly hostile to the Article 86 (go EEC) procedure, which can exclude their input entirely. Yet at the same time, interested actors were not all adverse to the prospect of telecommunications liberalization itself. As a result, the use of Article 86 (go EEC) could serve as a means to promote preferred reforms, despite serious reservations about the method. The following discussions demonstrate that the offensive legal procedure was politically tolerable in telecommunications because the substantive content of liberalization measures became acceptable to a wide range of interests.

Toward a Competitive European Regulatory Framework for Telecommunications The European Commission successfully invoked its legal authority to liberalize telecommunications markets across the European Union. Repeatedly exercising independent regulatory power through its own directives under Article 86 (go EEC), the European Commission opened all facets of the telecommunications industry to competition. As the ECJ upheld the actions of the European Commission, the Council of Ministers adopted the legislation necessary to construct a progressively liberal regulatory framework for telecommunications. However, the legal obligations articulated by the European Commission and Eq developed in close tandem with the political will of national governments. The substance of each directive under Article 86 (go EEC), formally an expression of supranational authority, broadly satisfied the demands of Member States. Furthermore, the specification of a harmonized European regulatory framework in Council directives accommodated the particular interests of national governments. Member Commission 1994a, 3o-31. "'Legally, you are right, but politically, you are wrong": Debates of the European Parliament," Official journal of the Eumpean Cammunities, Annex no. 3-381 ( l l October tg8g): ll5· 21. 22.

From Jurisprudence to Policy

103

States ultimately controlled much of the pace and content of the new telecommunications regime, gradually phasing in its application to an expanding range of activities. The following chronology of the policy process demonstrates that the evolution of a broad political consensus was critical to the realization of reforms propelled by supranational legal instruments. Interest in a common European regulatory framework for telecommunications dates from the early 1g8os, but a consensus to liberalize emerged gradually from 1984 to 1993. National representatives in the Council of Ministers adopted a recommendation to harmonize regulation in this field in 1g84. 23 The Council adopted the first binding measure to promote liberalization in a 1g86 directive, creating the initial framework for a mutual recognition of standards for terminal equipment (telephones, facsimile machines). 24 The European Commission promoted further liberalization with its 1g87 Green Paper, advocating the liberalization of markets interminal equipment and value-added services (high-speed data transmission).25 Then, rather than proposing a directive for Council approval as it typically does, the European Commission capitalized on its legal powers under Article 86 (go EEC) to open the market in terminal equipment on its own. In May 1g88 the European Commission issued a directive under Article 86 (go EEC) to abolish restrictions on competition in the sale of terminal equipment.'" The basic substance of this directive posed no problems for Member States, who endorsed the goals of the 1987 Green Paper in a June 1988 Council Resolution. 27 However, Member States did object to the European Commission's legislative use of Article 86 (go EEC). France led a legal challenge to this regulatory action under Article 86 (go EEC) before the ECJ, joined by Belgium, Germany, Greece, and Italy. 28 In disputing the legal basis of the directive, Member States sought to preclude any further unilateral exercise of power by the European Commission. Although the terminals directive was essentially innocuous, the prospect of future directives under Article 86 (go EEC) posed a significant threat to the Council's role as legislator. Ifthe European Commission could legally open the telecommunications field, it could impose liberalization on other monopolized public services as well. 23. Council Recommendation (84/549/EEC) of 12 November 1984, Official journal of the European Communities L-298 (16 November 1984): 49·

24. Council Directive (86/361/EEC) of 24]uly 1986, Official journal of the European Communities L-21 7 (5 August 1986): 21. 25. Commission, Towards a Dynamic European Economy: Green Paper on the Development of the Common Market for Telecommunications Services and Equipment, COM (87) 290 final, 1987. 26. Commission Directive 88/301/EEC. 27. Council Resolution 88/C257/01 of 30 June 1g88, Official journal of the European CommunitiesC-257 (4 October 1988): 1. 28. France v Commission C-202/88, 1991.

104

Justice Contained

Appealing to criticisms about the democratic deficit at the European level, the French denounced the exclusion of the Council and European Parliament from the legislative process. 29 Meanwhile, the United Kingdom avoided joining the legal challenge despite its disapproval of the legislative use of Article 86 (go EEC). British leaders strongly supported the substance of the Article 86 (go EEC) directive and did not want to slow down progress toward the liberalization of telecommunications. Having already liberalized many of the "public services" potentially vulnerable to liberalization under Article 86 (go EEC), U .K leaders faced fewer paten tial challenges to their regulatory regimes than most other Member States. Confident of its legal authority, the European Commission did not wait for the ECJ's ruling on this legal challenge before issuing a second directive under Article 86 (go EEC) to introduce competition in value-added services in 1ggo. 30 This directive also reflects a preexisting political consensus: the June 1g88 Council Resolution had already accepted liberalization in value-added services/1 and the Council adopted a directive to implement the open-network provision (ONP) for value-added services on the same day that the European Commission approved its directive. The open-network provision provides the necessary means to create an internal market for telecommunications services, allowing competitors access to infrastructure.'12 Yet despite their approval of the substance of the European Commission's directives, Member States continued to reject the European Commission's efforts to pass legislation without Council approval. As a result, Spain initiated a legal challenge before the ECJ, joined by Italy and Belgium and supported by French intervention.'" The ECJ upheld both of the European Commission's legislative actions under Article 86 (go EEC) in 1991 and 1992. The ECJ did declare a few provisions of each directive void but supported the basic right of the European Commission to issue competition directives related to public undertakings and enterprises with exclusive rights.' 4 Meanwhile, the Council had 29. Interviews at Commission Legal Service, Brussels, 1 June 1995; Foreign Ministry of France, Paris, 21 February 1996; Direction du Gaz, de l'Electricite, et du Charbon (DIGEC), Paris, 2 1 February 1996; Henri Revol, Senateur, "Marc he interieur de I' electricite et du gaz nature!: que lie politique pour Ia France," Rapport d 'Information, no. 491 ( 1993-1994); M. Jacques Oudin, Senateur, "L'Europe et les services publics," Rapport d'lnformation, no. 459 ( 1993-1994); M. Franck Borotra, Depute, "Faut-il defendre le service public?" Rapport d'Information, no. 2260 (6 October 1995). 30. Commission Directive go/388/EEC. 31. Council Resolution 88/257/01. 32. Council Directive (go/ 387 /EEC) of 28 June 1990, Official Journal of the European Communitieslr192 (24July 1990): 1. 33· SpainandothersvCommissionC-271, 281,289/90,1992. 34· France v Commission C-202/88, 1991; Spain and others v Commission C-271, 281, 28g/go, 1992-

From Jurisprudence to Policy

105

continued to expand the scope for liberalization, approving a directive to extend the open-network provision to leased lines, which firms use for internal communications and links to clients. 35 Legislation to construct a competitive telecommunications market progressed steadily, with a careful balance between European Commission directives to open up selected telecommunications fields and Council directives to harmonize the regulatory framework for liberalized areas. Two months after the Council adopted its ONP directive for leased lines injune 1992, the European Commission promoted the liberalization of basic voice telephone services with a proposal to apply ONP to voice telephony.36 By 1993, the Council advocated the further development of the telecommunications market and accepted the goal to liberalize traditional voice telephony by January 1, 1gg8. The Council also accepted goals related to the liberalization of cable networks and mobile and personal communications.37 In the next two years the European Commission continued to rally support for its telecommunications agenda, issuing the Bangemann report on "Europe and the Global Information Society" and Green Papers on mobile and personal communications, infrastructure, and cable television networks.38 Mter extensive consultation with interested parties, 39 the European Commission opened cable networks and mobile and personal communications with directives under Article 86 (go EEC) in 1995 and tgg6. 40 35· Council Directive 92/ 44/EEC of 5 June 1992, Official journal of the European Communities L-165 (19]une 1992): 27. 36. Commission, "First Proposal for a European Parliament and Council Directive on the Application of Open Network Provision to Voice Telephony," COM (92) 247 (27 August '992h). 37· Council Resolution 93/C213/01 of 22july 1993, Official journal of the European CommunitiesG-213 (6August 1993), 1. 38. Bangemann Report of the Commission, "Europe and the Global Information Society: Recommendation to the European Council," (26 May 1994); Commission, Green Paper on a Common Approach to Mobile and Personal Communications in the European Union, COM (94) 145 final ( 17 April 1994h); Commission, Green Paper on the Liberalization uf Telecommunications Infrastmt:ture and Cable TV Networks-Part I: Principle and Timetable, COM (94) 440 (25 October 1994c); Commission, Green Paper on the Liberalization of Telecommunications Infrastructure and Cable TV Networks-Part II: A Common Approach to the Provision ofInfrastructure in the European Union, COM (94) 682 (25]anuary 1995a). 39· Commission, Communication on the Consultation on the Green Paper on Mobile and Personal Communications in the European Union, COM (94) 492 final (23 November 1994d) and Commission, Communication on the Consultation on the Green Paper on the Liberalisation of Telecommunications Infrastructure and Cable Television Networks, COM (95) 158 final (3 May 1995b); European Parliament Resolution A{-{)063/95 of 7 April 1995, Offitial journal of the European Communities C-109 ( 1 May 1995): 310; European Parliament Resolution A4-{)111/95 of 19 May 1995, Official journal of the European Communities C-151 (9]une 1995): 479; European Parliament Resolution A4-{)097 I 95 of 19 May 1995, Official journal of the European Communities C151 (19]une 1995): 473· 40. Commission Directives 95/51 /EC and 96/ 2/EC.

106

Justice Contained

The European telecommunications regime progressed toward full liberalization as the European Commission solidified support to open infrastructure and voice telephony to competition. The Council accepted the principles and timetable for the liberalization of infrastructure in late 1994!1 agreed to the implementation of the future regulatory framework for telecommunications in 1995, 42 and responded positively to the European Commission's second proposal for a directive to apply ONP to voice telephony' by adopting a directive in 1995. 44 Mter consulting interested parties on its draft directive to liberalize voice telephony and the use of infrastructure for voice services, 15 the European Commission issued another directive under Article 86 (go EEC) for the full liberalization of telecommunications markets on March 13, 1996. 46 By mid-1997, the Council of Ministers and European Parliament together agreed to a directive to regulate interconnection to ensure universal service and interoperability through the application of the open-network provision. 47 Ongoing legislation by the Council of Ministers, European Parliament, and European Commission has proceeded to specifY and amend the regulatory framework in telecommunications. In 1997, the Council and European Parliament approved a directive to regulate authorizations for licenses in telecommunications services'8 and another to regulate the protection of privacy associated with the transmission of personal data.4 9 The Council and European Parliament approved a second directive to regulate the mutual recognition of standards for equipment in 1998 50 and yet another directive on the mutual recognition of standards in equipment in 41. Council Resolution 94IC379Io3 of 22 December 1994, Officialfournal of the Europran Communities C'r379 (31 December 1994): 4· 42. Council Resolution 95IC188Io2 of 29june 1995, Official journal of the European Communities C-188 (22july 1995): 3; Council Resolution 95ICzs8lo1 of 18 September 1995, Of ficial]ournal of the European CommunitiesC-258 (3 October 1995): 1. 43· Official journal of the European Communities C-122 (31 March 1995): 4· 44· Council Directive 95I6IEC of 13 December 1995, Official journal of the European CommunitiesL-321 (30 December 1995): 6. 45· London Representation of the European Commission, Background Report: Telecommuni· cations LiberalizationBI1I96 Qanuary 1996), 1. 46. Commission Directive 96/I9IEC. 4 7. European Parliament and Council Directive 97 I 33IEC of 30June 1997, Official journal oftheEuropean CommunitiesL-199 (z6July 1997): 32. 48. European Parliament and Council Directive 97 I 13IEC, O[ficialfournal of the Eumpean CommunitiesL-117 (7 May 1997): 15. 49· European Parliament and Council Directive 97 1661EC of 15 December 1997, Official Journal of the European CommunitiesL-24 (3ojanuary 1998): 1. 50. European Parliament and Council Directive 98/I3IEC of 12 February 1998, Official journal of the European Communities L-74 (12 March 1998): 1.

From Jurisprudence to Policy

107

1ggg. 51 The European Commission also amended its previous directive on cable television networks by adopting a directive in 1ggg to ensure that telecommunications and cable television networks that are owned by a single operator are separate legal entities. 52 With each Article 86 (go EEC) directive since 1ggo, the European Commission extensively consulted the Council, European Parliament, and private and public actors in its transnational policy network on the content and timing offurther liberalization measures. 53 While rallying political support behind the general liberalization effort, the European Commission provided the final push into the market with its Article 86 (go EEC) directives. Meanwhile, the Council and European Parliament's harmonizing directives articulated conditions for the ensuing market, including exceptions and delayed implementation for particular Member States. The specter of competition directives under Article 86 (go EEC) facilitated the reform process by encouraging Member States to broker compromises for a harmonized regulatory framework before the European Commission opened markets." The ECJ contributed to this policy process by legitimat~ ing the European Commission's regulatory authority, denying Member State demands for a purely intergovernmental process in the Council. Finally, most Member States have proceeded to implement EC provisions in this field, and a competitive telecommunications market is actually emerging across the EU. 55 By November 1ggg, the European Commission concluded that "21 months after the introduction of full competition, the [Eu51. European Parliament and Council Directive 99/ 5/EC of 9 March 1999, Official Journal of the European Communities L-g1 (7 April1999): 10. 52. Commission directive gq/64/EC of 23June 1999, Official journal of the F:uropean CommunitiesL-175 (wJuly 1999): 39· 53· Interviews confirmed the process of consultation traced in official documentation: Direction Generale des Postes et Telecoms, Paris, 22 February 1996; Deutsche Telekom, Bonn, 23 January 1 gg6; Commission Legal Service, Brussels 2 June 1 995; Commission Legal Service, telephone interview with Brussels, 18 March 1996; Center for European Policy Studies (CEPS), Brussels, !June •995· 54· Interviews at Commission Legal Service, telephone interview with Brussels, 18 March 1996; Commission Directorate General IV-B, Brussels, 30 May 1995; a private consultant on telecommunications regulation, London, 17 July 1995; a private consultant on European regulation, Brussels, 31 May 1995· 55· "Debates of the European Parliament," Addresses by Vice President of the Commission Bangemann, Official Journal of the European Communities, Annex no. 3-441 (1994): 42, and no. 3-442 ( 1994): 24; Commission, "Fourteenth Annual Report on Monitoring the Application of Community Law-1gg6," Official Journal of the Eumpean Communities C-332 (3 November 1997): 33-34, 47-49; Commission, "Sixteenth Annual Report on Monitoring the Application of Community Law-1gg8," COM (gg) 301 final (gJuly 1999), g6-37; 53-55; Commission, "Seventeenth Annual Report on Monitoring the Application of Community Law-1ggg," COM (2ooo) 92 final (23June 2000), introduction, 39-42,57-59.

108

Justice Contained

Table 4.2. European Regulatory Framework for Telecommunications

Market Opening Terminal equipment

Value-added services

Cable networks

Mobile and personal communications Full liberalization: Infrastructure and voice telephone services

European Commission

Council of Ministers

European Parliament

Directive 88/ 301/EEC Directive 94/ 46/EC Directive 90/ 388/EEC Directive 94/ 46/EC

Resolution 88/C257/01 Directive 91/263/EEC Directive 98/13/EC Directive 99/5/EC Resolution 88/C257/01 Directive 90/387/EEC Directive 92/44/EEC Directive 97/66/EC Directive 97/13/EC

Directive 98/13/EC Directive 99/5/EC Directive 97/66/EC Directive 97/13/EC

Directive 95/ 51/EC Directive 99/ 64/EC Directive 96/ 2/EC Directive 96/ 19/EC

Resolution 93/C213/01

Resolution M-0063/95 Resolution M-0111/95 Resolution M-0097/95 Resolution M-0063/95 Resolution M-0111/95 Directive 97/33 EC

Resolution 93/C213/01 Resolution 95/C188/02 Resolution 93/C213/01 Resolution 94/C379/03 Resolution 95/C258/01 Directive 95/62/EEC Directive 97/33 EC

ECJ C-202/88, 1991 Upholds Commission Directive 88/301/EEC C-271, C-281, and C-289/90, 1992 Upholds Commission Directive 90/388/EEC

ropean] regulatory framework in place drives telecommunications services markets in the Member States with an accelerating growth rate, large numbers of market entrants and falling tariffs. "56 Table 4.2 charts the policy process in telecommunications and illustrates the parallel development of political consensus and independent legislative actions by the European Commission. As the next section shows, this dynamic process of reform rested on an evolution of interests and mobilization of political pressure that emerged over a decade.

Interests, Institutions, and the Mobilization of Support for liberalization The process of telecommunications reform was far from an automatic response to legal obligations. Member States were not initially inclined to embrace European-level action in the telecommunications field with enthusiasm. Traditionally associating telecommunications with internal security 56. Commission

2000,

57·

From Jurisprudence to Policy

109

and domestic social and economic goals, Member States historically pursued national strategies and considered telecommunications to be exempt from European coordination." Wayne Sandholtz demonstrates that Member States responded positively to European initiatives for cooperative research and development ventures in telecommunications only after they exhausted exclusively national strategies. 58 And, when the European Commission initially promoted the introduction of competition in the telecommunications field in 1984, the British and Dutch governments were the only immediate allies among the Member States. The application of the European competition regime was controversial because the liberalization of Europe's telecommunications sector necessitated confrontations with monopolistic public enterprises that were uncompetitive by international standards. Furthermore, the restructuring associated with the introduction of competition threatened the job-security of well-organized public-sector workers in an era of high unemployment and recession. As a result, the European Commission's effort to extend the EC competition regime to nationally regulated service monopolies in telecommunications inspired conflict among concentrated, resourceful actors with intense interests. The management and labor force of monopolist enterprises, aspiring market entrants, and corporate consumers are all relatively concentrated actors whose preferences relative to liberalization are intense. The traditional importance of the communications sector to national economic competitiveness and security also generates intense policy preferences within governments. Liberalization tends to affect the diffused and relatively weak preferences of individual consumers as well, but these actors generally fail to influence the policy-making process directly. Competing demands among Member States and concentrated, intensely interested actors, therefore, generated the dynamic process of regulatory reform in this field. The evolution of a consensus and opportunities to marginalize sources of resistance facilitated liberalization efforts. In this section, I specifY the preferences and mobilization of political pressure that influenced the policy process and regulatory outcome in telecommunications. The Evolution of Interests Related to Telecommunications Liberalization

Telecommunications providers originally operated effectively as "natural monopolies" that managed large network infrastructures to provide service 57· Wayne Sandholtz, High-Tech Europe (Berkeley: University of California Press, 1992), 59-112; Volker Schneider, Godefroy Dang-Nguyen, and Raymund Werle, "Corporate Actor Networks in European Policy-making: Harmonizing Telecommunications Policy," Journal of Common Mark£t Studies 3l! (1994): 481-82. 58. Sandholtz 1992, 113-42.

110

Justice Contained

universally. Network infrastructures in telecommunications were traditionally a highly specific asset, costly to move out of their current use in the event of changing market opportunities. Because such network infrastructures were prohibitively expensive for multiple providers to develop and maintain, governments everywhere initially encouraged their provision by formally restricting entry, monopolizing the means of production within public enterprises, or granting exclusive rights to private producers. Only such authoritative organization of the sector could promote the long-term contracting that was necessary to recoup returns on investment in highly specific assets. 59 Recognizing this relationship, actors in asset-specific industries typically hold intense, cohesive preferences to preserve barriers to entry.60 Yet as long as initial investment costs were high, telecommunications providers did not even have to rely on formal barriers to protect their exclusive status. Exclusive provision became contested only after technological changes led to the development of both alternative means of providing telecommunications services and new uses for existing network assets. For instance, cellular phone technologies and cable television each created mechanisms to provide telecommunications services that bypassed traditional network infrastructures. And advances in information technologies such as highspeed data transmission expanded the range of uses for traditional network infrastructures. These technological developments generated new opportunities for profitable exchange, which created incentives to replace politically protected regimes with market mechanisms61 and generated cleavages between public and private actors with interests in the industry. Ultimately, the capacity to bypass traditional infrastructures and the rapidly increasing reliance of advanced economies on communications technologies empowered supporters of liberalization over opponents. The progressive liberalization of European telecommunications cannot be understood in isolation from this evolution of interests."2 Yet, the successful push to open 59· Oliver Williamson, MarkeL' and Hierarchies (New York: Free Press, 1975); Oliver Williamson, Economic Institutions of Capitalism (New York: Free Press, tg8s); Philip Cerny, "Globalization and the Changing Logic of Collective Action," International Organization 49 (autumn 1995): 6oo-1, 6os-6. 6o. Jeffry Frieden, Debt, DeveWpment, and Democracy: Modern Political Economy and Latin America, 1965-1985 (Princeton, NJ.: Princeton University Press, 1991); S. Magee, 'Three Simple Tests of the Stopler-Samuelson Theorem," in Issues in International Economics, ed. P. Oppenheimer (London: Oriel, 1 g8o). 61. Cerny 1995, 6o2; Jeffry Frieden and Ronald Rogowski, 'The Impact of the International Economy on National Policies: Analytical Overview," in Internationalization and Domestic Politics, ed. Robert Keohane and Helen Milner (Cambridge: Cambridge University Press, Igg6), 26,36-38,45· 62. Ian Bartle, 'Transnational Interests in the European Union: Globalization and Changing Organization in Telecommunications and Electricity," .Journal of Common Market Studies 37 (September tggg): 363-83.

From Jurisprudence to Policy

111

telecommunications markets across the entire EU, in states with more and less competitive telecommunications industries, is also inconceivable outside the EU institutional context, where aspiring liberalizers could exploit liberal treaty provisions, exceptional supranational autonomy in the field of competition, and a binding legal regime to achieve relatively rapid and widespread reforms. Interests in pursuing liberalization in Europe originally developed in response to U.S. and japanese advances in information technologies. Before any concrete European initiatives existed, the United Kingdom embarked on its own domestic program of liberalization. British telecommunications markets progressively opened from 1981 to 1994· Mercury received a license to build and operate an independent network in 1982, and the traditional public monopoly, British Telecom, privatized in 1984. To promote the international expansion of its telecommunications industry, the British government supported telecommunications liberalization throughout the EU. 63 This "early" British liberalization is consistent with U.K institutional structures that facilitated comprehensive legislative reform and the marginalization of opponents. Both the absence of veto points in the legislative process and the structural weakening of British labor through Margaret Thatcher's trade union reforms increased the capacity of Conservative governments to respond to new interests in liberalization. 64 Over the course of the 198os, Conservatives proceeded to liberalize multiple sectors of the British economy ahead of Europe, including telecommunications, despite the substantial dislocations that liberalization entailed for less competitive and technologically backward segments of British industry. Meanwhile, most other Member States and their public telecommunications providers were initially opposed to the prospects of liberalization in their national markets. 65 Slower responses to incentives to liberalize in France and Germany were consistent with institutional structures that provided potential opponents with more voice in policy processes. Periods of cohabitation in the divided French executive and constitutional challenges can require the recognition of competing interests in French reform efforts, while the shared competencies of Germany's cooperative federalism and coalition governments necessitate compromise and consensus building among diverse interests. More specifically, the organization of labor re63. Department of Trade and Industry, Telecommunications Liberalization in the UK: Key Elements, History, and Benefits (London, March 1995), 5-14. 64. Geoffrey Garrett and Peter Lange, "Internationalization, Institutions, and Political Change," International Organization 49 (1995): 627-55. 65. Schneider, Dang-Nguyen, and Werle 1994, 484; interviews at Commission Legal Service, Brussels, I june 1995; Deutsche Telekom, Bonn, 23]anuary 1996; Commission legal service, telephone interview with Brussels, 18 March 1996; private consultant on telecommunications regulation, London, 17 July 1995.

112

Justice Contained

lations in France and Germany generated pressures to accommodate the demands of potential losers in a liberalized telecommunications market. Extensive organization of French public-sector labor, combined with minimal organization of French private-sector labor, privileges the interests of public-sector workers whose conditions are subject to government policy rather than any market constraints. By contrast, the extensive and centrally coordinated organization of German labor in the private and public sectors creates countervailing pressures to reconcile the interests of those who must compete in markets with those who do not. 66 While institutionally privileged French public-sector workers would be more capable of producing genuine impediments to liberalization, supporters of liberalization in both countries would need to accommodate the interests of organized labor to some degree. Despite these delays relative to the United Kingdom, however, the interests of public and private actors evolved rapidly during the 198os, leading to transformations in France and Germany that would also eventually lead Europe. The French government began promoting telecommunications liberalization in 1986 and launched a domestic program of reform. By 1988 French measures to liberalize provision in cable television, mobile and value-added services, and terminals equipment preceded European measures. Since 1994, the French government accelerated the pace of further telecommunications liberalization, often ahead of European deadlines. 57 Meanwhile, official German interest in liberalization originated in the Bundeskartelamt (federal antitrust authority) from 1980 and was followed by interest within the coalition of Christian Democrats and Liberals who came into power in 1982. By 1986, the German government established an advisory commission to design a coherent program of reform. A reform adopted in 1989 initiated liberalization in value-added services and terminal equipment, which was followed by liberalization in mobile communications, satellite services, and corporate networks in the 1990s. Parallel to the French, the German government accelerated its program of domestic reform after 1994, opening markets largely ahead of the European schedule. 68 This acceptance of liberalization developed as national governments 66. Garrett and Lange 1995, 636-640. 67. Volker Schneider, "Institutional Reform in Telecommunications: The European Union in Transnational Policy Diffusion," in Transfarming Europr: Europeanization and Domestic Change, ed. Maria Green Cowles, James Caporaso, and Thomas Risse (Ithaca, N.Y.: Cornell University Press, 2001); J.P. Chamoux, Tt!f£coms: La fin des priviV!ges (Paris: Presses Universitaires de France, 1993); inten~ews with a private telecommunications la"'Yer and former Commission official from Directorate General IV, Brussels, 14 September 1995; a private telecommunications consultant, Namur, 15 September 1995· 68. Schneider 2001; intemews at Deutsche Telekom, Bonn, 23 January 1gg6; private telecommunications consultant, Namur, 15 September 1995·

From Jurisprudence to Policy

113

came to perceive the sector in the context of globalization and industrial competitiveness. National leaders' views converged with the European Commission as they observed the international expansion of U.S. firms and impact ofliberalization in the United Kingdom and Japan. Skeptical about the capacity of their public monopolies to keep pace with technological change, Member States decided to promote innovation and expansion in the field by opening their markets to competition. As discussed in the previous section, Member States confirmed their support of the European Commission's major policy initiatives in telecommunications with Council Resolutions from 1g88 to 1gg5."" The Council Resolution of 1g88 followed the European Commission's corresponding directive under Article 86 (go EEC) by six weeks, while Council Resolutions from 1gg3 to 1gg5 preceded the European Commission's adoption of corresponding directives under Article 86 (go EEC). The mobilization of support from a set of concentrated actors with increasing interests in liberalization informed and facilitated this shift in the position of national governments. A critical source of support for liberalization developed within the leadership of French and German telecommunications operators. The successful example of the privatized British Telecom, subject to emerging competition in domestic and international markets, interested the French and German telecommunications administrations and managers of France Telecom and Deutsche Telekom. Managers recognized the potential for profitable expansion in a competitive, internationalized telecommunications regime. They could use the need to compete in open markets as justification for changes in traditional practices that inhibited efficiency, including the phasing out of lifetime employment in civil service contracts. Hiring workers under private contracts instead would facilitate more flexible adjustment to technological and market change. 7° Furthermore, an interest in joint ventures, whose approval the European Commission connected with domestic liberalization, led France Telecom and Deutsche Telekom to pressure their governments for faster, more extensive liberalization. 71 69. Council Resolution 88/C257/o1; Council Resolution 93/C213/o1; Council Resolution 94/C379/o3; Council Resolution 95/C188/o2; Council Resolution 95/C258/cn. 70. Josef Esser and Ronald Noppe, "Private Muddling through as a Political Programme? The Role of the European Commission in the Telecommunications Sector in the 1980s," West European Politics 19 (1gg6): 548, 551-52; interviews at Secretariat general de comite interministeriel pour les questions de cooperation economique europeenne (SGCI), Paris, 16 October 1995; Direction generale des Postes et Telecoms, Paris, 22 February 1996; Commission Legal Service, Brussels, 1 June 1995; Deutsche Telekom, Bonn, 23January 1996; private consultant on telecommunications regulation, London, 17 July 1995; private telecommunications consultant, Namur, 15 September 1995. 71. Deutsche Telekom AG, "Position of Deutsche Telekom AG Concerning the Draft Directive Amending the Commission Directive Regarding the Implementation of Full Competition in Telecommunications Markets," 7 December 1995, W2;.foint press release by Deutsche

114

Justice Contained

Other relatively concentrated actors who demanded reform from national governments included manufacturers of equipment, aspiring entrants, and corporate users. Volker Schneider, Godefroy Dang-Nguyen, and Raymund Werle argue that the European Commission helped coordinate this pressure for reform by building a transnational policy network among these actors: the European Commission consulted interested parties in order to forge a broad alliance that would support its initiatives at the national and European level. Individual multinational firms, national peak and sectoral organizations, the Union oflndustrial and Employers Confederations of Europe, the Roundtable of European Industrialists, and the Association of European Telecommunications Producers represented producer interests in consultations with the European Commission. Individual firms, national chambers of commerce, and user associations at the national, European, and international level represented the interests of corporate consumers. Schneider, Dang-Nguyen, and Werle trace a process of"reverse lobbying,"where the European Commission facilitated the collective action of business users as a means to disseminate support in national policy arenas. 72 Clearly identi.f}'ing winners is difficult in most instances ofliberalization/' but incentives for liberalization were strong for many actors. Both aspiring entrants and corporate users were interested in a global market for telecommunications equipment that had more than doubled in value in constant dollars from 1977 to the mid-1g8os, when it reached$go billion U.S. dollars. The market for value-added services had been growing from 25 to 30 percent annually, reaching $505 billion in 1992. By 1993 the largest twenty-five public telecommunications operators in the developed world were more profitable than the largest hundred commercial banks. In addition to being a lucrative field to enter, access to more sophisticated services at falling costs would enhance the competitiveness of industries that rely on information and communication. Indeed, advances in the telecommunications sector contribute to technological change across all sectors of the economy. 74

Telekom, France Telecom, and Sprint, "Telekom Top Info," 31 January 1gg6; FrancE' Telecom, "France Telecom, Deutsche Telekom et Sprint lancent Global One," La kttre d1! France Tef£com, no. 181, 5 February 1gg6, 1-4. 72. Schneider, Dang-Nguyen and Werle 1994, 487-g1; Wayne Sandholtz, "Institutions and Collective Action: The New Telecommunications in Western Europe," World Politics 45 ( 1993): 242-43· 254· 73· Andrew Moravcsik, "Preferences and Power in the European Community: A Liberal Intergovernmentalist Approach," in Economic and Political Integration in Europe: Internal Dynamics and Global Context, ed. Simon Bulmer and Andrew Scott (Oxford: Blackwell, 1994), 44-46; Andrew Moravcsik, "Liberal Intergovernmentalism and Integration: A Rejoinder," Journal of Common Market Studies 33 (December '99.')): 612-14. 7 4· Sandholtz 1992, 55-57; European Commission Delegation Office of Press and Public Affairs, "Commission Accelerates Telecoms Liberalization, Stresses Importance of Universal Service," European Union News (Washington, D.C.), no. 13/96 (1 March 1996).

From Jurisprudence to Policy

115

Meanwhile, identifYing obvious losers among existing producers was not easy in the telecommunications sector. Uncertainty about the consequences of policy changes generally moderated the demands that governments faced from less competitive providers and their labor forces. Expanding opportunities in the telecommunications industry, where enterprises could provide an increasing variety of services and workers could pursue employment in a growing range of activities, diluted opposition even from the producers who anticipated imminent losses from competition. Because it was difficult to predict longer-term losses, potential losers failed to generate sustained opposition to liberalization. The telecommunications personnel of national operators constituted the major potential source of concentrated opposition to liberalization. The fear of large-scale redundancies in the wake of liberalization could have generated intense opposition to policy change. Organized in publicsector unions and privileged through civil-service statutes, telecommunications personnel possessed the capacity for collective action in many Member States. Indeed, a strike of 75 percent of French telecommunications employees delayed a 1993 attempt to privatize France Telecom. 75 Yet the U.S. and U.K. experiences in telecommunications liberalization partially diluted negative expectations in the early stages of policy debate. Expanding employment opportunities with new entrants in the U.S. and U.K. markets and the achievement oflabor reductions through voluntary retirement obscured the specter of mass layoffs. The relative efficiency with which France Telecom operated in the early 1990s also reduced the need for redundancies. While Deutsche Telekom faced the prospect of a much higher rate of layoffs, German labor organizations had also been much more constrained to respond cooperatively to market adjustments. Most significantly, France Telecom and Deutsche Telekom strategically neutralized the most threatening form of organized opposition by retaining privileges for existing civil servants and hiring all incoming employees under private contracts. This policy pacified existing fonctionnaires and Beamte (French and German civil servants) and resulted in their attrition, leading to the gradual transformation of the labor force. 76 Little political resistance to telecommunications liberalization developed among consumer constituencies in the broader public, despite the potential for increases in local rates. Local telephone services have traditionally been subsidized by high charges for long-distance services in most European countries. The disaggregation of local and long-distance providers 75· John Horrocks, European Guide to Telecommunications Liberalization (Surrey: Horrocks Technology Publisher, 1993), France page 1. 76. Interviews at Foreign Ministry of France, Paris, 6 October 1995; SGCI, Paris, 16 October 1995; Direction generale des Postes et Telecoms, Paris, 22 February 1996; Deutsche Telekom, Bonn, 23]anuary 1996.

116

Justice Contained

eliminates local subsidies, potentially creating the need to raise local phone rates. This is most problematic for Member States with the lowest GDPs per capita, where real costs for local services are prohibitive for many individuals. By contrast, local services in wealthier Member States, including France and Germany, primarily paid for themselves. 77 In summary, the European Commission advocated a process of liberalization in telecommunications that gained the political support of Member States and organized, resourceful societal interests. Initially opposed to reform, France and Germany began opening their domestic telecommunications markets ahead of European initiatives. The leadership of French and German telecommunications enterprises neutralized the only significant source of concentrated opposition by grandfathering the privileges of highly organized civil servants. Law and the Mobilization of Political Pressures

Policy reform in telecommunications progressed primarily through political channels due to the prominence of concentrated, intensely interested actors. Legal interpretation was at the foundation of the European Commission's directives under Article 86 (go EEC), but actors affected by policy change pursued their aims overwhelmingly in political arenas. Two instances of legalized bargaining preceded proposals for reform by the European Commission: in 1978 the Society for Worldwide Information and Fund Transfer (SWIFT) issued a complaint with European Commission Directorate General IV (DG-IV) about public telecommunications providers' restrictions on the use of international leased lines and prohibitive tariffs. The international consortium of banks represented by SWIFT successfully negotiated the use of public telecommunications infrastructure to operate an independent network for data exchange and abandoned the formal legal complaint with DG-IV. SWIFT attained essentially what the European Commission sought fifteen years later with ONP for leased lines. 78 Societe internationale de Telecommunications Aeronautiques (SITA), which operates a system of air traffic control between countries, also successfully negotiated the independent use of public infrastructure for information exchange.79 However, these examples failed to inspire copycat claims. The vast majority of corporate users and aspiring entrants have been loathe to pursue their goals through legal means, whether informal or formal. This 77· Interview at Department of Trade and Industry, London, 14]uly 1995· 78. Schneider, Dang-Nguyen, and Werle 1994, 483. 79· Interview with a private consultant on telecommunications regulation, London, 17 July 1995·

Fig. 4.1. Competing Interests and Policy Outcomes in Telecommunications ECJ legal interpretation asserts applicability of competition rules to public enterprises and upholds the European Commission's legislation to open telecommunications markets unilaterally.

Liberalization promises benefits of intense interest to concentrated actors and threatens losses of intense interest to concentrated actors.

European Commission defends its legislation successfully before the ECJ and mobilizes networks of political support.

Intensely interested, concentrated beneficiaries of liberalization mobilize to support reform overwhelmingly through political arenas. Potential losers neutralized by uncertainty and the cooptation of the most organized segments of labor.

Supranational measures liberalize telecommunications while intergovernmental measures regulate market relations.

118

Justice Contained

stance reflects the dilemmas associated with access to justice in the EU. Business skepticism about the possibility of judicial redress in national courts is prevalent across Europe. 80 Private lawyers, telecommunications consultants, and European Commission officials all confirm that parties interested in opening telecommunications markets have viewed challenges before national courts as a risky and inefficient strategy. Rather than initiating suits in courts in each Member State, interested parties preferred to seek the European Commission's assistance with a European-wide opening of markets. 81 The reluctance to wage direct legal battles also reflects the preferred strategy of actors who interact frequently and enjoy access to political arenas. Legal challenges threaten to poison mutually beneficial relations that already exist and subject all parties to the potentially unpredictable decision of a court. Conversely, political negotiations for those with access to policy networks offer a much more flexible and controlled setting to arrange compromises that are satisfactory to all sides. As a result, most parties pursued their complaints to the European Commission on an informal basis, being unwilling to initiate formal claims against national governments.82 Both aspiring entrants and corporate users also avoided legal confrontations with monopolist telecommunications providers. Many potential entrants in telecommunications among other utilities enjoyed exclusive rights in their own domain and hesitated to initiate legal claims that could come back to haunt their own sector. Other potential entrants relied on public telecommunications enterprises as purchasers of their products, in addition to being completely dependent as telecommunications consumers. Even large U.S. firms that had experience with legal battles on the home front, such as IBM and AT&T, abstained from formal complaints and legal challenges. 83 So. Kieran Bradley and Alastair Sutton, "European Union and the Rule of Law," in Maastricht and Beyond: Building the European Union, ed. Andrew Duff, John Pinder, and Roy Pryce (London: Routledge, 1994), 2_t)l-52. 81. Interviews with a private telecommunications lawyer and former official in Commission Directorate General IV, Brussels, 14 September 1995; private lawyer, Brussels, 30 May 1995; private legal consultant on European regulation, Brussels, 31 May 1995; private legal advisor on European competition law, Brussels, 29 May 1995; Commission Directorate General IV-B, Brussels, 30 May 1995· 82. Interviews at Commission Directorate General IV-B, Brussels, 30 May 1995; private telecommunications consultant, Namur, 15 September 199.5; International Telecommunications Users Group (INTUG),18july 1995; Department of Trade and Industry, London, 18july 1995· 83. Interviews with a private telecommunications consultant, Namur, 15 September 199.5; Commission Legal Service, phone interview with Brussels, 13 March 1996; Commission Directorate General IV-B, Brussels, 30 May 1995; private legal consultant on European regulation, Brussels, 31 May 1995·

From Jurisprudence to Policy

119

During the process of reform, only two disputes between enterprises over competition abuses resulted in ECJ litigation, and only three parties "voluntarily" disputed competition issues with Member States or their agents before the ECJB4 Most litigation in this field results from Member States' prosecution of entrants flaunting national restrictions or European Commission prosecution of a handful of Member States for their failures to implement telecommunications directives. Five ECJ preliminary rulings involved national criminal proceedings against entrants, who had simply entered the market on the sly rather than go bankrupt while they waited for legal clearance. The defendants invoked the European Commission directive on terminal equipment as justification in these cases. 85 Seven ECJ cases involved the European Commission's prosecution of infringements in a few Member States, including Belgium, Greece, Ireland, and Luxembourg.'6 Litigation before national courts has not been much more prevalent, and once again, a concentration of cases involves national prosecution of entrants who ignored national restrictions. British courts have fielded only three cases related to EU telecommunications competition, German courts nine, and French courts twenty-seven. In the French case, twenty-five of twenty-seven disputes involved criminal proceedings against private operators. National court litigation primarily involves European telecommunications directives. Article 86 (go EEC) litigation has been trivial, producing only one case related to telecommunications before the French Conseil d'lhat. 87 84. Sociite alsacienne et lorraine de telecommunications et d'etectronique (Alsatel) v SA Novasam C24718611988, (1990) E.C.R., II-231; Rigie des teligraphes et des teliphones v GB-lnno-BM SAC18/88, (1991) E.C.R., l-5941; SA CentreBelge d'etudes de marche-telimarketing (CBEM) v SA Compagnie luxembourgeoise de telidiffusion (CLT) et SA Information publicite Benelux (IPB) C-311/84, (1985) E.C.R., 3261; Te/emarsicabruzzo SpA v Circostel et Ministero delle Paste e delle Teleromunicazioni et Ministero della Difesa C-320-322/ 90, ( 1993) E.C.R., l-393; The Queen v Secretary of State for Trade and Industry, ex parte British Telecommunications plr C-302l94, ( 1996) E.C.R., 1-6417. 85. Procureur du Roi v]ean-Marie Lagauche et autres C-46l9o and C-93l91, (1993) E.C.R., I5267; Procedure pmale v Francine Decoster; epouse Gillon C-69/91, ( 1993) E.C.R., 1-5335; Procedure Pmale v Annick Taillandier; epouse Neny G92l91, (1993) E.C.R., I-5383; Procedure Pmale v Franrois Rouffeteau et !Wbert Badia C-314/ 93, ( 1994) E.C.R., l-3257; Criminal Proceedings against Thierry Tranchantand TeliphoneStoreSARLC-91194, (1995) E.C.R., l-3911. 86. Commission v Belgium C'r218l94, (1995) E.C.R., l-89; Commission v Luxembourg Cr 22ol94, (1995) E.C.R., I-1589; Commission v C.reeceC-259194, (1995) E.C.R., l-1947; Commission v Greece C-26ol94, (1995) E.C.R., l-2603; Commission v Luxembourg C-221l94, (1996) E.C.R., l-5669; Commission v Ireland C-239/94, (1996) E.C.R., l-g83; Commission v Luxembourg C-59lg8, (1 999 ) E.C.R., 1-!!81. 87. Lexis search on 22 November 2000, yielded three cases: British Telecommunications plc v Director General of Telecommunications, Queen's Bench Division (Crown Office List), CO I 1506I 2ooo (4 August 2ooo); R v Director of Telecommunications, ex parte British Telecommunications plc, Queen's Bench Division (Crown Office List), COI3596/g6 (20 December 1996); R (on the application of Orange Personal Communications Ltd) v Secretary of State for Trade and Industry,

120

Justice Contained

Now that the new regulatory framework for telecommunications is operational and competition is developing, litigation may eventually play a greater role in resolving disputes over market access and abuse. The comprehensive legislative framework and the greater number of actors together decrease the risks associated with future litigation. Parties can assess their chances before courts on the basis of established legal codes and need not worry they are alienating the only source of a particular good or service. Yet such "private enforcement" of legal obligations may be unnecessary as long as the European Commission remains committed to realizing reforms in this field. Mter having pushed European legislative reform, the European Commission appears determined to ensure the timely and adequate implementation of telecommunications directives at the domestic level. The European Commission established a team to monitor the transposal of telecommunications directives in 1997, which involves both official investigation and the commissioning of audits from independent consultants. The resulting infringement proceedings have steadily maintained pressure to eliminate lingering restrictions, which have primarily targeted Member States that had already negotiated delayed schedules of implementation (mostly Mediterranean countries) and/or that habitually have difficulties coming into conformity with European legislation (Belgium, Italy, Greece). 88 In conclusion, the enforcement of dormant legal obligations in the Treaty of Rome contributed to a process of reform that liberalized telecommunications markets in Europe. However, the ECJ's interpretation of competition rules and the European Commission's application of these rules did not independently compel policy change. Iflegal obligations alone can force policy change, monopolistic public provision of telecommunications services should have ended with the implementation of the Treaty of Rome or at least with the reassertion of treaty obligations in ECJ case law dating from the 1970s. Instead, the positive political response of concentrated, intensely interested actors was a fundamental component behind the opening of telecommunications markets across Europe. The European ComQueen's Bench Division (Administrative Court), C0/4231/1999 (2.5 October 2ooo);juris search on 4 December 2000 yielded nine cases in administrative courts from 1997 to 20oo;Jurifrance searches on 15-16, 20, 23 November 2000 yielded twenty-seven cases, two of which derived from the administrative Conseil d'Etat and the remaining twenty-five from the criminal chambers of the Cour de Cassation. This is not necessarily an exhaustive list of national litigation because the databases do not include all unreported decisions from lower courts. However, these data bases provide the best available indicator of national litigation levels, and they provide a good comparison with national litigation in other case studies because the underreporting of lower court decisions affects all sectors in each country. 88. Commission 1999, COM (gg) 301 final, 36-7, 53-5; Commission 2oooa, COM (2000) 92 final, introduction, 39-42, 57-9·

From Jurisprudence to Policy

121

mission did not attempt to force change solely through legal means as it theoretically could, but consistently worked to create political consensus and incrementally eliminated exclusive rights. Formal legal challenges ultimately represented a small fraction of the policy process in telecommunications, where the vast majority of reform occurred through politically negotiated legislation. The shadow of the law, which emerged through ECJ interpretation and Article 86 (go EEC) directives, speeded along this negotiation and broadened the scope ofreforms to the entire EU. But theresulting reforms reflect the willing consent of national governments nonetheless.

5 European Law and What States Made of It: Intergovernmental Reform of Electricity

In

the midst of telecommunications reform, the European Commission tried to extend its supranational method of liberalization under Article 86 (go EEC) 1 to the electricity sector. Both the functional and legal logic for liberalization were broadly parallel in the two sectors. The European Cornmission linked the prevalence of exclusive rights to provide electricity with high prices and insufficient technological innovation. Competition in an internal market for electricity would lead to more efficient production of energy, improving the productivity of European industry as a whole. Legally, the supranational approach to regulatory reform appeared promising as a means to open electricity markets to competition. The European Court of justice had upheld the European Commission's competence to issue competition directives in telecommunications, and there was no obvious legal reason for the electricity sector to be treated any differently. As a result, the European Commission announced its intention to liberalize electricity provision, drafted a directive under Article 86 (go EEC) to complete "the internal market in electricity and gas," and initiated infringement proceedings against import and export monopolies in nine Member States. Despite clear legal justification for these actions, the European Commission met with massive political resistance to its program of reform. Although the European Commission sustained its infringement proceedings against import and export monopolies in electricity, it rapidly abandoned 1. Treaty of Amsterdam (2 October 1997); Treaty Establishing the European Economic Communities (25 March 1957).

122

European Law and What States Made alit

123

the supranational approach to regulatory reform and proposed draft directives under Article g5 (1ooa EC) 2 that are subject to approval by national representatives in the Council of Ministers and Members of the European Parliament (MEP). Nearly six years of negotiation between national governments and MEPs culminated in the adoption of an electricity directive that accommodated the demands of the most hesitant liberalizers, significantly diluting the competition regime in electricity. This process of policy-making indicates that reforms that defY political consensus cannot simply be legally imposed. Instead, legal and political pressures mediated the compromise that ultimately led to partial liberalization of electricity markets in the EU. In this chapter I show that the concentration and intensity of interests affected by liberalization in electricity contributed to a political process that limited the scope of policy change. In this case, Member States retained the lion's share of control over the pace and extent of liberalization, although the European Commission and ECJ pushed reform further than it ever could have been expected to progress in the absence oflegal pressure. The chapter proceeds by chronicling the evolution of legal obligations, tracing the policy responses, and specifYing how the constellation of competing interests is connected to the mobilization of pressure and policy outcomes.

Legal Innovations: Rebalancing Services of General Interest against Trade Article 86 (go EEC) addresses any sector characterized by provision through public enterprises or enterprises granted special rights by Member States. As a result, the supranational Article 86 (go EEC) approach was available to liberalize energy markets. Indeed, the European Commission consistently insisted on its legal capacity to use this tool across sectors, including electricity.' A legal adviser on competition policy observed that an electricity directive required only the substitution of the words in existing telecommunications directives from "telecommunications" to "electricity. "• Two years after the European Commission balked at imposing this approach, the ECJ proceeded to complicate the legal situation by recognizing potentially legitimate exceptions to the application of competition rules in 2. Single European Act, Article 18 amends Article 100 EEC (January 1987). 3· Commission, "Eleventh Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law-1993," Official journal of the European CommunitiesCrl54 (6June 1994): 30-31; interview at Commission Directorate General IV-C, Brussels, 27 September 1995. 4· Interview at Commission Directorate General IV-B, Brussels, 30 May 1995.

124

Justice Contained

preliminary rulings in 1993 and 1994· In its Corbeau and Almelo rulings, 5 the ECJ emphasized the competing demands of the general economic interest clause and the restriction on trade in Article 86,2 (90,2 EEC): "Undertakings entrusted with the operation of seiVices of general economic interest or having the character of a revenue-producing monopoly shall be subject to the rules contained in this Treaty, in particular to the rules on competition, insofar as the application of such rules does not obstruct the performance, in law or in fact, of the particular tasks assigned to them. The development of trade must not be affected to such an extent as would be contrary to the interests of the Community. "6 In Corbeau, the ECJ reasoned that a restriction of competition may be justified where it is necessary to ensure the "financial equilibrium" of enterprises that are obligated to perfoml seiVices in the general economic interest. In Almelo, the ECJ instructed the national court to consider the "economic conditions" under which an enterprise must operate. The ECJ left the precise criteria for the "proportionality test," which balances the general economic interest against the damage done by the restriction on competition and trade, to the national court that had referred the question. This ECJ case law did little to resolve the interpretive ambiguity in the treaty, but subtly shifted the burden of proof. Previously, the ECJ had not readily accepted exceptions to competition rules, requiring national officials to prove that the application of competition rules would be "incompatible" with the performance of tasks in the general economic interest. In Corbeau and Almelo, the ECJ indicated that it was more receptive to the possibility of exceptions under the general economic interest clause, where national officials could try to convince national courts that competition might threaten the economic stability of the enterprise with exclusive rights. The European Commission carefully considered the implications of this case law for its liberalization strategy, particularly because Almelo specifically addressed trade in electricity. An official from the Competition Directorate General explicitly recognized the political controversy surrounding the issue, chiding the ECJ for "hiding behind the national court."7 A legal adviser from the Directorate General administering European electricity policy anticipated that the ECJ's "leniency," in this instance, in not denouncing a trade restriction outright, derived from the restriction having been imposed by one undertaking on another, rather than being imposed by national legislation itself. An individual restrictive agreement between 5· Procedure penale v Paul Corbeau C-3>w/g1, ( 1993) European Court Reports (E.C.R.), l-2533; Gemeente Almew and Others v Energiebedrijf Ijsselmij NV C-393/ 92, ( 1994) E.C.R., I-14 77. 6. Article 86,2 (go, 2 EEC). 7· Interview at Commission Directorate General IV, Brussels, 30 May I99.'i·

European Law and What States Made of It

125

undertakings would be less threatening to the operation of the internal market than blanket restrictions imposed by Member States on all undertakings.8 Regardless of the ECJ's precise motives, these rulings reasserted the possibility of invoking exceptions to competition rules, providing opponents of liberalization with a legitimate legal claim to counter those who wanted to introduce competition. Yet, at the same time that the ECJ resurrected an escape clause within the treaty, it confirmed that electricity is a "good" in Almelo. As a result, the electricity sector is subject to Article 31 (37 EEC), which requires Member States to abolish restrictions on the free import of goods that compete with state monopolies of a commercial character. 9 In verifYing that electricity is a good, the ECJ put prohibitions against import restrictions on the strongest possible legal basis. By contrast, liberalization in telecommunications "services" lacked the explicit obligations present for goods under Article 31 (37 EEC). In contrast to the indeterminate reading of Article 86, 2 (go,2 EEC), this application of Article 31 (3 7 EEC) fueled optimism among supporters of liberalization. A classic example of judicial decision-making, the ECJ handed down a decision that allowed all parties to claim victory despite their fundamentally incompatible positions on electricity liberalization. The European Commission claimed that the Almelo ruling suggested that the introduction of a certain degree of competition in the electricity sector could be compatible with public-service obligations. 10 The French, who opposed the European Commission's proposals for electricity liberalization, praised Almelo as a balanced and pragmatic interpretation of Article 86 (go EEC) that respected the principle of subsidiarity enshrined in the Treaty on European Union (TEU). 11 Although the ECJ managed to satisfy opposing sides of a debate, its ruling did nothing to settle a fundamental policy dispute about the appropriate scope of liberalization in public-service sectors. And, by directing national courts to balance competing treaty goals, in the absence of specific criteria to determine whether particular restrictions on trade and competition are proportionate to their public-service goals, the ECJ did nothing to stop the emergence of divergent outcomes across Member States. Instead, the ECJ left the precise substance of policy reform to others. While disagreements about electricity liberalization persisted, the Euro8. Interview at Commission Directorate General XVII-A, Brussels, 14 September 1995· 9· Pubblico Ministero v Manghera C-59/ 75, ( 1976) E.C.R., 91. 10. European Commission Spokesman's Service, "LaCour de Justice se prononce sur l'applicabilite des regles de concurrence au secteur de l'e!ectricite," Brussels (29 April1994a). 11. Interviews at Direction du Gaz, de l'Electricite, et du Charbon, Paris, 21 February 1gg6; Secretariat General de Comite Interministeriel pour les Questions de Cooperation Economique Europeenne (SGCI), Paris, 18 October 1995·

126

Justice Contained

pean Commission and Member States continued to recognize the legal status of supranational regulatory authority under Article 86 (go EEC). Even those Member States most opposed to electricity liberalization under Article 86 (go EEC) recognized the European Commission's formal competence to issue Article 86 (go EEC) directives to open the electricity sector. French officials conceded that "juridiquement, c'est claire, c'est possible."12 Yet in contrast to the telecommunications case, legal recognition coincided with overt political hostility to the extensive liberalization of electricity markets. As a result, the unilateral supranational approach under Article 86 (go EEC) receded into the background of negotiations over reforms. Although periodically invoked as a threat, the European Commission ultimately refrained from using the legally justifiable procedure, which had become a fundamentally illegitimate mechanism to force change in the absence of consensus.

Appeasing the Malcontents The European Commission's efforts to liberalize the electricity sector under Article 86 (go EEC) inspired an overwhelmingly negative political response. The policy process that the European Commission initiated in this field confirms the extent to which major reforms depend on political consensus and institutional cooperation. Highly variable commitments to liberalization in the electricity field paralyzed the potential utility of Article 86 (go EEC), and a concentrated bloc of opposition to particular reforms protected many elements of electricity provision from competition. A prolonged period of debate and substantial dilution of the competition regime for electricity markets demonstrate that national governments are not powerless in the wake of European judicial review. The following chronology of the policy process in electricity liberalization reveals that legal mechanisms cannot necessarily resolve serious policy disputes. Inspired by the success of its strategy in telecommunications, the European Commission targeted parallel fields of public service for liberalization under Article 86 (go EEC). One month after the ECJ confirmed the European Commission's competence to issue telecommunications directives under Article 86 (go EEC) in March 1gg1,13 Competition Commissioner Sir Leon Brittan suggested that the European Commission would use the same 12. "Legally, it's clear, it's possible," interviews at Direction du Gaz, de l'Electricite, et du Charbon (DIGEC), Paris, 21 February 1996; Direction Generale des Postes et Telecoms, Paris, 22 February 1gg6. 13. Francev CommissionC-202/88, (1991) E.C.R., l-1223.

European Law and What States Made of It

127

Table 5.1. Evolution of Legal Obligations under Articles 86 and 31 (90 and 37 EEC) Decisions of Supranational Institutions

Emphasis of European Legal Obligations

ECJ case law 1974-1993

Exemptions from competition rules are exceptional and not readily accepted. Exemptions require the application of competition rules to be incompatible with the performance of tasks serving the general economic interest. The European Commission may issue its own directives to ensure the application of competition rules. The European Commission will open electricity markets to competition with its own unilateral legislation.

European Commission circulates a draft directive under Article 86 (90 EEC), July 1991 European Commission initiates infringement proceedings against import and export monopolies in electricity, August 1991 European Commission withdraws its draft directive under Article 86 (90 EEC), October 1991

Corbeau ruling, 1993

Almelo ruling, 1994

The European Commission will enforce competition rules in the electricity sector, prosecuting restrictions on trade that violate treaty obi igations. The European Commission acknowledges a fundamental lack of consensus to liberalize electricity markets and initiates a consultative, intergovernmental process of negotiations. Restrictions of competition are justifiable to ensure the achievement of public-service obligations. National courts must consider the impact of competition on the financial equilibrium of enterprises with exclusive rights that also perform services in the general economic interest. Restrictions of competition are justifiable to ensure the achievement of public-service obligations. National courts must consider the economic conditions under which enterprises with exclusive rights operate. Electricity is a good, which requires state monopolies to accommodate import competition.

legal basis to introduce competition in the electricity sector. 14 During June 1991, Energy Commissioner Cardoso Cunha expressed a firm commitment to develop an internal market in electricity15 and proclaimed that commercial relations between electricity producers and distributors would not be left to the exclusive discretion of the Member States. 16 Following through on these announcements, the European Commission 14. European Commission Spokesman's Service, "Competition in the European Community's Electricity and Gas Markets," speech given by Commissioner Sir Leon Brittan at the Institution of Civil Engineers, London, 18 Aprilrgg1. 15. Caroline Monnot, "Le commissaire europeen a l'energie veut liberer Ie marche de l'Hectricite," Le Montie (12june 1991a). 16. Aline Richard, "Guerre de positions entre Bruxelles et les distributeurs d'Hectricite," La Tribune de I'Expansion ( 11 June 1991).

128

Justice Contained

prepared and privately circulated a draft directive under Article 86 (go EEC) for electricity competition during July 19g1 Y The European Commission's draft directive eliminated exclusive rights for electricity generation, imports and exports, the construction of transport or distribution lines, and connections with the transmission and distribution systems. It also imposed third-party access (TPA) rights on electricity grids, suppressing all special or exclusive rights for the utilization of individual grids. 18 Supplementing this supranational legislative approach, the European Commission began its assault against particular import and export monopolies in August 1991, initiating formal infringement proceedings against Belgium, Denmark, France, Greece, Ireland, Italy, the Netherlands, Spain, and the United Kingdom. 19 Susanne Schmidt argues that the European Commission adopted this unilateral supranational approach precisely in order to bypass the opposition to liberalization that it anticipated in the electricity sector. 20 The effort to neutralize resistance failed, however, as these actions immediately inspired an uproar from electricity suppliers, most Member States, and a majority of the European Parliament. The Association of European Electricity Producers and Distributors (UNIPEDE) considered Commissioner Cunha's statements to be a "declaration of war," and the European Grouping of the Electricity Supply Industry (Eurelectric) demanded the initiation of a democratic process of reform to include consultation with affected parties. 21 Overwhelming opposition to the European Commission's use of Article 86 (go EEC) in the Council of Ministers and European Parliament seriously threatened the European Commission's capacity to act: legally unable to stop an electricity directive under Article 86 (go EEC), the Council could nonetheless stall all proposals for Council legislation, and the European Parliament could censure the European Com17 . .July 1991 draft directive on "Completion of the Internal Market in Electricity and Gas." 18. "Contents of the Directive of DC-IV on the Basis of Article 90.3 of the Treaty," Internal consultation briefing, Madrid (16 July 1991 ); Commission, "First Draft Electricity Directive," Brussels (30 June 1991); M.Jacques Oudin, Senateur, "Electricite et gaz: pour un marche interieur respectuex du service public." Rapport d'Infurmation, no. 459 (1993-1994): s-6; Henri Revol, Senateur, "Marche interieur de I' electricite et du gaz nature!: quelle politique pour Ia France?" Rapport d1nformation, no. 491 (1993-1994): 37; interviews at Commission Legal Service, telephone interview with Brussels, 18 March 1996; Direction du gaz, de l'electricite, et du charbon (DIGEC), Paris, 21 February 1996. 19. Commission, "Tenth Annual Report to the European Parliament on Commission Monitoring of the Application of Community Law-1992 ," Official journal iew at Commission Directorate General XVII-A, Brussels, 14 September 1995. 51. Commission 1997, C-332, 49· 52. Interviews at Commission Directorate General IV-C, Brussels, 14 September 1995; Commission Directorate General XVII-C, Brussels, 14 September 1995; Commission Directorate General XVII-A, Brussels, 14 September 1995; Permanent Representation of the U.K, Brussels, 28 February 1996; Direction du gaz, de l'electricite, et du charbon (DIGEC), Paris, 21 February 1996; Bundesministerium fiir Wirtschaft, Bonn, 18 January 1996; European Grouping of the Electricity Supply Industry (Eurelectric), Brussels, 27 September 1995.

134

Justice Contained

mise, the internal market directive under Article g5 ( 1ooa EC) also accommodates all major French demands: Member States are free to organize national electricity provision according the Single-Buyer model and may deny public distributors the status of eligible consumers with TPA rights. These provisions, along with permission to regulate prices for customers located in particular areas, 53 enable the French to maintain their tradition of public service in the electricity sector. Table 5.2 charts the evolution of proposals for electricity liberalization, which reflects the substantial dilution of the regime originally crafted by the European Commission under Article 86 (go EEC) and the congruence of the ultimate outcome with French demands. Despite the relatively limited degree of liberalization required by the internal market directive for electricity, implementation has been problematic nonetheless. France and Luxembourg both failed to transpose the directive into domestic law by the deadline of February 1g, 1ggg, German measures fail to guarantee access for new entrants into the electricity market, and many complaints have reached the European Commission about the lack of fair access to distribution grids. In terms of formal market opening, about two thirds of total EU electricity demand has become subject to competition, where large industrial users have a choice of suppliers. Market opening varies substantially across Member States, with a low of 30 percent in France, Greece, and Portugal, to a high of 100 percent in Finland, Germany, Sweden, and the United Kingdom. The opening of markets has not yet produced much of an EU market in electricity, however, given that only 8 percent of the EU's electricity output is currently being traded across borders. 54 Both the diluted regime of competition and its limited realization reflect fundamental policy disputes that have not been particularly amenable to resolution through legal mechanisms. The mobilization of opposition effectively eliminated the European Commission's universally acknowledged competence to legislate unilaterally under Article 86 (go EEC) and forced major concessions on a program ofliberalization theoretically under supranational legal control. Meanwhile, the ECJ shunned the opportunity to resolve underlying substantive conflicts by failing to articulate any clear proportionality tests to balance competing treaty provisions. The European Commission ultimately achieved more from threats to litigate and negotiations with the Council and European Parliament than it did from the ECJ, 53· European Parliament and Council Directive 96/92/EC of 19 December 1996, Official journal of the EurofJean Communities L-27 (30 January 1997): 20. 54· Commission, "Seventeenth Annual Report on Monitoring the Application of Community Law-1999," COM (2ooo) 92 final (23]une 2000), introduction, 50; Daniel Dombey and David Buchan, "Energy Liberalisation Leaves EU Members Divided," Financial Times (14 March 2001), 2.

European Law and What States Made of It

135

Table 5.2. The Evolution of Reform Proposals for Electricity Liberalization

Legislative Proposals and Outcomes Draft directive under Article 86 (90 EEC), July 1991; European Commission abandons, October 1991 First proposal for a directive under Article 95 (1 ooa EC), February 1992

Generation, Import, and Export of Electricity Open to Competition?

Third-party Access Rights to Networks Distributing Electricity

Single Buyer of Electricity an Accepted Means to Regulate Competition?

Yes

Regulated access for large industrial users and distributors serving residential consumers Regulated access for large industrial users and distributors serving residential consumers Negotiated access for large industrial users and distributors serving residential consumers

No

Negotiated access for large industrial users only Negotiated access for large industrial users only

Yes

Yes

Second proposal for a directive under Article 95 (1 ooa EC), December 1993

Yes

French Single-buyer Proposal, May 1994

Single-buyer purchases electricity from competing providers Yes, but Single-buyer model is permitted as well

Directive 96/92 EC under Article 95 (100a EC), December 1996

No

No

Yes

which dismissed enforcement actions against import and export monopolies in France, Italy, the Netherlands, and Spain as ill-founded in 1997. 55 In these cases, the ECJ indicated its unwillingness to impose any definitive settlement by further relaxing requirements to justify exclusive rights and demanding proof from the European Commission that competition and trade would be viable. Extending the logic from Corbeau and Almelo, the Eq argued that exceptions to treaty rules are sufficiently justified in the case of an undertaking granted exclusive rights if treaty rules would obstruct the performance of services provided in the general economic interest, ''without its being necessary for the survival of the undertaking itself to be under threat." The ECJ went on to accept the idea that a Member State may justify restrictions on the basis of evidence that the performance of services in the general economic interest under "economically acceptable 55· Commission v France C-159/94, (1997) E.C.R., I-5815; Commission v Italy C-158/94, (1997) E.C.R., I-.~789; Commission v NetherlandsC-157/94, (1997) E.C.R., I-5699; Commission v SpainC-160/94, (1997) E.C.R., I-5851.

136

Justice Contained

conditions" would be jeopardized by the elimination of exemptions. The burden of proof could not be so extensive as to require a Member State to "prove, positively, that no other conceivable measure, which by definition would be hypothetical, could enable those tasks to be performed under the same conditions." The ECJ found that the European Commission had not provided evidence to prove that the maintenance of exclusive rights was not justifiable. Legal argumentation about the inapplicability of Article 86,2 (go,2 EEC) to the cases was insufficient because "it is not for the Court, on the basis of observations of a general nature, to undertake an assessment, necessarily extending to economic, financial and social matters, of the means which a Member State might adopt in order to ensure that the country is supplied with electricity on the basis of costs which are as low as possible and in a socially responsible manner." Furthermore, the European Commission needed to "show how, in the absence of a common policy in the area concerned, development of direct trade between producers and consumers, in parallel with the development of trade between major networks, would have been possible having regard in particular to the existing transmission and distribution capacities and facilities. "56 The following section explains why the pattern of interests in this sector led to such a decisive mobilization of opposition and concludes that the uncertainties associated with litigation ultimately provided much more leverage for policy change than the actual imposition of any legal obligations by supranational institutions.

Interests, Institutions, and Mobilization against Supranational Reform of Electricity The process of reform in electricity involved enduring opposition to the supranational application of European legal obligations. EU Member States historically pursued independent energy policies based on national regulation, which was similar to the traditional situation in telecommunications. European initiatives in electricity threatened national autonomy in a sector that has important implications for national security, industrial production, and social welfare. Cooperation had eluded Member States during the energy crisis of the 1970s, and most national governments did not initially welcome the European Commission's efforts to include electricity within the scope of the internal market. Parallel to the early reaction to European initiatives in telecommunications, the United Kingdom was among the few Member States to support the European Commission initially. 56. Commission v Netherlands C-1 ii7 /94, 1997.

European Law and What States Made of It

137

France had also pushed early plans to open access to external electricity networks as a means to export French electricity to other Member States. However, the French soon abandoned their support when they realized changes would affect electricity distribution within France. 57 The application of the European competition regime became controversial once again, this time because the liberalization of European electricity provision invited a face down with many enterprises that preferred exclusive provision in captive markets to expansion into new markets. As in telecommunications, industrial restructuring associated with the introduction of competition to electricity threatened the job security of well-organized public-sector workers. Therefore, the European Commission's effort to extend the European competition regime to electricity also inspired conflict among concentrated, resourceful actors with intense interests. National governments, electricity providers with exclusive rights, corporate consumers, aspiring entrants, and labor unions responded to European initiatives with competing demands. More diffuse interests, such as residential consumers, were again absent from policy debates. Yet in contrast to the telecommunications sector, structural features of the electricity sector generated a strong base of opposition to liberalization of electricity. In this section, I specify the policy preferences of contending actors and relate the mobilization of these competing interests to the regulatory outcome in this field. The .t'volution of Interests Related to Electricity Liberalization

Cleavages between supporters and opponents of liberalization characterize the electricity sector, but a more variable specificity of assets, higher degree of certainty about policy consequences, and more cohesive set of preferences among managers and organized labor in segments of this industry concentrated and intensified opposition to liberalization. In contrast to the telecommunications case, the evolution of interests in the electricity debate generally empowered opponents ofliberalization over supporters. First, the variable specificity of assets in the electricity sector led to diverse interests among producers. Electricity can be generated from multiple sources, each of which requires different levels of initial investment and reliable access to particular resources to be profitable. For instance, the assets necessary to produce nuclear energy are highly specific and require massive initial investments. Once operational, nuclear plants generate a continuous, high volume of electricity. Authoritative control of a large captive market provides an efficient means to guarantee returns on previous 57· Schmidt 1997, 246-47.

138

Justice Contained

investments. By contrast, gas turbine technology, which produces a rapid return on investments, is compatible with a competitive market system that may involve fluctuations in demand. Not all modes of electricity generation are universally accessible due to the unequal distribution of particular resources: hydroelectric plants are not viable without rivers, and scarcity in fossil fuels renders some modes of generation riskier than others. Technological developments in the electricity sector that led to efficiency improvements and new modes of generation stimulated cleavages between holders of more and less specific asset'>. Meanwhile, a single network of distribution has continued to provide the most efficient means to transmit electricity. In contrast to the situation in telecommunications, attempts to bypass existing electricity grids are prohibitively costly on a large scale. This distinction is important because the capacity for physical control over a single distribution network enables existing providers to restrict access to particular markets. As a result, those who anticipate losses on previous investment in highly specific assets can mobilize more effective resistance to the introduction of market mechanisms. So, while the variable specificity of assets to generate electricity leads to different interests relative to liberalization, the continued high specificity of assets to distribute electricity empowers enterprises that control networks of transmission. Second, greater certainty about the negative consequences of liberalization intensified opposition to reforms. Among producers, identifying losers in a competitive regime has been relatively easier in the electricity sector than it was in telecommunications. Unlike the telecommunications sector, in which goods and services have proliferated, the electricity sector continues to produce a single commodity, whose price is subject to downward pressures from efficiency improvements. As a result, losses in market share in this exclusive commodity cannot be compensated by gains in market share in related services or products. Without new niches to exploit within the industry, less competitive electricity enterprises and their workers can project more permanent losses. Among consumers, the pattern of cross-subsidization in the electricity sector threatened to shift costs from a diffuse set of residential consumers to more concentrated industrial consumers. High residential electricity rates have traditionally subsidized industrial rates or funded particular public services. This stmcture of cross-subsidization is the inverse to the situation in telecommunications, where long-distance revenues from corporate consumers traditionally subsidized local rates for residential consumers. The introduction of market competition eliminates subsidies, benefiting whoever underwrote the costs of these subsidies. In electricity, more con-

European Law and What States Made of It

139

centrated actors such as industrial enterprises and public agencies would bear the burden while more diffuse actors such as household consumers would benefit from lower electricity rates. A high degree of certainty about these losses intensified opposition to reforms among concentrated actors. Third, a more cohesive set of preferences among managers and organized labor in segments of the electricity sector prevented the marginalization of organized labor as a factor in the process of electricity reform. In contrast to telecommunications, technological changes were not leading to universal pressures for a more flexible labor force within the management of electricity providers. Technological changes that offered new means of generating electricity were only relevant to enterprises with reliable access to the resources necessary for these technologies, for instance, natural gas. For enterprises generating and distributing electricity with highly specific assets, the interests in sustaining existing practices remained consistent among the management and labor force. Cross-national variations in the specificity of assets associated with electricity provision, expectations about policy consequences, and position of organized labor generated a highly conflictual policy debate over electricity liberalization. In France, the coincidence of highly specific assets, negative expectations about the European Commission's program of liberalization, and privileged position of labor in this sector, created a particularly concentrated and intense source of opposition to full-scale liberalization. The following discussion specifies the particular interests of actors in the United Kingdom, Germany, and France. Once again, the United Kingdom was in the process ofliberalizing its domestic electricity market when the European Commission initiated plans for competition in European electricity provision. The 1g8g Electricity Act established the foundations for the privatization of the nationalized industry, introducing competition into electricity generation and sales to final consumers. The United Kingdom introduced a system of independent regulation to ensure access to the network for transmission and distribution activities.'" The British government supported the creation of an internal market in electricity from the beginning, arguing that the absence of any derogation for electricity subjects the field to treaty rules. 59 Domestic natural gas reserves had enabled U.K. providers to take advantage of new gas turbine technologies that are efficient and clean, without any vulnerability associated with trade disruptions. This opportunity, along with a generally 58. Overseas Relations Service, "The British Electricity System" (London: Electricity Association Services, 1994), 2. 59· Interview at Permanent Representation of the U.K., Brussels, 28 February 1gg6.

140

Justice Contained

diversified base of electricity generation, made the United Kingdom an ideal site for a market-based regime.uo The German government was initially opposed to the liberalization of its electricity market. Germany had granted the electricity sector exemptions from its domestic competition law since 1957. 61 German electricity suppliers include private and public undertakings at the national, regional, and municipal level. National law had historically enabled these providers to agree to geographically exclusive rights among themselves. Local communities also conferred exclusive distribution rights to these firms in exchange for financial concessions, which have funded a variety of public services at the municipal and regionallevel. 62 However, the German position shifted by 1994, when the government led by Helmut Kohl proposed the elimination of competition exemptions for the electricity sector. 63 This change reflected shifts in the preferences of Germany's largest national electricity providers as well as a growing concern with high industrial electricity prices and the incentives that these prices generated for relocation among energy-intensive industries. 64 Located in a state where electricity prices have been three times the world average, German industrial users paid the highest rates for electricity in Europe. 65 By 1994, the official German position on an internal market broadly corresponded to the proposals of the European Commission: Germany supported the system of negotiated TPA for eligible consumers including large industry and local distributors. Despite the early French interest in gaining access to external grids to export its own electricity to other Member States, France consistently opposed the program of electricity liberalization advocated by the European Commission. Structural features of the French electricity sector inspired particularly intense and cohesive political opposition to a market-based system of provision. The French had developed a nuclear program to accommodate national electricity requirements because they lacked natural resources for

fio. Overseas Relations Service, 1; interviews at Commission Directorate General XVII-C, Brussels, 14 September 1995; Permanent Representation of the U.K, Brussels, 28 February 1996; European Grouping of the Electricity Supply Industry (Eurelectric), Brussels, 27 September 1995· 61. Article 103, Gesetzgegen Wettbewerbsbeschriinkungen (27 July 1957). 62. Gerhard Holm, "Struktur der deut,chen Elektrizitiitswirtschaft," Strombasiswissen, no. 115 (Frankfurt: Bender and Kelkheim, 1995); Joachim Grawe, "Zentrale und dezentrale Energieversorgung," Elektrizitiitswirtschaft, no. 2 2 ( 1990): 1207- dl. 63. Gesetz zur Neureglung des Energiewirtschaftsrechts, Bonn, 15 February 1994· 64. Interviews at Bundesministerium fiir Wirtschaft. Bonn, 18January 1996; Commission Directorate General XVII-C, Brussels, 14 September 1995. 65. Eurostat, Statistics in Focus: Energy and Electricity, no. 8 (Luxembourg: Office for Official Publications of the European Communities, 1995); interview at Vereinigung Deutscher Elektrizitiitswerke, Bonn, 19 January 1996.

European Law and What States Made of It

141

electricity generation. Attaining self-sufficiency in electricity through their large nuclear program has required extensive, long-term investments. To recoup these sunk costs, French nuclear plants have needed a large market for the enormous volume of electricity they produce. And because electricity cannot be stored, a captive market guarantees relative stability in demand. L'Electricite de France (EDF), a public enterprise created through a bill of nationalization in 1946, operates the nuclear program that has nearly monopolized the French electricity sector. The French state has controlled the organization of the system: EDF has held a total monopoly for the transport of electricity over the state-owned network, and EDF has produced and distributed 95 percent of all French electricity (So percent of which is generated by nuclear plants). The remaining 5 percent has been generated and distributed by municipal systems that existed prior to the 1946 nationalization program and independent producers who generally consume all of the electricity they generate. 66 The magnitude of sunk costs in the nuclear sector and a continued desire to avoid dependence on foreign suppliers offossil fuels precluded French interests in technologies that facilitate the introduction of market mechanisms in the electricity sector, for instance, turbines that run on natural gas. Active resistance to complete liberalization has derived from the French desire to avoid disruptions in demand that could reduce the viability of the nuclear program. 67 As a result, the French government persistently opposed TPA rights for electricity distribution, with the exception of rights for large industrial users. 66 The EDF's control of over 95 percent of distribution to communities has facilitated the long-term planning necessary for nuclear plants. Moreover, the virtual distribution monopoly allows the French to maintain their distinctive tradition of service public. The French concept of public service involves a system of uniform prices for service in all areas of the country, whether urban or rural. Official French commitment to its tradition of public service has remained strong: in its public report of 1 994, the Conseil d'Etat defended the French tradition of public service as an appropriate response to market failure and denounced what it viewed as an extreme European conception of competition that makes inadequate distinc66. Interviews at Direction du Gaz, de l'Electricite, et du Charbon (DIGEC), Paris, 21 February 1996. 67. Interviews at Foreign Ministry of France, Paris, 21 February 1gg6; Direction du Gaz, de l'Electricite, et du Charbon (DIGEC), Paris, 21 February 1996; SGCI, Paris, 18 October 1995; European Grouping of the Electricity Supply Industry (Eurelectric), Brussels, 27 September 1995; Commission Directorate General XVII-C, Brussels, 14 September 1995· 68. Direction du Gaz, de l'Electricite, et du Charbon 1994; interviews at Direction du Gaz, de l'Electricite, et du Charbon (DIGEC), Paris, 21 February 1996; SGCI, Paris, 18 October 1 995·

142

Justice Contained

tions between the private and public realms. 69 Reports and resolutions of the French Senate and National Assembly also reflected resolute opposition to the end of French public-service obligations (i.e., including nationally uniform prices) in the electricity sector. 70 The French have remained relatively resistant to the notion that the market necessarily outperforms authoritative control in many economic sectors. Unlike many other EU Member States, the quality of service provided by public enterprises has remained quite high in France. Multiple observers contributed scathing critiques of "Europe's" inability to understand and respect the French tradition of public service in an issue of Le Monde released as the debate over electricity reform raged. 71 Indeed, concerns over the future of French public service in Europe spawned a major cottage industry among official and academic policy analysts. 72 The positions of the British, German, and French governments have been largely consistent with the preferences of concentrated, resourceful domestic actors who have held intense interests relative to the organization of electricity provision. British electricity producers, who have been competing for market share in a liberalized regime, saw exports to Europe as their major opportunity for expansion. As a result, British electricity producers strongly supported the creation of an internal electricity market. 73 Electricity producers in most other Member States, who have largely enjoyed captive markets, expressed little to no interest in exporting electricity beyond the limited border area exchanges that predated reform efforts. Unlike the situation in telecommunications, the electricity market has not been attracting many aspiring entrants because initial investment demands are much higher and growth projections do not approximate those in 69. Conseil d'Etat, Rapport Public I 994, no. 46 (Paris: La Documentation fran