Downloaded from https://academic.oup.

com/book/5362 by guest on 01 May 2025

 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

T H E E U R O P E A N CO U R T O F J U S T I C E
AND THE POLICY PROCESS

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

 Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

The European Court

of Justice and the
Policy Process

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

The Shadow of Case Law

SUSANNE K. SCHMIDT

1
 3
Great Clarendon Street, Oxford, OX2 6DP,
United Kingdom
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide. Oxford is a registered trade mark of
Oxford University Press in the UK and in certain other countries
© Susanne K. Schmidt 2018
The moral rights of the author have been asserted
First Edition published in 2018
Impression: 1
Some rights reserved. No part of this publication may be reproduced, stored in

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

a retrieval system, or transmitted, in any form or by any means, for commercial purposes,
without the prior permission in writing of Oxford University Press, or as expressly
permitted by law, by licence or under terms agreed with the appropriate
reprographics rights organization.

This is an open access publication, available online and distributed under the terms of a
Creative Commons Attribution – Non Commercial – No Derivatives 4.0
International licence (CC BY-NC-ND 4.0), a copy of which is available at
http://creativecommons.org/licenses/by-nc-nd/4.0/.
For any use not expressly allowed in the CC BY-NC-ND licence terms, please contact
the publisher at the address above.
Published in the United States of America by Oxford University Press
198 Madison Avenue, New York, NY 10016, United States of America
British Library Cataloguing in Publication Data
Data available
Library of Congress Control Number: 2017946417
ISBN 978–0–19–871777–5
Printed and bound by
CPI Group (UK) Ltd, Croydon, CR0 4YY
Links to third party websites are provided by Oxford in good faith and
for information only. Oxford disclaims any responsibility for the materials
contained in any third party website referenced in this work.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

For Torsten, Jasper, and Arthur

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

 Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Acknowledgements

This book consolidates work I have conducted over many years, dating back to
my time at the Max Planck Institute for the Study of Societies in Cologne,
where I worked until 2005. The results of several projects feed into this text:
‘The Domestic Impact of European Law’, funded from 2004 to 2007 by the
European Commission’s Sixth Framework Programme in the NEWGOV
consortium, coordinated by the European University Institute in Florence;
project A6 of the Collaborative Research Centre 597, which was funded by the

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

German Research Foundation (DFG) from 2008 to 2014; and the project
TransJudFare, funded by NORFACE from 2015 to 2018. I am very grateful
for the financial support I have received from the German Research Founda-
tion and the European Commission, and I would like to thank all the interview
partners who shared their time and insights. Michael Blauberger, Wendelmoet
van den Nouland, Julia Spreen, Björn Schreinermacher, Rike Krämer, Tilman
Krüger, Daniel Seikel, Angelika Schenk, and Benjamin Werner have collabo-
rated or are currently collaborating with me on these projects, which has been
a great pleasure. For this book, I was able to draw on the interviews that Björn
Schreinermacher conducted for his Ph.D. dissertation in the context of project
A6. My close collaboration with Michael across all these projects has been a
particular joy. Different student assistants have provided support over the
years; among them, I would like to mention Daniel Kosak, Sascha Göbel,
Torben Schmengler, Martin Losch, Mario Dentice, Lukas Antoine, Pauline
Anton, and Lea Feldhaus. Britta Plote provided language support and is a great
secretary in all respects. I was very fortunate to be able to draw on the support
of Ben Robbins for the copy-editing of the final text.
Working on a topic for such a long time means that numerous colleagues
have provided me with feedback over the years. Drawing on work which has
been published over a number of years implies that parts have been presented
on many occasions. I am sorry that I cannot acknowledge all the helpful
comments that I have received. Where I rely on previously published work,
these texts have been substantially revised and updated. Without a doubt, my
greatest intellectual debt is to Fritz Scharpf. I am extremely grateful for all his
encouraging comments, his support of my work since the 1990s, and our
many conversations throughout my career. All scholars who have met him can
appreciate how fortunate I have been. Numerous others have helped me over
the years to sharpen my thinking about the way the European Court of Justice
impacts policymaking at the European and member-state levels. Among them
are: Dorte Martinsen, Gareth Davies, Martin Höpner, Adrienne Héritier, Dan
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

viii Acknowledgements
Kelemen, Christian Joerges, Damian Chalmers, Miriam Hartlapp, Cliff
Carrubba, Marie-Pierre Granger, Sabine Saurugger, Gerda Falkner, and
Michelle Egan. Philipp Genschel has been a close colleague and friend with
whom I have had inspiring conversations both in Cologne and Bremen.
I would also like to thank my colleagues in Bremen, particularly Peter
Mayer, Philip Manow, and Kerstin Martens, for making university life as
good as it gets. The InIIS, Bremen’s Institut für Interkulturelle und Inter-
nationale Studien, provides me with a very agreeable working environment.
Special thanks go to Stephan Leibfried for his support of this book project.
Finally, I have been very lucky that several colleagues kindly gave me
detailed feedback on the manuscript: Adrienne Héritier, Michael Blauberger,
Dan Kelemen, Martin Höpner, Daniel Naurin, Benjamin Werner, Angelika

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Schenk, Fritz Scharpf, and Hans Micklitz. Their comments and enthusiasm
were a great help in enabling me to manage the final months of the project.
I would also like to thank Dominic Byatt and Olivia Wells from Oxford
University Press, Lakshmanan Sethuraman and Joshua Hey for their very
careful and professional support in the production process, as well as the
three anonymous reviewers, who gave me very helpful comments on the book
proposal and the final manuscript.
Combining work on a long manuscript with university and family life is not
always easy. I am grateful for the support of my family, my parents, Wilma
and Horst Schmidt, and my parents-in-law, Elisabeth and Joachim Siol. My
friends Annette, Karen, Annette, and Carolin have helped me to balance the
exercise of writing a book with running and walking in the park. Most of all,
I am indebted to my husband Torsten, and my sons Jasper and Arthur, to
whom I dedicate this book. I wouldn’t be where I am today without them!
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Contents

List of Figures xiii

List of Tables xv

1. Introduction 1
The Book’s Argument 4
A Skewed Balance of Powers 6
Understanding Case-Law Development 7
Case-Law Constraints on Policymaking 10

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Case Selection and Methodology 13
The Book’s Structure 17
2. The European Court of Justice as a Political Actor 23
Courts as Political Actors 24
The ECJ as a Motor of Integration 26
The Court, its Judges, and its Professional Environment 28
The Commission 34
The ECJ’s Supporting Coalition: Member-State Courts
and Private Litigants 36
The Court and the Masters of the Treaty 39
Conclusion: Shortcomings in the Literature 47
3. Case-Law Development between Path Dependence and Legal
Uncertainty 50
The Distinctive Approach towards Integration in the EU 52
Legal Uncertainty 54
Legal Texts as a Source of Legal Uncertainty 55
Adjudication as a Source of Legal Uncertainty 56
The Multilevel System as a Source of Legal Uncertainty 59
Applying the Concept of Path Dependency to Case Law 60
Path-Dependent Jurisprudence on the Fundamental Freedoms 63
The Transfer of Logic to Other Freedoms 69
Services 70
Establishment 72
The Free Movement of Workers 74
Capital 77
Citizenship 78
A Unified Restrictions Approach 82
The Evolving Interpretation of Case Law 85
Conclusion 89
4. The Interaction of Judicial and Legislative Policymaking 93
Negotiating Legislation in the Shadow of the Court 95
Pressurizing the Legislative Process: Strategies of the Commission 100
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

x Contents
Making an Offer that Can’t be Refused: The Lesser-Evil Strategy 100
Breaking the Resistance in the Council: Divide and Conquer 103
The Services Directive 106
The Content of the Directive 107
Negotiating the Services Directive 109
What we can Learn from the Directive 113
The Regulation for Goods 114
The Content of the Regulation 117
The Burden of Proof in the Case Law 118
Explaining Member States’ Consent 120
Summarizing the Argument 122
Conclusion: Is the Judiciary Overshadowing the Legislature? 123
5. Reaching Beyond the Market into State Responsibilities 126

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

The Citizenship Directive 127
The Historical Origins of Citizenship Rights 128
Case-Law Development 131
Negotiating the Citizenship Directive (2004/38/EC) 132
Case-Law Development After the Directive 136
Conclusion 139
The Patient Mobility Directive 141
Existing Case Law on Patient Mobility 142
Negotiating the Patient Mobility Directive 145
Conclusion 148
Failed Codification 150
The Failed Monti II Regulation 151
Integrating Member States’ Tax Policies 153
Litigating Direct Domestic Taxes at the ECJ 154
An Impossible Quest? 158
The Court’s Response—Switching to a Lower Gear 161
Legislative Responses at the EU Level 163
Conclusion: No Exit from the Joint-Decision Trap 165
Conclusion 166
6. Europeanization With and Against the Odds: The Cases of
Meilicke and Zambrano 169
Research on Europeanization 171
The Meilicke cases 175
Meilicke 175
The Facts Concerning Meilicke I 176
The Court’s Ruling 178
Meilicke II 178
Assessing Meilicke 179
The Court’s Activism on Citizenship Rights 180
The Zambrano Case 181
Fine-Tuning Zambrano 182
Member States’ Reactions 185
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Contents xi
The Irish Reaction 186
Citizenship Law in Ireland 186
Explaining Proactivity 188
The Reaction in Other Member States 190
Conclusion 194
7. The Europeanization Effects of Case Law 197
Explaining Member States’ Reactions to Case Law 199
The Cognitive Dimension of Responses to Case Law 201
Actors’ Reactions 201
Executive Reactions 202
Administrative Accommodation 202
Executive Resistance 206
Executive Compensation 209

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Legislative Accommodation 212
The Reform of Limited Companies 212
The Posting of Workers 214
The Laval Quartet 215
Reform of Social Benefits in the United Kingdom 216
Judicial Reactions 219
Domestic Judicial Activism 220
Domestic Judicial Restraint 222
Societal Actors 224
Conclusion 226
8. Conclusion 231
Activism? 236
Time 240
Legitimacy 241
The Rule of Law beyond the Nation State 246
Where Should Political Science Go from Here? 248

Bibliography 253
Index 283
 Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

List of Figures

1.1. The book’s argument 12

2.1. Weak courts 44
2.2. The strength of the Court 47
3.1. Different sources of legal uncertainty 59
3.2. The non-discrimination and non-restriction approaches 66
3.3. Legal uncertainty and path dependence 69

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

3.4. Completed legal cases according to freedom, 1958–2012 70
3.5. Several freedoms in a ruling, 1990–2014 84
4.1. Lesser evil 102
5.1. Infringement cases and preliminary rulings on direct
taxation (by year) 160
5.2. Infringement cases and preliminary rulings on direct taxation
(by member state) 160
6.1. The blind spot of research on Europeanization 174
7.1. Direct and indirect responses to case law 200
 Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

List of Tables

1.1. Legal uncertainty in an emerging multilevel polity 8

4.1. Types of default condition and bargaining power in European legislation 96
4.2. The interrelationship between judicial and legislative policymaking 99
7.1. Overview of responses at the member-state level to direct
and indirect pressures 227

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

 Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Introduction

The European Court of Justice (ECJ)1 established itself very early on as an

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

important actor in the European integration process. Institutionally, the ECJ
created the direct effect and supremacy of European law in its rulings in Case
26/62 Van Gend en Loos in 1963 and in Case 6/64 Costa v ENEL in 19642 which
de facto constitutionalized European law (Shapiro 1992: 126; Stein 1981; Weiler
1991: 2415). Yet the consequences of these steps are still largely unrecognized.
Implicitly, the Treaty Establishing the European Economic Community
(EEC Treaty/TEEC) received constitutional status as European law overrules
national law. Because of this move, the European Union (EU) and its member
states operate under a constitution that is very different from those we know
for nation states. The constitutions we know describe state organization and
fundamental individual rights. Also, the EU constitution focuses on ‘state’
organization, describing the rights of the European Commission, the Council
of Ministers, and the like. Yet, instead of individual, liberal rights, which
formally only entered the EU with the Charter of Fundamental Rights in
2009, the EEC Treaty contains ample policy goals, most notably the famous
four freedoms and European competition law. This is unusual, as national
constitutions normally abstain from policy prescriptions. International trea-
ties, in contrast, are agreements on common policy objectives, and the differ-
ent revisions of the EEC Treaty have constantly expanded and refined
these goals, but they have not deleted competences from the Treaty. The EEC
Treaty was renamed Treaty Establishing the European Community (TEC) at
Maastricht (1992). Given many changes, the numbering of the TEC was revised
at Amsterdam (1997). In 2009, the Treaty was renamed the Treaty on the Fun-
ctioning of the European Union (TFEU) at Lisbon, again renumbering articles.
Despite this complexity, I will often simply speak of the ‘Treaty’ in this book,

1
Increasingly, it is common to refer to the Court of Justice of the European Union (CJEU).
However, this new term describes the whole court system of the EU, including the Court of
Justice and the General Court. As this book only analyses the Court of Justice, I stick to the
familiar ECJ.
2
With the Treaty of Lisbon, Declaration 17 now codifies the case law on the supremacy of the
Treaty (Beck 2012: 168).
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

2 The ECJ and the Policy Process

given that the legal matter that interests me most, the four freedoms, has
been part of all these treaties since the founding of the European Economic
Community (EEC).3 In combination with decades of the ECJ’s jurisdiction,
this constitutionalized international treaty implies that policymaking in the
EU is highly constrained by case law. When constitutional courts rule how to
interpret the constitution, their rulings add to the constitution.
Along with the Treaty, the Court’s interpretations of it have constitutional
status. Therefore, member states have to be careful not only to implement EU
secondary law, but also to abide by the ECJ’s interpretations of the Treaty. And
yet, scholarship on the EU largely turns a blind eye to the momentous impact
of the ECJ’s constitutionalized policymaking, wherein, increasingly, policy
options are withdrawn from majoritarian decision-making at the European

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

and national levels.
Different from national constitutions, the treaties are not confined to those
provisions that reflect the functions of a constitution. They are full of provisions
that would be ordinary law in the member states. . . . the constitutionalization of
the treaties, immunizes the Commission and particularly the ECJ against any
attempt by the democratically responsible institutions of the EU to react to the
Court’s jurisprudence by changing the law. (Grimm 2015: 470–1)
Grimm is one of the very few scholars to discuss critically this ‘over-
constitutionalization’ (Grimm 2015: 469). The surprising neglect of this
topic in the scholarship is not due to a lack of debate about the ECJ. The
power of the Court is broadly discussed and largely recognized. However, in
line with EU integration research in general, which appears to be still inter-
ested in the intergovernmentalism/supranationalism debate, the analysis of
the Court often centres on the question of whether the Court is activist and
how far the Court takes into account member states’ preferences. This juxta-
position of member states and the EU ignores the consequences of the
multilevel polity. Member states are not enclosed entities with governments
that have a gatekeeping role in relation to the EU. In contrast, private actors
can address domestic courts, claiming European legal positions when these
appear more advantageous than domestic law. Domestic courts, in fact,
partly favour such European law positions themselves. The European legal
order’s supremacy and direct effect are part of the domestic political strug-
gle, and those finding support in EU law are in a privileged position by being
able to realize their interests. Possible opposition of governments is medi-
ated by the fact that important domestic constituencies support the policies
of European law. Policy change through case law is a slow process, and as
it progresses actors’ preferences change as well. Constitutionalized case law

3
In view of the renumbering, I cannot avoid referring to the different versions of the Treaty,
as citations I draw on will rely on different versions.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Introduction 3
itself, however, is relatively immune to change, unless the ECJ alters its
interpretation of EU law.
European Union research on the ECJ concerns itself mainly with the political
importance of this actor for the process of European integration. In exploring
this issue, however, the ongoing focus on the importance of member states’
interests has captured analytical interest in a debate that risks deciding whether
the glass is half full or half empty. There are good arguments for detecting both
‘supranational’ judicial activism and the ‘intergovernmental’ act of listening to
member states’ preferences in the operation of the Court. A principal-agent
approach can persuade us that the Court as an agent (or a trustee) has a great
deal of leeway as the twenty-eight principals of the members states all now have
to agree when they wish to rein the Court in. At the same time, the Court is what

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

one could call a ‘reactive’ actor (Conant 2012). It has to wait until being called
upon, as it can only decide upon single cases that are brought before it.
Moreover, its rulings are pointless if they are not implemented. Repeated non-
responsiveness to its adjudication undermines a court’s role, making it mean-
ingless for anyone to seek judicial redress there in the first place. All courts are
therefore sensitive to general support for their adjudication. That does not mean
that all rulings conform to the dominant actors’ preferences. But to enjoy
continued support, courts cannot become too controversial.
The debate on the ECJ has focused a lot on actors’ preferences and little on
the policy implications of the constitutionalized EU legal order. An early
exception to this rule is Conant (2002), who discovered the restricted impact
of case law on policymaking at the European and member-state levels. More
recently, most work has focused on showing that the ECJ responds to member
states’ preferences (Martinsen 2015; Larsson and Naurin 2016; Carrubba and
Gabel 2014), countering claims that the Court has a high degree of power (Alter
2009), and the impact of judicialization (Stone Sweet 2010). However, tracing
how the Court’s rulings pay tribute to the preferences of member states seems
to overlook the most obvious point: to have a supranational constitutionaliza-
tion of material policy is a highly supranational outcome. Because the Court’s
case law that interprets the Treaty adds to the EU’s constitution, every time the
Court rules expansively, this ruling carries a long way. This is not counteracted
by many more rulings where the Court may be attentive to member states’
preferences. Governments come and go, and democratic majorities advancing
different policy preferences change. Yet a supranational constitution is likely to
stay. Since 1957, we have seen many Treaty renegotiations, notably with a
European Parliament that has changed its shape and increased its competences.
However, the fundamental material content of the Treaty that is mainly
embodied in the four economic freedoms (of goods, services, persons, and
capital) and in competition law has mostly been altered by the Court. By
overlooking over-constitutionalization, the debate on the Court has focused
on the wrong questions in determining the degree of the Court’s power.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

4 The ECJ and the Policy Process

What lies behind such a lack of attention being paid to over-
constitutionalization may be the predominantly Europhile attitude of most
scholars of European integration. This has been criticized in relation to
political science (Majone 2014) as well as EU law (Conway 2012: 84). As the
constitutionalization of EU policy locks integration in, transferring compe-
tences from the member states to the EU level, this is an outcome welcomed by
many. But, concomitantly, policymaking is withdrawn from democratic
decision-making and transferred to courts, which requires critical discussion.
The significant amounts of attention paid to the ways in which the democratic
legitimation of the EU can be furthered (Føllesdal and Hix 2006; Hix et al.
2007; Moravcsik 2002; Olsen 2013) surprisingly turns a blind eye to the extent
of policy constitutionalization that leaves ever fewer policy choices to major-

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

itarian decision-making.
I will leave the normative discussion of my positive analysis to the conclud-
ing chapter, but let me note from the outset that I perceive the normative
predicament of over-constitutionalization in the withdrawal of policy options
from majoritarian decision-making. With its origins as an intergovernmental
Treaty, the EU’s constitution contains detailed policy provisions, in particular
for market-making, while being weak on the basic rights protecting minorities,
a traditional function of constitutions. Closely related here is another norma-
tive concern. Fundamentally, as an economic treaty, its constitutionalization
distorts the balance between state and market. Negative integration is favoured
over positive integration, as Scharpf put it convincingly in the mid-1990s
(Scharpf 1996). This has important political economy implications for mem-
ber states. While my empirical analysis focuses on forms of market-making
that are furthered by case law, thereby interpreting the four freedoms (and EU
citizenship), I will leave the important question of these political economy
implications to other scholars. My interest concerns what one may term
judicialization,4 which has resulted in an over-reliance on courts in deciding
policies. Possibly, the normative contentions surrounding negative integration
have diverted the attention of scholars away from the mechanism underlying
negative integration, namely over-constitutionalization.

THE BOOK’S ARGU ME NT

This book analyses the Court’s power. It argues that the key to understand-
ing this power lies in the policy implications of having a supranational

4
I follow the definition of Stone Sweet: ‘The “judicialization of politics,” concerns how
judicial lawmaking—defined as the law produced by a judge through normative interpretation,
reason-giving, and the application of legal norms to facts in the course of resolving disputes—
influences the strategic behavior of non-judicial agents of governance’ (Stone Sweet 2010: 7).
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Introduction 5
constitution that, according to its drafters’ intentions, is an international
treaty. An international treaty contains different policy goals that are subject
to further efforts of cooperation. But with the constitutionalization of the
Treaty, these policy goals receive constitutional status. Every polity legitimizes
itself also through its policy output (Scharpf 1970). If this output is highly
constrained by constitutional rigidities, it means that policy options that
would be in the ‘win-set’ of participating actors’ preferences may no longer
be an option for majoritarian decision-making. This constraint of constitu-
tionalizing policymaking limits democratic decision-making at a time when
globalization and government debt have already decreased policy options
(Streeck and Mertens 2011).
Given the long record of research on the ECJ, it may seem unconvincing

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

that one can advance an argument that illuminates a significant blind spot of
existing research. An example may be helpful here. In 2015 and 2016, much of
the discussion about Brexit focused on the question of EU citizens’ access to
social benefits in the UK. One might ask why the UK, as one of the masters of
the Treaty, is politicizing a policy to which it must have consented in the
Council? However, much of the European policy relating to EU citizens’ access
to benefits is shaped by the case law of the Court, as I will discuss in the
following chapters. The constitutional status of this case law, drawing directly
on the Treaty’s provisions on free movement and EU citizenship, heavily
constrains the EU legislature’s decisions on the EU’s mobility regime. The
UK, with its tradition of parliamentary sovereignty, may be regarded as being
particularly sensitive to increasing transfers of sovereignty. With its common-
law tradition, it is also particularly well-suited to attentiveness to the way that
court rulings determine policy. In this sense, it is surprising that the Brexit
debate did not put more emphasis on the general problem of constitutional-
ized EU policymaking. Instead, again, the question was framed in terms of
subsidiarity, namely national parliaments’ means of showing ‘yellow’ or ‘red’
cards to the Commission’s policy proposals. And yet, as we will see, the EU’s
secondary law is heavily influenced by ECJ case law. Controlling the Com-
mission is of little help. While research on the ECJ has time and again shown
the Court’s significant impact on policy (Alter and Vargas 2000; Cichowski
2007; Stone Sweet 2004), the insights of this literature have not been taken up
in the analysis of policymaking in the EU. The latter barely pays attention to
constitutional constraints on policy options but assumes that everything
depends on participating actors’ preferences.
Historically, it was quite early in the process of European integration that
the ECJ established itself as an important force, and therefore as a political
actor in its own right. I argue that the impact of the direct effect and
supremacy of European law is still not well understood. With these principles,
which were later followed by further rulings that extended their reach to
different kinds of secondary law and established state liability for breaches
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

6 The ECJ and the Policy Process

of EU law (C-6/90 Francovich), the Court constitutionalized the rights of the
Treaty, thereby transforming the political system of the EU. Stone Sweet
(2007) even argues that this amounts to a ‘juridical coup d’état’. It is this
transformation, and its impact on policymaking on the European and national
levels, that is the focus of this book.

A Skewed Balance of Powers

Compared to the situation at the national level, the balance of powers at the
supranational level is skewed. When studying the EU as a political system

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

from the perspective of comparative politics, as was advocated in an influential
contribution by Hix (1994), this has to be taken into account. With a stronger
court, the legislative is weaker. Weiler was early to point out the interdepen-
dence between legislature and judiciary at the European level (Weiler 1981,
1991). In the booming field of EU legislative studies, researchers tend to ignore
the implications of case law, the constitutional nature of which they are rarely
aware of. An example of this can be found in an important book by Thomson
(2011) on the Council of Ministers. Despite being the result of one of the
largest empirical exercises in EU studies, which analyses the preferences of
member states in EU legislative decision-making, the index does not even
contain an entry for the Court.
The constitutionalized Treaty of the EU makes case law, alongside legisla-
tion, vital to EU policies. But case-law development presupposes case load.
Only if a sufficient number of cases reach a court can it incrementally develop
case law. Two legal procedures are of particular relevance for the case load of
the ECJ: infringement procedures that are handed by the Commission to the
Court when member states fail to meet their obligations from European law
(Article 258 of the TFEU) and preliminary-ruling procedures (Article 267 of
the TFEU), where national courts ask the ECJ for advice in the interpretation
of European law. Through the preliminary-ruling procedure, the Court estab-
lishes rules for interpreting EU law in the context of a specific case. As such,
the decision on the legal case is taken by the referring court, taking into
account the ECJ’s interpretive guidelines. This procedure gives private actors
indirect access to the Court. Member states thus cannot control the case load
of the ECJ. They can, however, address observations to the Court and thereby
make their views known to the judges. On this basis, scholars have analysed
how member states’ preferences influence the Court. Since 1994, preliminary
rulings have been more important than direct actions (which are mostly
infringement procedures initiated by the Commission alongside annulment
procedures) in most years. In recent years, preliminary rulings have by far
outnumbered all other procedures; there were 436 in 2015 and 428 in 2014 in
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Introduction 7
comparison to 48 and 74 direct actions in 2015 and 2014 respectively (Court
of Justice of the European Union 2016: 94).
But how can court rulings have a significant impact if the ECJ is as
responsive to governmental positions as some authors are able to show
empirically (Carrubba and Gabel 2014; Larsson and Naurin 2016)?
A combination of factors makes this plausible. First, authors fail to take into
account that all rulings interpreting the Treaty have as much constitutional
status as the Treaty itself. Small constitutional changes go a long way. The
Court’s high case load allows it to develop case law incrementally. Although
many rulings are deferential to member states’ preferences, this cannot offset
the impact of even a few cases if these have constitutional status. Secondly,
case-law development and its impact on policymaking is a long drawn-out

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

process. We, therefore, have to be attentive to the time element. This matters
for governmental preferences, which are in part fluid. Governments do not
necessarily know what their preferences will be in a few years’ time, they
cannot foresee how the many different Court cases will possibly affect them,
and they partly use the EU level as a means to ensure ‘blame avoidance’
(Weaver 1986) for unpopular measures. Only on a very general level can we
assume the existence of an antagonistic relationship, where governments want
to safeguard their sovereignty while the Court aims to foster integration. We
are analysing a multilevel system, not an international regime where member
states are the sole gatekeepers to the supranational level. The largely private
actors who turn to European law through preliminary preferences are, at the
same time, an important constituency for national governments. Thus, there
may be silent support for the ECJ and, whether this support exists or not, its
rulings matter, as politically contentious measures could not be realized
without the blame being shifted in this way.

Understanding Case-Law Development

If the Court’s power lies in the influence of its case law on policymaking, it is
necessary to understand how this case law develops and how actors integrate it
within their strategies. Case-law development needs to proceed incrementally
in a case-specific way and emphasizing precedent, as courts are not legitimated
to set general rules in the way a legislature is. Precedent and case law play a
particular role in common-law systems as one source of law alongside the
legislative text. Also civil-law systems, which emphasize guidance by legislative
statutes, have to honour precedent (Beck 2012: 274; Jacob 2012; De Somer
2016). It is only by adjudicating similar cases in the same way that justice can
be established.
When called upon in preliminary procedures, the ECJ establishes guidelines
as to the interpretation of EU law that direct the application of the law in
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

8 The ECJ and the Policy Process

general beyond the specific case. While these guidelines steer the interpre-
tation of law, such as the term ‘worker’ under the Treaty, they rarely give clear-
cut rules, for which the Court would not be legitimated. Thus, for the term
‘worker’, the Court does not prescribe a number of hours or an income
necessary for an individual to qualify as a worker but argues that work has
to be ‘genuine and effective’ but not ‘purely marginal and ancillary’ (53/81
Levin), the assessment of which is then left to the national court.
I argue that for understanding the implications of EU case law for policy-
making, it is necessary to take into account the legal uncertainty that such case
law generates. I use the term ‘legal uncertainty’ to point to the absence of the
possibility to predict law, which is a central element of legal certainty. Thus,
I do not focus on other elements that equally contribute to legal certainty, such

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

as procedural safeguards. Compromises and rule contestations characterize
social life in general and are not unique to the EU. Why should one draw
specific attention to it in the context of the EU, if law never offers complete
legal certainty? How a court reaches a decision is always uncertain, otherwise
there would be no need to address the court in the first place. Yet I argue that
legal uncertainty is exacerbated when dealing with European law. If we see legal
certainty and legal uncertainty on a continuum, there are several reasons why
supranational legal questions are subject to more legal uncertainty than national
ones. These reasons relate to the European Union as an emerging multilevel
polity. We can distinguish causes of legal uncertainty at the national and at the
EU levels, which are connected to the institutional structure and to actors’
interests. Table 1.1 summarizes these sources of legal uncertainty in the EU.
By constitutionalizing the Treaty, its supremacy brings some member-state
laws into conflict. Institutionally, the ever-wider Union has resulted in great
institutional heterogeneity at the member-state level. Consequently, it is
increasingly difficult to judge how a Court case—establishing or interpreting
a legal principle in the light of the situation in one member state—translates to

Table 1.1. Legal uncertainty in an emerging multilevel polity

Institutions Actors

EU level Direct effect and supremacy of EU law; Self-interests of ECJ and COM to use
constitutionalizing the EU Treaties EU law as a motor for integration;
intergovernmental decision-making
leading to joint-decision traps and
empty compromises subject to
competing interpretations
MS level Heterogeneity of MS’s institutional Interests of litigants and MS courts in
conditions; uncertainty as to what case an alternative legal order; incentives to
law concerning one MS means for escape national legal restrictions
another MS
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Introduction 9
similar, but also different, conditions in other member states. This adds legal
uncertainty and may provide incentives for further litigation to challenge
national rules. European case law has a fundamentally different status to the
case law of many other international courts. Its relevance is not restricted to
the parties of the case (inter partes), but it takes general effect (erga omnes).
The interpretation provided by the Court adds to the rule’s meaning.
On the EU level, the Court as well as the Commission have their own
interests in furthering integration, with the help of the constitutionalized
Treaty. As the Treaty sets the goal of an ‘ever closer union’, the Court often
interprets rules in a teleological fashion, in pursuit of this aim. Given their
heterogeneous interests (and demanding decision rules) member states cannot
rein in the Court. Because of the joint-decision trap (Scharpf 1988), a minority

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

of governments, or even a single one when case law is based on the Treaty, can
block a change. Moreover, EU law often has multiple meanings given its
nature as a compromise. It is on the basis of a highly differentiated back-
ground and without a shared culture-based understanding (Streeck and
Thelen 2005: 14) that member states agree to political compromises that are
likely to incorporate more diversity than those struck at the domestic level.
This leaves a great deal of leeway for the Court, as well as the Commission, in
interpreting these rules.
For private actors, legal uncertainty grants opportunities for litigation that
advances the impact of EU law further. At the level of the member states, we
find that this constitutionalized supranational order provides frequent oppor-
tunities for private litigants to turn to an alternative legal setting, overcoming
the constraints of national law, under the precondition that they have standing
at a court to make their claims. Ever since the ECJ established the direct effect
and supremacy of European law, this predominantly economic constitution
grants subjective rights to individuals (Vauchez 2008). Additionally, public
actors, most of all national courts, may well take an active interest in applying
European rules rather than national ones. For member states, there are,
therefore, incentives for codifying EU case law into secondary European law
(in the form of directives or regulations), in order to strengthen legal certainty
and possibly to stop case-law development in an attempt to signal their
preferences to the Court. Codification means that the regulatory principles
that the Court has developed in its case law are written into secondary law. It is
not unusual for directives to directly cite principles that the Court has devel-
oped. If an area is only regulated by case law, it is much more difficult for
actors to know their rights and duties. And as member states’ administrations
respond to case law to different degrees, codification promises member states
more of a level playing field amongst each other. European Union legislation
may, thus, be embedded in case-law development. This implies that the impact
of case law cannot be reduced to the question of compliance with single rulings
but reaches into legislative responses.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

10 The ECJ and the Policy Process

Legal uncertainty is intimately linked to institutional and policy change
(Streeck and Thelen 2005: 26–7). Actors have to adapt to new rules and
interpret them according to their needs. Given that the extent of institutional
change is connected to the pace of European integration, legal uncertainty
accompanies integration. The importance of legal uncertainty in the EU is
interesting in view of the way that the ECJ is taken as a model for the new kind
of international courts. Alter’s recent work shows that the ECJ has lost its
former sui generis status. It has become the premier example of a new kind of
international court that opens up ‘the new terrain of international law’—which
is the title of her book (Alter 2014). As international legalization is increasingly
analysed in terms of obligation, precision, and delegation (Abbott et al. 2000),
the EU demonstrates that there are inherent limits to precision, as supra-

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

national law thrives on its expansive interpretation.
As I will show when I analyse the development of case law, legal uncertainty
is embedded in a path-dependent method of interpreting the Treaty’s prin-
ciples. Interpretations that originate in one area of the Treaty are being
transferred to other areas, as actors advance parallel arguments in order to
further their interests. The path dependency of the Treaty’s interpretation
provides necessary stability, despite its expansive nature.

Case-Law Constraints on Policymaking

Case law impacts actors’ strategies in the legislative process. Legislative pol-
icymaking in the EU takes place on the basis of a changing default condition
that is defined by case-law development. The Commission bases its legislative
proposals on this interpretation. When negotiating over European secondary
law, member states do not necessarily choose between integrating a policy or
leaving it under national legislative prerogative. Often the choice is between
Europeanizing a policy politically through decision-making in the Council
and the European Parliament or leaving it to the ECJ and its interpretation of
the Treaty in case law. If judicial policymaking matters, much depends on
whether member states assume that the ECJ will further their preferred policy
or whether they rely on compromise in the Council to realize their prefer-
ences. If the Court defines the default condition of decision-making in the
Council through judicial policymaking, member states find themselves in a
very different bargaining position than if this were not the case.
ECJ case law not only impacts legislative policymaking at the EU level, it is
also directly binding on member states. Member states’ immediate reactions to
EU case law are not only important as a neglected side to Europeanization
research and compliance, the impact of case law on member states will also
influence their position on the codification of case law in European secondary
law. How far member states still enjoy the autonomy of regulating their society
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Introduction 11
and economy ultimately depends on the extent to which the Court rules that the
Treaty applies to the national situation. European and national laws overlap.
Domestic economic activity is not solely regulated by national law, which
includes the transposition of existing secondary European rules. Activities
for which no specific European rules seem to exist can still be influenced by
European law, as certain demands may be inferred from the Treaty due to
supremacy and direct effect. Well-established domestic rules may thus suddenly
infringe European law. As this continually developing order takes supremacy
over different national ones, there is recurrent uncertainty as to its effects.
Moreover, the specifics of case-law development leave their mark on the
kind of policy at the European and national levels. Courts decide upon single
cases that are mostly characterized by quite idiosyncratic aspects. In prelim-

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

inary procedures, the ECJ establishes interpretative guidelines in view of a
specific case. Giving constitutional status to these decisions raises serious
problems of generalization, as this book will show. The rulings are relevant
erga omnes and add onto the Treaty. Member states cannot overturn them,
short of a Treaty revision (Syrpis 2012: 13). Neither can secondary law, in the
form of directives or regulations, modify primary law. However, single rulings
and paying attention to the multiple aspects of a dispute are both difficult
to generalize into secondary law when there is codification, and it is difficult to
transform them into general administrative procedures when implementing
case law at the national level. Going back to the example of the term ‘worker’,
the Court’s interpretative guidelines for work as that which is ‘genuine and
effective’ but not ‘purely marginal and ancillary’ neither allow the EU legisla-
ture to set down simply in secondary law that the term worker is reserved for
those who are employed for at least, say, twenty hours a week nor can the
national administration easily revert to such rules, as will be discussed in
Chapter 7. Conant’s (2002) contrasting finding of a lack of impact partly rests
on her classification of case law as inter partes (63). While she traces a
‘dramatic legal development’ (186), she nevertheless expects that ‘automatic
policy changes’ (45) should follow, interpreting instances of resistance that
follow as ‘contained compliance’ (32) and not as member states’ legitimate
struggle to get closer to the original agreement.
Figure 1.1 summarizes the argument of the book. This book aims to
increase the attention paid in studies of the EU to the relevance of ECJ case
law in policymaking at the EU and domestic levels. Moreover, it warns against
too much enthusiasm among scholars about the increasing power of inter-
national courts in a world where the difference between rich and poor
countries depends on the rule of law and functioning political institutions
(Acemoglu and Robinson 2012). Much of the scholarship focuses on whether
international courts can constrain the executive or whether governments are
supreme and do not allow international courts to have significant impact
(Alter 2014; Posner and Yoo 2005). But, for the EU, the assumed dichotomy
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

12 The ECJ and the Policy Process

Treaty

Direct effect and supremacy of the Treaty

Over-constitutionalization of the Treaty and case law

Actors

Alternative legal regime = opportunity structure for private actors

and lower courts
Legal uncertainty about the reach of the EU legal order
Path-dependent case-law development oriented to precedent

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Fluid preferences of governments

MS Level
EU Level
Policy options are closed
Codification
Compliance
COM generalizes case-law
Resistance
constraints
Compensatory measures

Figure 1.1. The book’s argument

between both approaches obfuscates the independent role of the ECJ. Courts
are ‘reactive’ (Conant 2012) actors that cannot drive along political and
societal change alone, as Posner et al. are right to emphasize (Posner and
Sykes 2011; Posner and Yoo 2005). Also, newer empirical research on the ECJ
reveals its dependence on member states’ preferences (Carrubba and Gabel
2014; Larsson and Naurin 2016). However, this does not mean that it has no
significant judicial impact. In a world of increasing transborder activities,
and particularly in the multilevel system of the EU, the interests of govern-
ments are not that fixed. The rulings of international courts, as we can study
in the case of the ECJ, impose constraints. They do that by giving constitu-
tional status to the rights they protect. That is why they are welcomed. But
these are not only constraints on deficient rule of law systems that fail to
honour human rights. In the case of the EU, there are often significant
constraints on democratic legislatures at the national and the European
levels. Under the supranational rule of law, the question is: what is the
law? As the Supreme Court Judge Charles Evans Hughes famously said:
‘We are under a Constitution, but the Constitution is what the judges say
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Introduction 13
it is.’5 The underlying treaties, regulations, and directives are bound to be
complex compromises to which different parties attach different meanings.
This exacerbates the problem of legal uncertainty and court discretion,
making it paramount to analyse the resulting bias in the balance of powers.
It is not clear that there is really room for the enthusiasm ascribed to the
independent role of international courts, as constitutionalized case law
removes policy options from legislatures (Bellamy 2007).

CASE S ELECTION AND METHODOLOGY

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

For quite a while, many overviews of the evolution of EU legislation have been
published (Maurer 2003; König 2007). It would be nice to be able to point out
which portion of legislation is heavily influenced by case law, but such data do
not exist. As I previously mentioned, there has been widespread neglect of the
study of EU legislation in terms of how the constitutionalized Treaty con-
strains policy options. Strategically, as we will see, the Commission is able to
heighten the chances of its proposals being accepted by applying pressure
through case law. Full transparency might undermine this strategy. In the field
of social policy, Martinsen (2015: 77) has analysed the number of Commission
proposals that refer to case law from 1958 to 2014. Of 123 proposals in total,
forty referred to case law and, in twenty-two of these, the Commission aimed
to justify legislative activity with case law. As we do not know whether these
forty proposals are important or marginal legislative proposals, it is difficult to
judge these numbers. This is generally the case with quantitative overviews of
legislative activities, as encompassing legislative acts are included alongside
wholly marginal ones. Given that member states have largely defended their
welfare systems from integration, agreeing only to the coordination and not to
the harmonization of social policy, we know, however, that social policy is not
the most likely area for EU legal influence.
Focusing on the role of the Commission, Hofmann (2013) has tried to
combine case studies on the importance of the Commission’s judicial recourse
on legislation with quantitative data concerning overall legislation and the
use of infringement procedures. He struggles to show the importance of case
law for legislation in quantitative terms, although in his case studies based
on process tracing the impact can be shown. One explanation he gives is that
his focus on infringement procedures represents only a partial picture. The
Commission joins all preliminary procedures with observations and often uses
case-law development in this area as a basis for future legislation. One may

5
<http://c250.columbia.edu/c250_celebrates/remarkable_columbians/charles_hughes.html>
[accessed 17 October 2017].
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

14 The ECJ and the Policy Process

also question whether it is relevant to look for quantitative logic. Often, there
need not be that many rulings—it suffices for just one to push for codification
(Weiss and Blauberger 2015).
Thus, we do not know which parts of legislation are heavily influenced by
case law. Yet, it is also important to keep in mind that if we were to know the
percentage of EU law that is influenced by the Court, its actual importance
would be hidden behind the difference in importance of laws. It is, therefore,
useful to approach the issue from the angle of case-law development and to
ask: in which policy fields has the Court been particularly active? This is
foremost the case in the area of the single market, which has been at the
centre of European integration for decades and the core of the European
integration project, as Beck expresses: ‘The internal market remains the core

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

area of substantive EU law’ (Beck 2012: 171). It is where it started and whence
further integratory steps, such as the monetary union, were taken. Case law is
particularly important in this area of EU law (Derlén and Lindholm 2013: 673;
Stone Sweet and Mathews 2008: 142).
My focus in this book is on the four freedoms: the free movement of goods,
services, persons (comprising worker mobility and the establishment of com-
panies), and capital, which allows the analysis of a central area of case-law
development in the EU. I also include an analysis of citizenship rights,
introduced in the Maastricht Treaty, as its legal interpretation parallels that
of workers. There is an important normative dimension to this area of case-
law development in the loosening of the nexus between member states and
their nationals (Bellamy 2008b: 599).
My first empirical task in this book is to explain the dynamic development
of case law. Case-law development is the initial dependent variable that this
book is interested in. I have to show that the Court indeed interprets the
Treaty in a way that places constraints on member states. Thus, Chapter 3
traces the interpretation of the four freedoms as an incomplete contract by the
Court, arguing that this development is characterized by legal uncertainty,
which provides an opportunity structure to private actors, lower courts, and
the Commission. In the attempt to broaden the reach of the freedoms through
litigation, reasoning by analogy plays an important role, which transfers
arguments from one freedom to another. This orientation of courts and
litigants towards precedent leads to a path-dependent legal interpretation,
which gives necessary stability.
After having developed an explanation for the development of case law, the
remainder of the book concerns itself with its impact, as an important
independent variable, on policymaking at the European and member-state
levels. At the EU level, focusing on the four freedoms allows me to select case
studies from EU legislation where case-law development has led to legislation,
as well as where it failed to do so (Geddes 1990). Note that my overall
argument relates to the neglect within research on European integration to
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Introduction 15
study the implications of constitutionalizing the Treaty. I do not include cases
where Court interpretations of secondary law require secondary law reform.
These follow a different logic, given the absent constitutional nature of case
law. While legislation in the single-market area abounds, there are some major
legislative initiatives that correspond to the market freedoms. Thus, there are
three examples of legislative acts that codify large areas of case-law develop-
ment: the Services Directive relating to the freedom to provide services and the
freedom of establishment; the regulation for the mutual recognition of goods;
and the Citizenship Directive. While quantitative studies can use representa-
tive sampling, they cannot account for the political and economic salience of
the legislation. The Services Directive rates highly on both accounts, as it led
to unprecedented political mobilization while covering a high percentage

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

of member states’ Gross Domestic Product (GDP). For the free movement
of services, I will also analyse the recent Patient Mobility Directive. It was
originally included in the Services Directive, but then separated from it. It is
important because new member states took a prominent role in its negoti-
ation, and member states were highly divided about codification. Moreover,
member states have reserved prerogatives for health in Article 152(5) of the
TEC, so that the Patient Mobility Directive demonstrates the difficulty of
protecting policies from single-market freedoms. In order to discuss the limits
of the proposed mechanisms of case law in impacting legislation, I also analyse
two cases where legislation at the EU level failed. First, I briefly discuss
the Monti II regulation, which was meant as a reply to the contentious Case
C-438/05 Viking and Case C-341/05 Laval cases of the ECJ that subjected the
right to collective action to the need to respect single-market freedoms in
2007. Secondly, I discuss in greater detail the failed legislative response in
direct company taxation, following up on the extensive case law of the free
movement of capital and of establishment. While case law in this area
seriously constrains member states’ tax policies, agreement on secondary
law was never achieved.
By focusing on the impact of the four fundamental freedoms, I leave out
important areas of the Treaty that have also had a big impact on policymaking
at the European and national levels. In particular, competition law and anti-
discrimination rules come to mind. Competition law gives the Commission
broad administrative powers and has supported many liberalization processes,
particularly in relation to utilities or transport. Because of the de minimis rule,
there is a threshold at which competition law becomes relevant (Conway 2012:
39), which protects member states somewhat. Next to economic rights, anti-
discrimination rights have a strong foundation in the Treaty and have led to
significant advances, particularly in relation to equal treatment between
women and men since the 1970s (Cichowski 2007), and resulted in different
anti-discrimination directives. In their study on position formation in the
European Commission, Hartlapp et al. (2014: 75–6) mention that Cases
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

16 The ECJ and the Policy Process

C-7/93, C-147/95, C-206/00, and C-351/00 had a direct influence on the
internal negotiation of secondary law proposals on anti-discrimination.
With the foundation and judicial development of EU citizenship rights since
Maastricht, rights have been broadened beyond the previous dominance of
economic aspects, partly allowing EU citizens to take equal advantage of
member states’ welfare systems (Schmidt 2012). As citizenship rights are not
only linked to non-discrimination but also to persons’ rights of free movement,
an analysis of them takes up part of this book.
After having analysed the impact of case law on EU policymaking, the book
turns to the member-state level and to their reactions to European case law.
Just as in the study of the integration process, research on Europeanization has
largely neglected the implications of case law. In an increasingly heteroge-

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

neous Union, a systematic study of these effects is far beyond the reach of any
monograph. To give an example, there are indications that the new member
states are more reluctant to recognize the erga omnes effect of case law. This is
apparent, for instance, in the negotiations on the Patient Mobility Directive
discussed in Chapter 5. This makes increasingly heterogeneous effects across
member states and policy fields likely. In my analysis of member-state
responses to case law, I only aim to synthesize what we know from existing
research on case law’s Europeanization effects. On the one hand, the goal is to
provide more systematic avenues for future research on Europeanization. On
the other hand, the impact of case law at the member-state level is important,
as it influences member-state positions on the codification of case law at the
European level, as will be discussed in Chapters 4 and 5.
Case studies face two fundamental problems: the problem of generalizabil-
ity and the problem of conclusive argumentation. There are thousands of EU
legislative acts. If I can show for a few of them that case law matters, is this a
relevant finding? It is not possible to indicate how generalizable my findings
are, as we lack sufficiently detailed information on the broader set of cases.
It is, however, possible to show that my findings are relevant in terms of
economic significance and in terms of the diversity of policy areas that are
affected. Given the common ignorance in political science of the constitutional
nature of case law, and the detailed, material policy prescriptions of the ECJ,
one could argue that legislative analyses of EU policymaking would need to
show that case law does not matter.
Additionally, if a judicial shadow over EU legislation is not in fact the rule,
which we cannot know, I describe important mechanisms for understanding
the European integration project. For instance, pressure from the ECJ is an
important exit from the joint-decision trap (Falkner 2011). Qualitatively,
legislation such as the Services Directive or the Citizenship Directive has a
much greater economic impact than most legislative projects. Moreover, and
on a more fundamental point, the following case studies can raise understand-
ing on how legal precedent may matter.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Introduction 17
In summary, the analysis of the impact of case law on policymaking at the
European and the member-state levels allows a more differentiated assessment
of the ways in which the ECJ makes a difference. As the analysis will show,
again and again constitutionalized case law withdraws policy options from
legislative decision-making.
Turning from the problem of generalization to the question of conclusive
argumentation, the analysis is based on case studies (Gerring 2007), employing
what one might call a ‘standard’ set of methods, in other words process tracing of
decision-making with respect to the relevant actors (George and Bennett 2005),
their preferences, strategies, and institutions, as well as making comparisons
across cases (Collier 1993; Hall 2003; Lijphart 1979; Sartori 1991). Alongside
comparisons, the analysis of case studies draws on inductive reasoning. Coun-

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

terfactual arguments can provide a check here (Fearon 1991; Levy 2010). My
analysis is grounded in actor-centred institutionalism (Scharpf 1997; Scharpf
2000a), with an interest in establishing mechanisms that explain institutional
change (Mayntz 2004). The quantitative turn of political science (King et al.
2004) has led to an underappreciation of what we can learn from single cases
(Rueschemeyer 2003) and from mere description (Gerring 2012). Increasingly,
methodological refinement in political science focuses on case studies, possibly
underestimating what we can learn through competing analyses of the same
political phenomena using different data, methods, and, importantly, theories.
Along with my argument that research on European integration fails to pay
sufficient attention to the policy implications of the constitutionalization of the
Treaty, I discuss different studies critically in terms of this deficiency.
Newspapers, press reports, and documents from the EU, national governments,
and private actors were analysed, partly with the help of MAXQDA. Given this
book’s research interest in the implications of ECJ case law, the review of case notes
and legal analyses has been a major part of the work. I was able to draw on different
interviews conducted within the context of different research projects. In addition,
quantitative data on case-law development have been gathered, for instance
concerning the path-dependent nature of EU law, by tracing how case law
increasingly refers to several of the fundamental freedoms. At the national level,
data have been assembled on the citation of ECJ landmark cases by domestic
courts as a way to trace how case law impacts the legal orders of member states
when there is no codification in EU secondary law. Specific references to the kind
of data analysis are made in the appropriate places throughout the book.

THE BOOK’S S TRUCTURE

My argument is that the supranational judiciary has far-reaching implications

for policies at the EU and national levels, which research on European
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

18 The ECJ and the Policy Process

integration has largely ignored. An intergovernmental treaty forged in order to
pursue specific integration aims has a very different thrust than a national
constitution designed to allow political order and to safeguard individual
rights. In order to show these implications, three steps are necessary. First,
we have to understand case-law development. How does the Court interpret
the Treaty? Only if case law adds specification and if it matters on its own is it
necessary for the analysis of legislation to pay attention to it. Secondly and
thirdly, the question becomes how case law makes a difference to policy-
making at the European and member-state levels. This book analyses this
impact with a special emphasis on the European level.
In order to advance my argument, Chapter 2 discusses research on the ECJ
as a political actor. The discussion of the Court has long been characterized by

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

two contrasting positions. What Alter (2014: 337) terms the ‘executive control
thesis’ in line with the intergovernmentalist paradigm has been challenged by
claims of ‘judicial activism’. I summarize how scholars have explained the
power of the Court and its limitations, discussing the institutional rules that
govern the Court and questioning what we know of judges’ preferences.
Important to the Court’s development of case law has been the support of
the EU’s legal community. Alongside this community, the Commission is
also an important source of support, as are the interests of private actors who
litigate and member-state courts that address the Court through the prelim-
inary procedure. Chapter 2 discusses how these actors can push case-law
development further. Member states, as masters of the Treaty, closely
observe the cases that come before the Court, and the latter is aware of its
dependence on member states’ implementation of its rulings. Yet, by inter-
preting the constitution, incremental case-law advances also go a long way.
The chapter closes with a summary of the shortcomings of the literature that
the book addresses.
Chapter 3 turns to an analysis of case-law development. The overlapping of
EU and domestic legal orders, coupled with the great material detail of the EU
Treaty, leads to a state of legal uncertainty, I argue, as to the extent to which
EU law impacts member states’ policies. These policies become subject to
potential challenge. Some private actors can draw on this supranational,
alternative legal setting, which promises them benefits. As I show in respect
of the interpretation of the four freedoms and citizenship, legal uncertainty in
terms of the Treaty’s scope, which is almost always broadening, is embedded
in a path-dependent method of interpreting rights. Principles that are estab-
lished in one area are transferred to other areas, as some private actors
perceive benefits in such a transfer and legitimize their claims through estab-
lished principles. The interpretation of the four freedoms as a prohibition of
restrictions, I will show, was first applied to the free movement of goods. This
was an accidental establishment of a path, hinging on the fact that the trade of
goods was first taken up. From here, the argument that the four freedoms not
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Introduction 19
only require that member states grant non-discrimination rights but that
member states have to justify all restrictions on these freedoms was transferred
to services, establishments, the free movement of workers, capital, and also
citizens.
It is not easy for political scientists to examine the course of legal argumen-
tation in case-law development. This is particularly true as case law relates in
great detail to the specific conditions of the decision at hand to an extent that
cannot be of interest to political scientists. By drawing on examples from the
four freedoms of goods, services, persons (free movement of labour, establish-
ment, citizenship), and capital in order to show parallels in juridical argumen-
tation, I hope that the discussion will be of wider general interest. At the same
time, the detailed analysis of ECJ case law will serve to open this terra

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

incognita more widely to political science. As I show in Chapter 3, the Court’s
interpretation of the freedoms gives detail and substance to their meaning. In
light of the constitutionalization of European law, policymaking at the EU and
member-state levels has to be attentive to this meaning.
As case law complements the Treaty, it is one basis on which the Commis-
sion proposes its secondary law. In this way, case law takes generalized effect.
Chapters 4 and 5 move on to this analysis. Chapter 4 starts by focusing on the
different ways that the Court’s case-law development interacts with legislative
policymaking. It then turns to an analysis of the Services Directive and the
lesser known regulation on the mutual recognition of goods, showing how
principles of case law that are motivated by the specific circumstances of
individual cases constrain the design of general rules, given the constitutional
status of case law. Secondary law cannot modify constitutional principles. At
the most, the legislature can hope to signal to the Court its political prefer-
ences. While the Court as a reactive actor, and so dependent on implementa-
tion, cannot ignore these, the Services Directive in particular shows how
political contention does not necessarily intimidate the Court. Generalizing
the prohibition of restrictions approach from the free movement of goods to
services raised concerns about the redistributive implications of the kind of
regulatory approach that was chosen. Such political contention can be inter-
preted as an ‘inefficient’ outcome in the context of a path-dependent expla-
nation of case-law development. Despite being confronted with a political
reaction to its legal principles, the Court went on to broaden the reach of its
jurisdiction with the contentious Viking and Laval rulings on the limits on
collective action.
The Services Directive is at the core of the single market and, despite the
political contention surrounding it, one may argue that the legitimation
requirements for market regulation are not so high. Thus, case-law constraints
may not matter so much after all. Chapter 5 addresses policies that are more
sovereign sensitive. The Citizenship Directive and the Patient Mobility Directive
were both highly influenced by case-law development, although member states
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

20 The ECJ and the Policy Process

have largely reserved their right to define citizenship and their welfare systems.
These exceptions in the Treaty have not restrained the Court, which consistently
holds that member states have to respect the four freedoms also in the areas
where they have defined exemptions in the Treaty (Azoulai 2011). Citizenship
and patient mobility are very interesting because member states defined com-
mon policies for coordinating their welfare systems at an early stage. Although
member states had agreed in secondary law how they wanted to translate the
policy goals of the Treaty, this did not inhibit the Court from partly designing an
alternative policy, drawing on its understanding of the Treaty’s requirements.
This case law has constitutional status and cannot be overruled. In particular, the
Patient Mobility Directive shows how EU policies are becoming contradictory
and dysfunctional as a result.

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Case law cannot establish general rules as courts are only legitimated to
rule in an individual case. Member states often have an interest in agreeing on
secondary law once a certain amount of case law has developed, as this
improves legal certainty compared to a situation where regulation of a field
is predominantly case law driven. Regulation through case law has a tendency
to create inequalities. For private actors, the ‘law of the land’ is more difficult
to understand if it is based on case law and not on secondary law, favouring
those who are able to afford legal advice. Across member states, there is no
resulting level playing field if administrations and local courts have to base
their decisions partly on case law, as member states understand and react to
case law unevenly. If member states agree to codify case law in the Council,
this cannot be taken to mean that it conforms to their preferences, as legal
scholars assume (Syrpis 2012: 7). Rather, case law may set in motion a process
of policy change that results in the necessary majority preferring codification
that would not have been there without the case law.
However, existing case law does not guarantee the necessary majorities for a
common policy. Corporate tax policy is an example where agreement on
secondary law does not emerge despite ample case law. I discuss this failure
of codification in Chapter 5. The tax competition that the Court has fostered
has increasingly been recognized as a problem for public revenues. But member
states’ interests remain diverse. Interestingly, recent moves by the G20 and
the Organization for Economic Cooperation and Development (OECD) to
draw up international rules against tax competition and base erosion have also
come into conflict with the constraints of EU case law that limits the win-set
of agreement.
In light of these growing constraints, it is surprising that political science is
not paying more attention to this special status of EU law. However, the
constraints could be a mere paper tiger, and political science would be right
to pay little attention to them. Chapters 6 and 7, therefore, consider the
member-state level to investigate compliance with case law. The literature
on Europeanization that deals with the consequences of EU membership for
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Introduction 21
member states has rarely analysed the domestic impact of case law. The
dominant argument is that member states are ‘contained compliers’ (Conant
2002) attending to case-law constraints only when there is pressure from
interested constituencies. In Chapter 6, I discuss two case studies that deal
with taxes and with the residence of third-country nationals. Both should be
least likely cases for the compliance of member states, and yet we find far-
reaching responses to the residence rights of third-country nationals and
resistance in the case of taxes. Member-state administrations do respond to
ECJ case law in a rule of law fashion, changing their administrative guidelines
as they would in response to national constitutional law. But they do not always
do it. They may also resist, as they wait for further case law and hopefully narrow
down the scope of EU law requirements.

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

On the basis of these inconclusive findings, Chapter 7 seeks to summarize
what we can derive from different case studies about member-state responses.
There can be no expectation that case law has a one-dimensional impact,
I argue. Member-state actors differ in their preferences for policy reform as
backed by case law and their interests in keeping to the status quo. Actors,
moreover, differ in their cognizance of the constitutional nature of case law
and the interpretation of case-law constraints. The domestic impact of case
law, therefore, is bound to be diverse and will be one consideration of many in
shaping domestic policy processes. In order to give examples of domestic
responses, I differentiate between executive, legislative, and judicial reactions.
Lastly, societal actors are important in the way they further the domestic
impact of EU law, and their cooperation is partly required in responding to
case law. As case law does not prescribe general rules in the way secondary law
does, there are direct and indirect responses. In part, domestic policies are
required to be changed as a direct effect of new case-law development. In part,
domestic policy changes respond to indirect effects, for instance if case law
fosters regime competition. Another indirect response arises when domestic
actors aim to compensate for policy instruments that are taken away by case-
law constraints. To sum up, it is a diverse and somewhat messy picture that
emerges when analysing the domestic impact of EU case law. As constitution-
alized case law takes away policy options, narrowing the realm of democratic
politics, it is, however, important to take note of these constraints. Moreover,
it is not only at the EU level that the need to heed case law hampers the
designing of general policy rules. The British response to EU citizens’ access to
social benefits that I discuss shows how single rulings with a constitutional
status do not translate well into universally applicable rules for administra-
tions (Blauberger and Schmidt 2017).
Chapter 8 concludes by providing a summary of the book’s argument. First,
it discusses the book’s findings with regard to the assertion that the ECJ is
activist. Contrary to this claim, I argue that the impact of the ECJ does not
depend on its activism. Rather, its impact is due to the constitutional nature
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

22 The ECJ and the Policy Process

of its case law. Secondly, in a temporal dimension, I contend that the
different durations of political and judicial processes imply that the impact
of case law is generally overlooked because of its incremental nature. Thirdly,
I discuss the normative dimension of the ECJ’s impact on policy. On this
basis, I ask what we can learn that can be applied to the study of international
courts. I close with a few remarks on where research on European integration
may go from here.

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

The European Court of Justice

as a Political Actor

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

In this book, I explain how the constitutional status of the European Court of
Justice’s (ECJ) case law impacts policymaking at the European and member-
state levels. This chapter will summarize how political science considers courts
in general, and the ECJ in particular. This allows me to map out the added
value of my own approach within this research field. Specifically, I argue that the
literature in this area has thus far ignored ‘over-constitutionalization’ (Grimm
2015: 469). This term implies the ways in which detailed policy prescriptions
in the Treaty offer ample opportunities for judicial interpretation: case law
not only reveals constitutional constraints on public authority, but it gives
an interpretation of policy that is subsequently embedded in secondary law.
Codification in secondary law generalizes case law. Even if the Court is often
attentive to governments’ preferences, small steps taken in constitutionalized
case law go a long way.
Analyses of courts as political actors can be roughly distinguished into two
categories: those which downplay or emphasize their impact. Scholars who
deny that courts can have any significant impact typically stress the idea that
courts are essentially ‘reactive’ (Conant 2012: 11) actors that are dependent on
both litigants bringing cases before them and on governments and legislatures
complying with their rulings (Hönnige and Gschwend 2010). Scholars who
regard courts as active or dynamic political actors typically point out that law
is an incomplete contract in need of interpretation. This gives leeway for
judges to pursue policy preferences. Courts’ principals would have to reach
agreement in order to ‘punish’ courts for going astray. They are also con-
strained by the ethos of the rule of law not to interfere with independent
courts. We can find both approaches in relation to the ECJ in the form of the
‘executive control thesis’ (Alter 2014: 337) and the ‘judicial activism’ claim
(Stone Sweet 2004). With convincing empirical support for both judicial
impact and constraints upon it, research is increasingly focusing on ‘the
political boundaries of judicial discretion’ (Larsson and Naurin 2016: 1).
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

24 The ECJ and the Policy Process

In the following section, I summarize to what extent the literature in this
area regards courts as political actors. Are courts more than the neutral umpire
implied within the concept of balance of power? How is it possible that they
can have an independent influence on policymaking? I begin with the question
of whether courts are conceptualized as political actors in both political
science and international relations and then turn to the abundant literature
on the ECJ. As we will see, the two perspectives on courts as being either
activist or largely reactive, dependent actors permeate the discussion. This
chapter’s review of the scholarship concludes by discussing the shortcomings
of the literature and leads on to describe the contribution of this book.

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

COURTS AS POLITICAL ACTORS

Are courts political actors? The common assessment of the political role of
courts draws on the standard assumption that as part of the separation of
powers courts adjudicate disputes on the basis of laws that are legitimated by
the legislative. The role of the adjudicator, who is guided by texts that are
decided elsewhere, at first sight does not carry much political significance. It is
only when one takes into account the notion that laws are generally incom-
plete contracts, which do not regulate everyday situations in complete detail
but require interpretation, that the independent impact of courts becomes
apparent. In fact, the US Supreme Court demonstrated very early in its history
that courts may well function as political actors. In its decision in Marbury v
Madison (1803), the Court gave itself the right to judicial review and to declare
laws unconstitutional. If courts can interfere in this way and overrule the
legislature, they are clearly political actors.
But are courts really in a position to do so? Courts receive their institutional
rules from the legislative, which detail the way they operate (Voeten 2013:
423–4). These include how judges are to be appointed. As part of the ruling
elite, they are not likely to stray off course, particularly as they may be fearful
of changes to their operational rules if they do so. Moreover, following the
school of political realism, courts are essentially reactive actors, which rely on
the support of multiple other actors. They have to be called upon by litigants.
More importantly, they cannot themselves enforce the full implementation of
their decisions by the executive and the legislature. The recent disempower-
ment of the Polish and the Hungarian Constitutional Courts and the similar
earlier fate of the Russian Constitutional Court (von Steinsdorff 2010;
Blauberger and Kelemen 2016; Kelemen and Orenstein 2016) serve as
examples of the very limited power courts have to interfere in politics. If there
is some scope for their independent political impact, this is clearly circum-
scribed by the political context in which the court operates (Conant 2006: 84).
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

The ECJ as a Political Actor 25

But the necessary political support is not restricted to the political system.
Courts also require sufficient societal support for their interpretations (Voeten
2013: 427). In the US, the relative weakness of courts was widely discussed after
Rosenberg (1993) claimed that it is a ‘hollow hope’ to expect social change from
court rulings, which he demonstrated using the examples of civil and women’s
rights. In his study of Germany, Vanberg (2005) additionally finds that the
public can give crucial support in overcoming the compliance problem that
courts face when adjudicating against the preferences of governments. If con-
tentious court rulings find public support, and executive inactivity is visible to
the public, there are potential electoral costs for governments in relation to non-
compliance.
In international relations, the debate concerning courts is similarly struc-

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

tured. In this area, independent courts hold the promise of transforming a
dominant state of anarchy into a system under the rule of law. Accordingly,
processes of legalization are broadly discussed (Abbott et al. 2000; Zangl
2005). Again, some authors reject the notion of any independent impact at
this level. Goldsmith and Posner (2006), Posner and Yoo (2005), and Posner
and Sykes (2011) are prominent advocates of the position that international
courts’ authority is bound by the preferences of governments. Rational gov-
ernments decide whether to bind themselves to international law and courts.
The latter are, therefore, fundamentally dependent on the preferences of power-
ful governments as they will only comply when it suits their interests (Alter
2014: 45–6). Alter (2014: 337), however, rejects what she calls the ‘executive
control thesis’ for several reasons. In contrast to the American context, where
political parties can influence the composition of the federal and state judiciary,
this kind of streamlining is absent at the international level. More fundamen-
tally, she rejects the preoccupation with executive control and the strong
rational-choice assumptions that governments cannot bind themselves to this
extent. After all, to achieve empirical support, the argument faces the problem of
revealed preferences: ‘falsifying this executive control thesis would require a
shocking correlation in which judges systematically and generally ignore gov-
ernment opinion. This would be surprising indeed’ (Alter 2014: 338). Rather
than being preoccupied with the means of executive control, she argues that it is
more important to show with which causal mechanisms international courts
can influence politics. A concept that resonates with this argument is that of
judicial power, elaborated by Staton and Moore (2011), which is defined as
the process through which ‘courts endowed with formal powers to review the
actions of states come to constitute binding constraints on governments’ (Staton
and Moore 2011: 555).
The lack of a legitimized monopoly of force at the international level may
appear to distinguish the analysis of international courts. However, as Carrubba
and Gabel (2014: 214) emphasize, it is just as questionable why governments at
the national level should obey court rulings that violate their preferences.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

26 The ECJ and the Policy Process

Similarly, Staton and Moore (2011) argue that the analysis of judicial power
gains from a unified approach for national and international courts.
To study the ECJ, one can thus draw on results for both domestic and
international courts. One approach that does so is Stone Sweet’s influential
theory of judicialization. It describes the increasing importance of the judi-
ciary in comparison to the legislature for the determination of collectively
binding decisions (Stone Sweet 1999). For judicialization to proceed, Stone
Sweet points to certain preconditions: a sufficient case load needs to reach a
court so that it is able to develop case law. Furthermore, precedent needs to
be honoured. Finally, courts have to give reasons in their rulings, as this
allows litigants to draw on case law to further their interests. As Stone Sweet
shows, the importance of precedent in orienting the future behaviour of

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

courts and litigants leads to the path-dependent development of case law
(Stone Sweet 2004: ch. 1). I will expand on this idea in the next chapter. We
will now proceed to a more detailed discussion of the institutionalist foun-
dations of the ECJ’s power (for an extensive discussion, see Stone Sweet
2010), paying due attention to instances in which member states continue
successfully to exercise control.

THE E CJ AS A MOTOR OF INTEGRATION

The literature on the ECJ mirrors the general arguments that are made about
courts. The question of its power was intensely discussed in a debate in the
1990s between Garrett on one side and Burley (later Slaughter) and Mattli on
the other. The idea of the ECJ’s intergovernmentalist position, in which it
merely follows the preferences of the member states and particularly the
larger ones, was introduced by Garrett (1995). He opposed the neofunction-
alist analysis advanced by Burley and Mattli (1993), who had argued that the
Court could use the ‘mask’ of law to further European integration so it did
not appear as a political actor. They interpreted the support of private
litigants and lower courts as processes of functional and political spillover.
Garrett’s argument that governments will only accept ECJ rulings if they
conform to their interests was heavily criticized by Mattli and Slaughter
(1995, 1998). Later, Garrett refined his argument in a joint article with
Kelemen and Schulz, arguing that the ECJ would also rule against the
preferences of powerful member states whenever the legal line of argumen-
tation was well established (Garrett et al. 1998). They found that the Court
backs down in the face of political opposition when its legal interpretation
does not meet with support. However, in cases where there is sufficient
precedent, the Court will uphold contentious decisions against the member
states. Garrett et al. (1998) thus assume that the Court is responsive to
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

The ECJ as a Political Actor 27

the political context from which it receives its ultimate political legitimacy.
Yet, law and precedent have a significant influence, as a court cannot
contradict its own rulings too many times.
By acknowledging the independent impact of (case) law, this position
comes close to the observations of Alter, who supports the dynamic view of
the Court. Alter explains the political power of courts by taking into account
the time dimension. She finds that new interpretations are normally tested in
relatively insignificant cases (Alter 1998: 131). It is only later that the conse-
quences materialize, but by this time there would be significant political costs
in interfering with an independent judgment, which carry much more weight
than the policy losses of a single case (cf. Kelemen 2001).
Based on the work of Alter, Mattli, and Slaughter, as well as Stone Sweet,

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Kelemen, and others, the dominant perspective within the literature on
European integration is that the ECJ is an important political actor in its
own right, as it has been highly influential in advancing European integration.
Phelan even argues that the field is unusual within social science as there is
no prominent alternative explanation (Phelan 2011: 769). Nevertheless, the
argument that member states’ preferences are paramount for the Court’s
rulings is often voiced.
It is more than plausible that the ECJ’s decisions actually reflect the opinion of a
majority of judges and, by imperfect proxy, a majority of member state govern-
ments that appointed them. Thus, an alternative claim is that the Court is acting
within its discretion and has not been curtailed because the preferences of the
judges are not at odds with those of the member states. (Malecki 2012: 62)

Carrubba and Gabel (2014) show in their book that the ECJ’s rulings do in fact
reflect the observations that member states submit in court cases. Larsson and
Naurin (2016) make a similar argument. Based on a separation-of-powers
model, they show that the Court is influenced by the potential for a legislative
override. With a focus on the executive and its means of non-compliance,
Conant (2002) argued that the justice administered by the ECJ is ‘contained’,
as member states only comply to the extent that there is a supporting coalition
furthering these rights. Consequently, the Court can only count on member
states’ compliance when rulings keep close to their preferences.
Thus, as is true for courts in general, the ECJ’s role is subject to continued
debate concerning the scope of its discretion. In order to understand how the
ECJ can be a politically influential actor, it is important to analyse its founda-
tional institutional rules, as well as the support it enjoys from different actors.
In line with the separation-of-powers approach, much of the research focuses
on the potential for member states to override court rulings (Carrubba and
Gabel 2014; Larsson and Naurin 2016). In the following section, I will sum-
marize what we know from the literature, beginning with the Court itself and
its professional environment. I then go on to discuss what can be seen as the
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

28 The ECJ and the Policy Process

Court’s supporting coalition, with the Commission on the one hand and the
ECJ’s cooperation with the member states’ courts and the interests of private
litigants on the other. These are the main parameters of the institutional
setting that permit an activist court, and I will summarize the debate on
judicial activism on this basis. Member states still remain important actors.
I will, therefore, close by discussing the influence that the masters of the Treaty
have on its interpretation. Given decades of research on the ECJ’s operation,
what follows cannot aspire to a complete literature review but seeks to draw
out the main lines of research in order to elucidate the absence in the literature
that my contribution seeks to fill.

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

The Court, its Judges, and its Professional Environment

To answer the question of whether the Court can be a political actor, it is good
to begin by looking at the court itself. How are its judges chosen and under
what constraints do they operate? What do we know about judges’ prefer-
ences, and to what extent is the Court supported within the legal profession?
One ECJ judge is appointed for each member state by common agreement
of the member states for a renewable term of six years (Article 253 of the
Treaty on the Functioning of the European Union (TFEU)). With the Treaty
of Lisbon (Article 255), a general hearing procedure was introduced in order
to screen candidates for their suitability. In general, renewable terms make
judges susceptible to political pressure and are therefore regarded as a sign of
reduced judicial independence. In terms of the ECJ, it is recorded that some
judges may be reminded that salary levels in their home country are much
lower and that reappointment is not automatic.1 But the fact that no dissent-
ing opinions are published and decisions are taken in secret shields the judges
at the ECJ. Moreover, as Carrubba and Gabel (2014: 120) argue, judges
are hardly able to predict which government will be in office upon their
reappointment, which limits strategic behaviour.
For the question of whether judges are put under pressure from their
member states, it is interesting to take a look at the Advocate Generals (AGs)
(Solanke 2011). Appointed on the same conditions as the judges for a six-year
renewable term, they are asked by the Court to give a statement on pending
cases that raise new legal issues. Their opinion lays out the evidence of the case,
pays attention to the applicable precedent and the relevant literature, and
makes an argument for how the case should be decided in the general interest.
Since they expose themselves with these opinions, and in light of the impres-
sion that the Court often follows their arguments, the AGs presumably face

1
Reinhard Müller, ‘Noch ein “Ja, aber”?’, FAZ 10 June 2015.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

The ECJ as a Political Actor 29

similar pressures to judges. However, it does not appear that the five AGs of the
large member states (Poland, as the sixth largest member state, has only
participated since 2013, and the other AG positions rotate among the smaller
member states) aim to avoid controversy with their opinions. Advocate Gen-
erals are frequently reappointed, despite opinions that extend the reach of EU
law and encroach on member states’ sovereignty. Thus, a number of AGs who
took up their positions in 2006 are now in their third term: France is sending
Yves Bot, Italy Paolo Mengozzi, and the United Kingdom Eleanor Sharpston.
The British AG Francis Jacobs held office from 1988 to 2006. The German AG
Juliane Kokott is also in her third term and has been in the post since 2003. In
the legal literature, AG opinions are time and again cited as marking important
departure points for the future development of EU law. One example is AG

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Sharpston, who has repeatedly argued for EU law to prohibit reverse discrim-
ination against national citizens. Such discrimination may result from the
freedoms not applying to purely national situations. If the Court judges that
national regulations, for instance regarding services provision, restrict
the four freedoms, it results in EU citizens being regulated less stringently
than national citizens. Should EU law prohibit this reverse discrimination against
national citizens, the sovereignty member states possess to regulate their markets
would be reduced, which is certainly not in line with the UK government’s
preferences. Nevertheless, AG Sharpston was reappointed and is now serving her
third term, which testifies to weak member-state control.
The ECJ decides by simple majority. Like courts in other political systems,
the ECJ is a very efficient decision-maker. This is the source of its power in the
relatively slow decision-making system of the EU, which has many veto points
(Maduro 2007). It sits in chambers of three or five judges, depending on the
issue. Important decisions are reserved for the Grand Chamber of thirteen
judges. Case C-62/14 Gauweiler, the first preliminary reference for the German
Constitutional Court, which deals with the Outright Monetary Transactions
(OMT) of the European Central Bank, is such a case. The plenum of all twenty-
eight judges very rarely decides a case; a recent example that required all
judges to reach a decision was Case C-370/12 Pringle on the ESM (European
Stability Mechanism) (Court of Justice of the European Union 2012: 112). In
2015, chambers of five judges decided 58 per cent of all cases (2014: 55 per
cent), chambers of three judges 34 per cent (2014: 36 per cent), and the
Grand Chamber 8 per cent (2014: 8.7 per cent) (Court of Justice of the
European Union 2016: 10).
Given that decisions are taken in secret, and because there are no dissenting
opinions, we lack an empirical basis from which to assess the significance of
judges’ preferences. In the research on courts in the US, the attitudinal model is
important, analysing the impact judges’ policy preferences have on judgments
(Segal 2008; Rehder 2006). Malecki (2012) has analysed the Court’s judgments
quantitatively with regard to the composition of the responsible chamber in
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

30 The ECJ and the Policy Process

order to shed some light on internal decision-making. Differentiating whether
the judgment supports or negates the Commission position, Malecki shows that
judges do in fact pursue different preferences; in other words, it makes a
difference which judges sit in a chamber. Beyond more general statements
such as these it is, however, scarcely possible to learn more through quan-
titative analyses. Based on a qualitative analysis of a set of particularly
contentious and seminal judgments, Höpner finds that neither party affili-
ation nor the background of judges within their home countries in different
kinds of capitalist systems contributes much to explaining the judgments
they reach (Höpner 2010).
As difficult as it is to gain insight into judges’ preferences, we can also look
at the institutional rules under which the Court operates and ask how likely it

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

is that judges can bring their preferences to bear upon rulings. I would argue
that there are good reasons to assume that the policy preferences of judges are
less rather than more significant at the EU level, given the rules that govern the
allocation of cases to rapporteurs and chambers. This allocation is the respon-
sibility of the president, who seems to be guided by ad-hoc considerations
(Malecki 2012).2 There is no specialization among chambers or judges in
terms of specific cases.3 This makes it less likely that a judge can develop a
specific line of argumentation across a series of cases. We can assume that the
rapporteur for a case is the most influential for its resolution. Judges are never
allocated cases that concern their home country. In general, one would expect
judges to pursue their policy preferences if they had a connection to the polity
concerned. While judgments influence other cases through precedent (Knight
and Epstein 1996), it is likely that policy preferences play a lesser role at the EU
than at the national level because the effect of case law is more indirect.
For want of ways to empirically assess judges’ preferences further, I there-
fore assume that the ECJ largely pursues its own interest to stabilize and
extend its position as a corporate actor (Schneider and Werle 1990). Self-
interest is represented by the expansion of their own competence and
resources to the monopolization of competence and the enlargement of their
autonomous decision-making (Schimank 1992: 264). To the extent that judges
pursue individual policy preferences, the institutional set-up means that they
are unlikely to have a coherent impact, reducing their preferences to the level of
empirical ‘noise’. While the pay-off for pursuing individual policy preferences
is quite low, the pay-off for building the EU’s legal doctrine is larger.

2
Rules of Procedure, Article 15 § 1: ‘As soon as possible after the document initiating proceed-
ings has been lodged, the President of the Court shall designate a Judge to act as Rapporteur in
the case.’
3
In contrast, at the German constitutional court, judges have specific portfolios. This allows
single judges with a specific assignment to shape case law significantly. An example is Udo di
Fabio, who was responsible for European integration between 1999 and 2011 and as rapporteur
notably shaped the ruling on the Lisbon Treaty, as well as the case about financial aid to Greece.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

The ECJ as a Political Actor 31

It is not a new phenomenon for the community of EU lawyers to provide
arguments for an expansive interpretation of EU law. As recent historical
studies show, from early on the community of EU lawyers has been actively
looking for cases in order to develop case law (Alter 2009: ch. 4; Vauchez 2010;
Vauchez 2015; Pollack 2013). Thus, a group of lawyers working at the
Commission’s legal service at the Court or who were members of the European
Law Association FIDE discussed possibilities for strengthening the ambit of
EU law before van Gend established direct effect. They then publicized the
decision broadly, emphasizing its transformative character and the new role
that the Court should assume for integration. The decision on supremacy in
the Costa case was, thus, clearly anticipated. Schepel and Wesseling show how
lawyers working for European institutions have been unusually active in

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

publishing in the major European law journals, using these venues to influence
the debate on European law (Schepel and Wesseling 1997: 172–3). Vauchez
(2012) analyses how a transnational ‘esprit de corps’ developed through
conferences, Festschriften, and eulogies that have instilled consistency into
the Court’s operation. This helped to overcome the problem of the Court itself
having no influence on the appointment of its judges, meaning they had to
socialize judges from highly heterogeneous environments.
Martin Höpner (2011) has shown in great detail how EU lawyers have
cultivated an ethos in which the transfer of competence to the European level
is regarded as a good thing per se. European legal scholars are not highly
concerned by the fact that the Court cannot on the whole be overruled. They
adhere to a view of the transformationist role of European law. This telos
provides legitimation for extensive case law. Conway (2012: p. xv) even speaks
of a ‘language of love’ between EU lawyers and the ECJ; he argues that
‘A reluctance to criticise the ECJ is marked in academic literature’ (Conway
2012: 84). In view of the overall supportive stance of the legal profession in
broadening the reach of European law, allegations of legal activism have
generally been largely denied and downplayed by the legal community. This
consensus resembles research on European integration in general, where
most researchers have long welcomed further integration. Majone has been a
vocal critic of this in recent years, arguing that European scholars have
turned a blind eye to the pitfalls of European integration. He argues that
more integration is still regarded as the only insurance against war, ignoring
how the dysfunctional aspects of European integration have led to new
conflicts among member states, the failed euro policy being the best example
(Majone 2014).
Within the legal profession, Rasmussen was the first to condemn the
Court’s legal activism (Rasmussen 1986). But his criticism did not meet with
agreement among the mainstream of European lawyers (Conway 2012: 64–9).
In part, defenders of the Court revert to semantic differentiations and argue
that the Court is more entrepreneurial than activist (Solanke 2011). Eeckhout
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

32 The ECJ and the Policy Process

(1998: 17) argues that critics would have to show the usurpation of ‘the role of
the legislature’. For this, the Court would need to divert from European law in
its interpretations, which Eeckhout cannot detect and therefore denies. Cru-
cially, he states ‘the interpretation which [the Court] prescribes is not contrary
to the intentions of the legislature. Those intentions are in any event most
difficult to determine in Community law.’ Another argument points to the fact
that the EU’s developed legislative system indirectly legitimates the Court, as
the legislature could step in (Dobler 2008: 530). However, such legitimation
overlooks the reality that the constitutionalized case law largely does not
permit overrule, given the multiple veto points that can lead to a joint-
decision trap (Scharpf 2006).
At times, criticism has become more vocal, diffusing beyond the expert

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

circles of lawyers and EU integration specialists to the general public (Basedow
2012). Interestingly, the debate was particularly pronounced in Germany,
despite the country’s tradition of having a powerful constitutional court.
Here, the former president of the country, and also of the constitutional
court, Roman Herzog, publicly pledged in 2008 to ‘stop the ECJ’ (Pop 2008).
Contentious rulings like those in the Laval and Viking cases, which con-
strained union activity by arguing that strikes hampered the economic free-
doms of the Treaty, were received very critically, particularly in Northern
European countries (Seikel 2014b; Blauberger 2012). Other cases, such as Case
C-372/04 Watts, which allowed a British patient to circumvent the waiting
times of the British National Health Service, demonstrated the far-reaching
domestic policy implications of supranational case law. In fact Kelemen (2016:
139) perceives that there is a ‘proliferation of critical voices’ and asks, ‘Might
the European legal field, which has so long supported the authority of the ECJ,
come to act as an external constraint of the Court?’
Whichever way the relationship develops in the future, its position of being
embedded in the EU’s legal community has for a long time allowed the ECJ to
expand the reach of EU law. Yet, even in the context of a legal community
pursuing common aims, filling in legal gaps can be contentious. Why was one
interpretation chosen over another? Often the underlying laws give little indica-
tion, as Keeling has argued in relation to the free movement of goods in Articles
30 and 36 of the Treaty Establishing the European Economic Community (EEC
Treaty/TEEC). It may, therefore, be difficult to avoid criticisms of activism.
The bulk of the task was, however, left to the Court. It was the Court that had to
determine the scope of the prohibition decreed by Article 30. It was the Court that
had to decide in what circumstances a measure caught by Article 30 was justified
on the grounds set out in Article 36. . . . The expression ‘creative jurisprudence’,
which is often used in mock disparagement of courts that give non-obvious
answers to questions for which there is no obvious answer, is especially absurd
in this context; for whatever the Court did with such scant material, its jurispru-
dence was bound to be creative. (Keeling 1998: 512)
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

The ECJ as a Political Actor 33

Contentious case law undermines the legitimation of the ECJ. One can,
therefore, assume that the Court has its own interest in preventing conflictual
rulings (Shapiro and Stone Sweet 2002: 128). But the institutional conditions
of its functioning as a supranational court make this difficult. In view of the
great institutional heterogeneity amongst the twenty-eight member states, it is
impossible for the Court to assess all the repercussions of its rulings when
called upon in preliminary proceedings. Larsson and Naurin agree that the
Court faces significant uncertainty as to the political reactions to its judgments
(Larsson and Naurin 2016: 2). Even if member states submit observations, this
is unlikely to provide a full picture of all concerns. Generally, courts have to
decide those cases that are brought before them. As the European Union has
not made all competences supranational, at times the ECJ could principally

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

point to its limited competence and deny adjudication (Roth 2008). In par-
ticular, as the Court acts as a constitutional court, it should be careful which
cases to decide upon, taking into account that its rulings are added to the EU’s
constitution (cf. Scharpf 1966: 528). However, the Court takes the view that it
has to solve all disputes that are brought before it (Keeling 1998: 512). Reach-
ing decisions by simple majority, it is also better placed to do so in comparison
to the Council or Parliament (cf. Bellamy 2008a). And often conflicts are
delegated to the Court that cannot be solved politically due to their conten-
tious nature (Maduro 2007: 824).
Therefore, it is difficult for the ECJ to avoid conflict. The dilemma is
manageable because of the way that the Court’s case law develops. Since
most of its cases are preliminary rulings, the Court does not decide upon
legal disputes by itself but gives indications to member-state courts on how to
interpret the Treaty. It is up to those courts to use these guidelines in
adjudication. Moreover, case law is typically developed incrementally. Given
the weak legitimation of courts to create rules, this allows the Court to await
reactions from the legal community and the member states in order to assess
whether there is sufficient support for the chosen interpretation (von
Bogdandy 1995: 25). The incremental development of case law provides
precedent, upon which later rulings can build (De Somer 2016). Stone Sweet
argues that case law is heavily path dependent (Stone Sweet 2004: 35–6), an
idea I will develop further in the next chapter. Precedent shields the Court
from the reproach of political judgments and makes it easier for the Court to
maintain contentious judgments reached against the member states (Garrett
et al. 1998). By objecting to these judgments, member states would simulta-
neously endanger the principle of legal independence and the general impor-
tance of the rule of law. Incrementally, the ECJ may establish doctrines which
member states would have objected to had they known the consequences of
these rulings for their domestic polities from the beginning (Alter 1998: 131).
In summary, while member states appoint the judges, the ECJ enjoys great
institutional independence. Its operation is shielded from oversight. It can
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

34 The ECJ and the Policy Process

draw on a broad network of EU lawyers, sympathetic to the expansive reach
of EU law, who are also willing to bring cases actively to the ECJ that allow
the establishment of new principles. Case-law development takes place incre-
mentally. Precedent and the value of the rule of law shield the Court from
interventions.

The Commission

Alongside the legal profession, the Commission is a particularly important

party in Court cases. As the two main supranational actors with their own
corporate interests in furthering integration, the Commission and the Court

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

are close partners. As the guardian of the Treaty, the Commission has
privileged access to and knowledge about the Court (Hofmann 2013). The
Commission can hand infringement procedures to the Court when member
states fail to meet their obligations under European law (Article 258 of the
TFEU). In the context of this book, these are of interest only when they are
connected to the Treaty. Their relevance in the monitoring of member states’
compliance with secondary law is discussed in implementation studies (Börzel
et al. 2012). The Commission’s Legal Service has been a particularly powerful
actor in the development of European law since the early days of the European
Economic Community (EEC) (Rasmussen 2012; Alter 2009: ch. 4; Hofmann
2008). Additionally, if the Commission is not part of the legal proceedings, it
normally intervenes and puts its opinion to the Court. Quantitative studies have
shown that the Court sides with the opinion of the Commission in 75 per cent of
cases (Carrubba and Gabel 2014: 134), making the Commission’s position the
single most important predictor of the Court’s ruling (Malecki 2012: 61, 65).
Carrubba and Gabel (2014: 136) argue, however, that this is only because the
Commission reflects member states’ positions towards a case, implying that it
does not have an independent impact on the ruling. Both authors also find that
the Commission refrains from infringement action if it expects non-compliance:
‘The Commission disproportionately pursues enforcement actions when it
expects compliance with a favourable ruling’ (Carrubba and Gabel 2014: 135).
With its right to initiate infringement procedures, the Commission can
instigate case-law development in new areas. When pushing for the expansion
of European law by initiating cases, the Commission normally enjoys the
support of private actors. An example of this is the Commission’s repeated
legal action against the shareholder privileges of the state of Lower Saxony in
the Volkswagen company, which led to two cases (C-112/05, C-95/12). Signifi-
cantly for the Commission, the German company Porsche wanted to strengthen
its influence on Volkswagen but could not achieve this in the German political
context. Porsche, therefore, addressed the Commission to gain its backing
(Werner 2013). European law offers private litigants a very favourable
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

The ECJ as a Political Actor 35

environment for pursuing their interests. On this basis, they may further
integration or push through domestic reforms (Schmidt 2000).4
The literature on adversarial legalism, in particular, has shown how the
Commission, the Parliament, and also the ECJ are keen to support the
enforcement of European law through private actors (Kelemen 2006, 2011).
In its reliance on the administration of member states for compliance, the
Commission needs the support of those actors who profit from European law
in order to enforce it. As such, the control of implementation of European
law relies more on the ‘whistle blowing’ of private actors than on the system-
atic ‘police control’ of national implementation by the Commission (Pollack
1996). Next to instrumentalizing private actors in this way, there is another
point of consideration. Given the unsolved problem of legitimating the European

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

order, the active use of European rights by private actors is seen to be legitimacy
enhancing by the European institutions. After all, it is mainly due to the
Commission, Court, and Parliament that these European rights exist and are
being expanded through case law. By using these rights, private actors indirectly
legitimate the actions of those building them (Kelemen 2006: 123; Cichowski
2006: 51; Slepcevic 2009).
A characteristic of most analyses of the Commission in relation to the Court
is the focus on its initiation of infringement procedures or its interventions in
preliminary references. There is much less attention paid to how the Com-
mission combines case law with its right of initiation (Schmidt 2000). Behind
this neglect, I would argue, lies the lack of attention given to the policy
implications of over-constitutionalization. The ECJ is incorrectly regarded as
a ‘normal’ constitutional court that does not make detailed policy prescrip-
tions. But secondary law—directives or regulations that are proposed by the
Commission—cannot overrule primary law, namely the Treaty and its inter-
pretations. To the extent to which the Court adds to the Treaty’s meaning
through its case law, the Commission is likely to take it up in its secondary law
proposals. This codification generalizes the impact of case law.
Hartlapp et al. (2014) give numerous examples of how references to case law
are important for reaching decisions on internal organizational conflicts
concerning policy within the Commission. In general, however, analyses of
policymaking in the EU pay scant attention to case law. This is the argument
of this book, and I will present my analysis of the importance of the interaction
between the Commission and the Court in Chapter 4. There I will also discuss
the European Parliament, which is not a relevant actor in Court procedures
and is therefore neglected in this chapter.

4
Sometimes (quasi-)public actors also have recourse to changing the legal regime by
reverting to EU law. An example is the attempt of the regional pharmacy board in Saarland to
challenge pharmacy regulation in Germany (C-171–172/07) (Schreinermacher 2013: 27).
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

36 The ECJ and the Policy Process

The ECJ’s Supporting Coalition: Member-State

Courts and Private Litigants

Private actors enjoy direct access to European courts only when appealing to
European decisions that affect them directly, for instance in competition law.
However, individuals or companies may argue before national courts that
European law needs to be applied to their case. This presupposes, of course,
that potential litigants find legal standing, which is often a demanding pre-
condition. When the implications of EU law are sufficiently clear and beyond
doubt, member-state courts can apply this law directly (acte clair doctrine), as
a uniform implementation across the EU can be expected. If there are doubts
about the legal issue, national courts have to address the ECJ through the

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

preliminary procedure. This gives private parties indirect access to the Court.
Litigants may be concerned with trade or general policy interests, or they may be
motivated to address the ECJ for reasons that are more loosely associated with
the case at hand. Thus, NGOs or private law firms may litigate as an advertise-
ment for their general activities or pursue a case for ideational reasons. An
example here is Elaine Vogel-Polsky, the lawyer who pushed the Defrenne
rulings combating discrimination against women (Hoskyns 1996: 68–73). Little
research has been conducted on litigants in the EU (Vanhala 2006, 2009).
Anecdotal evidence points to the relevance of Galanter’s (1974) distinction
between ‘one-shotters’ and ‘repeat players’ in the legal system of the EU too.
While the literature rightly emphasizes the partly prohibitive preconditions for
legal standing (Kelemen 2011), this should not distract from the significant
prize of successful litigation that is analysed in this book: constitutional change
throughout the Union. And the development of ECJ case law shows that, as
restrictive as the situation may appear to the individual litigant, for the Union as
a whole, case load is sufficient to bring about far-reaching changes.
A fundamental aspect of this possibility of challenging the national legal
order emerged in the 1960s as part of the constitutionalization of individual
rights through the ECJ with the doctrines of direct effect and supremacy
(Vauchez 2008). In fact, the preliminary reference procedure was initially
introduced to provide lower courts with the means to seek help in the
interpretation of EU law. But direct effect and supremacy transformed the
procedure into one allowing private actors to challenge national law in terms
of the degree to which it conformed with EU law (Mancini 1989: 606; Alter
1998: 133–5). This developed into a mechanism for decentralized control of
member states’ compliance with European law. The institutional provision of
preliminary rulings, as Weiler argues, leads to a unitary system of judicial
decision-making in the EU (Weiler 1981: 300). As the rulings of the ECJ are
directly translated into the rulings of member states’ judicial systems, govern-
ments struggle to oppose European law without questioning the judicial
independence of their courts.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

The ECJ as a Political Actor 37

But why do member-state courts address the ECJ and help to overturn the
domestic legal system with European law? In fact, it took a few years until the
courts began to take up this opportunity, with the first case reaching the ECJ in
1961 (Mancini 1989: 605). For lower courts, the preliminary procedure is quite
attractive, as it strengthens their independence from the domestic court
hierarchy to which they belong. Basing their decision on supreme European
law shields them from the risk of being overturned by the next higher court
(Weiler 1994: 518–20). And lower courts are said to be less concerned with the
coherence of the national legal system (Alter 2001: 49).
Domestic courts often frame the questions raised in the preliminary pro-
cedure in a way that gives clear indications of the sort of answer they would
like to receive (Nyikos 2006). The ECJ has every incentive to follow this lead,

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

which gives assurance that its verdict will be taken up rather than ignored by
the lower court. Were lower courts not to follow its rulings, this would
undermine the authority of the ECJ. The cooperation with lower national
courts is particularly important for the ECJ, as they provide it with a case load
and integrate European law into the judicial systems of the member states
(Dyevre 2010: 323). Acknowledging the important role of lower courts in
actively asking for certain rulings, Gareth Davies has argued that the ECJ
should not be blamed for its activism (G. Davies 2012). The contentious Case
C-144/04 Mangold ruling on age discrimination shows how litigators may
strategically address courts so as to derail domestic policies with the help of
European law, thereby relying on the explicit framing of the issue by the lower
court judge. In this case, the litigator had first attempted to influence the policy
in the German Parliament. After this had failed, he reverted to fabricating a
case and litigating (Stone Sweet and Stranz 2012). In a quantitative analysis,
Chalmers and Chaves show how preliminary rulings only partly monitor
compliance with EU law and partly push for what they call ‘thickly evaluative
norms’ in the areas of economic liberalism, equal opportunities, and labour
rights (Chalmers and Chaves 2012).
Discussing Brexit, Davies (2016b) has recently called upon domestic courts
to engage in more critical debate with the ECJ, thereby requesting a challenge
to what Conant (2002: 50, 67), for example, criticizes as an often contained
compliance with case law. Davies argues that ‘unquestioning obedience’ is not
loyalty but subversion, as only a judicial dialogue can lead to an acceptable
development of case law.
Still, there are significant differences in the use of the reference system in the
member states (Mayoral 2016). Analysing the British case, Golub pointed to
disincentives to refer cases, which allows policy outcomes to be controlled
while slowing down the integration process (Golub 1996: 381). Marlene Wind
argues that the small number of references from Danish and Swedish courts
have to be explained through the absence of a tradition of judicial review in
majoritarian democracies. There is the assumption that only the highest courts
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

38 The ECJ and the Policy Process

should request preliminary rulings. In general, she finds that domestic courts
interpret EU law themselves, arguing with the acte clair doctrine that the
meaning of EU law is sufficiently clear (Wind 2010: 1046, 1048, 1053, 1055).
Despite such support for the ECJ, the relationship with national courts is
not necessarily an easy one, particularly if supreme or constitutional courts are
involved. The ECJ depends on their cooperation, but member-state courts
may not always be interested in furthering the applicability of European law
when they address the ECJ. Beach (2001) gives the example of the Irish
abortion case (C-159/90) in which the issue was whether freedom of services
could undermine the Irish prohibition of abortion. Irish student associations
had published information about abortion clinics in the UK, which were illegal
under Irish law. In their defence, they referred to the freedom of services as

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

allowing them to disseminate information about services in other member
states (Beach 2001: 20; see also Nyikos 2006: 528–9). When the Irish High
Court addressed the ECJ, it was clear that it would not have followed a ruling
that undermined the Irish prohibition of abortion. In fact, the Irish Supreme
Court openly threatened non-compliance should the ECJ take such a decision
(Beach 2001: 21). The ECJ, therefore, in a Solomonic judgment, established,
in principle, the relevance of freedom of services for abortion but denied its
relevance in the case at hand. This gave Ireland the opportunity to include a
protocol on abortion to the ongoing negotiation of the Maastricht Treaty.
Supreme and constitutional courts lose power to the ECJ at the apex of the
judicial hierarchy. This has led to many conflicts between the ECJ and the
highest national courts, both historical and ongoing. For the highest national
courts, it cannot be taken for granted that the supremacy of European law will
be accepted. Thus, the Czech Constitutional Court declared a ruling of the ECJ
to be ultra vires in 2012 (Anagnostaras 2013). A notable—and widely cited—
critic of the ECJ is the German Constitutional Court (GCC). Only very
recently has the GCC posed its first preliminary reference (Case C-62/14
Gauweiler), questioning the competence of the European Central Bank
(ECB) to enact the Outright Monetary Transactions (OMT). Although the
GCC has made clear in its reference that it regards the decision of the ECB to
be ultra vires, in its judgment it largely followed the ECJ. Again, the ECJ faced
a difficult situation. On the one hand, it had already affirmed the euro rescue
plan in the Irish Pringle case. On the other hand, it risked provoking another
instance of open opposition from the GCC, as it had done in the early 1970s
when the ECJ did not deem the protection of basic rights in the Community to
be sufficient (B. Davies 2012: ch. 5). In its ruling on 16 June 2015, the ECJ
backed the ECB plan, but it at least followed the GCC in the assessment that
the ECB should be subject to judicial oversight.
In terms of the ECJ’s case-law development, conflict with the highest
member-state courts has been important. Notably, its establishment of its
own human rights’ jurisdiction can be seen as an answer to the objections of
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

The ECJ as a Political Actor 39

the Italian and German constitutional courts in the 1960s and 1970s. They had
threatened to subject European law and court rulings to judicial review for as
long as there was insufficient fundamental rights protection at this level
(Weiler 1986: 1119; B. Davies 2012).
For my analysis of the policy implications of case law, the interests of
private actors and lower courts in turning to an alternative legal setting is of
central importance. It is facilitated by the great detail contained in constitu-
tionalized case law based on a Treaty that provides much more differentiated
rules than national constitutions do.

The Court and the Masters of the Treaty

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

To what extent are member states able to guide the Court to follow their
wishes more closely? Member-state governments are the masters of the Treaty.
In discussions within political science, the discretion of the Court in its
relationship to the member states receives the most attention. Member states
shape the institutional rules under which the Court operates. Moreover, there
is quantitative data on the way the Court responds to member states’ obser-
vations in Court proceedings.5 As Kelemen (2016: 128) argues,
For the most part, the literature on interactions between member governments
and the ECJ has long since arrived at a general consensus that member govern-
ments set the outer bounds of how far the ECJ can push both its authority and the
scope of European law but that within these bounds the ECJ has substantial room
for maneuvering that it can use to promote deeper integration.
In relation to these outer bounds, I will first discuss how member states shape
the Court’s institutional foundations, before turning to the more difficult
question of how they influence its rulings.

Institutional foundations
In terms of their institutional foundations, there are convincing arguments for
why it would be difficult to show supranational courts what their politically
acceptable limits are. Principal-agent theory explains that agents’ discretion
rises with the number of principals that have heterogeneous preferences
(Hammond 1996; Tsebelis and Garrett 2001). And principal-agent theory
also tells us that the ECJ has a very broad sphere of action, as it has to please,

5
Formally, there is the distinction between interventions in direct actions, i.e. infringement
procedures, and those in observations to preliminary references. In the former, intervening
governments join the case. In the latter, this is not possible as the national court poses the
question (Carrubba and Gabel 2014: 63).
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

40 The ECJ and the Policy Process

in extreme cases, only a single member state in order to be immune to
sanctions (Kassim and Menon 2003; Tsebelis and Garrett 2001). The more
the EU grows, and the more heterogeneous it becomes, the greater the
discretion of the ECJ becomes. The EU polity’s decision rules are demanding,
requiring a qualified majority of member states’ votes for secondary law, and
unanimity to change the Treaty. High-order decision rules imply that a
minority of member states can lock court-induced changes in. Twenty-eight
member states are unlikely to agree on how to rein the ECJ in. The joint-
decision trap kicks in, which protects the Court (Scharpf 1988, 2006). In fact,
Larsson and Naurin (2016: 25–6) find that the Court is less responsive to
member states’ observations when an override requires unanimity.
However, member states not only have preferences concerning policy but

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

also in relation to the institutional balance of the European Union. Govern-
ments may turn against judgments that further their policy interests, as they
know that a court that is too strong may violate their own preferences in the
future (Moravcsik 1995). It is hard to predict whether member states will
follow more homogeneous institutional or more heterogeneous policy inter-
ests. From this perspective, the independence of the Court, stemming from the
heterogeneous policy preferences of member states, is limited by member
states’ homogeneous institutional preferences.
Expanding on the literature on principal-agent theory, Alter and Stone
Sweet have argued in favour of taking a ‘trustee’ perspective on courts (Alter
2008a; Stone Sweet and Brunell 2013). While they differ in their precise
argumentation, the general idea is that member states have to ‘trust’ their
courts and delegate independent power to them in order to safeguard the rule
of law and the independent third-party arbitration of courts. Accepting the
discretion of courts is part of this delegation: ‘A trustee is a particular kind of
agent, possessing the power to govern the rulers themselves. In the most
common situation, the trustee court exercises fiduciary responsibilities with
respect to the constitution, in the name of a fictitious entity: the sovereign
People’ (Stone Sweet and Mathews 2008: 86). After all, it is fundamental to the
balance of powers and the rule of law for courts to rule against executives
(Caranta 2008: 195). Otherwise, independent courts would have no use. From
this perspective, the ECJ pretty much does what it is intended to do.
At several times during treaty negotiations, the Court has become an issue
in debates, which shows that the trustee relationship is not without contention
for member states. But again, these instances can be taken as either the
exception that proves the rule that member states are generally content with
the case law of the Court or as the tip of the iceberg at points when govern-
ments finally had an opportunity to intervene.
The member states, thus, tried to constrain the implications of the ECJ’s
ruling in Case 262/88 Barber through a protocol to the Maastricht Treaty that
limited the retroactive effect of the ruling. In this ruling, the ECJ argued that
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

The ECJ as a Political Actor 41

Article 119 of the EEC Treaty (now Article 157 of the TFEU) also applied to the
equal treatment of women under private pension schemes. There were enor-
mous financial implications, as women working part-time were often excluded
from such schemes. Because the Court interprets the EU law as it stands, its
rulings generally elucidate how things should have been all along, implying that
its findings should take full retroactive effect. The reaction to the Barber ruling
can be taken as a clear sign to the ECJ that member states were not willing to
accept an activist court. Following the ruling, the Court was perceived to be
practising judicial self-restraint (Garrett et al. 1998: 166, 168; Dehousse 1998:
148–56). The UK had also brought far-reaching reform proposals into the
negotiations on the Maastricht Treaty, attempting to curtail the activism of
the Court. However, this received some support from Germany and France but

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

not the necessary backing from all member states (Tallberg 2000: 114–15).
In the context of the Lisbon Treaty, Austria successfully bargained for a
five-year moratorium in the conflict about German medical students studying
in Austria. Austria had tried to curb the influx of incoming German students
who had not found a place in German medical schools which had more
restrictions on the intake of students. However, the ECJ ruled that the prin-
ciples of non-discrimination and free movement prohibited Austria from
treating German students any differently from Austrian ones (C-147/03).
Austria would thus be forced to pay for part of German medical training
(Scharpf 2008: 91–2). Following this bargain, it was ensured that a quota of
75 per cent of medical student places had to be reserved for Austrians in
Austria for five years, which was later prolonged until 2016. Following another
ECJ ruling concerning French students in Belgium, the Court accepted
restrictions in cases where member states are able to prove a lack of profes-
sionals due to students returning to their home country (Damjanovic 2012).
A survey in 2014 showed that only 21 per cent of Germans who completed
medical studies in Austria stayed in the country to work.6
In terms of the strength of the Court, it is crucial to keep in mind that it
operates within a multilevel political system and that the ECJ receives direct
support from litigants and courts in member states. As I have argued, litigants’
support for alternative policies at the EU level is also significant for the electoral
considerations of member-state governments. And because of the preliminary
ruling procedure, member states are scarcely able to ignore European law.
By openly opposing it, they would not only question the independence of
the EU judiciary, which is so central to the rule of law, but also their domestic
court system.
Based on experiences in the US, Dan Kelemen has analysed why member
states have such difficulty using court-curbing mechanisms against the ECJ

6
‘Ausländische Medizinstudenten gehen fort’, ORF, Tirol, 1 Dec. 2014. <http://tirol.orf.at/
news/stories/2683711> [accessed 10 May 2017].
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

42 The ECJ and the Policy Process

(Kelemen 2012). As we have just seen, the joint-decision trap prevents member
states from stripping the court of its competences, changing its mandate, or
reducing its resources. Given that jurisdiction stripping is not really possible,
member states have signalled their criticism of the Court by keeping policy
areas outside of its jurisdiction, such as immigration and asylum in the
Maastricht Treaty (Kelemen 2012: 46). Interestingly, at Lisbon, immigration
and asylum was added to the Court’s jurisdiction. Member states had seen that
they need the Court to enforce agreements. This instance shows that, while
member states are critical of the Court, their overall assessment is still positive.
Drawing on an argument made for the national context that courts depend on
public support (Vanberg 2005), Kelemen argues that the extent of public
support (‘trust’) the Court enjoys, as measured by Eurobarometer data, shields

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

it from court-curbing measures. Even when comparisons are made to the
national judicial system, he finds high levels of support. Surprisingly, countries
that were struck by highly controversial rulings, such as the cases of Viking and
Laval that curtailed union rights, did not show a lowering of public support for
the ECJ (Kelemen 2012: 50). Kelemen also discusses whether member states can
attempt to influence the Court by appointing eurosceptic judges. Only euro-
sceptic countries are likely to do so, and Kelemen asks whether such judges
could make a difference. They would have to be the median judge in decision-
making, given the simple majority rule in the Court. Looking at the composition
of the different chambers at the ECJ, Kelemen concludes that it is very unlikely
that eurosceptic member states could make their influence felt by appointing the
median judge (Kelemen 2012: 53–4). Changing the underlying institutional
rules is always something of a nuclear option. Such an extreme reaction is not
necessary if the ‘masters’ of the Treaty are deferred to in the Court’s everyday
operation. We will now turn to this question.

The influence of the member states

Carrubba and Gabel (2014) recently argued in their quantitative study that the
Court largely follows the member states’ position. Courts ‘cannot successfully
push interpretations of international law that are inconsistent with underlying
government preferences’ (Carrubba and Gabel 2014: 49), they argue. Carrubba
and Gabel show that the Court responds to member states’ threats of non-
compliance, as becomes apparent in the balance of third-party government
briefs that are submitted. They analyse legal issues rather than court cases, as
cases often address several legal questions. By taking the opinions of the AG to
reflect the legal merit of the case (Carrubba and Gabel 2014: 97), they show
that the Court does in fact respond to the threat of non-compliance by
member states and does not decide in a certain way purely based on legal
arguments. Thus, Carrubba and Gabel (2014: 213) conclude that ‘Everything
happens conditionally on implicit government acquiescence.’
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

The ECJ as a Political Actor 43

Using similar but more recent data in an argument that refers to override
rather than non-compliance in a separation-of-powers model, Larsson and
Naurin (2016) arrive at similar conclusions that the Court is responsive to
member states’ signals. This does not mean determination, however, as the
results of their empirical analysis show that there is considerable room for
Court discretion. Although Carrubba and Gabel appear to emphasize the role
of member states more strongly, they also hold that ‘within limits, it is
perfectly possible for international courts to innovate and change the func-
tioning of an international organization’ (Carrubba and Gabel 2014: 49). Thus,
the difference between the various authors appears to be one of degree, as few
would say that governments’ preferences do not matter at all to international
courts. Alter emphasizes that it would be surprising to find that international

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

courts consistently violated governmental preferences (Alter 2014: 338). Inter-
estingly, Carrubba and Gabel find—counter to expectations that preliminary
rulings have a greater impact in the integrated European court system—no
difference in the significance of government briefs for ECJ rulings in infringe-
ment versus preliminary rulings (Carrubba and Gabel 2014: 164). Thus, they
see no evidence that the Court relies heavily on the support of lower national
courts to bring its verdict into the judicial systems of member states. What
could one add, from a qualitative viewpoint, to these quantitative analyses of
how governments’ interventions influence the Court?
Relating to actors, Carrubba and Gabel assume that governments rationally
intervene in all those cases that matter to them. Larsson and Naurin (2016),
however, emphasize that the Court has to operate with significant uncertainty
as to whether member states are seeking to override processes. Additionally,
case studies show that member states do not necessarily always intervene
when their interests are at stake. One example is Case C-292/04 Meilicke on
taxes, which I discuss in Chapter 6. In this example, the legal issue of dividend
taxation had already been discussed in Case C-319/02 Manninen, but the
German government had not intervened in Manninen, although it claimed
large tax losses of €5 billion in Meilicke. Thus, governments do not necessarily
intervene in rulings that affect them. It is possible that they are not aware of
the implications. Alternatively, they may hope that their infraction of EU law
will not come to light too soon if they keep quiet. Interventions in cases, thus,
do not adequately reflect political salience.
Moreover, the preferences of governments are somewhat fluid. Not only is it
difficult for governments to assess the consequences of all possible turns that a
ruling can take, but it is also difficult to foresee how a ruling could favour or
hamper future policy choices. In the European multilevel system, the European
legal system offers societal actors the option to turn to an alternative legal
framework. For governments, this means that they can observe which of the
policy options are preferred by domestic actors, which may be important
constituents for them. Moreover, within coalition governments, there are
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

44 The ECJ and the Policy Process

frequently differences in policy preferences. As the EU often serves the
purpose of avoiding blame, not all rulings which imply shifts in domestic
policy are necessarily unwelcome. It is important to note that even if govern-
ments concur with court-driven changes, this does not imply that these
changes would have been realized without the influence of the Court. In
relation to government briefs then, we cannot assume that all required policy
changes are brought to the attention of the Court (cf. Larsson and Naurin
2016). Figure 2.1 summarizes the arguments for the weakness of courts.
Although qualitative studies often find that many rulings contradict the
observations and policies of a high number of member states, which indicates
that the Court has a large amount of leeway and the member states have little
control (Cichowski 2004: 496), building on Conant (2002), Martinsen (2015)

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

questions the Court’s power, emphasizing the political constraints under
which it operates. Whereas Carrubba and Gabel, in addition to Larsson and
Naurin, do not observe compliance or override as such, but rather analyse how
the Court’s rulings respond to member-state governments’ interventions, in a
qualitative study this interaction between rulings and legislation can be traced.
In a very systematic way, Martinsen compares the impact of case law on the
legislative process in three social-policy case studies: the Working Time
Directive, the Patient Mobility Directive, and the failed Monti II Regulation.
She differentiates between four different legislative reactions: codification,
modification, non-adoption, and override. Only codification is taken to indi-
cate significant judicial impact. Martinsen’s research interests are similar to
my own, given her focus on judicial and legislative policymaking. By focusing
on EU legislation, Martinsen’s work addresses an important gap in the liter-
ature that neglects this interaction. Martinsen’s starting point is, however, that
the ECJ is much weaker than is assumed in the literature, whereas I would
argue that its actual impact has been neglected.

Reactive actors

• Courts need to be addressed, as they do not have own initiative

• Are dependent on government compliance with their rulings

Member states’ oversight

• Court’s mandate subject to political control

• Political appointment of judges
• Common interest of governments to avoid activist court

Figure 2.1. Weak courts

 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

The ECJ as a Political Actor 45

How can these different interpretations be explained? I would like to point
to two reasons here: Martinsen’s neglect of the importance of over-
constitutionalization and her expectation that Court rulings are able to deter-
mine policies. First, by considering pure secondary case law (the Working
Time Directive) to have the same level of importance as her other two case
studies, Martinsen does not examine whether member states can revise the
underlying directive or regulation or whether they would need to alter the
Treaty. The Working Time Directive, which is known for its opt-out provi-
sion, is, additionally, a case that cannot tell us much about constraints from
the Court. Member states can simply opt out of the directive. Secondly, there is
attention needed to the specifics of case law, which only gives directions to
solve very distinct disputes. In analysing judicial influence, Martinsen expects

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

the full endorsement of Court rulings in legislation, arguing otherwise that
legislators did not ‘abide’ (2015: 87, 122) by the Court’s rulings. For Martinsen,
as soon as the influence of the Court is mediated, this is taken as a sign that
politics has the upper hand. However, as case law is so sketchy, it normally
has to be complemented by codification. Martinsen wrongly assumes that
Court rulings can devise whole policies that consist of general rules. Not taking
the partial nature of case law into account leads to an overemphasizing
‘modification’, and with it political influence, thereby downplaying the im-
portance of the Court.
There are two further qualitative findings that are interesting within the
context of the influence of member states on the Court’s rulings. Maduro
(1998) analyses the case law on the free movement of goods and argues
that the Court pursues a majoritarian activist position on the basis of
information on the regulatory status quo in member states. He believes
that the Court is only prepared to strike down member states’ regulations if
these are not in line with the regulations of the majority of member states.
A kind of judicial harmonization process results from majoritarian activism.
There has been no parallel study of the other freedoms to test this argument to
my knowledge. In fact, we will see when we discuss the Court’s tax case law in
Chapter 5 that authors explicitly argue against Maduro’s view (Graetz and
Warren Jr 2006: 1193). This could also mean that the Court has become less
cautious since the 1990s. An indicator for this is that some authors deny that
the Court still pursues a majoritarian approach in other areas of case law
(Barnard 2009: 598–9). This may mean that additional considerations have
become stronger for the Court, such as its wish to improve the legitimation of
the EU through a strengthening of individual rights (cf. Cichowski 2013).
Obermaier has argued, on the basis of the case law on patient mobility, that
the Court ‘fine-tunes’ its rulings over a succession of cases. This allows the
Court to take into account member states’ reactions, thereby adapting its case
law to their institutional conditions. Consequently, he argues that the impli-
cations of case law remain marginal (Obermaier 2008b: 746, 751). As we turn
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

46 The ECJ and the Policy Process

to tracing case-law development in Chapter 3, we will see how important it is
for the Court to develop case law over a series of rulings. There may be seminal
judgments that stand out, but as single judgments cannot provide general
rules, they are typically embedded in a group of similar cases that collectively
delineate the scope of European law. Within such a series of cases, fine-tuning
implies that the Court takes into account member states’ considerations,
aligning its rulings further with their concerns. We could distinguish it from
an extreme reaction that we might call ‘reversion’, where the Court implicitly
or explicitly overturns its earlier rulings. We could thus see a continuum of
case law between activism, where new principles are established that encroach
upon member states’ competences, fine-tuning, where the limits of the reach
of such activist case law are elucidated, and reversion, where the Court, in light

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

of opposition from member states, reverts to a previous judicial position, re-
establishing member states’ competences.
Placing fine-tuning in such a continuum emphasizes a general feature of
these processes and leads me to an interpretation that differs from Obermaier.
For the implications of case law to be marginal, there would need to be
reversion, since, otherwise, the seminal, activist case lives on. Fine-tuning
only clarifies the scope of the conditions for case law’s relevance. The general
point to keep in mind is that, because of its erga omnes effect, activist case law
does not only affect a single legal dispute. Its implications continue to be
significant, in particular if case law interprets the Treaty. These implications
may concern the immediate duties of implementation. But the constitutional
nature of case law also means that certain policy options at the domestic and
European levels are foreclosed. When debating the Services Directive, for
instance, national treatment was no longer an option given past case law.
The ECJ has a significant influence on non-decisions (Bachrach and Baratz
1962) in the way it closes off policy options that were previously available. And
the Commission often enshrines such case law in proposals for secondary law,
as will be analysed in Chapters 4 and 5.
The fact that case law is not only relevant for the individual case but lives
on also leads us to reconsider quantitative analyses of case law. In general,
quantitative analyses do not allow authors to differentiate between the impor-
tance of issues. Thus, if a ruling involves three legal issues, many member
states intervene, and the Court consequently backs down on two of the
three issues, it may be the case that these two issues are comparatively
insignificant. Significant changes may result from rulings on a few issues in
which the Court decides against the member states. Moreover, we could even
question the findings on purely quantitative grounds. We could say that,
whether the Court issues one activist judgment in ten cases, or only one in a
hundred, as long as there is no reversion, a step taken towards more integra-
tion is still made. In this respect, the high case load that results from the
indirect access of private actors is significant, and member states cannot
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

The ECJ as a Political Actor 47

Multilevel system/compliance
Over-constitutionalization
constituency

• Detailed policy goals • Governments do not control access to

• Unanimity for Treaty change Court
• Heterogeneous MS preferences, • Support of lower courts and litigants
joint-decision trap • Fluid governmental preferences
• Incremental development of case law mediated by interest in blame
driven by path dependence, legal avoidance
uncertainty, and the telos of ‘ever • Governmental interest in rule of law
closer union’ shields court
• Constraints on domestic and EU
policies (‘non-decisions’)

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Link to legislation

• Commission generalizes case law in legislative proposals (codification)

• Member states conform to or ignore/resist case law at national level

Figure 2.2. The strength of the Court

control this access. Figure 2.2 summarizes the arguments for the strength
of the Court.
Where does this leave us? The Court can be shown to be responsive to
member states’ interventions. But we cannot judge to what extent it is lenient
towards insignificant issues while it pushes through more significant ones.
Member states may also fail to interfere reliably in relation to their prefer-
ences. There may be overriding interests in the EU’s legal system, conflicting
preferences within government, or insecurity about future voter preferences.
A high case load allows the Court to incrementally develop case law.
In Carrubba and Gabel’s (2014) view, this evolves in line with governmental
preferences. Larsson and Naurin (2016) show that override matters, but
accord the Court significant discretion in furthering integration. According
to Alter (2014), international courts have much greater leeway. Disagreement
among these positions is, however, a matter of degrees. All fall short, however,
of emphasizing the constitutionalization of case law, where even infrequent
activist rulings go a long way.

CONCLUSIO N: S HORTCO MIN GS IN TH E L ITERATURE

As we have seen, there is a significant amount of research on multiple aspects of

how the ECJ functions. It is broadly recognized that the Court is an important
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

48 The ECJ and the Policy Process

motor of integration, while it remains a matter of debate to what degree the
Court can have an impact that contradicts the preferences and intentions of
member states. As my starting point, the discussion so far has largely focused
on the relationship of the judiciary to member-state governments. If the
implications of the ECJ’s case law for policymaking are analysed, this is
done with the view that governments have the means to mediate its impact.
The expectation here is one of ‘automatic policy changes’ (Conant 2002: 45),
which thereby interprets any alterations as ‘modification’ (Martinsen 2015)
and, therefore, as a counterbalance to the Court. The impact of case law is not
sufficiently captured by analysing whether and how member states respond to
single rulings. Case law becomes part of the Treaty and thereby influences
future secondary law, as well as member states’ remaining policy options. This

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

is the topic of this book. Even if we find that the Court is essentially a reactive
actor that takes heed of and aligns itself with governments’ opinions, this does
not mean that its impact is largely marginal.
In this respect, the over-constitutionalization of the Treaty plays a decisive
role. As an international treaty that fosters economic integration, primary law
in the EU is at once very detailed and only partially elaborated in comparison
with national constitutions. Though periodical Treaty revisions have broad-
ened rights, this has not fundamentally altered the thrust of economic inte-
gration in the Treaty. The Treaty favours economic rights over others and
supports liberal rather than republican ideas about legitimation (Scharpf 2009).
More importantly, despite the wide recognition that the European Treaties have
fundamentally changed their character over a long period through the doctrines
of direct effect and supremacy, there has been no discussion about reducing and
simplifying the content of the Treaties in order to regain flexibility at the level of
EU secondary lawmaking, as well as at the level of the member states, which are
bound by the supremacy of EU law.
If the meaning of the constitution is altered in one ruling, this change may
go a long way, even if the Court submits to member states’ preferences in
several rulings that follow. Courts need to respond to political pressures and
uphold the independent rule of law. Case-law development unfolds in a path-
dependent way. Its incremental, case-specific development makes its implica-
tions uncertain. Member states’ positions may not be the single best indicator
as to the impact of ECJ case law. Not only are governments not really able to
foresee the implications of single rulings, they also have somewhat fluid
preferences, as actors that are important constituents litigate for policy change
at the European level.
Thus, focusing on member states’ preferences can only partly explain the
Court’s behaviour. Its relationship to its supporting coalition of lower courts,
litigants, and the legal profession is also relevant here. In particular, we also
have to account for the importance of legal arguments, as it is implausible that
case law is only determined by the preferences of judges and member states.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

The ECJ as a Political Actor 49

Should this idea hold, there would be no explanation for why rational actors
would aim to influence the policy process in the first place, as the meaning of
the resulting laws would be irrelevant in subsequent court proceedings.
This chapter has shown that research on the ECJ is preoccupied with
juxtaposing governmental control of the Court and judicial activism, analysing
these two options according to degree and therefore largely failing to capture
the real relevance of the Court in terms of its influence on material policy. In
order to understand this influence, it is necessary to analyse the development
of case law by the Court in the next chapter.

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Case-Law Development between Path

Dependence and Legal Uncertainty

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

This book argues that research on the European Union (EU) has neglected the
policy implications of European Court of Justice (ECJ) case law. Although
the importance of the ECJ in developing EU legal doctrine is recognized, the
literature largely overlooks how this directly impacts policymaking at the EU
and member-state levels. Implicitly, the constraints that the ECJ places are
compared to the constraints that national constitutional courts pose. However,
one finding of this book is that the rule of law and independent courts have a
different quality at the supranational as opposed to the national level. The
extent of detail within the material rules of the constitutionalized European
Treaties is underestimated (Grimm 2015).
In comparison to national constitutions, the EU Treaties do not predom-
inantly focus on rules of state organization, setting out the structure of the
polity, and sets of individual rights. Rather, we find quite detailed rules in
relation to different economic freedoms and competitive markets. The case
law pertaining to these rules complements the Treaties and results in a dense
web of material rules that mainly focus on market regulation, but increasingly
also on social rights. It is true that modern constitutions are characterized by a
growing set of individual social rights, which aim to protect not only individ-
ual liberty and political participation but also a decent quality of life, in terms
of rights to shelter and labour. But the extensive individual social rights of
modern constitutions do not prescribe which policies to pursue as the Court
does in its case law.
This chapter presents an explanatory framework for understanding the
case-law development of the Court. My starting point is that the EU needs
to be conceptualized as a multilevel system of governance (Jachtenfuchs 1995;
Marks et al. 1996). One consequence of this system is that governments are no
longer gatekeepers to the EU polity. They remain important players, but
different actors may seek to further their interests at the EU level, rather
than within a domestic context. Governments acting in such an environment
only partly have a set of fixed preferences. They profit from the multilevel
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Case-Law Development 51
system in different ways, including through blame avoidance (Weaver 1986).
They can observe different policy solutions across the EU. Additionally, they
get to know which domestic policies their prospective voters will attempt to
abolish through litigation. We can, therefore, expect governments not to
intervene reliably in the ECJ when they fear a significant impact from ECJ
rulings. They do not always know which policies they will support in two
years’ time when the ECJ gives its judgment.
The ECJ itself is committed to an ‘ever closer union’. In this chapter, I argue
that the detailed constitutionalized Treaty rules, the multilevel character of the
EU, and the incremental progression of case-law development combine in a
way that characterizes this development as one that is simultaneously marked
by legal uncertainty and path dependence. It is important to emphasize from

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

the start that these terms are in a tension and appear contradictory. I argue
that these concepts are helpful for understanding case-law development in the
EU, which manages to bridge this exact contradiction. The Court takes a
difficult task upon itself. On the one hand, it has to foster an integrated court
system where lower national courts apply unfamiliar European law alongside
national statutes. This requires stability and predictability in European law.
On the other hand, the Court has to provide impetuses for further integration.
It needs to be responsive to a community of EU lawyers that has long
continued to present legal concepts for advancing integration. Litigants pick
up these ideas and provide further impetus. Reconciling these demands for
stability and change is possible, I argue, by pursuing, on the one hand, a path-
dependent interpretation of the Treaty oriented towards precedent. The
necessary dynamism, I argue, is provided, on the other hand, by legal uncertainty
concerning the actual reach of EU law.
The extent to which the Treaty grants legal positions is evolving and
uncertain. Legal uncertainty arises from the unclear implications of the con-
stitutionalized Treaty overlapping with domestic legal orders and the balanc-
ing approach that the ECJ employs to mediate between the different legal
orders. In reconciling the demands of integration and the legitimate policy
interests of the member states, the Court employs a proportionality test,
implying that the specific outcome of such balancing is uncertain. Legal
uncertainty also results from case law’s focus on the situation of the affected
member state, with uncertain implications for similar but slightly different
rules in other member states. Additionally, when the Court gives guidelines on
interpretations in preliminary rulings, these do not amount to clear policy
prescriptions, for which the Court is not legitimated. Moreover, reading
unequivocal requirements out of the Treaty would conflict with the Court’s
interest in providing continuing impetuses for integration.
The path-dependent development of case law complements its uncertain
implications for the EU and the domestic legal orders. Path dependence
reflects an emphasis on precedent and a unified approach to interpreting the
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

52 The ECJ and the Policy Process

four freedoms of goods, services, persons, and capital, as I will show. Legal
innovations in one area, creating a new path, may spread to other areas as
litigants adopt new arguments that are beneficial to them. Litigants provide
positive feedback, but how the Court rules cannot be predicted. Though
seemingly contradictory, both path dependence and legal uncertainty thus
characterize case-law development.
In the following section, I will start by discussing the distinctive approach to
integration in the EU, underlying the extensive reach of EU law. I will then
discuss legal uncertainty. Laws are incomplete contracts that are open to
interpretation and development of their meaning. To understand how case
law can impact legislation, it is necessary to show how in its development it
can acquire autonomy from the control of political actors. In this chapter,

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

I aim to explain why the jurisdiction of the ECJ could evolve in such a way,
proving that political efforts to counteract the development are insufficient.
In light of over-constitutionalization and the detailed policy aims of the
Treaty, case-law development is an important determinant of policymaking,
alongside the political preferences of legislative, executive, and societal
actors. Whether the Court follows these arguments, and which consequences
result from policies at the EU and member-state levels, remains to some
extent uncertain. In developing this argument, I will summarize case-law
development for the four freedoms in the hope of further familiarizing
readers within political science with this important and overlooked deter-
minant of EU policymaking.
My explanation is the basis for discussing the impact of case law on
policymaking at the EU and member-state levels in the following chapters.

THE DISTINCTIVE APPROACH TOWARDS

IN TEG RA TION IN TH E EU

The EU is not only about concerns with trade among member states. To
understand the impact of ECJ case law on policymaking, it may be best to
begin with what may be a common misunderstanding. In many quantitative
analyses of case-law development, case load at the Court is related to trade. For
instance, in a major contribution to the field, Stone Sweet and Brunell take the
growing case load at the ECJ as an indication for a growth in transborder
conflicts that are related to trade (Stone Sweet and Brunell 1998; Stone Sweet
and Caporaso 1996; cf. Schepel 1998). This supports their neofunctionalist
explanation of European integration as based on private actors acting in
unison with supranational ones. In addition, Carrubba and Gabel consider
how trade flows change and show that rulings have a long-term effect on trade
(Carrubba and Gabel 2014: 178). This assumes that trade is the main reason
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Case-Law Development 53
that litigants address the Court. However, Court cases only partly deal with
transborder disputes (Alter 2000: 500).1
The focus on trade could reflect a misunderstanding rooted in experiences
in the US, which follows a different philosophy. The EU legal order substi-
tutes the national order, rather than simply complementing it, as is the case in
the US, where interstate commerce adheres to different rules (Barnard 2009).
European harmonization seeks to regulate state and transborder activities
alike. Though the US interstate commerce clause directs federal laws to
questions of interstate rather than state commerce, it has been interpreted
along different lines across the decades. We can see, for instance, that the
freedom to provide services is much less developed in the US than it is in the
EU. This concerns highly regulated sectors, such as the insurance sector, as

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

well as basic services, such as hairdressing (Schelkle 2017: 250; Egan 2015;
Goebel 2002; Pfennigstorf 1987). In the EU, even though it would often
have been easier to agree on rules that were directed exclusively towards
transborder activities, the Commission has generally favoured an integrated
approach (Troberg 1997b: 1414, no. 32). The advantage is that companies
that wish to take up transborder activities in the single market do not have to
comply first with additional regulatory requirements. This would burden the
single market and, therefore, is something the Commission seeks to avoid
(Miersch 1996: 137–8). Thus, the comprehensive coordination approach,
which goes beyond merely facilitating interstate commerce, actually results
in a high degree of privileging of single-market activities. In order to facilitate
an at times very low percentage of market activity, all regulation is changed
and all market activities have to adapt. A vivid example is regulation 924/2009,
which requires the use of the international bank account number (IBAN) for
all transactions in euros, although only a fraction of these are transborder.
The EU legal order does not complement but substitutes the legal orders of
member states.
This comprehensive order, first of all, is concerned with the harmonization
of laws in EU legislation. There is a certain tension between the latter and the
four freedoms, which formally exclude situations that are ‘wholly internal’
(Shuibhne 2002) to a member state. The four freedoms require a cross-border
element to become relevant. However, the broad interpretation of the free-
doms through the ECJ, following the Dassonville formula (see further on) and
targeting all domestic rules that ‘are capable of hindering, directly or indir-
ectly, actually or potentially, intra-Community trade’, has little regard for the
autonomy of member states. Barnard argues that this contrasts with the non-
discriminatory approach of the US Supreme Court that only intervenes in

1
Conant also criticizes this point, not in relation to the comprehensive approach, which is
discussed here, but rather the fact that non-trade-related references (such as those concerning
social policy) are included in the data set of Stone Sweet and Brunell (Conant 2001: 114).
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

54 The ECJ and the Policy Process

favour of interstate commerce if state rules discriminate against interstate
trade (Barnard 2009: 586–90).
There are, thus, differences between the US and the EU, and within the EU
between harmonization and the direct judicial application of the four free-
doms. The broad interpretation of the four freedoms implies, however, that a
potential effect on trade is sufficient. As the ECJ is called upon to interpret
primary and secondary EU law, it follows that case law is not only rooted in
economic interdependence. Often private actors turn to EU law in purely
domestic cases, seeking to benefit from a more favourable legal order.
Empirically, this may be demonstrated through an analysis of all cases at the
ECJ between 1984 and 2011 that dealt with insurance services and can be
researched in the EUR-Lex database using. Insurance services are subject to

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

the freedom to provide services and, additionally, there are different EU
insurance directives regulating the sector. The ECJ cases are distinguished
according to whether they are infringement procedures (thirty-four) or pre-
liminary rulings (twenty-nine). We then checked whether the dispute involved
a cross-border aspect, including foreign-owned insurance companies. As the
Commission often initiates cases on behalf of complainants, infringement
procedures were taken to imply a cross-border conflict, with the exception of
those cases that unambiguously deal with purely domestic issues. The analysis,
therefore, overstates transborder conflicts, as all procedures that deal with
implementation deficits are included as well. Nevertheless, we found thirty-
four cases that dealt with transborder conflicts against twenty-nine cases that
dealt with purely domestic issues. This shows how often European law is used to
target domestic regulatory changes.
The focus of the literature in this area on interstate trade effects tends to
overlook the far-reaching policy consequences of the EU legal order. The fact
that the background to much important scholarship is the United States may
explain the relative neglect of the policy implications of case law within the
literature. I will now turn to a demonstration of how legal uncertainty within a
context of path dependence provides a framework for the dynamic develop-
ment of case law.

LEGAL UNCERTAINTY

As was noted in the introduction, I take legal uncertainty to imply that the
precise meaning of law cannot be predicted. To provide legal certainty is one
of the core functions of the rule of law. At the same time, legal certainty can
never be absolute. Legal scholars often speak of the indeterminacy of law
(Stone Sweet 2004: 32, 34, 38), also at the national level. Treaty articles
cover general topics with comparatively few sentences, implying that their
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Case-Law Development 55
precise implications are—to an extent—open. Rather than speaking of legal
indeterminacy, which would connect my argument to a related but essentially
different discussion about the theory of law by legal scholars (Everson and
Eisner 2007: 53), I prefer the term legal uncertainty. The justification for this
is, first of all, disciplinary. The aim of this book is to pay attention to the
importance of the judicial system for analysing policy processes. For this,
the term legal indeterminacy comes with a lot of legal theoretical baggage,
while the term uncertainty is well established in political science, for instance
when analysing decision-making under conditions of uncertainty (Streeck and
Thelen 2005). Uncertainty also seems to be the preferred term in light of the
widespread use of rational choice theory, and the problems of bounded
rationality and incomplete information. Moreover, Beck (2012) has put the

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

term at the centre of his critical analysis of the Court’s legal reasoning.
If a degree of legal uncertainty characterizes any legal order, why should
legal uncertainty be a promising focus of analysis at the supranational level?
Sources of legal uncertainty can be detected, first of all, in the legal texts
themselves, particularly as we are dealing with supranational law, which is
full of compromises. Secondly, the way the Court adjudicates adds to legal
uncertainty. Thirdly, the context of the multilevel system heightens the ambiguity
of the implications of EU law. These sources of uncertainty will be discussed
in succession.

Legal Texts as a Source of Legal Uncertainty

Case law develops within the context of laws that are necessarily incomplete
contracts, since they cannot cater for all eventualities: ‘The idea of the incom-
plete contract is basic to a wide range of approaches to delegation and to
courts. Generally, contracts are said to be “incomplete” to the extent that there
exists meaningful uncertainty as to the precise nature of the commitments
made’ (Stone Sweet 2004: 24).
Legal uncertainty, Beck (2012: 5) argues, ‘is a necessary feature of all legal
systems’, but its extent varies. As was argued in the introduction to this book,
there is a continuum between legal certainty and legal uncertainty. Within the
EU, legal uncertainty is pronounced. For one thing, arguments that have been
made for the international context apply here. Member states are forced to
draw up incomplete contracts that are flexible enough to deal with future
contingencies, so legal uncertainty becomes a necessary condition for integra-
tion (Abbott and Snidal 2000: 433, 456). Goldstein and Martin argue that
governments may not be able to enter international obligations without a ‘veil
of ignorance’ over the precise future consequences of international regimes, in
order to prevent the mobilization of domestic lobbying groups (Goldstein and
Martin 2000: 606). Similar arguments of incomplete contracting have been
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

56 The ECJ and the Policy Process

made for national constitutions (Stone Sweet 2000: 44). Uncertain rules, thus,
facilitate international cooperation.
The logic of political decision-making in the EU further demonstrates that
legal uncertainty appears to be a much more significant problem at the
supranational level. Decision-making in the European Union is hampered
by the heterogeneity of the member states’ interests as well as demanding
decision rules. Decisions require a qualified majority or unanimity, in addition
to the involvement of the European Parliament. Due to common compro-
mises and incomplete legislative specifications, there is great need for legal
clarification by the Court (Everling 2000: 221). The extent of national hetero-
geneity is distinctive at the European level. Member states’ heterogeneous
readings of legal compromises are backed by the value pluralism that they

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

represent (Beck 2012: 175–6).

Adjudication as a Source of Legal Uncertainty

The ECJ and the Commission have to absorb the legal uncertainty that stems
from the partly empty or inconclusive compromises and the incomplete
contracts from the Council of Ministers and the European Parliament (EP).
Given the indeterminacy of legal texts, the Court enjoys a certain degree of
freedom when interpreting European law. Stone Sweet calls this the zone
of discretion (Stone Sweet 2004: 26). The Court’s ‘discretion increases broadly
in proportion to the extent of legal uncertainty in the norms relevant to the
specific legal problems it is asked to resolve’ (Beck 2012: 8). Legal uncertainty
resulting from intergovernmental problems of agreement offers a chance for
supranational actors to strengthen their own position by opting for interpre-
tations that enhance their own competences. Consequently, supranational
actors can be assumed to have their own interest in legal uncertainty, which
serves them as an opportunity structure.
Self-interest can explain why legal uncertainty is not only absorbed but
indeed also generated by supranational actors. In this context, it is relevant to
note how rapidly the case law of the Court has been developing, as integration
in the EU has matured quickly (Beck 2012: 237). Were the ECJ to define the
limits of national and European competences in its judgments once and for all,
it would hamper its ability to adjust future rulings towards the course of
further integration. It is easier to maintain this capacity if some ambiguity
remains in the ECJ’s judgments, allowing the provision of positive impetuses
later. Otherwise, progress towards further integration would hinge exclusively
on the decisions of the Council and the Parliament regarding secondary law.
In fact, Kranenpohl reports a similar finding from his interviews with German
constitutional judges, who state that judgments have to remain ambiguous in
order to allow for appropriate judgments in the future (Kranenpohl 2009).
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Case-Law Development 57
If even national constitutional judges see themselves as being faced with
contradictory demands to provide stability while remaining flexible enough
for future imponderables, this must be all the more relevant within a dynam-
ically developing and expanding union.
However, it is not only self-interest that leads supranational actors to
generate legal uncertainty. Legal uncertainty also derives from the nature of
case law. Courts make decisions on contentious issues. In order to do so,
they must interpret rules. If these rules do not determine the issue at hand,
judges use a method of interpretation to adapt the existing rules to their
current problem. There are different methods of interpretation, but the ECJ
gives particular weight to the teleological method that is oriented towards
the goal of further integration (Pescatore 1983). As courts are only legiti-

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

mated to adjudicate and not to legislate, case law develops incrementally by
slowly closing existing gaps in the law (Shapiro and Stone Sweet 2002: 91–2)
and always focusing on the specific problem of interpretation arising from
the case. The legitimation of judge-made law is significantly bound by its
incremental character. Typically, case law is originally developed in a series
of cases, which incorporate new legal principles in a rather unclear way. It is
also common to establish a new interpretation but to deny its relevance in
the case at hand (Alter 1998: 131). This mode of development makes it
possible for the Court to get to know the reactions of national courts, legal
commentators, and important political actors, allowing it to modify its
rulings in the face of strong dissent (Bleckmann 1983: 81). By proceeding
in such an incremental fashion, sudden changes to the course of legal
interpretation can be avoided, thereby defending the consistency of the
legal system (von Bogdandy 1995: 25).
The openness of legal texts, which provides a zone of discretion for courts,
is, thus, incrementally closed. Precedent serves to orient and to legitimate
jurisprudence: ‘precedent performs its most significant function as a modality
of argumentation. In performing this function, it helps to limit the range
of possible decisions and permissible arguments’ (Gerhardt 2005: 909). By
acknowledging precedent, judges avoid appearing to rule arbitrarily or polit-
ically (Jacob 2012). Precedent fosters path dependence and provides stability,
as I will explain later. It also supports the Court’s autonomy (De Somer 2016).
However, most case law includes some ambiguity. ‘Precedents thus create
their own legal uncertainty’ (Beck 2012: 92). Moreover, the Court does not
really signal when it changes its interpretation (Conway 2012; Beck 2012: 255).
Derlén et al. (2012: 518) argue that, in contrast to common-law countries, the
ECJ has no known method of establishing its case law and ‘nor does it state
when and why it deviates from existing case law’. This makes it impossible to
know which direction the Court is likely to take in a case. There is a tradition
of relatively short judgments with few explicatory justifications, in which
earlier case law is only cited when it backs the ruling (Derlén et al. 2012: 519).
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

58 The ECJ and the Policy Process

In relation to the specific reasoning of the ECJ, the importance it gives to
the proportionality principle also has to be mentioned. Typically, propor-
tionality analysis, involving four steps, is used to assess whether a govern-
ment intervention into a right is lawful. The Court asks whether the
intervention is legitimate (1), in other words, whether the government was
authorized to do it. The suitability test (2) questions whether the policy
objective and intervention are related. The necessity test (3) considers
whether it is the least restrictive measure. Finally, the proportionality test
(4), in the narrow sense, assesses the overall costs and benefits of the intervention
(Stone Sweet and Mathews 2008: 76). As a principle, proportionality has
German origins and has been developed since the late eighteenth century
(Stone Sweet and Mathews 2008: 98–112). With the establishment of a strong

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

constitutional court in the Federal Republic after the Second World War,
proportionality analysis was further elaborated and diffused to other coun-
tries, as well as being adopted by international courts (Stone Sweet and
Mathews 2008; Knill and Becker 2003). Its increasing relevance is tied to
the constitutional courts’ growth in importance and to constitutionalization
processes more generally. Proportionality analysis allows for the managing of
conflicting goals. Assigning this analysis to courts places them in the role of
ultimate arbiter. Importantly, there are few right or wrong decisions in
proportionality analysis, but only the question of which court is competent
to do the balancing (Wahl 2006: 491). There is ‘near complete freedom
enjoyed by the judiciary in balancing’ (Beck 2012: 87). In this context, it is
important to remember that the constitutionalized Treaty is special in terms
of the extent of detailed rules that it prescribes. This means there is a greater
need to balance between different rules. Beck (2012: 77) also emphasizes that
the absence of a hierarchy of norms aggravates legal uncertainty in the
EU. ‘As a mode of judicial governance, PA [proportionality analysis] casts
a deep shadow on the lawmaking of non-judicial actors, while providing
judges with a flexible means of managing sensitive legal questions in poten-
tially explosive political environments’ (Stone Sweet and Mathews 2008: 162).
This adds uncertainty, and fuels litigation and case-law development. Stone
Sweet and Mathews (2008: 142) call proportionality analysis the most im-
portant institutional innovation of the ECJ after supremacy and direct effect.
At the same time, proportionality analysis is ‘the most intrusive form of
review’ (Stone Sweet and Mathews 2008: 80), leaving less autonomy to other
governmental branches than the strict scrutiny test, for example, which the
US Supreme Court employs. As Martinsen (2011) shows, control of the ECJ
by proportionality has increasingly led to EU law even interfering in member
states’ welfare and asylum policies, although governments have always been
keen to keep these policies outside the reach of the EU as much as possible.
The proportionality test, moreover, is another example of a rule that the ECJ
introduced through case law that later became codified. Article 52(1) of the
 OUP CORRECTED PROOF – FINAL, 6/12/2017, SPi

Case-Law Development 59
EU Charter of Fundamental Rights refers to it (Stone Sweet and Mathews
2008: 141).

The Multilevel System as a Source of Legal Uncertainty

A final cause of legal uncertainty is the multilevel character of the European

political system. European law is superimposed on the different national legal
orders of member states. Accordingly, case law has different national impli-
cations in the member states, and it is often hard to predict what a preliminary
ruling will mean for other member states (Hatzopoulos 2002: 728). National
rules may be similar, but they differ in some—possibly decisive—ways.

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Another important consequence of superimposing a set of rules on a variety
of circumstances in different member states is that the resulting diversity gives
private actors ample ideas for possible litigation strategies, under the precon-
dition that they have legal standing. They can try to transfer legal arguments
across member states, as well as across different policies and fundamental
freedoms. Finally, part of the multilevel system is its multilingual nature. Legal
texts exist in the different languages of the Community, with multilingualism
adding to legal uncertainty as different meanings are conveyed (Beck 2012:
174). Figure 3.1 summarizes the different sources of legal uncertainty.
The rule of law has to provide legal certainty, taking into account that this
cannot be absolute. As I have given several reasons why legal uncertainty is
heightened in the EU, I now aim to show that path dependence provides the
necessary stability, drawing on the seminal work of Stone Sweet (2002).

• Law as incomplete contract

Law • Multiple veto points result in empty
compromises, which are subject to
interpretation

• Ever closer union and expansive Treaty

interpretation
Court • Self-interest in legal uncertainty to maintain
autonomy and keep case-law development open
• Balancing according to the proportionality test

• Indirect access, benefits of an alternative

legal order, potential of analogous
Multi-level system reasoning with path dependence
• Transfer of reasoning from similar
situations of other MS

Figure 3.1. Different sources of legal uncertainty

OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

60 The ECJ and the Policy Process

In the following section, I begin by showing how the concept of path
dependence can be applied to case-law development by focusing on the four
freedoms, as explained in Chapter 1. I then go on to show how the path
established for interpretation of the freedom of goods was incrementally
transferred to the other fundamental freedoms of services, persons (workers,
establishment), and capital. The resulting extensive interpretation of the
freedoms significantly constrains policymaking at the EU and national levels,
as I will show in later chapters.

APPL YING THE CONCEPT OF PATH

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

DEPENDENCY TO CASE L AW

There are four attributes of path-dependent processes (Beyer 2005). First, the
beginning of a new path is contingent. It has often been noted that historical
accidents provide critical junctures whence a new path develops (Capoccia
and Kelemen 2007). Secondly, the sequence of actions is important. Thirdly,
positive feedback stabilizes the path, making a path reversal increasingly
unlikely (Mahoney 2000; Pierson 2000). Fourthly, and last, a ‘lock in’ is the
result, even though the institutional solution that characterizes the path may
be inefficient. Developments that are particularly long term are often analysed,
such as Pierson’s explanation of the welfare state as path-dependent (Pierson
2000). A particularly famous example is that of the QWERTY keyboard for
typewriters. Even though it is considered to be less suited to typing quickly
than alternative keyboard settings, it has remained the most prevalent setup,
and once it was established, it could not be replaced by a more efficient
keyboard layout (David 1985).2
For the concept of path dependence to be meaningful, it cannot only signify
that ‘history matters’ (Beyer 2005; Werle 2007). The notion of positive feed-
back is the mechanism that stabilizes a path. Three sources of such positive
feedback can be distinguished and illustrated by the QWERTY example:
positive economies of scale due to sinking costs of production; technical
complementarity between hard- and software, referring in this example to
the keyboards and the secretaries trained to type on them; and, thirdly, the
non-reversibility of investment (David 1985: 334–5). The emphasis on the
sequence of events and the contingent beginning of a path are less distinctive
than positive feedback, as they are general neo-institutionalist explanations.

2
The relevance of path dependence was rejected early on by Liebowitz and Margolis (1990),
who argued that QWERTY would have been abandoned had there been efficiency gains. They
argue that rational economic actors would then have invested in new typing machines and the
necessary retraining of secretaries.
 OUP CORRECTED PROOF – FINAL, 6/12/2017, SPi

Case-Law Development 61
Increasingly, path dependence is used to analyse jurisprudence, focusing
mainly on the effect of precedent on subsequent judgments (Hartmann 2016;
Fon et al. 2005; Hathaway 2003). Jurisprudence has, as Shapiro and Stone
Sweet argue, an inherent tendency to produce path dependence: ‘Legal insti-
tutions are path dependent to the extent that how litigation and judicial rule-
making proceeds, in any given area of the law at any given point in time, is
fundamentally conditioned by how earlier legal disputes in that area of the law
have been sequenced and resolved’ (Shapiro and Stone Sweet 2002: 113). Beck
(2012: 239) notes that the ECJ makes clear reference to precedent, quoting its
own case law more frequently than it refers to the wording of rules or to
purposive arguments. Reasoning by analogy and precedent legitimizes the
Court; the teleological approach orients its rulings.

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

For the rule of law, the unity of the legal order is an important juridical goal
in order to ensure that law develops without contradictions. This supports
precedent and path dependence. In the judicial hierarchy of the EU, the ECJ
depends on the cooperating member-state courts in order to be effective.
However, lower courts will only be able to apply European law independently
if the case law is consistent. This is also one incentive for the US Supreme
Court to honour stare decisis (Bartels 2009: 475). It has to be noted that the cases
that reach the ECJ or other constitutional or supreme courts are those where
precedent does not give a clear indication of the outcome. Otherwise the cases
would not need to be handled at this level. Moreover, courts, in particular
courts of last instance, are not bound by precedent at all times (Gerhardt
2005). They also have the responsibility to arrive at ‘just’ decisions.
While precedent and path dependence orient litigants and lower courts, the
ECJ itself is free to enact changes in its development of case law, possibly
resulting in a critical juncture from which a new path can develop (Capoccia
and Kelemen 2007: 367). For such a break in the path to occur, it is likely that
particular incentives or mounting opposition from relevant actors will be of
significance. In contrast to historical critical junctures, which rarely occur,
occasions for new judicial interpretations abound.3 To what extent are there
incentives for legal innovation in the court system to bring about such a
critical juncture and start a new path? Occasional landmark judgments initiate
new interpretations and show that judges are free to bring in new consider-
ations. Successful new interpretations in the legal literature, in judgments, or
in the opinions of the general advocates that initiate a new development in
case law are disproportionately cited afterwards. This can be an internal legal

3
Note that Capoccia and Kelemen (2007: 367) argue that even the rulings that establish direct
effect and supremacy do not have the status of critical juncture, as they could equally have come
at another time. I disagree. The consequences of these two rulings are enormous and still little
understood. It therefore seems more useful to argue that critical junctures for case law are more
easily brought about—the question is whether positive feedback is provided.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

62 The ECJ and the Policy Process

incentive to deviate from precedent, alongside external pressures that push
towards a critical juncture. If there is subsequent positive feedback, a new path
can be solidified. For case law, path deviations or reversals appear to be less of
a problem than they are for other path-dependent processes.
In Chapter 2 we saw, when legal activism and the difficulty of court curbing
was discussed, that member states cannot directly steer the course of juris-
prudence. They can only hope to influence it through handing their opinions
and observations to the Court. But from within the legal profession, path
deviations can be fostered. The development of EU law is the best example as
the doctrines of the direct effect and supremacy of European law were
purposefully promoted and established by Commission officials of the legal
service, professors, and judges (Vauchez 2008; Alter 2009).

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

As we will see later, in terms of the freedom of goods, the Dassonville Case 8/74
(11 July 1974) and Cassis de Dijon Case 120/78 (20 February 1979) cases
established a new path in the late 1970s. Instead of being interpreted narrowly
as a non-discrimination rule, freedom of goods was interpreted broadly as a
prohibition of restrictions. An orientation towards precedent does not prevent
about-turns. ‘It is difficult to defend the assertion that precedents impose binding
rules on judges, compelling them to decide individual cases in certain ways’
(McCown 2003: 979). Subsequently, processes of positive feedback have trans-
ferred this broad interpretation to the other freedoms, as we will see. Private
litigants give crucial positive feedback. A broad interpretation of the basic free-
doms gives them freedoms of choice that they have not enjoyed before, if they can
construct a cross-border characteristic of their economic activity (Oliver and
Enchelmaier 2007), which is the prerequisite for EU law to apply—once they
have access to courts, which can prove to be a serious constraint, of course. It is
important to note, however, that restrictions on legal standing hardly lessen the
impact of my argument. As complicated as it may be for the individual litigant to
draw on possible EU Treaty rights, once a claim can be established, the resulting
ruling of the Court has a treaty-like status and is applicable across the Union. As
we will see in the discussion of case law for the four freedoms, constraints on the
legal standing of individual litigants have not hindered the necessary case load for
far-reaching case-law development. In this process, the interpretation of the
freedom of goods serves as a ‘focal point’, in Schelling’s definition (1960: 57).
The more private actors assert their interests through European law, the
more general knowledge about these European rights is strengthened. Success
and failure in court cases give positive or negative signals to actors in a similar
situation, thereby reinforcing either litigation or non-litigation (Fon et al.
2005: 44, 47, 51). Litigants orient their legal argumentation towards precedent
because presenting ‘arguments rooted in the Court’s past interpretations is
extremely effective’ (McCown 2003: 981).
Yet, litigants do not provide positive feedback by themselves. They find a
support structure in the cooperation between lower courts and the ECJ, in the
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Case-Law Development 63
community of EU lawyers that push for the broader impact of EU law, and in
the Commission, which frequently instigates new case law with infringement
proceedings. Often, complaints by private actors who choose not to litigate
themselves stand behind initiatives of the Commission. But litigation and the
actions of the Commission could not be so successful were there not a
community of EU lawyers who often favour the broadening of EU law.
Moreover, private litigants would have to rely on the Commission exclusively
should domestic courts be less willing to hand preliminary references to the
ECJ or to apply EU law directly.
The interest of private actors in a more favourable legal order, the interest
of European institutions in improved implementation and legitimation of
European rights, as well as the orientation of jurisprudence towards cases of

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

precedent and internal coherence, consequently, all contribute to the positive
feedback received by a parallel interpretation of the basic freedoms. Referring
back to the QWERTY example, one could draw a parallel between the positive
returns to scale and the positive feedback on successful litigation. If private
actors can enhance their position through litigation, this sends a positive signal
to other litigants. The path of case law and the interest of supranational actors in
strengthening their own legitimacy might be compared with the hardware and
the litigation strategies of private actors with the complementary software.
Finally, the relevance of precedent parallels the non-reversibility of technical
investments. The more established a line of legal reasoning, the less likely it is to
be altered. However, the precise scope of an interpretative path remains uncer-
tain. Given the differences between the fundamental freedoms, there are ample
possibilities to find a rationale for a differentiated treatment. Therefore, there
are not only good reasons for a transfer of the argumentation for the freedom of
goods to the other freedoms but also against it.
Furthermore, central to the concept of path dependence is the notion that
once a path has been ‘locked in’, it cannot be changed despite existing
dysfunctionality. If the freedoms can be interpreted broadly as prohibiting
restriction, economic actors can pick the member state with the lowest
regulatory requirements for the freedoms’ establishment, raising fears of a
regulatory race to the bottom (Sun and Pelkmans 1995). These distributive
consequences of case law can be taken as a dysfunctional lock-in, as those
actors that are negatively affected by case law cannot alter the path’s course.

PATH-DEPENDENT JURISPRUDENCE ON
THE F UNDAMENTAL FREED OMS

As I have shown, the concept of path dependence suits the analysis of

jurisprudence. It is particularly meaningful within the context of EU law, for
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

64 The ECJ and the Policy Process

which we wish to explain the significant expansion of case-law development
and its impact on policies. For dynamic case-law development to unfold, it was
important that the ECJ initiated a new path for the freedom of goods, as I will
now discuss. Originally, the fundamental freedoms were interpreted as form-
ing a prohibition against discrimination. Under this non-discrimination rule,
goods had to comply with the rules of the importing (or host) state, which is
also called national treatment. Non-discrimination puts EU nationals on an
equal footing with nationals and protects the regulatory capacity of the state
where goods are traded. Nicolaïdis and Shaffer have referred to the saying
‘When in Rome do as the Romans do’ as characterizing this approach
(Nicolaïdis and Shaffer 2005: 268).
By interpreting the fundamental freedoms as prohibiting discrimination,

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

one defers to the autonomy of the member states. Each member state
remained relatively free to determine the quality of goods and services traded
in its territory. If trade was not to be hampered by these different national
requirements, it was necessary to harmonize regulations at the European level.
Private actors had few incentives to turn to EU law if they did not have to
counter discrimination.
This changed radically in the 1970s when the ECJ broadened its interpre-
tation of the freedom of goods first in its Dassonville ruling and later in the
Cassis judgment. The issue was Article 28 of the Treaty Establishing the
European Community (TEC), which reads: ‘Quantitative restrictions on
imports and all measures having equivalent effect shall be prohibited between
Member States.’ In Dassonville (No. 5), the Court ruled that these equivalent
effects were characterized as follows: ‘All trading rules enacted by Member
States which are capable of hindering, directly or indirectly, actually or
potentially, intra-Community trade are to be considered as measures having
an effect equivalent to quantitative restrictions.’
The emphasis for the freedom of goods changed from the prohibition of
discrimination to the prohibition of restrictions to trade more generally. By
including indirect and potential effects, the new scope of the freedom of goods
was enormous, ‘maximizing the right for individuals to participate on the
market on whatever terms they choose, and providing them with a vehicle to
challenge any national rule which—even potentially and indirectly—stands in
their way’ (Barnard 2013: 74–5).
Cassis de Dijon dealt with German import restrictions on French liqueur,
given that these imports did not meet German regulations on the alcohol
content of liqueur. The case illustrates differences in interpretation. Under the
former interpretation of narrow non-discrimination, Germany could bar
French liqueur for violating German rules. But in Cassis, the Court ruled
that member states had to recognize products ‘lawfully produced and mar-
keted in one of the Member States’ (No. 14). This implied that member states
were obliged to take into account the regulatory requirements of products in
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Case-Law Development 65
their origin country. Germany had to recognize the French regulation of
liqueur as being equivalent. Additional regulatory burdens may only be
imposed when they are proportionate and justified by overriding public
interests, as they pose ‘a double burden’ for market participants who are
already regulated in their home country. Consequently, the member state is
no longer able to regulate EU foreigners according to its own discretion but
has to recognize mutually the functionally equivalent regulations of other
member states. In moving the interpretation of the law towards a prohibition
of restrictions, a general assumption of such equivalence (the home-country
rule) was introduced. This loosens the link between territory and legal order
(Schmidt 2007), and private actors need no longer be tied to their national
regulatory framework. European law allows the extraterritorial validity of the

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

most preferential national legal order (Steindorff 1999).
It is useful at this point to discuss another concept that will be relevant
throughout the book: the potential for reverse discrimination against nation-
als. This happens if nationals are regulated more stringently than economic
actors from other member states who are active in the national market based
on their equivalent home-country regulation. The four freedoms aim to
facilitate trade; it is not possible for nationals to claim a violation of an internal
market freedom with respect to their own government if there is no transbor-
der aspect. In terms of its own nationals, the government is free to pursue
stricter rules; however, it may be that there are national constitutional prin-
ciples that prohibit such discrimination. As we will discuss at the end of this
chapter, as long as reverse discrimination against nationals remains possible,
the reach of the four freedoms does not include purely national markets.
Should reverse discrimination against nationals be prohibited, as has some-
times been discussed in the legal profession, the four freedoms impose a
barrier on all market regulations, and also those that are purely domestic.
Along with the expanded interpretation of the non-restrictions approach,
member states could seek broader exceptions to the reach of the fundamental
freedoms. Next to the exemption in Article 30 of the TEC, which refers to
public security, Cassis introduced an open list of exemptions in the form
of ‘mandatory requirements’ (summary 2) for the general public interest,
allowing member states to impose their host-country regulatory requirements.
However, the ECJ interprets this possibility very narrowly (Oliver and
Enchelmaier 2007), and the measures need to be proportionate. The inter-
pretation of the freedoms as a prohibition of restrictions makes the regulations
of host states subject to external scrutiny as to their proportionality (Schneider
1996: 515). Democratic rules are checked for their market consequences. The
differences between the non-discrimination and the non-restriction approach
are depicted in Figure 3.2.
The Court’s balancing approach of proportionality can be neatly summarized
through the example of Case C-55/94 Gebhard. National regulatory measures
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

66 The ECJ and the Policy Process

non-discrimination non-restriction =
= host-country rule home-country rule

Products legally
marketed in any MS
All products marketed
A–Z may be marketed in
in MS A have to adhere
MS A (Exception:
to its rules
‘mandatory
requirements’)

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Within a territory,
Within a territory, all products regulated
products are regulated according to standards
in the same way: A–Z:
AAAAAAAAAAA ABCDEFGHIJKLM
N O P Q R ST UVW XY Z

= mutual recognition
= regard for autonomy undermines the
of member states regulatory autonomy of
member states

Figure 3.2. The non-discrimination and non-restriction approaches

must meet ‘four conditions: they must be applied in a non-discriminatory

manner; they must be justified by imperative requirements in the general
interest; they must be suitable for securing the attainment of the objective
which they pursue; and they must not go beyond what is necessary in order to
attain it’ (No. 37 of C-55/94). The introduction of proportionality, Stone Sweet
and Mathews (2008) emphasize, broadens the scope for judicial review, limiting
parliamentary sovereignty in general. Since the autonomy member states still
have to regulate is checked by the Court’s proportionality test, legal uncertainty
becomes an inherent part of this extensive interpretation (Randelzhofer and
Forsthoff 2001: No. 54; Hatzopoulos 2000: 43–4).
Why is the non-restriction interpretation an important new path? In
principle, this interpretation subjects all national regulatory measures to
scrutiny that will possibly hinder market freedoms. At the same time, Cassis
accorded a minimum of regulatory autonomy to member states by introduc-
ing mandatory requirements as a rule of reason. Implicitly, member states
traded enhanced market access for a loss of their regulatory autonomy
(McCall Smith 2000). This began, however, in a Community of nine member
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Case-Law Development 67
states with similar levels of economic development, where member states
could assume an equivalent level of market regulation. Although ongoing
rounds of enlargement have eroded this precondition, the judicial path of
interpreting the fundamental freedoms was subsequently broadened to include
all freedoms and deepened to allow fewer exceptions to the national regulatory
prerogative. It is only most recently, in light of the EU’s crisis, that the Court
has again accorded more leeway to member states, most noticeably in the
greater amount of freedom they have been given to protect their welfare
systems, as I will discuss later.
Alter and Meunier demonstrated early on that the Commission took up the
Cassis judgment to provide itself with legal support on which it could build its
single-market initiative (Alter and Meunier-Aitsahalia 1994). Though the

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Court had not spoken of ‘mutual recognition’, the Commission established
this term. Building the single market would be much easier if deadlock-prone
harmonization could be restricted to those areas where there were narrow
exceptions to the freedom. Under the broad interpretation of non-restriction,
mutual recognition would be the rule.
In the following section, I argue that the interpretation of the freedom of
goods served, subsequently, as a ‘focal point’. Different actors pushed for
similar leeway for the other freedoms (cf. Garrett and Weingast 1993). In
theoretical terms, the initial interpretation of the freedom of goods represents
a critical juncture. Interpreting the fundamental freedoms as prohibiting
restriction is a very different path to the obligation of non-discrimination.
With the much greater leeway to pursue economic activities that this inter-
pretation offers, there are significant incentives for litigants to argue analo-
gously for the other freedoms. Such positive feedback is the core of path
dependence. We can also consider it to be a contingent event that goods
markets internationalized first. As we will see later, the transfer of the non-
restriction interpretation to the other freedoms has significant redistributive
repercussions. The political costs of redistribution render the lock-in of non-
restrictions dysfunctional. It is, therefore, highly unlikely that this interpretative
path would have been chosen had the process begun with the interpretation of
another freedom.
Before we analyse how the reasoning for a prohibition of restrictions
transferred to the other freedoms, it is important to note that the freedom of
goods also offers an example of a critical juncture that did not result in a path
that structured the interpretation of the other freedoms.
The Keck ruling (Cases C-267 and C-268/91, 24 November 1993) broke
with precedent in the interpretation of the freedom of goods (Reich 1994). It is
one of the rare examples where the Court explicitly decided for a juncture in
its case law (Beck 2012: 252). Previously, litigants in the UK had made ever
further inroads into national regulatory autonomy in relation to shop opening
hours (Rawlings 1993). The broad interpretation of the freedom of goods as a
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

68 The ECJ and the Policy Process

prohibition of restrictions had confronted the ECJ with all sorts of questions.
By establishing in Keck that selling arrangements do not fall under the
freedom of goods, the ECJ limited the reach of the freedom of goods as a
prohibition of restrictions.

Keck can only be understood against the background of the considerable confu-
sion which had been created over the previous few years as a result of the Court
treating all types of import restriction as falling under Article 28, whether or not
they discriminated in any way against imports. The risk of abuse became increas-
ingly obvious and the Court was driven to ever more far-fetched solutions. By its
ruling in Keck, which partially reversed the pre-existing case law, the Court
sought to set proper bounds to the scope of Article 28. (Oliver 1999: 793)

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

The Court drew a distinction between rules that relate to the production and
distribution of goods and those that relate to certain selling arrangements (like
shop opening hours, pricing, or ruinous competition). As long as the latter
concern all economic actors in a state, and both national and EU foreign
goods, they are not covered by the freedom of goods and member states retain
the autonomy to regulate the organization of their markets (Troberg 1997a:
1471 No. 33). How does Keck relate to the idea of fine-tuning, which was
introduced in the last chapter? Keck restricted the reach of the freedom. In
doing so, it fine-tuned the application of the non-restrictions approach, as its
expansion into selling arrangements was halted. It was a partial reversion, as
few earlier steps in this direction had been taken. Potentially, Keck could have
served as a critical juncture had this argument about the bounds of non-
restriction been applied to the other freedoms. But this path was not estab-
lished. There was no positive feedback on the Keck ruling. As was argued
earlier, the Court and the legal profession are well placed to provide new legal
interpretations that can establish new paths, but private actors and the Com-
mission will only provide positive feedback on those interpretations that are in
their interest.
How could one refute the presence of a path-dependent process? Case law is
often ambiguous, as this book argues. As such, different scholars classify cases
in different ways—where one scholar sees a non-discrimination approach
(Davies 2003), another sees a prohibition of restriction (Barnard 2010),
while yet another argues for an evolving market-access test (Tryfonidou
2010). Yet, beyond such differences, there is widespread agreement among
legal scholars on a convergent interpretation of the fundamental freedoms
through the ECJ (Davies 2003; Tryfonidou 2010; Oliver and Roth 2004;
Barnard 2013). A path-dependent argument could be refuted if one could
prove that there are distinct elaborations of reasoning on the different funda-
mental freedoms rather than identical types of arguments in different areas.
Figure 3.3 summarizes how legal uncertainty and path dependence relate to
each other.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Case-Law Development 69

• Balancing between EU goals and legitimate MS interests

• Transfer of legal reasoning from one freedom to another and
legal of one MS-specific ruling to another MS creates legal
uncertainty uncertainty

• Precedent orients courts and litigants

positive • Litigants, lawyers, and courts transfer legal arguments
feedback

• Path dependent, incremental case-law development

• Context of the ‘ever closer union’
path
• Transfer to other freedoms, reasoning by analogy

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

dependence

Figure 3.3. Legal uncertainty and path dependence

T H E T R A N S F E R OF L O G I C T O O TH E R F RE E D O M S

In the following section, I will show how the ‘logic of goods’ was transferred to
the other fundamental freedoms, building on the litigation strategies of private
actors. Initially, the ECJ ruled cautiously outside of the area of freedom of
goods and treated the different freedoms distinctly.
In the 1970s and 1980s the prevailing view was that the scope of the four
freedoms should be drawn along different lines: Article 28 EC on the free
movement of goods was to be considered as a far-reaching prohibition of any
measures potentially hindering or restricting the import of goods, whether
indistinctly applicable or not. . . . In contrast, the provisions on the free movement
of persons—workers, establishment and services—were generally regarded as an
expression of the general principle of non-discrimination on grounds of nation-
ality, as set forth in Article 12 EC. (Oliver and Roth 2004: 411)
An analysis of the types of cases that come before the ECJ over time, presented
in Figure 3.4,4 shows which freedom was interpreted at which point in time.
Initially, the freedom of goods and the free movement of workers were respon-
sible for a high number of cases. Because of the need to coordinate social-
security systems, secondary law structured the free movement of workers from
the very beginning so that it was less subject to case-law development, as based
on the Treaty, than the freedom of goods. Then, towards the end of the 1990s,

4
This is the result of a EUR-Lex search of preliminary and infringement procedures for the
relevant freedom in the case summary.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

70 The ECJ and the Policy Process

50

45

40

35

30

25

20

15

10

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

5

0
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
Free movement of services Free movement of goods
Free movement of capital Freedom of establishment
Free movement of workers Free movement of citizens

Figure 3.4. Completed legal cases according to freedom, 1958–2012

Source: Adapted from Susanne K. Schmidt, Gefangen im ‘lock in’? Zur Pfadabhängigkeit der Rechtsprechung
des Europäischen Gerichtshofs. Der moderne Staat: Zeitschrift für Public Policy, Recht und Management, 3
(2) (2010), 477.

the freedom of services led to more cases, with the freedom of establishment
following a few years later. The free movement of capital came last, as this was
only declared to be directly effective after the Maastricht Treaty.

Services

For a long time, the ECJ interpreted the freedom of services restrictively
(Hatzopoulos 2000: 63–4; Roth 2002: 20; Davies 2007b: 14). Article 57 of the
TFEU notes that ‘the person providing a service may, in order to do so,
temporarily pursue his activity in the Member State where the service is
provided, under the same conditions as are imposed by that State on its own
nationals.’ By limiting the freedom of services to temporary activities, with
explicit reference to the ‘same conditions’ of its ‘own nationals’, the freedom
did not seem to have much scope, so freedom of establishment appeared
necessary. Although the ECJ had already argued in Case 33/74 van Binsbergen
for a limited mutual recognition of rules, so that the requirements for tem-
porary activities were not prohibitive, this restrictive line on services was still
dominant. One example is the well-known Case C-113/89 Rush Portuguesa,
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Case-Law Development 71
which allowed France to apply its minimum wage to workers posted from
Portugal. Instead of the home-country rule and Portuguese wage levels, the
host-country principle was deemed relevant, and trade in services could be
restricted in favour of domestic regulatory interests.
It is only since the 1990s and, particularly, the 2000s, that the Court has
emphasized the duty of member states to abolish hindrances to trade in
services more forcefully. Accompanying the more liberal approach, interest
in litigation rose. The first ruling to apply the argumentation in the Cassis
case to services was that arising from Case C-76/90 Säger in 1991. It is a
typical example of a private actor seeking to overcome the restraints
of national law. A British company was offering patent renewal services in
Germany, where only registered patent attorneys were allowed to provide this

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

service. An incumbent competitor sought legal assistance, and the German court
addressed the ECJ (Barnard 2013: 386–7). The ECJ argued in parallel to the
freedom of goods:

Article 59 of the Treaty requires not only the elimination of all discrimination
against a person providing services on the grounds of his nationality, but also the
abolition of any restriction, even if it applies without distinction to national
providers of services and to those of other Member States, when it is liable to
prohibit or otherwise impede the activities of a provider of services established in
another Member State where he lawfully provides similar services. In particular,
a Member State may not make the provision of services in its territory subject
to compliance with all the conditions required for establishment and thereby
deprive of all practical effectiveness the provisions of the Treaty whose object is,
precisely, to guarantee the freedom to provide services. (Summary 1)

Dynamic case-law development shows the increasing motivation of private

actors to seek the assistance of courts in proceedings against national regula-
tory restrictions. By shifting the interpretation from a non-discrimination to a
non-restrictions approach, existing regulations of other member states have to
be mutually recognized. As private actors found an alternative legal order, the
number of cases rose. Hatzopoulos (2012: 179) tracks the growth in case law
based on the freedom to provide services (Article 56 of the TFEU): while the
ECJ decided on only forty cases between 1995 and 1999 (an average of eight
cases per year), there were 140 cases between 2000 and 2005 (an average of
twenty-three cases per year), and between 2006 and 2010 (with the accession
of many more member states) there were 137 cases (an average of twenty-
seven cases per year). A high number of cases enables courts to develop case
law incrementally. ‘Only disputes that are actually litigated are capable of
generating legal precedent’ (Fon et al. 2005: 45).
The scope that the freedom of services offers is also due to its application to
service recipients alongside service providers (Menéndez 2009: 8). As early as
1975, the Commission argued that German delegates of Amnesty International
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

72 The ECJ and the Policy Process

who sought to have lunch in France could not be denied entrance to the country
as they were recipients of a service (Hofmann 2013: 214–15). Recently, this same
argument was taken up to extend the rights of third-country nationals. One
preliminary reference concerned a Turkish citizen, who, after having been denied
a visa, argued that she had intended to receive services (a haircut) in Berlin, and
therefore should not need a visa under the Turkish Association Agreement. For
service providers (a lorry driver), this argument had been successful (Case
C-228/06 Soysal) (Wiesbrock 2013a), but the Court denied its relevance for the
passive freedom of services (Case C-221/11 Demirkan).
With view to the Services Directive (2006/123/EC), which will be analysed
in Chapter 4, Hatzopoulos (2012: p. x) holds that the case law on the freedom
of services is unique as ‘the Court has set the principles . . . the legislature

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

has intervened ex post in order to consolidate, codify, rationalize, and,
often, restrict the scope of principles and rules introduced by the Court’
(Hatzopoulos 2012: p. xi).

Establishment

The development of case law for the freedom of establishment is comparable

to that for the freedom of services. Originally, it appeared that member states
were able to maintain their national company-law restrictions, following
the ruling in Case 81/87 Daily Mail in 1988 (Freitag 1999: 269). In 1999,
a fundamental change to the case law came about with the ruling in Case
C-212/97 Centros. A Danish couple took up the non-restrictions reasoning on
the fundamental freedoms and established a limited company in the UK, in
order to exploit the much lower British capital requirements, in comparison to
Denmark (only approximately DKR 1,000 instead of DKR 200,000), to estab-
lish a company. Again, this shows how litigants provide positive feedback,
transferring legal arguments from one fundamental freedom to another. The
couple had no intention of becoming economically active in the UK. They
only wanted to circumvent the stricter Danish rules. The Danish authorities
declined to register the Danish branch office and a preliminary proceeding
ensued. The ECJ diverted from its earlier jurisprudence in the case of the Daily
Mail (Hoor 1999; Freitag 1999: 270) and extended the non-restrictions
approach to the freedom of establishment (Article 49 and Article 54 of the
TFEU). Companies, thus, had a ‘right to have the most preferable legal order’
(Steindorff 1999).
Not surprisingly, companies took up this new regulatory flexibility and
several rulings followed, allowing the ECJ to concretize its new case law. The
cases of C-208/00 Überseering and C-167/01 Inspire Art again extended the
freedom of establishment. Parallel to the freedom of goods, the ECJ detailed
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Case-Law Development 73
member states’ obligations to mutually recognize the regulations of other
member states.
Interestingly, the important case of Überseering is an example of how case
law can come about quite accidentally. Überseering basically abolished the
German seat theory in company law, dealing a serious blow to German
corporate governance. In this case, a Dutch company was active in Germany
but had only transferred its administrative centre and not its legal seat to
Germany. It had been denied legal standing in a court case. According to
German seat theory, which governs company law, the company would have
needed to be re-established under German law once it had lost its Dutch links
in order to have its legal seat in Germany. The German court asked whether
this would not effectively prohibit the freedom of establishment, a view which

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

the ECJ shared. Interestingly, this is not a case in which active domestic courts
aimed to instrumentalize European law. The chamber of the supreme court
(BGH) that put forward the reference was responsible for buildings and not
for company law. The company law chamber had criticized the posing of the
preliminary reference as it deemed the legal questions to have been settled
through its own case law.5 Therefore, the case came about coincidentally. After
Überseering, this chamber also had to change its line of interpretation.
The ECJ took its turn in its ruling on Case C-210/06 Cartesio in late 2008, in
which it employed caution by not extending its liberal jurisprudence on
company mobility further. After a string of cases that expanded the case law,
the ECJ reverted to fine-tuning. In this case, a Hungarian company wanted to
move its administration to Italy while keeping its seat of incorporation in
Hungary. Surprisingly to most observers, the ECJ ruled that the freedom of
establishment does not require member states to allow companies to change
their administrative seat while maintaining, at the same time, their status
according to the law of the member state of incorporation. The very liberal
jurisprudence was applied only to companies moving into member states and
not to cases of companies moving out of a member state (Barnard 2013: 329).
Interestingly, the English, German, and French translation of the Hungarian
reference to the ECJ included contradictions about whether the transfer of the
registered office or the ‘real seat’ was the issue. This led the Irish government
to demand a reopening of the oral procedure (Gerner-Beuerle and Schilling
2010: 307–8), which the Court declined (C-210/06, No. 41–53). Thus, the case
exemplifies the legal uncertainty that flows from the multilingual nature of
the legal order.
A broad interpretation of the freedom allows actors to pick the most efficient
regime for establishing companies. Often, the freedom of establishment is used

5
Thank you to Rike Krämer for pointing this out to me. Ulrich Forsthoff, ‘Mehr Wahlmög-
lichkeiten für Unternehmen, Europarichter: Deutschland muß ausländische Rechtsformen
anerkennen’, FAZ 13 Nov. 2002, p. 18.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

74 The ECJ and the Policy Process

in tandem with the freedom of services. The Services Directive reflects this in its
application to both freedoms. In addition to allowing lenient establishment
rules for companies, the freedom of establishment, together with the free
movement of capital, has become relevant for tax competition, which will be
discussed in Chapter 5. Concerning the codification of the case law on estab-
lishment at the EU level, I only discuss the Services Directive in Chapter 4.
There are also several company directives that have been agreed in the EU since
the 1960s.6 Explicitly in relation to the Cartesio judgment, the European
Parliament has called on the Commission to pursue its plans for a fourteenth
Company Law Directive on the cross-border transfer of company seats in 2012
in order to give ‘the necessary clarification’, emphasizing that ‘it is for the
legislators and not for the Court of Justice to establish on the basis of the Treaty

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

the relevant measures to accomplish the freedom of a company to transfer its
seat’ (2011/2046 (INI)). However, the institutional diversity of member states’
regimes has not allowed progress in this area.

The Free Movement of Workers

For several reasons the case law on the free movement of workers is much
more complicated than that for the other freedoms. I will present this in
greater detail in Chapter 5, in addition to a historical account of European
citizenship and its subsequent codification in the Citizenship Directive, but
here I only focus on the switch to a non-restrictions approach. In contrast to
the other freedoms, this freedom was shaped by secondary law from the very
beginning. In order to ease the free movement of workers, the participation of
workers and their families in social assistance and insurance schemes needed
to be regulated. Case law, therefore, often deals with secondary law alongside
interpretation of the Treaty.
The freedom of movement of persons is special because it is explicitly linked
to non-discrimination (Article 45 II of the TFEU): ‘Such freedom of move-
ment shall entail the abolition of any discrimination based on nationality
between workers of the Member States as regards employment, remuneration
and other conditions of work and employment.’ An approach of non-
discrimination is linked to the host state treating nationals and EU nationals
alike. There is no equivalent in this article to the prohibition of quantitative
restrictions, which allowed the switch to the non-restrictions approach for the
freedom of goods (Barnard 2001: 38). A different development of case law can
also be expected here since the argument concerning a ‘double burden’ of
regulatory requirements, subject to the regulations of the home and host state,

6
<http://ec.europa.eu/internal_market/company/official/index_en.htm> [accessed 10 May 2017].
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Case-Law Development 75
is only significant for goods and services but not for natural or legal persons,
who are only integrated into the state of economic activity and so its rules
apply (Barnard 2013: 244).
Building on regulations 3 and 4 of 1958, the freedom of movement for
workers and their families was further spelled out in legislation in 1968 (EEC
No. 1612/68, now 492/2011; 68/360/EEC, now 2004/38/EC) by granting rights
to non-discrimination and national treatment. The right of residence in a
member state after the termination of employment followed in 1970. In 1990,
three directives further regulated the residence of workers, students, the
retired, and self-employed workers (90/364/EEC; 90/365/EEC; 90/366/EEC).
The affiliation to different national social insurance schemes was coordinated
in regulation 1408/71/EEC (now 883/2004).

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Originally, rights of residence and movement in the EU were strictly tied to the
status of a worker. In the 1980s, a series of cases (C-53/81; C-139/85; C-196/87)
broadened the application of these rights, making it sufficient to be economically
active to any extent, even if one was not necessarily earning enough to make a
living (Davies 2003: 195). The Court tested for indirect and covert discriminatory
measures (Oliver and Roth 2004: 416), including the presence of a mere potential
effect following the Dassonville formula (Barnard 2001: 38–9).
The shift to the non-restriction path took place with the ruling in Case
C-415/93 Bosman, which involved a football player. The transfer fees required
when players changed clubs was held to be ‘an obstacle to the freedom of
movement’ (No. 103), though the system did not discriminate along national
lines. Thus, the Court established an interpretation of the prohibition of
restrictions that also applied to the free movement of workers. The restrictions
path has by now become the dominant interpretation for the movement of
workers and the other freedoms.7
Whatever nomenclature used in that jurisprudence, be it manifested in impeding
access/hindrance to trade or the terminology of ‘liable to hamper or to render less
attractive’, it is crucial to appreciate that the common objective in all such
jurisprudence is the removal of the restriction to exercise the free movement
right. . . . The removal at the national level of such restrictions is clearly at the
focus of recent free movement jurisprudence. (Connor 2010: 177)
As was the case for the other freedoms, the change to the non-restriction path
for free movement led to numerous cases where actors sought to improve their
situation, sometimes including rather far-fetched arguments. Case C-190/98
Graf concerned a German who had resigned from his employment contract

7
Barnard (2009: 592) mentions the following in relation to workers in Case C-464/02, no. 45:
‘It is settled case-law that Article 39 EC prohibits not only all discrimination, direct or indirect,
based on nationality, but also national rules which are applicable irrespective of the nationality of
the workers concerned but impede their freedom of movement (Graf, No. 18, and Weigel, No. 50
and 51).’
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

76 The ECJ and the Policy Process

after four years in Austria and litigated for compensation, due to the fact that
Austrian workers receive this when they are laid off after more than three
years. The ECJ did not follow this logic and argued that the possibility of
hindrance was ‘too uncertain and indirect’ (No. 24). At the time, it was hoped
that this could undermine the positive feedback of further litigation and help
to ‘stem the flow of cases that seek to challenge key pillars of national welfare
and social provision internal to the Member States which were never intended
to interfere with free movement’ (Barnard 2001: 50). Another example is the
Weigel case in which a German couple, after having moved to Austria, filed a
lawsuit because the Austrian annual tax on automobiles was higher than it was
in Germany, arguing that this was a restriction of their freedom of movement.
So too here, the Court rejected the claims, thereby fine-tuning what it under-

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

stands to be non-restriction:
However, the Treaty offers no guarantee to a worker that transferring his
activities to a Member State other than the one in which he previously resided
will be neutral as regards taxation. Given the disparities in the legislation of the
Member States in this area, such a transfer may be to the worker’s advantage in
terms of indirect taxation or not, according to circumstance. (C-387/01, No. 55)

These instances of fine-tuning can be regarded as points when negative

feedback is given to litigants, as the Court offers no incentive for further
litigation. It leaves one wondering what motivated litigants and lower courts
to pursue these cases before the ECJ. But within such extreme cases, the Court
takes a broad approach to restrictions, which does not require the introduction
of hindrance, making it somewhat likely that national rules will appear to
restrict a certain freedom (Barnard 2010: 258).
Beginning with Case C-224/98 d’Hoop, the Court has argued that any measure
that makes the take-up of rights to free movement less attractive is a restriction.
D’Hoop involved a Belgian who had obtained secondary-school education in
another member state and, therefore, did not qualify for the Belgian ‘tide over’
allowance available to school-leavers. Case C-212/05 Hartmann is also part of
this series of cases. In this case, a family lost the German child-raising allowance
after their move to Austria due to the residence principle, though the father was a
German civil servant. Such financial losses clearly made it less attractive to take
up residence abroad. The Court recognized movement unconnected to employ-
ment, thereby broadening the right to free movement, which the AG had
dismissed (No. 88). At the same time, the Court agreed that member states can
demand that a ‘link’ to their territory must be proven where social benefits are
concerned. But residence was not the only possible way to prove this, as ‘a
substantial contribution to the national labour market also constituted a valid
factor of integration into the society of that Member State’ (No. 36).
A further step was taken with Case C-287/05 Hendrix (Verschueren 2012:
193). Mr Hendrix, a recipient of the Dutch incapacity benefit for young
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Case-Law Development 77
people, had moved to Belgium while continuing to work in the Netherlands.
The action of cutting his benefits was in line with the existing social-security
regulation (1408/71, now 883/2004), establishing the territoriality principle for
this kind of social benefit. But the Court challenged this hindrance to free
movement, thereby overruling the legislative decision of the member states:
‘the development in Hendrix can be seen as a way of facilitating free move-
ment by encouraging the home state to bear some of the costs of the emigra-
tion of its citizens, and not just the host state’ (Barnard 2013: 292).
As is the case for previous paths of development, we can see the necessita-
tion of many far-reaching changes at the member-state level if disincentives to
moving abroad are counted as restrictions to the free movement of workers
and, additionally, if member states are even called upon to facilitate the

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

movement of their citizens. The case law on the free movement of workers
is particularly interesting, as this area was shaped by the secondary law of the
Council. This demonstrates the low levels of deference with which the Court,
at times, addresses the preferences of member states (Martinsen and Falkner
2011; Schmidt 2012).

Capital

Finally, the free movement of capital diverges to some extent from the other
freedoms, as it has been regarded ‘for many years the poor relation to the
other three’ (Barnard 2013: 579). The freedom, laid down in Article 67 of the
Treaty Establishing the European Economic Community (EEC Treaty/TEEC),
was shaped legislatively by two early directives in the 1960s. In an early
judgment (Casati, C-203/80), the ECJ established that this freedom did not
enjoy direct effect. In the context of the single market, the Council liberalized
capital movements from 1990 onwards with directive 88/361/EEC. Shortly
afterwards, the governments agreed in the Maastricht Treaty to prohibit all
restrictions on capital and payments in Article 73b(1) (now Article 63(1) of
the TFEU), rephrasing the freedom and opening it up for direct effect. The
Court affirmed the direct effect of the freedom in Sanz de Lera (joined cases
C-163/91, C-165/94, and C-250/94).
Following the Treaty change, since 1994, the number of cases reaching the
ECJ has risen significantly, enabling the ECJ to bring its interpretation into
line with the restrictions approach employed within the other freedoms
(Murphy 2012: 290). In the first case on golden shares in relation to Portugal,
it ruled—largely in parallel to the aforementioned quotation from the Säger
ruling—that non-discriminatory rules violated European law as they were
‘liable to impede’ the acquisition of shares and therefore likely ‘to dissuade
investors’ from putting money into other member states (C-367/98, No. 45)
(Barnard 2013: 592).
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

78 The ECJ and the Policy Process

Barnard (2013: 594–6) asserts that the case law of the Court deals with five
different kinds of national regulation: (1) property purchase and investment;
(2) cross-border transactions; (3) loans; (4) investments in companies and
shares; and (5) golden shares giving privileges to certain shareholders. By
largely adopting the language of ‘restriction’, the Court follows the approach
established in the Cassis and Dassonville jurisprudence. Member states have to
justify their regulations objectively, even if they apply without discrimination
‘either under the judicially developed public-interest requirements or, more
commonly, under the broad express derogations’ (Barnard 2013: 593).
Alongside the agreement on direct effect, the Maastricht Treaty introduced a
new exception that was to allow member states to discriminate for tax purposes
between place of residence and investment (Article 65(1)(a) of the TFEU),

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

which could be seen as less liberal than the earlier directive (Barnard 2013:
607–8).8 However, the Court also interpreted this exception narrowly: ‘Gone
was the cautious and uncharacteristically deferential Court which sought to
leave matters of capital liberalisation to the Member States in Council’
(Murphy 2012: 291). Thus, in 2009 the ECJ stated that member states may
not restrict tax deductions for charitable donations to domestic bodies, which
render donations to foreign charities less attractive (Persche, C-318/07). The
governments of Germany, France, the UK, Ireland, Spain, and Greece had
submitted opinions arguing that tax deductions needed a link to the pursuit of
public interest in the same territory. Article 65(1)(b) of the TFEU includes
another derogation for ‘all requisite measures to prevent infringements . . . in
the field of taxation and the prudential supervision’, which also referred to
public policy and public security (Barnard 2013: 610). While member states
could at times successfully justify regulations, I will discuss in Chapter 5 how
their tax policies have come under pressure from the expanded interpretation
of the free movement of capital combined with the freedom of establishment,
which has given private corporate actors many incentives to litigate against
domestic tax rules. As the Court interprets restrictions narrowly and does not
generally honour fiscal arguments, this is a particularly difficult development
for member states. In light of the complexity of the case law on taxes, which
concerns both freedoms, I will discuss it in detail in Chapter 5, where I will also
discuss the failure to codify this case law.

Citizenship

Citizenship rights were a last-minute addition to the Maastricht Treaty and

were added without significant political discussions (Aziz 2009: 291). EU

8
Barnard (2013: 608) argues that this article codifies the Court’s decisions in Case C-279/93
Schumacker and Case C-203/90 Bachmann.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Case-Law Development 79
citizenship has a mere supplementary status for those that possess it, for which
all citizens of all the member states automatically qualify. As an essential part
of this citizenship, all EU citizens may move and reside freely within the EU;
this right is explicitly subject to secondary law provisions. In terms of basic
political rights, EU citizens may stand and vote in elections for municipalities
and the European Parliament in all member states. In third countries, member
states provide consular protection for each other (Wind 2009: 254). These
rights, mentioned in Article 18 of the TEC inter alia, are just examples (Barnard
2010: 420). Article 12 of the TEC, which establishes the principle of non-
discrimination, is also relevant for citizenship rights. Given their partial
overlap, the free movement of workers is lex specialis, and Article 18 only
functions as a safety net. Based on this addition to citizenship, case law has

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

incrementally extended rights from workers to economically inactive citizens.
In due course, the language of restrictions was transferred to citizenship law.
Over the years, the interpretation of Article 18 has resulted in comprehensive
citizen rights, despite their intended supplementary nature. Gareth Davies goes
so far as to argue that citizenship rights have now subsumed the other freedoms
(see also Somek 2007: 794). This convergent development merits a discussion
of citizenship alongside the four freedoms.
[S]ince [Articles 18 and 12 of the TEC] are as broadly worded as the economic
rights, the economic Treaty articles do become entirely superfluous. One may rely
on them as a matter of lex specialis legal form, but the limits of their scope become
entirely unimportant, since where they stop citizenship will pick up.
(Davies 2003: 189)

The first case to show that citizenship rights can be decisive was Case C-85/
96 Martinez Sala. It established the principle of non-discrimination in the
granting of child-raising allowance to an economically inactive EU citizen
lawfully residing in Germany (Tryfonidou 2010: 39). In Case C-184/99 Grzelcyk,
decided upon in 2001, the Court established that the right of residence also
applied to economically inactive EU citizens—in this case a student—that
require assistance. This case was very influential in the negotiation of the
Citizenship Directive (see Chapter 5). If member states want to withdraw
the right of EU nationals to be treated in the same way, they have to show
evidence of an ‘unreasonable burden’ on their welfare state that results from it
(Tryfonidou 2010: 39). Hereby, the Court contradicted existing secondary law
and broke with previous case law (Case C-197/86 Brown), which had ruled out
benefits for EU students (Wind 2009: 258). In Case C-413/99 Baumbast, the
Court declared Article 18 I of the TEC to be directly applicable, thereby giving
EU citizens the right of residence and movement (Wind 2009: 256, 259).
Without going too much into the details of this case-law development,
suffice it to say that the addition of citizenship rights gave the Court, the
EU’s legal community, and civil-society actors a basis on which to question
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

80 The ECJ and the Policy Process

difficult European political compromises on how to differentiate between
economically and non-economically active EU citizens in the allocation of
social benefits. Case C-138/02 Collins is important here, as AG Colomer
tackled the need to define limits to the rights of EU citizenship for the first
time. According to AG Colomer, the requirement that member states treat all
EU citizens in the same way undermines the national orientation of welfare
systems, although the EU Treaty considers welfare to be chiefly the responsi-
bility of the member states:
In the present case, however, I take the view that a condition as to residence,
which is intended to ascertain the degree of connection with the State and
the links which the claimant has with the domestic employment market, may
be justified in order to avoid what has come to be known as ‘benefit tourism’,

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

where persons move from State to State with the purpose of taking advantage
of non-contributory benefits, and in order to prevent abuses. I do not believe
that that condition goes beyond what is necessary to attain the objective pursued
since it is applied after examination of claimants’ personal circumstances in
each case.
(No. 75)
The Court, however, was not ready to follow the recommendation of the AG
(see No. 63), emphasizing instead that the general principle of non-
discrimination (Articles 6 and 8 of the TEC) implied the right to access
‘benefits of financial nature’ (No. 50). In the case at hand, which concerned
access to jobseeker’s allowance for a newly established Irish American in the
UK, however, the Court perceived there to be a lack of ‘a reasonable period of
residence’. Thus, this serves as an example for the finding that new legal
principles (‘a reasonable period of residence’) are often introduced in a case,
where their actual relevance is denied, making it easier to establish precedent.
In another case from the time, AG Jacobs was the first to argue for a broader
non-restriction approach to citizenship:
But discrimination on grounds of nationality, whether direct or indirect, is not
necessary in order for Article 18 to apply. In particular, it is not necessary to
establish that, for example, a measure adversely affects nationals of other Member
States more than those of the Member State imposing the measure.
(Pusa (C-224/02, No. 18))
The Court had followed this line in Pusa, as well as in other cases. In Case
C-406/04 De Cuyper, the Court argued that it was a restriction and clearly fell
under Article 18 (No. 39), since a national incurred a disadvantage by simply
exercising their freedom to move. In this case, the residence clause for Belgian
unemployment benefits was considered to be justified by the public interest
(No. 48). The restriction served to assure the availability to the labour market.
The case is, thus, another example in which the broad scope of a freedom is
stated by the Court, while affirming an exception.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Case-Law Development 81
The language of De Cuyper is similar to that of Cassis de Dijon. It is the
restrictiveness of the measure which is considered to be egregious, but it will
be considered lawful if it is in the public interest and proportionate. Such
language exposes a variety of national measures to challenge on grounds of
their restrictiveness—anti-terrorism detention orders requiring individuals to
remain within an address, bail restrictions, jury service, anti-social behaviour
orders containing restrictions on movement, or even restrictions on residence
which are imposed to prevent tax evasion. (Chalmers and Monti 2008: 127)
Alongside the incremental establishment of rights that pertain to EU citizens,
the Court has expanded eligibility for these rights. For instance, it has allowed
parents to partake in the citizenship rights of their children, which will be
discussed further in Chapter 6, and has restricted member states’ power to

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

expel EU citizens for reasons of public security (C-482/01). Thus, the ECJ has
instituted high barriers for member states in instances where they use the
public policy provision as a means to impede the rules of free movement.
To summarize this development, citizenship rights have had an astonishing
journey from their status as a rather meaningless addition to the Treaty to
the point where they have become a central tenet of EU law. For some time,
the ECJ has pursued an approach that upholds equal rights through its
non-restriction approach, making the nationality of citizens seem increas-
ingly unimportant. The Court aimed to align its case law to the rights of
workers and non-workers (Wind 2009: 243). By applying the principle of
non-discrimination on the basis of nationality, alongside the prohibition
of restriction, the case law on citizenship made inroads into national welfare
systems. An increasingly contentious issue was whether member states could
still privilege their own nationals according to EU law. Increasing economic
heterogeneity, after rounds of Eastern enlargement and the Eurozone crisis,
politicized potential ‘welfare migration’ (Blauberger and Schmidt 2014). With-
in this context, and that of the UK’s Brexit discussions, the Court has become
more reluctant to emphasize the equal rights EU citizens have to access
benefits. Under mounting political debate and increasing opposition from
rich Western member states, the Court has begun to halt the expansion
of rights, fine-tuning the definition of member states’ obligations to non-
economically active EU citizens.
In Case C-333/13 Dano, which was decided upon in November 2014, the
ECJ held that a Romanian who had neither worked nor intended to do so
could be denied social assistance in Germany. In Case C-67/14 Alimanovic,
Germany was allowed to deny unemployment benefit for over six months to
those who have worked for less than a year. Case C-299/14 Garcia Nieto,
another preliminary reference from Germany, permitted further restrictions.
In Case C-308/14 Commission v UK, AG Villalon held the contentious British
‘right to reside’ test to be compatible with EU law, and the Court followed this
assessment shortly before the Brexit referendum. In view of mounting political
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

82 The ECJ and the Policy Process

opposition, the expansion of EU rights was halted (Kramer 2016). Before
Dano, members of the German social-benefit administration expected the
ECJ to abolish all distinctions between nationals and EU citizens (interview
Senatorin für Soziales, Bremen, 27 April 2016). Halting the expansion of case
law does not imply reversion; it remains to be seen whether the political
pressure on the Court will become so high that this step is taken. At present,
the Court has put its interpretations of the four freedoms and citizenship on a
common path. If it refrains from further extending EU rights in this area, legal
uncertainty will decline and so too will the incentives litigants have to use EU
law to pursue an extension of rights.

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

A Unified Restrictions Approach

Thus far, I have laid out how the early interpretation of the freedom of goods,
which established a prohibition of restrictions approach, began a path that
provided incentives for many private actors to argue analogously for the other
fundamental freedoms. Such positive feedback broadened the path, helping to
establish it for all the freedoms—goods, services, establishment, free move-
ment of workers, capital, and, finally, citizens. This path-dependent process of
convergent interpretation is mainly driven by litigants, with the support of
the Commission and the EU law community. Because of direct effect and
supremacy, the fundamental freedoms can be used by litigants to bypass
national regulations. As is the case for the other freedoms, the prohibition of
restriction is coupled with the question of whether there are public interests to
justify it (‘mandatory requirements’) and, should this be the case, whether the
measure is proportionate. The results of this proportionality assessment
cannot easily be predicted, making the actual scope of the fundamental
freedoms legally uncertain. Derlén et al. (2012: 534–5) also consider this
path of convergent interpretation, which is developing in the case law, to
justify restrictions, based on a network analysis of how all Court rulings cite
each other: ‘Using resampling, we conclude that Member States Justifying
Restrictions of Basic Freedoms constitutes an independent area of law separate
from the case law dealing with other issues regarding those basic freedoms
which, at this juncture, remain independent from each other’ (Derlén et al.
2012: 531, original emphasis). ‘[T]he Court appears to take a more unified
approach as regards justifications’ (Derlén et al. 2012: 535). My analysis of the
Court’s interpretation as a path-dependent process can be seen to present the
microfoundations that underlie this clustering of citations.
Next to the positive feedback from litigants who work towards convergence,
there are internal legal reasons that explain the transfer of the interpretative
logic applied to the free movement of goods to the other freedoms. In reading
the Court’s rulings, it is striking how often several of the freedoms are relevant
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Case-Law Development 83
to a decision. An analysis of all court cases pertaining to the fundamental
freedoms reveals how often a freedom is taken up by the Court in combination
with other freedoms. It appears that the freedom of goods is the only freedom
frequently applied on its own. Therefore, it is easier to pursue a slightly
different legal path in this area. Once the Dassonville logic spreads through
positive feedback to one of the other freedoms, however, there is indeed
pressure to interpret all freedoms in the same way.
If several basic freedoms are relevant to a case and they are interpreted
analogously, this strengthens the coherence of case law (Lenaerts and
Corthaut 2008). Figure 3.5 shows how common it is for the different freedoms
to be simultaneously relevant in a single ruling.9
Once two or more basic freedoms are relevant to a case, legal coherence can

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

only be maintained if the legal reasoning does not differ. The opinion of AG
Colomer in the first Volkswagen case emphasized that the parallel interpre-
tation of the freedoms makes it ultimately irrelevant whether the free move-
ment of capital or the freedom of establishment applies to the case (Blauberger
et al. 2012: 54).
In any event, I see no point in delving any deeper into an incorrect legal
classification of the alleged infringement, which is of no great consequence,
since the Court of Justice subjects both Community freedoms to similar scrutiny,
and I propose to apply that methodology below in order to establish whether the
infringements complained of have taken place ( . . . ). (C-112/05, No. 60)
In a similar line of reasoning in his opinion to Cases C-158/04 and C-159/04
Alfa Vita, AG Maduro criticized the restriction of the Keck argumentation to
the freedom of goods as undermining the consistency of the law:
Thirdly, it has been apparent that the rule in Keck and Mithouard is not easily
transposed into the fields of the other freedoms of movement. The Court has
never in fact adopted the ‘selling arrangement’ classification in its case-law
relating to the other freedoms. . . . This difference in approach raises a problem
of consistency in the case-law. This problem appears to be even greater as many
national measures examined by the Court from the perspective of the free
movement of goods can also be treated as restrictions on the other freedoms of
movement. (No. 33)
The litigation behaviour of private actors, and the attempts to ensure the
consistency and coherence of the legal order, act as positive feedback to a path-
dependent development, I have argued. The judges themselves are relatively
free to rule either in favour of similarity or difference. The example of the Keck
jurisprudence may help to show the different weight of both factors. From the

9
This is the result of a EUR-Lex search of preliminary and infringement procedures for the
relevant freedom in the case summary. The analysis discusses whether other freedoms are
relevant to the rulings. An earlier version with less information appeared in Schmidt (2010).
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

84 The ECJ and the Policy Process

30

25
number of cases

20

15

10

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

0
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
Free movement of goods - without other freedoms Free movement of goods - with other freedoms
Free movement of services - without other freedoms Free movement of services - with other freedoms
Freedom of establishment - without other freedoms Freedom of establishment - with other freedoms
Free movement of workers - without other freedoms Free movement of workers - with other freedoms
Free movement of capital - without other freedoms Free movement of capital - with other freedoms
Free movement of persons - without other freedoms Free movement of persons - with other freedoms

Figure 3.5. Several freedoms in a ruling, 1990–2014

Source: Adapted from Susanne K. Schmidt, Who Cares about Nationality? The Path-Dependent Case Law of
the ECJ from Goods to Citizens. Journal of European Public Policy, 19(1) (2012), 13.

perspective of legal dogma, a transfer of the Keck argumentation would

strengthen the coherence of the legal order. However, there is little positive
feedback for such an extension of Keck to the other freedoms, as actors who
prefer a restrictive reading of the basic freedoms will not turn actively to the
ECJ (Scharpf 2010: 221). They may only make a Keck-like argument when
they have to defend national regulations before the ECJ. This role is restricted
to governments that are either involved in the cases directly or as intervening
parties. The transfer of the more restrictive Keck argumentation to the other
freedoms depends, therefore, solely on the Court’s assessment of whether it
needs such a restriction, which is only likely in the face of significant political
contention.
What does this dynamic, which moves towards convergence and a legally
coherent interpretation, imply? I have shown earlier that the Court, supported
by the legal profession, is well placed to bring legal innovations into case law.
Yet not all innovations are critical junctures leading to a new path of case-law
development. Private litigants only provide positive feedback to those inter-
pretations that extend their rights. Therefore, it is important to take a look at
those areas where the interpretation of the fundamental freedoms is still fluid.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Case-Law Development 85
Before turning to this question, I would like to emphasize how easily
national policy decisions can be seen to interfere with the four freedoms
according to the restrictions path. I will give a few more examples here.
Opponents of the UK’s 2004 Hunting Act—this prohibits the hunting of
large mammals, such as foxes, with dogs—argued for a reference to the ECJ
by claiming that the import of dogs and horses from Ireland was suffering
(Ryland and Nurse 2013: 105). Cases that did reach the ECJ were the follow-
ing: C-110/05 Italian trailers and C-142/05 Mickelsson. The first case con-
cerned an infringement procedure, where the Commission took Italy to court
over prohibiting motorcycles that tow a trailer, as this effectively bans imports
from other member states. The Court agreed that such a prohibition violated
Article 34 of the TFEU; however, it could be justified in the interest of road

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

safety. Barnard (2009: 576) argues that this case may mean that the Court is in
the process of reconsidering the Keck restrictions on the free movement of
goods, replacing it with a restrictions analysis. Mickelsson was a preliminary
procedure that questioned the almost complete Swedish ban on jet-skis in
navigable waterways. Again, the Court regarded the ban as a ‘measure having
an equivalent effect to quantitative restrictions’. This being a preliminary
procedure, it was up to the domestic court to balance the justification of
environmental protection and the restriction on the freedom. Both cases
were notable as the AGs had discussed in their opinions whether the Keck
rule could be transferred to justify restrictions on use (Barnard 2013: 141–2).

THE E VOLVING INTERPRETATION OF CASE L AW

Depending on the breadth of the interpretation of the rights of free movement,

the sphere of the market is strengthened at the expense of member-state
governments’ ability to regulate their markets. Given the need to strike a
balance in the assessment of the proportionality of measures, there is persis-
tent legal uncertainty regarding the reach of the freedoms vis-à-vis the scope of
the exemptions. But it is not only the act of balancing that contributes towards
legal uncertainty. In addition, the interpretation of the freedoms is not fixed.
While the prohibition of restrictions is well recognized, additional incipient
developments can be mentioned here, which can lead to a significant broad-
ening of the application of the fundamental freedoms. These are: the horizon-
tal effect of the freedoms; a prohibition of reverse discrimination against
nationals; and, closely related to these, the application of the freedoms to
purely internal situations.
First, it is a matter of contention to what extent the fundamental freedoms
have a horizontal direct effect and are binding, not only for public but also
for private actors. This would be another critical juncture. In its rulings in Case
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

86 The ECJ and the Policy Process

C-341/05 Laval and C-438/05 Viking, the ECJ supported such an understand-
ing and assessed the actions of the unions with regard to their effects on the
freedoms of establishment and services (Joerges and Rödl 2008; Höpner and
Schäfer 2010; Höpner and Schäfer 2012). These much-discussed cases were
decided upon in December 2007. In Laval, a Swedish union had boycotted a
building site where a company had posted Latvian workers, refusing to sign a
collective agreement with the Swedish unions. In Viking, a ferry route from
Finland to Estonia had reflagged the ships from Finland to Estonia, leading to
a union boycott. In both cases, the ECJ ruled that the right of unions to take
industrial action—while being a fundamental right—needs to be proportion-
ate; in other words, it must be suitable for the achievement of the given
objectives, as strikes restrict private employers’ fundamental freedoms. AG

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Maduro even argued that the freedom of establishment should have a hori-
zontal direct effect, but the Court did not completely share his opinion in the
Viking case. As a result, it is not clear whether and to what extent Article 21(1)
of the TFEU on citizenship, Article 49 of the TFEU on establishment, and
Article 56 of the TFEU on the freedom to provide services have a horizontal
direct effect (Barnard 2013: 242).
For the free movement of goods, the question appeared more settled. Thus,
Barnard (2013: 77) denied the existence of such a horizontal effect, referring
only to the indirect responsibility of member states to guarantee the free
movement of goods. The Court took the position in Case C-265/95 Spanish
strawberries that Article 34 of the TFEU prohibits state action and state
inaction vis-à-vis private restrictions. In this case, the Commission had brought
France before the Court for its lack of action in the face of French farmers’
recurrent obstruction of agricultural products from other member states,
notably Spain. However, Case C-171/11 Fra.bo, which was decided upon in
September 2012, has now established the horizontal effect of the free move-
ment of goods. The case concerned an Italian manufacturer of copper fittings
for water and gas pipes.10 It had brought a legal action against a German
standardization body, the German Technical and Scientific Association for
Gas and Water (DVGW), which resulted in the preliminary proceeding. The
DVGW, a private, not-for-profit body, issues standards for water pipes that are
relevant for marketing products in Germany. Fra.bo had received a preliminary
certificate for five years in 2000, which the DVGW did not renew because a test
result for a more demanding technical specification was missing. The DVGW
argued that as a private body it was not bound by the provisions of the free
movement of goods. Moreover, it did not pursue economic activities. The
Court, however, reiterated its interpretation of rules ‘capable of hindering,
directly or indirectly . . . intra-Community trade’. Article 28 of the TEC also

10
There is a documentation of the case carried out by the company. See: <http://www.frabo.
com/deu/frabo/der-fall-frabo> [accessed 10 May 2017].
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Case-Law Development 87
applied to the standardization and certification activities of a private law body,
if national legislation referred to it (Heller 2012; Kubicki and Obajtek 2012).
The case demonstrates quite well the predicament in which the Court finds
itself. If states delegate regulatory competence to private actors, this cannot
circumvent their responsibility towards the single market. However, binding
private actors by the four freedoms significantly extends the reach of EU law.
The free movement of workers has a horizontal direct effect, at least when it
comes to questions of non-discrimination. This was first established in Case
36/74 Walrave in 1974 for reasons of nationality (Schepel 2012: 177–8;
Chalmers et al. 2010: 797) and was confirmed for all matters of non-
discrimination in Case C-281/98 Agonese, which was decided upon in 2000
(Barnard 2013: 77, 241). What should also be noted here is the ruling in Case

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

C-415/93 Bosman on football transfers. The free movement of capital (Article
63 of the TFEU), as a latecomer, is generally less developed than the other
freedoms. Here the Court could have referred to a horizontal direct effect in
the Volkswagen case (C-112/05) in relation to the private agreement between
workers and trade unions of 1959, but it refrained from doing so (Barnard
2013: 586). Schepel (2012) argues that extending the doctrine of a horizontal
direct effect to the free movement of capital would undermine the different
varieties of capitalism currently present in the EU.
Secondly, the scope of application of EU law would increase greatly were
EU law interpreted as prohibiting reverse discrimination against nationals.
As mentioned earlier, when explaining the difference between the non-
discrimination and non-restriction approaches, reverse discrimination against
nationals occurs when EU foreigners profit from a more beneficial regulatory
regime than nationals. In upholding stricter market regulation, governments
then discriminate against their own nationals. A well-known example is the
German beer-purity regulation that constrains German brewers, but not other
breweries who sell beer in Germany. It is not uncommon for EU lawyers to
argue that the different treatment resulting from reverse discrimination
against nationals is prohibited by EU law (Tryfonidou 2009). After all, it
may seem absurd that a legal order establishing non-discrimination on the
basis of nationality among EU citizens leads to discrimination within member
states (AG Sharpston C-34/09, Zambrano, No. 144). So far, however, the ECJ
has not established this principle, which member states very much oppose,
judging from the opinions they present to the ECJ. Thus, EU citizens living in
a member state may be subject to more beneficial rules concerning, for
instance, family unification with third-party nationals than those citizens
who are subject to national law. If EU law rendered discrimination against
nationals impossible, this would imply that all national regulations needed to
be abolished as soon as laxer regulations from other member states were
mutually recognized. Such an interpretation would severely curtail the poten-
tial to pursue policy goals through regulation at the national level.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

88 The ECJ and the Policy Process

In the absence of prohibition of reverse discrimination at the EU level,
however, national fundamental rights may be violated. For instance, if more
weakly regulated goods or services do not pose a risk to the general interest, it
is not really possible to justify restrictions to the right of professional freedom
(Riese and Noll 2007). As we shall see in Chapter 7, Italy is an example of a
member state where reverse discrimination is prohibited by the constitution
(Conant 2002: 14). Consequently, EU law has a far-reaching domestic impact.
If an extension of EU law towards the prohibition of reverse discrimination
against nationals were to take place, there would no longer be regulatory
competition, but national regulations would be struck down immediately,
alongside which the least restrictive European regulation would be allowed.
Again, this would mean a critical juncture resulting most probably in extensive

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

positive feedback for a new path of legal interpretation.
Thirdly, and closely related to the prohibition of reverse discrimination, an
extension occurs if the ECJ applies the freedoms to cases that are purely
internal to a member state. Traditionally, the application of the basic freedoms
requires a cross-border element, and nationals cannot draw on European law
for purely domestic cases. Otherwise, reverse discrimination against nationals
could not occur, as they would have the same rights as EU citizens. In its
jurisprudence on the freedom of goods and services, however, the ECJ has, at
times, violated this principle (Hatzopoulos 2000: 58–62). Thus, the ECJ
delivered the following ruling in Pistre concerning the question of the free
movement of goods within France (C-321-324/94, No. 44): ‘Accordingly,
whilst the application of a national measure having no actual link to the
importation of goods does not fall within the ambit of Article 30 of the
Treaty . . . Article 30 cannot be considered inapplicable simply because all
the facts of the specific case before the national court are confined to a single
Member State.’
It is easy to see that the potential to turn to European law, even in purely
domestic, internal cases, motivates private actors to improve their situation
further by turning to the ECJ. For the freedom of services, the rule has been
eroding (Barnard 2013: 368). An example is Case C-60/00 Carpenter, which
was decided upon in 2002. In this case, the Filipino wife of a British citizen had
been denied residence in the UK. The ECJ argued, however, that since
Mr Carpenter was offering services in other member states, Mrs Carpenter’s
right of residence could be derived from EU law, according to which her
husband’s provision of services should not be inhibited.
Two examples from the area of freedom of movement demonstrate the
incentives that such a broadening of EU rights can offer.11 Case C-535/08
Pignataro (26 March 2009) dealt with the preconditions for eligibility in

11
I am grateful to Ulrike Liebert and Tatjana Evas for telling me about these cases.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Case-Law Development 89
regional elections. Ms Pignataro could not stand for elections in Sicily, as she
did not have her domicile in that region at the time. She then tried to argue
that she had suffered inadmissible discrimination under European law, as
nationals from other member states, who could also stand in the elections,
were not required to take up a domicile in Sicily. The ECJ, however, denied the
relevance of this application of EU law to this purely internal situation. In
another case, that of Mariano (C-27/08, 17 March 2009), an Italian woman
litigated, as she received a very small pension after her (unmarried) partner’s
fatal accident, arguing that an EU foreigner living in Italy would have been
treated more favourably under the same circumstances. Again, the ECJ denied
this application, invoking the argument that it was a purely internal situation.
It is apparent that these and other cases take up the opportunities provided

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

by the ECJ’s decisions in cases like Pistre or Carpenter, where the freedom of
goods and services was applied to purely internal situations. At the same time,
if the ECJ denies a broadening of the applicability of case law in the area of
freedom of movement and citizenship, it becomes less likely that additional
litigants will attempt to invoke this argumentation. This shows the possibilities
not only for positive but also for negative feedback in litigation proceedings.
The examples given show how private litigants orient themselves towards
precedent. This transfers legal arguments from one freedom to another. The
ECJ, however, is neither a slave to its own case law nor to a desirable coherence
in the interpretation of the different freedoms. Moreover, striking a balance
according to the proportionality principle allows it to argue in favour of either
European or national interests. It may present arguments for differences or
parallels between the different freedoms. Whenever it does not further the
scope of the freedoms, as in cases such as Cartesio, Pignataro, and Mariano, or
even when it presents a new legal argument that limits the reach of the
fundamental freedoms, as was the case in Keck, litigants are not rewarded.
In cases such as these, which lack positive feedback, the Court sets out
different priorities concerning its legitimation: while member states may
hold the Court in higher esteem when it protects their autonomy, the com-
munity of EU legal scholars, private litigants, and the Commission reward the
Court for advancing the applicability of EU law.

CO NCLUSION

Under no rule of law is legal certainty absolute. However, I have argued that
there are good reasons to assume that European law is characterized, to an
unusual extent, by legal uncertainty. The compromising character of European
legal acts, as they are decided upon by a heterogeneous set of actors, puts a
heavy interpretative load on the ECJ. The ECJ, as it is committed to the telos of
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

90 The ECJ and the Policy Process

European integration, is bound to interpret these norms in a way that
strengthens the European level. In order to have the scope to make suitable
future decisions, it is likely to let its case law remain legally uncertain, thereby
allowing for future integrative steps, rather than deciding finally on an issue.
Private actors find a set of individual rights in the European legal order
based on the supremacy and direct effect of European law that offer an
alternative to their domestic legal obligations. They draw on precedent, help
to consolidate it, and transfer it to other legal areas. The interpretation of a
prohibition of restrictions that was first established for the freedom of goods
serves, I have argued, as a ‘focal point’ for different actors, initiating a path-
dependent process. Positive feedback results from mobile private actors who
seek out a more preferential legal order and present arguments in favour of

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

analogous developments for the other freedoms as has occurred for the
freedom of goods. Even if private actors face restrictions in legal standing,
successful litigation grants the significant prize of a change to EU law that has
constitutional status. For the Commission and the Court, this use of European
rights fosters the supranational institutions’ legitimacy. And a parallel inter-
pretation of the different fundamental freedoms furthers the coherence of the
legal order, which facilitates the cooperation of the national courts, who apply
EU law autonomously, whenever precedent is sufficiently clear and unambigu-
ous (Alter 2008b: 218).Too many changes in its case law and a differentiated
approach to the different freedoms would complicate and hamper the diffusion
of EU law as the ‘law of the land’. The combination of legal uncertainty and
path-dependent case-law development allows the Court to respond to the
contradictory demands of providing both impetus for integration through
expansive rulings and judicial stability.
It is also possible to detect both a ‘coincidental’ beginning of a path and the
resultant inefficiencies that occur at a later stage. For the freedom of goods, the
Cassis logic does not imply that there will be significant redistributive conse-
quences. This is different for the other freedoms. It is, therefore, highly unlikely
that a similarly far-reaching interpretation could have been established if it had
not been for the ‘coincidence’ of goods markets internationalizing first. Given
the distributional issues that are raised by this path, it can be regarded
as inefficient and dysfunctional in a political sense. In terms of the freedom
of establishment, for instance, the transfer of a far-reaching prohibition of
restrictions undermines market regulations that are regarded to be of central
importance for the national form of capitalism in some member states (Höpner
and Schäfer 2010). A broad interpretation of the freedom of establishment
enables regulatory arbitrage of company law and taxes. Similarly, the extensive
interpretation of the freedom of movement and of union citizenship would lead
to universal access to different kinds of national social benefits (Haltern 2005:
97). The transfer of the non-restrictions approach from goods to other areas,
thus, poses a dilemma. While free trade requires non-discrimination between
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Case-Law Development 91
national and foreign products, this kind of equal treatment between foreigners
and nationals strips national citizenship rights of much of their content.
Ultimately, without any form of exclusion, it would be questionable whether
a state could impel its citizens to fulfil the duties (taxes, defence) necessary for
the state’s survival (Scharpf 2009). This became particularly apparent in the
jurisprudence on the free movement of persons and union citizenship (cf.
Hilpold 2008) in which the Court recently responded with fine-tuning rather
than extending the equal treatment of EU citizens further in light of waning
public support for European integration.
It is important to recognize yet another aspect of these redistributive conse-
quences of path-dependent case-law development: some actors are favoured
over others. Mobile private actors, partly supported by national political actors

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

who favour similar (normally liberalizing) reforms, can find legal positions in
European law that offer them more beneficial regulation in comparison with
the status quo, which is oriented towards the national situation. Actors who
profit from the regulatory status quo, in contrast, cannot defend their interests
with EU law; nevertheless, they are privileged by national law. At the national
level, the actors who support ECJ-driven liberalization are those who cannot
achieve such changes domestically, due to the veto power of other actors. These
vetoes are overruled with the help of European law.
As for the Court, it is not imprisoned in its legal doctrine. The judges may
deviate from precedent and its attendant path. There are, however, strong
positive incentives to honour precedent, as there are to follow a convergent
approach in the interpretation of all fundamental freedoms, for instance, by
facilitating the application of EU law through the lower courts. Path deviations
are initiated mainly from within the legal system and are hardly ever the result
of political steering. These occasional shifts are not predictable. At what point
exactly the ECJ draws a line between the need to honour European obligations
and national regulatory concerns is, then, somewhat unclear to the participat-
ing actors. This points to the relevance of the concept of legal uncertainty.
By not following certain arguments, the Court, after a while, demotivates
actors from pursuing a certain line of reasoning in litigation since there is
negative feedback.
The advantage of a path-dependent argument is that it takes ‘the law
seriously’ (Joerges 1996) and allows for the independent political influence
of legal doctrine (Smith 2008). If the content of legal texts scarcely contributes
to the Court’s decision-making, as some scholars of rational choice appear to
assume in the way they simply relate rulings to governmental power or judges’
political preferences, it becomes difficult to understand why rational actors try
to influence the legislative process in the first place, or why it should make a
difference to courts that they can be overruled legislatively (Dyevre 2010: 311).
Having discussed legal uncertainty and path dependence, we can look
at the question of legal activism or judicial power once more. Due to
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

92 The ECJ and the Policy Process

over-constitutionalization, highly dynamic case-law development can also be
produced by a court that is not very activist. Operating between legal uncer-
tainty and the path-dependent development of case law, the Court does not
require a pronounced activist orientation to push the process of case-law
development along. In having to honour precedent, legal consistency, and a
cooperative spirit with lower courts, the Court answers individual litigants with
the aim of improving their regulatory position. The Court’s interest in the
consistency of the law leads to converging interpretations of the four freedoms
and citizenship rights, and, given the complexity of the European polity, the
Court must be largely ignorant of the implications of its decisions in the
heterogeneous polities of the current twenty-eight member states.
On the basis of these considerations, I will turn now to the implications of

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

constitutionalized case-law development for EU policymaking.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

The Interaction of Judicial

and Legislative Policymaking

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

The last chapter explained the dynamic expansion of case law, which progresses
in a path-dependent fashion and is driven by interested actors and legal
uncertainty concerning the boundaries of European law. We will now turn
to the legislative process and analyse the shadow that case law casts over it.
By analysing how case law impacts legislation in the European Union (EU),
we can explain why the European Court of Justice’s (ECJ) importance
is not solely reliant on the direct implementation of its rulings. The over-
constitutionalization of the Treaty implies that there are far-reaching constraints
on EU policymaking.
As we have seen, the broadening of the jurisprudence on the fundamental
freedoms develops incrementally through case law—it is not explicitly decided
upon by the member states in the Council or by the European Parliament (EP).
Unlike other international regimes, the EU has a relatively effective legislative
process that makes it possible to tie case law to the political process through
consensual decision-making. This possibility indirectly legitimizes the far-
reaching case law of the ECJ (Dobler 2008: 530). As Scharpf (2009) argues,
the possibility to rein in the Court through legislative acts constitutes the
ultimate input legitimacy to the output legitimacy of court judgments. But
within the EU, the legislative process cannot overrule the Court if it has based
its rulings on the Treaty. The constitutional nature of case law does not allow it.
By embedding case law subsequently in secondary law proposals, the Commis-
sion ensures that case-law obligations are enforced. This is one important
solution to the puzzle of why member states do not simply respond with non-
compliance to the considerable redistributive consequences of case law.
Case law can be equivalent to legislation. Rules are then determined by
courts and not by legislatures. However, regulation through case law leaves a
lot of legal uncertainty, since judge-made law is oriented towards single cases.
Though Weiler demonstrated long ago that in the EU judicial policymaking
is an alternative to legislation (Weiler 1981, 1991), the interaction between
legislative and judicial policymaking is analysed as more of an exception
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

94 The ECJ and the Policy Process

than a rule in European integration studies. Over-constitutionalization is
ignored, with its many direct implications for the content of European
policies. There is the implicit assumption that the status quo in EU legis-
lation is set by the member states, as liberal intergovernmentalism argues
(Moravcsik 1993), or by an existing piece of European legislation. Often,
however, the ECJ’s judicial policymaking defines the default condition of
non-agreement, and lawmaking takes place in the shadow of the Court.
This chapter systematizes the different ways that judicial policymaking can
impact upon European legislation. Depending on how the interpretation of
European law develops, driven by private litigants and the Commission, the
legislative process can be pressured to respond. The European Commission is
crucial in this respect. It enjoys a formal monopoly on initiative. It drafts the

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

proposals and introduces its interpretation of the Court’s judgments under the
purview of its powerful legal service. The Commission here has the role of a
gatekeeper, as it is the only actor that can systematically broaden the case-
specific consequences of rulings into general implications. Moreover, the
Commission may actively foster judicial pressure, combining its legislative
role with its role as a guardian of the Treaty, emphasizing that codification is
needed for greater legal certainty. The term codification denotes that regula-
tory principles that have been developed in case law are written into secondary
law. This provides general rules, while the implications of case law are difficult
to judge by those it concerns. Acknowledging codification adds to EU research
in two important ways. First, it shows how EU legislation is embedded in case-
law development. Secondly, this codification reveals that the impact of case
law cannot be reduced to the question of compliance with single rulings.
I argue that the relative neglect of judicial policymaking in research is
rooted in the difference that we find between the rule of law at the national
and the supranational levels. The ‘comparative turn’ in EU studies (Hix
1994) has resulted in treating the EU like any national political system.
However, since it has an intergovernmental Treaty as a constitution, this
provides a very different basis. When referring to the Treaty, case law has
constitutional status and EU secondary law has to abide by it—a constraint
that is generally overlooked by political scientists. Courts are much more
efficient decision-makers than legislators, particularly at the supranational
level. Given the multilevel European system, case law is developed in
response to cases that are situated in very different national contexts, making
it difficult to foresee its implications. EU legislation is embedded within this
context of case-law constraints.
Secondary law (legislation) cannot overrule primary law (the Treaty).
However, as marked by legal uncertainty as the latter’s implications are, and
as much the Court is an actor dependent on the support of not only lower
courts and litigants but also governments, in making the will of the legislature
explicit, there is the hope of influencing the interpretation of the Court.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Judicial and Legislative Policymaking 95

This chapter proceeds as follows: I begin by discussing how judicial
policymaking can be brought into the legislative process of the EU, differentiat-
ing between several types of judicial ‘shadow’ on the legislative process, analysing
their features through small case studies. After first focusing on the way that the
Commission may combine its legislative and judicial roles, I then move on to
discuss member states’ preferences when they are confronted with case law in the
legislative process, focusing on the Services Directive and the regulation on
the mutual recognition of goods. This chapter presents examples of policies
that liberalize markets. The next chapter will analyse cases that deal with issues
that are more sensitive to sovereignty, such as individual social rights, healthcare,
and taxes.
Before delving into the impact of case law on legislative policymaking, I will

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

discuss expectations relating to how legislative actors position themselves vis-
à-vis case law. This principally concerns the member states’ governments, but
also the Commission and the EP.

NEGOTIATING L EGISLATION IN THE S HADOW

OF THE COURT

Decision-making in any arena is highly dependent on the question of whether

some or all actors have alternative means to realize their interests. This puts
the focus on the question of the default condition: what happens in instances
of non-agreement (Ostrom 1986)? Many legislative studies of the EU implic-
itly or explicitly assume that the default condition of legislation in the EU is
continued national prerogative. This implies that member states’ bargaining
positions depend on the extent to which they favour cooperation as opposed
to national, independent decision-making. However, European legislation is
frequently not concerned with new areas of cooperation but about reforming
existing ones. One example is the case mentioned in the original joint-decision
trap article, the Common Agricultural Policy (CAP) (Scharpf 1988). Once
there is a policy in place in the EU, those actors who favour its continuation
are in a particularly strong bargaining position, given that a minority can block
a reform that is favoured by the majority; in cases of unanimity rule, this can
be even a single member state. Most importantly for our context, the default
condition can be defined by the ECJ’s case law due to over-constitutionalization
and the abundant policy content of the Treaty. Davies (2014: 1593) even speaks
of ‘the legislative competence of the Court’, which he argues to be ‘broader than
the legislative competence of the legislature’.
Table 4.1 shows the consequences of the decision rule of qualified majority
voting or unanimity at the supranational level: those who favour the status
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

96 The ECJ and the Policy Process

Table 4.1. Types of default condition and bargaining power in European legislation
Type Default condition Bargaining power lies with:

A) National legislation → Those favouring non-cooperation.

B) Existing EU legislation → Those favouring legislative status quo
C) Existing case law → Those favouring existing or likely case law, which
is either extensive or restrictive (= type A).

quo are always in the stronger bargaining position, reflecting that it is easier
to maintain the status quo than to change it. If there is case law (Type C), and
the Court does not simply back national competence (Type A), member
states that favour existing or likely case-law development have a stronger

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

bargaining position.
How can we expect case law that details policy requirements to influence
decision-making in the EU? Assuming that there is case-law development and
the Court does not give a large amount of leeway to member states’ regulatory
objectives, legislation is constrained by it. First of all, it becomes an issue of
how the Commission integrates case law into policy proposals in order to
codify it into secondary law. Secondly, the bargaining power of member states
will differ according to the interpretation of the Court. Thirdly, member-state
governments can be expected to have diverse preferences regarding codifica-
tion, depending on their preferences on policy content, their normative
understandings of the role of legislatures, the uneven use of Court-made rights
by private actors, the unequal implementation of case law among member
states, and the importance of legal uncertainty.
First of all, the impact of case law on legislation depends on how the
Commission picks it up in its agenda setting. With its formal monopoly on
initiation, it has a gatekeeping role. Moreover, of all legislative actors, it is
the only one that also systematically takes part in the judicial process, as the
guardian of the Treaty. Thus, it may simultaneously be a player in the
legislative and judicial realms. The Court structures decision-making by
foreclosing or opening up certain options at the proposal stage for all legisla-
tive actors, including the Commission. For example, Hartlapp et al. (2014:
88–91) show that the Directorate-General (DG) for the Environment could
successfully refer to several cases (C-31/87; C-324/93) and the then pending
Case C-513/99 to convince the DG for the Internal Market to include green
award criteria in public procurement. In these rulings, the Court allowed
member states exceptions (see Type A). Because the single market would
have remained fragmented without a common agreement on policy, regula-
tory interests were strengthened over liberalization. As much as the Court
supports integration through its development of case law, we can expect the
Commission to take up these incentives for legislation, partly codifying case
law into general rules.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Judicial and Legislative Policymaking 97

Secondly, for the Commission and all actors involved in the legislation,
everything depends fundamentally on whether the Court interprets the rights
of the Treaty broadly or not. Member states were made aware of this at an
early stage when the Cassis de Dijon judgment overhauled the building of
the single market (Alter and Meunier-Aitsahalia 1994). While before Cassis
member states could rely on their host-country rules as long as there was no
European harmonization—so that in the Council negotiations the choice was
between the national status quo and a European rule—after Cassis there is the
presumption that the single market already exists and member states mutually
recognize each other’s rules (Schmidt 2007).
Where the Court interprets the Treaty freedoms broadly, and the Commis-
sion takes this case law up in its proposal, the more liberal states in the Council

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

are favoured (Gormley 2012: 52–3). They may threaten simply to await further
case law in line with their preferences, rather than agreeing on secondary law.
In contrast, where the Court interprets the Treaty freedoms more narrowly,1
those member states that favour a high level of regulation have a better
bargaining position in the Council. Retaining regulatory competence allows
them to realize their preferences unilaterally. This threatens other member
states with market fragmentation. Consequently, these may agree to a higher
level of regulation than they would have actually favoured. However, if market
fragmentation is not that significant, and if the option of exemptions for
those member states that have a higher degree of regulation (Article 95,
4–10 of the Treaty Establishing the European Community (TEC)) applies,
European agreement at a lower regulatory level becomes possible.
Hubschmid and Moser give the example of car emissions standards for such
a logic of reregulation in the Council. The ECJ allowed member states to
pursue stricter environmental standards in its Case 302/86 Danish Bottles.
This changed the default condition, labelled ‘reversion policy’ by Hubschmid
and Moser. Those member states that favour stricter environmental standards,
like the Netherlands, Denmark, and Germany, could pursue them after Danish
Bottles on a unilateral basis and started to do so (Hubschmid and Moser 1997:
238; De Witte 2012: 31). Because they wanted to avoid market fragmentation,
member states like France and Italy, who had blocked the decision before,
agreed to stricter environmental standards. Similarly, Kilpatrick (2012: 211)
sees the restrictive Rush Portuguesa ruling of 1990 as having favoured stricter
regulation in the original Posted Workers Directive. A wide or narrow inter-
pretation thus strengthens actors who favour liberalization or regulation in the
negotiations. But this is not the only impact the Court has.

1
It allows exceptions for member states due to health, environmental, or other reasons that
follow the mandatory requirements. It also allows member states to draw on the right of
exception, which Article 95 (paragraphs 4–10) of the TEC stipulates.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

98 The ECJ and the Policy Process

Much depends, thirdly, on member states’ preferences regarding the codifi-
cation of case law. There are benefits and disadvantages for the legislature in
responding to case-law development. Depending on how case law develops,
member states may be interested in using codification to signal their policy
preferences to the Court. If case-law development is rapid and the number of
potential litigants is high, member states may hope that their agreement on
secondary law will play a role in the considerations of the Court (‘signalling
effect’) (Garrett et al. 1998). From these considerations we can surmise that
those member states whose preferences conflict with case-law development
are interested in codification. In particular, when case law shapes policy issues
of greater political salience, we may expect the legislature to be interested in
participating in the shaping of policy through codification. The Council and

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

the EP have a much more legitimate authority than the Court to shape
European policies. In areas of important case-law development, this may
also lead to the normative conviction that policymaking cannot be left to
the Court.
Codification generalizes rights that are established by case law. As case law
is difficult to understand, private actors will take up these rights unevenly
(Conant 2002), raising concerns about equality before the law. Differences in
implementation among member states give an incentive to support codifica-
tion. If case law is implemented more unevenly than secondary law, those
member states that implement case law will likely push towards a level playing
field. In contrast, those member states that reject the policy implications of
case law may resist codification, as a low take-up of rights implies that it has
restricted impact. Reflecting on these considerations, we can surmise that
those member states that implement case law according to the logic of the
rule of law are likely to favour codification. This promises greater equality
before the law, as well as a level playing field vis-à-vis other member states.
Member states that aim to constrain the impact of case law will, conversely,
resist codification.
Legal uncertainty, finally, may be another important consideration here.
Patchy case-law development may make it unclear what exactly European law
is demanding from the regulation of a policy field. A breach of EU law is not
always without cost. It can lead to liability claims, for instance in public
procurement. Those who have to make long-term investment decisions may
need to know how stable certain regulatory privileges are (Schmidt 2000).
Therefore, there may be an interest in codifying case law and setting up
general rules, allowing the re-establishment of legal certainty. It may be,
however, that only private actors pay the price for legal uncertainty, making
codification less of a priority, particularly if member states prefer rights not to
be diffused and used. We can deduce from these points that member states
have an incentive to codify if they perceive there to be costs related to legal
uncertainty. If legal uncertainty that arises from case law only impacts
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Judicial and Legislative Policymaking 99

relatively marginal actors, the incentive may be smaller. If there are benefits
to be derived from legal uncertainty, as the rights established by case law are
barely used, this may make non-codification attractive.2
To sum up, member states’ positions in relation to codification depend on
several factors. What are the normative costs of regulation carried out by the
Court? What are the gains and losses of an uneven implementation of case-law
regulations? How rapid is case-law development, and what is the worst-case
scenario that is associated with it? The policy interests of member states have
to be weighed against these factors. When legal uncertainty is highly disad-
vantageous, one can expect member states to be willing to forego their policy
preferences to a certain extent in order to achieve legal certainty.
In addition to the member states and the Commission, the European

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Parliament is equally affected by the interplay of case law and legislation.
Along with the Commission and the Court, the EP is generally believed to be
an actor that is pro-integration. In line with the general neglect of over-
constitutionalization by legislative analyses of the EU, not much is known
about how the EP positions itself in relation to ECJ case law. In fact, members
of the EP seem to be little aware of it. The Services Directive, analysed in this
chapter, shows this clearly.
If the Commission takes it up, case law can be generalized into secondary
law. Table 4.2 (adapted from Schmidt 2008) shows continuity in the relation-
ship between judicial and legislative policymaking. In extreme cases, there is
judicial dominance, in which case law replaces legislative action, and legisla-
tive dominance, in which case law has no influence. The latter is the standard
assumption in EU legislative studies. I will not discuss these two extreme
positions in this book. Between these two poles, the Commission takes case

Table 4.2. The interrelationship between judicial and legislative policymaking

Judicial Legislative
Interaction
dominance dominance

Case law Active pressure: Passive linkage of Case law has no

replaces Lesser evil, divide case law with influence =
legislative and conquer legislation: Services standard
action Dir., Goods Reg., assumption of
Citizenship Dir., EU legislative
Patient Mobility Dir. studies

Source: Adapted from Susanne K. Schmidt, Europäische Integration zwischen judikativer und legislativer
Politik. In Martin Höpner and Armin Schäfer (eds), Die Politische Ökonomie der europäischen Integration
(Frankfurt a.M.: Campus 2008), 111.

2
Blauberger (2014) makes a similar argument for the member-state level, whether they react
to case law with anticipatory obedience (out of fear of liability claims and/or further litigation) or
contained compliance, if such pressure is absent.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

100 The ECJ and the Policy Process

law not only generally to give support to its preferred policy position, but as a
specific instrument to force through changes. I summarize this active role first,
before I then turn to a more extensive analysis of how legislation is shaped
under the shadow of case law.

PRESSURIZING THE L EGISLA TIVE PROCESS:

STRATEGIES OF THE COMMISSION

On several occasions, I have described the way that the Commission can use
its privileged position in relation to the Court to push its legislative proposals

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

through (Schmidt 2000, 2004, 2011a). I will therefore only provide a summary
here. By utilizing its role as a guardian of the Treaty, in possible combination
with its executive competences in competition law,3 the Commission can
actively foster case law in those policy fields where it wishes to pass legislation.
Two main strategies can be distinguished here (Schmidt 2000): ‘lesser evil’ and
‘divide and conquer’. Reaching an agreement on legislation in the Council
becomes a ‘lesser evil’ for member states when the Commission initiates cases
at the Court that threaten case law and legal uncertainty. But the Commission
may also advance proposals incrementally through a strategy of ‘divide and
conquer’. It may use infringement procedures to initiate domestic changes in
some member states, thereby breaking down opposition in the Council and
winning allies.

Making an Offer that Can’t Be Refused:

The Lesser-Evil Strategy

In the case of lesser evil, governments can avoid a lingering worst-case

scenario by reaching an agreement in the Council. I became aware of this
strategy when analysing electricity liberalization in the 1990s (directive 96/92/
EC) (Schmidt 2000). Although, as a single-market measure, the Council only
required a qualified majority, given the important role France had as the main
electricity exporter, the case was one of de facto unanimity. France firmly
opposed liberalization, but could not be marginalized on this issue. As there
seemed little chance of reaching an agreement in the direction of liberalization,
the Commission initiated infringement proceedings in 1994, putting pressure

3
Note that while in the rest of this book I focus on the four freedoms, competition law also
becomes relevant in terms of the strategies that are employed. The Commission can, to some
extent, choose whether to use competition-law powers against a dominant position or to
strengthen the fundamental freedoms.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Judicial and Legislative Policymaking 101

on member states. It argued that the import and export monopolies of five
member states, including France, infringed upon the market freedoms (Slot
1994: 525). For member states, this posed the threat that their electricity
monopolies might incrementally fall under European competition law and
the market freedoms. Given the legal uncertainty resulting from case law
and the long-term investment needs of the sector, this was highly undesirable.
This credible threat of litigation changed the default condition for the
member states. While they preferred their national policy to a common European
one, they preferred the latter over a stepwise dismantling of monopolies
through the Court. France therefore cooperated, allowing the Council to
reach an agreement in 1996, just before the Court ruling set out the member
states’ Treaty obligations. But what happened next shows the inherent unpre-

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

dictability of case law due to the proportionality test, namely the legal uncer-
tainty that arises from the Court’s balancing of the Treaty’s free-market
principles and legitimate exemptions for member states. In its ruling from
October 1997 (C-157-160/94), the Court largely sided with the member states
and not with the Commission. Had member states known this would happen,
it is unlikely that they would have agreed to liberalization.
Another example here is the liberalization of road haulage (Schmidt 2004).
Despite the dedicated chapter for transport in the Treaty, the European
Economic Community remained largely inactive in the 1960s and 1970s
(Héritier et al. 2001). It proved difficult to overcome the traditional system
of bilateral quotas for road haulage, and harmonization was unsuccessful.
Change came in 1985 when the Court ruled, after instigation by the EP, that
the Council had failed to implement the transport chapter of the Treaty.
Member states took this ruling as a threat that ‘the Court would directly
apply the Treaty . . . which could have meant the instantaneous liberalization
of the road haulage market’ (Young 1994: 6).
Actors such as the Netherlands and the Commission, who supported
liberalization, were strengthened in their bargaining position and repeatedly
threatened to refer the matter back to the Court if the Council failed to lift the
restrictions quickly. In light of this, the Council reached an agreement to
liberalize all transit transport from the beginning of 1993 (Regulation 1841/88).
Comparing the votes of two coalitions in the Council shows how important
the Court’s threat was. The original opposition from France, Germany, Italy,
Spain, Portugal, and Greece mustered forty-eight votes, while the supporters
of liberalization—the UK, Belgium, the Netherlands, Denmark, Ireland, and
Luxembourg—could only gather twenty-eight votes in this community of
twelve member states (Young 1994: 15). With the threat of litigation for
liberalization, reaching agreement in the Council became a lesser evil for the
supporters of the status quo. Figure 4.1 depicts the lesser-evil strategy.
Liberalization of air transport is another case where the Commission’s powers
were used at an early stage to bring about phases of liberalization, for which
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

102 The ECJ and the Policy Process

Veto coalition Commission
proposal
A–D E–F G H
SQ0 SQ1

SQ0 SQ1
Regulated markets Liberalization

Commission and ECJ

A–H
Hypothetical MS positions

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

SQ = Status quo

Figure 4.1. Lesser evil

Source: Adapted from Susanne K. Schmidt, Only an Agenda Setter? The European Commission’s Power over
the Council of Ministers. European Union Politics 1(1) (2000), 52, and Susanne K. Schmidt, The European
Commission’s Powers in Shaping European Policies. In Dionyssis G. Dimitrakopoulos (ed.), The Changing
European Commission (Manchester: Manchester University Press, 2004), 111.

the necessary majorities in the Council were lacking (Argyris 1989). Dobson
has recounted how the Commission, with the support of a few member states
and the Court, managed to liberalize the market between 1987 and 1997 in
several stages (Dobson 2010: 1135). Case C-209-213/84 Nouvelles Frontières
was crucial here, in which the Court affirmed that competition rules applied to
air transport. The Commission then drafted letters to ten airlines, containing
the threat of high fines, which coerced the Council into agreeing on liberal-
ization (Kassim and Stevens 2010). In the early 2000s, the Commission
was again helped by the Court when it was struggling against the bilateral
open-skies agreements that had been made by different member states with
the US (Dobson 2010: 1144–5). Similarly, Héritier and Karagiannis have
analysed this case, without differentiating, however, between lesser-evil and
divide-and-conquer strategies (Héritier and Karagiannis 2011: 160).
Rauh and Schneider give general support for the relevance of case law to
regulation. They have analysed investors’ responses to regulatory decisions on
the basis of stock-market information. They find that investors are sensitive
to court cases that indicate imminent regulatory changes (Rauh and Schneider
2013: 1138).
Another important example of the lesser-evil strategy is given by Blauberger
and Weiss (2013). They analysed the Council’s decision on the Defence
Procurement Directive in 2009, after it had refused to make this step a few
years earlier in 2005. In this field, which is sensitive to sovereignty, member
states supported their national policy prerogative by referring to Article 346
of the Treaty on the Functioning of the European Union (TFEU) that allows
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Judicial and Legislative Policymaking 103

an exemption from common-market rules where essential security interests
are concerned. But a ruling of the ECJ (C-414/97) eroded this argument by
interpreting the exception in a narrow way. The Commission picked up this
interpretation after some time and made a proposal for a directive, which
member states accepted in light of the danger of further court cases.
A preliminary procedure (C-615/10) that followed shortly afterwards empha-
sized that this was not merely a theoretical danger but a very acute one
(Blauberger and Weiss 2013). Despite the agreement on legislation, the Com-
mission initiated a second round of threats using the Court, thereby attaining
further concessions from the member states that went beyond the directive
(Weiss and Blauberger 2015).
Other examples may be given here. The way that Bulmer (1994) describes

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

the adoption of the merger-control regulation also shows how the Commis-
sion successfully threatened member states with a worst-case scenario. Public
procurement is a more recent example where case law has had significant
‘influence in the law-making process’ (Bovis 2006: 462) of the public sector
directive (2004/18).

Breaking the Resistance in the Council: Divide and Conquer

In cases of lesser evil, the default condition of negotiations in the Council is

different. Member states no longer decide whether they prefer a common
European policy to their national prerogative, but only whether they prefer it
to regulation through case law. This is not, however, the only way that the
Court, as a political actor, can interfere with legislation in the EU. In a process
that unfolds in stages, the Commission can aim to overcome a blocking
minority in the Council by targeting some of the member states with infringe-
ment procedures or a competition law action. If these member states then alter
their domestic policies in response to this divide-and-conquer strategy, they
are likely to switch sides in the Council and support the Commission’s
proposal for a general policy change.
I originally discovered this strategy when analysing the liberalization of
ground handling at airports. Here, the Commission overcame opposition
by targeting airport monopolies in several member states with its competition
law powers (Schmidt 1998: 286–92). Subsequent legislation was successful.
The various European liberalization policies of previously highly regulated
sectors give further examples of this strategy. These show that the Commission
has used its power to approve state aid as a lever to foster the liberalization of
telecommunications in Italy, and it has made the granting of its approval for
international mergers of electricity providers and air-transport companies
subject to liberalization consent (Thatcher 2007: 193, 216, 231). The third
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

104 The ECJ and the Policy Process

liberalization package for electricity (and gas) also profited from this combi-
nation of the Commission’s competences. Member states opposed the Com-
mission’s plans for ‘ownership unbundling’, which requires operators to sell
their networks, in order to assure non-discriminatory access for competitors.
In particular, Germany attempted to build a veto coalition within the Council,
but the Commission secured a deal with E.ON, the largest German electricity
provider. The Commission had been pursuing cartel charges due to an abuse
of dominant positions in the industry that had occurred through cross-
subsidization and exorbitant retail prices, and had raided the offices of the four
large German electricity companies E.ON, RWE, Vattenfall, and EnBW in
December 2006. E.ON agreed to sell its high-voltage transmission network in
exchange for the dropping of these charges.4 This helped the Commission to

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

deliver a significant blow to the German opposition and break resistance among
the other electricity companies that were opposed to change.
Though these are examples of the Commission using its executive powers
under competition law, they can be subsumed, nevertheless, under judicial
policymaking. After all, the significant independence and breadth of the
Commission’s powers depend on their broad interpretation through the
ECJ. If the ECJ practises judicial self-restraint, the Commission is much
weaker. In such cases, its far-reaching interpretations of its powers under
European competition law are no longer credible to private and public actors
and do not change their behaviour.
The gambling sector provides an example of such a weakening of the
Commission in its use of the divide-and-conquer strategy. After its first
attempt, the Commission had to exclude gambling from the Services Directive.
Based on the freedom of services, there had already been several Court cases
on gambling monopolies at the time (Case C-243/01 Gambelli, 6 November
2003; Cases C-338/04, C-359/04, and C-360/04 Placanica, 6 March 2007)
(Adam 2015). On this basis, the Commission began infringement procedures
against restrictions on sports betting in 2006. Originally targeting nine mem-
ber states (Denmark, Finland, Germany, Hungary, Italy, the Netherlands,
Austria, France, and Sweden) (IP/06/436 and IP/06/1362), the Commission
continued the cases against Germany, Sweden, Greece, and the Netherlands in
2008. With the initiation of additional preliminary proceedings by interested
competitors, liberalization appeared imminent (SWD (2012) 345 final, p. 22).
However, the Court then seemed to back down. In a Portuguese reference
from September 2009 (C-42/07), it questioned whether member states were
required to recognize service providers licensed in other member states mutually
if there was no prior harmonization, since member states may not be sufficiently
assured that consumers are protected against fraud (Hatzopoulos 2013: 494).

4
FT 28 Feb. 2008, FAZ 29 Feb. 2008, FAZ 1 Mar. 2008.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Judicial and Legislative Policymaking 105

This surprising interpretation was followed in later cases, concerning Dutch
(C-258/08), Austrian (C-347/09), and Italian (C-660/11 and C-8/12) references.
For the first time, the Court excluded a sector from the need of mutual recognition,
while emphasizing, however, that member states cannot prevent all cross-border
activity (Hatzopoulos 2013: 495–6). The Commission received much less support
in jurisprudence than it had originally thought it would.
Nevertheless, the Commission pursued its policy goals. In May 2010, it
closed the infringement file against Italy after it had liberalized online gam-
bling (IP/10/504). In October 2010, the case against France was dropped,
which had also liberalized (IP/10/1597). In autumn 2012, the Commission
published a policy document on games of chance, which laid out its reading of
the legal situation (COM (2012) 596 final; IP/12/1135). Although the Court is

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

cautious in its demands, the Commission requires member states at least to
justify their restrictions using a proportionality test. In November 2013, the
Commission announced that it had closed the proceedings against Finland but
requested information from Belgium, Cyprus, the Czech Republic, Lithuania,
Poland, and Romania. It also notified Sweden that it perceived there to be
violations of the freedom to provide services, announcing that it would soon
turn to the Court if Sweden refused to alter its policy (IP/13/1101).
In its attempt to follow a divide-and-conquer strategy for gambling services,
the Commission has not only faced an unusually restrictive Court. The
committees in the European Parliament too have repeatedly been sceptical
about liberalizing online gambling (2012/2322 (INI)). Consequently, liberal-
ization is slow to proceed.
Another example of this strategy is sports policy. Sports originally appeared
outside the Treaty and the single market, until in 1995 the Court judged
professional football’s transfer system according to the free movement of
workers in Case C-415/93 Bosman. Some other cases based on free movement
and competition law followed (Barani 2005; Rincón 2007), as well as the
issuing of a White Paper by the Commission in 2007 (COM (2007) 391
final). This case law has supported moves towards legislation in this area,
with the parliamentary committee for sports and culture demanding guide-
lines from the Commission ‘on the application of Community competition
and internal market law to sport’.5 Additionally, the Lisbon Treaty gives
further competences for sports policies to the European level. In a communi-
cation (COM (2011) 12 final), the Commission detailed several measures it
intends to pursue, among them the further assessment of the consequences of
the Treaty’s non-discrimination rules—on the grounds of nationality—for
sports (p. 11).

5
Euractiv, ‘MEPs Adopt Report on Future of EU Sport Policy’, 3 Apr. 2008. <http://www.
euractiv.com/section/science-policymaking/news/meps-adopt-report-on-future-of-eu-sport-pol
icy> [accessed 11 May 2017].
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

106 The ECJ and the Policy Process

Regarding the free movement of capital, from 2002 onwards, the Court
issued several rulings against Golden Shares (Höpner and Schäfer 2010:
355–7; Werner 2016). Werner (2013: ch. 5) traces these incremental attacks
by the Commission on Golden Shares in different countries. In its most recent
ruling against the German Volkswagen law in October 2013 (C-95/12), how-
ever, the Court let the special veto rights of Lower Saxony pass. Following this
ruling, the Commission once again has less support in case law to liberalize
corporate governance rules.
These examples, where the Commission strategically combines its judicial
and legislative roles, or turns to its broad administrative powers in competi-
tion law to promote legislation, are not the only cases where case law is
significant for legislation. Preliminary references make up the bulk of the

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

ECJ’s activities and therefore account for much of the development of case
law. This may give incentives for the Commission and the member states to
codify case law in legislation. While I treat both active pressure and passive
linkage (Table 4.2) as analytically distinct, they do overlap empirically. Thus,
while not the result of a grand strategy, in the example of the Services
Directive, to which we now turn, the Commission also took up some judicial
activities in preparation (Schreinermacher 2013: 166).

THE S ERVICES DIRECTIVE

The Services Directive has been characterized as ‘the most salient and
contested piece of legislation passed in the EU in the last decade’ (Lindberg
2008: 1185). It can be seen as a crucial case (Eckstein 1975) for analysis, which
shows that the shadow case law casts is also relevant in important legislative
projects. The Services Directive has attracted significant academic attention
(Lubow 2017). Strikingly, most scholars overlook how case law influenced the
directive but interpret this legislative process as a question of preferences and
power differentials in the EP and in the Council. Of course, this could mean
that I am mistaken. However, my analysis that points out the relevance of case
law supports my claim that existing analyses have thus far ignored the
importance of judicial policymaking. I see this neglect as being less rooted in
the failure of individual scholars, and rather in the treatment of the EU as a
normal political system, an interpretation that became common after Hix’s
seminal early article (Hix 1994). As important as it has been to open the study
of the EU to comparative politics, and to move it away from an explicit
international relations perspective, this glosses over a crucial difference: at
the supranational level, the balance of power is skewed due to a fragmented,
weak legislature and the very detailed constitutionalization of rights. Namely,
there is over-constitutionalization.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Judicial and Legislative Policymaking 107

As was mentioned in Chapter 3, for a long time the Court interpreted the
freedom of services narrowly. It was only in the 2000s that it emphasized the
necessity to facilitate transborder trade in services on the basis of a growing
number of court cases (Hatzopoulos and Do 2006: 923). Against this back-
drop, the Commission launched its proposal for a Services Directive in early
2004. While services had been part of the original 1992 programme, the intra-
EU trade of services was not as significant as their role in domestic economies.
The single-market programme only succeeded in some sector-specific direct-
ives, specifically for financial services, such as insurance. As long as the Court
interpreted the freedom to provide services restrictively, there was little pres-
sure on member states to agree on single-market rules.
Schreinermacher (2013: 166) shows that the Commission had prepared the

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

ground for the Services Directive with some infringement procedures: between
2002 and 2006, it opened sixty-six infringement procedures against member
states in the area of free movement of services, compared to only forty-seven
in the period from 1997 to 2001. In addition, preliminary references increased
from seventy-three to ninety-seven. In particular, France was the target of
thirteen infringement procedures in 1997–2001 and nine procedures in
2002–6. The number of French preliminary references was low in comparison
(four and seven in these two periods respectively). The opposite is the case in
Germany: there were three and seven infringement procedures in the two
periods respectively, but, in comparison, there were nine and sixteen refer-
ences, putting Germany under much more pressure from European law, as
utilized by private actors, than France.

The Content of the Directive

The story of the Services Directive is well known, but let me quickly reca-
pitulate its main points. Since decision-making on sector-specific directives for
services had proven cumbersome, the Commission proposed a horizontal
approach for a Services Directive, which would apply to all services where
specific legislative measures had not yet been taken (COM (2004) 2 final, p. 3).
This unusually broad target covered about 50 per cent of all economic activity,
which led to a process of unprecedented politicization. The directive aims to
realize the freedom of establishment and the freedom of services. Originally,
only lotteries and genuine not-for-profit public services (such as education or
cultural activities) were left out of the directive, but health and social services
were included. In order to achieve a harmonized horizontal approach to
services regulation, the draft directive relied on the principle of home-country
control. Service providers had to abide by the regulations of their home state
and enjoyed the freedom to provide services on this basis. This implied that all
member states had to recognize services that were regulated in other member
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

108 The ECJ and the Policy Process

states as being equivalent to their domestically regulated services. At the same
time, member states had to review their national regulatory requirements and
abolish excessive requirements. To ensure the required level of cooperation
between home- and host-country authorities, the directive introduced a duty
of cooperation so that host-country authorities could obtain information from
home-country authorities concerning the legality of companies and details of
their regulation.
The proposal was a significant radicalization (De Witte 2007: 9–10) of the
incipient changes in the Court’s case law. While the ECJ had been more
generous in its exceptions for the freedom of services than for the freedom
of goods, when balancing the freedom with public interests, the Commission
proposed the country-of-origin principle as a general rule. This would abolish

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

opportunities to impose host-country rules, which the Court still had granted,
and gave the directive a considerable deregulatory potential. Once unions and
NGOs became aware of the consequences, their protests eventually led to the
failure of the constitutional treaty in France and the Netherlands (Howarth
2007: 94; Grossman and Woll 2011: 99–100).
Due to its broad scope, it was not generally possible for public and private
actors to assess all the implications of the directive (Davies 2007a: 241–2). In
relation to the breadth of necessary regulatory change, it is important to keep
in mind the EU’s specific approach for integration, which was discussed
earlier. Although the Treaty and the fundamental freedoms only target
transborder activities and not internal matters, secondary law pursues a
unified approach for domestic and transborder activities. In addition, there
was an overlap with existing services law, in particular the Posted Workers
Directive (96/71/EC) of 1996. The services draft loosened some restrictions on
posted workers in its Article 24, such as the need to carry papers for local
controls in the host country and the obligation to appoint a national repre-
sentative for the posting company, making control by the host country more
difficult. The high number of transposition measures, which were required in
member states once the directive was agreed, testifies to its breadth and to
the difficulty of assessing its effects: Hungary had to adopt 333 measures,
Germany 222, France 84, Austria 52, the UK 15, Italy 10, and Greece 107
(Hatzopoulos 2013: 461). As a side note, these highly diverse numbers sub-
stantiate doubts about whether one can assess the domestic impact of European
integration quantitatively by counting laws (Töller 2010).
Indeed, several analyses have shown (Miklin 2009) that many relevant
political actors were slow to realize the enormous implications of the directive.
The detailed preparation of legislation in the Commission is actually meant to
prevent such politically contentious proposals reaching the public. In this
sense, the Services Directive was an accident of internal policy coordination.
In addition, member states failed to notice the political salience of the issue in
their feedback on various preparatory policy documents. In this ‘accident’,
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Judicial and Legislative Policymaking 109

case law played an important role. As Hartlapp et al. show in their detailed
empirical work, as the directive was being prepared, the DG for the Internal
Market repeatedly pointed out that all it was doing was codifying case law: no
changes to the policy status quo were intended, merely a consolidation of all
the existing requirements for case law and a greater transparency of regulation
(Hartlapp et al. 2014: 104–7). It was particularly the case within the Commis-
sion that the argument for a mere codification of case law kept the concerns of
other DGs at bay. The contrast between this perspective on ‘simple codifica-
tion’ and the political controversies the directive attracted shows how
much the public underestimates the reach of EU law as it is interpreted by
the ECJ, and, possibly, how much the ECJ overestimates the public’s support
for its rulings.

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Negotiating the Services Directive

The political reactions to the draft emphasized the significant redistributive

consequences of the regulatory status quo as shaped by case law. If we consider
that the directive merely codified the case law, this elucidates the ‘inefficiency’
part of my path-dependency argument. The proposal was published in early
2004, a few months before the Eastern enlargement. Consequently, the labour-
intensive nature of many services raised fears of wage competition, most of all
in those countries that rely on collective wage agreements, such as Germany.
Only minimum wages are automatically binding for service providers from
other member states, following the Rush Portuguesa ruling; collective agree-
ments are not.
As Davies points out, home-country rule for services provision is very
challenging, as it violates expectations in relation to two relationships: the
privileges of nationals vis-à-vis their own nation state, an issue we discussed in
Chapter 3 in terms of reverse discrimination against nationals, and the equal
treatment of citizens.
An individual who is present in the jurisdiction but not subject to its regulation,
and operating under a more beneficial regime, is a direct challenge to the content
of citizenship—national or European—and its associated guarantees of equality
and privilege. His domestic competitor sees his most privileged position as a
national citizen undermined, while the two competitors, working side by side,
operate under different legal regimes with different rights, despite a shared EU
citizenship. (Davies 2007b: 7)
After all the demonstrations and protests that the ‘Bolkestein directive’ gave
rise to, which undermined the legitimacy of the EU project, the directive was
extensively modified. In this process, the EP and its rapporteur Evelyne
Gebhardt took on a leadership role. The political constellation of the Services
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

110 The ECJ and the Policy Process

Directive was unusual in several respects. Member states were divided. One
coalition was mainly comprised of old member states, the other of the new
members, though including both the UK and the Netherlands. But despite these
divisions, the Council was united in supporting the idea that a Services Directive
was needed for economic growth as there were more and more hindrances to
trade in services.6 Moreover, while the Commission and the EP are normally
natural allies, here they were opposed to one another. This extraordinary
constellation among the EU institutions gave the EP an unprecedented promin-
ence. The divided Council could not agree unanimously on a revised proposal. If
the EP could achieve the necessary unity to agree on a revision with the support
of a broad majority, the Commission and the Council would be bound to go
along with it and conclude this contentious but important piece of legislation.7

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

While the Commission, in its discussions with the Council, repeatedly
referred to the constraints of case law,8 the rapporteur in the EP largely denied
them. It is fair to say that case-law restrictions were scarcely reflected in the
EP’s discussions. The socialists had invited the German Advocate General
(AG) Kokott to explain the legal dimension of the directive, which Gebhardt
considered to be rather loose (interview MEP PES 1 December 2010; interview
German Trade Union Confederation 20 November 2009). The Parliament,
as the conflict around the services directive clearly shows, ignored over-
constitutionalization. In the course of negotiations, Gebhardt argued for
replacing the principle of home-country control with the principle of mutual
recognition. Although many did not perceive a difference (interview German
Trade Union Confederation 20 November 2009), the reference to ‘recognition’
seemed to imply a broadening of rights for the host country and to put the
burden of proof of equivalent regulation on the private party offering the
product, not on the member state (interview French Financial Ministry
25 November 2009). As was widely promoted by Gebhardt, the EP wanted
to see the rights of the host country strengthened, but settling for host-
country control would have thwarted the building of a services market and
would have been next to meaningless given the case law of the Court and
its overriding constitutional status.
In the end, the central Article 16, which originally included home-county
rule, simply discusses the obligation to enable freedom of services. By includ-
ing a list of measures that host states are not allowed to impose, such as special
duties on registering in the host country or ex ante certification, as well as
prescriptions for materials and tools used, Article 16 indirectly establishes

6
Interviews: Council General Secretariat 27 Oct. 2010; German Federal Ministry of Economics
19 Nov. 2009; German Trade Union Confederation 20 Nov. 2009.
7
Interviews: German Federal Ministry of Economics 19 Nov. 2009; MEP PES 28 Oct. 2010.
8
Interviews: French Financial Ministry 25 Nov. 2009; Council General Secretariat
27 Oct. 2010.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Judicial and Legislative Policymaking 111

home-country regulation (Nicolaïdis and Schmidt 2007). It should be addi-
tionally noted that the list of justifications in Article 16(3)—that host countries
may use to achieve ‘overriding reasons relating to the public interest’—is
much narrower than ECJ case law. Lawyers have considered this to be a
‘deregulatory shift’, which was brought about by the directive (Davies 2007b:
12, 18; De Witte 2007: 12).9 Other legal commentators, however, considered
the directive as falling behind the case law of the ECJ (Editorial Comments
2006: 309), since the Court had clearly established the country-of-origin
principle. This raised the interesting question of whether the Court would
take into account the preferences the legislature had signalled and adapt its
interpretations in the future.
The change in language, if not in substance, allowed agreement to be

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

reached on the directive, which was desired by many but was highly politi-
cized. Significantly, the European Trade Union Confederation (ETUC)
backed the compromise in terms of the interests of its Eastern members,
easing member states’ fears of further protests (interview German Trade
Union Confederation 20 November 2009). The broad mobilization of unions
had helped the Socialists to abolish all references to the contentious home-
country principle in the directive (interview MEP PES 28 October 2010), but
this was understood by many to be only a symbolic change.10 The com-
promise on Directive 2006/123/EC of 12 December 2006 between the
Socialists and Christian Democrats in the EP was also made possible by
the more restricted scope of the directive, which exempted health services,
utilities, public transport, social and security services, temporary workers,
gambling and lotteries, waste, audiovisual services, electronic communica-
tion, and financial and legal services. However, again, it is crucial to consider
what this exemption means when one examines not only the legislative
process but also the judicial route. Those who asked for the exemptions
assumed that they would keep these policies under national purview. In fact,
narrowing down the directive only meant that these areas were subject to
the Treaty rules, turning the exemption into an easy concession for the
supporters of liberalization (interview German Trade Union Confederation
20 November 2009). Much to the displeasure of the Eastern European

9
For instance, consumer protection is not part of the exceptions in Article 16(3), which only
refer to ‘public policy, public security, public health or the protection of the environment’.
Consumer protection is mentioned, however, in Article 3(2) where it is considered to fall under
host-country rules. In its case law, the Court hardly ever recognized it as an exception, and those
actors that wanted the directive felt that the liberalization of services would not succeed if
member states could refer to consumer protection as an exemption (interview MEP EPP 27 Oct.
2010). See also Recital 40.
10
Interviews: French Financial Ministry 25 Nov. 2009; Council General Secretariat 27 Oct.
2010; Federal Ministry of Economics 19 Nov. 2009; MEP PES 28 Oct. 2010; German Trade
Union Confederation 20 Nov. 2009; MEDEF 11 Dec. 2009.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

112 The ECJ and the Policy Process

member states, the facilitation of the Posted Workers Directive was deleted.
As it had done for health services (see next chapter), the Commission
promised a separate follow-up. The fact that the posting of workers enforce-
ment Directive (2014/67/EU) was only adopted in May 2014 demonstrates
that this issue continued to be politically contentious.
The legislative process of the Services Directive is generally hailed as a
significant success for the EP (Loder 2011), and for social movements and
unions in general (Leiren and Parks 2014).11 The political discussions, as is
also the case for academic analyses, have largely overlooked the shadow that
case law casts.
The Court proved these assumptions to be wrong. The Laval and Viking
judgments of December 2007 showed shortly afterwards that leaving the

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Posted Workers Directive out of the Services Directive did not alter its
subjection to the freedom of services. The Court presented itself as being
unimpressed by the heightened political contentiousness of the Services
Directive and seriously interfered with the prerogatives of societal self-
regulation (Joerges and Rödl 2009; Höpner and Schäfer 2010: 354). It clearly
sided with the observations of the new member states and the UK in favour
of the freedom of services and against the right to industrial action
(Lindstrom 2010: 1316, 1320), thereby arguing for the horizontal effect of
Article 43 of the TEC.12
Other provisions of the proposal survived the political debate more or less
untouched. By including far-reaching provisions on administrative cooper-
ation (Chapter VI; Articles 28–36), the Services Directive lays the foundation
for administrations to operate transnationally. But the difficulty of breaking
the national hierarchy of command has been little discussed. While the
Minister has the ultimate political responsibility (Döhler 2001), administra-
tions are obliged to comply with horizontal demands that originate in other
member states’ administrations. This is not our focus here, but it is important
to recognize that these far-reaching changes are a requirement that stems from
the case-law principle of non-restriction. As trade in services often relies on
the presence of the service provider in the host state, its regulation in the home
state necessitates far-reaching administrative exchanges.

11
See Lindberg (2008) for a detailed analysis of inner-parliamentary decision-making that
makes no reference to the constraints of case law, however.
12
Lindstrom sees the Services Directive as an example of the Council and the EP scaling back
the market, and concludes her article by arguing that the Laval and Viking cases show how much
secondary law is needed (Lindstrom 2010: 1322–3). Such an argument can only be made under
the assumption that secondary law could constrain the Treaty, disregarding its constitutional
status.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Judicial and Legislative Policymaking 113

What we can Learn from the Directive

To sum up, in the negotiations actors differed in their awareness of the

constraints of case law embedded in the Commission’s proposal for codifica-
tion. As Stone Sweet and Mathews (2008) argue, proportionality analysis
subjects parliamentary sovereignty to judicial review. Since a general frame-
work cannot do justice to the balancing the Court performs in individual
cases, the proposal radicalized the case law, giving more scope to the freedoms
to provide services and of establishment, and less scope to member states’
regulatory interests. The dominant political discourse that the EP established
largely ignored the case law, although more liberal supporters in the Europäische
Volkspartei (EVP) were clearly aware of its constraints. Nevertheless, the

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

position of negligence advanced by Gebhardt, who argued that there are many
legal interpretations but that these should not overshadow the political process,
also had its merits. After all, the Commission instrumentalizes case law to a large
extent strategically, emphasizing certain interpretations over others. By ignoring
over-constitutionalization, inadvertently or not, the EP can be seen to be revert-
ing to an extreme form of signalling: showing the Court that its policy prefer-
ences are uninhibited by the policy constraints of the Treaty. The extent to which
political actors ignore over-constitutionalization is also apparent in political
calls to exclude policies from the directive, which totally misjudges that this
would mean that the Treaty applies directly. What could have been hoped for—
that the Court would note the lack of political and societal support for an
extensive interpretation of individual rights that followed from the fundamental
freedoms—did not materialize. The Court simply continued to push through
its case law; this was true for the patient mobility case law (see Chapter 5) as well
as for Laval and Viking. However, the case law on services is still difficult to
predict. Thus, for a ruling that concerned German regulatory restrictions on
pharmacies (Doc Morris; C 171–172/07; 19 May 2009), most commentators
would not have foreseen that the Court would have come to such a restrictive
interpretation of the freedom of services.
Additionally, the Commission seemingly remained unaffected by the polit-
ical sensitivities that emerged during the legislative process. This becomes
evident when one considers the Commission’s Handbook on Implementation
of the Services Directive that was published in 2007 in all the Community’s
languages (European Commission 2007). It gives member states guidance on
implementing the directive. Interestingly, the Commission interprets the cen-
tral Article 16 (pp. 36–40) with extensive reference to the existing case law on
the freedom of services. In these few pages alone, it cites the case law of the ECJ
twenty-one times, referring to specific rulings in the footnotes sixty-eight times.

It should be noted that the terms ‘public policy’, ‘public security’ and ‘public
health’ are concepts of Community law which stem directly from article 46 of the
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

114 The ECJ and the Policy Process

EC Treaty [Treaty Establishing the European Community]. These concepts have
been consistently interpreted by the ECJ in a narrow sense, meaning that there
must be a genuine and serious threat to a fundamental interest of society and it is
for the Member State invoking these public interest objectives to demonstrate the
risks involved. (European Commission 2007: 37)
While secondary law cannot override primary law, it appears that the Com-
mission is not easily troubled by the political process, as long as it has the
constitutional nature of case law to support its interpretation. This is the
baseline from which the legislative process cannot divert. In the course of
implementation, the Commission has begun several infringement proce-
dures to pressure member states into compliance. In the summer of 2010, it
approached twelve member states, as a result of their insufficient implementation

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

of the directive, with a reasoned opinion (Belgium, Germany, France, Greece,
Ireland, Luxembourg, Austria, Poland, Romania, Slovenia, the UK, and Cyprus,
later followed by Sweden; IP/10/821). The cases against Germany, Greece, and
Austria were referred to the Court in October 2011 (IP/11/1283) but closed in
May 2012 (IP/12/534). After all these original cases were closed, in the summer of
2013, a new case against Hungary was sent to the Court, which involved
luncheon vouchers that had been introduced in 2012 (IP/13/578).
In summary, the Services Directive was unusual in the degree to which it was
politicized. Interestingly, political contention left the Court and the Commis-
sion unswayed in their supranational, hierarchical mode of decision-making,
despite the higher legitimacy of the joint-decision mode that involves the
legislature (Scharpf 2000b, 2006: 851). It is striking that the literature in the
area of political science on the Services Directive—a widely analysed piece of
legislation—largely ignores the power of judicial constraints. Following the
activism of the EP, most research has equally overstated the political nature and
understated the judicial nature of the Services Directive. The legislature is taken
to be sovereign, in parallel to the situation at the national level. The different
nature of the balance of powers in the EU has not been understood. Here, the
legislature does not have many options at its disposal, as the EU suffers from
too many constitutional constraints.

THE REGULATION FO R GOODS

The regulation on mutual recognition of goods (764/2008)13 is much less well

known than the Services Directive, although it is similar in its thrust. It was
decided after the Services Directive. The lack of controversy that characterized

13
This part of the chapter draws on Schmidt (2011b).
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Judicial and Legislative Policymaking 115

the decision-making process for the regulation of goods indirectly supports
my interpretation of the path dependency of case law. It shows that the Cassis
logic of regulation meets with relative acceptance in terms of goods, but once it
is transferred to services or other freedoms, it becomes controversial because
of the redistributive issues it raises.
The Commission announced a new package for the internal goods market
in February 2007 (COM (2007) 35 final), which consisted of four different
measures, among them the regulation on mutual recognition of goods.14 It is
now called a ‘new legislative framework’.15 Similar to the Services Directive,
the regulation builds directly on the case law of the ECJ and specifies the
obligations of member states under mutual recognition. Like the Services
Directive, it codifies the case law and introduces single contact points to

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

facilitate the marketing of products for EU suppliers. However, given the
extent of harmonization in this area, it only applies to the approximately
20 per cent share of the goods market that is non-harmonized. Additional
differences concern the long-standing development of case law for goods—
Commission documents typically mention some 300 cases here (2003/C 265/
02, p. 18)—and the extent of standardization of goods that facilitates high
amounts of trade. Services are not standardized to this extent. Although the
regulation is described in interviews as an extremely complex piece of legis-
lation that is not sufficiently understood by many,16 with its horizontal, all-
inclusive approach, the whole internal market package passed quickly, as it
was proposed in February 2007 and agreed upon in June 2008. The British
House of Commons, for instance, complained about the insufficient time they
had to consider the proposal thoroughly.17 Only Italy abstained from voting.18
This time, the Commission and EP were on the same side (interview Com-
mission 24 June 2009). One parliamentary reading sufficed in February 2008
(Council 6 611/08, 27 February 2008, pp. 1–2). It was particularly the German
Presidency, in the first half of 2007 that pushed the directive through.

14
‘A proposal for a Regulation of the European Parliament and of the Council laying down
procedures relating to the application of certain national technical rules to products lawfully
marketed in another Member State and repealing Decision 3052/95/EC.’ (Decision 3052/95/EC
had established a procedure for the exchange of information in cases where member states
prohibited the marketing of products under mutual recognition. However, while initially mem-
ber states gave notice of measures taken, after the year 2000 the Commission was hardly ever
contacted.)
15
<http://ec.europa.eu/growth/single-market/goods/new-legislative-framework/index_en.htm>
[accessed 11 May 2017].
16
Interviews: Permanent Representation Germany 22 June 2009; Commission 24 June 2009.
17
See House of Commons, Select Committee on European Scrutiny Twenty-Second Report;
under: <http://www.publications.parliament.uk/pa/cm200607/cmselect/cmeuleg/41-xxii/4106.htm>
[accessed 11 May 2017].
18
Conseil 11075/08, 23 June 2008. This was due to a question concerning precious metals,
which had been discussed a few times in the Working Group (Council 16213/07, 10 Dec. 2007,
p. 2; 5317/08, 18 Jan. 2008, p. 18).
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

116 The ECJ and the Policy Process

The regulation codifies existing case law. Like the services directive, it even
goes beyond case law in that member states sacrifice additional national
regulatory autonomy. Again, the case-specific balancing the Court performs
between the freedom of goods and member states’ regulatory interests could not
be generalized. Its outcome is uncertain. If member states want to re-establish
legal certainty with codification, they cannot generalize the exceptions granted
from balancing, but are forced to go beyond the rulings.
Since the Cassis ruling, case law has been crucial for the internal market for
goods. Following this judgment, the Commission issued a ‘Communication
from the Commission concerning the consequences of the judgment given by
the Court of Justice on 20 February 1979 in case 120/78 (“Cassis de Dijon”)’
(OJ No. C 256/2, 3 October 1980), which laid down its interpretation of the

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

ruling. This was the first time that the Commission took up case law to build
policy, thereby radicalizing its implications (Alter and Meunier-Aitsahalia
1994). The Commission simply assumed that the member states had a clear
one-sided duty to recognize the legal orders of each other by stating that ‘Any
product lawfully produced and marketed in one Member State must, in
principle, be admitted to the market of any other Member State.’ The possible
exceptions from the free movement of goods included in both Article 30 of the
TEC and the mandatory requirements, which the Court had introduced, were
simply neglected. Already in this first instance, we find the problem that
the Court can balance between a freedom and the public interest but that
these exceptions are case-specific and hard to generalize. However, member
states did not share the Commission’s view, and their subsequent practice did
not match the Commission’s hopes for a far-reaching mutual recognition of
goods regulations (Pelkmans 2007).
The history of the single market for goods illustrates well the difficulty of
relying on case law for regulation. Case law cannot define an issue of policy in
a conclusive way. Focused as it is on the case at hand, case law has unclear
implications for similar, but slightly different, additional cases. This suggests
that there is less legal certainty than when secondary law explicitly describes
regulatory intent. As there are only EU case law and national product speci-
fications to refer to, companies that market their products in other member
states face problems when claiming their rights only based on case law. In the
late 1990s, the Commission started to require that member states add mutual
recognition clauses to their national product regulations (COM (2002) 419
final, p. 26). This was meant to serve as a legal reference point for companies
from other member states when they demanded market access for equivalent
products. The background to this was the ECJ’s Foie Gras case (C-184/96,
22 October 1998), where the Commission insisted that foie gras that was
legally marketed in other member states could be sold in France. The Court
agreed in this case with the Commission’s argument that member states were
required to insert a mutual recognition clause into their technical regulations
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Judicial and Legislative Policymaking 117

(in para. 28) (Craig and De Búrca 2007: 716–17). However, a few years later in
Case C-24/00, which dealt with the authorization of a food additive, and in Case
C-95/01 Greenham and Abel, the Court did not uphold the Commission’s
argument that Article 28 of the TEC required the inclusion of this clause. In
both cases, AG Mischo had similarly argued against this obligation (Bartels 2005:
693–6). This shows the limited guidance of case law for regulatory requirements,
an experience that motivates the Commission to introduce legislation.

The Content of the Regulation

What is the content of the regulation? It is notable that companies have clear

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

requirements (Article 4 of Regulation 764/2008) to provide ‘relevant informa-
tion concerning the characteristics of the product or type of product in
question’ as well as ‘relevant and readily available information on the lawful
marketing of the product in another Member State’. However, this does not
need to be in the language of the country of destination (Article 4b). The
central Article 6 delineates the burden of proof concerning the equivalence of
regulation. Mutual recognition presupposes equivalent regulation. The ques-
tion addressed under Article 6 is who bears the burden of proof whenever
member states doubt the presence of equivalent regulation and therefore want
to deny market access. The regulation thereby clarifies a point where the case
law had left the highest legal uncertainty, according to the Commission (COM
(2012) 292 p. 5). The regulation solves the matter with a reversal of the burden
of proof, which now falls exclusively on member states. If member states wish
to invoke an exception and to prevent the marketing of a product, they have to
specify ‘the technical rule on which the decision is to be based and [set] out
technical or scientific evidence to the effect that . . . (a) the intended decision is
justified’ and ‘appropriate’ (Article 6).
Companies are allowed to immediately put their products on the market. If
authorities intend to object to the marketing, they face clear deadlines in the
regulation: the supplier has twenty working days to respond to the objections
of the authority, which then has another twenty days to justify its final decision
and communicate it to the economic operator and the Commission (Article 6).
The EP and the Council included the alteration that authorities may bar a
product only in the case of danger. Altogether, authorities are confronted with
a potentially complex situation as products regulated in the other twenty-
seven member states may enter each national market. Underlying harmoni-
zation is only provided by the General Product Safety Directive (2001/95/EC
and 92/59/EEC) that requires producers to market only safe products (Inter-
pretative communication, 2003/C 265/02, p. 5). In addition to the information
in various languages provided by the companies, authorities can inform
themselves at member states’ national contact points, which are established
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

118 The ECJ and the Policy Process

by the regulation. In Recital 30, the regulation encourages member states to
provide information in different languages.
Whenever national authorities object to foreign products, they bear the
burden of proof and have to demonstrate the proportionality of their rules
(Recital 23 and Article 6b). This was closely discussed in the Council, and
member states made sure to state explicitly that they do not have to justify their
national regulations as such (Council 9610/07, limite, 15 May 2007, pp. 2, 7).
Another issue in the working group was the question of ‘technical or scientific
evidence’ in Article 6, as member states were concerned about the level of
scientific support they would need when rejecting a product (Council 5673/08,
25 January 2008, p. 22).
The EP had included several modifications to the regulation (interview

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

General Secretariat Council 25 June 2009). It requested that the effect on
free movement should be specified more precisely than in the proposal by
replacing ‘affecting’ free movement with ‘hindering’ it (amendment 33, Article 1,
p. 24). Supported by the efforts of the German and Portuguese presidencies,
the Parliament also included the requirement for the operator to provide
information on the products (Article 4) and the possibility of temporarily
suspending the marketing of a product (now Article 7) (see Council 5312/08,
1 February 2008, p. 2). But in stark contrast to the discussions around the
Services Directive, the EP did not focus on dangers of insufficient regulation.
The liberal Alexander Stubb, who shortly afterwards became the foreign
minister of Finland, acted as rapporteur. In its report (A6-0489/2007), the
EP emphasized the need to enhance legal clarity and its assumption of
equivalent regulations in the different member states: ‘This approach (mutual
recognition) is perfectly understandable because it can be assumed that the
public interest, e. g. protection of health or of the environment, is safeguarded
in a similar way in the different Member States’ (p. 50).

The Burden of Proof in the Case Law

The regulation, I have argued, improves legal certainty for companies and
authorities by one-sidedly putting the burden of proof on member states. This
burden of proof is crucial, as it manifests the underlying assumptions of the
regime. If the assumption is that the member states still possess the sovereign
rights to regulate their markets, companies have to show that they meet
‘equivalent’ regulatory requirements. If the assumption is that the European
free market prevails, member states have to justify the basis of their regulatory
goals. It is surprising that the member states have scarcely discussed this
relevant issue of the burden of proof. Why would they consent to such a
restriction of their regulatory autonomy? To answer this question, I will now
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Judicial and Legislative Policymaking 119

outline how the Commission, the member states, and the ECJ perceived the
burden of proof.
The Commission never discussed a reversal of the burden of proof but
simply a clarification on an existing legal uncertainty (e.g. SEC (2007) 112,
p. 18). This is because it had already understood what the onus on member
states was, as it described in its communication after the Cassis judgment
(interview Commission 24 June 2009). Regulation 764/2008 finally codified
this position. In its 1999 communication on mutual recognition, the Com-
mission had stated that
The protection of health cannot be used without a valid reason, for this would
lead to borders being reintroduced within the Single Market. Therefore, it is for
the national authorities to prove that their regulations are necessary in order to

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

truly protect the interests cited in Article 30 (ex Article 36), and in particular that
placing the product in question would pose a risk to public health.
(SEC (1999) 1106, p. 10)
Moreover, it is interesting that while the Commission cites the case law of the
ECJ on many occasions for other aspects, it does not do so for the burden of
proof. It just mentions the general lack of legal certainty, given that the mutual
recognition regime is structured around more than 300 rulings on the freedom
of goods. It also mentions that the case law does not give the Commission any
means to require the member states to take up specific measures (SEC (2007)
112, pp. 18, 20).
The question of the burden of proof is hard to solve with current legal tools. . . . It
should be noted in this context that the case-law of the Court of Justice does not
necessarily allow the Commission to require much more from Member States
than a mere reference to Articles 28 and 30 EC Treaty in their national rules: the
Court has indicated that Member States enjoy a margin of discretion in deter-
mining what measures are most appropriate to eliminate barriers to the imports
of products. (SEC (2007) 112, pp. 19–20)
The Council, in contrast, clearly stated that the regulation brings about a
change to the burden of proof: ‘It also transfers the “burden of proof ” from the
economic operator to the administration, thus making it more difficult for a
member state to deny the marketing of products that are already accepted
in another member state.’19 In the same way, the German Parliament was
informed about this matter (see Baddenhausen 2007: 2). Stubb, as the EP’s
rapporteur, also spoke of such a reversal (A6-0489/2007).
The case law includes divergent statements on the burden of proof. The beer
purity case (178/84) of 1987 states that ‘it is for those authorities to demonstrate

19
Luxembourg 23 June 2008, 11062/08 (Presse 186): ‘Council approves rules to improve free
movement of goods in the EU’, p. 2.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

120 The ECJ and the Policy Process

that the prohibition is justified on grounds relating to the protection of the
health of its population’ (summary, No. 4; compare also grounds No. 46).
Similarly, the Court had argued before in the Case C-251/78 Denkavit
(8 November 1979, Rec. 24) that the national authority referring to the
restrictions of Article 30 of the TEC had to bear the burden of proof
(Dammann 2007: 100). These rulings support the position of the new regula-
tion. But the Court did not always burden the member states consistently. In
the famous case on woodworking machines (C-188/84), the Court denied the
claim that member states have to approve machines that offer less safety for
users, which seemed to imply that the Commission had the burden of proof
(grounds 17–22; Dammann 2007: 101).
Legal theories of burden of proof support the idea that there is not a one-

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

sided obligation (Lenaerts et al. 2006: 24-073; Dammann 2007: 116). As AG
Tesauro argued in Case C-362/95, the party that takes a certain position needs
to provide the proof. Also, the Court stated in Case C-127/92 (No. 13) that ‘It
is normally for the person alleging facts in support of a claim to adduce proof
of such facts.’ On this basis, the producer using Article 28 of the TEC for
marketing its products would have to bear the burden of proof concerning the
equivalence of regulation.20 In addition, the member states that draw on
Article 30 of the TEC or on mandatory requirements would have to bring
proof for the need to bar the product.

Explaining Member States’ Consent

In internal market decisions, member states face a trade-off between retaining

regulatory authority and ensuring easier market access for their companies.
The benefits of regulatory authority are counterbalanced with free-trade
benefits. The complicated mutual recognition regime for goods did not lead
to a level playing field. Companies that are denied market access could
complain to the Commission, address the Courts, or approach the SOLVIT
network,21 but they cannot refer to concrete legal positions. It was particularly
small- and medium-sized enterprises (SMEs) that were disadvantaged, as they
could not pay lawyers as easily as large companies or adapt their products to
national specifications. Consequently, it is not surprising that Germany, which
is strong in exports and SMEs, pushed for the regulation during its presidency.
Moreover, the case-law-based regime led to further inequalities as member states
implemented case law unevenly, often following an approach of ‘contained

20
I would like to thank Rike Krämer for pointing this out to me.
21
SOLVIT is a network of national officials, coordinated by the Commission, which was
established in 2002 with the aim of providing simple, informal problem-solving in issues of
mutual recognition that are below the level of a formal infringement procedure.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Judicial and Legislative Policymaking 121

compliance’ (Conant 2002) where the relevance of case law was assumed to be
only inter partes (interviews Federal Ministry of Economics 5 and 19 November
2009). The Commission explained the complicated case-law regime in numerous
communications. Different Council and Parliament decisions aimed to enhance
compliance.
It is important to understand that, in light of the situation-specific nature of
case law, a less radical solution, which results in further responsibilities for
companies, might not have achieved the goal of enhanced legal certainty.
Legislation had to extract a general rule out of partly contradictory case law.
In order to ensure that the Commission and the Court would have no reason
to intervene any further, it was necessary to regress to the lowest permissible
level of national competence (Schmidt 2011b). Similar to the Services Directive,

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

not all participants in the negotiations understood the implications of the
complex regulation.22 Even after the agreement, some member states assumed
that they still retained regulatory control.23
In the negotiations, major discussions also revolved around the question of
prior authorization procedures, which the regulation does not abolish. These
particularly affect precious metals. The UK controlled jewellery before releas-
ing it on the market, and this practice had been passed by the Court (C-293/93).
Italy, as the largest exporter of jewellery, wanted to abolish these procedures
through the regulation. As Italy was unsuccessful, it voted against it.24 The
Commission stated, in its first report to follow the regulation, that most
notifications have come up in this area: of the 1,524 notifications between
May 2009 and December 2011, 99 per cent concern precious metals and 1,378
stem from one country (COM (2012) 292 fin, p. 8).
Another contentious area is food supplements. The Commission has issued
a reasoned opinion on the basis of the regulation against Poland from 2013,25
which had put the burden of proof on economic operators, even though food
supplements were already marketed in other member states. It is most likely
that this question would have been significant in the negotiations had the
Commission not begun infringements procedures in a divide-and-conquer
fashion. As Schreinermacher (2013: 215) has shown, compared to the period
from 1999 to 2003, during which the Commission launched eighteen infringe-
ment procedures against member states that dealt with the free movement of
goods, in the period from 2004 to 2008, these went up to thirty-four. This
compares to 109 and 110 preliminary procedures in these two periods
respectively. Following the ruling in Case C-24/00 Commission v France,

22
Interviews: Permanent Representation Germany 22 June 2009; Commission 24 June 2009.
23
Interviews: Permanent Representation UK 22 June 2009; Commission 24 June 2009.
24
Interviews: Commission 24 June 2009; General Secretariat Council 25 June 2009.
25
21 Feb. 2013. <http://europa.eu/rapid/press-release_MEMO-13-122_en.htm> [accessed
10 May 2017].
OUP CORRECTED PROOF – FINAL, 6/12/2017, SPi

122 The ECJ and the Policy Process

which dealt with a prohibition on selling the energy drink Red Bull in France,
France simplified its procedures (Décret no 2006-352, 20 March 2006), lead-
ing to requests for the commercialization of 8,000 to 9,000 new products
(Schreinermacher 2013: 229). Another relevant case is C-192/01 Commission
v Denmark; Denmark had barred food additives in cases where there was no
nutritional demand among the population for such additives. The Court
required Danish authorities to prove that these products posed a danger to
health (Schreinermacher 2013: 190).

SUMMARIZING THE ARGUMENT

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

The path of an interpretation based on non-restriction began with the
interpretation of the free movement of goods. Many years of a regulatory
regime based on case law followed, at least for those 20 per cent of goods for
which there was no specialized secondary law. The disadvantages of the
regime were the degree of legal uncertainty in terms of the rights and obliga-
tions of private and public actors. This resulted in an uneven take-up of the
rights, as large companies were better placed to deal with legal uncertainty, as
well as the uneven implementation of case-law requirements across member
states, which harmed the idea of a level playing field. The lack of political
salience this area attracted, compared to the previous Services Directive, shows
that trade in goods governed by mutual recognition does not raise as many
issues for redistribution.
For the question of how the negative consequences of legal uncertainty
influence actors’ bargaining positions, I argue that legislating in the shadow of
the Court will advance extreme positions (namely the most liberal or those
that advocate the most regulation, in the rare cases where the Court interprets
the Treaty narrowly). The individual balancing the Court performs cannot
be generalized. Secondary law does not overrule primary law. At most, the
legislature can hope to signal to the Court how it wishes to understand the
Treaty’s implications. The Court, and particularly the Commission, interpret
secondary law in the light of their understanding of the Treaty. If secondary
law is kept behind the case law, this poses the danger that it will not be
interpreted as it was intended and nothing will be gained in terms of legal
certainty. Moreover, incentives for litigation persist.
That the Services Directive and the regulation on mutual recognition of
goods go beyond case law testifies that legal certainty can only be established
when one takes the most extreme positions of the ECJ into account. The
examples that I have analysed further liberalization. The cases for car emis-
sions and the Posted Workers Directive from 1996, which were mentioned
previously, are examples of when the exception granted by the Court furthers a
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Judicial and Legislative Policymaking 123

reregulatory logic in the Council. For member states, the legal certainty of
codification strengthens equality before the law among private actors and
ensures a more level playing field among member states. Member states are
likely to implement ECJ case law unevenly. Some will implement it in a rule-
of-law fashion. Others will deny its erga omnes effect. It is not surprising that
Germany made a considerable push for the regulation on the mutual recog-
nition of goods. Member states that export fewer goods would profit from legal
uncertainty, as individuals forgo some of their rights under such a condition.
In the next chapter, I will discuss the Citizenship Directive and the Patient
Mobility Directive. These show that if member states’ policy interests are
markedly opposed to the direction of case law, they may favour unclear
codification in order to constrain its impact.

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

CONCLUSION: IS THE JUDICIARY
OVERSHADOWING THE LEGISLATURE?

While the European Union can be analysed in the same way as other political
systems, it has specific features that are not present at the national level.
Notable among these is the great amount of material detail that the constitu-
tionalized treaty contains, which implies that judicial policymaking partly
becomes an alternative to legislative policymaking, as it needs to be included
in the analyses of the latter as the default condition.
Regulation based on case law has certain constraints. The example of
mutual-recognition clauses most clearly shows that where case law in one
instance seems to provide the basis for the requirement of certain regulations,
in another instance this basis can suddenly be withdrawn. By taking the
example of the regulation on mutual recognition, I have argued that secondary
law that seeks to ensure legal certainty has to radicalize its meaning. Where
the Court takes member states’ regulatory objectives into account with its
proportionality principle, the Commission cannot translate this balancing into
general rules, as national regulatory goals are only justified under very specific
conditions.
In addition to the policy preferences of member states, their position on legal
uncertainty has to be taken into account. Legal uncertainty hampers long-term
private or public investment. But even if legal uncertainty is beneficial to
states, as individuals do not exercise their rights to a great extent, there is an
incentive to re-establish legal certainty. This strengthens equality before the
law. Codification creates a level playing field among member states, and
member states can hope to influence the Court’s interpretation of the Treaty
through secondary law (the ‘signalling effect’). But once case law is established,
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

124 The ECJ and the Policy Process

and if preferences to re-establish legal certainty are high, the legislature has to
be oriented towards the most far-reaching interpretation of the ambiguous
case law.
There are fundamental differences between regulating by legislation and
regulating through case law. The former establishes general guidelines, while
the latter draws conclusions from general principles for a specific dispute. In
the national context, if the legislature wants to act, it can do so. In the EU,
however, the constitutional nature of case law constrains the legislature. Its
distinctive mode of balancing safeguards the interests of the general public in a
single case, but this balanced outcome cannot be transformed into general
rules. It is an individual assessment. The relationship between case law and
codification differs in nature between the national and the supranational level,

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

given the EU’s over-constitutionalization.
By taking the two examples of the Services Directive and the regulation for
goods into account, what do we learn about legislative actors’ awareness of the
constraints of case law? The Commission is known to combine its role as a
guardian of the Treaty with a formal monopoly for agenda setting (Schmidt
2000). The Services Directive and the regulation for goods show how the
Commission bases legislative proposals on earlier infringement procedures,
using these for potential strategic preparation for a divide-and-conquer
approach. But what about the EP and the Council? Both have more legitimacy
than the Court to determine the course of integration. It is, therefore, not clear
why they should accept that the Court has more ‘sovereignty’ than themselves.
The EP, as was made apparent in the Services Directive, partly ignores the
extent to which case law predetermines legislative outcomes.
Member states appear more cognizant of the impact of case law, but they
differ in their national legal systems, traditions of judicial review, and relative
lengths of EU membership. Thus, for the UK, which has a common-law
system and a polity based on parliamentary sovereignty, the idea that case
law can be codified is strange, as is the acceptance of case-law constraints
on the legislature (interview Permanent Representation UK 22 June 2009).
Germany, in contrast, which has a strong tradition of judicial review carried
out by the Constitutional Court, is much more likely to accept such con-
straints, as well as the idea of codification (interviews Permanent Repre-
sentation Germany 22 June 2009; German Federal Ministry of Economics
19 November 2009).
However, it is not necessary for all participating actors to be aware of the
impact of case law through the default condition of legislation. The Services
Directive shows that a highly politicized legislative process will nevertheless
end up being very close to the case law. How the Commission acts is crucial
here, as it drafts the proposals and interprets the secondary law afterwards, by
initiating infringement proceedings in cases of non-compliance. The Com-
mission’s legal services and the Council are deeply involved in the drafting of
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Judicial and Legislative Policymaking 125

legislation, as they are able to point out the limits that case law imposes.
Although, according to their understanding of themselves and their legiti-
mation, the Council and the EP are the ones who should decide on the
framework that the Court then interprets, the lawyers in the Council and
Commission’s legal services will point out when proposals ‘do not fly’ (inter-
view Commission 24 June 2009), thereby steering the legislative process in a
direction that conforms to case law.
It should finally be mentioned that there is a discrepancy in political
salience between these two cases. Hardly any political attention was paid to
the regulation of goods. The country-of-origin principle has very different
implications for services than it does for goods. It is politically much less
contentious when goods that are regulated differently have to compete on the

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

same market, since regulations are decided upon politically and implemented
in the respective home countries. Goods travel by themselves; they are in
direct price competition, but manufacturers do not work side by side. Differ-
ences in production are hidden. Service providers, in contrast, may work side
by side in a single location, although they are regulated very differently. This
raises significant issues concerning equal treatment among EU citizens. As a
result, governments can no longer guarantee equal treatment to their own
citizens and are forced to allow reverse discrimination against nationals if they
want to maintain stricter domestic regulations than those of other member
states. Arising from these injustices, lawyers have discussed whether or not EU
law prohibits reverse discrimination against nationals. In addition, the difficulty
of controlling home-country rules in service provision is exacerbated further, as
parts of this control can only be exercised in the host country, where the home-
country administration has no jurisdiction. This, in turn, also explains the
frequent rule violations in trade in services (Mankowski 2004: 388–9). Finally,
posted workers often offer their services at a wage that can only allow them to
subsist in their home country, but not in the host country (Streeck 2000). Thus,
home-country regulation, and an approach to the interpretation of the funda-
mental freedoms based on non-restriction, clearly raises significant redistributive
issues when it is transferred from goods to services. This shows the inefficiencies
of the path-dependent development of case law.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Reaching Beyond the Market into

State Responsibilities

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

The last chapter showed through the example of some individual but empirically
important cases—the liberalization of former utility services and the liberaliza-
tion of goods and services markets—how the ECJ’s case-law development can
drive EU legislation forward. The EU’s over-constitutionalization implies that
policymaking falls under the shadow of case law. The codification of this case
law promises greater legal certainty and improved equality before the law, and
re-establishes the responsibility of the legislature for policy development. At the
same time, such codification is under pressure to take interpretations of case law
to extremes in order to avoid the all too clear incentives for future litigation
by private actors.
This chapter assesses whether this argument also fits other cases. If market
regulation is overshadowed by the judiciary, this might not be so relevant, after
all. While regulatory policies also have redistributive implications, which were
clearly apparent in the debate around the Services Directive, the regulation of
markets is arguably not very sensitive to sovereignty and legitimacy. Many
instances of market regulation are delegated to private actors’ self-regulation,
for example in standardization organizations. Distributive and redistributive
policies are more closely tied to state sovereignty and are in need of greater
legitimation. If they are shaped by the judiciary, this is more problematic.
Welfare-state policies are redistributive, as are the taxes that are needed to
finance them partly. Equally closely connected to the core of state responsi-
bilities are questions concerning citizenship rights, which define those that
have rights and duties in a given state territory. In this chapter, I discuss how
case law also influences these sensitive issues.
The Citizenship Directive of 2004 and the Patient Mobility Directive of 2011
are examples of secondary law that are heavily influenced by case law. The
Citizenship Directive specifies when EU citizens acquire the same set of rights
so that they are treated on a par with nationals. The Patient Mobility Directive
is only concerned with a small area of member states’ health markets: the cross-
border demand for goods and services. However, the precondition for allowing
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Reaching Beyond the Market 127

such reimbursement has far-reaching consequences for health systems, in
particular those that follow a benefits-in-kind model, which now have to attach
prices to their goods and services.
In this book, I analyse how constitutionalized case law, as an independent
variable, impacts EU policymaking, as the dependent variable. To provide
conclusive evidence and to avoid selection bias, it is beneficial to include cases
that give different values for the dependent variable, instead of only analysing
cases of one kind (Geddes 1990). Luckily, there are prominent examples in
which continued case-law development failed to stimulate legislation: the
Monti II regulation, which aimed for reregulation after the contentious
‘Laval quartet’ rulings, and the case of direct company taxation. The discussion
of these cases concludes this chapter and allows me to be more precise about

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

the impact of case law on legislation.
Before analysing these case studies, it should be noted that the increasing
relevance of the ECJ to these core areas of statehood has also been interpreted
favourably. Caporaso and Tarrow have made the argument for a ‘Polanyi in
Brussels’, viewing the ECJ’s case law as supportive of the development of a
social-market economy (Caporaso and Tarrow 2009). Countering Scharpf ’s
and others’ criticism of a negative integration bias, Caporaso and Tarrow
emphasize that there has been a Polanyian re-embedding of markets. They
see ‘a structure of supranational embedded liberal compromises’ (emphasis in
original, p. 594) emerging through the Court’s case law on the free movement
of labour. Drawing on the arguments of Conant (2002) and Cichowski (2007),
they regard the transnational mobilization behind case-law development as an
alternative way to legitimize the EU (p. 613).
Contrary to the optimistic analysis from Caporaso and Tarrow, Höpner and
Schäfer (2012) have argued that policies that move non-discrimination for-
ward and enforce the market, which the ECJ can advance, only further
individual rights at the expense of solidarity. Höpner and Schäfer thus make
an argument similar to that of Scharpf, who believes that liberal rights are
strengthened at the cost of republican ones (Scharpf 2009). I will return to this
discussion after an analysis of the cases. The analysis begins with the two
examples of successful codification, the Citizenship and the Patient Mobility
Directives, and then turns to Monti II and taxes.

T H E CI T I Z E N SH IP D I R E C T I V E

With the provisions for the free movement of workers and the freedom of
establishment, the Treaty grants those citizens who are economically active
far-reaching rights to non-discrimination throughout the Union. The inclu-
sion of EU citizenship in the Treaty of Maastricht laid the foundation for the
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

128 The ECJ and the Policy Process

expansion of these rights to all EU citizens. Such an extension had been
debated since the 1960s, thereby making it necessary to adopt a more histor-
ical perspective. Case law played an important role in this whole process by
repeatedly interacting with legislation, as we will see. At Maastricht, member-
state governments had aimed to keep control over their citizenship rights:
‘The question whether an individual possesses the nationality of a Member
State will be settled solely by reference to the national law of the Member State
concerned.’1 But the Court’s dynamic case-law development has meant that
member states have partly lost their earlier autonomy.

The Historical Origins of Citizenship Rights

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

As it was originally founded as an economic Treaty, workers were privileged.
Article 48 of the Treaty Establishing the European Economic Community
(EEC Treaty/TEEC) demanded the abolition of discrimination against workers
on the basis of nationality for Employment, remuneration, and other conditions
of work and employment. Employment within the public sector was exempted
(Article 48(3) of the TEEC). Article 51 of the TEEC targeted the coordination
of national social-security systems, which are subject to unanimous agreement
in the Council. If rights to social benefits cannot be aggregated across member
states, this is an important barrier to free movement. The veto power of each
member state was continued in the Treaty of Amsterdam, which only granted
co-decision powers to the European Parliament (EP). The Treaty of Lisbon
changed this to the ordinary legislative procedure of a qualified majority but
adopted an emergency brake (Article 48 of the Treaty on the Functioning of
the European Union (TFEU)) for member states that disagreed (Hofmann
2013: 203).
The historical foundation of the free movement of workers lies in the
interests of five member states in importing labour and in Italy’s labour
surplus (Menéndez 2009: 4). Starting in 1958, different regulations and di-
rectives were adopted that detailed the rights of migrants and their families to
gain non-discriminatory access to employment, social advantages and bene-
fits, training, housing, taxes, and trade-union membership (Hofmann 2013:
207). ‘The case law of the European Court of Justice enlarged the scope of the
provisions on free movement of workers by expanding the understanding of
who was entitled to the right to free movement, and by means of recharacter-
izing the value basis of free movement of workers’ (Menéndez 2009: 5–6). As
Hofmann describes, in these early decades the Commission itself often
sided with member states in judicial disputes that arose from preliminary

1
Council of Heads of State and Heads of Government at their meeting in Edinburgh, Dec.
1992, cited in Jessurun d’Oliveira 2011: 146.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Reaching Beyond the Market 129

procedures. The first important conflict concerned the definition of ‘wage
earner’, which the Court viewed as subject to community definition in Case
75/63 Hoekstra in order to prevent member states from undermining the
rights of migrant workers. The Court opted for a broad definition, which
included all those who are covered by the different social-security systems
(Hofmann 2013: 207–8). Regulation 1408/71 on the coordination of the
social-security systems took up this case law. In the 1970s, the Commission
followed suit, arguing that exemptions to rights to free movement had to be
small. While member states had considered the public sector to be exempted
from the Treaty provisions, the Court, now backed by the Commission,
argued in Case 152/73 Sotgiu, which concerned the German postal sector,
that the public-sector exemption exclusively applies to those parts of the sector

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

that deal with public authority. Subsequently, the Commission aimed to
enforce this narrow definition in an infringement procedure against Belgium
(149/79), with Belgium receiving the support of France, Germany, and Britain
through interventions (Hofmann 2013: 210–11). On the basis of the Court’s
support here, the Commission then acted to liberalize employment in the
public sector (Hofmann 2013: 212).
The idea that European citizenship confers rights irrespective of economic
activity has been intermittently discussed since the early 1960s by Commis-
sioners and in the EP, and reached the summit in Paris in 1972 (Hofmann
2013: 217). Rights to free movement were seen ‘as a vehicle of both economic
and political integration’ (Menéndez 2009: 10). In 1977, the EP asked the
Commission to act and they submitted a directive on citizens’ rights of
residence to the Council in 1979. At this time, the EP were already arguing
that member states should refrain from demanding sufficient resources as a
precondition for rights to free movement, leading member states to fear
potential burdens on their social-security systems (Hofmann 2013: 221).
This legislative proposal stalled over the next decade, and progress was left
to the Court. An example was Case 53/81 Levin, which concerned a British
woman living in the Netherlands in part-time employment and earning less
than the Dutch minimum wage. Her application for a residence permit failed
for her third-country national partner. Against the Dutch and also the Danish
governments, who defined the status of being a worker as the ability to be self-
subsistent, the Commission argued for a broad definition of worker, which
was backed up by the Court (Hofmann 2013: 225–6).2 The Court held that
work had to be ‘effective and genuine’ and not ‘purely marginal and ancillary’,
attributes that remain important to this day. Case 139/85 Kempf dealt with
social benefits. A German music teacher had applied for an income supple-
ment in the Netherlands, as he earned less than the Dutch minimum wage.

2
Hofmann mentions the following similar cases, which were all preliminary references:
66/85, 197/86, 196/87, 344/87 (2013: p. 227, n. 95).
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

130 The ECJ and the Policy Process

Again, the Court sided with the Commission against the Dutch and the
Danish governments (Hofmann 2013: 227).
Despite these judicial advances, the legislative process stalled, leading the
Commission to split up its proposal on citizenship rights into three directives:
for students, pensioners, and a general residence directive for those who have
sufficient resources and health insurance (Hofmann 2013: 229–30). They were
agreed upon in 1990 (directives 90/364, 90/365, 90/366). Shortly afterwards,
the heads of the member-state governments included EU citizenship in the
Treaty of Maastricht. Article 8a of the Treaty Establishing the European
Community (TEC) reads that ‘Every citizen of the Union shall have the
right to move and reside freely within the territory of the Member States,
subject to the limitations and conditions laid down in this Treaty and by the

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

measures adopted to give it effect.’ For member states, this article was a merely
symbolic addition (Aziz 2009), but for the Court and the Commission, it laid
the foundation for the broadening European rights.
One central aim of the Commission was to expand the social-security
coordination rules of regulation 1408/71, which only covered economically
active persons at that point, to include all insured citizens. It submitted a
proposal to the Council in 1991 (Hofmann 2013: 234–5), which also stalled.
New impetus was achieved when at Amsterdam the Council agreed on an
action plan for the full implementation of the single market, which led the
Commission to propose the different pieces of legislation that dealt with the
free movement of workers, aligning it with existing case law (Hofmann 2013:
235–6). In 1998, proposals followed to expand social-security coordination to
all insured EU citizens, resulting in the revisions of Regulation 1408/71 (now
883/2004) and Regulation 1612/68 (now 492/2011).
The 1998 proposal for an amendment of Regulation 1612/68 emphasizes
the advances made by case law as follows:

The Commission intends to ensure that these proposals succeed in their aim of
improving conditions for freedom of movement to reflect the spirit expressed by
case-law. The latter is a basic step forward for the European citizen and the
Commission will ensure that discussions in the Council do not lead to the loss of
the headway made by case-law. (COM 98/394, p. 6; see also Hofmann 2013: 240)
The rules are now 30 years old, however. Over this period, the Court of Justice has
repeatedly ruled on the texts and interpreted them. As a result, a whole corpus of
case-law to interpret the wording of the legislation has been created. To strength-
en the security and transparency of the law on behalf of the citizen, it is now time
to bring the texts into line with existing case-law. (COM 98/394, p. 4)

Nevertheless, reaching an agreement proved difficult and protracted, and was

only successfully established in 2004 and 2011 (Hofmann 2013: 237). Parallel
to these negotiations, the Commission submitted a proposal in 2001 that
consolidated the legislation on residence rights, bringing together what it had
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Reaching Beyond the Market 131

not managed to do in the 1980s. The resulting Citizenship Directive (2004/38/
EC) was adopted at the same time as Regulation 883/2004, and we will address
it in a moment.

Case-Law Development

While negotiations stalled in the 1990s, the Court’s case law accrued. The
Commission’s behaviour is highly interesting in this respect: Hofmann empha-
sizes that it largely refrained from infringement procedures and only joined
preliminary references, in contrast to the market for goods and services
(Hofmann 2013: 233). Moreover, for quite some time, the Commission only

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

favoured a partial extension of rights in its observations. While pushing to expand
residence rights, it often sided with member states’ restrictive position when
litigants demanded access to social benefits (Hofmann 2013: 239, 242–3, 245).
The first important case here was C-85/96 Martinez Sala, which concerned
a Spaniard with legal residence in Germany, whose application for child-
raising allowance had been denied since she was not working and, therefore,
was not covered under the right to free movement (Menéndez 2009: 12–13).
After Maastricht, the Commission argued, the Treaty covered her rights of free
movement and residence. The Court sided with the Commission and against
interventions by the German, French, and British governments (Hofmann
2013: 238–9). Moreover, the Court held that an economically inactive EU
citizen lawfully residing in Germany had to have the same access to child-
raising allowance that Germans have (Tryfonidou 2010: 39).
Case C-413/99 Baumbast concerned a German who was married to a
Colombian and wanted to stay in the UK after his job there had ended. The
Commission argued that he needed health insurance and sufficient resources
to gain residence rights in another member state according to the Treaty
(Hofmann 2013: 239).3 But the Court found that EU citizens enjoy the right
of residence and free movement through the direct application of Article 18(1)
of the TEC (Wind 2009: 256, 259; Menéndez 2009: 13–15; Wollenschläger
2011). As Menéndez holds, the ‘main concrete implications of Martinez
Sala and Baumbast for the supranational citizen has been access to non-
contributory social benefits’ (Menéndez 2009: 18). Nevertheless, he does not
consider these cases to be ‘revolutionary’ but rather to push the existing case
law to its ‘logical conclusion’ (Menéndez 2009: 26). This view supports my
path-dependent analysis.
Also, in Case C-184/99 Grzelczyk, the Commission came down against an
extension of rights by arguing that students cannot claim social benefits

3
The Court diverted here from Lebon (1987) in which it had rejected an instance of non-
discriminatory access to unemployment benefits (Hailbronner 2004: 2186, 2189).
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

132 The ECJ and the Policy Process

(Hofmann 2013: 239). A French student had applied for Minimex social
support for his last year of study in Belgium. The ruling established that
right of residence also applied to economically inactive EU citizens who
required assistance, and this was decided in the midst of the negotiations on
the Citizenship Directive (see next section). The decision was reached that
only in cases where EU citizens place an ‘unreasonable burden’ on a member
state can this right be withdrawn; up until that point, they have to be treated
on a par with nationals in terms of social assistance (Tryfonidou 2010: 39).
The Court thus diverted from the student directive (93/96/EEC), which denied
social assistance to students. It even argued in C-184/99 No. 39 that, although
the directive did not establish entitlement to assistance, it also did not ‘pre-
clude’ such benefits (Hailbronner 2004: 2186). In Case C-209/03 Bidar, the

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Commission again denied that students should have access to maintenance
grants (Hofmann 2013: 240). This case concerned a Frenchman who had
moved to the UK to live with his grandmother. After completing secondary
education there, he started university, but his application for a student loan
was denied. In support of the Commission, the UK, Denmark, Germany,
France, the Netherlands, Austria, and Finland all participated with interven-
tions that argued for the exclusion of students from host state’s social assis-
tance and study grants. Diverting from the previous Case C-39/86 Lair and
Case C-197/86 Brown in 1988, the ECJ ruled that assistance needed to be given
as the Maastricht Treaty had included education (Wind 2009: 243, 262). Case-
law development for citizenship thus illustrates how the expansion of rights
does not depend on central orchestration by the Commission. Preliminary
references can drive the process forward as well.

Negotiating the Citizenship Directive (2004/38/EC)

As we have seen, the Commission’s initial attempt to legislate rights to free

movement in a comprehensive way failed in the 1980s and early 1990s. The
Citizenship Directive realizes this comprehensive legislation by consolidating
nine different directives, as well as amending Regulation 1612/68 by explicitly
taking the case law into account. In its first proposal for this directive (COM
(2001) 257 final), the Commission referred to the ECJ in its explanation for
fourteen articles (6, 8, 12, 13, 19, 21–3, 25, 28–31, 33), as well as in Recital 21.
Negotiations in the beginning were slow; member states were reluctant to see
the broad proposal as pertaining to the single market, and several Council
working groups felt responsible.4 In addressing the rights of third-country
national family members to accompany EU nationals to other member

4
Interviews: Permanent Representative Austria 26 June 2009; German Federal Ministry of
Labour and Social Affairs 17 Nov. 2009.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Reaching Beyond the Market 133

states, immigration questions were touched upon that were already part of
Regulation 1612/68. And by dealing with conditions of expulsion on grounds
of public security or public order, the directive engaged with matters of justice
and home affairs.
In the following section, my focus will be on EU citizens’ access to social
benefits, which became important in the negotiations after the Grzelczyk
ruling. This made the negotiations more contentious in comparison to the
original emphasis on rights to free movement that member states had sup-
ported (Schreinermacher 2013: 88). In its revised proposal (COM (2003) 199),
the Commission responded to eighty-two amendments that were put forward
by the EP. Notably, the EP called for non-discriminatory access to welfare
assistance after six months of residence, a priority it had pursued for a long

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

time (A5-0009/2003). This put the member states that wanted a directive on
rights to free movement under pressure, as they could only revise the Com-
mission’s proposal unanimously.
The Commission’s original emphasis on rights of free movement, instead
of access to welfare, is in line with its opinions in ECJ cases from that time.
Both the amendments to the directive that the EP requested and the devel-
oping case law then led the Commission to change its position
(Schreinermacher 2013: 85–6). The Commission justified the revision of
its proposal with several references to recent case law, most notably to:
Case C-459/99 MRAX (2002) on family members not needing visas (Article
9(2a)); Case C-413/99 Baumbast (2002), which concerned residence rights
for parents (Article 12(3)); and Case C-184/99 Grzelczyk (2001), in terms of
the important Article 21(2) on equal treatment. Relating to welfare assis-
tance, it rescinded the original exclusion of economically inactive persons
with no permanent residence status. In the revised proposal (COM (2003)
199), Article 21(2) only excluded maintenance grants for students, whereas
before the sentence had included the underlined section of the following: ‘the
host Member State shall not be obliged to confer entitlement to social
assistance on persons other than those engaged in gainful activity in an
employed or self-employed capacity or the members of their families nor
shall it be obliged to award maintenance grants to persons having the right of
residence who have come there to study.’
As one might suspect, member states did not simply accept the new
emphasis in free movement on equal treatment in the access of welfare
assistance. Agreement on the directive only succeeded shortly before the
accession of ten new member states in May 2004. Interestingly, the Commis-
sion failed to align the rights of economically active and inactive citizens, as it
had hoped to do. As a result, Recital 9 speaks of ‘a more favourable treatment
applicable to job-seekers’. Distinctions between economically active and in-
active EU citizens pervade the directive and are also mentioned in Recitals 16,
19, and 21, and in Article 24(2).
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

134 The ECJ and the Policy Process

The directive differentiates residence rights according to citizens’ length of
stay in a member state. The terms here changed during the course of negoti-
ation. Originally, the Commission had planned to grant unconditional resi-
dence rights for six months and to allow permanent residence after four years.
The final directive distinguishes rights in the first three months, when every
EU citizen may reside in any EU member state. After five years, permanent
residence sets in. In the interim period, EU citizens and their families can stay
for more than three months on the condition that they have health insurance
and sufficient resources in order not to become ‘an unreasonable burden on
the social assistance system of the host Member State’, according to Article 7
(1). Overall, the directive mentions that EU citizens should not become such a
burden eight times. This very much reflects the Court’s argument in the

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Grzelczyk case (Hailbronner 2004: 2187) and later in Trojani (Wind 2009:
243, 262), but also draws on the fourth Recital of Directive 90/364 (Hofmann
2013: 171).
In the final directive, Article 24 spells out the extent of rights to equal
treatment. ‘Article 24 was widely understood as an attempted override’ of the
case law, Davies (2014: 1600) argues. Paragraph 2 sets out that member states
are not obliged to grant social assistance during the first three months or,
‘where appropriate, the longer period provided for in Article 14(4)b’. The
latter demands that EU citizens may not be expelled if they continue to seek
work and ‘have a genuine chance of being engaged’. This is another reference
to case law (Case 75/63 Unger). Host states are not obliged to give out student
grants or loans before permanent residence is granted. It becomes apparent
that the directive is a response to case law but, as such, it does not eliminate the
resulting legal uncertainty. As Schreinermacher (2013: 92) argues, member
states deliberately left the rights to welfare benefits of economically inactive
EU citizens—between month four and the granting of the final residence
status after five years—open in the directive. After all, Grzelczyk was able to
access benefits after three years, whereas in the Citizenship Directive this was
only possible after five years. Following this judgment, it was no longer
possible simply to exclude EU citizens from benefits, as the Court had spoken
of a certain amount of financial solidarity between the citizens of different
member states (No. 44) (Giubboni 2007). As Article 14(3) determines, ‘An
expulsion measure shall not be the automatic consequence of a Union citizen’s
or his or her family member’s recourse to the social assistance system of the
host Member State.’ Secondary law could not reverse case law, given the
latter’s constitutional nature (Wasserfallen 2010: 1140). ‘Representatives
from Luxemburg, the UK, Belgium, Ireland and Denmark still remained
critical about this legislative confirmation of the Court’s jurisprudence. But
throughout the policy-making process, the leverage of reluctant member
states continuously declined to the point where this coalition was too weak to
prevent the institutionalization of case law’ (Wasserfallen 2010: 1141). At the
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Reaching Beyond the Market 135

same time, Wasserfallen (2010: 1142) points out that member states took the
legislative process as an opportunity to signal to the court that they did not
approve of EU nationals’ rights to social assistance.
It is plausible that member states felt under pressure. The upcoming Eastern
enlargement would have made agreement on a directive much more difficult
(Interview German Federal Ministry of Labour and Social Affairs 17 Novem-
ber 2009). At the same time, once the Commission had begun to move across
to a position that favoured access to benefits, there was the danger that it
would pressure member states with infringement procedures to push them in
this direction. In Case C-138/02 Collins, the Commission argued that EU
citizens must be permitted access to jobseeker’s allowance, even if citizens
had no connection to the given member state (AG Colomer 10 July 2003,

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

No. 57) (Schreinermacher 2013: 101). In autumn 2003, it handed an infringe-
ment procedure against Belgium to the Court (C-408/03). In it, the Commis-
sion opposed the Belgian rules that required EU citizens to have sufficient
personal means as a precondition for their right of residence. Schreinermacher
(2013: 101) argues that this communicated the Commission’s greater asser-
tiveness to the member states. Advocate General (AG) Colomer, however,
took the problem of abuse of citizenship rights seriously in the opinion he gave
to Collins, arguing that member states could restrict access to welfare services
through transparent, non-discriminatory means (No. 75).
Moreover, it has to be kept in mind that member states were put under
pressure from preliminary procedures. The long-term restraint of the Com-
mission had no parallel in domestic courts. Preliminary procedures drove
case-law development. As Schreinermacher (2013: 109) shows, German courts
were actively taking up the reasoning of the ECJ through fifty-six references to
the Grzelzcyk ruling between 2001 and 2011. In these cases, however, German
courts often denied the relevance of the ruling. But a restrictive stance in
domestic courts may be only temporary. There was always the fear that, by
forwarding a preliminary reference, German courts could undermine crucial
elements of the German welfare-state system.5 Britain was directly involved
with the Collins (2004) and Bidar (2005) court cases, which concerned benefits
for students. Its domestic courts also took up European case law to push for
restriction (Schreinermacher 2013: 112–13). In France, domestic courts largely
backed the actions of the administration, as is made apparent by the fact that
there were fewer preliminary references (four) than infringement procedures
(seven) between 1995 and 2004 that concerned rights of free movement
(Schreinermacher 2013: 116). French courts did not refer to the ECJ’s crucial
case law from Grzelczyk, Collins, or Trojani (Schreinermacher 2013: 117). This
significantly lessened the pressure on the French government.

5
This happened from 2013 onwards with the reference of the German federal social court
concerning the restriction of not paying Hartz IV social benefit to EU citizens; see Chapter 7.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

136 The ECJ and the Policy Process

As we have seen, the case law on EU citizens’ access to welfare services took
some years to develop. The first case, Martinez Sala, was a very specific one, as
the woman had already lived in Germany for a long time, and member states
did not need to worry about EU citizens’ rights to benefits, as long as they
themselves determined the conditions of entry to and expulsion from their
territory. However, gradually these rights were increasingly defined at the
European rather than the national level. Thus, in Collins, the Court argued
that the rights of residence granted by member states are merely declaratory
and not legally constitutive (Groß 2005: 83), since residence rights are ‘con-
ferred directly by the Treaty’ (No. 40). For member states that were anxious to
restrict access to their welfare services, a directive that remained opaque as to
the extent of entitlements was far better than one granting far-reaching rights

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

after a three-month stay. Given member states’ disagreement on many aspects
of the directive, the text was bound to be ambiguously worded in many
instances, reflecting the underlying political compromises.6 Schreinermacher
(2013: 234–5) argues that case law may pay more regard to autonomy than
secondary law, as it does not generalize legal rights but partly hides them
under a screen of legal uncertainty. Actors that favour far-reaching rights and
liberalization will therefore opt for codification, while actors that favour
restrictions will prefer case law.

Case-Law Development After the Directive

As I have mentioned, some governments hoped that the Citizenship Directive

would signal member states’ preferences to the Court. Were they successful?
The case-law development that followed the directive shows that the Court
was not responsive for a long time.
One might have expected the Court’s approach to the conditions in the Directive
to be inspired by a desire to understand what compromise exactly the legislature
sought to make. Instead, to a significant extent, it has placed the Directive within
its own framework of purposive free movement principle, and interpreted the
Directive in that light, with the result that several of its conditions have been
deprived of at least some of the consequences they were intended to have.
(Davies 2014: 1599)
Although the member states had largely incorporated the existing case law
into the Citizenship Directive of 2004, the Court’s activism could not be
curbed. In Cases C-22/08 and C-23/08 Vatsouras (4 June 2009), the Court
went so far as to overrule the recent directive by arguing that financial benefits

6
Interviews: Permanent Representative Austria 26 June 2009; German Federal Ministry of
Labour and Social Affairs 17 Nov. 2009; European Civil Action Service 27 Oct. 2010.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Reaching Beyond the Market 137

intended to facilitate access to the labour market were not ‘social assistance’ as
defined by Article 24(2) of the directive but that the equal-treatment provision
of Article 45(2) of the TFEU applied instead. A ‘real link’ of citizens to the
member states was, however, required (No. 38–40) (Barnard 2010: 285).
Thanks to Vatsouras social assistance appears to have become a residual category.
It is perhaps ironic that measures taken precisely in order to make domestic
access to benefits harder and stricter now have the legal effect of making them
easier to obtain for migrants. (Davies 2014: 1600)
Case C-310/08 Ibrahim (February 2010) was a landmark ruling in the en-
hancement of migrants’ access to social assistance. This case concerned a
Somali woman who had remained in the UK after her Danish husband, who

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

had worked there previously, left the country. She was granted the right to
social assistance given that her children—Danish nationals—were going to
school in the UK. It is interesting that the legal proceedings were paid for by
Shelter, a UK charity that supports homeless people.7 The activities of non-
governmental organizations (NGOs) and the Commission, in its attempts to
broaden access to EU rights, may offer support to individuals who do not have
access to the financial means of corporate actors.
It was not only access to social benefits that generated contentious case-law
development. The ECJ also proceeded to claim its competence in defining the
residence rights of EU citizens’ third-country national family members. The
ruling in Case C-127/08 Metock (25 July 2008) was particularly contentious
here, given that Article 35 of the Citizenship Directive allows member states to
challenge the abuse of rights or fraud.8 This case concerned four third-country
nationals who were illegally resident and then married EU citizens who were
not nationals of Ireland. Previously, in Case C-109/01 Akrich (23 September
2003), the ECJ had argued that the right of third-country nationals to accom-
pany an EU citizen was dependent on having prior legal residence. This
condition was now abolished. Metock overturned member states’ agreement
in the Citizenship Directive, which allowed third-country national family
members to join EU citizens who were residing in another member state.
Metock and other cases transformed this, granting privileged residence con-
ditions to EU citizens’ third-country national family members as soon as there
is a cross-border dimension and, therefore, EU law applies. This gives incen-
tives for migrants to use the ‘Malmö route’ or the ‘Dublin hop’, through which
Danish or UK families move for a limited time to qualify as EU citizens in
order to profit from family unification (Martinsen 2011: 957; Wind 2014:

7
They recruited Nicola Rogers, a solicitor who works for Garden Court Chambers, a large
barrister firm, to argue this case at the ECJ: <http://www.gardencourtchambers.co.uk/ecj-rules-
on-right-of-residence-of-parent-caring-for-child-of-migrant-worker> [accessed 9 May 2017].
8
Interviews: Permanent Representative Germany 24 June 2009; German Federal Ministry of
Labour and Social Affairs 17 Nov. 2009.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

138 The ECJ and the Policy Process

166–7). An important issue here is the requirement for the amount of time
that has to be spent abroad. Two preliminary rulings from the Netherlands on
Cases C-456/12 and C-457/12 asked whether family reunification must also be
granted to frontier workers and to those who have lived in another member
state for a short time. In her opinion, AG Sharpston argued for a broad
approach, which would include rights derived from the freedom to provide
services. In its rulings in 2014, the Court held that residence in another
member state needed to be ‘sufficiently genuine’ in order for third-country
nationals to have rights conferred upon them.
Recently, the ECJ has also favoured more restriction in the opening up of
non-contributory social benefits to economically inactive EU citizens. In Cases
C-333/13 Dano, C-67/14 Alimanovic, and C-299/14 Garcia Nieto, the Court

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

held that Germany could restrict its non-contributory Hartz IV social benefit
to nationals under certain conditions. I will discuss these cases in Chapter 7. It
is noteworthy here that, in being more constrained, the Court has not aban-
doned its path of interpreting the fundamental rights according to a principle
of non-restriction. It has not argued for a distinct interpretation of citizenship
rights thus far. However, it is refraining from extending its interpretation
further, thereby reducing the legal uncertainty related to the reach of EU
law and taking away incentives for further litigation.
Some years ago, Kadelbach warned that case law may have ambivalent
effects. If it broadened social entitlements, member states would likely restrict
residence rights in response (Kadelbach 2003: 563). After the introduction of
the Citizenship Directive, the ECJ has made it much more difficult for member
states to expel EU citizens or their family members. This may give incentives
for cutting down on social benefits in general (Scharpf 2009; Wiesbrock 2012:
93). Conant (2002: 194) mentions a scrapped German plan for a supplemen-
tary old-age pension as an example of this: ‘Ostensibly “progressive” case law
contributed towards a marked lack of progress for poor pensioners in Ger-
many.’ The debate on the Brexit referendum similarly brought general cuts
such as these onto the agenda. Out of its struggle with the non-discrimination
provision of the Treaty, and in line with incumbent party ideology, the UK’s
Conservative government has cut student assistance for UK and EU students
alike (Schenk 2016). During the Brexit discussion it aimed to bar all EU
citizens from access to benefits during their first four years of residence
in the UK.
Thus, the development shows that the ECJ has not only repeatedly deviated
from established case law, for instance in Grzelczyk from Brown, and in
Baumbast from Lebon, but it has also frequently departed from the secondary
law of the Council and the EP. Wollenschläger (2012: 327) has even argued
that the Court has avoided ‘open conflict’ as it could have declared the
legislation as ‘void’ by infringing primary law. The explicit reference to
secondary law in Article 21 of the TFEU stands in opposition to this; the
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Reaching Beyond the Market 139

Article would mean that the Court must ‘articulate a retreat of almost all of its
other citizenship case law’ (Shuibhne 2012: 360).
In continuously extending the entitlements of EU citizens, the ECJ has been
supported by a legal profession that has been united in its teleological legal
interpretations. Citizenship rights have come to play a central role here, as
these promise to transfer individuals’ allegiance from the nation state to
the EU.
Notwithstanding this, the political elite of Europe has mismanaged the EU’s
constitutional process by failing to place the citizen at the center. This failure is
especially important, since without the support of its citizens, European integra-
tion as a democratic project becomes meaningless. The same cannot be said
regarding the judicial elite, who have championed the cause of citizenship to a

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

considerable extent. (Aziz 2009: 282)
Whether this will play out as planned once citizens realize that there will be a
concomitant loss in the national ability to shape the welfare state democrat-
ically is questionable. Research shows that support for the Union is low when
citizens perceive their national welfare systems to be under threat from the EU
(Beaudonnet 2014). In the Brexit referendum, the perceived loss of influence
over the conditions of free movement and citizenship led the British electorate
to vote to leave the EU.

Conclusion

The development of citizenship rights has been hailed as an important move

away from the exclusive market orientation of the EU, by embedding the oft-
criticized neo-liberal EU integration project in strengthened social rights
(Caporaso and Tarrow 2009). Lawyers generally support an expansive inter-
pretation of EU citizenship. One of the few pronounced critics is Menéndez,
who points to the increasing ‘judicialisation of major areas of national law and
policy’ (Menéndez 2009: 27) as a process in which the Court takes power from
the national representative institutions, and therefore from nations’ ability to
democratically shape the integration process. Moreover, and in a similar vein
to Scharpf (2009), he criticizes individualistic, as opposed to collective and
solidarity-based, outcomes: ‘Pretending that the extension of welfare rights
always lead to a better protection of the welfare objective is simply illusionary,
because the key point of any redistributive program is to use the taxes collected
from some to comply with the obligation of distributive justice they had
towards others’ (Menéndez 2009: 30). Mobile individuals are strengthened at
the expense of immobile ones. Though the individuals in these cases do not
appear privileged, by being mobile and able to enforce their EU rights, they
belong to the better off (Menéndez 2009: 30–1). Only a minority of citizens has
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

140 The ECJ and the Policy Process

the skills and the means to operate within the scope of community law.
‘Individualism explains the substance of European citizenship’ (Somek 2007:
815). If member states lose the power to determine the recipients of their non-
contributory benefits and cannot control the costs, it may serve as a disincen-
tive for the expansion of their welfare systems.
Bellamy (2008b) provides another argument against the increasingly inde-
pendent status of citizenship, which had been planned as a complementary
status. As he argues, belonging, rights, and participation constitute national
citizenship and are all closely intertwined: ‘regular and open-ended interaction
between repeat-players reduces the incentives for free-riding and defection
and builds confidence in the possibility of collaboration. It helps engender the
bonds of reciprocity needed to produce benefits that are diffuse and public

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

rather than direct and purely personal’ (Bellamy 2008b: 599). The EU, in
contrast, has instilled hardly any sense of belonging in citizens (Bellamy
2008b: 603). Contrary to Habermas’s belief that rights can be a basis for
solidarity and replace a sense of belonging, Bellamy sees them as ‘at best a
necessary rather than a sufficient condition of subjective legitimation’
(Bellamy 2008b: 604). He therefore contends that EU rights ought to be
supplementary and should not ‘disrupt the rights enjoyed by national
citizens—not least with regard to their access to domestic services’ (Bellamy
2008b: 606). In contrast to European lawyers who assume that meaningful
rights can only be granted at the European level (Kochenov 2013), constructed
by lawyers and courts rather than the democratic process, Bellamy gives a
strong argument for taking the member states seriously: ‘Unlike subjects,
citizens are equal before the law because they enjoy an equal influence over
the making of the laws through being participants in a democratic process’
(Bellamy 2008b: 600). In contrast to Kochenov, he argues for citizenship rights
to be coupled with duties (Bellamy 2015).
If member states have been put under sufficient pressure by case law to
codify far-reaching citizenship rights, what do we know about the implemen-
tation of the directive? In its first report on the directive, the Commission
noted immense implementation deficits. ‘In the thirty months since the
Directive has been applicable, the Commission has received more than 1800
individual complaints, 40 questions from the Parliament and 33 petitions on
its application’ (COM (2008) 840/3, p. 9). The Commission initiated nineteen
infringement procedures against member states between mid-2006 and early
2007 (COM (2008) 840/3: 3). Among other problems, twelve member states
transposed the ‘sufficient resources’ requirement incorrectly (p. 6), and thir-
teen member states allowed expulsion of EU citizens as an automatic response
when social benefits were taken up (p. 7).
But in its 2013 action plan, the Commission is much more positive and
observes that ‘90% of transposition issues’ have been solved due to its ‘rigorous
enforcement policy’ (COM (2013) 837 final, p. 2). The European Citizen
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Reaching Beyond the Market 141

Action Service (ECAS), an NGO that runs the Your Europe Advice helpline
for EU citizens and is sponsored by the Commission, is much more negative in
its assessment. EU citizens find help here in exercising their citizenship rights;
they received more than 17,000 inquiries, which were handled by sixty legal
experts, in 2012 (ECAS 2013: 10). In its report titled Mind the Gap, which was
published in 2013, ECAS argues that it ‘does not see improvements on the
ground’ (p. 10), due to a ‘lack of political will on the part of Member States’
(p. 10), and that the citizenship directive has ‘resulted in more, rather than
fewer restrictions on the free movement of citizens’ (p. 9).

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

THE P ATIENT MOBILITY DIRECTIVE

Health services are another important case where the internal market reason-
ing became relevant through case law, despite member state opposition (Greer
2006; Martinsen 2005). It is also interesting because Martinsen discussed it in
her book, where she uses it to refute the activist court perspective (Martinsen
2015).9 Originally, the Services Directive draft from January 2004 included an
article on patient mobility, which aimed to codify the existing case law. Given
the general debate around the Services Directive, patient mobility was dropped
after a resolution of the EP in 2005 argued for a separate approach to
healthcare (OJ C 124 E, 25 May 2006, p. 543). The EP then called upon the
Commission ‘for a codification of existing case law on the reimbursement of
cross-border health care law in order to ensure the proper application of the
case law by all Member States’ (A6-0173/2007, No. 35). After a consultation
process, the Commission published its proposal for a directive on patient
mobility in 2008, leading to its adoption in March 2011 (Rosa 2012: 26–7).
The close link to ECJ rulings since the late 1990s pervaded the whole
legislative process. In fact, of all the examples discussed, the rationale of having
to codify case law was strongest in the case of the Healthcare Directive (Rosa
2012; Hancher and Sauter 2010: 117). ‘The directive proposal thus aims to
codify the court’s interpretation of the scope and limits of intra-European
healthcare’ (Martinsen 2005: 1049). Contrary to Martinsen’s (2015) more
recent argument, the EU legislature could not overcome or modify these
judicial constraints in any meaningful way, as I will show.
As early as 2002, a high-level group of experts from the member states
had been assigned the task of exploring ways to increase legal certainty in view
of the ECJ’s case law on patient mobility (Rosa 2012: 25). Due to the hetero-
geneous health systems of the member states, which often treated health

9
I have profited greatly from my collaborative work with Marzena Kloka on this directive
(Kloka and Schmidt 2015).
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

142 The ECJ and the Policy Process

services as benefits-in-kind, the implications of case law were difficult to
determine. In the politics of health, judicialization resulted from individual
complaints. It is striking how the case-law development was driven exclusively
by preliminary procedures and that hardly any infringement measures were
initiated by the Commission (IP/07/1515). The restrictive Commission po-
sition is reminiscent of its position on citizens’ rights. Obermaier argues that
the Commission opted to follow up the case law with codification, instead of
intensifying infringement procedures (Obermaier 2009: 146, 151).
In the following section, I will first discuss the case-law development that
put pressure on the legislature to act. Secondly, I will analyse the Council
negotiations, as these show that member states reacted in different ways to the
constraints of existing case law on legislative options. Some governments

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

wanted to seize the opportunity to influence future case law through second-
ary law, moulding it more closely to their political preferences. Other govern-
ments, notably those of the new member states, opposed the idea of subjecting
their health services to greater market freedoms. They seemed to prefer to let
case law reign, rather than generalizing these legal positions through second-
ary law, which would require far-reaching changes to their systems.

Existing Case Law on Patient Mobility

The Treaty (Article 152(5) of the TEC, now Article 168(7) of the TFEU)
regards health as a competence that is shared between the EU and member
states. It protects member states’ autonomy to regulate their health systems,
giving limited competence to the EU for public-health measures (Hatzopoulos
2002: 685). In order to allow for the treatment of patients in other member
states, member states agreed on coordination rules at an early stage: Article 22
of Regulation No 1408/71 established a system (called E112)10 that allowed a
patient in need of treatment during a stay abroad to access ‘the benefits in kind
which the host state provides to its own insured persons or to cash benefits
which are to be paid by the home state and which permit patients to defray the
cost of healthcare in the host State’ (Ackermann et al. 2008: 1326). For patients
actively seeking treatment abroad, the regulation includes a procedure for
prior authorization.
From the late 1990s onwards, this status quo, which was regulated by
secondary law, changed, as several cases reached the Court through prelim-
inary procedures. Disrespecting the prior authorization required by secondary
law, patients sought treatment in other member states and attempted judicial
redress against the limited or entirely absent reimbursement offered by their

10
Regulation 883/2004 adapted it into the European Health Insurance Card (EHIC).
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Reaching Beyond the Market 143

health insurance. Remarkably, instead of simply referring to the existing
framework of secondary law and the rules that the legislature thought should
govern patient mobility, the Court ignored these in favour of ‘creating an
alternative source of rights in the same field’ (Davies 2014: 1603). This case law
culminated in the directive.
As Dorte Martinsen (2005: 1037–8) points out, national healthcare systems
had already been challenged by Cases C-117/77 and C-182/78 Pierik in the late
1970s, which required the member states to refund treatment abroad even if
the domestic health services did not provide for this. The treatment only
needed to be internationally recognized as necessary and effective. The ruling
thus took away member states’ authority to determine the scope of their health
services. It resulted in an instance of legislative overrule in which the Council

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

reacted by unanimously amending the secondary law in question so as to
restrict reimbursement to only those treatments that were agreed upon by the
national health system (Martinsen 2005: 1038–9).
As early as 1984, the ECJ established that healthcare fell under the freedom
of services (Luisi and Carbone 286/82 and 26/83). This was reiterated in Case
C-159/90 Grogan, but it did not have much impact (Hatzopoulos 2002: 688).
Momentum only began to gather in 1998 when Cases C-120/95 and C-158/96
Kohll and Decker took place, which were both initiated by citizens of Luxem-
bourg. Decker had bought glasses abroad, while Kohll’s daughter had had dental
treatment that would have been available in Luxembourg. Both cases related to
the passive freedom of services. Neither had sought prior authorization. When
their reimbursement was refused, they litigated. With the submission of opinions
by Luxembourg and eight other member states, member states signalled clearly
to the Court that they attributed high levels of importance to the prior author-
ization procedure and wanted to keep their social-security systems separate from
the single market’s principles. But the Court argued that the freedom of services
applied here, turning prior authorization into a disproportionate burden on
patients. It is interesting and fairly typical of such rulings that the Court
nevertheless referred to ‘the powers of the Member States to organise their social
security systems’ (No. 17 Kohll, No. 21 Decker), as if the need to comply with
market freedoms would not impact this autonomy. The existing regulation,
which required such authorization, did not stop the Court.
Case C-157/99 Smits/Peerbooms in July 2001 was the next one of relevance.
Two Dutch patients left the country for treatment: one to Germany and one to
Austria. However, it was not part of Dutch standard treatment and was
therefore not reimbursed. The situation was comparable to the early Piriek
cases (Martinsen 2005: 1040). As Smits/Peerbooms concerned hospital care,
ten member states intervened against the application of single-market rules,
given that care was ‘provided in the context of a social security scheme’
(Hatzopoulos 2002: 691), but the Court argued that these were services like
any other and prior authorization was a restriction. The authorization procedure
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

144 The ECJ and the Policy Process

could only be seen as proportionate, according to the needs of hospital planning,
if it was transparent, timely, and based on non-discriminatory criteria. Moreover,
standard treatment needed to be based on ‘international medical science’
(Martinsen 2005: 1042). The case law was affirmed in Case C-56/01 Inizan
(Martinsen 2005: 1043).
Case C-368/98 Vanbraekel was decided on the same day as Smits/Peer-
booms. It raised another noteworthy point. The Court argued that in those
cases where non-domestic treatment was less expensive, the patient had a right
to claim the surplus. Otherwise, the application of the freedom of services
could be hampered (No. 45, 53) (Hatzopoulos 2002: 690; Krajewski 2010: 172;
van de Gronden 2009: 714–16).
Case C-385/99 Müller-Fauré & Van Riet followed in May 2003. It con-

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

cerned an instance where reimbursement in the Netherlands had been denied
for dental treatment and hospital care because of a lack of authorization. Nine
member states, together with Norway and Iceland, submitted opinions. The
Court saw no justification for an authorization procedure for non-hospital
care. Jeopardy to the financial balance of the social-security system needed to
be proven, but the planning that was required could possibly justify prior
authorization for hospital services, allowing member states greater leeway. The
case is also relevant as it explicitly referred to the reimbursement of providers
that did not have a contract with the national health insurance programme
(named ‘non-contracted providers’). These became a major concern for Eastern
member states in the negotiations of the directive.
Case C-372/04 Watts, which was decided upon in May 2006, concerned a
woman named Ms Watts, who had avoided the waiting times in the NHS by
going to France to get a hip replacement. The case is significant, as the NHS
is a pure benefits-in-kind system and, as such, had hoped to evade the market
logic of the freedom of services. Though it does not attach prices to its services,
it was subjected to the market freedoms. The Court’s ruling forces member
states to come up with transparent cost calculations for their hospital services
in order to allow the necessary transborder reimbursement (van de Gronden
2009: 720). Moreover, the ECJ dealt with the issue of waiting times, providing
criteria for what it held to be undue delay. Nevertheless, Ms Watts could
not recoup her travel costs. Even after Watts it was still unclear to what extent
tax-funded national health services, such as the NHS, fall under the freedom of
services (No. 91) (Krajewski 2010: 170).
In the judgment on Case C-444/05 Stamatelaki, reached in April 2007, the
Court held that member states cannot simply refuse to reimburse hospital
costs from other member states on the basis that they were incurred in a
private hospital without a contract with the national health system. After
all, these non-contracted providers are subject to the quality controls of the
home member state, and the professionals working for them have to comply
with formal qualifications (No. 37). Thus, under single-market rules, fewer
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Reaching Beyond the Market 145

constraints are put on cross-border than domestic service provision, as non-
contracted providers have access to cross-border reimbursement.

Negotiating the Patient Mobility Directive

After the failed attempt to include the provision and consumption of health
services in the general Services Directive, the Commission proposed its directive
on patient mobility in 2008 (COM (2008) 414 final). Inside the Commission,
responsibility for the directive had changed from the Directorate-General (DG)
for Internal Market to the DG for Health, which began to pursue a broader
approach to health than simply strengthening the market (Hartlapp et al. 2014:

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

110–13). The DG for Employment was also involved and was responsible for
Regulation 883/2004. The DG for Health found itself positioned between the
market-making and case-law codification approach of the DG for Internal
Market and the regulatory approach, guided by the conviction that health should
be an exception from market principles and come under the common good,
which was advanced by the DG for Employment.
Despite the fact that the DG for Health pursued a broader approach to the
Patient Mobility Directive, there were ample references to the case law in the
proposal for the directive, leaving the distinct impression that legislation was
being driven by the Court (COM (2008) 414 final). The rulings of the ECJ were
referred to twenty-two times in the proposal, and the Commission repeatedly
emphasized that there was a lack of legal certainty as to the level of reim-
bursement patients should receive for medical treatment in other member
states, which the directive had to address.
The proposal was part of a larger initiative, the programme on ‘the social
dimension of the internal market’. The interpretation of the ECJ’s case law that
the Commission put forward left member states with little control over
medical tourism. The proposal differentiates between the member state to
which the citizen is affiliated and the host member state, where the treatment
takes place. Their respective rights and duties are set out. Most importantly,
this concerns the member state of affiliation’s right of ex ante authorization,
which is limited to those cases where the financial viability of the health service
would otherwise be put at risk. The host state has the duty to ensure non-
discrimination in the treatment of EU citizens (p. 12 of the proposal, No. 12
and 13).11 The relation of the directive to existing regulations is also important.
The directive does not replace but complements the authorization procedure
that Regulation 883/2004 establishes. Should patients choose to follow the

11
In the final directive, the requirement of non-discrimination according to nationality is
weakened, so as not to increase patient waiting times (No. 21), which allows member states
measures to ensure access to healthcare in their territory (Article 4(3)).
OUP CORRECTED PROOF – FINAL, 6/12/2017, SPi

146 The ECJ and the Policy Process

latter procedure, they will be fully reimbursed for the treatment. If, in contrast,
they exercise their right to the passive freedom of services, as detailed in the case
law and the directive, the patient will be reimbursed only up to the cost of
medical treatment in their member state of affiliation.
As occurred in the other examples that have been discussed, the motivation
to reinstate legal certainty made the Commission’s proposal more restrictive
than the case law at times. In relation to Vanbraekel, the proposal required that
patients would need to be reimbursed up to the equivalent cost of treatment in
their member state of affiliation (pp. 7, 13). Further deviations of the Com-
mission’s proposal should be noted. The possibility to refer to the common
good in the authorization of non-hospital services was prohibited, while the
Court had only subjected it to adequate proof of negative financial conse-

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

quences (Krajewski 2010: 178). Furthermore, the Commission requires clear
proof that patients seeking external treatment somehow jeopardize the finan-
cial viability of the healthcare system, whereas the Court had only set out the
requirement that it should be made plausible. This amounts to a reversal of the
burden of proof, which implies that costs generally have to be reimbursed
(Krajewski 2010: 179).
Next to the codification of case law, the directive (2011/24/EU) includes
new patient rights in the realm of health services in relation to quality and
safety standards according to international best practice, as well as the right
to a complaint and redress procedure (Article 4). In the end, these additions
may prove to be of great significance (Sauter 2009: 121–2). Chapter IV of
the proposal originally provided for further supranationalization through
enhanced cooperation in healthcare, but, as it lacked sufficient support from
member states, it was dropped during the negotiations. The Council reached a
political agreement within only twenty-three months, engaging with the
presidencies of France, the Czech Republic, Sweden, and Spain. The relative
speed at which the agreement was reached, however, should not be taken to
mean that the dossier was not contentious.
As Kloka (2013) shows, it was particularly the UK, the Netherlands, France,
Sweden, and Germany that pushed for a directive. They wanted to regain legal
certainty through codification, with the additional hope that political agree-
ment would put an end to the string of cases coming before the ECJ, which
necessitated continual changes to domestic health policy. In contrast, the new
Eastern and the Southern European member states, in particular, opposed the
idea of partly liberalizing their health systems, given that the Treaty only
granted limited rights at the EU level.
Germany, for its part, was the most keen to ensure that long-term care
would not be covered by the directive. It succeeded, and the exclusion was
written into the final directive in Article 1(3). Even though such an exclusion
only means that the general rules of the Treaty apply, and there was existing
and pending case law on this issue (C-208/07 Chamier-Glisczinski), the
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Reaching Beyond the Market 147

minister at the time, Ulla Schmidt, fought fiercely for this exclusion, though
her own ministry remained sceptical (Kloka and Schmidt 2015). While this
serves as an example that the consequences of the Treaty in terms of policy are
not understood, it may be a case of successful signalling. Subsequently, the
Court argued in Case C-562/10 Commission v Germany (No. 57) that since
there was only a coordination of social systems among member states, one had
no right to the same level of treatment when moving between member states.
The Commission failed in its argument that the free movement of services
demanded the inclusion of the full portability of rights in German long-term
care regulation. For the UK, it was important that the directive did not
undermine the central role of the general practitioner (GP) in the NHS, who
decides on the rights of patients to treatment. In Recital 37 of the directive, the

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

UK received this assurance (Kloka and Schmidt 2015).
In contrast to the Western European supporters of the directive, the Eastern
and Southern member states were not subject to health-related preliminary
references. Therefore, they were less familiar with the case law, for which they
rejected erga omnes obligations. They objected to the introduction of market
principles into healthcare, denied the competence of the EU in this matter, and
also pushed to base at least parts of the proposal on Article 168 of the TFEU
pertaining to public health. In particular, they feared domestic repercussions
and large costs if they had to reimburse non-contracted service providers from
other member states. Germany, however, had managed the reimbursement
of such providers established in other member states in 2003 without such
negative consequences (Krajewski 2010: 195–6). But this question was not
only a matter of European law. National non-discrimination rules may pro-
hibit reverse discrimination against nationals and may therefore demand such
reimbursements. Given the existing case law, it was impossible to exclude non-
contracted providers from the directive altogether, so the legal service of the
Council and the Commission argued, although a majority of member states
would have preferred this step (Kloka and Schmidt 2015).
The Southern and Central Eastern European (CEE) member states that
opposed the directive could have mustered a blocking minority, as Marzena
Kloka (2013) argues, but they did not succeed in countering the older member
states that were pushing for legislation. Poland, Slovakia, Portugal, and
Romania were eventually outvoted, as they were not able to secure a com-
promise on non-contracted providers. As they were not in favour of applying
market freedoms to health services, these member states preferred to have case
law, rather than general secondary law, set out the rules. A case-law-based
regime would limit the application of the freedom of services much more than
general secondary rules that made rights clear to all (Kloka and Schmidt 2015).
In summary, the Patient Mobility Directive (2011/24/EU) is another
example in which ongoing case-law development, fuelled by preliminary
procedures, made it preferable for most member states to agree on a directive
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

148 The ECJ and the Policy Process

as a ‘lesser evil’. To leave the shaping of this policy field in the hands of the
Court risked ongoing case-law development and continued legal uncertainty.
Moreover, in the eyes of the member states, to stay inactive in an important
area was a significant reproach to the representation of ‘social Europe’, as
they were instead leaving policymaking to the Court. For instance, the
Commissioner-designate for Public Health, Androula Vassiliou, emphasized
that ‘the proposal was needed soon to avoid policy being decided by the Court
of Justice and lawyers rather than politicians.’12

Conclusion

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

The implementation deadline of the directive passed in October 2013. On the
basis of her detailed research, Martinsen (2015) finds that it has not resulted in
much cross-border demand for health services. Greer perceives there to be
insufficient support among actors for the directive’s material policy aims
(Greer 2013), resulting in minimum adaptation, leaving open the question
of how the directive will actually work. He suspects that this will give rise to
a new round of litigation, as patients are bound to sit on their bills due to the
unclear procedures for reimbursement and the reality that going to court is
their only recourse. He also expects that patient associations for rare diseases
will push case-law development further, when seeking access to new medica-
tions that are only available in a few member states (Greer 2013: 417–18).
In terms of the agreement on the directive, Martinsen downplays the
importance of the Court rulings by arguing that member states succeeded in
modifying the impact of case law through the directive (Martinsen 2015: 175,
178). However, she does not attribute much significance to the constitutional
nature of case law. By searching for instances where the Court dictates policy,
she refutes its impact, since ‘the Court pays attention to politics’ (Martinsen
2015: 180). ‘Where the “judicialisation of politics” theoretical approach
assumes politics to be the receptive part of judicial dialogues, we find that
politics also influences the judiciary’ (Martinsen 2015: 180). However, the fact
that judges are receptive to political preferences does not discount the grave
impact of constitutionalized case law on secondary law. Court rulings neces-
sarily focus on specific disputes, around which the legislature builds general
rules. Since it cannot overrule the Court or assert legislative will in place of
rules established by case law, the Court remains important. The conflict on
non-contracted providers gives strong evidence for this.
The Commission’s report from February 2014 (COM (2014) 44 final) is
particularly interesting in relation to Martinsen’s (2015) conflicting analysis

12
Summary of hearing of Androula Vassiliou, Commissioner-designate for Health, 1 Apr.
2008. REF.: 20080331IPR25292.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Reaching Beyond the Market 149

that views case law as being subject to political modification. In the report, the
Commission emphasizes the difficulty of having two parallel procedures for
reimbursing transborder treatment, according to either the established regu-
lation or the directive. As we have discussed, each allows for different sorts of
reimbursement. In accordance with the regulation, all planned treatment
needs to be authorized; in accordance with the directive, only some kinds of
treatment have this requirement. Based on the directive, all sorts of treatment
can be reimbursed, but not necessarily at full cost to the patient; based on the
regulation, the full cost is covered, but only for those medical services under
contract within the public health system (p. 7). There are additional difficulties
in relation to pensioners who live in a host country but are insured by the
home country, as for them each kind of legislation provides a different

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

solution (pp. 8–9). This shows how difficult it can be to integrate case law
into a general legislative solution.
As for the Court itself, some see a shift in its case law, which gives more
scope to member states’ regulatory authority. This is also what Martinsen
(2015: 173) argues. In cases that have been ruled upon in parallel to the final
negotiations, the Court has resisted interfering any further with member
states’ rights. As a result, it backed the need for preauthorization for the
non-hospital use of major medical equipment in Case C-512/08 Commission v
France, which was decided upon in October 2010. It also argued not to
undermine Regulation 883/2004 in Case C-211/08 Commission v Spain,
which was decided upon in June 2010. Hatzopoulos and Hervey, therefore,
identify a new restraint on the Court. They argue that in the patient mobility
cases the Court has opened the door to new rights. However, in light of
low-level demand for these rights, and given the response of the legislature
in the Patient Mobility Directive and the revision of Regulation 883/2004, the
Court now is reluctant to interfere still further in the field (Hatzopoulos and
Hervey 2013). Their analysis is taken up, for instance, by Larsson and Naurin
(2016: n. 72) as an argument for the receptiveness of the Court to member
states’ preferences.
However, in Case C-173/09 Elchinov, which was also decided upon in
October 2010, the Court followed its previous line of assuming the authority
to decide on standards of treatment. The case concerned a Bulgarian citizen
suffering from a rare form of eye cancer. Without prior authorization, he
sought out treatment in Berlin that was not available in his home country. In
Bulgaria, the decision by the lower court to grant reimbursement was over-
ruled by the higher court. Instead of succumbing to the judicial hierarchy, the
lower court turned to the ECJ, where the AG argued that this was a question
that concerned the domestic constitutional order. However, the Grand Chamber
found that lower courts were right not to abide by the rulings of their high
courts when coming into conflict with EU law. In addition to reading into EU
law that the duty of member states was to reimburse the most advanced
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

150 The ECJ and the Policy Process

treatment within the community, the Court interfered in the organization of
the member states’ judicial systems.
When there is restraint, it is important to realize that any receptiveness on
the Court’s part only concerns future case-law development. The Court has
hardly ever been seen to alter its existing case law. If the Court is seen to be
‘receptive’, it only means that it will not extend the reach of case law any
further. This more restrained position taken by the Court is also apparent in a
few cases where it was not patients but health-service providers that litigated
under EU competition law.13 Contrary to the expectation that competition law
would provide a further important lever against member states’ regulation of
their health systems, the Court refrained from taking this step (Hancher and
Sauter 2010).

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

In summary, the likely future development of case law is hard to predict.
The Court has adopted a more restrained position, a development which can
also be noted in the realm of citizenship and social rights since 2008, and
particularly since 2014. There has not been much support from patients
claiming their rights. Nevertheless, it could well be that the directive achieves
the opposite of what member states intended. They wanted to curb the further
development of case law. The directive could give way to a new wave of
litigation, as Greer expects.

FAILED CODIFICATION

With the Citizenship Directive and the Patient Mobility Directive, the EU
legislature codified case law, thereby broadening access to the rights and,
to some degree, improving legal certainty. Davies (2016a) takes these two
examples, along with the Services Directive and the regulation on the mutual
recognition of goods, which I discussed in Chapter 4, to argue provocatively
that the EU legislature is an ‘agent’ of the Court. While this term bluntly
expresses the problem that constitutionalized case law poses to the legislature,
as it needs to endorse certain policies, ‘agent’ also implies that the Court as
principal can direct it. But a legislative body is not a lone actor. What happens
if a legislative body does not command the necessary majorities?
In the following section, I will discuss two examples of such a situation. The
Monti II regulation was proposed in 2012 as a follow-up to the contentious
Laval and Viking rulings, which limited unions’ ability to revert to industrial

13
See Asklepios (T-167/04) versus the Commission concerning state aid for hospitals and
BUPA (T-289/03), in Feb. 2008, versus the Commission, again concerning state aid in Ireland
(Greer 2012: 280).
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Reaching Beyond the Market 151

action in the rulings’ requirement that measures had to be proportionate with
regard to the four freedoms. National parliaments enacted the new yellow-
card subsidiarity procedure, and the proposal was withdrawn. This case is also
discussed by Martinsen (2015), and I will therefore discuss it succinctly. The
second example is tax competition. Given the complexities of the case law that
pertains to my second example, tax competition, a discussion of this area
merits greater detail.

The Failed Monti II Regulation

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

The rulings in the Viking and Laval cases certainly count among the Court’s
most politically contentious. By arguing that the freedom of establishment and
the free movement of services also have a horizontal direct effect, these rulings
have wide repercussions (Fabbrini and Granat 2013: 127–8). Private actors
have to see that their actions can be justified under the proportionality
principle if the freedoms are interpreted in this way. As strikes directly impede
the free provision of services and/or the freedom of establishment, as had been
the situation in the Viking and Laval cases, unions, in particular, were par-
ticularly concerned by the ruling. Not surprisingly, critics of the Court’s
interpretation mounted demands to counter the Court politically so as not
to allow the unions’ power to be curtailed judicially.
Two legislative proposals were inspired by these wishes. The Posted Work-
ers Directive needed a revision, as the Court had interpreted its provisions as
providing for maximum protection, whereas the political consensus was that
Article 3 provided for minimum protection rules. I will not focus on this
process here, which would be a long case study in its own right. Martinsen
(2015) discusses the reform of the Posted Workers Directive, which resulted in
Directive 2014/67. The process is ongoing with a new proposal in 2016. Even
more interesting from our perspective of over-constitutionalization is the
attempted legislative reaction to the freedoms’ horizontal direct effect with
the proposal for a Council regulation ‘on the exercise of the right to take
collective action within the context of the freedom of establishment and the
freedom to provide services’ (COM (2012) 130 final). The proposal was named
after the former European Commissioner Mario Monti and was dubbed
‘Monti II’. The name Monti II refers to a report by Monti from 2010 and
relates to the Monti I Regulation (2679/98/EC) on the functioning of the
internal market (Fabbrini and Granat 2013: 131).
As Martinsen (2015: 199) shows, the EP had been particularly vocal in
demanding a political response to the case law in the debate about the rulings.
It actually tied its support for the re-election of Barroso to this question, and
the Commission president promised to deliver such a response in 2009. It was
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

152 The ECJ and the Policy Process

delayed until March 2012, at which point the Commission presented the
Monti II Regulation, demonstrating the difficulty of drawing up a proposal.
Within the Commission, the DG for Internal Market and the DG for
Employment struggled to strike a compromise between the constitutional
constraints of case law and the political exigencies of granting the right to
strike. This mirrored the conflict around social partners in which employers
largely welcomed and unions opposed the case-law development (Martinsen
2015: 200). The compromise, which took so long to be reached within the
Commission, however, did not calm the critics. Unsurprisingly in light of our
analysis so far, the Commission could not alter the constitutionalized case law
with secondary law. Those who made the criticism that the right to strike has
to safeguard fundamental freedoms after Viking and Laval could not find

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

consolation in the Monti II proposal.
In it, the Commission closely followed the case law. It emphasized that the
Court had recognized the right to take collective action as a fundamental right,
while it set the requirement for a proportionality test to respect the funda-
mental freedoms. However, it was precisely this ‘balancing’ that unions
opposed. The proposal also included some details on dispute resolution and
proposed a new EU-wide alert mechanism for strikes that disrupted the
internal market.
The political reaction, once again, was pronounced. Those that were
unhappy about the case law remained so (Ewing 2012). All the Commission
had done—and could only do—was codify the case law. This political debate,
which was coordinated by the Danish Parliament, resulted in the first yellow-
card procedure under the subsidiarity protocol of the Maastricht Treaty
(Martinsen 2015: 203; Fabbrini and Granat 2013: 135). Fabbrini and Granat
criticize the way in which the procedure was misused, as national parliaments
did in fact disagree with the contents of the regulation on political grounds
and did not act based on reasons that concerned subsidiarity (Fabbrini and
Granat 2013).
In September 2012, the Commission withdrew the proposal. Legal uncer-
tainty subsists concerning the limits of collective action as defined by the
Treaty (Martinsen 2015: 204). I disagree with Martinsen’s expectation that
non-adoption will ‘push the Commission to adapt its position and to present
a proposal that modifies the course of legal integration. This is a likely future
scenario for the relationship between free movement and the right to
take collective action’ (Martinsen 2015: 220). Modification, in contrast to
Martinsen’s argument, is not an option open to the legislature if primary law
is at issue. All member states can aspire to do is to give intimations of their
political consensus on the policy to the Court to inform its subsequent
decisions. It is generally uncertain whether the Court will follow the legisla-
ture as often as the Court reinterprets secondary law in light of the Treaty. By
leaving the contentious area ruled by case law, rather than generalizing
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Reaching Beyond the Market 153

the rule it establishes through codification, the subsisting legal uncertainty
may restrict the use of the market freedoms against strikes.14 Therefore, by
neither codifying case law nor generalizing the rule it establishes, the best
solution was probably reached, short of a treaty change.
It is striking that, although the cases of Viking and Laval illustrated the
danger of constitutionalized material policy positions so clearly, and the
Commission’s failed Monti II Regulation demonstrates the impossibility of
altering the Court’s interpretation politically, the limits of modification are still
not recognized. One can only support Gareth Davies’s observation that most
commentators and politicians still fail to acknowledge the problematic nature
of judicial governance in the EU (Davies 2016a: 16).

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Integrating Member States’ Tax Policies

The power to tax is a core competence of the sovereign nation state and is
crucial to its operation and existence. It is therefore not surprising that EU
member states have largely resisted moves towards integration of this crucial
policy. Harmonization measures still depend on unanimity in the Council,
and the EP is restricted to consultation rights. It is only in relation to indirect
taxes, which notably includes value added tax, that member states have agreed
on several harmonization measures. Indirect taxes have the closest relation-
ship to the single market (Genschel 2007).
Since the 1990s, an increasing number of ECJ cases have put member states’
power over direct taxation in jeopardy. Given the heterogeneous interests of
member states, such pressure from the Court has not succeeded, however, in
gaining Council agreement on the harmonization of direct taxes (Genschel
2011). This makes tax policy very interesting within the context of this book.
What happens if the Court’s case law pushes for integration but member states
do not comply?
In the following section, I will map the development of case law for direct
taxation. Given the extreme technicality of the subject and the complicated
facts in the relevant cases, this can only be an attempt to convey the basics.
I will show how court cases, often based on the preliminary procedure, have
progressively undermined the autonomy of member states to decide on direct
taxation. However, the Commission’s attempts to achieve the harmonization
of direct taxes based on this track record have failed. As Wasserfallen (2014)

14
It should be noted that Martin Höpner argues that secondary law could be used in this case
to overrule the Court, given that the right to strike is a fundamental one. Fundamental rights, in
his interpretation, allow the EU legislature to engage in full harmonization, subsequently binding
judicial interpretation (Höpner 2016).
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

154 The ECJ and the Policy Process

has shown, after the Eastern enlargement, the increasing heterogeneity of
member states makes it highly unlikely that harmonization will ever succeed.
In this situation, some have perceived a lessening in pressure from the Court,
as the Court has, in some cases, accorded the member states greater leeway in
the design of their tax systems (Genschel et al. 2011). Clearly, it is hardly
possible for the Court to take responsibility for a deteriorating ability of
member states to tax. Yet member states’ autonomy is markedly circum-
scribed by the EU’s legal interpretations.

Litigating Direct Domestic Taxes at the ECJ

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

While indirect taxes, in particular value added tax, have been harmonized
since the 1960s, member states kept control over direct taxes.15 However,
within an open economy, tax coordination becomes important. Companies
may evade taxation, or they may be taxed in both their home and their host
state, neither of which is desirable. Governments may discriminate against
companies as host states, or home governments may restrict the freedom of
establishment of their companies. To avoid double taxation in cases of trans-
border private and business transactions, there is a coordination system of
bilateral tax treaties under the general purview of the Organization for Eco-
nomic Cooperation and Development (OECD). For member states, it became
apparent in the mid-1980s that the possibility of blocking harmonization
decisions on direct taxes did not mean they would retain full sovereignty
over this area. In 1986, the ECJ decided in Case 270/83 Avoir Fiscal that the
principles of community law also apply to the direct tax systems of the
member states. It is important to note what is self-evident from the beginning
here: tax case law has ‘an inherent tax reduction bias’ (Genschel et al. 2011:
599), as litigants only sue on the basis of EU law to reduce their tax payments.
Avoir Fiscal ‘marked the date of birth of European direct tax law’ (Pistone
2008: 713). The background to the infringement procedure is illuminating. In
1977, the Commission put forward a proposal for company taxation in order
to eliminate discriminatory practices among the member states. The Council,
however, rejected a negotiation of the proposal. Sometime later, the Commis-
sion attempted to put judicial pressure on the member states (Genschel 2002:
176–89, 267). Thus, we find another example here of the Commission’s ability
to combine its legislative functions with its capacity to initiate infringement
procedures, as discussed in Chapter 4. The Commission pursued Case Avoir

15
I am very thankful that I was able to draw on a report on tax case law prepared by Tilman
Krüger within the framework of our project. Special thanks also go to Philipp Genschel for
helping me with his expertise.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Reaching Beyond the Market 155

Fiscal against France for not giving shareholder tax credit, which was granted
to French companies, to French branches of insurance companies established in
other member states. This discriminated against these companies and violated
the freedom of establishment (O’Shea 2008: 261). France argued that, as long
as there was no tax harmonization, the distinction between residents and non-
residents that underlies international tax treaties could also be upheld in the
European Community. The Court rejected this argument (O’Shea 2008: 262),
setting out the requirement that France, as a host state, had to treat branches
of French and other EU companies alike. In subsequent case law, the Court
confirmed this position. As O’Shea (2008: 270) notes, the Court applies a
national treatment test in the different cases that relate to discrimination
against the host state, which requires that national and EU companies have

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

to be treated alike.
Another set of cases concerns origin or home member states that hinder
their companies’ freedom of establishment. This series of cases begins with
Case 81/87 Daily Mail in 1988 (O’Shea 2008: 271–5). The Daily Mail company
wanted to transfer its seat from the UK to Luxembourg for tax reasons, but the
Treasury refused to grant the necessary permit for this move. The company
considered its freedom of establishment to have been violated, and a prelim-
inary ruling followed. The Court did not uphold all the company’s arguments.
While member states could not restrict the freedom of establishment, the lack
of harmonization had to be taken into account. The UK could set its own rules
for companies that want to relocate their headquarters while preserving their
legal personality as a UK company (O’Shea 2008: 271). The Marks & Spencer
ruling, which I will discuss later as a crucial turning point in the case law that
gave more leeway to the member states, also belongs in this category. The case
concerned the rules for group relief that applied to domestic subsidiaries of
UK companies and were not granted to subsidiaries of other member states.
Another case, C-471/04 Keller Holdings, which was decided upon in 2006,
dealt with the German imputation system. It treated domestic indirect sub-
sidiaries and those in other member states, in this case Austria, differently.
Germany justified the difference by arguing that only the dividends of indirect
domestic subsidiaries were taxed in Germany, so the same tax advantage could
not be given to indirect subsidiaries in other member states that paid their
taxes elsewhere. However, the ECJ rejected this argument, as it has done in
other cases, arguing that it does not matter where an indirect subsidiary is
established (O’Shea 2008: 273–4).
The essence of these arguments is that they undermine the link between
taxation and territory. This is a radical step for nation states whose existence
depends on fiscal autonomy over a defined territory. Taken as a whole, the
ECJ’s case law has tackled different discriminatory practices. It has ensured that:
non-resident EU and resident companies have to be treated equally; personal
allowances for non-resident EU citizens have to be the same as those for
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

156 The ECJ and the Policy Process

residents; deductions have to be made for contributions paid to non-resident
EU pension funds, as well as to interest payments to non-resident EU companies;
and group-relief arrangements have to include non-resident EU companies
(De Goede 2003: 206–7).
O’Shea (2008: 268) argues that the Court’s case law has led to a ‘mantra’
with which the Court justifies its competence over direct taxation, despite it
being formally within the member states’ jurisdiction: ‘although direct tax-
ation falls within the competence of the Member States, the latter must none
the less exercise that competence consistently with Community law and
therefore avoid any overt or covert discrimination on grounds of nationality’
(Case C-250/95 Futura, No. 19). This mantra reflects what we know from
other areas, such as health, where the Court also holds that single-market

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

principles have to be respected, although member states have kept certain
competences exclusively for themselves.
The early court cases concerned the principles of company taxation, which
had to be aligned with the freedom of establishment and later with the free
movement of capital. Following the move to the single market, companies
were motivated to use member states’ heterogeneous systems of taxation to
their advantage. After a delay of a few years, individual litigants also chal-
lenged their tax bills, with the help of EU law, based on the free movement of
workers. Central here is Case C-279/93 Schumacker. In fact, Schumacker is
one of the cases the Court itself cites most, as Derlén and Lindholm show
in their network analysis (Derlén and Lindholm 2013: 673). In this case, a
non-resident worker earned almost all his income in Germany. This barred
him from applying for exemption from taxes in his country of residence,
Belgium. Workers who do not live and work across borders consequently
received more tax benefits. International tax law grants the responsibility
for tax exemptions to the country of residence. However, if hardly any taxes
are paid there, there is no scope for exemptions. Similar to the assessment
of company taxation, the Court analyses whether the principle of non-
discrimination or the fundamental freedoms are violated by personal tax
rules and whether there are sufficient justifications for restrictions (Malherbe
et al. 2011: 25–6). As O’Shea (2008) shows, the case law concerning personal
taxation can also be divided along lines of discrimination from the perspective
of the host or home state.
In general, companies are much better placed to exploit the overlapping EU
legal order than private individual actors, given the significant expenditure of
time and money necessary to pursue preliminary reference proceedings.
Nevertheless, private actors have also instigated important court cases,
which have partly led to significant tax losses for member states. In addition
to Schumacker, Cases C-292/04 Meilicke I and C-262/09 II of 2007 and 2011
are noteworthy, the German response to which will be analysed in Chapter 6.
The issue here was the old imputation system of company taxation on
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Reaching Beyond the Market 157

dividends that Germany only granted for German shares.16 Because of the
general retroactive application of case law—after all, the Court only elucidates
what EU law has required all along—rulings may lead to retroactive entitle-
ments. In the Meilicke cases, which built on the free movement of capital that
had become directly effective in 1988, the German government feared signifi-
cant tax loss because of retroactive claims.
More recently, the Commission has started to analyse tax distortions that
prove to be disincentives for cross-border workers (Agence Europe, 3 April
2012, 20 January 2014), which may mean that the Commission will also put
greater emphasis in the future on pursuing infringement procedures related to
personal direct tax distortions.
To date, the ECJ has decided more than one hundred cases involving income tax

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

issues, with the vast majority striking down member states’ tax provisions on the
ground that they violate either one of the four freedoms guaranteed by the treaties
or the treaties’ bar against discrimination on the basis of nationality.
(Graetz and Warren Jr 2006: 1186)
In their analysis of these cases, Graetz and Warren Jr explicitly reject the idea
that case law can be explained through the Court’s alignment with majoritar-
ian activism, as Maduro’s theoretical argument (1998) holds (Graetz and
Warren Jr 2006: 1193). The Court does not push for harmonization according
to the preferences of the majority of member states. Time and again, the Court
has rejected arguments about budgetary concerns, such as revenue loss or an
erosion of the tax base (Cordewener et al. 2009: 1957–63).17 Member states
can only pursue their domestic taxation priorities when they are a justifiable
exception to the freedoms, and the Court defines these exceptions narrowly.
This results in a potentially devastating effect of case law. The familiar
arguments about the application of the Treaty in other policy fields are
relevant here as well:
According to the Court of Justice, overt discrimination may be justified by those
grounds set out explicitly in the EU Treaty (such as public policy, public security
and public health) whereas a restrictive measure is permissible ‘only if it pursues a
legitimate objective compatible with the Treaty and is justified by imperative
reasons in the public interest’. Furthermore, it must ‘not go beyond what is
necessary to attain the objective pursued’. (Malherbe et al. 2011: 21, No. 25)

16
Traditionally, member states avoided the problem of double taxation through imputation
systems, where investors, in addition to their dividends, receive a tax credit for the taxes the
company has paid. Given the case law of the Court, these systems were replaced with dividend
exclusions, where the dividends received are exempted from taxation (Graetz and Warren Jr
2006: 1208–12; Graetz and Warren Jr 2007: 1591).
17
Relevant cases here are: C-264/96, ICI C-307/97 Saint-Gobain; C-35/98 Verkooijen; Joined
Cases C-397/98 and C-410/98 Metallgesellschaft and Hoechst; C-136/00 Danner; C-385/00 De
Groot; C-319/02 Manninen; and C-196/04 Cadbury Schweppes.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

158 The ECJ and the Policy Process

This broad interpretation disregards member states’ preservation of their
prerogative for direct taxation in the EU treaties. The gist of the ECJ’s case
law until the mid-2000s can be summarized as the active and gradual removal
of discriminatory tax measures by making references to the fundamental
freedoms (Genschel et al. 2011: 600; Cordewener et al. 2009: 1951–2; Aujean
2007: 330; Pistone 2008: 713).

An Impossible Quest?

Considering the explicit wish of the masters of the Treaty to keep tax policy
under national control, it is not only the legitimacy of the Court’s case law that

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

needs questioning. Fundamentally, the Court lacks a reference tax system for
the pursuit of its non-discrimination approach. In order to avoid double
taxation, as well as non-taxation, controlling for non-discrimination has to
use the perspective of either the home or the host state. Graetz and Warren Jr
(2007) make this point.
We have shown here that the ECJ’s nondiscrimination jurisprudence reveals an
impossible quest: to eliminate discrimination based on both the origin and
destination of economic activity. We have also shown that this quest necessarily
must fail in the absence of harmonized corporate income tax bases and rates
among EU member states. This implies that the court will find it necessary
somewhere along the way to retreat, creating not only legal uncertainty, but
ultimately doctrinal incoherence. (Graetz and Warren Jr 2007: 1253)
As the authors explain, tax neutrality can be realized as capital import neutrality,
where all economic activity within the country is taxed the same, regardless of
whether the capital is domestic or foreign. Alternatively, capital export neutrality
can be achieved by taxing the income of residents in the same way, whether it is
earned at home or abroad. As long as tax rates differ among countries, both kinds
of neutrality cannot be achieved at the same time. Thus, the Court would need to
decide on one principle over the other, but it has no basis for this choice (Graetz
and Warren Jr 2007: 1582). As the authors show, this problem of the Court’s lack
of a reference point leads to some significant contradictions between rulings,
where the Court decided in one or the other way.18
Similarly, Pantazatou (2013: 95) argues that member states’ tax systems
include few directly discriminatory measures, but that discrimination arises

18
For instance, they mention Denkavit and ACT Test Claimants (Graetz and Warren Jr 2007:
1615). The criticism of the lack of a reference system was taken up by the Advocate General in a
proceeding against the annulment of a Commission decision on state-aid allegations against
Gibraltar (C-106/09). The Court did not follow the argument, however. (EU) EU/CJEU: ‘Advo-
cate General proposes to uphold 2004 Gibraltar corporate tax ruling’, Brussels, 7 Apr. 2011
(Agence Europe).
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Reaching Beyond the Market 159

from the interaction of two different systems. Critically, Graetz and Warren Jr
argue that this tax case law means that the EU is becoming ‘a place where
corporate income can easily escape tax. . . . Thus, the ECJ decisions raise the
possibility of a United States–European race to the bottom in corporate
income taxation’ (Graetz and Warren Jr 2006: 1252–3).
In addition, the Commission foresaw the shortcomings of a court-driven
European tax policy, despite its general interest in taking advantage of the
momentum generated by case law to foster integration. Frits Bolkestein,
Internal Market Commissioner from 1999 to 2004, complained that ‘company
tax law must not be made in court’ and that the ECJ’s ‘taking over the role of
lawmaker on crucial tax issues in Europe, not least in the area of company
taxation’ was unacceptable (Financial Times, 21 October 2003). But of course,

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

the Commission itself was not innocent of the Court’s activities. After the
Court had made inroads through case law into member states’ tax sovereignty
on the basis of preliminary rulings, there was a ‘considerable increase in the
number of infringement procedures . . . in the field of direct taxes’ (Pistone
2008: 727) from the 1990s.
The Commission re-emphasized its willingness to put member states under
pressure through infringement procedures in 2001 (COM (2001) 260 final,
pp. 21–2). Even after the Court became more careful in 2005 (see next
section), the Commission did not refrain from using infringement procedures
against member states in the field of direct taxation. Fifty-six infringement
cases were completed up until March 2016. Figures 5.1 and 5.2, which are
based on the Commission’s report on important ECJ cases in the area of direct
taxation,19 show when and where the Commission has intervened, putting
several member states under pressure, some of them repeatedly. In this way,
the Commission partly compensates for the uneven development of prelim-
inary proceedings across the EU, within which some domestic courts refer
frequently and others not at all. Pantazatou (2013: 115) even argues that courts
in those member states that rely on tax competition, such as Ireland or Cyprus,
were careful not to undermine the business model of their national economy
by presenting a preliminary reference to the ECJ. Figure 5.2 differentiates tax
cases according to member states and shows that this is not the case for
Luxembourg. However, Pantazatou also mentions that the courts of the
member states that were hit by fiscal debt (the GIPSI countries: Greece,
Ireland, Portugal, Spain, and Italy) did not respond by changing their refer-
encing behaviour (Pantazatou 2013: 116). Thus, with respect to tax competi-
tion, member states’ courts should not be assumed to simply follow the
‘national’ interest.

19
Date of document: 9 Mar. 2016, on file with author. The current version can be found
under: <http://ec.europa.eu/taxation_customs/infringements/eu-court-justice-case-law_en>
[accessed 9 May 2017].
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

160 The ECJ and the Policy Process

30

4
25 7
7
9
3
20 6
3 4

2
15 4

10 3

5 1
1

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

1
1 2 2 2 3 9 6 3 8 7 6 5 15 22 18 13 13 13 16 10 17 14
1 1 1 1 3 1
0
1986
1987
1988
1990
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
Preliminary ruling Infringement proceedings

Figure 5.1. Infringement cases and preliminary rulings on direct taxation (by year)

70
7
60

50

40 10
2

30
3
20
2
2 4 1 6 1
2 55 9
10 28 32 4
3 2 2
9 13 9 6 10 17
8 1 7 1
2 3 1 3 1
0
Be ria
D ium

Es rk

Fi ia

Fr d
G nce

G y
H ce

ry

m y
he rg

Po s
Po nd

Sl gal

ia

Sw n
en

K
xe Ital
an

nd
an

ai

U
n

en
a

ga

et ou
e

la

ed
st

rtu
to

Sp
m

m
lg

re

rla
nl

ov
Au

un

b
en

er

Lu

Preliminary ruling Infringement proceedings

Figure 5.2. Infringement cases and preliminary rulings on direct taxation (by
member state)
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Reaching Beyond the Market 161

The Court’s Response—Switching to a Lower Gear

Despite the pressure the case law places on member states’ tax systems, and
the inherent contradictions of the case law, member states did not manage to
agree on secondary law as a means to intervene in the case-law development.
The Court thus found itself in a difficult situation. It had instigated the
application of EU law to member states’ direct tax systems—a process
which gathered significant momentum, fuelled by the interests of mainly
corporate litigants and their ability to use the preliminary procedure. While
the Court had to deliver judgments on the cases brought to its attention,
it did not find any guidance in secondary law that could have legitimated
its approach.

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

In this situation, the Court took up a strategy that was familiar from other
areas and engaged in the fine-tuning of its case law, which has been inter-
preted by some as a break from the previous line of reasoning. Its judgment
in Case C-446/03 Marks & Spencer (13 Dec. 2005) is regarded in the
literature as the turning point, after which the Court began to grant member
states greater leeway in deciding on their tax policy (Tiedtke and Mohr 2008:
428; Sutter 2006: n. 2). The company Marks & Spencer had aimed to deduct
losses incurred through discontinued subsidiaries in France, Germany, and
Belgium from its UK tax. However, the British group-relief rules viewed
this possibility as only being for domestic subsidiaries. This violated the
freedom of establishment in the Court’s opinion, but it conceded that
EU member states could, in principle, exclude the deductibility of losses
incurred by subsidiaries in other member states (No. 59). It even held that
the non-deductibility of losses could be the rule rather than the exception
(Lang 2006: 54). It added a balanced allocation of the power to tax (No. 43–46)
to its ‘overriding reasons in the public interest’ (No. 51), avoiding the risk that
losses might be deducted twice (No. 47–8) or taxes avoided altogether (No. 49).
However, measures had to be proportionate and the deductibility of losses could
not be refused if there was no other way to take them into account (No. 55).
By broadening the scope of legitimate objectives, the Court diverted from the
opinion of AG Maduro, and it surprised EU tax law experts (Lang 2006: 54). EU
finance ministers were relieved that the Court had awarded them some more
legitimate reasons for protecting their tax base (Jochum 2006: 621), but the final
losses of EU foreign subsidiaries had to be offset as a tax exemption for the
income of domestic parent companies, though these subsidiaries could not be
taxed. ‘The big difference, of course, is that the domestic subsidiaries are subject
to tax at the same rate as the parent, while the foreign subsidiaries can be in
Estonia, where there is no corporate tax, or in Ireland, where the tax rate is only
12,5%’ (Avi-Yonah et al. 2007: 468).
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

162 The ECJ and the Policy Process

After Marks & Spencer, other rulings have confirmed the Court’s revised
position (Lang 2009).20 ‘The “outright activism” of the years 1995 to 2004 gave
way to a certain deference to the fiscal sovereignty of the Member States (since
2005)’ (Wattel 2008: 205). Member states lost fewer cases (Genschel et al.
2011). But where exactly does the fiscal sovereignty of the member states end,
and what are the limits of the freedom of establishment and the free move-
ment of capital? In particular, questions of cross-border loss relief or the
deductibility of foreign losses, incurred by a company in the discontinued
operation of subsidiaries and permanent establishments in other EU member
states, were subject to many decisions. Thus, it is unclear when losses have to
be regarded as final to become deductible (Jochum 2006: 622; BDI and PWC
2011). Secondly, the Marks & Spencer decision contains three justifications

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

that member states have to invoke (the balanced allocation of the power to tax,
the risk of a renewed deduction of losses, and tax avoidance), which raises the
question whether all justifications need to be present simultaneously. Subse-
quent case law gave further guidance. In Case C-231/05 Oy AA (No. 44-60),
the ECJ found that two justifications were needed to restrict freedom of
establishment legally, namely the balanced allocation of tax jurisdiction and
a risk of tax avoidance. In Case C-414/06 Lidl Belgium (No. 38-42), the ECJ
required again only two justifications: next to the balanced allocation of tax
jurisdiction, the risk of a renewed deduction of losses. In Case C-337/08 X
Holding (No. 33), the ECJ found that the balanced allocation of tax jurisdiction
alone was sufficient to justify the discriminatory measure. As a study of the EP
shows, it becomes apparent again that policy guided by case law implies
significant legal uncertainty (Malherbe et al. 2011: 21, No. 25).
Other open issues can be mentioned here, such as the question of whether
subsidiaries and permanent establishments have to be treated alike (Lang
2009: 98–106). The result was a constant stream of cases, which, however,
has produced ‘an intractable ball of unacceptably inconsistent case law, as the
court was regularly backing out—without saying so—of consequences of its
previous vigorous case law’ (Wattel 2008: 205). For private litigants, the
ambiguities of case law provide sufficient incentives to go to court, as the
continuous flow of preliminary rulings shows (Figure 5.1), but for member
states legal uncertainty reigns over the limits of their fiscal sovereignty. The
extent to which the fundamental freedoms have eroded their ability to tax—
and the scope companies have to avoid taxation—create an urgent need to
establish common rules. The ECJ’s tax cases ‘substantially inhibit the flexibility
of EU Member States to address the vexing issues of multiple corporate taxation,
economic taxation, and international double taxation. It is impossible, in our

20
These include: Lidl Belgium and C-157/07 Krankenheim Ruhesitz am Wannsee on subsid-
iaries; Oy AA and X Holding on permanent establishments; and even judgments on corporate
income tax (such as C-418/07 Papillon).
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Reaching Beyond the Market 163

view, to identify consistent tax policy norms that would explain these results’
(Graetz and Warren Jr 2007: 1618).

Legislative Responses at the EU Level

Political agreement on tax coordination is highly difficult to achieve, as is

apparent from the persistence of the unanimity rule in the Council on tax
matters. For member states, their fiscal autonomy is too central to their func-
tioning for them to give up their veto power (Genschel and Jachtenfuchs 2011).
Nevertheless, there is some secondary law on direct taxation. Instruments of

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

positive integration that have been adopted include the code of conduct for
business taxation (1997), the Parent-Subsidiary Directive (1990/2003), the Fiscal
Merger Directive (1992/2006), the Interest and Royalty Directive (2003), and
the Savings Directive (2003). While these steps are limited in scope—they
typically deal with the balancing of tax jurisdictions and with certain tax
avoidance practices—they indicate that national rules on direct taxation have
an increasingly visible European dimension. It is widely recognized in the
literature that the pressure on the legislature emanates from the case law. ‘It is
widely believed that the ECJ ruled the way it did in order to force the political
branches of the EU to move toward corporate tax rate harmonization, as the
Commission has advocated (to no avail) for many years. But here the ECJ can
learn a lesson from the US Supreme Court: deciding cases in order to force
action by the legislature can be dangerous’ (Avi-Yonah et al. 2007: 468–9).
Given the issue of case law undermining the tax sovereignty of member
states, while lack of agreement prohibits unanimous decisions on a common
tax policy, the Commission drew up plans to harmonize the corporate tax base
in the 2000s, as a way to provide a minimum level of harmonization. Member
states would still be free to set their tax rates, but the tax base, namely the
kind of corporate profits to be taxed, would be harmonized. The Common
Consolidated Corporate Tax Base (CCCTB) has been the most significant
project carried out by the Commission. The scheme aims to avoid discrimi-
nation, double taxation, and non-taxation, while lowering the compliance
costs of firms that are subject to different tax jurisdictions (Agence Europe,
20 December 2006). According to the scheme, companies that do business in
more than one member state can opt to apply a single set of tax rules and keep
up contact with a single tax authority. Losses and profits made in different
member states can be balanced. Goods transferred between different subsid-
iaries no longer need to be priced separately. Double taxation and adminis-
trative burdens can be avoided (Agence Europe, 10 February 2011). What
sounds convincing here may, however, have significant problems of incen-
tivization among member states. If a single member state administers tax
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

164 The ECJ and the Policy Process

collection on the behalf of others, there is the danger of a ‘competition in
laxity’ (Genschel 2007).
The Commission first envisaged publishing such a proposal in 2006, with
the aim of making it effective in 2008, based on a communication on company
taxation from October 2001. However, it was already known then that the UK,
Ireland, Slovakia, Malta, Lithuania, Latvia, and Estonia objected to the plan
(Euractiv, 4 April 2006). To cut a long story short, a decade later the issue is
still on the agenda. In the meantime, the Commissioner responsible for this
area repeatedly attempted to gain the support of the Economic and Financial
Affairs (ECOFIN) Council.
Parallel to the debate on harmonizing the tax base, the Commission devel-
oped plans from 2001 onwards for the experimental use of ‘home-state tax-

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

ation’ for small- and medium-sized enterprises (SMEs), as for them the tax
issues that trans-border transactions entailed often proved prohibitive. The
idea here was that the company calculates its taxes according to the tax-base
rules of its home country, and the taxes are then paid to the host country
according its tax rate. SMEs and member states were invited to opt for this pilot
scheme, based on a bi- or multilateral treaty among member states (Agence
Europe, 11 January 2006). However, member states did not take up this idea.
The financial crisis of 2008 and its aftermath appeared simultaneously as an
opportunity to advance harmonization and to bring more pressing issues than
the CCCTB onto the ECOFIN Council’s agenda.21 Ireland was put under
pressure by some member states to raise its corporate taxes of 12.5 per cent as
part of the bailout—but it successfully resisted (Agence Europe, 15 March
2011). France and Germany in particular aimed to advance the issue of tax
harmonization, engaging in bilateral plans to bring about a common corporate
tax beginning in January 2013 (Bundesministerium 2012b). The change in the
French government after the election of President Hollande brought this to a
halt. The Commission published its ‘Proposal for a Council Directive on a
Common Consolidated Corporate Tax Base (CCCTB)’ (COM (2011) 121/4)
in 2011, but the proposal stalled.
A more recent priority for the Commission has been the fight against tax
evasion at EU and member-state levels (COM (2015) 302 final). After a long
period of fostering tax competition, this may seem surprising. Nevertheless, the
financial crisis and the large extent of VAT ‘carousel fraud’, which reached new
heights after the Eastern enlargement, has shifted the Commission’s attention.22
The Commission realized that the member states’ insufficient fiscal revenues are

21
Euractiv, 7 Jan. 2010, ‘Parliament Sees Crisis as “Opportunity” for Tax Harmonisation’,
<www.euractiv.com/innovation-enterprise/parliament-sees-crisis-opportuni-news-223363>
[accessed 10 May 2017].
22
Euractiv, 12 Sept. 2012, ‘Šemeta: Fiscal Union Involves Tax Cooperation’, <http://www.
euractiv.com/section/euro-finance/interview/emeta-fiscal-union-involves-tax-cooperation>
[accessed 10 May 2017].
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Reaching Beyond the Market 165

also an EU problem. Accordingly, the Commission began an initiative to fight
letter-box companies which are established solely to save taxes. Such an active
approach to the ‘abuse’ of the freedoms is new,23 as the Commission and the
Court have for a long time predominantly argued against the view that the
freedoms can also be abused (Engsig Sorensen 2006; Kjellgren 2002; Schammo
2008). Reports in late 2013 on large multinational corporations’ tax evasion in
the EU, by the likes of Apple, Google, Starbucks, or Amazon, renewed interest in
the CCCTB.24 Interestingly, these discussions moved into different arenas, as the
OECD and the G20 also took up the issue. Tax competition in the EU not only
impacts EU member states with high taxation. The US has the highest corporate
tax rates of any industrialized country (Genschel and Schwarz 2011) and is
therefore highly concerned by the judicial facilitation of tax evasion in the

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

EU. At the same time, the multinational companies whose tax practices were
critically discussed in the EU operate globally.
Consequently, from 2013 onwards, the talks at the OECD on base erosion
and profit shifting (BEPS) dominated supranational attempts to harmonize
tax policy, resulting in concrete policy recommendations in 2015. It is inter-
esting for our context that the constitutionalized case law of the ECJ was also
significant for the discussion of policy recommendations in these OECD talks.
To the surprise of non-EU OECD members, in some of the action areas, policy
options that seemed desirable to other participants could not be considered,
given that countervailing ECJ case law does not permit it (interview German
Federal Ministry of Finance, 28 September 2015). The serious constraints case
law poses were particularly apparent in the discussion of BEPS Action 3 on
strengthening controlled foreign company (CFC) rules (OECD 2015) and
BEPS Action 6 on preventing treaty abuse (Kemmeren 2014). It is interesting
that, in addition to constraints on policymaking at the European and domestic
levels, the constitutionalization of policies is now also making itself felt in
international negotiations. Since global tax evasion is a serious problem, it has
also become a concern beyond the EU that member states have lost a great
deal of policy options.

Conclusion: No Exit from the Joint-Decision Trap

Through its case law, the Court can introduce powerful incentives for legislation
at the EU level through which deadlock in decision-making—the joint-decision

23
Euractiv, 26 Nov. 2013, ‘Commission Starts Fight Against “Letter-Box” Companies’,
<http://www.euractiv.com/section/innovation-industry/news/commission-starts-fight-against-
letter-box-companies> [accessed 10 May 2017].
24
Euractiv, 19 Nov. 2013, ‘Furore Over Tax Evasion Opens Door to New EU Proposal on
Corporate Tax’, <http://www.euractiv.com/section/innovation-industry/news/furore-over-tax-
evasion-opens-door-to-new-eu-proposal-on-corporate-tax> [accessed 10 May 2017].
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

166 The ECJ and the Policy Process

trap—can be overcome (Falkner 2011). But, in terms of taxation, pressure from
the Court is not sufficient for the alignment of the preferences of an increasingly
diverse set of member states, which are different sizes and are at very different
stages of economic development. The Court’s far-reaching case law to foster tax
competition, therefore, has not led to an agreement on tax harmonization. In
view of legislative deadlock, the Court started granting more leeway to member
states, beginning with its Marks & Spencer ruling. However, this about-turn does
not mean that the problem of case law interfering with member states’ tax
policies has vanished. Many options that were previously available to member
states through tax policies have now been foreclosed. The constraints of consti-
tutionalized case law even hamper international agreements, and the high
number of preliminary references shows that, notwithstanding the Court’s

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

more cautious position, private actors are still motivated to litigate in order to
lessen their tax bills.
In addition, the implications of the general retroactive effect of case law are
particularly striking in this area. If the basis of tax policy can be altered
retroactively, enormous sums may be at stake. In Case C-475/03 Banca
Populare, €120 billion were claimed to be in question (Pantazatou 2013:
107–8). Of course, member states will overstate the damage incurred in
court proceedings, but it is also then not entirely possible for the Court to
take political responsibility for such sums of money.
Member states thus find themselves in a predicament. They cannot agree on
harmonization, but nor can they live well under the constraints of case law.
Much, therefore, depends on the leeway that domestic courts allow the
executive, by referring or not referring cases, and whether the domestic
administration follows all the rulings. We will turn to the crucial member-
state level in the next chapter. It is interesting that, in this situation, pressure
from international organizations like the OECD and the G20, which are much
more loosely institutionalized than the EU, becomes relevant. Given the
dilemma of judicial pressure and the political inability to act, tax policy may
be an area where alternative policy measures need to be sought out. In view of
the limited policy options available under case-law development, the more
recent attempts to find a solution appear to have led to placing transparency
obligations on multinationals for the paying of taxes in different jurisdictions
and placing those same obligations on governments for tax deals that have
been granted to them.

CONCLUSIO N

The cases in this chapter have shown how the market freedoms encroach on
policies that member states have attempted to keep separate from the EU
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Reaching Beyond the Market 167

through special provisions in the Treaties. The fundamental freedoms are
interpreted broadly and are enforced by the Court, which is also the case in
areas where member states have retained competences. To this end, the Court
repeatedly reverts to formulations such as ‘the powers retained by the Member
States must nevertheless be exercised consistently with Community law’
(Schumacker C-279/93 No. 21), which is called the ECJ’s ‘retained powers
formula’ by Azoulai (2011). ‘[A]lthough direct taxation falls within their
competence, the Member States must none the less exercise that competence
consistently with Community law’ (C-374/04, No. 36).
In the area of citizenship and patient mobility, the Court has additionally
repeatedly disregarded secondary law, as decided by the Council and the
EP. In relation to the free movement of labour in particular, it should not be

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

forgotten that the Court has time and again intervened against forms of
discrimination that have arisen against EU citizens working in other member
states. The Court, therefore, provides a necessary corrective. However, the
extent of reverse discrimination against nationals demonstrates that the Court
continuously seeks to broaden rights as a way of fostering integration. In doing
so, it demands a ‘certain amount of financial solidarity’ from the member
states. Can this be taken to imply that a re-embedding of markets has been
taking place, as Caporaso and Tarrow hold? In such a process, the Court
fosters social policy that member states would not agree upon, given the
significant heterogeneity of their interests (Caporaso and Tarrow 2009). As
Höpner and Schäfer (2012: 447) rightly point out, the Court neglects the
necessary reciprocity between benefits and duties. It is likely, therefore, that
the extended access to social services will result in a cutback in their provision.
In negotiating the directive, the member states’ governments were not willing
to yield to the Court in the way they had done in relation to the Services
Directive and the mutual recognition of goods regulation, which were dis-
cussed in the last chapter. Therefore, they left the rights of non-economically
active EU citizens to access social benefits deliberately open in the directive.
They opted for continued legal uncertainty rather than codification.
This chapter not only confirmed that taking a re-embedding perspective
on the ECJ is overly optimistic the Patient Mobility Directive also showed how
difficult it can be to codify case law, even if there is a sufficient majority. The
new legal status quo raises many questions and falls far short of a desired level
of legal certainty. The Patient Mobility Directive is also interesting, as it is the
only case I have discussed in which the new member states had a particular
position. They had not been targeted by case law at the time, in contrast to
the old member states that were pushing for the directive. They objected to the
right of the EU to interfere in this policy area. Some of them denied that case
law has erga omnes effects. From what we know from the old member states,
who also took time to understand the implications of direct effect and
supremacy, it is plausible that the new member states differ in their cognizance
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

168 The ECJ and the Policy Process

of case law. However, the preference to keep the polity reined in through case
law is a rational strategy for containing its impact, as secondary law generalizes
the effects of case law. We could therefore hypothesize that member states that
oppose the direction of case-law development will resist its codification—in
particular, if they do not trust the Court to be attentive to member states’
signals. From the divergent assessments of EU law due to its broad expansion
in membership in recent years, we can hypothesize that there are additional
incentives for codification. The more member states differ in whether they
follow the supreme and directly effective case law, the more it ought to be in
the interest of those member states that adhere to case law to opt for codifi-
cation, as a way to ensure a level playing field.
The failed Monti II regulations and the taxation case, finally, show the limits

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

of pressure applied by the Court. Though case law set incentives for codifica-
tion, member states could not agree on how to respond. In the case of Monti II,
the frustration that secondary law could make a difference to constitutionalized
case law resulted in failure. Unions and all those actors who disagreed with
Laval and Viking preferred to live with case law than with codification that set
out these constraints on collective action openly. For taxation, the Court
backed down somewhat in 2005, in an about-turn that possibly amounted to
reversion, lessening the legal pressure. However, the continued flow of prelim-
inary references shows that private actors still have reason to expect a lessening
of their tax bills with the help of the Court. In this situation where tax
competition provokes more and more political contestation, we see repeated
attempts to cope with this state of affairs, for instance by the Commission
requesting that the member states’ governments publicize their tax deals with
large companies or by demanding that multinationals make how much tax they
pay transparent in each of the member states. Reversion would set in were the
Court to grant member states leeway that had been taken away before.
Nevertheless, the situation is puzzling. On the one hand, it has been claimed
that single judgments by the Court can cost member states more than the
bailout of Ireland (Pantazatou 2013: 78). On the other hand, collective action
through legislation will fail. Secondary law can give the Court indications of
member states’ policy priorities, but it cannot divert from case law that
interprets the Treaty. Moreover, the Court often does not follow secondary
law but applies the Treaty, as was apparent in the case law on patient mobility
where a regulation that set out a policy existed. Therefore, there is also a
rationale to go along with case law that is not codified and generalized. It is
puzzling that the far-reaching constraints of constitutionalized case law are
not discussed more broadly in political terms. However, we have not yet
analysed how member states react to case law in domestic politics when
codification is absent. It may be the case that it does not impose many constraints
in this arena with implications for their preferences on EU legislation. The
next two chapters will focus on this question.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Europeanization With
and Against the Odds
The Cases of Meilicke and Zambrano

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

The European Court of Justice’s (ECJ) case law, as we have seen in the
previous chapters, can deeply influence the course of European legislation,
or even replace it. However, the European legislator does not necessarily
respond, as the example of taxation showed. In addition to the European
level, European case law may have significant domestic effects. By interpreting
European law as supreme and having direct effect, and by attributing broad
meaning to its rules, such as the four fundamental freedoms (of goods,
services, persons, and capital), the Court imposes potentially significant con-
straints on the possibilities for member states to govern their economies and
welfare systems.
Because of its constitutional nature, European case law is supposed to take
immediate effect in the member states. Domestic courts must apply EU law
directly if its meaning is beyond doubt (acte clair doctrine) or once its
implications have been settled through ECJ case law (acte éclairé), without
beginning another preliminary procedure. Established member states’ policies
are therefore vulnerable to becoming incompatible with European law in the
course of ongoing case-law development. From this perspective, given the
extent of competences that have been transferred to the EU, it has become
increasingly difficult for member states to use their remaining national com-
petences (Scharpf 2009). In cases of dispute, what remains of national auton-
omy to act depends on the ECJ (Everling 2003: 878).
However, courts are ‘reactive’ actors (Conant 2012), dependent on the
implementation of their rulings. Member states could simply insist that they
are not concerned. They may argue that case law only relates to specific cases
and has no generalized erga omnes effect, as it is only valid inter partes for the
parties involved in the judicial dispute. If domestic courts turn a blind eye—and
member-state governments, administrations, and parliaments do not react—
the impact of case law will be small if the Commission does not apply pressure.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

170 The ECJ and the Policy Process

Research on Europeanization analyses how integration affects member states.
However, it focuses predominantly on the implementation of European sec-
ondary law. It mirrors the research on European integration processes as both
share a tacit assumption that integration only proceeds when member states and
the EU’s political institutions explicitly agree to pool policies. Similar to the
European level, the impact of the Court is also neglected at the member-state
level. Such a narrow focus could be entirely suitable for Europeanization
research should far-reaching case law only be significant when it is codified in
secondary law. This argument is advanced by Wasserfallen:
When we take implementation problems and contained compliance seriously, it
becomes evident that the potential of judge-made law is limited. Ipso facto,
I argue that activist judicial decisions in salient policy fields will influence

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

integration and the everyday lives of Europeans after being incorporated into
new legislation. (Wasserfallen 2010: 1142)
In salient policy fields, activist Court decisions cannot by themselves effectively
influence Europeans’ everyday lives. (Wasserfallen 2010: 1142)

Following Wasserfallen’s argument, we should not be able to detect any direct

effects of Europeanization, as derived from case law, on member-state policies.
This might be regarded as the equivalent to a null hypothesis for analysing the
impact of case law on member states. According to Wasserfallen, case law’s
sole importance is its impact on the process of European integration, which
I analysed in the previous chapters. From this perspective, Europeanization
research justifiably has a narrow focus. The opposite assumption would be that
member-state executives, legislatures, and courts do in fact abide by the ECJ’s
case law, given its supreme nature. Thus, member-state administrations, in
their understanding of themselves as operating under the rule of law, would
automatically adapt to supreme European case law, just as they would take up
the rulings of their own domestic high courts. Consequently, research on
Europeanization would have a blind spot.
One can thus set out to study the Europeanization effects of case law with
very contradictory expectations. Research on Europeanization may justifiably
only focus on implementation research when there is no direct reaction from
the member state to case law. There is ‘contained compliance’, as Lisa Conant
(2002) has argued. However, if there are national responses to EU case law, the
impact is much more complex. In contrast to secondary law, which is shaped
by the member states collectively in the Council, the implications of case law
for member states are more legally uncertain. Secondary law consists of rules
that have to be implemented. Case law rarely makes prescriptions in this sense;
it results from the examination of single issues, in one of the current twenty-
eight member states, and has uncertain implications for similar but distinct
conditions in other member states. In this way, member states’ domestic
regulations face challenges to their legality under European law. Moreover,
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Europeanization With and Against the Odds 171

when the ECJ is called upon to rule on an instance of incompatibility
between a domestic and a European norm, it often prohibits certain existing
practices that are followed by member states or forecloses previously avail-
able political options. Examples of ‘non-action’ are, however, difficult to
research (Bachrach and Baratz 1962; Schmidt et al. 2008). When policy
options are foreclosed, member states may also search for alternative
means to realize their policy objectives. These compensatory measures,
then, are also a Europeanization effect.
In the following section, I briefly recapitulate how research on European-
ization has a blind spot when it comes to the impact of European case law.
Case law can necessitate decisions, as well as non-decisions (Bachrach and
Baratz 1962). Given higher levels of legal uncertainty, and the possible double

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

emphasis of both demanding and prohibiting certain actions, it is likely that
case law serves as an opportunity structure for different actors. In order to
approach the topic, I will contrast two case studies in which markedly different
member-state reactions to case law arose: the German reaction to the Meilicke
tax cases and the Irish reaction to the Zambrano case. Germany had feared
high tax losses from the Meilicke tax cases. As might be expected from a
contained compliance perspective, this was resisted as much as possible.
Meilicke thus confirms the ‘null’ hypothesis that case law has no or little direct
impact on the member-state level. In a second judgment, the Court fine-tuned
its case law, thereby limiting its potential implications. A much more unex-
pected development resulted from the Zambrano case in Ireland. Although it
was a case from Belgium in a politically salient policy field—the citizenship
rights of third-country nationals—Ireland responded with several measures.
As a result, this case allows me to refute the ‘null’ hypothesis by demonstrating
that member states do respond directly to case law. The juxtaposition of these
two cases provides the foundation for studying the issue of member-state
reactions to case law more systematically in the chapter that follows this one. It
should not be forgotten that the question of member states’ responses to case
law also feeds back into the processes discussed in Chapters 4 and 5. If
member states implement changes to domestic policies in response to case
law, they will pursue different preferences in relation to codification than if
they simply ignore the implications of case law.

R E S E A R C H O N E U R O P E A N I ZA TI O N

Research on Europeanization analyses the impact of integration on member

states. This research, however, has largely concerned itself with how member
states comply with European secondary law and therefore focuses mostly on
the implementation or, more specifically, the transposition of directives. This
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

172 The ECJ and the Policy Process

shows how deeply integration research is still rooted in the intergovernmen-
talist paradigm: member states, it is assumed, give themselves rules at the
European level, which they then have to implement. That European integra-
tion also develops outside of the control of member states is acknowledged
abstractly, in particular when the ECJ’s power is discussed, but little attention
is paid to its implications. In one of the few early articles to explicitly refer to
the domestic impact of European law, Knill et al. (2009) only mention that
regime competition between administrative systems has an indirect effect.
They regard this as one ‘neglected face of Europeanization’.
The narrow focus of research on Europeanization is surprising. Regulations
that are already directly effective, although they make up the bulk of European
secondary law, are rarely analysed (Treib 2014: 16). This probably has meth-

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

odological reasons. Data on infringement procedures against member states
are as readily available as parliamentary documents on the national transpo-
sition of directives (Töller 2010). Regulations that are directly effective, in
contrast, would need to be researched in relation to their implementation,
namely the way in which their stipulations are handled by the domestic
administrations of the current twenty-eight member states, which is much
more difficult. In his overview of research on Europeanization, Treib (2014)
distinguishes between four different waves of research. Research originally
assumed that member-state compliance was largely determined by the ‘good-
ness of fit’ between domestic policies and European demands (Knill and
Lenschow 1998; Börzel and Risse 2003; Héritier 1995). This institutionalist
approach proved too simple, however, as it assumed that all domestic actors
prefer the status quo policy. As many domestic actors seize on EU legal
requirements as an opportunity structure, it is relevant to focus on their
preferences. In his research on compliance with EU directives, Treib (2014)
shows that research on Europeanization has become very complex, no longer
lending itself to simple hypotheses, such as the one on ‘goodness of fit’
between EU law and domestic conditions that determine compliance.
The impact of case law is particularly difficult to research (Schmidt et al.
2008). Treib (2014) includes this research under the most recent fourth wave.
The ECJ’s case law, typically, decides on the compatibility of a specific
member-state rule with supreme European law. Due to institutional hetero-
geneity among member states, the implications for other member states are
often contested and unclear at the national level until further court cases
evolve. Moreover, case law does not necessarily only require decisions, but
also non-decisions, where governments are asked to halt certain policies
(Schmidt et al. 2008). Alternatively, they may not be directly called upon to
take action at all; instead, they may only face strong indirect pressures, as is the
case in regulatory competition.
When considering these differences between the transposition of directives
and the impact of case law, the typologies of member states’ reactions, which
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Europeanization With and Against the Odds 173

are identified in the literature on Europeanization, may need some modifica-
tion. They do not capture the ambivalence of case law, which does not give
general policy directions. They do not pay attention to non-decisions
and neither do they take the indirect effects of case law into account, such
as regime competition or compensatory measures. To give an example,
Radaelli, in a much-cited typology, distinguishes between inertia (no domes-
tic response), absorption (limited response), transformation (far-reaching
change), and retrenchment (active resistance) (Radaelli 2003).1 Börzel and
Risse distinguish between different degrees of compliance in the form of
absorption, accommodation, and transformation (Börzel and Risse 2007). In
her early research on the impact of case law, Conant found contained
compliance (Conant 2002). While single judgments may be far-reaching,

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

member states draw no general policy lessons from them, Conant argues;
consequently, their impact scarcely surpasses an individual case. She distin-
guishes between contained compliance, restrictive application in secondary law,
and pre-emption, which avoids future judicial interference, as the dominant
responses to case law (Conant 2002: 32–3). This finding resonates with Was-
serfallen’s argument, cited earlier. Should this prove to be the case, a typology of
member states’ reactions to case law would have to focus more on retrenchment
and inertia rather than on absorption and transformation.
With an explicit focus on the impact of case law, Blauberger (2012) differ-
entiates between escape routes, through which member states attempt to find
alternative means of realizing original policy goals, and regulatory surrender.
When adapting to case law, member states may choose ‘emulation’ as a
strategy, where they adopt policies which have been vetted by the Commission
and/or the Court in other member states (Blauberger 2012). In an attempt to
come to grips with the question of when member states are likely to follow up
on ECJ rulings—and when they are more likely to simply ‘contain justice’—
Blauberger (2012) holds that it is crucial to ask to what extent member states
can tolerate the legal uncertainty that results from ongoing case-law develop-
ment. Thus, he includes the ambiguity of case law, which is absent from other
Europeanization typologies. Blauberger argues that member states respond
partly with contained compliance and partly with ‘anticipatory obedience’.
Member states are unlikely to react if the challengers of the regulatory status
quo have to bear the cost of legal uncertainty. However, if legal uncertainty
and ongoing case-law development imply costs for member states, it is much
more likely that they will respond proactively. The costs of legal uncertainty
depend on ‘Time constraints, the population of similar cases and each party’s
worst case scenario for an eventual ECJ ruling’ (Blauberger 2014: 458).

1
For an application of these distinctions, see the example of the implementation of the case
law from Rüffert in the German states as discussed by Sack (2012).
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

174 The ECJ and the Policy Process

Furthermore, he specifies: ‘[W]hen legal uncertainty undermines political
planning capacity or even involves great financial risks, and when the
spectrum of potential litigants becomes too vast, member state govern-
ments anticipate future legal challenges by adjusting domestic regulation’
(Blauberger 2014: 472).
Thus, Blauberger focuses on the cost of ignoring case law and of risking
further case-law development. For member states, it is crucial, if the worst-
case scenarios that result from case-law development include liability issues or
other costs that are incurred from periods of disobeying case law. As has been
mentioned, given that the ECJ interprets EU law as it stands, rulings normally
have retroactive effect. This does not matter for all legal issues in an equal way.
For instance, one would expect an individual third-country national who is

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

denied certain rights to be much less likely to claim compensation than a
company if tax privileges are at stake. In addition, special attention must be
given to non-decisions as reactions and to the indirect effects of case law.
Against this background, summarized in Figure 6.1, I will now analyse two
case studies concerning member states’ reactions to ECJ case law. Both deal
with an area of significant political salience: corporate taxation and the
residence rights of third-country nationals. These two case studies, which
have very different outcomes, show that member states’ reactions to case
law do not follow a simple contained compliance approach. On this basis,
the following chapter will systematize the findings on the Europeanization
of case law.

Europeanization studies typically focus on directives

Europeanization effects: inertia (no domestic

response), absorption (limited response),
transformation (far-reaching change), and
retrenchment (active resistance)
Börzel/Risse: absorption, accommodation,
transformation

This focus overlooks: decisions and non-

decisions, legal uncertainty of case law,
anticipatory reactions, compensation, and
regulatory competition

Figure 6.1. The blind spot of research on Europeanization

 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Europeanization With and Against the Odds 175

THE MEILICKE C AS E S

The Meilicke tax cases (C-292/04, 6 March 2007, and C-262/09, 30 June
2011) are a suitable starting point for our discussion of the effects of case law
on member states, as they present a national worst-case scenario. The cases
were referred by the lower tax court of Cologne and dealt with the question
of discriminatory tax discounts for the dividends of EU companies under the
imputation system that was valid in the 1990s. In the second referral, the
Cologne court asked procedural questions that arose from the first judgment.
In the Meilicke case, an individual litigant (a lawyer) gave the German tax
system a serious blow. Germany claimed possible tax losses of €5–10 billion
(interview Federal Ministry of Economics, 12 February 2013). Therefore, the

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

cases are an example of how member states deal with judgments that
potentially impose high budgetary costs upon them, while illustrating the
motivation of litigants to seek the assistance of European law. If the Court is
assumed to be responsive to member states’ worries about case law, as an
intergovernmentalist position would hold, the result would be reduced
incentives for litigants to take the time and cost of litigation upon them-
selves. The more the Court is prepared to harm member states’ interests to
the benefit of litigants, the higher incentives for litigation will be. Meilicke,
therefore, provides evidence of the threat individual litigants pose in bring-
ing down member-state policies with the help of EU law.
In the following, I will start with an outline of the litigant: Meilicke. I will
then go on to give the factual background for the case and will continue by
discussing the Court’s rulings and their consequences.

Meilicke

The heirs of Heinz Meilicke initiated the case that led to the two references.
Heinz Meilicke was a tax lawyer, practising in Bonn. His son, Wienand
Meilicke, was of the same profession and a partner in the same office. As a
specialist in tax law, Wienand Meilicke had already gathered experience in
using EU law as a lever against German tax law. The first case was C-83/91
Meilicke v ADV/ORGA AG and was related to the second company law
directive. This case was about shareholders’ rights to information concerning
the financial health of the company. As AG Tesauro noted in his opinion on
the case, Meilicke had initiated it in order to receive a comment from the ECJ
on ‘his writings concerning the legality of the theory of disguised non-cash
subscriptions’ (Arnull 1993: 616). In doing so, the ECJ would also provide a
comment on the Bundesgerichtshof (Federal Court of Justice), which had
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

176 The ECJ and the Policy Process

developed the case law that Meilicke was criticizing in his writings. The ECJ,
however, refused to rule on the case, which it considered to be hypothetical. In
parallel to the cases on dividend taxes, Meilicke also initiated Case C-76/05
Schwarz and Gootjes-Schwarz (11 September 2009), which dealt with the
discriminatory taxation treatment of boarding-school expenses. German tax
law only allowed deductions for German schools but not for expenses incurred
from schools in other member states, in this case Scotland (Herlinghaus et al.
2010). Moreover, Meilicke was active in pushing the Commission to start an
infringement procedure against Germany for not implementing the conse-
quences of the Centros and Überseering rulings, which related to certain kinds
of non-resident companies that had been excluded from its fiscal-unity regime
(Organschaft) for corporation tax. This case was closed once Germany

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

changed its law (Meilicke 2009).2

The Facts Concerning Meilicke I

At issue in the case were the dividends paid by Danish and Dutch companies
to the late Mr Meilicke between 1995 and 1997. Before 2001, Germany taxed
dividends according to the imputation system. To avoid the double taxation
that arises when companies pay taxes on their profits and shareholders pay
taxes on the dividends, shareholders received a tax rebate on their income
tax of three-sevenths of the dividends for the company taxes that had been
paid. These were only granted, however, to dividends paid by German
companies. This discriminatory rule violated the free movement of capital,
making an investment in companies in other EU member states less attrac-
tive. As the free movement of capital had been directly effective since 1988,
Germany would have had to compensate for discriminatory tax treatment
since that date. The heirs to the late Mr Meilicke addressed the tax author-
ities in October 2000, after the ruling in Case C-35/98 Verkooijen. In this
case, the Court had ruled against a discriminatory income tax exemption on
dividends. When the German tax authorities rejected income tax credit for
dividends from companies in other member states, Meilicke initiated the
proceeding at the lower tax court in Cologne, whence the request for a
preliminary ruling resulted (Rainer 2005). It should be mentioned that,
during the legal procedure, Meilicke repeatedly aimed to expand his claims,

2
The case reference number was 2008/4909. In Mar. 2012, the Commission handed the
case to the ECJ (IP/12/283). The proceeding was then closed, after Germany changed its
legislation.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Europeanization With and Against the Odds 177

including dividends from companies situated in the European Economic
Area (EEA) and also overseas.3
Germany changed the law in 2001, ending the discriminatory treatment
with the Halbeinkünfteverfahren.4 The question raised in the Meilicke case was
to what extent shareholders could claim tax credit for the dividends of EU
companies between 1988 and 2001. The case was unusual, as opinions from
two Advocate Generals (AGs) were given. In the first opinion in November
2005, AG Tizzano argued that dividends had to be treated in a non-
discriminatory way, but that the retroactive effects of the case should only
go as far back as mid-2000, when the Verkooijen case had been ruled upon.
This would have been a good outcome for Germany, but because limiting the
temporal effects of the case would have been a bold judicial step, the case was

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

transferred to the Grand Chamber, which reopened the oral procedure.
A second hearing took place, in which eleven member states (Germany, the
Czech Republic, France, the Netherlands, Denmark, Greece, Spain, Hungary,
Austria, Sweden, and the UK), the Commission, and Mr Meilicke all gave
opinions (AG Stix-Hackl, No. 5a). AG Stix-Hackl issued her opinion in
October 2006 (Rainer 2006). She argued against a retroactive limitation of
the case. Such a limitation could only be granted in exceptional circumstances
as the Court had done in the Defrenne II ruling (No. 12). An ‘overriding
reason in the general interest’ could not be related to the budget, she argued
(No. 16). A limitation of the temporal effect needed to be restricted to the
member states to which it was granted (No. 14) and could only be given in a
judgment that interpreted the contentious norm for the first time (in this case,
the Verkooijen and Manninen (2004) rulings, No. 19). She also claimed that
the German government had not provided sufficient evidence as to the ‘serious
economic repercussions’ (No. 7, 62) of the retroactive ruling. Moreover, she
pointed out that the fact that the German government had submitted the draft
law to change the taxation of dividends in 2001, before the Verkooijen ruling,
showed the government’s awareness that it had breached EU law (No. 55).
This opinion gave rise to a very critical press release from the German
Federal Ministry of Finance.5 In this statement, the Ministry argued that it
could face losses of up to €5 billion in 2005 and 2006, endangering fiscal
consolidation, which was characterized in the AG’s opinion as grotesk
(preposterous).

3
European Commission, Schriftsatz Rechtssache C-262/09, 27 Nov. 2009, pp. 28ff. JURM
2313 (Meilicke II, Opinion of the Commission, on file with the author). However, the Cologne
court refused to expand the case.
4
The half-income procedure charges income tax on only half of the dividends received.
5
Bundesministerium der Finanzen, Pressemitteilung 121/2006: ‘Verfahren vor dem EuGH
gefährdet massiv deutsche Haushaltskonsolidierung’, 5 Oct. 2006, <http://presseservice.pre
ssrelations.de/standard/result_main.cfm?aktion=jour_pm&r=250777&quelle=0&pfach=1&n_
firmanr_=100106&sektor=pm&detail=1> [accessed 10 May 2017].
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

178 The ECJ and the Policy Process

The Court’s Ruling

In its ruling on the Meilicke I case in 2007, the Grand Chamber emphasized
that it had already decided on the violation of the freedom of capital through
an imputation system in 2004 in the Finnish Case C-319/02 Manninen.
Articles 56 and 58 of the Treaty Establishing the European Community
(TEC) (Articles 63 and 65 of the Treaty on the Functioning of the European
Union (TFEU)) precluded a restriction of tax credit on dividends paid by
German companies. The real relevance of the ruling was, therefore, not that
the imputation system violated the Treaty but that the Court refused to limit
the retroactive scope of the ruling, despite the fact that Germany claimed
significant tax losses. The Commission and the eleven intervening member

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

states had argued in favour of such a limit, when member states could claim
serious economic harm. In addition, AG Tizzano had argued for limiting the
temporal effect back to the date of the Verkooijen ruling in mid-2000, while
AG Stix-Hackl had taken another view.
The litigant himself interpreted the ruling as requiring the Court to determine
any retroactive effects when first deciding on a case (in this instance, Manninen).
Only in this way could discriminatory treatment among member states be
avoided. Otherwise, there was a danger that some member states would change
their tax rules only to find that others would be later exempted from a retroactive
effect, as the Court was lenient with the largest violators of treaty obligations. As a
consequence, after the Meilicke case, member states would have to state any
serious economic consequences of tax rulings the first time that the ECJ rules on
a question. Germany, in contrast, had foregone the opportunity to submit an
observation to the Manninen case, despite the similarity between the German
and Finnish imputation systems (Meilicke 2007).

Meilicke II

In the second Meilicke case, the German court asked how it should apply the
previous ruling, as the amounts of company tax paid in other member states
were difficult to determine. The question was whether shareholders could
claim a lump sum or act on an estimate, what kind of documentation they
would have to provide, and whether German tax authorities were in any way
obliged to seek the assistance of tax authorities in other member states in order
to determine what taxes had been paid there.
A further question in the Meilicke II case concerned a change of Section 175
para. 2 of the German General Tax Act. In this section, a second sentence had
been inserted after Case C-319/02 Manninen on the Finnish imputation
system, which prohibited the late submission to the authorities of further
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Europeanization With and Against the Odds 179

evidence on taxes paid in other member states. Effectively, this hindered
taxpayers from claiming retroactive tax credit that was due after Meilicke I.
In its ruling on Meilicke II in 2011, the Court stated that Germany had violated
the effet utile of the Treaty by this addition to the Tax Act. Tax authorities
could not apply this rule when they dealt with companies from other member
states. Shareholders had to have a transitional period for claiming their tax
credit. It was for the member-state court to establish the modalities. Tax
authorities are not allowed to simply act on an estimate or a lump sum.
They are also not required to acquire information themselves on the basis of
the cooperation agreement between member states’ authorities. Instead, share-
holders must submit documentation on the taxes paid in other member states,
and the tax credit that would be claimed was limited by this amount.

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

The Federal Fiscal Court (Bundesfinanzhof ) gave its final judgment in the
Meilicke case in early 2015 (Sydow 2015).6 It demanded the documentation of
taxes paid in other member states that were of an extent that the litigants were
unable to meet. Despite the originally bold ruling of the ECJ, there were only
meagre direct benefits for the Meilicke heirs. As a consequence, Meilicke sent a
complaint to the European Commission requesting an infringement procedure7
and also initiated a constitutional complaint at the German Constitutional
Court (Meilicke 2015). With many tax statements still open, the final
consequences of the Meilicke ruling remain unclear. The considerable time
lag between the start of the litigation in the year 2000 and the final judgment
fifteen years later points to an additional factor that makes research into the
Europeanization effects of case law difficult. While EU secondary law comes
with clear transposition dates, the drawn-out process of case-law develop-
ment ameliorates and blurs its effects to a great degree.

Assessing Meilicke

It is difficult to assess the real importance of the Meilicke cases. By imposing

relatively strict requirements for documentation on shareholders, the Court
cushioned the Meilicke ruling. While the Court outlined that demands for
documentation must not be prohibitive, it put the burden of proof (for the
company taxes that had been paid) on the shareholders that claimed income
tax credit. This limited the fiscal repercussions. The Court, it appears, shied
away from the high figures that had been claimed by Germany, tax losses
for which it was difficult to take political responsibility. However, Germany
had to adapt its tax rules, as Meilicke was another example of an ECJ

6
BFH, Urteil v. 15 Jan. 2015—I R 69/12; published 10 June 2015.
7
Available at: <http://www.meilicke-hoffmann.de/Beschwerde%20EU-Kommission.pdf>
[accessed 10 May 2017].
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

180 The ECJ and the Policy Process

ruling that restricted member states’ leeway. In light of the risk of rulings
having retroactive effects, member-state governments need to take great care
in their monitoring of tax cases at the ECJ.
The German reaction to Meilicke conformed to theoretical expectations, but
only in part. Germany appeared to be relatively passive, as it clearly opposed
the fiscal implications of the Meilicke case. This follows the argument that
member states ignore and oppose case law whenever this conflicts with their
policy preferences. Nevertheless, after Case C-35/98 Verkooijen in mid-2000,
Germany changed its dividend tax law in 2001 in the Steuersenkungsgesetz
(StSenkG), which was part of the general company law reform of the red–green
coalition. It had thus realized the erga omnes effects of Verkooijen, although
this was not noted in the legal proposal or the parliamentary debates.8 Graetz

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

and Warren Jr claim that several member states abandoned their imputation
system in anticipation of Manninen (Graetz and Warren Jr 2006: 1211). In
addition, a study of the EP argues that, after the Manninen case in 2004, the
German legislator was well aware of the incompatibility of the German tax
system with the Treaty but deliberately stayed passive. In allowing member
states to limit the retroactive effects of rulings, as requested by the German
government in the Meilicke case, incentives for non-compliance would be
created (Malherbe et al. 2011: No. 51).
Despite the significant potential fiscal costs, there were no apparent moves to
avoid the case-law development of Meilicke I and II. This points to an additional
relevant consideration: in very high-cost situations for governments, it may be
advisable to wait for additional case law rather than to avoid it (interview Federal
Ministry of Finance, 28 September 2015). This allows the opportunity for the
Court to fine-tune its case law and ameliorate its impact.
We will now turn to the Zambrano case, and the Irish reaction to it. Since it
concerns the residence rights of third-country nationals, the case similarly
dealt with a politically salient issue. The case originated in Belgium, so one
would not have expected there to be large repercussions in other member
states like Ireland. Nonetheless, there was a marked Irish response, which is
interesting in how it contrasts to theoretical expectations, as well as in light of
the findings in the Meilicke case.

T H E CO U R T ’S A C T I V I S M ON CI T I Z EN SH IP R I G H T S

As we saw in Chapters 3 and 5, the Court has developed EU citizenship into a

status that provides comprehensive rights in other member states. The Court

8
See the documents on this law at: <http://dip.bundestag.de/extrakt/14/019/14019321.html>
[accessed 10 May 2017].
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Europeanization With and Against the Odds 181

has done this in two ways. On the one hand, it has extended the privileges of
free movement for workers to those who only engage in limited work and are
not able to cover the costs for their subsistence. On the other hand, it has
transformed symbolic EU citizenship, arguing that it ‘is destined to be the
fundamental status of nationals of the Member States’ (Grzelczyk C-184/99
No. 31). It is based on an analysis of these Court-driven citizenship rights that
Wasserfallen (2010) has argued that case law is only significant when it is
codified in secondary law.
Might member states simply ignore the case law on citizenship rights if
there is no legislative follow-up? Do they only respond when they are under
pressure from the Commission? Or are they obedient compliers, making the
necessary changes that arise from case law as the rule of law requires? The

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Zambrano case can be regarded as one of the most notable instances of judicial
activism in recent years. The Court significantly broadened the residence
rights of third-country nationals (TCNs) in a purely internal situation,
where EU law does not normally apply. This ruling, with its high disregard
for sovereignty, makes it a least-likely case for member-state compliance
(Gerring 2007). The same can be said in relation to the addressees: individual
actors are less apt to enforce their rights than corporate actors are. This should
be particularly true for TCNs. As a least-likely case, we would expect that
if case law leads to domestic adaptations here, case law should generally
be influential.

THE Z A M B R A N O CASE

Case C-34/09 Zambrano (8 March 2011)9 was a reference from a Belgian

court. A Colombian couple, Mr and Mrs Zambrano, had lived in Belgium,
initially without refugee status and then with registered residence status; they
had not been deported because of the civil war in Colombia. Mr Zambrano
had worked since 2001 for extended periods without a work permit while
paying social security. Their two children, born in 2003 and 2005, had Belgian
nationality, which had been granted at the time by the Belgian Nationality
Code (Article 10(1), see Opinion AG Sharpston No. 16).10 When Mr Zambrano
lost his job, his lack of a work permit meant that he did not receive unemploy-
ment benefits. The Brussels court posed several questions to the ECJ, concerning
the right of residence and to work of Mr Zambrano. The case’s allocation to the

9
This part of the chapter draws on Schmidt (2014).
10
From Dec. 2006 onwards, this possibility no longer existed in cases where parents could
apply for another nationality for their child (Opinion AG Sharpston, No. 17), which the
Zambranos could have done.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

182 The ECJ and the Policy Process

Grand Chamber underlined its legal relevance. Six governments participated in
the case with observations (Germany, Austria, Denmark, the Netherlands,
Poland, and Greece). All governments argued—along with the Commission, it
should be noted—that this was a purely internal case with no transborder
element, having no relation to EU law. However, in her opinion, AG Sharpston
supported the position of Mr Zambrano and pointed out several possible legal
justifications. Her far-reaching interpretation of EU legal entitlements attracted
a good deal of attention. Most notably, she argued against the view that it was a
purely internal case and for a prohibition of reverse discrimination against
nationals. This would imply a major extension of the relevance of EU law, as
well as the ECJ’s case-law development.
With the argument that children could not be deprived of the benefits of

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

their EU citizenship, the Court followed the AG in its ruling, giving
Mr Zambrano the right of residence and to work. However, the Court was
silent on which parts of the AG’s reasoning it followed. While rights
established for cases with a cross-border element were extended to internal
situations, the scope of these rights remained unclear (Van den Brink
2012: 286).

Fine-Tuning Zambrano

Drawing on Obermaier’s finding that the Court fine-tunes its case law in order
to keep its implications within acceptable limits (Obermaier 2008b: 746, 751),
it is relevant to ask how case law developed after Zambrano. The implications
of the Zambrano case would clearly be very different for member states,
depending on which TCN family members could indirectly profit from the
EU citizenship of relatives. Between the decision on Zambrano in the spring of
2011 and late 2012, several cases were decided upon, specifying the ruling’s
meaning and effectively narrowing down its scope (see also De Somer 2016).
Case C-434/09 McCarthy (5 May 2011) followed soon after Zambrano.
A British citizen, who also held Irish citizenship, had never left Northern
Ireland, where she lived from social assistance. After getting married to a
Jamaican citizen, who had no right of residence in the UK, she got an Irish
passport in order to argue that she was an EU citizen living in the UK with the
right to family reunification. The UK, Denmark, Estonia, Ireland, and the
Netherlands intervened. The third chamber of the Court denied McCarthy’s
claim, as it could not be established that she would be forced to leave the EU,
while emphasizing that EU citizens cannot be deprived of the ‘genuine enjoy-
ment’ of their EU citizenship. Again, some issues remained unclear. As Wray
(2011) argues, the ECJ failed to clarify whether McCarthy could not be joined
by her spouse on the grounds that she was economically inactive or because
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Europeanization With and Against the Odds 183

she had never moved. However, the implications of the Zambrano ruling were
narrowed—or fine-tuned—(Obermaier 2009: ch. 9) by this judgment.
The same can be said of Case C-256/11 Dereci and others (15 November
2011). There were five joined cases handed to the ECJ by the Austrian
administrative court, and they were ruled upon, as Zambrano had been, by
the Grand Chamber. In all these cases, which had different factual circum-
stances (one also involved the association agreement with Turkey), Austria
had not allowed the TCN spouses of Austrian citizens residence or had denied
it to the grown-up children of TCN Austrian residents, although these chil-
dren were economic dependants. Again, several member states joined the
case: the governments of Austria, Denmark, Germany, Ireland, Greece, the
Netherlands, Poland, and the UK all argued that this was a purely internal

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

situation with no relevant connection to EU law. For the member states, this
was an important case, and the fact that the Court had allocated it to the
Grand Chamber could have signified important new legal developments.
A transfer of the Zambrano argument on dependent children to different
kinds of TCN family members would have had significant quantitative con-
sequences for migration, and member states were still concerned about AG
Sharpston’s arguments in the Zambrano case that reverse discrimination
against nationals was prohibited in EU law. This would have seriously ham-
pered member states’ regulatory autonomy. The Court again emphasized
genuine enjoyment of EU citizenship, which, however, only sets in if ‘the
territory of the Union as a whole’ had to be left. EU law did not grant the right
of family unification as such. Therefore, as far as all the cases were concerned,
right of residence in Austria did not follow from EU citizenship rights. AG
Mengozzi had argued similarly.
Case C-40/11 Iida, for which a decision was reached in November 2012,
concerned a referral from a German court and was ruled upon by the third
chamber. The governments of Germany, Belgium, the Czech Republic,
Denmark, Italy, the Netherlands, Poland, and the UK joined the case, dem-
onstrating once more the great importance governments attach to citizenship
cases. Mr Iida, a Japanese national, worked in Germany and had joint custody
of his German child with his German wife, from whom he had separated
and who had moved to Austria with their daughter. He could have received
right of residence on the basis of his work permit, but he demanded it as a
spouse of an EU citizen. The Court, however, ruled that Mr Iida could not
derive right of residence from the EU citizenship of his family.
Cases C-356/11 and C-357/11 O and S Maahanmuuttovirasto (6 December
2012) questioned whether the principles established in the Zambrano ruling
also cover a TCN stepfather who is an economic dependant. These two cases,
which were decided upon by the second chamber, were referred by a Finnish
court, and the Finnish, Danish, German, Italian, Dutch, and Polish govern-
ments joined the case. The Court ruled that Article 20 of the TFEU did not
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

184 The ECJ and the Policy Process

deny member states the right to refuse a residence permit, but that the court
that had referred the case had to ascertain whether the genuine enjoyment of
the children’s EU citizenship rights was violated or not, also taking into
account the provisions of the Family Reunification Directive. As a result, the
Zambrano reasoning was again not extended but rather handed over to the
domestic court for it to reach its own decision.
Several cases also followed in 2013 that refined the Zambrano reasoning still
further. Case C-87/12 Ymeraga, which was decided upon in May 2013,
involved a situation similar to that in the Dereci case. On this occasion,
Luxembourg had denied residence to a third-country national parent. Again,
the Court saw no general right to family reunification. Case C-86/12 Alokpa
from October 2013 concerned children of French nationality, who had never

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

resided in France and had no contact with their French father. The question
was whether Luxembourg could expel the TCN mother, if this resulted in the
children having to leave the EU. As they had the possibility to live in France,
the Court backed Luxembourg’s right to deny residence.
Cases C-456/12 O and C-457/12 S, which were decided upon in March
2014, questioned to what extent transborder movement is necessary to enjoy
the privilege of family reunification with TCN family members under EU law.
AG Sharpston had raised this point in her opinion in the Zambrano case. In
relation to its purely internal nature, she had asked whether it would have
made a difference if neighbours had taken the children to ‘Parc Astérix in
Paris’ (No. 86), which raises the difficult question of whether a single visit is
sufficient for establishing relevance to Community law. In these cases, a Dutch
national had lived abroad for some time, and another Dutch national had
worked abroad on a regular basis. The Court clarified that residence for more
than three months in another member state is necessary to establish residency
rights for TCN family members and that regular weekly trips to another
member state are insufficient.
In summary, this series of cases closely corresponds to Obermaier’s argument
on the fine-tuning of jurisprudence. The original Zambrano case, taken together
with AG Sharpston’s far-reaching argument and the scant reasoning that the
Court itself provided, threatened to constrain member states’ migration regimes
in a serious way. However, between March 2011 and December 2012, the ECJ
limited the impact of the Zambrano ruling to cases where EU citizen minors
would be forced to leave the EU were TCN family members not allowed to stay.
The Court confirmed that EU citizenship does not encompass a right to live in a
particular member state with all family members but that it is up to the member
states whether they grant these rights. The speed of this fine-tuning is very
interesting. It shows how the EU’s legal community, along with lawyers and
courts in the different member states, quickly took up the opportunity provided
by new case law and sought to establish the scope of its meaning.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Europeanization With and Against the Odds 185

However, the fine-tuning should not be mistaken for a reversal of the
previous case law. With the Zambrano ruling, a bold step in broadening the
implications of EU law was taken. Subsequent fine-tuning did not reverse this
case; the Court only elucidated which family members could profit from the
legal reasoning, thereby refraining from extending the reasoning beyond
dependent children. Nevertheless, in light of the research finding that case
law has a restricted impact when it is not followed up by secondary law, we
should expect the Zambrano ruling to have little impact on member states.
However, this highly controversial ruling led Ireland to react immediately,
without playing for time.

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

MEMBER STATES ’ REACTIONS

TCNs are comparatively weak political actors with few rights. Governments
do not have to fear liability claims when they restrict their rights. However,
they may possibly be fearful of politicization through active NGOs and pro-
bono legal services. This situation is basically the same in all member states.
However, member states were not all affected in the same way by the Zambrano
judgment. The largest impact results from ius soli nationality being granted
at birth (Honohan 2010), which was only the case in Ireland up until 2004.
Member states’ citizenship regimes are quite complex, generally combining ius
soli and ius sanguinis, and have different conditions attached, such as prior
parental residence or parental birth in the country.
Alongside citizenship, some other legal positions are relevant. The Family
Reunification Directive (2003/86/EC) gives TCNs the option of applying for
the right of residence of their spouse and minor children, under the condition
that the applicant has had a residence permit for at least one year; this includes
the option of long-term residence. Ireland, however, had opted out of this
directive (Becker 2014: 105). Article 8 of the European Convention on Human
Rights (ECHR) and Article 7 of the Charter of Fundamental Rights of the EU
defend ‘respect for private and family life’. The ECHR is relevant for purely
domestic situations, while the Charter only applies when member states
implement EU law (Article 51(1)) (Van den Brink 2012: 283). However, in
its interpretation of Article 8, the European Court of Human Rights (ECtHR)
argues for a narrow interpretation, in which childbirth alone does not grant an
entitlement to stay, given that the family could live together in another place
(Van den Brink 2012: 285).
This proactive Irish stance allows for the further analysis of the conditions
under which member states comply with case law.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

186 The ECJ and the Policy Process

The Irish Reaction

Of all the member states, Ireland was the one that was most affected by the
Zambrano ruling, since it had granted unrestricted ius soli rights up until 2004.
In the observation submitted in the Zambrano case, the government had
warned of the ‘floodgates’ being opened were the Court to grant EU citizen-
ship rights in this purely internal situation (AG Sharpston, No. 114). ‘Counsel
for Ireland painted a dramatic picture of the wave of immigration by TCNs
that would inevitably result if Mr Ruiz Zambrano were held to enjoy a
right of residence derived from his children’s Belgian nationality’ (No. 112).
Coincidentally, one day after the ruling on 8 March 2011, the government
changed and the new Minister for Justice from the Fine Gael party, Alan

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Shatter, immediately announced the initiation of a reassessment of all cases
(approximately 120 of them) where deportation had been considered before
the courts. The National Naturalization and Immigration Service (INIS), the
administration responsible for this area, even looked into cases where depor-
tation had already taken place. This is surprising, as one should expect little
immediate government reaction to far-reaching court cases when no powerful
actors are involved. To understand the Irish reaction, it is necessary to begin
by giving some background on the development of citizenship rights in the
country (Handoll 2012).

Citizenship Law in Ireland

Citizenship law in Ireland is closely connected to its division as a country

and the wish to give everyone born in Northern Ireland the opportunity to
gain Irish citizenship. As a common-law country, landmark judgments have
shaped the citizenship regime. Of all the member states, Ireland stands out
for its lack of family reunification legislation (Becker 2014: 105), which gives
the Minister for Justice, Equality, and Law Reform a wide margin of discre-
tion under the Nationality and Citizenship Act, subject to judicial review
(Handoll 2012: 17–19). In 1989, the Supreme Court granted far-reaching
residence rights to a couple who had been living illegally in Ireland for eight
years and who had a daughter with Irish citizenship (Fajujonu v Minister for
Justice (1989)). This allowed migrants giving birth in Ireland to receive
residence rights (Handoll 2012: 8; Grossman 2004). Statistics on birthright
citizenship are limited, and, since they never include data on the movement
and residence of those who have obtained citizenship in this way, they are
also hard to interpret (Grossman 2004: 111). Between 1996 and 2003, approxi-
mately 10,500 residence permits were given to the parents of Irish-born
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Europeanization With and Against the Odds 187

children.11 It has to be kept in mind that, during this period, when the
economy was dubbed the ‘Celtic tiger’, immigration into Ireland rose in
general, with work permits issued to migrants outside of the EEA rising
from 5,750 in 1999 to 40,504 in 2002 (Mancini and Finlay 2008: 577).
In 2003, the Supreme Court supported a more restrictive decision by the
minister to deny unconditional residence rights to parents who were illegally
resident (Handoll 2012: 8).12 As a consequence, the minister reviewed the
residence procedure, which affected 11,493 outstanding cases.13 Shortly after-
wards in 2004, a referendum brought about a constitutional change that
abolished the pure ius soli right. Since 2005, a parent has to be Irish, British
in Northern Ireland, or to have been legally resident for three out of four years
before the birth for children to qualify for citizenship (Handoll 2012: 11).

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

This reform calmed political debate about the growing numbers of non-EU
nationals giving birth in Ireland for citizenship reasons (Grossman 2004).
Interestingly, the change in Irish citizenship law also had a European
dimension. Case C-200/02 Zhu Chen had revealed the externalities of the
Irish ius soli right as they applied to other member states, which AG Tizzano
emphasized in his opinion (De Somer 2012: 16; Mancini and Finlay 2008: 582).
Here the Court decided that (Chinese) parents living in the UK could partake
in the citizenship rights of their (Irish-born) daughter, whose EU citizenship
rights would be violated if her parents were not allowed to live with her. The
parents had deliberately arranged to give birth in Northern Ireland to profit
from the Irish citizenship rule in order to take up their intended residence in
the UK. AG Tizzano stated the following: ‘The fact is that the problem, if
problem there be, lies in the criterion used by the Irish legislation for granting
nationality, the ius soli, which lends itself to the emergence of situations like
the one at issue in this case’ (No. 124). Importantly, these externalities of the
liberal Irish regime were discussed at the same time that the Supreme Court
case of 2003 had made the regime in Ireland more restrictive (Handoll 2012: 9;
Mancini and Finlay 2008: n. 38). The altered citizenship law of 2005 limited
the impact of Zambrano.
The quantitative implications of the Zambrano ruling were continuously
reported in the Irish press and discussed in the Dáil (lower chamber). In
just four newspapers, twenty-two articles on Zambrano and its effects were
published.14 In no other member state has comparable attention been paid
to the judgment. In October 2012, The Irish Times reported that 2,100

11
Bode (A Minor) v Minister for Justice, Equality, & Law Reform & Ors (2007).
12
A.O. & D.L. v Minister for Justice (2003).
13
Bode (A Minor) v Minister for Justice, Equality, & Law Reform & Ors (2007).
14
These are findings based on a search of the webpages of the examined newspapers and the
Factiva database (keyword ‘Zambrano’), Mar. 2013.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

188 The ECJ and the Policy Process

applications had been received after the Zambrano ruling; in response to these,
1,184 persons were allowed to remain in Ireland, 84 cases were refused, and
about 800 still had to be settled.15 In January 2013, the minister also elucidated
the following criteria:
each applicant parent must be a TCN who is residing in the State with their Irish
born minor citizen child or children, they must be playing a significant role in the
upbringing of their Irish born minor citizen child or children and the applicant
parent’s immigration circumstances must be such that if a decision was taken to
refuse him or her a right of residency, the Irish born minor citizen child or
children would be at risk of being expelled from the State and, by extension, the
EU and, as such, they would not be able to enjoy the substance of their rights as an
Irish and EU citizen.16

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

At the end of 2013, policy guidelines gave greater legal certainty on non-EEA
family reunification, without altering the minister’s discretion.17

Explaining Proactivity

The government’s immediate reaction shows that there was no pressure from
domestic courts to advance the application of EU law. This has to be explained
politically. With the change of government prior to the ruling, the Fine Gael
minister explained his reaction with reference to his disagreement with the
previous government’s immigration policy.18 His predecessor from the Fianna
Fáil party, Brendan Smith, had, in fact, presided over the highest number of
deportations ever, deporting 416 migrants in only six weeks of office, possibly
under the influence of AG Sharpston’s opinion in the Zambrano case, which
was delivered in October 2010.19
Ireland has an active civil society that informs TCNs about their rights, such
as the Immigrant Council of Ireland, an advocate for migrants and their
families. The Council stepped up its activities in the aftermath of the judgment
and also noted a considerable increase in demand for its services.20 Moreover,

15
Pamela Duncan, ‘State Pays €1.2m to Settle Cases with Non-EU Parents’, Irish Times,
15 Oct. 2012.
16
Houses of the Oireachtas 30 Jan. 2013, Written Answer No. 149—Citizenship Applications,
<http://oireachtasdebates.oireachtas.ie/debates%20authoring/debateswebpack.nsf/takes/
dail2013013000075?opendocument#WRW01150> [accessed 9 May 2017].
17
INIS, ‘Policy Document on Non-EEA Family Reunification’ (Department of Justice and
Equality, Dec. 2013). On file with author.
18
This link is made in the following article: Carol Coulter and Jamie Smyth, ‘Case Review for
Non-Irish Parents of Irish Citizens’, Irish Times, 22 Mar. 2011.
19
Jim Cusack, ‘Smith Deported 416 in Six Weeks as Justice Minister’, Irish Independent,
3 Apr. 2011.
20
Immigrant Council of Ireland 6 Jan. 2013, ‘Citizenship Queries Top Issue on Immigrant
Council Helpline, Over 5,000 Inquiries from 145 Nationalities, Migrants Showing Commitment
to Ireland’. 2 Feb. 2012, ‘50,000 People Accessed Information on Migrant Issues’.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Europeanization With and Against the Odds 189

the Irish Human Rights Commission often participates in cases as amicus
curiae. There is also the Irish Immigrant Support Centre (NASC).
Additionally, before Zambrano, Ireland had already experienced transitional
migration regimes. In 2005, the Department of Justice introduced the Irish
Born Child Scheme (IBC/05) to allow the parents of Irish children without a
residence status to apply for residence under a more favourable procedure
than normal, which was initially granted for two years. Many of those who
would potentially benefit from the Zambrano ruling already fell under this
scheme. About 17,000 persons were granted residence.21 Similar to Zambrano,
it was often only the mother who could stay with the Irish children, who would
possibly be joined by siblings without rights of residence in the EU.22 As a
result, many cases had already been settled. In comparison, Zambrano did not

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

have such a big impact, resulting in only around 1,200 cases. While he
followed a political approach that took ‘the best interests of the eligible
minor Irish citizen children’23 into account, the minister also justified his
proactive approach by avoiding the costs of unnecessary court cases from the
very beginning:
One possible approach in these matters is to wait for pending cases to be
determined by the Irish Courts and for the Courts to interpret and apply the
Court of Justice ruling. That is an entirely justifiable approach from a legal
standpoint. However, in this case the Government has agreed that there needs
to be a more proactive approach and that it should make a clear statement of its
intention to take early action in these cases, insofar as it is unnecessary to await
rulings of the Courts. We should not tie up the courts unnecessarily or ask eligible
families to wait longer than necessary.24
As one of the eurozone countries that was bailed out, there were also incen-
tives to comply with European legal demands.25 Nevertheless, in late 2011, the
government was facing legal expenses of over €100 million in 2,000 cases
where it was fighting deportation injunctions that were granted by the High
Court, although these cases were explicitly related to the Zambrano ruling.26
Clearly, the government was employing a mixed strategy of trying to avert
unnecessary court cases, while pursuing an approach that was not too liberal,

21
Bode (A Minor) v Minister for Justice, Equality, & Law Reform & Ors (2007).
22
Carol Coulter, ‘Review Sought of Scheme Allowing Migrant Parents to Live in Ireland’,
Irish Times, 23 Mar. 2012.
23
Department of Justice, Equality, and Defence, written answers Asylum Applications,
18 Jan. 2012. On file with author.
24
Statement by Minister for Justice, Equality, and Defence, Mr Alan Shatter, TD, on the
implications of the recent ruling of the Court of Justice of the European Union in the case of Ruiz
Zambrano. 21 Mar. 2011. On file with author.
25
I thank John Handoll for pointing this out to me.
26
These costs mainly arise from the government having to take over legal expenses in lost
cases. Jim Cusack, ‘State Facing €100m Bill to Fight 2,000 Asylum Seeker Deportation Injunc-
tions’, Irish Independent, 11 Dec. 2011.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

190 The ECJ and the Policy Process

which could take into account the fine-tuning of the case law after the
Zambrano ruling. Despite the proactive Irish approach, the settling of indi-
vidual cases took a relatively long time. In this respect, it has to be noted that
the High Court had a backlog of about 1,000 asylum cases, which led to a delay
of about two years (Becker 2014: 109).
Interestingly, when one searches for explicit court references to Zambrano,
the wide application of this ruling is not evident, either because the government
grossly overestimated its implications or because the courts’ citations of
Zambrano underestimate its impact. The Zambrano judgment (or the opinion
of AG Sharpston) was cited eighteen times (sixteen times by the High Court
and twice by the Supreme Court) between January 2011 and 2013 in cases that
concerned the residence rights of TCNs with dependent Irish citizens. Of these

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

cases, Zambrano was causally linked to the right of residence or to judicial
review in four cases.27 In at least seven of the cases, Zambrano did not explicitly
apply, but this did not necessarily exclude the granting of the right of residence.
For instance, in one case, the right of a father, who was separated from the
mother of the family, to rely on the Zambrano ruling was denied by the High
Court, as the mother had refugee status. Nevertheless, an interlocutory injunc-
tion was granted to the father on the basis of the child’s constitutional right to
enjoy the company and care of their parents.28 Thus, given the extensive rights
granted by the Irish constitution, Zambrano did not necessarily grant addi-
tional further rights. It appears that both the High Court and the Supreme
Court have not readily established a link to Zambrano if it can be avoided.
The active role of immigration NGOs, as well as recent legal changes, have
meant that the issue is politically salient and has received political and media
attention. The government has played an ambiguous role here. While it
immediately implemented Zambrano, it clearly aimed to constrain rights.
One parent suffices for the genuine enjoyment of EU citizenship. Interestingly,
the right of contact to both parents could be argued on the basis of the Irish
constitution. At the same time, the government’s attempts to avoid the costs of
court cases is also notable.

The Reaction in Other Member States

Due to its citizenship law, Ireland was particularly affected by the Zambrano
judgment. The Zambrano ruling only has an impact on member states when

27
The findings on judgments that reference the Zambrano case are based on a keyword
search in 2013 on the website for the Courts Service of Ireland, <http://www.courts.ie/
Judgments.nsf/advancedsearch?openform&l=en>. Keyword ‘Zambrano’, search Mar. 2013.
28
Immigrant Council of Ireland, ICI News Bulletin, 102, 27 Sept. 2012. On file with author.
E.A & Anore v Minister for Justice & Anor (2012); Irish Times, 11 Dec. 2012, ‘Rights of the Child
Mean Father Should Not Be Deported’.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Europeanization With and Against the Odds 191

their own citizens claim the right to live with a TCN family member and this
right is not granted domestically. It is not necessary to refer to the ruling when
EU citizens have taken advantage of freedom of movement and want to be
joined by their TCN family members, which is clearly granted by EU law. In
general, all families with a domestic and a TCN spouse can be concerned,
where children have EU citizenship, and where the TCN spouse may want to
claim rights from Zambrano, in particular after a separation from the partner.
How did the governments of the other member states react to the ruling
(Fernhout 2012)? In the following discussion, I contrast the Irish experience
with that in Belgium, which is the country where the Zambrano case origi-
nated; I go on to consider the UK, France, and Germany as countries with ius
soli elements in which the case law could well have had a larger impact, given

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

their population sizes, and finally the Netherlands.
In focusing first of all on press reports until early 2013, Ireland clearly
stands out. In contrast to Ireland’s twenty-two articles on the topic, hardly any
attention was paid to it in the other member states; there was one article in six
different papers in Belgium, one in forty papers in Germany, two in thirteen in
Austria, and three in eighteen in the UK.29
Despite being the country where the Zambrano ruling originated, Belgium
was not highly concerned by it, as the underlying citizenship law had already
been changed. In addition, the administration interpreted the ruling restric-
tively from the beginning as only concerning dependent minors and not
situations such as the one that was ruled upon in the McCarthy case (De
Somer 2012: 15). In the courts, seventeen rulings cited the case, but only in
three of them did the Zambrano reasoning make a positive difference until
May 2013.30
In Germany, there were twenty-two rulings in this period that cited the
Zambrano case,31 and in four it made a positive difference. That the ruling
would only be of marginal relevance was not apparent in the beginning, as is
evident in a press statement from a European law professor just after the
ruling, which claimed that up to one million people could be affected by it in
Germany, since Germany had introduced restricted ius soli citizenship rights eleven
years earlier and about 100,000 children are born each year with a non-German

29
These findings are based on a search of the webpages of the examined newspapers and the
Factiva database (keyword ‘Zambrano’), Mar. 2013.
30
The findings on judgments that reference the Zambrano case are based on a keyword
search on the website of the Conseil du Contentieux des Étrangers. <http://www.rvv-cce.be/rvv/
index.php/fr/arresten/arresten-rvv>. Keyword ‘Zambrano’, search May 2013.
31
The findings on judgments that reference the Zambrano case are based on a keyword
search on the websites of the Federal Administrative Court and the Beck-Online database.
<http://www.bverwg.de/entscheidungen/entscheidungen.php> and <http://beck-online.beck.
de/default.aspx>. Keyword ‘Zambrano’, search May 2013.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

192 The ECJ and the Policy Process

parent.32 Although many of these parents are EU citizens or enjoy the right of
residence for other reasons, Professor Thym nevertheless expected the case to
be relevant for a few thousand cases.
The German legal profession was very responsive to the new legal develop-
ment in the Zambrano ruling. It tested the limits of the new development, as
the Iida case shows (Schönberger and Thym 2014: 580). Once subsequent case
law had contained the revolutionary potential of the Zambrano ruling, the
attention that was paid to it died down. The case also had less relevance in
Germany, as courts already interpreted national law in light of the ECHR’s
requirements.
In the UK, the media attention that was paid to the ruling was a bit more
pronounced (it resulted in three articles), and NGOs were also more aware of

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

the ruling. Interestingly, the administration adapted its handling of cases
immediately to the new legal background,33 even before an official policy
was devised in November 2012 (Sibley et al. 2013: 46).34 The UK government
also periodically reports the number of residences granted on the basis of the
Zambrano case. Between 1 March 2011 and 30 September 2013, 115 cases out
of 4,895 were granted on the grounds of the Zambrano case, with 1,470 still
pending.35 Another 105 cases36 were granted in England and Wales between
1 March 2014 and 30 September 2014. Up until May 2013, there were twenty-
seven rulings that cited Zambrano.37 The residence status that was ultimately
granted was partly based on Zambrano in only five of these cases; in all other
cases, its relevance was denied. One parent’s residence was generally deemed
sufficient for the child to be able to stay. In one case, the court referred to

32
Bleiberecht für ausländische Eltern: ‘Wenn Luxemburg keine Ausnahmen zulässt, wären
die Konsequenzen enorm’, Legal Tribune Online, 10 Mar. 2011, <http://www.lto.de/persistent/
a_id/2731> [accessed 10 May 2017].
33
UK Border Agency, Letter from Jonathan Devereux, Head of European and Nationality
Operational Policy to Sophie Barrett-Brown, Chair of the Immigration Law Practitioners
Association. Home Office, Derivative rights of residence—Ruiz Zambrano cases, 12 Dec. 2012.
<https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/257807/zambrano.
pdf> [accessed 9 May 2017].
34
The Immigration (European Economic Area) (Amendment) (No. 2) Regulations 2012
(SI 2012/2560) of Nov. 2012 includes the Zambrano criteria in § 15A.
35
Home Office, Mar. 2014: ‘Zambrano Cases Processed and their Outcome’, <https://www.
gov.uk/government/publications/zambrano-applications-processed-and-their-outcome-from-
2011-to-2013/zambrano-applications-processed-and-their-outcome-from-2011-to-2013>
[accessed 10 May 2017].
36
Home Office, 11 Feb. 2015: ‘People Granted the Right of UK Residence under the
Zambrano Ruling from 1 March 2014 to 30 September 2014’, <https://www.gov.uk/government/
publications/people-granted-the-right-of-uk-residence-under-the-zambrano-ruling-from-1-
march-2014-to-30-september-2014/people-granted-the-right-of-uk-residence-under-the-
zambrano-ruling-from-1-march-2014-to-30-september-2014> [accessed 9 May 2017].
37
The findings on judgments that reference the Zambrano case are based on a keyword
search on the website of the British and Irish Legal Information Institute. <http://www.bailii.org/
form/search_multidatabase.html>. Keyword ‘Zambrano’, May 2013.
 OUP CORRECTED PROOF – FINAL, 6/12/2017, SPi

Europeanization With and Against the Odds 193

British constitutional principles to grant residence.38 Since then, references to
the Zambrano ruling in court cases have become ubiquitous, with 297 UK
court rulings citing the case in early 2016.39 While ‘Zambrano carers’ receive a
residence and work permit, their access to social assistance was restricted in
November 2012, excluding them from child and housing benefit and income
support in order for them not to be treated on a par with EU citizens.40 Since
lack of resources may lead to a situation where the child will be deprived of
the benefits of their EU citizenship, the Zambrano ruling gave rise to several
more court cases. In February 2015, the Court of Appeal held in the case of
Sanneh and Secretary of State for Work and Pensions that ‘Zambrano carers’
do not have a right to non-discriminatory access to social benefits. Authorities
only have to ascertain that this does not impede the EU citizenship rights of

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

the child.
In France, a circular from September 2010, issued after the Metock judg-
ment, had already granted right of residence to the TCN parents of an EU
child (Fernhout and Wever 2011: 73). In the courts, there have only been five
cases up until May 2013 that have referred to the Zambrano ruling, which
served as a basis for residence in one of them.41
In the Netherlands, the government took a very restrictive approach in
implementing the consequences of the Zambrano case, even arguing that it
would suffice for grandparents to take care of the child, which the domestic
court rejected. The judicial division of the Council of State passed two judg-
ments in March 2012 elucidating guidelines for how the ‘genuine enjoyment’
of EU citizenship can be safeguarded (De Hart et al. 2012: 61–2); they hold
that it is sufficient for one parent to be present. According to the court, even if
this parent has problems caring for the child due to a medical or a psycho-
logical condition, there is no reason to grant residence to the TCN parent as
public assistance could be sought out (Fernhout 2012: 10). Currently, there is a
preliminary reference at the ECJ (Cases C-133/15 Chavez-Vilchez and Others),
which is concerned with TNC mothers who face expulsion from the Netherlands,
despite the fact that they are the primary carers of their children. Although
the fathers are partly absent or not really available to raise the children—some
also do not have custody—the authorities have argued that for the raising

38
England and Wales Court of Appeal (2012) EWCA Civ 1363. 23 Oct. 2012.
39
<http://www.bailii.org/form/search_cases.html>, keyword ‘Zambrano’, Jan. 2016.
40
NRPF (no recourse to public funds) network factsheet, Birmingham Community Law
Centre, <http://www.nrpfnetwork.org.uk/Documents/Zambrano-Factsheet.pdf> [accessed
10 May 2017]. Department for Work and Pensions, Oct. 2012, ‘Access to Benefits for Those
Who Have a “Zambrano” Right to Residence and Work’, <https://www.gov.uk/government/
uploads/system/uploads/attachment_data/file/220300/eia-zambrano-right-to-reside-and-work.pdf>
[accessed 10 May 2017]. The Social Security (Habitual Residence) (Amendment) Regulations
2012, No. 2587, <http://www.legislation.gov.uk/uksi/2012/2587/pdfs/uksi_20122587_en.pdf>
[accessed 10 May 2017].
41
Cour Administrative d’Appel de Lyon—11LY01612. 7 June 2012.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

194 The ECJ and the Policy Process

of the children it is sufficient for these, mainly Dutch, fathers to be present in
the country (Biersteker et al. 2015). It remains to be seen how these follow-up
cases will be decided upon. The Dutch case shows that, if governments are
reluctant to implement them, very few rights may follow from the Zambrano
ruling. However, it also shows that courts do impose constraints.
In general, we can thus conclude that member states respond to case law
beyond what one would expect from a contained-compliance perspective. The
report of the Fédération Internationale pour le Droit Européen (FIDE) on
citizenship rights, which documents reactions in twenty-two member states,
shows clearly how administrations and courts take up ECJ case law. New
member states, such as Bulgaria, may face difficulty in knowing how to apply
case law alongside secondary law (Kornezov 2014: 351). The report from

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Denmark is also of particular interest in our context, since the country takes
a very restrictive stance on immigration. As domestic courts rarely address the
ECJ, direct pressure through the judicial system is low (Jacqueson 2014: 476).
Nevertheless, the administration is responsive, as is apparent, for instance, in
the briefing note from the Ministry of Integration in May 2011, which clarified
that Zambrano grants residence rights if the child would otherwise be forced
to leave (Jacqueson 2014: 470). The official homepage also sets out these
guidelines in English.42

CONCLUSIO N

In the literature, it is a point of contention whether case law has implications

for member states’ policies outside of its codification in generalized policies.
This chapter has analysed member states’ reactions to two court cases, show-
ing that we find both rejections in the form of inertia, on the one hand, and
active responses, on the other. While the way that member states respond
cannot be explained in a monocausal way, it is apparent that research on
Europeanization needs to pay more attention to the effects of case law.
The German reaction to the Meilicke case is marked by inertia. Given the
potentially significant fiscal costs, the executive responded with a two-pronged
approach: legislative adaptations were made relatively early on to restrict the
retroactive impact of the case law. And the executive aimed to contain justice,
including waiting for further case law in the hope of more favourable fine-
tuning. This strategy paid off. With significant requirements for documentation,

42
See the official website of the Ministry of Refugee, Immigration, and Integration Affairs, ‘new
to Denmark’: <https://www.nyidanmark.dk/en-us/coming_to_dk/familyreunification/family_re
unification_under_eu-law/reopening_cases.htm> [accessed 9 May 2017].
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Europeanization With and Against the Odds 195

and the time that has elapsed, the impact of the case law is more manageable in
Germany. As case law may take a very long time to materialize within the
hierarchy of courts, member states can absorb its impact incrementally, allowing
them to cope more easily. Nevertheless, the general constraints that case law
puts on policy choices are grave and should not be underestimated.
Case law that reduces fiscal revenue is unambiguously negative for execu-
tives. Opposition to it is, therefore, not surprising. The Zambrano case
dealt with another salient policy field: the citizenship rights of third-country
nationals. However, the range of reactions demonstrates some proactivity. The
costs of responding were clearly not as high as they were in the tax case, and
there were benefits involved as well, given that human rights are politically
salient. Thus, the new Irish government explicitly wanted to present itself in a

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

different light to its predecessor. For some lawyers, litigants, and lower courts,
the Zambrano case opened up ‘a whole new world of EU law-argumentation’
(Shuibhne and Shaw 2014: 149). It led to significant judicial activity in many
member states, as it gave an incentive to use a ‘Union light at the end of the
national tunnel’ (Meduna et al. 2014: 284).
Executive reactions to the Zambrano case can appear puzzling. It is not so
much that there are striking differences in reaction among the member states,
but rather that member states react with a strange mixture of responsiveness
and resistance. One might have expected a reaction of contained compliance,
since TCNs are relatively weak litigants and political actors. However, there
are numerous examples where member-state administrations have published
circulars that respond to Zambrano or similar cases. On the one hand, this is
in response to the activities of lawyers and litigants, as well as the attempts of
lower courts to establish the precise meaning of new case-law developments.
The EU is a multilevel system, in which member-state governments have long
lost their gatekeeping role. On the other hand, executives operate within a
rule-of-law context, which results in administrative circulars that take up new
case-law developments, which is additionally the case where pressure from
domestic courts is low. There appears to be no fundamental difference
between when an administration adapts to a national high-court ruling or to
one from the ECJ. The implications of ECJ rulings are real, even in the case of
TCNs where political mobilization is restricted and there are no significant
costs to the executive from legal uncertainty. However, it is likely that there
will be a wish to keep follow-up court cases to a minimum in order to
constrain the evolution of further case law. As the many interventions at the
ECJ show, member states do not respond to case-law development, because
they endorse its thrust politically. Instead, they aim to contain it.
In the Zambrano case, this strategy was successful, as case law was not
extended to other family members. In addition, domestic courts appear to be
ambivalent about the ruling. They apply it, but often deny its relevance,
restricting its impact to the right of only one parent to stay in the relevant
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

196 The ECJ and the Policy Process

member state. If they act more generously, there is a tendency to link this
generosity to national constitutional principles or the ECHR.43
Although the Irish reaction to the Zambrano case appears to be strong at
first sight, it was not so different after all, given the extent to which Ireland was
actually affected by it. In political terms, the Zambrano case could be used to
demonstrate the pursuit of a more liberal approach after a change in govern-
ment. Nevertheless, taken as a whole, the differences among member states are
slight. While the administrations appear to operate in a multilevel system,
without differentiating between European and national high-court decisions,
governments and lower courts seem to regard the ECJ’s legitimacy to develop
such far-reaching case law as being more contested. Governments have inter-
vened steadily in a large number of cases. They have also signalled that the

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Citizenship Directive requires changes. The ECJ has fine-tuned its case law
in response. Additionally, lower courts have applied the Zambrano ruling
restrictively.
Finally, this discussion gives some insight into the motivations of individual
litigants. To pursue preliminary references takes a long time, and it is expen-
sive. To the extent that the Court is responsive to member states’ worries, as
some authors find (Carrubba and Gabel 2014; Larsson and Naurin 2016), it is
questionable why litigants pursue this route at all. However, the example of the
Meilicke case shows that some litigants’ interests can reach beyond the case at
hand. If tax lawyers initiate litigation this also allows them to promote their
services to potential clients. The same proves true if the cases themselves do
not secure individual gains but result in general legislative changes. While the
logic of collective action would suggest that many interests will not be pursued,
due to insufficient returns, there are added incentives because of the reach of
the ECJ’s rulings. In light of the erga omnes effect of the ECJ’s rulings, the
rationale to undertake litigation does not solely have to be supported by
the individual case, as the broader implications may well be worth pursuing.
The broader these effects are, given the size of the Union, the more likely it
becomes that there will in fact be actors who opt for litigation. A specialized
tax lawyer would gain from such a process through the advertisement it
provides for their company and through the opportunity to demonstrate
their legal skills. Large corporate actors may be in a position where the
likelihood of gains from an individual case justify the risks and costs of
litigation, and NGOs and lawyers offering pro-bono services may face still
different incentives.

43
A similar finding was identified by Stone Sweet and Stranz, who argued that the German
Federal Constitutional Court did not challenge the activism of the Mangold case, as this could
have meant that the German Basic Law was positioned behind EU law for the question of
fundamental rights (Stone Sweet and Stranz 2012).
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

The Europeanization Effects of Case Law

In Chapter 6, I argued that the literature on Europeanization focuses in a

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

one-sided fashion on the question of whether member states comply with EU
secondary law in general and the implementation (or merely transposition)
requirements of EU directives more specifically. The typologies within
this work overlook the specifics of member states’ reactions to case law. An
example of this is Radaelli’s differentiation between inertia (no domestic
response), absorption (limited response), transformation (far-reaching change),
and retrenchment (active resistance) (Radaelli 2003). The literature that dis-
cusses the implications of case law partly denies that member states are
responsive when social groups do not mobilize for their rights. Such opposition
could be seen in the previous chapter in relation to the Meilicke case, but the
resistance of the German government was also motivated by the wish to fine-
tune the case law. The Zambrano case, in contrast, showed that there may be
proactive responses that are also driven by the hope of avoiding further case law.
Typologies geared towards the specifics of the legislative process cannot capture
this kind of interaction between member-state governments and the Court. As
I argued in Chapter 6, anticipatory obedience (Blauberger 2014: 458) may be as
important as attention to non-decisions and indirect effects, as are instances
where member states find alternative means to realize policy objectives once
earlier policy options are foreclosed by the Court.
Compliance with secondary law requires the implementation of specific
policies, but case law cannot make comprehensive policy prescriptions.
Instead, it normally prohibits certain member-state policies by declaring
them to conflict with EU law. Following the Zambrano ruling, member
states are not allowed to expel the third-country national (TCN) parents of
children who are EU citizens if this deprives them of their EU citizenship.
Following the Meilicke ruling, member states are not allowed to treat cross-
border dividends differently from dividends from domestic shares. Thus,
case law also requires non-decisions from member states, leaving the exact
type of response much more open than secondary law does, due to its design
of certain policies.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

198 The ECJ and the Policy Process

The previous chapter showed that we cannot assume that member states
simply ignore case law that does not match their preferences, as the
contained-compliance perspective would argue. Case law, as my analysis of
the Zambrano and Meilicke cases shows, gives rise to multiple Europeaniza-
tion effects. In order to better understand its impact on the national level, it
is important to be aware of the variety of different possible reactions. It is
only on this basis that we can begin to understand them.
Since the Europeanization effects of case law have been researched to a
much lesser extent than secondary law, I will give an overview of the
different reactions, which I structure according to the different relevant
actors. While compliance with EU secondary law predominantly requires
the executive and legislature to act, responses to case law are not only

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

restricted to these actors. The domestic judiciary fosters or constrains the
domestic impact of EU law (G. Davies 2012), and the same is true for societal
actors that make active use of the alternative EU legal order. For the
empirical work, I partly draw on my own research and partly refer to existing
studies. Due to the relative neglect of the Europeanization effects of case law,
there is a somewhat German bias in the analysis. However, the traditional
literature on Europeanization that analyses the transposition of directives
also has a bias towards the large, old member states, with Germany, France,
and Britain disproportionately represented (Treib 2014).1 From an empirical
perspective, I will focus again on the four freedoms and citizenship rights.
I will only delve into anti-discrimination law for one example because of
the availability of a particularly interesting case study (Stone Sweet and
Stranz 2012). However, it should be clearly pointed out here that I do not
discuss the wide range of possible domestic responses to anti-discrimination
law; the same is true for EU competition law. Examples that could be given
here are the different liberalization measures for utilities but also, for instance,
the reform of the German public banks (Smith 2005; Seikel 2014a) and
procurement law.
In the following section, I start by giving a simple explanation of member-
state reactions to case law. I will then elucidate the cognitive foundation for
these responses. On this basis, I discuss examples of executive, legislative, and
judicial reactions, turning last of all to societal actors. The chapter concludes
by summarizing the variety of European responses to case law.

1
The larger membership of the EU and the increased heterogeneity of member states imply
that Europeanization effects will become less uniform. There are few studies on the new member
states. While their transposition of secondary law was assured under the conditions of accession,
studies have found this to result in a world of ‘dead letters’ (Falkner et al. 2008), where European
law hardly plays a role in day-to-day administrative practice.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Europeanization Effects of Case Law 199

EXPLAINING MEMBER STATES ’ REACTIONS

TO CASE L AW

Whether actors respond to a stimulus, we can assume, depends on the prior

recognition of the stimulus, and on the costs and benefits of reacting to it
(Conant 2002: 32). If actors neither comprehend the supremacy and direct
effect of the Treaty nor realize that the Court’s interpretation of the Treaty in
an individual case has erga omnes effect for all actors that are bound by the
Treaty, we cannot expect any response. This cognitive aspect is not trivial, as
the history of the EU contains many examples of actors having first of all to
learn of the Treaty’s direct relevance to them.
Once actors become aware of the relevance of case law, their reactions are

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

mediated by different interpretations of its reach, given legal uncertainty. Case
law serves as an opportunity structure. It affects the power balance between
different coalitions of actors at the domestic level. By favouring the position of
certain actors over others, they can advance their interests more easily in the
domestic policymaking arena.
What determines actors’ responses? Put abstractly, whether a response
characterized by inaction (namely, the policy status quo) will be an actor’s
preference depends on the related normative or material costs, as well as the
benefits, that are attached to either maintaining the status quo or changing
policy instruments. The costs of maintaining the status quo, for instance,
may relate to further case-law development that actors might seek to evade;
to enforcement from the Commission demanding a generalized response;
to the costs of unequal access to rights based on case law; or to the cost of regime
competition, which results from a broad interpretation of the fundamental
freedoms. There are, therefore, direct responses that either implement or
resist judgments, as well as indirect responses to the effects of judgments,
most notably regime competition. A central finding of the research on
Europeanization is all the more relevant to the impact of EU case law, namely
that there is no single dominant independent variable, such as policy misfit, that
can explain member states’ reactions (cf. Treib 2014). Different reactions,
therefore, can be explained successfully ex post, but generalizations remain
difficult.
The possible costs of a response may be that the original status quo receives
broader societal backing and/or provides a better solution to a policy problem.
However, the opposite may also be true if the status quo is only supported by a
powerful minority veto. Needless to say, the different kinds of relevant actors
rarely share the same set of preferences. For some actors, there are costs of
maintaining the status quo, and for others there are costs of reform.
With this general explanation in mind, the rest of this chapter will sum-
marize empirical findings on member states’ reactions to case law, paying
OUP CORRECTED PROOF – FINAL, 6/12/2017, SPi

200 The ECJ and the Policy Process

specific attention to the four fundamental freedoms. Much more than is
the case for implementations of secondary law, responses to case law (or the
anticipation or avoidance of further case law from the Court) also include
strategic considerations. It is a response to an external actor rather than the
implementation of joint decisions, and all three branches of government
are relevant alongside societal actors.
When considering actors’ reactions, it is important to keep in mind that
case law may demand either decisions or non-decisions, or it may lead to
indirect effects. Because of the liberal bias of the fundamental freedoms,
actors can overcome national regulatory restrictions and choose the most
permissive regulatory order within the EU. Regime competition is a relevant
indirect response here, as long as the interpretation of the fundamental

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

freedoms does not prohibit reverse discrimination against nationals, which
allows member states to preserve restrictive national regulations. Since
member states remain politically responsible for their domestic policies,
while they are constrained by EU law (Scharpf 2009), there is an incentive
to use institutional equivalents as escape routes to pursue their policy goals.
The institutional equivalents that are available may include pacts with
societal actors or contracts with member-state governments outside the EU
legal regime, as we will see later.
While typologies geared towards implementation differentiate between
degrees of compliance (Radaelli 2003), such as absorption, accommodation,
or transformation (Börzel and Risse 2007), the greater variety of responses
to case law leads me to distinguish indirect from such direct responses. Indirect
reactions include regime competition or measures to compensate for lost policy
instruments. Direct responses may require decisions or non-decisions, and they
can be characterized by resistance, accommodation, and also anticipatory
obedience. Figure 7.1 summarizes these direct and indirect effects.

Direct responses (decisions and non-

decisions): resistance, accommodation,
anticipatory obedience

Case law

Indirect responses: regime competition,

compensatory measures

Figure 7.1. Direct and indirect responses to case law

 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Europeanization Effects of Case Law 201

THE COGNITIVE DIMENSION OF RESPON SES

TO CASE L AW

If actors’ responses to case law depend on the way that they perceive the costs
and benefits that arise from a given ruling, how actors interpret the relevance
of case law is decisive. Compared to the implementation of secondary law,
there are not only more diverse reactions to case law from more actors to be
kept in mind, but the cognitive perceptions of the domestic legal requirements
also differ to a larger extent. As I have argued, the implications of case law are
marked by a higher degree of legal uncertainty, as it is case-specific and actors
have not participated in the legislative process that designs the rules. Actors’
reactions are always mediated by their world views. This is particularly

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

significant for the impact of case law, which is very open to interpretation.
In not having complete information about all the relevant developments
that influence the realization of their preferences, actors always act with
uncertainty. However, in comparison to EU secondary law, the implications
of case law are much more uncertain, which makes the way that relevant
actors interpret case law decisive. To give an example, the UK government
long disputed the relevance of the case law on the freedom of services as it
applied to its National Health Service (NHS). After all, there are no market
transactions in the NHS, since it is a pure benefits-in-kind system (Obermaier
2009: 137). The difficulty of analysing to what extent national reforms can be
seen either as responses to case law or to be domestically motivated is related
to the uncertain implications of rulings. To stick with the same example, the
introduction of some elements of marketization into the NHS was regarded by
some as a response that had been induced domestically to tackle the politically
contentious long waiting times patients had to endure. However, there are also
clear indications that the government was under pressure to accommodate the
ECJ’s case law in this way (Obermaier 2009: 130–1).
Given that the implementation requirements of case law are, in general, more
open to interpretation than secondary law, we should expect to find significant
differences between member states. As cognitive differences may under- or
overestimate the pressures of EU law, it may be empirically difficult to distin-
guish cognitive aspects from preferences. Actors may take up policy instruments
that are no longer available under EU law or they may not be aware of EU legal
constraints. On this basis, we will turn now to the responses of different actors.

ACTORS ’ REACTIONS

In the following section, I will begin by discussing executive reactions to case

law, before going on to consider the legislature and the judiciary. Last of all,
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

202 The ECJ and the Policy Process

I will pay attention to societal actors. Needless to say, in empirical cases there
will often be an interplay between these different reactions from actors.
Discussing them separately will increase awareness of the variety of different
responses. I will distinguish between accommodation, resistance, and, where
relevant, the indirect effects of compensation and regime competition.

E X E C U T I V E RE A C T I O N S

Member-state governments can accommodate EU case law either through

administrative guidelines or by initiating legislation. Alongside direct responses,

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

there is the indirect option of compensation, which draws on alternative means
of action in order to alleviate the pressure of European case law. Increased
administrative controls or bilateral agreements may help to counter, for instance,
incentives for illegal activities that result from liberalization. Alternatively, the
government may foster private modes of governance to compensate for legal
restrictions. Bearing in mind that the executive is broadly involved in domestic
reactions, I will begin by discussing executive responses in the narrow sense.
Reactions will be costly if the case law does not happen to push for reforms
that correspond to the policy preferences of the executive. However, as we saw
in the previous chapter, conforming to the supranational demands of the rule
of law is a consideration of its own. In the extreme example of the Meilicke
case, where a ruling imposed fiscal costs, the executive may aim for concret-
ization or fine-tuning in a subsequent ruling. Nevertheless, a case like Zam-
brano, which had less pronounced costs, shows that violating the rule of law is
not an easy option for executives. In addition to governmental preferences, the
kind of pressure that is applied is significant, whether this is internal (from
domestic litigants and courts) or external (from the Commission, regulatory
competition, or evolving case law).

Administrative Accommodation

If EU case law cannot be contained, member states have to respond by

accommodating the case law, safeguarding their original policy goals as
much as possible. Case law is not subject to the same need for transposition
as is the case for European secondary law. By publishing an administrative
circular, executives can order their administrations to change their adminis-
trative practice, doing justice to the rule of law in the multilevel system. The
Irish reaction to Case C-34/09 Zambrano (8 March 2011), which I discussed in
the previous chapter, is a case in point, demonstrating such changes to
administrative practices.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Europeanization Effects of Case Law 203

One advantage of accommodating case law is that additional European case
law can be avoided. Obermaier (2009: 136–7) found that, following the ECJ’s
case law on health services, British authorities began a generous administrative
practice of reimbursement for health services from other member states in
order to avoid further ECJ rulings.2 Such a strategy is only plausible when
domestic institutional conditions are so particular that the case law that
originates from other member states will not have the same impact. However,
we saw in the follow-up case law for the Zambrano case, and also for Meilicke,
that further court cases may also lead to fine-tuning and a deradicalization of
the implications of case law, which can produce incentives to provoke further
rulings by the ECJ.

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Residence rights of third-country national family members
The executive reaction to Zambrano was not exceptional, as can be demon-
strated when one looks into member states’ reactions to the Court’s case law
concerning the residence rights of third-country national (TCN) family mem-
bers of EU citizens (De Somer 2016). The residence rights of TCNs are
restricted by many member states, for instance as a measure to combat
marriages of convenience. In 2003, the ECJ argued in Case C-109/01 Akrich
that the right of a TCN family member to accompany an EU citizen depended
on the prior legal residence. However, soon afterwards, in 2008, in Case
C-127/08 Metock, which was a preliminary ruling from Ireland, the ECJ
ruled against this requirement. Against interventions from ten member states,
the Court argued that the Community, and not the member states, is author-
ized to regulate the conditions of lawful residence for TCN nationals when
they join EU citizens who have taken advantage of their right to freedom of
movement. As the Treaty requires a cross-border element, immobile member-
state nationals cannot rely on EU law.
The Metock case is highly interesting, as it provides a rare example in the
literature in this area where the national consequences of case law have been
discussed. After Akrich, Denmark, the UK, the Netherlands, and Finland
formulated the requirement that prior lawful residence was a precondition
for family reunification (Thorp 2009: 5). Also after Metock, the conditions for
family reunification were eased in Denmark (Wind 2014: 169–70), and the
number of individuals that were granted the right to family reunification
tripled in 2009 in comparison to 2008 (Martinsen 2011: 958). In Germany,
administrative instructions from the federal government issued on 29 July
2009 referred to Metock and lifted the general requirements, such as basic
knowledge of German, for this group of people. Additionally, Italy and Ireland

2
I thank Michael Blauberger for pointing this out to me.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

204 The ECJ and the Policy Process

swiftly adapted their legal and administrative practices, as did Finland and the
UK (Fernhout and Wever 2011: 15, 54–5; Costello 2009). UK courts even
extended eligibility for the Metock-type rule to non-dependent family mem-
bers (Fernhout and Wever 2011: 62).
Since the Court interprets restrictions on EU citizens’ TCN family members
as constraints on EU citizenship, EU citizens may be privileged in cases of
family reunification in comparison to nationals. The Citizenship Directive
codifies these privileges (for instance, Article 20 and Article 16(2)). In their
transposition of the Citizenship Directive, several member states have explicitly
agreed to reverse discrimination against their nationals (the UK, Denmark,
Germany, Sweden, the Netherlands, Austria, Slovenia, Ireland, Poland, and
Estonia). Others have opted to grant their nationals the same right (Spain,

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Portugal, Italy, Hungary, Malta, Cyprus, Finland, Croatia, and the Czech Republic)
(Shuibhne and Shaw 2014: 74).
When opting in favour of or opposition to discrimination against nationals,
member states face a trade-off between pursuing their domestic policy objectives
against interference from the ECJ and their responsibility for the well-being
of their own citizens. As AG Sharpston put it in her opinion on Cases C-456/12
and C-457/12:

Why a Member State would wish thus to treat its own nationals less favourably
than other EU citizens (who, except for their nationality, might very well be in
identical or similar circumstances) is curious. So is the fact that, by denying
residence, that Member State might be at risk of de facto ‘expelling’ its own
nationals, forcing them either to move to another Member State where EU law
will guarantee that they can reside with their family members or perhaps to leave
the European Union altogether. (No. 86; Shuibhne and Shaw 2014: 150)

Reverse discrimination against nationals may not only be politically insensi-

tive, but may also be constitutionally problematic to regulate nationals in
a stricter way than migrants. Regulation restricts personal freedoms, and
constitutional law may require that there are sufficient reasons for such a
restriction. If EU nationals can use the fundamental freedoms to become
economically active under laxer rules, a restriction of rights is hard to justify.
Courts in Austria, Italy, and France have found constitutional problems in
reverse discrimination against nationals (Tryfonidou 2009: 121–2). Conse-
quently, Italy is one of the countries that has opted to oppose discrimination
against nationals by extending the benefits of family reunification for the TCN
family members of EU citizens to nationals (Fernhout and Wever 2011: 57–8).
If this route is taken, case law simultaneously changes domestic regulations
and EU rules. As a result, in Italy, the implications of the ECJ’s case law on
migration are much more far-reaching.
As I discussed in Chapters 5 and 6, reverse discrimination against nationals
gives incentives for citizens to take the ‘Malmö route’ or do the ‘Dublin hop’,
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Europeanization Effects of Case Law 205

with Danish or UK families moving abroad for a limited time to profit from
more favourable regulations for EU citizens (Martinsen 2011: 957). This raises
questions concerning the necessary time that must be spent abroad in order
to qualify for EU legal benefits. In the aforementioned Cases C-456/12 and
C-457/12, the Court held that neither the passive services freedom nor regular
visits suffice, but that residence for more than three months in another
member state is required.

Child benefits
Another example of prompt administrative accommodation of case law con-
cerns the German payment of child benefit for children living in other

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

member states. Regulations 883/2004 (formerly 1408/71) and 987/2009 out-
line which member state has to pay this social benefit. In a nutshell, the
member state where someone is employed and pays taxes is responsible for
paying child benefit. Otherwise, the country of residence takes on the
responsibility.
The ECJ required Germany to pay child benefit to seasonal and posted
workers, who were paying taxes in Germany, for children living in their home
member states. The key issue in Cases C-611/10 and C-612/10 Hudzinski and
Wawrzyniak, which were decided upon in June 2012, is that, according to
regulation 883/2004, the home state would be responsible for paying benefits
to seasonal and posted workers, and Germany would have no responsibility.
Because Germany gives benefits to everyone subject to income tax, irrespective
of the country of residence, it could not exclude those receiving benefits
already in another member state. The Court held that Germany could go
beyond the provisions of the regulation as this was benefiting workers, but
excluding all those being eligible for benefits in other member states was
infringing the free movement principle. With this ruling, the ECJ changed
its earlier interpretation of an exclusive competence of one member state,
based on the social security coordination regulation, to an orientation on the
most favourable solution with view to the free movement principle. By
changing its case law, the Court undermined the regulation’s attempt at
coordination (Weimar 2014: 1, 3).
The ruling attracted significant media attention. The rate of child benefit
in Germany is relatively high, at €184 per month for the first two children,
€190 for the third, and €215 for the fourth and all further children. This is in
the range of the median monthly income in Central and Eastern European
member states, which is €273 in Bulgaria and €176 in Romania (Blauberger
and Schmidt 2014: 5). The government set aside €200 million in November
2012 as a compliance cost for the judgment (Bundesministerium der Finanzen
2012a). In the highly politicized debates on welfare migration in 2014, the cost
of child benefit for EU children not living in Germany was often put
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

206 The ECJ and the Policy Process

at €1 billion, since child benefits can be claimed for four years retroactively.3
The number of administrative positions dealing with these claims had to
double between 2013 and 2014.4
In order to prevent abuse, the government changed the law to include the
requirement for a tax identification number and more extensive proof of
the existence of children in other member states. Interestingly, however,
in the political discussion there has been no mention of the fact that, according
to the European social regulations, Germany could also deny its responsibility
for claims from seasonal and posted workers. The Federal Fiscal Court, in its
judgment after Cases C-611/10 and C-612/10 Hudzinski and Wawrzyniak,
openly stated that it was up to the legislature to change the law.5 To the extent
that posted and seasonal workers claim child benefits for children living in

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

other EU member states, the generous benefits are a cross-subsidy for poorly
paid jobs (Schelkle 2017: 262).
In line with the generally politically contentious discussions about welfare
migration (Blauberger and Schmidt 2014), the conditions for eligibility to
child benefits has become important in other Western member states as
well. For example, the UK’s EU referendum debate focused on this issue,
with the Conservatives aiming to ban EU migrants’ access to child benefits,
along with other social benefits, for four years.

Executive Resistance

The discussion so far has shown that governments do follow case law, even if this
imposes costs and requires greater administration, which is very different to
Conant’s finding (2002: 69) that administrations are unlikely to introduce
broader changes. However, if their policy preferences largely diverge from the
demands of case law, the costs of opposing the demands of EU law and violating
the general principles of the rule of law may appear lower than the costs of
implementing policy choices that are not preferred. This is particularly the case if
legal uncertainty gives competing interpretations the benefit of the doubt. An
example that illustrates executive resistance particularly well is the language test

3
Tagesschau, 13 May 2014, ‘Verwirrspiel mit geschätzten Zahlen. Kindergeld für EU-
Saisonarbeiter’, <https://www.tagesschau.de/inland/kindergeld132.html> [accessed 9 May 2017].
Spiegel online, 12 May 2014, ‘Anspruch von EU-Ausländern: Eine Milliarde Kindergeld für
Saisonarbeiter’, <http://www.spiegel.de/wirtschaft/soziales/kindergeld-fuer-eu-auslaender-
deutschland-muss-milliarden-zahlen-a-968828.html> [accessed 9 May 2017].
4
BMF, 25 June 2014, Letter to MP Franziska Brantner, on file with author.
5
‘Somit obliegt die Entscheidung, den Kindergeldanspruch solcher als unbeschränkt ein-
kommensteuerpflichtig behandelter Personen, die gemäß Art. 13 ff. der VO Nr. 1408/71 den
deutschen Bestimmungen über soziale Sicherheit nicht unterliegen, durch Änderung oder
Aufhebung des § 62 Abs. 1 Nr. 2 Buchst. b EStG zu versagen, allein dem Gesetzgeber.’ BFH-
Urteil 16 May 2013, III R 8/11 (published 28 Aug. 2013).
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Europeanization Effects of Case Law 207

for TCN family members that some member states, such as the Netherlands and
Germany, use as a requirement before residence is permitted (De Somer 2016:
233–4). It is important to note that this example is based on secondary law,
namely on the Family Reunification Directive (2003/86/EC), which regulates the
right of TCNs living in the EU to have their family members join them, if
they have sufficient financial means and health insurance and pose no threat
to public order. The directive gives member states the option to demand
integration measures according to national law (Article 7(2)). Legally and polit-
ically, it is contentious whether language tests count amongst these measures, in
particular as member states like the Netherlands, France, and Germany often
subject only some TCNs to this requirement, while exempting others, such as
citizens of the USA. I am including this case, despite its legal basis in secondary

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

law, as it is a notable example of executive resistance.
An interesting case in this context is C-155/11 Imran, which concerned an
Afghan woman with eight children who wanted to join her husband in the
Netherlands. While her children could move to the Netherlands, she was
denied residence after failing the test. Following legal action, the case reached
the ECJ in 2011. For the Dutch government, but also for the governments in
Germany, Austria, and France, this posed a threat to their attempts to make
family reunification for TCNs subject to more demanding integration require-
ments. Shortly before, in Case C-578/08 Chakroun (March 2010), the Court
had held that such measures are not allowed to undermine the rights of the
Family Reunification Directive. Although the processing of Mrs Imran’s
application had been going on for almost two years, within days of the case
reaching the ECJ, the responsible Dutch minister used his discretion to grant
the residence permit in order to avoid the constraints of a court case (Arcarazo
and Geddes 2013: 189–91). Case C-513/12 Ayalti was also cancelled; it con-
cerned a reference from Germany that was withdrawn after the woman
involved in the case passed the language test. Here German inaction was
certainly assisted by the domestic courts. The German Federal Administrative
Court, as well as the German Constitutional Court, rejected a preliminary
proceeding in 2010 and 2011,6 with the Federal Administrative Court arguing
that the question was sufficiently clear and fell under the acte clair doctrine.
During this time, the Commission emphasized in 2011 that it regarded the
language test as a violation of EU law.7 In the Bundestag, the Social Demo-
cratic and Green party groups launched a legislative proposal to abolish the
language-test requirement that had been introduced in 2007 (BT Drs. 17/8921;

6
BVerwG, Urteil of 30 Mar. 2010, Az. 1 C 8.09; BVerfG, Beschluss of 25 Mar. 2011, Az. 2 BvR
1413/10.
7
Der Tagesspiegel, 1 Aug. 2011: ‘Sprachtest darf Zusammenführung von Familien nicht
verhindern’, <http://www.tagesspiegel.de/politik/integration-sprachtest-darf-zusammenfuehrung-
von-familien-nicht-verhindern/4453702.html> [accessed 9 May 2017].
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

208 The ECJ and the Policy Process

BT Drs. 17/1626). Members of the ruling Christian Democrat/Free Democratic
(liberal) coalition argued in the parliamentary debate that the test facilitates
integration and enhances women’s emancipation. Predominantly female Muslims
have made use of subsequent family migration. Knowledge of the national
language facilitates their societal integration.8
Member states could not ultimately prevent a Court ruling. In 2014, the ECJ
ruled upon Case C-138/13 Dogan, which was a reference from the Berlin
Administrative Court involving a Turkish woman, who was illiterate and
had failed the test (§ 30 Abs. 1 Nr. 2 AufenthG). Drawing in part on the
Association Agreement with Turkey, the ECJ demanded that cases be
assessed individually, and found an automatic denial of residence to be out
of proportion with the policy goals of achieving integration and hampering

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

forced marriages.
In Germany, a member of the Bundestag from Die Linke, Sevim Dagdelen,
who herself has Turkish roots, has been particularly actively engaged in the
question of language tests for TCN family members, repeatedly initiating oral
requests for information about the actions of the German administration in
response to the constraints of EU law (BT Drs. 17/14046; BT-Drs. 18/937;
interview Federal Ministry of Economics, 12 February 2013). While the
government has introduced administrative adaptations, meaning that a failed
language test no longer entirely prohibits migration into Germany, TCN
family members must show that they have made efforts to pass the integration
test for at least one year. Following the Dogan case, the Commission addressed
the German government with a pilot procedure (3395/12/ELAR), requesting
information on how the government implements the ruling, which led to an
official letter from the Commission.9 While the government responded to the
Dogan case with an administrative circular introducing a proportionality test
in August 2014, it continued to support the language requirement. Exemp-
tions are only granted by the Foreign Office directly. Thereby, the German
government resisted, as much as possible, the pressure from EU law (BT Drs.
18/4598, pp. 12–13).
More recent cases may have settled this contentious matter. In Case C-579/13,
mandatory integration measures for TCNs, who have long-term residence
status in the Netherlands, were passed by the ECJ, as long as these measures
are proportionate and do not hinder the aims of the Family Reunification
Directive. Case C-153/14, which was also decided upon in 2015, concerned the
Dutch language test, which also examines knowledge about the country. The
Court allowed member states to use the requirement of a language test if it is

8
Plenarprotokoll 17/240, 16 May 2013, pp. 30337–43. <http://dipbt.bundestag.de/doc/btp/
17/17240.pdf> [accessed 9 May 2017].
9
BT Drs. 18/4001. <http://dip21.bundestag.de/dip21/btd/18/040/1804001.pdf>, pp. 23–4.
BT Drs. 18/4598, p. 16.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Europeanization Effects of Case Law 209

proportionate and pays attention to the specificities of individual cases. The
German legislator subsequently included the possibility of adding exceptions
into the law (Thym 2015). The issue of language tests for TCNs appears,
therefore, as a further example of fine-tuning in which member states can
reconcile their policy priorities with case law by emphasizing their policy
preferences and resisting the pressures of EU law. As I have mentioned,
however, this example is based on secondary law provisions. For the Court,
therefore, the threat of override is stronger than is the case when it is interpret-
ing the Treaty. This may be the reason why I am not aware of a comparable
example for successful executive resistance to the interpretation of the Treaty.
Executives not only have the option of resistance as a means to pursue their
own policy preferences. They can also seek compensatory measures, to which

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

I now turn.

Executive Compensation

Europeanization through case law not only requires specific acts of imple-
mentation. Some policy options are foreclosed, non-decisions are required,
and governments may have to seek out alternatives. In his analysis of member-
state reactions to the case law on golden shares, Werner (2016: 14) speaks of
‘autonomy-protecting equivalents’. The example of the freedom to provide
services, and its impact on Germany, is interesting in this respect, as it shows
how member states can try to compensate for the implications of European
law. After the Eastern enlargement in 2004, Germany had used the option of
postponing the free movement of workers for 2 + 3 + 2 years until May 2011.
However, soon after the accession of new member states in 2004, it became
clear that the posting of workers from Eastern Europe posed a significant
challenge, as it fell under the freedom of services provision and the Posted
Workers Directive. Different to the free movement of workers, where workers
are integrated into the regulatory environment of the new home state, under
the freedom of services provision, this state is the host state, and the pay and
labour conditions of the home country apply. Posted workers can only take up
work on a temporary basis, interpreted as up to two years, in the host country
and are not allowed to be fully integrated into the work processes there;
otherwise, the provision for the free movement of workers would be relevant.
Important here is that Germany did not have a general minimum wage until
2015 but relied heavily on collective agreements. While the host country’s
legally binding minimum wages apply to posted workers, collective agree-
ments cannot be enforced. They only commit the relevant parties to the
agreement. As they do not cover all employers in the host country, it would
be discriminatory to require them from posted workers.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

210 The ECJ and the Policy Process

With an average German hourly wage, inclusive of social-security contribu-
tions, of €26.90 in 2004, Eastern enlargement applied great pressure as a result of
posted workers. At the time, the average Polish wage was €4.74 (European
Commission 2008: 133). This competitive pressure resulted in legislation for
different sector-specific minimum wages—these are discussed later in greater
detail as a legislative reaction—which preceded the general German minimum
wage that was introduced in 2015. However, this was not the only reaction.
The freedom to provide services through posted workers also facilitated fraud,
as the regulation and administrative control of this area is the responsibility of
the home country, not the host country where services are provided, and
where the costs of non-compliance accrue (Schmidt 2007, 2009). In addition,
this complicated division of labour between the different member-state

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

administrations was ill understood. The Eastern European administrations
responsible were quoted in the press as claiming that it was not their duty to
control their companies’ activities in the West.10 Instances of fraud were
reported, as mere letterbox companies were posting workers on a permanent
instead of a temporary basis; as such, these workers were fully integrated into
the world of work and were thereby replacing regular workers. Maximum
working times and minimum wages were often violated, and workers were
crowded into camp accommodation that was overpriced, to cite some of the
complaints (Interview Ver.di, 30 March 2006, Berlin). Even where detected,
German authorities often could not act on fraud, as responsibility lay with
the home state. The ECJ emphasized in a ruling that the host state even had to
accept certificates (E-101 for posted workers) that were obviously forged, due
to the rules on administrative cooperation in the EU. Germany, as a host state,
would need to address the home state directly through an infringement
procedure brought before the ECJ (C-2/05, 26 January 2006). Consequently,
we have a striking example here of an instance where case law required a
non-decision. Germany could no longer implement certain controls and was
deprived of important policy instruments. In this situation, Germany reacted
with compensatory measures.
Given the formal administrative responsibility of the home state, Germany
bolstered its administrative controls on illegal activities. The ECJ (C-490/04,
6 October 2007) allows member states to examine some translated documents
(labour contract, details of wages, and evidence of working time and the
payment of wages). In late 2003, the previously separate controls that were
carried out by the labour administration and customs were merged in the
Finanzkontrolle Schwarzarbeit (FKS; financial control of illegal employment).
In mid-2004, the law to intensify the control of illegal work and tax evasion
(SchwarzArbG) bundled together and harmonized existing rules. In early

10
Taz (13 Nov. 2003), Das miese Geschäft mit den Schlachtern.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Europeanization Effects of Case Law 211

2005, the Federal Ministry for Economics and Labour and the Federal
Ministry of Finance built a joint taskforce to control abuse of the cross-
border freedom to provide services and the freedom of establishment.11 In
addition to intensifying the cooperation of the state and federal administra-
tion, its task was to enter into bilateral negotiations with the administrations of
other member states. A new law required that all E-101 certificates for posted
workers had to be sent to the central office for old-age insurance in Würzburg,
which was tasked with detecting forged certificates (BT-Drs. 16/5098, p. 18).
Moreover, the Federal Ministry of Finance published a guide to the rules
concerning the exercising of the freedom to provide services and the freedom
of establishment (Bundesministerium der Finanzen 2006).
In addition to strengthening its domestic administration, Germany engaged

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

in regular bilateral talks with old (Denmark, the Netherlands, and Austria)
and new (Poland, Hungary, the Czech Republic, and Romania) member states
in order to reach a common understanding of the conditions for the legal
posting of workers, as well as on the criteria for determining their abuse, for
instance through letterbox companies. It has to be emphasized that the ECJ
has shown little concern about abuse. ‘Therefore, in conclusion the free
movement rights generally protect the use of letterbox companies and have
done more to promote their use than to limit it’ (Sorensen 2015: 93). One goal
of the talks was to reach an agreement on the joint control of illegal activities.12
It is interesting that member states revert to bilateral or multilateral agree-
ments not only in the absence of political consensus at the European level but
also in view of legal difficulties.
Finally, the problems that arose from the posting of workers within the
context of the Eastern enlargement also gave rise to legislative negotiations at
the EU level. The revision of the Posted Workers Directive (2014/67/EU),
which also stood in the context of the Laval and Viking rulings, was agreed in
spring 2014, and it includes specifications for the ‘genuine’ establishment and
posting of workers in order to counter fraud (Article 4). The directive,
moreover, pays attention to the problem of subcontracting liability, allowing
member states to introduce non-discriminatory measures of control (Sorensen
2015: 99–100). With another proposal for a revision tabled in 2016 (COM
(2016) 505 final), difficulties appear ongoing.
In summary, it is not unusual for executives simply to comply with the
demands of supranational law. They do this in areas where one can assume
that they have other policy preferences too. We also find an example of

11
‘Task Force zur Bekämpfung des Missbrauchs der grenzüberschreitenden Dienstleistungs-
und Niederlassungsfreiheit.’
12
Interview BMAS (29 Mar. 2006); interview BMWi (29 Mar. 2006); Die Welt (12 Apr. 2005),
‘Lohndumping: Regierung verhandelt mit Polen’; Financial Times Deutschland (26 Apr. 2005),
‘Dienstleistungen: Rüge aus Polen’; Bundesministerium der Finanzen 2008.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

212 The ECJ and the Policy Process

successful resistance in the case of the language test for TCN family members.
This success most likely hinges on the fact that member states generally
comply with rulings, so instances of resistance are taken seriously by supra-
national actors and are consequently met with responses. Given that the
legality of the language tests was based on secondary law, and not on the
Treaty, the more credible threat of legislative overrule is likely to have helped
the cause of member states’ executives. Lastly, we also find indirect responses
in the form of compensation.

LEGISLATIVE ACCOMMODATION

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Given the parliamentary systems of EU member states, reactions from the
legislature that are not supported by the executive hardly ever occur. Legisla-
tive reactions will normally be prepared by the executive, but the executive also
reacts by drawing on its administrative powers, as we just saw, which makes it
valid to discuss legislative reactions separately.

The Reform of Limited Companies

A notable example of legislative reform is the response to regulatory compe-

tition that resulted from Case C-212/97 Centros. In the case, the ECJ ruled that
a Danish couple who had founded a limited company in the UK, solely for the
purpose of circumventing stricter capital requirements in Denmark, could
rightly demand the authorization of their branch office in Denmark. As a
result, the ECJ affirmed that the freedom of establishment protects regulatory
arbitrage (Steindorff 1999). Before we look in greater detail at this instance
of company reform, it is important to ask what kind of Europeanization
response we can identify here. With the ruling on the Centros case, the ECJ
did not impose requirements on member states to react in a certain way. It
simply allowed regime competition, by enforcing mutual recognition. It is,
thus, an indirect Europeanization effect and does not fall within notions of
compliance. Company law regulates how companies can be established and
operated. These rules are important for the different stakeholders, the man-
agement, shareholders, creditors, employees, and also consumers (Blauberger
and Krämer 2014: 788).
Following the Centros ruling from the ECJ, the establishment of British
limited companies became very popular in Germany and also in other mem-
ber states. Different to the US, in the EU, competition for incorporations is not
fuelled by franchise taxes, but there are political interests in not losing control
over the regulation of companies employing and trading in the member state
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Europeanization Effects of Case Law 213

(Becht et al. 2008). The status of being established as a limited company does
more than allow the bypassing of German co-determination rules for large
companies. More relevant here are liability issues, which are settled outside of
company law in Britain. By transferring only parts of a regulatory system, in
this case rules of establishment, regulatory gaps may result. As the German
rules on the liability of limited companies do not apply to limited companies
established in the UK, these companies are insufficiently regulated in Ger-
many (Horn 2004: 900).
In response to the significant regime competition fostered by the Centros
ruling, several member states lowered the minimum legal capital requirements
in their legislation. Blauberger and Krämer (2014: 794) list the following
reforms: France 2003/2008; Finland 2006; Germany 2008; Poland 2009;

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Denmark 2010; Sweden 2010; Bulgaria 2010; Portugal 2011; the Netherlands
2012; Italy 2012; Greece 2012; Austria 2013; Luxembourg 2013; and the Czech
Republic 2014 (see also Bratton et al. 2009; Ringe 2011). In Germany, capital
requirements to establish a limited company were originally set at €25,000.
After the Centros ruling, consulting firms offered help in founding limited
companies in Germany and disseminated information about this possibility
through the internet (Höpner and Schäfer 2010: 360).
The German MoMiG reform13 relaxed requirements in 2008, introducing
the option of a ‘mini-GmbH’ that can be established with upwards from €1,
but 25 per cent of annual profits have to be put aside until reserves of €25,000
are reached. In addition to rapidly evolving regime competition, there were
also domestic motives for the reform. Partially as a result of this legislative
response, regime competition has become less of an issue in German company
law. The costs of being established in the UK as a private company limited by
shares became more widely known. While these companies are quick and easy
to establish, they are comparatively difficult to maintain as they have signifi-
cant yearly reporting and testifying duties. Consequently, the Chamber of
Crafts in Cologne reported in late 2010 that of the 220 limited companies that
had originally been established in the area, only 78 continued to exist. The
issue is that a company is automatically erased from the register if it fails to
comply with their reporting duties. Once this happens, owners are personally
liable and find themselves in the exact situation they wanted to avoid by
founding a limited liability company in the first place.14
In their analysis of company law, Blauberger and Krämer (2014: 795) argue
that the reform of limited companies is an extreme case of regulatory compe-
tition, and also of legislative response. It is rare that case law triggers such a
race to the bottom.

13
Gesetz zur Modernisierung des GmbH-Rechts und zur Bekämpfung von Missbräuchen
(MoMiG).
14
Handwerksblatt, Nov. 2010: Good bye limited, Tschüss Geld.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

214 The ECJ and the Policy Process

The Posting of Workers

Earlier in this chapter, I discussed German attempts to counter the increase in

fraud following the posting of workers after the Eastern enlargement through
administrative controls and bilateral agreements.
In Germany, the posting of workers brought the introduction of minimum
wages onto the agenda since, in contrast to collective agreements, they can be a
requirement for posted workers. After Eastern enlargement, the pressure on
low-skilled German jobs was significant. The meat industry union complained
in 2005 that one-third of all jobs had been replaced by posted Eastern
Europeans, which amounted to 26,000 jobs.15 The accession of Bulgaria and
Romania worsened the situation; the union spoke of hourly wages of €1.90 and

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

estimated that 90 per cent of posted workers worked illegally.16 According to
the union, 69 per cent of workers in the meat industry were posted (Hassel
et al. 2016: 1225). German wage advantages also led the Danish meat companies
Danish Crown and Tulip to relocate production to Germany. Employment in
meat production in Denmark almost disappeared, leading Danish politicians
to support the introduction of a minimum wage in Germany (Hassel et al.
2016: 1230).
A prolonged discussion surrounded the general introduction of a legally
binding minimum wage. In the domestic arena too, the traditional system of
collective agreements had come under pressure due to decreasing membership in
employer associations and unions. After the Eastern enlargement, several sectors
were added to the list of those that had a minimum wage (the building sector was
the first in 1997, and was then joined by cleaning, postal services, security,
mining, laundry services, the waste industry, including street cleaning, public
training, and care services). However, it took until August 2014 to agree on a
minimum wage in the meat industry, the sector which had the biggest problems.
In an unusual move, the Belgian government handed a complaint about
Germany to the Commission in 2013, as the lack of a German minimum wage
was ruining the competitiveness of Belgian slaughterhouses.17
Finally, from 2015 onwards, Germany introduced a general minimum
wage, which was a significant change to the domestic institutions of capitalism
because of the importance of collective agreements. In this context, I only wish
to point out that Europeanization added incentives for this major reform. The
introduction of a minimum wage is one possible unilateral move that is
acceptable within the ECJ’s interpretation of the Treaty.

15
FTD (9 Feb. 2005), 27; Der Spiegel 7/2005, 32–5.
16
FTD (24 May 2007), Hängepartie im Schlachthof.
17
Süddeutsche Zeitung, 10 Apr. 2013, ‘Bulgarische Billig-Fleischer erzürnen Belgiens Metzger’,
<http://www.sueddeutsche.de/wirtschaft/protest-gegen-minijobs-in-deutschland-bulgarische-
billig-fleischer-erzuernen-belgiens-metzger-1.1645144> [accessed 9 May 2017].
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Europeanization Effects of Case Law 215

The Laval Quartet

The series of rulings on the Laval, Viking, Rüffert, and Luxembourg cases,
which were issued in late 2007 to 2008, restricted domestic options to foster
collective agreements against wage dumping, as they interpreted the provi-
sions of the Posted Workers Directive as rules for the maximum protection of
workers. In particular, the ECJ argued that the right to strike needed to be
assessed against its implications for the freedom to provide services and the
freedom of establishment, and therefore necessitated a proportionality test. As
strikes are only effective if they hurt economically, this was heavily criticized
by the unions (European Trade Union Confederation 2008). The rulings
attracted significant political attention, and they required domestic legislative

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

changes. At a time when memories of the conflicts around the Services
Directive were still fresh, the rulings backed up criticisms that accused inte-
gration of having a neoliberal bias.
It is not possible to do justice here to the wide range of literature that
recounts member-state reactions to this series of cases (Malmberg and
Sigeman 2008; Zimmer 2011; Bücker and Warneck 2010). I will only outline
some of the reactions. In Sweden, legislation had to be changed that had
allowed unions to take collective action against companies that adhered to
collective agreements under foreign law (Blauberger 2014: 467–8). As a result,
the protection of posted workers deteriorated (Seikel 2015: 1176–7). Reform in
Sweden was difficult as employers backed a loosening of the protections for
workers (Davesne 2009; Woolfson and Sommers 2006; Rönnmar 2010).
Compared to Sweden, Denmark managed to react much more quickly in an
example of anticipatory obedience (Blauberger 2014: 468; Blauberger 2012: 117).
Reform in Denmark was greatly helped by the larger societal consensus
among unions and employers on regulating posted workers (Refslund 2015;
Seikel 2015).
The Rüffert ruling concerned the Public Procurement Act of the German
state of Lower Saxony. Construction firms had to commit to collective
agreements for winning public tenders. A German firm subcontracting to a
Polish company failed to make sure that sufficient wages were being paid,
which resulted in a fine and the annulment of the contract. This led to the
preliminary ruling in which the ECJ found that the Public Procurement
Act violated the freedom of services and the Posted Workers Directive. As
a consequence, several federal states reformed their Public Procurement
Acts as well. As Sack has shown, party differences in the governments
played out in these reforms (Sack 2010, 2012; Blauberger 2012: 120).
Centre-right governments were happy to take up the opportunity to
abolish collective-agreement declarations (Blauberger 2012: 121). However,
governmental concerns about the possible liability claims of companies
who were unlawfully excluded from public tenders were also an important
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

216 The ECJ and the Policy Process

consideration in enacting reforms after the Rüffert ruling (Blauberger 2014:
470; Seikel 2015: 1178–81).
The Luxembourg case was the only infringement procedure of the four.
Luxembourg had extended all the terms and conditions of employment
regulation to posted workers, thereby violating the freedom of services and
the Posted Workers Directive. As Blauberger (2012: 122) argues, the Luxem-
bourg government had deliberately tested the waters with its law by interpret-
ing the Posted Workers Directive in the broadest possible sense. After the
ruling, the government gave in and reformed the regulations. However, what
appeared to be regulatory surrender, Blauberger argues, still allowed the
defence of much of the government’s original policy aims. He calls the strategy
‘emulation’, through which member states adopt a policy solution that the

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Court has already vetted as a template for their own policies (Blauberger
2012: 114). This happened in the Luxembourg case, as the country used the
different policy solutions member states had reached after the Laval and
Rüffert rulings for orientation.

Reform of Social Benefits in the United Kingdom

In the context of the referendum on Brexit, EU citizens’ access to social

benefits in the UK has been a central cause for concern. In line with the
country’s increasingly eurosceptic political climate, the UK government has
continuously reduced EU citizens’ access to different social benefits through
various legislative reforms. Just as the Centros ruling led to several legal
reforms for domestic companies, these cuts are also a response to the different
ECJ rulings that have facilitated EU citizens’ access to non-contributory social
benefits in other member states. These rulings have, on the one hand, broad-
ened the definition of who enjoys worker privileges under the free movement
of workers and, on the other hand, linked more entitlements to the status
of being an EU citizen under the Treaty’s provisions. Chapters 3 and 5
discussed this.
After the Eastern enlargement in 2004, the UK was one of the few member
states (in addition to Ireland and Sweden) not to use the transitional option of
suspending the free movement of labour for 2 + 3 + 2 years, in the expectation
that net migration from the Eastern accession countries would amount up to
13,000 a year (Drew and Sriskandarajah 2007). However, by September 2006
almost half a million people had registered to work in the UK (Drew and
Sriskandarajah 2007). With access to non-contributory benefits for EU citi-
zens being originally generous, the UK government has struggled with ECJ
case law in carrying out cuts (O’Brien 2015), repeatedly demanding the right
to deny access to social benefits during the first four years of EU citizens’
residence in the country. In a period of austerity that followed the financial
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Europeanization Effects of Case Law 217

crisis, access to social benefits became a dominant topic in the EU referendum
debates. It is important to observe how case law structured these reforms,
although the Citizenship Directive only grants EU citizens’ equal access to
benefits after five years of residence. However, there must be an ‘individual
assessment’ (C-140/12 Brey), and benefits can only be denied to citizens who
are an ‘unreasonable burden’ (C-184/99 Grzelczyk) on the state. While there is
a requirement for a ‘certain degree of integration into the society of that State’
(C-209/03 Bidar), case law demands a ‘certain degree of financial solidarity’
(C-184/99 Grzelczyk), and the privileged status of workers with the right
of free movement applies as soon as work is not ‘only marginal and ancillary’
(C-53/81 Levin). Moreover, the Court requires authorities to assess the specific
situation of each EU citizen on a case-by-case basis, making it very difficult to

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

handle mass applications (Blauberger and Schmidt 2017). The debate is
interesting because it demonstrates the difficulty of translating rules that
have been introduced by case law into general policy.
Under the Conservative government in 1994, the UK had already intro-
duced a habitual residence test, in order to counter fears of possible ‘benefit
tourism’, as income support had previously been granted irrespective of the
length of stay in the UK. The notion of habitual residence follows Regulation
883/2004, where it establishes the responsible member state. The definition of
habitual residence draws heavily on case law. The test does not apply to EEA
migrant workers to the UK. Since its introduction, British citizens returning to
the UK also have to pass the test in order not to violate EU non-discrimination
rules. In fact, a review of the test under the Labour government in the late
1990s criticized the way in which returning UK citizens were the ones most
affected by the requirement. Residence requirements were then cut from five
to two years. In 2008, almost 3,000 British citizens failed the test (Kennedy
2011b: 3, 6–8). The underlying legislation does not clearly define habitual
residence, but the test takes into account the length of residence, prospects of
employment, existing links to the UK, and the intention to settle there.
In 2004, the test was complemented by the right to reside test, which was
directly linked to the Eastern enlargement. In contrast to habitual residence,
UK nationals are exempted from the right to reside test. The right to reside test
is applied before the test for habitual residence and is relevant for different
benefits, such as jobseeker’s allowance (JSA) and child and housing benefits,
and seeks to ensure that those EU citizens who settle in the UK are either self-
sufficient or economically active. Because of the discriminatory nature of the
test, the Commission began an infringement procedure in 2010 (Kennedy
2011a: 4, 15–16), which was taken to the Court in 2013 (Case C-308/14).
While the Commission originally considered there to be a case for discrimi-
nation against EU nationals in relation to different benefits (Pension Credit,
income-based JSA, Child Benefit, and Child Tax Credit), after the ruling in
Case C-140/12 Brey, it restricted the infringement procedure to Child Benefit
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

218 The ECJ and the Policy Process

and Child Tax Credit.18 AG Cruz Villalón, in his opinion on the case in
October 2015, argued that the test was proportionate and justified by the
need to protect public finances. Shortly before the referendum, the Court
followed this view in June 2016.
During the pending court case, the UK continued to restrict EU citizens’
access to social benefits. In these attempts, it is striking how closely ECJ case
law was followed. In late 2013, the government introduced a revised habitual
residence test, which had more individualized questions and was guided by an
intelligent IT system; it included questions about the individual’s efforts to
gain employment. This fulfilled the requirement for a case-by-case assessment
process (Kennedy 2015: 8–9). From 2014 onwards, jobseeker’s allowance has
been cut for the first three months in the UK.19 The allowance is only paid for

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

the following three months, as member states may limit the right of residence
of jobseekers to a period of six months; beyond this period, the jobseeker has
to provide evidence of ‘genuine chances of being engaged’ (Case C-292/89
Antonissen, No. 21) in a Genuine Prospect of Work test introduced in mid-
2014 for new applicants for JSA (Kennedy 2015: 17–19).20
Moreover, since 2014, having the status of a jobseeker no longer entitles an
individual to housing benefits.21 The UK government argued that, in view of
the 3,000 EEA citizens who were claiming housing benefits, this could imply
savings of about £10 million a year (Kennedy 2015: 24). Access to child
benefits and tax credits was also cut in 2014 for the first three months of
residence in the UK. The government set a minimum earning threshold of
£150 a week in order to assess whether work is not ‘only marginal and
ancillary’. Those earning below this threshold are assessed individually as to
whether they qualify as workers that fulfil the ECJ requirement. Detailed
guidelines from the Department of Work and Pensions (DWP), which is
responsible for this area, structure the assessments (Kennedy 2015: 21, 31).
Since 2013, Universal Credit has been introduced in stages, replacing six
major means-tested benefits.22 Its classification as social assistance makes it
easier to exclude EU citizens, in comparison to unemployment-related bene-
fits where the free movement of workers becomes relevant. Since June 2015,

18
See Kennedy 2011a: 17–20 for the British response to the case and prognoses as to the fiscal
costs of waiving the right to reside test.
19
Jobseeker’s Allowance (Habitual Residence) Amendment Regulations 2013 (SI 2013/3196),
The Immigration (European Economic Area) (Amendment) (No. 3) Regulations 201434.
<http://webarchive.nationalarchives.gov.uk/20140519141601/https:/www.gov.uk/government/
uploads/system/uploads/attachment_data/file/269505/m-28-13.pdf> [accessed 9 May 2017].
20
The Immigration (European Economic Area) (Amendment) Regulations 2014; SI 2014/1451.
21
Housing Benefit (Habitual Residence) Amendment Regulations 201443.
22
Income-based Jobseeker’s Allowance, Income-related Employment and Support Allow-
ance, Income Support, Working Tax Credit, Child Tax Credit, and Housing Benefit.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Europeanization Effects of Case Law 219

legislation has detailed that EU citizens who are seeking a job do not qualify to
receive it (Kennedy 2015: 33).
Within our context, the way the UK has reformed non-economically active
EU citizens’ access to social benefits is interesting in several respects. First, it
shows the difficulties of translating principles of case law into daily adminis-
trative practice. Courts have to make decisions on individual cases in terms of
their eligibility, since no social policy legislation can capture all individual
circumstances (Blauberger and Schmidt 2017). However, the Court’s unwill-
ingness to accept the five-year threshold of the Citizenship Directive and its
demand for case-by-case assessments create a significant administrative bur-
den, as well as a lack of legal certainty for those seeking social assistance. With
its infringement procedure, the Commission pushed the non-majoritarian

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

nature of these rights further, at a time when exit from the EU was increasingly
being discussed as an alternative to not being granted a voice in discussions
(Hirschman 1970).
To sum up, the different examples show that legislative responses to EU
case law are common. In part, we find domestic responses that are directly
triggered by case law, as in the examples of the Laval, Viking, and Rüffert cases.
For the responses, national party-political reactions are significant. To
some extent, case law heavily constrains domestic reforms, as was shown
in the example of UK social benefits. Here, the difficulty of translating
individual rulings into general rules is particularly pronounced. Finally, to
different degrees, case law is only one factor among others. The reform of
company law had additional motivations. The German minimum wage is
co-determined by EU case law, but this link was absent in the public
political debate.

JUDIC IAL REACTIONS

When analysing the Europeanization effects of EU case law, domestic courts

are crucial actors that can push or constrain the importance of EU law, thereby
defining the political leeway accorded to domestic executive and legislative
actors. ‘In areas governed by direct effect and supremacy, national judges are
expected to act as agents of the EU legal order, not the national legal order’
(Stone Sweet and Stranz 2012: 95). When the issue is sufficiently settled either
by EU law (acte clair) or by an ECJ ruling (acte éclairé), domestic courts are
held to ignore national law that contradicts European entitlements. If domes-
tic courts are non-responsive, in that they do not generally apply European
law and address few demands for preliminary rulings to the ECJ, the pressure
on domestic governance is much less than if domestic courts are activist,
seeking to change the domestic legal setting through EU law. Domestic courts
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

220 The ECJ and the Policy Process

are one important gatekeeper for the influence of EU law at the member-state
level. I will discuss examples for both alternatives. If domestic courts build up
pressure for policy change, reform deadlock at the national level may poten-
tially be overcome at the European level, but the latter may also simply help to
realize the policy goals of minorities. The prerequisite is the extent to which
litigants take advantage of the opportunity structure that EU law provides to
change their regulatory setting (Knill and Lehmkuhl 1999: 2, 8).
Member states differ in their judicial traditions, which influences the way
domestic courts embrace the European legal order. Wind (2010) has argued
that, because of the Scandinavian tradition of parliamentary sovereignty, lower
courts in these countries are highly reluctant to set up preliminary proceed-
ings, as they assume that the highest courts should decide on references. This

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

leads to less pressure for domestic policy changes in comparison to a system of
constitutional sovereignty, where it is common to accept judicial review and
constraints on domestic policy (Abromeit 1995). By way of contrast, in
another system of parliamentary sovereignty, but one that has a common-
law tradition, such as the UK, there seems to be a great deal of awareness of the
constraints of EU case law, as courts are an established source of rights
(Blauberger and Schmidt 2017). Notwithstanding the different court reactions,
the discussion of executive reactions has shown that Danish, German, Dutch,
and British executives all implement ECJ rulings in the context of the rule of
law. If lower courts refrain from initiating preliminary proceedings, as in the
Scandinavian tradition, it is less likely that specific domestic questions will
become subject to judicial review.
Finally, we can also hypothesize that the differences among member states
relate to the length of their membership in the EU. It is plausible that newer
member states, in particular those that have only regained full sovereignty
recently, face more difficulty in accepting the constitutional nature of case law.
This was apparent when discussing the example of the Patient Mobility
Directive in Chapter 5 and the Eastern European member states. It is, there-
fore, likely that, with no external pressure, these member states are more likely
to contain the impact of European case law. In fact, this may give an incentive
for all other member-state governments to codify case law in EU secondary
law in order to level the playing field.

Domestic Judicial Activism

Domestic courts may be activist and help to build and diffuse emergent
European legal entitlements. Recent case law on age discrimination may
serve as an illustration here. Although this relates to another area of European
law, and not to the four freedoms, I shall present this example briefly, as it
provides a very interesting case study. As Stone Sweet and Stranz (2012) show,
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Europeanization Effects of Case Law 221

German labour courts have an activist tradition, which began in the area of sex
discrimination. Progressive labour courts, which disagreed with the litigation
on rights by the German Federal Constitutional Court (GFCC), addressed the
ECJ at an early stage in order to pursue a line of adjudication that was not
shared by the GFCC and went far beyond the constitutional court’s line of
reasoning. The important Cases 170/84 Bilka and 181/88 Rinner-Kühn were
the result of references from the German labour courts. ‘In their references,
German judges made it clear that they considered these practices, and the
national statutes that permitted them, to be a form of indirect sex discrimi-
nation, and therefore unlawful under ex-Article 119 TEEC. The ECJ respond-
ed by giving the referring courts what they wanted: a mandate to root out
indirect discrimination in the workplace and in Germany’s social policy’

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

(Stone Sweet and Stranz 2012: 95). The authors argue that ‘Put differently,
the ECJ’s key rulings in the area strengthened what labour judges were already
pre-disposed to do’ (p. 100).
Against this backdrop, a significant broadening of European discrimination
rights was achieved through Cases C-144/04 Mangold and C-555/07
Kücükdeveci. The Mangold case attracted a great deal of attention in many
areas. For once, it was a clearly constructed case. A lawyer, who had been
politically vocal about his opposition to a labour-market reform, which
allowed more lenient rules for limited labour contracts for those over 52
years of age, hired another lawyer under this reform with the explicit goal of
challenging the rule in court. At the time, the deadline for implementing the
2000 EU directive on anti-discrimination had not yet passed. Judge Hauf of
the repsonsible Munich labour court of the first instance nevertheless took the
opportunity to refer the case to the ECJ. As the judge could not yet argue with
the directive, she referred to the EU Charter of Fundamental Rights, which
was not yet in force either. ‘Taking its cues from Judge Hauf, the Court held
that the German provision under review discriminated against older employ-
ees. . . . The ECJ then gave Judge Hauf what she wanted, authorization to
enforce EU fundamental rights against the national rule under review’
(Stone Sweet and Stranz 2012: 101).
Another example here is Case C-555/07 Kücükdeveci. In this case, the
German rule at issue stated that working times for those under 25 years of
age did not have to be included when calculating the length of notice for
terminating work contracts. The Higher Labour Court in Düsseldorf referred
the case to the ECJ. Its reference baldly stated that, given the GFCC’s juris-
prudence, a reference to the GFCC would be unlikely to help Ms Kücükdeveci:
‘the judges made it clear that they believed that the part of the Civil Code
under review violated the EU’s age discrimination principle, and could not be
justified under the EU’s Framework Directive’ (Stone Sweet and Stranz 2012:
103). The ECJ confirmed the contentious Mangold age-discrimination prin-
ciple. In general, active labour courts that have linked up with the ECJ have
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

222 The ECJ and the Policy Process

fundamentally transformed German labour law through European law.
Domestic judicial activism, it becomes apparent through this case study, can
make a difference in fostering ECJ rulings.
The reaction from the Swedish labour court after it received the ECJ’s Laval
ruling can be seen in a similar activist light. The court established an indem-
nity claim for the employer, although it stated ‘that there was no explicit
support in the case law of the ECJ for the proposition that an individual is to
pay damages on an EU law basis to another individual upon a violation
specifically of Article 49 TEC (now Article 56 TFEU)’ (Rönnmar 2010: 282).
Thus, despite the fact that the legal matter had been so unclear as to require the
local court to address the ECJ—after which the ECJ produced an expansive
ruling in a way that surprised and outraged the legal and political community—

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

and that there was no clear EU legal requirement for liability, the union had to
pay punitive damages of €55,000.23
The potential to be liable for the economic consequences of a strike may put
a serious brake on industrial action. Thus, when British Airways (BA) planned
to set up a subsidiary in another member state in 2008, the British Airline
Pilots’ Association (BALPA) wanted to strike in order to avoid social dumping
impacting their members. BA, however, requested an injunction from the
court, arguing that a strike would breach the freedom of establishment of
Article 49 of the Treaty on the Functioning of the European Union (TFEU)
and would, therefore, give grounds for claiming damages of approximately
£100 million a day. Consequently, BALPA could not risk strike action
(Zimmer 2011).

Domestic Judicial Restraint

Domestic judicial restraint in following a broad interpretation of case law

lessens the pressure from EU law. If domestic courts are reluctant to refer to
European law, justice may be ‘contained’, as executives and legislators are
under little pressure to modify domestic laws in accordance with European
demands. However, we have seen also that it appears to be quite common for
executives to respond to ECJ rulings—this is also true for cases where their
country is not directly involved—without gambling first on the reactions from
the judicial system. Perinetto (2012) analyses why the important judgments of
Laval and Viking have had no effect in Italy. As is the case in Scandinavia, in
Italy it is up to the social partners to set tariffs, so one would expect the rulings
to have a noticeable impact here. However, since wage levels between Italy and
the new Eastern European member states do not differ so much, there is less of

23
I thank Michael Blauberger for reminding me of this case. See Ewing (2012) for details.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Europeanization Effects of Case Law 223

a wage-differential to exploit. Moreover, labour courts demonstrate a great
commitment to guarding the social rights of workers in their judgments, so
they cannot be seen as natural allies of the ECJ in this respect. Perinetto,
therefore, expects that the Italian Constitutional Court would explicitly
reject the supremacy of EU law in this case, should Italian courts be pressed
to follow the case law from the Laval ruling.
Another interesting example of an at least temporary restraint is the
reaction to the case law on European citizenship rights. As was noted in
Chapters 3 and 5, the ECJ has, over the years, built up an impressive body
of case law, transforming an initially meaningless treaty article into far-
reaching citizenship rights, including EU citizens’ access to national schemes
of social assistance that were previously reserved for nationals (Schmidt 2012).

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Importantly, the general article on non-discrimination is interpreted as partly
prohibiting legal provisions that favour nationals over EU-citizens. As we have
seen earlier, the Citizenship Directive of 2004 largely codifies the existing case
law from the time, which was greatly influenced by the Grzelczyk ruling.
As Björn Schreinermacher (2013: 103–15) shows, by analysing the court
cases in the UK and in Germany that referred to the Grzelzcyk case, it becomes
clear that domestic courts supported a more constraining interpretation of the
case law for some time. Numerous UK court proceedings have dealt with the
habitual residence and right to reside tests. In mid-2011, the UK Supreme
Court agreed in the Patmalniece case that the exemption of UK nationals from
the right to reside test is indirectly discriminatory, but it argued for its
proportionality, as granting equal benefits would result in the placing of an
undue financial burden on the UK. Such a restrictive interpretation of national
discriminatory principles clearly shields national administrations.
For some time, German courts excluded EU citizens who fell short of the
five-year residence requirement from social assistance by making reference to
the will of the European legislature.24 Schreinermacher (2013: 106) argues that
these largely restrictive interpretations by the courts allowed the German
federal government to maintain its restrictive legislation that equates EU
citizens with foreigners and only reserves non-contributory social assistance
for Germans.
In late 2013, the first newspaper reports drew attention to social courts that
opposed the legislative status quo and administrative practice.25 Increasingly,
courts turned to EU law to argue that Germany could not withhold social

24
Landessozialgericht (LSG) Nordrhein-Westfalen, Beschluss of 15 June 2007, Az. L 20 B 59/
07 AS ER, L 20 B 59/07LSG Niedersachsen-Bremen, Beschluss of 02 Aug. 2007, Az. L 9 AS 447/
07 ER, LSG Hessen, Beschluss of 14 Oct. 2009, Az. L 7 AS 166/09 B ER. Cases where assistance
was granted are: LSG Berlin-Brandenburg, Beschluss of 25 Apr. 2007, Az. L 19 B 116/07 AS ER,
SG Wiesbaden, Beschluss of 15 Jan. 2008, Az. S 16 AS 690/07 ER, 16 AS 690/07
(cf. Schreinermacher 2013: 103–15).
25
FAZ 2 Dec. 2013, ‘ “Hartz IV” für Arbeitslose aus Rumänien’.
OUP CORRECTED PROOF – FINAL, 6/12/2017, SPi

224 The ECJ and the Policy Process

assistance from EU citizens automatically but that there needed to be case-
by-case assessments, according to ECJ case law. While the German social
administration was bound by exclusionary German law, once EU citizens
increasingly sought judicial redress, a heterogeneous picture developed. Some
social courts maintained restrictions, taking their guide from domestic social
law, and some were permissive, following the EU non-discrimination rule. In
2014 and 2015, courts handed cases to the ECJ, asking for guidelines on when
EU nationals may be excluded from benefits (Cases C-333/13 Dano, C-67/14
Alimanovic, and C-299/14 Garcia Nieto) (Blauberger and Schmidt 2014). In
late 2014, the Dano case was decided upon, which involved a Romanian who
had moved to Germany and never intended to work. The ECJ drew the
line here and affirmed Germany’s right to deny social benefits in this case. In

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

the Alimanovic case that followed in September 2015, the Court supported the
rule that unemployment benefits can be restricted to six months for those that
have worked for less than a year. In these two cases, the ECJ refrained from
engaging in judicial activism by permitting limits to ‘financial solidarity’. At least
some questions concerning the right of access to benefits are thereby settled.
In summary, the national judiciary has a significant degree of freedom to
react to case law in responsive or restrictive ways. It is difficult to predict which
way the judiciary will turn. If European law offers judges preferable legal
dogma against domestic judicial opposition, there are significant incentives
to turn to supreme European law. If judges, in contrast, defend upholding
differences between EU nationals and nationals, or are inclined to guard
national political autonomy, the impact can be contained to a greater degree.
It then depends on external pressures, such as interventions from the Com-
mission to change domestic policies, or on the executive and the legislature.

SOCIETAL A CTORS

Societal actors may drive or constrain the development of case law, putting
pressure on domestic governance or helping to contain it. To the extent that
the ECJ grants direct horizontal effect to the fundamental freedoms, private
actors are also the direct addressees of EU case law, having to change their
behaviour. Finally, private actors may be important collaborative partners for
governments in containing the negative implications of case law.
It is more common in the US to regard litigation as one option, alongside
lobbying in the parliamentary process that private actors can use to realize
their interests (Conant 2006; Kelemen 2006). By actively seeking out the
benefits of EU law, actors advance its applications, partly with the help of
the Commission. Consequently, the Commission’s intervention against
German public banks on the basis of European competition law, most notably
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Europeanization Effects of Case Law 225

the Westdeutsche Landesbank (West LB), was supported by complaints from
German private banks (Smith 2005; Seikel 2013). Additionally, the action
taken against Volkswagen’s privileges was partly motivated by Porsche’s
complaint to the Commission (Blauberger 2014). Many interventions in the
name of integration, therefore, reflect domestic political conflicts. As such, to
answer the question of whether member states have political autonomy, it is
paramount to ascertain whether the existing governance structure is broadly
supported by domestic societal actors. Blauberger’s (2012) analysis of the
Danish and Swedish reactions to the Laval case shows that governments can
be certain that domestic court cases will not challenge domestic laws when
societal actors are sufficiently densely institutionalized and in agreement. In
Sweden, the Confederation of Swedish Industry welcomed the pressure the

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

ruling put on collective agreements. It even financed the legal dispute in the
first place (Asteriti 2013: 78). When Sweden wanted to reach a new legislative
consensus for posted workers, striking a compromise between the social
partners proved difficult and led to a much more narrow definition of min-
imum pay than had been established in Denmark, where reform was also
deemed necessary and was much easier to realize due to the broad consensus
that existed between the social partners (Blauberger 2012: 118; Refslund 2015).
As Seikel (2015) shows, the difference between the Danish and the Swedish
response can be successfully explained through the different structures of the
building industries in both countries. The Danish building industry consists of
small- and medium-sized firms, which are united in their interest in avoiding
wage competition created by posted workers. In Sweden, in contrast, a few big
players dominate the scene. For them, wage competition from posted workers
is beneficial. By establishing a horizontal direct effect, the Laval and Viking
rulings also address unions directly. In considering strikes, they have to ensure
that the implications for the freedom to provide services and the freedom of
establishment are proportionate. As mentioned earlier, as a result, a strike
against BA in 2008 was halted through a court injunction.
The horizontal effect of the free movement of goods, established through Case
C-171/11 Fra.bo, offers another example. In this case, an Italian manufacturer of
copper fittings for water and gas pipes took legal action in Germany against a
German standardization body, the Deutscher Verein des Gas- und Wasserfaches
(DVGW), for denying certification under their standards. The Court backed Fra.
bo’s interpretation that this hindered market access and that the standardization
body is bound by the free movement of goods. After this ruling in 2012, it took
until May 2015 for an out-of-court settlement to be made, in which the DVGW
promised to adapt its standards to meet Fra.bo’s needs.26 Up until that point, the

26
DVGW, 22 May 2015: ‘DVGW und Fra.bo beenden über 10 Jahre andauernde rechtliche
Auseinandersetzung’, <https://www.dvgw.de/der-dvgw/aktuelles/presse/presseinformationen/dvgw-
presseinformation-vom-22052015> [accessed May 2017].
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

226 The ECJ and the Policy Process

court case had been pending at the German Federal High Court of Justice
(Bundesgerichtshof ). When the ECJ grants horizontal effect to the fundamental
freedoms, compliance demands that societal actors change their behaviour. In
this specific case, the German government also had to change some statutes that
referred to private standards. Should the ECJ broaden the horizontal effect of the
fundamental freedoms in the future, Europeanization effects that concern private
actors will become more important.
Private actors are not only relevant as potential litigants or addressees of
case law. They may also act as partners to governments in the sense that they
allow escape routes from the pressures of EU law.27 In the German govern-
ment’s struggle against illegal activities related to the posting of workers,
discussed earlier, the ministries responsible for this area sought to forge

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

pacts against illegal work with the employers’ associations and labour unions
in those sectors that were particularly vulnerable to fraud (Schmidt 2009: 179).
Pacts were agreed with the building industry (2004), the transport sector
(2006), the meat industry (2007), and cleaning services (2008), all with the
aim of increasing understanding of the negative implications of illegal work
and raising awareness of the existing regulations.

CONCLUSIO N

In this chapter, I have argued that member states’ reactions to case law can be
explained with reference to the costs and benefits imposed on different actors.
The preferences of the executive, legislature, and judiciary are as diverse as
their responses. The costs for member states may be related to the impact
of regulatory competition, liability claim cases, and further threats from
European case law. The pressure to respond to EU law is low if the orientation
of domestic courts and a high degree of societal consensus result in little
litigation. It is obviously the case that governments may also welcome case-law
induced reforms, which will give rise to a swift reaction. In processes of
domestic reform, case law is often only one consideration among many. It is
therefore difficult to tie domestic institutional changes exclusively to specific
judgments, and, compared to the transposition of secondary law, the impli-
cations of case law are more fraught with legal uncertainty. That does not

27
Töller found societal agreements in German environmental politics in which domestic
environmental goals were achieved through self-regulation, as it was unclear whether legislation
would be possible under European law. An example here is the prohibition of dangerous
materials, such as asbestos, which could have been interpreted as a distortion of the free
movement of goods. In order to evade such legal uncertainty, an agreement was sought with
the relevant economic actors concerning the phasing out of this material (Töller 2004).
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Europeanization Effects of Case Law 227

mean that case law is unimportant, however. Though EU law’s exceptional
nature is widely recognized, its consequences are still poorly understood
(Davies 2016a). As we saw, in particular in Chapter 6 with the Meilicke case,
the sometimes very drawn-out period of time until the implications of EU law
are specified and domestic reactions are institutionalized helps to blur the link
between case law and policy responses.
In order to make the multiple responses to case law more transparent,
I differentiated between the reactions of executives, legislatures, domestic
courts, and private actors. For all these actors, we can find direct and indirect
reactions to case law. In part, there are also reactions that compensate for its
constraints. Table 7.1 summarizes the findings.
In light of the contained-compliance perspective, the extent to which

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

member states do in fact accommodate the demands of EU case law is surprising.
This happens even if there is not a great deal of pressure from domestic courts.
The residence rights of TCN family members and EU citizens’ access to social
benefits are costly and do not meet governments’ preferences. Despite this,
governments do respond, and they appear to set priorities. Some decisions
they clearly resist, in particular when they have potentially large fiscal implica-
tions, such as in the Meilicke case. In such cases it is preferable to elicit another
ruling from the ECJ in the hope of achieving fine-tuning and the circumscription
of the impact of EU law. Other instances of resistance, such as the language
test for TCN family members, are clearly politically motivated. However, resist-
ance is by no means the rule, but rather the exception. When it does occur, the
Court appears to be forthcoming and fine-tunes its case law. Such a reaction is

Table 7.1. Overview of responses at the member-state level to direct and indirect
pressures
Executives Legislature Judiciary Societal actors

Accommodation: Accommodation: Activism: Accommodation:

- TCN family members - Reform of - Mangold, age - Laval in Denmark
- Child benefits limited discrimination - Fra.bo
companies
- Posting of
workers
- Laval quartet
- UK social
benefits
Resistance: Restraint:
- Integration test for - EU citizens’ access
TCNs to social benefits
- Meilicke
Compensation: Compensation:
- Control of posted - Pacts against illegal
workers in Germany posted workers
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

228 The ECJ and the Policy Process

probably more likely when resistance concerns secondary law, as was the case for
the language test. These contrasting reactions of accommodation and resistance
explain why both the perspectives of judicial activism and contained compliance
persist in the literature in this area. Rather than attempting to solve this
theoretical tension in one direction, the analysis performed in this chapter
makes it more plausible to regard them as two sides of the same coin. Since
executives normally implement case law in a rule-of-law fashion, they can trust
that the Court will give them some leeway to be able to follow their policy
preferences in cases of protracted resistance. This does not mean, however, that
case-law development is supported by the political consensus of member-state
governments, as some analyses put forward (Carrubba and Gabel 2014; Larsson
and Naurin 2016).

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

The ECJ’s case law can have a very disturbing influence on the member
states’ domestic institutions, as is often criticized. However, the ECJ is a
reactive political actor and requires the support of litigants and lower courts,
or the Commission. The domestic impact of European law shows that inter-
locutors from member states often have an interest in challenging the domes-
tic policy status quo with the help of EU law. For the member states’ courts,
the question is whether they perceive there to be legal advantages in the
broader application of EU law or whether they are interested in keeping the
influence of EU law at a low level, which gives rise to containment.
Similarly, societal consensus is crucial in safeguarding the domestic insti-
tutional setting. If there is broad agreement about domestic institutions, they
are less likely to falter, as no litigants will emerge, and the Commission is also
more reluctant to turn to the Court in the absence of complaints. Under a
condition of broad agreement, domestic institutions are more easily main-
tained. However, if there are vocal actors who stand to profit from a regime
shift, it may be that a minority that could not have induced changes through
legislation at the national level uses EU law to bring about institutional change.
Age discrimination in the Mangold case and the liberalization of German
public banks are clear examples of this, as are the reforms of limited liability
companies. Given the leeway that case law can give a minority to push for
reforms, it may be the case that homogeneous small member states are at an
advantage. In larger countries that have a more heterogeneous economic
structure, the heterogeneity of interest will be more pronounced, making the
status quo, arguably, more vulnerable to internal contention. The comparison
of the Danish and Swedish reactions to the Laval case may point to that. The
reactions to the quartet of Laval cases, in which party positions and the degree
of societal consensus made a significant difference, also show that there are
some policy options when member states accommodate case law.
The problem of translating the demands of single judgments into general
principles should also be noted, particularly in the case of the British social-
policy example. The process of balancing that courts employ to fulfil the
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Europeanization Effects of Case Law 229

requirements of the individual case cannot be generalized into universal rules.
Due to their constitutional status, however, the principles derived from single
cases cannot simply be ignored. This leads to increasing constraints on
policymaking. In an administration that deals with mass application, it is
simply impossible to take into account an unlimited range of individual
aspects in order to enact a case-by-case decision. Moreover, it is highly
questionable whether this is in fact desirable. Those who need social support
are interested in clear rules as to their eligibility. They do not want to be
referred on to the ever-changing development of case law, to which the
Austrian website for public student support refers, for example (Schulten
2013: n. 124). The UK, in its quest to limit access to social benefits, aimed to
design tests that reconciled individual assessment with mass application,

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

which received heavy criticism from the Commission: ‘Determining whether
or not a particular person has a “right to reside” can be far from straightfor-
ward. Anyone refused benefits or tax credits on the grounds because they are
deemed not to have a right to reside should seek professional advice’ (Kennedy
2011a: 20).
In addition to case law leading to direct changes to domestic policies, EU
case law also triggers indirect responses. This is particularly the case if
regulatory competition is furthered by it, resulting in more restrictive national
regulations coming under pressure. The Centros ruling is a remarkable
example of this. It also shows that reactions may have been exaggerated and
premature. After several years of limited companies being established on the
continent, they now appear much less attractive than was initially the case
when their reporting requirements were still ignored. In addition, the fraud
related to the posting of workers can be regarded as an indirect effect of case
law. As we saw, there are several examples of compensatory measures. The
German government augmented its administrative controls, but it also forged
pacts with societal actors and other member states.
Given the limits of my empirical basis and space constraints, I could only
provide a preliminary overview on the Europeanization effects of case law.
Empirical evidence is also highly biased in favour of the old member states,
particularly Germany. To what extent can the German experience be gener-
alized? On the one hand, Germany is the most important and powerful
member state. This could lead one to assume that it is a least-likely case, as
Germany would be less willing to accept case-law constraints outside of
its political control than weaker member states. On the other hand, Germany
has a strong constitutional court and a political system marked by constitu-
tional sovereignty, in which responding to judicial constraints is normal.
This tradition would make it into a most-likely case for abiding with case
law. It seems very plausible that member states with weaker administrative
capacities and weaker traditions of the rule of law would present a different
picture.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

230 The ECJ and the Policy Process

In a multilevel system, actors’ political preferences are subject to multiple
influences. On account of directly effective and supreme EU law, actors do at
times have the incentive to draw on this alternative legal order. ECJ case law
rarely mandates the implementation of very specific policy measures in the
way that secondary law does. Therefore, we find a much more complex picture
when analysing the Europeanization effects of case law. There are often
multiple ways to respond, including compensatory measures, which are
more influenced by legal uncertainty. Options for blocking responses may
be more varied than is the case for the implementation requirements of
secondary law. Consequently, when analysing domestic reforms, reactions to
case law may be just one consideration among many. How relevant was the
impact of the posting of workers for the introduction of a general minimum

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

wage in Germany? Such a question is difficult to answer.
The EU Treaty is unique in its multiple and far-reaching policy conse-
quences. By summarizing various examples of its impact, this chapter has
aimed to draw attention to its relevance in the process of pursuing a more
systematic analysis. Clearly, a disregard for ECJ case law in political science
cannot be justified, since member states do submit to sovereignty constraints.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Conclusion

Research on European integration readily recognizes the importance of the

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

European Court of Justice (ECJ). However, in the study of European Union
(EU) policymaking processes, its role is largely ignored. The Court is
predominantly analysed in the context of research into compliance with
secondary law, as infringement procedures are an important indicator of
compliance problems. But beyond making agreements binding and credible,
the activist development of case law by the ECJ influences many EU policy
processes. The wide reach of the four freedoms into different policy fields
integrates policies through case law, even if there is no European secondary
law in sight. My book emphasizes how this largely neglected aspect of the
Court’s power casts a shadow over the policymaking processes of both the EU
and the member states.
In order to make the argument that the Court has an important influence
on policymaking, I had to show the following. First, it was necessary to
substantiate the notion that dynamic case-law development based on the
Treaty adds to its effect. If the Treaty had pretty much the same content
that its masters had already agreed upon, the influence it has would scarcely be
problematic. Secondly, I had to demonstrate that this case law is a constraint
on legislation at the EU level. If member states could change the import of
rulings by agreeing on secondary law, again, the constraint would be less
significant. Thirdly, I shifted my focus to the member-state level to confirm
that, independent of its codification in EU legislation, case law also has an
impact when its content is controversial at the national level. This feeds back
into the European decision-making process.
In terms of the dynamic development of case law, I argued that the extent of
legal uncertainty as to the reach of the Treaty’s rules is crucial. Direct effect
and supremacy have established a constitutional supranational order, the
implications of which are constantly changing. Given the enormous material
details of this supranational legal order, in comparison to national constitu-
tions, private actors may find alternative legal positions in this arena that they
prefer to those they can pursue in the national legal setting. Through the
preliminary procedure, private actors have indirect access to the ECJ. They
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

232 The ECJ and the Policy Process

may find allies in national lower courts and collaborate to bring supranational
law into effect. The ECJ, as is the case for all courts, is only legitimated to solve
legal disputes. Courts may not draw up general policy rules. In preliminary
procedures, the Court gives general guidelines as to the interpretation of EU
law, but such principles of interpretation are not the same as clear policy rules.
How the ruling on a specific legal case has to be generalized across situations
in other member states is often hard to discern; this is particularly true in light
of the increasing heterogeneity of member states after successive processes of
enlargement. The detailed provisions of the Treaty that deal with material
policy can therefore be used in multiple ways to overcome regulatory restric-
tions at the national level. The four freedoms, which were my central focus,
touch upon all national regulation of economic activities for goods, services,

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

persons, companies, and capital. The free movement of workers, taken
together with more recently established citizenship rights, moreover, has
direct implications for the national welfare state.
The Court’s orientation towards a teleological interpretation of the Treaty,
that is, ‘towards an ever closer union’, means that the respective competences
of the national and the EU level are kept in flux. Legal uncertainty provides
incentives for those actors that see potential benefits in using an expansive
interpretation of EU law to address national courts. The ECJ, as I have shown,
responds with a path-dependent interpretation of the four freedoms and
citizenship, thereby providing doctrinal coherence and strengthening legal
certainty. With its constitutional status, the Court’s case law cannot easily be
overruled politically, as Treaty changes are virtually impossible to achieve.
Secondary law cannot modify case law that is based on the Treaty. Member
states can only hope to signal their preferences to the Court through secondary
law, alongside the possibility of intervening in Court proceedings with their
observations. It is up to the Court to decide whether to modify its case law or
even to make an about-turn by taking a new path. Most notably, this happened
with the Keck ruling on the free movement of goods. While an extension of
rights receives positive feedback through the process of litigants addressing the
Court, this feedback is absent when the Court limits the reach of the Treaty
and practises judicial restraint. There is no use in addressing the Court to limit
EU law; such arguments are only made in interventions by member states
when litigants are aiming for expansion. Judicial restraint is therefore up to the
Court, and in the ECJ’s institutional set-up, there is little support for its
practice. The community of EU lawyers, which broadly favours an expansion
of EU law, is important in this context.
Legal uncertainty also gives governments an incentive to codify case law in
legislation, to re-establish legal certainty, and possibly to prevent future
unwelcome case-law developments. As it is focused on the case at hand, case
law by itself cannot ensure the complete regulation of a field, but it elucidates
the relevant principles of the Treaty. A worker is not someone who works a
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Conclusion 233
certain number of hours but someone whose work is ‘not marginal and
ancillary’. In particular, the single market for goods demonstrates the difficulty
of regulation based on case law, since mutual recognition was codified at a
relatively late stage. Secondary law cannot overrule the Treaty’s interpretations
but has to codify them. In doing so, secondary law necessarily adds onto case
law, as single judicial cases cannot provide general rules. In contrast to
Martinsen (2015), I argue that this does not mean that one should speak of
legislative ‘modification’ of case law. The Commission, as the guardian of the
Treaty, ensures that its proposals comply with case law and interprets political
compromises on secondary law in light of this body of law. Case law takes
generalized effect through codification in secondary law. As member states
respond differently when implementing case law, several of them have inter-

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

ests in codifying case law in order to ensure a level playing field—even if they
disagree with its content.
This interaction of case law with policymaking has not received sufficient
attention in research on integration. Why is such an important feature of the
EU polity relatively neglected? In contrast to American political science,
European political science does not pay much attention to courts (Bellamy
2008a; Rehder 2006). The burgeoning literature on legislative EU policy-
making (Hörl et al. 2005; Thomson 2011) neglects the judicial roots of legis-
lation. European legal scholars, in contrast, do in fact look at the ECJ’s case law
closely. Distinct from political scientists, they are less interested in the condi-
tions of legislative decision-making. Moreover, it is likely that the turn to
comparative politics in European studies (Hix 1994) has blinded scholars as to
the specific nature of EU case law. Additionally, in political systems such as
the German one, which are characterized by a strong constitutional court,
it is very common for court rulings to overshadow the legislative process
(Hassemer 2008). However, in the EU’s political system, case law cannot easily
be overturned, and over-constitutionalization implies detailed material policy
provisions. The interaction between judicial and legislative politics at the
European level is very different from what we know from the national context.
Political science’s ignorance of the way that case law overshadows the
legislative process is particularly noticeable in the discussion of the Services
Directive. Most scholars who analyse this legislative process have regarded it
as an example of politicization, through which the European Parliament (EP)
and trade unions made their preferences felt. While the political debate
around the Services Directive brought the political disagreement about the
course of case-law development into the open, this manifestation was largely
symbolic. The directive could not redress case-law development, although
members of the EP assumed that they could decide politically whether they
preferred home-country or host-country rule. The Commission, bound by the
Treaty, made clear that the directive needed to be interpreted along the lines
the Court had previously set out.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

234 The ECJ and the Policy Process

In addition, the Citizenship Directive revealed the constraints of case law. In
this case, member states were reluctant to follow the Court in its expansive
interpretation of rights and so left some aspects of the directive deliberately
open. The attempt to ‘rein in’ the Court has not necessarily been successful.
Member states upheld EU legal privileges for the status of workers over EU
citizens, not wanting equal access to non-contributory social benefits, but
aiming to safeguard the privileged treatment of their nationals. Nevertheless,
the subsequent actions of the ECJ and the Commission disregarded some
specifications of the Citizenship Directive and continued for some time to
push for the increasingly equalized status of EU citizens and nationals. The
resulting legal uncertainty, concerning when economically inactive EU citi-
zens can access national welfare services, has kept courts busy. It was in light of

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

heightened economic heterogeneity after Eastern enlargement, and the fear
that access to welfare-state services provides incentives for the ‘abuse’ of rights
to free movement, that the Court has taken a more restrictive position in its
more recent rulings. The Court agreed that member states can clearly exempt
some categories of EU citizens from eligibility for social services, without
having to undertake an individual assessment as to the extent that the indi-
vidual is integrated into the relevant member state. Nevertheless, the UK chose
to exit the EU.
Member states were divided on patient mobility. Some governments
opposed codification altogether, but those hoping to re-establish legal cer-
tainty and stop case-law development gained the upper hand. While it is not
clear yet to what extent the Court will take into account the wishes of the
member states in subsequent case law, the legislative process clearly failed to
re-establish legal certainty. Since the case law is based on the freedom of
services and cannot be changed through secondary law, there are now two
alternative processes that authorize cross-border healthcare that are not easily
reconcilable with each other.
Of all the examples, the regulation on mutual recognition of goods is the
least controversial. It is an area where sovereignty restraints are met with
acceptance and where the political salience of these constraints is comparably
low. For goods, the Commission managed to achieve relatively straightforward
codification, as the case law was largely uncontroversial. Legislation, however,
granted more extensive rights than the previous case law had done.
In all of the cases I have discussed, with the exception of the regulation of
goods, codification was politically controversial. The Council’s response was
motivated, on the one hand, by the wish to codify existing case law in order to
heighten legal certainty and to establish a level playing field, and, on the other
hand, by the wish to pre-empt further case law by signalling its preferences to
the Court. In all cases, the EU legislature faced the difficulty of accommodat-
ing existing case law based on the Treaty. It was confronted with rulings that
concerned the details of individual cases that did not lend themselves to
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Conclusion 235
general rules. Courts have to perform a case-specific assessment. In deducing
abstract rules, the legislature may generalize the rights that the Court has
granted to all cases that exhibit some of the same features. Such a broadening
of rights from case law took place in the implementation of the regulation on
the mutual recognition of goods, namely the decision that shifted the burden
of proof of sufficient regulation from the manufacturer to the regulator. In the
Citizenship Directive, member states did not want to generalize case-law
rights in this way, leaving the directive with insufficient specification. And,
in the case of patient mobility, the need to respect case law and the wish to
establish general rules led to a fairly contradictory regime of two alternative
options for enjoying cross-border healthcare.
Case law gives rise to a patchy and legally uncertain regulatory framework.

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Yet it cannot guarantee that the legislature will respond and provide general
rules. This happened in the case of company taxation, where existing case law
imposed constraints on member states, but they were unable to agree on a
common policy. Although the Court subsequently practised more judicial
restraint, this does not mean that sufficient policy options have been kept
open. The recent attempt to counter global tax avoidance at the Organization
for Economic Cooperation and Development (OECD) through common rules
demonstrated this forcefully. Due to ECJ case law, EU member states could
not agree on policy options that other OECD member states considered to be
feasible and appropriate.
Case law also has implications at the domestic level for member states.
Similar to research on European integration, research on Europeanization that
analyses the impact of integration at the member-state level pays little atten-
tion to Europeanization through case law (Treib 2014). Direct responses to EU
case law at the member-state level are not only important as a neglected
element of research on Europeanization. Differences in member states’ policy
reactions to case law also feed back into their preferences for the codification
of case law at the European level.
The EU relies on the implementation of its laws by its member states. By
itself, the EU has little administrative power. If member states ignore EU case
law, contained justice is the result, as Conant (2002) found. My detailed
discussion of the reaction to two notable examples of case law, the Zambrano
and Meilicke cases, can already demonstrate that the assumption of ‘contained
compliance’ does not capture member states’ reactions to case law. Within the
context of the supremacy of EU law, it appears more plausible that member
states respond to case law in a rule-of-law fashion, unless there are significant
costs involved and subsequent ECJ rulings offer the promise of fine-tuning.
My analysis, which ranged across different cases, showed the variety of
responses at the member-state level. The national legislature may change the
law to implement case law, similar to the way it would react to European
directives, but it may also seek to escape its constraints. While the literature on
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

236 The ECJ and the Policy Process

Europeanization often looks almost exclusively at domestic parliaments, this is
not sufficient when researching the impact of case law. Comparable to regu-
lations that are directly effective, administrations may simply adjust their
procedures. Independent of whether they adapt or not, they may also seek
compensation for lost policy instruments or the negative implications of
European rules, partly in cooperation with societal actors. In addition, the
domestic judiciary is an important actor. Independent of whether the execu-
tive or legislature acts, the judiciary may simply follow European case law. It is
only in the cases that it also rejects it that there is likely to be no domestic
impact from ECJ case law. Finally, societal actors are important. Whether or
not they are aware of their rights and push for them may be the decisive
difference. If societal actors are not interested in their European rights, it is

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

much easier for governments to turn a blind eye.
Thus, there is a significant array of Europeanization responses to case law.
Surprisingly, the Western member states, if not the others, often appear to
implement even politically contentious judgments in a straightforward rule-
of-law fashion. Such standard implementation responses to case law, in the
form of legislation or administrative circulars and information for potential
beneficiaries on government websites, appear as a foundation from which
governments hope to push successfully for the fine-tuning of case law, when
they have different important policy preferences.
After the summary of the book’s argument that I have just provided, my
conclusion will now discuss the book’s significance for several important
themes in research. I will start by asking whether the ECJ is activist, which is
a common topic in discussions on the Court’s power. In light of this book’s
analysis, it seems questionable whether activism captures the Court’s impor-
tance. Secondly, I turn to the time dimension, which is particularly relevant.
Thirdly, I discuss the ECJ’s legitimacy in determining policy to this extent.
I then turn to the question of what this book’s analysis implies for the
empowerment of international courts, given that the ECJ is taken as a model
in this respect. I close the discussion by asking where research in political
science should go from here.

A C T I V I SM ?

Much of the debate on the importance of the ECJ focuses on the question of
whether it is activist (G. Davies 2012; Keeling 1998; Solanke 2011; Saurugger
and Terpan 2017). Analyses of the Court’s activism vary according to what
extent they show the Court deferring to member-state preferences (Carrubba
and Gabel 2014; Larsson and Naurin 2016), and indeed my analysis of case-
law development has shown that there are different paths of interpretation
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Conclusion 237
that the Court can take. An interpretation of the four freedoms as prohibiting
discrimination, I argued, grants much more autonomy to member states than
an interpretation of them prohibiting restrictions. As the latter subjects all
national regulations for economic activity to a proportionality analysis, the
reach of European law would be much broader than is the case when employ-
ing a non-discrimination approach.
Since judicial activism is a matter of degree, it is therefore clearly present in
the ECJ’s case law. However, I argue that activist case-law development does
not require an activist court. Proceeding in a path-dependent way, the Court’s
interest in the consistency of the law leads towards converging interpretations
of the four freedoms and citizenship rights. Individual litigants with easy
access to the Court fuel this development, as they steadily try to improve

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

their regulatory position. The Court does not require much of an activist
orientation to move this process along—it simply has to honour precedent and
legal consistency, and maintain a cooperative spirit with lower courts. It is
itself largely ignorant of its decisions’ precise implications in the heteroge-
neous polities of twenty-eight member states. Operating within its setting, it is
difficult for the Court to practise restraint. Throughout this book, I have
emphasized that the ECJ is not a ‘lone ranger’ pushing the development of
case law along but that there are multiple types of support for its expansionary
case-law development. Its interlocutors push it towards broad interpretations:
litigants address it in order to realize the more extended reach of EU law, and
they are supported by lower courts that hand cases to the ECJ, as well as by the
community of EU legal scholars, who provide arguments to defend the ever-
increasing reach of EU law. Arguments for judicial restraint play a smaller
part in this setting, which is why Davies (2016b) argues that domestic courts
need to take on a more critical role so that the EU legal order has fewer
disruptive effects.
Member states do in fact submit observations to this effect. The Court, as
quantitative analyses can show, does take these into account. Qualitatively,
however, we see in the case of many ‘path-breaking’ decisions that a significant
number of member states intervened but could not prevent case law from
expanding into areas where member states would like to maintain their policy
competences: patient mobility, the residence rights of third-country nationals,
EU citizens’ access to welfare services, and taxation policy are all examples that
were discussed in this book. It is interesting to ask how these different
quantitative and qualitative findings can be reconciled. It is likely that this
difference mirrors the process of fine-tuning case law, where the Court initially
may subject a new policy field to case law and then subsequently adapts and
refines its rulings in order to take member states’ objections into account.
The ruling on the Zambrano case, which I discussed in Chapter 6, allows us
to observe this process of fine-tuning. In the Zambrano case, the Court
granted residence rights to third-country national family members of an EU
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

238 The ECJ and the Policy Process

citizen within purely internal situations, where the EU Treaty normally does
not apply. In the many follow-up cases that came after Zambrano, the Court
took member states’ objections into account and did not broaden the case law.
Instead, they elucidated that it only applies to dependent children of EU
citizens who would otherwise have to leave the EU. The Zambrano case
may, thus, support the findings of quantitative studies, namely that cases
where the Court takes member states’ concerns into account outnumber the
ones where it makes progress in broadening the reach of EU law. But as long as
the Court does not engage in reversion and subsequently denies the relevance
of EU law altogether—something which it is hardly ever known to do—its
responsiveness to member states’ objections does not mean that there has not
been significant case-law development. My analysis has given multiple examples

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

of the constraints that EU law places on policy options. Despite expectations
that member states simply try to ‘contain justice’ (Conant 2002), the study
of member states’ responses shows that EU case law is implemented by
member-state administrations in what appears to be a straightforward rule-of-
law fashion. Based on case studies, it is not possible to say whether these findings
can be generalized. My hunch would be that such compliance is quite common,
and that member states only openly oppose case law when they absolutely
feel that further fine-tuning is necessary. The second Meilicke ruling serves as
an illustration of this. It should be noted, however, that all my examples are
from Western states that have a long tradition of membership and strong
administrations; newer member states may react differently.
So case law makes a difference, and member states respond to it. Even if the
Court provides answers to the concerns of member states in general, the
minority of cases where it does not do so may make an important difference
to the acquis communautaire. In explaining the demands of the Treaty,
much of the Court’s impact depends on the way case law interacts with
policymaking at the EU and member-state levels. As we saw, in proposing
legislation the Commission has to base its proposals on the Treaty, of which
the case law is a part.
Translating case law into general rules in secondary law poses specific
challenges. Courts base their rulings on case-by-case assessments, taking
into account the multiple aspects that make the case unique. The ECJ takes
the broader perspective of devising more general guidelines for interpretation
that are intended to direct national courts in applying EU law. Courts are not
legitimated to devise general policy rules—but legislatures are. However, given
the constitutional nature of case law, the legislature cannot be more restrictive
than the Court. Otherwise, there is a risk of further litigation. But if the Court
has decided that under conditions A, B, C, D, and E someone should have
access to social benefits, what is the legislature supposed to make of it? The
Court often keeps its reasoning opaque, as it is also motivated by the wish to
keep the Treaty’s meaning flexible in order to further future integration. It is
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Conclusion 239
therefore not clear whether conditions A to E all need to be present, whether
any of these conditions are sufficient, or whether condition C is the decisive
one, as it is both necessary and sufficient. In the legislative process, there is
consequently the option to either generalize case law in a way that broadens its
impact, as was done in the regulation on the mutual recognition of goods,
leaving the burden of proof to fall upon the state authorities, or to leave
matters deliberately open, as was the case in the Citizenship Directive. If
there is no EU codification, and member states respond directly with admin-
istrative changes, for instance, they face problems as the Court’s guidelines for
interpreting the Treaty are difficult to generalize and require a case-by-case
approach. Translating constitutionalized case law into general rules, I argue,
has expansive effects.

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

My analysis has shown the considerable policy implications of case law.
These are larger if the Court is particularly activist, but they are also present if
the Court practises some restraint. Rather than searching for activism—which
does, of course, occur at times—it seems to be more useful to analyse judicial
power, following the arguments of Staton and Moore (2011). ‘[B]inding
constraints on governments’ (Staton and Moore 2011: 555) accumulate in
the EU because of the constitutional nature of case law. It is because the EU
has a constitution that consists of a Treaty in which member states have
inscribed multiple policy goals that the ECJ is so important. Since there is
no parallel example at the national or supranational levels of polities being
constrained to this degree in their policy decisions by case law, analysts
overlook the constraints on the EU and the member-state level.
Consequently, we can observe a ‘judicial coup d’état’ (Stone Sweet 2007) at the
EU level, the consequences of which are rarely recognized. However, we should
not be misled about the extent to which courts can change societal development.
This remains a ‘hollow hope’ (Rosenberg 1993). Courts are essentially reactive
actors, dependent on others to bring cases before them and to implement their
rulings. The ECJ is therefore also responsive to the way that member-state
governments intervene, but this is not enough if case law lives on as part of the
constitution. Although much of it is codified and implemented as standard rule-
of-law practice, there is the latent danger that its constraints will foster opposi-
tion. The EU referendum debate reflected this. Part of the UK’s discontent was
being forced to a judicially driven position of non-discrimination against EU
citizens, alongside having to engage in cumbersome administrative procedures
related to EU citizens’ access to social benefits. To the extent that leaving the EU
and having a voice within it are related, following Hirschman (1970), the UK’s
inability to express themselves influentially on the laws of citizenship must be
seen as one factor that drove their exit from the EU. Within this context of
growing political opposition, the Court has recently practised self-restraint in
relation to EU citizens’ access to non-contributory social benefits in host coun-
tries where they have recently established themselves, surprising observers that
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

240 The ECJ and the Policy Process

had expected more activist rulings. With such self-restraint, the Court has not
abandoned its path of interpreting the Treaty, but it has refrained from extending
its meaning further to cover EU citizens’ full and equal access to benefits. This
strengthens legal certainty about the reach of EU law vis-à-vis national law. The
example shows that the Court can indeed choose between activism and self-
restraint, but this choice should not overshadow the importance of over-
constitutionalization and codification.

TIME

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Differences in time horizons between the judiciary and political actors, who
are dependent on the electoral cycle, are a central aspect in understanding
judicial power. Time matters in several respects. It is known to be important
for the development of legal doctrine. Alter argued early on that the Court
may use time in its favour when it establishes new interpretations in cases
where it first denies their relevance (Alter 1998: 131). In this way, the Court
can test the waters. It is expected that opposition from member states will be
lower if the new interpretation does not have an immediate impact. In later
cases, then, the Court can refer to established precedent if it brings the
principle to bear upon them, which shields the Court from political interven-
tion (Kelemen 2001).
Taking into account how time can help us to understand the central puzzle
that my analysis raises, we can ask the following important question: if
constitutional case law restricts European policymaking, as I demonstrated
in the analysis I conducted in Chapters 4 and 5, and also imposes multiple
constraints at the national level, as I discussed in Chapters 6 and 7, how can we
explain the way in which the research on integration largely neglects this fact?
Why is there no coordinated effort from member-state governments to halt
the EU’s processes of over-constitutionalization? (And why do member states,
in contrast, continue to inscribe further policy goals into the Treaty? This a
question that I will discuss in the next subsection.)
The constraints of case law accumulate over time. Case law develops
incrementally, and, as is true for all incremental changes, it is difficult to
muster the political will to respond to such development. As part of these
incremental shifts, the general implications of case law are not initially clear.
In contrast to legislation, requirements from case law do not come with
implementation deadlines. The Meilicke cases, which were discussed in
Chapter 6, illustrate this in a striking way. Fifteen years passed between the
first preliminary ruling that was addressed to the ECJ and the national court’s
final ruling. Even today, the judicial process around the Meilicke case may
continue due to the complaint that was handed to the Commission. With such
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Conclusion 241
long periods of legal uncertainty, during which the precise implications of EU
law for national policy are left vague, it is likely that incremental steps will be
taken to lessen vulnerability to a worst-case scenario at the national level. Such
steps are part of ‘anticipatory obedience’ (Blauberger 2014: 468), which is
difficult to link to case law.
In addition, political and academic debate on the Services Directive demon-
strated the neglect of the constraints of case law. However, even if some political
actors fail to comprehend the constraints of case law or are unwilling to accept
them, as we saw, the Commission, in its proposals for secondary law, honours the
case-law constraints. However, there is no need to resort to conspiracy theories.
Many governmental actors are aware of the constraints the Treaty places on
policy options. Not all constraints are unwelcome in day-to-day governmental

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

policymaking. In addition, and more importantly, member-state governments’
preferences should not be regarded as unmovable and fundamentally opposed to
those at the EU level. The EU has long developed into a multilevel system
in which few national preferences are as clear-cut as intergovernmentalism
assumes. Moreover, member states are not gatekeepers to the supranational
system, as different actors have access to EU policymaking—and to EU litigation.
Most of these actors who push for EU legal constraints on domestic policies are
also important actors for national governments. As I have argued, governmental
preferences in the medium to long term are fluid and responsive to the case-law
constraints that arise. Litigation at the Court shows governments the policy
preferences of important national actors, and the constraints of EU case law
facilitate the necessary national reforms. ‘Blame avoidance’ (Weaver 1986),
traditionally, is an important asset that European integration offers to member-
state governments. Edgar Grande has termed this a ‘paradox of weakness’, as
governments are strengthened vis-à-vis societal actors by being able to point to
EU constraints (Grande 1996).
Incremental change, the long drawn-out judicial process, fluid governmen-
tal preferences, and the interests of governments in facilitating domestic
reforms by being able to draw on EU constraints all consequently explain
our puzzle, in which political interventions against over-constitutionalization
are absent. Importantly, if governments can live with the outcomes of ECJ case
law in most instances—if only at a particular moment, in comparison to the
cost of intervention against an independent court—this does not equate to
their consent, or to an existing majority to back the development.

LEGITIMACY

Is the constitutionalization of case law problematic? After all, it is member

states’ governments that agree to add detailed material policy aims to the
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

242 The ECJ and the Policy Process

Treaty, Treaty reform after Treaty reform. Since supremacy and direct effect
date back to the early 1960s, governments have had ample time to observe the
consequences of having such detailed Treaty rules. As governments are well
legitimated, there is no reason to problematize the workings of the ECJ, which
essentially does what it has to do: it interprets relatively vague Treaty articles
and derives solutions to the dispute at hand from them. Christoph Möllers
(2015) makes this argument. In light of what I have said about the importance
of time, I would hesitate to legitimize over-constitutionalization in this way.
Given their more short-term political perspective, member-state governments
do not attach much political importance to defending more long-term policy
options. In agreeing to and pushing for detailed rules during Treaty reforms,
they often respond to the demands of important constituents. Since national

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

constitutions contain significantly different constraints to the European one,
societal actors have little basis for understanding the implications of writing
their concerns into the Treaty.
An analysis of Treaty reforms was not the subject of this book. However,
we have seen in the examples of the Services Directive and the Patient
Mobility Directive how societal actors aspire to exempt certain areas from
a directive: social services in the example of the Services Directive, long-term
care in the example of the Patient Mobility Directive. But exempting an area
from a directive does not mean that it falls under national prerogative. As the
Court states again and again through its ‘retained powers formula’, member
states also have to comply with EU law in the areas where they exercise
their retained powers, for instance in the fields of taxation, education, or
social services.

Literally this formula means that the scope of application of EU law extends
beyond the subject areas over which the EU has been given jurisdiction. By
dissociating the existence of state powers from the exercise of such powers, the
Court legitimizes the application of EU law in any domain that is not a priori
within the Union’s scope of intervention.
(Azoulai 2011: 194, emphasis in original)

This stands in marked contrast to the attempts of member states to circum-

scribe the reach of EU law, for instance with the subsidiarity principle. Despite
frequent Treaty revisions, member states do not legitimize the extensive reach
of EU law. There are additional arguments from the rich debate on the
legitimation of European integration that are relevant in this context.
Joerges makes an interesting argument for why the EU strengthens legit-
imation. He refers to the democratic deficit member states experience when
their decisions have cross-border effects without those that are affected having
a say in these decisions. To reconcile these different interests, he argues for a
conflict-of-laws approach and regards the supremacy of EU law as compen-
sating for this deficit of national decision-making (Joerges 2007). Neyer (2012)
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Conclusion 243
expands on this argument, embedding it in political theory. However, the
extensive broadening of EU law through the legal profession can hardly be
encapsulated within this argument. Joerges himself agrees that, ultimately,
input legitimacy is decisive for supranational law. Similarly, Conway makes
the criticism that the extensive interpretation of the Treaty by the Court is
political and not legitimized by output legitimacy.
Innovative interpretation downplays traditional rule of law and democratic
objections to an ‘activist’ judiciary or ‘gouvernement de juges’. Its main justifica-
tion relates to consequentalist or output-oriented legitimacy. This often implicit
reliance on output legitimacy, is, however, problematic, since the outcome that is
claimed to be good in the EU, enhanced integration, has no apolitical validity.
Conversely, conserving or originalist interpretation privileges input-legitimacy,

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

which dovetails with a traditional rule of law emphasis on formal legality under-
stood as requiring certainty and predictability in the law. (Conway 2012: 113)
I demonstrated in Chapter 3 how the extensive interpretation of the free
movement of goods serves as a focal point for the subsequent interpretation
of the other freedoms. The far-reaching case-law development culminates
in the extension of citizenship rights, which link citizens directly to the
EU. Setting up and strengthening this direct relationship, citizenship rights
receive particular normative elevation. Wiesbrock has shown how legal
scholars and practitioners, most notably the AGs, establish a core of citizen-
ship rights through self-citation that are far removed from what member states
originally intended.
The interpretation and analysis of the Court’s judgments by legal scholars has to a
large extent served to justify the expansive case law of the Court. EU legal scholars
have generally defended the ‘quasi-legislative’ role of the Court. . . . EU legal
scholars, who have an active interest in the extension and increasing importance
of EU law, have thought to present the progression towards more extensive free
movement rights as an inevitable step towards further integration.
(Wiesbrock 2013b: 148)
My analysis of the difficulty of translating case law, which has a constitutional
status, into general secondary law and/or administrative practice adds a
further legitimacy concern to supremacy: the danger of inequality before the
law. Legal uncertainty about the reach of EU law automatically leads to
heterogeneous decisions. The rationale of the preliminary procedure is the
necessity of ‘preserving the unity of EU law’ (Chalmers et al. 2010: 160–2).
However, the sheer complexity of the regulatory framework, which is driven
by case law, is actually resulting in ‘dis-unity’, as national administrations and
EU citizens will make sense of and practise the complicated legal regimes in
very different ways. When applying ECJ case law at the member-state level, the
requirement for case-by-case assessments on the basis of complex, unfamiliar
EU law leads to highly divergent decisions among administrators.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

244 The ECJ and the Policy Process

Within the EU polity, the EP is the actor that claims that it is strengthening
the legitimation of the integration process. It should therefore be the actor that
politically confronts the judicialization of politics, criticizing the withdrawal of
policy options from majoritarian rule. Within the cases I have discussed, the
EP took a particularly prominent position in the debate on the Services
Directive. It largely ignored the case-law constraints, but the Commission’s
implementation document for the directive reflected more closely the position
of the Court than that of the EP (European Commission 2007). It is doubtful
whether the EP could develop into a bigger opponent of the ECJ. In general,
one could argue that the EP, as a whole, is less concerned with the material
impact of policy decisions than the Council would be. This can be identified in
its successful strategic extension of its competences, as analysed by Farrell and

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Héritier (2003). They show that, wherever possible, the EP puts its own
institutional interests before material policy interests. Apparently, the EP is
less likely to be punished in elections for failing to pursue material policy
interests than member-state governments, otherwise it would be impossible to
pursue this strategy. And since the EP will always favour more rather than less
integration—as long as eurosceptic parties do not dominate—it is highly
unlikely that it will openly oppose the constraints on policy options set by
the Court.
The accumulating constraints of case law imply that the politicization of
EU politics will encounter difficulties in fostering legitimation (Zürn 2016).
Politicization is expected to benefit European integration by showing that
there are different options for shaping integration politically. In this way, it
may help to overcome the ‘constraining dissensus’ of public opinion (Hooghe
and Marks 2009) and avert the danger of disintegration (Webber 2014). The
literature on the democratic deficit has argued strongly in favour of increased
politicization (Føllesdal and Hix 2006; Kriesi 2009). The Services Directive is
a prominent example here. Clearly, though, analysing this case makes it
questionable why one would aim to increase legitimation through politiciza-
tion, if what matters in the end is the ECJ’s interpretation of the Treaty.
Bartolini (2006) emphasized at an early stage that the ECJ’s case law serves
as a counter-argument to politicization. A mere symbolic discussion of policy
options, given legal constraints, rather delegitimizes the political order of
the EU.
It is also useful to compare the EU level to the national level. In those
member states that have strong constitutional courts, their legitimation to
protect basic liberal rights is not contentious (Syrpis 2012), but constitutional
courts tread much more carefully when dealing with social, positive rights, in
relation to which they heed judicial self-restraint (Sieberer 2006). It is not just
for courts to decide how taxpayers’ money should be spent. In relation to
social rights, national constitutional courts do in fact intervene when they wish
to guarantee non-discrimination (Sieberer 2006). In a way, this is also what the
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Conclusion 245
ECJ is doing, as much of its citizenship case law responds to discriminatory
practices in the granting of social rights under conditions of increasing
transborder mobility. However, in granting social rights, the ECJ intervenes
in the allocation of welfare-state services, which rely on an informal contract
among creditors and beneficiaries within a system characterized by national
and not European membership. This risks a domestic political backlash.
In terms of the legitimacy of the EU, therefore, a restrictive interpretation of
the Treaty appears to be more desirable from a normative perspective.

The perspective argued for in this chapter is that conserving or originalist

interpretation is both epistemically possible and normatively preferable than
evolutive interpretation. The latter rests on essentially political preferences as to
what is a desirable outcome and is difficult to reconcile with the requirement of

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

predictability as a key feature of formal legality and with the democratic authority
of the law-maker (and the comparative lack of democratic or representative legit-
imacy of the judiciary, despite occasional, strained arguments to the contrary).
(Conway 2012: 106)

Moreover, it should not be forgotten that my analysis highlighted how the

supremacy of EU law threatens to undermine legitimacy at the national level.
Mobile actors are, at times, more equal than others and can use EU law as an
alternative set of rights, which partly results in conspicuous inequalities, for
instance when one considers the rights of third-country national family
members, as I discussed in Chapter 7.
Finally, reference should be made to the difference between negative and
positive integration. I have not emphasized this difference; as such, it has
remained a somewhat hidden dimension of what I have analysed throughout
this book. Scharpf has repeatedly shown how there is a bias towards nega-
tive integration in the EU, as liberalization policies can draw directly on the
Treaty and the Court, while positive integration, in the form of harmoniza-
tion policies, requires demanding legislative processes (Scharpf 1996, 1999,
2000b). This bias between negative and positive integration in the EU is
relevant not only because some policies are favoured over others at the
European level, namely liberal over republican ones (Scharpf 2009, 2010),
it is, moreover, normatively important as member states with their hetero-
geneous political-economic institutions are constrained in very different
ways by this kind of European integration (Callaghan and Höpner 2005;
Höpner and Schäfer 2010, 2012). Although I have not emphasized these
implications, this does not mean that I do not share these concerns. How-
ever, my suspicion is that the debate on positive and negative integration has
partly hidden the importance of over-constitutionalization and the Court. The
normative dimension of positive and negative integration, it appears to me,
has led observers to take sides, either welcoming or criticizing liberalization
effects. What has been overlooked is the importance of the judicialization of
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

246 The ECJ and the Policy Process

policymaking, through which the constitutional nature of case law takes
political decisions away from majoritarian decision-making. To me, this is
the more fundamental normative dimension, and it is most likely to be one
around which critics and supporters of negative integration could unite.

THE RULE OF L AW BEYOND THE NATION S TATE

What can we learn from my analysis of the discussion about international

courts? International courts have grown tremendously in importance. The
ECJ, with its far-reaching competence and relatively low threshold of access, is

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

a ‘model’ for international courts (Alter 2014). What does this model provide?
A Treaty, as a constitution, presents very different challenges than those we
are familiar with in a national constitution. The latter focuses on rules for state
organization and the protection of individual liberal rights. A Treaty lays
down material policy goals. It is likely that the specific institutional conditions
of the EU allow for greater dynamism than we see in other areas of supra-
national law. The thresholds to access the ECJ are lower than are found
elsewhere, resulting in more case-law development. However, under less
favourable conditions there may also be important constitutionalized case-
law development, as the example of the European Court of Human Rights
shows (Stone Sweet 2012). Due to international anarchy and massive human-
rights violations, actors in international relations predominantly discuss
increasing judicialization as a positive development (Zangl 2005; Zangl and
Zürn 2011). However, there are also important studies on the costs of this
development (Hirschl 2008; Bellamy 2013).
My analysis has shown how the ECJ’s judicial power leads to far-reaching
constraints on policymaking in the EU. One might argue that judicial inde-
pendence is instituted in order to foster such constraints. However, in the EU
the constraints are so far-reaching as to remove otherwise feasible policy
options from the agenda (Scharpf 2016). This is the consequence of the
detailed policy prescriptions in the Treaty. As a result of the ECJ’s institutional
set-up, highly activist case-law development may follow without the need for
judges to overstretch their mandate constantly, as I have shown. Supremacy
and direct effect constitutionalize the material policy provisions of the Treaty,
and the case law that spells out its content is difficult to translate into
meaningful general administrative rules or secondary law. Each ruling
provides for a process of balancing in a single and often unique situation.
The different conditions that lead to a decision do not provide a coherent
whole, and the latter cannot be designed politically under the constitutional
constraints of this case law. While it is convincing to legitimate supremacy and
direct effect with the need to compensate for the negative externalities of
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Conclusion 247
national democratic rules for other member states, as Joerges and Neyer do
(Joerges 2007; Neyer 2012), the consequences for problem-solving are over-
looked. This is the main point this book wishes to communicate about
international law that aims to create binding, constitutionalized obligations
for members. However, the EU has a unique legislature, which allows for the
interaction of judicial and legislative policymaking. With less dynamic case-
law development in other international regimes, and no legislature to codify
these rulings, pressures will be lower.
Supranational constraints arise as a result of a supranational legal order
supplanting national legal orders. Such overlapping of legal orders also has
normative consequences. I have emphasized that the dynamic interpretation
of supranational law results in legal uncertainty as to the reach of the supra-

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

national Treaty. Legal uncertainty, as a possible consequence of integration in
the EU, is interesting because it contrasts with expectations in the literature on
international relations of increased legal certainty resulting from international
trade regimes (Zürn 2001). This argument relates to the increased certainty of
trade under an international regime in comparison to the previous situation of
relatively unconstrained national competence over the conditions of market
access, complete with all its surprises. The expectation of increased legal
certainty may depend, however, on an unrealistic assumption of relatively
clear borders between different legal regimes. As this book has shown,
dynamic supranational case-law development constantly questions such de-
marcations. In addition, when they use their retained competences, member
states have to comply with the demands of the Treaty. A comparable situation
of legal uncertainty arose, for instance, when Canada challenged the French
ban on asbestos as a barrier to trade under World Trade Organization rules in
the 1990s (Howse and Tuerk 2001). Increased legal certainty with regard to
transborder transactions coincides with increased uncertainty about the legal-
ity of domestic rules.
The lack of legal certainty that results from the unclear demarcation of
different legal regimes is also problematic in another respect. It makes some
actors more equal than others, as they can seek the legal regime that is most
appropriate for them. In the EU, mobile actors are favoured, and while the four
freedoms are rightly needed to compensate for discrimination against transbor-
der situations by national rules, whenever they lead to reverse discrimination
against nationals, they overcompensate and invite rule arbitration. If some
people are able to pick and choose the legal regime that is most beneficial to
them, within the context of overlapping regimes, inequality is the result. To the
extent that the rule of law constrains public authority, overlapping legal regimes,
by adding legal justifications, expand the exercise of state power.
Reflecting on the rule of law in the EU, recent contentious reforms of
the judiciary in Hungary and Poland have highlighted the limits of the
EU’s competence when confronting member states’ deviations from judicial
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

248 The ECJ and the Policy Process

independence. Despite the details it contains, in this central task the EU’s
constitution lacks competence and the tools of intervention for strengthening
the rule of law in member states (Blauberger and Kelemen 2016; Kelemen and
Orenstein 2016). This failure reveals the peculiar nature of the EU’s constitu-
tion, as it is strong on material policy goals and weak on classic elements of
the rule of law.

WHERE S HOULD P OLITICAL SCIENCE

GO FROM HERE?

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Constitutions ensure the rule of law and constrain majoritarian decision-
making. Over-constitutionalization hinders democratic politics. After half a
century of case-law development, the disadvantages of EU constitutional
constraints shaping material policy decisions in significant detail have become
increasingly apparent. Any constitution, particularly if it is only implicitly
recognized, as the Treaty is, can only play its role when it concentrates on
those principles that require constitutional status. For the EU, this would
imply taking the four freedoms and competition law out of the Treaty,
which would be a very far-reaching step indeed. If we try to imagine how
European integration would have progressed without these fundamental
cornerstones of the Treaty, it shows this clearly. Where would the single
market be without the Cassis ruling? In its absence, it is most likely that
there would be no mutual recognition, for instance. At the same time, it
should not be forgotten that the harmonization of rules is much more far-
reaching in the EU single market than it is in the US (Egan 2015).
In times of crisis, as these are, it is particularly unlikely that the Treaty’s
scope will be cut back to such an extent. However, there are also less radical
ways to counter the development described in this book. Some years ago,
Scharpf argued that the European Council should be empowered to check the
Court’s case law. Whenever a member state feels unduly constrained by case
law, Scharpf argues, the state should make the ruling’s recognition subject to
the backing of a majority of governments in the Council. In this way, case-law
development could be checked, while keeping the credible commitment func-
tion of the Court intact (Scharpf 2009, 2010).
By providing a check on the judiciary through the European Council, this
proposal could be criticized for violating the balance of powers and breaching
judicial independence. In light of this book’s analysis, this concern appears to
be less relevant. In the EU, taking over-constitutionalization into account, the
Court encroaches upon legislative decision-making to an extent that goes
beyond what occurs in the national realm. The balance of powers in the EU
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Conclusion 249
is already skewed. Therefore, it does not seem problematic to subject the Court
to enhanced political control.
The analysis has shown, moreover, that there are not many alternatives.
Secondary law does not overrule case law that is based on the Treaty. Member
states can use secondary law to signal their political preferences to the Court,
but the latter is free to decide as it sees fit. The case law on patient mobility
illustrates very clearly how little the Court feels itself to be bound by the
legislature’s wishes. In this area, secondary law provided rules on how cross-
border healthcare should be reimbursed. With its different rulings, the Court
established an alternative regime alongside the regulation. In order to provide
greater legal certainty and improve equality before the law, member states
codified these rules in the Patient Mobility Directive. However, since it was

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

impossible to alter the constitutionalized case law, there is now a regime in
place that is overly complicated and contains several contradictions between
regulation 884/2004 and the directive. If we take this example and assume that
those governments that intervene at the Court would have backed an inter-
vention against these rulings in the European Council, this would have been
an early signal to the Court to stop this line of case law.
In Cases C-120/95 Decker and C-158/96 Kohll, which began the develop-
ment, seven governments (Luxembourg, Belgium, Germany, Spain, France,
the Netherlands, and the UK) intervened in Decker and six governments
(Luxembourg, Germany, Greece, France, Austria, and the UK) intervened in
Kohll. In the next significant case, C-157/99 Smits/Peerbooms from July 2001,
ten governments intervened (the Netherlands, Belgium, Denmark, Germany,
France, Ireland, Portugal, Finland, Sweden, and the UK, alongside Iceland and
Norway in the European Economic Area). Since there were only fifteen
member states at the time, at this point it would have at least been possible
to rein in the Court.
In light of the limits of political intervention, as things stand everything
depends on whether the Court fine-tunes its case law once it learns of
undesirable consequences. In the case of patient mobility, this fine-tuning
did take place, according to the assessment of Obermaier (2008a), who
analysed this area. Taking into account the overly complex legal regime that
currently exists, however, it has to be judged as too little, too late. In the recent
decisions the Court has made in the area of EU citizens’ access to non-
contributory social services, we can witness another instance of fine-tuning.
After a long period of continuously expanding EU citizens’ right to access
social services in other member states, the Court has become more cautious
since the Dano case. The Court previously always demanded individual
assessments of whether a person was sufficiently integrated into a member
state or whether an undue burden on the member state was the result of
broadening access to social benefits. Recently, however, the Court has acqui-
esced to the legislator’s wishes and has made reference to the period of
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

250 The ECJ and the Policy Process

residence, as set down in the Citizenship Directive. In cases where EU citizens
are not seeking work at all or have only arrived in a member state recently,
member states can refuse to offer support.
However, it is questionable whether it is sufficient to rely on the Court’s
own assessment of where to draw the line in terms of the extension of rights.
In the area of social services, fine-tuning took place against mounting criti-
cism, of which the UK’s EU referendum discussions were the most prominent
sign. As we have seen, through its right to reside test, the UK aimed to restrict
EU citizens’ access to benefits, but they encountered significant legal con-
straints in doing so. The UK’s administrative response to the case law on social
benefits demonstrates the difficulty of implementing the Court’s requirements
in daily administrative practice. Although the Court is the actor that is least

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

held back from doing about-turns, there is no reason to assume that this
question should be left solely in its hands. Davies (2016b) argues that member-
state courts have to be much more critical when engaging with the ECJ.
Scharpf (2015) has proposed the ‘deconstitutionalization of European law’,
meaning there would no longer be an economic constitution. In addition to
different rules for opt-outs, which would allow differentiated integration,
member states would be able to repoliticize the current acquis, opening it up
to majority rule and the political preferences of member states. Related to this
change would be the need to make political decisions on how to order markets,
rather than hiding behind judicialization. In another paper, he has argued that
infringement and preliminary procedures should be restricted to secondary
law, as a way to contain case-law development and make it subject to major-
itarian decision-making to a greater degree (Scharpf 2016: 14). In order to
enable a larger win-set, he also proposes cutting the option of Commission
vetoes by abolishing its monopoly of legislative initiative.
Academic monitoring of the development of judicial power in the EU that is
more critical will certainly help lead towards its more sensible use. Of the
different cases that I have analysed to show the interaction of ECJ case law and
EU legislation—the Citizenship Directive, the Services Directive, the regula-
tion on the mutual recognition of goods, and the Patient Mobility Directive—
it is notable that the broad attention that has been paid to the Services
Directive in political science shows how little the constitutionalized nature
of case law is understood within the discipline. For the study of legislative
processes in the EU, it is common to assume that—as the analogy to domestic
political systems has long suggested—the default condition of European
decision-making is sustained national competence or previous EU legislation,
but not case law. Moreover, even in instances where case law is taken into
account, and the interaction of judicial and legislative policymaking is explic-
itly analysed (Martinsen 2015), it is normally assumed that politics has the
upper hand, and secondary law can overrule the judicial interpretation of the
Treaty. EU lawyers, in contrast, are aware of the constraints, to the extent that
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Conclusion 251
they declare the legislature to be an ‘agent’ of the Court (Davies 2016a).
However, the teleological interpretation of the Court is generally supported,
and critical analyses of European legal integration are rare (Conway 2012).
The same pretty much holds for political science in which increased integra-
tion has largely been welcomed (Majone 2014). It may be that the debacle
around the euro, a stage in integration that few politicians and political
scientists would probably support again, will bring about a change here.
In the national context, we find very different conditions. Constitutional
courts are embedded in the national political discussion. To the extent that
their rulings have policy implications, judges can assess these quite well. Since
national constitutions are much leaner than the Treaty, national constitution-
alized case law has fewer policy implications. If case law is not supported by

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

politics, it will be overruled by secondary law in the national context. In the
EU, in contrast, this is not possible. Treaty changes are hardly feasible. The
inscription of national prerogatives into the Treaty are always subject to
the Treaty’s goals and provide little help. However, the inability of politics
to rein in the Court does not make the latter more powerful in the long run. In
the lack of incremental political feedback the Court receives on its case-law
development, there is the clear and present danger of radical responses that
question European integration in itself.
 Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Bibliography

Abbott, Kenneth W., and Duncan Snidal, 2000: Hard and Soft Law in International
Governance. International Organization, 54(3), 421–56.
Abbott, Kenneth W., et al., 2000: The Concept of Legalization. International Organ-
ization, 54(3), 401–19.
Abromeit, Heidrun, 1995: Volkssouveränität, Parlamentssouveränität, Verfassungs-
souveränität: Drei Realmodelle der Legitimation staatlichen Handelns. Politische
Vierteljahresschrift, 36(1), 49–66.
Acemoglu, Daron, and James A. Robinson, 2012: Why Nations Fail: The Origins of

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Power, Prosperity, and Poverty. New York: Crown Business.
Ackermann, Thomas, et al., 2008: Editoral Comments: Towards an Improved Frame-
work for Cross-Border Healthcare. Common Market Law Review, 45(5), 1325–33.
Adam, Christian, 2015: Gambling: The Erosion of Domestic Sports Betting Monop-
olies. In Christoph Knill, Christian Adam, and Steffen Hurka (eds), On the Road to
Permissiveness? Change and Convergence of Moral Regulation in Europe. Oxford:
Oxford University Press, 206–33.
Alter, Karen J., 1998: Who are the ‘Masters of the Treaty’? European Governments and
the European Court of Justice. International Organization, 52(1), 121–47.
Alter, Karen J., 2000: The European Union’s Legal System and Domestic Policy:
Spillover or Backlash? International Organization, 54(3), 489–518.
Alter, Karen J., 2001: Establishing the Supremacy of European Law: The Making of an
International Rule of Law in Europe. Oxford: Oxford University Press.
Alter, Karen J., 2008a: Agents or Trustees? International Courts in their Political
Context. European Journal of International Relations, 14(1), 33–63.
Alter, Karen J., 2008b: The European Court and Legal Integration: An Exceptional
Story or Harbinger of the Future? In Keith E. Whittington, R. Daniel Kelemen, and
Gregory A. Caldeira (eds), The Oxford Handbook of Law and Politics. Oxford:
Oxford University Press, 209–29.
Alter, Karen J., 2009: The European Court’s Political Power: Selected Essays. Oxford:
Oxford University Press.
Alter, Karen J., 2014: The New Terrain of International Law: Courts, Politics, Rights.
Princeton: Princeton University Press.
Alter, Karen J., and Sophie Meunier-Aitsahalia, 1994: Judicial Politics in the European
Community: European Integration and the Pathbreaking Cassis de Dijon Decision.
Comparative Political Studies, 26(4), 535–61.
Alter, Karen J., and Jeannette Vargas, 2000: Explaining Variation in the Use of
European Litigation Strategies: European Community Law and British Gender
Equality Policy. Comparative Political Studies, 33(4), 452–82.
Anagnostaras, Georgios, 2013: Activation of the Ultra Vires Review: The Slovak
Pensions Judgment of the Czech Constitutional Court. German Law Journal,
14(7), 959–73.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

254 Bibliography
Arcarazo, Diego A., and Andrew Geddes, 2013: The Development, Application and
Implications of an EU Rule of Law in the Area of Migration Policy. Journal of
Common Market Studies, 51(2), 179–93.
Argyris, Nicholas, 1989: The EEC Rules of Competition and the Air Transport Sector.
Common Market Law Review, 26(5), 3–32.
Arnull, Anthony, 1993: Case Law—Case C-83/91, Wienand Meilicke v. ADV/ORGA
F.A. Meyer AG, Judgment of 16 July 1992. Common Market Law Review, 30(3), 613–22.
Asteriti, Alessandra, 2013: Social Dialogue, Laval-Style. European Journal of Legal
Studies, 5(2), 58–79.
Aujean, Michel, 2007: The Future of Non-Discrimination: Direct Taxation in Com-
munity Law. In Reuven S. Avi-Yonah, James R. Hines Jr, and Michael Lang (eds),
Comparative Fiscal Federalism: Comparing the Court of Justice and the US Supreme
Court’s Tax Jurisprudence. Alphen aan den Rijn: Kluwer Law International, 321–30.

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Avi-Yonah, Reuven S., James R. Hines Jr, and Michael Lang, 2007: Comparative Fiscal
Federalism: Comparing the European Court of Justice and the US Supreme Court’s
Tax Jurisprudence. Alphen aan den Rijn: Kluwer Law International.
Aziz, Miriam, 2009: Implementation as the Test Case of European Union Citizenship.
Columbia Journal of European Law, 15(2), 281–98.
Azoulai, Loïc, 2011: The “Retained Powers” Formula in the Case Law of the European
Court of Justice: EU Law as Total Law? European Journal of Legal Studies, 4(2),
192–219.
Bachrach, Peter, and Morton S. Baratz, 1962: The Two Faces of Power. American
Political Science Review, 56(4), 947–52.
Baddenhausen, Heike, 2007: Der Grundsatz gegenseitiger Anerkennung im freien
Warenverkehr—Dassonville, Keck, Cassis de Dijon: Haben die Klassiker ausgedient?
Deutscher Bundestag—Wissenschaftliche Dienste, 14(7). <https://www.bundestag.
de/blob/504298/ee6936378a53234f5f0ed014c37f9c4d/freier-warenverkehr-data.pdf
(accessed 17 October 2017).>
Barani, Luca, 2005: The Role of the European Court of Justice as a Political Actor in the
Integration Process: The Case of Sport Regulation After the Bosman Ruling. Journal
of Contemporary European Research, 1(1), 42–58.
Barnard, Catherine, 2001: Fitting the Remaining Pieces into the Goods and Persons
Jigsaw? European Law Review, 26(1), 35–59.
Barnard, Catherine, 2009: Restricting Restrictions: Lessons for the EU from the US?
Cambridge Law Journal, 68(3), 575–606.
Barnard, Catherine, 2010: The Substantive Law of the EU. The Four Freedoms. 3rd edn.
Oxford and New York: Oxford University Press.
Barnard, Catherine, 2013: The Substantive Law of the EU. The Four Freedoms. 4th edn.
Oxford and New York: Oxford University Press.
Bartels, Brandon L., 2009: The Constraining Capacity of Legal Doctrine on the
U.S. Supreme Court. American Political Science Review, 103(3), 474–95.
Bartels, Lorand, 2005: The Legality of the EC Mutual Recognition Clause under WTO
law. Journal of International Economic Law, 8(3), 691–720.
Bartolini, Stefano, 2006: Mass Politics in Brussels: How Benign Could it Be? Zeitschrift
für Staats- und Europawissenschaften, 4(1), 28–56.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Bibliography 255
Basedow, Jürgen, 2012: The Judge’s Role in European Integration: The Court of Justice and
its Critics. In Hans-W. Micklitz and Bruno De Witte (eds), The European Court of Justice
and the Autonomy of the Member States. Cambridge: Intersentia Publishing, 65–79.
BDI and PWC, 2011: Verlustberücksichtigung über Grenzen hinweg: Vergleichende
Gegenüberstellung der Verlustverrechnungsmöglichkeiten in 35 Ländern. Berlin and
Frankfurt a.M.: BDI and PWC.
Beach, Derek, 2001: Between Law and Politics: Taking the Law Seriously in Rationalist
Models of Judicial Autonomy in the EU. Paper presented at the ECSA Seventh
Biennial International Conference, Madison, 31 May–2 June.
Beaudonnet, Laurie, 2014: A Threatening Horizon: The Impact of the Welfare State on
Support for Europe. Journal of Common Market Studies, 53(3), 1–19.
Becht, Marco, Colin Mayer, and Hannes F. Wagner, 2008: Where do Firms Incorporate?
Deregulation and the Cost of Entry. Journal of Corporate Finance, 14(3), 241–56.

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Beck, Gunnar, 2012: The Legal Reasoning of the Court of Justice of the EU. Oxford and
Portland, OR: Hart.
Becker, Hilkka, 2014: The Law on Family Reunification in Ireland. Irish Journal of
European Law, 17(1), 103–10.
Bellamy, Richard, 2007: Political Constitutionalism: A Republican Defence of the
Constitutionality of Democracy. Cambridge: Cambridge University Press.
Bellamy, Richard, 2008a: The Democratic Constitution: Why Europeans Should Avoid
American Style Constitutional Judicial Review. European Political Science, 7(1), 9–20.
Bellamy, Richard, 2008b: Evaluating Union Citizenship: Belongings, Rights and Par-
ticipation within the EU. Citizenship Studies, 12(6), 597–611.
Bellamy, Richard, 2013: The Democratic Qualities of Courts: A Critical Analysis of
Three Arguments. Representation, 49(3), 333–46.
Bellamy, Richard, 2015: A Duty-Free Europe? What’s Wrong with Kochenov’s
Account of EU Citizenship Rights. European Law Journal, 21(4), 558–65.
Beyer, Jürgen, 2005: Pfadabhängigkeit ist nicht gleich Pfadabhängigkeit! Wider den
impliziten Konservatismus eines gängigen Konzepts. Zeitschrift für Soziologie, 34(1),
5–21.
Biersteker, Anouk, Lukas Dziedzic, and Lorena Navia Rodriguez, 2015: Right of
Residence under Article 20 TFEU of the Primary Carer of a Minor Union Citizen:
Summary of an Expert Opinion on Issues Arising from the Pending Case of Chavez-
Vilchez and Others(C-133/15). 1st edn. Amsterdam: Migration Law Clinic.
Blauberger, Michael, 2012: With Luxembourg in Mind … The Remaking of National
Policies in the Face of ECJ Jurisprudence. Journal of European Public Policy, 19(1),
109–26.
Blauberger, Michael, 2014: National Responses to European Court Jurisprudence.
West European Politics, 37(3), 457–74.
Blauberger, Michael, and R. Daniel Kelemen, 2016: Can Courts Rescue National
Democracy? Judicial Safeguards Against Democratic Backsliding in the EU. Journal
of European Public Policy, 24(3), 321–36.
Blauberger, Michael, and Rike U. Krämer, 2014: Europeanisation with Many
Unknowns: National Company Law Reforms After Centros. West European Politics,
37(4), 786–804.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

256 Bibliography
Blauberger, Michael, Tilman Krüger, and Susanne K. Schmidt, 2012: Die Pfadabhän-
gigkeit internationaler Verrechtlichung. EU und WTO im Vergleich. Zeitschrift für
Internationale Beziehungen, 19(1), 37–63.
Blauberger, Michael, and Susanne K. Schmidt, 2014: Welfare Migration? Free Movement
of EU Citizens and Access to Social Benefits. Research and Politics, 1(3), 1–7.
Blauberger, Michael, and Susanne K. Schmidt, 2017: Free Movement, the Welfare
State, and the European Union’s Over-Constitutionalization: Administrating
Contradictions. Public Administration, 95(1), 1–13.
Blauberger, Michael, and Moritz Weiss, 2013: ‘If you can't beat me, join me!’ How the
Commission Pushed and Pulled Member States into Legislating Defence Procure-
ment. Journal of European Public Policy, 20(8), 1120–38.
Bleckmann, Albert, 1983: Die Rolle der richterlichen Rechtsschöpfung im
Europäischen Gemeinschaftsrecht. In Gerhard Lüke, Georg Ress, and Michael

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

R. Will (eds), Rechtsvergleichung, Europarecht und Staatenintegration. Cologne:
Carl Heymanns Verlag, 61–81.
Börzel, Tanja A., Tobias Hofmann, and Diana Panke, 2012: Caving in or Sitting it out?
Longitudinal Patterns of Non-Compliance in the European Union. Journal of
European Public Policy, 16(4), 454–71.
Börzel, Tanja A., and Thomas Risse, 2003: Conceptualizing the Domestic Impact of
Europe. In Keith Featherstone and Claudio M. Radaelli (eds), The Politics of
Europeanisation. Oxford: Oxford University Press, 57–82.
Börzel, Tanja A., and Thomas Risse, 2007: Europeanization: The Domestic Impact of
European Union Politics. In Knud Erik Jørgensen, Mark A. Pollack, and Ben
Rosamond (eds), Handbook of European Union Politics. London: Sage Publications,
483–503.
Bovis, Christopher H., 2006: Developing Public Procurement Regulation: Jurispru-
dence and its Influence on Law Making. Common Market Law Review, 43(2),
461–95.
Bratton, William W., Joseph A. McCahery, and Erik P. M. Vermeulen, 2009: How does
Corporate Mobility Affect Lawmaking? A Comparative Analysis. American Journal
of Comparative Law, 57(2), 347–86.
Bücker, Andreas, and Wiebke Warneck, 2010: Viking—Laval—Rüffert: Consequences
and Policy Perspectives. Brussels: European Trade Union Institute.
Bulmer, Simon J., 1994: Institutions and Policy Change in the European Communities:
The Case of Merger Control. Public Administration, 72(3), 423–44.
Bundesministerium der Finanzen, 2006: Handbuch: Vorschriften für die Erbringung
von Dienst- oder Werkleistungen im Bereich der EU-Dienstleistungs- und Niederlas-
sungsfreiheit. Berlin: BMF. <http://www.budapest.diplo.de/contentblob/1703698/
Daten/74850/Handbuch_DownloadDatei.pdf>
Bundesministerium der Finanzen, 2008: Ergebnisse und Initiativen des Bundesminister-
iums der Finanzen (BMF) und der Finanzkontrolle Schwarzarbeit der Zollverwaltung
(FKS) zur Bekämpfung der Schwarzarbeit und der illegalen Beschäftigung 2007 ein-
schließlich eines Ausblicks auf das weitere Vorgehen. Berlin: BMF. <http://www.
olafscholz.hamburg/media/public/db/media/1/2010/12/191/bmffks2007anlage.pdf>
Bundesministerium der Finanzen, 2012a: Ergebnisse der Steuerschätzung vom 29. bis 31.
Oktober 2012, Monatsberichte. Berlin: BMF.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Bibliography 257
Bundesministerium der Finanzen, 2012b: Grünbuch der Deutsch-Französischen
Zusammenarbeit: Konvergenzpunkte bei der Unternehmensbesteuerung. Berlin: BMF.
Burley, Anne-Marie, and Walter Mattli, 1993: Europe Before the Court: A Political
Theory of Legal Integration. International Organization, 47(1), 41–76.
Callaghan, Helen, and Martin Höpner, 2005: European Integration and the Clash of
Capitalisms: Political Cleavages over Takeover Liberalization. Comparative European
Politics, 3(3), 307–32.
Capoccia, Giovanni, and R. Daniel Kelemen, 2007: The Study of Critical Junctures:
Theory, Narrative, and Counterfactuals in Historical Institutionalism. World Politics,
59(3), 341–69.
Caporaso, James A., and Sidney Tarrow, 2009: Polanyi in Brussels: Supranational
Institutions and the Transnational Embedding of Markets. International Organiza-
tion, 63(4), 593–620.

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Caranta, Roberto, 2008: On Discretion. In Sacha Prechal and Bert Van Roermund
(eds), The Coherence of EU Law: The Search for Unity in Divergent Concepts.
Oxford: Oxford University Press, 185–215.
Carrubba, Clifford, and Matthew Gabel, 2014: International Courts and the Perform-
ance of International Agreements: A General Theory with Evidence from the European
Union. Cambridge: Cambridge University Press.
Chalmers, Damian, and Mariana Chaves, 2012: The Reference Points of EU Judicial
Politics. Journal of European Public Policy, 19(1), 25–42.
Chalmers, Damian, Gareth Davies, and Giorgio Monti, 2010: European Union Law:
Cases and Materials. 2nd edn. Cambridge: Cambridge University Press.
Chalmers, Damian, and Giorgio Monti, 2008: European Union Law: Text and Mate-
rials: Updating Supplement. Cambridge: Cambridge University Press.
Cichowski, Rachel A., 2004: Women’s Rights, the European Court, and Supranational
Constitutionalism. Law and Society Review, 38(3), 489–512.
Cichowski, Rachel A., 2006: Courts, Rights, and Democratic Participation. Compara-
tive Political Studies, 39(1), 50–75.
Cichowski, Rachel A., 2007: The European Court and Civil Society: Litigation, Mobili-
zation and Governance. Cambridge: Cambridge University Press.
Cichowski, Rachel A., 2013: Mobilisation, Litigation and Democratic Governance.
Representation, 49(3), 321–32.
Collier, David, 1993: The Comparative Method. In W. Finifter Ada (ed.), Political
Science: The State of the Discipline II. Washington, DC: American Political Science
Association, 105–19.
Conant, Lisa, 2001: Europeanization and the Courts: Variable Patterns of Adaptation
among National Judiciaries. In Maria Green Cowles, James A. Caporaso, and Thomas
Risse (eds), Transforming Europe. London: Cornell University Press, 97–115.
Conant, Lisa, 2002: Justice Contained. Law and Politics in the European Union. Ithaca,
NY, and London: Cornell University Press.
Conant, Lisa, 2006: Individuals, Courts, and the Development of European Social
Rights. Comparative Political Studies, 39(1), 76–100.
Conant, Lisa, 2012: Compliance and What EU Member States Make of it. In Marise
Cremona (ed.), Compliance and the Enforcement of EU Law. Oxford: Oxford
University Press, 1–30.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

258 Bibliography
Connor, Tim, 2010: Goods, Persons, Services and Capital in the European Union:
Jurisprudential Routes to Free Movement. German Law Journal, 11(2), 159–209.
Conway, Gerard, 2012: The Limits of Legal Reasoning and the European Court of
Justice. Cambridge: Cambridge University Press.
Cordewener, Axel, Georg Kofler, and Servaas van Thiel, 2009: The Clash between
European Freedoms and National Direct Tax Law: Public Interest Defences Avail-
able to the Member States. Common Market Law Review, 46(6), 1951–2000.
Costello, Cathryn, 2009: Metock: Free Movement and ‘Normal Family Life’ in the
Union. Common Market Law Review, 46(2), 587–662.
Court of Justice of the European Union, 2012: Annual Report 2011. Luxembourg:
Publications Office of the European Union.
Court of Justice of the European Union, 2016: Annual Report 2015. Luxembourg:
Publications Office of the European Union.

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Craig, Paul, and Gráinne De Búrca, 2007: EU Law. Text, Cases, and Materials. 4th edn.
Oxford: Oxford University Press.
Damjanovic, Dragana, 2012: ‘Reserved Areas’ of the Member States and the ECJ: The
Case of Higher Education. In Hans-W. Micklitz and Bruno De Witte (eds), The
European Court of Justice and the Autonomy of the Member States. Cambridge:
Intersentia Publishing, 149–74.
Dammann, Jens C., 2007: Materielles Recht und Beweisrecht im System der Grundfrei-
heiten, Habilitationsschrift. Tübingen: Mohr Siebeck.
Davesne, Alban, 2009: The Laval Case and the Future of Labour Relations in Sweden,
Les Cahiers Européens de Sciences Po, 01/2009. Paris: Sciences Po.
David, Paul A., 1985: Clio and the Economics of QWERTY. American Economic
Review, 75(2), 332–7.
Davies, Bill, 2012: Resisting the European Court of Justice: West Germany’s Confron-
tation with European Law, 1949–1979. Cambridge: Cambridge University Press.
Davies, Gareth, 2003: Nationality Discrimination in the European Internal Market,
European Monographs. London and New York: Kluwer Law International.
Davies, Gareth, 2007a: The Services Directive: Extending the Country of Origin Principle
and Reforming Public Administration. European Law Review, 32(2), 232–45.
Davies, Gareth, 2007b: Services, Citizenship, and the Country of Origin Principle,
Mitchell Working Paper Series 2/2007. Edinburgh: Edinburgh Europa Institute.
Davies, Gareth, 2012: Activism Relocated: The Self-Restraint of the European Court of
Justice in its National Context. Journal of European Public Policy, 19(1), 76–91.
Davies, Gareth, 2014: Legislative Control of the European Court of Justice. Common
Market Law Review, 51(6), 1579–1608.
Davies, Gareth, 2016a: The European Union Legislature as an Agent of the European
Court of Justice. Journal of Common Market Studies, 54(1), 1–16.
Davies, Gareth, 2016b: Brexit and the Free Movement of Persons: A Plea for National
Legal Self-Assertiveness. European Law Review, 2016(6), 925–37.
de Goede, Jan, 2003: European Integration and Tax Law. European Taxation, 43(6), 203–9.
De Hart, Betty, Tineke Strik, and Henrike Pankratz, 2012: Family Reunification:
A Barrier or Facilitator of Integration? Country Report of the Netherlands. Brussels:
European Commission. <research.icmpd.org/fileadmin/Research-Website/Project_
material/Family_Reunification_EIF/Dutch.pdf> (accessed 15 May 2017).
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Bibliography 259
De Somer, Marie, 2012: Enhanced Competences for the European Court of Justice:
‘Re-shuffling’ the Dynamics of EU Migration Policy-Making?, Migration Studies Unit
Working Papers 12/01. London: London School of Economics and Political Science.
De Somer, Marie, 2016: Autonomy through Precedent: A Longitudinal Analysis of the
EU Court of Justice’s Case Law on Family Reunification Immigration. Maastricht
University, Ph.D. thesis.
De Witte, Bruno, 2007: Setting the Scene: How Did Services Get to Bolkestein, Mitchell
Working Paper Series 3/2007. Edinburgh: Edinburgh Europa Institute.
De Witte, Bruno, 2012: A Competence to Protect: The Pursuit of Non-Market Aims
through Internal Market Legislation. In Phil Syrpis (ed.), The Judiciary, the Legis-
lature and the EU Internal Market. Cambridge: Cambridge University Press, 25–48.
Dehousse, Renaud, 1998: The European Court of Justice. Houndmills: Macmillan.
Derlén, Mattias, and Johan Lindholm, 2013: Goodbye van Gend en Loos, Hello

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Bosman? Using Network Analysis to Measure the Importance of Individual CJEU
Judgments. European Law Journal, 20(5), 667–87.
Derlén, Mattias, et al., 2012: Coherence Out of Chaos: Mapping European Union Law
by Running Randomly through the Maze of CJEU Case Law. Europarättslig Tid-
skrift, 2013(3), 517–35.
Dobler, Philipp, 2008: Legitimation und Grenzen der Rechtsfortbildung durch den
EuGH. In Günter H. Roth and Peter Hilpold (eds), Der EuGH und die Souveränität
der Mitgliedsstaaten: Eine kritische Analyse richterlicher Rechtsschöpfung auf ausge-
wählten Rechtsgebieten. Vienna: Linde, 509–60.
Dobson, Alan, 2010: Civil Aviation and European Integration: Creating the Seemingly
Impossible SEAM. Journal of Common Market Studies, 48(4), 1127–47.
Döhler, Marian, 2001: Das Modell der unabhängigen Regulierungsbehörde im Kontext
des deutschen Regierungs- und Verwaltungssystems. Die Verwaltung, 34(1), 59–91.
Drew, Catherine, and Dhananjayan Sriskandarajah, 2007: EU Enlargement in 2007:
No Warm Welcome for Labor Migrants. Washington, DC: Migration Policy Institute.
Dyevre, Arthur, 2010: Unifying the Field of Comparative Judicial Politics: Towards a
General Theory of Judicial Behaviour. European Political Science Review, 2(2),
297–327.
ECAS, 2013: Mind the Gap: Towards a Better Enforcement of European Citizens’ Rights
of Free Movement. Brussels: European Citizen Action Service.
Eckstein, Harry, 1975: Case Study and Theory in Political Science. In Fred
I. Greenstein and Nelson W. Polsby (eds), Handbook of Political Science, vii.
Reading, MA: Addison-Wesley, 79–137.
Editorial Comments, 2006: The Services Directive Proposal: Striking a Balance between
the Promotion of the Internal Market and Preserving the European Social Model?
Common Market Law Review, 43(2), 307–11.
Eeckhout, Piet, 1998: The European Court of Justice and the Legislature. In Piet
Eeckhout and Takis Tridimas (eds), Yearbook of the European Law. Oxford: Oxford
University Press, 1–28.
Egan, Michelle P., 2015: Single Markets. Economic Integration in Europe and the
United States. 1st edn. New York: Oxford University Press.
Engsig Sorensen, Karsten, 2006: Abuse of Rights in Community Law: A Principle of
Substance or Merely Rhetoric? Common Market Law Review, 43(2), 423–59.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

260 Bibliography
European Commission, 2007: Handbook on the Implementation of the Services
Directive. Luxembourg: Office for Official Publications of the European Communities.
European Commission, 2008: Europe in Figures: Eurostat Yearbook 2008. Luxem-
bourg: Office for Official Publications of the European Communities.
European Trade Union Confederation, 2008: Viking and Laval Cases. Explanatory
Memorandum ( for Executive Committee of the ETUC, 4 March 2008). Brussels:
European Trade Union Confederation.
Everling, Ulrich, 2000: Richterliche Rechtsfortbildung in der EG. Juristenzeitung,
55(5), 217–27.
Everling, Ulrich, 2003: Die Europäische Union im Spannungsfeld von gemeinschaft-
licher und nationaler Politik und Rechtsordnung. In Armin von Bogdandy (ed.),
Europäisches Verfassungsrecht. Berlin: Springer, 847–92.
Everson, Michelle, and Julia Eisner, 2007: The Making of a European Constitution:

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Judges and Law Beyond Constitutive Power. Abingdon: Routledge-Cavendish.
Ewing, Keith D., 2012: The Draft Monti II Regulation: An Inadequate Response to
Viking and Laval. 1st edn. Liverpool: Institute of Employment Rights Briefing.
Fabbrini, Federico, and Katarzyna Granat, 2013: ‘Yellow Card, But No Foul’: The Role
of the National Parliaments under the Subsidiarity Protocol and the Commission
Proposal for an EU Regulation on the Right to Strike. Common Market Law Review,
50(1), 115–44.
Falkner, Gerda, 2011: In and Out of EU Decision Traps: Comparative Perspectives. In
Gerda Falkner (ed.) The EU’s Decision Traps. New York: Oxford University Press,
237–58.
Falkner, Gerda, Oliver Treib, and Elisabeth Holzleithner, 2008: Compliance in the
Enlarged European Union: Living Rights or Dead Letters? Farnham: Ashgate.
Farrell, Henry, and Adrienne Héritier, 2003: Formal and Informal Institutions under
Codecision: Continuous Constitution-Building in Europe. Governance: An Inter-
national Journal of Policy and Administration, 16(4), 577–600.
Fearon, James D, 1991: Counterfactuals and Hypothesis Testing in Political Science.
World Politics, 43(2), 169–95.
Fernhout, Roel, 2012: European Report on the Free Movement of Workers in Europe in
2011–2012: Follow-Up of Case Law of the Court of Justice. Nijmegen: Centre for
Migration Law.
Fernhout, Roel, and Ries Wever, 2011: Thematic Report 2010–2011: Follow-up of
the Case Law of the Court of Justice of the European Union. Brussels: European
Commission.
Føllesdal, Andreas, and Simon Hix, 2006: Why there is a Democratic Deficit in the EU:
A Response to Majone and Moravcsik. Journal of Common Market Studies, 44(3),
533–62.
Fon, Vincy, Francesco Parisi, and Ben Depoorter, 2005: Litigation, Judicial Path-
Dependence, and Legal Change. European Journal of Law and Economics, 20(1), 43–56.
Freitag, Robert, 1999: Der Wettbewerb der Rechtsordnungen im Internationalen
Gesellschaftsrecht. Europäische Zeitschrift für Wirtschaftsrecht, 10(9), 267–70.
Galanter, Marc, 1974: Why the ‘Haves’ Come Out Ahead: Speculations on the Limits
of Legal Change. Law and Society Review, 9(1), 95–160.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Bibliography 261
Garrett, Geoffrey, 1995: The Politics of Legal Integration in the European Union.
International Organization, 49(1), 171–81.
Garrett, Geoffrey, R. Daniel Kelemen, and Heiner Schulz, 1998: The European Court
of Justice, National Governance, and Legal Integration in the European Union.
International Organization, 52(1), 149–76.
Garrett, Geoffrey, and Barry R. Weingast, 1993: Ideas, Interests, and Institutions:
Constructing the European Community’s Internal Market. In Judith Goldstein
and Robert O. Keohane (eds), Ideas and Foreign Policy: Beliefs, Institutions and
Political Change. Ithaca, NY: Cornell University Press, 173–206.
Geddes, Barbara, 1990: How the Cases you Choose Affect the Answers you Get:
Selection Bias in Comparative Politics. Political Analysis, 2(1), 131–50.
Genschel, Philipp, 2002: Steuerwettbewerb und Steuerharmonisierung in der Euro-
päischen Union, Schriften des Max-Planck-Instituts für Gesellschaftsforschung Köln,

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

44. Frankfurt a.M.: Campus.
Genschel, Philipp, 2007: Why No Mutual Recognition of VAT? Regulation, Taxation
and the Integration of the EU’s Internal Market for Goods. Journal of European
Public Policy, 14(5), 743–61.
Genschel, Philipp, 2011: One Trap, Many Exits, But No Free Lunch: How the Joint-
Decision Trap Shapes EU Tax Policy. In Gerda Falkner (ed.), The EU’s Decision
Traps. New York: Oxford University Press, 54–72.
Genschel, Philipp, and Markus Jachtenfuchs, 2011: How the European Union Con-
strains the State: Multilevel Governance of Taxation. European Journal of Political
Research, 50(3), 293–314.
Genschel, Philipp, Achim Kemmerling, and Eric Seils, 2011: Accelerating Downhill:
How the EU Shapes Corporate Tax Competition in the Single Market. Journal of
Common Market Studies, 49(3), 585–606.
Genschel, Philipp, and Peter Schwarz, 2011: Tax Competition: A Literature Review.
Socio-Economic Review, 9(2), 339–70.
George, Alexander L., and Andrew Bennett, 2005: Process-Tracing and Historical
Explanation. In Alexander L. George and Andrew Bennett (eds), Case
Studies and Theory Development in the Social Sciences. Cambridge, MA: MIT
Press, 205–32.
Gerhardt, Michael J., 2005: The Limited Path of Dependency of Precedent. Journal of
Constitutional Law, 7(4), 903–1000.
Gerner-Beuerle, Carsten, and Michael Schilling, 2010: The Mysteries of Freedom of
Establishment After Cartesio. International and Comparative Law Quarterly, 59(2),
303–23.
Gerring, John, 2007: Case Study Research, Principles and Practices. Cambridge: Cambridge
University Press.
Gerring, John, 2012: Mere Description. British Journal of Political Science, 42(4),
721–46.
Giubboni, Stefano, 2007: Free Movement of Persons and European Solidarity. European
Law Journal, 13(3), 360–79.
Goebel, Roger J., 2002: The Liberalization of Interstate Legal Practice in the European
Union: Lessons for the United States? In Mads Andenas and Wulf-Henning Roth
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

262 Bibliography
(eds), Services and Free Movement in EU Law. Oxford: Oxford University Press,
413–58.
Goldsmith, Jack, and Eric A. Posner, 2006: The New International Law Scholarship.
Georgia Journal of International Law, 34(2), 463–84.
Goldstein, Judith, and Lisa L. Martin, 2000: Legalization, Trade Liberalization, and
Domestic Politics: A Cautionary Note. International Organization, 54(3), 603–32.
Golub, Jonathan, 1996: The Politics of Judicial Discretion: Rethinking the Interaction
between National Courts and the European Court of Justice. West European Politics,
19(2), 360–85.
Gormley, Laurence W., 2012: Free Movement of Goods and EU Legislation in the
Court of Justice. In Phil Syrpis (ed.), The Judiciary, the Legislature and the EU
Internal Market. Cambridge: Cambridge University Press, 49–61.
Graetz, Michael J., and Alvin C. Warren Jr, 2006: Income Tax Discrimination and the

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Political and Economic Integration of Europe. Yale Law Journal, 115(6), 1186–1255.
Graetz, Michael J., and Alvin C. Warren Jr, 2007: Dividend Taxation in Europe: When
the ECJ Makes Tax Policy. Common Market Law Review, 44(6), 1566–623.
Grande, Edgar, 1996: Das Paradox der Schwäche: Forschungspolitik und die Ein-
flußlogik europäischer Politikverflechtung. In Markus Jachtenfuchs and Beate
Kohler-Koch (eds), Europäische Integration. Opladen: Leske & Budrich, 373–99.
Greer, Scott L., 2006: Uninvited Europeanization: Neofunctionalism and the EU in
Health Policy. Journal of European Public Policy, 13(1), 134–52.
Greer, Scott L., 2012: Polity-Making without Policy-Making: European Union Health-
care Services Policy. In Jeremy Richardson (ed.), Constructing a Policy-Making
State? Policy Dynamics in the EU. Oxford: Oxford University Press, 271–91.
Greer, Scott L., 2013: Avoiding Another Directive: The Unstable Politics of European
Union Cross-Border Health Care Law. Health Economics, Policy and Law, 8(4),
415–21.
Grimm, Dieter, 2015: The Democratic Costs of Constitutionalisation: The European
Case. European Law Journal, 21(4), 460–73.
Groß, Helene, 2005: Das Gesetz über die allgemeine Freizügigkeit von Unionsbürgern.
Zeitschrift für Ausländerrecht und Ausländerpolitik, 25(3/4), 81–6.
Grossman, Andrew, 2004: Birthright Citizenship as Nationality of Convenience. In
Proceedings of the Third Conference on Nationality. Strasbourg: Council of Europe,
109–21.
Grossman, Emiliano, and Cornelia Woll, 2011: The French Debate over the Bolkestein
Directive. Comparative European Politics, 9(3), 344–66.
Hailbronner, Kay, 2004: Die Unionsbürgerschaft und das Ende rationaler Jurispru-
denz durch den EuGH? Neue Juristische Wochenschrift, 31, 2185–9.
Hall, Peter A., 2003: Aligning Ontology and Methodology on Comparative Research.
In James Mahoney and Dietrich Rueschemeyer (eds), Comparative Historical Anal-
ysis in the Social Sciences. Cambridge: Cambridge University Press, 373–404.
Haltern, Ulrich, 2005: Das Janusgesicht der Unionsbürgerschaft. Swiss Political Science
Review, 11(1), 87–117.
Hammond, Thomas H, 1996: Formal Theory and the Institutions of Governance.
Governance, 9(2), 107–85.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Bibliography 263
Hancher, Leigh, and Wolf Sauter, 2010: One Step Beyond? From Sodemare to Doc-
Morris: The EU’s Freedom of Establishment Case Law Concerning Healthcare.
Common Market Law Review, 47(1), 117–46.
Handoll, John, 2012: Ireland: Country Report, EUDO Citizenship Observatory. Florence:
EUI. <http://eudo-citizenship.eu/docs/CountryReports/Ireland.pdf> (accessed
15 June 2016).
Hartlapp, Miriam, Julia Metz, and Christian Rauh, 2014: Which Policy for Europe?
Power and Conflict Inside the European Commission. Oxford: Oxford University
Press.
Hartmann, Lucas, 2016: Das Konzept der Pfadabhängigkeit—Bedingungen und Grenzen
seiner Rezeption durch die Wissenschaften vom Öffentlichen Recht. In Pfadabhän-
gigkeit hoheitlicher Ordnungsmodelle. Mainz: Nomos Verlagsgesellschaft, 71–94.
Hassel, Anke, Jette S. Knudsen, and Bettina Wagner, 2016: Winning the Battle or

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Losing the War: The Impact of European Integration on Labour Market Institutions
in Germany and Denmark. Journal of European Public Policy, 23(8), 1218–39.
Hassemer, Winfried, 2008: Politik aus Karlsruhe? Juristenzeitung, 63(1), 1–10.
Hathaway, Oona A., 2003: Path Dependence in the Law: The Course and Pattern of
Legal Change in a Common Law System, Public Law and Legal Theory Research
Paper 270. New Haven: Yale Law School.
Hatzopoulos, Vassilis, 2000: Recent Developments of the Case Law of the ECJ in the
Field of Services. Common Market Law Review, 37(1), 43–82.
Hatzopoulos, Vassilis, 2002: Killing National Health and Insurance Systems But
Healing Patients? The European Market for Health Care Services After the Judg-
ments of the ECJ in Vanbraekel and Peerbooms. Common Market Law Review,
39(4), 683–729.
Hatzopoulos, Vassilis, 2012: Regulating Services in the EU. Oxford: Oxford University
Press.
Hatzopoulos, Vassilis, 2013: The Court’s Approach to Services (2006–2012): From
Case Law to Case Load? Common Market Law Review, 50(2), 459–502.
Hatzopoulos, Vassilis, and Thien Uyen Do, 2006: The Case Law of the ECJ Concerning
the Free Provision of Services. Common Market Law Review, 43(4), 923–91.
Hatzopoulos, Vassilis, and Tamara Hervey, 2013: Coming into Line: The EU’s Court
Softens on Cross-Border Health Care. Health Economics, Policy and Law, 8(1), 1–5.
Heller, Hans, 2012: EuGH v. 12.7.2012, C-171/11 ‘Fra.bo SpA/Deutsche Vereinigung des
Gas- und Wasserfaches e. V.(DVGW)’—Anwendbarkeit der Warenverkehrsfreiheit
auf private Normungs- und Zertifizierungstätigkeiten. EWeRK, 12(5), 179–82.
Héritier, Adrienne, 1995: ‘Leaders’ and ‘Laggards’ in European Clean Air Policy. In
Frans van Waarden and Benjamin Unger (eds), Convergence or Diversity? Inter-
nationalization and Economic Policy Response. Avebury: Aldershot, 278–305.
Héritier, Adrienne, and Yannis Karagiannis, 2011: The New Institutions of Transat-
lantic Aviation. Global Policy, 2(2), 152–62.
Héritier, Adrienne, et al., 2001: Differential Europe. Lanham, MD: Rowman & Little-
field Publishers.
Herlinghaus, Andreas, et al., 2010: Vorwort. In Andreas Herlinghaus et al. (eds),
Festschrift für Wienand Meilicke. Baden-Baden: Nomos Verlagsgesellschaft, 7–8.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

264 Bibliography
Hilpold, Peter, 2008: Unionsbürgerschaft und Bildungsrechte oder: Der EuGH-Richter
als ‘Künstler’. In Günter H. Roth and Peter Hilpold (eds), Der EuGH und die
Souveränität der Mitgliedsstaaten: Eine kritische Analyse richterlicher Rechtsschöp-
fung auf ausgewählten Rechtsgebieten. Vienna: Linde, 11–54.
Hirschl, Ran, 2008: The Judicialization of Politics. In Keith E. Whittington, R. Daniel
Kelemen, and Gregory A. Caldeira (eds), The Oxford Handbook of Law and Politics.
Oxford: Oxford University Press, 119–42.
Hirschman, Albert O., 1970: Exit, Voice, and Loyalty: Responses to Decline in Firms,
Organizations, and States. Cambridge, MA: Harvard University Press.
Hix, Simon, 1994: The Study of the European Community: The Challenge To Com-
parative Politics. West European Politics, 17(1), 1–30.
Hix, Simon, Abdul Noury, and Gerard Roland, 2007: Democratic Politics in the
European Parliament. Cambridge: Cambridge University Press.

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Hofmann, Andreas, 2008: Litigation Strategy and Legal Dynamics: The European
Commission Before the Court of Justice. Paper presented at the Fourth Pan-
European Conference on EU Politics, Riga, 25–27 September Riga.
Hofmann, Andreas, 2013: Strategies of the Repeat Player: The European Commission
between Courtroom and Legislature. Berlin: epubli GmbH.
Hönnige, Christoph, and Thomas Gschwend, 2010: Das Bundesverfassungsgericht im
politischen System der BRD—ein unbekanntes Wesen? Politische Vierteljahres-
schrift, 51(3), 507–30.
Honohan, Iseult, 2010: The Theory and Politics of Ius Soli. <http://eudo-citizenship.eu/
docs/IusSoli.pdf> (accessed 15 June 2016).
Hooghe, Liesbet, and Gary Marks, 2009: A Postfunctionalist Theory of European
Integration: From Permissive Consensus to Constraining Dissensus. British Journal
of Political Science, 39(1), 1–23.
Hoor, Gerd, 1999: Das Centros-Urteil des EuGH und seine Auswirkungen auf die
Anknüpfung des Gesellschaftsstatuts. Neue Zeitschrift für Gesellschaftsrecht, 20(2),
984–6.
Höpner, Martin, 2010: Warum betreibt der Europäische Gerichtshof Rechtsfortbil-
dung? Die Politisierungshypothese. Sozialer Fortschritt, 59(5), 141–51.
Höpner, Martin, 2011: Der Europäische Gerichtshof als Motor der Integration: Eine
akteursbezogene Erklärung. Berliner Journal für Soziologie, 21(2), 203–29.
Höpner, Martin, 2016: Das Soziale Fortschrittsprotokoll des Europäischen
Gewerkschaftsbunds: Ein Vorschlag zur Weiterentwicklung. WSI-Mitteilungen,
37(4), 245–53.
Höpner, Martin, and Armin Schäfer, 2010: A New Phase of European Integration:
Organized Capitalisms in Post-Ricardian Europe. West European Politics, 33(2),
344–68.
Höpner, Martin, and Armin Schäfer, 2012: Embeddedness and Regional Integration:
Waiting for Polanyi in a Hayekian Setting. International Organization, 66(3),
429–55.
Hörl, Björn, Andreas Warntjen, and Arndt Wonka, 2005: Built on Quicksand?
A Decade of Procedural Spatial Models on EU Legislative Decision-Making. Journal
of European Public Policy, 12(3), 592–606.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Bibliography 265
Horn, Norbert, 2004: Deutsches und europäisches Gesellschaftsrecht und die EuGH-
Rechtsprechung zur Niederlassungsfreiheit—Inspire Art. Neue Juristische Wochen-
schrift, 57(13), 893–901.
Hoskyns, Catherine, 1996: Integrating Gender: Women, Law and Politics in the
European Union. London and New York: Verso.
Howarth, David, 2007: Internal Policies: Reinforcing the New Lisbon Message of Com-
petitiveness and Innovation. Journal of Common Market Studies, 45(1), 89–106.
Howse, Robert, and Elisabeth Tuerk, 2001: The WTO Impact on Internal Regulations:
A Case Study of the Canada–EC Asbestos Dispute. In Gráinne de Búrca and Joanne
Scott (eds), The EU and the WTO. Legal and Constitutional Issues. Oxford: Hart,
283–328.
Hubschmid, Claudia, and Peter Moser, 1997: The Co-operation Procedure in the EU:
Why was the European Parliament Influential in the Decision on Car Emission

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Standards? Journal of Common Market Studies, 35(2), 225–42.
Jachtenfuchs, Markus, 1995: Theoretical Perspectives on European Governance. European
Law Journal, 1(2), 115–33.
Jacob, Marc, 2012: Precedents: Lawmaking through International Adjudication. German
Law Journal, 12(5), 35–68.
Jacqueson, Catherine, 2014: Denmark. In Ulla Neergaard, Catherine Jacqueson, and
Nina Holst-Christensen (eds), Union Citizenship: Development, Impact and Chal-
lenges. The XXVI FIDE Congress in Copenhagen, 2014. Congress Publications, ii.
Copenhagen: DJØF Publishing, 453–79.
Jessurun d’Oliveira, H. U., 2011: Decoupling Nationality and Union Citizenship?
European Constitutional Law Review, 7(1), 138–60.
Jochum, Heike, 2006: Zur Fortsetzung der Rechtssache Marks & Spencer: Gestattet das
europäische Gemeinschaftsrecht eine ‘geltungserhaltende Reduktion’ des nationa-
len Steuerrechts? Internationales Steuerrecht, 15(18), 621–3.
Joerges, Christian, 1996: Taking the Law Seriously: On Political Science and the
Role of Law in the Process of European Integration. European Law Journal, 2(2),
105–35.
Joerges, Christian, 2007: Reconceptualizing the Supremacy of European Law: A Plea
for a Supranational Conflict of Laws. In Beate Kohler-Koch and Berthold Rittberger
(eds), Debating the Democratic Legitimacy of the European Union. Lanham, MD:
Rowman & Littlefield Publishers, 311–28.
Joerges, Christian, and Florian Rödl, 2008: Das soziale Defizit des Europäischen
Integrationsprojekts. Kritische Justiz, 41(2), 149–65.
Joerges, Christian, and Florian Rödl, 2009: Informal Politics, Formalised Law and the
‘Social Deficit’ of European Integration: Reflections After the Judgments of the ECJ
in Viking and Laval. European Law Journal, 15(1), 1–19.
Kadelbach, Stefan, 2003: Unionsbürgerschaft. In Armin von Bogdandy (ed.), Europäisches
Verfassungsrecht: Theoretische und dogmatische Grundzüge. Berlin and Heidelberg:
Springer-Verlag, 539–82.
Kassim, Hussein, and Arnand Menon, 2003: The Principal-Agent Approach and the
Study of the European Union: Promise Unfulfilled? Journal of European Public
Policy, 10(1), 121–39.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

266 Bibliography
Kassim, Hussein, and Handley Stevens, 2010: Air Transport and the European Union:
Europeanization and its Limits. Basingstoke: Palgrave Macmillan.
Keeling, David, 1998: In Praise of Judicial Activism: But What does it Mean? And has
the European Court of Justice Ever Practised it? In Seminario Giuridico della
Università di Bologna (ed.), Scritti in onore di Giuseppe Federico Mancini. Milan:
Dott. A. Giuffrè Editore, 505–36.
Kelemen, R. Daniel, 2001: The Limits of Judicial Power: Trade–Environment Disputes
in the GATT/WTO and the EU. Comparative Political Studies, 34(6), 622–50.
Kelemen, R. Daniel, 2006: Suing for Europe. Adversarial Legalism and European
Governance. Comparative Political Studies, 39(1), 101–27.
Kelemen, R. Daniel, 2011: Eurolegalism. The Transformation of Law and Regulation in
the European Union. Cambridge, MA: Harvard University Press.
Kelemen, R. Daniel, 2012: The Political Foundations of Judicial Independence in the

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

European Union. Journal of European Public Policy, 19(1), 43–58.
Kelemen, R. Daniel, 2016: The Court of Justice of the European Union in the Twenty-
First Century. Law and Contemporary Problems, 79(1), 117–40.
Kelemen, R. Daniel, and Mitchell A. Orenstein, 2016: Europe’s Autocracy Prob-
lem. Foreign Affairs (7 Jan.). <https://www.foreignaffairs.com/articles/poland/
2016-01-07/europes-autocracy-problem> (accessed 27 June 2016).
Kemmeren, Eric, 2014: Where is EU Law in the OECD BEPS Discussion? EC Tax
Review, 23(4), 190–3.
Kennedy, Steven, 2011a: EEA Nationals: The ‘Right to Reside’ Requirement for Benefits.
London: House of Commons Briefing Paper SN05972.
Kennedy, Steven, 2011b: The Habitual Residence Test. London: House of Commons
Briefing Paper SN00416.
Kennedy, Steven, 2015: People from Abroad: What Benefits Can they Claim? London:
House of Commons Library Briefing Paper 06847.
Kilpatrick, Claire, 2012: Internal Market Architecture and the Accommodation
of Labour Rights: As Good as it Gets? In Phil Syrpis (ed.), The Judiciary, the
Legislature and the EU Internal Market. Cambridge: Cambridge University Press,
205–42.
King, Gary, Robert O. Keohane, and Sidney Verba, 2004: The Importance of Research
Design. In Henry E. Brady and David Collier (eds), Rethinking Social Inquiry.
Lanham, MD: Rowman & Littlefield, 181–92.
Kjellgren, Anders, 2002: On the Border of Abuse: The Jurisprudence of the European
Court of Justice on Circumvention, Fraud and Abuses of Community Law. In Mads
Andenas and Wulf-Henning Roth (eds), Services and Free Movement in EU Law.
Oxford: Oxford University Press, 245–77.
Kloka, Marzena, 2013: Business as Usual? Negotiation Dynamics and Legislative
Performance in the Council of the European Union After the Eastern Enlargement
Bremen International Graduate School of Social Sciences, Ph.D. thesis.
Kloka, Marzena, and Susanne K. Schmidt, 2015: Legislative and Judicial Politics in the
Post-Maastricht Era. In Christopher J. Bickerton, Dermot Hodson, and Uwe Puetter
(eds), The New Intergovernmentalism: States and Supranational Actors in the Post-
Maastricht Era. Oxford: Oxford University Press, 227–44.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Bibliography 267
Knight, Jack, and Lee Epstein, 1996: The Norm of Stare Decisis. American Journal of
Political Science, 40(4), 1018–35.
Knill, Christoph, and Florian Becker, 2003: Divergenz trotz Diffusion? Rechtsver-
gleichende Aspekte des Verhältnismäßigkeitsprinzips in Deutschland, Großbritannien
und der Europäischen Union. Bonn: Max-Planck-Projektgruppe Recht der
Gemeinschaftsgüter.
Knill, Christoph, and Dirk Lehmkuhl, 1999: How Europe Matters: Different Mechan-
isms of Europeanization. European Integration Online Papers, 3(7), 1–19.
Knill, Christoph, and Andrea Lenschow, 1998: Coping with Europe: The Impact of
British and German Administrations on the Implementation of EU Environmental
Policy. Journal of European Public Policy, 5(4), 595–614.
Knill, Christoph, Jale Tosun, and Michael W. Bauer, 2009: Neglected Faces of
Europeanization: The Differential Impact of the EU on the Dismantling and

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Expansion of Domestic Policies. Public Administration, 87(3), 519–37.
Kochenov, Dimitry, 2013: The Right to Have What Rights? EU Citizenship in Need of
Clarification. European Law Journal, 19(4), 502–16.
König, Thomas, 2007: Divergence or Convergence? From Ever-Growing to Ever-
Slowing: European Legislative Decision Making. European Journal of Political
Research, 46(3), 417–44.
Kornezov, Alexander, 2014: Bulgaria. In Ulla Neergaard, Catherine Jacqueson, and
Nina Holst-Christensen (eds), Union Citizenship: Development, Impact and Chal-
lenges. The XXVI FIDE Congress in Copenhagen, 2014. Congress Publications, ii.
Copenhagen: DJØF Publishing, 339–61.
Krajewski, Markus, 2010: Grenzüberschreitende Patientenmobilität in Europa
zwischen negativer und positiver Integration der Gesundheitssysteme. Europarecht,
45(2), 165–87.
Kramer, Dion, 2016: Earning Social Citizenship in the European Union: Free Move-
ment and Access to Social Assistance Benefits Reconstructed. Cambridge Yearbook
of European Legal Studies, 18, 270–301.
Kranenpohl, Uwe, 2009: Die Bedeutung von Interpretationsmethoden und Dogmatik
in der Entscheidungspraxis des Bundesverfassungsgerichts. Der Staat, 48(3),
387–409.
Kriesi, Hanspeter, 2009: Notes and Comments: Rejoinder to Liesbet Hooghe and Gary
Marks, ‘A Postfunctional Theory of European Integration: From Permissive Con-
sensus to Constraining Dissensus’. British Journal of Political Science, 39(1), 221–4.
Kubicki, Philipp, and Magdalena Obajtek, 2012: Die DVGW-Entscheidung. EuGH,
Rs. C-171/11(Fra.bo SpA ./. DVGW), Urteil des Gerichtshofs vom 12. Juli 2012.
DeLuxe, 27(4), 1–7.
Lang, Michael, 2006: The Marks & Spencer Case: The Open Issues Following the ECJ’s
Final Word. European Taxation, 46(2), 54–67.
Lang, Michael, 2009: Recent Case Law of the ECJ in Direct Taxation: Trends, Tensions,
and Contradictions. EC Tax Review, 18(3), 98–113.
Larsson, Olof, and Daniel Naurin, 2016: Judicial Independence and Political Uncer-
tainty: How the Risk of Override Impacts on the Court of Justice of the EU.
International Organization, 70(2), 1–32.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

268 Bibliography
Larsson, Olof, et al., 2016: Speaking Law to Power: The Strategic Use of Precedent of
the Court of Justice of the European Union. Comparative Political Studies, 50(7),
879–907.
Leiren, Merethe D., and Louisa Parks, 2014: When Trade Unions Succeed: Cases
of Blocked Liberalisation in the Common Market. European Journal of Political
Research, 53(3), 1–15.
Lenaerts, Koen, and Tim Corthaut, 2008: Towards an Internally Consistent Doctrine
on Invoking Norms of EU Law. In Sacha Prechal and Bert Van Roermund (eds),
The Coherence of EU Law: The Search for Unity in Divergent Concepts. Oxford:
Oxford University Press, 495–515.
Lenaerts, Koen, et al., 2006: Procedural Law of the European Union. London: Sweet &
Maxwell.
Levy, Jack S., 2010: Counterfactuals and Case Studies. In Janet M. Box-Steffensmeier,

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Henry E. Brady, and David Collier (eds), The Oxford Handbook of Political Meth-
odology. Oxford and New York: Oxford University Press, 627–44.
Liebowitz, S. J., and Stephen E. Margolis, 1990: The Fable of the Keys. Journal of Law
and Economics, 33 (Apr.), 1–25.
Lijphart, Arend, 1979: Comparative Politics and the Comparative Method. American
Political Science Review, 65(3), 682–93.
Lindberg, Björn, 2008: Are Political Parties Controlling Legislative Decision-Making
in the European Parliament? The Case of the Services Directive. Journal of European
Public Policy, 15(8), 1184–1204.
Lindstrom, Nicole, 2010: Service Liberalization in the Enlarged EU: A Race to the
Bottom or the Emergence of Transnational Political Conflict? Journal of Common
Market Studies, 48(5), 1307–27.
Loder, Jeff, 2011: The Lisbon Strategy and the Politicization of EU Policy-Making: The
Case of the Services Directive. Journal of European Public Policy, 18(4), 566–83.
Lubow, Alexis, 2017: Taming Regulatory Competition: Interest Groups v. Joint-Decision
Trap. Four EU Policy Cases on Workers Mobility. European University Institute,
Ph.D. thesis.
McCall Smith, James, 2000: The Politics of Dispute Settlement Design: Explaining
Legalism in Regional Trade Pacts. International Organization, 54(1), 137–80.
McCown, Margaret, 2003: The European Parliament Before the Bench: ECJ Precedent
and EP Litigation Strategies. Journal of European Public Policy, 10(6), 974–95.
Maduro, Miguel P., 1998: We the Court: The European Court of Justice and the
European Economic Constitution. A Critical Reading of Article 30 of the EU Treaty.
Oxford: Hart.
Maduro, Miguel P., 2007: So Close and Yet So Far: The Paradoxes of Mutual Recog-
nition. Journal of European Public Policy, 14(5), 814–25.
Mahoney, James, 2000: Path Dependence in Historical Sociology. Theory and Society,
29(4), 507–48.
Majone, Giandomenico, 2014: Rethinking the Union of Europe Post-Crisis: Has Inte-
gration Gone Too Far? Cambridge: Cambridge University Press.
Malecki, Michael, 2012: Do ECJ Judges All Speak with the Same Voice? Evidence of
Divergent Preferences from the Judgments of Chambers. Journal of European
Public Policy, 19(1), 59–75.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Bibliography 269
Malherbe, Jacques, et al., 2011: The Impact of the Rulings of the European Court of
Justice in the Area of Direct Taxation 2010, Directorate-General for Internal Policies—
Policy Department A: Economic and Scientific Policy. Strassburg: European
Parliament.
Malmberg, Jonas, and Tore Sigeman, 2008: Industrial Actions and EU Economic
Freedoms: The Autonomous Collective Bargaining Model Curtailed by the European
Court of Justice. Common Market Law Review, 45(4), 1115–46.
Mancini, G. Federico, 1989: The Making of a Constitution for Europe. Common
Market Law Review, 26(4), 595–614.
Mancini, Joanne M., and Graham Finlay, 2008: Citizenship Matters: Lessons from the
Irish Citizenship Referendum. American Quarterly, 60(3), 575–99.
Mankowski, Peter, 2004: Wider ein Herkunftslandprinzip für Dienstleistungen im
Binnenmarkt. Praxis des Internationalen Privat- und Verfahrensrechts, 24(5),

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

385–95.
Marks, Gary, Liesbet Hooghe, and Kermit Blank, 1996: European Integration from the
1980s: State-Centric v. Multi-Level Governance. Journal of Common Market Stud-
ies, 34(3), 341–78.
Martinsen, Dorte S., 2005: Towards an Internal Health Market with the European
Court. West European Politics, 28(5), 1035–56.
Martinsen, Dorte S., 2011: Judicial Policy-Making and Europeanization: The Propor-
tionality of National Control and Administrative Discretion. Journal of European
Public Policy, 18(7), 944–61.
Martinsen, Dorte S., 2015: An Ever More Powerful Court? The Political Constraints of
Legal Integration in the European Union. Oxford: Oxford University Press.
Martinsen, Dorte S., and Gerda Falkner, 2011: Social Policy: Problem-Solving Gaps,
Partial Exits, and Court-Decision Traps. In Gerda Falkner (ed.), The EU’s Decision
Traps. New York: Oxford University Press, 128–44.
Mattli, Walter, and Anne-Marie Slaughter, 1995: Law and Politics in the European
Union: A Reply to Garrett. International Organization, 49(1), 183–90.
Mattli, Walter, and Anne-Marie Slaughter, 1998: Revisiting the European Court of
Justice. International Organization, 52(1), 177–209.
Maurer, Andreas, 2003: The Legislative Powers and Impact of the European Parlia-
ment. Journal of Common Market Studies, 41(2), 227–47.
Mayntz, Renate, 2004: Mechanisms in the Analysis of Social Macro-Phenomena.
Philosophy of the Social Sciences, 34(2), 237–59.
Mayoral, Juan A., 2016: In the CJEU Judges Trust: A New Approach in the Judicial
Construction of Europe. Journal of Common Market Studies, 55(3), 417–671.
Meduna, Michal, et al., 2014: Institutional Report. In Ulla Neergaard, Catherine
Jacqueson, and Nina Holst-Christensen (eds), Union Citizenship: Development,
Impact and Challenges. The XXVI FIDE Congress in Copenhagen, 2014. Congress
Publications, ii. Copenhagen: DJØF Publishing, 227–316.
Meilicke, Wienand, 2007: Nachlese zum EuGH-Urteil vom 6.3.2007—RS. C-292/04,
Meilicke. Der Betrieb, 19(12), 650–1.
Meilicke, Wienand, 2009: Vertragsverletzungsverfahren wegen Diskriminierung
von nach ausländischem Recht errichteten Organgesellschaften. Der Betrieb,
21(13), 653.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

270 Bibliography
Meilicke, Wienand, 2015: Nach dem BFH-Schlussurteil ‘Meilicke’: Was tun? Der
Betrieb, 68(45), 2601–2.
Menéndez, Agustín José, 2009: European Citizenship After Martínez Sala and Baum-
bast. Has European Law Become More Human But Less Social?, RECON Online
Working Paper Series 2009/05. Oslo: Arena Centre for European Studies University
of Oslo.
Miersch, Gerald, 1996: Versicherungsaufsicht nach den Dritten Richtlinien. Aufsichtsbe-
fugnisse und Inländerbenachteiligungen. Karlsruhe: Verlag Versicherungswirtschaft.
Miklin, Eric, 2009: Government Positions on the EU Services Directive in the Council:
National Interests or Individual Ideological Preferences? West European Politics,
32(5), 943–62.
Möllers, Christoph, 2015: Krisenzurechnung und Legitimationsproblematik in der
Europäischen Union. Leviathan, 43(3), 339–64.

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Moravcsik, Andrew, 1993: Preferences and Power in the European Community:
A Liberal Intergovernmentalist Approach. Journal of Common Market Studies, 31(4),
473–524.
Moravcsik, Andrew, 1995: Liberal Intergovernmentalism and Integration:
A Rejoinder. Journal of Common Market Studies, 33(4), 611–28.
Moravcsik, Andrew, 2002: In Defence of the ‘Democratic Deficit’: Reassessing Legit-
imacy in the European Union. Journal of Common Market Studies, 40(4), 603–24.
Murphy, Ryan, 2012: Changing Treaty and Changing Economic Context: The
Dynamic Relationship of the Legislature and the Judiciary in the Pursuit of Capital
Liberalisation. In Phil Syrpis (ed.) The Judiciary, the Legislature and the EU Internal
Market. Cambridge: Cambridge University Press, 274–301.
Neyer, Jürgen, 2012: The Justification of Europe: A Political Theory of Supranational
Integration. Oxford: Oxford University Press.
Nicolaïdis, Kalypso, and Susanne K. Schmidt, 2007: Mutual Recognition ‘on Trial’: The
Long Road to Services Liberalization. Journal of European Public Policy, 14(5),
717–34.
Nicolaïdis, Kalypso, and Gregory Shaffer, 2005: Transnational Mutual Recognition
Regimes: Governance without Global Government. Law and Contemporary Prob-
lems, 68(3), 267–322.
Nyikos, Stacy A., 2006: Strategic Interaction among Courts within the Preliminary
Reference Process—Stage 1: National Court Preemptive Opinions. European Jour-
nal of Political Research, 45(4), 527–50.
O’Brien, Charlotte, 2015: The Pillory, the Precipice and the Slippery Slope: The
Profound Effects of the UK’s Legal Reform Programme Targeting EU Migrants.
Journal of Social Welfare and Family Law, 37(1), 111–36.
O’Shea, Tom, 2008: Freedom of Establishment Tax Jurisprudence: Avoir Fiscal Re-
visited. EC Tax Review, 17(6), 259–75.
Obermaier, Andreas J., 2008a: Fine-Tuning the Jurisprudence: The ECJ’s Judicial
Activism, Working Paper 02/2008. Vienna: Institute for European Integration
Research.
Obermaier, Andreas J., 2008b: The National Judiciary—Sword of European Court of
Justice Rulings: The Example of the Kohll/Decker Jurisprudence. European Law
Journal, 14(6), 735–52.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Bibliography 271
Obermaier, Andreas J., 2009: The End of Territoriality? The Impact of ECJ Rulings on
British, German and French Social Policy. Farnham: Ashgate Publishing.
OECD, 2015: BEPS Action 3: Strengthening CFC Rules. Paris: OECD.
Oliver, Peter, 1999: Some Further Reflections on the Scope of Articles 28–30
(Ex 30–36) EC. Common Market Law Review, 36(4), 783–806.
Oliver, Peter, and Stefan Enchelmaier, 2007: Free Movement of Goods: Recent Devel-
opments in the Case Law. Common Market Law Review, 44(3), 649–704.
Oliver, Peter, and Wulf-Henning Roth, 2004: The Internal Market and the Four
Freedoms. Common Market Law Review, 41(2), 407–41.
Olsen, Johan P., 2013: The Institutional Basis of Democratic Accountability. West
European Politics, 36(3), 447–73.
Ostrom, Elinor, 1986: An Agenda for the Study of Institutions. Public Choice, 48(1), 3–35.
Pantazatou, Katerina, 2013: Economic and Political Considerations of the Court’s Case

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Law Post Crisis: An Example from Tax Law and the Internal Market. Croatian
Yearbook of European Law and Policy, 9, 77–118.
Pelkmans, Jacques, 2007: Mutual Recognition in Goods: On Promises and Disillusions.
Journal of European Public Policy, 14(5), 699–716.
Perinetto, Patrick A., 2012: Viking and Laval: An Italian Perspective—a Case of No
Impact. European Labour Law Journal, 3(4), 270–99.
Pescatore, Pierre, 1983: La carence du législateur communautaire et le devoir du juge.
In Gerhard Lüke, Georg Ress, and Michael R. Will (eds), Rechtsvergleichung—
Gedächtnisschrift für Léontin-Jean Constantinesco. Cologne: Heymanns, 559–80.
Pfennigstorf, Werner, 1987: Versicherungsaufsicht ohne Niederlassung—europäischer
Auftrag und amerikanische Erfahrungen. Versicherungsrecht, 13, 331–6.
Phelan, William, 2011: Why do the EU Member States Accept the Supremacy of
European Law? Explaining Supremacy as an Alternative to Bilateral Reciprocity.
Journal of European Public Policy, 18(5), 766–77.
Pierson, Paul, 2000: Increasing Returns, Path Dependence, and the Study of Politics.
American Political Science Review, 94(2), 251–67.
Pistone, Pasquale, 2008: European Direct Tax Law: Quo Vadis? In Luc Hinnekens and
Philippe Hinnekens (eds), A Vision of Taxes within and outside European Borders.
Alphen aan den Rijn: Kluwer Law International, 713–28.
Pollack, Mark A., 1996: The New Institutionalism and EC Governance: The Promise
and Limits of Institutional Analysis. Governance, 9(4), 429–85.
Pollack, Mark A., 2013: The New EU Legal History: What’s New, What’s Missing?
American University International Law Review, 28(5), 1257–1310.
Pop, Valentina, 2008: Former German President Bashes EU Court, Euobserver.
<https://euobserver.com/news/26712> (accessed 20 June 2016).
Posner, Eric A., and Alan Sykes, 2011: Efficient Breach of International Law: Optimal
Remedies, ‘Legalized Noncompliance’, and Related Issues. Michigan Law Review,
110(2), 243–94.
Posner, Eric A., and John C. Yoo, 2005: A Theory of International Adjudication.
California Law Review, 93(1), 1–72.
Radaelli, Claudio, 2003: The Europeanization of Public Policy. In Kevin Featherstone
and Claudio Radaelli (eds), The Politics of Europeanization. Oxford: Oxford University
Press, 27–56.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

272 Bibliography
Rainer, Anno, 2005: EC Tax Scene—ECJ Hears Case on Cross-Border Tax Credits.
Intertax, 33(11), 553–4.
Rainer, Anno, 2006: EC Tax Scene: Are Finnish Cross-Border Group Contributions
Tax-Deductible for the Contributor? Intertax, 34(12), 639–41.
Randelzhofer, Albrecht, and Ernst Forsthoff, 2001: Art. 49/50 EGV: Dienstleistungen.
In Eberhard Grabitz and Meinhard Hilf (eds), Das Recht der EU: Kommentar.
Munich: C. H. Beck.
Rasmussen, Hjalte, 1986: On Law and Policy in the European Court of Justice.
Dordrecht: Martinus Nijhoff Publishers.
Rasmussen, Morten, 2012: Establishing a Constitutional Practice of European Law:
The History of the Legal Service of the European Executive, 1952–65. Contemporary
European History, 21(3), 375–97.
Rauh, Christian, and Gerald Schneider, 2013: There is No Such Thing as a Free Open

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Sky: Financial Markets and the Struggle over European Competences in Inter-
national Air Transport. Journal of Common Market Studies, 51(6), 1124–40.
Rawlings, Richard, 1993: The Eurolaw Game: Some Deductions from a Saga. Journal of
Law and Society, 20(3), 309–40.
Refslund, Bjarke, 2015: Adjusting the Danish Industrial Relations System After Laval:
Recalibration rather than Erosion. European Review of Labour and Research, 21(2),
247–51.
Rehder, Britta, 2006: What is Political about Jurisprudence? Courts, Politics, and
Political Science in Europe and the United States, MPIfG Working Paper 07/5.
Cologne: Max-Planck-Institut für Gesellschaftsforschung.
Reich, Norbert, 1994: The ‘November Revolution’ of the European Court of Justice:
Keck, Meng and Audi Revisited. Common Market Law Review, 31(3), 459–92.
Riese, Christoph, and Paul Noll, 2007: Europarechtliche und verfassungsrechtliche
Aspekte der Inländerdiskriminierung. Neue Zeitschrift für Verwaltungsrecht, 26(5),
516–21.
Rincón, Alfonso, 2007: EC Competition and Internal Market Law: On the Existence of
a Sporting Exemption and its Withdrawal. Journal of Contemporary European
Research, 3(3), 224–37.
Ringe, Wolf-Georg, 2011: Sparking Regulatory Competition in European Company
Law: The Impact of the Centros Line of Case-Law and its Concept of ‘Abuse of Law’.
In Rita de la Feria and Stefan Vogenauer (eds), Prohibition of Abuse Law: A New
General Principle of EU Law. Oxford: Hart Publishing, 107–25.
Rönnmar, Mia, 2010: Laval Returns to Sweden: The Final Judgment of the Swedish
Labour Court and Swedish Legislative Reforms. Industrial Law Journal, 39(3), 280–7.
Rosa, Stephane de la, 2012: The Directive on Cross-Border Healthcare or the Art of
Codifying Complex Case Law. Common Market Law Review, 49(1), 15–46.
Rosenberg, Gerald N., 1993: The Hollow Hope: Can Courts Bring about Social Change?
American Politics and Political Economy. Chicago: University of Chicago Press.
Roth, Günter H., 2008: Der EuGH und die Souveränität der Mitgliedsstaaten: Zusam-
menfassung und Schlussfolgerungen. In Günter H. Roth and Peter Hilpold (eds), Der
EuGH und die Souveränität der Mitgliedsstaaten: Eine kritische Analyse richterlicher
Rechtsschöpfung auf ausgewählten Rechtsgebieten. Vienna: Linde, 561–608.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Bibliography 273
Roth, Wulf-Henning, 2002: The European Court of Justice’s Case Law on Freedom to
Provide Services: Is Keck Relevant? In Mads Andenas and Wulf-Henning Roth
(eds), Services and Free Movement in EU Law. Oxford: Oxford University Press, 1–24.
Rueschemeyer, Dietrich, 2003: Can One or a Few Cases Yield Theoretical Gains? In
James Mahoney and Dietrich Rueschemeyer (eds), Comparative Historical Analysis
in the Social Sciences. Cambridge: Cambridge University Press, 305–33.
Ryland, Diane, and Angus Nurse, 2013: Mainstreaming After Lisbon: Advancing
Animal Welfare in the EU Internal Market. European Energy and Environmental
Law Review, 22(3), 101–15.
Sack, Detlef, 2010: Europäisierungsdruck und Parteiendifferenz in den deutschen
Bundesländern: Die Rechtsprechung des EuGH und die Novellierung von Tarif-
treueregelungen. Politische Vierteljahresschrift, 51(4), 619–42.
Sack, Detlef, 2012: Europeanization through Law, Compliance, and Party Differ-

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

ences: The ECJ’s ‘Rüffert’ Judgment (C-346/06) and Amendments to Public
Procurement Laws in German Federal States. Journal of European Integration,
34(3), 241–60.
Sartori, Giovanni, 1991: Comparing and Miscomparing. Journal of Theoretical Politics,
3(3), 243–57.
Saurugger, Sabine, and Fabien Terpan, 2017: Measuring Judicial Activism: Is the Court
of Justice of the European Union an Activist Court? Paper prepared for the EUSA
Conference, Miami.
Sauter, Wolf, 2009: The Proposed Patients’ Rights Directive and the Reform of (Cross-
Border) Healthcare in the European Union. Legal Issues of Economic Integration,
36(2), 109–31.
Schammo, Pierre, 2008: Arbitrage and Abuse of Rights in the EC Legal System.
European Law Journal, 14(3), 351–76.
Scharpf, Fritz W., 1966: Judicial Review and the Political Question: A Functional
Analysis. Yale Law Journal, 75(4), 517–97.
Scharpf, Fritz W., 1970: Demokratietheorie zwischen Utopie und Anpassung. Konstanz:
Universitätsverlag.
Scharpf, Fritz W., 1988: The Joint-Decision Trap: Lessons from German Federalism
and European Integration. Public Administration, 66(3), 239–78.
Scharpf, Fritz W., 1996: Negative and Positive Integration in the Political Economy of
European Welfare States. In Gary Marks (ed.), Governance in the European Union.
London: Sage, 13–39.
Scharpf, Fritz W., 1997: Games Real Actors Play: Actor-Centered Institutionalism in
Policy Research. Boulder, CO: Westview Press.
Scharpf, Fritz W., 1999: Governing in Europe: Effective and Democratic? Oxford:
Oxford University Press.
Scharpf, Fritz W., 2000a: Institutions in Comparative Policy Research. Comparative
Political Studies, 33(6/7), 762–90.
Scharpf, Fritz W., 2000b: Notes toward a Theory of Multilevel Governing in Europe.
Scandinavian Political Studies, 24(1), 1–26.
Scharpf, Fritz W., 2006: The Joint-Decision Trap Revisited. Journal of Common
Market Studies, 44(4), 845–64.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

274 Bibliography
Scharpf, Fritz W., 2008: Individualrechte gegen nationale Solidarität. In Martin
Höpner and Armin Schäfer (eds), Die Politische Ökonomie der europäischen Integra-
tion. Frankfurt a.M.: Campus, 89–100.
Scharpf, Fritz W., 2009: Legitimacy in the Multilevel European Polity. European
Political Science Review, 1(2), 173–204.
Scharpf, Fritz W., 2010: The Asymmetry of European Integration, or Why the EU
Cannot be a ‘Social Market Economy’. Socio-Economic Review, 8(2), 211–50.
Scharpf, Fritz W., 2015: After the Crash: A Perspective on Multilevel European
Democracy. European Law Journal, 21(3), 384–405.
Scharpf, Fritz W., 2016: De-Constitutionalization and Majority Rule: A Democratic
Vision for Europe, MPIfG Discussion Paper 14. Cologne: Max-Planck-Institut für
Gesellschaftsforschung.
Schelkle, Waltraud, 2017: The Political Economy of Monetary Solidarity. Oxford:

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Oxford University Press.
Schelling, Thomas C., 1960: The Strategy of Conflict. Cambridge, MA: Harvard
University Press.
Schenk, Angelika, 2016: Students Welcome? Politicisation in the Course of Implementing
European Case Law on Student Mobility and Study Finance. Paper presented
at Tagung des Projektverbundes ‘Europäische Wirtschafts- und Sozialintegration’,
Eberhard Karls Universität Tübingen, 21/22 January.
Schepel, Harm, 1998: The Mobilisation of European Community Law. In Jürgen
Brand and Dieter Strempel (eds), Soziologie des Rechts: Festschrift für Erhard
Blankenburg zum 60. Geburtstag. Baden-Baden: Nomos, 443–55.
Schepel, Harm, 2012: Constitutionalising the Market, Marketising the Constitution,
and to Tell the Difference: On the Horizontal Application of the Free Movement
Provisions in EU Law. European Law Journal, 18(2), 177–200.
Schepel, Harm, and Rein Wesseling, 1997: The Legal Community: Judges, Lawyers,
Officials, and Clerks in the Writing of Europe. European Law Journal, 3(2), 165–88.
Schimank, Uwe, 1992: Spezifische Interessenkonsense trotz generellem Orientierungs-
dissens. In Hans-Joachim Giegel (ed.), Kommunikation und Konsense in modernen
Gesellschaften. Frankfurt a.M.: Suhrkamp, 236–75.
Schmidt, Susanne K., 1998: Liberalisierung in Europa: Die Rolle der Europäischen
Kommission, Schriften des Max-Planck-Instituts für Gesellschaftsforschung Köln 33.
Frankfurt a.M.: Campus.
Schmidt, Susanne K., 2000: Only an Agenda Setter? The European Commission’s
Power over the Council of Ministers. European Union Politics, 1(1), 37–61.
Schmidt, Susanne K., 2004: The European Commission’s Powers in Shaping European
Policies. In Dionyssis G. Dimitrakopoulos (ed.), The Changing European Commis-
sion. Manchester: Manchester University Press, 105–20.
Schmidt, Susanne K., 2007: Mutual Recognition as a New Mode of Governance.
Journal of European Public Policy, 14(5), 667–81.
Schmidt, Susanne K., 2008: Europäische Integration zwischen judikativer und legisla-
tiver Politik. In Martin Höpner and Armin Schäfer (eds), Die Politische Ökonomie
der europäischen Integration. Frankfurt a.M.: Campus, 101–28.
Schmidt, Susanne K., 2009: Wechselseitiges Misstrauen? Die Grenzen der
Europäischen Dienstleistungsfreiheit. In Andreas Fischer-Lescano, Florian Rödl,
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Bibliography 275
and Christoph U. Schmid (eds), Europäische Gesellschaftsverfassung: Zur Konstitu-
tionalisierung sozialer Demokratie in Europa. Baden-Baden: Nomos, 167–84.
Schmidt, Susanne K., 2010: Gefangen im ‘lock in’? Zur Pfadabhängigkeit der
Rechtsprechung des Europäischen Gerichtshofs. Der moderne Staat: Zeitschrift für
Public Policy, Recht und Management, 3(2), 471–89.
Schmidt, Susanne K., 2011a: Law-Making in the Shadow of Judicial Politics. In Renaud
Dehousse (ed.), The ‘Community Method’: Obstinate or Obsolete? Basingstoke:
Palgrave Macmillan, 43–59.
Schmidt, Susanne K., 2011b: Overcoming the Joint-Decision Trap in Single-Market
Legislation: The Interplay between Judicial and Legislative Politics. In Gerda
Falkner (ed.), The EU’s Decision Traps. New York: Oxford University Press, 38–53.
Schmidt, Susanne K., 2012: Who Cares about Nationality? The Path-Dependent Case
Law of the ECJ from Goods to Citizens. Journal of European Public Policy, 19(1), 8–24.

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Schmidt, Susanne K., 2014: Judicial Europeanisation: The Case of Zambrano in
Ireland. West European Politics, 37(4), 769–85.
Schmidt, Susanne K., Michael Blauberger, and Wendelmoet van den Nouland, 2008:
Jenseits von Implementierung und Compliance. Die Europäisierung der Mitglied-
staaten. In Ingeborg Tömmel (ed.), Die Europäische Union: Politische Vierteljahres-
schrift, Sonderheft 40. Wiesbaden: VS Verlag für Sozialwissenschaften, 275–96.
Schneider, Harmut, 1996: Zum Funktionswandel der Grundfreiheiten des EGV und zu
seinen Auswirkungen auf das nationale Recht. Neue Justiz, 50(10), 512–15.
Schneider, Volker, and Raymund Werle, 1990: International Regime or Corporate
Actor? The European Community in Telecommunications Policy. In Kenneth
Dyson and Peter Humphreys (eds), The Political Economy of Communications.
London: Routledge, 77–106.
Schönberger, Christoph, and Daniel Thym, 2014: Germany. In Ulla Neergaard, Cather-
ine Jacqueson, and Nina Holst-Christensen (eds), Union Citizenship: Development,
Impact and Challenges. The XXVI FIDE Congress in Copenhagen, 2014. Congress
Publications, ii. Copenhagen: DJØF Publishing, 569–85.
Schreinermacher, Björn, 2013: Die Aushandlung Europäischer Gesetzgebung vor dem
Hintergrund der Rechtsprechung des Europäischen Gerichtshofs. University of Bremen,
Ph.D. thesis.
Schulten, Lucia, 2013: Das Studienförderungsgesetz im Lichte des Unionsrechts: Zum
Anspruch der ‘NC-Flüchtlinge’ auf Leistungen aus dem StudFG. In Werner Hauser
(ed.), Hochschulrecht. Jahrbuch 2013. Vienna: NWV Verlag, 453–67.
Segal, Jefrey A., 2008: Judicial Behaviour. In Keith E. Whittington, R. Daniel Kelemen,
and Gregory A. Caldeira (eds), The Oxford Handbook of Law and Politics. Oxford:
Oxford University Press, 19–34.
Seikel, Daniel, 2013: Der Kampf um öffentlich-rechtliche Banken: Wie die Europäische
Kommission Liberalisierung durchsetzt. Frankfurt a.M.: Campus.
Seikel, Daniel, 2014a: How the European Commission Deepened Financial Market
Integration: The Battle over the Liberalization of Public Banks in Germany. Journal
of European Public Policy, 21(2), 169–87.
Seikel, Daniel, 2014b: Nationale Anpassungsstrategien an das Fallrecht des Europäischen
Gerichtshofes: Europäisierung durch Richterrecht und innenpolitische Veto-Positionen.
Düsseldorf: WSI-Diskussionspapier 189.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

276 Bibliography
Seikel, Daniel, 2015: Class Struggle in the Shadow of Luxembourg: The Domestic
Impact of the European Court of Justice’s Case Law on the Regulation of Working
Conditions. Journal of European Public Policy, 22(8), 1166–85.
Shapiro, Martin, 1992: The European Court of Justice. In Alberta M. Sbragia (ed.),
Europolitics: Institutions and Policymaking in the ‘New’ European Community.
Washington, DC: Brookings Institutions, 123–56.
Shapiro, Martin, and Alec Stone Sweet, 2002: On Law, Politics, and Judicialization.
Oxford and New York: Oxford University Press.
Shuibhne, Niamh N., 2002: Free Movement of Persons and the Wholly Internal Rule:
Time to Move on? Common Market Law Review, 39(4), 731–71.
Shuibhne, Niamh N., 2012: The Third Age of EU Citizenship: Directive 2004/38 in the
Case Law of the Court of Justice. In Phil Syrpis (ed.), The Judiciary, the Legislature
and the EU Internal Market. Cambridge: Cambridge University Press, 331–62.

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Shuibhne, Niamh N., and Jo Shaw, 2014: General Report. In Ulla Neergaard, Catherine
Jacqueson, and Nina Holst-Christensen (eds), Union Citizenship: Development,
Impact and Challenges. The XXVI FIDE Congress in Copenhagen, 2014. Congress
Publications ii. Copenhagen: DJØF Publishing, 65–226.
Sibley, Eleanor, Emma Fenelon, and Nuala Mole, 2013: Family Reunification Require-
ments: A Barrier or Facilitator to Integration? London: AIRE Centre.
Sieberer, Ulrich, 2006: Strategische Zurückhaltung von Verfassungsgerichten. Gewal-
tenteilungsvorstellungen und die Grenzen der Justizialisierung. Zeitschrift für Poli-
tikwissenschaft, 16(4), 1299–323.
Slepcevic, Reinhard, 2009: The Judicial Enforcement of EU Law through National
Courts: Possibilities and Limits. Journal of European Public Policy, 16(3), 378–94.
Slot, Piet J., 1994: Energy and Competition. Common Market Law Review, 31(3), 511–47.
Smith, Mitchell P., 2005: States of Liberalization: Redefining the Public Sector in
Integrated Europe. Albany, NY: State University of New York Press.
Smith, Rogers M., 2008: Historical Institutionalism and the Study of Law. In Keith E.
Whittington, R. Daniel Kelemen, and Gregory A. Caldeira (eds), The Oxford
Handbook of Law and Politics. Oxford: Oxford University Press, 46–63.
Solanke, Iyiola, 2011: ‘Stop the ECJ’?: An Empirical Analysis of Activism at the Court.
European Law Journal, 17(6), 764–84.
Somek, Alexander, 2007: Solidarity Decomposed: Being and Time in European Citi-
zenship. European Law Review, 32(6), 787–818.
Sorensen, Karsten E., 2015: The Fight Against Letterbox Companies in the Internal
Market. Common Market Law Review, 52(1), 85–118.
Staton, Jeffrey K., and Will H. Moore, 2011: Judicial Power in Domestic and Inter-
national Politics. International Organization, 65(3), 553–87.
Stein, Eric, 1981: Lawyers, Judges, and the Making of a Transnational Constitution.
American Journal of International Law, 75(1), 1–27.
Steindorff, Ernst, 1999: Centros und das Recht auf die günstigste Rechtsordnung.
Juristenzeitung, 54(23), 1140–3.
Stone Sweet, Alec, 1999: Judicialization and the Construction of Governance. Com-
parative Political Studies, 32(2), 147–84.
Stone Sweet, Alec, 2000: Governing with Judges: Constitutional Politics in Europe.
Oxford and New York: Oxford University Press.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Bibliography 277
Stone Sweet, Alec, 2002: Path Dependence, Precedent, and Judicial Power. In Martin
Shapiro and Alec Stone Sweet (eds), On Law, Politics, and Judicialization. Oxford
and New York: Oxford University Press, 112–35.
Stone Sweet, Alec, 2004: The Judicial Construction of Europe. New York: Oxford
University Press.
Stone Sweet, Alec, 2007: The Juridical Coup d’État and the Problem of Authority.
German Law Journal, 8(10), 915–28.
Stone Sweet, Alec, 2010: The European Court of Justice and the Judicialization of EU
Governance. Living Reviews in European Governance, 5(2), 1–50.
Stone Sweet, Alec, 2012: A Cosmopolitan Legal Order: Constitutional Pluralism and
Rights Adjudication in Europe. Global Constitutionalism, 1(1), 53–90.
Stone Sweet, Alec, and Thomas L. Brunell, 1998: Constructing a Supranational Con-
stitution: Dispute Resolution and Governance, in the European Community. American

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Political Science Review, 92(1), 63–81.
Stone Sweet, Alec, and Thomas L. Brunell, 2013: Trustee Courts and the Judicialization
of International Regimes: The Politics of Majoritarian Activism in the European
Convention on Human Rights, the European Union, and the World Trade Organ-
ization. Journal of Law and Courts, 1(1), 61–88.
Stone Sweet, Alec, and James A. Caporaso, 1996: From Free Trade to Supranational
Polity: The European Court and Integration, Center for German and European
Studies Working Paper 2.45. Berkeley, CA: University of California.
Stone Sweet, Alec, and Jud Mathews, 2008: Proportionality Balancing and Global
Constitutionalism. Columbia Journal of Transnational Law, 47(1), 73–165.
Stone Sweet, Alec, and Kathleen Stranz, 2012: Rights Adjudication and Constitutional
Pluralism in Germany and Europe. Journal of European Public Policy, 19(1),
92–108.
Streeck, Wolfgang, 2000: Vorwort: Europäische? Sozialpolitik? In Werner Eichhorst,
Europäische Sozialpolitik zwischen nationaler Autonomie und Marktfreiheit: Die
Entsendung von Arbeitnehmern in der EU. Frankfurt a.M.: Campus, 19–35.
Streeck, Wolfgang, and Daniel Mertens, 2011: Fiscal Austerity and Public Investment:
Is the Possible the Enemy of the Necessary?, MPIfG Discussion Paper 11/12. Cologne:
Max-Planck-Institut für Gesellschaftsforschung.
Streeck, Wolfgang, and Kathleen Thelen, 2005: Introduction: Institutional Change in
Advanced Political Economies. In Wolfgang Streeck and Kathleen Thelen (eds),
Beyond Continuity: Institutional Change in Advanced Political Economies. Oxford:
Oxford University Press, 1–39.
Sun, Jeanne-Mey, and Jacques Pelkmans, 1995: Regulatory Competition in the Single
Market. Journal of Common Market Studies, 33(1), 67–89.
Sutter, Franz P., 2006: Pauschales Verbot grenzüberschreitenden Verlustabzugs—
Verstoß gegen Niederlassungsfreiheit. Europäische Zeitschrift für Wirtschaftsrecht,
17(3), 85–8.
Sydow, Sabine, 2015: Anmerkung zu BFH-Schlussurteil Meilicke. Internationales
Steuerrecht, 24(15), 551–4.
Syrpis, Phil, 2012: Theorising the Relationship between the Judiciary and the Legisla-
ture in the EU Internal Market. In Phil Syrpis (ed.), The Judiciary, the Legislature
and the EU Internal Market. Cambridge: Cambridge University Press, 3–24.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

278 Bibliography
Tallberg, Jonas, 2000: Supranational Influence in EU Enforcement: The ECJ and the
Principle of State Liability. Journal of European Public Policy, 7(1), 104–21.
Thatcher, Mark, 2007: Internationalisation and Economic Institutions: Comparing
European Experiences. Oxford: Oxford University Press.
Thomson, Robert, 2011: Resolving Controversy in the European Union: Legislative Decision-
Making Before and After Enlargement. Cambridge: Cambridge University Press.
Thorp, Arabella, 2009: Immigration: The Metock Case and its Implications for UK
Rules on Family Members of EEA Citizens. London: House of Commons Library.
<http://www.webcitation.org/6CUXVM0Ng> (accessed 28 June 2013).
Thym, Daniel 2015: Ende der Debatte, Legal Tribune Online. Available at <http://
www.lto.de/persistent/a_id/16169> (accessed 1 June 2016).
Tiedtke, Klaus, and Martin Mohr, 2008: Die Grundfreiheiten als zulässiger Maßstab
für die direkten Steuern. Europäische Zeitschrift für Wirtschaftsrecht, 19(14), 424–8.

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Töller, Annette Elisabeth, 2004: The Europeanization of Public Politics: Understand-
ing Idiosyncratic Mechanisms and Contingent Results. European Integration Online
Papers, 8(9), 1–25.
Töller, Annette Elisabeth, 2010: Measuring and Comparing the Europeanization of
National Legislation: A Research Note. Journal of Common Market Studies, 48(2),
417–44.
Treib, Oliver, 2014: Implementing and Complying with EU Governance Outputs.
Living Reviews in European Governance, 9(1), 1–47.
Troberg, Peter, 1997a: Dienstleistungen. In Hans von der Groeben, Jochen Thiesing,
and Claus-Dieter Ehlermann (eds), Kommentar zum EU-/EG-Vertrag. Baden-
Baden: Nomos, 1441–1523.
Troberg, Peter, 1997b: Niederlassung. In Hans von der Groeben, Jochen Thiesing, and
Claus-Dieter Ehlermann (eds), Kommentar zum EU-/EG-Vertrag. Baden-Baden:
Nomos-Verlag, 1303–1440.
Tryfonidou, Alina, 2009: Reverse Discrimination in EC Law. Alphen aan den Rijn:
Kluwer Law International.
Tryfonidou, Alina, 2010: Further Steps on the Road to Convergence among the Market
Freedoms. European Law Review, 35(1), 36–56.
Tsebelis, George, and Geoffrey Garrett, 2001: The Institutional Foundations of Inter-
governmentalism and Supranationalism in the European Union. International
Organization, 55(2), 357–90.
van de Gronden, Johan W., 2009: Cross-Border Health Care in the EU and the
Organization of the National Health Care Systems of the Member States: The
Dynamics Resulting from the European Court of Justice’s Decisions on Free Move-
ment and Competition Law. Wisconsin International Law Journal, 26(3), 705–60.
Van den Brink, Martijn, 2012: EU Citizenship and EU Fundamental Rights: Taking
EU Citizenship Rights Seriously? Legal Issues of Economic Integration, 39(2), 273–90.
Vanberg, Georg, 2005: Verfassungsgerichtsbarkeit und Gesetzgebung: Zum poli-
tischen Spielraum des Bundesverfassungsgerichts. In Steffen Ganghof and Philip
Manow (eds), Mechanismen der Politik: Strategische Interaktion im deutschen
Regierungssystem. Frankfurt a.M.: Campus, 183–213.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Bibliography 279
Vanhala, Lisa, 2006: Fighting Discrimination through Litigation in the UK: The Social
Model of Disability and the EU Anti-Discrimination Directive. Disability and
Society, 21(5), 551–65.
Vanhala, Lisa, 2009: Anti-Discrimination Policy Actors and their Use of Litigation
Strategies: The Influence of Identity Politics. Journal of European Public Policy,
16(5), 738–54.
Vauchez, Antoine, 2008: ‘Integration-through-Law’: Contribution to a Socio-History of
EU Political Commonsense, EUI Working Papers RSCAS 2008/10. Florence: European
University Institute.
Vauchez, Antoine, 2010: The Transnational Politics of Judicialization: Van Gend en
Loos and the Making of EU Polity. European Law Journal, 16(1), 1–28.
Vauchez, Antoine, 2012: Keeping the Dream Alive: The European Court of Justice and
the Transnational Fabric of Integrationist Jurisprudence. European Political Science

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Review, 4(1), 51–71.
Vauchez, Antoine, 2015: Brokering Europe: Euro-Lawyers and the Making of a Trans-
national Polity. Cambridge: Cambridge University Press.
Verschueren, Herwig, 2012: The EU Social Security Co-Ordination System: A Close
Interplay between the EU Legislature and Judiciary. In Phil Syrpis (ed.), The
Judiciary, the Legislature and the EU Internal Market. Cambridge: Cambridge
University Press, 177–204.
Voeten, Erik, 2013: International Judicial Independence. In Jeffrey L. Dunoff and Mark
A. Pollack (eds), Interdisciplinary Perspectives on International Law and Inter-
national Relations. New York: Cambridge University Press, 421–44.
von Bogdandy, Armin, 1995: Rechtsfortbildung mit Artikel 5 EG-Vertrag. Zur Zuläs-
sigkeit gemeinschaftsrechtlicher Innovationen nach EG-Vertrag und Grundgesetz.
In Albrecht Randelzhofer, Rupert Scholz, and Dieter Wilke (eds), Gedächtnisschrift
für Eberhard Grabitz. Munich: Beck, 17–28.
von Steinsdorff, Silvia, 2010: Verfassungsgerichte als Demokratie-Versicherung?
Ursachen und Grenzen der wachsenden Bedeutung juristischer Politikkontrolle.
In Klemens H. Schrenk and Markus Soldner (eds), Analyse demokratischer Regie-
rungssysteme. Wiesbaden: VS-Verlag, 479–98.
Wahl, Rainer, 2006: Das Bundesverfassungsgericht im europäischen und internatio-
nalen Umfeld. In Robert Chr. Van Ooyten and Martin H. W. Möllers (eds), Das
Bundesverfassungsgericht im politischen System. Wiesbaden: VS Verlag für Sozialwis-
senschaften, 477–93.
Wasserfallen, Fabio, 2010: The Judiciary as Legislator? How the European Court of
Justice Shapes Policy-Making in the European Union. Journal of European Public
Policy, 17(8), 1128–46.
Wasserfallen, Fabio, 2014: Political and Economic Integration in the EU: The Case of
Failed Tax Harmonization. Journal of Common Market Studies, 52, 420–35.
Wattel, Peter J., 2008: Commentary: Judicial Restraint and Three Trends in the ECJ’s
Direct Tax Case Law. Tax Law Review, 62(1), 205–19.
Weaver, R. Kent, 1986: The Politics of Blame Avoidance. Journal of Public Policy, 6(4),
371–98.
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

280 Bibliography
Webber, Douglas, 2014: How Likely is it that the European Union Will Disintegrate?
A Critical Analysis of Competing Theoretical Perspectives. European Journal of
International Relations, 20(2), 341–65.
Weiler, Joseph H. H., 1981: The Community System: The Dual Character of Supra-
nationalism. Yearbook of European Law, 1(1), 267–306.
Weiler, Joseph H. H., 1986: Eurocracy and Distrust: Some Questions Concerning the
Role of the European Court of Justice in the Protection of Fundamental Human
Rights within the Legal Order of the European Communities. Washington Law
Review, 61, 1103–42.
Weiler, Joseph H. H., 1991: The Transformation of Europe. Yale Law Journal, 100(1),
2402–83.
Weiler, Joseph H. H., 1994: A Quiet Revolution: The European Court of Justice and its
Interlocutors. Comparative Political Studies, 26(4), 510–34.

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Weimar, Tanja, 2014: Aktuelle Entwicklungen bei der grenzüberschreitenden Fami-
lienbesteuerung. Zeitschrift für Internationales Steuerrecht, 3(1), 1–7.
Weiss, Moritz, and Michael Blauberger, 2015: Judicialized Law-Making and Oppor-
tunistic Enforcement: Explaining the EU’s Challenge of National Defence Offsets.
Journal of Common Market Studies, 54(2), 444–62.
Werle, Raymund, 2007: Pfadabhängigkeit. In Arthur Benz et al. (eds), Handbuch
Governance: Theoretische Grundlagen und empirische Anwendungsfelder. Wiesbaden:
VS Verlag für Sozialwissenschaften, 119–31.
Werner, Benjamin, 2013: Der Streit um das VW-Gesetz: Wie Europäische Kommission
und Europäischer Gerichtshof die Unternehmenskontrolle liberalisieren, Schriften aus
dem Max-Planck-Institut für Gesellschaftsforschung 79. Frankfurt a.M.: Campus.
Werner, Benjamin, 2016: National Responses to the European Court of Justice Case
Law on Golden Shares: The Role of Protective Equivalents. Journal of European
Public Policy, 23(7), 1–17.
Wiesbrock, Anja, 2012: Granting Citizenship-Related Rights to Third-Country Nationals:
An Alternative to the Full Extension of European Union Citizenship? European
Journal of Migration and Law, 14(1), 63–94.
Wiesbrock, Anja, 2013a: Political Reluctance and Judicial Activism in the Area of Free
Movement of Persons: The Court as the Motor of EU–Turkey Relations? European
Law Journal, 19(3), 422–42.
Wiesbrock, Anja, 2013b: The Selfperpetuation of EU Constitutionalism in the Arena
of Free Movement of Persons: Virtuous or Vicious Cycle? Global Constitutionalism,
2(1), 125–50.
Wind, Marlene, 2009: Post-National Citizenship in Europe: The EU as a ‘Welfare
Rights Generator’? Columbia Journal of European Law, 15(2), 239–64.
Wind, Marlene, 2010: The Nordics, the EU and the Reluctance towards Supranational
Judicial Review. Journal of Common Market Studies, 48(4), 1039–63.
Wind, Marlene, 2014: The European ‘Rights Revolution’ and the (Non) Implementa-
tion of the Citizenship Directive in Denmark. In Lee Miles and Anders Wivel (eds),
Denmark and the European Union. Abingdon and New York: Routledge, 158–74.
Wollenschläger, Ferdinand, 2011: A New Fundamental Freedom beyond Market
Integration: Union Citizenship and its Dynamics for Shifting the Economic Para-
digm of European Integration. European Law Journal, 17(1), 1–34.
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Bibliography 281
Wollenschläger, Ferdinand, 2012: The Judiciary, the Legislature and the Evolution of
Union Citizenship. In Phil Syrpis (ed.), The Judiciary, the Legislature and the EU
Internal Market. Cambridge: Cambridge University Press, 302–30.
Woolfson, Charles, and Jeff Sommers, 2006: Labour Mobility in Construction: European
Implications of the Laval un Partneri Dispute with Swedish Labour. European Journal
of Industrial Relations, 12(1), 49–68.
Wray, Helena, 2011: Family Life and EU Citizenship: A Commentary on McCarthy
C-434/09. EUDO CItizenship Observatory. <http://eudo-citizenship.eu/citizenship-
news/479-family-life-and-eu-citizenship-a-commentary-on-mccarthy-c-43409-5-may-
2011> (accessed 15 June 2016).
Young, Alasdair R., 1994: Interests and Institutions: The Politics of Liberalisation in
the EC’s Road Haulage Industry, Working Paper, European Institute. Brighton:
University of Sussex.

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Zangl, Bernhard, 2005: The Rule of Law: Internationalization and Privatization. Is
there an Emerging International Rule of Law? European Review, 13(1), 73–91.
Zangl, Bernhard, and Michael Zürn, 2011: Make Law, Not War: Internationale und
transnationale Verrechtlichung als Baustein für Global Governance. In Berthold
Meyer (ed.), Konfliktregelung und Friedensstrategien: Eine Einführung. Wiesbaden:
VS Verlag für Sozialwissenschaften, 523–44.
Zimmer, Reingard, 2011: Labour Market Politics through Jurisprudence: The Influ-
ence of the Judgements of the European Court of Justice (Viking, Laval, Rüffert,
Luxembourg) on Labour Market Policies. German Policy Studies, 7(1), 211–34.
Zürn, Michael, 2001: Sovereignty and Law in a Denationalised World. In Richard
P. Appelbaum, William L. F. Felstiner, and Volkmar Gessner (eds), Rules of
Networks: The Legal Culture of Global Business Transactions. Oxford: Hart, 39–71.
Zürn, Michael, 2016: Opening up Europe: Next Steps in Politicisation Research. West
European Politics, 39(1), 164–82.
 Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Index

absorption 173, 174, 197, 200 anti-discrimination 15, 16, 198, 221
accommodation anticipatory reaction 174
administrative 202–6 approach
acte clair 36, 38, 169, 207, 219 integrated 53
acte éclairé 169, 219 asbestos 226 n. 27, 247
activism assumption
Court 21, 32, 37, 41, 136, 162, 180, 236, rational-choice 25
239–40 Austria
judicial activism 3, 18, 23, 28, 49, 181, 220, German medical students 41

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

222, 224, 228, 237 autonomy
legal 31, 62, 91 national 116, 169, 224
majoritarian 45, 157 Avoir Fiscal 154
actor Ayalti 207
corporate 30, 78, 137, 181, 196 Aziz, Miriam 78, 130, 139
private 2, 6, 7, 9, 14, 17, 18, 20, 34–6, 39, 46, Azoulai, Loïc 20, 167, 242
52, 54, 59, 62–5, 68, 69, 71, 78, 83, 85, 87,
88, 90, 91, 96, 98, 104, 107, 108, 122, 123, backlog 190
126, 151, 156, 166, 168, 224, 226, 227, 231 bailout 164, 168
societal 21, 43, 52, 198, 200, 202, 224–7, balance of power 6–7, 13, 24, 40, 106, 114, 248
229, 236, 241, 242 balancing 51, 58, 59, 65, 69, 85, 101, 108, 113,
administration 116, 122–4, 152, 163, 229, 246
Eastern European 210 Banca Populare 166
member-state 21, 169, 170, 195, 238, 243 Barber 40, 41
administrative control 202, 210, 214, 229 Barnard, Catherine 45, 53, 54, 64, 68, 71,
adversarial legalism 35 73–9, 75 n. 7, 78 n. 8, 85–8, 137
Advocate General 28, 29, 110, 135, Bartolini, Stefano 244
158 n. 18, 177 base erosion and profit shifting (BEPS) 165
AG Colomer 80, 83, 135 basic rights 4, 38
AG Cruz Villalón 218 Baumbast 79, 131, 133, 138
AG Jacobs 80 Beck, Gunnar 1 n. 2, 7, 14, 55–9, 61, 67
AG Maduro 83, 86, 161 behaviour
AG Mischo 117 strategic 4 n. 4, 28
AG Sharpston 29, 138, 181, 181 n. 10, 182–4, Bellamy, Richard 13, 14, 33, 140, 233, 246
186, 188, 190, 204 ‘benefit tourism’ 80, 217
AG Stix-Hackl 177, 178 BGH 73
AG Tesauro 120, 175 Bidar 132, 135, 217
AG Villalon 81 bilateral talks 211
agent 3, 4 n. 4, 39, 40, 150, 219, 251 Bilka 221
Agonese 87 blame avoidance 7, 47, 51, 241
agreement Blauberger, Michael 14, 21, 24, 32, 81, 83, 99
bilateral 102, 202, 211, 214 n. 2, 102, 103, 173, 174, 197, 203 n. 2, 205,
air transport 101–3 206, 212, 213, 215–17, 219, 220, 222 n.
Akrich 137, 203 23, 224, 225, 241, 248
alert mechanism 152 blind spot 5, 170, 171, 174
Alimanovic 81, 138, 224 blocking minority 103, 147
Alokpa 184 Bolkestein directive 109
analysis Bolkestein, Frits 159
proportionality 58, 113, 237 Börzel, Tanja A. 34, 172, 173, 174, 200
anarchy 25, 246 Bosman 75, 87, 105
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

284 Index
Brexit 5, 37, 81, 138, 139, 216 third-country national 137, 171, 195, 197,
British Airways 222, 225 203–5, 227, 237–8
Brown 79, 132, 138 Citizenship Directive 15–16, 74, 79, 123,
Bulmer, Simon J. 103 126–7, 131–2, 134, 136–7, 141, 150, 196,
Bundesfinanzhof 179 204, 217, 219, 223, 234–5, 239, 250
Bundesgerichtshof 175, 226 civil-law 7
Bundestag 207, 208 civil society 80, 188
burden of proof 110, 117–21, 146, 179, codification 9–12, 14–17, 20, 23, 35, 44, 45,
235, 239 47, 74, 94, 96, 98, 99, 109, 113, 116, 123,
Burley, Anne-Marie 26 124, 126, 127, 136, 141, 142, 145, 146,
150–68, 171, 194, 231, 233–5, 239, 240
Caporaso, James A. 52, 127, 139, 167 coincidence 90
car emissions standard 97 collective agreement 86, 109, 209, 214,
Carbone 143 215, 225
Carpenter 88, 89 Collins 80, 135, 136
Carrubba, Clifford 3, 7, 12, 25, 27, 28, 34, Commission

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

39 n. 5, 42–4, 47, 52, 196, 228, 236 guardian of the Treaty 34, 94, 96, 100,
Cartesio 73, 74, 89 124, 233
Casati 77 monopoly on initiative 94
case-by-case assessment 218, 219, opinion 34, 62, 114, 121, 133, 135, 177, 177
238, 243 n. 3
case-by-case decision 229 pressure 13, 94, 100, 106, 114, 133, 135,
case law 140, 154, 159, 181, 202
ambiguity 55, 57, 162, 173 role 13, 14, 94, 96, 106, 124
attention 11, 18, 20, 28, 35, 39, 45, 106, 161, Commission v UK 81
174, 182, 192, 194, 200, 209, 233, 250 Common Consolidated Corporate Tax Base
coherence 37, 63, 83, 84, 90 (CCCTB) 163–5
policy implication 3, 4, 32, 35, 39, 50, 54, common-law country 57, 186
98, 239, 251 common-law tradition 5
shadow 16, 93–100, 106, 112, 122, 126 comparative politics 6, 106, 233
case-law constraints 10–13, 19, 21, 94, 112 n. comparative turn 94
11, 124, 229, 241, 244 comparison 7, 17, 26, 33, 42, 48, 50, 72, 91,
case-law development 6, 7–15, 17–19, 21, 34, 107, 133, 189, 201, 203, 204, 218, 220,
38, 46, 48, 50–92, 96, 98, 99, 126–8, 228, 231, 241, 247
131–2, 135–9, 142, 147, 148, 150, 152, compensation
161, 166, 168, 169, 173, 174, 179, 180, executive 209–12
182, 195, 199, 228, 231–4, 237, 243, compensatory measure 12, 171, 173, 200, 209,
246–8, 250, 251 210, 229, 230
case load 6, 7, 26, 36, 37, 46, 47, 52, 62 competition law 1, 3, 15, 36, 100, 100 n. 3,
case study 13, 14, 16, 17, 21, 43, 44, 45, 95, 101, 103–6, 150, 198, 224, 248
127, 151, 171, 174, 198, 220, 222, 238 complaint 63, 140, 142, 146, 179, 210, 214,
Cassis de Dijon 62, 64, 81, 97, 116 225, 228, 240
Centros 72, 176, 212, 213, 216, 229 compliance 9–12, 20, 21, 25, 27, 34–8, 44, 47,
Chakroun 207 71, 93, 94, 99 n. 2, 114, 121, 163, 170–4,
Chamier-Glisczinski 146 181, 194, 195, 197, 198, 200, 205, 212,
change 226–8, 231, 235, 238
court-driven 44 compromise 8–10, 13, 55, 56, 59, 80, 111, 136,
child benefit 205–6, 217–18, 227 147, 152, 225, 233
Christian Democrats 111, 208 Conant, Lisa 3, 11, 12, 21, 23, 24, 27, 37, 44,
Cichowski, Rachel A. 5, 15, 35, 44, 45, 127 48, 53 n. 1, 88, 98, 121, 127, 138, 169, 170,
citizen 173, 199, 206, 224, 235, 238
inactive 79, 131–4, 138, 234 Confederation of Swedish Industry 225
citizen rights conflict
comprehensive 79, 180 cross-border 54
citizenship conflict of laws 242
Irish 171, 182, 186–90 ‘contained compliers’ 21
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Index 285
consensus 31, 39, 151, 152, 211, 215, 225, support of lower national courts 37, 43
226, 228 weakness 25, 44
constituents 43, 48, 242 court-curbing mechanism 41, 42, 62
constitution critical juncture 60, 61, 61 n. 3, 62, 67, 68, 84,
national 1, 2, 18, 39, 48, 50, 56, 231, 242, 85, 88
246, 251 criticism 31, 32, 42, 127, 152, 158 n. 18, 215,
economic 9, 250 219, 250
constitutional cross-border element 53, 88, 182, 203
change 7, 36, 187 Czech constitutional court 38
court 2, 24, 29, 30 n. 3, 32, 33, 35, 38, 39, 50,
58, 124, 179, 196 n. 43, 207, 221, 223, 229, D’Hoop 76
233, 244, 251 Dagdelen, Sevim 208
status 1, 2, 5, 7, 11, 12, 19, 20, 21, 23, 90, 94, Daily Mail 72, 155
110, 112 n. 12, 229, 232, 243, 248 Danish Bottles 97
constitutionalization 2–5, 12, 17, 19, 23, Dano 81, 82, 138, 224, 249
35, 36, 45, 47, 48, 52, 58, 92–5, 99, 106, Dassonville

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

110, 113, 124, 126, 151, 165, 233, 240–2, formula 53, 75
245, 248 Davies, Gareth 37, 68, 70, 75, 79, 95, 108, 109,
constitutionalized Treaty 6, 9, 13, 51, 111, 134, 136, 137, 143, 150, 153, 198,
58, 123 227, 236, 237, 250, 251
constitutionalizing 5, 8, 15 De Cuyper 80
constraints dead letter 198 n. 1
political 44 decision making
contained compliance 11, 37, 99 n. 2, 170, majoritarian 2, 4, 5, 246, 248, 250
171, 173, 174, 194, 195, 198, 227, 228, 235 decision-making process 4, 6, 10, 17, 29, 30,
convergence 82, 84 36, 55, 56, 91, 93, 95, 96, 107, 112 n. 11,
corporate tax 20, 158 n. 18, 161–5, 114, 115, 165, 231, 233, 242, 248, 250
165 n. 24, 174 Decker 143, 249
Costa v ENEL 1 deconstitutionalization 250
country-of-origin principle 108, 111, 125 default condition 10, 94–7, 101, 103, 123,
court 124, 250
activist 2, 21, 28, 41, 44, 45, 46, 141, 170, Defence Procurement Directive 102
219, 220, 221, 236, 237, 239, 240 Defrenne 36, 177
discretion 13, 27, 39, 40, 43, 47, 56, 57, 119, democratic deficit 242, 244
186 Denkavit 120, 158 n. 18
disempowerment 24 Department of Work and Pensions 218
domestic 2, 17, 37, 38, 41, 63, 73, 85, 135, deportation injunction 189, 189 n. 26
159, 166, 169, 184, 188, 193–5, 207, 219, Dereci 183–4
220, 222, 223, 225–7, 237 Derlén, Mattias 14, 57, 82, 156
implementation 18, 36, 93, 200 direct effect
independent 12, 13, 23, 24, 25, 27, 40, 48, and supremacy 1, 5, 8–9, 12, 36, 61 n. 3, 62,
50, 241 82, 167, 219, 231
institutional rules 18, 24, 27, 30, 39, 42 horizontal 85–7, 151, 225
limited power 24 Directorate-General
leeway 3, 9, 23, 44, 47, 67, 96, 154, 161, 166, Employment 145, 152
168, 228 Health 145
lower 12, 14, 26, 36, 37, 39, 43, 47, 48, 51, Internal Market 96, 109, 145, 152
61, 62, 76, 91, 92, 94, 149, 175, 176, 195, ‘dis-unity’ 243
196, 220, 228, 232, 237 discretion 13, 23, 27, 39, 40, 43, 47, 56, 57, 65,
member-state 18, 33, 36–9, 44, 61, 179, 236 119, 186, 188, 207
political actors 23–49, 103 discrimination against workers 128
political role 24 disempowerment 24
reactive actor 3, 19, 24, 44, 48, 169, 239 dispute
receptive 148–50 legal 33, 46, 61, 225, 232
references 119, 158, 166, 181, 190, 220, 249 divide and conquer 100, 103–6, 121, 124
strength 41, 47 doctrinal coherence 232
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

286 Index
doctrine 30, 33, 36, 38, 48, 50, 62, 87, 91, 169, European Economic Area (EEA) 177, 192 n.
207, 240 34, 218 n. 19, 218 n. 20, 249
Dogan 208 European Economic Community (EEC) 1,
domestic effects 2, 169 2, 32, 34, 41, 75, 77, 79, 101, 117, 128,
domestic impact 21, 88, 108, 172, 198, 132, 221
228, 236 European law
domestic parliament 236 domestic impact 21, 88, 172, 198, 228, 236
domestic repercussion 147 European Parliament 3, 10, 35, 56, 74, 79, 93,
double burden 65, 74 95, 98, 99, 101, 105, 106, 109, 110–15, 112
double taxation 154, 157 n. 16, 158, 162, n. 12, 115 n. 14, 117–19, 124, 125, 128,
163, 176 129, 133, 138, 141, 151, 153, 162, 167,
Dublin hop 137, 204 180, 233, 244
Europeanization
Eastern enlargement 81, 109, 135, 154, 164, effects 16, 170, 171, 174, 179, 197–230
209, 210, 211, 214, 216, 217, 234 Europhile 4
ECAS see European Citizen Action Service ever closer union 9, 47, 51, 59, 69, 232

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

(ECAS) exception
ECJ to freedom 20, 65–7, 78, 80, 96, 97 n. 1,
activist 21, 24, 41, 236, 237, 246 103, 108, 111, 116, 122, 145, 157
chamber 29, 42, 137 n. 7 ‘executive control thesis’ 18, 23, 25
conflict 33, 38 exit 16, 165–6, 219, 234, 239
dependence 12, 61 expectation 21, 43, 45, 48, 95, 109, 150, 152,
highest national court 38 170, 180, 216, 238, 247
institutional independence 33 explanatory framework 50
judge 27, 28, 30, 33, 221, 246 expulsion 133, 134, 136, 140, 193
jurisdiction 2, 19, 38, 42, 52
political actor 5, 18, 23–49, 52, 228 Fabbrini, Federico 151, 152
power 2, 26, 27, 38, 104, 167, 172, 236, 246 Falkner, Gerda 16, 77, 166, 198 n. 1
responsive 7, 239 family reunification 138, 182, 184, 186, 188,
relationship with national court 38 188 n. 17, 203, 204, 207
weaker 44, 104 Family Reunification Directive 184, 185,
ECOFIN Council 164 207, 208
economic constitution 9, 250 Farrell, Henry 244
economic heterogeneity 81, 234 Federal Ministry of Economics 110 n. 6, 110
economic impact 16 n. 7, 111 n. 10, 121, 124, 175, 208
EEC Treaty 1, 32 Fédération Internationale pour le Droit
Elchinov 149 Européen (FIDE) 31, 194
electricity financial solidarity 134, 167, 217, 224
ownership unbundling 104 Finanzkontrolle Schwarzarbeit 210
equality before the law 98, 123, 126, 249 fine-tunes 45, 180, 182, 228, 249
erga omnes 9, 11, 16, 46, 123, 147, 167, 169, fine-tuning 46, 68, 73, 76, 81, 91, 161, 182–5,
180, 196, 199 190, 202, 203, 209, 227, 235–8, 249, 250
‘esprit de corps’ 31 fiscal sovereignty 162
ethos 23, 31 floodgates 186
ETUC 111 focal point 62, 67, 90, 243
EU Charter of Fundamental Rights 59, 221 food additive 117, 122
EU citizenship 4, 5, 16, 80, 109, 127, 130, 139, food supplement 121
180–4, 186, 187, 190, 191, 193, 197, 204 Foie Gras 116
EU lawyers 31, 34, 51, 63, 87, 232, 250 four freedoms
Eurobarometer 42 cross-border 53, 162
European Citizen Action Service freedom
(ECAS) 140–1 abuse 68, 135, 165, 211, 234
European Commission capital 3, 19, 52, 60, 77–8, 82, 169, 178, 232
Directorate-General 96, 145, 152 establishment 14, 15, 60, 63, 69, 70, 72–4,
European Convention on Human Rights 78, 82, 83, 86, 90, 107, 127, 151, 154, 155,
(ECHR) 185, 192, 196 156, 161, 162, 211, 212, 215, 222, 225
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Index 287
goods 3, 14, 15, 19, 52, 60, 62, 63, 64, 67, 68, preferences 3, 5, 7, 12, 23, 25, 29, 42, 43, 47,
69, 71, 72, 74, 82, 83, 88–90, 108, 119, 96, 202, 227, 241
169, 232 rational 25
persons 3, 14, 52, 60, 69, 169, 232 Graetz, Michael J. 45, 157, 157 n. 16, 158,
services 3, 14, 19, 38, 52, 53, 60, 69, 70–2, 158 n. 18, 159, 163, 180
74, 82, 86, 88, 89, 104, 107, 108, 110, 112, Graf 75, 75 n. 7
113, 115, 138, 143, 144, 146, 147, 169, Granat, Katarzyna 151, 152
201, 205, 209, 210, 211, 215, 216, 225, Grand Chamber 29, 149, 177, 178, 182, 183
232, 234 Green Party 207
workers 14, 19, 60, 69, 74–7, 82, 127, 209 Greenham and Abel 117
Fra.bo 86, 225, 225 n. 26, 227 Greer, Scott L. 141, 148, 150, 150 n. 13
fraud 104, 137, 164, 210, 211, 214, 226, 229 Grimm, Dieter 2, 23, 50
Free Democrats 208 Grogan 143
free movement ground handling 103
hindering 118 group relief 155, 156, 161
principle 41, 77, 136, 205 Grzelczyk 131, 133–5, 138, 181, 217, 223

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

free movement of persons
non-discrimination 16, 41, 74 Halbeinkünfteverfahren 177
freedom of services harmonization 13, 45, 53, 54, 67, 97, 101, 104,
temporary 70 115, 117, 153, 153 n. 14, 154, 155, 157,
passive 72, 143, 146, 205 163, 164, 166, 245, 248
fundamental freedom Hartlapp, Miriam 15, 35, 96, 109, 145
convergent interpretation 68, 82 Hartmann 76
exemption 20, 65, 85 Hatzopoulos, Vassilis 59, 66, 70, 71, 72, 88,
parallel interpretation 63, 83, 90 104, 105, 107, 108, 142, 143, 144, 149
health services 32, 111, 112, 141–8, 150,
G20 20, 165, 166 201, 203
Gabel, Matthew 3, 7, 12, 25, 27, 28, 34, 39 n. 5, Héritier, Adrienne 101, 102, 172, 244
42–4, 47, 52, 196, 228, 236 Hendrix 76–7
gambling 104, 105, 111, 222 High Court 38, 149, 170, 189, 190, 195,
Garcia Nieto 81, 138, 224 196, 226
Garrett, Geoffrey 26, 33, 39, 40, 41, 67, 98 Hilpold, Peter 91
gatekeeping 2, 96, 195 Hix, Simon 4, 6, 94, 106, 233, 244
Gauweiler 29, 38 Hoekstra 129
Gebhardt 110 Hofmann, Andreas 13, 34, 72, 128, 129,
Gebhardt, Evelyne 109, 110, 113 129 n. 2, 130, 131, 132, 134
general interest 19, 28, 66, 88, 159, 177 ‘hollow hope’ 25, 239
general practitioner 147 home-country control 107, 110
General Product Safety Directive 117 home state 77, 107, 112, 142, 156, 164, 205,
generalizability 16 209, 210
Genschel, Philipp 153, 154, 154 n. 15, 158, Höpner, Martin 30, 31, 86, 90, 106, 112, 127,
162–5 153 n. 14, 167, 213, 245
genuine and effective 8, 11 host state 64, 65, 74, 77, 110, 112, 132, 134,
Genuine Prospect of Work test 218 142, 145, 154, 155, 158, 209, 210
German Basic Law 196 n. 43 House of Commons 115, 115 n. 17
German beer-purity 87 human right’s jurisdiction 38
German Constitutional Court 29, 30 n. 3, 38, Hunting Act 85
39, 179, 207 hypothesis 170, 171
German Federal Ministry of Finance
165, 177 IBAN 53
German General Tax Act 178 Ibrahim 137
German labour court 221 Iida 183, 192
Golden Shares 77, 78, 106, 209 illegal activities 202, 210–11, 226
Golub, Jonathan 37 Imran 207
government Immigrant Council of Ireland 188, 188 n. 20,
Conservative 138, 217 190 n. 28
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

288 Index
Implementation 18–19, 24, 34–6, 46, 54, 63, Irish abortion case 38
93, 96, 98–9, 113–14, 122, 130, 140, 148, Irish Born Child Scheme 189
169–72, 173 n. 1, 197, 200–1, 209, 230, Irish constitution 190
235–6, 240, 244 Irish Immigrant Support Centre (NASC) 189
imputation system 155–6, 157 n. 16, 175–6, Italian trailers 85
178, 180 ius soli 185–7, 191
inaction 86, 199, 207
incentive jewellery 121
to codify 98, 232 job-seekers 133
incomplete contract 14, 23–4, 52, 55–6, 59 Joerges, Christian 86, 91, 112, 242,
incomplete information 55 243, 247
inconclusive compromise 56 joint-decision trap 8–9, 16, 42, 47, 95, 165
incremental nature 22 judge
indication 16, 32–3, 37, 52, 61, 168, 201 Eurosceptic 42
indirect tax 76, 153–4 policy preferences 23, 29, 30
individual assessment 124, 217, 229, 234, 249 judgment

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

inductive reasoning 17 ambiguity 56
inefficiency 109 contentious 30, 33, 177, 236
inertia 173–4, 194, 197 landmark 61, 186
influence judicialization
independent 24, 91 cost 27, 163, 205
infringement procedure 6, 13, 34–5, 39 n. 5, theory 26
54, 69, 83 n. 9, 85, 100, 103–4, 107, 114, judicial dominance 99
120 n. 21, 121, 124, 129, 131, 135, 140, judicial influence 45
142, 154, 157, 159, 172, 176, 179, 210, judicial policymaking 10, 93–5, 104, 106, 123
216–17, 219, 231 judicial power 25–6, 91, 239–40, 246, 250
Inizan 144 judicial pressure 94, 154, 166
Inspire Art 72 judicial review
institutional change 10, 17, 226, 228 tradition of 37, 124
institutionalism judicial self-restraint 41, 104, 244
actor-centred 17 judicial system 36–7, 42–3, 55, 150,
insurance 31, 53–4, 74–5, 107, 130–1, 134, 194, 222
142 n. 10, 143–4, 155, 207, 211 judiciary
integration domestic 198, 236
distinctive approach 52 ‘juridical coup d´état’ 6, 239
negative 4, 127, 245, 246 justice
positive 4, 163, 245 contained 235
inter partes 9, 11, 121, 169
intergovernmentalism Kadelbach, Stefan 138
liberal 94 Keck 67–8, 83–5, 89, 232
paradigm 18, 172 Keeling, David 32, 33, 236
intergovernmentalism/supranationalism 2 Kelemen, Dan 24, 26, 27, 32, 35, 36, 39, 41, 42,
internal market 60, 61, 61 n. 3, 224, 240, 248
social dimension 145 Keller Holdings 155
international courts 9–13, 22, 25–6, 43, 47, Kloka, Marzena 141 n. 9, 146, 147
58, 236, 246 Knill, Christoph 58, 172, 220
international relations 24–5, 106, 246–7 Kochenov, Dimitry 140
international tax treaties 155 Kohll 143, 249
interpretation Kokott, Juliane 29, 110
broad 53, 54, 62, 67, 73, 90, 104, 158, 199, Kranenpohl, Uwe 56
222, 237 Kücükdeveci 221
expansive 10, 31, 139, 232, 234
interpretative guideline 11 Lair 132
interstate commerce 53–4 language test 206–9, 212, 227–8
interventions 34–5, 39, 43–4, 47, 129, 131–2, Larsson, Olof 3, 7, 12, 23, 27, 33, 40, 43, 44,
195, 203, 224–5, 232, 241 47, 149, 196, 228, 236
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Index 289
Laval 15, 19, 32, 42, 86, 112, 112 n. 12, 113, gambling 105
127, 150–3, 168, 211, 215–16, 219, 222–3, telecommunication 103
225, 227–8 Lidl Belgium 162, 162 n. 20
‘law of the land’ 20, 90 limited company 72, 212–13
lawful residence 203 Lindholm, Johan 14, 156
least-likely case 181, 229 Lindstrom, Nicole 112, 112 n. 12
leeway 3, 9, 23, 44, 47, 67, 96, 144, 154–5, 161, link territory
166, 168, 180, 219, 228 legal order 65, 155
legal argument literature
transfers 14, 59, 63, 69, 72, 84, 89, 183 shortcomings 18, 24, 47–9
legal argumentation 19, 62 litigant
legal certainty 8–9, 20, 54–5, 59, 89, 94, 98–9, incentives 8, 67, 76, 78, 82, 175
116, 118–19, 121–4, 126, 141, 145–6, 150, interests 8, 26, 28, 36, 63, 161, 175, 196
167, 188, 219, 232, 234, 240, 247, 249 litigation
legal clarification 56 credible threat of 101
legal community 18, 31–3, 80, 184 strategies 59, 63, 69

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

legal doctrine 30, 50, 91, 240 lobbying 55, 224
legal expenses 189, 189 n. 26 ‘lock in’ 60
legal framework long-term care 146–7, 242
alternative 43 lower court judge 37
legal indeterminacy 55 Luisi 143
legal innovation 52, 61, 84 Luxembourg 101, 114, 119 n. 19, 143, 155,
legal order 159–60, 184, 213, 215–16, 249
alternative 8, 59, 71, 198, 230
territory link 65, 155 Maastricht Treaty
legal principle Protocol 38, 40, 152
new 57, 80 McCarthy 182, 191
legal profession 28, 31, 34, 48, 62, 65 McCown, Margaret 62
legal service Maduro, Miguel P. 29, 33, 45, 157
pro-bono 196 Majone, Giandomenico 4, 31, 251
legal standing 36, 59, 62, 73, 90 majoritarian activism 45, 157
legal system majority
European 36–7, 43, 47, 55, 91, 124 simple 29, 33, 42
legal text Malmö route 137, 204
openness 57 Mancini 36–7, 187
legal uncertainty mandatory requirement 65–6, 82, 97 n. 1,
supranational level 8, 55, 56, 57, 247 116, 120
legalization 10, 25 Mangold 37, 196 n. 43, 221, 227–8
legislative policymaking 10, 19, 44, 93, 99, Manninen 43, 157 n. 17, 177–8, 180
123, 247, 250 ‘mantra’ 156
legislative process 10, 44, 91, 93–5, 100, 106, ‘marginal and ancillary’ 8, 11, 129, 217–18, 233
111–14, 124–5, 130, 135, 141, 197, 201, Mariano 89
233–4, 239, 245, 250 market fragmentation 97
legitimacy market-making 4, 145
input 93, 243 Marks & Spencer 155, 161–2
output 93, 243 marriages of convenience 203
legitimation 4, 19, 31–3, 45, 48, 57, 63, 89, Martinez Sala 79, 131, 136
125–6, 140, 242, 244 Martinsen, Dorte S. 3, 13, 44–5, 48,
lesser evil 99–103, 148 58, 77, 137, 141, 143–4, 148–9, 151–2,
level-playing field 9, 20, 98, 120, 122, 123, 203, 205, 233, 250
168, 233, 234 Mattli, Walter 26–7
Levin 8, 129, 217 mechanism 4, 15–17, 25, 29, 36, 41, 60, 152
lex specialis 79 media attention 190, 192, 205
liberalization Meilicke 43, 156–7, 169, 175–7, 177 n. 3,
air transport 101–3 178–80, 194, 196–8, 202–3, 227, 235,
electricity 100, 103–4 238, 240
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

290 Index
Meilicke tax cases 171, 175 normative discussion 4
member state, member states Nouvelles Frontières 102
bargaining power 96
Eastern 111, 144, 146, 147, 154, 205, 220, 222 O and S 183
heterogeneous policy preferences 40 Obermaier, Andreas 45, 46, 142, 182, 183,
homogeneous institutional preferences 40 184, 201, 203, 249
institutional heterogeneity 8, 33, 172 observations 6, 13, 27, 33, 39, 39 n. 5, 40, 44,
observations 6, 27, 33, 39, 40, 44, 62, 112, 62, 112, 131, 153, 178, 182, 186, 232, 237
131, 237 OECD 20, 154, 165, 166, 235
prerogative 10, 15, 95, 103, 112, 158, 242 ‘one-shotters’ 36
Southern 146–7 opinion
Typologies 172–3, 197 AG 42, 61, 83, 85–6, 135, 138, 161, 175,
Menéndez, Agustín José 71, 128–9, 131, 139 177, 181–2, 184, 187–8, 190, 204, 218
merger-control regulation 103 dissenting 28–9
Metock 137, 193, 203–4 member state 48, 62, 78, 87, 143–4
Mickelsson 85 reasoned 114, 121

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

minimum protection 151 opportunity structure 12, 14, 56, 171, 172,
minority 9, 40, 95, 103, 139, 147, 199, 228, 238 199, 220
model opposition
separation of powers 27, 43 court 38, 224
modification 44–5, 48, 118, 149, 152–3, government 2, 195, 197
173, 233 member state 26, 46, 81, 101, 104, 141,
Monti II Regulation 15, 44, 127, 150–3, 168 204, 240
motor of integration 26, 48 organization 1, 20, 43, 50, 68, 126, 137, 150,
MRAX 133 154, 166, 235, 246, 247
Müller-Fauré & Van Riet 144 outright monetary transaction 29, 38
multilevel system 7, 12, 43, 47, 50, 55, 59, over-constitutionalization 3, 4, 12, 23, 35, 47,
195–6, 202, 230, 241 48, 52, 92, 94, 95, 106, 113, 124, 126, 151,
multilingualism 59 233, 240, 241, 242, 245, 248
multinational corporations 165 override 27, 40, 43, 44, 47, 114, 134, 209
Muslims 208 overriding reason 111, 161, 177
mutual recognition 15, 19, 66–7, 70, 95, 105,
110, 114, 115, 115 n. 14, 116–20, 122–3, Pantazatou, Katerina 158, 159, 166, 168
150, 167, 212, 233–5, 239, 248, 250 ‘paradox of weakness’ 241
mutual recognition clause 116, 123 Parc Asterix 184
parliament 3, 5, 10, 33, 35, 37, 56, 74, 79, 93,
national treatment 99, 105, 110, 115 n. 14, 118, 119, 121, 128,
test 155 140, 151, 152, 164 n. 21, 169, 233, 236
Naurin, Daniel 3, 7, 12, 23, 27, 33, 40, 43, 44, parliamentary sovereignty 5, 66, 113, 124, 220
47, 149, 196, 228, 236 path-dependence 47, 50–92
negative integration 4, 127, 245, 246 path-dependent
network analysis 82, 156 attributes 60
Neyer, Jürgen 242, 247 path deviation 62, 91
NGOs 36, 108, 137, 141, 185, 190, 192, 196 Patient Mobility Directive 15, 16, 20, 44, 123,
NHS 144, 147, 201 126, 127, 141–50, 167, 220, 242, 249, 250
Nicolaïdis, Kalypso 64, 111 Patmalniece 223
non-action 171 permanent residence 133, 134
non-adoption 44, 152 Pierik 143
non-compliance Pignataro 88–9
threat 38, 42 Pistre 88–9
non-decision 46, 47, 171–4, 197, 200, 209, 210 ‘Polanyi in Brussels’ 127
non-discrimination police control 35
rules 62, 64, 66, 105, 147, 217, 224 policy
non-restriction 65–8, 71, 72, 74–6, 80, 81, 87, alternative 20, 41, 166
90, 112, 122, 125, 138 policy change
normative dimension 14, 22, 245, 246 domestic 2, 21, 220
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Index 291
policy field 14, 16, 98, 100, 148, 157, 170, 171, public policy 78, 81, 111 n. 9, 113, 157
195, 231, 237 public security 65, 78, 81, 111 n. 9, 113,
policy implication 3, 4, 17, 32, 35, 39, 50, 54, 133, 157
98, 239, 251 public support 25, 42, 91
policy option 2, 4, 5, 12, 13, 17, 21, 43, 46, 48, purely marginal and ancillary 8, 11, 129
165, 166, 171, 197, 209, 228, 235, 238, Pusa 80
241, 242, 244, 246
policymaking qualified majority voting 95
judicial 10, 93–5, 104, 106, 123 quality 50, 64, 144, 146
legislative 10, 19, 44, 93–125, 247, 250 quantitative analyses 30, 37, 43, 46, 52, 237
political realism 24 quantitative restriction 64, 74, 85
political salience 43, 98, 108, 122, 125, QWERTY 60, 60 n. 2, 63
174, 234
political science 4, 16, 17, 19, 20, 23, 24, 39, race to the bottom 63, 159, 213
52, 55, 114, 230, 233, 236, 248–51 Radaelli, Claudio M. 173, 197, 200
political mobilization 15, 195 rapporteur 30, 30 n. 2, 30 n. 3, 109, 110,

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

Pollack, Mark A. 31, 35 118, 119
positive feedback 52, 60, 61 n. 3, 62, 63, 67, Rasmussen, Hjalte 31
68, 69, 72, 76, 82, 83, 84, 88, 89, 90, 232 Rasmussen, Morten 34
posted workers 97, 108, 112, 122, 125, 151, ‘reactive’ actor 3, 19, 24, 44, 48, 169, 239
205, 206, 209, 210, 211, 214–16, 225 ‘real seat’ 73
Posted Workers Directive 97, 108, 112, 122, reasoning by analogy 14, 61, 69
151, 209, 211, 215, 216 redistribution 67, 122
precedent 7, 12, 14, 16, 26, 27, 28, 30, 33, 34, reform
51, 57, 61, 62, 63, 67, 69, 71, 80, 89, 90, 91, domestic 35, 219, 226, 230, 241
92, 237, 240 MoMiG 213
preferences national 201, 241
revealed 25 Regulation
preliminary-ruling procedure 6, 41, 54, 176 492/2011 75, 130
presidency 115, 118, 120, 146 883/2004 75, 77, 130, 131, 142 n. 10, 145,
president 30, 30 n. 2, 32, 151, 164 149, 205, 217
pressure 1408/71 75, 77, 129, 130, 142, 205,
active 99, 106 206 n. 5
external 62, 224 1612/68 75, 130, 132, 133
judicial 94, 154, 166 mutual recognition of goods 15, 19, 95,
principal-agent approach 3 114–16, 122, 123, 150, 167, 234, 235,
principal-agent theory 39, 40 239, 250
principle regulation
general 80, 124, 206, 228 equivalent 110, 117, 118
legal 19, 57, 80 regulatory competition 88, 172, 174, 202, 212,
new 34, 46 213, 226, 229
regulatory 9 regulatory control 121
Pringle 29, 38 regulatory requirement
primary law 11, 35, 48, 94, 114, 122, 152 equivalent 118
privilege 34, 78, 81, 98, 109, 174, 181, 184, Reich, Norbert 67
204, 216, 225, 234, 243 reimbursement 127, 141–5, 147–9, 203
problem solving 120 n. 21, 247 ‘repeat players’ 36, 140
process tracing 13, 17 residence
professional environment 27–34 habitual 193 n. 40, 217, 218, 218 n. 19, 218
proportionality n. 21, 223
analysis 58, 113, 237 requirements 217, 223
principle 58, 89, 123, 151 rights 21, 130, 131, 133, 134, 136–8, 174,
test 51, 58, 59, 66, 101, 105, 152, 208, 215 180, 181, 186, 187, 190, 194, 203–5,
public banks 198, 224, 228 227, 237
public health 111 n. 9, 113, 119, 142, 147, 148, resistance
149, 157 executive 206–9
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

292 Index
response Shapiro, Martin 1, 33, 57, 61
direct 21, 199, 200, 202, 235 Shatter, Alan 186, 189 n. 24
indirect 21, 86, 199, 200, 212, 229 Shelter 50, 137
responsiveness 195, 238 signal 9, 19, 42, 43, 57, 62, 63, 98, 111, 113,
restrictions 122, 135, 136, 143, 147, 168, 196, 232,
approach 19, 77, 82–5 234, 249
prohibition of 18, 19, 62, 64, 65, 67, 68, 75, signalling effect 98, 123
81, 82, 85, 90 single contact point 115
retained powers formula 167, 242 single market 14, 15, 19, 53, 67, 77, 87, 96, 97,
retrenchment 173, 174, 197 100, 105, 107, 115 n. 15, 116, 119, 130,
reverse discrimination against nationals 29, 132, 143, 144, 153, 156, 233, 248
65, 85–8, 109, 125, 147, 167, 182, 183, Slaughter, Anne-Marie 26, 27, 214
200, 204, 247 SMEs 120, 164
reversion 46, 68, 82, 97, 168, 238 Smith, Brendan 188, 188 n. 19
Rinner-Kühn 221 Smits/Peerbooms 143, 144, 249
Risse, Thomas 172–4, 200 social assistance 74, 81, 132–5, 137, 182, 193,

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

road haulage 101 218, 219, 223
Rosenberg, Gerald N. 25, 239 Social Democratic Party 207
Rüffert 173 n. 1, 215, 216, 219 social rights 50, 95, 139, 150, 223, 244, 245
rule of law 11, 12, 23, 25, 33, 34, 40, 41, 47, 48, social-security system 69, 128, 129, 143, 144
50, 54, 59, 61, 89, 94, 98, 170, 181, 195, societal actor 21, 43, 52, 198, 200, 202, 224–7,
202, 206, 220, 229, 243, 246–8 229, 236, 241, 242
rule-of-law fashion 21, 228, 235 societal self-regulation 112
rule of reason 66 SOLVIT 120
rules Somek, Alexander 79, 140
general 7, 19, 20, 21, 45, 46, 94, 96, 98, 108, Sotgiu 129
121, 123, 124, 146, 148, 219, 233, 235, sovereignty
238, 239 constitutional 220, 229
rulings parliamentary 5, 66, 113, 124, 220
series of 46, 215 Spanish strawberries 86
Rush Portuguesa 70, 97, 109 sports policy 105
stability 10, 14, 29, 51, 57, 59, 90
safety standard 146 Stamatelaki 144
Säger 71, 77 standardization 86, 87, 115, 126, 225
sanction 40 stare decisis 61
Sanneh 193 state organization 1, 50, 246
Schäfer, Armin 86, 90, 99, 106, 112, 127, 167, status of a worker 75
213, 245 status quo
Scharpf, Fritz W. 4, 5, 9, 17, 32, 33, 40, 41, 48, regulatory 45, 91, 173
84, 91, 93, 95, 114, 127, 138, 139, 169, Steindorff, Sylvia von 65, 72, 212
200, 245, 246, 248, 250 stimulus 199
Schelling, Thomas 62 Stone Sweet, Alec 3, 4 n. 4, 5, 6, 14, 23, 26, 27,
Schepel, Harm 31, 52, 87 33, 37, 40, 52, 53 n. 1, 54–9, 61, 66, 113,
scholarship 2, 11, 24, 54 196 n. 43, 198, 219–21, 239, 246
Schreinermacher, Björn 35 n. 4, 106, 107, 121, ‘stop the ECJ’ 32
122, 133–6, 223, 223 n. 24 Stranz, Kathleen 37, 196 n. 43, 198,
Schumacker 78 n. 8, 156, 167 219–21
self-interest 8, 56, 57, 59 Streeck, Wolfgang 5, 9, 10, 55, 125
selling arrangement 68, 83 Stubb, Alexander 118, 119
separation of powers 24, 27, 43 sufficient resources 129–31, 134, 140
Services Directive 15, 16, 19, 46, 72, 74, 95, 99, sui generis 10
104, 106–14, 112 n. 12, 118, 121, 122, supranational actor 34, 56, 57, 63, 212
124, 126, 141, 145, 150, 167, 215, 233, supranational constraints 247
241, 242, 244, 250 Supreme Court 12, 24, 38, 53, 58, 61, 73, 163,
shadow 16, 58, 93–100, 106, 112, 122, 186, 187, 190, 223
126, 231 Swedish labour court 222
 OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

Index 293
Tarrow, Sidney 127, 139, 167 trustee 3, 40
tax Tryfonidou, Alina 68, 79, 87, 131, 132, 204
authority 163, 176, 178, 179
competition 20, 74, 151, 159, 164–6, 168 Überseering 72, 73, 176
evasion 81, 164, 165, 165 n. 24, 210 ultra vires 38
identification number 206 unanimity 40, 47, 56, 95, 100, 153, 163
lawyer 175, 196 unitary system 36
neutrality 158 unity of the legal order 61
policy 15, 20, 78, 153–4, 158, 159, 161, 163, Universal Credit 218
165, 166 unreasonable burden 79, 132, 134, 217
taxation US
direct company 15, 127 different philosophy 53
double 154, 157 n. 16, 158, 162, 163, 176 Supreme Court 24, 53, 58, 61, 163
telecommunications 103 usurpation 32
teleological approach 61
teleological interpretation 232, 251 value added tax 153, 154

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025

telos 31, 47, 89 value pluralism 56
Thelen, Kathleen 9, 10, 55 Van Binsbergen 70
theory Vanberg, Georg 25, 42
rational choice 55 Vanbraekel 144, 146
third-country national (TCN) variable
family members 132, 182–4, 191, 203–5, independent 14, 199
207, 208, 212, 227, 237, 245 Vatsouras 136, 137
Thomson, Robert 6, 233 Vauchez, Antoine 9, 31, 36, 62
Thym, Daniel 192, 209 veil of ignorance 55
time 7, 27, 51, 61, 61 n. 3, 69, 88, 107, 173, Verkooijen 157 n. 17, 176–8, 180
175, 179, 185, 190, 195, 196, 205, 227, veto point 29, 32, 59
236, 240, 242 Viking 15, 19, 32, 42, 86, 112, 112 n. 12, 113,
time horizons 240 150–3, 168, 211, 215, 219, 222, 225
Tizzano 177, 178, 187 Vogel-Polsky, Elaine 36
trade 18, 36, 52, 53, 54, 64, 65, 71, 75, 86, voice 32, 219, 239
87, 90, 107, 110, 110 n. 6, 111, 111 n. 10, Volkswagen 83, 87
112, 115, 120, 122, 125, 128, 204, 215, Volkswagen law 106
233, 247 voter preference 47
transborder conflict 52, 54
transborder trade 107 wage
transformation 6, 31, 173, 174, 197, 200 minimum 71, 109, 129, 209, 210, 214,
transitional period 179 219, 230
translation 73 Walrave 87
transposition 11, 108, 140, 171, 172, 179, 197, Warren Jr., Alvin C. 45, 157–9, 157 n. 16,
198, 202, 204, 226 158 n. 18, 163, 180
Treaty Wasserfallen, Fabio 134, 135, 153, 170, 173, 181
constitutionalized 6, 9, 13, 51, 58, 123 Watts 32, 144
constitutionalizing 8, 15 Weigel 75, 76
masters of the 5, 18, 28, 39–47, 158 Weiler, Joseph 1, 6, 29, 36–9, 93
revision 11, 48, 242 welfare migration 81, 205, 206
Treaty Establishing the European Community welfare state 13, 16, 20, 60, 67, 79, 126, 135,
(TEC) 1, 15, 64, 65, 79, 80, 86, 97, 112, 139, 140, 169, 232, 234, 245
114, 116, 117, 120, 130, 131, 142, 178, 222 Wesseling, Rein 31
treaty negotiation West LB 225
court 40 whistle blowing 35
Treaty on the Functioning of the European wholly internal 53
Union (TFEU) 1, 6, 28, 34, 41, 70–2, 74, Wiesbrock, Anja 72, 138, 243
77, 78, 85–7, 102, 128, 137, 138, 142, 147, win-set 5, 20, 250
178, 183, 222 Wind, Marlene 37, 38, 79, 81, 131, 132, 134,
Treib, Oliver 172, 198, 199, 235 137, 203, 230
OUP CORRECTED PROOF – FINAL, 5/12/2017, SPi

294 Index
Wollenschläger, Ferdinand 131, 138 X Holding 162, 162 n. 20
worker
definition 129, 216 Ymeraga 184
interest 128, 225
mobility 14 Zambrano
working group 115 n. 18, 118, 132 carers 193
working time 210, 221 quantitative implications 187
Working Time Directive 44, 45 Zhu Chen 187
worst-case scenario 99, 100, 103, 173, 175, 241 zone of discretion 56, 57

Downloaded from https://academic.oup.com/book/5362 by guest on 01 May 2025