The Routledge Handbook of Differentiation in the European Union 2021048438, 2021048439, 9780367149659, 9781032183824, 9780429054136

Citation preview

“Finally! Everything you ever wanted to know about differentiated integration, and then more. Wish I had this comprehensive handbook when I was writing my PhD on the subject in the 1990s. A wealth of well-structured, clear and accessible information on a complex subject”. Alexander Stubb, Professor and Director, EUI, Florence; Former Prime Minister of Finland “Differentiation is here to stay, and at the latest Brexit has made it clear that differentiation in Europe encompasses both integration and disintegration processes. This is the core message of this timely handbook, which shows the breadth of this phenomenon in detail thanks to an impressive list of contributors. It is an excellent go-​to resource for all scholars of European Studies seeking to understand the challenges of ‘unity in diversity’ ”. Sieglinde Gstöhl, Director of Studies, College of Europe, Belgium “This handbook is an invaluable resource on differentiation—​involving both integration and disintegration—​in the European Union. Its stellar list of contributors offers a plethora of insights into the many different ways in which the EU has not only differentially integrated with regard to institutions, policy areas and territory, but also dis​integrated, as in the case of Brexit. A must-​ read for all those concerned with the ins-​and-​outs of EU integration”. Vivien Schmidt, Jean Monnet Professor, Boston University, United States

The Routledge Handbook of Differentiation in the European Union

The Routledge Handbook of Differentiation in the European Union offers an essential collection of groundbreaking chapters reflecting on the causes and consequences of this complex phenomenon. With contributions from key experts in this subfield of European Studies, it will become a key volume used for those interested in learning the nuts and bolts of differentiation as a mechanism of (dis)integration in the European Union, especially in the light of Brexit. Organized around five key themes, it offers an authoritative “encyclopaedia” of differentiation and addresses questions such as: • • •

How can one define differentiation in the European Union in the light of the most recent events? Does differentiation create more challenges or opportunities for the European Union? Is Europe moving away from an “ever closer Union” and heading towards an “ever more differentiated Union”, especially as leading political figures across Europe favour the use of differentiation to reconcile divergences between member states?

This handbook is essential reading and an authoritative reference for scholars, students, researchers and practitioners involved in, and actively concerned about, research in the study of European integration. As European differentiation is multifaceted and involves a wide range of actors and policies, it will be of further interest to those working on countries and/​or in policy areas where differentiation is an increasingly relevant feature. Benjamin Leruth is an Assistant Professor in European Politics and Society at the University of Groningen, the Netherlands. Stefan Gänzle is Professor of Political Science and Head of the Department of Political Science and Management and a member of the Jean Monnet Centre of Excellence, University of Agder (UiA), Norway. Jarle Trondal is a Professor at the University of Agder, Department of Political Science and Management, and a Professor at the University of Oslo, ARENA Centre for European Studies, Norway.

The Routledge Handbook of Differentiation in the European Union

Edited by Benjamin Leruth, Stefan Gänzle and Jarle Trondal

Cover image: © Shutterstock First published 2022 by Routledge 4 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 605 Third Avenue, New York, NY 10158 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2022 selection and editorial matter, Benjamin Leruth, Stefan Gänzle and Jarle Trondal; individual chapters, the contributors The right of Benjamin Leruth, Stefan Gänzle and Jarle Trondal to be identified as the authors of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. With the exception of the Introduction and chapters 13, 21, 30, and 35, no part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. The Introduction and chapters 13, 21, 30, and 35 of this book is available for free in PDF format as Open Access from the individual product page at www.routledge.com. It has been made available under a Creative Commons Attribution-Non Commercial-No Derivatives 4.0 license. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing-​in-​Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-​in-​Publication Data Names: Leruth, Benjamin, editor. | Gänzle, Stefan, 1970– editor. | Trondal, Jarle, editor. Title: The Routledge handbook of differentiation in the European Union / edited by Benjamin Leruth, Stefan Gänzle and Jarle Trondal. Other titles: Handbook of differentiation in the European Union Description: First Edition. | New York : Routledge, 2022. | Series: Routledge International Handbooks | Includes bibliographical references and index. Identifiers: LCCN 2021048438 (print) | LCCN 2021048439 (ebook) | ISBN 9780367149659 (Hardback) | ISBN 9781032183824 (Paperback) | ISBN 9780429054136 (eBook) Subjects: LCSH: Europe–Economic integration–Political aspects. | Political participation–European Union countries. | International organization. ​ ​ ​ ​ Classification: ​ ​ LCC​ JN15 ​ .R688 2022 (print) | LCC JN15 ​ (ebook) ​ ​ | DDC ​ 341.242/2–dc23/eng/20220106 LC record available at https://lccn.loc.gov/2021048438 ​ available at https://lccn.loc.gov/2021048439 LC ebook record ISBN: 978-0-367-14965-9 (hbk) ISBN: 978-1-032-18382-4 (pbk) ISBN: 978-0-429-05413-6 (ebk) DOI: 10.4324/9780429054136 Typeset in Bembo by Newgen Publishing UK

Contents

List of figures List of tables Notes on contributors Preface and acknowledgements 1 Introduction: differentiation in the European Union as a field of study Benjamin Leruth, Stefan Gänzle and Jarle Trondal

xii xiii xiv xix 1

PART 1

Conceptualizing differentiation

17

Introduction to Part 1 Jarle Trondal, Benjamin Leruth and Stefan Gänzle 2 Studying differentiated integration: methods and data Marian Burk and Dirk Leuffen

23

3 On the legitimacy of differentiated integration Erik O. Eriksen

34

4 Differentiated (dis)integration beyond Europe: a comparative regionalism approach Stefan Gänzle and Jens Uwe Wunderlich

50

5 De facto differentiation in the European Union: circumventing rules, law, and rule of law Tobias C. Hofelich

66

6 Constitutive differentiation John Erik Fossum

81

vii

Contents

7 EU external differentiated integration and compliance: theoretical and legal aspects Maryna Rabinovych and Anne Pintsch 8 From integration to fragmegration: political symbols and the emergence of differentiated European identities Russell Foster 9 Differentiation and segmentation Jozef Bátora and John Erik Fossum

99

116 132

PART 2

Institutionally based differentiation

149

Introduction to Part 2 Jarle Trondal, Stefan Gänzle and Benjamin Leruth 10 An ever more fragmented Union? On the emerging relevance of the institutional implications of differentiated integration in the EU Alexander Schilin

155

11 Promise unfulfilled? Managing differentiated integration in EU secondary law through enhanced cooperation Daniela Kroll

170

12 Differentiation in the European Parliament: united about diversity? Guri Rosén 13 Differentiation and the European Central Bank: a bulwark against (differentiated) disintegration? Daniel F. Schulz and Amy Verdun

184

200

14 Differentiation and the European Commission Diane Fromage and Cristina Fasone

216

15 Differentiation and the European Court of Justice Sabine Saurugger and Fabien Terpan

231

16 Third-​country participation in EU agencies: towards “condominio”? Sandra Lavenex

248

17 The Council of the European Union: organizational and social dynamics of differentiation Jeffrey Lewis viii

264

Contents

PART 3

Policy-​based differentiation

281

Introduction to Part 3 Stefan Gänzle, Benjamin Leruth, and Jarle Trondal 18 Differentiated integration in EU Energy Market Policy Torbjørg Jevnaker

289

19 Brexit and the Common Fisheries Policy: opportunities for multi-​level differentiated (dis)integration? Arno van der Zwet, John Connolly, Christopher Huggins and Craig McAngus

310

20 The European Asylum Policy: core state powers, flexibility and differentiated integration Foteini Asderaki and Eleftheria Markozani

325

21 Differentiated integration in EU climate policy Elin Lerum Boasson, Merethe Dotterud Leiren and Jørgen Wettestad

340

22 Differentiated integration in European external action Thomas Henökl

355

23 PESCO: a formula for positive integration in European defence Steven Blockmans and Dylan Macchiarini Crosson

370

24 The increasingly differentiated European Single Market? David Howarth

391

25 Differentiation and social policy: a sustainable way forward? Benjamin Leruth and Sven Schreurs

408

26 Differentiation in EU security and defence policy Elie Perot and Stephan Klose

426

PART 4

Territorial differentiation

441

Introduction to Part 4 Benjamin Leruth, Stefan Gänzle and Jarle Trondal 27 The Nordic countries as pioneers of differentiation Benjamin Leruth and Jarle Trondal

449

ix

Contents

28 Risky advantageous differentiation: Iceland and the EEA Baldur Thorhallsson

464

29 The Swiss and Liechtenstein relations with the EU: an ongoing institutional challenge Christian Frommelt

482

30 Turkey’s external differentiated integration with the EU in the field of migration governance: the case of border management Ebru Turhan and Ayselin Yildiz

502

31 Visegrád Four and EU differentiated integration: activities, perception and self-​perception after the refugee crisis Petr Kaniok,Vratislav Havlík and Veronika Zapletalová

519

32 Poland as the (new) awkward partner: differentiated integration or differentiated disintegration? Agnieszka K. Cianciara

537

33 Trajectories of differentiated EU integration for the Western Balkans Marko Milenković

551

34 European Neighbourhood Policy: differentiated integration beyond the EU’s Eastern and Southern borders Zuzana Reptova

565

35 Differentiation at the local level: an overview of subnational authority networks in the EU Pier Domenico Tortola and Stefan Couperus

581

PART 5

Brexit

597

Introduction to Part 5 Stefan Gänzle, Jarle Trondal and Benjamin Leruth 36 Brexit as a phenomenon: national solidarity as a tool against the European project? Mikko Kuisma and Matthew Donoghue

605

37 (Post-​)Brexit: negotiating differentiated disintegration in the European Union Frank Schimmelfennig

619

x

Contents

38 International perceptions of Brexit Johanna Speyer, Natalia Chaban and Arne Niemann 39 Differentiation and power asymmetry: how Brexit is changing UK relations with Czechia and Slovakia Monika Brusenbauch Meislová and Andrew Glencross 40 Brexit and Northern Ireland David Phinnemore

633

649 664

41 Border conflicts and territorial differentiation after Brexit: the cases of Northern Ireland, Gibraltar and the UK Sovereign Base Areas in Cyprus Nikos Skoutaris

680

42 Growing apart together? Brexit and the dynamics of differentiated disintegration in security and defence Benjamin Martill and Monika Sus

696

43 Conclusion Benjamin Leruth, Stefan Gänzle and Jarle Trondal Epilogue: polycrisis and resilience in the European Union: Covid-​19 and avenues for future studies Marianne Riddervold, Akasemi Newsome and Jarle Trondal Index

712

722

735

xi

Figures

2.1 Total number of publications on DI from 1995 to 2020, as generated by a keyword search on ‘differentiated integration’ using the database ‘dimensions.ai’ 2.2 Total number of publications on DI over time, sorted by methodological approach. Own classification on ‘dimensions.ai’ ‘differentiated integration’ sample (1995–​2020) 2.3 Share of methodological approaches in DI publications (period 2007–​2020). Own classification on ‘dimensions.ai’ ‘differentiated integration’ sample 11.1 Enhanced cooperation in the first and third pillar in the Lisbon Treaty 11.2 Enhanced cooperation in the second pillar in the Lisbon Treaty 19.1 The internal borders of the UK’s Exclusive Economic Zone 25.1 Percentage of support for exclusive domestic competence in the field of health and welfare policy, per experience of horizontal differentiation 25.2 Percentage of support for exclusive domestic competence in the field of health and welfare policy, per region 26.1 The concept tree of differentiated cooperation vs differentiated integration 26.2 The concept tree of negative vs positive forms of differentiated cooperation 35.1 European subnational authority networks by decade of establishment 35.2 Thematic breakdown of subnational authority networks 35.3 Subnational authority network participation by country 35.4 EU vs. non-​EU membership of subnational authority networks

xii

26

26 27 173 175 315 419 420 428 428 583 584 588 590

Tables

1.1 Summary of key concepts 4.1 Scales of regionness 10.1 New governance facilities in EMU and how EU members outside the currency area are represented within them 10.2 The proposed dimensions for analysing the inclusiveness of EU organisations 11.1 Overview of the authorized cases of enhanced cooperation 16.1 EU Agencies, year of creation, function and legislative power 18.1 EU energy market laws 18.2 Opt-outs in the electricity directive (directive 2009/72/EC – recast as directive (EU) 2019/944) 18.3 Opt-outs in the electricity regulation (regulation (EC) no 714/2009 – recast as regulation (EU) 2019/ 943) 18.4 Opt-outs in the gas directive and the gas regulation 18.5 Opt-outs in other energy market laws 18.6 EEA participation in the third energy market package of 2009 (EEA, 2020) 19.1 Share of fishing industry per nation (2019) 21.1 A dimension of differentiated integration and operationalization of differentiation in the three policy stages 21.2 Categorizing differentiated integration of EU ETS over time 21.3 Categorizing differentiated integration of EU renewables policy and practices 23.1 Levels of differentiation in PESCO 23.2 European defence technological and industrial base and defence expenditures of PESCO-​participating member states 23.3 Strategic cultures within PESCO-​involved EU member states 29.1 Second-​order differentiation in the EEA 29.2 Explanatory factors for Liechtenstein’s differentiated integration in the EEA 31.1 Dimensions of differentiated integration and V4 countries’ activities/​perceptions 34.1 Expansion of the economic community: how ENP partners adopt EU rules 34.2 Differences in the terms of the EU’s offer for inclusion in the expanding economic community over time 35.1 Subnational authority network dataset –​summary codebook 38.1 Overview of global EU perceptions after Brexit

15 54 159 165 176 253 293 295 298 299 301 302 316 343 346 349 371 373 380 494 496 523 571 573 596 642

xiii

Contributors

Foteini Asderaki is an Associate Professor of European Integration and Jean Monnet Chair on

European Union’s Education, Training, Research and Innovation Policies at the Department of International and European Studies at the University of Piraeus. Jozef Bátora is a Professor at the Webster Vienna Private University and a Professor at the

Comenius University, Bratislava. Steven Blockmans is a Professor of EU External Relations Law and Governance at the University

of Amsterdam, Director of Research at the Centre for European Policy Studies (CEPS) and Editor-​in-​Chief of the European Foreign Affairs Review. Elin Lerum Boasson is an Associate Professor at the Department of Political Science at the

University of Oslo and Senior Researcher at the CICERO Center for International Climate Research in Oslo. Marian Burk is a Research Fellow at the German Stifterverband für die deutsche Wissenschaft.

He holds an MA from the European Master in Government program at the University of Konstanz and the Pompeu Fabra University, Barcelona. Natalia Chaban is Professor and Jean Monnet Chair at the Department of Media and

Communications at the University of Canterbury, New Zealand, President of the Ukrainian Studies Association of Australia and New Zealand and Co-​Editor of the peer-​reviewed Australia and New Zealand Journal of European Studies. Agnieszka K. Cianciara is an Associate Professor at the Institute of Political Studies, Polish

Academy of Sciences Warsaw. John Connolly is a Professor of Public Policy at the School of Education and Social Sciences at

the University of the West of Scotland and Co-​Editor-​in-​Chief of Contemporary Social Science. Stefan Couperus is an Associate Professor of European Politics and Society at the University of

Groningen. Dylan Macchiarini Crosson is Research Assistant at the Centre for European Policy Studies

(CEPS) and recently completed his Master’s in European Affairs at Sciences Po (Paris).

xiv

Contributors

Matthew Donoghue is an Assistant Professor and Ad Astra Fellow in Social Policy at the

University College Dublin. Erik O. Eriksen is a Professor at the ARENA Centre for European Studies at the University

of Oslo. Cristina Fasone is an Assistant Professor of Comparative Public Law at LUISS Guido Carli

University, Rome. John Erik Fossum is a Professor at the ARENA –​Centre for European Studies at the University

of Oslo. Russell Foster is a Lecturer in British and European Politics Education at King’s College London. Diane Fromage is a Marie Sklodowska-Curie Individual Fellow at Sciences Po Paris. Christian Frommelt is Director and Head of Research Politics at the Liechtenstein institute. Stefan Gänzle is a Professor of Political Science and Head of the Department of Political

Science and Management and a member of the Jean Monnet Centre of Excellence, University of Agder (UiA). Andrew Glencross is an Associate Professor of Political Science and Deputy Director for

International Affairs at the European School of Political and Social Sciences, Université Catholique de Lille. Vratislav Havlík is an Associate Professor at the Department of International Relations and

European Studies at Masaryk University. Thomas Henökl is an Associate Professor of Public Policy at the University of Agder in

Kristiansand, Norway, and Senior Research Associate at the German Development Institute in Bonn. Tobias C. Hofelich is a PhD Research Fellow at the Department of Political Science and

Management at the University of Agder, Norway. David Howarth is a Professor in Political Science at the University of Luxembourg and one of

three editors of the Routledge/​UACES book series on Contemporary European Studies. Christopher Huggins is an Associate Dean for Learning, Teaching and Student Experience and

Associate Professor in Politics at the University of Suffolk. Torbjørg Jevnaker is a PhD candidate at the University of Oslo and Research Fellow at the

Fridtjof Nansen Institute. Petr Kaniok is an Associate Professor at the Department of International Relations and European

Studies at Masaryk University.

xv

Contributors

Stephan Klose is a Postdoctoral scholar at the Brussels School of Governance (BSoG) at the

Vrije Universiteit Brussel (VUB). Daniela Kroll is responsible for inter parliamentary cooperation and monitoring of opinions of

the Bundestag on EU Matters. Mikko Kuisma is Program Lead and Lecturer in Comparative Public Policy at the University of

Tübingen. Sandra Lavenex is a Professor of European and International Politics at the University of

Geneva. Merethe Dotterud Leiren is a Research Director at the CICERO Center for International

Climate Research in Oslo. Benjamin Leruth is an Assistant Professor in European Politics and Society at the University of

Groningen and editor of the journal Acta Politica. Dirk Leuffen is a professor of International Politics at the Department of Politics and Public

Administration, University of Konstanz. He is also a principal investigator at the Cluster of Excellence on “The Politics of Inequality”. Jeffrey Lewis is a Professor in the Department of Political Science at Cleveland State University. Eleftheria Markozani is a PhD candidate and Researcher at the Department of International

and European Studies at the University of Piraeus. Benjamin Martill is a PhD and Lecturer in Politics and International Relations at the University

of Edinburgh, School of Social and Political Science. Craig McAngus was Lecturer in Politics at the School of Education and Social Sciences at the

University of the West of Scotland. He passed away on 16 December 2020 at only 34 years of age. He will be remembered as an excellent scholar and colleague. Monika Brusenbauch Meislová serves as an Assistant Professor at the Department of International

Relations and European Studies at Masaryk University in Brno. Marko Milenković is a Research Fellow at the Institute of Social Sciences Belgrade; Affiliate

researcher –​Center for Constitutional Studies and Democratic Development (CCSDD), School of Advanced International Studies (SAIS) Bologna. Akasemi Newsome is Associate Director at the Institute of European Studies and Executive

Director at the Peder Sather Center for Advanced Study, University of California, Berkeley, and Senior Fellow at the Inland Norway University of Applied Sciences. Arne Niemann is Professor of International Politics at the Department of Political Science,

Johannes Gutenberg University Mainz, Germany.

xvi

Contributors

Elie Perot is a PhD candidate at the Brussels School of Governance (BSoG) at the Vrije

Universiteit Brussel (VUB). David Phinnemore is a Professor of European Politics and a Dean of Education at Queen’s

University Belfast and a visiting Professor at the College of Europe in Bruges. Anne Pintsch is an Associate Professor at the Department of Political Science and Management

at the University of Agder. Maryna Rabinovych is an Assistant Professor at the Public Policy and Governance Department

of Kyiv School of Economics. Zuzana Reptova (Novakova) is a PhD Researcher at the Erasmus University Rotterdam. Marianne Riddervold is a Professor of Political Science/​International Relations at the Inland

School of Economics and Social Sciences at the Inland Norway University of Applied Sciences and at the Norwegian Institute of Foreign Affairs. Guri Rosén is an Associate Professor at Oslo Metropolitan University and Senior Researcher at

ARENA Centre for European Studies. Sabine Saurugger is a Professor of Political Science and Director of Sciences Po Grenoble. Alexander Schilin is a PhD candidate at the Institute of Political Science at the University of

Leiden. Frank Schimmelfennig is a Professor of European Politics, member of the Centre for

Comparative and International Studies, and Director of Studies of the BA program in Public Policy at ETH Zurich. Sven Schreurs is a PhD student at the European University Institute. Daniel F. Schulz is a Postdoctoral Research Fellow in European Studies at the University

of Agder. Nikos Skoutaris is an Associate Professor in European Union Law at the University of East

Anglia. Johanna Speyer is a PhD candidate at the Chair of International Politics, Johannes Gutenberg

University Mainz, Germany. Monika Sus is PhD and Research Fellow at the Centre for International Security at Hertie

School, Berlin, and Associate Professor at the Polish Academy of Sciences in Warsaw. Fabien Terpan is an Associate Professor (EU Law) and Jean Monnet Chair –​Sciences Po

Grenoble.

xvii

Contributors

Baldur Thorhallsson is a Professor of Political Science and Jean Monnet Chair in European

Studies at the University of Iceland. Pier Domenico Tortola is an Assistant Professor in European Politics and Society at the

University of Groningen. Jarle Trondal is a Professor at University of Agder, Department of Political Science and

Management and a Professor at University of Oslo, ARENA Centre for European Studies. Ebru Turhan is an Assistant Professor at the Department of Political Science and International

Relations, Turkish-​German University, Istanbul. Jørgen Wettestad is a Research Professor at the Fridtjof Nansen Institute in Oslo. Jens Uwe Wunderlich is a Lecturer in International Relations at the School of Social Sciences

and Humanities at Aston University in Birmingham. Arno van der Zwet is Senior Lecturer in Politics and Public Policy at the School of Education

and Social Sciences at the University of the West of Scotland. Amy Verdun is a Professor of Political Science and the founding Director of the European

Studies Program at the University of Victoria (UVic). Ayselin Yıldız is an Associate Professor and UNESCO Chair on International Migration at the

Department of International Relations,Yasar University. Veronika Zapletalová is an Assistant Professor at the Department of International Relations and

European Studies at Masaryk University.

xviii

Preface and acknowledgements

The Handbook is the result of over three years of collaboration between over 60 researchers working on the issue of differentiation in the European Union. This collaboration started in 2017 with the establishment of a formal Research Network on “Differentiated Integration and Disintegration in a post-​Brexit Europe”, funded by the University Association for Contemporary European Studies (UACES). However, the project itself has evolved over even some more time; it began at the UACES conference in Bilbao, 7–​9 September 2015, when the future network convenors, Benjamin Leruth, Stefan Gänzle and Jarle Trondal, met to discuss potential avenues for research on Europe’s increasingly differentiated political order. Just a few days before the Bilbao conference, the German government decided late in the night of 4 September to welcome thousands of refugees that were stuck in Hungary to Germany. It was a strong decision on a matter that continues to divide EU member states today. The decision by the German chancellor Angela Merkel also played a major role in setting the stage of the United Kingdom’s (UK) popular referendum on the country’s membership in the EU. When the British people voted, by a small margin, to leave the Union on 23 June 2016, an unprecedented withdrawal process started.The referendum eventually led to the UK’s exit agreed on 31 December 2020, ending the country’s European Union membership after almost fifty years. In early 2020, this cumbersome process got even more complex due to a virus that spread all over the globe: Covid-​19. In the midst of these extremely challenging times for all of us due to Covid-​19, we eventually finalized our Handbook. We know that some of our contributors had to meet some hardship during that time. We are therefore grateful to all of them for putting quite some effort in producing excellent chapters, and we would like to thank them once again for their patience, resilience and dedication to this ambitious project. This Handbook would not have been made possible without generous financial support. Funding and assistance were provided by UACES as well as the European Commission via the UiA Jean Monnet Chair, which Stefan Gänzle held from 2017 to 2020, as well as the University of Agder’s support scheme for international conferences, workshops and funding of PhD fellows who assisted us in the production of the Handbook. Jarle Trondal also acknowledges funding by the EU’s Horizon 2020 project “EU differentiation, domination and democracy (EU3D)”. We are very grateful to PhD students Frans af Malmborg, Rukia Pazi and Tobias C. Hofelich (University of Agder) for their assistance throughout this project, to Andrew Taylor (Routledge) for his trust and to Sophie Iddamalgoda (Routledge) for her invaluable advice.

xix

newgenprepdf

Preface and acknowledgements

We dedicate this volume to the memory of one of our contributors, Craig McAngus (1986–​2020). We encourage our readers to look at the obituary written by his friends and co-​authors Arno van der Zwet and John Connolly: www.psa.ac.uk/​psa/​news/​memoriam-​drcraig-​mcangus. Benjamin Leruth, Stefan Gänzle and Jarle Trondal Groningen and Kristiansand, 30 June 2021

xx

1 Introduction Differentiation in the European Union as a field of study Benjamin Leruth, Stefan Gänzle and Jarle Trondal

Introduction On 24 June 2016, the European Union (EU) woke up facing a situation which was as much unprecedented as unexpected. For the first time in the history of European integration, citizens of a member state had cast a vote in favour of leaving the Union. The outcome of the popular referendum held in the United Kingdom (UK) sent shockwaves across the continent and the world. At that time the EU found itself centre stage of a series of challenges, including the Eurozone crisis, which had followed the global financial and economic crisis of 2007–​2008, the Ukrainian crisis of 2013–​2014 and so-​called refugee crisis of 2015. These crises have subsequently been complemented by the Covid-​19 pandemic challenging the EU since early 2020. Thus, the EU is likely to remain prone of what has accurately been coined a ‘polycrisis’ (Zeitlin 2016); in addition, throughout the 2010s, the EU has become increasingly exposed to high levels of public and party-​based Euroscepticism in many of its member states (see, e.g., Leruth et al. 2018). Rather than breaking down, previous studies suggest that the EU has become resilient to crises owing to its ability to adapt and absorb, and if necessary, muddling-​through (Riddervold et al. 2021). The Brexit vote resulted in a lengthy and cumbersome withdrawal process that came to a close with the so-​called transition period elapsing on 31 December 2020.Therefore, Brexit, in itself, is best understood as a process rather than a single event that is transforming the EU. In retrospect, Brexit was not fully unexpectable either; at least not for those who had been questioning to what extent ‘Britain was European’ (Ash 2001) for a while. Ever since its successful bid for an exclusive reduction of its share to the Community budget –​the famous ‘British rebate’ –​in the mid-​1980s the UK nurtured a reputation of being an ‘awkward partner’ (George 1998) in Europe.The state’s attitude was in line with some of its long-​standing foreign policy traditions which set the UK somewhat ‘apart’ –​underpinning its mutually reinforcing insular and ‘splendid’ isolation: British exceptionalism and differentiation from continental Europe –​‘where the weather comes from’ as allegedly put by Winston Churchill –​have, in the aftermath of the Treaty of Maastricht, been underwritten by opt-​outs covering policy areas as economic and monetary union (EMU) as well as justice and home affairs. Second, the EU and its member states themselves had come a long

DOI: 10.4324/9780429054136-1

1

Benjamin Leruth, Stefan Gänzle and Jarle Trondal

way in recognizing that the core idea of its political order ultimately revolved around the reconciliation of two –​at first sight –​dichotomous principles, namely ‘unity in diversity’. This principle paralleled one of the EU’s strongest original objectives which consisted in the realization of ‘ever closer union’. Controversial questions surrounding member states’ rights to withdraw from the EU were legally settled through the ratification of the Lisbon Treaty in 2009 through what is now Article 50 of the Treaty on EU. Article 50 describes the procedural steps a member state is obliged to take in order to lawfully withdraw from EU membership (European Parliamentary Research Service 2016).Yet, only a few observers at that time would have guessed that this article was to become a point of reference in less than a decade already. The unpredictable character of Brexit has led scholars to rethink its implications for the future of the EU. Some studies have focused on particular policy areas, such as trade (e.g. Dhingra et al. 2016a, 2016b), environment (e.g. Burns et al. 2019), climate change (e.g. Hepburn and Teytelboym 2017), labour market (e.g. Fagan and Rubery 2018) and foreign and security policy (e.g. Duke 2019; Martill and Sus 2018). Other studies have focused more broadly on implications of Brexit on the future of European integration (e.g. Rosamond 2016; Jones 2018; Cardwell 2019). In this vein, the European Commission (EC) ignited scholarly interest through its response to Brexit: in the ‘White Paper on the Future of Europe’ in 2017, the Commission sketched out several broad scenarios for the EU’s way ahead. A total of five scenarios were presented: ‘1: Carrying on’, i.e. following the existent path of muddling through without any major changes and reforms; ‘2: Nothing but the Single Market’ excluding areas such as migration, security and defence; ‘3: Those who want more do more’ based on coalitions of the willing; ‘4: Doing less more efficiently’ with a strong focus on further market integration leaving non-​ market-​related affairs aside, and, eventually, ‘5: Doing much more together’ across a wide range of areas (European Commission 2017: 15–​25).The importance of the White Paper, as we argued elsewhere (Gänzle et al. 2019), does not lie so much in capturing each scenario per se and in isolation, but in the remarkable fact that there is a choice for scenarios at the detriment of a single grand vision as well as the nature of these scenarios. The White Paper meticulously avoids references to the term of differentiation and carefully maintains that ‘the starting point for each scenario is that the 27 Member States move forward together as a Union’ (European Commission 2017: 15) based on the unity of the single market. Differentiation is implicitly present, when calling for further cooperation where ‘a group of countries, including the euro area and possibly a few others, chooses to work much closer notably on taxation and social matters’ (European Commission 2017: 20) using, for instance, the legal mechanism of enhanced cooperation more actively. Two scenarios, in turn, call for a ‘spill-​back’ in several policy areas, ‘such as regional development, public health, or parts of employment and social policy not directly related to the functioning of the single market’ (European Commission 2017: 22). These differ from state-​based ‘opt-​outs’ and are best conceived of as varieties of disintegration. Therefore, the scenarios ultimately flesh a wide range of differentiation encompassing both integrationist and disintegrationist processes and strategies (see Gänzle et al. 2019). In this sense, Brexit has revived the debate over a well-​known yet often misunderstood concept: differentiation as an attribute to integration. Developed in the 1970s and gaining momentum with the ratifications of the Maastricht and Amsterdam Treaties in the 1990s, scholars have increasingly paid attention to the causes and consequences of differentiated integration (see, e.g., Holzinger and Schimmelfennig 2012; Leruth et al. 2019a for literature overviews). The core objective of this Handbook is therefore to demonstrate that differentiation in the EU has become a persistent phenomenon and should therefore be considered as a systemic feature of European integration. The establishment of the Eurozone in the late 1990s as well as the

2

Introduction

subsequent ‘big bang enlargement’ of 2004 constitute the EU’s most far-​reaching initiatives in terms of integration led to an increase in the use of flexibility mechanisms. By the early 2010s, less than half of EU’s 18 major policy areas still applied uniform integration (Leuffen et al. 2013). Although differentiation has been understood as the exception to the rule (as the favoured approach fostered among EU institutions to apply EU policies uniformly across member states), it is arguably a core and structural part of the European integration project. In sum, after almost 70 years of deepening and widening processes, the core dependent variable in European integration studies has shifted from integration to differentiation –​to put forth the main proposition and ambition of this volume. Despite a vast literature, scholars still struggle to come to grips with the full consequences on differentiation in the EU (Leruth et al. 2019b). By bringing together over 50 leading and early career scholars from different disciplines, this Handbook demonstrates the breadth and depth in the study of differentiation in the EU, and the diverging approaches taken to understand the phenomenon. The introduction to this volume sets the stage for the subsequent chapters. It starts by offering a brief review of the existing literature and a deliberatively loose notion of differentiation, as scholars interpret and apply the notion in various ways.We then expand on the mechanisms that foster differentiation drawing on the supply and demand models developed by Schimmelfennig and Winzen (2020). The structure of the Handbook is then presented. As the volume covers a wide range of mechanisms of differentiated integration and disintegration, this chapter also concludes with a glossary or ‘memo’ which may help the reader to understand the differences and nuances between key concepts used throughout the volume. Before proceeding, a common understanding of the notion of ‘integration’ is needed, as it has not often been provided in the literature. The meaning of the term ‘integration’ varies across theoretical perspectives in literature and will subsequently vary across the chapters in this volume. Overall, we choose a less attended and general definition of integration suggested by James G. March (1999: 134) who sees integration as the imagination of ‘a world consisting of a set of parts. At the least, integration is gauged by some measure of the density, intensity, and character of relations among the elements of that set’. Subsequently, he suggests three parameters for integration: consistency among the parts, interdependence among the parts and structural connectedness among the parts. On this basis, disintegration would imply a lower degree of density and intensity of the consistency, interdependence and structural connectedness among these parts.

From studying ‘integration’ to ‘differentiated integration’ to ‘differentiation’ –​a cursory review of literature1 Differentiation, as discussed earlier, is not a new phenomenon in European integration. What is relatively new, however, is that it has become more conventional to include both integration and disintegration as possible variants of differentiation (Schimmelfennig 2018; Gänzle et al. 2019; Leruth et al. 2019a). Moreover, differentiation can take various forms which are often intertwined (see, e.g., Stubb 1996; Dyson and Sepos 2010). This Handbook thus suggests that differentiation constitutes a ‘normal’ state of affairs because of the Union’s institutional architecture and character as a composite polity. The early literature on European integration did not include differentiation conceptually to make sense of the integration process. At the beginning of the European integration project in the aftermath of World War II –​when the three Benelux countries, France, Italy and Western Germany, established the European Coal and Steel Community (ECSC) in 1952 –​policymakers

3

Benjamin Leruth, Stefan Gänzle and Jarle Trondal

emphasized the need for unity in Europe. This general attitude was further exacerbated by the role both the United States and the Soviet Union assumed as ‘external federators’ vis-​à-​vis the ECSC and, later, the European Economic Community (1957). In areas where ECSC members were unable to eventually agree and proceed jointly, such as in the proposed European Defence Community (EDC) of 1954, the option of choice was not to engage in forms of differentiated integration in the first place, but to ‘transfer’ to other international organizations. In the area of ‘hard politics’, one could argue the North Atlantic Treaty Organization (NATO) ultimately accounts for ‘European Security’ in structural terms until to date –​despite the advent of the EU Common Foreign and Security and its variants, including, most recently, enhanced cooperation on defence in the frame of Permanent Structured Cooperation (PESCO) in 2017. European integration initially remained centred around a set of policy areas where a small number of founding member states, which at that time also converged politically in terms of their conservative, Christian democratic majorities, could ultimately agree. Consequently, this situation provided unusually favourable circumstances in Western Europe to cultivate ‘spill-​overs from one functional arena to another and from lower to higher levels of common authority’ (Schmitter and Lefkofridi 2016: 2) in the terminology of neo-​functionalism.The political agreement of both the political elite and the populations translated into a ‘permissive consensus’ vis-​à-​vis European integration until the 1980s.The process of integration was passively approved and has allowed to perceive European integration almost as a one-​way street for a long time. Although certain limited elements of (legal) differentiation were present in the Treaty of Rome (see, e.g., Hanf 2001), they were not seriously invoked in the theory-​building of European integration. None of the established grand schools of thought such as neo-​functionalism and liberal intergovernmentalism proposed major research programs to account for differentiation. Neo-​functionalism indeed reminded us of reverse processes of disintegration or even spill-​back (Schmitter 1969).Yet, cases of disintegration and spill-​back were primarily examined outside the context of European integration and focused on cases of regional disintegration in Eastern Africa, such as the one of the East-​African Community (e.g. Nye 1965; see Gänzle and Wunderlich in this collection).

Retracing the evolution of differentiation in the EU Differentiation as a genuine strategy of integration finds its roots in the Tindemans (1975) report, which laid the foundations of a ‘multi-​speed Europe’ without referring to this notion in explicit terms (Stubb 1996).The broad concept of (temporary) differentiation appeared for the first time in the primary Community law in 1986, as stated in Article 8c of the Single European Act (now Article 27 of the Treaty on the Functioning of the EU [TFEU]): When drawing up its proposals with a view to achieving the objectives set out in Article 7a [now Article 26 TFEU, author’s note], the Commission shall take into account the extent of the effort that certain economies showing differences in development will have to sustain for the establishment of the internal market and it may propose appropriate provisions. If these provisions take the form of derogations, they must be of a temporary nature and must cause the least possible disturbance to the functioning of the internal market. While both the Tindemans Report and the Single European Act triggered only a few articles reflecting on differentiation (Ehlermann 1984; Grabitz 1984; Wallace et al. 1983; Wallace and Ridley 1985), academic discussions on differentiated integration eventually started in the early 1990s. This can be attributed to three main reasons. First, several opt-​outs from the Maastricht Treaty were granted to the UK and Denmark in 1992, leading towards more settled forms of 4

Introduction

differentiation that were based on de jure clauses, raising questions on the future of European integration. Second, the end of the Cold War opened the door to the future ‘big bang enlargement’, creating new challenges for the future of European integration with the potential diversification of national interests and possibilities of further temporary if not non-​ traditional differentiation (Centre for Economic Policy Research 1995). Accordingly, discussions on the constitutionalization of differentiated integration in the Treaty of Amsterdam arose, leading to the introduction of the ‘enhanced cooperation’ mechanism which, to date, has only been used in three cases (divorce law in 2010, unitary patent in 2013 and property regimes of international couples in 2016; see Philipart and Edwards 1999; Fabbrini 2012). One of the very first attempts to conceptually grasp differentiation was made by Alexander Stubb (1996: 283). He conceived differentiated integration as ‘the general mode of integration strategies which try to reconcile heterogeneity within the European Union’. Stubb’s study is the first to conceptualize differentiated integration by listing about thirty models and classifying the mechanism into three categories linked to general concepts of European integration: ‘time’, with ‘multi-​speed Europe’ as the main concept based on different temporal stages of integration (temporal differentiation); ‘space’, with ‘variable geometry Europe’ as the main concept based on differentiated integration of member states (territorial differentiation); and ‘matter’, with ‘à la carte Europe’ as the main concept focusing on differentiated integration across policy domains (sectoral differentiation). In this early literature, terms such as ‘differentiated integration’ and ‘flexible integration’ often remain used interchangeably (see, e.g., Kölliker 2001, 2006; Warleigh 2002). Some studies did not include an explicit definition of the term (see, for instance, Andersen and Sitter 2006; de Neve 2007; Warleigh 2002). When trying to assess the idea of differentiated integration, some were particularly critical and called it as a ‘non-​project’, which could lead to irreconcilable divergences in terms of managing boundaries between legal orders, political efficacy, democratic credentials and self-​legitimation: ‘[c]‌ontingency, ambiguity and disagreement, rather than design, certainty and consensus, are key motifs in the composition of the new differentiated structure’ (Walker 1998: 374). Kölliker stated that although temporary differentiated mechanisms can trigger centripetal effects on ‘reluctant’ member states, that only applies where policy design can ‘change the fundamental character of a common pool resource or a public good’ (2001: 147). Warleigh argued that ‘flexibility offers the most useful means of balancing different (national) interests and thereby allowing progress to be made for (and in) the EU as a whole’ (2002: 2). A series of case studies were also published between the late 1990s and early 2000s. These studies were influenced by the first generation of studies of Europeanization of the nation state (see Mény et al. 1996; Olsen 1996; Hanf and Soetendorp 1998; Knill 2001; Zeff and Pirro 2001; Featherstone and Radaelli 2003). Most of these empirically driven studies of differentiation focused on the relations between the Nordic countries and the EU (Mouritzen 1993; Egeberg and Trondal 1999). Petersen (1998) examined Denmark’s integration policy in what he called a ‘dilemma’ between influence capability and stress sensitivity. Gstöhl (2002a, 2002b) also published studies on ‘reluctant Europeans’, i.e. European countries that did not join the EU (i.e. Norway and Switzerland) or did not join the EMU (i.e. Sweden). Much like Kölliker, she also argued for the need to theorize differentiated integration following the ratification of the Treaty of Amsterdam (Gstöhl 2000). Interestingly, and unlike this particular interest on the Nordic countries, there were few country-​specific studies focusing on the UK as a case of differentiated integration, one notable exception being the UK-​based report of Centre for Economic Policy Research (1995) advocating flexibility to shape the future of European integration. The introduction of the third stage of the EMU and the 2004 ‘big bang enlargement’ effectively led to an increase in temporal differentiation, and to the emergence of what many will dub 5

Benjamin Leruth, Stefan Gänzle and Jarle Trondal

a ‘two-​speed Europe’ (see Piris 2012). By 2010, more than half of EU policies were implemented in different ways. Majone (2009: 205) acknowledged that the EU was evolving into a ‘number of, often overlapping, state groupings established for cooperation in a variety of fields’ (see also Jensen and Slapin 2012). Scholarly contributions extended the work conducted by the first generation of scholars by improving the theoretical and empirical depth of what started to become a sub-​field of European studies. From a theoretical perspective, many studies focused on the scope and limits of differentiated integration in the EU. Andersen and Sitter (2006) asked ‘how much differentiation can the EU accommodate?’ and proposed a typology of European integration with four models based on homogeneous integration, aligned integration, deviant integration and autonomous integration. In line with the macro-​sociological tradition associated with Meyer, Rowan, Powell and colleagues, they argued that differentiation now is ‘a common and normal phenomenon’ (ibid.: 327) and that its study should also include formal and informal arrangements. However, their work has never really been picked up in the subsequent literature on differentiation. De Neve (2007: 516) asked whether differentiated integration is reshaping ‘the European polity into what increasingly resembles a multi-​layered European Onion’ and whether there could be ‘too much’ differentiated integration (an issue that is still being debated in a post-​Brexit context). Following the first Irish vote on the Lisbon Treaty, Jensen and Slapin (2012) focused on the efficiency of the ‘multi-​speed approach’ and suggested a model under which opt-​outs could lead to cascades, i.e. a kind of ‘domino effect’ under which member states opt out because of other member states’ decisions to opt out.The latter study, however, reflects some of the semantic confusion in the existing literature, as it somewhat contradicts Stubb’s original categorization of differentiated integration by using ‘multi-​speed integration’ as a synonym of differentiation (see also Leruth and Lord 2015). The varied ideas about differentiation led Johan P. Olsen (2007) to generalize the question of what kind of political order Europe was in search of. The EU was depicted as ‘a conceptual battleground and an institutional building site’ (Olsen 2010: 81) with a varied mix of organizational forms, governance patterns and ideas about legitimate forms and speeds of integration. Institutional differentiation was eventually understood as ‘new institutional spheres have split off from older ones and developed their own identities’ (ibid.: 142). Dyson and Sepos defined differentiated integration as the process whereby European States, or sub-​units, opt to move at different speeds and/​or towards different objectives with regard to common policies, by adopting different formal and informal arrangements, whether inside or outside the EU treaty framework, and by assuming different rights and obligations. 2010: 4 This extends Kölliker’s definition by including formal and informal arrangements in the framework of differentiation, which were first introduced by Andersen and Sitter (2006). Nevertheless, both definitions only emphasize the demand side of differentiated integration by member states, and not its supply side (i.e. the role of EU institutions to shape differentiated integration). With the Great Recession of 2007–​2008, the EU entered a new multifaceted polycrisis (Leruth 2017; Riddervold et al. 2021). The future of European integration became an increasingly debated issue, and so did the issue of differentiated integration. The possibility of scenarios such as Grexit (i.e. Greece leaving either the Eurozone) and Brexit (re)emerged during the Euro crisis, and Eurosceptic political parties became increasingly influential across Europe (in most cases through more, and sometimes disproportionate, media exposure; see de Vreese 2007). And so did the potential for European disintegration (Vollaard 2014). 6

Introduction

These studies of differentiated integration have been dominated by a ‘Swiss-​German’ school (Gänzle et al. 2019: 9), with numerous scholars attempting to ‘tidy up’ the existing literature and expand knowledge on the implications of differentiation. Holzinger and Schimmelfennig outlined some of the existing shortcomings in this field of study: ‘empirical analysis has been limited to a few important cases of treaty law (such as EMU and Schengen), but there are no comprehensive data sets’ (2012: 293). They highlighted that differentiation always has territorial and sectoral impacts (see Egeberg and Trondal 1999), and that purely functional conceptions are not included in this categorization. They suggested, in turn, a categorization into six dimensions: (1) permanent vs. temporary differentiation, (2) territorial vs. purely functional differentiation, (3) differentiation across nation states vs. multilevel differentiation, (4) differentiation takes place within the EU treaties vs. outside the EU treaties, (5) decision-​making at EU level vs. at regime level (i.e. intergovernmental decisions) and (6) only for member states vs. also for non-​member states/​areas outside the EU territory. Referring to several empirical examples, the authors underline that ‘differentiated integration comes in an astonishing variety of forms and […] the concepts of differentiated integration can and should be used systematically to describe these forms and their frequency’ (ibid.: 297). A major attempt at categorizing differentiated integration was also undertaken by Leuffen et al. (2013), describing the EU is a system of differentiated integration, i.e. ‘one Europe with a single organizational and member state core and a territorial extension that varies by function’ (Schimmelfennig et al. 2015: 767). Basing their study on differentiation of primary law, they argue that differentiated integration varies primarily along two dimensions: variation in the level of centralization across policies (vertical differentiation) and variation in territorial extension across policies (horizontal differentiation). Furthermore, they classify horizontal differentiation into four subcategories: (1) no horizontal differentiation, where all EU rules apply uniformly to all EU member states (e.g. pre-​Maastricht Europe); (2) external differentiation, where EU rules apply uniformly to all EU member states and where non-​member states also can adopt these rules (e.g. the European Economic Area, EEA); (3) internal differentiation, where EU rules do not apply uniformly to all EU member states (e.g. Denmark through the Edinburgh Agreement or the enhanced cooperation procedure); (4) internal and external differentiation, where EU rules from which some EU member states opted out, while non-​member states opted in (e.g. Schengen). Between the mid-​and late 2010s, studies have focused on a variety of aspects of differentiated integration, such as constitutional differentiation (Schimmelfennig and Winzen 2014, 2020, the effects of EU enlargement on differentiated integration (Schimmelfennig 2014; Schimmelfennig and Winzen 2017) and how differentiation affects EU governance (Schimmelfennig 2016a, 2016b). Studies have also focused on differentiated integration within EU legislation, which demonstrate the increasing complexity of EU law and law-​making (e.g. Kroll and Leuffen 2015; Duttle et al. 2017). Special collections in the Journal of European Public Policy, Comparative European Politics and the Journal of Common Market Studies reflected on the evolution of the literature on differentiated integration and included further theoretical and empirical work, notably reflecting on the future of the EU (see, e.g., Fossum 2015; Leruth 2015; Lord 2015; Warleigh-​Lack 2015; Gänzle et al. 2019; Fabbrini and Schmidt 2019). De Wilde and colleagues have argued that a differentiated EU leads to differentiated politicization across times, countries and settings (De Wilde et al. 2016). In sum, these studies have provided theoretical and empirical contents to the literature on European differentiation and suggested conceptual perspectives for studying the phenomenon. Moreover, this line of literature conceives of differentiation as a persistent and ‘normal’ feature of European integration. Yet, the study of differentiation still harboured a myriad of concepts and definitions of the phenomenon as well as theories to explain it. 7

Benjamin Leruth, Stefan Gänzle and Jarle Trondal

Beyond differentiated integration in a post-​Brexit Europe With Brexit, the EU was once again at a crossroads, and so was the scholarly literature on European differentiation. The UK’s vote to leave the EU and the British government’s subsequent decision to trigger Article 50 meant that the Union was facing a series of unprecedented challenges in uncharted territories. As a result, scholars have attempted to explain how and why Brexit happened as well as likely consequences for the future of the EU. For the first time in the history of the EU, one country chose to leave the Union, thus leading not only to differentiated European integration but eventually towards a form of European disintegration. Studies of European disintegration are relatively scarce, mostly because of the lack of empirical evidence pre-​Brexit (Zielonka 2014). The first studies of European disintegration were produced by Vollaard (2014) and Webber (2014), written in the context of the Euro crisis. Douglas Webber (2014) was one of the first academic articles to discuss the possibility of European disintegration, with a strong focus on the role of the states. In so doing, his work also makes a broader plea for mid-​range theories which explain how exogenous shocks are absorbed in ways that reflect the endogenous bias of already existing rules and routines. When applied to how the EU adapts to crisis, institutional perspectives have focused on how institutional segmentation of the EU fosters differentiated crisis sensitivity and crisis management within different policy domains. Consequently, crisis in one policy area, it is argued, does not necessarily spill over to neighbouring policy areas, thus not reverberating across entire systems. In broader terms, ‘bad’ solutions may therefore be implemented in parts of organizations without ruining it all (Ansell and Trondal 2018). Similar ideas from Genschel and Jachtenfuchs (2018) suggest that processes of (dis)integration may unfold differently in policy domains of core-​state powers (through institutional capacity-​building) and non–​core state market integration (through regulatory measures). Some of the recent literature on crisis, disintegration and differentiation in Europe also combines explanations based on collective actors’ cost-​benefit calculations –​such as the promotion of equality of opportunity among EU members (Jones 2018) and institutionalist explanations focusing on how crises are channelled through and mediated by pre-​existing institutional frameworks and resources (e.g. Bátora and Fossum 2020). Consistent with the conclusions of this Handbook, both Vollaard (2018) and Bátora and Fossum (2020) suggest that the EU mainly muddles through crisis (Riddervold et al. 2021), either by means-​end calculating member states balancing different strategies of exit, voice and loyalty (Vollaard 2018) or through institutional lock-​in mechanisms influenced by pre-​existing segmented institutional orders (Bátora and Fossum 2020). Similarly, by combining insights from studies of European disintegration, post-​functionalism and differentiated integration, attempts have recently also been made to draw on the Brexit crisis as a groundbreaking case of differentiated disintegration to explore the mechanisms underlying these processes (Leruth et al. 2019a). Brexit has reinvigorated differentiated integration as a key focus of research in EU studies. Common to this literature is the idea of crisis as a catalyst of increased European differentiation. According to Bátora and Fossum (2020), Schimmelfennig (2017, 2018) and a Symposium in the Journal of Common Market Studies by Leruth et al. (2019b), differentiation is a persistent and embedded phenomenon in the EU –​a systemic feature and not a mere episode in the history of integration. The process of European integration is abundant with examples of fundamental crises, such as the ones triggered by the failure of the EDC in 1954, the empty chair crisis of 1965–​1966 or the ‘Eurosclerosis’ of 1970, to name but a few.Yet, the full disintegration of the Union has never happened and is in line with the findings of this volume not likely to happen according to Vollaard (2018: 259). Theorizing this phenomenon, 8

Introduction

scholars have argued that differentiation is driven by the need to find functional or constitutional compromises (Leuffen et al. 2013; Schimmelfennig 2017, 2018) and linked mechanisms of supply and demand: those on the demand side mostly consist of the national governments of one country or a group of countries that do not wish to follow the integrationist path taken by the inner core of the EU, while the supply side consists of pro-​integrationist governments from member states that accept the demands to move away from uniformity (Leruth et al. 2019b). Differentiation, arguably, not only covers processes where groups of member states proceed with more integration but also processes under which a member state withdraws from participation in the process of European integration (full exit; Leruth et al. 2019b), or component parts of member states withdraw (partial exit), leading to processes of differentiated disintegration (Vollaard 2018: 233). One caveat may be added to this discussion: differentiation should not be just understood as yet another form of or response to crisis (Saurugger and Terpan 2016; Riddervold et al. 2021; Brack and Gurkan 2021). The process of European integration is abundant with examples of fundamental crises. Differentiation is not a crisis per se; it needs to be understood as a variant of integration. It should also be noted that differentiation is neither a feature that is unique to the EU. Even in terms of differentiated disintegration, several regional organizations such as the Eurasian Economic Community (EAEC) –​predecessor of the Eurasian Economic Union –​or the Economic Community of West African States (ECOWAS) have seen states withdrawing from cooperation (Gänzle 2019). Despite the regional focus of this collection, it also offers accounts that may affect how other regional and international organizations deal with diverging interests across member states, periods of integrational stagnation or threats of disintegration.

An inclusive definition of differentiation The above-​mentioned section has shown that differentiation has taken different meanings in the literature (Leruth and Lord 2015; Leruth et al. 2019a; Gänzle et al. 2019). It was often used interchangeably with terms such as ‘flexible integration’ (e.g. Kölliker 2001; Warleigh 2002) and translations of the term also vary (e.g. Crivat 1997). As outlined earlier, Stubb (1996: 283) was among the first ones to offer a clear-​cut definition of differentiated integration and gave it a clear goal, namely ‘to reconcile heterogeneity within the European Union’. Subsequent studies have offered complementing (or sometimes competing) definitions of the term. As differentiation progressively has become a persistent feature of European integration affecting over half of EU policy areas, Leuffen et al. (2013) eventually argued that the EU should be understood as a system of differentiated integration. The semantic confusion and lack of catch-​all definition are perhaps best explained by the fact that differentiation has been a moving target over the past decades. Going back to the roots of differentiation with the ratification of the Treaty of Rome in 1957, differentiation was a purely legal matter (Hanf 2001). It then became an abstract proposal to overcome the period of Eurosclerosis in the 1970s, with Belgian Prime Minister Léo Tindemans (1975) calling for a ‘multi-​speed’ Europe (Koenig 2015). Such temporary form of differentiation later appeared in the Single European Act, before gaining momentum in the early 1990s with the complex ratification of the Maastricht Treaty, the end of the Cold War and subsequent planning of the 2004 big bang enlargement. As any future rounds of EU enlargement were going to lead to a growing diversification of national interests (Centre for Economic Policy Research 1995), differentiated integration was seen as a tool to reconcile such heterogeneity, as covered by the definition of Stubb (1996). The introduction of the enhanced cooperation mechanism through the ratification of the Treaty of Amsterdam, the de facto EMU opt-​out given by the EC to Sweden in 9

Benjamin Leruth, Stefan Gänzle and Jarle Trondal

2003 and discussions surrounding the eventual British withdrawal from the EU meant that differentiation took various meaning that went beyond its original meaning. Subsequently, the Brexit vote triggered an unprecedented process of differentiated disintegration, the latest form of differentiation. Crucially, however, this Handbook documents that differentiation must not be used as a synonym of differentiated integration. Broadly speaking, differentiation is conceptually best conceived in terms of heterogeneity and does not describe a movement towards more or less cooperation. Differentiation as a term therefore covers both differentiated integration and differentiated disintegration. Furthermore, differentiation is an umbrella term covering a wide range of (dis)integrationist techniques such as multi-​speed Europe, variable geometry or à la carte Europe (Stubb 1996; Holzinger and Schimmelfennig 2012). This Handbook includes contributions from authors who offer different perspectives on the roots, scope and normative impact of differentiation. The inclusive definition used in this Handbook is that differentiation is an umbrella term referring to heterogeneous modes of integration and disintegration in the EU.

How differentiation works: a supply and demand model Several authors have developed a supply and demand model of differentiated integration anchored in rational choice theory. Moreover, this model expands the rational choice premises developed by liberal intergovernmentalism. Schimmelfennig and Winzen (2020) have recently developed this line of arguments and established a theory of supply and demand for the study of differentiation, which is particularly useful in order to understand how this phenomenon became increasingly influential in the EU. For that reason, we will refer to it more extensively. The demand side generally comes from national governments of one or a group of countries that do not wish to follow the integrationist path taken by the ‘inner core’ of the EU (e.g. the UK and Denmark, the two ‘champions’ of differentiated integration), while the supply side mostly consists of pro-​integrationist governments, the so-​called insider group. Yet, it should be noted that governments are not alone in requesting and accepting differentiation in the EU: other actors are also directly or indirectly involved in shaping both supply and demand. Supranational actors such as the EC or the European Parliament (EP) can promote or hinder differentiation as the ‘way forward’ (recall the White Paper discussed in the first section). At the domestic level, political parties, civic movements and public opinion can also shape a government’s position on European integration, as is the case with other highly politicized issues such as migration. In addition, demand for differentiation can be driven by the public especially when referendums are held. Some countries’ constitutions require such referendums to be held (as is the case in Denmark and Ireland), while in other countries, government may opt to hold advisory plebiscites for electoral gains or internal divisions (this happened in France, the Netherlands, Sweden and the UK). Where referendums focus on a country’s participation in specific EU policies, the people’s decision may be hard to overturn, even in the case of advisory referendums, as this may fuel otherwise Eurosceptic sentiments (Leruth et al. 2018). Schimmelfennig and Winzen (2020) further analyze the circumstances that affect both supply and demand for differentiated integration and further contribute towards the theorization of the circumstances that create the conditions for differentiation in the EU and beyond. They argue that demand for differentiated integration is driven by three dimensions of heterogeneity. The first one is the heterogeneity of preferences among member states, according to which governments do not share similar values or disagree on supporting specific policies (as was the case, for instance, with issuing ‘corona bonds’ as a response to the coronavirus crisis). The heterogeneity of preferences tends to be linked to the salience of national identity, as support for 10

Introduction

European integration varies depending on the exclusiveness of identity: ‘the more exclusively an individual identifies with an ingroup, the less that individual is predisposed to support a jurisdiction encompassing outgroups’ (Hooghe and Marks 2009: 12). Accordingly, one would expect that countries where exclusive national identities prevail demand more differentiation, as is the case with the UK. The second type is the heterogeneity of dependence, under which governments are not affected in similar ways by one or several factors (such as cross-​border pollution). Such heterogeneity of dependence means that some countries may benefit from participation in EU policies more than others, with the latter group being more disposed to demanding differentiation.The British and Irish opt-​outs of Schengen further illustrate such heterogeneity of dependence, given their insular location.The third form of heterogeneity identified by Schimmelfennig and Winzen is the heterogeneity of capacity, under which governments lack the financial and/​ or technological means to cooperate. The Polish opt-​out of the European Council’s agreement towards carbon neutrality by 2050 is an example of such heterogeneity of capacity, as the Polish government argued that the country’s reliance on coal (on top of political divisions on the matter) was non-​negotiable.The authors further argue that the two dimensions of heterogeneity create conditions for an ‘à la carte’ (or, in the authors’ words, a ‘multi-​menu’) Europe, while the third one creates conditions for multi-​speed differentiation supposing that countries demanding differentiation will eventually rejoin the inner core of integrationist countries. Schimmelfennig and Winzen further identify three supply-​side factors that determine whether differentiated integration will effectively take place. The first one is the size of the insider group, which needs to be large enough to create patterns of integration. The second factor is whether differentiation creates positive or negative externalities. If proposed differentiation creates negative conditions for the supply-​side group and hinders the effectiveness of a policy at the EU level, then differentiation is less likely to be accepted by the integrationist group.The third factor is the institutional context such as decision-​making rules (to determine whether differentiation is allowed) and integration norms. It is the combination of all these factors that help explain why EU member states agreed to grant opt-​outs to the UK in the early 1990s. Differentiation also creates path-​dependency: once agreed, opt-​outs are difficult to roll back and may even spill over other policy areas, which (partly) explains how differentiation becomes so prominent in the 2010s. One factor that is perhaps understudied in the field is the role institutions play in shaping supply differentiation, for instance with regards to the decision-​making mechanisms used and/​ or favoured by actors. Hence, the second part of this Handbook is dedicated to these institutions.

Overview of the Handbook The volume is outlined in five separate parts, each with a section introduction that introduces and summarizes the most important contributions of each section. Part 1 outlines the core set of theoretical approaches that have become important to the study of differentiation in the EU. Part 2 focuses on institutional differentiation, with chapters covering differentiation EU institutions such as the EC, the EP, the Council, the European Central Bank, the European Court of Justice and EU agencies. Part 3 focuses on policy differentiation, with chapters on differentiation across an array of policy fields such as fisheries, market, competition, social policy, asylum, energy, climate, foreign, defence and security policy. Part 4 focuses on territorial differentiation, with chapters on the Nordic countries, Turkey, and the Western Balkans. Finally, Part 5 is devoted to Brexit. The Handbook finally closes with two chapters: first one concluding chapter that brings the Handbook to a close and second one an epilogue chapter on the role of crisis in the study of differentiation and with an empirical study of how the corona crisis has impacted the EU. 11

Benjamin Leruth, Stefan Gänzle and Jarle Trondal

Note 1 This section draws on Leruth et al. (2019).

References Andersen, S.S. and Sitter, N. (2006) ‘Differentiated Integration: What Is It and How Much Can the EU Accommodate?’, Journal of European Integration 28(4): 313–​330. Ansell, C. and Trondal, J. (2018) ‘Governing Turbulence: An Organizational-​Institutional Agenda’, Perspectives on Public Management and Governance 1: 43–​57. https://​doi.org/​10.1093/​ppm​gov/​gvx​013. Ash, T. (2001) ‘Is Britain European?’, International Affairs 77(1): 1–​ 14, https://​doi.org/​10.1111/​14682346.00174. Bátora, J. and Fossum, J.E. (eds.) (2020) Towards a Segmented European Political Order: The European Union’s Post-​Crises Conundrum. Abingdon; New York, NY: Routledge. Brack, N. and Gurkan, S. (eds.) (2021) Theorizing the Crises of the European Union. London: Routledge. Burns, C.J., Gravey,V., Jordan, A. and Zito, A. (2019) ‘De-​Europeanising or Disengaging? EU Environmental Policy and Brexit’, Environmental Politics 28(2): 271–​292. Cardwell, P.J. (2019) ‘The End of Exceptionalism and a Strengthening of Coherence? Law and Legal Integration in the EU Post-​Brexit’, Journal of Common Market Studies 57(6): 1407–​1418. Centre for Economic Policy Research (1995) ‘Flexible Integration: Towards a More Effective and Democratic Europe’, in Monitoring European Integration, Vol. 6. London: Centre for Economic Policy Research. Crivat, L. (1997) ‘Mythes et réalités de l'intégration différenciée dans l'Union européenne’, Master: Université de Genève. De Neve, J.-​E. (2007) ‘The European Onion? How Differentiated Integration is Reshaping the EU’, Journal of European Integration 29(4): 503–​251. de Vreese, C. H. (2007) ‘A Spiral of Euroscepticism: The Media's Fault?’, Acta Politica 42: 271–​286. De Wilde, P., Leupold, A. and Schmidtke, H. (2016) ‘Introduction: The Differentiated Politicization of European Governance’, West European Politics 39(1): 3–​22. Dhingra, S., Huang, H., Ottaviano, G., Pessoa, J., Sampson, T. and Van Reenen, J. (2016a) The Costs and Benefits of Leaving the EU. London: Centre for Economic Performance Technical Report. Dhingra, S., Ottoviano, G., Sampson, T. and van Reenen, J. (2016b) The Consequences of Brexit for UK Trade and Living Standards, LSE Research Online Documents of Economics, London School of Economics and Political Science. Duke, S. (2019) Will Brexit Damage our Security and Defence.The Impact on the UK and EU. London: Palgrave Macmillan. Duttle, T., Holzinger, K., Malang, T., Schläubli, T., Schimmelfennig, F. and Winzen, T. (2017) ‘Opting Out from European Union Legislation: The Differentiation of Secondary Law’, Journal of European Public Policy 24(3): 406–​428. Dyson, K. and Sepos, A. (2010) ‘Differentiation as Design Principle and as Tool in the Political Management of European Integration’, in K. Dyson and A. Sepos (eds.) Which Europe? The Politics of Differentiated Integration. Basingstoke: Palgrave Macmillan, pp. 3–​23. Egeberg, M. and Trondal, J. (1999) ‘Differentiated Integration in Europe: The Case of EEA Country, Norway’, Journal of Common Market Studies 37(1): 133–​142. Ehlermann, C.-​D. (1984) ‘How Flexible is Community Law? An Unusual Approach to the Concept of “Two-​Speeds”  ’, Michigan Law Review 82(5/​6): 1274-​1293.European Parliamentary Research Service (2016) ‘Article 50 TEU: Withdrawal of a Member State from the EU’. Available at: www.europarl. europa.eu/​RegData/​etudes/​BRIE/​2016/​577971/​EPRS_​BRI(2016)577971_​EN.pdf [Accessed 20 November 2020]. European Commission (2017) White Paper on the Future of Europe. Brussels: European Commission. Fabbrini, F. (2012) ‘The Enhanced Cooperation Procedure: A Study in Multispeed Integration’, in Centro Studi sul Federalismo Research Paper October 2012. Torino: Centro Studi sul Federalismo. Fabbrini, S. and Schmidt, V. (2019) ‘Introduction: Differentiation and Beyond’, Comparative European Politics 17: 173–​175. Fagan, C. and Rubery, J. (2018) ‘Advancing Gender Equality through European Employment Policy. The Impact of the UK’s EU Membership and the Risks of Brexit’, Social Policy and Society 17(2): 297–​317. 12

Introduction

Featherstone, K. and Radaelli, C.M. (eds.) (2003) The Politics of Europeanization. Oxford: Oxford University Press. Fossum, J.E. (2015) ‘Democracy and Differentiation in Europe’, Journal of European Public Policy 22(6): 799–​815. Gänzle, S. (2019) ‘Differentiated (Dis)Integration in Europe and Beyond. Historical and Comparative Perspectives’ (Chapter 12), in S. Gänzle, B. Leruth and J. Trondal (eds.) Differentiated Integration and Disintegration in a Post-​Brexit Era. London: Routledge, pp. 202–​218. Gänzle, S., Leruth, B. and Trondal, J. (eds.) (2019a) Differentiated Integration and Disintegration in a Post-​Brexit Era. London: Routledge. Gänzle, S., Leruth, B. and Trondal, J. (2019b) ‘Differentiation, Differentiated Integration and Disintegration in a “Post-​ Brexit-​ Era”’, in S. Gänzle, B. Leruth and J. Trondal (eds.) Differentiated Integration and Disintegration in a Post-​Brexit Era. London: Routledge, pp. 1–​18. Gänzle, S., Leruth, B. and Trondal, J. (2019c) ‘Conclusion: Whither Differentiated (Dis)Integration in a “Post-​Brexit-​Era”?’ (Chapter 14), in S. Gänzle, B. Leruth and J.Trondal (eds.) Differentiated Integration and Disintegration in a Post-​Brexit Era. London: Routledge, pp. 236–​248. Genschel, P. and Jachtenfuchs, M. (2018) ‘From Market Integration to Core State Powers: The Eurozone Crisis, the Refugee Crisis and Integration Theory’, Journal of Common Market Studies 56: 178–​196. George, S. (1998) An Awkward Partner. Britain and the European Community, Oxford: Oxford University Press. Grabitz, E. (ed.) (1984) Abgestufte Integration: Eine Alternative zum herkömmlichen Integrationskonzept?. Strasbourg: N.P. Engel Verlag. Gstöhl, S. (2000) ‘The European Union after Amsterdam:Towards a Theoretical Approach to (Differentiated) Integration’, in M.G. Cowles and M. Smith (eds.) The State of the European Union Volume 5: Risks, Reform, Resistance, and Revival. Oxford: Oxford University Press, pp. 42–​63. Gstöhl, S. (2002a) Reluctant Europeans: Norway, Sweden, and Switzerland in the Process of Integration. London: Lynne Rienner. Gstöhl, S. (2002b) ‘Scandinavia and Switzerland: Small, Successful and Stubborn towards the EU’, Journal of European Public Policy 9(4): 529–​549. Hanf, K. and Soetendorp, B. (eds.) (1998) Adapting to European Integration. London: Longman. Hanf, D. (2001) ‘Flexibility Clauses in the Founding Treaties, from Rome to Nice’, in B. De Witte, D. Hanf and E.Vos (eds.) The Many Faces of Differentiation in EU Law. Oxford: Hart Publishing, 3–​26. Hepburn, C. and Teytelboym, A. (2017) ‘Climate Change Policy after Brexit’, Oxford Review of Economic Policy 33(1): S144–​S154. Holzinger, K. and Schimmelfennig, F. (2012) ‘Differentiated Integration in the European Union: Many Concepts, Sparse Theory, Few Data’, Journal of European Public Policy 19(2): 292–​305. Hooghe, L. and Marks, G. (2009) ‘A Postfunctionalist Theory of European Integration: From Permissive Consensus to Constraining Dissensus’, British Journal of Political Science 39: 1–​23. Howorth, J. (2019) ‘Diferentiation in security and defence policy’, Comparative European Politics 17: 261–277. Jensen, C.B. and Slapin, J.B. (2012) ‘Institutional Hokey-​Pokey: The Politics of Multispeed Integration in the European Union’, Journal of European Public Policy 19(6): 779–​795. Jones, E. (2018) ‘Towards a Theory of Disintegration’, Journal of European Public Policy 25: 440–​451. Knill, C. (2001) The Europeanization of National Administrations. Cambridge: Cambridge University Press. Koenig, N. (2015) A Differentiated View of Differentiated Integration. Berlin: Jacques Delors Institute. Kölliker, A. (2001) ‘Bringing Together or Driving Apart the Union? Towards a Theory of Differentiated Integration’, West European Politics 24(4): 125–​151. Kölliker, A. (2006) Flexibility and European Unification:The Logic of Differentiated Integration. Oxford: Rowman & Littlefield. Kroll, D.A. and Leuffen, D. (2015) ‘Enhanced Cooperation in Practice. An Analysis of Differentiated Integration in EU Secondary Law’, Journal of European Public Policy 22(3): 353–​373. Leruth, B. (2015) ‘Operationalizing national preferences on Europe and differentiated integration’, Journal of European Public Policy 22(6): 816-​835. Leruth, B. (2017) ‘The Europeanization of the Welfare State: The Case for a Differentiated European Social Model’, in P. Taylor-​Gooby, B. Leruth and H. Chung (eds.) After Austerity: Welfare State Transformation in Europe after the Great Recession. Oxford: Oxford University Press, pp. 180–​201. Leruth, B. and Lord, C. (2015) ‘Differentiated Integration in the European Union: A Concept, a Process, a System or a Theory?’, Journal of European Public Policy 22(6): 754–​763. Leruth, B., Startin, N. and Usherwood, S. (2018) The Routledge Handbook of Euroscepticism. London: Routledge. 13

Benjamin Leruth, Stefan Gänzle and Jarle Trondal

Leruth, B., Gänzle, S. and Trondal, J. (2019a) ‘Exploring Differentiated Disintegration in a Post-​Brexit European Union’, Journal of Common Market Studies 57(5): 1013–​1030. Leruth, B., Gänzle, S. and Trondal, J. (2019b) ‘Differentiated Integration and Disintegration in the EU after Brexit: Risks and Opportunities’, Journal of Common Market Studies 57(6), 1383–​1394. https://​doi.org/​ 10.1111/​jcms.12957. Leuffen, D., Rittberger, B. and Schimmelfennig, F. (2013) Differentiated Integration: Explaining Variation in the European Union. Basingstoke: Palgrave Macmillan. Lord, C. (2015) ‘Utopia or Dystopia? Towards a Normative Analysis of Differentiated Integration’, Journal of European Public Policy 22(6): 783–​798. Majone, G. (2009) Europe as theWould-​beWorld Power:The EU at Fifty, Cambridge: Cambridge University Press. March, J.G. (1999) ‘A Learning Perspective on the Network Dynamics of Institutional Integration’, in M. Egeberg and P. Lægreid (eds.) Organizing Political Institutions. Oslo: Scandinavian University Press, pp. 129–​155. Martill, B. and Sus, M. (2018) ‘Post-​Brexit EU/​UK Security Cooperation: NATO, CSDP+​, or ‘French Connection’?’, The British Journal of Politics and International Relations 20(4): 846–​863. Mény,Y., Müller, P. and Quermonne, J.-​L. (eds.) (1996) Adjusting to Europe. London: Routledge. Mouritzen, H. (1993) ‘The Two Musterknaben and the Naughty Boy: Sweden, Finland and Denmark in the Process of European Integration’, Cooperation and Conflict 28(4): 373–​402. Nye, J.S. (1965) Pan-​Africanism and East-​African Integration. Cambridge: Harvard University Press. Olsen, J.P. (1996) ‘Europeanization and Nation-​State Dynamics’, in S. Gustavsson and L. Lewin (eds.) The Future of the Nation-​State. London: Routledge, 157–​188. Olsen, J.P. (2007) Europe in Search of Political Order. Oxford: Oxford University Press. Olsen, J.P. (2010) Governing through Institution Building. Oxford: Oxford University Press. Petersen, N. (1998) ‘National Strategies in the Integration Dilemma: An Adaptation Approach’, Journal of Common Market Studies 36(1): 33–​54. Phillipart, E. and Edwards, G. (1999) ‘The Provisions on Closer Co-​ Operation in the Treaty of Amsterdam:The Politics of Flexibility in the European Union’, Journal of Common Market Studies 37(1): 87–​108. Piris, J.-​C. (2012) The Future of Europe:Towards a Two-​Speed EU?. Cambridge: Cambridge University Press. Riddervold, M., Trondal, J. and Newsome, A. (eds.) (2021) The Palgrave Handbook of EU Crises. Houndmills: Palgrave Macmillan. Rosamond, B. (2016) ‘Brexit and the Problem of European Disintegration’, Journal of Contemporary European Research 12(4): 865–​871. Saurugger, S. and Terpan, F. (2016) Crisis and Institutional Change in Regional Integration (Routledge Studies on Challenges, Crises and Dissent in World Politics, Vol. No. 2). London: Routledge. Schimmelfennig, F. (2014) ‘EU Enlargement and Differentiated Integration: Discrimination or Equal Treatment?’, Journal of European Public Policy 21(5): 681–​698. Schimmelfennig, F. (2016a) ‘Good Governance and Differentiated Integration: Graded Membership in the European Union’, European Journal of Political Research 55: 789–​810. Schimmelfennig, F. (2016b) ‘A Differentiated Leap Forward: Spillover, Path-​Dependency, and Graded Membership in European Banking Regulation’, West European Politics 39(3): 483–​502. Schimmelfennig, F. (2017) ‘Theorising in European Integration’, in D. Dinan, N. Nugent and W.E. Paterson (eds.) The European Union in Crisis. London: Palgrave, pp. 316–​336. Schimmelfennig, F. (2018) ‘Liberal Intergovernmentalism and the Crises of the European Union’, Journal of Common Market Studies 56: 1578–​1594. Schimmelfennig, F., Leuffen, D. and Rittberger, B. (2015) ‘The European Union As a System of Differentiated Integration: Interdependence, Politicization and Differentiation’, Journal of European Public Policy 22(6): 764–​782. Schimmelfennig, F. and Winzen, T. (2014) ‘Instrumental and Constitutional Differentiation in the European Union’, Journal of Common Market Studies 52(2): 354–​370. Schimmelfennig, F. and Winzen, T. (2017) ‘Eastern Enlargement and Differentiated Integration: Towards Normalization’, Journal of European Public Policy 24(2): 239–​258. Schimmelfennig, F. and Winzen, T. (2020) Ever Looser Union? Differentiated European Integration. Oxford: Oxford University Press. Schmitter, P.C. (1969) ‘Three Neo-​Functional Hypotheses about International Integration’, International Organization 23(01): 161–​166.

14

Introduction

Schmitter, P.C. and Lefkofridi, Z. (2016) ‘Neo-​Functionalism As a Theory of Disintegration’, Chinese Political Science Review 1:1–​29. Schneider, C. (2007) ‘Enlargement Processes and Distributional Conflicts: The Politics of Discriminatory Membership in the European Union’, Public Choice 132(1/​2): 85–​102. Stubb, A. (1996) ‘A Categorisation of Differentiated Integration’, Journal of Common Market Studies 34(2): 283–​295. Tindemans, L. (1975) ‘European Union: Report by Mr Léo Tindemans, Prime Minister of Belgium, to the European Council’, Bulletin of the European Communities, Supplement 1/​76. Brussels: European Commission. Vollaard, H. (2014) ‘Explaining European Disintegration’, Journal of Common Market Studies 52(5): 1142–​ 1159.Vollaard, H. (2018) European Disintegration: A Search for Explanations. London: Macmillan. Wallace, H. and Ridley,A. (1985) Europe the Challenge of Diversity (Chatham House Papers). London: Routledge. Wallace, H.,Wallace,W. and Webb, C. (eds.) (1983) Policy Making in European Community. London: John Wiley. Walker, N. (1998) ‘Sovereignty and Differentiated Integration in the European Union’, European Law Journal 4(4): 355–​388. Warleigh,A. (2002) Flexible Integration:Which Model for the European Union?. London: Sheffield Academic Press. Warleigh-​Lack, A. (2015) ‘Differentiated Integration in the European Union: Towards a Comparative Regionalism Perspective’, Journal of European Public Policy 22(6): 871–​887. Webber, D. (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–​365. Zeff, E.E. and Pirro, E.B. (eds.) (2001) The European Union and Member States. London: Boulder. Zeitlin, J. (2016) ‘EU Experimentalist Governance in Times of Crisis’, West European Politics 39(5): 1073–​ 1094. Doi:10.1080/​01402382.2016.1181873. Zielonka, J. (2014) Is the EU Doomed?. Cambridge: Polity Press.

Appendix: a glossary of differentiation Differentiation is a complex phenomenon that takes a wide range of forms, which are often confusing or (wrongly) used interchangeably. This Handbook makes reference to a wide range of concepts and models of differentiation, which all have varying scopes, causes and effects. In order to help the reader understand the key differences between these different concepts and models, Table 1.1 offers a summary of key concepts that are used by contributors throughout this volume. Table 1.1  Summary of key concepts Notion

Definition (and key reference)

Constitutional differentiation

Differentiation adopted following treaty revisions transferring further power to the EU by member states concerned about national sovereignty and identity (Schimmelfennig and Winzen 2014) Differentiation not legally enshrined in EU treaty, but de facto acknowledged by European institutions, e.g. Sweden’s ‘opt-​out’ from EMU (Andersen and Sitter 2006) Differentiation legally enshrined in EU treaty, such as protocols, allowing member states to permanently opt out, e.g. Denmark with regards to EMU (Leruth et al. 2019b) Form of instrumental differentiation under which new member states are excluded from some EU membership benefits for a limited time (Schneider 2007) Form of instrumental differentiation under which new member states are exempted from specific membership obligations (Schimmelfennig 2014) Refers to areas where EU non-​member states can also adopt rules EU rules which apply uniformly to all EU member states, e.g. in the context of the European Economic Area, EEA (Leuffen et al. 2013) (continued)

De facto differentiation

De jure differentiation

Discriminatory differentiation Exemptive differentiation External differentiation

15

Benjamin Leruth, Stefan Gänzle and Jarle Trondal Table 1.1  Cont. Notion

Definition (and key reference)

Horizontal differentiation Internal differentiation

Refers to variation in territorial extension across policies (Leuffen et al. 2013)

Internal and external differentiation Instrumental differentiation Positive differentiation

Negative differentiation

Vertical differentiation

16

Refers to areas where EU rules do not apply uniformly to all EU member states, e.g. Denmark through the Edinburgh Agreement or the enhanced cooperation procedure (Leuffen et al. 2013) Refers to areas where EU rules from which some EU member states opted out, while non-​member states opted in, e.g. Schengen (Leuffen et al. 2013) Transitional arrangements between new member states and the EU institutions to pave the way towards full membership without delaying the accession process (Schimmelfennig and Winzen 2014) Often with normative undertone, allowing other member states to proceed without impacting negatively on those who do not join, e.g. in the framework of enhanced cooperation In highly politicized intergovernmental policy areas, such as the Common Foreign and Security Policy (CFSP), differentiation in cooperation has been the starting point – rather than integration (Howorth 2019) Refers to variation in the level of centralization across policies (Leuffen et al. 2013)

Part 1

Conceptualizing differentiation Jarle Trondal, Benjamin Leruth and Stefan Gänzle

A decade of poly-​crisis has contributed to what can be understood as ‘institutional soul-​searching’ and the raising of questions about the nature of res public among European states (Emery and Giauque 2014: 24; Riddervold et al. 2021). Perhaps unsurprisingly for a political system in the making, the institutional soul searching has been particularly intense in the EU. Academic literature has been equally interested in understanding if and how crisis affects stable and robust political order as well as analysing how varieties of political order are affected differently by crisis. Beyond the study of crisis and political order, the challenge of understanding social and political order has been enduring to the social sciences (Waldo 1992: 149). Unveiling political order has involved disentangling causes of order formation and decay (Bartolini 2005; Fukuyama 2014), consequences of order formation and how the accumulation of political power challenges already pre-​existing orders (Olsen 2007) and finally understanding how political order can be theoretically conceptualized (Bauer and Trondal 2017). With increasing differentiation of the EU, endeavours towards conceptualizing differentiation have become central to EU studies.This section outlines ways of conceptualizing differentiation from a variety of EU studies literatures. One enduring challenge in conceptualizing political order in general, and differentiation of the EU in particular, is to establish meaningful categories that capture essential and enduring characteristics of the phenomenon. Our starting assumption is therefore that adequate analytical categories to classify contemporary differentiation should offer more empirical variation between than within categories. Acknowledging that both analysts and practitioners face problems of meaningful categorization that capture essential aspects of social order in general (Dahlstrom et al. 2012; Painter and Peters 2010: 6; Waldo 1992: 37), and the European political-​administrative system in particular, this section outlines a variety of ways of conceptualizing differentiation. These range from ways of methodologically studying differentiation and examining available data sources from theoretically grounded discussions of differentiation (Chapter 2). Attempts to theoretically discuss differentiation include analysis of the legitimacy of differentiation (Chapter 3), comparing differentiation across international organizations (Chapter 4), examining de jure and de facto differentiation (Chapter 5), the constitutive principles of differentiation (Chapter 6), the internal and external dimension of differentiation and compliant behaviour by non-​EU states involved in external differentiation arrangements (Chapter 7), the role of political symbolism and identities in differentiation (Chapter 8) and finally differentiation from the perspective of segmented political order (Chapter 9). Together, Part 2 offers a varied DOI: 10.4324/9780429054136-2

17

Jarle Trondal, Benjamin Leruth and Stefan Gänzle

conceptual menu to understand differentiation and that is helpful to the succeeding sections of this Handbook.

Chapter 2 Studying differentiated integration: methods and data This chapter by Marian Burk and Dirk Leuffen critically takes stock of methods and data sources used in the study of differentiated integration (DI). They first catalogue and exemplify prominent methods of DI research, before quantitatively tracing the usage of these methods over time. The analysis of a DI bibliography, spanning the period 1995–​2020, reveals that DI predominantly has been a playground for qualitative research in the past. However, more recently, an upsurge of methodological pluralism becomes detectable in the data. They take this as an indication for a growing maturity of the field. Before concluding the chapter, they also introduce and discuss a number of datasets that are already used, or that may potentially be useable, in the study of DI.

Chapter 3 On the legitimacy of differentiated integration DI is a puzzle from the point of view of legitimacy as outlined by Erik O. Eriksen. On the one hand, DI is associated with illicit rule as people are subjected to laws, the making of which they cannot influence on an equal basis. Under circumstances of far-​reaching legal, administrative and economic integration, DI means disenfranchisement undermining the fundamental conditions of democratic self-​rule. Some forms of DI are in breach with the principle of equal membership in a self-​governing republic. On the other hand, DI allows for cooperation even when there is disagreement and when mutual dependence is unequal. DI enables cooperation in specific policy fields, lowers transactions costs and reaps some of the benefits of European public goods. In order to square this circle, this chapter derives an indirect legitimation theorem from Kant’s category of permissive law of public right. This theorem comes with a time and knowledge index, which ties the legitimacy of DI to the level of knowledge and to a particular point in time. This theorem establishes conditions under which DI would enjoy the presumption of being legitimate. Differentiation in this perspective is not perceived as a threat to the European project, but rather as a vehicle of integration.

Chapter 4 Differentiated (dis)integration beyond Europe: a comparative regionalism approach By studying several cases of member-​states’ withdrawal from regional organizations, such as Mauritania’s exit from the Economic Community of West African States in 2000, this chapter by Stefan Gänzle and Jens Uwe Wunderlich offers a comparative regionalism approach to understand differentiation in systems of regional integration. It demonstrates that DI is not exclusively constrained to European integration and the EU, but a more generic feature of regional

18

Conceptualizing differentiation

integration and regional organizations in general. At the same time, however, this chapter also observes that processes of disintegration have remained rare exceptions. The chapter pleads for a far more nuanced understanding of differentiation by including comparative perspectives beyond Europe and the EU.

Chapter 5 De facto differentiation in the European Union: circumventing rules, law, and rule of law Despite differentiation has become a common feature in European integration and has attracted considerable scholarly attention, the vast majority of academic literature focuses on formal (de jure) differentiation, in which opt-​outs or other derogations are codified in EU law.This chapter by Tobias C. Hofelich elaborates on the understudied and ill-​defined phenomenon of informal (de facto) differentiation which is not recognized or supported by EU law. Combining knowledge from several disciplines of EU studies, this chapter conceptualizes de facto differentiation and distinguishes three distinct types: de facto differentiation by non-​compliance, cooperation outside the EU legal framework and unilateral opt-​ins to the acquis. These theoretical foundations are subsequently discussed in an illustrative study of three cases in Economic and Monetary Union, comprising all three types of de facto differentiation. Ultimately, the contribution of this chapter is a conceptual framework of de facto differentiation that invites further research in this area.

Chapter 6 Constitutive differentiation Constitutive differentiation refers to the fundamental architecture of a political system in differentiation terms: how the territorial, functional, hierarchical and person-​incorporation dimensions are configured. This chapter by John Erik Fossum starts by spelling out the distinct challenges associated with clarifying the nature of the EU’s differentiation configuration, and why we need to relate back to the nation-​state. Thereafter, the chapter presents the constitutive principles of the nation-​state, of which sovereignty is essential, and shows how sovereignty shapes the distinct nation-​state differentiation configuration.The chapter argues that the EU transforms state sovereignty, but that the lack of an agreement on which polity configuration should constitute the EU in turn makes it difficult to spell out distinctly what constitutes the differentiation configuration of the EU. Comparing the EU’s actual differentiation pattern with the nation-​state configuration is problematic, due to the uncertain status of the EU configuration in constitutive terms (including the EU’s democratic deficit). In order to make up for that uncertainty, the chapter presents two EU constitutional models. The first has roots in intergovernmentalism. The chapter unpacks it with reference to the differentiation dimensions that were discerned from the nation-​ state model. Thereafter, the chapter lists some of the intergovernmentalism-​inspired accounts of the EU and illuminates that such accounts must exceed well beyond intergovernmentalism to account for present-​day EU. The second constitutional model has roots in federalism, and the chapter shows that this model must also be stretched to take the EU’s patterns of differentiation properly into account.

19

Jarle Trondal, Benjamin Leruth and Stefan Gänzle

Chapter 7 EU external differentiated integration and compliance: theoretical and legal aspects This chapter by Maryna Rabinovych and Anne Pintsch addresses the so far less considered legal dimension of external differentiation. Its point of departure is the challenge that the homogeneity of EU law and the achievement of external differentiation objectives may not be secured due to compliance problems of non-​EU members. Compliance problems may arise because, in contrast to the boundaries of the rules, the boundaries of the institutions are not shifted under external differentiation, so that the usual EU-​internal infringement procedure does not apply. The chapter first introduces two theoretical approaches and related practical suggestions regarding international compliance. Based on this overview, the second part of the chapter presents legal tools that are in place to ensure compliant behaviour by non-​EU states involved in external differentiation arrangements. It concludes that compliance in external differentiation requires both a firm legal framework and, depending on the concrete case, non-​legal instruments such as dialogue and capacity-​building measures. It ends with a call for empirical research on compliance in external differentiation.

Chapter 8 From integration to fragmegration: political symbols and the emergence of differentiated European identities From its inception as the European Coal and Steel Community in 1951, symbols have been a prominent but understudied dimension of European integration. As outlined by Russell Foster in this chapter, academic studies of European integration have focused on the roles of policymakers and institutions at national and transnational levels, while theories of integration such as neofunctionalism, intergovernmentalism and liberal intergovernmentalism have emphasized ‘rational’ dimensions of statecraft. With the Brexit process and the rise of politically powerful Eurosceptic movements across the EU in the aftermath of the Global Financial Crisis and Migration Crisis, this chapter argues that academic studies of European integration must also account for European disintegration, a realm in which symbols are as important, if not more so, than European integration from the 1950s to the 2010s. This chapter argues that post-​functionalist theory which illuminates how the EU’s symbols are integral not only to the processes of European integration, but to the differentiated paths of European disintegration.

Chapter 9 Differentiation and segmentation It is widely recognized that the EU that emerged from the financial and refugee crises of the last decade has become more differentiated. Such a development, as argued by Jozef Bátora and John Erik Fossum in this chapter, brings forth important questions about the nature and character of the EU as a political system, and the kinds of processes and mechanisms that drive its development. An important problem is that neither differentiated integration nor differentiation say much about the positive character or the distinguishing features of the EU as a political system. This chapter argues that the notion of the EU as a segmented political system provides 20

Conceptualizing differentiation

a more apt and precise characterization of the EU as a (differentiated) political system. In addition, the notion of segmentation helps to capture some of the distinct dynamics that propel the EU’s differentiated developments. The chapter starts by providing a brief overview of how EU studies refer to differentiated integration and differentiation. It thereafter defines segmentation and segmented political order and shows how and in what sense we may talk of the EU as a segmented political order. The third and final part of the chapter spells out the most important similarities and differences between on the one hand differentiation/​differentiated integration, and on the other segmentation/​segmented political order.

Looking ahead Based on this section of the Handbook, future conceptual studies of differentiation would be potentially manifold. However, differentiation is of particular relevance for studying how federal-​type systems –​such as the EU –​integrate, and particularly the institutional dimension of such processes. Conceptualizing differentiation is also of relevance for studying the settlement of emergent organizations and organized systems. Institutions and organized systems are temporary settlements of rules and procedures, demography, locations, beliefs and norms. Different approaches to differentiation as outlined in this section are particularly helpful in conceptualizing the settlement of emergent political orders that involves multiple tiers of authoritative decision-​making. Finally, we should also emphasize the attractiveness of studying the sustainability of a differentiated European political system in times of turbulence. The sustainability and resilience of institutions are often taken for granted during periods of stability (Olsen 2007). During periods of turbulence, established institutions are subject to contestation and quest for reform. Sustainability, as argued in this section, relies not merely on the effectiveness and efficiency of institutions to deliver public value, but equally on its legitimacy and perceived value to citizens. Moreover, this section has discussed how differentiation not necessarily undermines but functions as a vehicle for integration and the long-​term sustainability of the EU. Future studies on the sustainability of a differentiated EU should thus in particular study how EU institutions run in situations of stress and turbulence. Turbulent times are marked by the lack of ‘order, rationality, control and predictability’ (Cohen et al. 2012: 7). Turbulence is coined by long-​term unruliness and tensions within environments and organizations that might go unnoticed by practitioners and observers but that may challenge long-​run sustainability of institutions (Ansell and Trondal 2018). During turbulence, decision makers face choices that need to be made under unfamiliar degrees of uncertainty (Tamuz and Lewis 2008: 158). As seen in the Eurozone crisis in Europe, the Euro was marked by unsettledness already from the launch of the Euro area, but the sudden crisis that hit Greece –​and, subsequently, other EU member-​states –​unveiled the latent turbulent nature of the arrangement. As such, the crisis unlocked latent turbulence of the Euro area, which in turn led to differentiated responses from the member-​states. Turbulent times can thus reveal the fragility of institutions and produce surprising cascading dynamics that test the sustainability of existing governance arrangements. We suggest that turbulent times represent an underappreciated opportunity to examine the resilience of differentiated political orders.

References Ansell, C., and J. Trondal (2018) ‘Governing turbulence. An organizational-​institutional agenda’, Perspectives on Public Management and Governance, 1(1): 43–​57. Bartolini, S. (2005) Re-​Structuring Europe. Oxford: Oxford University Press. 21

Jarle Trondal, Benjamin Leruth and Stefan Gänzle

Cohen, M.D., J.G. March and J.P. Olsen (2012) ‘“A Garbage Can Model” at forty: A solution that still attracts problems’, in A. Lomi and J.R. Harrison (eds.) The Garbage Can Model of Organizational Choice: Looking Forward at Forty. Bingley: Emerald, pp. 19–​32. Dahlstrom, C., V. Lapuente and J. Teorell (2012) ‘Public administration around the work’, in S. Holmberg and B. Rothstein (eds.) Good Government. Cheltenham: Edward Elgar, pp. 45–​65. Emery, Y. and D. Giauque (2014) ‘The hybrid universe of public administration in the 21st century’, International Review of Administrative Sciences, 80(1): 23–​32. Fukuyama, F. (2014) Political Order and Political Decay. New York, NY: Farrar, Straus and Giroux. Olsen, J.P. (2007) Europe in Search of Political Order. Oxford: Oxford University Press. Painter, M. and B.G. Peters (eds.) (2010) Tradition and Public Administration. Houndmills: Palgrave Macmillan. Riddervold, M., J.Trondal and A. Newsome (2021) The Palgrave Handbook of EU Crises. Houndmills: Palgrave Macmillan. Tamuz, M. and E.T. Lewis (2008) ‘Facing the threat of disaster: Decision making when the stakes are high’, in G.P. Hodgkinson and W.H. Starbuck (eds.) The Oxford Handbook of Organizational Decision Making. Oxford: Oxford University Press, pp. 155–​173. Trondal, J. and M.W. Bauer (2017) ‘Conceptualizing the European multilevel administrative order. Capturing variation in the European administrative system’, European Political Science Review, 9(1): 73–​94. Waldo, D. (1992) The Enterprise of Public Administration. Novato, CA: Chandler & Sharp Publishers.

22

2 Studying differentiated integration Methods and data Marian Burk and Dirk Leuffen

Introduction1 Since their beginnings, the European Communities have struggled with the question of how to deal with internal heterogeneity. The empty chair crisis is just one, albeit very prominent example, highlighting the potential of individual member states to hold back the entire integration project. In response to the dilemma of wanting to be fair to both sides, i.e. those states that are ready to proceed with further integration and those that are more reluctant, the issue of differentiated or flexible integration has been on the political agenda at least since the 1970s. For instance, the Tindemans report (1976) promotes ‘a new approach’ to prevent Europe ‘to continue to crumble away’, promoting a ‘multi-​speed’ concept of differentiated integration (DI) (Stubb, 1996). Another loud, and more radical, voice against the uniformity dogma of the time was Ralph Dahrendorf, who in his Jean Monnet Lecture at the European University Institute, discussed the ‘Europe à la carte’ model and deplored the ‘strange puritanism not to say masochism which underlies much of Community action’ (Dahrendorf 1979), which was traditionally geared towards uniformity. In the early days of the Community, empirical realizations of DI were mostly limited to temporary restrictions or exceptions, oftentimes used in the context of the accession of new member states, primarily affecting the area of secondary law (Duttle et al. 2017). The Schengen treaty, initially negotiated outside the treaty framework, started a period of more permanent differentiations in the area of primary law. Since then DI has become a core characteristic of the European Union’s setup, leading observers to even describe it as a ‘system of differentiated integration’ (Leuffen et al. 2013). The development of DI has led to an increase in academic interest in the topic. Initially, it was mostly legal scholars addressing the topic, oftentimes with clear practical orientations (cf., e.g., Ehlermann 1984, 1995, 1998; Tuytschaever 1999; Thym 2005, 2006a, 2006b; Piris 2011, 2012). In a review article published with the Journal of European Public Policy in 2012, Holzinger and Schimmelfennig (2012) diagnose ‘many concepts, sparse theory, few data’. In the meanwhile, the field has significantly matured and become a vibrant area of EU studies. Moreover, as there are currently three EU-​financed Horizon 2020 consortia, which study differentiation, the output on the topic is likely to increase significantly in the coming years.2 Against this backdrop,

DOI: 10.4324/9780429054136-3

23

Marian Burk and Dirk Leuffen

this chapter critically takes stock of methodological developments and data sources used in DI research. The chapter starts out by sketching a catalogue of methods used in the study of DI. We then present a quantitative review of the methods used in the study of DI.The descriptive quantitative analysis supports the intuition that the study of DI traditionally was a field of qualitative research. However, more recently, we witness increasing methodological pluralism. We consider this a positive development, underlining the maturity of the field. Moreover, the methodological pluralism signals that the field is methodologically ripe to provide answers to some pressing problems in the academic and political debate. For example, DI research can now contribute to substantiating debates on the future of Europe and the role of DI, therein. In particular, scholars are now able to provide evidence-​based assessments on the consequences of DI at different levels, partly but not exclusively by applying the potential outcomes framework of causal inference. Before concluding the chapter, we introduce a number of datasets suitable for usage in DI research. We thereby hope to inspire readers to consider new approaches to the study of DI.

Cataloguing the methods used in DI research Conceptually, we can broadly distinguish theoretical and empirical contributions to the study of DI. We briefly discuss the theoretical strand before moving to its empirical counterpart. Theoretical research on DI covers conceptual, explanatory and normative approaches. From a methodological perspective, the theoretical work can be classified into formal and non-​formal approaches. Formal modelling approaches are (still) the exception in the study of DI; in fact, just very few examples of formal work on DI stand out (cf., e.g., Jensen and Slapin 2012, Holzinger and Tosun 2019). At the same time, a part of the theoretical literature indirectly links up to formal, rationalist models of politics. For instance, Kölliker (2001) in his theoretical treatment on the emergence (or not) of DI builds on the public or collective goods literature, stressing the importance of the externality structure of goods. As already highlighted above, the early contributions were often motivated by studying the legal but also political feasibility of DI (e.g. Tuytschaever 1999; Ehlermann 1984; Thym 2005; Törö 2005; Walker 1998), and a number of legal scholars continue to be active in the field today (cf. De Witte et al. 2017; Cardwell 2019; Avbelj 2013). The contributions from the social sciences and most importantly from political science, at first, were strongly concept-​oriented, usually, however, with the aim of grasping some core mechanism and logic of DI (e.g. Stubb 1996, 2002; de Neve 2007; Emmanouilidis 2007; Holzinger and Schimmelfennig 2012; Frey and Eichenberger 1996; Fossum 2020). An example of non-​formal explanatory work is perhaps Schimmelfennig et al. (2015), which attempts at harnessing theories of European integration with the purpose of better understanding the patterns of horizontal and vertical differentiation; in the end resulting in a synthetic model of DI. Lastly, normative accounts of DI have quite steadily been on the rise during the past five years or so; examples are Fossum (2015), Lord (2015), Bellamy and Kröger (2017), Eriksen (2018, 2019), Curtin and Fasone (2017), Heermann and Leuffen (2020) and Leuffen (2021). These works have contributed to providing normative justifications of DI; for instance, by pointing out that DI may be a suitable way to reach the EU’s justification of fighting dominance in the form of negative externalities (Lord 2015). Others have critically assessed the emergence of, from their perspective, pathological forms of differentiation (e.g. Eriksen 2018). In a highly stylized, and admittedly somewhat outdated, reading (Goertz and Mahoney 2012), the empirical literature can be subdivided into qualitative and quantitative work. As we will show in more detail below, the latter has increased more recently with new data on the market –​most notably provided by the EUDIFF datasets (Duttle et al. 2017; Schimmelfennig and Winzen 24

Methods and data

2014, 2020). Partly, researchers have also linked theoretical expectations and knowledge on DI to other data sources, as, for instance, in Zhelyazkova’s (2014) analysis of implementation records. Before introducing different datasets on DI research below, it is noteworthy that most quantitative work on DI uses techniques such as regression analysis, more recently, taking insights from the causal inference ‘revolution’ more rigorously into account (Burk and Leuffen 2019; Schraff and Schimmelfennig 2020). An innovative approach is proposed by Naurin and Lindahl (2010), who use network data retrieved from elite telephone interviews to better understand the integration of the states, which had opted out of certain EU policies, into the EU’s informal decision-​making structures, thereby nicely seconding qualitative work by Adler-​Nissen (2009, 2011, 2014).There still is very limited work on the public opinion about DI (Leuffen et al. 2020; de Blok and de Vries 2020), but more research in this area is in the pipeline, conducted, for instance, in the context of the above-​mentioned Horizon 2020 projects. The qualitative empirical literature basically consists of (comparative) case study research, using both, co-​variational and causal-​process tracing logics of inference (Blatter and Haverland 2012; Beach and Brun Pedersen 2013; Rohlfing 2012; Gelhaus and Leuffen 2020). Most comparative work consists in comparing cases of DI, inside the EU, as, for example, between different policy areas, or over time (Rittberger et al. 2014; Dyson and Sepos 2010). However, there also are some examples of comparative work stretching out to comparative regionalism (Warleigh-​ Lack 2015; Murray and Brianson 2019; Leuffen 2013; See Gänzle and Wunderlich, chapter 4 in this volume) and to federalism (Fossum 2019; Fabbrini 2015; Bifulco 2020; Trein et al. 2019). Certainly, this approach has proven to be promising, not least by bringing the study of DI back into comparative politics, and this strand of literature will most likely grow in the years to come. In the following, we outline the use of different methods and approaches used in the study of DI over time.

Methods of DI research over time Using a quantitative content analysis, we here examine journal articles on the topic of DI from 1995 to 2020. For our analysis, we rely on data collected through a keyword search on the term ‘Differentiated integration’, using the abstract search of the powerful online search engine ‘dimensions.ai’.3 In particular, we scanned the titles and abstracts of a selection of major international journals on European integration from 1995 to 2020.4 In a second step, we manually checked the collected articles and discarded those from our dataset that may not be associated with the field of DI.5 We are not claiming data completeness as we miss out, for example, contributions that were published in books.Through our keyword search, we also do not register articles, which may address the topic of DI while not necessarily using the same terminology. Nevertheless, we trust that our data is able to make visible trends over time, and we have no grounds for assuming strong biases in our data (i.e. we would not expect that book publications, in principle, differ from journal articles in terms of the methodologies used). In total, our dataset contains 88 publications from 1995 to 2020. Figure 2.1 reports the total count of publications on ‘differentiated integration’ for each year. As a general result, we observe a growing body of literature dealing with the topic of DI and we diagnose an increasing number of publications per year. Starting with the seminal work of Stubb (1996) only a few articles addressed the topic during the 1990s.6 In fact, until 2007, we identified only one or two journal publications per year. Following a higher number of articles in 2007, we perceive an increasing body of literature on DI from 2013 onwards with peaks reached in 2015 and 2019. In order to examine the methods used in the DI literature, we group the articles in our dataset according to their methodological approaches. Our primary goal with this analysis is not to 25

Marian Burk and Dirk Leuffen 16

Total count of publications per year

14 12 10 8 6 4 2 0 1995

1998

2000

2002

2004

2006

2008

2010

2012

2014

2016

2018

2020

Year

Figure 2.1  Total number of publications on DI from 1995 to 2020, as generated by a keyword search on ‘differentiated integration’ using the database ‘dimensions.ai’

Figure 2.2  Total number of publications on DI over time, sorted by methodological approach. Own classification on ‘dimensions.ai’ ‘differentiated integration’ sample (1995–​2020)

provide an exact examination of the specific methods used by the individual articles, but rather to provide an overview of the methodological trends in the literature. Taking up the general and simple framework, described above, we distinguish between theoretical, qualitative and quantitative accounts and group articles according to their methodological focus in the respective groups. For the classification, we hand-​coded the articles in our sample. Theoretical articles include conceptual, explanatory and normative contributions, which do not have a strong empirical focus. Qualitative approaches, for example, use single or comparative case studies, expert interviews or process tracing, or, in other words, empirical research that is not quantitative. Lastly, we code articles as quantitative if they mainly rely on descriptive and/​or inferential statistics using large-​n datasets.7 Figure 2.2 provides an overview of the different approaches, as used over time. In general, we find that qualitative approaches make up the largest share of DI literature in the covered time period. Up until 2014, the articles mostly consist of theoretical and/​or qualitative 26

Methods and data

100%

90%

Share in total count of publications

80%

70%

60%

50%

40%

30%

20%

10%

0% 2007

2009

2011

2013

2015

2017

2019

Year Theoretical

Qualitative

Quantitative

Figure 2.3  Share of methodological approaches in DI publications (period 2007–​2020). Own classification on ‘dimensions.ai’ ‘differentiated integration’ sample

approaches. From 2014 onward, however, we witness a surge of contributions that rely on quantitative methods, alongside the use of qualitative and theoretical approaches. Importantly, however, qualitative accounts still hold the largest share during most years. This pattern also becomes visible in Figure 2.3, which depicts the share of each approach in the overall sum of publications.8 The general result here is that we observe more methodological pluralism. While still relying on a solid basis of theoretical work, and complementing the qualitative study of DI, the use of quantitative methods has surged in recent years.

Towards methodological pluralism in the study of DI The analysis of the sampled DI articles has revealed that DI was largely a field of qualitative research in the past. However, we diagnose an increase of other approaches that complement 27

Marian Burk and Dirk Leuffen

the qualitative approaches. This development is promising, since it allows a better understanding of DI as a multidimensional and multilevel phenomenon. In fact, at the aggregate level, the methodological pluralism brings about the advantages of multi-​method approaches to the social sciences (Seawright 2016; Collier and Elman 2008; Leuffen and Puetter 2021). For instance, it combines the strength of strong internal validity, often linked to small-​n research, and external validity ascribed to large-​n research. Or, in other words, ‘working together’ (Poteete et al. 2010) pays off! Elsewhere, we have highlighted that more recent DI research has taken the direction of moving from causes-​to-​effects to effects-​of-​causes approaches (Burk and Leuffen 2019).Without wanting to open the delicate Pandora’s box of causality, our claim is that, not least for practical purposes (cf. European Commission 2017), it is indispensable to get a better understanding of the workings and consequences of DI. Here, the use of new data, as well as a stronger appreciation of the potential outcome model of causal inference (Angrist and Pischke 2009; Morgan and Winship 2014) may be useful.

Datasets for DI research While the qualitative strand of literature has mostly used ‘causal-​ process observations’, an increased use of ‘data-​set observations’ (Brady 2010) is possible in the study of DI today. In the following, we briefly introduce a number of datasets currently employed in DI research. Moreover, we mention other datasets that have not been used extensively in the study of DI, despite their potential for it. As the field is developing fast, many of the mentioned data resources are work in progress and we hope that they will be made available to the public in the coming years; right now, only some of them meet this criterion. We here provide some examples of datasets, without claiming completeness but with the hope that those examples may serve as an inspiration to the future designers of DI research. Most importantly, the EUDIFF1 and EUDIFF2 datasets provide the best empirical foundations for DI in primary (also termed ‘treaty-​based differentiation’) and secondary law (‘legislative differentiation’), respectively. EUDIFF2 records all legislative exemptions and opt-​outs granted to member states from 1958 to 2012 (an updated version is used in Schimmelfennig and Winzen (2020)).Various forms of DI are covered –​the authors distinguish temporary versus permanent, actual versus potential and general versus specific differentiation. The dataset is introduced in depth in Duttle et al. (2017) and can be accessed and downloaded in the accompanying data archive.9 EUDIFF1 is based on article-​years, EUDIFF2 on legislative act-​years (Schimmelfennig 2014; Schimmelfennig and Winzen 2014). These datasets provide information on longitudinal patterns of DI and allow for comparisons across policy areas and countries. Moreover, they can be used for identifying specific instances of differentiation. In sum, they form a particularly useful foundation for the quantitative study of DI.The EUDIFF datasets have also been linked to other data in past research, an example is Zhelyazkova (2014) that links information on differentiation to implementation records (cf. also Thomann and Zhelyazkova 2017). Interestingly, to our knowledge, the DEU I and II (Thomson et al. 2012) or EMU Choices (Wasserfallen et al. 2019) datasets so far have not systematically been used in the study of DI.That is surprising, given that their focus on preferences and saliences could provide important insights into the mechanisms of decision-​making leading to, or resulting from, differentiation. DEU type of data could be merged, for instance, with data on the legislative process, as provided by the European Union Policy-​Making (EUPOL) dataset, constructed by Frank Häge.10 There are a number of data collection projects currently in progress in the above-​mentioned Horizon 2020 consortia. For instance, one Work Package in EU3D collects data on parliamentary debates on differentiation, a group around Richard Bellamy and Sandra Kröger (of 28

Methods and data

the InDivEU consortium) collects data on elite and partisan opinions about DI (Kröger et al. 2021), and there are a number of projects on public opinion (e.g. by Funda Tekin [EU IDEAS], Catherine de Vries [InDivEU] and Dirk Leuffen [EU3D]). De Blok and de Vries (2020) provide an overview of existing public opinion data about DI (cf. also Leuffen et al. 2020). In sum, there is a rich data basis emerging that will allow for a more careful investigation of the causes and consequences of DI.

Conclusion This chapter has reviewed DI research with a particular perspective on methods. After typologizing and exemplifying different approaches used in the study of DI, we present a quantitative content analysis of journal articles on DI. As a first result, our data shows that the field of DI has grown especially in the more recent years. While in no way claiming completeness, the analysis has further corroborated the general impression that DI started out as a largely qualitative field of research, but in the last decade has moved towards methodological pluralism. In our view, this is a positive development highlighting the maturity of the research area, as well as the continuous curiosity created by DI. Moreover, we find that the different approaches nicely complement one another ensuring more solid understandings of the patterns and the mechanism of different forms of DI. Echoing the claim we made in Burk and Leuffen (2019), we see further scope for a more systematic taking into account of the potential outcomes model of causal inference. In the chapter, we also discuss a few examples of datasets already employed in DI research, as well as others that so far have been less used in the study of DI.These datasets can be fruitfully integrated into mixed methods approaches to the study of DI and open up the field for new research questions. In sum, we have reason to be optimistic that the quality of research on DI is likely to improve even further in the coming years. Not least, this development may also lead to better policy recommendations on DI, as a salient and fascinating issue of present and future European politics.

Notes 1 We thank Jarle Trondal for helpful comments. Leuffen’s work was generously supported by the European Union’s Horizon 2020 Research and Innovation Programme under Grant Agreement 822419. 2 The most up to date information on these consortia (EU3D, EU IDEA, InDivEU) and their research output can be found online under www.eu3d.uio.no/​, https://​eui​dea.eu/​ and http://​indi​veu. eui.eu/​. 3 ‘Dimensions.ai’ is increasingly used for such purposes, cf., e.g.,Yin et al. (2021). Coevolution of policy and science during the pandemic. Science, 371, 128–​130. 4 In particular, we searched the following outlets: Journal of European Public Policy, Journal of European Integration, JCMS Journal of Common Market Studies West European Politics, European Union Politics, Journal of European Union Studies, European Journal of Political Economy, European Journal of Political Research, Comparative European Politics, European Law Journal. 5 An example for such a false positive is DOIDGE, M. (2007). Joined at the hip: Regionalism and interregionalism. European integration, 29(2), 229–​ 248. This article uses a perspective rooted in interregionalism and does not use the term Differentiated Integration. 6 The above-​mentioned restrictions imposed by the keyword search may particularly apply for the earlier period of inquiry. However, by the year 2000, the DI terminology seems solidly established. 7 As we are interested in the applied research designs, we do not include articles such as book reviews or editorials in the methodological analysis. This reduces the count of publications to 82. 8 As the number of contributions until 2007 is comparatively low, we only depict the shares for the years from 2007 to 2020. 29

Marian Burk and Dirk Leuffen

9 https://​tandf.figshare.com/​articles/​dataset/​Opting_​out_​from_​European_U ​ nion_​legislation_​the_​differentiation_​of_​secondary_​law/​4650190 10 This dataset is accessible via Dataverse: https://​dataverse.harvard.edu/​dataset.xhtml?persistentId=​ doi:10.7910/​DVN/​AZCAFA

Bibliography Adler-​Nissen, R. (2009) Behind the scenes of differentiated integration: circumventing national opt-​ outs in Justice and Home Affairs. Journal of European Public Policy, 16(1), pp. 62–​80. doi: 10.1080/​ 13501760802453239. Adler-​Nissen, R. (2011) Opting out of an ever closer union: the integration Doxa and the management of sovereignty. West European Politics, 34(5), pp. 1092–​1113. doi: 10.1080/​01402382.2011.591102. Adler-​Nissen, R. (2014) Opting Out of the European Union. Diplomacy, Sovereignty and European Integration, Cambridge, Cambridge University Press. Angrist, J.D. and Pischke, J.-​S. (2009) Mostly Harmless Econometrics. An Empiricist’s Companion, Princeton, NJ, Princeton University Press. Avbelj, M. (2013) Differentiated integration –​farewell to the EU-​27? German Law Journal, 14(1), pp. 191–​ 212. doi: 10.1017/​S2071832200001760. Beach, D. and Brun Pedersen, R. (2013) Process-​Tracing Methods: Foundations and Guidelines, Ann Arbor, MI, University of Michigan Press. Bellamy, R. and Kröger, S. (2017) A democratic justification of differentiated integration in a heterogeneous EU. Journal of European Integration, 39(5), pp. 625–​639. doi: 10.1080/​07036337.2017.1332058. Bifulco, R. (2020) Differenziazione e asimmetrie nella teoria federale contemporanea. In: Fasone, C., Lupo, N. and Vauchez, A. (eds.) Parlamenti e democrazia in Europa. Federalismi asimmetrici e integrazione differenziata. Bologna: Il Mulino, pp. 47–​76. Blatter, J. and Haverland, M. (2012) Designing Case Studies. Explanatory Approaches in Small-​N Research, Houndmills, Palgrave Macmillan. Brady, H.E. (2010) Data-​set observations versus causal-​process observations: the 2002 U.S. presidential election. In: Brady, H.E. and Collier, D. (eds.) Rethinking Social Inquiry. Diverse Tools, Shared Standards Lanham: Rowman & Littlefield, pp. 237–​242. Burk, M. and Leuffen, D. (2019) On the methodology of studying differentiated (dis)integration: or how the potential outcome framework can contribute to evaluating the costs and benefits of opting in or out. JCMS: Journal of Common Market Studies, 57(6), pp. 1395–​1406. doi: 10.1111/​jcms.12958. Cardwell, P.J. (2019) The end of exceptionalism and a strengthening of coherence? Law and legal integration in the EU post-​Brexit. JCMS: Journal of Common Market Studies, 57(6), 1407–​1418. doi: 10.1111/​ jcms.12959. Collier, D. and Elman, C. (2008) Multi-​method research: organizations, publication, and reflections on integration. In: Box-​Steffensmeier, J., Brady, H.E. and Collier, D. (eds.) The Oxford Handbook of Political Methodology. Oxford: Oxford University Press. doi: 10.1093/​oxfordhb/​9780199286546.003.0034. Curtin, D. and Fasone, C. (2017) Differentiated representation: is a flexible European Parliament desirable? In: De Witte, B., Ott, A. and Vos, E. (eds.) Between Flexibility and Disintegration. The Trajectory of Differentiation in EU Law. Cheltenham: Edward Elgar. pp. 118–​145. Dahrendorf, R. (1979) RE: A Third Europe? Type to Lecture, T. J. M. De Blok, L. and De Vries, C. (2020) A blessing and a curse? Examining public preferences for differentiated integration. In: InDivEU Workshop. De Neve, J.-​E. (2007) The European onion? How differentiated integration is reshaping the EU. Journal of European Integration, 29(4), pp. 503–​521. doi: 10.1080/​07036330701502498. De Witte, B., Ott,A. andVos, E. (eds.) (2017) Between Flexibility and Disintegration.The Trajectory of Differentiation in EU Law, Cheltenham: Edward Elgar. Duttle,T., Holzinger, K., Malang,T., Schäubli,T., Schimmelfennig, F. and Winzen,T. (2017) Opting out from European Union legislation: the differentiation of secondary law. Journal of European Public Policy, 24(3), pp. 406–​428. doi: 10.1080/​13501763.2016.1149206. Dyson, K. and Sepos, A. (eds.) (2010) Which Europe? The Politics of Differentiated Integration, Houndmills, Palgrave. Ehlermann, C.-​D. (1984) How flexible is community law? An unusual approach to the concept of “two-​ speeds”. Michigan Law Review, 82, pp. 1274–​1293. doi: 10.2307/​1288478. 30

Methods and data

Ehlermann, C.-​D. (1995) In: Paper, E.W. (ed.) Increased Differentiation or Stronger Uniformity. Badia Fiesolana and San Domenico, CA: European University Institute. Ehlermann, C.D. (1998) Differentiation, flexibility, closer co-​operation: the new provisions of the Amsterdam treaty. European Law Journal, 4(3), pp. 246–​270. doi: 10.1111/​1468-​0386.00052. Emmanouilidis, J.A. (2007) Institutional consequences of differentiated integration. CAP Discussion Paper, February 2007. Eriksen, E.O. (2018) Political differentiation and the problem of dominance: segmentation and hegemony. European Journal of Political Research, 57(4), pp. 989–​1008. doi: 10.1111/​1475-​6765.12263. Eriksen, E.O. (2019) Contesting Political Differentiation: European Division and the Problem of Dominance, Basingstoke, Palgrave. European Commission (2017) White Paper on the Future of Europe. Reflections and Scenarios for the EU27 by 2025. Brussels: EU Commission. Fabbrini, S. (2015) Which European Union? Europe after the Euro Crisis, Cambridge, Cambridge University Press. Fossum, J.E. (2015) Democracy and differentiation in Europe. Journal of European Public Policy, 22(6), pp. 799–​815. doi: 10.1080/​13501763.2015.1020838. Fossum, J.E. (2019) Europe’s triangular challenge: differentiation, dominance and democracy. EU3D Research Papers. Fossum, J.E. (2020) The institutional make-​up of Europe’s segmented political order. In: Bátora, J. and Fossum, J.E. (eds.) Towards a Segmented European Political Order. The European Union’s Post-​Crises Conundrum. London: Routledge. pp. 22–​44. Frey, B.S. and Eichenberger, R. (1996) FOCJ: competitive governments for Europe. International Review of Law and Economics, 16(3), pp. 315–​327. doi: 10.1016/​0144-​8188(96)00019-​1. Gelhaus, L. and Leuffen, D. (2020) Case selection. In: Morin, F., Olsson, C. and Atikcan, E.Ö. (eds.) Research methods in the social sciences: a A-​Z of key concepts. Oxford: Oxford University Press. pp. 33–​38. Goertz, G. and Mahoney, J. (2012) A tale of two cultures: qualitative and quantitative research in the social sciences, Princeton, NJ, Princeton University Press. Heermann, M. and Leuffen, D. (2020) No representation without integration! Why differentiated integration challenges the composition of the European Parliament. JCMS: Journal of Common Market Studies, 58(4), pp. 1016–​1033. doi: 10.1111/​jcms.13015. Holzinger, K. and Schimmelfennig, F. (2012) Differentiated integration in the European Union: many concepts, sparse theory, few data. Journal of European Public Policy, 19(2), pp. 292–​305. doi: 10.1080/​ 13501763.2012.641747. Holzinger, K. and Tosun, J. (2019) Why differentiated integration is such a common practice in Europe: a rational explanation. Journal of Theoretical Politics, 31(4), pp. 642–​659. doi: 10.1177/​0951629819875522. Jensen, C.B. and Slapin, J.B. (2012) Institutional Hokey-​ Pokey: the politics of multispeed integration in the European Union. Journal of European Public Policy, 19(6), pp. 779–​ 795. doi: 10.1080/​ 13501763.2011.610694. Kölliker, A. (2001) Bringing together or driving apart the union? Towards a theory of differentiated integration. West European Politics, 24(4), pp. 125–​151. doi: 10.1080/​01402380108425468. Kröger, S., Lorimer, M. and Bellamy, R. (2021) The democratic dilemmas of differentiated integration: the views of political party actors (March 13, 2021). Swiss Political Science Review. doi: 10.1111/​spsr.12449. Leuffen, D. (2013) European Union as a blueprint? Nine hypotheses on differentiated integration in a comparative perspective. In: WAI-​ZEI Paper. Bonn: Center for European Integration Studies. Leuffen, D. (2021) Member state equality and procedural fairness in differentiated integration: a republican perspective. In: Fromage, D. (ed.) (Re-​)defining Membership: Differentiation in and Outside the European Union. Oxford: Oxford University Press. Leuffen, D. and Puetter, U. (2021) Studying integration and EMU’s choices. In: Puntscher-​Rieckmann, S. and Wasserfallen, F. (eds.) The Politics of Eurozone Reform. Colchester: ECPR Press, pp. 103–​122. Leuffen, D., Rittberger, B. and Schimmelfennig, F. (2013) Differentiated Integration. Explaining Variation in the European Union, Basingstoke, Palgrave. Leuffen, D., Schuessler, J. and Gómez Díaz, J. (2020) Public support for differentiated integration: individual liberal values and concerns about member state discrimination. Journal of European Public Policy, pp. 1–​20. doi: 10.1080/​13501763.2020.1829005. Lord, C. (2015) Utopia or dystopia? Towards a normative analysis of differentiated integration. Journal of European Public Policy, 22(6), pp. 783–​798. doi: 10.1080/​13501763.2015.1020839. Morgan, S.L. and Winship, C. (2014) Counterfactuals and Causal Inference. Methods and Principles for Social Research, Cambridge, Cambridge University Press. 31

Marian Burk and Dirk Leuffen

Murray, P. and Brianson, A. (2019) Rethinking Britain’s role in a differentiated Europe after Brexit: a comparative regionalism perspective. JCMS: Journal of Common Market Studies, 57(6), pp. 1431–​1442. doi: 10.1111/​jcms.12961. Naurin, D. and Lindahl, R. (2010) Out in the cold? Flexible integration and the political status of Euro opt-​ outs. European Union Politics, 11(4), pp. 485–​509. doi: 10.1177/​1465116510382463. Piris, J.-​C. (2011) It is time for the Euro area to develop further closer cooperation among its members. In: Jean Monnet Working Paper 05/​11. New York, NY: New York University. Piris, J.-​C. (2012) The Future of Europe.Towards a Two-​Speed EU?, Cambridge, Cambridge University Press. Poteete, A., Janssen, M.A. and Ostrom, E. (2010) Working Together. Collective Action, the Commons, and Multiple Methods in Practice, Princeton, NJ, Princeton University Press. Rittberger, B., Leuffen, D. and Schimmelfennig, F. (2014). Differentiated integration of core state powers. In: Genschel, P. and Jachtenfuchs, M. (eds.) Beyond the Regulatory Polity? The European Integration of Core State Powers. Oxford: Oxford University Press. pp. 189–​210. Rohlfing, I. (2012) Case Studies and Causal Inference: An Integrative Framework, Basingstoke, Palgrave Macmillan. Schimmelfennig, F. (2014) EU enlargement and differentiated integration: discrimination or equal treatment? Journal of European Public Policy, 21(5), pp. 681–​698. doi: 10.1080/​13501763.2014.897744. Schimmelfennig, F. and Winzen, T. (2014) Instrumental and constitutional differentiation in the European Union. Journal of Common Market Studies, 52(2), pp. 354–​370. doi: 10.1111/​jcms.12103. Schimmelfennig, F., Leuffen, D. and Rittberger, B. (2015) The European Union as a system of differentiated integration: interdependence, politicization and differentiation. Journal of European Public Policy, 22(6), pp. 764–​782. doi: 10.1080/​13501763.2015.1020835. Schimmelfennig, F. and Winzen, T. (2020) Ever Looser Union? Differentiated European Integration, Oxford, Oxford University Press. Schraff, D. and Schimmelfennig, F. (2020) Does differentiated integration strengthen the democratic legitimacy of the EU? Evidence from the 2015 Danish opt-​out referendum. European Union Politics, 21(4), pp. 590–​611. doi: 10.1177/​1465116520949698. Seawright, J. (2016) Multi-​Method Social Science. Combining Qualitative and Quantitative Tools, Cambridge, Cambridge University Press. Stubb, A. (1996) A categorization of differentiated integration. JCMS: Journal of Common Market Studies, 34(2), pp. 283–​295. doi: 10.1111/​j.1468-​5965.1996.tb00573.x. Stubb, A. (2002) Negotiating Flexibility in the European Union. Amsterdam, Nice and Beyond, Basingstoke, Palgrave. Thomann, E. and Zhelyazkova, A. (2017) Moving beyond (non-​)compliance: the customization of European Union policies in 27 countries. Journal of European Public Policy, 24(9), pp. 1269–​1288. doi: 10.1080/​ 13501763.2017.1314536. Thomson, R., Arregui, J., Leuffen, D., Costello, R., Cross, J., Hertz, R. and Jensen, T. (2012) A new dataset on decision-​making in the European Union before and after the 2004 and 2007 enlargements (DEUII). Journal of European Public Policy, 19(4), pp. 604–​622. doi: 10.1080/​13501763.2012.662028. Thym, D. (2005) “United in diversity” –​the integration of enhanced cooperation into the European constitutional order. German Law Journal, 6(11), pp. 1731–​1747. doi: 10.1017/​S2071832200014656. Thym, D. (2006a) The political character of supranational differentiation. European Law Review, 31, pp. 781–​799. Thym, D. (2006b) Supranationale Ungleichzeitigkeit im Recht der europäischen Integration. Europarecht, 41(5), pp. 637–​655. doi: 10.5771/​0531-​2485-​2006-​5-​637. Tindeman, L. (1976) European Union. Report by Mr Leo Tindemans, Prime Minister of Belgium, to the European Council. In: Bulletin of the European Communities. Brussels: Commission of the European Communities. Törö, C. (2005) The latest example of enhanced cooperation in the constitutional treaty: the benefits of flexibility and differentiation in European security and defence policy decisions and their implementation. European Law Journal, 11(5), pp. 641–​656. doi: 10.1111/​j.1468-​0386.2005.00280.x. Trein, P., Thomann, E. and Maggetti, M. (2019) Integration, functional differentiation and problem-​solving in multilevel governance. Public Administration, 97(2), pp. 339–​354. doi: 10.1111/​padm.12595. Tuytschaever, F. (1999) Differentiation in European Union Law, Oxford, Hart. Walker, N. (1998) Sovereignty and differentiated integration in the European Union. European Law Journal, 4(4), 355–​388. doi: 10.1111/​1468-​0386.00058.

32

Methods and data

Warleigh-​ Lack, A. (2015) Differentiated integration in the European Union: towards a comparative regionalism perspective. Journal of European Public Policy, 22(6), pp. 871–​ 887. doi: 10.1080/​ 13501763.2015.1020837. Wasserfallen, F., Leuffen, D., Kudrna, Z. and Degner, H. (2019) Analysing European Union decision-​ making during the Eurozone crisis with new data. European Union Politics, 20(1), pp. 3–​23. doi: 10.1177/​ 1465116518814954. Yin, Y., Gao, J., Jones, B.F. and Wang, D. (2021) Coevolution of policy and science during the pandemic. Science, 371(6526), pp. 128–​130. doi: 10.1126/​science.abe3084. Zhelyazkova, A. (2014) From selective integration into selective implementation: the link between differentiated integration and conformity with EU laws. European Journal of Political Research, 53(4), pp. 727–​746. doi: 10.1111/​1475-​6765.12062.

33

3 On the legitimacy of differentiated integration Erik O. Eriksen

Introduction Brexit has been a shock, awakening us to the instability of the present European political order.1 Instability may lead to more political differentiation. It can pave the way for a multi-​speed or a two-​tiered Europe. Several European leaders have supported the idea of formalizing the concept of a multi-​speed Europe, where member states could choose the speed of their integration. According to the European Union (EU), ‘Coalitions of the willing’ may join forces and go together in specific areas (European Commission 2017). Integration as a one-​way street, if it ever has been the case, may thus end. We may see a change from a union of different speeds to one of different statuses –​a permanent two-​tiered Europe. Differentiated integration (DI) is a pragmatic response to political challenges of a fundamental character. It is a way to keep together a controversial system by sectioning off particular policy areas and countries from centralized rule. DI has become a functional means of handling (or bypassing) various forms of crises and various forms of opposition to integration. The consequences of DI are, however, not ‘pragmatic’; they are not trivial. Types of differentiation affect identities, political statuses and self-​rule and threaten the integrity and viability of the EU. Misguided differentiation jeopardizes citizens’ capacity ‘to gain and maintain control over their collective fates’ (Offe and Preuss 2016: 21). It undermines democratic autonomy and increases heteronomy in the form of arbitrary rule, i.e., dominance. This finding may come as a surprise for proponents of differentiation and also of disintegration of the EU. What may be a benign plea for sovereignty and independence, in fact, turns out otherwise. Many fight supranationalism and the EU, or they opt for political differentiation in the name of freedom and self-​rule. They see differentiation as a way to strengthen the autonomy of the member states. For many, differentiation means flexibility and the dispersion of democratic control. It would reduce the need for veto power and bargaining muscles to protect national interests (see e.g., Christiano 2010: 132). It may prevent integration contrary to the citizens’ preferences (Pettit 2012: 27). However, under conditions of far-​reaching legal, administrative and economic integration, political differentiation of sorts has malignant consequences. It means decisional exclusion, disenfranchisement and hence arbitrary rule.There is the spectre of new divisions and dividing lines in Europe: of semi-​colonial states and of first-​and second-​class Europeans. 34

DOI: 10.4324/9780429054136-4

Legitimacy

Orders bereft of legitimacy –​orders that are not acceptable for subjected parties –​raise the problem of stability. A multi-​speed Europe may not be acceptable for the parties, or it may only be acceptable for conflicting reasons. Laggards and integrationists prototypically have different reasons for supporting opt-​outs and differential treatment. Disagreement on a common course of action, and unwillingness to cede sovereignty, are the original impetuses for relegating a group of states to the periphery. While laggards do not want more integration, integrationists accept opt-​outs and side-​lining of some members in order to make more integration possible. Orders that are not justified with mutually acceptable reasons face legitimacy problems and would be unstable. In this article, I first clarify the democratic problem associated with DI, why there is a legitimacy problem. However, one may also wonder whether a Europe of different speeds could be justified as a necessary means to a future end –​a united and democratic Union. Jürgen Habermas and Jacques Derrida saw this kind of differentiation as the only way to convince the populations of member states that want to hold on to their sovereignty that the European project makes sense.2 Could political differentiation be justified as a necessary step back in order for Eurosceptics to change their minds? From Kant’s category of permissive law of public right, I derive an indirect legitimation theorem with a time and knowledge index. It establishes conditions under which DI would enjoy the presumption of being legitimate. Differentiation, in this perspective, is not perceived as a threat to the European project, but rather as a vehicle of integration. I proceed by addressing the legitimacy problems of DI both in the horizontal and the vertical dimension of integration. The associated non-​members’ agreement with the EU and the Eurozone crisis arrangements represent extreme forms of DI in the respective dimensions.Then, I will address the question of political differentiation as a vehicle of integration. The question is whether political differentiation, which can undermine democratic self-​rule and cause dominance, could still, under certain conditions, be legitimate. But first of all, what does legitimacy mean today?

On legitimacy What is puzzling about many analyses of differentiation and pleas for more differentiation is the lack of concern for democratic legitimacy. Or rather, legitimacy is related to outcome-​oriented standards. If differentiation pays off in functional and economic terms, or if it calms the EU sceptics, then it is conceived of as a good thing. The policy effects rather than the polity effects are in focus. Instead of seeing differentiation as solely a policy question or the technical task of managing externalities, as a question of Pareto optimality, one should see it as deeply political, related to the basic political structure –​to justice and democracy. In this reading, political orders are about more than satisfying preferences and improving welfare; they are about a polity securing the equal freedom of all. Under modern conditions, political order is essentially about the ability of all subjected to be able to participate in controlling their fate, of shaping a common future. Initially, the legitimacy of European integration was derived from the member states; in democratic terms, the legitimacy was a question of conferral of powers.3 The euro-​polity’s own legitimacy was seen to derive from its outputs: its ability to produce outcomes beneficial to the contracting partners. ‘Output oriented legitimation’ pertains to ‘governing for the people’, whereas democracy’s proper meaning is ‘government by the people’ (Scharpf 1999). The EU is, however, more than an international organization under international law. International organizations do not invoke own standards and criteria for legitimacy. Their legitimacy stems from the agreement between its members. In contrast, the EU is a power-​wielding polity, 35

Erik O. Eriksen

wielding power over its constituent members, and one that has established own standards of legitimacy.4 In contrast to international organizations, the EU has its own source of legitimacy through representative bodies that make decisions by qualified majority vote (QMV), and through the Court of Justice of the European Union (CJEU) having monopoly on the interpretation of treaties. The EU’s supranationalism started with the acceptance of a constitutional reading of the founding treaties, already in the 1950s, and explicitly in the 1960s. This reading transformed the European Economic Community (EEC) from an international regime into a quasi-​federal legal system based on the precepts of higher law constitutionalism. It was generated by the European Court of Justice in the combined doctrines of direct effect, which affirms the full legal character, under certain conditions, of EEC norms; and supremacy of Union law, stating that national norms must give way to Community ones when an irreducible conflict arises within the scope of application of the Treaties.5 The EU is in fact in many areas a superior political community affecting core state powers. There are hardly areas of core state powers left in the EU (Genschel and Jachtenfuchs 2018).The EU makes law to which most Europeans are bound through institutions committed to the Union itself. The Union amounts to a powerful commanding height (Eriksen 2014: 87). Wielding power, affecting citizens’ choice situations, governments generally stand under the obligation of legitimation. Due to the original right to freedom that attach to every person qua human being, when dominating power is exercised, when freedoms are affected, legitimation is required. Governments generally ‘impinge upon the freedom in two ways: they threaten fundamental rights and liberties of the sort liberals are concerned with, and they impose duties on us, in the way republicans traditionally have been concerned with’ (Richardson 2002: 34). The EU’s powers are in need of legitimation but so is DI, as it affects citizens’ autonomy. Hence, there is need for stipulations of the conditions under which DI, mutatis mutandis, would be mutually acceptable. Students of modern politics subscribe to the tenet that democracy is the sole remaining legitimation principle of political power. Of the long-​established authorities –​religion, monarchy, democracy and tradition –​it is only democratically enacted law that has survived the corrosion process of modernity (Frankenberg 2000). Notwithstanding, recent examples of backsliding into ‘illiberal democracy’, religion and tradition are spent resources. They do not make up the constitutional essentials of modern states. Rather, procedural forms of legitimation have replaced substantive, theocentric forms. Democracy establishes a procedure for legitimate legislation and has become the sole principle of political legitimation. Modern normative theories hold that the ‘exercise of political power is proper and hence justifiable only when it is exercised in accordance with a constitution’ the essentials of which all reasonable and rational citizens are expected to endorse (Rawls 1993: 217). Only reasonably acceptable laws may claim legitimacy.6 Democratic autonomy requires that those subjected to laws are at the same time their authors. When those subjected do not have the same rights and duties, when different policy fields and groups of countries are governed by different rules, there is a risk not only of legal uncertainty but of arbitrary rule as well; the rule of others, i.e., dominance. The basic legitimation problem, which political differentiation raises, is the exercise of arbitrary power as parties are obliged to follow laws, the making of which they cannot influence on an equal basis. The core of dominance is dependence on others’ unauthorized discretion. Dominance is an empirical indicator of illegitimate rule. It identifies legitimacy problems, and by implication also contains the resources of opposition, i.e., the critical standards evocative of freedom, of the right not to be dominated, of equal citizenship in a political community.

36

Legitimacy

Varieties of differentiation DI depicts institutional variation and different forms of association.7 It describes differences in sectoral, territorial or temporal dimensions, such as the concepts of Variable Geometry or Europe à la carte, Core Europe or the Europe of Concentric Circles and two or multi-​speed Europe, respectively. The temporal dimension is of particular interest from the point of view of legitimacy, as the different forms of association have to date been seen as interim arrangements in the advent of EU membership based on equal rights and statuses. If these arrangements are not actually stepping-​stones towards membership, we find ourselves in uncharted territory.That being the case, we can talk of a uniquely structured political order, in which the present level of political differentiation is upheld or increased even more with consequences for democratic self-​r ule. The idea of a Europe of different speeds is not a new one (see Curtin 1993). The Treaty of Rome contained several safeguard clauses and protocols dealing with derogations. In 1993, the Treaty of Maastricht introduced a pillar structure, giving rise to structural viability, with different policy sectors being governed by different rules. The creation of the European Monetary Union (EMU) led to a higher degree of differentiation. The enhanced cooperation model was entrenched in the Treaties of Amsterdam and Nice in 1997 and 2001, respectively; the former has a general clause opening for differentiation in some areas. Today, the EU consists of multiple overlapping groupings: not all countries are members of the Euro or of the Schengen. Currently, the Eurozone consists of only 19 out of 28 EU member states. In the Schengen Area of passport-​free travel, not all EU member states are included but some non-​members are. There are also transitional arrangements, which limit the newcomers’ access to the rights and benefits of membership. What is more, groups of countries can work together on defence, or decide to pass a new law because of flexibility in the EU rulebook. In the treaties, there are two primary legal provisions for DI. The first is Art. 114(4) Treaty on the Functioning of the European Union (TFEU), which allows member states, under strict conditions, to apply national rules after they have been outvoted by a qualified majority in the Council. Under certain conditions, members can opt out after a vote has been taken.The second is the provision for closer cooperation in Art. 20 of the Treaty on European Union (TEU). This provision allows a majority of member states to pursue deeper integration as a last resort, i.e., when there is no other way to pursue integration within the EU framework. According to Art. 329 TFEU, member states in the Council may unanimously vote to invoke Art. 20 upon a proposal from the Commission (requested by those states who wish to pursue enhanced cooperation) with the consent of the European Parliament. Art. 20 TEU addresses how decisions are taken when not all member states participate in the decision-​making process. It states that ‘only members of the Council representing the member states participating in enhanced cooperation shall take part in the vote’. Dimensions of differentiation can be separated. Vertical differentiation –​‘where policy areas have been integrated at different speeds and reached different levels of centralisation’ –​has to be distinguished from horizontal differentiation along a ‘territorial dimension’ in which ‘many integrated policies are neither uniformly nor exclusively valid in the EU’s member states’ (Schimmelfennig et al. 2015: 767). The EU’s internal horizontal differentiation is reflected in its external relations with its non-​members, ranging from the European Neighbourhood Policy to Turkey’s Customs Union, Switzerland’s bilateral approach, and the multilateral European Economic Area Agreement (EEA) for Norway, Iceland and Lichtenstein.

37

Erik O. Eriksen

Self-​inflicted harms Not only has the EU’s internal order become differentiated, so has the geographical scope of the Union’s political authority. Some non-​members are more EU members than others, and some are subjected to EU’s legal acts even though they are not members. The associated non-​members who are exempted from membership in core institutions are not exempted from the effects of these institutions’ decisions. What is more, they are also bound by them. The EEA Agreement is not merely an international agreement pertaining to ordinary politics. It is also a case of constitutional politics as it affects political autonomy: citizens’ possibilities to decide about the basic structure and competences of their political systems. The EEA Agreement is unusual, and more than just an international agreement in terms of status and reach. In effect, the EEA law is EU law, and when transposed into national law, it enjoys supremacy over other national legislation when there is conflict.8 The dynamic aspect of the EEA Agreement reflects the Agreement’s basic intention: the maintenance of a homogeneous market throughout the whole area. A principle of homogeneity applies. This principle trumps sovereignty by excluding the legal norms of the EEA countries as a legal basis for the interpretation of EU market rules (see Eriksen 2015). Only member state law counts as EU law. Regulations and directives are to be uniform and have the same effect across all EU member states as well as the EEA member countries. The homogeneity requirement can be found in the preamble to the EEA Agreement, but is also an ‘unwritten rule’ governing the Schengen and other agreements with the EU. By adopting the EEA Agreement, the associated non-​members have voluntarily subjected themselves to the EU by seeking access to the Single Market and committing themselves to adopting all the relevant EU regulations and accepting the jurisdiction of the CJEU.9 Rejecting the adoption of EU regulations would put the whole agreement in danger. At first, it may seem strange to describe the associated non-​members as subjected to the EU.10 In fact, they are often seen as rich free riders on European public goods that other Europeans have laboured with such difficulty to establish (cf. Tilly 1986). Nonetheless, these non-​members have voluntarily subjected themselves to the EU, which as an unintended consequence has become a hegemon. The EU was never intended to become a hegemon, quite the opposite, and it has invited associated non-​members to become members. It is also paradoxical that it is the states with the strongest and best-​entrenched democratic traditions –​states that are wealthy, well organized, and that qualify for EU membership –​that have opted not to become members. It may also seem strange to talk about self-​inflicted harms as subjection or dominance. Dependence on an arbitrary will is the defining feature of dominance and the main contrast to freedom. Acts of interference may be instances of dominance, even if they are for our own good. Kant’s dictum is that right must never bend its knee before expediency by prioritizing considerations of advantage over political freedom. That persons should be protected against humiliation as well as against being instrumentalized for other purposes, implies protection of a person against their own actions. The associated non-​members have restricted their freedom for economic reasons when other options were available. Dominance or subjection is the effect, regardless of how voluntary or beneficial the arrangement may be. One may be subjected to or dominated by someone or something whether or not one has consented to it.11 The EEA countries encroach on their own freedom because they prioritize access to the European common goods while not being EU members. They manage their relations with the EU through a contractual framework under international law; a framework premised on the formal equality between the contracting states. The EEA members are unable to back

38

Legitimacy

their claims towards the Community with credible threats –​external sanctions or reciprocity –​ whereas the EU could unilaterally destroy the whole arrangement with negligible costs. The associated non-​members are free to annul their arrangements with the EU but have chosen not to face the costs of such an act. Exit options have so far been perceived as unrealistic. The EEA Agreement is an asymmetric power relationship. The EEA Agreement is a contract between very different parties: a Union with about 450 million inhabitants, on the one hand, and certain very small countries that are dependent on the EU for economic reasons on the other. The associated non-​members are in an inferior position because there is no parity of power to render the use of threats and counter-​measures credible under international law, nor can they participate in joint decision-​making to wield influence over, or demand justifications under EU law. They can only make their voice heard through membership in some expert committees, through consultancy mechanisms and lobbying. The arrangements curtail the sovereignty of their citizens and infringe upon their right to self-​determination. The EEA members are thus ruled by others’ law –​made by EU members –​not by themselves; hence, their semi-​colonial status. However, exclusion and disenfranchisement are self-​imposed, which is hardly the case in the next case under consideration.

Pre-​emption of choice Differentiation in the Eurozone has been seen as temporary (see Adler-​Nissen 2014). The states that are not members of the core, of the Eurozone, have been under an obligation to join at some later stage. The Eurozone crisis reinforced DI, and the development of the Monetary Union in its aftermath is the clearest example of DI in the vertical dimension. Before the 2008 financial crisis, differentiated arrangements of integration were not considered a defining trait of the system, since the exemptions states had obtained were temporary. Integration was a ‘one-​way street’, albeit with a variable speed limit. The Eurozone crisis changed everything. In terms of agreements, the Treaty on Stability, Coordination and Governance in the Economic and Monetary Union –​the Fiscal Compact –​formalized the distinction between members and non-​members of the Eurozone, while fundamental issues regarding the EU’s competences have been executed outside EU treaties. The Eurozone-​specific regime of economic governance amounts to a segment of specialized, not general, decision-​making structures (Eriksen 2018). It involves an incongruence between the Single Market and the single currency as well as between debtors and creditors in the Eurozone. As a consequence, the citizens’ representatives are highly uneven depending on which representatives they represent. The ability to control the political agenda is not level. The debtors’ freedom of choice is limited because their unbalanced budgets compel them to accept conditions imposed by the financial assistance programs. Loans and credit are conditional on reforms that are not initiated by citizens’ representatives or justified to the affected citizens. Quite the contrary, governments are required to write a doctrine of fiscal responsibility into binding legislation. Some choice opportunities and alternatives have been removed from the political agenda. There is a pre-​emption of choice. These arrangements for monitoring and ensuring compliance thus severely limit the fiscal policy space of the member states, which have already ceded national control of interest rates and currency exchange rates by adopting the Euro. Crisis-​r idden states that are members of the Eurozone cannot devaluate their currency and have little room for overspending in order to recover from recession. Macro-​economic alternatives are no longer available.To choose between macroeconomic alternatives like that of Keynesianism versus Monetarism, investment versus austerity, is not possible. In this structure, the debtors are placed in a subjected position: they are 39

Erik O. Eriksen

unable to control or moderate the causes of their own vulnerability. However, this development goes against the normative expectation of the European integration process, where every integrationist step usually is complemented with a democratic upgrade.

Hindrance to co-​legislation The EU is not a state with a monopoly on taxation and power, but an experiment in curbing dominance in international relations with peaceful means. It aims to overcome the anarchic state of international affairs, to prevent war, foster trust and secure peace through encompassing cooperation. Interdependence and lack of legal constraints increase states’ ability to impose negative externalities on other states and decrease their ability to provide their own citizens with the positive externalities of public goods that benefit more than just those who pay for them (Collignon 2003; Lord 2017). Greater interconnectedness and interdependence without due process affect the ability of citizens to govern themselves through institutions they can control. DI creates new opportunities for some actors or groups to exploit others by free-​riding on their provision of positive externalities. In order to reduce moral hazard, political integration is required. Binding collective decision-​making within politically integrated orders reduces transaction costs and free-​r iding, overcomes commitment and collective action problems and founds public goods, including peace and rule-​based resolution of conflicts. Only democratic orders can, however, banish dominance. Due to the Eurozone crisis, the protracted European democratization process has come to a standstill and political differentiation is catching on. Problems, solutions and participants are linked together in specialized decision-​ making structures, in segments. A segmented political order in Europe is a source of arbitrary power because it decouples the decision-​making structure from the (democratic) access structure. When the mapping of choices onto decision-​makers is decoupled from the mapping of problems onto choices, there is arbitrary rule. Eriksen 2019:13 The changes to the Eurozone have segmented the management of decisions by removing them from parliamentary agendas and by compartmentalizing them in convoluted, technocratic-​ driven decision-​making processes under the auspices of the European Council (see Eriksen 2019, 71ff; Bátora and Fossum 2020). Thus, in the EU, dominance emerges as arbitrary interference in zones of public autonomy, viz., the type of autonomy constituted of basic liberties that are indispensable to free and equal persons coordinating their common affairs through the public use of reason and egalitarian structures of decision-​making. Decisional exclusion and pre-​emption of choice resulting from the handling of the Eurozone crisis testify to a type of dominance that ultimately affects citizens’ public autonomy. Segmented structures are hindrances to co-​legislation as they undercut citizens’ democratic rights to participate on an equal basis in opinion and will-​formation processes. They affect the very political status of the citizens.12 Decisional exclusion is the general problem with enhanced cooperation, which allow only members of the Council representing the member states participating in it the right to vote (Art. 20 TEU). There are two types of decisional exclusion: in the external, horizontal dimension of DI, the associated non-​members are excluded from the decision-​making fora; they are decision takers, not decision makers. In the vertical dimension, DI creates large asymmetries of power and voting rights between member states. Some of them get to be de facto excluded from decisions that are in practice largely made the powerful. 40

Legitimacy

When the citizens lack the instruments of influence, their political rights are being curtailed. There is, however, opposition also from within the institutional complex of the EU. Already in 2013, The European Parliament stated that: [A]‌ny formal differentiation of parliamentary participation rights with regard to the origin of Members of the European Parliament represents discrimination on grounds of nationality, the prohibition of which is a founding principle of the European Union, and violates the principle of equality of Union citizens as enshrined in Article 9 TEU. European Parliament 2013, Art. 2913 Moreover, the Fiscal Compact states that within five years at most following the entry into force of this Treaty, on the basis of an assessment of the experience with its implementation, the necessary steps shall be taken […] with the aim of incorporating the substance of this Treaty into the legal framework of the European Union. European Council 2012: Art. 16 Hence, differentiation is not meant to be permanent. Thus, can DI and the democratic setbacks it represents be legitimate?

Temporary delays As mentioned, DI describes differences in sectoral, territorial or temporal dimensions. The later dimension is of particular interest, as the different forms of association have to date been seen as interim arrangements in the advent of EU membership based on equal rights. If these arrangements are actually stepping-​stones towards membership, we may see political differentiation as a vehicle for political integration. Kant, who is known for offering only a vision of a perfectly just society, opens, in fact, for the moving of societies out of injustice and into an ideal prescribed by justice –​a rightful political order –​by considering empirical constraints. By introducing the category of permissive law of public right, Kant made possible the integration between morality and political knowledge.14 Public right concerns the juridical relationship between a state and its own members (or between states) and involve ‘a permissive principle (lex permissiva) of practical reason’ (Kant 1797/​1996: 406). This principle authorizes the temporary delay of a necessary reorganization of an unjust order when the ‘implementation of immediate reform would counteract the ruler’s duty to reform the legal order as a whole’ (Weinrib 2013: 108). True politics draw on empirical knowledge and prudent judgement of the circumstances under which the existing legal system can be brought into conformity with its own standard of adequacy. Since the severing of a bond of civil or cosmopolitan union even before a better constitution is ready to take its place is contrary to all political prudence, which agrees with morals in this, it would indeed be absurd to require that those defects be altered at once and violently; but it can be required of the one in power that he at least take to heart the maxim that such an alteration is necessary, in order to keep constantly approaching the end (of the best constitution in accordance with laws of right) Kant 1795/​1996: 8:372 41

Erik O. Eriksen

Permissive rights exempt action from necessity and permit the postponement of reform until conditions are favourable. It does so without losing sight of the end. This permissive principle justifies delays but not status quo. It makes clear that suboptimal solutions may be legitimate. This principle can be applied to DI and be specified as a legitimation theorem with a time and knowledge index: DI would be acceptable for affected parties at a particular moment in time and corresponding to the actual level of knowledge.15 This makes clear that under certain conditions DI, with its democratic deficiencies, would satisfy the presumption of being legitimate, it would be seen to represent a common interest. The time index of this theorem awaits the requisite solidaristic basis for integration to come about, viz., the we-​feeling, the learning and enlightenment about what is at stake and what is in the citizens’ long-​term interest. In other words, to await the creation of the civic bonds necessary to sustain democratic process of integration. DI would only be legitimate for a period of time. The knowledge index links the legitimacy of DI to the current level of information, about what is known about popular sentiments, power constellations, the management of externalities, the reaping of benefits of public goods, about new avenues for problem-​solving, etc. Under such conditions, DI could be seen to embody a common interest, that is, it is valid for everyone affected. There are thus arguments in favour of the internal differentiation between a political Euro-​ Union working ever closer together –​a Core Europe –​and a periphery of hesitant member states that can join the core at any time provided that the integration ‘telos’ is not undermined. However, the right to freedom, to not being dominated, makes integration in Europe a duty.16 The level of interdependence and interconnectedness, the externalities and the general affectedness of citizens, the many cross border problems that need to be addressed, among other things, cater for integration. To paraphrase Kant, actors whose conduct regularly affects others’ rights, and who refuse to abolish an unjust order –​the unregulated ‘state of nature’ –​and bring about a rightful political-​legal order, are collectively breaking their natural duty of justice (see also Waldron 2002). When people live close by and/​or cannot avoid affecting one another’s rights, they have a duty to establish a fair scheme of cooperation; a rightful order. Thus, any group of individuals whose acts affect one another’s rights and who do not currently have a set of legal and political institutions are obliged to create them (Stilz 2009: 199). Therefore, the normative achievements of the European integration project must not be forfeited. These achievements include the removal of European states’ self-​help means of reciprocity and countermeasures. The member states can no longer take the law into their own hands. The EU has thus domesticated international relations in Europe. Moreover, as mentioned the EU makes law to which most Europeans are bound through institutions committed to the Union itself. The EU has established representative bodies that make European wide decisions. QMV, the abolishment of veto in many areas, means that the Members have accepted to be outvoted. The states are in principle no longer veto players in the regulation of practical affairs in Europe. Unlike international organizations, the EU has a constitutional identity and a legal personality. The European Treaties are vital for rights’ protection and have the function of a constitution: they establish both a unitary European citizenry distinct from national ones and a set of autonomous European bodies. The EU has established rights and provisions that amount to more than non-​excludable public goods.17 Rights involve principles, which are inalienable by their deontological nature, and are, as such, conditions for the normative regulation of practical affairs. These rights and provisions make up a European basic structure, viz., ‘a system for producing benefits by making possible and encouraging forms of interaction through which material benefits are produced’ (Scanlon 2018: 141).18 Such a structure made up of inviolable rights –​erga omnes norms –​and a

42

Legitimacy

political system premised on the separation of powers constitutes a public coercive framework for claims making, justification and decision-​making.19 A return to a Europe of independent nation states would eradicate these normative achievements. Under present conditions of cascading interdependence, the proliferation of problems and interstate conflict, disintegration would increase transaction costs, negative externalities and moral hazard. It would also have democratic costs. No public coercive framework would be in place capable of tracking the interests, views or wills of the citizens, nor of ensuring equal political rights of affected parties European wide. Nor would there be European public goods to reap the benefits of.

Conditional legitimacy From a normative perspective, disintegration would thus not be allowed: ‘Once democratic rights and institutions have been established and individuals have been awarded a supranational citizen status, there is no legitimate way back to the status quo ante’ (Patberg 2020: 594). From this perspective, DI would only be conditionally legitimate. The type of differentiation, which entails infringements of democratic rights, is intrinsically linked to the possibility of and the need for renewed integration (ibid). The permissive principle, which justifies delays but not status quo, makes clear that only under certain conditions would differentiation be legitimate. Ultimately, the presumption of legitimate DI is warranted if the consequences and side effects of its observance under unchanging conditions can be accepted by subjected parties. When specified as a legitimation theorem, conditions for legitimate, temporary variants of DI can be arrived at, i.e., as long as opt-​outs/​escape clauses do not: • • •

obstruct the duty to establish a fair scheme of cooperation; a rightful order foreclose future membership and/​or Treaty change undermine the reasons for seeking membership/​for Treaty change20

The benefits of DI should not outweigh the benefits of integration; they should not undermine the creation of European public goods. Moreover, DI would have to: • • •

be open and transparent also on the downsides of not being full member involve ongoing debate about EU membership or Treaty change include institutionalized possibilities for review and revision of position

Under such conditions, DI would be a temporary phenomenon, and seen as part of a long-​term plan for abolishing the present forms of illicit divergence and dominance. In due time, the presumption is that people will learn and appreciate the effects of European integration and abolish the self-​imposed abandonment of being involved in politically shaping the future. One may question whether present arrangements comply with these conditions. The condition of institutionalized possibilities for review and revision of position is hardly complied with. With reference to vertical integration, Norway’s last referendum was in 1994, and a new one is not on the table. The same is the case with the much needed and promised Eurozone Treaty change. In 2017, the Commission put forward a number of ideas for reform of EMU. These include a European Monetary Fund to replace the intergovernmental ESM, a specific euro area budget; unified Eurozone external representation; a full-​time permanent chair for the

43

Erik O. Eriksen

Eurogroup; and integration of the Fiscal Compact into EU law (European Commission 2017). It was met with resistance from the northern member states (Middelaar 2019: 38). There is the danger that temporary arrangements, which are means of handling various forms of crises and various forms of opposition to integration at particular moments in time, become permanent. Still, the decision July 2020 of the European Council to adopt common European bonds for a recovery fund may be the right move to bring about change.21

Second-​class Europeans The indirect legitimation theorem derived from the permissive law of public right allows for the moral possibility of an intermediate order and provisory arrangements. This view sits with Kant’s general understanding of being in a transitional stage –​between international and cosmopolitan law. Here an avant-​garde of peaceful republics could take the lead: ‘For if good fortune should ordain that a powerful and enlightened people can form itself into a republic […], this would provide a focal point of federative union for other states […] it would gradually extend further and further’ (Kant 1795/​1996: 327, 8:356). For today’s avant-​garde spear headed by France and Germany, and underscored by the 2017 European Commission White Paper on the Future of Europe,22 DI is also an important strategic option for the reform and future development of the EU (Schimmelfennig 2019: 176). In this perspective, the avant-​gardists may have an obligation to slow down integration rather than catching up in order to avoid excessive widening of differences between those fully in the shared system of rule and those who are not. However, the idea of a two-​speed Europe is not shared by all European leaders. States that risk being left on the periphery are reacting negatively to such an idea. There is fear of new divisions and dividing lines in Europe, of first-​and second-​class Europeans.23 Leaders from Hungary, Czech Republic and Poland have expressed concern over the possible shift to a multi-​speed Europe, worried that it could sideline their views (Strzelecki 2017). Denmark and Greece have also objected to the idea of a two-​tier Europe, fearing they would be treated as outsiders and excluded from important decision-​making forums. Another concern regarding the idea of a differentiated Europe is that if the EU members are not able to agree on common principles, the Union will eventually fall apart. For some, differentiation is thus the road to disintegration. The reactions to a potential formalization of a two-​tiered Europe attest to the struggle for recognition. Those who would be left in the periphery are worried. They fare that not only their views and interests would be sidelined and curtailed, but also their ability to partake in the shaping of a future. Such a shift towards a two-​tiered Europe would affect members’ status; it would violate the principle equal European citizenship and frustrate expectations. There is also a historic background for the reactions. In the words of EU President Donald Tusk at the ceremony of the 60th anniversary of the Treaties of Rome 25/​03/​2017: I lived behind the Iron Curtain for more than half of my life, where it was forbidden to even dream about those values.Yes, back then, that really was a two-​speed Europe.24 A two-​speed Europe entails that European citizens and their representatives would not enjoy equal standing –​their statuses would be different, hence there is a problem of recognition. DI may then become the reason for those in the periphery to seek membership in Core Europe. Could the force misrecognition cause change, and hence could differentiation become a vehicle for integration (Eriksen 2019: 235ff)? There may, however, also be reasons for accepting political differentiation as a temporary arrangement. 44

Legitimacy

Mitigating reasons The indirect legitimation theorem with a time-​and knowledge index makes DI’s acceptability conditional on parties experience and pertinent knowledge of power constellations. The legitimacy of DI depends on the knowledge of its consequences for parties’ statuses, interests and values. When it is well-​known that interdependence is intense and asymmetric and value conflicts thrive, when capacity is low and interests diverse, DI may be a justifiable solution to the predicaments. DI would enable cooperation in specific fields, lower transactions costs and reap some of the benefits of European public goods. At the same time, it would reduce the need to use the veto power and bargaining muscles to protect national interests. It would reduce the danger of intense and permanent minorities. Consequently, DI may be seen to express a common interest. Some mutually acceptable reasons for DI with a time limit can be identified, such as: • • • • • • •

when deep national disagreement prevails, when popular sentiments and care for identities and ‘sovereignty’, for a specific way of life, persist, when DI would reduce the danger of intense and permanent minorities, when DI would make possible choices actors would otherwise be unable to make, when DI is a means for managing diversity between the actors without recourse to power, when DI does not allow cherry-​picking public goods, when DI prevents race-​to-​the-​bottom deregulation, harmful tax competition and competitive currency devaluations, when DI helps producing positive externalities; when DI assists addressing collective action problems such as climate change,

Even though DI subjects parties to a constitutional order they cannot amend, they may find that preferable to non-​association. This is so, first, because the essentials of the EU’s constitutional order is not alien, it corresponds to national constitutions, and parties are in principle free to abolish DI arrangements. Second, because DI gives parties access to a legally regulated international order, to a rights-​based order encompassing most Europeans (see Blichner 2008). DI lends support to the principle of a legally domesticated international order. Apart from the many economic and political reasons that count in favour of EU links, organized relationships with a voluntary union of European people, where entrenched rules have replaced might, would ceteris paribus be preferable to being dependent on power-​based international orders. The associated non-​members have become free riders on a rights-​based political order, an order which assigns rights also to laggards, which ensures stability and predictability, which solves disputes through institutionalized procedures. Differentiation would thus, under certain conditions, be perceived not as a threat to the European project, but rather as a vehicle of integration.The indirect legitimation theorem makes DI arrangements operate under the presumption of legitimacy only as far as it attends to collective interests for the time being and according to the present state of knowledge.

Conclusion DI is a puzzle from the point of view of legitimacy. On the one hand, DI is associated with illicit rule as people are subjected to laws, the making of which they cannot influence on an equal basis. Some of those affected by the decisions of the polity are excluded and disenfranchised. Therefore, the political autonomy of the members is affected; hence there is a case for dominance. Dominance occurs when citizens are subjected to the will of others, when they do not 45

Erik O. Eriksen

have equal opportunity to wield political influence. Differentiation is thus not an innocent instrument for handling conflicts in interconnected contexts. It undermines the fundamental conditions of democratic self-​rule. On the other hand, DI allows for cooperation even when there is disagreement and when mutual dependence is unequal. DI enables cooperation in specific fields, lowers transactions costs and enables collective action. From the point of view of efficiency, DI is highly recommended. Moreover, as it fosters cooperation and sometimes learning about the European integration process, DI may also have normative value. Specifying the permissive law of public right into a legitimation theorem with a time and knowledge index makes it possible to establish the conditions under which DI may enjoy the presumption of being mutually acceptable for affected parties. This index ties the legitimacy of DI to the level of knowledge and to a particular point in time. The legitimation theorem is an antidote to utopian prescriptionism commonly associated with Kant. Suboptimal arrangements may be legitimate.

Notes 1 The research leading to these results stems from the EU3D project (EU Differentiation, Dominance and Democracy), which has received funding from the European Union’s Horizon 2020 Research and Innovation Programme under Grant Agreement No. 822419. It draws on Eriksen 2019 and Eriksen 2022, forthcoming. I am grateful for comments by Christopher Lord and Markus Patberg. 2 They argued in favour of a “Core Europe” (Habermas and Derrida 2003). 3 According to the principle of conferral, the EU is a union of member states, and all its competences are voluntarily conferred on it by its member states. 4 One may question whether the ability to produce outcomes constitute a legitimation standard at all, as outputs themselves are in need of legitimation. The ability to produce outcomes says something about the effectiveness and efficiency of a system not its legitimacy.That is, irrespective of how effective a system is, it must be reformed if it is unjust (Rawls 1971). Illegitimate orders are unjust, and thus unstable. 5 See the leading cases 26/​62 Algemene Transporten Expeditie Onderneming Van Dend en Loos v. Nederlandse Administratie der Belastingen [1963] and 6/​64 Flaminio Costa v ENEL [1964]. 6 ‘Only those statutes may claim legitimacy that can meet with the assent of all citizens in a discursive process of legislation that in turn has been legally constituted’ (Habermas 1996: 110). 7 For an overview of political differentiation, see Leruth and Lord (2015), see further Fabbrini (2015); Leuffen, Rittberger, and Schimmelfennig (2013); Lord (2017); Fossum (2015); Genschel and Jachtenfucs (2018); De Witte, Ott and Vos (2017); Schimmelfennig (2019). 8 Henrik Bull refers to a long list of sources claiming the EEA Agreement involves the same principles of precedence and direct effect as EU law, but the EFTA Court has not been willing to accept this (Bull 2011: 260). 9 About 75 per cent of EU laws and regulations, none of which Norwegians have an opportunity to influence, apply to Norway. Since 1994, Norway has adopted 11,013 EU Directives, and rejected none (Norwegian Ministry of Foreign Affairs, 2016: 9). 10 Other states involved in partnership arrangements, for example, Ukraine and Turkey, are in a more vulnerable position, and are also affected by, and indeed dependent on, the EU. However, when the affected countries are not members of the Single Market, of the Eurozone or of Schengen, they are not subjected to EU’s regulative framework. Hence, their political or constitutional status is not affected even though their interests are. 11 For the associated non-​members, their own choices caused their subservience. Hence, we have to do with self-​inflicted dominance.The associated non-​members have prioritized access to the single market over political autonomy in terms of the ability to live by self-​determined laws. Such a self-​inflicted harm is in breach of the Kantian approach, in which the means do not justify the ends and freedom can only be restricted for the sake of freedom itself (Eriksen 2019: 111). 12 There are several examples of breaches of established principles of co-​decision and legal equality. The Greek bailout, the conditional financial assistance, the establishment of the EFSF in 2010, the role and operations of the so-​called Troika interfere with members’ rights in a way not prescribed by neither 46

Legitimacy

national nor EU law. The Excessive Imbalances Procedure (EIP) is seen to undermine the EU’s constitutional authority (Joerges and Weimer 2013). The Securities Market Programme (SMP) initiated in 2010 as well as the Outright Monetary Transactions (OMT) programme announced in 2012 have been subjected to harsh criticism. It has been seen as a manifest case of transgression of the ECB’s competencies as bond-​buying programmes represent indirect fiscal assistance (see e.g., Schelke 2014; Everson 2015). (See also Eriksen 2018, 2020). 13 The European Parliament has noted that ‘the democratic credibility of European integration has suffered enormously from the manner in which the euro crisis has been dealt with to date’ (European Parliament 2011: I). 14 By this move Kant counters the well-​ known allegation of impotence of ‘ought’, of powerless prescriptionism, in his moral theory. 15 For an analogy see Günther (1993: 36). 16 Holocaust itself can be seen as an independent reason for integration. In this vein, Theodore W Adorno reformulated Kant’s categorical imperative: “think and act in such a way that the history of Auschwitz will not be repeated” (Adorno 1980: 358, author’s translation). 17 For terminology see Kölliker (2001). 18 Rights are not simply ‘goods’: ‘They cannot be assimilated to distributive goods without forfeiting their deontological meaning’. They cannot be possessed like things as they ‘regulate relations between the actors’ (Habermas 1998: 54). See also Young (1990: 25). 19 Hence, ‘managing externalities’ (Lord 2015: 788) does not suffice as an evaluating standard for DI. Such a management presupposes a basic structure. It is the existence of a basic structure that makes the managing of externalities possible. 20 Escape clauses should not ‘limit the degree to which states integrate by offering them an opt-​out that would not otherwise exist’ (Jensen and Slapin 2012: 782). 21 The Franco-​German proposed Covid-​19 recovery fund build on long-​term EU borrowings, leaves the austerity politics of the Euro zone behind. The decision of the European Council to adopt common European bonds is remarkable in the history of European integration. 500 billion of euro grants aimed at helping the most needy member states in the form of non-​repayable grants can bring about the most important integrationist step since Maastricht. 22 The paper on the Future of Europe presented five possible scenarios one of which was labelled ‘those who want more do more’ and outlined a multi-​speed model of differentiation for the EU (European Commission 2017: 20f.). 23 The Visegrád group –​comprising Czech Republic, Hungary, Poland and Slovakia –​was alarmed (Wintour 2017). 24 www.consilium.europa.eu/​en/​press/​press-​releases/​2017/​03/​25/​tusk-​ceremony-​rome-​speech/​.

References Adler-​Nissen, R. (2014) Opting Out of the European Union, Cambridge: Cambridge University Press. Adorno, T. W. (1980) ‘Negative Dialektik’. In: T. W. Adorno (ed) Gesammelte Schriften bd 5, Frankfurt. Bátora, J. and Fossum, J. E. (2020) (eds) Towards a Segmented European Political Order, Abingdon: Routledge. Blichner, L. (2008) Five possible reasons for accepting the EEA agreement (Fem mulige grunner til å akseptere EØS-​avtalen), Nytt Norsk Tidsskrift 4, pp. 359–​367. Bull, H. (2011) ‘One Market,Two Courts; the Case Law of the EFTA Court’. In: F. Carbonell, A. J. Menéndez and J. E. Fossum (eds) Hope, Reluctance or Fear? The Democratic Consequences of Case Law of the European Court of Justice, ARENA Report 5/​11, Oslo: ARENA, pp. 253–​274. Christiano, T. (2010) ‘Democratic Legitimacy and International Institutions’. In: S. Besson and J. Tasioulis (eds) The Philosophy of International Law, Oxford: Oxford University Press, pp. 119–​137. Collignon, S. (2003) The European Republic, London: The Federal Trust, Bertelsmann Foundation. Curtin, D. (1993) The constitutional structures of the union: a Europe of bits and pieces, Common Market Law Review, 30, pp. 17–​69. De Witte, B., Ott,A. and Vos, E. (eds) (2017) Between Flexibility and Disintegration:The Trajectory of Differentiation in EU Law, Cheltenham: Edward Elgar Publishing. Eriksen, E. O. (2014) The Normativity of the European Union, London: Palgrave Macmillan. Eriksen, E. O. (2015) ‘Despoiling Norwegian Democracy’. In: E. O. Eriksen and J. E. Fossum (eds) The European Union’s Non-​Members: Independence under Hegemony?, London: Routledge, pp. 77–​101. 47

Erik O. Eriksen

Eriksen, E. O. (2018) Political differentiation and the problem of dominance: segmentation and hegemony, European Journal of Political Research, 57(4), pp. 989–​1008. doi: 10.1111/​1475-​6765.12263. Eriksen, E. O. (2019) Contesting Political Differentiation: European Division and the Problem of Dominance, London: Palgrave Macmillan. Eriksen, E. O. (2020) Banishing dominance in Europe: the case for regional cosmopolitanism. European Journal of International Relations, 26(3), pp. 742–​766, online first, doi: 10.1177/​1354066119882067. Eriksen, E. O (2022) Justifying the imperfect. Differentiation and the problem of the second best. Ratio Juris, forthcoming. European Commission (2017) White Paper on the Future of Europe: Reflections and Scenarios for the EU27 by 2025, COM(2017)2025 of 1 March 2017. European Council (2012) Treaty on the Stability, Coordination and Governance in the Economic and Monetary Union. Available at: www.consilium.europa.eu/​media/​20399/​st00tscg26_​en12.pdf (accessed 27 August 2020). European Parliament (2011) European Parliament Resolution of 1 December 2011 on the European Semester for Economic Policy Coordination 2011/​2071(INI). Available at: www.europarl.europa.eu/​sides/​getDoc. do?type=​TA&reference=​P7-​TA-​2011-​0542&language=​EN (accessed 27 August 2020). European Parliament (2013) European Parliament Resolution of 12 December 2013 on Constitutional Problems of a Multitier Governance in the European Union. Available at: www.europarl.europa.eu/​sides/​getDoc. do?type=​REPORT&reference=​A7-​2013-​0372&language=​EN (accessed 27 August 2020). Everson,M.(2015) An exercise in legal honesty:rewriting the court of justice and the Bundesverfassungsgericht, European Law Journal, 21(4), pp. 474–​499. doi: 10.1111/​eulj.12144. Fabbrini, S. (2015) Which European Union?, Cambridge: Cambridge University Press. Fasone, C. (2014) European Economic Governance and Parliamentary Representation: what place for the European Parliament?, European Law Journal, 20(2), pp. 164–​185. doi: 10.1111/​eulj.12069. Fossum, J. E. (2015) Democracy and differentiation in Europe, Journal of European Public Policy, 22(6), pp. 799–​815. doi: 10.1080/​13501763.2015.1020838. Frankenberg, G. (2000) The return of the contract: problems and pitfalls of European Constitutionalism, European Law Journal, 12(1), pp. 39–​60. doi: 10.1080/​09615768.2001.11427528. Genschel, P. and Jachtenfuchs, M. (2018) From market integration to core state powers: the Eurozone crisis, the refugee crisis and integration theory, Journal of Common Market Studies, 56(1), pp. 178–​196. doi: 10.1111/​jcms.12654. Günther, K. (1993) The Sense of Appropriateness: Application Discourses in Morality and Law, New York, NY: State University of New York Press. Habermas, J. (1996) Between Facts and Norms: Contributions to a Discourse Theory of Law, Cambridge, MA: MIT Press. Habermas, J. (1998) The Inclusion of the Other: Studies in Political Theory, Cambridge, MA: MIT Press. Habermas, J. and Derrida, J. (2003) February 15, or what binds Europeans together: a plea for a common foreign policy, beginning in the core of Europe, Constellations, 0(3), pp. 291–​ 297. doi: 10.1111/​ 1467-​8675.00333. Jensen, C. B. and Slapin, J. B. (2012) Institutional hokey-​pokey: the politics of multispeed integration in the European Union, Journal of European Public Policy, 19(6), pp. 779–​795, doi: 10.1080/​13501763. 2011.610694. Joerges, C. and Weimer, M. (2013) ‘A Crisis of Executive Managerialism in the EU: No Alternative?’. In: D. M. Trubek, G. Búrca, C. Kilpatrick and J. Scott (eds) Critical Legal Perspectives on Global Governance: Liber Amicorum David M Trubek, Oxford: Hart Publishing, pp. 295–​322. Kant, I. (1795/​1996) ‘Toward Perpetual Peace’. In: M. Greogor (ed) Practical Philosophy, Cambridge: Cambridge University Press, pp. 311–​352. Kant, I. (1797/​ 1996) ‘The Metaphysics of Morals’. In: M. Greogor (ed) Practical Philosophy, Cambridge: Cambridge University Press, pp. 37–​108. Kölliker, A. (2001) Bringing together or driving apart the Union? Towards a theory of differentiated integration, West European Politics, 24(4), pp. 125–​151 doi: 10.1080/​01402380108425468. Leruth, B. and Lord, C. (2015) Differentiated integration in the European Union: a concept, a process, a system or a theory?, Journal of European Public Policy, 22(6), pp. 754–​763. doi: 10.1080/​13501763. 2015.1021196. Leuffen, D., Rittberger, B. and Schimmelfennig, F. (2013) Differentiated Integration: Explaining Variation in the European Union, Basingstoke: Palgrave Macmillan.

48

Legitimacy

Lord, C. (2015) Utopia or dystopia? Towards a normative analysis of differentiated integration, Journal of European Public Policy, 22(6), pp. 783–​798. doi: 10.1080/​13501763.2015.1020839. Lord, C. (2017) An indirect legitimacy argument for a directly elected European Parliament, European Journal of Political Research, 56(3), pp. 512–​528. doi:10.1111/​1475-​6765.12204. Norwegian Ministry of Foreign Affairs (2016) ‘EU/​EØS-​håndboken’ [The EEA Handbook]. Available at: www.regjeringen.no/​globalassets/​departementene/​ud/​vedlegg/‌​‌europapolitikk/​eu_​eos_​handbok 2016.pdf (accessed 10 May 2017). Offe, C. and Preuss, U. K. (2016) Citizens in Europe. Essays on Democracy, Constitutionalism and European Integration, Colchester: ECPR Press. Patberg, M. (2020) Can disintegration be democratic? The European Union between legitimate change and regression, Political Studies, 68(3), pp. 582–​599. doi: 10.1177/​0032321719870431. Pettit, P. (2012) On the People’s Terms: A Republican Theory and Model of Democracy (The Seeley Lectures), Cambridge: Cambridge University Press. Rawls, J. (1971) A Theory of Justice, Oxford: Oxford University Press. Rawls, J. (1993) Political Liberalism, New York, NY: Colombia University Press. Richardson, H. S. (2002) Democratic Autonomy: Public Reasoning about the End of Policy, Oxford: Oxford University Press. Scanlon, T. M. (2018) Why Does Inequality Matter, Oxford: Oxford University Press. Scharpf, F. W. (1999) Governing in Europe: Effective and Democratic, Oxford: Oxford University Press. Schelke, W. (2014) ‘Fiscal Integration by Default’. In: P. Genschel and M. Jachtenfuchs (eds) Beyond the Regulatory Polity? The European Integration of Core State Powers, Oxford: Oxford University Press, pp. 105–​123. Schimmelfennig, F. (2019) The choice for differentiated Europe: an intergovernmentalist theoretical framework, Comparative European Politics, 17(2), pp. 176–​191. doi: 10.1057/​s41295-​019-​00166-​5. Schimmelfennig, F., Leuffen, D. and Rittberger, B. (2015) The European Union as a system of differentiated integration: interdependence, politicization and differentiation, Journal of European Public Policy, 22(6), pp. 764–​782. doi: 10.1080/​13501763.2015.1020835. Stilz, A. (2009) Liberal Loyalty, Freedom, Obligation, and the State, Princeton, NJ: Princeton University Press. Strzelecki, M. (2017) Poland Talks Tough against ‘Multi-​Speed’ EU Ahead of Rome Summit, Bloomberg, 23 March. Available at: www.bloomberg.com/​politics/​articles/​2017-​03-​23/​poland-​talks-​tough-​against-​ multi-​speed-​eu-​ahead-​of-​rome-​summit (accessed 24 May 2017). Tilly, C. (1986) European violence and collective action since 1700, Social Research, 53(1), pp. 158–​184. Van Middelaar, L. (2019) ‘The Lisbon Treaty in a Decade of Crises:The EU’s New Political Executive’. In: A. Södersten (ed) The Lisbon Treaty 10 Years on: Success of Failure, Stockholm: Sieps, pp. 17–​23. Waldron, J. (2002) Redressing historic injustice, University ofToronto Law Journal, 52(1): 136–​160. doi: 10.2307/​ 825930. Weinrib, J. (2013) Permissive laws and the dynamism of Kantian Justice, Law and Philosophy, 33(1), pp. 105–​136. Wintour, P. (2017) Plans for two-​speed EU risk split with ‘peripheral’ members, The Guardian, 14 February. Available at: www.theguardian.com/​world/​2017/​feb/​14/​plans-​for-​two-​speed-​eu-​r isk-​split-​with-​ peripheral-​members (accessed 24 May 2017). Young, I. M. (1990) Justice and the Politics of Difference, Princeton, NJ: Princeton University Press.

49

4 Differentiated (dis)integration beyond Europe A comparative regionalism approach1 Stefan Gänzle and Jens Uwe Wunderlich

Introduction ‘United in diversity’ This notorious motto –​both concise and seminal –​found its way into the official vocabulary of the European Union (EU) for the first time in 2000. In a nutshell,‘[i]‌t signifies how Europeans have come together, in the form of the EU, to work for peace and prosperity, while at the same time being enriched by the continent’s many different cultures, traditions and languages’.2 Yet, the expression itself certainly is much older. In the context of European integration, for example, it came to be used in one of the so-​called Marshall Plan films as part of the US-​sponsored European Recovery Program (ERP). These films were, produced between 1948 and 1953, designed to visually promote the positive effects of the ERP in supporting collaboration amongst European aid recipients whilst respecting the diversity of peoples in Europe.3 It is interesting to note that the emblematic motto implicitly suggests that unity and diversity ought to be brought into some form of balance.Yet it appears that over time this idea of constant recalibration of these intrinsically linked objectives has been increasingly absorbed by the expectation that European integration ‘would, and should, produce a uniform outcome’ (Warleigh-​Lack 2015: 871). This, too, can be traced back to the early days of European integration. The EEC Treaty, for example, contains the commitment to create ‘the foundations of an ever-​closer union among the peoples of Europe’ (EEC Treaty 1957: 17). The Maastricht Treaty on EU did eventually provide the legal foundations for permanent differentiation.This was reinforced with further treaty revisions such as, for example, the Amsterdam Treaty. European integration has thus emerged as a complex system of concentric circles at different speeds and à la carte with a core of club members surrounded by countries which had managed to negotiate opt-​outs from specific policy areas over time (Lavenex 2011); countries that have been temporarily excluded from certain policy areas, i.e. for not meeting convergence criteria (e.g. Romania and Bulgaria from Schengen and EMU); countries willing to join and various neighbourhood agreements involving both European and non-​European countries adamant for closer affiliation (see e.g. Gstöhl and Lannon 2018). From the beginning, the then 50

DOI: 10.4324/9780429054136-5

Comparative regionalism

European ‘union’ was more an aspiration of integration rather than a reflection of it. Looking further afield, differentiation appears to be even more common. The issue of differentiation has by now been explored across several dimensions in the context of European integration. In contrast, instances of differentiated disintegration –​with only a few exceptions (Schimmelfennig 2018; Leruth et al. 2019; Gänzle et al. 2019 for the most comprehensive account thus far) –​have not yet been fully addressed. This is somewhat surprising as it is not an entirely new phenomenon, nor is it confined to the realm of European integration –​a point well-​made by the comparative regionalism literature (e.g. Söderbaum 2016). Drawing from the definition put forth by Schimmelfennig, differentiated disintegration can be defined as the ‘selective reduction of a state’s level and scope of integration’ (2018: 1154).4 This applies to the EU as much as it does to other systems of regional integration in the world –​and results in a decrease in both the level and scope of integration with the (potentially) looming prospect of full departure of the member state in question. Hence, whereas differentiated integration refers to unequal growth of integration, differentiated disintegration points to unequal reduction in the level, scope or membership of a system of regional integration (Schimmelfennig 2018: 1156). Differentiated disintegration triggers effects on the departing member state and the system of regional integration. Ultimately, disintegration may entail the complete ‘falling apart’ of a regional institutional set-​up with important consequences for the as wider international system. Since the end of the Second World War, several attempts at regional integration, defined in terms of institution-​building, in Europe and elsewhere in the world have occasionally failed to live up to their initial objectives or at least exposed some disintegrative symptoms –​ultimately withdrawing in some way ‘from tasks already assigned to them or simply collapse[ing] altogether’ (Lefkofridi and Schmitter 2015: 3). Some prominent cases include the precursors for contemporary African regionalism: the failure to establish the East African Federation in 1963 (Nye 1965) and the subsequent collapse of the East African Community, ‘one of the most successful models of integration on the globe until its demise in 1977’ (Khadiagala 2016: 175). However, prior to the Brexit referendum of June 2016, which set the United Kingdom (UK) on a path towards withdrawal from the EU, neither the study of differentiation in general nor differentiated disintegration in particular has occupied a central role in European integration, new regionalism or comparative regionalism scholarship.5 Differentiated disintegration has hitherto remained an almost exceptional and odd interlude of otherwise successful trajectories of regional integration in the world or (normatively) considered, in a tragic view, ‘as integration gone wrong’ (Warleigh-​Lack 2015: 881). Methodologically, our chapter draws by and large from a critical reading of existing comparative regionalism scholarship focusing on differentiated disintegration as well as document analysis with regards to our selected cases. Consequently, our theoretical ambition is to outline the relative lack of systematic engagement with differentiated disintegration. The empirical section indicates, however, that there is a need to do just that –​and not just with the EU in mind. Comparative regionalism can offer useful insights –​but we need an analytical framework to facilitate such comparisons –​which we will be arguing for in our concluding section. Thus, the remainder of this chapter will proceed in the following way: we will first conceptualize terms such as ‘differentiation’ and ‘differentiated disintegration’ and then move on to provide an overview of theoretical accounts for (differentiated) disintegration. Second, we will discuss instances of associations from Europe, Eurasia, Africa and Latin America exhibiting forms of differentiated disintegration. Third, we will elaborate on the causes and consequences of differentiated disintegration –​before we ultimately conclude on the added value of Comparative Regionalism vis-​à-​ vis the study of differentiation in regional integration and disintegration. 51

Stefan Gänzle and Jens Uwe Wunderlich

Differentiation: concepts and theories of comparative regional integration Although differentiation is a ubiquitous feature of regional integration, it has been poorly understood thus far. Differentiation characterizes organizations of regional integration externally, i.e. by the many ways how these organizations relate to the international environment in which they are embedded. For East Asian and African regionalism, for example, the metaphor of a ‘spaghetti bowl’ (e.g. Byiers et al. 2019) has often been used to describe the complex reality of interwoven regional (multilateral) trade pacts and other institutional agreements. Most members of the African Regional Economic Communities –​such as the Common Market for South Eastern Africa (COMESA) –​participate in several schemes of regional integration. Because of different membership arrangements and policy scopes ‘inside’ a given regional integration system, differentiation also characterizes regional organizations internally. The differentiated polity of the EU, for example, has been distinguished in terms of horizontal and vertical integration –​both internally and externally (see Leuffen et al. 2013). Thus, rather than a system of (relative uniform) integration across a smaller range of policies in the past, the EU has morphed into a dynamic system of differentiated integration.6 Today, it is more accurate to describe the EU as a system of differentiation, including both aspects of differentiated integration and (differentiated) disintegration (Leruth et al. 2019; Gänzle et al. 2019; Schimmelfennig 2018: 1159). Scholarly interest in one critical feature of differentiation, namely ‘differentiated disintegration’ has been scattered throughout the literature. The realist spectrum of thought, for example, emphasizes the centrality of power in world politics. Balance-​of-​power considerations are, therefore, at the heart of this school of thought. By and large, realist scholars remain sceptical regarding the durability of institutional cooperation. States may be driven towards some form of collaboration by balance-​of-​power considerations. Regional institutions, conceptualized in such a manner, are hardly more than institutionalized expressions of an alliance against a common threat or a hegemonic power (Fry 2000: 129; Gilpin 1981; Hoffman 1982). This point has been well made by hegemonic stability theory (Kindleberger 1986; Gilpin 1987; Keohane 1984). However, these institutions are inherently vulnerable to real and perceived changes in the relative distribution of capabilities. Disintegration and complete disbandment consequently remain real options (Mearsheimer 1990). Neo-​functionalism emerged in the 1950s and 1960s as the most credible liberal challenge to the realist tradition of international relations (Rosamond 2000: 73). Stripping classical functionalism off its global telos, neo-​functionalism has studied post-​Second World War efforts of managing economic interdependence and de facto regional integration, explaining the steady expansion of integration by two critical factors: first the presence of spill-​over, by which integrative efforts in one economic sector expanded into adjacent policy sectors; second, the support of this development by a political elite embraced by interest groups on the one hand and supranational policy entrepreneurs and entrepreneurial institutions (e.g. the High Authority) on the other.7 Overall, it seems that the framework conditions in Western Europe have remained ‘unusually favourable to the generation/​cultivation of spill-​overs from one functional arena to another and from lower to higher levels of common authority’ (Schmitter and Lefkofridi 2016: 2) for quite some time. This has paved the ground upon which European integration unfolded –​and was passively approved by public opinion or at least not actively contested for most its time. Ultimately, such ‘permissive consensus’ has allowed European integration to be perceived as a one-​way street, which would result in a steady progression towards an ‘ever closer union’ of both peoples and states as expressed in the preamble of the Treaty on the EU. 52

Comparative regionalism

Still, neo-​functionalist reasoning remained sensitive enough to the necessity that any serious theory of regional integration needs to account for reverse processes, encapsulated in terms of the spill-​back concept (Haas and Schmitter 1964; Schmitter 1969). Neo-​functionalism remains a notable exception, however.The majority of the existing theoretical approaches were designed to explain integration (rather than disintegration) and are, subsequently, ‘ill-​equipped to go in reverse’ (Jones 2018: 440).8 Neo-​functionalism, falling out of favour during the 1980s, was revived in the 1990s and 2000s under the umbrella of neo-​institutionalism (March and Olsen 1989; Powell and DiMaggio 1991) and transactionalism (Stone Sweet and Sandholtz 1998). Following the ruptures caused by the ongoing Eurozone and the Mediterranean refugee crises as well as the 2016 British EU-​exit referendum, several scholars have started to focus on the forces that appear to drive the EU apart (Eppler et al. 2016; Hooghe and Marks 2009; Jones 2018; Leruth et al. 2019; Schimmelfennig and Winzen 2014;Vollaard 2014, 2018; Zielonka 2014). Conceiving disintegration as a multidimensional phenomenon, some authors explore the role of hegemonic stability theory (Webber 2019) and the dynamics of political community building (Vollaard 2014, 2018). Proposing a variety of causal mechanisms, others seek to establish new frameworks for operationalizing explanatory variables (Eppler and Scheller 2013) or devise scenarios for future developments of the EU towards a loosely connected ‘empire’ (Zielonka 2007, 2014). Identified as ‘the most powerful’ (Jones 2018: 441) theoretical account to date, post-​functionalism emerged in the fold of functionalist thinking (Hooghe and Marks 2009). In contrast to neo-​functionalism and liberal intergovernmentalism, post-​functionalist scholarship is rooted in research on identity and political contestation, emphasizing the disruptive clash between functional pressures and claims to exclusive identity (Hooghe and Marks 2009: 4). This is particularly interesting when it comes to differentiated disintegration in the age of identity-​ based forms of political mobilization and politicization. The emphasis on ideational factors is also a hallmark of new regionalism scholarship. Indeed, proponents of ‘new regionalism’ have been much more alert to the potential of differentiation as well as disintegration –​although here again explicit scholarly accounts of disintegration remain relatively rare, too. There have been several calls for a more sustained engagement between European integration and new regionalism studies (Warleigh-​Lack and Robinson 2011; Warleigh-​Lack and Van Langenhove 2012). These efforts have been reiterated by the burgeoning comparative regionalism literature (e.g. Söderbaum 2009; Warleigh-​Lack and Rosamond 2010). One of the benefits of comparative regionalism is to provide a better understanding of differentiated integration through an increase in case numbers (Warleigh-​ Lack and Van Langenhove 2012). Hettne and Söderbaum’s (2000) regionness scale offers an analytical framework for comparative studies. It outlines various phases of regional community and institution-​building. The focus is on the processes of economic and political transactions, social communications and institution-​building within a particular geographical space. It is worthwhile to note that the regionness scale represents an idealized representation of region-​ building for heuristic and comparative purposes (Hettne 2014: 58–​9; Warleigh-​Lack and Robinson 2011: 7). The regioness scale emphasises region-​building (i.e. ‘moving up along the scale’). It provides the potential to include various forms of regional institutions in different parts of the world. However, it is important to note that movement in both directions is possible. Disintegration (i.e. ‘moving down the scale’) remains a possibility –​and is much more naturally conceived as part of dynamic processes of region-​building and un-​building (Warleigh-​Lack and Robinson 2011: 7). A state could leave a set of institutional arrangements that describe the ‘region-​state’. Depending on the remaining or the replacing institutional infrastructure, it will, at the very minimum, remain embedded within the shared regional space. Differentiated disintegration, 53

Stefan Gänzle and Jens Uwe Wunderlich Table 4.1  Scales of regionness Scales of regionness

Features

Regional social space

Territory delineated by geography and politically organized by distinct groups of people Increased contacts lead to more durable relations between different groups and increase interdependencies Establishment of a set of rules and norms to make interstate relations more stable and predictable, analogous to the anarchical society (Bull, 1977) Creation of a formal or informal organizational framework promoting social communication and convergences of values, norms and expected behaviours, encourages identity formation at regional level and formation of a transnational civil society based on trust Fixed structure of decision-​making institutions at regional level, institutional capacity to be an international actor

Regional social system Regional international society

Regional community

Region as institutionalized polity/​region state Source: Adopted from Hettne (2011: 30).

therefore, should not be perceived as failure but as continuation of regional processes of cooperation. Deinstitutionalization is often followed by the establishment of alternative institutional arrangements. Work on comparative regionalism can be traced back to neo-​functionalism. Scholars such as Ernst B. Haas and Karl W. Deutsch aimed at comparing European integration to regional experiments in other parts of the world (Haas and Schmitter 1964; Nye 1968). However, with the declining fortunes of neo-​functionalism, interest in regional comparisons began to dwindle as well. In the late 1980s/​early 1990s scholarship on regionalism experienced somewhat of a renaissance. This triggered the evolution of new regionalism studies. Moreover, since the 2000s a revival of comparative regionalism can be observed,9 including a steady stream of studies comparing the European experience with other instances of regional cooperation.10 Comparative regionalism offers promising research avenues. It helps to overcome culture-​bound explanations and false generalizations that may arise from single case studies (Buzan and Wæver 2003: 468). Towards this background, the following section offers a brief overview of selected cases of differentiated disintegration.11

Differentiated disintegration –​empirical evidence beyond the European Union The Organization for African Unity (OAU), predecessor of the African Union (AU) established in 2002, witnessed its first departure of a member state in 1984, when Morocco quit over a dispute concerning insurgents based in Western Sahara. Western Sahara, a former Spanish colony situated in the Maghreb, was partially controlled by the Polisario Front and partially occupied by Moroccan troops. In 1976, the Polisario Front proclaimed the announcement of the Sahrawi Arab Democratic Republic (SADR). A large part of the territory claimed by the SADR is now controlled by Morocco (May 1984). This dispute over territorial sovereignty led Morocco to withdraw from the OAU when the organization accepted the SADR as a full member state. It has been argued that this effectively paved the road for the OAU and its affiliated agencies to be transformed ‘into advocacy entities for the SADR within the United Nations Security Council, the European Union, and other international institutions’ (Boukhars 2017: 1). It was only in 54

Comparative regionalism

1999 that, a Moroccan leadership change triggered a change in the country’s foreign policy outlook. The new king, Mohammad IV not only fostered closer relations with the EU, becoming the ‘EU’s “closest neighbor” in terms of integration’ (Lonardo 2016: 1), he also reached out to countries and regional groupings in Africa. These political ambitions were underpinned by the strength of the Moroccan economy.These initiatives have culminated in the country’s full reintegration into the AU in 2017. Moreover, Morocco filed a bid to join the Economic Community of West African States (ECOWAS). Several analysts have argued that Morocco’s disintegration from the OAU/​AU left the country at a strategic disadvantage. Boukhars (2017: 1) maintains, for example, that ‘[f]‌or decades, Morocco has sought to maintain influence in the AU while opting for the empty chair.Yet in recent years, it has become clear that an empty chair ceded the strategic advantage to its adversary’. The ECOWAS was formed in 1975. Mauritania, a founding member, exited ECOWAS in 2000, expressing concerns about ECOWAS plans to establish a currency union (BBC 1999). Mauritania joined the Arab Maghreb Union (AMU), which was established in 1989 based on a trade agreement aiming at economic and political integration in Northern Africa. However, the AMU has not been in a position to achieve tangible results because its members, in particular Morocco and Algeria, remained deeply split on the issue of Western Sahara. As a consequence, Mauritania signed a partnership agreement with ECOWAS in 2017 seeking full reintegration by the year 2019, an objective which it has yet to achieve. It is notable that Mauritania, despite its withdrawal from ECOWAS, has always maintained close bilateral agreements with several ECOWAS members, such as Niger, the Ivory Coast, Senegal, Guinea and Mali. It is, therefore, possible to reason that Mauritania’s departure from ECOWAS has taken a form of differentiated disintegration by which the exit was softened by means of upholding bilateral relations with some of the Community’s member states in a view of softening the negative externalities of the decision. The Union of South American Nations (UNASUR) is a relatively extreme example with a relatively large number of withdrawals. UNASUR was founded in 2008. Its origins can be traced back to Pan-​American ideas promoted by revolutionaries such as Francisco de Miranda, Simón Bolívar or José de San Martin who aimed to unite several Spanish American states into a loose federation/​confederation to resist outside interference. UNASUR, bringing together the members of Mercosur and the Andean Community, was set up to create a regional problem-​ solving mechanism in order to balance the influence of the United States (Borda 2014: 2). The initiative was spearheaded by Brazil and by Venezuela. At the time, most Latin American countries were governed by left-​leaning governments. However, in 2018 the organization lost half its membership after Argentina, Brazil, Chile, Colombia, Paraguay and Peru announced that they would (temporarily) suspend their membership. The immediate trigger was a leadership crisis ensuing in 2017 when UNASUR governments failed to reach a consensus on the succession of the general secretary of the organization. The crisis revealed deep-​seated divisions between left-​wing and conservative politics among UNASUR’s members.12 All six countries are part of the so-​called Lima grouping, a coalition of countries set up to mediate the ongoing crisis in Venezuela under the Bolivarian government. It was clear that the principle of Latin American solidarity in face of the US-​dominated Organization of American States (OAS) had not lived up to its expectations. Times have changed since 2008 and the political and economic divisions in the region have only widened. Countries remain split between their outlook across the Pacific to China and Mercosur.Various domestic issues have contributed to limited appetite for UNASUR’s integrative aims. Indeed, there has been deep-​seated resistance on that front in a region obsessed with sovereignty. There is little hope that the divisions will be overcome anytime soon. At the time of writing, Colombia, Brazil and Ecuador have formally informed UNASUR of their intention to leave to bloc permanently. In 2019, Brazil, Argentina, Chile, 55

Stefan Gänzle and Jens Uwe Wunderlich

Paraguay, Colombia, Ecuador, Peru and Guyana announced the foundation of a new association as an alternative to the leftist UNASUR –​the Forum for the Progress of South America (PROSUR). The Andean Community (CAN) emerged from the Andean Pact in 1996. The Andean Pact, in turn, was triggered by the failure of the Latin American Free Trade Association (LAFTA). In 1969, Bolivia, Chile, Columbia, Ecuador and Peru signed the Cartagena agreement. The primary aim was to promote regional economic development and growth. In terms of institutional design, the founding members took inspiration from Europe. As a result, the Andean Pact and its successor, CAN, appear to be highly institutionalized, even including a supranational dimension, at least on paper (Koff 2018: 67). Major changes in the wider global political economy implied that the member countries moved towards a model of open regionalism. In 1996, the Andean Pact was reconstituted as the Andean Community with the mission ‘to promote autonomous, balanced, and integral development through Andean, South American, and Latin American integration’ (ibid.). In 1973, a military coup overthrew Allende’s government. The military junta under General Pinochet took control of the country and, with Washington’s tacit support, embarked on a neoliberal economic reform course, characterized by privatization, deregulation, eliminating capital and trade barriers. In 1976, Chile withdrew from the Andean Pact, citing as main reason the protectionist nature of the bloc (Kandell 1976). However, this was not a complete withdrawal, Chile maintained observer status for 30 years. In June 2007, the country rejoined CAN as an associate member. At that time, CAN had emerged as a vehicle to promote free trade in the Andean region and beyond. This caused friction between CAN members and in 2006, Hugo Chavez withdrew Venezuela (which had joined CAN in 1973) in response to Peru and Columbia’s intentions of signing a free trade agreement with the United States, instead joining Mercosur in 2012. However, in 2016 its Mercosur membership was suspended indefinitely due to Venezuela’s unwillingness to implement key rules on trade and human rights into domestic law. This has led Venezuela to promote its own version of regional integration within the Bolivarian Alliance for the Peoples of Our America (or ALBA), brought into existence in 2004 on the initiative of Fidel Castro and Hugo Chavez (Koff 2018: 68). Differentiated disintegration, in the case of CAN, was triggered by diverging domestic political and economic priorities and the strategic differences between neighbouring countries, as well as their alignment to wider international dynamics (in particular relations with the United States). Chile’s revived membership in CAN may be just the catalyst for invigorating the organization’s capability to promote free trade not only in the Andean region and throughout South America, but also across the Pacific, a prime field of its interest. This is of particular significance at a time when the Andean Community faces uncertainty due to a lack of cooperation among some of its members as well as heated competition with other South American economic conglomerates (like the Mercosur bloc). The Eurasian Economic Union (EAEU) came into existence in 2014 succeeding the Eurasian Economic Community (EAEC) of 2000. Since the collapse of the Soviet Union in 1991/​92, there have been various attempts by the successor states ‘to reconnect former Soviet economies in order to offset the dire economic conditions’ (Roberts and Moshes 2016: 543; see also Blockmans et al. 2012). As early as 1994, the then-​president of Kazakhstan, Nursultan Nazarbayev, called for the establishment of a ‘Eurasian Union’ (Tarr 2016); however, with most regional leaders preoccupied with state-​building and ring-​fencing the newly established sovereignty, this idea was stalled. It was only in October 2000 that Belarus, Kazakhstan, Kyrgyzstan, Russia and Tajikistan agreed on the formation of the EAEC with the aim of ensuring and fostering multilateral economic cooperation among its member states and with the medium-​term objective of spelling out a customs union. In the following years, on the initiative of Vladimir Putin, freshly elected 56

Comparative regionalism

President of the Russian Federation, efforts were made to sponsor economic integration among the EAEC states –​disclosing the geopolitical component and in particular the perception that Eurasian integration is a manifestation of Russia’s hegemonic ambitions in the region (Roberts and Moshes 2016: 543). Eventually, in 2006, the EAEC expanded to include Uzbekistan and it was decided to establish the projected customs union within the framework of the EAEC, with Belarus, Kazakhstan and Russia as the founding members. In January 2010, the leaders of Russia, Belarus and Kazakhstan reaffirmed once more the desire to implement a customs union and agreed to launch the Eurasian Customs Union (ECU), including the implementation of a common external tariff policy.Two years later, the three countries subscribed to closer economic ties by signing an agreement to form a ‘common economic space’; this eventually led to the creation of the EAEU, which was joined by Armenia and Kyrgyzstan. In short, the EAEU has as its core objectives the free flow of goods, services, labour and capital among its members –​in other words, a single market. As with the EAEC, several international observers see the EAEU as an effort by Putin to cement his legacy by restoring Russia’s influence among the states of the former Soviet Union and creating a geopolitical Eurasian union that would make Russia the leader of a bloc of countries that competes with the EU for influence in Europe and maybe on the world stage. By that time, however, Uzbekistan had already withdrawn from the body. The Uzbek Ministry of Foreign Affairs informed the EAEC Secretariat of its decision to ‘suspend’ participation in the EAEC in October 2008. The Uzbek decision was not publicly discussed until the EAEC Secretariat acknowledged the diplomatic note in November that year. At the same time, a letter from Uzbek President Islam Karimov was sent to all the heads of state of the other EAEC countries –​Belarus, Kazakhstan, Kyrgyzstan, Russia and Tajikistan –​informing them of Uzbekistan’s decision. In a press interview on 17 November 2008, the Russian Foreign Minister Sergei Lavrov ‘acknowledged the Uzbek move, noting that Russia would respect Uzbekistan’s right to make independent judgments but felt that the decision was not in Russia’s best interests’ (Gleason 2008: 1). Gleason (2008) argued that the Uzbek government’s withdrawal from the EAEC exemplifies the reliance upon what has come to be known as the ‘Uzbek Path’. Uzbekistan has a long tradition of independence in policy and orientation, having pursued economic and foreign policies often at odds with its Central Asian neighbours and with Russia. But the withdrawal from participation in the foremost economic cooperation organization in the former Soviet region, the EAEC, is significant because it represents a major fissure in the solidarity among the former Soviet states so vigorously promoted by the Russian government. More recently, however, Uzbekistan has started to harmonize its import tariffs with EAEU norms (Tickett 2018: 1) as part of a larger reorganization of its customs service with a view to improve tax collection and the investment climate. The European Free Trade Agreement (EFTA) is an interesting case of what one might call ‘counter’ or ‘balancing’ regionalism. It was brought into being by the UK to balance and to undermine the EEC by creating a rival European trade bloc. EFTA brought together seven countries: Austria, Denmark, Norway, Portugal, Sweden, Switzerland and the UK. Later accessions include Iceland, Lichtenstein and Finland. The United States, however, continued to support European integration in form of the EEC as it bore far greater potential of generating trade flows between its member states and between the EEC and the United States. Moreover, the EEC offered what EFTA could not –​the possibility of firmly anchoring Germany within Europe. EFTA was seen for what it was –​a defensive and divisive arrangement with very little added value for European unity during the Cold War (Winand 1997: 167). Nor did EFTA generate the envisaged benefits in terms of trade between its members. As early as 1961, only one year after EFTA’s emergence (and almost five years after the Rome treaties), the UK, Norway and Denmark, applied for EEC membership. Only a decade later, the UK and Denmark left 57

Stefan Gänzle and Jens Uwe Wunderlich

EFTA in 1972, joining the EEC in 1973. Norway remained in EFTA after EEC membership had been rejected by a referendum in 1972. In 1976, the military junta that had been ruling Portugal had been replaced by a civilian government which quickly began accession negotiations with the EEC in order to consolidate democratic transition and to gain access to much needed economic assistance. In 1985, Portugal followed the UK and Denmark in leaving EFTA and joined the EC. This process was repeated after the Cold War when three more countries withdrew from EFTA in 1994. These countries, Austria, Sweden and Finland, had remained neutral during the Cold War preventing them from participating in regional arrangements designed to balance Soviet influence in Europe such as NATO and the EEC. All three countries used the opportunity offered by geopolitical changes following the end of the Cold War and the demise of the Soviet Union and joined the EU in 1995.To some extent, the EFTA case is unique. While geopolitical and domestic-​level changes were important, the decisive factor behind differentiated disintegration was the accession to another (larger and more comprehensive) regional association.13

Discussion: drivers of regional disintegration States that form and join regional cooperation agreements are motivated by a range of factors. The same is true for suspending membership or leaving a regional organization. It is possible to discern at least three broad and interrelated factor groups motivating disintegration. These include (1) intramural tensions between members of a regional association agreement –​also with regards to the scope of regional identification. As we have seen, these tensions can be the result of a change in policy priorities at the domestic level, territorial disputes or the rise of a regional hegemon.This goes some way to explain why Morocco left the OAU in 1984. It also accounts in part for the difficulties of UNASUR which appears to have been divided into two different camps, resulting in a significant exodus of members. The Uzbek decision to quit the EAEC can be explained by suspicions of Russia using the organization as platform to foster its own hegemonic ambitions in Eurasia. A second factor has been the (2) perceived lack of progress of the regional association agreement in attaining member state-​level economic and political cooperation goals; or ‘too much’ progress leading to an increasing mismatch between the general direction of regional cooperation and the preferences of individual member states. This is very much contingent on perceptions of the usefulness of the regional organization in question by domestic policy-​making elites within said member states. Cost-​benefit calculations are important in that context. An additional contributing factor, it seems, is the existence of alternative/​competitive regional arrangements. Mauritania, as outlined above, left ECOWAS and joined the AMU only to sign an association agreement with ECOWAS in 2019. Similarly, Britain and Denmark (and Portugal) left EFTA and joined the EC. Diverging priorities between individual member states and the direction of regional cooperation, triggered by a change in government, help to explain why Chile left the Andean Pact. After almost 30 years, it rejoined CAN as associate member.Venezuela’s decision to leave CAN and to join Mercosur falls also into this category. Indeed, this is also behind Venezuela’s ALBA ambition. A third factor are (3) changes in the geopolitical and geo-​economic framework conditions. Such considerations are driving states to join regional cooperation agreements. Hence, alterations of these conditions may not necessarily result in the collapse of a regional organization but may trigger instances of differentiated disintegration. Here, association agreements that span the old and new regionalism period, such as EFTA, are interesting. Sweden, Austria and Finland were freed from neutrality commitments by the end of the Cold War and left EFTA to join the EU. 58

Comparative regionalism

Finally, we need to note that regional spaces are often crisscrossed by various regional cooperation agreements. The analogy of a ‘spaghetti bowl’ above is an apt one. Differentiated disintegration does not mean the end of regionalism, a point well made by the new regionalism and the regionness scale. Former member states of specific regional organizations still inhabit the same regional space, remain subject to economic, social and political interdependencies (i.e. the regional social system) and often remain part of the wider regional normative infrastructure (i.e. the regional international society). Hence, to quote an old German song, ‘Niemals geht man so ganz’.14 In other words, differentiated disintegration should not be regarded as a failure of regional cooperation –​more often than not, it is the continuation of interaction and cooperation by other means. Herein lies the most important lesson –​regional integration is not an evolutionary process towards ‘ever closer’ union. We can note the instances of differentiated disintegration are scattered across different time periods, accounting for so-​called old and new regionalism, and can be found in almost every regional setting.15 A notable exception in this context is East Asia. There are no obvious explanations for this. It might be that East Asian regionalism is a relatively recent phenomenon. In some ways, it can be traced back to the mid-​1990s and the ASEAN Plus Three (APT) process.16 Prior to that, East Asian regional institutions had been confined to the regional periphery –​i.e. Southeast Asia. Here, ASEAN, the Association of Southeast Asian Nations, evolved as the core of an institutional infrastructure. Compared to European integration, ASEAN regionalism represents a set of very loose institutional arrangements. This is encapsulated in the so-​called ASEAN way, consisting of a set of procedural norms and principles such as a preference for informal dialogue and non-​binding commitments over formalistic and legalistic approaches, consensual decision-​making based on consultation and a firm emphasis on sovereign equality and non-​interference. Member states compromise very little, if any, decision-​making power.This is one of the factors facilitating ASEANs emergence at the core of the evolving wider East Asian regional institutional infrastructure in the form of the East Asia Summit and the APT.

Conclusion It is clear now that differentiated disintegration is by no means a novel phenomenon restricted to European integration and the case of ‘Brexit’. We have shown that ‘disintegration’ had been sidelined for both normative and empirical reasons in major European integration theories. While unusual, empirical cases can be found across the timeline, affecting a range of very different regional association arrangements. Comparative regionalism provides the appropriate analytical lenses to accommodate an idea of differentiation in integration which encompasses both integration and disintegration. We recognize that by way of a normative evaluation both the perspective and the scope of disintegration need to be taken into consideration. With regards to the former and from the position of international institutions as well as remaining member states, disintegration may always be perceived as a failure, as institutions intrinsically strive for (ever closer) union and cooperation rather than disunion, and in almost all cases the economic benefit of being part of an international institution rises with the number of member states. As for the latter, it is safe to say that internal disintegration from individual policies can be viewed less a failure than losses of membership. Furthermore, while Brexit can be perceived as a loss (for all parties involved), most scholars are likely to agree that it was not a failure of the EU –​nor does it fundamentally affect the aggregate regionness of the remaining Union. UNASUR and its member states in contrast, probably saw the loss of half of their members as a blatant failure of the integration project in itself. 59

Stefan Gänzle and Jens Uwe Wunderlich

This ultimately leads to the question whether differentiated disintegration can be theorized as general phenomenon. Are there certain factor groups that are similar across these cases? Or are these individual instances explained by unique circumstances that are specific to each case and, therefore, defy any attempt of generalization? More research is needed here. Ideographic research, for example, by and large questions the applicability of general factors across multiple cases. Each case is unique, requiring in-​depth, historically and socially grounded analysis to facilitate a contextualized understanding of the processes and dynamics characterizing each instance of disintegration. Such an approach emphasizes the profound historical, cultural, anthropological, political and sociopolitical particularities that characterize integration and, by extend, disintegration. This causes the ‘many variables/​small n-​problem’ –​the lack of comparable cases (see De Lombaerde et al. 2010). By its very nature then, the generalizability of the insights of ideographic research is limited. In contrast, comparative regionalism seeks to find general explanations by studying multiple cases –​and by doing so greatly fosters our understanding of differentiation in regional integration processess. The challenge is to develop a framework that is flexible enough to accommodate the very different empirical cases introduced under section three. The regionness scale (see Table 4.1) has the potential to provide such a framework. Here, we tentatively suggest that three explanatory clusters can be identified: first, member state-​level disagreement with regional integration objectives (possibly fostered by region-​sceptic political movements); second, dissatisfaction with regards to putative economic and political gains from regional integration and third major changes of the geopolitical and geo-​economic environment. However, these clusters are very broad and more research is needed to make them analytically useful and to systematically map individual cases of differentiated disintegration along the regionness scale.

Notes 1 The authors are grateful to the comments and suggestions by Jarle Trondal, Benjamin Leruth and Tobias C. Hofelich. 2 Both ‘Europeans’ and ‘EU’ are in bold in the original, available at (accessed 2 July 2020). 3 Such as in the 20-​minute picture ‘The Hour of Choice’ (London: 1951). 4 This may also apply to parts of a member state’s territory, too, see the cases of Greenland and Saint-​ Barthélemy in the case of the EU (Patel 2018; Gänzle 2020). 5 For historical examples of differentiated disintegration in the European Economic Community/​EU, see Gad (2013), Gänzle (2019, 2020), Harhoff (1983), Kobza (2016), Oraison (2008) and Patel (2018). 6 There is a growing literature on differentiated integration in the EU: Adler-​Nissen (2009, 2011, 2014), Andersen and Sitter (2006), de Neve (2007), Dyson and Sepos (2010), Egeberg and Trondal (1999), Gstöhl (2000), Kölliker (2001, 2006), Leuffen et al. (2013), Leruth and Lord (2015), Holzinger and Schimmelfennig (2012), Stubb (1996, 1997), Trondal (2002), Walker (1998), Warleigh (2002). 7 Neo-​functionlist theory has been closely associated with scholars such as Haas (1958), Lindberg (1966) and Schmitter (1970). 8 This is understandable given the research aim of explaining and analysing integration. 9 For examples see Acharya and Johnston (2008), Breslin et al. (2002), Buzan and Waever (2003), Laursen (2003), Farrell et al. (2005), Söderbaum (2009), Mattheis (2014). 10 See, for example, Katzenstein (2005), Beeson (2006),Telò (2007), Wunderlich (2007), Murray (2008, 2010), Jetschke and Murray (2012), Moxon-Browne and Murray (2013), Murray and Warleigh-Lack (2013), Brennan and Murray (2015). 11 The cases of Morocco, Mauretania and Uzbekistan have been discussed already by Gänzle (2019, 2020). 12 Underlying were also different visions of regionalism with post-​liberal and liberal regionalism. For post-​ liberal regionalism and UNASUR, see Gardini (2017). 13 The UK left the EU in 2020. The idea of rejoining EFTA has been floated at various times during the Brexit process. However, the four current members of EFTA, Norway, Switzerland, Iceland and 60

Comparative regionalism

Lichtenstein, are a vary about the disruption to intra-​EFTA and EFTA–​EU relations caused by possible post-​Brexit UK EFTA membership (Wintour 2019). 14 ‘You never go completely’ was a song by the late Trude Herr. The phrase is an apt analogy for differentiated disintegration (see Gänzle 2020). 15 The literature commonly distinguishes between two broad empirical waves of contemporary regional institution-​building. ‘Old’ regionalism refers to the period between the 1950s and 1970s. Regional institutions were influenced by geopolitical considerations (i.e. the Cold War) and decolonization.‘New’ regionalism refers to a post-​Cold War phenomenon. For the differences between both periods see Hettne (1996: 9–​11). 16 The APT is bringing ASEAN members together with South Korea and East Asia’s core powers –​Japan and China.

Literature Acharya, A. and Johnston, A. (2008) Crafting Cooperation –​Regional Institutions in Comparative Perspective. Cambridge: Cambridge University Press. Adler-​Nissen, R. (2009) Behind the Scenes of Differentiated Integration: Circumventing National Opt-​ Outs in Justice and Home Affairs, Journal of European Public Policy 16(1), pp. 62–​80. doi: 10.1080/​ 13501760802453239. Adler-​Nissen, R. (2011) Opting Out of an Ever Closer Union:The Integration Doxa and the Management of Sovereignty, West European Politics 34(5), pp. 1092–​113. doi: 10.1080/​01402382.2011.591102. Adler-​Nissen, R. (2014) Opting Out of the European Union: Diplomacy, Sovereignty and European Integration. Cambridge: Cambridge University Press. Andersen, S.S. and Sitter, N. (2006) Differentiated Integration: What Is It and How Much Can the EU Accommodate?, Journal of European Integration 28(4), pp. 313–​30. doi: 10.1080/​07036330600853919. Balassa, B. (1961) The Theory of Economic Integration. Westport, CT: Greenwood Press. BBC (1999) Mauritania to Quit ECOWAS, available at: http://​news.bbc.co.uk/​2/​hi/​afr​ica/​578​966.stm (accessed 20 August 2020). Beeson, M. (2006) Rethinking Regionalism: Europe and East Asia in Comparative Historical Perspective, Journal of European Public Policy 12(6), pp. 969–​85. doi: 10.1080/​13501760500270620. Blockmans, S., Kostanyan, H. and Vorobiov, I. (2012) Towards a Eurasian Economic Union: The Challenge of Integration and Unity, CEPS Special Report No. 75, December 14, 2012. Borda, S. (2014) The Union of South American Nations: Mapping Multilateralism in Transition No. 3. International Peace Institute, available online at: www.jstor.org/​stable/​resrep09508 (accessed 5 August 2020). Boukhars, A. (2017) Morocco and the African Union: Back into the Fold, available at: http://​carneg​ ieen​dowm​ent.org/​2017/​02/​25/​moro​cco-​and-​afri​can-​union-​back-​into-​fold-​pub-​68130 (accessed 5 August 2020). Brennan, L. and Murray, P. (2015) Drivers of Integration and Regionalism in Europe and Asia –​Comparative Perspectives. London: Routledge. Breslin, S., Higgott, R. and Rosamond, B. (2002) Regions in Comparative Perspective, in S. Breslin, C. Hughes, N. Phillips and B. Rosamond (eds) New Regionalisms in the Global Political Economy. London: Routledge, pp. 1–​19. Bull, H. (1977) The Anarchical Society: A Study of Order in World Politics. London: Macmillan. Buzan, B. andWaever, O. (2003) Regions and Powers:The Structure of International Security. Cambridge: Cambridge University Press. Byiers, B., Woolfrey, S., Medinilla, A. and Vanheukelom, J. (2019) The Political Economy of Africa’s ‘Spaghetti Bowl’ Regionalism –​Synthesis Report. Maastricht: ECDPM. de Lombaerde, P., Söderbaum, F., Van Langenhove, L. and Baert, F. (2010) The Problem of Comparison in Comparative Regionalism. Review of International Studies 36(3), pp. 731–​ 751. doi:10.1017/​ S0260210510000707. de Neve, J.-​E. (2007) The European Onion? How Differentiated Integration is Reshaping the EU, Journal of European Integration 29(4), pp. 503–​521. doi: 10.1080/​07036330701502498. Dyson, K. and Sepos, A. (2010) Differentiation as Design Principle and as Tool in the Political Management of European Integration, in K. Dyson and A. Sepos (eds) Which Europe? The Politics of Differentiated Integration. Basingstoke: Palgrave Macmillan, pp. 3–​23.

61

Stefan Gänzle and Jens Uwe Wunderlich

EEC Treaty (1957) Treaty Establishing the European Economic Community, available online at: http://​ eur-​lex.eur​opa.eu/​legal-​cont​ent/​DE/​TXT/​PDF/​?uri=​CELEX:119​57E/​TXT&from=​EN (accessed 13 March 2017). Egeberg, M. and Trondal, J. (1999) Differentiated Integration in Europe: The Case of EEA Country, Norway, Journal of Common Market Studies 37(1), pp. 133–​42. doi: 10.1111/​1468-​5965.00154.Eppler, A. and Scheller, H. (eds) (2013) Zur Konzeptionalisierung europäischer Disintegration. Zug-​und Gegenkräfte im europäischen Integrationsprozess. Baden-​Baden: Nomos. Eppler, A., Anders, L. and Tuntschew, T. (2016) Europe’s Political, Social, and Economic (Dis-​ ) Integration: Revisiting the Elephant in Times of Crisis October, IHS Political Science Series Working Paper No. 143. Farrell, M., Hettne, B. and Van Langenhove, L. (2005) The Global Politics of Regionalism: Theory and Practice. London: Pluto Press. Fry, G. (2000) A ‘Coming Age of Regionalism’?, in G. Fry and J. O’Hagan (eds) Contending Images of World Politics. Basingstoke: Macmillan, pp. 117–​131. Gad, U.P. (2013) Greenland: A Post-​Danish Sovereign Nation State in the Making, Cooperation and Conflict 49(1), pp. 98–​118. doi: 10.1177/​0010836713514151. Gänzle, S. (2019) Differentiated (Dis)integration in Europe and Beyond. Historical and Comparative Perspectives (Chapter 12), in S. Gänzle, B. Leruth and J. Trondal (eds) Differentiated Integration and Disintegration in a Post-​Brexit Era. London: Routledge, pp. 202–​18. Gänzle, S. (2020) Niemals geht man so ganz. Historische Beispiele für regionale Desintegration [Leaving Never Is for Good. Historical Examples for Regional Disintegration], Aus Politik und Zeitgeschichte 23–​25, pp. 47–​52. Gänzle, S., Leruth, B. and Trondal, J. (eds) (2019) Differentiated Integration and Disintegration in a Post-​Brexit Era. London: Routledge. Gardini, G. (2017) Towards Modular Regionalism –​The Proliferation of Latin American Cooperation, in U. Engel, H. Zinecker, F. Mattheis, A. Dietze and T. Plötze (eds) The New Politics of Regionalism –​ Perspectives from Africa, Latin America and Asia-​Pacific. Oxon: Routledge, pp. 19–​36. Gilpin, R. (1981) War and Change in World Politics. Cambridge: Cambridge University Press. Gilpin, R. (1987) The Political Economy of International Relations. Princeton, NJ: Princeton University Press. Gleason, G. (2008) Uzbekistan Charts a New “Uzbek Path”, MC Series: Security Insights Number 3, November 2008. Gstöhl, S. and Lannon, E. (2018) The Neighbours of the European Union’s Neighbours. Diplomatic and Geopolitical Dimensions beyond the European Neighbourhood Policy. London: Routledge. Gstöhl, S. (2000) The European Union after Amsterdam:Towards a Theoretical Approach to (Differentiated) Integration, in M.G. Cowles and M. Smith (eds) The State of the European Union Volume 5: Risks, Reform, Resistance, and Revival. Oxford: Oxford University Press, pp. 42–​63. Haas, E. (1958) The Uniting of Europe: Political, Social and Economic Forces, 1950–​57. Stanford, CA: Stanford University Press. Haas, E.B. and Schmitter, P.C. (1964) Economics and Differential Patterns of Political Integration: Projections about Unity in Latin America, International Organization 18(4), pp. 705–​37. Harhoff, F. (1983) Greenland’s Withdrawal from the European Communities, Common Market Law Review 20, pp. 13–​33. Hettne, B. (1996) Globalization, the New Regionalism and East Asia. Hayama: United Nations University.Hettne, B. (2011) The European Union as an Emerging Global Actor, in J.-​U. Wunderlich and D. Bailey (eds) The European Union and Global Governance –​A Handbook. London: Routledge, pp. 28–​36. Hettne, B. (2014) Regional Actorship: A Comparative Approach to Interregionalism, in F. Baert, T. Scaramagli and F. Söderbaum (eds) Intersecting Interregionalism –​Regions, Global Governance and the EU. London: Springer, pp. 55–​70. Hettne, B. and Söderbaum, F. (2000) Theorising the Rise of Regionness, New Political Economy, 5(3), pp. 457–​73. doi: 10.1080/​713687778. Hoffmann, S. (1982) Reflections on the Nation-​State in Western Europe Today, Journal of Common Market Studies 21(1), pp. 21–​38. Holzinger, K. and Schimmelfennig, F. (2012) Differentiated Integration in the European Union: Many Concepts, Sparse Theory, Few Data, Journal of European Public Policy 19(2), pp. 292–​305. doi: 10.1080/​ 13501763.2012.641747.

62

Comparative regionalism

Hooghe, L. and Marks, G. (2009) A Postfunctionalist Theory of European Integration: From Permissive Consensus to Constraining Dissensus, British Journal of Political Science 39(1), pp. 1–​23. doi: doi:10.1017/​ S0007123408000409. Jetschke, A. and Murray, P. (2012) Diffusing Regional Integration:The EU and Southeast Asia, West European Politics 35(1), pp. 174–​91. doi: 10.1080/​01402382.2012.631320. Jones, E. (2018) Towards a Theory of Disintegration, Journal of European Public Policy 25(3), pp. 440–​51. doi: 10.1080/​13501763.2017.1411381. Kandell, J. (1976) Chile’s Official Withdrawal Is Further Harm to an Effort at Economic Integration. The New York Times, available at: www.nytimes.com/​1976/​11/​21/​archives/​chiles-​official-​withdrawal-​is-​ further-​harm-​to-​an-​effort-​at-​economic.html (accessed 6 August 2020). Katzenstein, P. (2005) A World of Regions: Asia and Europe in the American Imperium. Ithaca, NY: Cornell University Press. Keohane, R. (1984) After Hegemony: Cooperation and Discord in the World Political Economy. Princeton, NJ: Princeton University Press. Khadiagala, G.M. (2016) Region-​Building in Eastern Africa, in D.H. Levine and D. Nagar (eds) Region-​ Building in Africa. Political and Economic Challenges, London: Springer, pp. 175–​90. Kindleberger, C. (1986) International Public Goods without Government, American Economic Review 76(1), pp. 1–​13. Kobza, P. (2016) European Union-​Greenland Relations after 2015 –​A Partnership beyond Fisheries, Journal of Military and Strategic Studies 16(4), pp. 130–​53. Koff, H. (2018) Outside-​In Region-​Building –​The Rile of Border Integration Zones in Andean Regional Integration, in E. Nadalutti and O. Kallscheuer (eds) Region-​Making and Cross-​Border Cooperation: New Evidence from Four Continents. Oxon: Routledge, pp. 62–​80. Kölliker, A. (2001) Bringing Together or Driving Apart the Union? Towards a Theory of Differentiated Integration, West European Politics 24(4), pp. 125–​51. doi: 10.1080/​01402380108425468. Kölliker, A. (2006) Flexibility and European Unification:The Logic of Differentiated Integration. Oxford: Rowman & Littlefield. Laursen, F. (2003) Comparative Regional Integration –​Theoretical Perspectives. Aldershot: Ashgate. Lavenex, S. (2011) Concentric Circles of Flexible ‘EUropean’ Integration: A Typology of EU External Governance Relations, Comparative European Politics 9(4–​5), pp. 372–​93. doi: 10.1057/​cep.2011.7. Lefkofridi, Z. and Schmitter, P.C. (2015) Transcending or Descending? European Integration in Times of Crisis, European Political Science Review 7(1), pp. 3–​22. doi: 10.1017/​S1755773914000046. Leruth, B. and Lord, C. (2015) Differentiated Integration in the European Union:A Concept, a Process, a System or a Theory?, Journal of European Public Policy 22(6), pp. 754–​63. doi: 10.1080/​13501763.2015.1021196. Leruth, B., Gänzle, S. and Trondal, J. (2019) Exploring Differentiated Disintegration in a Post-​Brexit European Union, Journal of Common Market Studies 57(5), pp. 1013–​30. doi: 10.1111/​jcms.12869. Leuffen, D., Rittberger, B. and Schimmelfennig, F. (2013) Differentiated Integration: Explaining Variation in the European Union. Basingstoke: Palgrave Macmillan. Lindberg, L. (1966) Integration as a Source of Stress on the European Community System, International Organization 20(2), pp. 233–​65. Lonardo, L. (2016) The EU’s “Diplomatic Accident” with Morocco Shows the Perils of Judge-​Led Foreign Policy, March 7, available at: http://​blogs.lse.ac.uk/​eur​oppb​log/​2016/​03/​07/​the-​eus-​dip​loma​tic-​accid​ ent-​with-​moro​cco-​shows-​the-​per​ils-​of-​judge-​led-​fore​ign-​pol​icy/​ (accessed 9 February 2021). March, J.G. and Olsen, J.P. (1989) Rediscovering Institutions. New York, NY: Free Press. Mattheis, F. (2014) New Regionalism in the South –​Mercosur and SADC in a Comparative and Interregional Perspective. Leipzig: Leipziger Universitätsverlag. May, C.M. (1984) Morocco Quits O.A.U. over Polisario, New York Times, 13 November, available at: www. nytimes.com/​1984/​11/​13/​world/​morocco-​quits-​oau-​over-​polisario.html (accessed 9 February 2021). Mearsheimer, J. (1990) Back to the Future –​Instability in Europe after the Cold War, International Security 5(1), pp. 5–​56. doi: 10.2307/​2538912. Moxon-​Browne, E. and Murray, P. (2013) The European Union as a Template for Regional Integration? The Case of ASEAN and its Committee of Representatives, Journal of Common Market Studies 51(3), pp. 522–​37. doi: 10.1111/​jcms.12008. Murray, P. (2008) Europe and Asia: Regions in Flux. Basingstoke: Palgrave. Murray, P. (2010) Comparative Regional Integration in the EU and East Asia: Beyond Integration Snobbery, International Politics 47(3–​4), pp. 308–​23. doi: 10.1057/​ip.2010.13.

63

Stefan Gänzle and Jens Uwe Wunderlich

Murray, P. and Warleigh-​Lack, A. (2013) Europe-​Asia Studies: The Contribution of Comparative Regional Integration, in The Palgrave Handbook of EU-​Asia Relations. Basingstoke: Palgrave, pp. 108–​23.Nye, J.S. (1965) Pan-​Africanism and East-​African Integration. Cambridge: Harvard University Press. Nye, J. (1968) Comparative Regional Integration: Concept and Measurement. International Organization 22(4), pp. 855–​80. Oraison, A. (2008) Le statut noveau de collectivité d’outre-​mer des iles de Saint-​Barthélemy et de Saint-​ Martin, Droit Public et de la Science Politique en France et à l’Etranger 1, pp. 153–​83. Patel, K. (2018) Something New under the Sun? The Lessons of Algeria and Greenland, in B. Martill and U. Staiger (eds) Brexit and Beyond. Rethinking the Futures of Europe. London: UCL Press, pp. 114–​20. Powell, W.W. and DiMaggio, P.J. (eds) (1991) The New Institutionalism in Organizational Analysis. Chicago, IL: Chicago University Press. Roberts, S.P. and Moshes, A. (2016) The Eurasian Economic Union: A Case of Reproductive Integration?, Post-​Soviet Affairs 32(6), pp. 542–​65. doi: 10.1080/​1060586X.2015.1115198. Robinson, N. and Warleigh-​ Lack, A. (2011) Conclusions –​Learning or Comparisons in the Study of Regions? Fruitful Dialogues and Research Directions, in A. Warleigh-​ Lack, N. Robinson and B. Rosamond (eds) New Regionalism and the European Union –​Dialogues, Comparisons and New Research Directions. Oxon: Routledge, pp. 257–​70. Rosamond, B. (2000) Theories of European Integration. Basingstoke: Macmillan. Schimmelfennig, F. (2018) Brexit: differentiated disintegration in the European Union, Journal of European Public Policy 25(8), pp. 1154–​1173. doi: 10.1080/​13501763.2018.1467954 Schimmelfennig, F. and Winzen, T. (2014) Instrumental and Constitutional Differentiation in the European Union, Journal of Common Market Studies 52(2), pp. 354–​70. doi: 10.1111/​jcms.12103. Schmitter, P.C. (1969) Three Neo-​Functional Hypotheses about International Integration, International Organization 23(1), pp. 161–​6. Schmitter, P.C. (1970) A Revised Theory of Regional Integration, International Organization 24, pp. 836–​68. Schmitter, P.C. and Lefkofridi, Z. (2016) Neo-​Functionalism as a Theory of Disintegration, Chinese Political Science Review 1, pp. 1–​29. doi: 10.1007/​s41111-​016-​0012-​4. Söderbaum, F. (2009) Comparative Regional Integration and Regionalism, in T. Landman and N. Robinson (eds) The Sage Handbook of Comparative Politics. London: Sage, pp. 477–​496. Söderbaum, F. (2016) Rethinking Regionalism. London: Palgrave. Stone Sweet, A. and Sandholtz,W. (eds) (1998) European Integration and Supranational Governance. Oxford and New York, NY: Oxford University Press. Stubb, A. (1996) A Categorisation of Differentiated Integration, Journal of Common Market Studies 34(2), pp. 283–​95. doi: 10.1111/​j.1468-​5965.1996.tb00573.x. Stubb, A. (1997) The 1996 Intergovernmental Conference and the Management of Flexible Integration, Journal of European Public Policy 4(1), pp. 37–​55. doi: 10.1080/​135017697344226. Tarr, D. (2016) The Eurasian Economic Union of Russia, Belarus, Kazakhstan, Armenia, and the Kyrgyz Republic: Can It Succeed Where Its Predecessor Failed?, Eastern European Economics 54(1), pp. 1–​22. 10.1080/​00128775.2015.1105672. Telò, M. (2007) European Union and New Regionalism –​Regional Actors and Global Governance in a Post-​ Hegemonic Era. London: Ashgate. Tickett, N. (2018) Reforming Customs, Uzbekistan Nods toward the Eurasian Economic Union: Tashkent Is Harmonizing Its Import Tariffs with EAEU Norms, The Diplomat, available at: https://​thed​iplo​mat. com/​2018/​04/​reform​ing-​cust​oms-​uzb​ekis​tan-​nod​stow​ard-​the-​euras​ian-​econo​mic-​union/​ (accessed 30 June 2019). Trondal, J. (2002) Beyond the EU Membership –​Non-​Membership Dichotomy? Explaining Supranational Identities among EU Decision-​Makers, Journal of European Public Policy 9(3), pp. 468–​87. doi: 10.1080/​ 13501760210139722. Vollaard, H. (2014) Explaining European Disintegration, Journal of Common Market Studies 52(5), pp. 1142–​ 59. doi: 10.1111/​jcms.12132. Vollaard, H. (2018) European Disintegration. A Search for Explanations. Basingstoke: Palgrave Macmillan. Walker, N. (1998) Sovereignty and Differentiated Integration in the European Union, European Law Journal 4(4), pp. 355–​88. doi: 10.1111/​1468-​0386.00058. Warleigh, A. (2002) Flexible Integration: Which Model for the European Union? London: Sheffield Academic Press. Warleigh-​Lack,A. and Rosamond, B. (2010) Across the EU Studies-​New Regionalism Frontier: Invitation to a Dialogue, Journal of Common Market Studies 48(4), pp. 993–​1013. doi: 10.1111/​j.1468-​5965.2010.02083.x. 64

Comparative regionalism

Warleigh-​ Lack, A. and Robinson, N. (2011) Introduction, in A. Warleigh-​ Lack, N. Robinson and B. Rosamond (eds) New Regionalism and the European Union –​Dialogues, Comparisons and New Research Directions. Oxon: Routledge, pp. 2–​17. Warleigh-​Lack, A., Robinson, N. and Rosamond, B. (2011) New Regionalism and the European Union –​ Dialogues, Comparisons and New Research Directions. Oxon: Routledge. Warleigh-​ Lack, A. and Van Langenhove, L. (2012) Rethinking European Studies: The Contribution of Comparative Regionalism, Journal of European Integration 32(6), pp. 541–​ 62. doi: 10.1080/​ 07036337.2010.518715. Warleigh-​Lack, A. (2015) Differentiated Integration in the European Union: Towards a Comparative Regionalism Perspective, Journal of European Public Policy 22(6), pp. 871–​ 87. doi: 10.1080/​ 13501763.2015.1020837. Webber, D. (2019) European Disintegration? The Politics of Crisis in the European Union. Basingstoke: Palgrave Macmillan. Winand, P. (1997) American “Europeanists”, Monnet’s Action Committee and British Membership, in G. Wilkes (ed) Britain’s Failure to Enter the European Community, 1961–​63: The Enlargement. Oxon: Routledge, pp. 164–​90. Wintour, P. (2019) EFTA Countries Wary of UK’s Interest in ‘Norway’ Option. The Guardian, 27 March, available at: www.theguardian.com/​politics/​2019/​mar/​27/​efta-​countries-​wary-​of-​commons-​interest-​ in-​norway-​brexit-​option (accessed 7 August 2020). Wunderlich, J.-​U. (2007) Regionalism, Globalisation and International Order –​Europe and Southeast Asia. Aldershot: Ashgate. Zielonka, J. (2007) Europe as Empire. The Nature of the Enlarged European Union. Oxford: Oxford University Press. Zielonka, J. (2014) Is the EU Doomed?. Cambridge: Polity Press.

65

5 De facto differentiation in the European Union Circumventing rules, law, and rule of law Tobias C. Hofelich

Introduction As a consequence of increasing socio-​economic and political heterogeneity among its member states, the European Union (EU) has to cope with divergent policy preferences in its pursuit of ‘ever closer union.’ The remedy of choice has often been to engage in various forms of differentiated integration which has become an integral part of EU governance and attracts considerable scholarly attention (Dyson and Sepos 2010; Leuffen et al. 2013; Schimmelfennig and Winzen 2020). More recently and in particular in light of Brexit, research has begun to turn to differentiated disintegration (Schimmelfennig 2018; Leruth et al. 2019a). The prevalence and multi-​directionality of differentiated processes of European integration has led to a more comprehensive concept of differentiation, seen as a key feature of European integration (Gänzle et al. 2020). This chapter contributes to this broadened scope by addressing a distinct form of differentiation that has long been overlooked. The vast majority of academic literature discusses formal arrangements of differentiation in primary and secondary EU laws (Leuffen et al. 2013; Duttle et al. 2017). Understudied in comparison remains differentiation that is not codified in EU law. This scholarly neglect is surprising, considering the high relevance of the policy areas affected. For example, in Economic and Monetary Union (EMU), both Sweden’s de facto opt-​out and the unilateral adoption of the euro by Kosovo and Montenegro constitute differentiation that is not recognized by EU law. Moreover, the original Schengen Agreement was conceived as an intergovernmental treaty and, signed by only five member states, created differentiation outside the EU legal framework. More research is needed to fully grasp this distinct type of differentiation in the EU. As yet, we know little about its origins and ramifications for EU integration. The key objective of this contribution is to lay the groundwork for a new research agenda that addresses these gaps in knowledge. To that end, the chapter proceeds as follows. In the next section, I review and assess the literature on differentiation in the EU. On that basis, I subsequently develop a typology and workable definition. Finally, I present three distinct cases of de facto differentiation in EMU and discuss their respective implications from the institutional perspective of the EU. 66

DOI: 10.4324/9780429054136-6

De facto differentiation

De facto differentiation in the literature The idea that differentiation can also be established outside of formal opt-​outs is not entirely new. It first entered academic debates after the EU’s enlargement by ten new member states in 2004. In view of limited resources, constrained administrative capacities, and the widely different institutional traditions among the predominantly post-​communist new member states, Andersen and Sitter (2006) called for including informal aspects of differentiation in the study of differentiated integration. Inspired by organizational theory and Europeanization literature, the authors sought to connect the dots between the dynamics of integration and variation in the actual impact of EU policy across sectors and states. Their main contribution is a typology1 of differentiated integration in secondary law, which includes three types of informal arrangements of policy opt-​outs. Most notable among these is the concept of deviant integration, whereby differentiation is established by circumventing EU law. Essentially based on some form of non-​ compliance, this type of differentiation was conceived to be the result of limited state capacity and/​or strong domestic resistance against the implementation of a certain EU policy. Few scholars of EU integration have since heeded their call. Notable exceptions include Dyson and Sepos (2010: 4) who mentioned informal arrangements in their definition of differentiated integration. In the same volume, Howarth (2010) listed non-​compliance as one aspect of informal differentiation in EU industrial policy and referred to breaches of the EU’s limit of state aid as an example. In a similar vein, Holzinger and Schimmelfennig (2012) reference Andersen and Sitter in a side note by acknowledging the possibility that member states may choose non-​compliance over negotiating differentiation to avoid costly policy obligations. Sweden’s reluctance to adopt the euro is often referred to as a de facto opt-​out in EMU (Adler-​Nissen 2009; Schimmelfennig et al. 2015). Following the result of a referendum in 2003, Sweden has since taken no steps to adopt the euro, despite being obliged to do so under the Maastricht Treaty. The case study of Jensen and Slapin (2012) on ‘cascade effects,’ in which the opt-​out of one state is said to encourage other member states to follow suit, specifically deals with this case. Surprisingly, the informal nature of Sweden’s opt-​out was neither discussed nor put in context with their findings. Using the same case as an example, Leruth et al. (2019b) conceptualize de facto differentiation as one of five empirical models of differentiation in the EU. In their study, de facto differentiation resembles the concept of deviant integration in that it depicts differentiation based on non-​compliance with EU law. Falling short of providing a full-​fledged definition for de facto differentiation, the article briefly discusses its limitations and risks. Primarily, de facto differentiation hinges on tolerance from the EU Commission. Its longevity is, therefore, uncertain because the legal questions raised by such arrangement will sooner or later be addressed. However, the authors argue that de facto differentiation may become permanent if the socio-​economic and political conditions for further integration are not met. The term de facto differentiation has also been mentioned in context with Poland and Hungary’s ongoing violations of the EU’s rule of law principles, albeit only as a sidenote (Schimmelfennig 2019). Thereby, Schimmelfennig referred to a recent contribution from Kelemen (2019), wherein he assessed the normative implications of differentiation in rule of law. Although acknowledging that Poland and Hungary have already established a national legal basis in violation of the EU’s rule of law principles, Kelemen did not further elaborate on whether the current situation constitutes some kind of informal differentiation even now. Eriksen (2019) views de facto differentiation from a spatial perspective that resembles the concept of variable geometry as defined by Stubb (1996). In that regard, Eriksen mentions 67

Tobias C. Hofelich

the Eurogroup, an EU Council subcommittee comprising the finance ministers of Eurozone member states, which indeed distinguishes itself from the EU-​27 by a different level of integration. Since the financial crisis hit Europe, Eriksen contends that non-​Eurogroup states have been ‘downgraded to a secondary status,’ thus establishing de facto differentiation. This assessment is grounded in the analysis of Avbelj (2013) who asserted that measures taken outside the EU legal framework such as the Fiscal Compact have created a new EU-​17 that could replace the original EU-​27. Although mostly just in passing, the literature acknowledges that differentiation can be established by means other than formal agreements. As yet, there exists no comprehensive concept that embraces the variety of associated cases, ranging from non-​compliance to international treaties. While the term de facto differentiation appears to have caught on recently, conceptual ambiguity prevails, and the potentially different causes and implications of this type of differentiation remain opaque. These questions shall be addressed, albeit tentatively, in section four. In the following section, I lay the conceptual groundwork by elaborating on the different types of informal arrangements of differentiation in the EU and by developing a workable definition.

Conceptualizing de facto differentiation For terminological clarity and brevity in describing these phenomena, I adopt the term de facto differentiation. Although Andersen and Sitter’s concept of deviant integration paved the way for this hitherto nascent sub-​discipline, it fails to capture the disintegrative direction which is now considered an inherent part of differentiation in the EU. Moreover, deviant integration is rather narrowly defined to describe informal differentiation that stems from non-​compliance, whereas the somewhat elusive term de facto differentiation still allows for broader conceptualization. In order to develop a comprehensive concept of de facto differentiation, it is first necessary to clarify the individual constituents of this composite term. This volume understands differentiation as an umbrella term that covers both differentiated integration and differentiated disintegration, as well as their respective modes multi-​speed, variable geometry, and à la carte (Stubb 1996). In that vein, Schimmelfennig (2018) offers an inclusive definition of differentiation as processes of an unequal increase or reduction in the centralization level, policy scope, or membership of the EU.2 Both EU member states and non-​members are included in this concept and represented in the different types of internal and external differentiation. The addition of the term ‘de facto’ allows for further distinction and refers to the nature of this particular type of differentiation. In legal studies, the term refers to practices or conditions that exist in reality but are not officially recognized by binding law, in this case EU law.This is what distinguishes de facto differentiation from de jure differentiation. In that sense, de facto differentiation describes cases that fall under the definition of differentiation but are not enshrined or recognized by EU law. That is not to say, however, that the practice is per se illegal. With basic terminology out of the way, it is time to address the two types of de facto differentiation identified in the literature. Both involve the circumvention of EU law. This can either be grounded in non-​compliance, as in Sweden’s opt-​out from EMU, or in some form of cooperation outside the EU legal framework, as in the Fiscal Compact. To develop a generalizable typology, this requires further specification. Clearly, not all kinds of cooperation outside the EU and not all acts of non-​compliance fall under the definition of differentiation. In the final subsection, I present a third type of de facto differentiation that has thus far been ignored.

68

De facto differentiation

De facto differentiation by non-​compliance Non-​compliance by nature creates a de facto suspension of otherwise valid EU law. In the most general sense, this can be considered differentiation because it effectively and unequally reduces the centralization level or membership in a certain policy area. Although EU law remains uniformly valid, its unequal application creates a situation in which citizens or businesses situated in compliant member states are affected by EU law, whereas those registered in non-​compliant member states effectively are not. This is, however, not to say that all cases of non-​compliance constitute or lead to de facto differentiation. Non-​compliance is a fairly common phenomenon in the EU, as well as in other international organizations. Recent research by Falkner (2018) attests to a ‘non-​trivial degree of non-​compliance with EU policy.’ Explanations for the existence of non-​compliance in international organizations are dominated by two schools of thought: the management approach and the enforcement approach (Tallberg 2002). The managerial school assumes that non-​compliance is involuntary and due to a lack of administrational capacity, vague wording, or inadequate timetables (Chayes and Handler Chayes 1993). By that logic, international organizations can manage non-​compliance by providing aid in implementation, but their resources are limited as well. In 2017, the EU Commission admitted that its capacities to enforce compliance were limited and that it strategically prioritized cases that obstruct the implementation of important policy objectives or risk undermining the four freedoms (European Commission 2017). In contrast, the enforcement approach posits that non-​compliance is intentional and serves states to avoid the costs of compliance (Downs et al. 1996). To enforce compliance, international organizations employ monitoring and sanctioning schemes to increase the costs of non-​compliance vis-​á-​vis the costs of compliance. The EU possesses several enforcement instruments, including the Commission’s infringement procedure, preliminary rulings and naming and shaming (Hartlapp 2007). As member states consider the costs of compliance, international organizations must take into account the costs of enforcement. In the same vein, König and Mäder (2014) contend that the EU Commission considers the likelihood and costs of enforcing compliance before taking action. Moreover, Steunenberg (2010) found that the Commission only acts if supported by the European Court of Justice (ECJ) or other member states. Whether non-​compliance constitutes de facto differentiation can be determined by applying a causal and a temporal criterium. In line with the enforcement and management approach, Tallberg (2002) divides the causes of non-​compliance in two main categories: non-​compliance as a preference and non-​compliance due to capacity limitations. In the latter case, non-​compliance is unintentional, and, crucially, states have no interest in differentiation. Furthermore, states have no incentive to protract non-​compliance, because the Commission’s infringement procedure incurs non-​compliance costs already during the early stages. Before being met with financial sanctions as the ultima ratio, non-​compliance can diminish reputation among other member states (Downs and Jones 2002). In addition, the negative press associated with court cases may incur domestic audience costs (Chaudoin 2014). Conversely, the picture can be very different if non-​compliance is a choice resulting from a rational evaluation of costs and benefits. Strong resistance from interest groups can incur significant compliance costs, which may entice states to prolong non-​compliance and drag out the infringement procedure (Hofmann 2018). By doing that, states create de facto differentiation that remains in effect until the state complies. Following Schimmelfennig and Winzen (2014), exemptions from EU legal rules qualify as differentiated integration if they last for at least one year. Although explicitly excluded in their contribution, de facto differentiation by non-​compliance should also be delimited by a temporal

69

Tobias C. Hofelich

criterium. Most registered cases of non-​compliance are due to failures of notification or untimely transposition, and most cases are resolved in the early stages of the EU Commission’s infringement procedure (European Commission 2020a). This resonates with the above-​mentioned causal differences between non-​compliance due to capacity issues and different preferences. Börzel et al. (2012) further estimate that by the time the EU initiates infringement procedures, member states that first ran into capacity issues should be able to redistribute resources to comply with the respective policy. In contrast, non-​compliance due to divergent preferences is more likely to go through more stages of the infringement procedure. As the intervals between the different stages are not uniform, setting the temporal criterium to a specific number of months or years is difficult. Instead, I suggest considering cases of non-​compliance de facto differentiation if non-​compliance endures after the Commission’s issuing of a reasoned opinion. At this second stage of the infringement procedure, the culprit state has failed to prove compliance and capacity issues could have already been addressed. To conclude, it is the wilful protraction of non-​compliance that can be considered de facto differentiation.The bulk of such cases affects secondary law. In fact, the EU Commission’s infringement database is rife with cases in which compliance with secondary law could be restored only after ruling by the ECJ. Some cases are even taken to court a second time, following which incessantly non-​compliant states may face financial penalties according to Article 260 (TFEU). However, Falkner (2018) demonstrates that even less wealthy states are not always deterred by financial penalties and choose to pay rather than comply. Although fewer in numbers, there are significant examples of de facto differentiation established by non-​compliance with primary law. This includes Sweden’s de facto opt-​out from EMU, which will be discussed in the following section. Poland and Hungary’s de facto differentiations in the rule of law or freedom of press technically belong in that category as well. Despite difficulties to pin this down to a specific EU policy, both states have passed national legislation undermining principles outlined in Article 2 (TEU).

De facto differentiation by cooperation outside the EU Aside from certain areas in which the EU has exclusive competences, e.g. international trade agreements, member states may engage in various types of international cooperation. Neoliberal institutionalists assert that an increasingly interdependent world economy demands cooperation within the framework of international organizations and regimes (Mitrany 1975; Keohane 1984). The EU is but one of many international organizations its member states participate in, and membership in other international regimes is not uniformly distributed among EU member states. It is, in fact, rather diverse. Following the neoliberal school, one could attribute this to states’ diverse and issue-​specific interests (Keohane 1986). If these interests cannot be accommodated within the EU, member states may be enticed to engage in other bilateral or multilateral regimes that comprise more likeminded allies.The same argument basically underlies the liberal intergovernmentalist interpretation of the emergence of differentiation within the EU (Schimmelfennig 2019). The diverse international regimes EU member states participate in range from bilateral agreements to membership in various types of international organizations. On the most basic level, states cooperate by way of bilateral agreements concerning specific issues. For instance, France and Germany have a long history in bilateral industrial cooperation. The two states have pooled competencies and resources i.e. to establish Airbus as a global player in the aviation industry or, more recently, to advance research and development in artificial intelligence. One step above that, EU member states increasingly engage in smaller multilateral networks to coordinate 70

De facto differentiation

policy in specific areas. For example, Denmark, Sweden, and Finland cooperate with other Nordic states in the Hanseatic League, and so do the central European member states Poland, Czechia, Slovakia, and Hungary within the framework of the Visegrád Group. Furthermore, subsets of EU member states are members in international organizations with different degrees of policy coordination and influence. For example, all EU member states except Austria, Ireland, Sweden and Finland are NATO members, only France, Germany, and Italy are in the G7, and only France holds a permanent seat in the United Nations (UN) Security Council. To be clear, while none of the above-​mentioned examples are codified in EU law, they do not qualify as differentiation. This is because neither the centralization level nor the policy scope or membership of the EU is affected. A positive case for de facto differentiation by cooperation outside the EU –​in this case, its predecessor, the European Communities (EC) –​is the 1985 Schengen Agreement between France, Germany, and the Benelux states. The abolition of passport controls for intra-​EC border crossings had long been discussed among the then ten member states of the EC. Due to concerns related to national sovereignty, consensus could not be reached among the EC-​10. To escape this deadlock, this integration step took the shape of an international treaty outside the EC legal framework.This type of de facto differentiation persisted until the Schengen Agreement was incorporated into EU law by the 1997 Treaty of Amsterdam.

De facto differentiation by unilateral opt-​ins Although not mentioned in the literature, de facto differentiation can also be established via unilateral opt-​ ins into the acquis communautaire by non-​ member states. The literature on Europeanization shows that the EU affects domestic policy, politics and polities not only among member states, but also beyond its borders (Sedelmeier 2011). Official and prospective candidates for EU membership adopt EU rules, standards and norms to meet the accession criteria defined by the EU Commission. But even in areas that are unlikely to be integrated into the EU in the foreseeable future, Europeanization can be observed in specific policy areas (Gänzle and Müntel 2011). The partial integration of non-​member states into the acquis is generally recognized as a form of external differentiation (Leuffen et al. 2013; Leruth et al. 2019b). Third states choose to ‘opt-​in’ either unilaterally or because they are induced to do so by the EU (Holzinger and Tosun 2019). Most research in that area deals with the latter, either in context with requirements for European Economic Area (EEA) members or the policy alignment targets of EU neighbourhood policy (Lægreid et al. 2004; Börzel 2011). While receiving some mention, the ‘de facto interaction’ with non-​member states or ‘de facto opt-​ins’ remain understudied (Lavenex 2015; Cianciara and Szymanski 2020). Such de facto opt-​ins are unilateral steps taken without the explicit approval from the EU. In most cases, they take the shape of adopting standards or norms. In some rare cases, unilateral opt-​ins allow for free-​riding and access to excludable collective goods. For instance, Kosovo and Montenegro have both unilaterally opted into parts of EMU by making the euro their official national currency.These cases will be discussed in further detail in the following section. In conclusion, de facto differentiation manifests in three distinct types. Each involves some way of circumventing the EU legal framework either by non-​compliance, cooperation outside the EU, or unilateral opt-​ins from non-​member states. Among these types, de facto differentiation by non-​compliance arguably bears the highest significance, because engaging in such differentiation actively contests the EU’s legal authority. Moreover, the numerous cases of lasting non-​compliance in secondary law make this type of de facto differentiation a far-​spread phenomenon. In comparison, the two other types are rather rare. 71

Tobias C. Hofelich

It was also the ambition of this section to propose a definition of de facto differentiation. A comprehensive definition must embrace all three types without becoming generic and, crucially, remain in line with the general understanding of differentiation. Taking this into account, I define de facto differentiation as a deliberate and enduring circumvention of the EU legal framework, which leads to an unequal increase or reduction of the centralization level, policy scope and/or membership of the EU.

De facto differentiation in economic and monetary union In this section, I present and discuss three empirical cases of de facto differentiation in EMU. EMU was established as part of the Maastricht Treaty in 1992 and comprises a set of policies and the adoption of the euro, through which the economic convergence of EU member states shall be facilitated. Crucially, EMU is a highly differentiated policy area. In addition to the UK’s and Denmark’s voluntary opt-​outs, its three-​staged integration process makes EMU one of only few policy areas that contain de jure differentiation by design. Only after a member state has moved into the third stage, it is permitted to adopt the euro and, thus, fully integrated. Full participation in EMU is further contingent on meeting five economic convergence criteria: member states must not exceed a predefined rate of inflation and long-​ term interest, stay below the 3% ratio of annual budget deficit to GDP and 60% overall debt to GDP, and participate in the exchange rate mechanism (ERM) which pegs member states’ national currency to the Euro. In the following subsections, I demonstrate that the high degree of differentiation in EMU extends to all three types of de facto differentiation. First, I introduce the three cases individually and elaborate on the circumstances under which they emerged and how they are today handled by the involved member states and EU institutions. In the final subsection, I take a more general perspective and discuss the implications of each type of de facto differentiation for European integration.

Sweden’s de facto opt-​out from EMU Sweden joined the EU in 1995 alongside Austria and Finland. As the Swedish government had not previously negotiated any opt-​outs, Sweden became subject to the entire acquis which, since the Treaty of Maastricht, obliges all EU member states except Denmark to adopt the Euro once the convergence criteria are met. Although mostly in favour of adopting the euro, the Swedish government was aware of the far-​spread public scepticism towards abandoning the krona for the euro and followed a cautious approach.3 In October 1995, the government appointed a commission tasked to assess the consequences of Swedish participation in EMU.The commission was headed by economist Lars Calmfors and comprised five economists and three political scientists. The report, delivered in October 1996, concluded that the economic arguments spoke against adopting the euro, whereas remaining outside the Eurozone would come at the loss of political influence (Calmfors et al. 1997). Ultimately, the Calmfors Commission suggested not to join EMU in the first wave in 1999 but to aim for future membership. In 1997, the Swedish Parliament (Riksdagen) acted accordingly and decided that Sweden would not join the ERM, which left one of the five convergence criteria deliberately unfulfilled. In order to settle the issue, the Swedish government put the question of introducing the euro to a legally non-​binding referendum, held in September 2003. With 55.9% of votes against the common currency and 42.0% in favour, the results were clear, and the government subsequently shelved any plans to adopt the euro. 72

De facto differentiation

Since then, Sweden has not undertaken any steps towards full participation in EMU and maintains the position that a future adoption of the euro must be decided by referendum. Thus, Sweden has established what is often called a de facto opt-​out that stands on no legal footing.The government-​funded Swedish Institute for European Policy Studies (SIEPS), however, argues that it is the government’s understanding that membership in ERM is voluntary (Campos et al. 2016). Conversely, the European Central Bank’s (ECB) annual convergence reports consistently stress that ‘Sweden has been under the obligation to adopt national legislation with a view to integration into the Eurosystem since 1 June 1998. As yet, no legislative action has been taken by the Swedish authorities […]’ (ECB 2020). From a legal viewpoint, the Swedish government’s position appears questionable. Because it is not inability but a deliberate decision not to fulfil the convergence criteria, Sweden’s opt-​out from monetary union can be considered a violation of its obligations as an EU member (Nergelius 2019). In that sense, Sweden has established de facto differentiation on the basis of non-​compliance with primary EU law. As of today, there are no indications on either side that this informal arrangement of differentiation will soon be dissolved or legalized by way of a formal opt-​out. The topic hardly ever comes up in Swedish politics, and recent opinion polls suggest that the public remains firmly opposed to adopting the euro.4 Likewise, the European Commission seems to have swept the issue under the carpet and has not taken any actions to enforce the adoption of the euro. A rare exception was former Commission President Jean-​Claude Juncker who stressed in his 2017 state of the union address: ‘The euro is meant to be the single currency of the European Union as a whole. All but two of our Member States are required and entitled to join the euro once they fulfil the conditions’ (Juncker 2017). In view of the loss of the United Kingdom as a strong ally with an opt-​out from EMU, Brianson and Stegmann McCallion (2020) caution that Sweden’s current status will become more difficult to maintain. Still, Sweden is far from alone in this issue. The Swedish method to put off ERM membership to avoid the adoption of the euro has been imitated by Poland, the Czech Republic and Hungary.

Kosovo and Montenegro’s de facto opt-​in to EMU Following the dissolution of the Socialist Federal Republic of Yugoslavia, Montenegro entered a political union with Serbia, which lasted until Montenegrin citizens voted for independence in the 2006 popular referendum. Between 1991 and 1994, the dinar underwent several episodes of hyperinflation, which caused economic distress and an increased use of foreign currencies, mostly US dollars and the German Deutsche Mark (DM). The dinar was subsequently replaced by the novi dinar which was pegged to the DM to achieve price stability. In 1999, the Montenegrin government decided to adopt the DM as a second official currency. This two-​pronged system was short-​lived, as the DM was raised to Montenegro’s sole legal tender in October 2000. This step was not coordinated with the German federal government but facilitated by the already widespread circulation of DM. Almost immediately after the EU began to issue euro banknotes and coins, the DM was replaced by the euro in June 2002. Similar to the use of the DM, the adoption of the euro was a unilateral decision by the Montenegrin government that had not been coordinated with the EU or the ECB, respectively. Kosovo adopted the euro as its official currency in January 2002, but on slightly different terms. After the war had ended in June 1999, one of the first regulations passed by the UN Interim Administration Mission in Kosovo (UNMIK) targeted monetary policy and allowed the use of DM and other foreign currencies. As in Montenegro, trust in the highly inflationary dinar was low, and the widespread use of the DM prompted its unilateral adoption as the de facto legal currency (Korovilas 2002). This step was not preceded by negotiations with the German central 73

Tobias C. Hofelich

bank (Bundesbank) or the ECB. In view of the EU’s impending cash changeover to the euro, the UNMIK authorized to adopt the euro as Kosovo’s official currency.5 Although the adoption of the euro was a unilateral step, the Kosovo Central Bank maintains that the process was supported by the ECB and EU member states (Tyrbedari 2006). The adoption of the euro in Montenegro and Kosovo clearly resembles de facto differentiation by unilateral opt-​in into the acquis that is not recognized by EU law. Moreover, the two states effectively bypass the convergence criteria and other instruments of fiscal policy coordination which, in addition to membership of the EU, mark a necessary precondition to using the euro. In that sense, Montenegro and Kosovo reap the benefits (and disadvantages) of a policy area that is designed as a collective good with exclusive access to EU member states. After eighteen years, it is safe to say that the de facto opt-​ins have turned into a tacit arrangement both Montenegro and Kosovo, as well as the EU seem to be content with.At the cost of monetary policy being made in Frankfurt without taking into consideration the economic conditions of non-​EU states (Padoa-​Schioppa 2003), the euro affords Kosovo and Montenegro stable inflation rates and has contributed to economic growth.6 At the same time, however, this arrangement creates a precedent for EU expansion and might affect both states’ prospective accessions. While Kosovo is currently only a potential candidate for EU membership, Montenegro has been an official candidate since 2010. Although the ‘unilateral euroization’ was initially a point of contention,7 the EU Commission’s latest report on future Montenegrin EU membership recognizes that the use of the euro was decided under ‘exceptional circumstances’ and does not specifically call for abandoning the chosen path (European Commission 2020b). For practical and political reasons, it appears unlikely that Kosovo and Montenegro will be required to abandon the euro only to reintroduce it sometime after becoming EU member states.

The fiscal compact The Treaty on Stability, Coordination and Governance in the EMU (TSCG) is most commonly referred to as the Fiscal Compact, after the title of the third and most significant section of the treaty. The TSCG is an intergovernmental treaty concluded outside of EU law, signed in March 2012 by all EU member states except the United Kingdom and the Czech Republic. It formed a major part of the EU’s response to the sovereign debt crisis that hit Europe in 2009. The TSCG expands the Stability and Growth Pact (SGP) and contains several measures to ensure and enforce fiscal discipline, establish closer economic policy coordination and institutionalize the governance of the Eurozone. All members of the Eurozone are bound by these measures, while non-​Eurozone signatories may opt-​in.8 Most notably, signatories must implement into national law –​preferably via constitutional amendment –​a provision for balanced government budgets and an automatic correction mechanism in case of deviation.The European Commission assumes the task to monitor compliance with the TSCG and shares the right with other contracting parties to bring non-​compliant states before the ECJ. If a breach of the TSCG is identified, the ECJ may impose penalty payments of up to 0.1% of the non-​compliant state’s GDP. These measures confer additional powers to the EU Commission that significantly expand its competences in monitoring fiscal policy as defined by the SGP (Craig 2012). For that reason, incorporating the TSCG into EU law would have required amendments to the EU treaties. Because the UK categorically vetoed any such amendments, the TSCG took the shape of an intergovernmental treaty outside the EU legal framework. Thus, the TSCG marks a clear case of de facto differentiation by cooperation outside the EU. It reinforces differentiation between the Eurogroup and member states not using the euro and contains in itself multiple layers of differentiated participation in the agreed-​upon measures. Ultimately, the EU Commission’s new 74

De facto differentiation

powers in monitoring and enforcing fiscal policy mark an unequal increase in the EU’s centralization level that goes beyond the monetary and economic policy instruments of EMU. Although the treaty foresees adoption into EU law at the latest five years after ratification, no notable steps in that direction have so far been undertaken.9 At this point, the integration of the TSCG is not a priority of the EU Commission or the member states. Understandably, the current agenda focuses on combatting the COVID-​19 pandemic and ameliorating the economic damage it caused.With the explicit support from the EU Commission, many member states have taken up additional sovereign debt to boost their economies. Hence, initiating legislative steps to ensure fiscal discipline would send a contradictive message and undermine these efforts. In the medium term, however, Brexit poses an opportunity to amend the treaties by adopting the TSCG. The United Kingdom had been the most fervent veto player in this regard, and since the Czech Republic’s opt-​in in 2019, all 27 member states of the EU have now ratified the TSCG.

Discussion The most basic observation the three cases in EMU invite to make is that they have each endured several years and appear to be ongoing. On the one hand, this demonstrates that de facto differentiation is not a recent phenomenon, but rather an ongoing and long-​lasting aspect of European integration. On the other hand, this begs the question whether the EU institutions see in de facto differentiation an intended means to create more flexibility, a tolerated deviation from the norm, or an undesired by-​product of disobedient member and non-​member states. At first glance, the two cases concerning the adoption of the euro seem to share being grounded in tacit arrangements between the involved state(s) and the EU institutions, whereas the EU Commission actively participated in the circumvention of EU law to deepen integration via the TSCG. But to address this question from a more general perspective requires looking at the purpose and potential consequences of each type of de facto differentiation. As the case of the TSCG showed, de facto differentiation that stems from cooperation outside EU law may help to escape legislative deadlock. Although not seen as an ideal solution, the EU tends to support such initiatives.10 This is understandable considering the EU’s guiding principle of ‘ever closer union.’ After all, closer cooperation outside the EU still effectively deepens integration, albeit not for all member states and not under the auspices of the EU. Moreover, the original Schengen Agreement has proven that intergovernmental treaties may eventually be incorporated into EU law. Nonetheless, such arrangements can be problematic, because integration outside the EU may fuel legal disputes regarding the primacy of EU law vis-​à-​vis international law. For instance, states charged with penalty payments according to the provisions of the Fiscal Compact might challenge that decision on the grounds that the ECJ and EU Commission have exceeded the capacities bestowed upon them by the Treaty of Lisbon. Similarly, de facto differentiation that stems from unilateral opt-​ins appears to be no major concern of the EU. Not only is policy alignment a prescribed step for candidate states and a key objective of EU neighbourhood policy, unilateral opt-​ins never did any real harm. This picture may be a bit different, if unilateral opt-​ins target collective goods inseparably linked to EU membership. In that case, the EU is wise to assess whether this produces negative externalities for the EU and its member states. In the case of Montenegro and Kosovo’s adoption of the euro, their small economies hardly affect EU monetary policy. Moreover, the Balkans’ unstable political situation in the early 2000s called for a cautious approach on the part of the EU, that facilitated economic recovery rather than asserting EU law. However, should unilateral opt-​ins indeed create negative externalities, the EU’s means to intervene are rather limited. Because the jurisdiction of the ECJ does not extend beyond the EU’s borders, the EU can only address unilateral 75

Tobias C. Hofelich

opt-​ins by applying political pressure, such as making their discontinuation a binding condition for accession to the EU. But this, of course, only applies to current and prospective candidate states and may counteract the interests of the Union. De facto differentiation established by non-​compliance is an entirely different beast, because it actively and deliberately undermines EU governance. The Swedish refusal to adopt the euro demonstrates that states use it to unilaterally accommodate divergent policy preferences. Due to the implied legal uncertainty, it can be assumed that states only resort to de facto differentiation by non-​compliance if de jure opt-​outs have proven or are assumed to be unavailable. More research is needed here. Because the EU has a natural interest in preserving the legal primacy of EU law, the EU Commission aims to inhibit differentiation created in such fashion. Its resources, however, are limited, and the enforcement of compliance may sometimes also present normative issues. If, for instance, the EU forced Sweden to adopt the euro, it would undermine the expressed will of Swedish citizens and potentially stoke Eurosceptic sentiments. To return to the question posed earlier, de facto differentiation is hardly ever a perfect solution, not for the EU and rarely for the involved states. In some cases, however, it may be the only way to accommodate insurmountable differences in policy preferences or to proceed integration. Because all three types pose legal challenges and threaten the EU’s authority, the EU is best advised to use or tolerate de facto differentiation, especially if established by non-​compliance, only under certain circumstances. With the powers bestowed upon the EU Commission and ECJ by Articles 258 and 260 (TEU), the EU possesses relatively sharp tools to combat de facto differentiation by non-​compliance. But, as mentioned in the third section, the lengthy legal procedure may entice states to protract non-​compliance and maintain de facto differentiation as long as the associated benefits outweigh the costs. In the end, the decision for de facto differentiation lies firmly in the hands of individual member states. Unlike de jure differentiation, the EU cannot prevent it but only react once it has already been established.

Conclusion In this chapter, I sought to anchor arrangements of differentiation operating outside EU law in the overall framework of differentiated integration in the EU. My contribution is largely grounded in knowledge that had been scattered across various disciplines of EU studies and complemented by simple empirical observations. On that basis, I developed a three-​pronged typology subsumed under the term de facto differentiation. I demonstrated this theoretical backbone by way of three exemplary cases in EMU. In conclusion, I argue that de facto differentiation is a distinct feature of European integration that deserves to be studied in its own right as well as in conjunction with de jure differentiation. My brief analysis of three cases in EMU has shown that de facto differentiation is not unproblematic, especially if grounded in non-​compliance. This chapter, however, only scratched the surface of the various reasons for and consequences of de facto differentiation. Rather than to tackle all these aspects head-​on, it was the intention of this chapter to lay the conceptual groundwork for future empirical research on de facto differentiation. The potential avenues for future research are manifold. Beyond the field of EU integration studies, de facto differentiation touches on several related subject areas and calls for more interdisciplinary cooperation. Related to the three subtypes, compliance research can contribute to a better understanding of the motivations and consequences of the long-​term manifestation of non-​compliance. Insights from Europeanization studies can further illuminate the spread of EU policies across the Union’s borders. Furthermore, de facto differentiation ties into research on informal governance, which can help clarify the EU’s motivations to participate in, or tolerate, 76

De facto differentiation

such arrangements. What is more, the democratic backsliding of Poland and Hungary can and should be studied from the perspective of de facto differentiation. Having effectively and unilaterally established differentiation by applying the norms and values specified in Article 2 (TEU) only partially, this presents a particularly harmful case of de facto differentiation that calls into question the EU’s authority. Ultimately, more empirical research is needed to address the two main questions associated with de facto differentiation: firstly, why and how does it come into being and secondly, how do the EU and the concerned member state(s) handle the legally unstable situation it creates?

Notes 1 Andersen and Sitter present four types of integration. ‘Homogenous integration’ does not depict differentiation, but cases in which EU policy uniformly affects all member states. ‘Aligned integration’ and ‘autonomous integration’ are each models in which EU law is specifically designed to provide leeway in the modalities of implementation or to accommodate divergent national preferences, respectively. While both account for variation in the transposition and application of EU law (cf. Zhelyazkova and Thomann 2021), they do not constitute differentiation if understood as the unequal validity of EU law across member states. In that sense, only the fourth type ‘deviant integration’ can be considered differentiation because the validity of EU law is de facto suspended. In other words, citizens or businesses located in states engaged in deviant integration remain unaffected by the piece of EU law in question. 2 Of course, the literature offers various other definitions (cf. Gänzle et al., this volume) and settling for one inevitably creates a bias. I adopted the Swiss-​German school’s strongly legal perspective, as it is most suitable to introduce and discuss a form of differentiation in which the circumvention of the EU legal framework plays a central role. 3 The European Commission’s polling instrument Eurobarometer found that 54% of Swedish survey participants rejected the common European currency, whereas only 29% spoke out in favour (Standard Eurobarometer 44, 1996). 4 The latest Eurobarometer found that 22% of the Swedish survey participants are ‘very much against the introduction [of the euro]’ and 40% ‘rather against the introduction’ (Flash Eurobarometer 487, 2020). 5 For more information, see UNMIK Administrative Direction No. 2001/​24. 6 For the 2019 fiscal year, the annual report of the central bank of Montenegro noted a 3.6% growth in GDP. Likewise, the central bank of Kosovo measured real growth of 4.2%. https://​bqk-​kos.org/​wp-​ cont​ent/​uplo​ads/​2020/​10/​CBK_​AR_​2​019.pdf; www.cbcg.me/​slike_​i_​fajlovi/​eng/​fajlovi/​fajlovi_​ publikacije/​god_​izv_​o_​radu/​cbcg_​annual_​report_​2019.pdf. 7 The annex of a COREPER document concerning the signature of the Stabilization and Association Agreement between the EU Commission and Montenegro reads: ‘The Council recalls that unilateral “euroisation” is not compatible with the Treaty, which foresees the eventual adoption of the euro as the endpoint of a structured convergence procedure within a multilateral framework.’ https://​data.consil​ ium.eur​opa.eu/​doc/​docum​ent/​ST-​13831-​2007-​REV-​1/​en/​pdf. 8 Denmark and Romania have chosen to opt-​in and to comply with the entire set of measures. Bulgaria maintains a partial opt-​in and is only bound to the fiscal measures. 9 According to the European Parliament, the EU Council has made ‘no visible progress’ in integrating the TSCG into EU law. www.europarl.europa.eu/​legislative-​train/​theme-​economic-​and-​monetary-​ affairs-​econ/​file-​integration-​of-​the-​fiscal-​compact-​into-​secondary-​eu-​law. 10 Günther Oettinger, at the time the German EU Commissioner for Budget and Human Resources, noted that the TSCG was a ‘good, second-​best solution.’ www.spiegel.de/​international/​europe/​split-​ summit-​the-​birth-​of-​a-​two-​speed-​europe-​a-​802703.html.

References Adler-​Nissen, R. (2009) Behind the scenes of differentiated integration: circumventing national opt-​ outs in Justice and Home Affairs. Journal of European Public Policy 16 (1), pp. 62–​80. doi: 10.1080/​ 13501760802453239. 77

Tobias C. Hofelich

Andersen, S. S., & Sitter, N. (2006) Differentiated integration: what is it and how much can the EU accommodate? European Integration 28 (4), pp. 313–​330. doi: 10.1080/​07036330600853919. Avbelj, M. (2013) Differentiated integration –​farewell to the EU-​27? German Law Journal 14 (1), pp. 191–​ 211. doi: 10.1017/​S2071832200001760. Börzel, T. A. (2011). When Europe hits … beyond its borders: Europeanization and the near abroad. Comparative European Politics 9 (4/​5), pp. 394–​413. doi: 10.1057/​cep.2011.8. Börzel, T. A., Hofmann, T., & Panke, D. (2012) Caving in or sitting out? Longitudinal patterns of non-​ compliance in the European Union. Journal of European Public Policy 19 (4), pp. 454–​471. doi: 10.1080/​ 13501763.2011.607338. Brianson, A., & Stegmann McCallion, M. (2020) Bang-​a-​boomerang? Sweden, differentiated integration and EMU after Brexit. In Gänzle, S., Leruth, B. and Trondal, J. (eds.), Differentiated integration and disintegration in a post-​Brexit era (pp. 146–​164). Abingdon: Routledge. Calmfors, L., Flam, H., Gottfries, N., Jerneck, M., Lindahl, R., Haaland Matlary, J., Nordh Berntsson, C., Rabinowicz, E., & Vredin, A. (1997) EMU –​a Swedish perspective. Dordrecht: Kluwer Academic Publishers. Campos, N. F., Coricelli, F., & Moretti, L. (2016) Sweden and the Euro: the neglected role of EU membership. SIEPS European Policy Analysis 2016, p. 15. Chaudoin, S. (2014) Promises or policies? An experimental analysis of international agreements and audience reactions. International Organization 68 (1), pp. 235–​256. doi: 10.1017/​S0020818313000386. Chayes, A., & Handler Chayes, A. (1993) On compliance. International Organization 47 (2), 175–​205. Cianciara, A. K., & Szymanski, A. (2020) Differentiation, Brexit and EU-​Turkey relations. In Gänzle, S., Leruth, B. and Trondal, J. (eds.), Differentiated integration and disintegration in a post-​Brexit era (pp. 182–​202). Abingdon: Routledge. Craig, P. P. (2012) The stability, coordination and governance treaty: principle, politics and pragmatism. European Law Review 37 (3), pp. 231–​248.Downs, G. W., Rocke, D. M., & Barsoom, P. N. (1996) Is the good news about compliance good news about cooperation? International Organization 50 (3), pp. 379–​406. Downs, G. W., & Jones, M. A. (2002) Reputation, compliance, and international law. Journal of Legal Studies 31 (S1), pp. 95–​114. Duttle, T., Holzinger, K., Malang, T., Schäuble, T., Schimmelfennig, F., & Winzen, T. (2017) Opting out from European Union legislation: the differentiation of secondary law. Journal of European Public Policy 24 (3), pp. 406–​428. Dyson, K., & Sepos, A. (eds.) (2010) Which Europe? The politics of differentiated integration. Basingstoke: Palgrave Macmillan. Eriksen, E. O. (2019) Contesting political differentiation –​European division and the problem of dominance. Switzerland: Springer Nature. European Central Bank (2020) Convergence Report. Frankfurt, June 2020, accessed 24.01.2021, www.ecb. europa.eu/​pub/​convergence/​html/​ecb.cr202006~9fefc8d4c0.en.htm. European Commission (2017) EU law: better results through better application. Official Journal of the European Union C 18/​02. https://​ec.eur​opa.eu/​info/​publi​cati​ons/​commun​icat​ion-​com​miss​ion-​eu-​ law-​bet​ter-​resu​lts-​thro​ugh-​bet​ter-​app​lica​tion​_​en. European Commission (2020a) Member States’ Compliance with EU Law in 2019: More Work Needed. Brussels, 31.07.2020, accessed 04.10.2020, https://​ec.eur​opa.eu/​info/​publi​cati​ons/​2019-​com​miss​ion-​ rep​ort-​and-​fac​tshe​ets-​mon​itor​ing-​appl​icat​ion-​eu-​law​_​en. European Commission (2020b) Montenegro 2020 Report. Brussels, 06.10.2020, accessed 24.01.2021, https://​eur-​lex.eur​opa.eu/​legal-​cont​ent/​EN/​TXT/​PDF/​?uri=​CELEX:5202​0SC0​353&from=​EN. Falkner, G. (2018) A causal loop? The Commission’s new enforcement approach in the context of non-​ compliance with EU law even after CJEU judgments. Journal of European Integration 40 (6), pp. 769–​784. doi: 10.1080/​07036337.2018.1500565. Flash Eurobarometer 487 (2020) Introduction of the euro in the Member States that have not yet adopted the common currency. Brussels: European Commission. https://​eur​opa.eu/​euroba​rome​ter/​surv​eys/​det​ ail/​2261. Gänzle, S., & Müntel, G. (2011) Europeanization ‘beyond’ Europe? EU impact on domestic policies in the Russian enclave of Kaliningrad. Journal of Baltic Studies 42 (1), pp. 57–​79. Gänzle, S., Leruth, B., & Trondal, J. (2020) Differentiation, differentiated integration and disintegration in a ‘post-​Brexit-​era.’ In Gänzle, S., Leruth, B. and Trondal, J. (eds.), Differentiated integration and disintegration in a post-​Brexit era (pp. 1–​19). Abingdon: Routledge.Hartlapp, M. (2007) On enforcement, management 78

De facto differentiation

and persuasion: different logics of implementation policy in the EU and the ILO. Journal of Common Market Studies 45 (3), pp. 653–​674. doi: 10.1111/​j.1468-​5965.2007.00721.x. Hofmann, T. (2018) How long to compliance? Escalating infringement proceedings and the diminishing power of special interests. Journal of European Integration 40 (6), pp. 785–​ 801. doi: 10.1080/​ 07036337.2018.1500564. Holzinger, K., & Schimmelfennig, F. (2012) Differentiated integration in the European Union: many concepts, sparse theory, few data. Journal of European Public Policy 19 (2), pp. 292–​305. doi: 10.1080/​ 13501763.2012.641747. Holzinger, K., & Tosun, J. (2019) Why differentiated integration is such a common practice in Europe: a rational explanation. Journal of Theoretical Politics 31 (4), pp. 642–​659. doi: 10.1177/​0951629819875522. Howarth, D. (2010) ‘Industrial’ Europe: the softer side of differentiated integration. In Dyson, K. and Sepos, A. (eds.), Which Europe? The politics of differentiated integration (pp. 233–​250). Basingstoke: Palgrave Macmillan. Jensen, C. B., & Slapin, J. B. (2012) Institutional Hokey-​Pokey: the politics of multispeed integration in the European Union. Journal of European Public Policy 19 (6), 779–​ 795. doi: 10.1080/​ 13501763.2011.610694. Juncker, J. C. (2017, September 13) President Jean-​Claude Juncker’s State of the Union Address 2017. European Commission. https://​ec.eur​opa.eu/​com​miss​ion/​pres​scor​ner/​det​ail/​en/​SPE​ECH_​17_​3​165. Kelemen, D. R. (2019) Is differentiation possible in rule of law? Comparative European Politics 17, pp. 246–​ 260. doi: 10.1057/​s41295-​019-​00162-​9. Keohane, R. O. (1984) After hegemony –​cooperation and discord in the world political economy. Princeton: Princeton University Press. Keohane, R. O. (1986) Neo-​realism and its critics. New York: Columbia University Press. König,T., & Mäder, L. (2014) The strategic nature of compliance: an empirical evaluation of law implementation in the central monitoring system of the European Union. American Journal of Political Science 58 (1), pp. 246–​263. doi: 10.1111/​ajps.12038. Korovilas, J. P. (2002) The economic sustainability of post-​conflict Kosovo. Post-​Communist Economies 14 (1), pp. 109–​121. doi: 10.1080/​14631370120116716.Lægreid, P., Steinthorsson, R. S., & Thorhallsson, B. (2004) Europeanization of central government administration in the Nordic states. Journal of Common Market Studies 42 (2), 347–​369. doi: 10.1111/​j.1468-​5965.2004.00491.x. Lavenex, S. (2015) The external face of differentiated integration: third country participation in EU sectoral bodies. Journal of European Public Policy 22 (6), pp. 836–​853. doi: 10.1080/​13501763.2015.1020836. Leruth, B., Gänzle, S., & Trondal, J. (2019a) Exploring differentiated integration in a post-​Brexit European Union. Journal of Common Market Studies 57 (5), pp. 1013–​1030. doi: 10.1111/​jcms.12869. Leruth, B., Gänzle, S., & Trondal, J. (2019b) Differentiated integration and disintegration in the EU after Brexit: risks versus opportunities. Journal of Common Market Studies 57 (6), pp. 1383–​1394. doi: 10.1111/​ jcms.12957. Leuffen, D., Rittberger, B., & Schimmelfennig, F. (2013) Differentiated integration –​explaining variation in the European Union. Basingstoke: Palgrave Macmillan. Mitrany, D. (1975) The functional theory of politics. London: St. Martin’s Press for the London School of Economics and Political Science. Nergelius, J. (2019) The constitution of Sweden and European influences: The changing balance between democratic and judicial power. In Albi, A. and Bardutzky, S. (eds.), National constitutions in European and global governance: democracy, rights, the rule of law (pp. 315–​358). The Hague: Asser Press. Padoa-​Schioppa, T. (2003) The Euro goes east. Comparative Economic Studies 45, pp. 215–​231. doi: 10.1057/​ palgrave.ces.8100016. Schimmelfennig, F., & Winzen, T. (2014) Instrumental and constitutional differentiation in the European Union. Journal of Common Market Studies 52 (2), pp. 354–​370. doi: 10.1111/​jcms.12103. Schimmelfennig, F., Leuffen, D., & Rittberger, B. (2015) The European Union as a system of differentiated integration: interdependence, politicization and differentiation. Journal of European Public Policy 22 (6), 764–​782. doi: 10.1080/​13501763.2015.1020835. Schimmelfennig, F. (2018) Brexit: differentiated disintegration in the European Union. Journal of European Public Policy 25 (8), pp. 1154–​1173. doi: 10.1080/​13501763.2018.1467954. Schimmelfennig, F. (2019) The choice for differentiated Europe: an intergovernmentalist theoretical framework. Comparative European Politics 17 (2), pp. 176–​191. doi: 10.1080/​13501763.2018.1467954.Schim melfennig, F., & Winzen, T. (2020) Ever looser union. Differentiated European integration. Oxford: Oxford University Press. 79

Tobias C. Hofelich

Sedelmeier, U. (2011) Europeanisation in new member and candidate states. Living Reviews in European Governance 6 (1), doi: 10.12942/​lreg-​2011-​1. Standard Eurobarometer (1996) Report Number 44. Brussels: European Commission. https://​eur​opa.eu/​ euroba​rome​ter/​surv​eys/​det​ail/​1416. Steunenberg, B. (2010) Is big brother watching? Commission oversight of the national implementation of EU directives. European Union Politics 11 (3), pp. 359–​380. doi: 10.1177/​1465116510369395. Stubb, A. (1996) A categorization of differentiated integration. Journal of Common Market Studies 34 (2), pp. 283–​295. doi: 10.1111/​j.1468-​5965.1996.tb00573.x. Tallberg, J. (2002) Paths to compliance: enforcement, management, and the European Union. International Organization 56 (3), pp. 609–​643. doi: 10.1162/​002081802760199908. Tyrbedari, S. A. (2006) Euroisation outside euro-​zone: assets and challenges –​the experience of Kosovo. Banking and Payments Authority of Kosovo, Working Paper No. 2. Zhelyazkova, A., & Thomann, E. (2021) ‘I did it my way’: customisation and practical compliance with EU policies. Journal of European Public Policy. doi: 10.1080/​13501763.2020.1859599.

80

6 Constitutive differentiation John Erik Fossum

Introduction Constitutive differentiation refers to the manner in which differentiation –​in territorial, functional, hierarchical and person-​incorporation terms –​is configured in the key principles and institutional arrangements that make up the political system.1 Constitutive differentiation is therefore about the fundamental architectonic of differentiation that underpins the political system. Constitutive differentiation, as the name implies, has a prescriptive dimension, in the sense that it directs attention to how we need to configure differentiation for the political system to be properly constituted and to function.2 In a modern world, which is committed to the principle of democracy, the polity’s differentiation configuration has implications for the character and quality of democracy. At the same time, our very notion of what democracy is and what it should be is profoundly affected by the prevailing (nation-​state-​based) differentiation configuration. Since the European Union (EU) is committed to democracy, our discussion of constitutive differentiation in the EU context includes democracy. The utility and practical relevance of the notion of constitutive differentiation hinges on specification: how the key dimensions of differentiation (territorial, functional, hierarchical and incorporation of persons through such means as citizenship and rights) manifest themselves in concrete constitutional and institutional arrangements and procedures. Specification brings up the issue of the nature of the relationship –​the gap –​between constitutive principles and practical reality. On the one hand, each polity be it state, or international organization, will come with a distinct differentiation configuration, which refers to how the core constitutive dimensions are configured. In the context of the sovereign state, these radiate from the political centre; in the context of the international organization, these radiate from the states that are the members of the international organization. On the other hand, even within a specific category of polity, such as a state, there is considerable variation in how the dimensions are configured, and how they operate, in practice. Our approach must therefore be attentive to principle and practice. As I will show, this is a particularly important concern in the context of the EU. The chapter consists of three parts. In the first part, I start by spelling out the distinct challenges associated with clarifying the nature of the EU’s differentiation configuration, and the need to relate back to the nation-​state. In the second part, I outline the constitutive principles DOI: 10.4324/9780429054136-7

81

John Erik Fossum

of the nation-​state, key to which is sovereignty, and show how that shapes the distinct nation-​ state differentiation configuration. I thereafter show that the EU transforms state sovereignty. Nevertheless, the lack of agreement on which polity configuration should constitute the EU makes it difficult to spell out a distinct EU differentiation configuration. Comparing the EU’s actual differentiation pattern with the nation-​state configuration is lopsided, due to the uncertain status of the EU configuration in constitutive terms (including the EU’s democratic deficit). In order to make up for that uncertainty, in the third and final substantive part, I present two EU constitutional models. The first has roots in intergovernmentalism. I unpack it with reference to the differentiation dimensions that I discerned from the nation-​state model. Thereafter, I list some of the intergovernmentalism-​inspired accounts of the EU. The exercise shows that such accounts must exceed well beyond intergovernmentalism to account for present-​day EU. The second constitutional model has roots in federalism. I show that this model also must be stretched to take the EU properly into account. Including both these models pay attention to the fact that there is contestation over whether the EU is mainly an organization steeped in intergovernmental principles or an organization steeped in supranational –​federal-​type –​principles. The EU embodies intergovernmental and supranational principles in its organizational and structural make-​up. The issue is whether these can be reconciled or whether the EU needs to be reformed to cohere better with an intergovernmental or federal constitutional model.

Constitutive differentiation and the EU: what are the challenges? One important challenge in terms of understanding constitutive differentiation in the EU relates to the hegemonic role of the nation-​state, not the least because it has figured so centrally in the normative imagination as the mainstay of democracy. This resonates with social science’s methodological nationalism (Beck 2003; Wimmer and Glick-​Schiller 2002). In principle, federalism, subsidiarity and cosmopolitanism, all of which are principles that predate the modern state could each have given rise to an alternative set of constitutive differentiation principles, and analysts have discussed them as such. These efforts have encountered their own problems and pitfalls. The development of modern federalism, especially the American federal experience, has infused the modern federal template with a strong underlying statist logic. The oddness of all federal theory…is due to its having seized on the very concept (sovereignty) which it actually opposed…its oddity is due to federalism’s having defined the autonomy of its territorial units in terms of sovereignty, whereas in fact it would have done better to try to overturn the idea of sovereignty; all efforts to divide what could only be conceived precisely in terms of total unity drew federal theory into constant paradox and contradiction. Riley 1973: 88 The fact that modern federalism has been reconfigured to cohere with state-​based sovereignty means that federalism’s combination of shared rule and self-​rule has to be worked out within the confines of a sovereign state framework.The upshot is that federalism copies much of the differentiation configuration of the nation-​state even if federalism is based on a more complex communal logic –​compound communities –​than nationalism’s unity-​centred approach to community. Subsidiarity posits that the higher level is subsidiary to the lower level and has a duty to help the lower level to realize its full potential. A polity organized along the lines of the principle of 82

Constitutive differentiation

subsidiarity would be based on a configuration of differentiation that diverges sharply from the nation-​state-​based one. Subsidiarity figures centrally in the EU, as spelled out in the EU Treaties (Art 5 and 12 TEU). Rather than serve as the key to structuring the EU’s differentiation configuration, along all relevant levels of governing, including the regional (sub-​state level), member states have largely appropriated it (Føllesdal 2000). They can use it as an important check on EU integration. Some analysts have tried to set out a cosmopolitan alternative to the nation-​state and have claimed that the EU is a regional cosmopolitan vanguard.3 The absence of actually existing cosmopolitan polities makes this an uneven comparison. We are very familiar with nation-​ state-​based constitutive differentiation –​even down to the smallest detail –​but have no similar experience of how to get from rather abstract cosmopolitan principles to those similarly specific institutional and procedural arrangements that would qualify as cosmopolitan. What in practice would figure as constitutive differentiation in a cosmopolitan context is thus shrouded in uncertainty. These observations give credence to the notion that the EU is an entity that ‘dear not speak its name’. There is no agreed-​upon conception of the EU qua polity that spells out its constitutive features in a polity sense. There is no EU-​official declaration that clarifies the EU’s status qua polity. Numerous analysts underline that the treaties make up the EU’s constitution. That is a constitution in a material albeit not in a formal or normative-​democratic sense, however (Fossum and Menéndez 2011). Further, the member states do not acknowledge that the EU has Kompetenz-​Kompetenz, in other words, the power to determine its own competence.The effect is to sustain an ongoing debate on the character of the EU: is it a particularly compelling form of international organization; a state-​in-​the-​making; or a cosmopolitan vanguard? This situation has given rise to the notion of the EU as a distinctly different entity or an entity sui generis. The danger here is falling into the sui generis trap, which is about reifying what is unique and distinctive about the EU. When considered from a political science perspective, the notion of the EU as an entity sui generis posits that the EU is a unique type of political entity: neither an international organization, nor a state.4 From its inception several of the EU’s architects spoke of the EU as an entity sui generis,5 which entailed that it was unique and one of its kind.6 This designation of the EU still figures strongly in the system’s self-​description, and in descriptions of policy instruments and institutional arrangements.7 What does the claim that the EU is a distinctly new entity signify? The most obvious response would be that the EU is characterized by features that set it apart from any known type of entity. If so, there has to be a distinct set of conceptual categories that can properly depict the EU. How distinct must they be? Does sui generis status presuppose a distinct vocabulary of association and of polity? The question is relevant given that a distinct vocabulary associated with the nation-​state has emerged.When the nation-​state was established in Europe: “the whole European vocabulary of association … [was] ransacked for suitable expressions with which to describe and to appraise the formal character of a modern European state” (Oakeshott 1975: 320). Nevertheless, this was not an entirely new vocabulary; it came about mainly through redefining the meaning of existing terms. In a similar manner, something may be framed as novel simply because it is composed in a different manner. The critical point is that there is a difference between novelty through and through and novelty through putting together known and familiar building-​blocks.We can make new types of buildings with well-​known and tried-​out building materials. That matters for the discussion of EU differentiation. The EU is a novel type of governing system but is not novel through and through because it embodies traits of the member states that established and operate it.This state imprint must be considered in connection with the fact that the EU’s member states exhibit institutional resilience. Even if the states that entered the EU have been transformed from 83

John Erik Fossum

nation-​states to member states, they have not ceased to be states, neither has the EU escaped them (Bickerton 2012: 182). Another way of putting this is to say that the EU has contributed to transforming their stateness without transcending it. The EU to a large extent piggybacks on their stateness in the effectuation of EU policies and programs of action. There is no question that the EU’s member states wield a significant influence on the EU, including in terms of shaping and controlling its distinct differentiation configuration. Art 4.2 TEU compels the EU to respect the equality of Member States before the Treaties as well as their national identities, inherent in their fundamental structures, political and constitutional, inclusive of regional and local self-​government. It shall respect their essential State functions, including ensuring the territorial integrity of the State, maintaining law and order and safeguarding national security. With these observations in mind, we see that the absence of an official EU ‘polity script’ effectively means that any attempt to discuss the EU’s constitutive features cannot refer back to an officially sanctioned EU benchmark that at the same time has clear normative standing to warrant the designation of prescriptive. The problem is compounded by two elements: (a) that the vocabulary of normatively legitimate political systems is still steeped in the vocabulary of the modern state, and (b) that empirical developments and efforts at conceptualization often proceed at different paces. Feudalism was only conceptualized as such in the 17th century well after it had ceased to exist (Spruyt 1994: 36).8 For a prescriptive notion that is institutionally thick and thus not confined to abstract principles, analysts will almost inadvertently refer back to the vocabulary associated with the dominant nation-​state conception in search of benchmarks. This vocabulary then forms the basis for establishing how similar or different the EU is. Given this nation-​state normative hegemony and the fact that the development of prescriptive polity schemes is not devoid of power politics, we need to outline the constitutive character of the EU’s differentiation configuration in several steps.The first is to unpack the nation-​state configuration by spelling out four differentiation dimensions that clarify the relationship between differentiation and democracy. Thereafter we outline relevant constitutional models. The fact that the EU is a composite of intergovernmental and supranational (federal) traits compels us to clarify both models, paying special attention to how and in what sense the EU coheres and diverges from these. This approach combines principle and practice. Before proceeding to outline the state-​based differentiation configuration, it is important to note that in the EU context, there is quite a bit of conceptual confusion in that analysts use differentiation and differentiated integration almost interchangeably.The two terms are nevertheless different. Differentiation is broader and encompasses moves in an integrationist and dis-​integrationist direction. Differentiation refers to the structural make-​up of the EU, which is the main concern when we discuss constitutive differentiation.The literature on differentiated integration is less fixated on structure and more process-​oriented, when it discusses such relevant categories as multiple speeds, moves towards core Europe, and questions of variable geometry.9 In addition, there is the issue of scale, with both terms encompassing macro and micro-​level phenomena.The terms are used to focus on different phenomena also at the micro-​level: differentiation focuses on the broader patterns, whereas differentiated integration zooms in on deviations: states gaining opt-​outs and exemptions from EU policies and legal provisions, be they primary or secondary laws, permanent or temporary provisions.10 This chapter understands constitutive differentiation as referring to matters of polity or those macroscopic traits that are basic to the configuration of the political system; it does not really 84

Constitutive differentiation

engage with policy-​or issue-​specific exemptions, exceptions and opt-​outs and opt-​ins.The main focus is therefore on differentiation, not differentiated integration as presently studied.

State sovereignty and differentiation The first part of this section outlines state sovereignty and unpacks the distinct differentiation configuration that is associated with that; the second part spells out how the notion of sovereignty is being reconfigured in the EU context, and the implications for constitutive differentiation. The statist conception of the political system refers to how the political system is configured along the three core constitutive dimensions of the territorial, the functional and the hierarchical (Bartolini 2005; Leuffen et al. 2013; Rokkan 1975).This structuring of the political system is complemented with a fourth dimension, which brings in the citizen and further highlights the democratic dimension. That fourth dimension relies on a distinction between state and society, a division between public and private, and the legal-​institutional arrangements (including rights and obligations) that constitute the person as a democratic citizen. The constitutive principle of the nation-​state configuration is state sovereignty.11 It posits that the state is sovereign in the sense that external actors recognize the state’s right to wield control of a given territory. Sovereignty entails “internal supremacy over all other authorities within a given territory and external independence of outside authorities” (Keohane 2002: 746). Functional differentiation is territorially delimited and is operated by a hierarchically structured system of government. Hierarchy also entails that the political centre asserts its differentiation design on the overall system: it does so at the central level, in its relations with sub-​units, and to a considerable extent even within the sub-​units. State sovereignty has significant implications for democracy, which is based on the notion of popular sovereignty. In a world of states, popular sovereignty must be adapted to state sovereignty. Thus, we find that modern nation-​state-​based democracy posits that the people that the state rules over is able to control the system of rule; in other words, that they are able to understand themselves as the authors of the laws that they are subject to. Hence, in a world of states, democracy is mainly a matter internal to each state.

The nation-​state model of differentiation unpacked Modern multilevel representative democracy –​to function –​is premised on at least four types of differentiation, along decisional, functional, territorial and citizen incorporation lines.These four dimensions are required for understanding how democracy and differentiation relate to each other in modern political systems (Fossum 2019). The four differentiation dimensions encompass democracy’s onus on governing capability and citizen participation/​representation. Decisional or lawmaking differentiation refers to division of power and lawmaking. Democracy depends on some institutional differentiation to guide and steer the lawmaking process. Democracies vary in terms of how well-​entrenched this mode of differentiation is. High levels of de-​differentiation through concentration of all three functions within one governing branch are democratically deleterious. For instance, will executive dominance mark an overly strong concentration in the executive; a concentration in the judiciary is associated with untrammeled juridification; and an overly strong concentration in the legislature is associated with un-​governability. Competence-​based functional differentiation reflects the fact that modern political systems are functionally differentiated. Modern democracy relies on functional differentiation. Having said that, it matters how that is structured, including who possesses the knowledge, how knowledge 85

John Erik Fossum

is brought to bear, what type of knowledge is available and where, how accessible expert knowledge is to democratic decision-​makers, and whether expertise is properly subjected to democratic controls. Knowledge and expertise accumulation can increase governing capacity and capability and can add to legitimacy by making good decisions. At the same time, an overly strong specialization that is not subject to proper democratic controls can engender technocracy; undermine coordination; generate and entrench strong epistemic biases; and as such undermine governing effectiveness and legitimacy. Territorial differentiation refers to the fact that every political system is territorially based and bound to a given space. That in turn has two aspects: the former pertains to internal–​external relations; the latter to vertical relations across levels of governing. With regard to the former, territorial confinement, all modern states are territorially confined and bounded, as is stipulated in the modern doctrine of state sovereignty. The extent to which a state is capable of exercising exclusive control of a given territory varies considerably, especially in today’s globalized world where states and societies are profoundly interwoven. An important consideration for state-​based democracy is how permeable the political system is to its surroundings, and how well the governing arrangements are able to regulate exit and entry to the system (of ideas, persons, groups and territories).The more permeable, the greater the scope for cross-​border law-​making, functional and territorial differentiation. High permeability violates the democratic norm of congruence between what the governing institutions relate to and how, on the one hand, and the citizenry’s ability to instruct and hold the governments to account, on the other. It follows that the greater the amount of interdependence and interweaving of territorially based governing entities, the more complex the democratic challenges. The question is therefore if or how such arrangements can be democratically legitimate. Must contiguity be restored for a system to be democratically viable? Or can arrangements that circumvent this still be democratically viable?12 The other aspect reflects the fact that all sizable modern political systems are based on some form of territorial differentiation. They vary greatly in structural composition and location of power and authority, from a continuum of unitary to federal systems: starting at central control and mere de-​concentration, moving to delegation, which is about loosening control somewhat, and then moving to the situation of non-​centralization that is the hallmark of federal systems. Federal systems are marked by the combination of shared rule and self-​rule, so that each sub-​ unit’s realm of competence and that of the central level are constitutionally guaranteed. It is generally accepted that the federal form of territorial-​functional differentiation is compatible with modern representative democracy. Asymmetrical, multinational and contested federal systems raise questions both for federalism and for democracy.13 The fourth differentiation dimension refers to citizen incorporation, and the manner in which modern political systems incorporate persons through citizenship.The status as citizen in a world of states yields a differentiated pattern: persons’ status, rights and entitlements vary considerably across different types of states. The very definition of a person as a citizen is legally entrenched and designates a specific status that gives persons rights to territorial access, a repertoire of protective and participatory rights and entitlements, and access to government and representative institutions. Citizenship gives legal credence to the right to have rights. Since the world is divided into states, modern citizenship formalizes and legalizes a distinct personal status that in turn is used as the basis for distinguishing between members and non-​members. The development of modern citizenship was at the same time a process of fusion and a process of functional differentiation of rights (Marshall 1950). Fusion took place through the manner in which the state took on the role of rights granter and guarantor under the heading of citizenship. Functional differentiation occurred through civil, political, economic, social and 86

Constitutive differentiation

cultural rights.These different categories of rights were tied up with different subsystems. Hence, citizens experienced different vulnerabilities depending on subsystem (social rights for instance are very exposed to the ups and downs of a globalizing capitalist system). States vary in terms of different citizenship incorporation rules; they vary in terms of whether they allow for singular or dual/​multiple citizenships; they vary in the precise bundle or composition of rights that they offer; they vary in what group-​based rights and protections they offer; and they vary in the relative importance of individual versus group-​based rights. All these dimensions speak to the relationship between differentiation and modern constitutional democracy. In today’s world, globalization and regional political integration complicate this situation. Globalization and large-​scale migration contribute to render national citizenship regimes more complex by introducing new categories of denizenship and the rights and entitlements that different categories of persons have. Thus, in a globalized and interconnected world, we face the prospect of different functional-​territorial configurations of rights as the state-​induced fusion gives way to a more differentiated system.

The EU and the transformation of sovereignty As noted above, the nation-​state differentiation configuration occupies a dominant –​even hegemonic –​role not least because it has long been considered the main (to many the only) frame for situating viable constitutional democracy. A comparison of the nation-​state differentiation configuration as outlined here with what we find in the EU would be fallacious because it would in effect compare a nation-​state prescriptive framework with a descriptive account of what is unfolding in the EU.The comparison would miss the fact that no nation-​state fully coheres with this prescriptive framework, and further that in today’s globalized world nation-​states are closely interwoven and interdependent. Functional differentiation is not nationally contained; states’ territorial control varies considerably; and hierarchical systems of command and control contend with less vertical and more horizontal forms of network-​based governance (Hooghe and Marks 2003). In effect, globalization’s fostering of interstate interweaving and interdependence has given rise to a comprehensive debate on the continued relevance of the sovereignty-​based nation-​state differentiation configuration. Having said that we need to be clear on more precisely how the EU alters state sovereignty in order to understand the implications for constitutive differentiation. Two factors stand out. One is hierarchical boundary control; the other is pooling and sharing of sovereignty. I will explain these in turn. On the first, Stefano Bartolini (2005: xvi) underlines the central role of boundary control in the state formation process: when an internal hierarchical order manages to control the external territorial and functional boundaries so closely that it insulates domestic structuring processes from external influences. In this case, the internal hierarchy presents itself as the single organizing principle of the internal domestic structuring and, at the same time, as the single autonomous centre for external relations. The EU is especially permeable, not the least because the EU-​level lacks much of the capacity and the instruments to assert a differentiation design of its own choice on the member states. The EU-​level does not make up a self-​standing independent centre that is capable of organizing the EU’s internal structure, nor does the EU-​level manage to insulate its domestic structuring processes from external influences, notably that of the EU’s member states. That is because the EU supranational system is solidly anchored in the member states. These states do not agree on 87

John Erik Fossum

what form of system the EU should become, not the least because many member states prefer a weak EU that harmonizes interstate interaction and cooperation rather than supranational integration. The strong member state presence in the institutions at the EU level also structures the integration process: it is a matter of fusing levels (EU and member state) and sharing competencies rather than singling out a distinct European level of government with exclusive competencies (Wessels 1997). States cede sovereignty not to a distant entity but to a common unit that they all participate directly in. In EU parlance this is generally referred to as pooling of sovereignty. This process of pooling has profound implications for the ensuing notion of sovereignty: States that are members of the European Union have broken sharply with the classical tradition of state sovereignty. Sovereignty is pooled, in the sense that, in many areas, states’ legal authority over internal and external affairs is transferred to the Community as a whole, authorising action through procedures not involving state vetoes […]. Under conditions of extensive and intensive interdependence, formal sovereignty becomes less a territorially defined barrier than a bargaining resource. Keohane 2002: 748 The implication is that constitutive differentiation is a matter of contestation between supranational and intergovernmental principles, both of which are embedded in supranational and intergovernmental EU institutions. The main thrust is not between levels of governing; the contestation is internalized in the EU as a tug-​of-​war between EU institutions that to different degrees are filled with member state officials and whose relations are tightly coupled or even fused with the member states and their administrations. The fact that there is this ongoing struggle between different visions of EU constitutive differentiation warrants the need for several constitutional models, the outlining of which is the topic for the next and final substantive section.

EU constitutional models and patterns of differentiation Since the EU-​level institutional arrangement incorporates both intergovernmental and supranational principles and arrangements, and there is no authoritative EU polity script, I start by outlining a bare-​bones intergovernmental model and thereafter show how analysts have sought to modify that to sit with the EU. Then, I do the same for the federal model. The point is to show how each version must be modified to capture the EU’s distinctive differentiation traits. Using simple labels to capture the complex EU is an invitation to double-​talk and easily leads to model overstretch.

The intergovernmental model The intergovernmental model posits that only the nation-​state can foster the type of trust and solidarity that is required to sustain a democratic polity. The model posits that the EU must be devised to ensure that the institutions at the EU level are accountable to the member states, which continue to serve as the main vehicles for ensuring the type of private and public autonomy that democracy requires. When we apply the four democracy-​differentiation dimensions to this model, we find that with regard to the first dimension, decisional or lawmaking differentiation, this dimension is primarily operating at the level of the member-​state, since the member states are the locus of democratic authority. It follows that the model does not envisage a democratic body at the EU-​level 88

Constitutive differentiation

that can claim to represent the collective body of citizens in the EU’s member states. The lines of authorization and accountability are mainly vertical, with the EU the agent and the member states the principal. The fact that each member state has the power of veto is meant to serve as a safeguard for national democracy. Nevertheless, precisely since each state has veto, the model has weak provisions for each member to intervene in the democratic affairs of the other member states, should they find them wanting. The EU’s legitimacy is indirect and hinges on its ability to perform in accordance with the prescriptions from the member states. The standard model understands democratic authorization by member states to take the form of intergovernmental bodies in which the contracting partners strike bargains on behalf of nationally fixed preferences and interests (Moravcsik 1998). With regard to the second, competence-​based functional differentiation dimension, the model provides that the member states authorize EU action and confine and delimit the EU’s range of operations through the provisions set out in the treaties, as well as through a set of institutions that permit each member state to exercise the power of veto. Thus, the model posits that the EU’s scope, its range of operations and its type and range of policy instruments are in principle determined by the member states. With regard to the third, territorial differentiation dimension, the intergovernmental model presumes, as noted above, that the member states delegate competence to the Union.The model does not leave space for the EU-​level institutions to determine what they should do perhaps even how they should do it (although it may be a matter of discussion whether the model is based on a strict delegation approach or whether it leaves space for the EU-​level to operate as a trustee). In any case, the model posits that the EU’s competence can in principle be revoked (Pollack 2003). To what extent this model allows for differentiated integration in terms of individual or groups of states opting-​in or opting-​out of arrangements; integrating at multiple speeds; seeking exemptions, exceptions and deviations is not entirely clear. The intergovernmental model opens up for such provisions to be decided in two different ways: by the states striking agreements in designated intergovernmental bodies or by each individual state determining this on its own accord. With regard to the internal–​external aspect of territorial differentiation, the intergovernmental model leaves no scope for external differentiation in the sense of EU norms and rules applying to states that are not EU-​members. That is because the boundaries of the association of sovereign states correspond with the boundaries of its member states. With regard to the fourth differentiation dimension, that of citizenship and differentiated rights, the model is based on the member-​states as the rights-​g ranting entities. Citizens’ status and rights will therefore vary with state, in accordance with its provisions. It follows that the host state sets the terms for citizens from other states, unless the member states have collectively decided to make common provisions. Such provisions again would be subject to least common denominator in that all states need to agree.

Intergovernmentalism adapted to the EU: the new intergovernmentalism The fact that the EU combines supranational and intergovernmental principles and organizational-​ structural arrangements testifies to the need to add or modify the intergovernmental model. One position that has sparked a lot of interest is the so-​called new intergovernmentalism (Bickerton et al. 2015), which presents an ‘integration paradox’: since Maastricht the EU has seen a significant increase in integration but without supranationalism. The additional EU integration, the argument goes, has taken the shape of more policy coordination through intergovernmental means rather than through uploading to the supranational EU institutions. From a constitutive 89

John Erik Fossum

differentiation perspective, this cannot be construed as an argument in support of the intergovernmental model. It is, on the one hand, a reminder to the effect that the EU balances supranational and intergovernmental traits and that the scale can tip in both directions. On the other hand, it can also be construed to suggest that increased intergovernmentalization can subvert supranationalism and that the traits we associate with intergovernmentalism in the EU have already been modified in a supranational direction.

Intergovernmentalism adapted to the EU: republican intergovernmentalism Theoretically and normatively speaking, the most advanced attempt to build on yet also to go beyond the standard intergovernmental model by taking globalization and Europeanization seriously –​not only functionally but also morally –​is Richard Bellamy’s position on republican intergovernmentalism. He explicitly situates himself in the intergovernmental camp by stating that: I see no need for the EU or similar international organisations to detract from the sovereignty of the member states and their peoples –​rather, they can and should help preserve such sovereignty and will be the more justifiable and effective for doing so. 2019: xiv Bellamy, as we shall see, has pitched the model against that of a federal Europe, understood as a European federal state. He notes, however, that his position is broadly federal in the Kantian sense of a federation of sovereign states (2019: 23).14 Bellamy’s point of departure is that today’s system of national democracy faces two main challenges: globalization and cosmopolitanism, associated respectively with a functional and a moral challenge. The functional challenge pertains to the fact that in today’s globalized world, nation-​states are either unable or are barred by transnational dynamics from properly addressing a whole range of externalities and border-​crossing issues and activities. In this circumstance, individual states’ ability to devise their own independent policies on such matters as security and socio-​economics is significantly constrained. States can be subject to externally generated forms of domination and can prove unable to harness the positive externalities of their activities. The functional challenge can only be addressed by interstate cooperation. The moral challenge pertains to the core cosmopolitan tenet that persons are moral equals. Cosmopolitans see the state system as a constraint on justice. These challenges cannot be adequately addressed without global and international institutions, which however for reasons of scale (constraining pluralism), capacity (limited ability to provide public goods) and complexity (reducing equity and legitimacy) cannot replace state-​based democracy. In effect, international institutions can exacerbate the problem of “democratic disconnect between the peoples of the constituent states and the inter-​and multi-​national decisions their domestic representatives make in their name…” (Bellamy 2019: 12). In a globalized and interconnected world, this disconnect cannot be abrogated by domesticating decision-​making because that renders people vulnerable to external domination. “Rather than subverting democracy at the national level, global institutions are in many respects vital to its continuing effectiveness and acceptability in an interconnected world” (2019: 3). The solution is however not democratizing global institutions. Instead, “they can acquire this legitimacy not by becoming themselves sources of democratic authority but through being under the democratically authorised and accountable control of the states that established them and regulate their interactions through them” (2019: 3). At the same time, international institutions –​notably 90

Constitutive differentiation

the EU –​affect the domestic democratic arrangements, so the challenge is to ensure workable democracy in an interdependent world –​rather than fostering a form of two-​levelled executive dominance (at EU and member state levels). The basic challenge facing the EU, then, is not to find a way to address the ‘no demos’ thesis at the EU-​level, which is what neo-​functionalists and federalists underline, but instead to find: “ways whereby the various national demoi can at one and the same time control their representatives at the EU level while acknowledging a duty to show equal concern and respect towards other demoi and to avoid dominating them” (2019: 99). As noted, a critical issue is to address the democratic disconnect between EU-​level decision-​making and the decision-​making processes that are taking place in the member states. Bellamy’s solution is a republican association of sovereign states, which states accede to on a voluntary basis. Such an association would ensure freedom as non-​domination insofar as two conditions are met: first, domestic democratic institutions must ensure political authority within a state is under their equal influence and control; and second, the state must be part of an association of democratic states in which the rules governing their mutual relations are under the equal influence and control of the elected representatives of those states. 2019: 4 The system is based on the logic of a ‘normative two-​level game’. That implies that the legitimacy of EU level decisions… would depend on them meeting a dual standard. These decisions must not only reflect the consent of each of the demoi to whom they apply but must not undermine the capacity for each of those demoi to give or withdraw that consent. 2019:69 The democratic conception of the EU that this position prescribes is one of demoicracy, which refers to “a form of democratic international governance among different peoples”. The implication is that “the focus for a theory of legitimacy for the EU is not the provision of fair terms of cooperation among free and equal individual citizens but among free and equal peoples” (2019: 18). It follows that this model opens up for a considerable scope of differentiated integration. The intergovernmental model posits that each member state is the ultimate judge of the quality of its democratic arrangements. Bellamy’s model is based on democratic states and does not allow for the rise of a repressive regime. Nevertheless, Bellamy is silent on how democratic regression can be dealt with. To what extent does the model allow for other states to intervene in the affairs of an increasingly repressive state? Does the model allow for the other states to use EU-​level institutions to impose democratic conditionality on a member state? The basic ambiguity surrounding Bellamy’s approach is whether it really qualifies as intergovernmental (in line with the constitutional model as presented above). One question is whether the normative model that Bellamy outlines is wholly consistent with intergovernmentalism. Another is whether his model can be reconciled with the EU’s combination of supranationalism and intergovernmentalism. This latter point is compounded by the fact that Bellamy does not prescribe dismantling the EU’s present institutional structure, but proposes rather modest small-​ scale proposals for transferring some of the decision-​making back to national parliaments.15 91

John Erik Fossum

The federal model A democratic federation is a system of shared rule combined with self-​rule embedded in a constitutional arrangement (as a core element of the federal pact). As noted above, federalism preceded the development of the modern nation-​state. The American Revolution and the ensuing development of American federalism represented a major innovation, as the United States was the first-​ever instance of founding a federal political system on popular sovereignty. The US experience has subsequently become the standard, or dominant, account, of democratic federalization (Beer 1993; Davis 1978; Ostrom 2008). The Forum of Federations lists 25 countries as federations today, and their populations make up 40 per cent of the world’s population.16 Even if the American federation has served as an important standard-​setter, federations vary considerably in the specific institutional arrangements for embedding popular sovereignty and for structuring relations between systems and levels of governing. If we unpack a baseline federal model against the four differentiation-​democracy dimensions outlined above, we find that with regard to the first lawmaking differentiation dimension, federal systems structure lawmaking at two main levels of governing, the federal and the subunit level. The division of powers along executive, legislative and judicial lines is symmetrical, in the sense that the same structure operates at both levels. For instance, in the United States, the checks and balance principle operates on both levels, whereas in parliamentary federations such as Canada and Australia, the parliamentary principle operates at both levels. Federal system is structured in accordance with the notion of compound representation: multiple, overlapping constituencies within one overarching polity. Citizens are directly represented in the central institutions, as well as collectively represented in the central institutions through their respective sub-​units (King 1982). With regard to the second functional differentiation dimension, this is in line with the statist principle of functional-​territorial contiguity structured so that the aggregate federation takes care of all relevant functional realms, but each level’s realm of functions is specified in accordance with the division of powers across levels of governing. Federations vary in the scope of functions undertaken at each level, and therefore also in the nature and range of policy instruments. They also vary in the relative capacity (fiscal and administrative) and expertise that is accumulated at each level of governing. With regard to the third territorial differentiation dimension, this is structured in accordance with the division of powers in the constitution. The model is based on the principle of non-​ centralization in the sense that neither level can remove the rights from the other; these are entrenched in the constitution that also normally includes provisions to prevent transgressions. Federations vary in terms of the extent to which levels are separate and interwoven; in terms of degree of sub-​unit autonomy; including asymmetrical status for subunits (for instance Quebec in Canada); and in the extent to which subunits have direct influence on the functioning of the central institutions (the German Bundesrat is the most explicit such instance). In terms of internal–​external relations, federations’ bounds and reach corresponds with the statist logic. With regard to the fourth rights and citizenship differentiation dimension, the federal level grants citizenship. All citizens of subunits are federal citizens. Citizens also have specific rights and obligations in relation to the subunits they live in, and for instance with regard to voting rights in subunit elections must obtain residence status in order to have the right to vote.

Federalism transposed to the EU In seeking to transpose federalism to the EU, what remains unclear is whether the main challenge is for the EU to become more compatible with federalism, or whether federal theory is insufficiently 92

Constitutive differentiation

tailored to the specific challenges facing the EU (Fossum and Jachtenfuchs 2017). The challenge includes specifying whether or the extent to which multilevel governance is compatible with federalism, and whether demoicracy is compatible with federalism. The EU’s present configuration of differentiation diverges along all four differentiation dimensions in the standard federal model. A key factor here is the EU’s inbuilt institutional asymmetry: a mixture of supranational and intergovernmental features, which gives lawmaking, functional and territorial differentiation a distinct shape. That is amplified by the EU’s distinct fusion of levels, and the EU level’s dependence on the member states for resources and effectuation of decisions. The EU’s multinational character also sets it apart (even if some federations are multinational, such as for instance Canada). Further, the EU is distinct in that it would be a fledgling ‘multilevel federation of federations’, given that a number of the EU’s member states are themselves federations and therefore engender a complicated three-​level dynamic.

Federalism transposed: the EU as a poly-​cephalous federation The EU’s ‘unfinished’ and contested nature brings in its own complications. I have argued elsewhere (Fossum 2017) that the EU is a fledgling poly-​cephalous (multi-​headed) federation. The main reason is the central role that the ‘collective executive’ plays in shaping and operating the EU’s distinct combination of shared EU rule and member state self-​rule. The more or less institutionalized forms of the collective executive correspond to the European Council (hereafter EC), its restricted formation as the European Council of the Eurozone (the so-​called Summit), the Council of Ministers, and, regarding again the Eurozone, the Euro-​g roup. The EU is distinctive in the manner in which this collective executive shapes and balances relations within and across levels of governing in the EU (balances shared rule and self-​rule). One key feature refers to the central role of the EC in Treaty changes and in setting down the main parameters for the EU’s nature and scope of operations (constitutive role).17 In the EC, the heads of the member states as the EU’s constituent units can negotiate and renegotiate the rules of co-​existence with considerable discretion. The second pertains to the EC’s many roles and its institutional flexibility in the interstices between driver of supranational integration and protector of national sovereignty (mediating role).18 These different roles show that it straddles across the spectre of shared rule and self-​rule in the roles it occupies. The third aspect refers to the European Council’s intervention in policy processes (policy-​making role). It engages in policy processes; it breaks log-​jams; and it signals to the other institutions what actions need to be taken (Werts 2008). Even if these are interventions in policy processes, their cumulative effects have ramifications for the relationship between shared rule and self-​rule. For instance, whatever decision bias there is in the European Council (in favour of member states or supranational institutions or market actors) translates into the balance between shared and self-​rule in the EU.

Federalism transposed: the EU as a Federal Union Sergio Fabbrini has long advocated the need for the EU to become a Federal Union. His point of departure is that “(t)he statist model is a formidable cognitive (as well as political) limitation on the future of Europe… Only by dissociating federalism from the state and from its government is it possible to put the future of Europe on realistic bases” (Fabbrini 2019: 87). Fabbrini’s proposed strategy is to combine de-​coupling with reforming the Union. De-​coupling means forming two separate organizations, and in doing so, the entire union would be reformed. The first organization is the exclusive Federal Union, based on a political agreement of constitutional import, built starting from the core member states of the Eurozone. The 93

John Erik Fossum

second organization is the inclusive economic community, open to all the European states that respect the basic principles of the rule of law and the fundamental criteria of the free market, based on an essential and functional interstate treaty. Fabbrini 2019: 134 Fabbrini offers a distinct proposal for how to convert the EU’s present system with its combination of intergovernmental and supranational features into a federation proper. Fabbrini’s position thus acknowledges the need to reform the EU for it to cohere with federal principles and constitutional-​institutional arrangements.

Conclusion In this chapter, I have defined what is meant by constitutive differentiation and have discussed how best to apply this notion to the EU. An important consideration was to pay explicit attention to democracy and therefore to explicate the relationship between differentiation and democracy. We normally associate differentiation with the structuring of the political system in territorial, functional and hierarchical terms but modern systems seek to square those with democracy and citizenship. The chapter has shown that this is far from straightforward. First, it was necessary to spell out the distinct challenges we confronted when trying to clarify the nature of the EU’s differentiation configuration. The pitfalls and problems compelled us to take a step back to outline the constitutive principles of the nation-​state. I did that with reference first to sovereignty, and thereafter by outlining the key differentiation dimensions that we associate with the democratic nation-​state. I thereafter showed how the EU transforms the notion of sovereignty and the implications that has for our thinking about constitutive differentiation. The contested nature of the EU required spelling out two EU constitutional models. The first was steeped in intergovernmentalism. I unpacked that with reference to the differentiation dimensions that I discerned from the nation-​state model and thereafter considered some of the adaptations of this model to the EU. In doing so, it became clear that such accounts must exceed well beyond intergovernmentalism to account for present-​day EU. The second constitutional model I presented was steeped in federalism. I unpacked this model as well and showed that this model also cannot be applied to present-​day EU without reforming the EU. This undertaking was confined to a rough sketch. A further more detailed examination along these lines should help us to hone in on the specific features and outline in further detail points of correspondence and divergence, all of which will be relevant for the ongoing discussion on the future of Europe.

Notes 1 Constitutive brings up two associations, related on the one hand to the process of constituting something and on the other to the principles, arrangements and practices that constitute the system. This chapter focuses exclusively on the latter association. 2 This is normative analytical in some contrast to normative evaluative which establishes a set of criteria by which to assess the normative (for instance with regard to democracy) quality of a process or a system of rule. 3 See for instance Archibugi (2008), Eriksen and Fossum (2012), Fossum et al. (2018), Habermas (2001, 2006), Held (1995). 4 The legal definition of sui generis is more narrow. See, for instance, www.law.cornell.edu/​wex/​sui_​ generis. 5 Many of the EU’s architects actively propounded the EU as an entity sui generis.Walter Hallstein noted that “the European Community is sui generis, a new kind of political animal” (1962: 28). 94

Constitutive differentiation

6 Etymologically speaking, sui generis is from Latin, where sui refers to ‘of its own’, and generis is the genitive form of genus, ‘kind’. http://​dic​tion​ary.refere​nce.com/​wordo​fthe​day/​arch​ive/​2001/​06/​14.html. 7 Consider Third Commission Report on Citizenship of the Union, COM (2001) 506 final at 9: When considering the scope of citizenship of the Union, attempts to draw parallels with national citizenship should be avoided. Because of its origins and the rights and duties associated with it, citizenship of the Union is sui generis and cannot be compared to national citizenship of a Member State. 8 The historical development of the nation-​state involved a comprehensive process of redefining and readapting designations of community and political order (Oakeshott 1975); and what is understood as a state has undergone significant changes over time (Skinner 2010). 9 There is a large body of literature on differentiated integration with more or less sharp distinctions between differentiated integration and differentiation. See, for instance, Adler-​Nissen (2014), Andersen and Sitter (2006), Bellamy and Kröger (2017), Bickerton et al. (2015), De Witte et al. (2017), Dyson and Sepos (2010), Fabbrini (2015), Eriksen (2018, 2019), Eriksen and Fossum (2015), Fossum (2015), Genschel and Jachtenfuchs (2014, 2016), Kelemen et al. (2014), Kölliker (2006), Leruth and Lord (2015), Leuffen et al. (2013), Lord (2017), Piris (2012), Schimmelfennig (2014), Schimmelfennig and Winzen (2014), Stubb (1996), Warleigh-​Lack (2015). 10 See, for instance, Schimmelfennig and Winzen (2019). 11 Robert Jackson (2000) posits state sovereignty as a key constitutive principle of the global society of states. 12 This is one of the most important and interesting democratic debates in the EU context and has given rise to a range of different perspectives, from standard notions of supranational federal governing to transnational and demoicratic proposals. 13 There is a large body of literature on multinational federations, and also on asymmetrical federations. See for instance Gagnon and Tully (2001), Gagnon and Iacovini (2007), Gagnon and Seymour (2012), Kymlicka (1995), Norman (2006), Webber (1994). 14 The problem with that position is that insofar as subunits in a federal structure are sovereign in accordance with modern-​day state sovereignty, such a system is incompatible with federalism. Modern state-​based federalism has been made compatible with state-​based sovereignty but only at the federal level.The position situates federalism in the structural arrangement and not in the consent of the federal consociates. Preston King (1982) has noted that we cannot have federation without federalism. 15 For an assessment of Bellamy’s model of republican intergovernmentalism in relation to the four differentiation dimensions, see Fossum (2021). 16 www.forumfed.org/​countries/​. 17 The EC’s main role in the EU is, according to the TEU (Art 15.1), that: “The European Council shall provide the Union with the necessary impetus for its development and shall define the general political directions and priorities thereof. It shall not exercise legislative functions”. The EC has interpreted this mandate very widely. 18 Wolfgang Wessels (2016) has shown how we should associate the EC with three quite distinct institutional roles: as a guardian of the sovereign nation-​state (the Presidency model); as a reluctant federator (the Council model), and as a fusion engine (the Fusion model).

Bibliography Adler-​Nissen, R. (2014) Opting Out of the European Union. Diplomacy, Sovereignty and European Integration, Cambridge: Cambridge University Press. Andersen, S.S. and Sitter, N. (2006) Differentiated Integration: What Is It and How Much Can the EU Accommodate?, Journal of European Integration, 28(4), pp. 313–​330. doi: 10.1080/​07036330600853919. Archibugi, D. (2008) The Global Commonwealth of Citizens: Toward Cosmopolitan Democracy, Princeton, NJ: Princeton University Press. Bartolini, S. (2005) Restructuring Europe: Centre Formation, System Building, and Political Structuring between the Nation State and the European Union, Oxford: Oxford University Press. Beck, U. (2003) Toward a New Critical Theory with a Cosmopolitan Intent, Constellations, 10(4), pp. 453–​ 468. doi: 10.1046/​j.1351-​0487.2003.00347.x. 95

John Erik Fossum

Beer, S. (1993) To Make A Nation –​The Rediscovery of American Federalism, Cambridge, MA: Harvard University Press.Bellamy, R. and Kröger, S. (2017) A Demoicratic Justification of Differentiated Integration in a Heterogeneous EU, Journal of European Integration, 39(5), 625–​639. doi: 10.1080/​ 07036337.2017.1332058. Bellamy, R. (2019) A Republican Europe of States, Cambridge: Cambridge University Press. Bickerton, C.J. (2012) European Integration –​From Nation-​ States to Member States, Oxford: Oxford University Press. Bickerton, C.J., Hoodson, D. and Puetter, U. (2015) The New Intergovernmentalism: States and Supranational Actors in the Post-​Maastricht Era, Oxford: Oxford University Press. Börzel, T. (2005) Mind the Gap! European Integration between Level and Scope, Journal of European Public Policy, 12(2), pp. 1–​20. doi: 10.1080/​13501760500043860. Davis, S.R. (1978) The Federal Principle –​A Journey Through Time in Quest of a Meaning, Berkeley, CA: University of California Press. De Witte, B., Ott, A and Vos, E. (eds) (2017) Between Flexibility and Disintegration, Cheltenham: Edward Elgar. Dyson, K. and Sepos,A. (eds) (2010) Which Europe?The Politics of Differentiated Integration, Basingstoke: Palgrave Macmillan.Eriksen, E.O. (2018) Political Differentiation and the Problem of Dominance: Segmentation and Hegemony, European Journal of Political Research, 57(4), pp. 989–​1008. doi: 10.1111/​1475-​6765.12263. Eriksen, E.O. (2019) Contesting Political Differentiation: European Division and the Problem of Dominance. Palgrave. Eriksen, E.O. & J. E. Fossum (eds.) (2012) Rethinking Democracy and the European Union. London: Routledge. Eriksen, E.O. and Fossum, J.E. (eds) (2015) The European Union’s Non-​Members: Independence under Hegemony?, London: Routledge. Fabbrini, S. (2013) Intergovernmentalism and Its Limits: Assessing the European Union’s Answer to the Euro Crisis, Comparative Political Studies, 46(9), pp. 1003–​1029. doi: 10.1177/​0010414013489502. Fabbrini, S. (2015) Which European Union? Europe after the Euro Crisis, Cambridge: Cambridge University Press. Fabbrini, S. (2019) Europe’s Future –​Decoupling and Reforming, Cambridge: Cambridge University Press. Føllesdal, A. (2000) Third Country Nationals as European Citizens: The Case Defended, The Sociological Review, 48(1), pp. 104–​122. doi: 10.1111/​j.1467-​954X.2000.tb03508.x. Fossum, J.E. (2015) Democracy and Differentiation in Europe, Journal of European Public Policy, 22(6), pp. 799–​815. doi: 10.1080/​13501763.2015.1020838. Fossum, J.E. (2017) “European federalism: Pitfalls and possibilities”, European Law Journal, 23(5): 361–​379. Fossum, J. E. (2019) “Europe's Triangular Challenge: Differentiation, Dominance and Democracy”, EU3D Research Papers 1/​2019. Available at: Europe’s Triangular Challenge: Differentiation, Dominance and Democracy by John Erik Fossum :: SSRN. Fossum, J.E. (2021) “EU Constitutional Models in 3D: Differentiation, Dominance and Democracy”, EU3D Research Papers 15/​2021. Available at: https://​pap​ers.ssrn.com/​sol3/​pap​ers.cfm?abst​ract​_​id=​ 3900​382. Fossum, J.E. and Jachtenfuchs, M. (2017) Federal Challenges and Challenges to Federalism. Insights from the EU and Federal States, Journal of European Public Policy, 24(4), pp. 467–​485, doi: 10.1080/​ 13501763.2016.1273965. Fossum, J.E. and Menéndez, A.J. (2011) The Constitution’s Gift –​A Constitutional Theory for a Democratic European Union, Boulder, CO: Rowman & Littlefield. Fossum, J.E., Kastoryano, R. and Siim, B. (eds) (2018) Diversity and Contestations over Nationalism in Europe and Canada, London: Palgrave. Gagnon, A. and Tully, J. (eds) (2001) Multinational Democracies, Cambridge: Cambridge University Press. Gagnon, A.G. and Iacovino, R. (2007) Federalism, Citizenship, and Quebec: Debating Multinationalism, Toronto: University of Toronto Press. Gagnon, A.G. and Seymour, M. (eds) (2012) Multinational Federalism –​Problems and Prospects, London: Palgrave. Genschel, P. and Jachtenfuchs, M. (2014) Beyond the Regulatory Polity? The European Integration of Core State Powers, Oxford: Oxford University Press. Genschel, P. and Jachtenfuchs, M. (2016) More Integration, Less Federation: The European Integration of Core State Powers, Journal of European Public Policy, 23(1), pp. 42–​59. doi: 10.1080/​ 13501763.2015.1055782. Genschel, P. and Jachtenfuchs, M. (2018) From Market Integration to Core State Powers: The Eurozone Crisis, the Refugee Crisis and Integration Theory, Journal of Common Market Studies, 56(1), pp. 178–​196. doi: 10.1111/​jcms.12654. 96

Constitutive differentiation

Habermas, J. (2001) The Postnational Constellation: Political Essays, Cambridge: Polity Press. Habermas, J. (2006) The Divided West, Cambridge: Polity Press. Hallstein, W. (1962) United Europe: Challenge and Opportunity, Cambridge, MA: Harvard University Press. Held, D. (1995) Democracy and the Global Order: From the Modern State to Cosmopolitan Governance. Cambridge: Polity Press. Hooghe, L. and Marks, G. (2003) Unravelling the Central State, but How? Types of Multi-​Level Governance, American Political Science Review, 97(2), pp. 233–​243. doi: 10.1017/​S0003055403000649. Jackson, R. (2000) The Global Covenant –​Human Conduct in a World of States, Oxford: Oxford University Press. Kelemen, D., Menon, A. and Slapin, J. (2014) Wider and Deeper? Enlargement and Integration in the European Union, Journal of European Public Policy, 21(5), pp. 647–​663. doi: 10.1080/​13501763. 2014.897745. Keohane, R.O. (2002) Ironies of Sovereignty: The European Union and the United States, Journal of Common Market Studies, 40(4), pp. 743–​765. doi: 10.1111/​1468-​5965.00396. King, Preston. 1982. Federalism and Federation. Baltimore: The Johns Hopkins University Press. Kölliker, A. (2006) Flexibility and European Unification: The Logic of Differentiated Integration, Lanham, MD: Rowman & Littlefield. Kymlicka, W. (1995) Multicultural Citizenship. A Liberal Theory of Minority Rights, Oxford: Clarendon. Leruth, B. and Lord, C. (2015) Differentiated Integration in the European Union: A Concept, a Process, a System or a Theory, Journal of European Public Policy, 22(6), pp. 754–​763. doi: 10.1080/​ 13501763.2015.1021196. Leuffen, D., Rittberger, B. and Schimmelfennig, F. (2013) Differentiated Integration: Explaining Variation in the European Union, Houndmills Basingstoke: Palgrave Macmillan. Lord, C. (2017) Segmentation, Differentiation and the Aims of European Integration, in Grimmel, A. (ed) The Crisis of the European Union: Challenges, Analyses, Solutions, London: Routledge. Chapter 13, pp. 243–​264. Marshall, T.H. (1950) Citizenship and Social Class and Other Essays, Cambridge: Cambridge University Press. Moravcsik, A. (1998) The Choice for Europe, London: UCL Press. Norman,W. (2006) Negotiating Nationalism: Nation-​Building, Federalism and Secession in the Multinational State, Oxford: Oxford University Press. Oakeshott, M. (1975) The Vocabulary of a Modern European State, Political Studies, 23(June), pp. 319–​341. doi: 10.1111/​j.1467-​9248.1975.tb00072.x. Ostrom, V. (2008) The Political Theory of a Compound Republic –​Designing the American Experiment, 3rd ed., Lanham, MD: Lexington Books. Piris, J.C. (2012) The Future of Europe.Towards a Two-​Speed EU?, Cambridge: Cambridge University Press. Pollack, M.A. (2003) The Engines of European Integration: Delegation, Agency and Agenda Setting in the EU, Oxford: Oxford University Press. Riley, P. (1973) The Origins of Federal Theory in International Relations Ideas, Polity, 6(1), pp. 87–​121. doi: 10.2307/​3234183. Rokkan, S. (1975) Dimensions of State Formation and Nation Building: A Possible Paradigm for Research Om Variations within Europe, Princeton, NJ: Princeton University Press. Schimmelfennig, F. (2014) European Integration in the Euro Crisis: The Limits of Post Functionalism, Journal of European Integration, 36(3), pp. 321–​337. doi: 10.1080/​07036337.2014.886399. Schimmelfennig, F. and Winzen, T. (2014) Instrumental and Constitutional Differentiation in the European Union, Journal of Common Market Studies, 52(2), pp. 354–​370. doi: 10.1111/​jcms.12103. Schimmelfennig, F. and Winzen, T. (2019) Grand Theories, Differentiated Integration, Journal of European Public Policy, 26(8), pp. 1172–​1192. doi: 10.1080/​13501763.2019.1576761. Skinner, Q. (2010) The Sovereign State: A Genealogy, in Kalmo, H. and Skinner, Q. (eds) Sovereignty in Fragments: The Past, Present and Future of a Contested Concept, Cambridge: Cambridge University Press, pp. 26–​46. Spruyt, H. (1994) The Sovereign State and Its Competitors: An Analysis of Systems Change, Princeton, NJ: Princeton University Press. Stubb, A.C.G. (1996) A Categorization of Differentiated Integration, Journal of Common Market Studies, 34(3), pp. 283–​295. doi: 10.1111/​j.1468-​5965.1996.tb00573.x. Warleigh-​Lack, A. (2015) Differentiated Integration in the European Union: Towards a Comparative Regionalism Perspective, Journal of European Public Policy, 22(6), pp. 871–​ 887. doi: 10.1080/​ 13501763.2015.1020837. 97

John Erik Fossum

Webber, J. (1994) Reimagining Canada, Montreal and Kingston: McGill-​Queen’s University Press. Werts, J. (2008) The European Council, London: John Harper Publishing. Wessels,W. (1997) An Ever Closer Fusion? A Dynamic Macropolitical View on Integration Processes, Journal of Common Market Studies, 35(2), pp. 267–​299. doi: 10.1111/​1468-​5965.00060. Wessels, W. (2016) The European Council, London: Palgrave. Wimmer, A. and Glick-​Schiller, N. (2002) Methodological Nationalism and Beyond: Nation-​State Building, Migration and the Social Sciences, Global Networks, 2(4), pp. 30–​334. doi: 10.1111/​1471-​0374.00043.

98

7 EU external differentiated integration and compliance Theoretical and legal aspects Maryna Rabinovych and Anne Pintsch

Introduction The contemporary European Union (EU) can be regarded as a differentiated rather than a single integrated system (Leruth, et al. 2022). Alongside the accommodation of member states’ divergent capabilities and preferences within the EU (Schimmelfennig 2017: 2), a crucial function of differentiated integration (DI) is to involve third countries into the EU’s integration structures without them acceding to the Union. While the DI constellations that only include EU member states are known as ‘internal DI’ (e.g. the Euro Area, enhanced cooperation between the EU member states), those that involve third states are called ‘external’ or ‘mixed’ DI (e.g. the Schengen Area, EU’s association agreements (AAs) with third countries). Recently, the research on external/​mixed DI became more active due to the challenge of redefining the EU’s relations with post-​Brexit Britain and amid the Future of Europe debate, more generally (Schimmelfennig 2018; Leruth et al. 2019). A less studied side of such flexibility, implied by differentiation, is compliance and the means by which the EU (or its core) can achieve it amid the proliferation of DI (Winzen and Schimmelfennig 2016: 634, but see Frommelt 2017 and Zhelyazkova 2014). Robust compliance in external differentiation is particularly challenging because, in contrast to the boundaries of the rules, the boundaries of the institutions are not shifted, so that the usual infringement procedure does not apply. In 2019, the consultations on the EU–​Switzerland Institutional Framework Agreement1 convincingly demonstrated how contested questions of how to ensure compliance can be (Schweizerische Eidgenossenschaft 2019). The same applies to the EU–​United Kingdom (UK) negotiations on a post-​Brexit agreement, in which the role of the European Court of Justice (ECJ) in settling future UK–​EU trade disputes had remained debated to the last minute (BBC 2020). The EU’s AAs with Eastern Neighbors, in turn, point to further challenges to rule abidance in the DI context, such as limited state capacity, insufficient incentives and legitimacy due to the missing EU membership perspective or the weak domestic compliance pressure, exerted by courts, electorates and civil society. Non-​compliance, however, poses a threat for both achieving the external DI objectives and the homogeneity of EU law.

DOI: 10.4324/9780429054136-8

99

Maryna Rabinovych and Anne Pintsch

In this vein, this chapter pursues a two-​fold aim. First, it shows how the key expectations formulated by the main theoretical approaches to compliance can be applied to the phenomenon of external DI. These are the external enforcement, management, domestic pressure and legitimacy approaches. Second, the chapter engages with the legal dimension of compliance in the realm of the key existing external DI constellations: the EU sectoral agreements with Switzerland, the AAs with Eastern Neighbors, the European Economic Area (EEA) and the Schengen Area, the draft EU–​Switzerland Institutional Framework Agreement, as well as the recently concluded EU–​UK Trade and Cooperation Agreement (TCA). Importantly, the study engages solely with the legal mechanisms of compliance, illustrating their linkage to the major theoretical approaches to compliance, and does not provide an insight into how compliance has functioned empirically in the above DI contexts. Thus, the work contributes to the emerging literature strand on the design, challenges and consequences of differentiation.

DI and theoretical approaches to compliance States’ compliance with international treaties, i.e. the extent to which they abide by norms after ratification or, put it differently, ‘the extent to which agents act in accordance with and in fulfilment of the conditions’, prescribed by such treaties (Checkel 2000) has been studied extensively from both the political science and the legal perspectives (e.g. Simmons 2010). There is, however, no definite canon of ‘theories of compliance’. While Tallberg (2002) refers to two major theoretical schools of compliance (the enforcement and management schools), Angelova et al. (2012) consider 12 theoretical accounts of compliance in the EU. The lines between various theoretical approaches are not always clear cut and some theoretical arguments may be subsumed under others. Moreover, the research on compliance has tried to integrate rather than separate approaches in order to explain states’ behaviour (see, for instance, Börzel et al. 2010). Hence, in the following overview we will concentrate on four theoretical approaches to compliance, frequently used by EU studies scholars (e.g. Tallberg 2002; Börzel et al. 2010), linking compliance to the external enforcement, management, domestic pressure and legitimacy issues.

External enforcement The rationalist enforcement approach assumes that states ‘are rational actors that weigh the costs and benefits of alternative decisions, when making compliance decisions in cooperative situations’ (Tallberg 2002: 611). Compliance or non-​compliance, respectively, is thus seen as a deliberate decision. States do not fulfil their commitments if the benefit of breaching a contract is higher than the costs of being discovered. Consequently, this approach focuses on monitoring and punishment. In brief, the argument is that an increasing probability of being discovered, and an increasing severity of the punishment decrease the incentive for non-​compliance. It is, hence, argued that ‘deeper’ cooperation requires more severe punishment. The depth of cooperation indicates the degree to which states have to adjust their behaviour in order to fulfil the treaty. Implementing ‘painful’ reforms entails high domestic costs, increasing the incentive to deviate. As a consequence, the more comprehensive the foreseen change of behaviour is, the more cooperation has to be safeguarded by punishments (Downs et al. 1996). This argument resonates with the ‘goodness-​of-​fit’ debate about the role of ‘policy misfit’ as a driver of non-​compliance of EU member states. The policy misfit is the larger, ‘[t]‌he more an EU policy challenges or contradicts the corresponding policy at the national level’ (Börzel 2000: 148). While the ‘goodness-​of-​fit’ argument came to be regarded as rather weak (Treib 2014: 8–​9, 23–​24), a meta-​analysis by Angelova et al. (2012) on compliance in the EU finds strong qualitative and 100

External differentiation and compliance

quantitative empirical support. Monitoring-​related arguments, however, were only found to be confirmed in case studies and thus require further analysis. From the external enforcement perspective, creating appropriate incentive structures would be the best way to prevent non-​compliance in DI. Effective monitoring mechanisms increase the probability that non-​compliance is detected, which actors take into account in their cost-​ benefit calculations. Harsh punitive measures make it unprofitable to break an agreement. These measures can be of a material or immaterial nature and include fines as well as reputational costs (Keohane 1984: 98ff.). Some scholars point out, however, that a state’s political or economic power can significantly mitigate the extent to which it is sensitive to these costs (e.g. Downs and Jones 2002). More powerful states are thus less likely to be concerned about being discovered and sanctioned for non-​compliance. Furthermore, enforcement authorities that ultimately depend on the states that delegate this task to them may be more reluctant to punish powerful principals (Börzel et al. 2012: 458–​459).

Management In contrast to the enforcement school, the management approach assumes that parties to a treaty do not violate the contract deliberately. Rather, a couple of factors can lead to unintended non-​compliance. First, ambiguous formulations in international treaties may lead to differences regarding their interpretation and, consequently, may entail different opinions about treaty violation. Second, aims that are overambitious in terms of timing or substance make it difficult for states to reach them. Third, non-​compliance may result from a lack of capacity in states that are parties to a treaty. When such treaties aim at behavioural changes of private persons or commercial actors, they presuppose a high degree of state regulatory capacity, including comprehensive financial means, scientific or technical expertise, as well as bureaucratic resources (Chayes and Chayes 1998: 9ff.). Researchers analysing compliance in the EU have added a second, autonomy-​related interpretation of ‘capacity’ to the field. They argue that the domestic institutional structure shapes a state’s capability to act. More specifically, the higher the number of veto players and thus the lower a government’s autonomy, the more likely is the violation of EU law (Börzel et al. 2010: 1369–​ 1370). While a few studies rejected this hypothesis, it was confirmed by many others (Treib 2014: 25–​26) and emerged as one of only two robust findings in the meta-​analysis presented by Angelova et al. (2012). Administrative capacity was also found to be conducive to compliance with EU law in many studies (Treib 2014: 26). Angelova et al. (2012: 1278), however, call for more empirical tests to substantiate the previous findings. Importantly, administrative capacity was found to increase through learning processes (Treib 2014: 26). With regard to the impact of complexity of EU legislation on compliance, studies do not present consistent findings, so that the effect remains unclear (Angelova et al. 2012: 1278; Treib 2014: 26–​27). In a recent study, Börzel and Buzogány (2019) argue that the decline in non-​compliance with EU environmental law can be better explained by the management approach than external enforcement. On the one hand, the increasing ratio of amending compared to new EU environmental legislation diminishes the danger of overambitious aims (see also Treib 2014: 24). On the other hand, the European Commission has invested in capacity-​building in the member states. With regard to compliance in external DI, the proponents of the management approach would suggest that treaties and the EU rules extended to third countries should be formulated in a precise manner (Jacobson and Brown Weiss 2000: 524 f.). Incremental and flexible treaty development, in turn, can help to prevent constraints regarding treaty implementation (Chayes and Chayes 1998: 225 ff.). Moreover, mechanisms of treaty interpretation and dispute settlement 101

Maryna Rabinovych and Anne Pintsch

should be in place (ibid.: 201 ff.). In order to counter the third problem of lacking capacity, the management approach suggests technical assistance and capacity-​building measures.This includes support for (administrative) institutions, the development of planning capacity, trainings, education and research (ibid.: 197 ff.). Review and assessment procedures form a more general basis for the success of the above-​mentioned measures aimed at improving compliance. Compared to the monitoring procedures suggested by the enforcement approach, however, they serve a different purpose. Based on a dialogue between the international organization and the parties to a treaty, their main goal is to find out where and how compliance can be improved (ibid.: 229 ff.).

Domestic pressure A third approach explaining compliance with international treaties focuses on domestic actors that pressure the government to fulfil the legal obligations. Tanja Börzel mentions political parties, non-​governmental organizations (NGOs) and economic interest groups as decisive actors for this pull mechanism (Börzel 2000: 148). This resonates with research on the impact of social movements on public policy. Even though results are mixed, there is evidence that social movements can exert a strong influence (Amenta et al. 2010). Research on transnational networks broadens this perspective on the role of domestic actors by including their alliances with foreign non-​governmental actors in the analysis (Risse et al. 1999). At the same time, Börzel points to the role of the media (2000: 148). Tallberg directs our attention to individuals and companies as ‘fire alarms’ in the EU’s decentralized compliance system (2002: 620). According to Xinyuan Dai, such domestic pressure is supported by international organizations, which supply domestic actors with important information about the government’s behaviour (2005: 363 ff.). Similarly, Börzel (2000) considers a ‘sandwich strategy’ combining pressure on the government from domestic actors with pressure from the European Commission. Beth Simmons underlines the role that competent and independent national courts play for a state’s compliance (2010: 273). The latter mechanism is a central component of the EU’s compliance system. It originates in the principles of direct effect and EC law supremacy created by the ECJ in the early 1960s and has developed over a few decades (Tallberg 2002: 620–​621).2 Importantly, the mentioned actors and mechanisms do not stand separately but do interact. Individuals or companies may sue national governments in court, potentially acting upon information international organizations or the media provide to them. Empirical evidence with regard to the influence of interest groups on compliance with EU law is heterogeneous. On the one hand, these groups can pressure governments into compliance when they benefit from such reforms. With regard to the EU’s eastern enlargement, Olga Avdeyeva found that despite international pressure, candidate countries showed different levels of compliance with EU rules on gender equality in the workplace. These differences could be traced back to the strength of women’s movements in the respective countries. The stronger these movements were, the more likely were both policy adoption and related institutional change (Avdeyeva 2010). On the other hand, interest groups negatively affected by the implementation of EU rules may exert counter pressures and thus hinder compliance (Treib 2014: 25). In general, it can be said that the domestic mechanism functions the better the more democratic the political system of the respective state is (Hathaway 2007). This is consistent with Courtney Hillebrecht’s (2014) finding that a constrained executive represents a central factor for a state’s abidance to agreed rules. According to this approach, compliance in external DI can be achieved through the following means: the EU, and in particular the Commission, should forward information on the government’s activities related to the fulfilment of the treaty to domestic non-​governmental 102

External differentiation and compliance

actors. The latter could also be directly involved into the monitoring or compliance process, e.g. by taking part in assessment meetings. If necessary, the EU should support the democratic development of states which it partially integrates, including civil society and free media. If DI entails large domestic reforms, societal expertise in reform sectors should be built up. DI and thus reform efforts should concentrate on those sectors that are characterized by the existence of domestic interest groups directly benefiting from the reforms. These groups and societal (transnational) coalitions should receive special attention from the EU.

Legitimacy The fourth approach builds on a constructivist perspective and claims that states obey to agreed rules because they are socialized to regard this rule-​following behaviour as appropriate. It assumes the legitimacy of rules to be a prerequisite for rule abidance. The community of states expects that members fulfil legitimate obligations (Franck 1997: 33). Legitimacy is defined as a property of a rule or rule-​making institution which itself exerts a pull towards compliance on those addressed normatively because those addressed believe that the rule or institution has come into being and operates in accordance with generally accepted principles of right process (Franck 1990: 24).The legitimacy of rules is founded on several factors: first, the rules have to be formulated in an unambiguous way. Clear rules define the expected behaviour and thus make it easier to act appropriately by following them. Second, the legitimacy of rules is strengthened through symbolic validation, which underlines the authenticity and authority of rules. Forms of symbolic validation are rituals and the ‘pedigree’ of a rule or a rule-​making institution. Third, rules must be rooted general principles and, as a consequence, be applied in a coherent way. Finally, legitimate rules should be based on preceding rules that among others determine the process of their formation. The legitimacy of a rule increases when it was created through an accepted procedure (Franck 1990). With regard to the legitimacy of the rule-​making institution, it has been hypothesized that compliance with EU law is better in states that show higher levels of public support for the EU (Börzel et al. 2010: 1371). Variation in rule violation can also be expected between countries with different domestic cultures of law-​abidingness. Thus, it has been hypothesized that compliance with EU law is more likely in countries with higher support for the rule of law (Berglund et al. 2006: 701; Börzel et al. 2010: 1370–​1371). Empirically, this would correspond to the ‘world of law observance’ identified by Falkner et al. (2007). While there is slight empirical support for the rule of law hypothesis (ibid.), more studies are needed to confirm it (Angelova et al. 2012: 1278). With regard to public support, the literature presents contradicting results (Börzel et al. 2010: 1380–​1381; Angelova et al. 2012: 1277–​1278). It follows from this approach that in order to achieve compliance in DI, rule-​making should follow an approved procedure and result in clear provisions. Rule application must be consistent or justified with underlying fundamental rules. Furthermore, the symbolic dimension of rules should be kept in mind. More generally, the rule of law should be strengthened where necessary. Finally, the EU would also have to care about its external image in order to ensure public support.

DI and legal instruments to ensure compliance This part of the analysis will focus on the key legal instruments the EU uses to ensure compliance in the external differentiation context. Apart from the legal instruments immediately provided for by the above-​mentioned bi-​and multilateral legal frameworks of external DI we will refer to some relevant EU unilateral instruments, such as the ones concerning EU’s financial 103

Maryna Rabinovych and Anne Pintsch

and technical assistance to third states. The spectrum of legal instruments, highlighted below, shall not be in any case regarded as exhaustive. It rather encompasses the instruments most commonly applied by the EU and, at the same time, most illustrative of the theoretical expectations, highlighted above. One legal instrument can be associated with several theoretical expectations, e.g. the provisions on civil society in Trade and Sustainable Development (TSD) chapters that speak to the external enforcement, domestic pressure and legitimacy theories. Vice versa, one theoretical expectation can be illustrated by several legal instruments, e.g. conditionality and monitoring are commonly used in the case of the external enforcement logic. The attribution of legal instruments to specific theoretical expectations shall not be regarded as an absolute. For example, the monitoring clauses would typically serve the objectives of external enforcement (i.e. revealing instances of non-​compliance) for the purposes of launching consultations or taking a recourse to the Dispute Settlement Mechanism (DSM). However, their application can also help the Parties to distinguish lacunas and legal uncertainties in the substantive rules to be dealt with via the management approach. Hence, the key tightly intertwined legal instruments to be addressed below include conditionality; monitoring and evaluation; civil society engagement; consultations; the DSM, incl. rules’ interpretation by the Court of Justice of the EU (CJEU), remedies or compensatory measures and sanctions for non-​compliance, and financial and technical support to the implementation of AAs.

Conditionality Conditionality is the principal tool the EU uses to promote third countries’ compliance with the EU’s values and rules. Substantively, political (or ‘democratic’) conditionality and acquis conditionality can be distinguished (Schimmelfennig and Sedelmeier 2020: 816). Moreover, it is worth distinguishing between positive conditionality (i.e. providing a reward in exchange of compliance) and negative conditionality (withdrawal of a reward or a punishment in case of non-​compliance). The typical ‘common values’ conditionality structure (based on negative conditionality), included in the EU AAs3 with third countries, comprises: • •

A clause specifying the core values the relationships between the parties are founded on (an ‘essential element’) and A ‘suspension clause’ that defines the procedure for suspending the Agreement in case of the violation of the ‘essential elements’ (Dolle 2015).

Such a structure was initially aimed at reacting to gross human rights violations in partner countries, so that they could be qualified as a ‘material breach’ of a treaty in terms of Art. 60 of the 1969 Vienna Convention on the law of treaties and, thus, justify its suspension and the application of sanctions (Hachez 2015). Later on, the scope of the ‘essential element’ clauses was broadened to include democracy, the rule of law and other context-​specific values. Although association relations with a third state typically presuppose close political links, the scope of the common values conditionality significantly varies across the AAs. Such a variation is typically attributable to the peculiarities of partner countries’ political regimes, coupled with the insufficiency of the EU’s bargaining power, as well as the ‘depth’ of an AA (Rabinovych 2019a: 296–​ 300). Therefore, the most comprehensive ‘essential element’ clauses are contained in the EU AAs with emerging democracies, such as the Stabilization and Association Agreements (SAAs) with the Western Balkans and the AAs with its Eastern Neighbors Ukraine, Moldova and Georgia (see, 104

External differentiation and compliance

e.g. EU-​Kosovo 2016, Art. 3; EU-​Ukraine 2014, Art. 2). From the perspective of the external enforcement approach, it may seem that the ‘essential element’ clauses represent a promising pathway to ensure compliance with the political conditions under the external DI constellations. In practice, however, the ‘essential element’ clauses are invoked rather seldom and in a selective manner (Hachez 2015; Moberg 2015), which is problematic from both the enforcement and the legitimacy perspective on compliance. Moreover, the application of the ‘essential element’ clauses is hampered by the lack of definitions of democracy and the rule of law and, subsequently, of what shall be regarded as a breach or a grave breach of these principles (Kochenov 2008). Again, this is not only problematic from the enforcement perspective but also with regard to legitimacy considerations. Hence, the common values conditionality in the EU external DI can be seen as a gateway to the application of further tools to ensure compliance (e.g. consultations), rather than a fully-​fledged compliance tool. The acquis conditionality (i.e. the conditionality related to the specific EU legal acts) first emerged in the EU’s external DI constellations with potential future members, such as during the Eastern enlargement process and in the SAAs with Western Balkans. Later on, it has become a crucial part of the EU’s AAs/​Deep and Comprehensive Free Trade Areas (DCFTAs) with associated Eastern Neighbours. As compared to the SAAs’ norms, the AAs’/​DCFTA’s legislative approximation provisions are marked by their more detailed nature and, as discussed in the next subsection of the analysis, a sophisticated monitoring mechanism (see, for instance, EU-​Kosovo 2016, Art. 74; EU-​Ukraine 2014, Art. 152–​154, Annexes). Given the uniqueness of the extent to which the AAs/​DCFTAs suggest integrating non-​member states into the EU Single Market (e.g. with regard to trade in services and public procurement), market access for associated Neighbors is made conditional on their specific progress with respect to legislative approximation (see, for instance, EU-​Ukraine 2014, Art. 152–​154, Annexes of Title IV). In relation to the EU–​Ukraine AA, Van der Loo thus uses the term market access conditionality (2016: 308). Compared to democratic conditionality, the acquis conditionality is considered to deliver better compliance results (e.g. Schimmelfennig and Sedelmeier 2020: 829). One of the key reasons for this includes lower compliance costs of the acquis-​related conditions. In a nutshell, conditionality represents an important feature of the EU AAs with third states as external DI constellations and a legal instrument to achieve compliance, rooted in the rationalist bargaining approach and consonant with the external enforcement compliance theory.

Monitoring and/​or evaluation In contrast to conditionality, monitoring and/​or evaluation mechanisms are contained not only in the EU agreements with candidate countries and associated Eastern Neighbors but also the EEA Agreement, the EU sectoral agreements with Switzerland and the Schengen Agreement. In theoretical terms, the monitoring and/​or evaluation clauses may speak to both the external enforcement and management schools of compliance. From the external enforcement perspective, conducting monitoring and/​or evaluation activities is necessary to unveil the instances of a party’s non-​compliance and decide on further steps to deal with them up to the recourse to the DSM and the application of sanctions. At the same time, monitoring and/​or evaluation activities help to distinguish uncertainties in the formulations of the provisions in question, as well as capacity challenges that represent the key sources of non-​compliance from the management perspective. In accordance with the EEA Agreement, Iceland, Norway and Liechtenstein as European Free Trade Association (EFTA) member states committed themselves to incorporating Internal Market regulations into their legal orders4 and implementing the Internal Market directives. 105

Maryna Rabinovych and Anne Pintsch

The monitoring throughout the EEA is exercised in close cooperation by the European Commission and the EFTA Surveillance Authority (ESA). According to Art. 5 of the Surveillance and Court Agreement (European Free Trade Association 2020), ESA performs three key functions: (i) ensuring the fulfilment by the EFTA States of their obligations under the EEA Agreement and the ESA/​Court Agreement, (ii) ensuring the application of the EEA rules on competition and (iii) monitoring the application of the EEA Agreement by the other contracting parties (i.e. EU member states) to this Agreement. Important EU–​EFTA joint instruments, in this vein, are the Internal Market Scoreboards that monitor how well both the EU and the EFTA States comply with their provisions under the EEA Agreement (see e.g. European Free Trade Association Surveillance Authority 2020). In the case of an EFTA State’s failure to fulfil an obligation under the EEA Agreement, ESA shall deliver ‘a reasoned opinion on the matter after giving the State concerned the opportunity to submit its observations’ (European Free Trade Association 2020, Art. 31). If the State fails to comply with the ESA opinion within the provided period of time, ESA may submit the matter for the consideration by the EFTA Court (ibid.). As shown later, in particular cases, disputes concerning the EEA Agreement can be also brought before the ECJ. In contrast to the EEA Agreement, the EU–​Switzerland economic relations are being governed by separate sectoral agreements, such as the 1972 EU–​Switzerland Free Trade Agreement or the 2002 EU–​Switzerland Agreement on Air Transport. As it can be exemplified by the latter, the EU–​Switzerland sectoral agreements do not contain monitoring and/​ or evaluation clauses as such (EU-​Switzerland 2002). They rather refer to joint bodies (e.g. the Community/​Switzerland Air Transport Committee) as responsible for administering the Agreement in question and capable of adopting decisions that are binding on the contracting parties (ibid., Chapter 5). A similar approach to monitoring is embedded into the novel EU–​ UK TCA, which designates numerous Specialized Committees to assist the Trade Partnership Council to exercise the monitoring of the Parties’ compliance with norms in specific domains, e.g. Intellectual Property Rights and Digital Trade (EU-​UK 2020, Art. INST.1–​2). More detailed and sophisticated monitoring provisions are included in the draft EU–​Switzerland Institutional Framework Agreement agreed at political level in 2018. The draft Agreement envisages joint monitoring of the application of the whole spectrum of the EU–​Switzerland agreements to be conducted by the European Commission and the responsible Swiss authorities in joint sectoral bodies (EU-​Switzerland 2018, Art. 6–​7). Despite the planned harmonization of the EU–​Switzerland agreements’ network through the draft Agreement, the existing joint bodies shall continue playing a pivotal role in the consultations and dispute settlement procedures. As for the Schengen Agreement and the Schengen acquis, the multi-​aspect monitoring and evaluation mechanism provided for by the Council Regulation 1053/​2013 has been in force since 2013 (Council of the European Union 2013). The Regulation constitutes part of the Schengen acquis and, hence, shall be incorporated in the national legal orders of the non-​EU Schengen countries Switzerland, Liechtenstein, Iceland and Norway. The Commission and the Schengen member states are jointly responsible for implementing the evaluation and monitoring mechanism. According to Art. 4 of the above Regulation, evaluations (that shall cover all aspects of the Schengen acquis) may consist of questionnaires and of both announced and unannounced site visits.The evaluation reports can assess Schengen member states as (i) compliant, (ii) compliant with improvements necessary or (iii) non-​compliant with respect to specific acquis provisions. In the two latter cases, they shall provide for remedial action (ibid., Art. 14). Based on the recommendations, the follow-​up and monitoring stage includes the preparation

106

External differentiation and compliance

of an action plan to address non-​compliance by the Schengen member state concerned, the Commission’s assessment of the action plan’s adequacy and possible on-​site revisits (ibid., Art. 16). The Commission reports to the Council and the European Parliament on the results of the evaluation and monitoring activities on a yearly basis (ibid., Art. 20). Last but not least, in-​detail provisions on monitoring are provided for in the EU AAs/​ DCFTAs with associated Eastern Neighbors. As it can be illustrated by the EU–​Ukraine AA, the Agreement contains both general and sector-​specific monitoring clauses. In general terms, monitoring shall include the ‘assessments of approximation of Ukrainian law to EU law as defined in this Agreement, including aspects of implementation or enforcement’ to be conducted individually by the EU Party or jointly (EU-​Ukraine 2014, Art. 475(2)). Apart from progress reports to be submitted by Ukraine, the monitoring activities can include on-​the-​spot missions, involving EU institutions, bodies and agencies, non-​governmental bodies, supervisory authorities, independent experts etc. (cf. ibid., Art. 475(3)). In the areas covered by market access conditionality, the results of monitoring activities constitute the foundation for a market opening, with the decision to be taken by the Association Council (ibid., Art. 475(6)). Sector-​specific monitoring clauses tend to be less detailed and stipulate the Parties’ obligations as to information exchange with respect to specific sectors, e.g. TSD (ibid., Art. 300). In sum, monitoring clauses can be regarded as an essential legal instrument, aimed at tackling non-​compliance and opening up the pathway for consultations, referring the matter to the DSM or the application of sanctions. Given its ‘intermediary’ nature, monitoring speaks to both the external enforcement and management schools of compliance.

Civil society engagement The mechanisms of civil society participation in the EU external DI constellations are rather novel, and, so far, primarily applicable to trade rules. Domestic Advisory Groups (DAGs) and joint civil society meetings are regarded by the Commission as participatory democracy instruments that enable the civil society actors to contribute to the respective implementation and monitoring activities (European Commission 2020). In theoretical terms, civil society engagement is primarily attributable to the domestic pressure theory, since civil society is generally considered as a key actor to exercise such a pressure. Civil society participation in the monitoring and implementation activities may also contribute to rules’ legitimacy. Moreover, if one considers civil society engagement as a monitoring tool as such, the above intermediary argument applies, and one can regard civil society engagement as speaking to both the external enforcement (i.e. potentially leading to the dispute settlement procedure and/​or the application of sanctions) and the management approach (i.e. opening up the pathway for consultations and/​ or capacity-​building/​assistance initiatives). The ‘Trade and Sustainable development’ chapter under the EU–​Ukraine AA is exemplary of civil society engagement. According to Art. 299 of the Agreement, ‘each Party shall designate and convene a new or existing Advisory Group on sustainable development with the task on advising on the implementation of this Chapter’. DAG members ‘conduct a dialogue encompassing sustainable development aspects of trade relations between the Parties’ at an open Civil Society Forum (CSF), conducted on a yearly basis (EU-​Ukraine 2014, Art. 299(3)). The Parties bear an obligation to inform the CSF on progress in the implementation of the ‘Trade and Sustainable development’ chapter (ibid., Art. 299(5)). The CSF submits its views, opinions and suggestions to the Parties directly or via the DAGs (ibid.). On a more general level, the Parties’ civil societies can exercise monitoring activities via the Civil Society Platform (ibid., Art. 469). Notably, the

107

Maryna Rabinovych and Anne Pintsch

provisions on the civil society engagement are significantly ‘weaker’, compared to the ones on monitoring of the approximation of Ukrainian law to the acquis communautaire. Such a situation responds to the scholarly concern about the lack of civil society’s leverage over monitoring and evaluating the Parties’ compliance with their commitments under EU external agreements (e.g. Ashraf and Seters 2020; Potjomkina 2018). Offering the Parties’ civil societies a chance to utilize complaint procedures and provide input in terms of the procedures under the DSM are seen as the key ways to boost their leverage.

Consultations Referring a matter of non-​compliance to the DSM or applying sanctions to the other Party is clearly detrimental for the EU’s relations with third countries participating in external DI constellations and the operation of such constellations as such. Therefore, all the considered Agreements suggest consultations in joint bodies as the first measure for the Parties to address compliance issues. Dependent on their scope and spirit, consultations can focus on incentives for compliance and/​or potential sanctions (an external enforcement perspective) or the ways to ensure legal certainty or improve capacity (the management perspective). As provided for by the EEA Agreement, the EEA Joint Committee shall represent a venue for consultations ‘on any point of relevance to the Agreement giving rise to a difficulty’, i.e. including compliance issues (EEA Agreement 1994, Art. 92(2)). The consultations are, however, obligatory in a limited number of situations, such as a Party’s intent to apply the unilateral safeguard measures (ibid., Art. 113(2)) or protective measures in the field of capital movements (ibid., Art. 45). Limitedness of the cases, when consultations in the joint bodies are obligatory, amid an unlimited scope of issues to which the consultations apply, is also characteristic for the EU–​ Switzerland sectoral agreements (e.g. EU-​Switzerland 2002, Art. 21, Art. 24–​26). The key difference, in this vein, is the lack of the umbrella joint body in the institutional architecture of the EU–​Switzerland relations due to the long legacy of separately functioning sectoral agreements between the EU and Switzerland. Given the ambitious scope of the EU AAs/​DCFTAs with associated Eastern Neighbors and the scale of required legislative approximation, the AAs/​DCFTAs specify numerous cases, when consultations are required. For instance, virtually all disciplines under the EU–​Ukraine DCFTA provide for consultation procedures to be conducted in the relevant sub-​committees of the Trade Committee (e.g. EU-​Ukraine 2014, Art. 50 bis, Art. 65, Art. 73). Some of the norms provide for regular consultations as to progress concerning the approximation of Ukraine’s legislation to the acquis communautaire, while other provisions are only invoked in specific cases (e.g. a Party’s intent to apply safeguard measures or trade remedies). In contrast to the EEA Agreement and the EU–​Swiss Air Transport Agreement, Art. 305(1) of the EU–​Ukraine AA obliges the Parties to ‘endeavour to resolve any dispute regarding the interpretation and application of the provisions of this Agreement…by entering into consultations in good faith with the aim of reaching a mutually agreed solution’ (i.e. in all the trade and trade-​related matters). Consultations are also obligatory before taking recourse to the DSM under the AA’s ‘Trade and Sustainable Development’ chapter (ibid., Art. 300(4)(6)). Similarly, under the EU–​UK TCA (EU-UK 2020, Art. INST.13.1), if a Party (‘the complaining Party’) considers that the other Party (‘the respondent Party’) has breached an obligation under this Agreement….the Parties shall endeavour to resolve the matter by entering into consultations in good faith, with the aim of reaching a mutually agreed solution. 108

External differentiation and compliance

The Parties can, however agree not to enter into consultations and proceed with arbitration (ibid., Art. INST.14.1c). As opposed to all the above external DI constellations, neither the Schengen Agreement nor the Convention implementing the Schengen Agreement provides for the establishment of joint bodies and consultations. The Convention, however, mentions several occasions, when the Parties are obliged to engage in consultations between themselves, e.g. the alleged incompatibility of an alert under the Schengen Information System (SIS) with national law (Schengen Convention 1985, Art. 41(9), Art. 94(4)). More detailed procedures as to consultations regarding non-​compliance are embedded into the Schengen evaluation and monitoring mechanism. In particular, its follow-​up and monitoring stage is marked by an intense exchange of information, recommendations and reports between the Schengen member state that was found non-​compliant during the evaluation stage, other member states, the Commission, the Council and the European Parliament (Council of the European Union 2013, Art. 16). In sum, consultations help the Parties to solve compliance-​ related matters via political channels and avoid detrimental effects on the functioning of the external DI constellations.

Dispute Settlement Mechanism (DSM)5 For the purposes of external DI research, the DSM can be understood as a structured process that helps concerned states to solve disputes, stemming from the relevant legal instruments that constitute the framework for external DI constellations. A DSM constitutes the foundation for an enforcement system under the DI constellations, thus, speaking to the external enforcement theory. The agreements under study offer different pathways to dispute settlement, yet share some of the peculiarities. Under the EEA Agreement and the EU–​Switzerland sectoral agreements (e.g. the Air Transport Agreement), the key and ‘last resort’ authority to settle disputes concerning the interpretation or application of these Agreements belongs to the single EEA Joint Committee or sectoral joint committees, respectively (EEA Agreement 1994, Art. 111; EU-​Switzerland 2002, Art. 29). In the case of the EEA Agreement, a judicial function is also being performed by the EFTA Court that is authorized to interpret the EEA Agreement with regard to EFTA States (EEA Agreement 1994, Art. 108; European Free Trade Association 2020, Part IV). In contrast, the draft EU–​Switzerland Institutional Framework Agreement grants the Parties the right to refer the dispute to the arbitration panel, composed of the EU’s and Switzerland’s representatives (in equal numbers), if the relevant joint committee could not find a solution within three months (European Union-​Switzerland 2018, Art. 10). Under the EU–​UK TCA, disputes are also being settled through arbitration (European Union-​United Kingdom 2020, Art. INST.14-​INST.15). Finally, the EU–​Ukraine AA provides for three dispute settlement procedures. In matters unrelated to trade, disputes are being resolved by the Association Council (EU-​Ukraine 2014, Art. 477). In trade and trade-​related matters, the matter is being referred to the arbitration panel, if no solution was achieved in terms of consultations (ibid., Part IV, Chapter 14). Under the DCFTA’s ‘Trade and Sustainable Development’ chapter, the Parties can request the dispute to be considered by the TSD Sub-​ Committee (ibid., Art. 300(6)). A crucial similarity, shared by the vast majority of the DI constellations in question, is that the joint committee (sub-​committee), an arbitration panel or the EFTA Court can request a ruling from the ECJ if a question concerns the interpretation of a norm of the EU law (e.g. EU-​Ukraine 2014, Art. 322(2)). An exception, in this vein, is the EU–​UK TCA, whereby the design of the arbitration procedure excludes both the ECJ and the UK courts (EU-​UK 2020, 109

Maryna Rabinovych and Anne Pintsch

Art. INST.14-​INST. 20). Under the two-​pillar DSM structure laid down by the EEA Agreement, the EFTA Court bears an obligation to interpret the EEA’s corresponding rules in conformity with the CJEU rulings delivered before the EEA Agreement was signed (EEA Agreement 1994, Art. 6). Furthermore, Art. 3 of the Surveillance and Court Agreement obliges the EFTA Court ‘to pay due account to the principles laid down by the relevant rulings of the ECJ’ in relation to the rulings delivered after the EEA Agreement was signed. Hereby, it is to note that the EFTA Court shall not be regarded as bound to make use of the ECJ rulings delivered after the EEA Agreement came into force, and there is no provision in the EEA Agreement that ‘suggests that the rulings of the ECJ regarding the internal market should prevail over the rulings of the EFTA Court’ (Valsson 2017). Hence, as one of the pillars under the EEA two-​pillar structure, the EFTA Court can decide not to make use of the ECJ precedents and come up with new ones in terms of EEA/​EFTA pillar (ibid.). In most of the cases, the agreements stipulate the binding nature of the decisions taken as a result of the dispute settlement (e.g. EU-​UK 2020, Art. INST.​21). Furthermore, as, for instance, in the EU–​Ukraine AA, the agreements can underline the Parties’ obligation ‘to comply in good faith with the arbitration panel ruling’ (EU-​Ukraine 2014, Art. 311). The cases of non-​compliance are being addressed through unilateral remedies or compensatory measures (to be examined below).

Unilateral remedies, safeguard and compensatory measures and suspension clauses A Party to an external DI constellation may unilaterally apply remedies, safeguard or compensatory measures to ensure the other Party’s compliance with its obligations under the treaty or a decision made as a result of the dispute settlement procedure. The application of such measures speaks to the external enforcement approach, with them being part of the enforcement systems under the agreements in question. In this vein, the EEA Agreement (1994, Art. 111(3)), for instance, utilizes the term ‘safeguard measure’. In international trade law, a ‘safeguard measure’ is defined as an ‘emergency action’ with respect to increased (potentially) damaging imports (World Trade Organization 2020). The EEA Agreement (1994, Art. 111(3)) does not, however, specify whether this meaning or a different one shall be attributed to safeguard measures that it enables a Party to apply in response to the other Party’s failure to comply with the decision of the EEA Committee. The right to utilize appropriate safeguard measures in case of the other Party’s non-​compliance with the decision of the relevant sectoral committee is provided for in Art. 31 of the EU–​Switzerland Air Transport Agreement and Art. 10(6) of the draft EU–​ Switzerland Institutional Framework Agreement. In contrast, Art. 315 of the EU–​Ukraine AA and Art. INST.24.1 and INST.24.3 of the EU–​UK TCA offer a two-​step compensatory mechanism for compliance failures in matters related to free trade, encompassing (i) a mutually satisfactory compensation, agreed in terms of consultations and (ii) the right of the other Party to unilaterally suspend obligations arising from the free trade-​related provision. Such a suspension shall be temporary, and only apply until the Party complained against notifies the other Party and the Trade Committee of the measures taken to achieve compliance (European Union–​Ukraine: 2014: 315; EU–​UK, Art. INST.24.13). Finally, the enforcement mechanism under the EU AAs with Eastern Neighbors contains a clause that enables the Party to suspend the Agreement in full or partly in response to the other Party’s violations of ‘essential elements’, explored above.

110

External differentiation and compliance

Financial and technical support to the implementation of the association agreements According to the management perspective, non-​compliance may stem from capacity issues rather than the Party’s wilful decision not to comply. To build the Eastern Neighbors’ capacity to comply with their obligations under the AAs/​DCFTAs and conduct ambitious legislative approximation activities, the EU channels unilateral financial and technical assistance to Ukraine, Moldova and Georgia. Between 2014 and 2020, the key legal pathway the EU utilized to deliver technical assistance and conduct capacity-​building measures in the Neighbourhood was the European Neighbourhood Instrument (European Parliament and The Council of the European Union 2014). In 2021, it has been replaced by the Neighbourhood, Development and International Cooperation Instrument. Given the continuing ‘crisis in and around Ukraine’,6 the country has been the key target for the EU’s support since early 2014. The EU’s assistance to Ukraine is marked by the combination of support to integration (the implementation of the AA/​DCFTA) and state-​building measures financed outside the scope of the European Neighbourhood Instrument (ENI) and the sectoral approach to reforms (Rabinovych 2019b: 332). The latter manifests itself in the EU’s support for sector-​specific on-​ site projects, combining technical assistance to the implementation of specific DCFTA chapters (e.g. competition, state aid, public procurement, sustainable development) and capacity-​building measures for professionals in these areas (Wolczuk and Zeruolis 2018: 11–​15). Besides, the EU is active in supporting small businesses (e.g. through the DCFTA SME Direct Support Facility) and civil society in the region. In this vein, civil society is viewed as a co-​creator and a ‘watchdog’ of reforms. Hence, apart from clearly correlating with the management perspective of compliance, the EU’s financial and technical support to the AAs’/​DCFTAs’ implementation may positively influence compliance through strengthening domestic pressure groups, such as business and civil society, and increasing the rules’ legitimacy, due to the pathways for civil society to co-​create the rules and the EU’s external image as a donor and benign partner.

Conclusion The aim of this chapter has been to explore theoretical and legal aspects in the EU external DI. In theoretical terms, compliance can be linked to several key factors, such as the costs and benefits (non)compliance entails, rules’ certainty, capacity, domestic pressures and rules’ legitimacy. Since the DI mode of relations with a non-​EU state precludes the Commission from launching infringement proceedings in the case of non-​compliance, the external DI constellations provide for a complex legal toolbox to promote/​ensure compliance. Notably, some of the considered legal instruments speak to several theoretical approaches to compliance (e.g. political consultations can be examined from both the external enforcement and management perspectives). While there are numerous legal tools to enforce compliance under the external DI, increasing the partner countries’ capacity to implement new norms, the engagement with domestic pressure groups and increasing rules’ legitimacy requires the daily dialogue between different stakeholders in the EU and partner countries, such as parliaments, the representatives of businesses and civil society. Hence, amid the post-​Brexit EU–​UK relations and the deepening of the EU’s links with associated Neighbors, the further evolution of external DI will require not only sophisticated legal tools to resolve disputes and enforce norms but ever new venues for multi-​stakeholder dialogue and cooperation, including the exchange between countries that actively take part in

111

Maryna Rabinovych and Anne Pintsch

external DI, i.e. Norway, Liechtenstein, Iceland, Switzerland, Ukraine, Moldova, Georgia and, prospectively, the UK. The key suggested direction for future research is to conduct empirical studies of compliance in various DI contexts.

Notes 1 As of 26 May 2021, the assessment, conducted by the EU and Swiss Parties demonstrated that the Institutional Framework Agreement cannot be signed due to the ‘substantial differences between Switzerland and the EU on key aspects of the agreement’, especially when it comes to wages and state aid. The Parties agreed to hold political dialogue on continued cooperation (Schweizerische Eidgenossenschaft, 2021). 2 It should be noted that Tallberg does not consider the domestic path as a separate approach to compliance but instead relates it to both external enforcement and management: ‘The decentralized compliance system fulfills both enforcement and management functions’ (2002: 621). Since we think that disaggregating approaches to compliance in more than two categories will give us a clearer picture, we subsume this part of Tallberg’s argument under the domestic path. 3 Hereby we understand the ‘association agreement’ as the legally binding agreement between the EU and a third country (countries), aimed at fostering closer relationships between the EU and this country (countries). The legal basis is constituted by Part V of Title V under the TFEU (Art. 216–​219 TFEU, Art. 218 TFEU, in particular). In this chapter, we focus on AAs that include the adoption of EU norms and EU rules beyond the EU’s territory, thus fostering differentiation. 4 According to the ESA, Norway’s and Iceland’s legislatures shall undertake formal incorporation procedures, while in Liechtenstein the incorporation happens automatically, since Liechtenstein recognizes EU regulations as directly applicable. 5 This section does not discuss the Schengen Agreement or the Schengen Convention, since these legal acts do not contain the DSM. 6 Hereby we apply the definition of the conflict suggested by the Organization for Security and Cooperation in Europe (OSCE). For more details, please visit: https://www.osce.org/ukrainecrisis.

Bibliography Amenta, E., Caren, N., Chiarello, E. and Su, Y. (2010) The Political Consequences of Social Movements. Annual Review of Sociology 36(1), pp. 287–​307. doi: 10.1146/​annurev-​soc-​070308-​120029. Angelova, M., Dannwolf, T. and König, T. (2012) How Robust Are Compliance Findings? A Research Synthesis. Journal of European Public Policy 19(8), pp. 1269–​1291. doi: 10.1080/​13501763.2012.705051. Ashraf, N. and van Seters, E. (2020) Making It Count: Civil Society Engagement in EU Trade Agreements. ECDPM. Available at: https://ecdpm.org/publications/making-count-civil-society-engagement-eutrade-agreements/​(accessed 8 November 2020). Avdeyeva, O. (2010) States’ Compliance with International Requirements: Gender Equality in EU Enlargement Countries. Political Research Quarterly 63(1), pp. 203–​217. doi: 10.1177/​1065912908327231. BBC (2020) Brexit: EU Offer ‘Unacceptable’ As Trade Talks Continue. Available at: https://www.bbc.com/ news/uk-politics-55290109 (accessed 14 December 2020). Berglund, S., Gange, I. and van Waarden, F. (2006) Mass Production of Law. Routinization in the Transposition of European Directives: a Sociological-​Institutionalist Account. Journal of European Public Policy 13(5), pp. 692–​716. doi: 10.1080/​13501760600808550. Börzel, T. A. (2000) Why There Is No ‘Southern Problem’. On Environmental Leaders and Laggards in the European Union. Journal of European Public Policy 7(1), pp. 141–​162. doi: 10.1080/​135017600343313. Börzel, T. A., Hofmann, T., Panke, D. and Sprungk, C. (2010) Obstinate and Inefficient: Why Member States Do Not Comply With European Law. Comparative Political Studies 43(11), pp. 1363–​1390. doi: 10.1177/​ 0010414010376910. Börzel, T. A., Hofmann, T. and Panke, D. (2012) Caving in or Sitting It Out? Longitudinal Patterns of Noncompliance in the European Union. Journal of European Public Policy 19(4), pp. 454–​471. doi: 10.1080/​ 13501763.2011.607338. Börzel, T. A. and Buzogány, A. (2019) Compliance with EU Environmental Law. The Iceberg Is Melting. Environmental Politics 28(2), pp. 315–​341. doi: 10.1080/​09644016.2019.1549772.

112

External differentiation and compliance

Chayes, A. and Chayes, A. H. (1998) The New Sovereignty. Compliance with International Regulatory Agreements. Cambridge, MA, Harvard University Press. Checkel, J. T. (2000) Compliance and Conditionality, ARENA Working Papers WP 00/​18. Available at: https:// www.sv.uio.no/arena/english/research/publications/arena-working-papers/1994-2000/2000/ wp00_18.htm (accessed 14 December 2020). Council of the European Union (2013) Regulation 1053/​ 2013 of 7 October 2013 Establishing an Evaluation and Monitoring Mechanism to Verify the Application of the Schengen Acquis and Repealing the Decision of the Executive Committee of 16 September 1998 Setting up a Standing Committee on the Evaluation and Implementation of Schengen. OJ L 295 of 6 November 2013. Dai, X. (2005) Why Comply? The Domestic Constituency Mechanism. International Organization 59(2), pp. 363–​398. Dolle, T. (2015) Human Rights Clauses in EU Trade Agreements: The New European Strategy in Free Trade Agreement Negotiations Focuses on Human Rights –​Advantages and Disadvantages, in N. Weiß and J. M. Thouvenin (eds) The Influence of Human Rights on International Law. London: Springer, pp. 213–​228. Downs, G. W., Rocke, D. M. and Barsoom, P. N. (1996) Is the Good News about Compliance Good News about Cooperation?. International Organization 50(3), pp. 379–​406. Downs, G. W. and Jones, M. A. (2002) Reputation, Compliance and International Law. The Journal of Legal Studies 31(1), pp. 95–​114. doi: 10.1086/​340405. European Commission (2020) Implementation of the Trade and Sustainable Development Chapter in Trade Agreements. TSD Committees and Civil Society Meetings. Available at: https://trade.ec.europa.eu/ doclib/press/index.cfm?id=1870 (accessed 8 November 2020). European Economic Area (1994) Agreement on the European Economic Area. OJ L 1 of 03.01.1994. European Free Trade Association (2020) Agreement between the EFTA States on the Establishment of a Surveillance Authority and a Court of Justice. Available at: https://www.efta.int/sites/default/files/ documents/legal-texts/the-surveillance-and-court-agreement/Surveillance-and-Court-Agreementconsolidated.pdf (accessed 27 November 2020). European Free Trade Association Surveillance Authority (2020) Internal Market Scoreboard. Available at: https://www.eftasurv.int/cms/sites/default/files/documents/gopro/IM%20Scoreboard%2046%20 -%20for%20publication.pdf (accessed 8 November 2020). European Parliament and the Council of the European Union (2014) Regulation (EU) No 232/​2014 of the European Parliament and of the Council of 11 March 2014 Establishing a European Neighbourhood Instrument. OJ L 77/​27 of 15 March 2014. European Union-​Kosovo (2016) Stabilisation and Association Agreement between the European Union and the European Atomic Energy Community, of the One Part, and Kosovo*, of the Other Part. OJ L 71 of 16 March 2016. European Union-​ Switzerland (2002) Agreement between the European Community and the Swiss Confederation on Air Transport. Final Act. Joint Declarations. OJ L 114 of 30 April 2002, pp. 73–​90. European Union-​Switzerland (2018) Abkommen zur Erleichterung der bilateralen Beziehungen zwischen der Europäischen Union und der Schweizerischen Eidgenossenschaft in den Bereichen des Binnenmarkts, an denen die Schweiz teilnimmt. 23 November 2018. Available at: www.eda.admin.ch/​dam/​dea/​de/​ documents/​abkommen/​Acccord-​inst-​Projet-​de-​texte_​de.pdf (last accessed 8 November 2020). European Union-​Ukraine (2014) Association Agreement between the EU and Its Member States, of the One Part, and Ukraine, of the Other Part. OJ L 161/​3 of 29 May 2014. European Union-​United Kingdom (2020) Trade and Cooperation Agreement between the European Union and the European Atomic Energy Community, of the One Part, and the United Kingdom of Great Britain and Northern Ireland, of the Other Part. OJ L 444/​14 of 31.12.2020. Falkner, G., Hartlapp, M. and Treib, O. (2007) Worlds of Compliance:Why Leading Approaches to European Union Implementation are Only ‘Sometimes-​True Theories’. European Journal of Political Research 46(3), pp. 395–​416. doi: 10.1111/​j.1475-​6765.2007.00703.x Franck, T. M. (1990) The Power of Legitimacy among Nations. New York, NY, Oxford University Press. Franck, T. M. (1997) Fairness in International Law and Institutions. Oxford, Oxford University Press. Frommelt, C. (2017) In Search of Effective Differentiated Integration: Lessons from the European Economic Area (EEA). Dr. Sc. ETH Zurich: Dissertation. Hachez, N. (2015) Essential Element Clauses in EU Trade Agreements Making Trade Work in a Way that Helps Human Rights. Leuven Centre for Global Governance Studies Working Paper No 158.

113

Maryna Rabinovych and Anne Pintsch

Available at: https://ghum.kuleuven.be/ggs/publications/working_papers/2015/158hachez (accessed 8 November 2020). Hathaway, O. A. (2007) Why Do Countries Commit to Human Rights Treaties? Journal of Conflict Resolution 51(4), pp. 588–​621. Hillebrecht, C. (2014) The Power of Human Rights Tribunals: Compliance with the European Court of Human Rights and Domestic Policy Change. European Journal of International Relations 20(4), pp. 1100–​ 1123. doi: 10.1177/​1354066113508591. Jacobson, H. K. and Brown Weiss, E. (2000) Assessing the Record and Designing Strategies to Engage Countries, in E. Brown Weiss and H. K. Jacobson (eds) Engaging Countries. Strengthening Compliance with International Environmental Accords. Cambridge: Cambridge University Press, pp. 511–​554. Keohane, R. O. (1984) After Hegemony. Cooperation and Discord in the World Political Economy. Princeton, NJ: Princeton University Press. Kochenov, D. (2008) The ENP Conditionality: Pre-​Accession Mistakes Repeated in Pioneer Europe?, in L. Delcour and E. Tulmets (eds) Testing EU Foreign Policy in the Neighbourhood. Baden-​Baden: Nomos, pp. 105–​120. Leruth, B., Gänzle, S. and Trondal, J. (2022) Differentiation in the European Union as a Field of Study, in: B. Leruth, S. Gänzle and J. Trondal (eds) The Routledge Handbook of Differentiation in the European Union. London: Routledge, pp. 1–​51. Leruth, B., Gänzle, S. and Trondal, J. (2019) Differentiated Integration and Disintegration in the EU after Brexit: Risks vs Opportunities. Journal of Common Market Studies 57(6), pp. 1383–​1394. doi: 10.1111/​ jcms.12957. Moberg, A. (2015) Can the European Union Use Agreements to Even Out the Global Imbalance in the Protection of Human Rights?, in A. B. Engelbrekt, M. Mårtensson, L. Oxelheim and T. Persson (eds) The EU’s Role on Fighting Global Imbalances. Cheltenham: Edward Elgar Publishing, pp. 138–​164. Potjomkina, D. (2018) Multistakeholderism in the EU’s Trade Governance. TEPSA Brief. Available at: http://tepsa.eu/wp-content/uploads/2018/11/TEPSA-Brief-Diana-Potjomkina_FINAL.pdf (accessed 27 October 2020). Rabinovych, M. (2019a) EU Free Trade Agreements as an Instrument of Promoting the Rule of Law in Third Countries: A Framework Paper, in M. Bungenberg, M. Krajewski, C. Tams, J. Terhechte and A. Ziegler (eds) European Yearbook of International Economic Law 2019, vol 10. Springer: Cham, pp. 285–​ 314. doi: 10.1007/​8165_​2019_​31. Rabinovych, M. (2019b) EU’s Development Policy vis-​à-​vis Ukraine after the Euromaidan: Securitisation, State-​ Building and Integration. Journal of East European Politics 35(3), pp. 332–​ 350. doi: 10.1080/​ 21599165.2019.1604338. Risse, T., Ropp, S. C. and Sikkink, K. (eds) (1999) The Power of Human Rights. International Norms and Domestic Change. Cambridge, Cambridge University Press. Schengen Convention (1985) Convention Implementing the Schengen Agreement of 14 June 1985 between the Governments of the States of the Benelux Economic Union, the Federal Republic of Germany and the French Republic on the Gradual Abolition of Checks at Their Common Borders. OJ L 239 of 22.09.2000. Schimmelfennig, F. (2017) Differentiation and Self-​Determination in European Integration. EU Borders Working Paper 01. Available at: www.ibei.org/​working-​paper-​01-​euborders_​145111.pdf (accessed 27 October 2020). Schimmelfennig, F. (2018) Brexit: Differentiated Disintegration in the European Union. Journal of European Public Policy 25(8), pp. 1154–​1773. doi: 10.1080/​13501763.2018.1467954. Schimmelfennig, F. and Sedelmeier, U. (2020) The Europeanization of Eastern Europe: The External Incentives Model Revisited. Journal of European Public Policy 27(6), pp. 814–​ 833. doi: 10.1080/​ 13501763.2019.1617333. Schweizerische Eidgenossenschaft (2021) No Signing of Swiss-​EU Institutional Agreement. Available at: https://www.admin.ch/gov/en/start/documentation/media-releases.msg-id-83705.html (accessed 14 December 2021) Schweizerische Eidgenossenschaft (2019) Bericht über die Konsultationen zum institutionellen Abkommen zwischen der Schweiz und der Europäischen Union. Available at: https://www.dfae.admin.ch/dam/ europa/de/documents/bericht_konsultationen_insta/BR-Bericht-Konsulationen-InstA-190607_ de.pdf (accessed 27 October 2020). Simmons, B. (2010) Treaty Compliance and Violation. Annual Review of Political Science 13, pp. 273–​296. doi: 10.1146/​annurev.polisci.12.040907.132713. 114

External differentiation and compliance

Tallberg, J. (2002) Paths to Compliance: Enforcement, Management, and the European Union. International Organization 56(3), pp. 609–​643. doi: 10.1162/​002081802760199908. Treib, O. (2014) Implementing and Complying with EU Governance Outputs. Living Reviews in European Governance 9(1), pp. 5–​33. doi: 10.12942/​lreg-​2014-​1. Valsson, O. (2017) EFTA Court Is Not Bound to Make Use of ECJ Rulings. Financial Times. Available at: https://www.ft.com/content/268fa1a2-ca2f-11e7-ab18-7a9fb7d6163e (accessed 8 November 2020). Van der Loo, G. (2016) The EU-​Ukraine Association Agreement and Deep and Comprehensive Free Trade Area: A New Legal Instrument for EU Integration without Membership. Leiden: BRILL. Vienna Convention (1969) Vienna Convention on the Law of Treaties, U.N.T.S., 1155, I-​18232. Winzen, T. and Schimmelfennig, F. (2016) Explaining Differentiation in European Union Treaties. European Union Politics 17(4), pp. 616–​637. doi: 10.1177/​1465116516640386. Wolczuk, K. and Zeruolis, D. (2018) Rebuilding Ukraine: An Assessment of EU Assistance. Chatham House Research Paper 2018/​ 08. Available at: https://www.chathamhouse.org/sites/default/files/ publications/research/2018-08-16-rebuilding-ukraine-eu-assistance-wolczuk-zeruolis.pdf (accessed 8 November 2020). World Trade Organization (2020) Technical Information on Safeguard Measures. Available at: https://www. wto.org/english/tratop_e/safeg_e/safeg_info_e.htm (accessed 8 November 2020). Zhelyazkova, A. (2014) From Selective Integration into Selective Implementation: The Link between Differentiated Integration and Conformity with EU Laws. European Journal of Political Research 53(4), pp. 727–​746. doi: 10.1111/​1475-​6765.12062.

115

8 From integration to fragmegration Political symbols and the emergence of differentiated European identities Russell Foster

To-​morrow the European sense of nationality will be awakened. Englishmen and Frenchmen, Germans and Italians, Poles and Spaniards, and all the other peoples of Europe will realize that they are children of a common civilization, a common race, and a common continent, heirs of a common tradition, bearers of a common mission, bound together forever in a community of destiny. Richard von Coudenhove-​Kalergi, 1940; Drace-​Francis 2013: 205 Since the emergence of proposals for a pan-​European project in twentieth-​century Europe, questions of identity, belonging, and emotion have been a visible, if not particularly prominent, discussion. While a frequent topic among policymakers, the relationship between integration and identity remained comparatively absent from European studies until the Maastricht Treaty’s transformation of the European Economic Community (EEC) into the European Union (EU) and accelerated following the introduction of the Euro currency in 2002 and the ‘Big Bang’ expansions of 2004 and 2007, which ignited fresh debates about ‘Europe’ as space, civilization, and identity. Since the beginning of the Global Financial Crisis in 2007 and the subsequent Great Recession, discussions of European identity and the EU have become more pronounced as the EU and its member states have staggered from one crisis to another; the Migration Crisis, Crimea, the breakdown of relations with Turkey, questions concerning the transatlantic relationship, and widespread dissatisfaction with the Union among politicians and populations, manifesting in the 2016–​2020 Brexit process and a wave of anti-​establishment, anti-​EU populists in the ‘verrechtssing’ (Mudde 2013) or ‘right-​turn’ of European politics since 2016. With the departure of the United Kingdom from the EU, and the prospect of the breakup of the United Kingdom and further surges in nationalism and anti-​EU politics due to future spontaneous external crises and systemic internal crises whose economic, constitutional, and cultural roots still have not been addressed, questions of identity are likely to become even more urgent in coming years. Of the many challenges facing the EU perhaps the most significant is what Samuel Pufendorf in 1683 had identified as the greatest threat facing the Holy Roman Empire –​an ungleichzeitigskeit des gleichzeitigen, or deficit of simultaneity. Put simply, the absence of a sufficiently strong identity is necessary to grant legitimacy, in the eyes of the population, to institutions and their power. 116

DOI: 10.4324/9780429054136-9

European identities

For Pufendorf, the Empire had imperial institutions but not enough of an imperial identity for people to accept the institutions and their rulings as legitimate, and would ultimately lead to the Empire breaking apart. Pufendorf was not entirely accurate –​the Empire lasted another century. But neither was he entirely inaccurate –​the end of the Empire in 1806 was met with neither celebration nor lamentation, but collective indifference to a project which, to its people, was already irrelevant (Foster 2015).The EU is perhaps experiencing the same phenomenon of centrifugal forces encouraging people to identify with their local communities even as centripetal forces encourage people –​ on the peripheries of the EU –​to identify with a Union that they are not part of. Contrary to the predictions of modernist theorists generally, and neo-​functionalist and intergovernmentalist theorists of the EU more specifically, the twenty-​first century has seen identities and identity politics not only becoming more visible but more impactful on EU politics, as disillusioned populations express their grievances and anxieties at the ballot box. This chapter makes three contributions to the study of European disintegration. First, it argues that postfunctionalist theory, while not unproblematic, is a crucial theory for understanding the growth of Euroscepticism in the twenty-​first century and the threat this poses to the integration (and existence) of the EU. Second, the chapter argues that despite being understudied EU symbols are the most significant source of European identity formation versus national identities, and that due to their affective (emotional) nature, EU symbols are extremely powerful mediators of identity. Third, the chapter argues that the EU’s symbols have indeed created a European identity, but have also led to a growing rejection of the EU among some Europeans, and that the symbols of the EU might now be as counter-​productive to a European identity as they are (or were) productive. Of the many structural and spontaneous problems which the EU has faced, faces and will face in the twenty-​first century, arguably one of the most significant is a resentment or rejection of the EU by its own citizens whose identities do not necessarily align with the vision of Europeanness shared by EU leaders since the 1950s.To understand this, it is essential to study the power of symbols.

Legitimacy and Europe The twenty-​first century has seen the EU face a variety of challenges. It has long been recognized that the EU occupies a uniquely difficult place in national and European political consciousnesses, rarely been given credit for the successes of its institutions while repeatedly being made into a scapegoat for problems not of the EU’s making. Since the onset of the Global Financial Crisis in 2007/​2008, this has manifested as Euroscepticism directed towards the aftershocks of the subprime mortgage crisis, the migration policies of member states, Russian military activity, and dissatisfaction with EU policies ranging from migration policies in 2015 to the 2021 ‘vaccine war’ between the EU and United Kingdom (The Economist 2021), and further tensions beyond the Covid-​19 pandemic. This reflects a severe problem for the EU, one which is not a new concern but is traceable throughout the process of integration, particularly since Maastricht. Before the Maastricht Treaty, the EEC was a distant and rarely considered entity in the everyday lives of Europeans, due not only to a smaller and less frequent presence in analogue media but also due to the limited competences of EEC institutions and impact on everyday lives. The 1990s–​2000s transition from newspapers and broadcast news to social media and 24-​hour digital communications, and the increasing presence of the EU in the governance of peoples’ lives, however, has transformed ‘Europe’ from a distant entity which received as much consideration as the United Nations (UN) receives in contemporary society into a concept which, in the 117

Russell Foster

twenty-​first century, has become a framing device and site of projection for economic, political, social, and cultural debates at local, national, regional, and continental levels. Symbols are significant here for three reasons. First, the EU’s symbols have become extremely visible, transforming from physical flags and icons which were rarely, if ever, encountered by EU inhabitants outside of Brussels and Strasbourg, into immediately recognizable icons1 which proliferate across digital communications and physical performances.2 Second, this proliferation of EU symbolism has caused the EU itself to become a symbol, a site of projection onto which people project positive and negative emotions (which, as argued below, is precisely how symbols function and is what EU symbols were intended to do). Third, the EU’s symbols reflect the emergence of competing and highly differentiated visions of integration and disintegration across Europe, a divide between those who seek ‘More Europe’, ‘Less Europe’, ‘No Europe’, and those who seek an ‘alter-​Europe’ ranging from the morally benign (but politically controversial) visions of European federalists, to the ‘Love Europe, Hate the EU’ rhetoric of the far left, to the visions of white nationalists seeking to transform the EU into a fortress built on their Thrawn philosophies. All of these illuminate the complex relationship between the EU, legitimacy, and symbols. Notwithstanding disagreements about whether the EU should have a democratic deficit or whether it is an unhelpful myth, academic consensus broadly agrees that at the root of the EU’s problems is an apparent lack of legitimacy, with frequent complaints from citizens, policymakers, and academics that the Union suffers from a democratic deficit which undermines the EU’s legitimacy in public eyes (Hix 2008).This relationship between Europeanness and political legitimacy has long been an issue. Prior to the Big Bang enlargements and a sequence of crises, the EU enjoyed procedural legitimacy resulting from its perception among member states as a progressive institution (Cafruny and Lankowski 1997). However, perennial discussions of the EU’s democratic deficit highlight a continuing and growing divide. With European citizens’ trust shaken, the EU’s legitimacy is increasingly challenged. This reflects a profound and widespread lack of citizen affiliation with the EU, with Eurobarometers 80–​90 recording a declining sense of Europeanness among member states populations, and coinciding with the rise of nativist, nationalist, antiglobalization politics in the twenty-​first century. This democratic deficit feeds into Europe’s deficit of simultaneity. While a European political entity exists, and there is some degree of corresponding identity to which some Europeans adhere, there are equally (and perhaps more popular) Europhobic or soft Eurosceptic identities which reject the EU in its current form. The proliferation of European symbolism in members or former members of the EU reflects an overt European identity, but at the same time, the EU’s symbols seem to be enhancing national rather than supranational affiliations among Europeans which illuminates a fracturing of polity and community, an ungleichzeitigskeit des gleichzeitigen. Ultimately, the relationship between EU and citizen is built not through formal political participation such as voting nor through EU legal procedures, but through visual communication and symbolic encounters in everyday life. It has long been recognized that a European community relies upon the emergence of a European identity (Haas 1963). As emphasized prominently by Habermas (2013: 122), ‘the European Union is at a crossroads’, while for former Dutch Prime Minister Jan Peter Balkenende (Poettering 2006: 31), ‘Europe cannot be built without people who feel European’. Since the onset of the EU’s ‘general crisis’ from 2008 onwards, scholars and policymakers have increasingly questioned the impact of the EU in citizens’ lives, asserting that the future of the Union rests in citizens’ support for, and political identification with, the EU (Foster 2015; McCormick 2013; Cram 2012). Support for the EU waxes and wanes in response to external and internal challenges, and consequently public support for EU membership, is consistently unstable. While 118

European identities

post-​Brexit opinion polls indicate that public support for the Union rose in response to Britain’s fraught departure, it is arguable that this was a temporary effect caused by publics’ and politicians’ desires to avoid replicating the social traumas of the Brexit process. Similarly, the global coronavirus pandemic from 2020 onwards resulted in a ‘rally around the flag’ effect around the world in 2020 (UK in a Changing Europe 2020), but tensions between the EU’s institutions and MS governments and populations, over vaccinations, revealed the continuation of Euroscepticism which manifests in response to any new emergency in the EU. The waxing and waning of Europhilia and Euroscepticism since 2000, what Michael Wintle (2011b: 5) terms a ‘rollercoaster’, is a phenomenon identifiable in Europe since the 1950s and is unlikely to disappear precisely because the EU acts as a site of projection for populations’ and politicians’ dissatisfaction with national politics. There are multiple causes of civil disillusionment with the EU, both exogenous and endogenous. Exogenous forces can be interpreted as spontaneous crises, including the aftershocks of the US subprime mortgage crisis, the refugee crisis following the Arab Spring, the ongoing war in eastern Ukraine, and the Covid-​19 pandemic. While these exogenous forces are potent and reveal severe inadequacies in supranational institutions and intergovernmental policymaking, they are arguably not the biggest challenge to the EU. Instead, endogenous problems or structural crises are much greater threats, including such problems as the unequal and unstable Eurozone, but much more urgently, citizens’ disaffection with an ostensibly distant, technocratic set of institutions with which people have limited emotional identification. More than exogenous threats or the Eurozone, this chapter argues, the greatest threat to the EU comes from citizenry who do not feel European, and who oscillate between grudging acceptance and overt resentment towards the Union. This is extremely difficult to manage. Partly this is due to the EU’s inability or refusal to acknowledge slow-​burning social resentments towards the EU, which are far less visible than external shocks; and partly because the emotional or affective nature of peoples’ lack of identity cannot be easily managed. As Sam Leith (2008) wrote of the European Parliament’s efforts to codify EP symbols into law: ‘you can manage taxes and borders and subsidies and rebates and bureaucratic structures, fine. But you can’t manage the collective unconscious’.

Unconscious Europe Theories of European integration have, historically, overlooked the role of individual and group identities on the integration and disintegration of the European project. Classical theories of integration, namely Ernst Haas’ neo-​functionalism, Stanley Hoffman’s intergovernmentalism, and Andrew Moravcsik’s liberal intergovernmentalism have privileged economics, the role of national and transnational institutions, transactions, territory, and the influence of policymaking, as the drivers of European integration. While these theories hold great merit, a common weakness is a focus on the ‘high politics’ of institutions and policymakers, and a consequent lack of engagement with the ‘low politics’ of identity formation among citizens. Lisbet Hooghes’ and Gary Marks’ postfunctionalism encourages an alternative approach to European integration not simply at the level of high politics, but examining three developments: first, the rapid politicization of European integration in national politics; second, citizens’ feelings about the EU influencing elections; third, the power of identity in shaping attitudes towards European integration (and arguably disintegration). Like competing theories, this postfunctionalist approach is not without criticism (Moravcsik 2019) as it can overemphasize identity while neglecting concrete policymaking’s impact on European integration. Similarly, criticism can be (and is) made towards all ‘grand theories’ of European integration, including postfunctionalism (Hooghe and Marks 2019) in which competing and perhaps incompatible elements of European integration 119

Russell Foster

are crowbarred into theoretical frameworks. There is no uncontestable theory of European integration, even less so regarding disintegration, but it is arguable that postfunctionalism is the most suitable theory for examining an under-​researched but crucial element of the centripetal and centrifugal forces affecting European identity since the 1950s –​symbols. European policymakers have sought to enhance Europeanness, through symbols and visual language, since the 1950s (Calligaro 2013). Despite academic arguments on the relationship between, and relative merits of, different theories of integration (and disintegration), it is arguable that in policymaking on symbols and identity, postfunctionalism’s focus on identity and the power of citizens’ emotions on voting behaviour is the most valid. However, the actual policymaking of European symbolic policy has been, and remains, dominated by neo-​functionalism. Since the signing of the Treaty of Rome, the drive for European unity has been predicated in this theory, stressing economic cooperation and assuming that the eventually required, supranational ‘European’ identity would emerge organically from the spillovers of rational and functional stages of economic and constitutional integration (McCormick 2007: 239–​312). Simultaneously, increasing intra-​European communication encouraged the perception among academics and policymakers that greater contact between Europeans would result in the lessening of parochialism while encouraging an identity as ‘European’. The EU has sought to build upon this presumed ‘sameness’ to engender a ‘European’ identity expressed through symbolism. Today, amidst continuing EU crises, this ‘Gospel of Jean Monnet’ (Barroso in Poettering et al. 2006: 14–​ 16) continues to form the basis of EU policy (European Parliament 2013). This may have been an astute stance during the nascence of unified Europe, but since the 1993 Maastricht Treaty’s transformation of a distant and technocratic EEC into a more political and social union with a presence in everyday life, particularly since the introduction of a common currency in 2002 alongside rapid expansion in territory and institutional competences, Europe has developed from an economic community to a political union whose sociopolitical impact is increasingly visible. Simultaneously, the expansion of communications technologies is enhancing the ocularization (Rose 2012) of European society, constructing an imagined community (Anderson 1991) that is created in, and expressed through, visual communication. As all communication is symbolic, composed of abstract sounds and arbitrary pictograms which convey meaning (meaning which is established only through social convention), the EU’s symbols convey distinct political connotations. However, the meaning conveyed by the EU’s current symbols is not agreed through organic social convention but imposed from above. For EU policymakers, these symbols communicate the concept of a united Europe. But for European populations, these symbols encountered in everyday life have multiple meanings, and these polysemic interpretations of the EU can, and do, influence political behaviour. Political behaviour is defined by symbols, for ‘all social and economic activities are carried on by means of symbols … society is held together by acceptance of, and reverence for, its symbols’ (Whittick 1960: 3). Visual language and symbols constitute the initial and most frequently encountered connection between citizen and the EU, particularly in newcomer and neighbouring states (Kuus 2007; Moisio et al. 2012: 77–​761). As their power stems from the very mundanity and omnipresence of these artefacts, symbols and visual rhetoric are saturated with embedded politics (Thrift 2004: 57–​78). Collectively, these symbols form a version of what Guy Debord (1990, 1967: 12–​13) terms the Spectacle; ‘not a collection of images, rather…the social relationship between people that is mediated by images’. Significantly, these images and symbols are, regardless of their makers’ intentions, representative of ‘a complex set of ideologies and behavior’ (Firth 1973: 329–​330). They convey, therefore, ‘spectacular expressions of state power’ (Painter 2006: 752–​774) which are charged with emotion because of the everyday context in which they appear. 120

European identities

EU symbols exist in the quotidian realm of everyday life, notably the visual technoculture of hypermedia: the interactive, multifaceted, everyday dimension of digital technologies, visual communications, and physical presence of artefacts of material culture, such as flags, clothing, and currency (Deibert 2011). This is the realm in which the greatest contact between EU and citizen occurs, and thus is the realm in which identity is formed through symbolic interactionism, whereby identities (and the meanings of symbols) are not static, but are constantly being formed and challenged (Street 2011). As Eurosceptic trends since the early discussions of ‘Grexit’ have demonstrated, political attitudes towards the EU, and political activity by populations and politicians, are not defined by purely rational thought but by emotions. This was particularly visible in the Brexit referendum campaign and the subsequent five years of social polarization and political deadlock in the United Kingdom, which saw logos superseded by pathos; feelings over facts. As national and elections across the EU from 2016 onwards demonstrated (and arguably will continue to demonstrate, beyond 2021), this phenomenon was, and is, far from unique to the British. In the context of a supranational entity whose governance remains distant from citizens, ‘political life is rooted in the manipulation of symbols’ (Cheles and Sponza 2001). EU policymakers continue to assume that a European (or more specifically, ‘EU-​ropean’) identity will emerge as a consequence of functionalist spillover and, with rare exceptions, scholarly studies of the European project (and indeed, mainstream political science and international relations theory) continue to approach symbols and iconography, and their reflections and creations of multiple and competing identities, to be ‘something of a quaint survival … a bit of puffery that has little influence on the real stuff of contemporary politics’ (Kertzer 1996: 150–​159). But this emphasis on what Max Weber termed the ‘iron cage’ (stahlhartes Gehäuse) of rational, systematic political activity as the determinant forces behind European integration and disintegration, fails to adequately capture the affective, emotive, and identitarian forces which, since the 2010s, have arguably been more responsible for disaffection with the European project than anything quantifiable and rational. As has been demonstrated by Brexit and widespread support for anti-​EU parties and politicians across the EU (and the broader West), formal, rational politics cannot compete with the fundamentally arational (i.e. unquantifiable and beyond the realm of static categorization) emotional responses to symbols and visual language, which act as extremely powerful mediators and communicators not only of identities, but of those emotions which are channelled and directed against the convenient scapegoat of the EU. Emotion is not an adjunct of politics but the very foundation,3 and it is through symbols that citizens’ emotions are funnelled into political behaviour. The emotional significance of EU symbols is masked by the quotidian or ‘everyday’ contexts in which they are used (Lüdtke 1995). A growing body of research since the performative and narrative turns (García 2017), though, demonstrates that politics at an EU level is not only about the state and state/​supranational institutions and intergovernmentalist bargaining, but also about everyday practices, habits, and imaginations. Different schools in the social sciences and humanities, namely the Annales school and the work of Henri Lefebrve (1991), Fernand Braudel (1992), and Reinhart Kosselleck (2002) have, through their focus on ‘Alltagsgeschichte’ or ‘everyday narratives’, helped us to understand the ‘prosaic banalities’ of everyday practices which not only resonate with high politics, but indeed create politics at its very foundation as what Laura Cram (2001) terms ‘Banal Europeanism’. Consequently, the most frequently encountered link between EU and citizen is not legislation, but symbols: flags, icons, currency, and other media. Thus, it is through symbols that Europeanness is formed. However, a problem exists. As demonstrated by rising Euroscepticism after the Global Financial Crisis, the EU’s symbols no longer work precisely as intended and are instead illustrative of a divergent, differentiated relationship between symbols and European identity. To many 121

Russell Foster

people inside and outside the EU, iconographies of the EU indeed reflect a positive identity (as demonstrated by the overt displays and performances of EU symbolism among British Remainers and Ukrainian citizens who wish to join the EU [Foster 2018]), exactly as they were intended. Yet to many other Europeans, the same iconographies represent, to them, an organization which is at best a distant and uncaring technocracy, and at worst an actively malevolent force intent on replacing peoples’ national or local identities through practices which Chomsky described as ‘manufacturing consent’ or ham-​fisted attempts at social engineering.4 Consequently, EU symbols reflect both centripetal and centrifugal forces in EU politics –​representing something positive or negative with little space in between. This is a reflection of the post-​2016 British phenomenon of sacralization, the splitting of the EU in public imagination into either entirely positive or entirely negative, with symbols of the EU (and member states) being deployed to overtly proclaim uncritical adherence to one identity (national or European), and complete condemnation of the other. EU citizens are not purely rational actors, and their relationship with the Union is not always defined by conscious and considered thought (Poyatos 2002). This is especially so among EU and pre-​/​post-​EU demographics who are disconnected from the Union either due to myths and narratives which construct the EU as something distant, alien, and hostile (i.e. British Leavers and Ukrainian anti-​EU populations, but a similar phenomenon is emerging among post-​Soviet populations in Hungary and Poland), and people disproportionately affected by the fallout from EU and national austerity measures. As a result, EU symbols come to embody what Albrecht Sonntag (2011: 125–​126) terms the ‘fossilization of social habitus’. The EU’s symbols have not changed in terms of what they represent (the EU), but they have changed in terms of what they mean (their emotional significance) to EU populations. Various factors have caused this divergence of symbolic identity and the transformation of EU iconographies into emotional vehicles of fiercely pro-​or anti-​EU sentiments, reflecting differentiated attitudes towards the EU’s integrative and disintegrative forces.While they still express what they were intended to (and arguably achieve this goal far more effectively than prior to the EU’s general crisis, as suggested by the reverence with which British Remainers and Ukrainian ‘joiners’ display, wear, and perform EU symbols), the EU’s current symbols are no longer capable of eliciting a single, pro-​EU Europeanness. Existing icons are inadequate vehicles for expressing the project of European unity as they are backwards-​looking, appealing to a manufactured myth or an archaic vision of Europeanness defined by Cold War ideology (Brague 2002; af Malmborg and Stråth 2002). In the absence of a clear Other with which to contrast the EU, these icons are not nearly as effective, and as demonstrated by groups seeking to create an alterity that defines their ‘European’ identity (British Remainers rejecting Leavers, Ukrainians rejecting Russians, far-​right ethnonationalists rejecting Muslims), any quest to manufacture an Other against which to symbolically contrast the ‘European’ Self is a dark path to tread.This is exacerbated by a lingering acknowledgement that the EU’s icons did not emerge in the primordialist past but were almost exclusively selected by elite policymakers in committees. With the expansion of the EU in the 1990s and 2000s –​geographically, socially, and institutionally as the Union encompassed more peoples and grew from a distant economic entity to a fact of daily life for its citizens (particularly through currency) –​the Union’s symbols were (and are) frequently interpreted by citizens and policymakers as unwelcome interlopers, symbols of the EU’s democratic deficit and the apparent erosion of national identity and national sovereignty. Related to this problem is that the EU’s current symbolic identity is backwards-​looking; using symbols that appeal to an imagined (and imaginary) unity in history (Stråth 2002: 387–​401) rather than embracing Europe’s geographical, cultural, and chronological diversity and looking forward to what the EU could become in the future.5 To approach the changing relationship between differentiated disintegration and symbolism, it is arguable that the EU’s current symbols are not entirely capable of appealing to emotions 122

European identities

and fostering affiliation to the Union, as they are the wrong category of symbols. EU symbols do indeed foster pro-​EU sentiments, but they also foster a variety of other identities. The EU’s current iconography does not transcend nationalism but reproduces nationalism. They belong to the realm of what we might call ‘Heroic Europeanness’: official symbols which are intended to behave like the traditional symbols of nation-​states by celebrating the past (Foster 2015; O’Shaughnessy 2004; Cram 2001), and effectively taking the place of national symbols which are entangled in complex and extremely emotional debates about history. EU policy on visual identity is predicated on the idea that the EU flag, anthem, iconography, and other forms of visual and symbolic communication can convey a myth of Europeanness (Bruter 2009). This barely functions for a national identity based in an imagined ethnic, primordial unity (Hobsbawm and Ranger 1983), and is ineffective work in a supranational entity such as the EU, which is too young to have acquired the necessary myth for people to emotionally identify with it. Within the EU, the surprise election of Eurosceptic (and often vitriolically anti-​EU) parties to the European Parliament in 2014 and 2019, alongside the electoral successes of anti-​EU parties in national elections, revealed the deep dissatisfaction of European citizens with the Union. Simultaneously, beyond the EU, the Union’s iconography has been appropriated by parties in the UK post-​Brexit, and in the Ukraine post-​2014, to represent different visions of the EU. British Remainers and Maidan protestors adopted EU iconography as a symbol of freedom and progress, while British Leavers and pro-​Russian Crimeans/​Ukrainian separatists continue to use the EU’s iconography to symbolize dissatisfaction, and to construct narratives of social and political corruption (Foster 2018). While the cases of the United Kingdom and Ukraine demonstrate that a strongly pro-​EU identity can be created rapidly, and expressed through the EU’s own symbolism, this is not a peaceful or positive process for two reasons. First, the UK and Ukrainian cases demonstrate that a strong pro-​EU sentiment and symbolic performances of Europeanness are only possible because of a traumatic political and social change (Brexit and the Crimea invasion, respectively); before the Brexit referendum symbols of Europeanness were barely visible in the United Kingdom, partly due to the UK’s historically ‘awkward’ relationship with the EU (George 1990), and partly as the EU was not at the forefront of political discourse and public consciousness for the British population. Second, overt and passionate pro-​EU identities, performed through symbolism, are a reaction to/​have created a polarized opposite identity, namely passionately anti-​EU identities. In the United Kingdom and Ukraine, the emergence of a pro-​EU symbolism has coincided with an anti-​EU symbolism as part of an all-​out war of identities (Shipman 2016) or, in Ukraine, actual military conflict. In both cases, the emergence of an EU identity that is performed through symbolism has only occurred and is maintained through a traumatic event and the polarization of society into mutually irreconcilable camps who communicate only through the most toxic of discourses –​this is not a positive phenomenon, and not one which it is desirable to imagine in other EU member states struggling with their own emerging polarization. These cases demonstrate a perhaps inevitable and very serious consequence of symbolic disintegration/​integration: peoples’ emotional identity remains with their Vaterland, not the zollverein of the EU. For British Remainers and Ukrainian pro-​EU activists, the EU is no longer a zollverein but has been reconstructed and reimagined as a new Vaterland which is imagined and treated with the same passions, ideologies, uncritical adoration, and violent, exclusionary rhetoric as any nineteenth-​or twentieth-​century nationalist discourse. Subsequently, the EU’s symbols either must compete with national symbols for citizens’ emotional attachment (a battle which EU symbols inevitably lose) or transform into symbols which, as the phenomenon of ‘toxic Remainers’ in the United Kingdom demonstrates (Guardian 2018), are symbols which replicate the symbols of nationalism –​exclusion, an imagination of superiority, and intolerance 123

Russell Foster

towards those constructed in discourse as ‘outsiders’ (in the British case; Leavers; and Remainers imagined to be insufficiently passionate towards the Remain/​Rejoin cause). The events of Ukraine and the United Kingdom, but also the post-​2016 shift towards nationalist and antiglobalist sentiment across the EU, clearly demonstrate how citizens’ identification with, or rejection of, the Union is not formed by top-​down institutional policies but is formed from the bottom up, through everyday symbols and performances. At the heart of this problem is that the EU’s current symbols, and their deployment, follow the national pattern and the assumption that an identity will ‘spill over’ from integration. This is demonstrably not the case. EU policymakers want to give the EU ‘a soul’ (Cloet et al. 2013); a political rallying-​point that appeals to Europeans’ emotions, expressed in symbols of what it means to be ‘European’. But this soul cannot be manufactured from abstract emblems and imposed from above in the hope that Europeans will identify with this vision.6 As identified above, symbols form a visual discourse and a ‘mundane omnipresence’ (Sonntag 2011) that is saturated with the normative goal of producing identity (Raento and Brunn 2005: 145–​163). Such symbols include everyday artefacts such as coins, banknotes, posters, billboards, political literature, logos on documentation, and performative manifestations such as political demonstrations (both within and beyond the EU), uses at electoral campaigns, festivals, and ceremonies. This visual discourse constructs an identity which through ongoing dialogue, continually reformed and reinforced (Carta and Morrin 2013: 1–​20; Müller 2008: 322–​338) through the everyday actions of citizens encountering the EU through symbols and thus reimagining the EU through how these symbols are used.7 Imagined and narrated in Eurosceptic discourses as an elite-​driven, economic-​centred, distant and uncaring technocracy, the Union lacks the emotional appeal to citizens whose identity, in the face of EU activities, is increasingly tied to nation-​states (Sonntag 2011) or, as British Remainers demonstrate, the EU is reimagined as a nation with the same tropes of exclusion and imagined superiority as nationalist rhetoric. The EU’s desired soul exists, but it is among Europeans’ emotions rather than policymakers’ visions.The EU is defined not by unity alone, but by unity in diversity. The EU’s current approach to its symbolic identity is to overlook this diversity while publicly expressing a myth of historical and contemporary unity across Europe: a myth which does not reflect the diverse cultures, histories and perceptions which make up everyday life in the EU, and which contrast sharply with the emotionally empty symbols produced by institutions. What it means to be ‘European/​European/​post-​national’ exists among EU (and post-​EU citizens) and forces of differentiated integration are expressed positively through symbolism, but equally, what it means to be ‘European/​non-​European/​national’ also exists among EU and post-​EU citizens, and forces of differentiated disintegration are expressed negatively (or, from their perspective, positively) through the same symbolism. The consequence is that the EU’s symbols, today, have diverged into two opposed, mutually contemptuous, mutually irreconcilable camps; one overtly anti-​EU and who express their identities and norms through old nationalist symbolism, and one overtly pro-​EU and who express their identities and norms through a new nationalist symbolism of the EU, in a fulfilment of David Mitrany’s 1963 (in Haas 1963: 51) warning that ‘the regional concept is not the triumph of the new internationalism, but only an extension of the old nationalism’.

Europa über alles? Existing research demonstrates the crucial importance of EU symbolism in the construction, reification, reinforcement, and expression of political discourses which simultaneously create and

124

European identities

reflect Europeanness (Theiler 2005; Calligaro 2013; Wintle 2009; Foster 2015). Like national iconographies, EU symbols mediate the concept of Europeanness through visual shorthand –​‘images that…can condense the meaning of a whole narrative without linguistic mediation’ (Bottici and Challand 2013). This is what lends symbols their supreme power: their ability to condense narratives into recognizable icons; a form of ‘visual shortcut to the part of our minds that understands them’ (Nozedar 2010: 13). Through this synecdochal mechanism, symbols appeal to socially constructed meanings and associations (Piercy 2013). These meanings (e.g. recognizing the state/​organization that a flag represents) are bound up in complex affective significances and emotional meanings among people who view them. As an example, the EU flag represents the same thing to a Leaver and a Remainer, but it means two completely different things, depending on the political beliefs and norms of the viewers. This is the essence of semiotic theory, the mechanisms by which symbols are interpreted and given meaning through socially constructed norms (Abdullah and Hübner 2006; Sorlin 1994; Packard 1980; Edelmann 1964; Bernays 1952). This is not unknown to the EU: the 1973 Copenhagen Declaration stressed that Europe is defined by a sameness expressed in symbols. At the same time, academic studies of political symbols’ emotional significance have an equal pedigree. However, such studies were situated almost exclusively in the rationalist tradition of Karl Deutsch which assumes almost static identities, or the Frankfurt School, interpreting EU symbols as part of a Debordian spectacle which deliberately or passively neutralizes citizens’ political behaviour. Few studies have approached EU identity as an ongoing dialogue mediated through symbols, rather than a static stance created by symbols. This illustrates a mythosdebatte that approaches symbols as fundamental to understanding the formation of political imaginations, and their consequences in the form of voting behaviour: ‘the force and effect of these mediating signs would remain a mystery if they were not ultimately rooted in…the very essence of consciousness’ (Cassirer 1953: 105–​106). Consciousness, mediated through symbols, is extremely powerful. As the British have demonstrated, not only is an EU identity fluid and flexible rather than static, but this fluidity is extremely volatile and can change extremely quickly, with EU symbols changing in the space of a few months from banal, barely observed icons on billboards in the background of everyday life and conscious awareness, to overt symbols of identity worn and painted on bodies, waved in the street and displayed on social media, expressing a passionate EU identity that is simultaneously inclusive and exclusionary, and which is one of several competing, incompatible imaginations of what ‘Europeanness’ means. However, the relationship between feeling and political behaviour –​mediated through symbols –​remains under-​addressed. This necessitates an overview of what the EU’s symbols actually are. Legally, the EU has no symbols.8 An attempt to grant formal, legal status to five symbols (the flag, anthem, motto, commemorative day, and Euro currency) in the proposed 2004 Constitutional Treaty ended with the Treaty’s quiet shelving following the 2005 Franco-​Dutch rejection in referenda, and the 2007 Lisbon Treaty dropped attempts to legalize the EU’s symbols.9 Officially, only the European Parliament has a formal iconography. However, despite their unofficial status, the EU’s symbols have become visible in everyday life through the expansion of material culture and mass media/​communication. As these symbols have not been changed since their creation or adoption from the 1960s to the 1990s (Beck and Grande 2008: 51), they have acquired a socially constructed significance, with a public consensus that these symbols represent the EU –​ but while there is a single social consensus on what they represent, there are multiple (frequently completely incompatible) agreements and disagreements on what they mean. A number of academic and policymaking investigations have inquired whether, and to what extent, the EU’s current symbolic repertoire is capable of reflecting Europeanness and encouraging citizens to affiliate with the Union, versus the extent to which these symbols convey

125

Russell Foster

a vision of the Union as hegemonic and exclusionary, symbolizing something from which people may (and do) wish to dis-​integrate rather than integrate (Theiler 2005; Fornäs 2013; Foster 2015). The European Commission, European Parliament, and European Central Bank are keenly aware of this (Fishman and Messina 2006), as demonstrated by the aforementioned attempts to legalize the Union’s visual identity. Subsequently, it can be argued that the EU has not only a democratic deficit (a concept rapidly returning to public discourse in the aftermath of Brexit and management of Covid-​19 countermeasures and vaccine supplies), but also a symbolic deficit. If the Union is to continue to develop or even exist in the face of growing dissatisfaction among European peoples, policies must reflect an awareness of the crucial importance of symbols in the formation and expression of European identities, Balkenende’s ‘people who feel European’, whose emotional connections (or lack thereof) to the Union are formed not through the politics of the polling station, but in the fluid and kinetic realm of identity, feeling, and affect. Where citizen behaviour towards the EU is infrequent (as in elections), citizen identification (or not) with the EU is much more frequent, perhaps several times a day in the case of people handling euro currency and witnessing, waving, or burning the EU flag. These extremely powerful mediators of identities are far more frequent than formal interactions with the EU through political participation, and it is ultimately through symbols and iconographies permeating and producing the mundanity of everyday life that peoples’ attitudes towards the Union are communicated, mediated, and reformed, and which ultimately manifest at the polling station. Ultimately, Europeanness is being made not by EU or national politicians, but by citizens interacting with visual symbols, being constructed and continually re-​affirmed in those prosaic, banal interactions which cement the concept of Europeanness in everyday life (Sassatelli 2017; Raento et al. 2004; Brunn 2000). Symbols chosen by EU elites –​either by professional designers or chosen by EU committees from submissions to public competitions–​are no longer symbols of a distant customs union or market area, but instead are symbols which are viewed as antithetical to national symbols and national identities, or (as the British case demonstrates), symbols of a new ‘national’ identity, which is imagined to be incompatible with existing nation-​state identities. The symbols of a united Europe are intended to promote a European identity, but the fact that these symbols are so prolific when a single European identity and European myth do not exist can give EU symbols a negative power (Kuus 2013: 30–​32). Thus, the EU is in an unsustainable position: its symbols are too recognizable and too old to be replaced, as there is universal social consensus that the EU is represented by the symbols and that these symbols represent the EU; but at the same time, its symbols are ineffective in that symbols that were designed to signify peace and unity have become divisive symbols of who does and does not, and who should and should not, belong. The EU faces the irreconcilable problem of needing a new symbolic identity that is formed in a different way to previous top-​down approaches which follow the nationalist paradigm, and which do little more than highlight the gulf between elite visions and everyday interpretations (or as illustrated in the United Kingdom, which highlight a divide between a nation of Remainers and a nation of Leavers who view the other as not only incompatible, but undeserving of sharing the same space), while simultaneously being unable to disentangle itself from its extant symbols which, after 30 years of social consensus and five years of often extreme levels of display and political polarization surrounding those symbols, are inextricable from the concept of the EU. In this way, the EU’s symbols themselves are symbolic of the EU’s differentiated processes of integration and of disintegration: they express the paradox of what Olaf Cramme (2016) calls ‘Europe Entrapped’, incapable of surviving or flourishing in its current form, but incapable of changing into something else.

126

European identities

Conclusion In the founding days of the European Coal and Steel Community, on 16 July 1951 a memorandum from Jacques Camille Paris, Secretary General of the Council of Europe, stated that: There are no ideals, however exalted in history, which can afford to do without a symbol. (CVCE 2021) The EU’s symbols are not an adjunct to integration and disintegration, but at the core. There is no single, adequate theory of European integration, and to attempt to understand why the EU has arguably entered a new era of disintegration, a synthetic theoretical framework and a diverse methodological toolkit are required. There remains merit in using liberal intergovernmentalist, historical institutionalist, and transactionalist approaches alongside postfunctionalism, as no single theory can adequately explain the diverse causes and manifestations of Euroscepticism in the twenty-​first century EU. However it is undeniable, as Brexit and Euroscepticism have demonstrated, that symbols play a crucial role in the integration and possible disintegration of the project, as symbols are the primary link between citizen and polity. To 447 million Europeans, the dominant link between individual and EU is symbolism, specifically the iconography of flag and (if in the Eurozone) currency, as the EU is not old enough to have acquired the other symbolic paraphernalia –​language, cuisine, architecture, historical heroes, jokes about the people over the border –​which defines identity against alterity, Self against Other (Fornäs 2013). This is a power which nation-​states do wield, and even in cases where a proto-​EU nation is forming, as is possible among pro-​EU populations deprived of EU membership, this is a worrying phenomenon because, as the British and Ukrainian cases demonstrate, such an overt EU identity is only possible because of the construction of an Other who exists in the same quotidian space, shops in the same supermarkets, works at the next desk, lives in the same homes, but who is symbolically castigated as the ausländer.This reappearance of nationalist narratives and myths in European politics, mediated through symbolism, is not something from which the EU is exempt. This resurrection of nationalism is a troubling phenomenon, as the history of European nationalism cannot but fail to remind us that increased nationalism results in the solidification and fortification of borders, the relegation in status or expulsion of those whose ethnicity differs, and the inevitable rise of violent ethnic politics. All of which, uncoincidentally, are visible in the present as the appeal of European supranationalism is challenged and eroded on a daily basis by competing imaginations not merely of integration, but of normative integration. Not just what the EU is and who does belong and how the EU actually operates, but what the EU should be, who shouldn’t belong, and how the EU ought to operate. It is arguably necessary to encourage a greater sense of transnational identity which, crucially, will not supplant national identity but co-​exist alongside it. This combination of national and transnational identity is difficult, but not far-fetched. As an example, all members of the EU are members of the UN, and most are also members of NATO, alongside a variety of other intergovernmental entities. But overt, symbolic objections to membership in the UN, NATO, FIFA, the Eurovision Song Contest, the Red Cross, or the International Postal and Telegraph Union are absent from the domestic and collective political discourses of European states. Processes of differentiated integration and disintegration of these organizations are not a matter of public debate, but occur in the background of consciousness,

127

Russell Foster

visible only on scheduled, infrequent occasions –​or not visible at all. And these clearly undemocratic activities go without comment.The EEC was arguably in this category, as was the EU in the 1990s.Yet through a combination of factors exogenous (the rise of digital communications) and indigenous (the EU’s institutions expanding into everyday life), the EU occupies a significantly greater position in the minds of Europhobes, Eurosceptics, Europhiles, and Euroalternatives, with descriptive and normative narratives of integration and disintegration manifest not in long treaties or (as the British demonstrated) scrutiny of legislation, but through feelings and symbols. In De statu imperii Germanici (1667), Samuel Pufendorf argued that the Holy Roman Empire was destined to fail as it could not reconcile its deficit of simultaneity; a debit of institutional power with a credit of identity –​an ungleichzeitigskeit des gleichzeitigen. EU elites have sought to pursue an artificial Sonderweg, an artificial path, characterized not by nationalism but a trans-​or meta-​nationalism, seeking to transcend parochialism in pursuit of a higher ideal. Yet this policy, of expressing identity via symbols, is indistinguishable from the nationalist narratives which the post-​war European project rejects. EU symbols are not only iconographies of integration, they are symbols of differentiated disintegration, simultaneous symbols of what ‘Europe’ was, what it is, and what, in the unity of its citizens’ diversity, it ought to be.

Notes 1 Technically, there is a slight semiological difference between ‘symbols’ and ‘icons’/​ ’iconographies’ (Whitehead 1927). However, this is of real relevance only to art historians, and this chapter uses the terms interchangeably. 2 Consider the proliferation of physical performances of EU symbolism in Greece in 2013–​2014, Catalonia in 2017, Scotland since 2014 and the United Kingdom since 2016. The United Kingdom in particular has experienced a startling symbolic transformation in an extremely short time, from a country which was at best reluctantly EU-​ropean and in which EU symbols were very rarely encountered, into a country in which EU flags, EU facepaint, EU clothes and a vast range of EU-​themed material culture from duvet sets to biscuits to contact lenses, is visible on a daily basis. As I argued above, this symbolism does not transcend nationalist tropes of exclusion but replicates the same themes of rejection and imagined community. 3 Indeed, this is acknowledged by EU policymakers who consequently call for greater symbolic communication between EU institutions and citizens in order to prevent EU citizens from voting according to emotions rather than rational thought. See Poettering et al. (2006). 4 The author’s ongoing project to redesign Euro currency (www.youreurope.org) with symbolism that better reflects what it means to be ‘European’ today, attracts significantly earthier criticisms on post-​ Brexit British social media. 5 Olaf Cramme (2009: 54) takes this further, arguing that ‘appealing to a culturally defined European identity would be like pouring oil on a fire’. Elite-​driven attempts to construct a European identity symbolically stress that ‘we’ are essentially the same. Yet this does little, besides antagonising vocal Eurosceptics, as diversity is ignored in favour of imagined histories of unity. As Pope Francis stated in his 2014 address to the European Parliament, unity is not the same as uniformity. If a European identity does exist, it is defined not by unity but by diversity, which current symbols do not convey. 6 Sonntag (2011: 122–​124) demonstrates that it is not possible to manufacture emotional support by creating symbols behind closed doors, and then expecting people to identify with this ‘visual overload’ in a rational manner. 7 As an example: the EU’s flag was intended to represent a peaceful European unification project.Yet the same flag is used in the campaign literature of Eurosceptic parties to represent their vision of the EU as a corrupt bureaucracy; is (and was, in the 2019 General Election) used by British Remainers/​Rejoiners to reject democracy (however ill-​advised and poorly informed the 2016 democratic vote was) and cancel Brexit; and is used by the Maidan revolutionaries to represent their vision of the EU as a project worth fighting and dying (and killing) for.The intentions of a symbol’s creator can bear little resemblance to the interpretations of those who use the symbols –​and it is this latter use which makes and remakes European identity.

128

European identities

8 In 2004, the Treaty Establishing a Constitution for Europe, Article I-​8, proposed legal status for the Union’s flag, anthem, motto, holiday, and currency iconography. The reason given was that symbols ‘would give Europe a soul and would help citizens to better identify with the Union’. Following the treaty’s rejection, attempts to legalize symbols were omitted from the 2007 Lisbon Treaty due to fears that legal symbols would erode national identity and signify a lessening of national sovereignty. 9 Curiously, the EU’s official stance on its (unofficial) symbols is that they are not symbols of just the EU, but of the whole of ‘Europe’. This metonymic synecdoche, through which the EU symbolically lays claim to the entire Western Eurasian landmass, raises the question of where ‘Europe’ is and who is and is not (and who should and should not) be considered ‘European’, and can be interpreted as an imperial belief in the EU’s manifest destiny (Foster 2015).

Bibliography Abdullah, R. and Hübner, R. (2006) Pictograms, Icons & Signs: A Guide to Information Graphics. London: Thames & Hudson. af Malmborg, M. and Stråth, B. (eds.) (2002) The Meaning of Europe. Oxford and New York, NY: Berg. Anderson, B. (1991) Imagined Communities: Reflections on the Origin and Spread of Nationalism. London: Verso. Beck, U. and Grande, E. (2008) Cosmopolitan Europe [trans. Ciaran Cronin]. Malden, MA: Polity. Bernays, E. (1952) Public Relations. Norman, OK: University of Oklahoma Press. Bottici, C. and Challand, B. (2013) Imagining Europe: Myth, Memory and Identity. Cambridge: Cambridge University Press. Brague, R. (2002) Eccentric Culture: A Theory of Western Civilization. South Bend, IN: St Augustine’s Press. Braudel, F. (1992) Civilization and Capitalism Vol. I: The Structures of Everyday Life. Berkeley, CA: University of California Press. Brunn, S. (2000) Stamps as iconography: Celebrating the independence of new European and Central Asian states, GeoJournal, 52, pp. 315–​323. doi: 10.1023/​A:1014307914500. Bruter, M. (2009) Time bomb? The dynamic effect of news and symbols on the political identity of European citizens, Comparative Political Studies, 42(12), pp. 1498–​1536. doi: 10.1177/​0010414009332465. Cafruny, A. and Lankowski, C. (eds.) (1997) Europe’s Ambiguous Unity: Conflict and Consensus in the Post-​ Maastricht Era. Boulder, CO: Lynne Reiner. Calligaro, O. (2013) Negotiating Europe: EU Promotion of Europeanness since the 1950s. Basingstoke: Palgrave Macmillan. Carta, C. and Morin, J. (2013) Struggling over meanings: Discourses on the EU’s international presence, Cooperation and Conflict, 49(3), pp. 1–​20. doi: 10.1177/​0010836713494995. Cassirer, E. (1953) The Philosophy of Symbolic Forms: Vol. I Language. New Haven, CT: Yale University Press. Cheles, L. and Sponza, L. (2001) The Art of Persuasion: Political Communication in Italy from 1945 to the 1990s. Manchester: Manchester University Press. Cloet, P., Legué, B. and Martel, K. (2013) United in diversity: Anthems and flags of the European Union, Study for Jacques Delors Institute. Available at: www.eng.notre-​europe.eu/​011-​17265-​United-​in-​diversity-​ Anthems-​and-​flags-​of-​the-​European-​Union.htm (accessed 26 April 2021). Cram, L. (2001) Imagining the Union: A Case of Banal Europeanism, in Wallace, H. (ed.) Whose Europe? Interlocking Dimensions of Integration. London: Palgrave Macmillan. pp. 233–​246. Cram, L. (2012) Does the EU need a navel? Implicit and explicit identification with the European Union, Journal of Common Market Studies, 50(1), pp. 71–​86. doi: 10.1111/​j.1468-​5965.2011.02207.x. Cramme, O. (ed.) (2009) Rescuing the European Project: EU Legitimacy, Governance and Security. London: Policy Network Press. Cramme, O. (2016) Europe Entrapped. London and New York, NY: Wiley. CVCE (2021) The Symbols of the European Union [orig.1951]. Available at: www.cvce.eu/​en/​collections/​ unit-​content/​-​/​unit/​552520d4-​f424-​4c9f-​922e-​da95b0667953 (accessed 26 April 2021). Debord, G. (1967) The Society of the Spectacle. New York, NY: Zone. Debord, G. (1990) Comments on the Society of the Spectacle. London:Verso. Deibert, R. (2011) Parchment, Printing and Hypermedia: Communication and World Order Transformation. New York, NY: Columbia University Press. Drace-​Francis, A. (2013) European Identity: A Historical Reader. Basingstoke: Palgrave Macmillan.

129

Russell Foster

Edeleman, M. (1964) The Symbolic Uses of Politics. Urbana, IL: University of Illinois Press. European Parliament (2013) Regulation (EU) No. 1291/​2013 of the European Parliament and of the Council, of 11 December 2013. European Union (2004) Conference of the Representatives of the Governments of the Member States, Treaty Establishing a Constitution for Europe, Brussels, 29 October 2004. CIG 87/​2/​04 REV 2. Firth, R. (1973) Symbols: Public and Private. London: George Allen and Unwin. Fishman, R. and Messina,A. (eds.) (2006) TheYear of the Euro. Notre Dame, IN: Notre Dame University Press. Fornäs, J. (2013) Signifying Europe. Bristol: Intellect Publishing.Foster, R. (2013) Tabulae Imperii Europaei: A Cartographic Approach to the Current Debate on the European Union as Empire, Geopolitics, 18(2), pp. 371–​402. doi: 10.1080/​14650045.2012.716466. Foster, R. (2015) Tabulae Imperii Europaei. London: Routledge. Foster, R. (2018) These are what Faustus most desires: Identity, iconography, and ‘Europe’ in the Crimea crisis. Journal of Contemporary European Research, 14(4), pp. 310–​323. doi: 10.30950/​jcer.v14i4.1039. García, L. (2017) Introduction: A narrative turn in European studies, Journal of Contemporary European Studies, 25(3), pp. 285–​290. doi: 10.1080/​14782804.2017.1348341. George, S. (1990) An Awkward Partner: Britain in the European Community. Oxford: Oxford University Press. Guardian (2018) ‘It’s not a done deal’: inside the battle to stop Brexit. 28 April 2018. https://​www.theguardian. com/​politics/​2018/​apr/​28/​brexit-​not-​a-​done-​deal-​battle-​to-​stay-​in-​eu-​second-​referendum Haas, E. (1958) The Uniting of Europe. Stanford, CN: Stanford University Press. Haas, E. (1963) Beyond the Nation-​State. Berkeley, CA: University of California Press. Habermas, J. (2013) The Crisis of the European Union: A Response [trans. Ciaran Cronin]. Cambridge and Malden, MA: Polity. Hooghe, L. and Marks, G. (2008) A postfunctionalist theory of European integration: From permissive consensus to restraining dissensus, British Journal of Politics and Society, 39(1), pp. 1–​23. doi: 10.1017/​ S0007123408000409. Hooghe, L. and Marks, G. (2019) Grand theories of European integration in the twenty-​first century, European Journal of Public Policy, 26(8), pp. 1113–​1133. doi: 10.1080/​13501763.2019.1569711. Hix, S. (2008) What’s Wrong with the European Union and How to Fix It, Cambridge and Malden, MA: Polity. Hobsbawm, E. and Ranger,T. (eds.) (1983) The Invention of Tradition. Cambridge: Cambridge University Press. Kertzer, D. (1996) Politics and Symbols: The Italian Communist Party and the Fall of Communism. New Haven, CT:Yale University Press. Kosselleck, R. (2002) The Practice of Conceptual History:Timing History, Spacing Concepts [trans. Todd Presner]. Stanford, CA: Stanford University Press. Kuus, M. (2007) Geopolitics Reframed: Security and Identity in Europe’s Eastern Enlargement. New York, NY: Palgrave Macmillan. Kuus, M. (2013) Foreign Policy and Ethnography: A Sceptical Intervention, Geopolitics, 18, pp. 115–​131. doi: 10.1080/​14650045.2012.706759. Lefebrve, H. (1991) The Critique of Everyday Life [trans. John Moore]. London:Verso. Leith, S. (2008) European Union symbols are a waste of time and public money, The Telegraph, 13 September 2008. Available at: www.telegraph.co.uk/​comment/​columnists/​samleith/​3562125/​European-​Union-​ symbols-​are-​a-​waste-​of-​time-​and-​public-​money.html (accessed 26 April 2020). Lüdtke, A. (1995) Histoire du Quotidien. Paris: Maison des Sciences de l’Homme. McCormick, J. (2007) Why Europe Matters. Basingstoke: Palgrave. McCormick, J. (2013) The European Superpower. Basingstoke: Palgrave. Miller,V. (2008) The European Union Symbols and their Adoption by the European Parliament, Report for Library of the House of Commons International Affairs and Defence Section. SN/​IA/​4874. Moisio, S., Bachmann, V., Bialasiewicz, L., dell’Agnese, E., Dittmer, J. and Mamadouh, V. (2012) Mapping the political geographies of Europeanization: National discourses, external perceptions and the question of popular culture, Progress in Human Geography, 37(6), pp. 737–​761. doi: 10.1177/​0309132512472093. Moravcsik, A. (2019) Preferences, power and institutions in 21st-​century Europe, Journal of Common Market Studies, 56(7), pp. 1648–​1674. doi: 10.1111/​jcms.12804. Mudde, C. (2013) Three decades of populist radical right parties in Western Europe: So what?, European Journal of Politics Research (52:1), pp. 1–​19. Müller, M. (2008) Reconsidering the concept of discourse for the field of critical geopolitics: Towards discourse as language and practice, Political Geography, 27(3), pp. 322–​338. doi: 10.1016/​j.polgeo.2007.12.003. Nozedar, A. (2010) The Illustrated Signs and Symbols Sourcebook. London: Sterling Publishing. 130

European identities

O’Shaughnessy, N. (2004) Politics and Propaganda: Weapons of Mass Seduction. Manchester: Manchester University Press.Painter, J. (2006) Prosaic geographies of stateness, Political Geography, 25(7), pp. 752–​774. doi: 10.1016/​j.polgeo.2006.07.004. Packard,V. (1980) The Hidden Persuaders. Brooklyn, NY: IG Publishing. Piercy, J. (2013) Symbols: A Universal Language. London: Michael O’Mara Books. Poettering, H. (ed.) (2006) Our Vision of Europe in 2020. Brussels: European Parliament. Poyatos, F. (2002) Nonverbal Communication across Disciplines Vol. III: Narrative Literature, Theater, Cinema, Translation. Amsterdam: John Benjamins. Pufendorf, S. (1667). De statu imperii Germanici. Geneva: Apud Petrum Columensium. Raento, P., Hämäläinen, A., Ikonen, H. and Mikkonen, S. (2004) Striking stories: A political geography of euro coinage, Political Geography, 23(6), pp. 929–​956. doi: 10.1016/​j.polgeo.2004.04.007. Raento, P. and Brunn, S. (2005) Visualizing Finland: Postage stamps as political messengers, Geografiska Annaler, 87, pp. 145–​163. doi: 10.1111/​j.0435-​3684.2005.00188.x. Rose, G. (2012) Visual Methodologies. London: Sage. Sassatelli, M. (2017) ‘Europe in your Pocket’: Narratives of identity in Euro iconography, Journal of Contemporary European Studies, 25(3), pp. 354–​366. doi: 10.1080/​14782804.2017.1348338. Shipman, T. (2016) All Out War: The Full Story of How Brexit Sank Britain’s Political Class. London: William Collins. Sonntag, A. (2011) False Expectations: The Counterproductive Side Effects of the EU’s Use of Political Symbols, in Lucarelli, S., Cerutti, F. and Schmidt, V. (eds.) Debating Political Identity and Legitimacy in the European Union. Abingdon: Routledge., pp. 115–​130. Sorlin, P. (1994) Mass Media. London and New York, NY: Routledge. Stråth, B. (2002) A European identity: To the historical limits of a concept, European Journal of Social Theory, 5(4), pp. 387–​401. doi: 10.1177/​136843102760513965. Street, J. (2011) Mass Media, Politics and Democracy. Basingstoke: Palgrave. Theiler, T. (2005) Political Symbolism and European Integration. Manchester: Manchester University Press. The Economist (2021) Vaccine protectionism endangers the global fight against Covid-​19. February 1st 2021. https://​www.economist.com/​graphic-​detail/​2021/​02/​01/​vaccine-​protectionism-​endangersthe-​global-​fight-​against-​covid-​19 Thrift, N. (2004) Intensities of Feeling:Towards a Spatial Politics of Affect, Geografiska Annaler, 86, pp. 57–​78. doi: 10.1111/​j.0435-​3684.2004.00154.x. UK in a Changing Europe (2020). Covid-​19 and the ‘rally-​round-​the-​flag’ effect. https://​ukandeu.ac.uk/​ covid-​19-​and-​the-​rally-​round-​the-​flag-​effect/​ van Gent, W., Mamadouh, V. and van der Wusten, H. (2013) Political reactions to the euro crisis: Cross-​ national variations and rescaling issues in elections and popular protests, Eurasian Geography and Economic, 54(2), pp. 135–​161. doi: 10.1080/​15387216.2013.811362. Whitehead, A. (1927) Symbolism: Its Meaning and Effect. New York, NY: Fordham University Press. Whittick, A. (1960) Symbols, Signs, and Their Meaning. London: Leonard Hill. Wintle, M. (2009) The Image of Europe. Cambridge: Cambridge University Press. Wintle, M. (2011a) Personifying the Past: National and European History in the Fine and Applied Arts in the Age of Nationalism, in Berger, S., Eriksonas, L. and Mycock, A. (eds.) Narrating the Nation: Representations in History, Media and the Arts. New York, NY: Berghahn. pp. 222–​245. Wintle, M. (2011b) Ideals, Identity and War: The Idea of Europe, 1939–​1970, in Wintle, M. and Spiering, M. (eds.) European Identity and the Second World War. London: Palgrave Macmillan. pp. 1–​20.

131

9 Differentiation and segmentation Jozef Bátora and John Erik Fossum

Introduction Since its inception in the late 1950s, the European Union (EU) has seen a dramatic rise in integration. The crises of the last decade, and no less importantly, the UK’s decision to leave the EU, have brought up new questions about the EU’s ability to sustain its historical integration momentum. At the same time, and even if accounts vary, there is little doubt that the EU that emerged from the Eurozone and refugee crises had integrated further (Bauer and Trondal 2015). Nevertheless, the tensions and divisions that the Eurozone and refugee crises highlighted testify to the need to look closer at the type of integration involved.1 The EU has experienced serious difficulties in pulling together in a unified manner. Hence, the focus is on differentiated integration. Even if the EU continues to integrate, there is less assurance than before that all states will integrate at the same pace, or even that all will eventually reach the same destination. A more differentiated EU raises a number of important questions about the nature and character of the EU as a political system, and the kinds of processes and mechanisms that drive its development. There is also terminological ambiguity involved, in that the literature uses the two terms differentiated integration and differentiation almost interchangeably, even if these two terms can refer to different phenomena and dynamics (there are both zones of overlap and divergence between them2.) The main purpose of this chapter is to show that these terms do not adequately account for the EU as a political system, nor do they contain sufficient explanatory power to account for the EU’s nature and development, especially during the last decade or so. Stating that in no way refutes the fact that the EU is a distinctly differentiated political system nor that it has become more differentiated in the last decade.The key point is that neither differentiated integration nor differentiation says much about the positive character or the distinguishing features of the EU as a political system. These terms are blunt instruments for properly capturing the EU’s defining features. For more precision, we should consider the EU as a segmented political system (Bátora and Fossum 2020).This term provides a more apt and precise characterization of the EU as a political system; it also introduces the notion of segmentation as a distinct dynamic. Indeed, as Héritier (2021) acknowledges, the formation of a segmented political order represents a distinct path in the formation of the EU qua polity. 132

DOI: 10.4324/9780429054136-10

Differentiation and segmentation

The first part of the chapter provides a brief overview of what in the EU literature is referred to as differentiated integration and differentiation. The second part defines segmentation and segmented political order and shows how and in what sense we may talk of the EU as a segmented political order. The third and final part spells out the most important similarities and differences between on the one hand differentiation/​differentiated integration, and on the other segmentation/​segmented political order.

The EU –​a distinctly differentiated polity In order to understand the distinctive features of the EU as a differentiated political system, we need to keep in mind that all modern political systems are differentiated –​along functional, territorial and hierarchical lines. The EU is distinctly differentiated, and the key term that is used to depict that is differentiated integration. In the EU context, differentiated integration refers to specific features of the EU integration process, EU structural features; and more specific EU rules and policies. Differentiated integration thus manifests itself in legal terms: in primary and secondary law; in institutional structures and constitutional arrangements; and in the use and scope of application of various types of policy instruments. Numerous concepts are used to describe the process and its (anticipated) result, such as multiple speeds, variable geometry or Europe à la carte, Core Europe or the Europe of Concentric Circles, and two-​or multi-​speed Europe, respectively.3 Further, the literature tracks member states’ exemptions, exceptions, opt-​ins and opt-​outs from EU institutional arrangements, laws and policies.4 A broad range of explanations has been set forth to account for the nature and causes of differentiated integration. Analysts have drawn on neo-​functional, (new) intergovernmental, institutionalist and constructivist approaches to explain processes of differentiated integration.5 It follows from the above that differentiated integration has been attributed with a variety of different things. It has even been interpreted and evaluated differently from one academic discipline to another (Holzinger and Schimmelfennig 2012: 293). Some analysts note that: “the complexity and plurality to studying DI [differentiated integration] is justified by the nature of DI itself ” (Leruth and Lord 2015: 761). The term is used to describe differences along vertical (across levels of governing), and horizontal (encompassing functional and territorial) dimensions, as well as along temporal lines (short, medium or long term).6 Nevertheless, it is clear that different forms of differentiated integration share a minimum common denominator, namely that within the scope of EU competences, not all member states are subject to the same uniform EU rules (Avbelj 2013: 193). As noted, differentiated integration is used to describe both legal-​and political-​institutional developments in the EU context. Structurally speaking, the main moment of institutional differentiation occurred with the Maastricht Treaty, which introduced an EU with three pillars. It represented a “major advance in the scope of activity of the EC/​EU [that] had to be paid for by abandoning the notion of a unitary Community legal order that applied evenly in all its Member States” (De Witte 2017: 11). Even if the pillars were formally abolished in the Lisbon Treaty, significant divisions in the EU system of governing remain (Fabbrini 2013, 2015; Fossum and Menéndez 2011). Studies have shown that the EU has become more differentiated over time, in response to enlargements, various forms of crises and various forms of opposition to integration. Both widening and deepening of integration are associated with differentiation.7 As EU integration has increasingly come to touch on core state powers (Genschel and Jachtenfuchs 2014), state interests and concerns have become activated, and differentiated integration has become a means 133

Jozef Bátora and John Erik Fossum

for handling (or bypassing) this. Post-​Maastricht the Union does not just exercise its own competence, it coordinates core state powers in areas of national competence (Heidbreder 2014). But, to combine the two tasks of exercising its own competence and coordinating national competence, the Union has had to differentiate both its own decision-​making methods and patterns of member state participation in particular policies. The effect is for the EU to incorporate two institutional systems or principles, respectively, the supranational Community System and the more intergovernmental Union System (Bátora and Fossum 2020; Bickerton et al. 2015; Fabbrini 2015, 2018). No other solution would have been compatible with maintaining the integrity of systems at the two levels. Nevertheless, Brexit is the starkest reminder thus far to the effect that the EU faces important disintegrative pressures; hence a more recent focus on (differentiated) disintegration.8 That is unto itself an important reason for distinguishing between differentiation and differentiated integration.9 Differentiation is not tied to a specific direction of development, whereas differentiated integration is about specific features of a process of coming together. Differentiation is the more general of the two terms in that it refers to the overall composition of differentiation in modern societies, from Talcott Parson’s (1951) notion of role differentiation as the point of departure, to the state formation literature’s preoccupation with differentiation along territorial, functional, social, economic and political lines (Rokkan 1975). These themes are also reflected in the EU literature (see, for instance, Bartolini (2005) and Leuffen et al. (2013)). This brief overview shows that differentiation and differentiated integration both overlap and refer to different things. In principle, it is possible to say that the phenomena captured under the rubric differentiated integration are all about differentiation but with two important provisos. One problem would be a loss of precision because differentiation does not specify the direction of change. In other words, a more differentiated system can mean a system where more of the integration is differentiated. Or, it can mean a system with pulls in both integrationist and dis-​integrationist directions. Or, it can refer to a move in a dis-​integrationist direction (akin to fragmentation). Further, if we focus mainly on differentiated integration, we need to be clear on whether we talk about this in structure or process terms. The EU’s distinct pattern of polity differentiation has bearings on the process of differentiated integration (Fossum 2019). At the same time, since the EU is structurally speaking differentially integrated, that shapes the manner in which differentiation –​along territorial, functional and hierarchical lines –​unfolds in the EU. We can only properly understand these dynamics insofar as we keep structure and process apart.

Segment, segmentation and segmented political order The main argument of this chapter is that we need to complement the debate on EU differentiated integration and differentiation with terms that help us to specify the character of the EU as a political system and that also shed further light on the distinct dynamics that propel the EU’s development.The relevant terms are segment, segmentation and segmented political order. “Segment” is a term that is familiar to a range of academic disciplines (sociology, economics and political science; Olsen (1988); Picot (2012)). Segments are normally discussed as specific traits or features of political systems, within certain sectors, at meso-​level. They are stabilized constellations of actors from government, parliament, regulatory agencies, NGOs, regional authorities and the private sector involved in recurrent practices of patterned information exchange and participation in policy formation (Christensen and Egeberg 1979).10 Under certain structural and institutional conditions, segments can solidify, and a political system where such structural elements and mechanisms are pervasive can be characterized as a segmented political system. Segments then become constitutive features of the political system. They become macro-​political phenomena 134

Differentiation and segmentation

in the sense that the logic of segmentation defines the structural make-​up: the pattern and composition of the political system’s functional, territorial and hierarchical dimensions. We can identify six traits that give meaning and shape to segmented political system (Bátora and Fossum 2020). Trait 1 is an entrenched set of ideas and ideologies that limits the search for alternatives and fosters cognitive closure. The systematic selection bias that we associate with a segment is a form of closure: actors within a segment systematically exclude alternative ways of understanding problems and shut out alternative types of expertise, alternative policy instruments and alternative solutions. Trait 2 refers to a particular (limited, narrow or strongly biased and with limited scope for switching between policy instruments) configuration of policy instruments that contributes to lock-​in forms of cognitive closure. Such a configuration typically makes up a distinct policy style (for this term, see Richardson 1982). Trait 3 refers to a set of organizational and procedural arrangements that sustains the segmental logic(s). The structure is imbued with a built-​in bias in the sense that it systematically organizes in certain ways of understanding issues, and systematically organizes out alternatives. Trait 4 refers to the fact that a segmented political order is fundamentally constrained. As such, segmented political orders resemble those pre-​modern political orders that lacked the magnitude, breadth of scope and type(s) of resources that mark modern democratic states (Tilly 1993). Trait 5 refers to external dependence. A segmented political order lacks the size, scope and access to resources that are available to the modern state and is therefore more dependent on factors and forces in the external world, such as other states and (global) markets, for instance. Finally, in accordance with Trait 6, a segmented political order is characterized by endogenous biases structuring policymaking. It has weak or deficient de-​segmenting arrangements, in other words arrangements that pry open segments and undermine built-​in biases and patterns of path dependence. Of particular importance are those arrangements that ensure consideration of a plurality of views, values and considerations (typically associated with arrangements that ensure democracy and transparency). The next section provides a brief overview of how the traits of a segmented political order manifest themselves in the EU.

The EU –​a fledgling segmented political order Segmented political orders may come about through original design –​with a built-​in bias from the outset –​or they may experience increased segmentation due to crises, upsets and ruptures. The EU contains a mixture of these: it was explicitly barred from becoming a state through clear constraints on integration in “core state powers” such as military defence, police and taxation.11 The main drivers of integration were executives and experts, and democratization was a matter of “catching up” with these (Crum and Fossum 2013). Nevertheless, the EU has, over time, developed important de-​segmenting arrangements.The poly-​crises of the last decade or so have led to an EU mutation that has, on the one hand, reinforced built-​in biases in a segmenting direction and, on the other, weakened and/​or sidelined de-​segmenting bodies/​arrangements, notably parliaments. In the following, we spell out what this entails with explicit reference to the six traits of segmented political order that we outlined above.

Trait 1: ideas and ideologies The first type of cognitive bias and ideological component has roots in the EU’s strong market orientation or what has also been referred to as “internal market rationality”.12 The EU, as noted above, and when compared to a state, has been a biased or lop-​sided system from its inception. Nevertheless, the strong market-​rational imprint has developed and solidified over 135

Jozef Bátora and John Erik Fossum

time, from the late 1970s (Menéndez 2013). The EU’s legal constitutional development has entrenched that (Grimm 2015; Isiksel 2016). What we see in connection with the EU’s orientation to the Eurozone crisis is a form of “segmental closure”, in that the EU’s crisis response was motivated by particular economic doctrines that systematically selected in certain crisis action handling repertoires, notably of an ordoliberal or, for some, neoliberal bent and that simultaneously systematically excluded (after the initial fiscal stimulus) Keynesian-​type action repertoires where “intellectual convergence around a flawed set of economic ideas continuously blinded EU policy-​makers to dangerous divergences in the real economy” (Tranøy and Schwartz 2020: 47–​48). The crises also brought forth a second type of bias associated with securitization or a propensity to think of issues and concerns as matters of security first and foremost. A securitization logic invests in this a specific element of fear and uncertainty. As Jef Huysmans has noted, “security practices … turn an issue like migration into a security problem by mobilizing specific institutions and practices”. (2000: 757) This logic has become far more pronounced in connection with the so-​called refugee crisis, where the refugee crisis was redefined from an initial focus on the humanitarian aspects of refugee suffering and the need for solidarity, to the concerns of states with borders and terrorist threats (Lazaridis and Wadia 2015).13

Trait 2: policy instruments and policy style The EU’s core realm of action lies in the field of regulation (Majone 1994), especially in the field of internal market.That in turn yields a distinct EU policy style, with a limited repertoire of policy instruments, and when coupled with the EU’s limited fiscal capacity, provides little room for flexibility in instrument choice (Genschel and Jachtenfuchs 2018). The limited range and built-​in biases in the EU’s policy measures have bearings on the EU’s institutional structure.This has at least since Theodore Lowi’s seminal work been an important insight. A recent important reminder is found in Orren and Skowronek (2017). It is also interesting to consider this in light of what Adam et al. (2019) have referred to as “the policy accumulation trap”, in other words, how responsiveness leads to an accumulation of policy and “renders policy content increasingly complex, which crowds out policy substance from public debates and leads to an increasingly unhealthy prioritization of politics over policy. Secondly, policy accumulation comes with aggravating implementation deficits, as it produces administrative backlogs and incentivizes selective implementation. Finally, policy accumulation undermines the pursuit of evidence-​based public policy, because it threatens our ability to evaluate the increasingly complex interactions within growing policy mixes”(Adam et al 2019: i). In this context, we may hypothesize that the greater or the more extensive the policy responsiveness in a context marked by a limited range of policy instruments and clear constraints on available resources and action capability the more troublesome the policy accumulation trap.

Trait 3: institutional and structural arrangements The EU’s institutional structure lends itself to the entrenchment of the market bias. An important reason for why it makes sense to talk about present-​day EU as a segmented political order stems from the fact that the two segmental logics correspond roughly to the two tracks of the EU’s institutional structure. We find one track in the supranational Community component, which we may, broadly speaking, refer to as supranational (expert-​based) regulation. The other is in the intergovernmental component (mainly associated with the Council of the EU and the European Council) and relies on intergovernmental interaction and bargaining. These are institutionally 136

Differentiation and segmentation

and functionally divided. Along each of these tracks, there was a process of segmental closure during the crises of the last decade. Institutionally speaking, the EU combines horizontal separation (of functional spheres: the supranational Community versus the Intergovernmental realm), with vertical fusion (of levels) (Fossum 2020). This combination represents a unique form of differentiation, in the sense that functional spheres are horizontally separated and operated by different institutional arrangements, one with a strong supranational tenor, the other with a strong intergovernmental tenor (the two arrangements also somewhat overlap). This is combined with a strong vertical fusion of levels; in other words, the EU institutions are tightly interwoven with the member states. This distinctive institutional configuration fosters segmentation by preventing or constraining horizontal coordination, market correction and fiscal stabilization at EU level, because, as Fritz Scharpf (2010) has underlined, the EU decision procedures in the institutions bent on market-​making are majoritarian, whereas the member states can exercise a veto in those institutions bent on market correction, and fiscal and tax harmonization. On the other hand, the structure fosters segmentation by implicating national officials in EU-​level decision-​making, and fusing levels of governing and administration by the development of stabilized constellations connecting participants, problem definitions, solutions and choice opportunities across a diverse group of actors both on the national and on the EU level. The combination of the specific institutional separation of tasks combined with the bringing together of executive and administrative officials across levels of governing are structural features of the EU that are conducive to segmentation, not least because these processes are removed from proper parliamentary oversight and control. The EU’s structural make-​up contains traits that are conducive to segmentation through the particular biases that they contain. The important point is that the Eurozone crisis has clearly re-​ enforced segmentation. The crisis has re-​enforced the ties among the components that make up the distinct socio-​economic policy stance that mark the Eurozone. Fiscal policy is increasingly coordinated across levels (and as such reflects an ever-​g reater pattern of fusion).14 In this situation, the scope for governments to pursue other policies is limited. Further, in dealing with its many crises and challenges, the EU has often resorted to intergovernmental means, with the European Council playing a central role through measures such as intergovernmental treaties (see the Treaty on Stability, Coordination and Governance in the Economic and Monetary Union) and informal intergovernmental bargains (notably between Germany and France). It should be noted that this does not necessarily entail that the supranational institutions have been sidelined or robbed of tasks. Aspects of the EU’s supranational components have been strengthened, notably in the areas of macroeconomic policy and banking regulation, in which the role of the Commission and the European Central Bank (ECB) has been considerably enhanced (Dehousse 2015). Nevertheless, in overall terms, what appears is that the more informal (less legally regulated and constrained) intergovernmental approach that we witnessed in the crisis handling may reprogramme the supranational structure and make it more attentive to the interests and views of certain core governments, especially Germany and its conception of how to deal with the crisis through a tough fiscal austerity policy with a neoliberal/​ordoliberal orientation. This supports the notion of crisis-​reinforced segmentation.

Trait 4: constraints and limits on capacity and capability The fourth trait that we associate with a segmented political order pertains to constraints. The fact that an entity is constrained does not –​in itself –​make it into a segmented political order. Only certain types of constraints and certain types of circumstances will foster segmentation, and both factors (types of constraints and specific circumstances) are relevant for the EU. The 137

Jozef Bátora and John Erik Fossum

EU was, from the very outset, prevented from becoming a state by the strong built-​in constraints on core state powers (military, diplomacy, policy, tax and fiscal). This, over time, has produced a curious paradox, wherein the EU has seen very little of the capacity build-​up that we find in all states. The EU is nevertheless a vital force in Europeanizing the Member (and affiliated) States. The EU’s development is therefore a powerful story of circumventing such constraints, but with strong built-​in biases. Consider the ECB’s “monetarization of fiscal policy at the EU-​ level”,15 which was attacked in the ruling of May 5, 2020 of the Second Senate of the German Constitutional Court on the ECB’s Public Sector Purchase Programme (PSPP). The Court ruling in turn opens further scope for national court challenges to EU measures, which was quickly noted by the Polish government after the German court ruling. This fits into the broader context of an EU regulatory policy style and political and administrative fusion, both of which leave the member states with a central role in effectuating EU action. This is most pronounced in the first track or the realm of the community system (which sees very little differentiation) and less so in the second track or the intergovernmental system where there is limited supranational integration, a lot of differentiation, weak democratic-​constitutional controls, and where statespeople play a predominant role. The second sovereignty-​cum-​securitization segment, which is institutionally situated in the intergovernmental track, is therefore quite differently constituted and constrained from the first single-​market-​based segment. The EU is constrained in a material sense; it is also constrained in an ideational sense, in two respects. First is that the member states as the EU’s constituent units have never agreed on a common normative script for the EU; in other words, they have never agreed on what type of entity it really is and what type of entity it should be. This would have been unproblematic if the EU had been an ordinary international organization, but it amounts to a normative deficit when the member states have conferred so many tasks on the EU. The other constraint is that the EU is institutionally barred from playing a significant socializing role; hence, confronts serious structural obstacles that work against generating the depth of citizens’ attachment that nation-​states can generate. In this respect, the EU is very different from all nation-​states, federal or not. These material and ideational constraints leave the EU highly vulnerable to criticism, which right-​wing Europhobes are exploiting to the full. This is very explicit in Hungary and Poland, which have embarked on a major onslaught on constitutional democracy, with significant ramifications for the EU in general. In structural terms, the combination of Trait 3 (institutions and procedures) and Trait 4 (limited resources) suggests that segments may vary considerably in terms of institutional arrangements and structures, as well as resources –​material and immaterial –​supporting them. Some segments can consist of tightly coupled institutional arrangements, while others consist of loosely coupled institutional arrangements.The latter is evident in the EU’s increased reliance on interstitial organizations –​organizations that emerge in the interstices between established institutional fields, tapping into the personnel, financial, legal and legitimacy resources of organizations belonging to different institutional fields, and re-​combining these to form new patterns of action and new organizational types (Bátora 2013, 2020, 2021a; Korff et al. 2015).16 Interstitial organizations emerge in situations in which there is a need for problem-​solving under strong institutional, legal, political and resource constraints, and interstitial organizations were a frequently resorted-​to EU crisis handling measure. Interstitial organizations or bodies are not a necessary manifestation of a segmented political order, but the relationship has thus far not been specified. In situations in which established institutional forms either fail to generate action capacity or are hindered from doing so, interstitial organizations provide action capacity through organizing across traditional institutional boundaries. This happens via creative re-​combination and transposition of norms, rules and 138

Differentiation and segmentation

procedures from across different institutional domains, which enable coordinated action and –​ as a bi-​product –​the generation of various kinds of ambiguities and uncertainties. The critical issue for segmentation pertains to whether interstitial organizations generate or sustain forms of cognitive and/​or decisional bias. Interstitial organizations are good illustrations of the sometimes very complex manner in which cognitive factors –​ideas and ideologies –​are institutionally entrenched.

Trait 5: dependence on external factors and patterns of external vulnerability The fifth trait that we may associate with a segmented political system refers to vulnerability and dependence on external factors. As was the case with Trait 4, this is not a defining feature, as such, but will, under certain circumstances, foster segmentation.With regard to the first segment, the financial crisis showed that the EU’s socio-​economic structure and built-​in imbalance –​ monetary union without an attendant fiscal union –​rendered it highly dependent on volatile international markets. This dependence on external factors was evident in the crisis responses, especially in the EU’s reliance on the International Monetary Fund (IMF) in the bailout of the EU’s member states. The central crisis handling role of the Troika (the European Commission, the ECB and the IMF) is a case in point. This type of dependence ensures that market-​based imperatives play a central role in the considerations of EU decision makers, and, as such, helps to entrench the market-​based segment, especially in so far as the criteria and conditions are, broadly speaking, aligned with a neoliberal austerity policy.17 The EU has also faced a significant external vulnerability and subsequent dependence in the other segment. The EU’s high level of vulnerability stems from the fact that it borders on such a large number of states, not least states that are weak or dysfunctional or deeply oppressive. The EU’s contrast to the United States, which has only two neighbours, and friendly ones to boot, is quite instructive. In the area of immigration and border controls, the EU’s vulnerability has been turned into structural dependence on Turkey for regulating the inflow of migrants to Europe. The EU-​ Turkey agreement has been widely criticized, and fosters segmentation in the sense that the EU has not only taken active measures to exclude migrants, but is also excluding them from European law (Spijkerboer 2017). Whether intended or not, the effect is for the securitization logic to crowd out humanitarian considerations across a broad range of institutions.

Trait 6: weak de-​segmenting arrangements The final trait of a segmented political order refers to its imbalanced nature, in the sense that the institutions and arrangements that foster and sustain segmentation are systematically more developed than those institutions and arrangements that can ensure de-​segmentation. In this regard, the EU has built-​in constraints, in the sense that its de-​segmenting features have consistently fallen short of those we find in democratic states. With regard to the European Parliament (EP), for instance, the EU’s two-​tracked structure generates a discrepancy between the EP’s realm of effective legislative and controlling action and the realm of tasks that are actually undertaken at EU level, by EU institutions. With the crises, we see a clear weakening of democratic systems of monitoring and control at all three key levels: EU, Member State and regional. Thus, those institutions that could open up and foster de-​segmentation are weakened. National parliaments have seen their fiscal sovereignty severely constrained, and the EP has not been given powers to fill the gap (Fasone 2014a,b).The EP, sidelined, at least in part, in the crisis response, has been one of the main losers, and the crisis 139

Jozef Bátora and John Erik Fossum

response has re-​enforced technocracy, in the sense that experts have obtained a freer role and are less encumbered by legal and democratic controls. An important hallmark of this system has been summed up in the new informality: far more is now settled in bargains among leaders and officials; less is subjected to proper, transparent procedures. To sum up thus far, the EU post-​crises have clear traits of a segmented political order.The two prevailing segmental logics are a marketization logic infused with a neoliberal/​ordoliberal twist propagating a rules-​based austerity mindset and a sovereignty-​cum-​securitization logic.The segmental logics are embedded in institutional arrangements that reinforce and sustain these cognitive and decision biases, through the EU’s distinct institutional configuration which combines a two-​tracked institutional structure (horizontal differentiation) with a lopsided process of the fusion of levels of governing.

Comparing/​Contrasting differentiation and segmentation The comparison of segmentation and segmented political order, on the one hand, and differentiated integration and differentiation, on the other, is complicated by at least four sets of factors. One is the relationship between process and structure. Both sets of concepts refer to distinct process dynamics and distinct structural configurations.The other complication in comparing the terms refers to the fact that differentiation and differentiated integration are not easy to relate to each other; both contain zones of convergence and divergence. The third complication refers to the fact that the terms may not be theory independent in the sense that analysts approach the terms from different theoretical traditions. The fourth complication is that the terms deal with phenomena at different levels of magnitude (polity versus policy), which in turn may affect their differences. We will deal with these aspects in the below. The teleology that marked much of the mainstream literature on European integration since the 1950s posits in various iterations the notion of an “ever closer Union” –​i.e. the idea that EU member states and their shared institutional arrangements would progressively grow more integrated and eventually encompass all parts of political governance.Yet, the reality of European integration has not corresponded with any form of statist teleology, neither in terms of process nor in terms of substance. Analysts have pointed to this deviation with reference to the notion of differentiated integration; the EU deviates in significant ways from both the functional and the territorial dimensions. Differentiated integration in its various forms has enabled participation of member-​and non-​member states in the EU’s political order and, hence, has sought to combine flexibility and difference accommodation with some modicum of stable and predictable forms of policy coordination and conduct among states in Europe. Although differentiated integration enables states with various interests and policy specifics to participate in flexible types of cooperation in institutionalized integration frameworks, a political order based on differentiated integration is –​necessarily –​highly unwieldy. If exceptions and vetoes are the rule and a condition based on which states can accept integration, political orders are prone to deadlock and –​ in crises –​prone to collapse.18 A similar logic applies to differentiation in general: the more differentiated the polity is in functional, territorial and hierarchical terms, the more unwieldy and difficult to govern it will be. Such a highly differentiated polity would exhibit value and institutional diversity and the predominance of local rationalities, with very little scope for any form of coherence. What we have seen in the last decade in the EU is that whereas it has been hit by a series of existential crises, it has not lost its steering-​and policy coordination capacity. On the contrary, in responding to crises, it has pursued a policy of fiscal restraint with dogged determination, has developed various new structural arrangements, policy instruments and initiatives 140

Differentiation and segmentation

all of which enable it to act in a concerted manner. For instance, the Eurozone crisis has been averted by a concerted effort of EU institutions and international monetary institutions such as the IMF, Eurozone member states’ finance ministries, finance ministries of non-​Eurozone EU member states’ governments, as well as new international organizations set up to deal with the Eurozone crisis, most notably the European Stability Mechanism (ESM) (Ban and Seabrooke 2017; Bátora 2020). The EU’s segmented character has enabled it to fashion quite coherent policy responses especially in the Eurozone. Segmentation allows stabilization of patterns of interaction among a constellation of actors sharing a specific view of what are appropriate types of responses to the given crisis –​in this case ordo-​cum-​neo-​liberalism as an ideational foundation for managing the bailout packages to Greece and other crisis ridden member states of the Eurozone (Tranøy and Schwartz 2020). The EU’s ability to fashion coordinated responses to thorny issues, whilst at the same time allowing for increased differentiation, underlines our main point, namely that a segmented political system is a distinctly differentiated polity. It operates according to a distinct logic of segmentation that may both serve integration and differentiation and do so in a distinct manner. The notion of segmented political system yields precision to the manner in which the EU is institutionally and structurally configured. In depicting the EU political and institutional system as a segmented political system, we obtain a more nuanced conceptual apparatus that helps us to specify different types of political system and how a given political system tags onto a given configuration of differentiation. The implication in structural terms is that we should abandon the statist analogy and the manner in which statists think about differentiation and instead build on what we presented above on the EU as a segmented political system, and spell that out in further detail with specific focus on how a segmented political system is configured in differentiation terms. We then need to analyse differentiated integration in relation to the EU’s segmented traits. That provides a much more specific vantage point for assessing the nature and dynamics of EU differentiation. In the following, we spell out in further detail the main points of convergence and divergence between differentiated integration and a segmented political order. As noted above, a complicating factor here is that the terms are far from theory independent; neither is their usage. Differentiated integration is primarily analysed from the vantage point of law and rational choice institutionalism, whereas our approach to segmentation and segmented political order is steeped in sociological and historical forms of institutionalism. We therefore include the theoretical difference in our brief assessment below. A further complication is that differentiated integration refers to policy or rule-​specific measures (exemptions, exceptions, opt-​outs and opt-​ins), the process of integration (different speeds and timescales) and the structure of the polity. The differences between the terms vary according to which of the three dimensions are in play. In the following, we discuss six sets of factors to explicate the similarities and differences between segmentation/​segmented political system on the one hand and differentiated integration/​differentiation on the other. Consider first demand and supply of differentiated integration in terms of exemptions, exceptions, opt-​ ins and opt-​ outs. Here students of differentiated integration focus on the role of the member state in the EU. They see differentiated integration as constituted by states and governments that negotiate participation in and/​or opt-​outs from various more or less institutionalized formats of cooperation. Indeed, differentiated integration confirms the role of the state as the constituting unit (Bellamy 2019). In the terminology introduced by Ruggie (1993), differentiated integration as a state-​driven process is state-​centric and homonymous. If 141

Jozef Bátora and John Erik Fossum

this state-​centred view is how analysts view integration, then differentiated integration would effectively preclude state formation on the polity level. A segmented political order is not state-​centric in the modern sense of the term. It is instead based on stabilized constellations of processes and interactions between various actors and structures across governments, international organizations, private enterprises, NGOs and civic groups. The closest a segmented political system comes to statehood is through the analogy to the rather peculiar pre-​modern form of statehood, which lacks the magnitude of territorial and functional penetration and control that we find in the modern state (Tilly 1993). Second, differentiated integration and segmentation vary with regard to integration. Segmentation can foster policy cooperation and joint solutions within segments. At the same time, segmented political orders do not necessarily emerge through integration but can emerge through partial disintegration. One option is by means of states collapsing into segmented political systems by limiting their functional repertoire and/​or losing the capacity to control territory. The other is through states coming together to form a new entity that is neither a state nor an international organization, but a partial and lopsided polity consisting of one or more segments –​ with limited ability to summon own resources and constraints on its scope and depth of action. Third, students of differentiated integration have a different conception of the institutional and procedural centre of gravity than do students of segmented political systems. Students of differentiated integration focus on (a) structures, institutions and formal rules of policy-​coordination; and (b) processes of states joining and enacting their participation in such integration frameworks and/​or, indeed, on opting out from these and enacting their non-​participation. Students of segmentation and segmented political orders focus on networks and interaction patterns stabilizing constellations of collaborating actors and structures, i.e. across governments, EU institutions, non-​state actors and private enterprises in stabilized constellations of collaboration within specific sectors (Bátora 2021b). Whereas differentiated integration is characterized by cooperation between hierarchies and based on hierarchical principles –​i.e. attention to formal rules and treaties, clear boundaries between participating organizations and actors’ position in formal structures, the logic of segmentation is based on network forms of cooperation characterized by informal ties, re-​current interactions, exchange of commodities with no stable value (e.g. information) and some sense of mutual obligation and trust.19 Fourth, differentiated integration differs from segmentation in terms of the key driving mechanisms. In differentiated integration, it is joint performance of selected policy functions. States participate in performing these together or they opt-​out. In terms of segmentation, in contrast, the key mechanisms are shared ideational frames and/​or ideologies that enable actors across different institutions and societal sectors to perceive policy problems in similar ways within distinct structural configurations that organize-​in particular policy solutions and organize-​out others. Examples of the latter include, for instance, the ordo-​cum-​neo-​liberal frame in dealing with the 2010s Eurozone crisis (Schwartz and Tranoy 2020); or the securitization frame in dealing with the migration crisis of 2015–​2016 (Bátora and Fossum 2020). Indeed, frames seem to be potent mechanisms for mobilizing attention and resources for particular types of policy solutions. As the study of crisis management in the EU’s recent crises shows, while EU member states easily end up abandoning joint precepts for the performance of joint policy functions when hit by crises and return to national solutions, mobilization of crisis response is helped by shared perceptions, frames and ideologies among key actors. Fifth, students of differentiated integration generally place less emphasis on the patterns of organizational learning and adaptation (for key sources, see March and Olsen 1976, Levitt and March 1988) than do students of segmentation. Students of segmentation see organizational

142

Differentiation and segmentation

learning as characterized by exploitation of existing frames across stabilized constellations of actors in governments, parliaments, EU institutions, private enterprises and NGOs. Changes in the environment are interpreted within such segments, and learning can involve path-​dependent adaptations. In this sense, organizational learning in segmented contexts is likely to generate coherence in adaptation patterns within the segment but not across the different sectors of the political order. Reliance on established solutions and pervasive exploitation can lead to competency traps or learner lock-​ins (Fossum 2020). In parallel, learning dynamics in EU-​wide segments may challenge patterns of exploitation in national contexts and dislodge existing competency traps there.20 Sixth, there is a difference between the effects that differentiated integration and segmentation have on democratic governance and how these effects can be dealt with from a democratic perspective. Students of differentiated integration insofar as they see democratic governance as anchored in the member states focus on structures and processes for conducting democratic scrutiny of governmental agencies’ participation in different formats and frameworks of differentiated integration. In terms of how cooperation is formed, we find that member state governments’ instrumental preferences play a key role (Moravcsik 1998). The democratic relation is often framed in principal-​agent terms, with the member states the principals and the EU-​level institutions the agent. An important democratic challenge then is agency drift, creeping competence and other developments that undermine the principals’ ability to control the agent. International institutions can exacerbate the problem of “democratic disconnect between the peoples of the constituent states and the inter-​and multi-​national decisions their domestic representatives make in their name…”(Bellamy 2019: 12). Nevertheless, in a globalized and interconnected world, this disconnect cannot be abrogated by domesticating decision-​making because that renders people vulnerable to external domination. As Richard Bellamy underlines, “(r)ather than subverting democracy at the national level, global institutions are in many respects vital to its continuing effectiveness and acceptability in an interconnected world”(2019: 3). The solution is however not democratizing global institutions. Instead, “they can acquire this legitimacy not by becoming themselves sources of democratic authority but through being under the democratically authorised and accountable control of the states that established them and regulate their interactions through them”(2019: 3). Differentiated integration is in this context intrinsic to the democratic character and operation of the multilevel EU system. Segmentation and a segmented political system pose serious challenges for democratic governance, but these are not the same as those we identify under the heading of differentiated integration, as spelled out above. Segments as stabilized constellations of actors promoting policies in various fields of governance boost collective action capacity in delivering policy solutions. This can strengthen output legitimacy (cf. Scharpf 1998). Nevertheless, a segmented political system is one where segments foster particular kinds of closure and generate –​and entrench –​ biases as well as sidelining democratic bodies. An important difference then is that students of differentiated integration operating from an intergovernmental perspective tend to downplay the role of such bodies as the EU parliament and, therefore, have problems accounting for the EP’s rising institutional role and presence in the EU. Students of segmented political systems underline how the rise of the EP was associated with de-​segmentation, whereas the EP was sidelined under the conditions of segmental closure that occurred during the financial crisis. As the case of the ESM and Troika involvement in implementation of the bailout packages in Greece after its 2011 financial breakdown indicates, ESM has brought in ordo-​cum-​neo-​liberal solutions that worked but, at the same time, decision-​making processes related to the bailout packages were seen as closed and impenetrable and were generating various types of hegemony

143

Jozef Bátora and John Erik Fossum

and dominance (Chatzistavrou 2020). Indeed, as some of the direct participants in the processes argued, the EU and its economic governance could be characterized as being under control of an unaccountable “deep-​establishment” (Varoufakis 2017). Operation of segments in EU governance is often characterized by informality, by interstitial organizational structures with special statuses, and that are situated on the fringes of and/​or outside the remits of the EU treaties. Given such features, it is not always straightforward to pinpoint where responsibility for particular decisions lays and how democratic scrutiny could be properly performed. De-​segmentation as restoring democracy is therefore not simply a matter of empowering national parliaments as intergovernmentalists underline, but equally about empowering the European Parliament to enforce proper control and accountability at the EU level. De-​segmentation in an EU context requires reinforcing the EU level as a relatively autonomous governing level; hence de-​segmentation requires both capacity-​building and strengthening of representative democracy at the EU level.

Conclusion In this chapter, we have taken as our point of departure that the EU that emerged out of the fiscal and refugee crises has continued to integrate, even if the pattern of integration has become more differentiated. Hence, as is frequently pointed out, there is less assurance than before that all states will integrate at the same pace, or even that all will eventually reach the same destination. Our concern in this chapter was to reflect on what a more differentiated EU entails in terms of the appropriate designation of the EU as a political system, and the processes that propel this entity. We sought to specify the phenomena and dynamics that the terms differentiated integration and differentiation referred to, and thereafter queried whether they could adequately account for the EU’s nature and development. Neither term offers a positive designation of the EU; hence, the difficulties in sorting out the different dimensions involved: policy-​specific versus system-​specific; structure-​specific versus process-​specific. In order to explicate the EU’s character qua polity, we introduced a third term, namely the notion of the EU as a segmented political system, and as part of that segmentation as a distinct action logic. We outlined the defining features of such a system and have underlined that the institutional arrangement at the EU level has obvious traits of a segmented political system. The designation of the EU as a segmented political system has clear bearings on how we understand and analyse differentiation and differentiated integration, given that a segmented political system is a distinctly differentiated polity that deviates in key ways from the manner in which we think of differentiation in the nation-​state context. In thus linking differentiation to the notion of segmented political system, we can hone in on and specify how we understand and analyse differentiated integration. The present literature relies on differentiated integration both to depict the EU’s distinct differentiation configuration and the process whereby states demand and the EU level supplies exceptions, exemptions, deviations from rules and institutional arrangements. By introducing the notion of segmented political system, we have a benchmark for assessing the implications of more or less territorial and functional differentiation. We can apply the same type of analysis to differentiated integration but now with the underlying phenomena and dynamics more clearly delineated. The overall implication of this analysis is that we should work from the positive designation of the EU as a segmented political order.That means abandoning the implicit statist analogy that crops up and muddies our conceptions of the EU. Ironically, such an approach renders more useful what statists think about differentiation because we can consider that in relation to a positive delineation of the EU qua segmented political order. 144

Differentiation and segmentation

Notes 1 Analysts then also speak of a certain shift in gravity towards the EU’s intergovernmental institutions. See in particular The New Intergovernmentalists (Bickerton et al. 2015). 2 For a more extensive elaboration of the similarities and differences between these two terms, see Fossum’s chapter in this book (2020) and Fossum (2019). 3 See Schmitter (1996, 2000), on “condominio” and “consortio”. 4 For a selection, consider Kölliker (2006); Warleigh-​Lack (2015); Stubb (1996); Andersen and Sitter (2006); Dyson and Sepos (2010); Fabbrini (2015); Schimmelfennig (2014); Leuffen, Rittberger and Schimmelfennig (2013); Piris (2012); Genschel and Jachtenfuchs (2014, 201); Bickerton, Hoodson and Puetter (2015); Adler-​Nissen (2014); Lord (2017); Schimmelfennig and Winzen (2014). 5 See, for instance, Leuffen, Rittberger and Schimmelfennig (2013). 6 See in particular Leuffen, Rittberger and Schimmelfennig (2013) for a specification of these dimensions. 7 “The conventional story attributes differentiation to an increase in heterogeneity among the member states triggered mainly by enlargement rounds but also by broadening the functional scope of EU-​ level policy-​making and the centralization or supranationalization of decision-​making”, Holzinger and Schimmelfennig (2012, 299). 8 There is a growing literature on EU disintegration. See, for instance, Jones, Kelemen and Meunier (2016);Vollaard (2014, 2018); Webber (2014; 2018); and Zielonka (2014). 9 For these distinctions, see Fossum (2015). 10 This conceptualization of segments builds on the “garbage can model” of organizational choice as proposed by Cohen, March and Olsen (1972). 11 For the notion of core state powers in the EU context, see Genschel and Jachtenfuchs (2014). 12 Marija Bartl presents it as: “a specific pattern of political action in the field of internal market, which has emerged gradually due to the confluence of three main factors: first, the EU’s functional institutional design; second, the processes of post-​national juridification; and third, a more contingent influence of ideas. In the interplay of those three factors, the interpretation of internal market has become overdetermined, restricting thereby the space of (democratic) politics in its regulation” (Bartl 2015: 572). 13 Further evidence of securitization are the practices of Frontex. In 2017–​2019, this agency was responsible for the protection of EU borders and management of refugee flows and has been actively in contact with defence industrial actors and has increasingly acquired various types of surveillance technologies and advanced defence equipment. See collection of official documents from 2017 to 2019 collected and published by German public TV station ZDF in 2021 at www.frontexfiles.eu (accessed on March 30, 2021). 14 That is reflected in the emergence of what Mark Dawson has labelled the Coordinative Method: “EU economic decision-​making is coordinative in that it is formed as a policy cycle based on a constant ‘back and forth’ between the EU and national levels… decision-​making never crystallises into a ‘once and for all’ agreement but is ongoing and revisable with the possibility of norms being adapted to changed factual circumstances” (Dawson 2015a: 53; see, also, 2015b). 15 Matej Avbelj, “The Right Question about the FCC Ultra Vires Decision”, Verfassungsblog (May 6, 2020). Available at: https://​verf​assu​ngsb​log.de/​the-​r ight-​quest​ion-​about-​the-​fcc-​ultra-​vires-​decis​ion/​. 16 Examples of interstitial organizations include the European External Action Service, European Border and Coast Guard or the European Stability Mechanism –​see Bátora (2020) and Bátora (2021a) 17 It needs to be noted, though, that the EU’s reaction to the Covid-​19 crisis and the agreement of member states on the 750 billion Euro heavy Recovery fund Next Generation EU is built on a combination of loans and grants with the latter set to flow, in particular, as direct subsidies to countries and regions hardest hit by the epidemic. The nature of these instruments still needs to be analysed. 18 A case in point is, for instance, is the liberum veto rule in the Polish kingdom prior to constitutional reform in 1791 based on the notion of full equality of all aristocratic representatives to the representative assembly (Sejm) and allowing every member to stop a piece of legislation by his own single vote. This often led to deadlocks and delays and, arguably, also to the first partition of Poland following Russian military intervention in 1767. 19 For models of hierarchies and networks as organizational paradigms, see Powell (1990). 20 Arguably, management of the financial/​Eurozone crisis in Greece by the Troika was an example of how external segmental logic building on ordo-​liberal frames dislodged existing patterns of performing fiscal and economic governance in Greece (Chatzistavrou 2020). In general, a change in organizational 145

Jozef Bátora and John Erik Fossum

structure can lead to dislodging existing competency traps (Levinthal and March 1993, Siggelkow and Levinthal 2005).

References Adam, C., Hurka, S., Knill, C. and Steinebach,Y. (2019) Policy Accumulation and the Democratic Responsiveness Trap, Cambridge: Cambridge University Press. Adler-​Nissen, R. (2014) Opting Out of the European Union. Diplomacy, Sovereignty and European Integration, Cambridge: Cambridge University Press. Andersen, S.S. and Sitter N. (2006) Differentiated Integration: What is it and How Much Can the EU Accommodate?, European Integration, 28(4), pp. 313–​330. doi: 10.1080/​07036330600853919. Avbelj, M. (2013) Differentiated Integration. Farewell to the EU-​27?, German Law Journal, 14(1), pp. 191–​ 211. doi: 10.1017/​S2071832200001760. Ban, C. and Seabrooke, L. (2017) From Crisis to Stability: How to Make the European Stability Mechanism Transparent and Accountable. Report. Brussels: Transparency International EU. Bartl, M. (2015) Internal Market Rationality, Private Law and the Direction of the Union: Resuscitating the Market as the Object of the Political, European Law Journal, 21(5), pp. 72–​598. doi: 10.1111/​eulj.12122. Bartolini, S. (2005) Restructuring Europe: Centre Formation, System Building, and Political Structuring between the Nation State and the European Union, Oxford: Oxford University Press. Bátora, J. (2013): “The ‘Mitrailleuse Effect’: The EEAS as an Interstitial Organization and the Dynamics of Innovation in Diplomacy“ in Journal of Common Market Studies, 51(4), pp. 598-​613 Bátora, J. (2020) Interstitial Organisations and Segmented Integration in EU Governance. In Bátora, J. and Fossum, J. E. (eds.) Towards a Segmented European Political Order: The European Union’s Post-​Crises Conundrum, London: Routledge, pp. 152–​174. Bátora, J. (2021a): “States, interstitial organizations and the prospects for liberal international order” in International Affairs, 97(5), pp. 1433-​1450 Bátora, J. (2021b): “Dynamics of Differentiated Integration in EU Defence: Organizational Field Formation and Segmentation” in European Foreign Affairs Review,Vol. 26, Special issue, pp. 63 –​86 Bátora, J. and Fossum, J. E. (eds) (2020) Towards a Segmented European Political Order, London: Routledge. Bauer, M. W. and Trondal, J. (eds) (2015) The Palgrave Handbook of the European Administrative System. Houndmills: Palgrave. Bellamy, R. (2019) A Republican Europe of States, Cambridge: Cambridge University Press. Bickerton, C. J., Hodson, D. and Puetter, U. (eds) (2015), The New Intergovernmentalism: States and Supranational Actors in the Post-​Maastricht Era, Oxford: Oxford University Press. Chatzistavrou, F. (2020):The current status of dominance in the EU system of economic governance: Drawing upon the Greek case, paper presented to the EU3D workshop, Bratislava, February 6, 2020. Christensen, T. and Egeberg, M. (1979) Organized Group-​Government Relations in Norway: On the Structured Selection of Participants, Problems, Solutions, and Choice Opportunities, Scandinavian Political Studies, 2(3), pp. 239–​260. Cohen, M. D., March, J. G., and Olsen, J. P. (1972) A Garbage Can Model of Organizational Choice, Administrative Science Quarterly, 17(1), pp. l–​25. doi: 10.2307/​2392088. Crum, B., and Fossum, J. E. (eds) (2013) Practices of Interparliamentary Coordination in International Politics:The European Union and Beyond, Colchester: ECPR Press. Dawson, M. (2015a) The Euro Crisis and its Transformation of EU Law and Politics. In Hertie School of Governance (eds), The Governance Report 2015, Oxford: Oxford University Press, 2015, pp. 41–​68. Dawson, M. (2015b) The Legal and Political Accountability Structure of ‘Post-​ Crisis’ EU Economic Governance, Journal of Common Market Studies, 53(5), pp. 976–​993. doi: 10.1111/​jcms.12248.De Witte, B. (2017) Variable Geometry and Differentiation as Structural Features of the EU Legal Order. In De Witte, B., Ott, A. and Vos, E. (eds) Between Flexibility and Disintegration, Cheltenham: Edward Elgar, pp. 9–​27. Dehousse, R. (2015) The New Supranationalism, Paper prepared for presentation at the ECPR General Conference, Montreal, 26–​29 August 2015. Dyson, K. and Sepos,A. (eds) (2010) Which Europe?The Politics of Differentiated Integration, Basingstoke: Palgrave Macmillan. Fabbrini, S. (2013) Intergovernmentalism and its Limits: Assessing the European Union’s Answer to the Euro Crisis, Comparative Political Studies, 46(9), pp. 1–​27. doi: 10.1177/​0010414013489502. Fabbrini, S. (2015) Which European Union? Europe After the Euro Crisis, Cambridge: Cambridge University Press. 146

Differentiation and segmentation

Fabbrini, S. (2018) Europe’s Future: Decoupling and Reforming, Cambridge: Cambridge University Press. Fasone, C. (2014a) Eurozone, Non-​Eurozone and ‘Troubled Asymmetries’ Among National Parliaments in the EU. Why and to what Extent this is of Concern, Perspectives on Federalism, 6(3), pp. 1–​41. Fasone, C. (2014b) European Economic Governance and Parliamentary Representation.What Place for the European Parliament?, European Law Journal, 20(2), pp. 164–​185. doi: 10.1111/​eulj.12069. Fossum, J.E. (2015): “Democracy and differentiation in Europe” in Journal of European Public Policy, 22(6): 799-​815 Fossum, J. E. (2017) Democracy and Legitimacy in the EU. Challenges and Options. InVai, L.,Tortola, P. D. and Pirozzi, N. (eds) Governing Europe –​How to Make the EU More Efficient and Democratic, Brussels: Peter Lang. Fossum, J. E. (2019) Europe’s Triangular Challenge: Differentiation, Dominance and Democracy, EU3D Research Papers 1/​2019. Available at: Europe’s Triangular Challenge: Differentiation, Dominance and Democracy by John Erik Fossum: SSRN. Fossum, J. E. (2020) The Institutional Make-​up of Europe’s Segmented Political Order. In Batora, J. and Fossum, J. E. (eds) Towards a Segmented European Political Order, London: Routledge, pp. 22–​46. Fossum, J. E. and Menéndez, A. J. (2011) The Constitution’s Gift –​A Constitutional Theory for a Democratic European Union, Boulder, CO: Rowman and Littlefield. Genschel, P. and Jachtenfuchs, M. (2014) Beyond the Regulatory Polity? The European Integration of Core State Powers, Oxford: Oxford University Press. Genschel, P. and Jachtenfuchs, M. (2018) From Market Integration to Core State Powers: The Eurozone Crisis, the Refugee Crisis and Integration Theory, Journal of Common Market Policy, 56(1), pp. 178–​196. doi: 10.1111/​jcms.12654. Grimm, D. (2015) The Democratic Costs of Constitutionalisation: The European Case, European Law Journal, 21(4), pp. 460–​473. doi: 10.1111/​eulj.12139. Heidbreder, E. (2014) Regulating Capacity Building by Stealth: Pattern and Extent of EU Involvement in Public Administration. In Genschel, P. and Jachtenfuchs, M. Beyond the Regulatory Polity? The European Integration of Core State Powers, Oxford: Oxford University Press, pp. 145–​165. Héritier, A. (2021) Quo Vadis Europa? Five Paths, their Plausibility, and Impact in EUI Working Papers, RSC 2021/​49, Florence: Robert Schuman Centre for Advanced Studies, EUI. Holzinger, K. and Schimmelfennig, F. (2012) Differentiated Integration in the European Union: Many Concepts, Sparse Theory, Few Data, Journal of European Public Policy, 19(2), pp. 292–​305. doi: 10.1080/​ 13501763.2012.641747. Huysmans, J. (2000) The European Union and the Securitization of Migration, Journal of Common Market Studies, 38(5), pp. 751–​777. Isiksel,T. (2016) Europe’s Functional Constitution:A Theory of Constitutionalism Beyond the State, Oxford: Oxford University Press. Jones, E., Kelemen, R. D. and Meunier, S. (2016) Failing Forward? The Euro Crisis and the Incomplete Nature of European Integration, Comparative Political Studies, 49(7), pp. 1010–​1034, doi: 10.1177/​ 0010414015617966. Kölliker, A. (2006) Flexibility and European Unification: The Logic of Differentiated Integration, Lanham, MA: Rowman and Littlefield. Korff,V. P., Oberg, A and Powell, W. W. (2015) Interstitial Organizations as Conversational Bridges, Bulletin of the American Society for Information Science and Technology, 41(2), pp. 34–​38 doi: 10.1002/​bult.2015.1720 410210.Lazaridis, G. and Khursheed,W. (eds) (2015) The Securitisation of Migration in the EU: Debates since 9/​11, Basingstoke: Palgrave Macmillan. Leruth, B. and Lord, C. (2015) Differentiated Integration in the European Union: A Concept, a Process, a System or a Theory, Journal of European Public Policy, 22(6), pp. 754–​763. doi: 10.1080/​ 13501763.2015.1021196. Leuffen, D., Rittberger, B. and Schimmelfennig, F. (2013) Differentiated Integration: Explaining Variation in the European Union, Basingstoke: Palgrave Macmillan. Levinthal, D. A. and March, J. G. (1993) The Myopia of Learning, Strategic Management Journal, 14(2), pp. 95–​112. doi: 10.1002/​smj.4250141009. Levitt, B. and March, J. G. (1988) Organizational Learning, Annual Review of Sociology. 14(2), pp. 319–​340. doi: 10.1146/​annurev.so.14.080188.001535. Lord, C. (2017) Segmentation, Differentiation and the Aims of European Integration. In Grimmel, A. (ed) The Crisis of the European Union: Challenges, Analyses, Solutions, Routledge, pp. 185–​198. Majone, G. (1994) The Rise of the Regulatory State in Europe, West European Politics, 17(3), pp. 77–​101. doi: 10.1080/​01402389408425031. 147

Jozef Bátora and John Erik Fossum

March, J. G. and Olsen, J. (1976) Ambiguity and Choice in Organizations. Bergen: Universitetsforlaget. Menéndez, A. J. (2013) The Existential Crisis of the European Union, German Law Journal, 14 (5), pp. 453–​ 525. doi: 10.1017/​S2071832200001917. Moravcsik, A. (1998) A Choice for Europe. Princeton, NJ: Princeton University Press. Olsen, J. P. (1988) Statsstyre og institusjonsutforming [State Governance and Institutional Design], Oslo: Universitetsforlaget. Orren, K. and Skowronek, S. (2017) The Policy State: An American Predicament, Harvard, MA: Harvard University Press. Parsons, T. (1951) The Social System, Chicago, IL: Free Press. Picot, G. (2012) Politics of Segmentation: Party Competition and Social Protection in Europe, Abingdon: Routledge. Powell, W. W. (1990) Neither Market nor Hierarchy: Network Forms of Organization, Research in Organizational Behavior, 12, pp. 295–​336. Richardson, J. (ed) (1982) The Concept of Policy Style, Abingdon: Routledge. Rokkan S. (1975) Dimensions of State Formation and Nation Building: A Possible Paradigm for Research on Variations within Europe. In Tilly, C. (ed) The Formation of National States in Western Europe, Princeton, NJ: Princeton University Press, pp. 562–​600. Ruggie, J. G. (1993) Territoriality and Beyond: Problematizing Modernity in International Relations, International Organization, 47(1), pp. 139–​174. Scharpf, F.W. (1998) Interdependence and Democratic Legitimation, MPIfG Working Paper, No. 98/​2, Cologne: Max Planck Institute for the Study of Societies. Scharpf, F. W. (2010) The Asymmetry of European Integration, or Why the EU cannot be a ‘Social Market Economy’, Socio-​Economic Review, 8(2), pp. 211–​250. doi: 10.1093/​ser/​mwp031. Schimmelfennig, F. (2014) European Integration in the Euro Crisis: The Limits of Post Functionalism, Journal of European Integration, 36(3), pp. 321–​337. doi: 10.1080/​07036337.2014.886399. Schimmelfennig, F. and Winzen, T. (2014) Instrumental and Constitutional Differentiation in the European Union, Journal of Common Market Studies, 52(2), pp. 354–​370. doi: 10.1111/​jcms.12103. Schmitter, P. (1996) Imagining the Future of the Euro-​Polity with the Help of New Concepts. In Marks, G., Scharpf, F. W., Schmitter, P. C. and Streeck, W. (eds) Governance in the European Union. London: Sage, pp. 121–​150. Schmitter, P. (2000) How to Democratize the European Union … And Why Bother? Lanham: Rowman and Littlefield. Siggelkow, N. and Levinthal, D. A. (2005) Escaping Real (Non-​Benign) Competency Traps: Linking the Dynamics of Organizational Structure to the Dynamics of Search, Strategic Organization, 3(1), pp. 85–​ 115. doi: 10.1177/​1476127005050521. Spijkerboer,T. (2017) Bifurcation of People, Bifurcation of Law: Externalization of Migration Policy before the EU Court of Justice, Journal of Refugee Studies, 31(2), pp. 216–​239. doi: 10.1093/​jrs/​fex038. Stubb, A. C. G. (1996) A Categorization of Differentiated Integration, Journal of Common Market Studies, 34(3), pp. 283–​295. doi: 10.1111/​j.1468-​5965.1996.tb00573.x. Tilly, C. (1993) European Revolutions, 1492–​1992, Oxford: Wiley-​Blackwell. Tranøy, B. S. and Schwartz, H. (2020) Illusions of Convergence: The Persistent Simplification of a Wicked Crisis. In Batora, J. and Fossum J. E. (eds) The EU Post-​crises: The Rise of a Segmented Political Order, London: Routledge, pp. 47–​71. Varoufakis, Y. (2017) Adults in the Room: My Battle with Europe’s Deep Establishment. London: The Bodley Head. Vollaard, H. (2014) Explaining European disintegration. Journal of Common Market Studies, 52(5), pp. 1142–​ 1159. doi: 10.1111/​jcms.12132. Vollaard, H. (2018) European Disintegration: A Search for Explanations. London: Palgrave. Warleigh-​Lack, A. (2015) Differentiated Integration in the European Union: Towards a Comparative Regionalism Perspective, Journal of European Public Policy, 22(6), pp. 871–​ 887. doi: 10.1080/​ 13501763.2015.1020837. Webber, D. (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), pp. 341–​ 365. doi: 10.1177/​1354066112461286. Webber, D. (2018) European Disintegration? The Politics of Crisis in the European Union. London: Palgrave. Zielonka, J (2014) Is the EU doomed? Cambridge: Polity Press.

148

Part 2

Institutionally based differentiation Jarle Trondal, Stefan Gänzle and Benjamin Leruth

The European Union (EU) serves as a living research laboratory for understanding the institutional basis of political order (Olsen 2007). A political order consists of a relatively stable arrangement of institutions that are both formalised and institutionalised, in which institutional structures serve as the action capacity of the system. Among these core institutions, public administration serves to stabilise the political order of the modern state by making the abstract power of the state effective (Beetham 1985). Public administration serves as a necessary component of political order. In a multilevel political order, public administration moreover systematically shapes policy outcomes across levels of executive authority. Without administrative institutions, public policies are less likely to be initiated, drafted, and implemented (Fukuyama 2013), and lesson drawing is less likely to be facilitated. Moreover, public administration is essential for ‘good governance’ since the well-​being of citizens is shown to be nurtured by societies administered by ‘impartial’ public bureaucracies (Rothstein 2012). Independent institutional capacities are central for making ‘good’ and ‘living’ political orders (March and Olsen 1989). Thus, examining the institutional architecture of differentiated political systems is important in order to understand efficient and legitimate public policymaking. A central feature of political order is their internal integration and ability to act coherently. One relevant question is thus how competing patterns of integration and differentiation and fragmentation and siloisation may co-​exist within and among EU institutions. Studies observe that common political orders do not necessarily act coherently. Orren and Skowronek (2004) suggested that different elements of administrative bodies tend to overlap, counteract, and be poorly coordinated rather than coordinated and well ‘ordered’. Patterns of integration and disintegration also tend to be cumulatively layered over time, contributing to compound structures and ways of governing (Thelen 2004). The public administration literature on institutional differentiation has argued that differentiation is a prerequisite for the emergence and sustainability of multilevel governance processes. Differentiation is argued to be an institutional precondition for multilevel governing processes to emerge and ensue (Egeberg and Trondal 2016). The argument is that differentiation of nation-​state governments and EU institutions establishes structured choice opportunities that are conducive to certain governing patterns to emerge across levels of governance –​e.g. governance processes between EU-​level agencies and domestic agencies working on related task portfolios (Trondal 2022). Such processes are moreover likely to emerge because institutional DOI: 10.4324/9780429054136-11

149

Jarle Trondal, Stefan Gänzle and Benjamin Leruth

choice architectures tend to mobilise attention and action capacity around certain problems and solutions while ignoring others, focus attention along particular lines of conflict and cooperation and so on (Simon 1983). Institutional differentiation is thus not merely an expression of symbol politics (Feldman and March 1981) but they create systematic and permanent biases in collective decision processes by directing and nudging collective choices towards certain problems and solutions, thereby making certain outcomes more plausible than others (Egeberg and Trondal 2020; Schattschneider 1975;Thaler and Sunstein 2009). As outlined in this part of the Handbook, the institutional differentiation of multilevel governance systems represents choice architectures that mobilise varieties of differentiation.

Chapter 10 An ever more fragmented Union? On the emerging relevance of the institutional implications of differentiated integration in the EU In this chapter, Alexander Schilin examines how differentiation affects patterns of cooperation among the EU member states. Existing scholarship has to date insufficiently taken into account the role of institutional designs. This chapter argues that in light of the developments within Economic and Monetary Union (EMU), scholars might study more systematically how governance processes are structured under the circumstances of differentiation and, in particular, how inclusive the institutional structures are to outsiders.The chapter reviews how EU member states outside the Eurozone are represented within the European Stability Mechanism, the Banking Union and the restructured Eurozone formats within the Council of the EU and European Council.The reviewed reform projects showcase the lack of organisational inclusiveness to Euro outsiders within post-​crisis EMU. Only on a very limited number of occasions are national officials representing those EU member states outside the Eurozone allowed to participate in meetings within the reviewed facilities. Considering these circumstances, shedding light on the institutional implications of differentiation in the EU becomes a pressing concern. In order to facilitate the study of this subject, the chapter also makes a proposal on how the inclusiveness of EU governance bodies to EU outsiders could be operationalised.The proposed operationalisation is based on two analytical dimensions: outsider representation and outsider access. While the former dimension entails considering how well outsiders are represented within an EU governance body, the latter one is directed towards determining the influence they can exert on decisions made in one EU organisation through their position within other places. The chapter concludes by highlighting two measures which are particularly well suited to ensure inclusiveness within EU institutions.

Chapter 11 Promise unfulfilled? Managing differentiated integration in EU secondary law through enhanced cooperation In this chapter, Daniela Annette Kroll examines enhanced cooperation (EnC). EnC is the legal procedure that allows a limited group of EU member states to pursue differentiated integration in EU secondary law. After showing how the procedure has evolved in the EU treaties over time, she describes the current legal rules on applying EnC in the different EU pillars.The ensuing literature review shows that most studies approach EnC from a legal perspective. Empirical studies with a theoretical argument are limited and often idiosyncratic case studies. Furthermore, EnC 150

Institutionally based differentiation

plays a secondary role in the literature. EnC has been implemented in only a handful of cases, although the procedure has been enshrined in the Treaty of Amsterdam.Yet, the chapter argues that the research focus on those positive cases is too short-​sighted, since it tends to ignore the major function of the procedure: EnC has been an instrument to establish consensus in EU negotiations. The potential use of EnC in the future has eventually led to its non-​use. Hence, the impact of EnC in EU negotiations often stays under the radar.

Chapter 12 Differentiation in the European Parliament: united about diversity? In this chapter, Guri Rosén explores how Members of the European Parliament (MEPs) approach differentiation. Differentiation in the EU has often been described and discussed as a matter for member states to accommodate their various concerns whilst being able to progress the integration process. At the same time, differentiation has broad implications not only for member states but also for the EU’s millions of citizens. Integrating measures of different speeds and an array of complex formats, as well as possibly disintegrating processes, trigger questions not only about cohesion and governance but also about social support and the very existence of the EU. Being directly elected by the EU citizenry, it is the role of MEPs to voice the viewpoints of their constituents. Therefore, this chapter addresses how MEPs experience and evaluate processes of differentiation. Although the European Parliament has proclaimed that the debate about differentiation should not be about pro-​differentiation versus anti-​differentiation, this chapter demonstrates that MEPs have opposing visions of the future of the EU, and they evaluate proposals for differentiation differently.The chapter shows how MEPs view the issue of differentiation as closely connected to the question of democracy in the EU. However, it is identified as both a remedy to a lack of democracy, and as a threat to it.

Chapter 13 Differentiation and the European Central Bank: a bulwark against (differentiated) disintegration? As the guardian of the euro, the European Central Bank (ECB) oversees a prime example of differentiated integration. Against the backdrop of the multiple crises of the euro’s second decade, Daniel F. Schulz and Amy Verdun ask how the ECB has dealt with differentiation. The chapter analyses both the historical development of the ECB’s relationship with euro outsiders –​ discerning between ‘old’ opt-​outs and ‘new’ accession countries in the context of European Union (EU) enlargement –​and how differentiation affects ECB policymaking across its various tasks. Specifically, the chapter analyses three logics of ‘deepening’ and ‘widening’: (1) is the ECB encouraging euro membership among the ‘outs’? (2) Does it seek to reduce the impact of differentiation by keeping the ‘outs’ on board as much as possible? (3) Or does the ECB further cement differentiation by excluding the ‘outs’ from decision-​making or deepening integration among the ‘ins’ only? The chapter finds that, in the past, the ECB has been hesitant to support ‘more Europe’.When the sovereign-​debt crisis posed a potentially existential threat, however, the ECB started adopting a more proactive role through both monetary policy and discursive acts. The COVID-​19 crisis appears to confirm that the ECB has shed its narrow technocratic focus in order to provide political leadership in the EU. Yet, this does not suggest that the ECB is a competence maximiser ‘hardwired’ to ever closer union. Rather, the evidence suggests that it merely 151

Jarle Trondal, Stefan Gänzle and Benjamin Leruth

accepted greater powers and a deepening of integration to avert the threat of (differentiated) disintegration.

Chapter 14 Differentiation and the European Commission Differentiation has become a main defining feature of the European integration process, and it may be expected to be one of the key mechanisms in allowing further integration in the future. It, however, sets the European institutions under pressure and begs the question of whether the EU institutional framework should mirror these evolutions. Thus, this chapter by Diane Fromage and Cristina Fasone specifically focuses on the question of whether the European Commission should become differentiated. The authors find that it should not because this would prevent it from assuming the role ascribed to it by the Treaties. This conclusion is based on the analysis of the composition and function set for the Commission in the current EU legal framework, and of its role in the emergence and the management of initiatives of differentiated integration.

Chapter 15 Differentiation and the European Court of Justice In this chapter, Sabine Saurugger and Fabien Terpan examine the place of the European Court of Justice (ECJ) in the increasingly differentiated political system of the EU. If a constitutional court’s role such as the ECJ’s is to guarantee the homogenous application of law in a political system, is increasing legal and judicial differentiation blocking its actions? The chapter starts with distinguishing two main sources of differentiation in the legal and judicial realm of the EU. Firstly, differentiation results from the resistance of national courts towards the constitutionalisation of European law, which led to constitutional pluralism. A second source of differentiation is linked to the fact that national governments have departed from the notion of a European rule of law homogenously applied to the EU territory, introducing specific types of differentiation –​vertical and horizontal –​with the aim to offer the Member States a way out of negotiating deadlock, and/​or preserve Member States’ sovereignty. Confronted with these dynamics, the Court has actively developed instruments to restrict legal and judicial differentiation through its activism in fields where its competences are limited at best and through the establishment of the sociological instrument of judicial dialogue. Using approaches stemming from legal, sociological as well as network studies traditions, this chapter shows that while legal differentiation is indeed increasing, the ECJ has reacted in order to keep this process at a minimum.

Chapter 16 Third-​country participation in EU agencies: towards ‘condominio’? In this chapter, Sandra Lavenex examines the role of EU agencies. The chapter conceptualises and illustrates the transgovernmental layer of external differentiated integration of non-​EU member states in EU agencies and adjacent administrative networks. It first introduces the distinct modes of governance applying to transgovernmental bodies and identifies the formal and 152

Institutionally based differentiation

informal institutional features that make EU agencies particularly permeable for flexible forms of differentiated integration with ramifications to overarching international regimes. This permeability is then illustrated for different EU agencies with regard to the involvement of third countries reaching from the closely associated West European (the EEA/​EFTA countries and Switzerland) candidate and neighbourhood countries (ENP) to more remote ones like Canada, Japan or the United States. While these transgovernmental ties echo Philippe Schmitter’s vision of a ‘condominio’ in which instead of a Eurocracy accumulating organisationally distinct but politically coordinated tasks around a single centre, there could be multiple regional institutions acting autonomously to solve common problems and produce different public goods. They evolve under the shadow of the EU’s central political bodies –​and in particular the Commission’s increasing quest for central control.

Chapter 17 The Council of the European Union: organisational and social dynamics of differentiation The Council of the European Union (EU) is a ‘complex’ of governance, one best depicted as a set of nested ‘sub-​societies’ rather than a monolithic politico-​administrative singularity, and each network of Council activity is embedded institutionally in its own social field of cooperation. As an organisational culture, Jeffrey Lewis shows in this chapter how the Council displays both time-​ worn ‘inherited traditions’ and ‘bricolage’ adaptability.This chapter examines how differentiation processes affect the Council system both institutionally and as a social field. Some differentiation patterns indicate a more variegated set of in-​group/​out-​g roup dynamics and excludable ‘club goods’ yet the Council’s ethos as transnational ‘community of practice’ is still keyed to a shared ‘system attitude’ that retains an outsize durability.

Key findings This section shows how varieties of organisational structures and institutional norms in the EU contribute to varieties of institutional differentiation. Whereas intergovernmental institutions such as the Council and administrative bodies such as EU-​level agencies tend to supply patterns of institutional differentiation, supranational institutions such as the European Commission and the European Central Bank are much less encouraging to differentiation. Yet, all chapters illuminate also how institutions tend to be complex systems embodying varieties of differentiation. Alexander Schilin examines how differentiation affects patterns of cooperation among the EU member states. Arguing that existing scholarship has insufficiently taken into account the role of institutional designs, he argues that scholars might study more systematically how governance processes are structured under the circumstances of differentiation and, in particular, how inclusive the institutional structures are to outsiders. The chapter shows how EU member states outside the Eurozone are also represented within the European Stability Mechanism, the Banking Union and the restructured Eurozone formats within the Council of the EU and European Council. Guri Rosèn also addresses differentiation from the viewpoint of the European Parliament and shows how Members of the European Parliament (MEPs) experience and evaluate processes of differentiation. Although the European Parliament has declared that the debate on differentiation should not be about pro-​differentiation versus anti-​differentiation, Rosèn demonstrates that MEPs have opposing visions of the future of the EU, and they evaluate proposals for differentiation differently. Jeffrey Lewis also shows how the intergovernmental 153

Jarle Trondal, Stefan Gänzle and Benjamin Leruth

institutional architecture of the Council is more supportive of differentiation, yet that internal variation is prominent. He observes both in-​group/​out-​g roup dynamics in the Council’s ethos as well as shared ‘system attitudes’. Similarly, this section establishes how administrative differentiation of the EU contributes to provide institutional architectures for differentiation. Sandra Lavenex shows how EU agencies are able to involve third countries such the EEA/​EFTA countries and Switzerland, candidate and neighbourhood countries, as well as countries like Canada, Japan and the United States. This section finally illustrates how the supranational institutions of the EU are institutional disruptions to processes of EU-​level differentiation. Daniel F. Schulz and Amy Verdun find that the European Central Bank has been hesitant to support ‘more Europe’. When the sovereign-​ debt crisis posed a potentially existential threat, however, the ECB has started adopting a more proactive role through both monetary policies and discursive acts. A fairly similar observation is made by Sabine Saurugger and Fabien Terpan, suggesting that while legal differentiation is indeed increasing, the European Court of Justice has reacted in order to keep this process at a minimum. Finally, Diane Fromage and Cristina Fasone focus on the question of whether the European Commission should become differentiated and find that it should not because this would prevent it from assuming the role ascribed to it by the Treaties.

References Beetham, D. (1985). Max Weber and the Theory of Modern Politics. Oxford: Polity Press. Egeberg, M. and J. Trondal (2016) `Why strong coordination at one level of government is incompatible with strong coordination across levels (and how to live with it): the case of the European Union', Public Administration, 94(3): 579–​592. Egeberg, M. and J. Trondal (2020) `The organizational basis for public governance', The Oxford Research Encyclopedia of Politics, 1–​28. Feldman, M.S. and J.G. March (1981) `Information in organizations as signal and symbol', Administrative Science Quarterly, 26(2): 171–​186. Fukuyama, F. (2013) `What is Governance?', Governance, 26(3): 347–​368. March, J.G. and J.P. Olsen (1989) Rediscovering Institutions. New York, NY: The Free Press. Olsen, J.P. (2007) Europe in Search of Political Order. Oxford: Oxford University Press. Orren, K. and S. Skowronek (2004) The Search for American Political Development. Cambridge: Cambridge University Press. Rothstein, B. (2012) `Good governance', in D.L. Faur (Ed.) The Oxford Handbook on Governance. Oxford: Oxford University Press, pp. 143–​154. Schattschneider, E.E. (1975) The Semisovereign People. Fort Worth, TX: Harcourt Brace Jovanovich College Publishers. Simon, H.A. (1983) Reason in Human Affairs. Stanford, CA: Stanford University Press. Thaler, R.H. and C.R. Sunstein (2009) Nudge. London: Penguin Books. Thelen, K. (2004) How Institutions Evolve. Cambridge: Cambridge University Press. Trondal, J. (2022) `The multilevel administrative state', in J. Tosun and P. Graziano (eds.) Encyclopedia of European Union Public Policy (forthcoming).

154

10 An ever more fragmented Union? On the emerging relevance of the institutional implications of differentiated integration in the EU Alexander Schilin

Introduction With differentiation becoming increasingly relevant to European Union (EU) politics concerns have grown about the damaging effects this might have on cooperation and cohesion among EU member states in the daily affairs of EU governance (European Parliament 2019). To study the phenomenon’s impact on EU governance, scholars have examined to what extent national representatives were able to maintain their informal contacts with their colleagues after their country had decided to opt out of a political project within the EU. In particular, in the Schengen Area (Adam et al. 2016; Adler-​Nissen 2009; 2014) and EMU (Howarth 2005; Naurin and Lindahl 2010), scholars have collected empirical evidence suggesting that the consequences of opt-​ outs for the professional networks of national representatives are almost negligible. Informal contacts and social norms of inclusion are still effective in facilitating deliberation and cooperation between all EU member states despite differentiation. While the consequences of differentiation on professional networks of member states’ representatives have been discussed extensively (see above), scholarship has fallen short of analysing the institutional structures which come with differentiation. In this chapter, I study the institutional structures associated with differentiation in EMU and how they affect the relationship between insiders and outsiders of the Eurozone. I review our knowledge of how outsiders of the Eurozone are integrated into the crucial gatherings within the European Stability Mechanism (ESM) as well as the Banking Union (BU) and in meetings of the EMU formats within the Council of the EU and the European Council. I showcase the unprecedented lack of organisational inclusiveness within the new facilities installed in the context of the sovereign-​debt crisis.While in other policy areas such as the Schengen Area, the forerunners in the integration process were careful to systemically include outsiders within the EU into processes of deliberation and governance; in EMU, the Eurozone outsiders have been excluded from new agencies’ governance boards or Council formats.This is observable for every Eurozone outsider in the EU regardless of whether the state is an accession state or a permanent outsider. As discussed in the introduction to the Handbook, most of the existing scholarship has DOI: 10.4324/9780429054136-12

155

Alexander Schilin

not focussed on the institutional implications of differentiation. Yet, these recent developments may require a shift in focus. I argue that scholars need to acquire a deeper understanding of the inclusiveness of institutional structures to outsiders. This would provide us with a more comprehensive picture of patterns of cooperation among EU member states under the circumstances of differentiation and thus enhance our understanding of the phenomenon’s implications on processes of EU governance. I further propose that in order to analyse the inclusiveness of EU governance structures for differentiation, scholars could consider outsider representation and outsider access.These analytical dimensions are based on a twofold operationalisation of inclusiveness within EU institutions which I deduct on the base of Lijphardt (1999). The chapter is divided into five sections. In the next section, I review the existing literature on the consequences of differentiation for processes of EU governance. Discussing different approaches of engaging with this topic I show that, with a few exceptions, scholarship has fallen short of examining the institutional implications of differentiation. In the third section, I analyse how Eurozone outsiders are represented in the institutional structures which the reform projects of the ESM, the BU and the reforms to the EMU formats in the two councils have produced. While in other policy fields, outsiders are systematically included (Adler-​Nissen 2009), in EMU, there is a clear departure from the overarching principle of institutional inclusiveness visible. In the fourth section, I propose an operationalisation of institutional inclusiveness consisting of outsider representation and outsider access. In the final section, I summarise the main conclusions and make concrete proposals on how in EMU inclusiveness to Eurozone outsiders could be improved.

Literature review Even if differentiation is not a new phenomenon within the process of European integration, scholarship knows little about the implications it has for governance and decision-​making processes in the EU. Studies have focused on exploring the notion of differentiation (Holzinger and Schimmelfennig 2012; Leruth and Lord 2015; Stubb 1996),explaining its emergence (Jensen and Slapin 2012; Kölliker 2001; Schimmelfennig et al. 2015; Schimmelfennig and Winzen 2020) or discussing its normative aspects (Bellamy and Kröger 2017; Eriksen 2018; Lord 2015). In examining the consequences of differentiation to EU governance, scholars mostly focus on whether governments from countries enjoying opt-​outs in specific policy areas would be able to maintain substantive influence on policy proposals. In this regard, scholars analysed how opt-​ outs affect the social networks of national officials representing their country’s interests in the Council of the EU. Yet, the way in which institutions and governance bodies are designed in light of differentiation has hardly been studied at this moment (Heermann and Leuffen 2020: 2; Schimmelfennig and Winzen 2020: 192). The first empirical studies on the situation that national representatives of an EU member state face in their daily working environment after their government had decided in favour of opting out from a step of integration were conducted in the context of the public discussion about the question of whether Sweden should join the common currency area. Swedish experts warned that after an opt-​out of monetary integration, Sweden’s representatives would lose their positions within the relevant networks and thus compromise on their political influence (Gidlund and Jarneck 1996). Other scholars supported this claim (Warleigh 2002). Campaigning in favour of Sweden’s accession to the Eurozone in the context of the national referendum in 2003, the Swedish deputy finance minister even argued that in case of an opt-​out, national diplomats would lose power in various policy fields beyond the immediate matters of fiscal and economic governance (Lund cited after Naurin and Lindahl 2010). Similar arguments were raised in the context 156

Governance structures

of the Danish referendum in 2000 (Marcussen and Zølner 2001) and the general election in the United Kingdom in 1997 (Bulmer 2008). However, Naurin and Lindahl (2010) presented data indicating that such apprehensions were largely unfounded (see also Manners 2008; ). They come to the conclusion that the (de facto) opt-​outs for monetary integration, which Denmark, the United Kingdom and Sweden have been granted with,1 did not have a significant influence on how the national representatives of these countries are embedded into professional networks. A more theoretical approach to this question has been provided by Adler-​Nissen (2008; 2014). Relying on the work of Bourdieu, she conceptualises the Council of the EU as a diplomatic field. When national representatives interact with each other, they refer to a shared system of meanings, behavioural norms and ontological views. This habitus includes a collection of unspoken rules determining what is appropriate when acting within the social context of the Council of the EU. National representatives learn and incorporate these rules within the regular practice of mutual interaction. With regard to the substance of this habitus, Adler-​Nissen (2014) considers two things particularly relevant: a commitment towards consensus-​seeking and a shared belief that more integration is the appropriate answer to today’s challenges of globalisation (integration doxa). Considering these social circumstances, the decision to opt out of a step of integration puts the officials from the concerned state in an unfortunate situation.They have to abstain from cooperation and therefore deviate from the social norms established within the Council of the EU. Representatives from other member states consider this decision of non-​assimilation non-​collegial. Therefore, in future interaction with the other member states’ representatives, the opt-​out becomes a stigma for the outsider state. Adler-​Nissen (2014: 64) defines a stigma as a disparaging emblem which members of one group can impose on others. Representatives whose states are confronted with a stigma suffer from it both in terms of its content and excluding impact. The relevance of the social stigma, however, is not predetermined. Instead, it is socially negotiated among national representatives themselves in the daily interaction within the committees of the Council of the EU (Svendsen and Adler-​Nissen 2019). There exist many strategies to manage a stigma (Adler-​Nissen 2014: 67–​69). How effective these strategies are in making the opt-​out less relevant for daily interactions within the Council of the EU depends on conceptions of the national government’s role, formal opt-​out rules and social norms relevant within the particular context (Adler-​Nissen 2009). This affects the capability of a country to maintain political influence despite an opt-​out. In assessing the consequences which differentiation has on EU governance, scholars are mostly concerned with the questions of how opt-​outs are perceived by national officials of other countries and whether member states subject to an opt-​out are able to maintain their political influence. Detailed analyses of the effects that differentiation has on the set-​up of EU institutions or governance processes at the same time are rare. They are published very sporadically and focus on individual policy fields. Tekin (2012; 2017) analyses the implementation of differentiation in the Area of Freedom, Security and Justice (AFSJ) as well as the Schengen Area. She concludes that while in the Schengen Area the exception rules for the United Kingdom and Ireland allow only for limited flexibility, in the AFSJ, the formal opportunities for opt-​ins allow for cost-​benefit decisions on a case-​by-​case basis. In the course of the negotiations on Brexit, these structures were reconsidered (see Foteini and Markozani 2019). The EMU reforms implemented in the context of the sovereign-​debt crisis have produced committees, agencies and boards which in terms of representation make a clear distinction between members and non-​members of the Eurozone. Looking into other EU policy areas, providing representatives from countries which have decided for an opt-​out with limited rights within bodies of governance and coordination is not a new approach. Yet, in those fields, the 157

Alexander Schilin

deprivation of voting rights on the basis of differentiation is carefully counterbalanced by various mechanisms of systematic inclusion ensuring that the views of the concerned countries are still incorporated. In post-​crisis EMU, however, the outsiders of the Eurozone are systematically excluded from processes of deliberation and coordination among the insiders. This is also applied to those EU member states which are still in the process of accessing the currency area in addition to those which officially secured an opt-​out from monetary integration. Thus, the degree to which the governance structures within post-​crisis EMU reforms exclude Eurozone outsiders within the EU is unique. These findings showcase the critical role that institutional structures can play in determining the degree to which outsiders can maintain influence in the respective policy area. Studies analysing how differentiation is reflected within the EU’s institutional structures and how this affects the formal opportunities for cooperation between insiders and outsiders are needed. This chapter makes a proposal on how to study the institutional implications of differentiation in the EU more systematically.

A union within the union? Reviewing EMU reforms in terms of their inclusiveness Before the sovereign-​debt crisis, EMU mainly relied on rules and soft coordination mechanisms (Schlosser 2019). Its main rulebook was laid down in the Stability and Growth Pact (SGP).Within this agreement, a combination of preventive and corrective measures was established to keep the fiscal balance sheets of all EU member states within explicitly determined margins (Heipertz and Verdun 2010). Furthermore, the SGP established the Eurogroup, in which the Finance Ministers of those member states inside the Eurozone could coordinate on their national policies as well as their positions on policy proposals discussed within the ECOFIN Council (Puetter 2006). Numerous scholars were concerned that this set of governance instruments would be insufficient (e.g.Verdun 2007). Indeed, when the crisis erupted, Eurozone members found themselves lacking the necessary resources to launch an appropriate response and to maintain the very existence of EMU they had to agree on far-​reaching reforms. In this section, I review those reforms which have significantly restructured EMU governance structures with a particular focus on the structures’ inherent inclusiveness to Eurozone outsiders. I demonstrate that the new mechanisms of governance and coordination within post-​crisis EMU lack inclusiveness to Eurozone outsiders. Table 10.1 offers a detailed overview of my findings.

Financial assistance –​the European Stability Mechanism (ESM) The lack of inclusiveness inherent to the structures, which have been added to EMU in the context of the sovereign-​debt crisis, is perhaps best illustrated by the ESM reforms.When the leaders of the EU member states prepared the launch of the Eurozone area, they agreed on including a special clause into the Maastricht Treaty; Article 125 TFEU prohibits sovereign bailouts among EMU member states.The dynamics of the sovereign-debt crisis, however, forced EMU members to introduce measures of mutual financial assistance. While the introduction of the Greek Loan Facility (GLF) solved the most immediate problems of the currency union, it was also clear that it would not be sufficient to restore trust in international markets for sovereign bonds on a long-​term basis. More permanent measures were needed, and with the GLF, the first precedent of financial resources getting collected for the purpose of mutual financial assistance was created, which opened paths towards more far-​reaching policies (Schlosser 2019: 76). In 2010, Eurozone members agreed on setting up the European Financial Stability Facility (EFSF) and put it under the authority of a public limited liability company based in Luxembourg.This legal construction 158

Governance structures Table 10.1  New governance facilities in EMU and how EU members outside the currency area are represented within them Area of reforms Financial assistance

Institutional structures introduced or reformed

-​  European Stability Mechanism (ESM) (new institution outside the EU framework) Banking -​  Single Supervisory Mechanism supervision and -​  Supervisory Board within resolution the ECB -​  Single Resolution Mechanism -​  Single Resolution Board (SRB) (new EU agency) -​  ECB Governing Council provided with new tasks regarding banking resolution Informal Eurozone -​  Euro Summit (new Eurozone formats within Composition of the European the two councils Council) -​  Eurogroup -​  Eurogroup Working Group

Representation of Eurozone outsiders which did not join individual projects -​  European Stability Mechanism (ESM) • Board of Governors → no representation • Board of Directors → no representation -​  Supervisory Board within the ECB • No representation, only under close cooperation -​  Single Resolution Board (SRB) • Board of Governors → no representation, only under close cooperation -​  ECB Governing Council • No representation, also not under close cooperation -​  Euro Summit •  Occasional representation if Eurozone outsiders have signed the Fiscal Compact and if the particular meeting is concerned with specific topics -​  Eurogroup • Inclusive format existing, but meeting rhythm and specific topics remain undetermined -​ Eurogroup Working Group • Inclusive format existing, but meeting rhythm and specific topics remain undetermined

enabled the Eurozone members to create a financial facility over €440 billion while at the same time maintaining the authority in their own hands. In 2012, the EFSF was in most of its tasks replaced by the ESM. How exclusive the organisational structures of the ESM are to Eurozone outsiders becomes obvious when looking into its governance structures. The Board of Governors (BoG) is in charge of the most important decisions within the fund.This board is staffed by the same national ministers who also gather in the Eurogroup. The Board of Directors (BoD) whose members are in charge of administration and implementation is staffed by the members of the Eurogroup Working Group (EWG). So, the new fund which Eurozone members have established for the purpose of mutual financial assistance is under the exclusive authority of the group’s own informal gatherings within the Council of the EU. National representatives from non-​Eurozone members have no access to the crucial boards of the ESM. Yet, when it comes to the inclusiveness of the ESM, two additional facts are worth mentioning. Firstly, with the European Commission a supranational institution is a key actor when it comes to matters of implementing and monitoring the procedures of financial assistance. Liquidity from the ESM does only come in combination with requirements for the countries requesting financial support. They have to commit themselves to fiscal and economic reforms. Here, the European Commission plays an important role. It undertakes an assessment of the country’s economic as well as fiscal situation, 159

Alexander Schilin

and on the basis of evidence found, it negotiates a Memorandum of Understanding (MoU). This document determines which reforms the indebted country needs to undertake. In cooperation with the European Central Bank (ECB) and the International Monetary Fund (IMF), the European Commission also monitors the implementation of this agreement (Bauer and Becker 2014). Secondly, there is a possibility for non-​Eurozone members to join the ESM. Indeed, Article 44 of the ESM treaty states that the BoG shall accept the application of any state which is already a member of the EU. After the successful completion of the accession process, the new members can receive shares from the fund.They can also appoint representatives to the BoG and the BoD. Still, as long as they are not a full member, there are no formal channels for Eurozone outsiders to influence ESM politics.

Banking supervision and banking resolution –​the Banking Union (BU) The organisational structures that emerged in the context of the establishment of the BU are another manifestation of the departure from the principle of inclusiveness within EMU. While the reforms with regards to financial assistance can be understood as incidents of path dependence and incremental change (Salines et al. 2012; Verdun 2015), the creation of the BU unfolded in a rather abrupt manner (Glöckler et al. 2017). During a meeting held in June 2012, the European Council decided that the BU should be based on five components: a single rulebook on banking and liquidity, a single framework for banking supervision, a single framework for the managed resolution of banks and financial institutions, a common deposit guarantee scheme and finally a common backstop for temporary financial support (Howarth and Quaglia 2014). At the time of writing (early 2021), three out of these five goals have been realised.The single rulebook and the single framework for banking supervision have been accomplished through the establishment of the Single Supervisory Mechanism (SSM). The respective legislative package entrusted the ECB with the task of coordinating and overseeing banking supervision in the EU. For this purpose, within the ECB, the Supervisory Board was established. It gathers representatives of the national supervision authorities within the BU member states and representatives of the ECB. In cooperation with the ECB Governing Council, the members of this board can take numerous decisions on supervision and bank resolution. The Single Resolution Mechanism (SRM), the main instruments of which are the Single Resolution Fund (SRF) and the Single Resolution Board (SRB), was established in line with the third goal. The SRB works as an independent agency outside the EU framework. It possesses its own budget and has a constantly increasing number of staff members. The agency’s main task is to create plans for resolving the most crucial banks in the EU. In sum, although the original plan drafted in June 2012 has not been completed yet (at the time of writing) with all the reforms described, Eurozone members installed a set of expert boards, funds and agencies, which collectively is able to supervise and resolve the largest banks in Europe. Similar to the ESM, in the BU, the boards at which the new tasks are carried out are inaccessible to Eurozone outsiders that have not joined the group through an agreement of close cooperation. The very first proposal for the regulation establishing the SSM even refrained from a formal procedure through which outsiders of the Eurozone could get involved in the BU. Eurozone outsiders were able to obtain such a mechanism only through strong resistance and building an alliance with the European Parliament (Howarth and Quaglia 2013: 114–​117). Under the current framework, Eurozone outsiders can ask for a close cooperation agreement. Yet, cooperation under this status remains asymmetrical (Schimmelfennig 2016). In fact, under this status, states gain the right to send delegates to the Supervisory Board within the ECB and the board of the SRB. However, they do not obtain a seat in the Governing Council of the 160

Governance structures

ECB in which only the national central banks from the Eurozone insiders are represented (see also Schulz and Verdun, this volume). This format has the main authority when it comes to the decision of whether a bank needs to be resolved. This could be one reason why so far only two Eurozone outsiders –​Bulgaria and Croatia –​have joined the BU.

Informal Eurozone formats in the two councils –​the Eurogroup Working Group (EWG), the Eurogroup and the Euro Summit As a final example for the non-​inclusive character of the institutional structures which have been added to EMU in the context of the sovereign-​debt crisis, I review the reforms which have been implemented to the EWG, the Eurogroup and the Euro Summit. Strengthening and reforming these informal Eurozone compositions of the Economic and Financial Committee (EFC), the ECOFIN Council and the European Council, the Eurozone members have established an independent institutional framework functioning as a cohesive machinery of deliberation and consensus production (Puetter 2016). The reforms included the systematic intensification and formalisation of intergovernmental coordination on all levels of hierarchy. The format of the Euro Summit –​a gathering of the heads of the Eurozone member states, the first meeting of which has been held in an ad hoc manner in 2008 –​has been formalised through the Fiscal Compact in 2012. Under Title V, it is determined that the leaders of the states within the Eurozone come together twice a year. Meetings are chaired by a self-​elected president and prepared by the Eurogroup.Through establishing the Euro summit, Eurozone members provided themselves with a formally acknowledged body which has the political power as well as the bureaucratic support to prepare and enforce far-​reaching decisions. This development was supplemented with various measures reinforcing the position of already existing informal bodies within the Council of the EU. The Eurogroup experienced considerable expansions of its tasks such as overtaking the function of the BoG of the ESM. Furthermore, in the Lisbon Treaty, the Eurogroup has been formally acknowledged as an informal composition of the Council of the EU. The EWG became also an integral part of the Eurozone’s governance machinery. In designing their reform packages, Eurozone governments increasingly relied on the expertise gathered in this Council committee (Schlosser 2019: 93–​95). This development not only manifested itself in the increase of the number of the committee’s official meetings but also through measures of formalisation: following an earlier decision by the Euro Summit (European Council 2011), in January 2012, Thomas Wieser became the first full-​time president of the EWG providing him with an office and support staff based in Brussels. With these reform efforts, Eurozone members established a comprehensive set of structures for informal deliberation and coordination, with the help of which they were able to agree on far-​reaching reforms (Puetter 2016). The institutions which the Eurozone members have restructured and established in the course of the crisis are almost exclusively accessible to their own national representatives. Eurozone outsiders are only allowed to join the meetings of the Euro summit when their country is a signatory party to the Fiscal Compact. Furthermore, the agenda of these meetings has to deal with matters related either to the competitiveness of the EU as a whole or with institutional reforms to EMU governance. This rule has been added to the Fiscal Compact as a concession to the United Kingdom, Sweden and Poland. In the context of the crisis, these countries were concerned about Eurozone members increasingly isolating themselves from the other EU members into their own spheres of governance. The then Polish Minister of Finance, Jacek Rostowski, was especially outspoken in rallying for more inclusion (Rettmann 2011). It was also the Polish government which put the issue on the agenda when negotiating the Fiscal Compact 161

Alexander Schilin

(Puetter 2014: 170). Despite this limitation, coordination and deliberation fora within post-​crisis EMU remain to be very exclusive to representatives from those EU member states outside the currency area. For the Eurogroup and the EWG, there exist inclusive formats allowing Eurozone outsiders to participate as observers, but there are no written rules determining topics or rhythms in which they need to be summoned.

Summary Inclusiveness, and, in particular, making sure that every member state still has access to institutional channels through which it can make its voice heard, has been a guiding principle in designing bodies of governance and coordination within policy areas affected by differentiation (Adler-​Nissen 2009). While EMU has always been an exception to that (Puetter 2006), through the reviewed reform projects this difference became significantly larger. In the context of the sovereign-​debt crisis, the governments of those EU member states within the Eurozone have established new facilities in which representation is assigned in accordance with the group patterns inherent to differentiation rather than EU membership. EU member states outside the Eurozone are not represented in the boards of the ESM, the SRB or the ECB Governing Council and only occasionally in the meetings of the Euro Summit or the Eurogroup. This constitutes a significant departure from the principle of inclusiveness which has been applied to other policy fields such as the Schengen Area or the Common Security and Defence Policy where representatives of all EU member states can attend and, in some cases, also vote in the relevant meetings both within the Council of the EU and the relevant agencies’ governance boards. Scholars have considered this inclusiveness of institutional structures to all member states crucial in maintaining the unity within the EU (Eriksen 2018). In the next section, I propose an operationalisation which might facilitate more systematic analyses of the inclusiveness of EU institutions to outsiders of differentiation.

How to analyse the inclusiveness of EU governance bodies in the context of differentiation? To understand the impact differentiation has on processes of EU governance, numerous contributions have examined the question of whether, after opt-​out decisions, the national representatives of the respective countries are able to maintain close relationships to national officials from the other states. Scholars have shown that this is a significant indicator to determine whether outsiders can remain involved in crucial process of deliberation, governance and decision-​ making (see above). However, showcasing that many reforms implemented to EMU governance in the context of the sovereign-​debt crisis have resulted in Eurozone outsiders being more excluded from crucial processes of governance, the above section has demonstrated that, in this discussion, scholars also need to consider the inclusiveness inherent to organisational set-​ups. In this section, I propose an operationalisation of inclusiveness of EU governance bodies to outsider states within the EU (internal differentiation). My conceptual reasoning is based on the work of Lijphart (1999), who compares the institutions within national democracies in terms of their inclusiveness to political minorities. Reviewing his conceptual thoughts, I deduct an operationalisation of inclusiveness which consists of two dimensions, outsider representation and outsider access. Discussing these two analytical dimensions in more detail, I suggest indicators for each of those. The usefulness of these indicators is illustrated by examples from differentiation in EMU, in some cases complemented by examples from the area of Justice and Home Affairs (JHA). It is worth noticing that this 162

Governance structures

operationalisation of inclusiveness constitutes only a first proposal which needs more discussion before being used in empirical studies. Before proposing an operationalisation of inclusiveness in EU organisations to outsiders of differentiation, defining the term ‘outsider’ is necessary. I consider ‘outsiders’ such EU member states which are not participating in a political project or a policy within the policy area affected by differentiation. Within the EU, one can further distinguish between two types of outsiders resulting from two different types of differentiation, constitutional and instrumental differentiation (see Table 1.1 in Leruth et al. this volume). EU member states that do not participate in projects or policies due to constitutional differentiation have explicitly asked to be exempted from transferring particular competencies to the EU level.These opt-​outs have been determined in the treaties. EU member states that do not participate in projects or policies due to instrumental differentiation are in this position because they do not satisfy specific criteria they are obliged to satisfy before joining. For instance, this is the case for the Eurozone or the Schengen Area. In these policy areas, outsiders have not chosen to abstain from integration, but to join the other states, they need to fulfil the accession criteria and remain in accession procedures until they have done so. Lijphart’s work is particularly useful to get a better understanding of the concept of inclusiveness. He developed an analytical framework which can be used to categorise national democratic systems in terms of their inherent inclusiveness to political minorities (Lijphart 1999). He distinguishes between two ideal types of democracy. While in the majoritarian system, the set-​up of political institutions makes the systematic inclusion of those parties, which represent political minorities in decision-​making processes, rather unlikely, in the consensus-​based system, efforts to include several political parties into such processes are almost inevitable. In order to estimate the spatial position of actual democratic systems on a scale between these two ideal-​type systems, Lijphart relies on two analytical dimensions. The executives-​parties dimension works as an estimator for the party-​political pluralism, which is observable within the main branches of a political system. Indicators within this dimension analyse the number of one-​party governments within a country or the effective number of parties in the national parliament.The federal-​unity dimension estimates the degree to which in a democratic system political power is distributed among different levels and places. Indicators within this dimension measure the degree to which competences as well as approval rights are shared between the federal and the regional level or determine to what extent different chambers are involved in the decision-​making processes. Looking at the theoretical ideas which lie behind Lijphart’s analytical dimensions and indicators, I deduct an operationalisation of inclusiveness within political institutions based on two principles. The first principle is representation. When political minorities are represented within those places in which key processes of deliberation and decision-​making take place, they are in a better position to make their voice heard and actually exert political influence. The other principle which inclusiveness within political institutions relies on is the decentralisation of power. If decisions made in a political institution need to be approved within other chambers or competences are shared between different levels of politics, the inclusion of political minorities does not rely on representation within this particular institution only. To ensure that their claims are lodged for political minorities can also use their position within the other political institutions which are involved in the particular processes of decision-​making. This makes it more likely that eventually the concerns of political minorities are incorporated even if they are not represented within the analysed institution. In the following, I demonstrate that, for the purpose of examining the inclusiveness of EU institutions to outsiders of internal differentiation, these two principles can be translated into analytical dimensions that I call outsider representation and outsider access. 163

Alexander Schilin

The relevance of the analytical dimension of outsider representation for the analysis of the inclusiveness of EU organisations is quite intuitive. To be included in a governance body’s daily businesses and get their political interests taken into account, outsiders need to have the opportunity to send their representatives to the places at which the crucial debates are held and the important decisions are made. For examining organisations regarding the dimension of outsider representation, I propose the following four indicators. First, I propose to analyse whether, in the examined organisation, outsiders are allowed to send representatives to those places at which the crucial processes of deliberation and decision-​making take place. For instance, when analysing the ESM in terms of inclusiveness, it is relevant to recognise that Eurozone outsiders are not represented within the facility, neither at the BoG nor at the BoD. When it comes to configurations of the Council or the respective preparatory bodies, it is relevant to analyse whether outsiders can send their national officials or their ministers to the meetings taking place within these formats. For instance, in the Fiscal Compact, it is determined that, for discussing certain topics, the Euro Summit must meet in an inclusive format in which Eurozone outsiders can fully participate. The outsider representatives’ rights regarding vote and voice within the examined institution should be considered a second factor. As a third indicator, I propose to track the hierarchical position of an outsiders’ representative within her home ministry. For instance, Eurozone outsiders have a higher capacity to significantly influence the debate in an inclusive meeting of the Eurogroup, if they send the minister instead of the state secretary as this indicates that they respect the other states and are committed to make constructive contributions to the debate. Finally, within agencies, it is relevant to consider the position of an outsider’s representative within the examined organisation. The usefulness of this indicator can be demonstrated by an example from the AFSJ. Between 2009 and 2018, Robert Wainwright, a British bureaucrat, has been the Executive Director of Europol. This is a manifestation of the important role which the United Kingdom used to possess within this EU agency despite its official opt-​out from the AFSJ (Carrapiço and Trauner 2013). The analytical dimension of outsider access is based on the second principle of the above-​ mentioned conceptualisation of inclusiveness within political institutions: decentralisation of power. Even if the outsiders are not directly represented within the respective boards or committees, there might be other tools and channels through which one can make sure that in the examined EU organisation political decisions are designed in accordance with their interests. In order to examine organisations with regard to the outsider access, I propose the following three indicators. The first indicator looks at the opt-​in rights which outsiders are provided with through special provisions in the treaty’s protocols, the relevant intergovernmental agreements or legislative acts. For instance, the regulation on the SSM foresees that Eurozone outsiders are allowed to join the BU through establishing a close cooperation and, recently, Croatia and Bulgaria have made use of that option (see above). As a second indicator for determining outsider access, scholars could consider the degree to which supranational institutions are involved in the decisions-​making processes carried out within the examined organisation. In particular, when it comes to legislative acts, for outsiders, which are EU member states, the European Commission and the European Parliament can be profitable channels to get involved in political decisions and compensate for relatively weak positions within the Council of the EU. This has been visible in the context of the discussions about the set-​up of the BU (Howarth and Quaglia 2013). The third indicator is especially relevant for EU agencies. I suggest that, within the analysed institutional structure, scholars examine to what extent outsiders can get involved in the main actions which are carried out within them. Agencies fulfil a variety of executive tasks within the EU. If outsiders can get involved in individual actions which are

164

Governance structures

carried out in the context of fulfilling these tasks or even can become an official member within the agencies, this provides them with a voice within these organisations even if there is a lack of direct representation. As an illustration for the usefulness of this indicator, you might consider an example from Frontex. Despite the United Kingdom’s opt-​out from the Schengen Agreement, during the times of the country’s EU-​membership, British border control and police forces could always become involved into individual missions carried out by the agency. The United Kingdom made extensive use of this option (Foteini and Markozani 2019: 98–​ 100).This commitment to Frontex missions provided the United Kingdom with a considerable amount of political influence within the organisation, even though at the Frontex management board, the British representatives did not have any voting rights (ibid.). Table 10.2 provides an overview of the dimensions and indicators which I propose to operationalise inclusiveness of EU institutions to outsiders of differentiation. When analysing the inclusiveness based on this operationalisation, two additional considerations must be taken into account. Firstly, the institutional conditions which determine the inclusiveness of an institution to outsiders might differ among the outsider states themselves. For instance, in the inclusive format of the JHA Council, the accession states to the Schengen Area, Bulgaria, Croatia, Cyprus and Romania have full voting rights, while the permanent outsiders, Ireland and (before Brexit) the United Kingdom, may participate in meetings but have no voting rights whatsoever. Yet, there can also be differences between the members of the same subgroups of outsiders. For example, the inclusive format of the Euro Summit is only accessible to those Eurozone outsiders that have signed the Fiscal Compact. Secondly, scholars might consider that there are compensatory interactions between the different dimensions of inclusiveness. Governance bodies, which score rather low in one dimension, might be very inclusive in the other one. For instance, while the ESM allows for no representation of non-​Eurozone members within the BoG and the BoD, the initiating treaty determines a procedure through which any EU member state can become a full member of this organisation. Frontex at the same time lacks an equivalent procedure while having a very inclusive policy when it comes to the EU member states which are represented on the agency’s management board.Thus, in different organisational Table 10.2  The proposed dimensions for analysing the inclusiveness of EU organisations Analytical Dimension

Corresponding principle in Lijphart

Proposed indicators

Outsider Representation

Representation

-​ Outsiders’ representation in agencies’ governance boards or council committees -​  Outsiders’ rights regarding voice and voting -​ Hierarchical rank of outsiders’ representatives at the home ministry -​ Position of outsiders’ representatives within the examined organisation -​ Options for becoming involved in individual missions of the examined organisation

Outsider Access

Decentralisation of power

-​ Outsiders’ rights to become a full member of the examined organisation (opt-​in) -​ Involvement of supranational institutions (only relevant, if outsiders are EU member states)

165

Alexander Schilin

contexts, there might exist different mechanisms through which inclusiveness is achieved and it is not clear which one is more effective or whether they are even comparable. Many questions remain open on how to recognise the different factors which can determine the inclusiveness of EU organisations to outsiders, leave out their individual effects on the actual reintegration of outsiders into decision-​making processes. In this section, I offered some initial thoughts on how an operationalisation might look like. Yet, before scholars can start with the empirical task of analysing individual organisations within the EU, analytical dimensions and indicators need to be refined.

Conclusion Examining how differentiation affects patterns of cooperation among member states within the processes of EU governance, scholarship has so far insufficiently considered the role of institutional designs. In this chapter, I argue that, in light of the developments within EMU, scholars should study more systematically how governance processes are structured under the circumstances of differentiation and the extent to which the institutional structures are inclusive to outsiders. I have reviewed how EU member states outside the Eurozone are represented within the ESM, the BU and the restructured Eurozone formats within the Council of the EU and the European Council. The reviewed reform projects showcase the lack of organisational inclusiveness to euro outsiders within post-​crisis EMU. Only on a very limited number of occasions, EU member states outside the Eurozone are allowed to send representatives to meetings within the reviewed facilities. Considering these circumstances enables us to recognise that within the academic debate there is a need for studies examining the institutional implications of differentiation in the EU more systematically. To facilitate such studies, I have proposed an operationalisation of inclusiveness of EU governance bodies to outsiders of internal differentiation. It is based on two analytical dimensions: outsider representation and outsider access. While the former dimension aims to measure how well outsiders are represented within an EU governance body, the latter one is directed towards determining the influence they can exert on decisions made in one EU organisation through their position within another. Attaining a more comprehensive understanding of how inclusive EU governance structures are to outsiders and how this factor can be moderated can also be useful for practitioners. Designing institutional structures under the circumstances of differentiation, EU politicians face a dilemma. From a perspective of democratic legitimacy, they need to make sure that, in a policy field affected by differentiation, they give influence to those member states that are also concerned by the political decisions, i.e. insiders of differentiation (Heermann and Leuffen 2020). On the other hand, EU politicians need to make sure that differentiation does not result in outsiders being totally excluded from political interaction processes as this might have long-​ term implications for the unity among the EU member states in general. Particularly in EMU, differentiation has developed a self-​reinforcing dynamic. Countries that have abstained from joining the currency area in the first place have also declined to participate in the new initiatives that emerged following the sovereign-​debt crisis, such as the BU (Schimmelfennig 2016). Contributions studying differentiation in JHA suggest that institutional inclusiveness mitigates such dynamics of path dependence (see above). In this chapter, I have demonstrated that there are many ways to design governance boards within agencies or informal venues within the Council of the EU more inclusively. Having this collection of measures at hand might make it easier for practitioners to find solutions which accommodate both concerns. In EMU, there are two concrete measures that practitioners might consider for ensuring more inclusiveness. Firstly, the opportunity for Eurozone outsiders to join individual projects such 166

Governance structures

as the BU should be supplemented by standards ensuring that outsiders who use this option also have the same say as insiders. Croatia and Bulgaria’s decision to join the BU indicates that openness, in general, can help to disrupt patterns of path dependency in EMU and thus maintain unity among the EU member states despite differentiation. However, countries joining the BU in a close cooperation agreement should also have an equal say regarding decisions on banking resolution at the ECB Governing Council.Thus, joining the BU through an agreement of close cooperation might become more appealing to Eurozone outsiders. Secondly, Eurozone insiders might create more occasions where outsiders are represented at the informal meetings in the Eurogroup or the EWG.They might also officially determine a precise rhythm or specific topics on which meetings in these two formats must be held in an inclusive format providing Eurozone outsiders with observer status. In the Covid-​19 crisis, the inclusive format of the Eurogroup has been critical to ensuring that the assistance from the ESM for Eurozone insiders was supplemented by the New Generation EU Fund organised on an EU-​27 level. Thus, inclusiveness in informal deliberations can disrupt path dependencies, and formalisation could help to make this measure more effective.

Note 1 Only Denmark and the United Kingdom have been granted de jure with an opt-​out from the third stage of EMU. Sweden’s opt-​out from replacing the national currency by the euro is de facto. Sweden abstains from fulfilling the last conditions necessary to join the Eurozone as the Swedish people have voted against this step of integration in the context of a national referendum in 2003. Considering the treaties only, Sweden is still obliged to join the Eurozone (see also Schulz and Verdun, this volume).

References Adam, I., Berg, J., Trauner, F., Tuley, M., and Westerveen, L. (2016) The UK in Justice and Home Affairs: the engaged outsider. Vrije Universiteit Brussel. IES Policy Brief Issue, (2016/​6). Available at: www.ies.be/​ node/​3654. (Access: 20 January 2020). Adler-​Nissen, R. (2008) The diplomacy of opting out: a Bourdieudian approach to national integration strategies, JCMS Journal of Common Market Studies, 46(3), pp. 663–​ 684. doi:10.1111/​j.14685965.2008.00799.x. Adler-​Nissen, R. (2009) Behind the scenes of differentiated integration: circumventing national opt-​ outs in Justice and Home Affairs, Journal of European Public Policy, 16(1), pp. 62–​80. doi: 10.1080/​ 13501760802453239. Adler-​Nissen, R. (2014) Opting Out of the European Union: Diplomacy, Sovereignty and European Integration. Cambridge: Cambridge University Press. Bauer, M. W., and Becker, S. (2014) The unexpected winner of the crisis: the European Commission’s strengthened role in economic governance, Journal of European Integration, 36(3), pp. 213–​ 229. doi: 10.1080/​07036337.2014.885750. Bellamy, R., and Kröger, S. (2017) A demoicratic justification of differentiated integration in a heterogeneous EU, Journal of European Integration, 39(5), pp. 625–​639. doi: 10.1080/​07036337.2017.1332058. Bulmer, S. (2008) New Labour and the European Union 1997-2007. A Constructive Partner?, Working Paper FG1 2008/05, Berlin: Stiftung Wissenschaft und Politik. Avaiable at: https://www.swp-berlin.org/ publications/products/arbeitspapiere/bulmer_KS.pdf (Accessed: 20 January 2020). Carrapiço, H., and Trauner, F. (2013) Europol and its influence on EU policy-​making on organized crime: analyzing governance dynamics and opportunities. Perspectives on European Politics and Society, 14(3), pp. 357–​371. doi:10.1080/​15705854.2013.817804. Eriksen, E. O. (2018) Political differentiation and the problem of dominance: segmentation and hegemony, European Journal of Political Research, 57(4), pp. 989–​1008.doi: 10.1111/​1475-​6765.12263. European Council (2011) Statement of the Euro Summit. Brussels: European Council. Available at: www. consilium.europa.eu/​uedocs/​cms_​data/​docs/​pressdata/​en/​ec/​125644.pdf (Accessed: 16 March 2021). 167

Alexander Schilin

European Parliament (2019) Report on Differentiated Integration. (2018/​ 2093(INI)). Brussels: European Parliament. Available at: www.europarl.europa.eu/​doceo/​document/​A-​8-​2018-​0402_​EN.html (Accessed: 16 March 2021). Foteini, A., and Markozani, E. (2019) European migration policy: Frontex, differentiation and Brexit. In Gänzle, S. and Leruth, B., Trondal, J. (Eds.), Differentiated Integration and Disintegration in a Post-​Brexit Era. London: Routledge pp. 91–​109. Gidlund, J. and Jerneck, M. (1996) Svenskt EMU-​medlemskap som proaktiv strategi och integrationsdilemma. Special Report No. 17, SOU 1996:158. Stockholm: Fritzes. Glöckler, G., Lindner, J., and Salines, M. (2017) Explaining the sudden creation of a banking supervisor for the euro area, Journal of European Public Policy, 24(8), pp. 1135–​1153. doi: 10.1080/​13501763. 2016.1184296. Heermann, M. and Leuffen, D., (2020) No representation without integration! Why differentiated integration challenges the composition of the European Parliament, JCMS Journal of Common Market Studies, 58(4), pp. 1016–​1033. doi: 10.1111/​jcms.13015. Heipertz, M., and Verdun, A. (2010) Ruling Europe: The Politics of the Stability and Growth Pact. Cambridge: Cambridge University Press. Holzinger, K., and Schimmelfennig, F. (2012). Differentiated integration in the European Union: many concepts, sparse theory, few data. Journal of European Public Policy, 19(2), pp. 292–​305. doi: 10.1080/​ 13501763.2012.641747. Howarth, D. (2005) The euro-​outsiders: conclusions. Journal of European Integration, 27(1), pp. 133–​140. doi: 10.1080/​07036330400030197. Howarth, D., and Quaglia, L. (2013) Banking Union as Holy Grail: rebuilding the single market in financial services, stabilizing Europe’s Banks and ‘Completing’ Economic and Monetary Union, JCMS Journal of Common Market Studies, 51(1), pp. 103–​123. doi 10.1111/​jcms.12054. Howarth, D., and Quaglia, L. (2014) The steep road to European Banking Union: constructing the single resolution mechanism, JCMS Journal of Common Market Studies, 52(1), pp. 125–​140. doi: 10.1111/​ jcms.12178. Jensen, C. B., and Slapin, J. B. (2012) Institutional Hokey-​Pokey: the politics of multispeed integration in the European Union, Journal of European Public Policy, 19(6), pp. 779–​795. doi: 10.1080/​13501763. 2011.610694. Kölliker, A. (2001) Bringing together or driving apart the union? Towards a theory of differentiated integration. West European Politics, 24(4), pp. 125–​151. doi: 10.1080/​01402380108425468. Leruth, B., and Lord, C. (2015) Differentiated integration in the European Union: a concept, a process, a system or a theory? Journal of European Public Policy, 22(6), pp. 754–​763. doi: 10.1080/​13501763. 2015.1021196. Leruth, B., Gänzle, S., Trondal, J. (this volume) Differentiation in the European Union as a field of study. Lijphart, A. (1999) Patterns of Democracy: Government Forms and Performance in Thirty-​Six Countries. New Haven, CT:Yale University Press. Lord, C. (2015) Utopia or dystopia? Towards a normative analysis of differentiated integration, Journal of European Public Policy, 22(6), pp. 783–​798. doi: 10.1080/​13501763.2015.1020839. Manners, I. (2008) The Danish Opt Outs from the European Union: Developments since 2000. Copenhagen: Danish Institute for International Studies. Available at: https://​forskn​ing.ruc.dk/​en/​publi​cati​ons/​de-​dan​ske-​ forbeh​old-​over-​for-​den europ%C3%A6iske-​union-​udviklingen-​sid (Accessed: 16 March 2021). Marcussen, M., and Zølner, M. (2001) The Danish EMU Referendum 2000: business as usual, Government and Opposition, 36(3), pp. 379–​402. doi: 10.1111/​1477-​7053.00071. Naurin, D., and Lindahl, R. (2010) Out in the cold? Flexible integration and the political status of Euro opt-​outs. European Union Politics, 11(4), pp. 485–​509. doi 10.1111/​1477-​7053.00071. Puetter, U. (2006) The Eurogroup: How a Secretive Circle of Finance Ministers Shape European Economic Governance. Manchester: Manchester University Press. Puetter, U. (2014) The European Council and the Council: New Intergovernmentalism and Institutional Change. Oxford: Oxford University Press. Puetter, U. (2016) The centrality of consensus and deliberation in contemporary EU politics and the new intergovernmentalism, Journal of European Integration, 38(5), pp. 601–​615. doi: 10.1080/​07036337. 2016.1179293. Rettmann, A. (2011) Poland takes part in Eurozone meeting amid worry “if it’s safe to join”. EU Observer. Available at: https://​euo​bser​ver.com/​pl2​011/​32581. (Accessed: January 8 2018).

168

Governance structures

Salines, M., Glöckler, G., and Truchlewski, Z. (2012) Existential crisis, incremental response: the Eurozone’s dual institutional evolution 2007–​2011. Journal of European Public Policy, 19(5), pp. 665–​681. doi: 10.1080/​ 13501763.2011.646777. Schimmelfennig, F. (2016) A differentiated leap forward: spillover, path-​dependency, and graded membership in European banking regulation. West European Politics, 39(3), pp. 483–​502. doi: 10.1080/​ 01402382.2016.1143244. Schimmelfennig, F., and Winzen, T. (2020). Ever Looser Union?: Differentiated European Integration. Oxford: Oxford University Press. Schimmelfennig, F., Leuffen, D., and Rittberger, B. (2015) The European Union as a system of differentiated integration: interdependence, politicization and differentiation, Journal of European Public Policy, 22(6), pp. 764–​782. doi: 10.1080/​13501763.2015.1020835. Schlosser, P. (2019). Europe’s New Fiscal Union. Basingstoke: Palgrave Macmillan. Schulz, D. and Verdun, A. (this volume). Differentiation and the ECB. Stubb, A. C. G. (1996) A categorization of differentiated integration, JCMS Journal of Common Market Studies, 34(2), pp. 283–​295. doi 10.1111/​j.1468-​5965.1996.tb00573.x. Svendsen, Ø., and Adler-​Nissen, R. (2019) Differentiated (dis)integration in practice: the diplomacy of Brexit and the low politics of high politics, JCMS Journal of Common Market Studies, 57(6), pp. 1419–​ 1430. doi: 10.1111/​jcms.12960. Tekin, F. (2012) Differentiated Integration at Work: The Institutionalisation and Implementation of Opt-​Outs from European Integration in the Area of Freedom, Security and Justice. Baden-​Baden: Nomos. Tekin, F. (2017) The Area of Freedom, Security and Justice: Brexit does not mean Brexit. Policy Paper 205. Berlin: Jaques Delors Institute. Available at: www.delorsinstitut.de/2015/wp-content/ uploads/2017/09/201709013_Brexit-and-the-AFSJ_Tekin.pdf. (Accessed: May 20 2018). Verdun, A. (2007) A historical institutionalist analysis of the road to economic and monetary union: a journey with many crossroads. In Meunier, S. and MacNamara, K. R. (Eds.), Making History: European Integration and Institutional Change at Fifty. Oxford: Oxford University Press pp. 195–​209. Verdun, A. (2015). A historical institutionalist explanation of the EU’s responses to the euro area financial crisis, Journal of European Public Policy, 22(2), pp. 219–​237.doi: 10.1080/​13501763.2014.994023. Warleigh, A. (2002) Towards network democracy? The potential of flexible integration. In Farrell, M., Fella, S., and Newman M. (Eds.), European Integration in the Twenty-​First Century: Unity in Diversity? London: SAGE Publications Ltd pp. 101–​118.

169

11 Promise unfulfilled? Managing differentiated integration in EU secondary law through enhanced cooperation1 Daniela Kroll

Introduction Differentiated integration (DI) seems to have become the key word in the debate on the future of the European Union (EU). Decision makers and academics alike suggest DI to be one future way to move the process of European integration forward. In his Sorbonne speech, French President Emmanuel Macron (2017) called to “embrace” DI, since it offers a way to overcome situations of gridlock. After the Brexit referendum, German Chancellor Angela Merkel ended the long-​standing German opposition to multispeed Europe (Marhold 2018: 35). Echoing President Macron, Merkel acknowledged the necessity of DI: “[W]‌e need to have the courage for some countries to go ahead if not everyone wants to participate. A Europe of different speeds is necessary, otherwise we will probably get stuck” (BBC 2017). In the Commission’s (COM’s) White Paper on the Future of Europe (2017), DI is mentioned as one scenario for the EU’s future development Closely linked to the debate on DI is the issue of enhanced cooperation (EnC). EnC is the procedure that allows DI to be pursued in EU secondary law. Introduced in the Treaty of Amsterdam, EnC was expected to bring new impetus to the integration process by increasing the use of DI.Yet, the outlook of EnC leading to a constant use of DI was unfulfilled. More recently, EnC has been prominent on the EU’s agenda. During the negotiations on the Multiannual Financial Framework (MFF) 2021–​2027 and the COVID-​19 recovery fund, Next Generation EU, Poland and Hungary threatened to veto the budget’s adoption unless the link between the Union’s funds and the respect for the rule of law would be removed. In order to break the deadlock, it was suggested that the budget could be alternatively adopted by EnC between the willing EU member states (Valero 2020). Furthermore, EnC was discussed to be used for some of the EU’s own resources in the MFF (Rios and Morgan 2020). In July 2020, the European Council (2020) stated that the future own resources system of the EU may include a financial transaction tax (FTT). In the same month, the Council appointed the national prosecutors for the European Public Prosecutor’s Office, making this case of EnC in which 22 countries are

170

DOI: 10.4324/9780429054136-13

Enhanced cooperation

participating to become operational (Council of the European Union 2020a). These examples show the procedure’s practical relevance as well as its potential to shape the future EU. This chapter provides an overview of the existing literature on EnC with the goal to identify the main findings and research gaps that could be closed by future research. Before the literature review, the procedure of EnC in the EU Treaties is described in Part 1 of the chapter. The description focuses on the Lisbon Treaty, since EnC has not been authorized in the pre-​Lisbon period. The following literature review shows that the limited occurrence of EnC should not be mixed up with the procedure having no impact on EU decision-​making. Instead, the few cases of EnC that have been authorized tell only one part of the story. Paradoxically, the dominant function of EnC as a negotiation instrument to exert pressure on reluctant member states results in the non-​use of the procedure. In consequence, EnC has contributed to decisions being made by consensus that mask existing disagreement. EnC is not only used rarely, but also at a late point in time during the negotiation process, reflecting it as a means of last resort. Member states’ concern that EnC will result in a division or even a break-​up of the Union has hindered the use of the procedure in practice for a long time.Yet, the former sceptical EU member states seem to have changed their opinion over time; EnC is increasingly considered as a useful procedure to move integration forward. Member states’ increasing preference heterogeneity and the strong politicization of EU affairs with seemingly insurmountable conflicts have contributed to this change. Because of the small number of authorized EnC cases, the existing literature is dominated by qualitative studies that often remain purely descriptive. The lack of methodological diversity and of medium-​, let alone large-​N, research results in few generalizable findings on EnC. The full potential of EnC as study subject for EU negotiations is therefore not used. Linked to the negotiations on EnC are issues of power, interests and institutions. More concretely, studying EnC in detail can inform EU research about the different interests of the member states, the basis of their interests, the coalitions that are build and how they are formed and cooperation and competition between EU institutions. Hence, EnC provides an interesting subject of study with issues ranging over the whole EU spectrum.

A history of enhanced cooperation in the EU Treaties EnC was institutionalized under the heading of “closer cooperation” in the Treaty of Amsterdam (1997). Whereas DI was granted in the form of opt-​outs in specific policy areas for certain member states in the pre-​Amsterdam period (Philippart and Edwards 1999: 7;Thym 2005: 1735; Peers 2017: 77), Amsterdam turned DI into a constitutional principle. After the institutionalization of EnC, many expected an increased use of DI. EnC was described as “Copernican Revolution” (Gaja 1998: 857) or “the biggest innovation since Maastricht in terms of EC/​EU methodology and constitution” (Philippart and Sie Dhian Ho 2000: 300). Providing a legal instrument for DI in EU secondary law was intended to fulfil two objectives: first, it would preserve the integration dynamic against the growing risk of stagnation in view of future EU enlargement (Amtenbrink and Kochenov 2009: 1; de la Serre and Wallace 1997: 5; Philippart and Edwards 1999: 87; Weatherill 1999: 21f.) and second, EnC would prevent DI arrangements being pursued outside of the EU’s institutional framework (de la Serre and Wallace 1997: 13; Fabbrini 2013: 206). The Treaty provisions on EnC can be divided into general and pillar-​specific provisions (Philippart and Edwards 1999). While the former defines the objectives, the scope and the procedure of EnC in general terms, the latter add conditions for EnC to be used in certain pillars.

171

Daniela Kroll

According to the Amsterdam Treaty, EnC could only be used in the first and third pillar and was not allowed in the second pillar of the Common Foreign and Security Pillar (CFSP). EnC aims at “furthering the objectives of the Union and […] protecting and serving its interests” (Treaty of European Union (TEU) 2012: art. K.15). Hence, EnC was clearly defined as a “pro-​integration” instrument (Fabbrini 2013: 206; see also Thym 2005: 1735f.). Furthermore, EnC would be an instrument of “last resort” if further integration could not be achieved otherwise, i.e. through uniform integration (TEU 2012: art. K.15). This ultima ratio character was intended to push the member states into the direction of compromises first before using EnC (Hall 2000: 7). Furthermore, it implies that EnC is always the “preliminary end” of a legislative procedure that has failed (Wessels and Gerards 2018: 19).Yet, the evaluation whether the last resort condition is fulfilled has been left open with a broad room for interpretation (Wessels and Gerards 2018: 19; Peers 2017: 83; Stubb 2002: 128). Moreover, Amsterdam provided the reluctant member states with an emergency brake. The authorization of EnC by qualified majority voting (QMV) in the Council could be halted if a member state would be against the use of EnC for important national reasons (TEU 2012: art. K.12; Treaty of Establishing European Community (TEC) 1997: art. 5a). In that case, the issue would be transferred to the European Council for a unanimous decision. The institutionalization of EnC in the Amsterdam Treaty was important for the integration process by recognizing the need to deviate from the traditional Community method officially for the first time (Philippart and Edwards 1999: 105). Yet, the strict requirements hindered the use of the procedure in practice (Gaja 1998; Philippart and Edwards 1999: 105; Amtenbrink and Kochenov 2009: 4; Becerril 2009: 13). The EnC provisions were therefore simplified and clarified in the next treaty revision (Shaw 2003). Most importantly, the Nice Treaty (2001) lowered the quorum from a majority of member states to an absolute number of eight participants and allowed EnC to be used in the CFSP. Still, the conditions for EnC remained quite high (Peers 2017). Political reasons and further initiatives of DI outside of the EU, e.g. the Prüm Treaty, convinced the member states to modify the procedure again in the Lisbon Treaty (2007). Figure 11.1 illustrates the procedure of EnC in the first and third pillar distinguishing between the initiation, the authorization and the implementation stage.2 According to the Lisbon Treaty, at least nine member states submit their request to use EnC to the COM (TEU 2012: art. 20(2)). The COM then decides whether to submit a legislative proposal to the Council. This reflects the COM’s role as guardian of the Treaties who protects the Union’s interests (Amtenbrink and Kochenov 2009: 6; Thym 2005: 1741; Cantore 2011: 8). Hence, the COM remains the “only gateway” for the initiation of EnC (Tekin and Wessels 2008: 29). If the COM submits a proposal for EnC, the Council decides whether to authorize EnC. After the European Parliament (EP) has given its consent, a qualified majority in the Council is necessary for the authorization of EnC. By turning the EP into a gatekeeper and requiring QMV in the Council, the EnC procedure was “communitarised” in the Lisbon Treaty (Tekin and Wessels 2008: 29; see also Becerril 2009: 20). Furthermore, the emergency brake of transferring a decision to the European Council became limited to specific policy areas in the third pillar (i.e. judicial cooperation in criminal matters, police cooperation and the European Public Prosecutor’s Office, see TEU (2012): arts. 82, 83, 86 and 87). If EnC is authorized, the COM will submit an implementing proposal on which only the “insider” countries that are participating in EnC will be allowed to vote on. Yet, the outsiders still retain the right to participate in the deliberations on EnC matters. This provision combines the idea of “no representation without taxation” with the aim to avoid the impression that EnC creates two classes of member states or even a hardcore (Wessels and Gerards 2018: 22; Stubb 2002: 132). The principles of transparency and openness are also reflected in the 172

COM submits proposal

EP gives consent

EP refuses consent

COUN rejects EnC

COM submits proposal

IMPLEMENTATION

COUN authorizes EnC by QMV

COM does not submit proposal

AUTHORIZATION

min. 9 Member States request EnC

INITIATION

Enhanced cooperation

EnC group g p votes on implementing proposal

Figure 11.1  Enhanced cooperation in the first and third pillar in the Lisbon Treaty Notes: EnC =​enhanced cooperation, COM =​Commission, EP =​European Parliament, QMV =​qualified majority voting, COUN =​Council of the EU. Source: Treaty of Function of European Union (TFEU) 2012: art. 329(1). Own depiction. 173

Daniela Kroll

provisions on how an outside member state can join an EnC group later (Thym 2005: 1741; Cantore 2011: 9). While the COM evaluates a request to join the EnC group, the participating states determine the conditions that the outsider will have to comply with (TEU 2012: art. 332). Hence, the insiders act as “gatekeepers” for any enlargement (Tekin and Wessels 2008: 30;Wessels and Gerards 2018: 23). Two so-​called passerelle clauses aim at making the implementation stage of EnC easier: first, the voting rule in the Council can be changed from unanimity to QMV and second, a special legislative procedure can be turned into an ordinary legislative procedure (TEU 2012: art. 333). As Figure 11.2 shows, the procedure in CFSP differs slightly. In contrast to the other two pillars member states that want to establish EnC submit their request to the Council and not the COM. Moreover, the Council decides by unanimity rather than by QMV (TEU 2012: art. 329(2)). Furthermore, the COM and the High Representative of the Union for Foreign Affairs give their opinion, while the EP is only informed. Although the unanimity rule at the authorization stage was maintained, the restrictions on the use of EnC were relaxed in the Lisbon Treaty (Cremona 2009). Since the Lisbon Treaty, EnC can be used for matters with military or defense implications and does not only implement a joint action or common position that had already been agreed upon. Furthermore, the Lisbon Treaty provides for a special form of DI in the area of defense through the so-​called permanent structured cooperation (PESCO). PESCO allows member states “whose military capabilities fulfil higher criteria” (TEU 2012: art. 42(6)) to move forward. Hence, the area of defense that had previously been excluded from DI was turned into a “privileged area for experimentation with a range of flexibility regimes” (de Witte 2019: 3). Compared to EnC, PESCO is a simpler procedure and can be used quicker (Piris 2010: 277). PESCO can be agreed by QMV, does not require a minimum number of participating member states and allows for the withdrawal or suspension of a participant (Wessels and Gerards 2018: 23f.). In consequence, more effective forms of DI than EnC are available in CFSP (Cremona 2009: 16). Overall, the description of the legal provisions shows that the hurdles for EnC are still quite high, although they have been lowered over time (Amtenbrink and Kochenov 2009).The Lisbon Treaty maintains a “regulated and legally constrained regime of flexible integration” (de Witte 2019: 4). As a result, EnC occurs in practice rarely. Despite its potential broad scope of application, the procedure is the form of DI that is least used. EnC has been therefore described as the “Cinderella” of the different DI forms (Peers 2017). Since EnC has been authorized in only a few cases, the procedure has also played a minor role in the literature on DI as the next section shows.

Enhanced cooperation in the EU literature EnC plays a subordinate role within the rapidly growing literature on DI, let alone in research on EU decision-​making. EnC has only been subject to limited research, most of which is atheoretical and limited in empirical scope. More specifically, the existing research can be divided into three categories: (1) legal studies, (2) think-​tank reports and (3) empirical, analytical studies. So far, the vast majority of the existing literature belongs to the first category in which the Treaty provisions on EnC are described from a legal perspective. In comparison, only a few empirical analytical studies that test a theoretical argument through real-​time data exist. In addition, the existing empirical research on EnC is dominated by qualitative, small-​N case studies. Zooming into the first literature category, legal studies examine the Treaty provisions on EnC by describing how the legal conditions for the procedure have evolved in the EU Treaty framework over time (e.g. Peers 2017; Thym 2018; Amtenbrink and Kochenov 2009; Cremona 174

Enhanced cooperation

COM and High Representative give opinion

INITIATION

min. 9 Member States request EnC

EnC group votes on decisions

COUN rejects EnC

IMPLEMENTATION

COUN authorizes EnC by unanimity

AUTHORIZATION

EP is informed

Figure 11.2  Enhanced cooperation in the second pillar in the Lisbon Treaty Source: TFEU 2012: art. 329(2). Own depiction.

2009; Becerril 2009; Tuytschaever 1999). Furthermore, some authors have examined the legality of the use of EnC in individual cases; with particular attention being paid to those two cases that have been legally challenged by non-​participating member states, i.e. the European patent and the FTT (Fiorini 2008; Peers 2010; Fabbrini 2013; Pistoia 2014; Englisch et al. 2013). 175

Daniela Kroll

Overall, the majority of the legal studies agrees on two main messages: first, the provisions are too restrictive and prevent the occurrence of EnC in practice (Peers 2017; Weatherill 1999; Amtenbrink and Kochenov 2009; Philippart and Edwards 1999; for a different view see Thym 2018; de Witte 2019) and second, EnC has no negative impact on the legal order of the EU but can be accommodated (Thym 2005, 2018; Cantore 2011: 18; Peers 2010). The main benefit of the legal studies is the description of the evolution of the legal provisions on EnC and the discussion of the consequences of EnC or DI in general for the legal order of the EU. Yet, those studies lack a theoretical argument and ignore the politics linked to the use of EnC. However, EnC and DI in general are “both a political tool and a legal instrument” (Stubb 2002: 166). The second category consists of reports that discuss the usefulness of EnC and make suggestions how the procedure should be modified and in which policy areas it should be used (de la Serre and Wallace 1997; Beneyto 2009; von Ondarza 2013; Closa Montero 2015; Gerards and Wessels 2019). In many studies, EnC is compared to other ways of integration, i.e. uniform integration or intergovernmental Treaties outside of the Union. A broad consensus seems to exist that EnC is preferable to outside DI arrangements, since the Court of Justice has the competence to review EnC decisions and EnC must be open any time to the non-​participating member states (de Witte 2019: 11ff; Amtenbrink and Kochenov 2009). Although these reports offer refreshing, timely perspectives on the desirability of EnC, they do not qualify as theory-​ driven, empirical research. The third and final literature category consists of empirical studies. Since the authorization of the first case of EnC occurred more than a decade after the institutionalization of the procedure and EnC has been authorized in only a handful of cases so far (see Table 11.1), few empirical analytical studies exist. Furthermore, the myriad of studies lack a theoretical framework. Methodologically, the literature on EnC is dominated by qualitative, in-​depth case studies that examine either a single or a few numbers of cases. Explicit cross-​case comparisons that go beyond description are still rare (for exceptions, see Kroll and Leuffen 2015;Wessels and Gerards 2018; Fabbrini 2012). Scholarly attention has been particularly strong on the first two cases of EnC, i.e. the Rome III regulation and the European patent.The case of the FTT has increasingly

Table 11.1  Overview of the authorized cases of enhanced cooperation Date authorizing proposal adopted

Date implementing proposal adopted

Number of Policy area participants (September 2020)

Legally challenged?

Rome III European patent

12 July 2010 10 March 2011

20 December 2010 17 December 2012

17 26

No Yes

Financial transaction tax Property regime rules for international couples European public prosecutor’s office

22 January 2013

–​

10

JHA Single Market Finance

9 June 2016

24 June 2016

18

JHA

No

12 October 2017

22

JHA

No

176

Yes

Enhanced cooperation

gained attention because of the controversy that surrounds the long-​standing non-​implementation of EnC (see Fabbrini 2014). Notwithstanding the advantages of qualitative research, the literature on EnC suffers from general conclusions that go beyond idiosyncratic, descriptive findings. The common qualitative approach can be partially explained by the small number of EnC cases. Until the time of writing (September 2020), EnC has been authorized in five cases (see Table 11.1). However, those authorized cases tell only one part of the story. Research would benefit from examining the “negative” cases of EnC in which the application of the procedure was seriously considered at some point in the negotiation process but did not eventually occur (see Kroll and Leuffen 2015). Analysing negative cases could tell us more about the dynamics that prevent DI and about the controversial side of DI in general. Furthermore, what is still missing in the literature is a study of the reasons that trigger the discussion on the use of EnC. What makes an issue in principle eligible for EnC? Having described the existing categories of literature, let’s turn to the main findings.

What we know so far about enhanced cooperation in practice and what is missing The procedure’s limited occurrence in practice should not be equated with EnC having no impact. EnC has been used as a negotiation tool to overcome situations of deadlock (Philippart and Sie Dhian Ho 2000; Cantore 2011: 10; Cremona 2009: 5; Thym 2005: 1737). As pointed out by Stubb (2002: 165), EnC is “like a nuclear weapon, it can be used as a threat but it is unlikely to be used in practice”. The sheer existence of the procedure, thus, affects the decision-​ making dynamic. This also implies that those cases in which EnC has been approved should be interpreted to be only the tip of the iceberg. The real impact of the procedure is hidden underneath the surface, often leaving few empirical traces. The majority of the literature has focused on explaining the use or non-​use of EnC. Besides, the high legal hurdles decision-​makers have deliberately refrained from using the procedure. “[P]‌olitical reasons” to avoid the emergence of a “two-​speed Europe” have prevented EnC (Piris 2010: 89). Wessels and Gerards (2018: 16) argue that the “output failure” of EnC has been due to member states’ strategic reluctance to use an instrument that calls the consensus culture of the EU into question (see also de Witte 2019: 6f.). For them, the use of EnC over time shows the “maturing of […] governments’ attitudes towards EnC as a not only a [sic!] legal but also a legitimate tool of problem-​solving at the EU level” (Wessels and Gerards 2018: 16). Drawing on the work of Kölliker (2006), some authors argue that the occurrence of EnC depends on the cost-​benefit calculations of the member states that vary according to the type of good. Based on a formal model, Bordignon and Brusco (2006) show that EnC can be Pareto-​ improving, if the outsiders are compensated for the change in the status quo. Likewise, Hvidsten and Hovi (2015) show in a game-​theoretic model that uniform integration persists, when it entails excludable benefits with positive externalities for the members. Kroll and Leuffen (2015) examine whether the occurrence of EnC depends on the expected externalities for the insiders and outsiders empirically in nine cases. In addition to three cases in which EnC has been authorized, they examine six negative cases in which EnC was highly likely but did not occur. Their cross-​case analysis shows that EnC is prevented and uniform integration prevails, if the outsiders expect negative externalities from the use of EnC. In contrast, EnC is likely in the case of neutral external effects. Most studies rely on member states’ preferences to explain the (non-​)occurrence of EnC. Kastner (2018) and Kalaitzake (2017) examine the impact of non-​state actors on EnC in the case 177

Daniela Kroll

of the FTT. The FTT is a unique case of EnC, insofar as the implementation is still pending at the time of writing,3 although the authorization has occurred already in 2013. The policy shift away from EnC has been explained by the lobbying success of the financial industry (Kalaitzake 2017; Kastner 2018). Similarly, Hardiman and Metinsoy (2019) show how the Irish government has been responsive to the arguments of the financial services industry. While evidence exists that interest groups have mattered for the FTT, the question whether non-​state actors have been active in other EnC cases as well remains open. With regard to the question of which countries participate in EnC, the existing cases show that EnC is not used by a fixed core group. Instead, the composition of the participants varies. The creation of a permanent, closed group is at the same time prevented by the Treaty that requires EnC to “be open any time to all Member States” (TEU 2012: art. 20(1)). In addition, decision-​making by EnC is transparent, as the outsiders are allowed to participate in the deliberations (TEU 2012: art. 20(3)). Overall, EnC is characterized by “third party friendly rules” that guarantee the outsiders that the unity of the Union will be preserved (Cantore 2011: 16f.). This corresponds to research findings on other DI forms that show that outsiders are not necessarily sidelined and may exert influence (Naurin and Lindahl 2010). Yet, the principle of openness that characterizes EnC is also risky. Latecomers might solely join the EnC group for the purpose to destroy it from within (Amtenbrink and Kochenov 2009: 14). However, while the issue of later accession to an existing EnC scheme has been explained by centripetal effects (Kölliker 2006), little has been said about the possibility of a member of an EnC group scheme to withdraw and the possible implications of such a decision. In fact, the evolution of an EnC group is still an open question, since studies usually end with the authorization decision of EnC. Yet, the withdrawals of Greece from the Rome III regulation and of Estonia from the FTT turn the question of withdrawal from a theoretical exercise into an empirical question. Likewise, the question whether the benefits differ between initial participants and latecomers has not yet been examined empirically (Bordignon and Brusco 2006). Despite the variation in the composition of EnC, some patterns seem to exist –​even though this needs to be treated cautiously given the small number of EnC cases. While the countries that belong to the Eurozone constitute the core of EnC, the countries with Treaty opt-​outs (i.e. Denmark, Ireland and the United Kingdom) form the periphery (Wessels and Gerards 2018: 26). Furthermore, Northern and Eastern member states participate less often in EnC than Western and Southern countries. Interestingly, France and Germany have participated in all EnC cases from the beginning. They are, therefore, “drivers of EnC” (Wessels and Gerards 2018: 26; see also Krotz and Schild 2018: 12). Another discussion in the literature concerns the policy areas in which EnC should be used. In general, it has been argued that EnC should only be used when member states disagree on whether the Union should act and not on how to act (Kuipers 2012: 213; Fabbrini 2012: 13). EnC is most useful in policy areas that fall under unanimity and in which the EU possesses clear competences, e.g. taxation, migration and criminal justice (de Witte 2019: 16). The latter criterion is also enshrined in the Lisbon Treaty, since EnC can only occur in policy fields of the Union’s non-​exclusive competence (TEU 2012: art. 20(1)). Economic governance has been put forward as a plausible candidate for EnC. EnC could be used to integrate the Fiscal Compact or the European Stability Mechanism that were adopted through intergovernmental Treaties outside the Union (Fabbrini 2012: 14; de Witte 2019: 9; Schwarz 2014).4 Empirically, EnC has been mostly used in policy areas in which primary law opt-​outs exist, with three of the five cases being in the area of Justice and Home Affairs (JHA). Since DI in primary law seems to act as a “catalyst” for DI in secondary law, EnC is likely to occur in those policy areas in the future as well (Wessels and Gerards 2018: 15). 178

Enhanced cooperation

Scholarly agreement seems to exist with regard to the policy areas in which EnC should not be applied. EnC should be excluded from foreign policy, as the added value of action at the EU level in that area lies in member states’ ability to speak with one voice (Thym 2018: 870f; Cremona 2009: 15). The use of EnC in the Single Market is controversial within the literature. While EnC should be used cautiously in Community policy areas to protect the rights of the EU institutions according to de la Serre and Wallace (1997: 19), EnC should be prohibited in the Single Market according to Thym (2005: 1736). EnC in the Single Market could lead to social dumping or protectionism. Instead, the procedure should be limited to areas linked to the creation of a political union, e.g. security and defence, JHA, social affairs and tax harmonization. For the EP (2019), DI depends on the level of politicization in a policy field. While DI is least likely in fields with low politicization, e.g. the internal market, it is “most likely to arise in policy areas characterised by deep political polarization, such as monetary policy, defense, border control, fundamental rights or taxation” (European Parliament 2019: G).

Concluding remarks Although DI is one of the most hotly debated ideas amongst EU officials and scholars, there is a lacuna in the literature dealing with EnC. Put simply, EnC leads a shadowy existence in the literature on DI. This disregards the procedure’s potential to shape future European integration and the impact it had on past EU negotiations. One central reason for the lack of scholarly attention is the limited occurrence of EnC in practice. Until this paper’s submission (March 2021), only five cases have been approved. In addition, thinking about EnC has been for a long time a theoretical exercise with the first case of EnC being approved more than a decade later after the institutionalization of the procedure in the Amsterdam Treaty. However, it is a fallacy to assume that EnC is meaningless because of its limited occurrence. EnC has been used as a negotiation tool. The possibility of some member states moving forward through EnC has sometimes resulted in decisions with all member states eventually participating. Hence, uniform integration has been created by the sheer prospect of EnC. EnC is not only a feasible option to move the integration process forward, but is it also an attractive option that offers clear benefits to the whole Union over DI arrangements concluded outside of the EU. Research on EnC is often atheoretical, legalistic and limited in empirical studies. Only a few empirical analytical studies that test a theoretical argument through empirical data exist. Methodologically, homogeneity exists in the field: qualitative studies of a single or a small number of cases are abundant. No one, to date, has conducted a large-​N analysis on EnC. Even though EnC has only been authorized in five cases and the topic therefore tilts towards qualitative research, a large-​N analysis could be conducted if the EnC cases would be put into a broader perspective with EU secondary law. The question why the authorization of EnC occurs or is discussed for some legislative files has not yet been considered. Previous analysis of the participating member states in EnC shows that the composition of the EnC group varies. Against initial fears, EnC is not used by a fixed number of member states that leave others deliberately behind. Instead, the transparency of the deliberations and the openness to admit outsiders later makes EnC prone to an encompassing number of participants. Furthermore, the existing cases show that the group of participating countries changes. Yet, a detailed analysis of how participation in EnC evolves over time is still missing. Future research might therefore engage more closely with the questions why initial outsiders join later or why some insiders leave the group. From a theoretical point of view, coalition building theories and negotiation theories in general might be useful here. Examining participation in EnC would also 179

Daniela Kroll

offer a fruitful line for future research as an analysis of the EnC cases could carry implications for other DI arrangements. If DI stays in the Union, but is ultimately a state of transition to the participation of all member states, taking the EnC cases as a study subject more seriously would benefit EU scholarship in general.

Abbreviations CFSP =​Common Foreign and Security Policy COM =​Commission COUN =​Council of the EU DI =​Differentiated integration EnC =​enhanced cooperation EP =​European Parliament EU =​European Union FTT =​Financial transaction tax JHA =​Justice and Home Affairs MFF =​Multiannual Financial Framework PESCO =​Permanent Structured Cooperation QMV =​Qualified majority voting

Notes 1 Disclaimer:This article does not represent the views or opinions of the German Bundestag, of any of its bodies or of the administration of the German Bundestag. The article represents solely the opinion of the author. 2 Despite acknowledging that the pillar structure was formally abolished in the Lisbon Treaty, the distinction is maintained here, since the rules and procedures in CFSP are still different (Piris 2010: 260; Cremona 2009: 7). 3 In November 2020, the COM declared that the negotiations will be either finalized by the end of 2022 or that it will put forward a proposal for a new FTT as an EU own resource (Council of the European Union 2020b). 4 Interestingly, EnC is explicitly mentioned as a future decision-​making mechanism in the Fiscal Compact. According to Article 10 of the Treaty, the participating countries “stand ready to make active use […] of enhanced cooperation […] on matters that are essential for the proper functioning of the euro area, without undermining the internal market” (European Union 2012).

References Amtenbrink, F. & Kochenov, D. (2009).Towards a More Flexible Approach to Enhanced Cooperation. In: A. Ott & E.Vos (eds.) 50 Years of European Integration: Foundations and Perspectives. The Hague: T.M.C. Asser Press, pp. 181–​200. BBC. (2017). EU Leaders Embrace Multi-​Speed Europe Amid Tensions, bbc.com. Available at: www.bbc. com/​news/​world-​europe-​39192045 [Accessed: 29 September 2020]. Becerril, B. (2009). The Regulation of Enhanced Cooperation and Its Reform in Lisbon: Towards a Model of Differentiation that Is Closer to the Community Method. In: J.M. Beneyto (ed.) Unity and Flexibility in the Future of the European Union:The Challenge of Enhanced Cooperation. Madrid: CEU Publishers: pp. 9–​22. Beneyto, J.M. (ed.) (2009). Unity and Flexibility in the Future of the European Union:The Challenge of Enhanced Cooperation. Madrid: CEU Publishers. Bordignon, M. & Brusco, S. (2006). On Enhanced Cooperation. Journal of Public Economics, 90(10–​11), pp. 2063–​2090. doi: 10.1016/​j.jpubeco.2006.02.001 Cantore, C.M. (2011).We’re One, but We’re Not the Same: Enhanced Cooperation and the Tension between Unity and Asymmetry in the EU. Perspectives on Federalism, 3(3), pp. 1–​21. Available at: http://​on-​fed​eral​ ism.eu/​atta​chme​nts/​103_​d​ownl​oad.pdf [Accessed: 9 April 2020]. 180

Enhanced cooperation

Closa, M.C. (2015). Flexibility Mechanisms in the Lisbon Treaty. European Parliament PE 536.474. Available at: www.europarl.europa.eu/​RegData/​etudes/​STUD/​2015/​536474/​IPOL_​STU(2015)536474_​EN. pdf [Accessed: 16 September 2020]. Council of the European Union. (2020a). EU Public Prosecutor’s Office (EPPO): Council Appoints European prosecutors. Brussels. Available at: www.consilium.europa.eu/​en/​press/​press-​releases/​2020/​07/​27/​eu-​ public-​prosecutor-​s-​office-​eppo-​council-​appoints-​european-​prosecutors/​ [Accessed: 29 September 2020]. Council of the European Union. (2020b). Note from Presidency to Delegations. Joint and Unilateral Declarations. 12793/​20. Brussels. Available at: www.consilium.europa.eu/​media/​47387/​st12793-​en20.pdf. [Accessed: 21 March 2021]. Cremona, M. (2009). Enhanced Cooperation and the Common Foreign and Security and Defence Policies of the EU. EUI Working Papers Law No. 21. Available at: https://​cad​mus.eui.eu/​bitstr​eam/​han​dle/​1814/​ 13002/​LAW_​2009​_​21.pdf?sequ [Accessed: 9 April 2020]. de La Serre, F. & Wallace, H. (1997). Flexibility and Enhanced Cooperation in the European Union: Placebo Rather than Panacea? Notre Europe. Available at: www.institutdelors.eu/​wp-​content/​uploads/​2018/​01/​etud2-​ delaserrewallace-​en.pdf [Accessed: 9 April 2020]. De Witte, B. (2019).The Law as Tool and Constraint of Differentiated Integration. Robert Schuman Centre for Advanced Studies Research Paper No. RSCAS, 47. Available at: https://​cad​mus.eui.eu/​bitstr​eam/​han​dle/​ 1814/​63604/​RSCAS_​2019​_​47.pdf?seque​nce=​1&isAllo​wed=​y [Accessed: 9 April 2020]. Englisch, J., Vella, J. & Cédelle, A. (2013). The Financial Transaction Tax Proposal under the Enhanced Cooperation Procedure: Legal and Practical Considerations. British Tax Review, 2, pp. 223–​259. Available at: https://​ssrn.com/​abstr​act=​2284​733 [Accessed: 11 April 2021]. European Commission. (2017). White Paper on the Future of Europe. Reflections and Scenarios for the EU-​27 by 2025. Brussels. Available at: https://​ec.eur​opa.eu/​com​miss​ion/​sites/​betapo​liti​cal/​files/​white_​paper​_​ on_​the_​futu​re_​o​f_​eu​rope​_​en.pdf [Accessed: 29 September 2020]. European Council. (2020). Special Meeting of the European Council (17–​21 July 2020) –​Conclusions. EUCO 10/​20. Brussels. Available at: www.consilium.europa.eu/​media/​45109/​210720-​euco-​final-​conclusions-​ en.pdf [Accessed: 21 March 2021]. European Parliament. (2019). Differentiated Integration. European Parliament Resolution of 17 January 2019 on Differentiated Integration P8_​TA(2019)0044. Strasbourg. Available at: www.europarl.europa.eu/​doceo/​ document/​TA-​8-​2019-​0044_​EN.pdf [Accessed: 17 September 2020]. European Union. (2012). Treaty on Stability, Coordination and Governance in the Economic and Monetary Union. T/​SCG/​en. Brussels. Available at: www.consilium.europa.eu/​media/​20399/​st00tscg26_​en12.pdf [Accessed: 18 September 2020]. Fabbrini, F. (2012). The Enhanced Cooperation Procedure: A Study in Multispeed Integration. Centro Studi Sul Federalismo. Available at: www.csfederalismo.it/​attachments/​article/​998/​CSF-​RP_​Fabbrini_​ ENHANCED%20COOPERATION_​October2012.pdf [Accessed: 9 April 2020]. Fabbrini, F. (2013). Enhanced Cooperation under Scrutiny: Revisiting the Law and Practice of Multi-​Speed Integration in Light of the First Involvement of the EU Judiciary. Legal Issues of Economic Integration, 40(2), pp. 197–​224. Available at: https://​kluw​erla​wonl​ine.com/​jou​r nal​arti​cle/​Legal+​Iss​ues+​of+​Econo​ mic+​Inte​g rat​ion/​40.3/​LEIE​2013​011 [Accessed: 11 April 2021]. Fabbrini, F. (2014). Taxing and Spending in the Euro Zone: Legal and Political Challenges Related to the Adoption of the Financial Transaction Tax. European Law Review, 39(2), pp. 155–​175. Available at: https://​ pap​ers.ssrn.com/​sol3/​pap​ers.cfm?abst​ract​_​id=​2578​710 [Accessed: 11 April 2021]. Fiorini, A. (2008). Rome III –​Choice of Law in Divorce: Is the Europeanization of Family Law Going Too Far? International Journal of Law, Policy and the Family, 22(2), pp.178–​205. doi: 10.1093/​lawfam/​ ebn006. Gaja, G. (1998). How Flexible is Flexibility under the Amsterdam Treaty? Common Market Law Review, 35(4), pp. 855–​870. Available at: https://​kluw​erla​wonl​ine.com/​api/​Prod​uct/​Cit​atio​nPDF​URL?file=​ Journ​ a ls\COLA\181​ 3 46.pdf&cas​ a _​ t o​ ken=​ b 9hVo​ r bH2​ a QAA​ A AA:Ny2S3DPpoYRLHpt5Ed_​ 5JbTRcIHmx8IZjYgiKvj8lIg4ppC​PYSn​pEYK​mI0X​buXy​YO5A​lLPk​Zdw [Accessed: 15 September 2020]. Gerards, C. & Wessels, W. (2019). Enhancing ‘Enhanced Cooperation’: Constraints and Opportunities of an Inflexible Flexibility Clause, College of Europe, CEPOB 1.19, March 2019. Available at: www. coleurope.eu/​system/​tdf/​research-​paper/​gerards_​wessels_​cepob_​1-​19_​0.pdf?file=​1&type=​node&id=​ 50002&force=​[Accessed: 16 September 2020].

181

Daniela Kroll

Hall, B. (2000). How Flexible Should Europe Be? Centre for European Reform Working Paper. Available at: www.cer.eu/​sites/​default/​files/​publications/​attachments/​pdf/​2011/​cerwp7-​2794.pdf [Accessed: 15 September 2020]. Hardiman, N. & Metinsoy, S. (2019). Power, Ideas, and National Preferences: Ireland and the FTT. Journal of European Public Policy, 26(11), pp. 1600–​1619. doi: 10.1080/​13501763.2018.1539117 Hvidsten, A.H. & Hovi, J. (2015). Why no Twin-​Track Europe? Unity, Discontent, and Differentiation in European Integration. European Union Politics, 16(1), pp. 3–​22. doi: 10.1177/​1465116514557964 Kalaitzake, M. (2017). Death by a Thousand Cuts? Financial Political Power and the Case of the European Financial Transaction Tax. New Political Economy, 22(6), pp. 709–​ 726. doi: 10.1080/​ 13563467.2017.1311850 Kastner, L. (2018). Business Lobbying under Salience –​Financial Industry Mobilization against the European Financial Transaction Tax. Journal of European Public Policy, 25(11), pp. 1648–​ 1666. doi: 10.1080/​ 13501763.2017.1330357 Kölliker, A. (2006). Flexibility and European Unification.The Logic of Differentiated Integration. Oxford: Rowman & Littlefield Publishers. Kroll, D.A. & Leuffen, D. (2015). Enhanced Cooperation in Practice. An Analysis of Differentiated Integration in EU Secondary Law. Journal of European Public Policy, 22(3), pp. 353–​373. doi: 10.1080/​ 13501763.2014.956781 Krotz, U. & Schild, J. (2018). France: Germany’s Indispensable Ally in European Policy-​Making. Institut für Europäische Politik, 18(1), pp. 1–​17. Available at: http://​iep-​ber​lin.de/​wp-​cont​ent/​uplo​ads/​2018/​ 01/​GEPS-​01-​18-​Fra​nce-​Germa​nys-​Indisp​ensa​ble-​Ally-​in-​Europ​ean-​Pol​icy-​Mak​ing.pdf [Accessed: 16 September 2020]. Kuipers, J.J. (2012). The Law Applicable to Divorce as Test Ground for Enhanced Cooperation. European Law Journal, 18(2), pp. 201–​229. doi: 10.1111/​j.1468-​0386.2011.00594 Macron, E.M. (2017). Sorbonne Speech of Emmanuel Macron. Full Text–​English Version, Ouest France, September 2017. Available at: http://​intern​atio​nal.blogs.ouest-​fra​nce.fr/​arch​ive/​2017/​09/​29/​mac​ron-​ sorbo​nne-​verba​tim-​eur​ope-​18583.html [Accessed: 29 September 2020]. Marhold, H. (2018). Differentiated Integration, Reconsidered. In: M.Wächter & H. Marhold (eds.) Europe –​ Against the Tide. Baden-​Baden: Nomos: pp. 35–​42. Naurin, D. & Lindahl, R. (2010). Out in the Cold? Flexible Integration and the Political Status of Euro Opt-​Outs. European Union Politics, 11(4), pp. 485–​509. doi: 10.1177/​1465116510382463 Peers, S. (2010). Divorce, European Style: The First Authorization of Enhanced Cooperation. European Constitutional Law Review, 6(3), pp. 339–​358. doi: 10.1017/​S1574019610300022 Peers, S. (2017). Enhanced Cooperation: The Cinderella of Differentiated Integration. In: B. De Witte, A. Ott & E. Vos (eds.) Between Flexibility and Disintegration. The Trajectory of Differentiation in EU Law. Cheltenham: Edward Elgar: pp. 76–​91. Philippart, E. & Edwards, G. (1999).The Provisions on Closer Co-​Operation in the Treaty of Amsterdam:The Politics of Flexibility in the European Union. Journal of Common Market Studies, 37(1), pp. 87–​108. doi: 10.1111/​1468-​5965.00151 Philippart, E. & Sie Dhian Ho, M. (2000). From Uniformity to Flexibility. The Management of Diversity and Its Impact on the EU System of Governance. In: G. de Búrca & J. Scott (eds.) Constitutional Change in the EU. From Uniformity to Flexibility? Oxford and Portland, OR: Hart, pp. 299–​336. Piris, J.C. (2010). The Lisbon Treaty: A Legal and Political Analysis. Cambridge: Cambridge University Press. Pistoia, E. (2014). Enhanced Cooperation as a Tool to … Enhanced Integration? Spain and Italy v. Council. Common Market Law Review, 51(1), pp. 247–​260. Available at: https://​hei​nonl​ine.org/​HOL/​Land​ingP​ age?han​dle=​hein.klu​wer/​cmlr0​051&div=​12&id=​&page=​ [Accessed: 29 September 2020]. Rios, B. & Morgan, S. (2020). Council Brokers Historic Stimulus Pact as Budget Cuts, Rule of Law Retreat Plague Deal, EurActiv.com. Available at: www.euractiv.com/​section/​politics/​news/​council-​ brokers-​historic-​stimulus-​pact-​as-​budget-​cuts-​r ule-​of-​law-​retreat-​plague-​deal/​ [Accessed: 29 September 2020]. Schwarz, M. (2014). A Memorandum of Misunderstanding –​The Doomed Road of the European Stability Mechanism and a Possible Way Out: Enhanced Cooperation. Common Market Law Review, 51(2), pp. 389–​424. Available at: https://​kluw​erla​wonl​ine.com/​api/​Prod​uct/​Cit​atio​nPDF​URL?file=​ Journ​als\COLA\COLA​2014​032.pdf&cas​a_​to​ken=​3Mmmy​mbxA​4UAA​AAA:CRE4K2oS3bC7kL t4VxtYYtGN7HJGpMn7rzQ2XVwQHoCiqAM​EbJJ​P1Ma​wluw​TAsf​EJ4Q​PxQ7​l4A [Accessed: 29 September 2020].

182

Enhanced cooperation

Shaw, J. (2003). Enhancing Cooperation After Nice: Will the Treaty Do the Trick? In: M. Andenas & J.A. Usher (eds.) The Treaty of Nice and Beyond: Enlargement and Constitutional Reform. London: Hart Publishing, pp. 207–​238. Stubb, A. (2002). Negotiating Flexibility in the European Union. Amsterdam, Nice and Beyond. Basingstoke: Palgrave Macmillan. Tekin, F. & Wessels, W. (2008). Flexibility within the Lisbon Treaty: Trademark or Empty Promise? Eipascope, 2008(1), pp. 1–​ 7. Available at: http://​aei.pitt.edu/​11044/​1/​20080​5091​8444​9_​SC​OPE2​ 008%2D1%2D5_​Teki​nand​Wess​els.pdf [Accessed: 9 April 2020]. Thym, D. (2005). “United in Diversity” –​The Integration of Enhanced Cooperation into the European Constitutional Order. German Law Journal, 6(11), pp. 1731–​1748. doi: 10.1017/​S2071832200014656 Thym, D. (2018). Supranational Differentiation and Enhanced Cooperation. In: R. Schütze & T. Tridimas (eds.) Oxford Principles of European Union Law.The European Union Legal Order.Volume I. Oxford: Oxford University Press, pp. 847–​883. Treaty of Amsterdam. (1997). European Parliament Resolution on the Amsterdam, europarl.europa. Available at: www.europarl.europa.eu/​about-​parliament/​en/​in-​the-​past/​the-​parliament-​and-​the-​treaties/​treaty-​ of-​amsterdam [Accessed: 13 April 2021]. Treaty of Establishing the European Community (TEC). (1997). Consolidated Version of the Treaty Establishing the European Community. Official Journal of the European Communities, p.C325. Available at: www.jus.uio.no/​english/​services/​library/​treaties/​15/​15-​02/​eu-​treaty.xml [Accessed: 13 April 2021]. Treaty of European Union (TEU). (2012). Consolidated Version of the Treaty on European Union. Official Journal of the European Union, p.C326/​13. Available at: https://​eur-​lex.eur​opa.eu/​resou​rce.html?uri=​ cel​lar:2bf14​0bf-​a3f8-​4ab2-​b506-​fd718​26e6​da6.0023.02/​DOC_​1&for​mat=​PDF [Accessed: 13 April 2021]. Treaty of Function of European Union (TFEU). (2012). Consolidated Version of the Treaty of Function of European Union. Official Journal of the European Union. Available at: https://​eur-​lex.eur​opa.eu/​Lex​UriS​ erv/​Lex​UriS​erv.do?uri=​CELEX:120​12E/​TXT:en:PDF [Accessed: 13 April 2021]. Treaty of Lisbon. (2007). Amending the Treaty on the European Union and the Treaty Establishing the European Community. Official Journal of the European Union, p.C306/​1. Available at: https://​eur-​lex.eur​ opa.eu/​eli/​tre​aty/​lis/​sign [Accessed: 13 April 2021]. Treaty of Nice. (2001). European Parliament Resolution on Treaty of Nice and Future of European Union, europarl.europa. Available at: www.europarl.europa.eu/​sides/​getDoc.do?pubRef=​-​//​EP//​TEXT+​TA+​ P5-​TA-​2001-​0302+​0+​DOC+​XML+​V0//​EN&language=​EN [Accessed: 13 April 2021]. Tuytschaever, F. (1999). Differentiation in European Union Law. Oxford and Portland, OR: Hart. Valero, J. (2020). Commission Considers Options for Recovery Fund without Hungary and Poland, EurActiv.com. Available at: www.euractiv.com/​section/​economy-​jobs/​news/​commission-​considers-​ options-​for-​recovery-​fund-​without-​hungary-​and-​poland/​ [Accessed: 21 March 2021]. Von Ondarza, N. (2013). Strengthening the Core or Splitting Europe? Prospects and Pitffals of a Strategy of Differentiated Integration, Stiftung Wissenschaft und Politik. Available at: www.swp-​berlin.org/​fileadmin/​ contents/​products/​research_​papers/​2013_​RP02_​orz.pdf [Accessed: 16 September 2020]. Weatherill, S. (1999). ‘If I’d Wanted You to Understand I Would Have Explained It Better’: What Is the Purpose of the Provisions on Closer Co-​Operation Introduced by the Treaty of Amsterdam? In: D. O’Keeffe & P.Twomey (eds.) Legal Issues of the Amsterdam Treaty. Portland, OR: Hart Publishing, pp. 21–​40. Wessels, W. & Gerards, C. (2018). The Implementation of Enhanced Cooperation in the European Union, europarl.europa. Available at: www.europarl.europa.eu/​RegData/​etudes/​STUD/​2018/​604987/​IPOL_​ STU(2018)604987_​EN.pdf [Accessed: 9 April 2020].

183

12 Differentiation in the European Parliament United about diversity?1 Guri Rosén

Introduction This chapter explores how Members of the European Parliament (MEPs) approach differentiation. Differentiation in the European Union (EU) has often been described and discussed as a matter for member states to accommodate their various concerns whilst being able to progress the integration process. Although the possibility of differentiated dis-​integration is becoming more pressing, the focus is still largely on differentiation as a mechanism for mediating member state disparities. However, differentiation both as differentiated integration and dis-​integration has broad implications not only for the relationship between member states, but also for the lives of the EU citizens. Being directly elected by the EU citizenry, it is the role of MEPs to voice the viewpoints of their constituents. Therefore, this chapter will address how MEPs experience and evaluate processes of differentiation. The EU has gone through several crises over the last decade. The financial crisis was an acute economic emergency that severely threatened the stability of the European integration project. The migration crisis challenged the EU as a political system placing the burden sharing between member states at the top of the agenda. Finally, Brexit is an even more existential crisis, the consequences of which we do not yet know. Parallel to these consecutive and partly overlapping crises, the EU itself has been met with an increasing amount of contestation and protest. Some have argued that European integration constitutes a new structuring conflict, increasingly putting the national political systems under strain (Kriesi 2007: 85). With their success in the last elections to the European Parliament (EP), Eurosceptic parties are one of the main driving forces behind the politicization of the EU (e.g. Hutter et al. 2016). Differentiated integration has been described as a constant companion of the European project, providing a way of coping with internal and external challenges. Multi-​speed integration, variable geometry and Europe à la carte are types of differentiation that have increased steadily with the widening and deepening of the EU project (e.g. Stubb 1996; Holzinger and Schimmelfennig 2012; Schimmelfennig and Winzen 2017; Duttle et al. 2017). Schimmelfennig and Winzen (2014: 355) distinguishes two main forms of differentiation –​instrumental and constitutional. The former is driven by enlargement, often temporary and chiefly about extending 184

DOI: 10.4324/9780429054136-14

European Parliament

market freedoms, while the latter is triggered by supranational treaty revisions that allow existing member states to retract from further integration. Following the introduction to this volume: “differentiation constitutes a ‘natural’ state of affairs because of the Union’s very character as a ‘composite polity’ ” (Leruth et al. 2022). Not only enlargement and treaty revisions, but also the multiple crises have contributed in bringing differentiation to the top of the EU’s agenda. The Eurozone is an example of differentiated integration in itself, but the Eurocrisis contributed to further differentiation, for example, through the establishment of the European Stability Mechanism (ESM) and the Fiscal Compact. With Brexit, however, the end result was not differentiated integration, but disintegration, which confronted both decision makers and students of the EU with a set of unprecedented challenges (See introduction to this volume). Earlier studies on a differentiated Europe have mainly been preoccupied with differentiated integration, which is multi-​speed or a Europe in different configurations, horizontally or vertically. Although member states chose different paths at a different pace, more integration was a common denominator. Recently, however, researchers have included differentiated dis​integration to the mix, acknowledging the possibility that differentiation might also lead to less integration, not invariably to more (Leruth et al. Cross-​reference). As a result, Brexit spurred yet another debate about the Future of Europe. In 2017, the Commission issued a White Paper setting out possible paths for the EU. The five suggested scenarios describe versions of the Union that range from “Nothing but the Single Market” to “Doing Much More Together” (European Commission 2017). Another alternative was “Those Who Want More Do More”, which “allows willing Member States to do more together in specific areas” (European Commission 2017). As argued by Gänzle et al. (2019), although the scenarios do not refer to differentiation as such, they do suggest a set of possible outcomes that include both integrationist and disintegration strategies. In the academic community, this process has reignited debates about potential future polity models for the Union, with some suggesting “a ‘hard core’ Europe around a restricted number of countries (…), through soft cores of overlapping memberships in different policy communities (…), all the way to a very loose confederation” (Fabbrini and Schmidt 2019: 174). Others have focused on the normative problems of differentiation and how differentiation can and should be justified (e.g. Lord 2015; Fossum 2015; Eriksen 2019). Heermann and Leuffen (2020) argue that differentiation requires a reform of the institutional set-​up of the EU as well, particularly the role of the EP. Common to these contributions is that they link differentiation to “the very existence of the EU and the project of integration” (Fabbrini and Schmidt 2019: 173). In the same vein, this chapter will attempt to bring these two debates together, by focusing on how differentiation features in the Future of Europe debates taking place in the EP in 2018–​2019. Differentiation –​whether as differentiated integration or dis-​ integration –​has broad implications not only for member states, but also for EU’s millions of citizens. Integrating measures of different speeds and an array of complex formats, as well as possibly dis-​integrating processes, “raises questions not only of cohesion and governing ability, but of social support” (Fossum 2019: 3). Even more so, some argue, because differentiation may lead to dominance, whereby one of the parties in a relationship is or can be subject to arbitrary interference or manipulation (Fossum 2019). Looking back at how the EU handled the Eurozone crisis, for instance, the EP was sidelined during decision-​making processes, calling into question the democracy of the ensuing arrangements (Fasone 2014). Between member states, moreover, the crisis amplified the already strained relations between North and South of Europe –​debtor/​lender states (Benz 2013). These findings are reproduced in studies of public opinion patterns, with for instance increasing divisions in support of the euro. While citizens in Eurozone countries 185

Guri Rosén

continue to be supportive, support among citizens in Non-​Eurozone is declining. Furthermore, citizens in rich creditor countries maintain their support, while in the Southern member states support is significantly lowered (Hobolt and de Vries 2016: 419). This is illustrative of the type of existential issues raised, and why it is important to include the bottom-​up perspective, when considering the democratic implications of differentiation. The aim of this chapter is to investigate to what extent and in what way the directly elected MEPs approach differentiation, as a proxy of such a bottom-​up perspective, on the assumption that it is the role of MEPs to voice concerns on behalf of their constituents. How do MEPs experience and evaluate processes of differentiation, and do they perceive of differentiation as related to challenges of dominance and democracy? Parliaments are the main arena for public, political debate as well as an arena for launching concrete proposals, including efforts to counter differentiation that is perceived as dominance. What dimensions of polity differentiation do MEPs hold to be most pressing and what characterizes the putative reforms suggested to correct any wrongs? Thus, this chapter aims to study how the MEPs envision the constitutional-​democratic outline of the EU in the future as well as how it perceives of the relationship between integration, differentiation, and dominance. In the following section, the analytical framework of the chapter is presented.The framework builds on the theoretical approach developed as part of the EU3D-​project, which rests on the presumption that the EU is facing a “triangular challenge” of differentiation, dominance and democracy (Fossum 2019).2 After presenting this framework, the data and methods used in the chapter are described. Subsequently, the analysis investigates which aspects of differentiation are addressed in EP-​plenary debates, how MEPs assess differentiation arrangements, to what extent they experience them as problematic and whether they suggest reforms. The final section of the chapter suggests some implications of the empirical study, arguing that it may not be straightforward for the EP to counter differentiation perceived as dominance (Fossum 2019).

Analytical framework The analytical framework in this chapter is built on the triangle of differentiation, dominance and democracy, which constitutes the core theoretical frame of the EU3D-​project (Fossum 2019). While differentiation has become a way of solving and containing internal conflict, it is argued that in the wake of recent crises, differentiation has raised concerns about a “two-​tiered Europe”, and that member states of the second tier are “excluded from EU decision-​making while nevertheless being subject to EU decisions due to Europe’s states and societies’ high degree of interdependence and interweaving” (Fossum 2019: 3). If this is the case, differentiation might solve some problems, but also create new ones. Differentiation constitutes a three-​way relationship with dominance and democracy. As described below, differentiation might cause dominance and thereby undermine democracy, but differentiation is also a precondition for democracy and is therefore neither a positive or negative at the outset. Dominance describes a set of relations between actors where one of the parties in the relationship is or can be subject to arbitrary interference or manipulation (Fossum 2019: 3ff). This conceptualization of dominance also encompasses structural dominance, which might entail inferior legal status, material deprivation or different forms of exclusion. Dominance is a challenge to democratic procedures, as is exemplified by various events following EU’s recent crises. During the Eurozone crisis, for example, the EP was sidelined from most decision-​making processes and was granted a diminished role in the resulting policies. Despite some disparity across different measures (Rittberger 2014), the crisis “clearly challenged the existing democratic order of the EU” (Olsen and Rosén 2021: 395). The Eurozone crisis also led to what one might describe 186

European Parliament

as a dominance relationship between member states, where citizens in debtor states were felt oppressed by lender states because of the measures forced upon their countries in return for financial aid, sidelining national democratic procedures (Benz 2013). These examples illustrate how differentiation, dominance and democracy are interrelated. Differentiation may in some cases cause dominance or vice versa, differentiation and dominance can undermine democracy, but differentiation is also necessary to sustain democracy (Fossum 2019). The question raised in this chapter is how these three challenges are perceived by EU lawmakers. More specifically, the three concepts of differentiation, dominance and democracy are used to get to an understanding of how MEPs envision the constitutional-​democratic outline of the EU in the future as well as how it perceives of the relationship between integration, differentiation and dominance. As described above, differentiation has a wider remit than differentiated integration in that it also encompasses disintegration. Differentiation is not particular to the EU, however, but “refers to how modern societies have become increasingly differentiated, along territorial, functional, social, economic and political lines” (Fossum 2019: 9). This conceptualization also broadens the focus on what processes of differentiation entail. It is not confined to differences in speed, geography or policy, but can be understood as territorial and functional, both which are essentials of the differentiation of any polity. Following Fossum (2019: 9), existing literature on EU differentiated integration has paid little attention to polity differentiation, and as a result fail to understand how “the EU’s distinct pattern of differentiation shapes EU outputs and the EU’s ability to grapple with problems and challenges”. A similar argument is made by Heermann and Leuffen (2020: 1017), who claim that the academic debates on differentiated integration have been “remarkably silent with respect to questions of institutional design and democratic legitimacy”. The above conceptualization not only explicates how differentiation has consequences for democracy, but also how and why differentiation is intimately tied to democracy. All democratic political systems of the modern age rest on different forms of differentiation (Fossum 2019: 12ff): (i) decisional, or law-​making differentiation, prevents the concentration of power and refers to, for example, the relationship between the executive and the legislative assembly; (ii) functional differentiation refers to the decision-​makers’ reliance on specialist knowledge. Of democratic importance is, for example, the balance between expert autonomy and political steering and control; (iii) territorial differentiation refers to the vertical relationship between the national, regional and local government, but also to the relationship to external actors; (iv) differentiation of citizen incorporation in the political system, which refers to basic rights as well as the degree of access to collective decision-​making arenas and processes. By encompassing the latter dimension as part of the conceptualization of differentiation, it is argued, enables a consideration of differentiation “not only from the system’s perspective but from the perspective of the person” (Fossum 2019: 13). At the same time, “some forms of differentiation are coherent with democracy, other forms are not”. Differentiation can sometimes be detrimental to democracy by accommodating dominance (Fossum 2019: 20). The concept of dominance is derived from Pettit (1997: 52), who defines dominance in the following way: “[o]‌ne agent dominates another if and only if they have a certain power over that other, in particular a power of interference on an arbitrary basis”. This entails not only the opportunity to interfere with the choice between a set of options, but also limit the options available (Fossum 2019: 21). Dominance can be manifest in several aspects of a relationship, including Formal legal status; limits or constraints on the actor’s choice options; vulnerability or susceptibility to external influences; deprivation (material and emotional such as sense of self-​worth); lack of or denial of recognition; undue impositions; and forms of exclusion. 187

Guri Rosén

Dominance may be intentional and deliberate, but it can also be unintentional and unanticipated Fossum 2019: 3 Following the four dimensions of differentiation delineated above, dominance can be a result of a lack of decisional differentiation in cases of biased power balance between the government branches. By contrast, it is argued that an institutionalization of majoritarian principles or legal-​constitutional constraints can counteract dominance (Fossum 2019: 23). With functional differentiation, a hierarchical relationship where technocrats dictate the democratically elected, one set of ideas have become hegemonic, results in dominance. Segmented political systems embody a similar logic, where selections biases in how issues are perceived are entrenched (Fossum 2019: 23). With regard to the territorial dimension, the EU obviously has some specific features compared to a nation state. On territorial boundaries, for example, the EU “stands out from states by clearly deviating from the principle of functional-​territorial contiguity” (Fossum 2019: 26). On a vertical dimension, centralization of governance can entail concentration of power and hence dominance, whereas “territorial divisions can serve as important bulwarks against dominance” (Fossum 2019: 26). Finally, on the fourth differentiation dimension –​citizens’ incorporation –​dominance can result from rights deprivations as well as inequalities and entitlements (ibid.). Summing up, as conceptualized above, differentiation denotes both a status and a process, which can be understood along four dimensions –​decisional, territorial, functional and citizen incorporation differentiation. Differentiation, moreover, is neither positive nor negative, but rather a mix of a process and inherent differences, of which the consequences will vary, and in turn lead to either a strengthening of democracy (which is deemed positive) or a rise in perceived dominance. In order to answer the question of how the MEPs envision the constitutional-​democratic outline of the EU in the future as well as how they perceive of the relationship between integration, differentiation and dominance, four more specific questions are posed3: 1 What are the experiences and subjective perspectives of EU differentiation, dominance and democracy expressed by the political representatives (the subjective dimension)? 2 What dimensions of polity differentiation are debated? 3 How are proposals for differentiated integration evaluated? 4 What rectifying measures are proposed to improve functioning of democracy? How this chapter goes about answering these four questions will be described in the following section.

Data and methods The empirical analysis in this chapter relies on data gathered from the Future of Europe-​debates, taking place in the EP between 2018 and 2019.These debates were organized by former President of the EP, Antonio Tajani, in the wake of Brexit, inviting Heads of State or Government in the member states to offer their take on the Future of Europe, but also to engage in “concrete discussion” with MEPs (European Parliament Plenary 2018). In the words of Fabbrini and Schmidt (2019: 173): “If the status quo is unsustainable, then a discussion about the future of the EU is necessary”. And the EP plenary debate is where one would expect such discussion to take place. Considering the goal of uncovering how the MEPs envision the constitutional-​democratic 188

European Parliament

outline of the EU in the future as well as how they perceive of the relationship between integration, differentiation and dominance, studying the Future of Europe-​debates, provides an opportunity to investigate prevalent narratives about the EU’s constitutional structure as well as patterns of diversity within the plenary. Plenary debates were selected as a source of empirical data because they constitute arenas where MEPs put forward ideas and potential solutions, represent their constituents, demonstrate their differences and together appraise and forge parliamentary positions in an open setting (Rosén and Stie 2020, see also Stie 2013). In this way, the parliamentary plenaries play a vital part in informing the European citizenry about current policy issues (cf. Auel and Raunio 2014: 13). Finally, plenary debates reflect the ideas that MEPs have, including their visions for the Union, and which of these visions they regard as possible, acceptable and desirable. In this way, plenary debates can help shed light on which ideas become dominant and why (cf. Mehta 2011). There were 20 debates in total between 2018 and 2019, each followed by a debate in the EP. An in-​depth analysis by the European Parliamentary Research Service (EPRS) showed that there were several common themes across the different debates, one being the need to face EU’s challenges together (European Parliament 2019a). The speakers did not agree on how to do this, however. Based on the EPRS-​analysis, two debates are selected for closer analysis, the speech by French President, Emmanuel Macron, taking place on the 17 April 2018 and the speech by Polish Prime Minister Mateusz Morawiecki, which took place on the 4 July the same year. The reason for choosing these particular debates is that they represent two, quite different views on the future of the EU. During the debates, Morawiecki was advocating a “Union of Nations 2.0”, while Macron spoke out for more European sovereignty (European Parliament 2019a). Macron is the leader of one of the big, founding member states, and one of the four member states who, at the Versailles summit in March 2017, advocated multispeed integration as a way forward for the EU (de la Baume and Herszenhorn 2017). Macron also spoke quite early in the series, he covered most of the topics addressed during the debate series, and most importantly, he seems to have his heart set on reforming the EU. In several speeches during his presidential campaign as well as immediately after he assumed office, he promoted the idea of differentiated integration on several occasions. During a speech at the Sorbonne University, for example, he declared: If we are to cultivate the desire to push ahead and ensure Europe’s progress benefits everyone, we need to constantly accommodate the driving ambition of some while allowing others to move ahead at their own speed Présidence de la République (2017) Poland is the fifth largest member of the EU, according to population. It entered the Union in 2004 and has been described as a model example of economic development (Piatowski 2018). Tensions between Poland and the EU grew in the wake of the Law and Justice-​party assuming power in 2015, the subsequent rejection by the Visegrád countries (Poland, Czech Republic, Hungary and Slovakia) of the attempts to establish a new migration policy in the wake of the refugee crisis, and the conflict over Poland’s recent judicial reforms. Compared to President Macron, Poland’s Prime Minister since December 2017, Mateusz Morawiecki, has regularly promoted a different vision of the Union. His version of the EU is one where the European project is considerably scaled back. Morawiecki has criticized the EU for being too integrationist and is an ardent advocate for strengthening national sovereignty. In his first speech to the Polish parliament as prime minister, he declared: “European Union should take care of what made it a great project and a great success –​the Europe of homelands” (Polish Government 2017). He also opposes Macron on the issue of differentiated integration. In the same speech he stated: “We 189

Guri Rosén

do not want the Union to develop at two different speeds. We do not want any further divisions or leaving some behind. We do not agree to dividing Europe into the superior and the inferior” (Polish Government 2017). Based on the theoretical framework described above, the EU3D-​ project developed a codebook for the analysis of parliamentary debates (Gora et al. 2020). The codebook was initially applied to a selected set of debates, and thereafter revised, taking into account the results of the test coding. There are two coding categories (Gora et al. 2020): (i) variables/​codes (V/​C) are used for each parliamentary intervention. These general codes are designed to capture to what extent and which aspects of differentiation, democracy and dominance are brought up by MEPs. (ii) codes (C) are more fine-​grained codes to be used within an intervention.These are codes on a detailed level, enabling a more nuanced interpretation of how MEPs analyze the current state of European integration, but also which measures they propose to deal with differentiation, democracy and/​or dominance. For V/​C-​codes the unit of analysis is the entire intervention, while for C-​codes the unit of analysis is a sentence or a sequence of sentences. None of the codes are mutually exclusive. The analytical framework described above delineates four differentiation dimensions “that can be disentangled from the nation state context and made amenable to the complex multilevel EU configuration” (Fossum 2019: 5).These are decisional, territorial, functional and citizen incorporation. In the codebook, differentiation is captured through a set of specific codes (Gora et al. 2020): •

•

•

•

Reforms of citizens’ rights: advances to change citizens’ rights and status in the EU, including changes in EU citizenship. More specifically, this may entail mentions of participation rights of EU citizens in elections, utterances that address freedom of movement or residence of EU citizens in other member states, or status of third-​country nationals. Change in competences between the EU and member states: proposals of changes that will affect the relations between levels of governing within the EU.This entails suggestions for re-​ distributing powers and competencies across levels of governing, i.e. vertical differentiation. More specifically, this may encompass propositions for more decision-​making competences for the EU (supranational or intergovernmental institutions) or for national institutions. EU capacity to act: references to the EU political system’s scope of competence and the degree of functional specialization. This may involve ideas to create new policies or policy instruments, for example, suggestions to develop new EU agencies or reform existing ones. Territorial differentiation: arguments for a territorial differentiation of EU policies or institutions, i.e. a set-​up, in which not all EU member states take part in a common policy or institutional arrangement. It can also refer to the status of non-​members. Such argumentation may also include proposals to develop various aspects of territorial differentiation, either temporarily (e.g. opt-​in/​opt-​out) or permanent (e.g. Schengen), as well as rejections of existing or proposed territorial differentiation.

The codes on democracy are (Gora et al. 2020): •

•

190

Malfunctioning of democracy in the EU: this code is used if an MEP speaks about problems related to democracy within the EU as a polity, including any aspect of malfunctioning democratic procedures. Proposals for rectifying measures: arguments for measures aimed to improve democratic malfunctioning. In the event that MEPs argues for reforms of democracy in the EU, additional codes can be added to indicate:

European Parliament

• •

At what level the measure is proposed: EU, national, regional or local. Alternatively, that the MEP calls for a strengthening of democracy but does not mention a particular level. Concrete reforms of EU interinstitutional relations: advocating a strengthening or weakening the role of the EP, redefining relations between the EP and the Commission, advocating a stronger role for national parliaments vis-​à-​vis EU institutions, claims made for a stronger or weaker Commission, a stronger or weaker Council, a stronger or weaker European Council, a stronger or weaker European Court of Justice. Some MEPs might also argue for an increased use of direct democracy, such as referenda or strengthening the citizens’ initiative. Others might simply make a case for maintaining the status quo. The latter code is used if MEPs do not make any suggestions for actual reforms, but urge that the existing rules are to be respected.

The codes on dominance are (Gora et al. 2020): •

Diagnosis of any form of dominance within the EU: this is to be applied if actors use the actual term dominance (or dominated/​dominant), but also to capture more implicit forms of dominance. Such references might be: • Exclusion: actors may argue that they are excluded from relevant information, access to decisions and decision-​making arenas that affect their choices, resources and status. Instances where powerful member states take decisions informally is one example of a situation that might count as exclusion. • Illicit hierarchy: arguments that refer to illicit hierarchy, i.e. institutional arrangements where decisions are made without being democratically authorized. • Fragmentation: MEPs might, for example, refer to a breakdown of coordination, which means that governance procedures no longer follow collectively decided-​upon rules. This encompasses arguments about a lack of order creating unpredictable situations both to public and private actors. • Lack of transparency: here, arguments might be voiced concerning situations where actors know that their interests and concerns will be affected but not by whom, when and how. • Inequality: arguments about inequality can refer to denied status or recognition, as well as the denial or deprivation of rights. Moreover, this code encompasses references to inequality that results from various forms of material deprivation or incorrect redistribution. With regard to the latter, this is only coded in instances where the negative distributive effects can be attributed to an intentional act or a structural-​institutional arrangement, and not, for example, a natural disaster.

The investigation below seeks to identify which of the ideas and positions of the MEPs on differentiation, democracy and dominance are more or less prevailing, but it also focuses on the discourse, meaning how ideas are reflected and made sense of during the debate. As described above, the codes in the codebook are designed to achieve both aims, i.e. to meet the need for both qualitative and quantitative information about the case, as well as the different dimensions that are to be covered.

Analysis The analysis in this chapter aims to shed light on to what extent and in what way MEPs approach differentiation. More specifically, it aims to answer how the MEPs envision the 191

Guri Rosén

constitutional-​democratic outline of the EU in the future as well as how they perceive of the relationship between integration, differentiation and dominance. The EP’s reaction to the Commission’s White Paper was somewhat divided. Some MEPs “welcomed the Commission’s decision to outline five possible paths for the EU to take in the coming years, while others criticized it for not picking a clear preferred path or providing concrete examples” (European Parliament 2017e). Representatives such as Esteban Gonzales Pons from the European People’s Party Group (EPP) agreed with the Commission that it was important to “align expectations with reality” and that member states had to “stop blaming Europe for what Europe cannot do because it does not have the tools” (European Parliament-​ plenary 2017). On the other hand, leader of the Progressive Alliance of Socialists and Democrats (S&D) group, Gianni Pittella argued that the five scenarios provided fuel to the flame to “all those who want to waken the European Union or even get rid of it”. From Pittella’s perspective, the only possible road ahead was “to work together as Europeans and do much more together” (European Parliament-​plenary 2017). Guy Verhofstadt from the Alliance of Liberals and Democrats for Europe (ALDE) echoed the EPP’s take on the Union’s lack of capacity, European Conservatives and Reformists-​group (ECR)’s representative, Ulrike Trebesius stated that EU should direct its efforts to becoming more efficient, while concentrating on fewer policy areas: “Times have changes, we need more flexibility and to adapt our institutional set-​up”. This illustrates not only that MEPs have opposing visions of the future of the EU, but also that they evaluate proposals for differentiation differently. The EP had started the work on three resolutions of its own, before the Commission issued its White Paper. The first resolution focused on making the most of the existing Lisbon Treaty. Among the main propositions was that the Council should become a full-​fledged second legislative chamber, relying solely on Qualified Majority Voting (QMV) as decision-​making procedure, to avoid blocking vetoes. On the subject of differentiation, the resolutions states that, in the context of a more democratic institutional set-​up for the EMU, the EP “[i]‌s of the opinion that differentiated integration should remain open to all Member States” (European Parliament 2017a). The second resolution looked at how the Lisbon Treaty could be reformed, evolving and adjusting the current institutional set up. According to the rapporteur, MEP Guy Verhofstadt (ALDE), “[t]hese reports give the blueprint of what a more perfect Union should look like” (European Parliament 2017d). The resolution states that “instead of fostering the Union”, by opening up to more “à la carte solutions”, the Lisbon Treaty has “increased the complexity of the Union and accentuated differentiation within it”. The resolution therefore proposes that “the next revision of the Treaties” should end or at least drastically reduce “the practice of opt-​outs, opt-​ins and exceptions for individual Member States at EU primary-​ law level” (European Parliament 2017b). The third resolution on the budgetary capacity for the Eurozone proposed bringing the euro area economies closer together and make them more resilient to outside shocks by establishing a higher level of fiscal and budgetary capacity (European Parliament 2017c). In an opinion to the report preparing the resolution, the rapporteur for the Committee on Constitutional Affairs proposed to incorporate a suggestion to involve non-​euro area member states “although in a differentiated way and depending on the design of the budgetary capacity” (European Parliament 2016). The suggestion was not part of the ensuing resolution, still, these diverse references to differentiation and differentiated integration show that to some MEPs, these issues at the top of the agenda, while to others, they are not. Furthermore, the quotes demonstrate how the issue of differentiation is closely connected to the question of democracy in the EU. Paradoxically, it is identified as both a remedy to lack of democracy, and as a threat to it. In the first resolution on how to improve the functioning of the 192

European Parliament

Union building on the potential of the Lisbon Treaty, drafted by MEPs Elmar Brok (EPP) and Mercedes Bresso (S&D), differentiated integration is mentioned in the context of making the EMU more democratic. In the resolution on how the Lisbon Treaty could be reformed, however, primary law opt-​outs are held to have “created an opaque system of intersecting circles of cooperation and impeded democratic control and accountability”. Thus, “a differentiated path is conceivable only as a temporary step on the way towards more effective and integrated EU policy making” (European Parliament 2017b). While the resolutions were part of a package that aimed to clarify the EP’s position on the Future of Europe on the occasion of the 60th anniversary of the Rome Treaty (European Parliament 2017d), the EP also hosted a debate series on the “Future of Europe”, in 2018–​2019. As described in the data and methods-​section above, this chapter analyzes the speeches of French President Emmanuel Macron and Polish Prime Minister Mateusz Morawiecki and the ensuing debates in the EP-​plenary. The speech of Macron to the EP took place on the 17 April 2018, while the speech of Morawiecki took place on the 4 July.4 As described above, both Macron and Morawiecki had presented their thoughts on how to reform the EU on several occasions prior to visiting the EP. The question is, when the future of the EU is debated with a more open agenda, to what extent, and in what way, MEPs raise issues linked to differentiation. The Macron-​debate consisted of 40 interventions by MEPs, with three interventions by Macron during the debate. In his introductory speech, Macron drew up a background scenario describing a somber and challenging reality, ranging from Brexit and the growth of illiberalism within Europe, to the consequences of climate change and opposition to multilateral problem-​solving (European Parliament plenary 2018). These multifaceted realities, he argues “force us to rethink the rules for our collective action”. He then goes on to describe “two strong convictions”: The first is that deciding to abandon our commitment to democracy, and all that goes with it in Europe, is the wrong path to take. The second is that, within this framework, we can, and must, build a new European sovereignty by the means of which we will provide a clear and firm response to our fellow citizens that, yes, we can protect them and provide a response to this global disorder. Then President of the Commission, Jean Claude Juncker, succeeded Macron, raising the concern for a rift between the East and West of Europe, arguing: “I believe that the fate of Europe cannot be written by dividing and mounting against each other. Europe is a whole”. Macron does not address differentiation specifically in his initial speech but returns later during the debate to state the following: I always wanted, sometimes breaking the front lines that had been established, to consider that we could move forward with a hard core form that we would have predefined. In recent months, I have seen several European leaders with real ambition, who were from countries that some say are small or in the East, or this or that. There are only member countries of the European Union and it is a project logic. But let’s try to move forward with a rule that whoever does not want to move forward cannot block others. Macron’s interventions project an image of, and a plan for, a stronger Europe, which echoes the message in his speeches immediately after his election. About four months after he assumed office, he gave a speech at the Sorbonne University, where he proposed several far-​reaching 193

Guri Rosén

reforms, while also reiterating his bid for a “strong and efficient core” to act as a vanguard of further integration (Briançon 2017). These former speeches are also referred to by several MEPs during the debate. The subsequent question is whether MEPs –​confronted with the French President’s vision of the EU –​also addresses issues of differentiation and if so, what dimensions of polity differentiation are debated? Immediately after Macron’s second intervention, MEP Adina-​ Ioana Vălean (EPP) took the floor: Mr President, I represent a country that the French love a lot: Romania. The Europe of the future, Mr President, is a Europe in which no one is left behind. Therefore, the discussion about a Europe with more speeds creates confusion and weakens the confidence of citizens in the European project. Her claim is echoed by several of her Romanian colleagues from other party groups, who also claim that it is hard to convince Romanian citizens that they are considered equal to other members, with a multispeed Europe, and that it might instead lead to more Euroscepticism (MEP Maria Grapini, S&D): “How can we reform Europe”, she asks Macron, “if you want this discrepancy between Member States”? MEP Doru-​Claudian Frunzulică (S&D) adds: “We cannot have, once again, differences in development between Western Europe and Central and Eastern Europe”. Several MEPs who discuss differentiation, concentrate on aspects of territorial differentiation, which is coded in the instances where MEPs mention set-​ups, in which not all EU member states take part in a common policy or an institutional arrangement (Gora et al. 2020). What is more, all who bring up territorial differentiation reject it, and make associations to aspects of dominance. This was also visible in the debate following Morawiecki’s speech, where MEP Zdzisław Krasnodębski from the ECR, denounced the idea, “often dressed up in pro-​European rhetoric, to reduce the Union and Europe itself to the size of a non-​optimal currency area”.The above quote by MEP Adina-​Ioana Vă lean (EPP) makes reference to exclusion –​ denoting a situation where actors are excluded from access to decision-​making arrangements that will affect their resources and status. In other words, they are “left behind”. MEP Doru-​ Claudian Frunzulică (S&D) refers to dominance qua inequality in the form of material deprivation. Moreover, one could argue that both statements refer to exclusion as inequality in status and recognition, warning that Romanian citizens perceive themselves as inferior in the proposed multispeed arrangement. Other MEPs engaged in debate about decisional differentiation, i.e. made proposals to change relations between levels of governing within the EU. The very first MEP to follow Macron and Juncker in the debate was MEP Mannfred Weber (EPP) who posed the question: “why, today, is Europe under so much pressure and even in crisis? My answer is: people do not feel involved”. His answer –​or rectifying measure to use the language of the codebook –​was not to invent anything new: “We simply have to practise what is the essence of Europe, what has been fought for over generations with blood, sweat and tears; we must dare to practise parliamentary democracy in the future”. MEP Weber’s advocacy for a stronger EP with the ability to control the selection of the President of the European Commission stands in stark contrast to the MEPs who argue for alternative vertical differentiation. Syed Kamall (ECR), for example, claims: “The growing dissatisfaction amongst voters has not been because the EU has too little power, but because it has too much”. What is interesting is that MEP Kamall’s bid for transferring powers back to the member states in areas such as immigration and social policy, it is combined with a rejection of territorial differentiation:

194

European Parliament

But when countries do agree that an EU solution is the best solution, all countries should stick to the rules and not pick and choose, as we saw with the Stability and Growth Pact or with the Dublin Regulation on asylum. Although some MEPs associated dominance with territorial differentiation, far more spoke of dominance in conjunction with decisional differentiation and democratic malfunction. The latter was –​by far –​the most frequently used code, but not all MEPs spoke of dominance as an explicit democratic problem. This was largely a trait of MEPs from Eurosceptic parties, left and right. MEPs from the European United Left-​Nordic Green Left (GUE/​NGL) spoke of the need to return to a “union of nations” (MEP Patrick le Hyaric). At the opposite end of the ideological spectrum, MEPs from European of Freedom and Direct Democracy (EFDD) and European of Nations and Freedom (ENF) make similar statements: MEP Florian Philippot (EFDD) argues that Frexit is the only way forward, and that Macron’s “European catechism” is “undermining national sovereignty, the only one that is valid in a democracy”. MEP Nicolas Bay from the ENF-​g roup echo this argument, accusing Macron of pushing for: An increasingly bureaucratic and ever less democratic construction, that of an amnesic Europe, without borders and without identity (…). The peoples express their rejection of this European Union which intends to decide everything but which is incapable of meeting the challenges where it is really expected. (…) the European Union is however omnipresent when it comes to undermine the freedom of nations. Some of these ideas were also prominent during the speech by Morawiecki, who declared, “we should not succumb to any form of determinism that would make us see only one future for the European Union, forcing us to simply transfer competences from the national level to the creation of some superstate”. Some MEPs welcomed his ideas, while MEP Mannfred Weber (EPP) was bluntly dismissive of the idea of a union of nations 2.0, stating: “We don’t need another Europe. I think our Europe is already working very well. (…). Europe of egoism and nationalism cannot deliver in the interests of the people of the European Union”. During the 32 interventions that followed his speech, however, only a few MEPs addressed the Prime Minister’s visions of the future of the EU. Instead, most expressed their concern about judicial reforms and democratic developments. During the Morawiecki-​ debate, there were far fewer statements about democratic malfunctioning in the EU, than during the Macron-​debate. Some interventions made reference to what can be considered dominant relationships and illicit hierarchies.5 This combination of codes was also prevalent in the Macron-​debate,6 but without diagnosing the democratic problems this raises in any detail, or explicitly suggesting a measure that could correct the wrong. In fact, most MEPs who proposed rectifying measures to correct for the democratic malfunctions they identified, suggested strengthening democracy at the national level. Now, this analysis is far from being a representative sample of EP-​plenary debates, still, while most MEPs in these debates did identify democratic problems, their suggestions to strengthen democracy did not refer to a specific governance level. Overall, then, what are the experiences and subjective perspectives of EU differentiation, dominance and democracy expressed by MEPs? To sum up, MEPs address most aspects of differentiation, except for the aspect of citizen incorporation, which is operationalized as mentioning of citizens’ rights. At the same time, democratic malfunctioning is the most prevalent code, and one could argue that some of the aspects of citizen incorporation dimension are captured by this

195

Guri Rosén

code. When MEPs bring up territorial differentiation during the debate with President Macron, it is rejected. Territorial differentiation is also clearly linked to dominance, with particular concerns raised about exclusion. Again, this does not mean that most MEPs reject territorial differentiation. In one of the resolutions on referred to above, for example, differentiated integration is mentioned in the context of making the EMU more democratic –​in positive terms. It does, however, indicate that MEPs are not united in their appraisal of territorial diversity. Dominance is also referred to alongside decisional differentiation and malfunctioning of democracy in the EU. A frequently proposed solution among left-​wing and right-​wing Eurosceptics is to strengthen democracy at the national level by reducing the EU’s competence. While other MEPs advocate a strengthening of democracy at the EU level, many who lament the democratic condition of the EU may mention the need to strengthen democracy, but are less specific about which level should be responsible for the strengthening. Taken together, both the finding that MEPs diverge on differentiation, and are rather vague on democratic solutions, suggest that –​to citizens following these plenary debates at least –​it is not immediately obvious that the EP is leading the way in bouncing back differentiation perceived as dominance.

Conclusion According to a European Parliament-​resolution (2019b), the debate about differentiation “should not be about pro-​differentiation versus anti-​differentiation, but the best way to operationalize differentiated integration –​which is already a political reality –​within the EU’s institutional framework”. What this chapter has shown, however, is that MEPs are not united in diversity. MEPs have opposing visions of the future of the EU, and they evaluate proposals for differentiation differently. Moreover, differentiation is not equally salient to all MEPs. This has bearing on the question about the EP’s role in countering dominance caused by differentiation (Fossum 2019). What can we expect from the EP? What the findings in this chapter indicate is that MEPs have diverging views on both differentiation and its democratic implications.The analysis shows that MEPs view the issue of differentiation as closely connected to the question of democracy in the EU. However, it is identified as both a remedy to lack of democracy, and as a threat to it. This goes for several aspects of differentiation –​law-​making differentiation, function differentiation and territorial differentiation. In other words, MEPs envision the constitutional-​democratic outline of the EU in different ways, including how they perceive of the relationship between integration, differentiation and dominance. These divisions mean that the EP might be less likely to push back against forms of differentiation that engenders dominance, at least as a united front. At the same time, the EP’s impact will depend on what aspect of differentiation that is at stake. The findings underline that importance of not perceiving of the EP as a unitary actor when considering the role of the EP in the discussion on differentiation. Investigating EP-​debates, Proksch and Slapin (2010) argue that most debates reflect partisan divisions over EU integration or national differences, rather than left-​r ight conflict. The debates analyzed in this chapter point in a similar direction. There are clear national differences in how concerned MEPs are about the consequences of territorial differentiation, with representatives from Central and Eastern Europe being quite explicit about their fears and apprehensions. On issues of functional and law-​ making differentiation, on the other hand, there are clear differences between those party groups who advocate a transfer of powers from supranational institutions to national parliaments and governments and those who want to strengthen democracy at the EU level. This demonstrates

196

European Parliament

that in order to better understand differentiation of the EU as a polity and its dynamics, it is valuable to disaggregate the concept of differentiation, while also taking into account the different cleavages of position-​taking in the EP.

Notes 1 The research leading to these results stems from the EU3D project (EU Differentiation, Dominance and Democracy), which has received funding from the European Union’s Horizon 2020 Research and Innovation Programme under Grant Agreement No. 822419. 2 EU3D is a research project funded by the EU’s Horizon 2020 programme, coordinated by ARENA Centre for European Studies University of Oslo. 3 These questions are developed within the EU3D-project (Gora et al. 2020). 4 Macron was the fourth Head of State or Government to visit the EP, while Morawiecki was number eight. 5 It should be noted that it makes little sense to speak of frequency of codes used in the case of the Morawiecki-​debate, because the majority of interventions addressed matters internal to Poland. 6 MEP Bogusław Liberadzki (S&D) expressed his concern about an EU that prioritizes the interests of rich countries taking advantage over poor member states, for example.

Bibliography Auel, K. and Raunio,T. (2014). Debating the State of the Union? Comparing Parliamentary Debates on EU Issues in Finland, France, Germany and the United Kingdom. The Journal of Legislative Studies, 20(1), pp. 13–​28. doi: 10.1080/​13572334.2013.871482. Benz, A. (2013). An Asymmetric Two-​level Game: Parliaments in the Euro Crisis, in B. Crum and J. E. Fossum (eds) Practices of Inter-​Parliamentary Coordination in International Politics: The European Union and Beyond. Essex: ECPR Press: pp. 125–​140. Briançon, F. (2017). 5 takeaways from Macron’s big speech on Europe’s future, Politico.Available at:: www.politico.eu/​article/​5-​takeaways-​from-​macrons-​big-​speech-​on-​europes-​future/​. (Accessed: 2 April 2021). de la Baume, M. and Herszenhorn, D. (2017). In Versailles, Eu’s big 4 back multispeed Europe, Politico, March 6, Available at: www.politico.eu/​article/​in-​versailles-​eus-​big-​4-​back-​multi-​speed-​europe-​italy-​france-​ germany-​spain/​. (Accessed: 2 April 2021). Duttle,T., Holzinger, K., Malang,T., Schäubli,T., Schimmelfennig, F. and Winzen,T. (2017). Opting out from European Union Legislation: The Differentiation of Secondary Law. Journal of European Public Policy, 24(3), pp. 406–​428. doi: 10.1080/​13501763.2016.1149206. Eriksen, E. O. (2019). Contesting Political Differentiation. European Division and the Problem of Dominance. Basingstoke: Palgrave Macmillan. European Commission (2017). White Paper on the Future of Europe, Brussels: European Commission. European Parliament (2016). Report on budgetary capacity for the Eurozone (A8-​0038/​2017), europarl. europa. Available at: www.europarl.europa.eu/​doceo/​document/​A-​8-​2017-​0038_​EN.html. (Accessed: 14 April 2021). European Parliament (2017a). European Parliament resolution of 16 February 2017 on improving the functioning of the European Union building on the potential of the Lisbon Treaty (TA-​8-​2017-​0049), europarl.europa. Available at: www.europarl.europa.eu/​doceo/​document/​TA-​8-​2017-​0049_​EN.html. (Accessed: 14 April 2021). European Parliament (2017b). European Parliament resolution of 16 February 2017 on possible evolutions of and adjustments to the current institutional set-​up of the European Union (2014/​2248(INI)), P8_​ TA (2017)0048, europarl.europa. Available at: www.europarl.europa.eu/​doceo/​document/​TA-​8-​2017-​ 0048_​EN.html. (Accessed: 14 April 2021). European Parliament (2017c). European Parliament resolution of 16 February 2017 on budgetary capacity for the euro area (2015/​2344(INI)), P8_​TA(2017)0050, europarl.europa. Available at: www.europarl. europa.eu/​doceo/​document/​TA-​8-​2017-​0050_​EN.html. (Accessed: 14 April 2021). European Parliament (2017d). Parliament sets out its vision for the future of Europe, European Parliament press release, europarl.europa. Available at: www.europarl.europa.eu/​news/​en/​press-​room/​ 20170210IPR61812/​parliament-​sets-​out-​its-​vision-​for-​the-​future-​of-​europe. (Accessed: 2 April 2021).

197

Guri Rosén

European Parliament (2017e). Future of the EU: MEPs discuss five scenarios set out by Jean-​Claude Juncker, European Parliament press release, europarl.europa. Available at: www.europarl.europa.eu/​news/​ en/​press-​room/​20170228IPR64287/​future-​of-​the-​eu-​meps-​discuss-​five-​scenarios-​set-​out-​by-​jean-​ claude-​juncker. (Accessed 2 April 2021). European Parliament (2019a). The Future of Europe debates in the European Parliament, 2018–​2019 –​ A synthesis of the speeches by EU Heads of State or Government, European Parliament Research Service, PE 637.948, europarl.europa. Available at: www.google.no/​searchq=​The+​Future+​of+​Europe+​ debates+ ​ i n+ ​ t he+ ​ E uropean+ ​ Parliament%2C+ ​ 2 0182019+ ​ % E2%80%93+ ​ A + ​ s ynthesis+ ​ o f+ ​ t he+​ speeches+ ​ by+ ​ E U+ ​ H eads+ ​ o f+ ​ S tate+ ​ o r+ ​ G overnment%2C+ ​ E uropean+ ​ Parliament+ ​ R esearch+​ Service&sxsrf=​ALeKk03DpzTHxP. (Accessed: 02 April 2021). European Parliament (2019b) European Parliament resolution of 17 January 2019 on differentiated integration (2018/​2093(INI)). European Parliament Plenary (2017). European Parliament Plenary Session, europarl.europa. Available at: https://​eur​opa.eu/​newsr​oom/​eve​nts/​europ​ean-​par​liam​ent-​plen​ary-​sess​ion-​50_​en. (Accessed: 14 April 2021). European Parliament Plenary (2018). The calendar of European Parliament part-​ sessions, europarl. europa. Available at: www.europarl.europa.eu/​news/​en/​press-​room/​20161117IPR51552/​european-​ parliament-​2018-​calendar-​of-​plenary-​sessions-​approved. (Accessed 02 April 2021). Fabbrini, S. and Schmidt, V. A. (2019). Introduction: Differentiation and Beyond. Comparative European Politics, 17, pp. 173–​175.doi: 10.1057/​s41295-​019-​00163-​8. Fasone, C. (2014). European Economic Governance and Parliamentary Representation: What Place for the European Parliament? European Law Journal, 20(2), pp. 164–​185. doi: 10.1111/​eulj.12069. Fossum, J. E. (2015). Democracy and Differentiation in Europe. Journal of European Public Policy, 22(6), pp. 799–​815. doi: 10.1080/​13501763.2015.1020838. Fossum, J. E. (2019). Europe’s triangular challenge: differentiation, dominance and democracy. EU3D Research Papers, No. 1, December 2019. Available at: www.researchgate.net/​profile/​JohnFossum/​publication/​338063883_​Europe%27s_​Triangular_​Challenge_​Differentiation_​Dominance_​and_​Democracy/​ links/​5e0629b14585159aa49f7119/​Europes-​Triangular-​Challenge-​Differentiation-​Dominance-​and-​ Democracy. (Accessed: 02 April 2021). Gänzle, S., Leruth, B. and Trondal, J. (eds) (2019). Differentiated Integration and Disintegration in a Post-​Brexit Era, London: Routledge. Góra, M., Thevenin, E., Zielińska, K., Fossum, J. E., Hermann, M., Ostatnik,V., Rosen, G., Slartmann, K. I., Steuer, M., Timkova, N., Umit, R., von Diepen, J., Zgaga, T. (2020), Codebook for analysis of debates on the EU reform proposals in the national and European parliaments. Prepared as a part of EU3D. Democracy, differentiation and dominance project funded by the European Union’s Horizon 2020 research and innovation programme under grant agreement no. 822419 (2019–​2023). Available upon request. Heermann, M. and Leuffen, D. (2020). No Representation without Integration! Why Differentiated Integration Challenges the Composition of the European Parliament. Journal of Common Market Studies, 58(4), pp. 1016–​1033. doi: 10.1111/​jcms.13015. Hobolt, S. B. and de Vries, C. E. (2016). Public Support for European Integration. Annual Review of Political Science, 19(1), 413–​432. doi: 10.1146/​annurev-​polisci-​042214-​044157. Holzinger, K. and Schimmelfennig, F. (2012). Differentiated Integration in the European Union: Many Concepts, Sparse Theory, Few Data. Journal of European Public Policy, 19(2), pp. 292–​305. doi: 10.1080/​ 13501763.2012.641747. Hutter, S., Grande, E. and Kriesi, H. (eds) (2016). Politicizing Europe. Integration and Mass Politics, Cambridge: Cambridge University Press. Kriesi, H. (2007). The Role of European Integration in National Election Campaigns. European Union Politics, 8(1), pp. 83–​108. doi: 10.1177/​1465116507073288. Leruth, B., Gänzle, S. and Trondal, J. (2022). Differentiation in the European Union as a Field of Study, in Leruth, B., Gänzle, S. and Trondal, J. (eds) Routledge Handbook of Differentiation in the European Union London: Routledge, pp. 1–16. Leuffen, D., Rittberger, B. and Schimmelfennig, F. (2013). Differentiated Integration. Basingstoke: Palgrave Macmillan. Lord, C. (2015). Utopia or Dystopia? Towards a Normative Analysis of Differentiated Integration. Journal of European Public Policy, 22(6), pp. 783–​798.

198

European Parliament

Macron, E. (2017). Initiative for Europe: Sorbonne Speech, diplomatie.gouv. Available at: at www. diplomatie. gouv.fr/​IMG/​pdf/​english_​version_​transcript_​-​_​initiative_​for_​europe_​-​_​speech_​ by_​the_​president_​ of_​the_​french_​republic_​cle8de628.pdf. (Accessed: 02 April 2021). Mehta, J. (2011).The Varied Role of Ideas in Politics: From ‘Whether’ to ‘How’, in D. Béland and R. H. Cox (eds) Ideas and Politics in Social Science Research. Oxford: Oxford University Press: pp. 45–​46. Olsen, E. D. and Rosén, G. (2021). The EU’s Response to the Financial Crisis, in M. Riddervold, J. Trondal and A. Newsome (eds) The Palgrave Handbook of EU Crisis. London: Palgrave Macmillan: pp. 381–​400. Pettit, P. (1997). Republicanism: A Theory of Freedom and Government. Oxford: Oxford University Press. Piatowski, M. (2018). Europe’s Growth Champion: Insights from the Economic Rise of Poland, Oxford: Oxford University Press. Polish Government (2017). Policy statement by Prime Minister Mateusz Morawiecki –​stenographic record, premier.gov. Available at: https://​archi​wum.prem​ier.gov.pl/​en/​pol​icy-​statem​ent-​by-​prime-​minis​ ter-​mate​usz-​mor​awie​cki-​steno​grap​hic-​rec​ord.html. (Accessed: 02 April 2021). Présidence de la République (2017). Initiative for Europe –​Speech by M. Emmanuel Macron, President of the French Republic, diplomatie.gouv . Available at: www.diplomatie.gouv.fr/​IMG/​pdf/​english_​version_​ transcript_​-​_​initiative_​for_​europe_​-​_​speech_​by_​the_​president_​of_​the_​french_​republic_​cle8de628. pdf. (Accessed: 02 April 2021). Proksch, S.-​O. and Slapin, J. (2010). Position Taking in European Parliament Speeches. British Journal of Political Science, 40(3), pp. 587–​661. doi: 10. 101 7/​S0007 123409990299. Rittberger, B. (2014). Integration without Representation? The European Parliament and the Reform of Economic Governance in the EU. Journal of European Public Policy, 52(6), pp. 1174–​1183. doi: 10.1111/​ jcms.12185. Rosén, G. and Stie, A. E. (2020). Balancing seclusion and inclusion: EU trilogues and democratic accountability. Journal of European Public Policy, pp. 1–​22. doi: 10.1080/​13501763.2020.1858326. Schimmelfennig, F. and Winzen, T. (2014). Instrumental and Constitutional Difference in the European Union. Journal of Common Market Studies, 52(2), pp. 354–​370. doi: 10.1111/​jcms.12103. Schimmelfennig, F. and Winzen, T. (2017). Eastern Enlargement and Differentiated Integration: Towards Normalization. Journal of European Public Policy, 24(2), pp. 239–​258. doi: 10.1080/​13501763.2016.1264083. Stie, A. E. (2013). Democratic Decision-​making in the EU:Technocracy in Disguise? Abingdon: Routledge. Stubb, A. (1996). A Categorization of Differentiated Integration. Journal of Common Market Studies, 34(2), pp. 283–​295. doi: 10.1111/​j.1468-​5965. 1996.tb00573.x.

199

13 Differentiation and the European Central Bank A bulwark against (differentiated) disintegration? Daniel F. Schulz and Amy Verdun1

If monetary union goes ahead, the European Union will be divided into two groups for the foreseeable future. Malcolm Rifkind, Zürich 19962

Introduction As the guardian of the euro, the European Central Bank (ECB) manages what has long been considered the prime example of differentiated integration (Verhelst 2013; Schimmelfennig 2014). Originally known by other names, such as ‘variable geometry’ or ‘two-​speed Europe’, differentiation arose when the Maastricht Treaty established Europe’s Economic and Monetary Union (EMU) with two member states (Denmark and the United Kingdom) receiving formal opt-​outs (Thygesen 1999). Importantly, their autonomous decisions not to join the single currency did not reflect a lack of capacity to meet requirements for joining. Rather this choice reflected deep-​seated concerns among political elites and electorates about losing sovereignty in an area of core state powers. The 1992 and 2000 referendums in Denmark and the 2003 referendum in Sweden offer the clearest examples of this dynamic (Leuffen et al. 2013: 149). The ECB has overseen both a significant widening and deepening of integration in its domain during the two decades since its birth. While it was confronted with the threat of disintegration in the context of the euro area crisis, it has played its part in successfully averting the scenario of one (or more) member states leaving the euro area (Spiegel 2014; Schoeller 2018). At the height of the Greek sovereign debt crisis, there was widespread concern that the departure of one member of the euro area could lead to a domino effect –​thereby calling into question the long-​term viability of the single currency itself (Chibber 2011; Kutter 2014). From a neofunctionalist perspective, one may argue that a supranational institution, such as the ECB, would seek to advance European integration as a supranational solution to transnational problems.The euro crisis offered a window of opportunity for pursuing this goal as the neofunctionalist logic suggests (Haas 1958; Schimmelfennig 2012; Hodson 2013; Niemann and Ioannou 2015). 200

DOI: 10.4324/9780429054136-15

European Central Bank

Ultimately, the sovereign debt crisis did result in a deepening of integration, firmly establishing the ECB as the lender of last resort for sovereigns in the euro area (Buiter and Rahbari 2012; Ban 2020). New institutions were established to provide financial stability, starting with the European Financial Stability Facility which later became the European Stability Mechanism (Meunier and Gocaj 2013; Verdun 2015). Furthermore, the ECB saw its competences increase when member state leaders bestowed upon it the role of the European Union (EU)’s main banking supervisor in 2013. These developments could be an indication of EMU spilling over into closer political and economic integration (‘deepening’). At the same time, however, one could argue that furthering euro adoption among more EU member states (‘widening’) also advances the European integration objective. However, since the post-​Maastricht EU has established differentiation as a permanent feature of European integration, clashes between the objectives of widening and deepening may become inevitable (Zimmermann 2016; Patel 2019). For example, one may consider what the deepening of integration in the context of banking union implies for euro outsiders (see Schimmelfennig (2016) and Schilin (in this volume)). We define ‘widening’ in this context as the formal adoption of the euro by outsiders –​in other words the formal extension of the Eurosystem (for example the eventual adoption of the euro by Bulgaria or Croatia, see European Commission 2020). We exclude from this definition the unilateral decision to use the euro as legal tender in a third country (euroization; i.e., using the euro as a de facto currency, which happens for instance in Kosovo and Montenegro). This process of widening within EMU reduces differentiation. Conversely, we define ‘deepening’ in the context of EMU as implying that existing euro area members integrate further. As this process leads to a bigger gap between euro insiders and outsiders, such deepening increases differentiation. Following Schimmelfennig (2018: 1154) differentiated disintegration implies ‘the selective reduction of a state’s level and scope of integration’ which may possibly lead to internal differentiation if a member state remains in the EU but exits from EMU. While much of the contemporary discussion around differentiated disintegration now focuses on the case of Brexit (Leruth et al. 2019; Glencross 2021), earlier contributions discussed the spectre of disintegration in the context of a potential Greek exit (‘Grexit’) from the euro area (see Vollaard 2014). While the threat of Brexit eventually materialized and thus captured the headlines, the earlier threat of Grexit could be averted –​in large part because it spurred the ECB into action (Spiegel 2014).We thus propose to extend the discussion of differentiated disintegration to also include cases where a state’s selective reduction of integration does not ultimately happen but, instead, produces significant policy changes to avert the scenario of disintegration. Against this background, this contribution asks how the ECB has dealt with differentiated integration and the potential threat of disintegration. It analyses both the historical development of the ECB’s relationship with euro outsiders –​discerning between ‘old’ opt-​outs and ‘new’ accession countries in the context of EU enlargement –​and how differentiated integration affects ECB policymaking across its various tasks. We analyse whether ECB policies follow one or more of three logics of ‘deepening’ and ‘widening’: (1) Is the ECB encouraging euro membership among the current ‘outs’? (2) Is the ECB seeking to reduce the impact of differentiation by keeping the ‘outs’ on board as much as possible? (3) Does the ECB contribute to further cementing differentiation, for instance by excluding the ‘outs’ from decision-​making or deepening integration among the ‘ins’ only? The structure of the paper is as follows. The next section examines the history, theory, and origins of differentiation in EMU. Section three provides an overview of ECB’s bodies and how they engage with euro outsiders across its variety of tasks: monetary policy, financial stability, and support for economic policies across the union. The Section ‘The ECB and the euro 201

Daniel F. Schulz and Amy Verdun

crisis: combatting disintegration?’ discusses the ECB’s responses to episodes of crisis and its approach to differentiation in the context of EMU reform. Our conclusions offer insights into the extent to which the ECB may be regarded as an ‘agent of change’ or ‘bulwark against disintegration’. We find that the ECB has long been hesitant to proactively push further integration but was prompted into action by the inaction of member state governments during critical moments. In counteracted the threat of disintegration through both monetary policies and discursive acts. The ECB has thus, albeit often reluctantly, accepted greater responsibilities within the EU and has gradually emerged as an agent of integration.

Origins of differentiation in EMU Although the term ‘differentiation’ was developed more recently, its practice dates back to earlier periods. ‘Variable geometries’ (‘two-​speed’ and ‘multi-​speed’ European integration) were considered in the 1990–​1991 intergovernmental conferences that led to the creation of the Maastricht Treaty (Dyson and Featherstone 1999; Verdun 2000). However, the plan to create an EMU in three stages built on earlier developments, such as the Werner Report (1970), the Tindemans Report (1976), and the developments of the European Monetary System (EMS) during the 1970s and 1980s (McNamara 1998). In its history, there had already been a recognition that there were different speeds and levels of ability to cooperate more. For instance, the Tindemans Report responded to instructions by the December 1974 Paris European Council; ‘It called for a new approach to EMU, arguing that the member states in a position to go ahead with the project should do so, whereas the others would still be committed to the final objective, but would proceed according to a different timetable’ (Bulletin of the European Communities, Supplement No 1/​1976, quoted in Dyson and Quaglia 2010: 266). Monetary integration was thus differentiated from the start and also fluctuated considerably over time (Leuffen et al. 2013: 145–​147). Already in the 1970s with the creation of the European Economic Community exchange rate mechanism (ERM), informally called ‘the snake in the tunnel’, some countries would participate whereas others would not. Building on that experience the EMS was created in 1979 with the European Currency Unit (ECU), the ERM, and the European Monetary Cooperation Fund (EMCF) at its core. The EMS was designed so that it was open to all member states of the European Community. Some could join the wider margin of the ERM (6%) –​the system of fixed but adjustable exchange rates –​whereas those that had been part of the snake would be at 2.25% margin. While all member states participated in EMS, the United Kingdom did not participate in the ERM. The British pound was part of the ECU however and the United Kingdom had been part of the EMCF since 1973.This shows how the United Kingdom took on a different perspective on European monetary cooperation already in the 1970s (Walsh 2000; Ikemoto 2011). The European Community members allowed this difference to be part of the institutional design of EMU (Ludlow 1982). But to be fair, the discussions in the 1970s and 1980s were as much about how to obtain policy outcomes as about the structure of institutions. Some of these differences can be traced back to differences in ideas over macroeconomic policymaking (and how to reduce inflation) and to power politics (who dominated EU institutional structures). The logic of trying to get the EU to stay close together (avoiding ‘Europe a la carte’ or ‘variable geometries’ as much as possible) was a major concern. Although the United Kingdom had had its exceptional position in the EMS, during the intergovernmental conferences (IGCs), negotiators nevertheless sought to avoid that this form of differentiation (some member states not participating in everything) would be institutionalized into the Treaty (Commission 1990; Commission 1993). It was known in advance, however, that starting discussions about EMU and 202

European Central Bank

an ECB could be a challenge. After all the decision to open the Intergovernmental Conferences on EMU and political union had met with opposition by UK Prime Minister Margaret Thatcher. The United Kingdom had put forward its own proposals for EMU (the ‘hard ecu’, see Verdun 2000: 87). Nevertheless, the United Kingdom eventually was formally given an ‘opt out’ from EMU and the Social Chapter in the 1992 Maastricht Treaty that entered into force in November 1993 (Sandholtz 1993; Dyson and Featherstone 1999; Verdun 2000). Yet the design of EMU institutions still included all member states. Similar to the EMS in the 1970s, the European Monetary Institute (EMI) as the front-​runner of the European System of Central Banks (ESCB) once again foresaw the representation of all member states. The Treaty envisaged in a separate protocol that some articles of the Maastricht Treaty would not apply to member states with a derogation (Protocol 11 and Protocol 12 annexed to the Treaty). Yet it was not only the United Kingdom that had some reservations about joining EMU. Denmark held a referendum about the Maastricht Treaty in 1992 and 1993. In response to the outcome of the 1992 referendum, in which a majority of 50.7% rejected the Maastricht Treaty as envisaged, it was given some assurances, including an opt-​out of EMU.This contributed to the outcome of a second referendum when a majority supported accepting the Maastricht Treaty (Worre 1995). After the creation of EMU, Denmark and Sweden held further referendums on whether to join the euro area, resulting in popular majorities against doing so –​a clear testimony of ongoing lack of popular support for the single currency among Danish and Swedish voters (Friis 2002: 383; Miles and Lindh 2004: 201). Since then, a number of other member states have chosen not to take the next steps towards joining the euro, even if they potentially could. These member states are formally called ‘member states with a derogation’. Informally Denmark and the United Kingdom are noted as having a de jure opt-​out whereas the countries that do not adopt the euro when they possibly could are considered having a de facto opt-​out (see Majone 2014). This process firmly established the notion of differentiation in the EU. Upon his arrival into office, UK Prime Minister Tony Blair discontinued his privilege to use the opt-​out in the social domain. At times, there were considerations that the United Kingdom might join the euro but it did not materialize, in part because of the political difficulties surrounding the Iraq War (cf. Schmidt 2006: 196). In the end, the United Kingdom became less rather than more interested in monetary integration and other forms of deepening (continuing to stay out of the Schengen Area for instance). Following the negative referendum in 2003, the Swedish government took it upon itself for the foreseeable future not to join the ERM –​a prerequisite to joining the euro. In 2004, ten more member states joined the EU. Since there was no official ‘opt out’ granted to these countries, they signed on to the notion that they would join the euro area once they met the criteria (Schadler et al. 2005; Dyson 2006; Epstein and Johnson 2010). Of these ten, as many as seven have since joined (all but Poland, Hungary, and the Czech Republic). Two more countries joined in 2007 (Romania and Bulgaria), followed by Croatia in 2013. In July 2020, the latter two have joined the ERM, which is an important step to joining the euro.3 We may hence distinguish pacesetters and laggards among the new EU member states when it comes to euro adoption. Juliet Johnson (2008) convincingly explains this pattern with the countries’ size and economic openness. While smaller and more trade-​dependent countries such as the Baltics benefit strongly from joining EMU in terms of international credibility and facilitating trade, this is less true for Poland, Hungary, and the Czech Republic, which are larger and less trade-​dependent. During the euro area crisis, these ‘laggards’ were content that they had not pegged their currency to the euro as they were now able to use the exchange rate as an instrument for adjustment. 203

Daniel F. Schulz and Amy Verdun

However, the seven countries that did join the euro did so enthusiastically: Slovenia joined in 2007, Cyprus and Malta in 2008, and Slovakia in 2009 (Dandashly and Verdun 2018). Lithuania was told in 2006 that it did not qualify as it missed the criteria by a small margin. However, once the worst of the crisis was over, it joined in 2015. Meanwhile, Estonia had joined in 2011 and Latvia in 2014. Some member states had previously been informed by the International Monetary Fund (IMF) that it could be better if they abandoned their interest in euro adoption and instead let their currencies float.Yet the Baltic states defied that recommendation and instead focused on internal devaluation as an adjustment mechanism (Dandashly and Verdun 2020). All these developments indicate that differentiation has a long history within EMU. It also suggests that the EU institutions have not a priori necessarily pushed for deepening if it meant that member states would be joining the final stage of EMU before being ready. Hence these institutions have had a checkered record of pushing integration in the context of differentiation. Let us now turn to a more detailed examination of the specific role of the ECB in this process.

The ECB’s bodies and tasks: differentiated decision-​making Twenty years after the ECB came into being, it has arguably developed into the most central and powerful supranational institution of our times (Curtin 2017). During this period, the adoption of the euro by several new member states has not only widened integration in the ECB’s domain. Following the financial crisis of 2008, the ECB’s adoption of unconventional policies has also extended its powers to provide monetary stability while the bank simultaneously acquired supervisory powers to safeguard financial stability. Despite these steps of widening and deepening, the ECB still oversees a highly differentiated policy area. Hence the ECB’s authority differs across the union depending on member states’ levels of monetary and financial integration. This section details how such differentiation impacts the ECB’s internal structures and decision-​ making procedures as well as its operations across its various tasks: monetary policy, banking supervision, and its support of the wider economic policies in the EU. The ECB’s General Council arguably represents the clearest imprint of differentiated integration on the institution. It is, formally, the third of three decision-​making bodies (Art. 44 of the Statute) whose only raison d’être derives from the fact that some EU member states have not (yet) adopted the euro (currently 8 out of 27). While the central bank governors from those euro outsider countries are therefore absent on the other two decision-​making bodies –​the Governing Council and the Executive Board –​they retain their seats and voting rights in this third body. The ECB itself thus describes the General Council as ‘transitional body’ which is to be dissolved once all EU member states have introduced the single currency. It is worth noting that this –​de jure –​view of differentiation as a temporary condition goes against the bulk of academic research with perceives it –​de facto –​as a permanent feature of the post-​Maastricht EU (Holzinger and Schimmelfennig 2012; Leruth et al. 2019). Hence the General Council is the ECB’s institutional embodiment of the ESCB (comprising the ECB and the central banks of all EU member states), while the Governing Council and the Eurosystem remain limited to euro insiders. Beyond its status as symbol of differentiation, however, most scholars pay little attention to the General Council given that has ‘no real executive powers’ (Dyson 2008: 120). In its key domain –​monetary policymaking –​the ECB grants euro outsiders only limited access.This might reflect the motive to make non-​participation a less attractive option for member states that remain ‘on the fence’ about joining (Smits 1997: 98). Hence, non-​members do not only stay on the outside when decisions are made at the top level (the Governing Council), they also do not participate in Eurosystem working groups which prepare top-​level decision-​making. There is a complex substructure of staff committees (and their associated working groups and 204

European Central Bank

task forces) which bring together staff from the ECB and the national central banks (NCBs), which may also include NCB experts from outside the euro area. These committees embody the Eurosystem’s key organizing principle of ‘centralized decision-​ making and decentralized implementation’, thus enhancing the ECB’s administrative capacity by including the expertise from the NCBs (Jung et. al. 2010). At its inception, there was even a concern that the ECB might be overwhelmed by powerful NCB governors and their large staffs, given that the NCB workforce collectively was about 100 times larger than that of the ECB in 1999 (Cecchetti and Schoenholtz 2008: 5). However, the ECB’s own workforce has since expanded significantly, and ECB staff can exert control by setting the agenda and chairing the committees. While relatively little is known about the political dynamics within those structures, the importance of expertise in central banking suggests that NCBs with large staffs (such as Germany’s Bundesbank, the Banque du France, or the Banca d’Italia) yield relatively more power over outcomes than smaller NCBs (see Schulz 2017: 43–​46 for a detailed discussion).4 By the same token, one might expect the euro outsiders to successfully make their voices heard in ESCB committees, where no formal differences exist between insiders and outsiders. After all, the outsiders include prestigious and well-​staffed institutions, including the two oldest central banks in the world (Sweden’s Riksbank and the Bank of England). In the much more influential Eurosystem committees, however, they have no seat at the table, for instance when it comes to preparing the macroeconomic projections as a key input for monetary policy decisions. The key committee –​the Monetary Policy Committee (MPC) –​and its substructures are equally limited to staff from central banks participating in EMU. Euro outsiders have thus little say over the ECB’s monetary policy stance, which nevertheless has important implications for monetary conditions in their countries. This is most obvious in cases where euro outsiders have decided to tie their national currencies to the euro at a fixed rate, as in Denmark. Yet, as Marcussen (2005) points out, Denmark and other small states have traditionally not enjoyed monetary sovereignty to any significant extent; they have always been more decision takers than decision makers. Furthermore, its status as an outsider has not prevented Denmark from following ‘sound’ fiscal and economic policies in line with EMU priorities, perhaps even more so than some euro insiders. Finally, Danish policymakers have tried to establish various indirect means to influence monetary policymaking in the euro area, including ESCB committees (Adler-​Nissen 2008: 675) and meetings among central bankers in the context of international organizations such as the IMF, the OECD, or the Bank for International Settlements (Marcussen 2005: 56). One of the key changes following the financial crisis of 2007/​2008 has been the creation of the European Banking Union (EBU) in 2012. During the negotiations of the Maastricht Treaty, national governments had been unwilling to surrender the supervisory duties over their national banks (James 2012), leading to a system of horizontal coordination and cooperation among national supervisors. Therefore, both central bankers (Draghi 2013) and academics (Howarth and Quaglia 2014) have described the shift towards supranational banking supervision as the most significant step towards deepening integration since the start of the euro. To be clear: What was unusual about the previous system was not so much the fact that the ECB lacked control over banking institutions operating in the euro area. Many states have historically assigned the powers to conduct monetary policy and banking supervision to separate institutions. Yet the pre-​2012 EMU was unique in terms of the asymmetry between supranational monetary policymaking and national banking supervision it implied (Issing 2008). These national arrangements for supervision remained quite diverse and often marked by bank-​ state ties at the national level, which contributed to the severity of the financial crisis (Epstein and Rhodes 2018). 205

Daniel F. Schulz and Amy Verdun

Against this background, the ECB used the opening presented by the crisis to actively pursue an extension of its powers (Davies and Green 2010: 202–​204). While its previous attempts to expand its supervisory powers had always met national resistance (Quaglia 2008: 137), the crisis experience clearly demonstrated the weaknesses associated with a fragmented supervisory system. Hence the ECB’s policy entrepreneurship proved successful this time around (Howarth and Quaglia 2014; De Rynck 2016; Skuodis 2018) as the Single Supervisory Mechanism (SSM) established the ECB as the sole supervisor for EMU’s biggest banks. The following discussions led to the Single Resolution Mechanism (SRM) as the second pillar of EBU, while the third pillar to ‘complete’ the banking union –​a common backstop or deposit insurance system –​still remains hotly debated (Howarth and Quaglia 2018). Numerous empirical studies examine the creation of the SSM and SRM by focusing on the role of the EU institutions –​most importantly the European Commission and the ECB itself –​ or the diverging interests between Northern and Southern EMU member states. However, the interests of euro outsiders in its creation should not be overlooked. As Frank Schimmelfennig (2016) argues, we may see the case of the banking union as a ‘first major instance of a spillover of differentiated integration from one policy to another’ which also suggests that a path-​ dependent continuation of differentiated integration can turn into a slippery slope over time. This explanation focuses on the differentiated effect that the creation of the single currency had on euro insiders and outsiders. For insiders, the euro undermined the ability of NCBs to safeguard financial stability because (1) they could no longer provide liquidity as lenders of last resort in a financial crisis and (2) their national resolution powers were constrained by European rules. This encouraged them to look to supranational solutions. Euro outsiders, however, still saw their lender of last resort functions intact in the crisis, which made the banking union a relatively less attractive prospect for them (Howarth and Quaglia 2014: 127). Outsiders therefore expressed concerns about a greater role for the ECB in banking supervision as they feared its dominance in setting technical rules (Hennessy 2014: 163). To overcome their concerns, the agreement on an EBU included some ‘sweeteners’, including protection in the voting procedures of the European Banking Authority (EBA) and the possibility for euro outsiders to voluntarily join the EBU through an arrangement called ‘close cooperation’.To date, however, only Bulgaria and Croatia have submitted a request for close cooperation while other outsiders remain on the fence. Initial assessments argued that this option should be particularly attractive to outsiders with large international banking groups such as Sweden and the United Kingdom (Schoenmaker and Siegmann 2013). While the United Kingdom vocally opposed joining early on, Sweden and Denmark have set up national taskforces to address the question of whether to join on a voluntary basis (Spendzharova and Bayram 2016; Schulz and Henökl 2020). The case of Sweden is illustrative in this regard. Subsidiaries of Swedish banks have long dominated large parts of the banking markets in all three Baltic states since the late 1990s. Given that the Baltic states joined EMU –​and hence the EBU –​the ECB now does not only supervise those systemically important banks operating in the euro area directly, it also participates in the supervision of non-​systemically important banks domiciled in countries outside the Eurozone (in the context of supervisory colleges managed by the EBA). Additionally, the Nordic region’s biggest bank –​Nordea –​announced the relocation of its headquarters from Stockholm to Helsinki in 2017 to position itself ‘on a par with its European peers’ through participation in the EBU.This move reopened the Swedish debate about joining EBU, which is supported by several important domestic players, including Sweden’s central bank. The ECB’s approach to differentiation came to light in a series of speeches on the EBU, which ECB policymakers gave to audiences in non-​EMU countries. While these speeches

206

European Central Bank

typically focused on the desirability of the project as such, they often specifically alluded to the benefits it may bring to countries such as Denmark, Sweden, and the United Kingdom. To cite the title of one of those speeches, the ECB’s perceives EBU as ‘essential for the ins [and] desirable for the outs’ (Asmussen 2013). The arguments focus both on the large size of the Danish, Swedish, and British banking sectors and, more importantly, their high degree of integration with the rest of Europe. While this discourse explicitly emphasized the benefits of joining and presenting the EBU as ‘the place to be’ (Lautenschläger 2018) for outsiders, it simultaneously stressed that they would incur no disadvantages relative to the status quo if they decided to remain on the outside. Hence, the ECB repeatedly stressed that EBU would benefit all member states which were part of the single financial market, irrespective of EBU membership. Since many euro outsiders, and especially the United Kingdom, were leading providers of financial services in Europe, ECB policymakers emphasized that they could ‘see only advantages for Britain from its creation’ (Constâncio 2013). In sum, however, the ECB appeared much more active in promoting EBU membership than EMU membership among the opt-​outs, as the comparison of such speeches with the typically neutral stance and factual tone of the ECB’s convergence reports shows.

The ECB and the euro crisis: combatting disintegration? The ECB’s monetary policies during the ‘Great Recession’ Following the ECB’s ‘happy childhood’ during the first ten years of its existence (Enderlein and Verdun 2009), the global financial crisis of 2007/​2008 set the stage for what would become very difficult teenage years (Howarth and Verdun 2020). The subsequent ‘Great Recession’ saw the ECB leap into the unknown, when interest rates approached zero and unconventional instruments were added to the toolkit to prevent the recession from worsening and a deflationary spiral from inflicting even more harm (Chang 2019). Of course, the ECB was not alone in confronting these challenges, but one of many central banks struggling with the ‘new normal’ in monetary policymaking. What made the ECB’s situation particularly dire, however, was the fact that markets suddenly woke up to specific but known problems of the euro area’s asymmetrical institutional structure (Verdun 1996; 2020; Schlosser 2019). When the incoming Greek government announced in 2009 that the budget deficit it inherited was much worse than the previous government had claimed, it set an avalanche in motion. Within the time span of only a few months, investors’ fears about the sustainability of governments’ finances spread from one Eurozone country to another, leading to sharp increases in interest rates for government bonds and hence threatening their access to bond markets to refinance their public debts. Importantly, this element of contagion also affected countries with previously low levels of public debt such as Spain and Ireland, invoking fears of a domino effect across the currency area. Thus, the ECB found itself not only fighting a recession of historic dimensions with largely new and untested instruments; it also had to confront the very real prospect of a disintegrating euro area. This existential crisis situation produced far-​reaching changes for the ECB’s interpretation of its own mandate (Schmidt 2016). In Draghi’s now famous ‘whatever it takes’ remarks, he offered the image of the asymmetrical euro area as a bumblebee that shouldn’t be able to fly even though for some years it miraculously had managed to stay up in the air. Going forward, however, he argued that ‘more Europe’ and steps towards political union were needed for the bumblebee to become a real bee:

207

Daniel F. Schulz and Amy Verdun

The euro is like a bumblebee.This is a mystery of nature because it shouldn’t fly but instead it does. So the euro was a bumblebee that flew very well for several years. And now –​and I think people ask “how come?” –​probably there was something in the atmosphere, in the air, that made the bumblebee fly. Now something must have changed in the air, and we know what after the financial crisis. The bumblebee would have to graduate to a real bee. And that’s what it’s doing. […] The last summit was a real success because for the first time in many years, all the leaders of the 27 countries of Europe, including UK etc., said that the only way out of this present crisis is to have more Europe, not less Europe. A Europe that is founded on four building blocks: a fiscal union, a financial union, an economic union and a political union Draghi (2012) This commitment to further integration (which also included vaguely worded commitments to structural reforms and a centralization of banking supervision) is clearly linked to concerns about the disintegration of the euro area. This became clear when ECB officials subsequently called fears that a country might exit the euro –​so-​called redenomination risks –​unfounded, and frequently reiterated that the euro was irreversible. The ECB’s announcement to do ‘whatever it takes’ to preserve the euro thus includes a clear commitment to further steps in European integration (combined with an increased role for the ECB). Crucially, it offered such a commitment only in the context of an existential threat, suggesting that the ECB’s change of heart was motivated by the fundamental interest in the survival of the euro area –​and the ECB itself. Following this crucial turning point in the euro area crisis, the ECB followed up with a program that potentially enabled unlimited purchase of government bonds of euro area countries in case governments were facing excessive borrowing costs (Verdun 2017; Schoeller 2019). While the Outright Monetary Transactions (OMT) program has not needed to be activated to date, its availability alone was sufficient to have the desired effect of dramatically reducing interest rate differentials for sovereign debt across the euro area. Hence, Mario Draghi lauded the program as ‘probably the most successful monetary policy measure undertaken in recent times’ (Steen 2013). Yet while its effectiveness remains undisputed, the same cannot be said about the OMT’s legality. The program launched a lengthy legal battle between the European Court of Justice (ECJ) and Germany’s Constitutional Court, owing to substantial ambiguity in the ECB’s mandate (see de Boer and van ’t Klooster, 2020). After all, the OMT de facto transformed the ECB into a lender of last resort to governments (de Grauwe 2013) –​a role which the Maastricht Treaty left undefined (Eijffinger 2005: 475) but which some would claim is a central bank’s main reason d’être (Goodhart 1988).The far-​reaching implications of the Weiss case for the EU’s legal order and the ECB’s democratic accountability are beyond the scope of this chapter.Yet we note that the OMT seemed to show that credibly promising to purchase government bonds may have been the most effective way to avoid having to actually make the purchases (Schulz, 2017: 144). However, the moment for bond purchases eventually did arrive in early 2015. While the OMT eliminated the immediate risk of ‘differentiated disintegration’, it was insufficient to end the economic crisis. Only after having exhausted all other avenues –​negative interest rates, forward guidance, and several rounds of cheap funding for banks –​the ECB followed the example of other central banks to launch a Quantitative Easing Program. This asset purchasing program provided monetary stimulus until 2018 and was quickly restarted during the COVID-​19 crisis in March 2020. The ECB’s response to this most recent crisis event reveals both significant learning and change vis-​à-​vis the sovereign debt crisis and continued concerns about the long-​term integrity of the euro area. On the one hand, the rapid response showed that it is now widely accepted 208

European Central Bank

that, in the face of sudden shocks, the EU needs an actor like the ECB to act fast (Jones 2019) and that asset purchases are a part of its toolkit. On the other hand, a remark by ECB President Christine Lagarde that the ECB was ‘not here to close spreads’ in sovereign debt markets quickly reawakened fears of fragmentation. The borrowing costs of hard-​hit Italy rose sharply and, echoing the contagion fears of the sovereign debt crisis, the government bonds of other Mediterranean EMU countries weakened seemingly in sympathy with Italy. While the ECB succeeded in walking back those comments in the following weeks, their effect on market sentiment showed that the evolution of its monetary policies alone was not enough to quell fears of differentiated disintegration in a lasting way.Thus, we now turn to the long-​term institutional reforms designed to overcome such concerns.

The ECB’s role in EMU reform Parallel to its emergency monetary policies, the ECB also became an important partner in discussions on the institutional structure of the EU in the wake of the immediate crisis (Hodson 2015; Niemann and Ioannou 2015). For example, ECB President Draghi was involved in drafting a blueprint for the future of EMU –​‘Towards a Genuine Economic and Monetary Union’ –​in cooperation with the presidents of the European Council, the Eurogroup, and the Commission (see European Council 2012).Three years later, in 2015, the ECB president was once again party to the so-​called five presidents’ report (including the president of the European Parliament) under the chairmanship of Commission President Jean-​Claude Juncker (Juncker et al. 2015). Again, the recommendations focused on steps towards deeper economic and monetary integration, including the aim to complete the banking union. It also called for a European Minister of Finance, a euro-​area budget, and a Capital Markets Union. These documents spelled out what were seen as the necessary steps to move towards deeper economic and monetary integration. Notably, the chairperson of the Eurogroup (first Juncker, later Dijsselbloem) relied heavily on the ECB president for expertise in these efforts (Verdun 2018). Despite the ECB’s involvement in the preparation of these key documents, Frank Schimmelfennig (2015: 188) finds that it ‘does not seem to have had a noteworthy agenda-​setting role in institutional reform’. In a similar vein, Thomas Warren (2020) argues that Draghi’s bold statements on ‘fiscal union’ mask the ECB’s limited reform ambition which was mainly focused on stricter enforcement of fiscal rules. However, the events of 2020 saw the ECB act as a strong advocate of a centralized fiscal capacity in the debate on the ‘Next Generation EU’ (NGEU). What is more, the ECB also suggested openness to making this one-​off response to the crisis a permanent feature of EMU (Arnold 2020). An early ECB research note suggested that NGEU holds lessons for the future of EMU which ‘still lacks a permanent fiscal capacity at supranational level for macroeconomic stabilization in deep crises’ (Giovannini et al. 2020). While it is early days for these debates, such openings may suggest that the ECB increasingly recognizes the costs of its institutional loneliness (Mabbett and Schelkle, 2019) when crisis fighting requires closer cooperation between fiscal authorities and the central bank.

Conclusion How has the ECB approached differentiated integration and the perceived threat of disintegration? This contribution has demonstrated that it has taken this young institution considerable time to accommodate the persistent differences among EU member states. Past research has even described the ECB as a ‘reluctant EU institution’ given that it often resisted attempts to enhance the Community dimension of EMU when it perceived such initiatives to be at 209

Daniel F. Schulz and Amy Verdun

odds with its primary mandate (Heisenberg and Richmond 2002; Hodson 2011, 2015). In this view, the ECB’s support for ‘more Europe’ has long been contingent on its overriding commitment to price stability. When the sovereign debt crisis posed a potentially existential threat, however, the ECB started adopting a more proactive role. As national governments lacked the capacity or the political will to act, the ECB accepted its role as a lender of last resort and provided funds to member states in need. Its major concern was that financial difficulties of one or more member states could jeopardize the integrity of the euro area as a whole and with it the rationale underlying European unity. This fear of ‘differentiated disintegration’ turning into full disintegration (of EMU or even the entire EU) fostered the ECB’s push for more integration during and after the sovereign debt crisis. From a historical perspective, differentiation in the context of EMU is less novel than might appear at first glance. This contribution has shown that differences in visions about institutional structure as well as the desirability of deeper monetary integration prevailed since the 1970s. The way to accommodate the differences has been to build institutions that respect the lack of political will of some governments to be part of the institutional structures but to still include them as much as possible (for instance in the EMS all member states were part of the ECU).The same occurred in the 1990s when the institutional design kept the door open to countries with an opt-​out to become part of the EMI. Regarding the ECB’s approach to differentiation, we examined whether the ECB tends to encourage outsiders to join, to keep them on board as much as possible, or to contribute to further differentiation by deepening integration only among the ‘ins’. We found elements of all three approaches in the ECB’s policies over the past twenty years. In its discursive engagement with euro outsiders, the ECB often explicitly highlights the benefits of euro membership for the current ‘outs’. There is also some evidence for the second avenue, namely that the ECB sought to reduce the impact of differentiation by keeping the ‘outs’ as much as possible on board –​yet not going the full length of accepting members which have not fully met the criteria (such as the case of Lithuania in 2005). Finally, we also witnessed policies which further cement the differentiated state of EMU. The creation of the EBU, for instance, increases differentiation between ins and outs. Hence our analysis suggests that the ECB has continuously tried to walk the tightrope between widening and deepening. However, when the chips are down, it has focused on the stability of the euro area as its main point of concern. The COVID-​19 crisis appears to confirm that the ECB has shed its narrow technocratic focus throughout the past decade in order to provide political leadership in the EU (Verdun 2017; Schoeller 2020). However, we note that it took several instances of potentially existential crisis to produce this evolution. From the perspective of the differentiation literature, the ECB’s concern with preventing ‘differentiated disintegration’ became paramount. This motive of preserving the integrity of the euro area, by keeping all euro insiders on board, has prompted stark changes –​both in the ECB’s monetary policies and its approach to EMU reform. Yet, in our view, this does not suggest that the ECB is a competence maximizer ‘hardwired’ to ever closer union. Rather, the evidence suggests that it merely accepted its greater powers and a deepening of integration to avert the threat of (differentiated) disintegration.

Notes 1 This chapter was supported in part by a Jean Monnet Network entitled “The Politics of the European Semester: EU Coordination and Domestic Political Institutions (EUROSEM)” Agreement number: 600110-EPP-1-2018-1-CA-EPPJMO-NETWORK (Grant Agreement No 2018-1359), with the support of the Erasmus+ programme of the European Union.

210

European Central Bank

2 Quoted in Apel (1998). 3 To date neither have met all the criteria. Once they do, they may ask the Commission to assess whether they meet the convergence criteria –​a condition that needs to be met before a country is eligible to join. 4 There are still few studies detailing the relative power of member states over the inner workings of the ECB. This might reflect the fact that most central banks are secretive institutions and notoriously difficult to study (Schulz, 2019). In the case of the ECB, this is particularly acute as the ECB officially always adopts an area-​wide perspective and plays down any divisions along national lines. Hence the ECB itself does not facilitate research done by nationality. In research focused on gender relations in the ECB, for instance, ECB services made data available but explicitly made sure no nationality could be identified (Verdun 2022).

References Adler-​ Nissen, R. (2008) The Diplomacy of Opting Out: A Bourdieudian Approach to National Integration Strategies. Journal of Common Market Studies 46(3), pp. 663–​ 684. doi: 10.1111/​j. 1468-​5965.2008.00799. Apel, E. (1998) European Monetary Integration 1958–​2002. London: Routledge. Arnold, M. (2020) ECB calls on Brussels to make recovery fund permanent. Financial Times, 23 September 2020. Available at: www.ft.com/​content/​5bc6dd62-​51e3-​414c-​b815-​5f964cb616db (Accessed: 10 June 2021). Asmussen, J. (2013) Banking Union –​essential for the ins, desirable for the outs! Speech at the Danske Bank Financial Forum 2013. Stockholm, 5 November 2013. Available at: www.semanticscholar.org/​paper/​ J%C3%B6rg-​Asmussen%3A-​Banking-​Union-​essential-​for-​ins%2CAsmussen/​6cd65ff3a0b37d6a34810 a7ec8b6938635a98df7 (Accessed: 10 June 2021). Ban, C. (2020) The Economic and Monetary Union: How Did the Euro Area Get a Lender of Last Resort? in: R. Coman, A. Crespy, and V. A. Schmidt (eds), Governance and Politics in the Post-​Crisis European Union. Cambridge: Cambridge University Press. pp. 179–​195. Buiter, W. and Rahbari, E. (2012) The European Central Bank as Lender of Last Resort for Sovereigns in the Eurozone. Journal of Common Market Studies, 50: 6–​35. Available at: https://doi.org/10.1111/j.14685965.2012.02275.x (Accessed: 10 June 2021). Cecchetti, S. G. and Schoenholtz, K. L (2008) How Central Bankers See It: The First Decade of ECB Policy and Beyond’, NBER Working Paper No. 14489. Available at: https://​ideas.repec.org/​p/​nbr/​nbe​ rwo/​14489.html (Accessed: 10 June 2021). Chang, M. (2019) The Constraints of Central Bank Independence: The European Central Bank's Unconventional Monetary Policy and Incremental Accountability in the Euro Crisis, Journal of Contemporary European Research 15(2):143-​161, DOI: 10.30950/​jcer.v15i2.1006 Chibber, K. (2011) The Domino Effect in Europe’s Debt Crisis. BBC News, 20 September, Available at: www.bbc.com/​news/​business-​14985256 (Accessed: 10 June 2021). Commission of the European Communities (1990) One Market, One Market’ European Economy, no 44. Europa.eu. Available at: https://​ec.eur​opa.eu/​econ​omy_​fina​nce/​publi​cati​ons/​pages/​pub​lica​tion​_​sum​ mary​7520​_​en.htm (Accessed: 10 June 2021). Commission of the European Communities (1993) ‘The Economics of EMU: Background Studies for European Economy, 44 “One Market, One Money” ’. Available at: https://​ec.eur​opa.eu/​econ​omy_​fina​ nce/​publi​cati​ons/​pages/​pub​lica​tion​7454​_​en.pdf. Constâncio, V. (2013) The Nature and Significance of Banking Union. Speech at the Conference Financial Regulation:Towards a Global Regulatory Framework?, London: Chatham House City Series. Curtin, D. (2017) Accountable Independence’ of the European Central Bank: Seeing the Logics of Transparency. European Law Journal 23(1–​2) pp. 28–​44. doi/​pdf/​10.1111/​eulj.12211. Dandashly, A. and Verdun, A. (2018) Joining the Euro Area? Central and Eastern European Countries and their Quest for Euro Adoption. Paper Submitted for Delivery at the ECPR-​SGEU Biennial Conference, Paris, 13–​ 15 June. Available at: https://​ecpr.eu/​Eve​nts/​Event/​Paper​Deta​ils/​39186 (Accessed: 10 June 2021). Dandashly, A. and Verdun, A. (2020) Euro Adoption Policies in the Second Decade –​The Remarkable Cases of the Baltic States. Journal of European Integration, 42(3), pp. 381–​ 397. doi: 10.1080/​ 07036337.2020.1730355.

211

Daniel F. Schulz and Amy Verdun

Davies, H. and Green, D. (2010) Banking on the Future: The Fall and Rise of Central Banking. Princeton, NJ: Princeton University Press. de Boer, N. and Van ’t Klooster, J. (2020) The ECB, The Courts and The Issue of Democratic Legitimacy After Weiss. Common Market Law Review, 57(6), pp. 1689–​1724. Available at: https://kluwerlawonline. com/journalarticle/Common+Market+Law+Review/57.6/COLA2020765 (Accessed: 10 June 2021). De Grauwe, P. (2013) The Political Economy of the Euro. Annual Review of Political Science 16, pp. 153–​170. doi:10.1146/​annurev-​polisci-​060911-​085923. De Rynck, S. (2016) Banking on a Union: The Politics of Changing Eurozone Banking Supervision. Journal of European Public Policy 23(1), pp. 119–​135. doi: 10.1080/​13501763.2015.1019551. Draghi, M. (2012)Verbatim of the Remarks Made by Mario Draghi. Speech at the Global Investment Conference, London, 26 July 2012. Available at: www.ecb.europa.eu/​press/​key/​date/​2012/​html/​sp120726.en.html (Accessed: 10 June 2021). Draghi, M. (2013) Address at the new year’s reception of the Frankfurt Chamber of Commerce and Industry. Frankfurt am Main, 22 Jan 2013. Available at: www.ecb.europa.eu/​press/​key/​date/​2013/​html/​sp130122. en.html (Accessed: 10 June 2021). Dyson, K. (2006) Euro Entry as Defining and Negotiating Fit: Conditionality, Contagion, and Domestic Politics, in: K. Dyson (ed) Enlarging the Euro Area: External empowerment and domestic transformation in East Central Europe. Oxford: Oxford University Press, pp. 7–​43. Dyson, K. (2008) The European Central Bank: Enlargement as Institutional Affirmation and Differentiation, in: E. Best,T. Christiansen and P. Settembri (eds) The Institutions of the Enlarged European Union. Continuity and Change. London: Edward Elgar, Chapter. Dyson, K. and Featherstone, K. (1999) The Road to Maastricht: Negotiating Economic and Monetary Union. Oxford: Oxford University Press. Dyson, K. and Quaglia, L. (2010) European Economic Governance and Politics. Volume I. Commentary on Key Historical and Institutional Documents, Oxford: Oxford University Press. Enderlein, H. and Verdun, A. (2009) EMU’s Teenage Challenge: What Have We Learned, and Can We Predict from Political Science?. Journal of European Public Policy, 16(4), pp. 490–​507. Eijffinger, S. C. W. (2005) The European Central Bank and Financial Supervision, in: D. Masciandaro (ed), Handbook of Central Banking and Financial Authorities in Europe: New Architectures in the Supervision of Financial Markets Cheltenham: Edward Elgar, pp. 455–​482. Epstein, R. and Johnson, J. (2010) Uneven Integration: Economic and Monetary Union in Central and Eastern Europe. Journal of Common Market Studies 48(5), pp. 1237–​1260. doi: 10.1111/​j.1468-​5965.2010. 02111.x. Epstein, R. and Rhodes, M. (2018) From Governance to Government: Banking Union, Capital Markets Union and the New EU. Competition & Change 22(2), pp. 205–​224. doi: 10.1177/​1024529417753017. European Commission (2020) Commission welcomes Bulgaria and Croatia’s entry into the Exchange Rate Mechanism II, ec.europa. Press Release, 10 July. Available at: https://​ec.eur​opa.eu/​com​miss​ion/​pres​scor​ ner/​det​ail/​en/​IP_​20_​1​321 (Accessed: 10 June 2021). European Council (2012) Towards a Genuine Economic and Monetary Union. Report by President of the European Council Herman Van Rompuy 2012. Brussels. EUCO 120/​12, PRESSE 296, PR PCE 102. Available at: www.consilium.europa.eu/​media/​21570/​131201.pdf (Accessed: 10 June 2021). Friis, L. (2002) The Battle Over Denmark: Denmark and the European Union. Scandinavian Studies 74(3), pp. 379–​396. Available at: www.researchgate.net/​publication/​241673088_​The_​Battle_​Over_​Denmark_​ Denmark_​and_​the_​European_​Union (Accessed: 10 June 2021). Giovannini, A., Hauptmeier, S., Leiner-​Killinger, N. and Valenta,V. (2020) The fiscal implications of the EU’s recovery package. ECB Economic Bulletin, Issue 6/​2020.Available at: www.ecb.europa.eu/​pub/​economic-​ bulletin/​focus/​2020/​html/​ecb.ebbox202006_​08~7f90a18630.en.html (Accessed: 10 June 2021). Glencross A. (2021) Managing Differentiated Disintegration: Insights from Comparative Federalism on Post-​Brexit EU–​UK Relations. The British Journal of Politics and International Relations 23(4), pp. 593–​608, doi:10.1177/​1369148120968516. Goodhart, C. A. E. (1988) The Evolution of Central Banks, Cambridge, MA: MIT Press. Haas, E. B. (1958) The Uniting of Europe: Political, Social, and Economic Forces, 1950–​1957. Stanford, CA: Stanford University Press. Heisenberg, D. and Richmond, A. (2002) Supranational Institution-​Building in the European Union: A Comparison of the European Court of Justice and the European Central Bank. Journal of European Public Policy 9(2), pp. 201–​218. doi: 10.1080/​13501760110120228.

212

European Central Bank

Hennessy, A. (2014) Redesigning Financial Supervision in the European Union (2009–​2013). Journal of European Public Policy 21(2), pp. 151–​168. doi: 10.1080/​13501763.2013.853601. Hodson, D. (2011) Governing the Euro Area in Good Times and Bad. Oxford: Oxford University Press. Hodson, D. (2013) The Little Engine that Wouldn’t: Supranational Entrepreneurship and the Barroso Commission. Journal of European Integration 35(3), pp. 301–​314. doi: 10.1080/​07036337.2013.774779. Hodson, D. (2015) De Novo Bodies and the New Intergovernmentalism:The Case of the European Central Bank, in: C. J. Bickerton, D. Hodson, and U. Puetter (eds), The New Intergovernmentalism: States and Supranational Actors in the Post-​Maastricht Era. Oxford: Oxford University Press, pp. 263–​288. Holzinger, K. and Schimmelfennig, F. (2012) Differentiated Integration in the European Union: Many Concepts, Sparse Theory, Few Data. Journal of European Public Policy 19(2), pp. 292–​305. doi:10.1080/​ 13501763.2012.641747. Howarth, D. and Quaglia L. (2014) The Steep Road to European Banking Union: Constructing the Single Resolution Mechanism. Journal of Common Market Studies 52(s1): 125–​140. Howarth, D. and Quaglia L. (2018) The Difficult Construction of a European Deposit Insurance Scheme: A Step Too Far in Banking Union?. Journal of Economic Policy Reform 21(3), pp. 190–​209. doi: 10.1080/​ 17487870.2017.1402682. Howarth, D. and A.Verdun (2020) Economic and Monetary Union at Twenty:A Stocktaking of a Tumultuous Second Decade: Introduction. Journal of European Integration, 42(3), pp. 287–​ 293. doi: 10.1080/​ 07036337.2020.1730348. Ikemoto, D. (2011) European Monetary Integration 1970–​79: British and French Experiences. Basingstoke: Palgrave-​Macmillan. Issing, O. (2008) The Birth of the Euro. Cambridge: Cambridge University Press. James, H. (2012) Making the European Monetary Union:The Role of the Committee of Central Bank Governors and the Origins of the European Central Bank. Cambridge, MA:The Belknap Press of Harvard University Press. Johnson, J. (2008) The Remains of Conditionality: The Faltering Enlargement of the Euro Zone. Journal of European Public Policy 15(6), pp. 826–​841. doi: 10.1080/​13501760802196564. Jones, C. (2019) The ECB after Draghi: ‘You need an actor who can act fast’. Financial Times, 13 Mar 2019. Available at: www.ft.com/​content/​911665fe-​41b8-​11e9-​9bee-​efab61506f44. (Accessed: 10 June 2021). Juncker, J., Tusk, D., Dijsselbloem, J., Draghi, M. and Schulz, M. (2015) Completing Europe’s Economic and Monetary Union, Brussels: European Commission (‘five presidents’ report’) ec.europa. Available at: https://​ ec.eur​opa.eu/​info/​publi​cati​ons/​five-​pre​side​nts-​rep​ort-​com​plet​ing-​euro​pes-​econo​mic-​and-​monet​ary-​ union​_​en (Accessed: 10 June 2021). Jung, A., Mongelli, F. P. and Moutot, P. (2010) How Are the Eurosystem’s Monetary Policy Decisions Prepared? A Roadmap. Journal of Common Market Studies 48(2), pp. 319–​ 345. doi: 10.1111/​j. 1468-​5965.2009.02054.x. Kutter, A. (2014) A Catalytic Moment: The Greek Crisis in the German Financial Press. Discourse & Society, 25(4), 446–​466. doi: 10.1177/​0957926514536958. Lautenschläger, S. (2018) European Banking Union –​the place to be? Biannual High-​Level Networking Seminar on Economic and Financial Issues Organised by Danmarks Nationalbank, Copenhagen, 14 May. Available at: www.ecb.europa.eu/​press/​key/​date/​2018/​html/​ecb.sp180514_​3.en.html (Accessed: 10 June 2021). Leruth, B., Gänzle, S. and Trondal, J. (2019) ‘Differentiated Integration and Disintegration in the EU after Brexit: Risks Versus Opportunities. Journal of Common Market Studies 57(6), pp. 1383–​1394. doi: 10.1111/​ jcms.12957. Leuffen, D., Rittberger, B. and Schimmelfennig, F. (2013) Differentiated Integration: Explaining Variation in the European Union. Basingstoke: Palgrave Macmillan. Ludlow, Peter (1982), The Making of the European Monetary System -​A Case Study of the Politics of the European Community (London: Butterworth Scientific). Mabbett, D. and Schelkle,W. (2019) Independent or Lonely? Central Banking in Crisis. Review of International Political Economy 26(3), pp. 436–​460. doi: 10.1080/​09692290.2018.1554539. Majone, G. (2014) Rethinking the Union of Europe Post-​Crisis: Has Integration Gone Too Far? Cambridge; New York: Cambridge University Press. Marcussen, M. (2005) Denmark and European Monetary Integration: Out but Far from Over. Journal of European Integration 27(1), pp. 43–​63. doi: 10.1080/​07036330400029918. McNamara, K. R. (1998) The Currency of Ideas. Monetary Politics in the European Union. Ithaca, NY: Cornell University Press.

213

Daniel F. Schulz and Amy Verdun

Meunier, S. and Gocaj L. (2013) Time Will Tell:The EFSF,The ESM, and The Euro Crisis. Journal of European Integration 35(3), pp. 239–​253. doi: 10.1080/​07036337.2013.774778. Miles, L. and Lindh, M. (2004) After the Referendum: A Swedish Euro Paradox? Cooperation and Conflict, 39(2), pp. 201–​206. doi: 10.1177/​0010836704042875. Niemann, A. and Ioannou, D. (2015) European Economic Integration in Times of Crisis: A Case of Neofunctionalism?.Journal of European Public Policy 22(2),pp.196–​218.doi:10.1080/1​ 3501763.2014.994021. Patel, K. (2019) Widening and Deepening? Recent Advances in European Integration History. Neue Politische Literatur 64, pp. 327–​357. Available at: link.springer.com/​article/​10.1007/​s42520-​019-​00105-​ 4 (Accessed: 10 June 2021). Quaglia, L. (2008) Central Banking Governance in the European Union: A Comparative Analysis. Abingdon: Routledge. Sandholtz, W. (1993) Choosing Union: Monetary Politics and Maastricht. International Organization 47(1) pp. 1–​39. Available at www.jstor.org/​stable/​pdf/​2706881.pdf (Accessed: 10 June 2021). Schadler, S., Drummond, P., Kuijs, L., Murgasova, Z. and van Elkan, R. (2005) Adopting the Euro in Central Europe: Challenges of the Next Step in European Integration International Monetary Fund Occasional Paper 234. Available at: www.bis.org/​review/​r070118c.pdf (Accessed: 10 June 2021). Schillin, A. (this volume) ‘An ever more fragmented Union? On the emerging relevance of the institutional implications of differentiated integration in the EU’. Schimmelfennig, F. (2012) Zwischen Neo-​und Postfunktionalismus: Die Integrationstheorien und die Eurokrise. Politische Vierteljahresschrift 53(3), pp. 394–​413. Available at: www.jstor.org/​stable/​24201881? (Accessed: 10 June 2021). Schimmelfennig, F. (2014) Differentiated Integration Before and After the Crisis, in: O. Cramme and S. B. Hobolt (eds), Democratic Politics in a European Union Under Stress Oxford: Oxford University Press, pp. 120–​134. Schimmelfennig, F. (2015) Liberal Intergovernmentalism and the Euro Area Crisis. Journal of European Public Policy, (22)2, pp. 177–​195. doi: 10.1080/​13501763.2014.994020. Schimmelfennig, F. (2016) A Differentiated Leap Forward: Spillover, Path Dependency, and Graded Membership in European Banking Regulation. West European Politics, 39(3), pp. 483–​502. doi: 10.1080/​ 01402382.2016.1143244. Schimmelfennig, F. (2018) Brexit: Differentiated Disintegration in the European Union. Journal of European Public Policy, 25(8), pp. 1154–​1173. doi: 10.1080/​13501763.2018.1467954. Schimmelfennig, F., Leuffen, D. and Rittberger, B. (2015) The European Union as a System of Differentiated Integration: Interdependence, Politicization and Differentiation. Journal of European Public Policy 22(6), pp. 764–​782. doi: 10.1080/​13501763.2015.1020835. Schlosser, P. (2019) Europe’s New Fiscal Union, Basingstoke: Palgrave. Schmidt,V. A. (2006) Democracy in Europe:The EU and National Polities. Oxford: Oxford University Press. Schmidt, V. A. (2016) Reinterpreting the Rules ‘by Stealth’ in Times of Crisis: A Discursive Institutionalist Analysis of the European Central Bank and the European Commission. West European Politics 39(5), pp. 1032–​1052. Schoeller, M. G. (2018) Leadership by Default: The ECB and the Announcement of Outright Monetary Transactions. Credit and Capital Markets, 51(1), pp. 73–​91. doi: 10.1080/​01402382.2016.1186389. Schoeller, M. G. (2019) Tracing Leadership: the ECB’s ‘Whatever it Takes’ and Germany in the Ukraine crisis. West European Politics 43(5), pp. 1095–​1116. doi: 10.1080/​01402382.2019.1635801. Schoeller, M. G. (2020) Germany, the Problem of Leadership, and Institution-​Building in EMU Reform. Journal of Economic Policy Reform, 23(3), pp. 309–​324. doi: 10.1080/​17487870.2018.1541410. Schoenmaker, D. and Siegmann A. H. (2013) Efficiency Gains of a European Banking Union. Tinbergen Institute Discussion Paper 13-​026/​IV/​DSF51. Available at: https://​pap​ers.tinber​gen.nl/​13026.pdf (Accessed: 10 June 2021). Schulz, D. F. (2017) Too Little, Too Late? How Central Bankers’ Beliefs Influence What They Do. Florence: European University Institute. Schulz, D. F. (2019) Researching Technocracy: Actor-​Centred Methodologies and Empirical Strategies for Studying EU Economic Governance in Hard Times. Journal of Contemporary European Research 15(2), pp. 228–​239. doi: 10.30950/​jcer.v15i2.1011. Schulz, D. F. and Henökl T. (2020) New Alliances in Post-​Brexit Europe: Does the New Hanseatic League Revive Nordic Political Cooperation? Politics and Governance 8(4), pp. 78–​ 88. doi: 10.17645/​pag. v8i4.3359.

214

European Central Bank

Skuodis, M. (2018) Playing the creation of the European banking union: what union for which Member States?. Journal of European Integration 40(1), pp. 99–​114. doi: 10.1080/​07036337.2017.1404056. Smits, R. (1997) The European Central Bank: Institutional Aspects. The Hague: Kluwer Law International. Spendzharova, A. B. and Bayram, I.E. (2016) Banking union through the back door? How European banking union affects Sweden and the Baltic States. West European Politics 39(3), pp. 565–​84. doi: 10.1080/​ 01402382.2016.1143245. Spiegel, P. (2014) How the Euro was saved, Financial Times, May 2014. www.ft.com/​content/​f6f4d6b4-​ ca2e-​11e3-​ac05-​00144feabdc0 (Accessed 10 June 2021). Steen, M. (2013) Draghi Lauds ‘Most Successful’ ECB Action. Financial Times, June 2013. www.ft.com/​content/​35f29c6c-​ce99-​11e2-​ae25-​00144feab7de8 Accessed 10 June 2021). Thygesen N. (1999) EMU and the Outsider Nations, in: Andersen T. M., Jensen S. E. H., Risager O. (eds), Macroeconomic Perspectives on the Danish Economy. Palgrave Macmillan, London. Tindemans Report (1976) European Union. Report by Mr. Leo Tindemans, Prime Minister of Belgium, to the European Council. Bulletin of the European Communities. Supplement 1/​76. Available at: http://​aei. pitt.edu/​942/​(Acces​sed: 10 June 2021). Verdun, A. (1996) An ‘Asymmetrical’ Economic and Monetary Union in the EU: Perceptions of Monetary Authorities and Social Partners. Journal of European Integration 20(1), pp. 59–​81. doi: 10.1080/​ 07036339608429045. Verdun, A. (2000) European Responses to Globalization and Financial Market Integration: Perceptions of Economic and Monetary Union in Britain, France and Germany. Basingstoke: Palgrave Macmillan. Verdun, A. (2015) A Historical Institutionalist Explanation of the EU’s Responses to the Euro Area Financial Crisis. Journal of European Public Policy, (22)2, pp. 219–​237. doi: 10.1080/​13501763.2014.994023. Verdun, A. (2017) Political Leadership of the European Central Bank. Journal of European Integration, 39(2), pp. 207–​221. doi: 10.1080/​07036337.2016.1277715. Verdun, A. (2018) Towards a deeper economic and monetary union: the five presidents’ report and EU socio-​ economic governance’ paper submitted to the workshop the workshop ‘governing the political future of the European Union’. NKWP Politicologenetmaal, Leiden, 7–​8 June. Available at: www.universiteitleiden. nl/​en/​events/​2018/​09/​towards-​a-​deeper-​economic-​and-​monetary-​union (Accessed: 10 June 2021). Verdun, A. (2022) Women’s Leadership in the European Central Bank, in: H. Müller and I. Tömmel (eds), Women and Leadership in the European Union, Oxford: Oxford University Press, Chapter 16, pp. 290-​307. Verhelst, S. (2013) Differentiated Economic and Monetary Integration: An Experimental Cocktail in Search of a New Recipe. Studia Diplomatica 66(3), pp. 19–​32. Available at: www.jstor.org/​stable/​26531583 (Accessed: 10 June 2021). Vollaard, H. (2014) Explaining European Disintegration. Journal of Common Market Studies 52(5), pp. 1142–​ 1159. doi: 10.1111/​jcms.12132. Walsh, J. I. (2000) European Monetary Integration and Domestic Politics: Britain, France and Italy, Boulder: Lynne Rienner Publishing. Warren, T. (2020) Explaining the European Central Bank’s Limited Reform Ambition: Ordoliberalism and Asymmetric Integration in the Eurozone. Journal of European Integration, 42(2), pp. 263–​279. doi: 10.1080/​ 07036337.2019.1658753. Werner Report (1970) Report to the Council and the Commission on the realization by stages of Economic and Monetary Union in the Community. “Werner Report” ’. Supplement to the Bulletin of the European Communities 11/​70. Council –​Commission of the European Communities. Available at: https://​ec.eur​ opa.eu/​econ​omy_​fina​nce/​publi​cati​ons/​pages/​pub​lica​tion​6142​_​en.pdf(Acces​sed: 10 June 2021). Worre, T. (1995) First No, Then Yes: The Danish Referendums on the Maastricht Treaty 1992 and 1993. Journal of Common Market Studies 33(2), pp. 235–​257. doi:10.1111/​j.1468-​5965.1995.tb00529.x. Zimmermann, H. (2016) The Euro Trilemma, or: How the Eurozone Fell into a Neofunctionalist Legitimacy Trap. Journal of European Integration, 3(4), pp. 425–​439. doi: 10.1080/​07036337.2016.1141904.

215

14 Differentiation and the European Commission Diane Fromage and Cristina Fasone

Introduction: (Re-​)situating the Commission in an increasingly differentiated European Union Differentiation “has proved essential for allowing the European integration to proceed” (De Witte 2017: 10) and has been one of its structural features ever since its very beginning. Indeed, inter se agreements amongst some European Union (EU) countries only, opt-​outs, derogations and flexibility in the enforcement of EU secondary law, and the “slow development of enhanced cooperation” (De Witte 2017: 9 ff.) have paved the way to an incremental process of differentiation, which thus takes numerous forms (Avbelj 2013: 192 ff.) and allows accommodating the existing diversity among member states. At this stage, (further) differentiation, understood as a non-​uniform application of EU norms, seems unavoidable. As Schmidt (2019: 294) puts it: “The future of Europe will be one of differentiated integration. The question is not whether but how that differentiation will develop, since the EU is already differentiated”. The past and future enlargements of the EU, the deepening of its competences and the Eurosceptic backlash visible in some member states (up to the point of the withdrawal from the Union in the British case), have made the reflections on differentiation ever more compelling. In view of the importance this trend has gained in the recent past, and is likely to continue to gain in the future, the issue of its implications for the EU’s institutional structure arguably deserves reflection. This chapter specifically focuses on the European Commission, one of the EU institutions with the most evident supranational pedigree, in its quality as the Guardian of the Treaties and as the promoter of the Union’s general interest. This inquiry is conducted against a background characterized by recent, yet profound, changes in the conceptualization of this phenomenon within the EU. The traditional categorization of the modes of differentiation –​“multiple speeds” or “two-​speed Europe” (Piris 2012), “federal core of Europe” and “flexibility à la carte” (Thym 2018: 848 ff.) and their long-​standing combination in practice (Markakis 2020: 490) –​has suffered an evolution that has affected the very concept of differentiation in the EU. Some have recently argued in favour of the creation of “clubs of Member States” (Demertzis et al. 2018), and the idea of a “soft-​core Europe” with “overlapping clusters of member-​states participating in the EU’s many different policy

216

DOI: 10.4324/9780429054136-16

European Commission

communities” has been advanced too (Schmidt 2019: 306). This has led some scholars to urge a change of paradigm and to suggest replacing the narrow notion of differentiated integration by that of differentiation at large, which better captures the state of today’s EU (Leruth, Gänzle and Trondal, in the Introduction). Indeed, differentiation not only encompasses the attempts of groups of member states to strengthen their cooperation while other EU countries remain unwilling to further integrate but also covers processes of (differentiated) disintegration whereby a Member State, or a part thereof, leaves the Union (see introduction to this volume). Therefore, according to the approach taken in this handbook, differentiation is used in this chapter as “heterogeneity” and does not necessarily describe a movement towards more or less cooperation (Leruth, Gänzle and Trondal, in the Introduction).This move towards increasing heterogeneity is an additional challenge for the EU’s institutional design and the EU’s relationship to its member states, where these issues are already at the core of the long-​standing debate on the EU’s legitimacy and effectiveness owing, among others, to the EU’s undefined nature between a state and an intergovernmental organization and its commonly used characterization as a sui generis entity. This is particularly visible in the domain of economic integration, where “[t]‌here is a host of questions regarding the composition and tasks of the EU/​Euro area institutions, agencies and bodies in an EU based on differentiation” (Markakis 2020: 503–​504). Depending on the domain of EU law at stake, the extent and breadth of the differentiation affect the choice of the legal instruments to be used, namely whether a Treaty amendment is needed, whether secondary law would suffice or whether an intergovernmental agreement could be resorted to. Additionally, the degree of flexibility that EU law allows in its implementation, especially in some policy areas like the EMU and the enforcement of the Schengen rules, is already remarkable (Markakis 2020: 504). The unity of the EU institutional framework, a mantra of the EU integration since the Merger Treaty of 1965, could though be endangered because of these dynamics. While the general preference in case of further differentiation seems to go in favour of keeping a single set of EU institutions (Schmidt 2019: 306), in some instances, like the Eurozone –​despite the expectation of its long-​term convergence with the EU27 –​there are already signs of ad hoc institutional arrangements in the form of the Eurogroup and the Euro summit, whose decisions nevertheless always need to be formally adopted by EU institutions in their own rights (that is, the European Council and the Council of the EU) to be valid within the EU. The question thus arises whether similar arrangements should be introduced with regard to other institutions, in particular the European Commission, as well as whether they should exist in all those areas in which the most severe forms of differentiation (i.e. enhanced cooperation and opt-​outs) are visible. In this chapter, we contend that, under the current Treaties, a “differentiated” Commission that would be mirrored either in its structure or in the manner in which it exercises its functions would distort the role this institution is called to play within the EU. This is the case notwithstanding its sometimes-​ambivalent attitude towards differentiation, and despite the fact that it has been traditionally assigned an important role in both the activation and the management of initiatives of differentiation. The next section serves to outline the attitude the Commission has adopted towards differentiation over time. An analysis of the role it may be called to play in its management follows. These two sections set the background for a third section, which highlights the reasons why the Commission itself should continue to be shielded from differentiation. The conclusion offers an overall reflection on the EU’s institutional framework in, and its potential adaptation to, a differentiated Union.

217

Diane Fromage and Cristina Fasone

The Commission’s ambivalent yet generally hostile attitude towards differentiation over time Somewhat paradoxically in view of its quality as the supranational institution par excellence, the Commission has also been at the origin of increased differentiation within the EU in some instances, that is differentiation has not only arisen on the member states’ initiatives. For instance, the EMU first introduced by the Maastricht Treaty is one of the fields in which differentiation among the EU member states is most far-​reaching. The EMU was the outcome of a process that started formally during the mid-​eighties when Commission President Jacques Delors relaunched the process of integration towards the completion of the single market with his 1985 White Paper. Relatedly, one could argue that the Commission indirectly played an important role in the emergence of differentiation, even if it was never Delors’ intention for all the member states meeting the convergence criteria to not become part of the Eurozone. Since the nineties, differentiation has, however, always been a concern for the Commission. When the discussions that preceded the adoption of the Amsterdam Treaty took place, France and Germany in particular “argued for a ‘constitutional’ recognition of differentiation, while the Commission and other member states tried to limit the impact of such a recognition in order to ensure a maximum of uniformity” (Hanf 2001: 4–​5; Tuytschaever 1999: 43 f). Along the same line, in a speech to the European Parliament on 3 October 2000, the President of the European Commission, Romano Prodi, stated that: [w]‌e need to simplify the mechanism for closer cooperation [now termed enhanced cooperation and introduced by the Amsterdam Treaty], at the same time leaving the door open to those Member States that wish to participate.The coherence of the acquis communautaire and the uniformity of the judicial framework must be preserved. Closer cooperation should be an inclusive not an exclusive instrument, but no-​one should prevent a group of Member States from achieving the closer union to which the Treaties explicitly refer and which should be properly regulated within the framework of the Union’s institutions. The increasing heterogeneity of aims, ambitions and agendas across the member states, linked to processes like the Eurozone crisis, the refugee crisis, Brexit and the rule of law crisis, have led the Commission to follow a more pragmatic –​though not particularly successful and perhaps also slightly ambiguous –​approach towards differentiation. A first illustration of this attitude of the Commission is visible in Jean-​Claude Juncker’s opening speech to the European Parliament session held on 15 July 2014. He recalled the Commission’s task to defend the general interest of the Union but immediately after admitted that “We do not necessarily all have to move at the same speed –​the Treaties provide for that and we have seen that we can work with different arrangements” (…). “This is particularly important in the Eurozone” (Juncker 2014: 12), the area in which differentiation has developed in its most stable and far-​reaching form. Another illustration of the Commission’s position towards differentiation is visible in the White Paper on the future of Europe it published on 1 March 2017 in the wake of the Brexit referendum, and just a few months before the creation of the permanent structured cooperation (PESCO, another initiative leading to increased differentiation). It aimed to provide different roadmaps on differentiation for the evolution of the Union of 27 member states1 and envisioned five scenarios (Cuvyers 2017; Avbelj 2017). The first scenario, “Carrying on”, proposed to keep the status quo. The single market would have been strengthened, for example by creating an energy union; the functioning of the single

218

European Commission

currency would have been improved, and financial supervision reinforced. Defence cooperation would have been deepened, progresses made towards the setting up of a common asylum system and in foreign policy the EU would have spoken with one voice. The second scenario, “Nothing but the single market”, less ambitious than the former, postulated the deepening of the integration in the single market only and a prospective differentiated disintegration in the other EU policies, mainly achieved through de-​regulation. The third scenario, “Those who want more do more” revolved around several and asymmetric “coalitions of the willing”, whereby, depending on the issue at stake, different groups of member states could have decided to strengthen their cooperation, while other States lagged behind, based on the status quo. The fourth scenario, “Doing less more efficiently”, close to the second one, was aimed to focus the EU-​27 on a selected number of policy priorities related –​though not limited –​to the internal market and to reduce EU competences in other fields, giving powers back to member states, through a sort of uniform and homogenous disintegration. The fifth scenario, “Doing much more together”, created expectations for an “ever closer Union” and the pooling of more powers and resources to be managed by the EU in all domains, including defence, taxation and social policy. Given the prospective strengthening of supranational institutions and of the Commission especially, President Juncker unsurprisingly explicitly endorsed the fifth scenario in his State of the Union Speech of 13 September 2017. The following year, he somewhat appeared to disown his previous plea for differentiation speaking of “European sovereignty” and of the need to “move forward as one” (Juncker 2018: 5 and 13). In fact, this plea in favour of unity was made again a few months thereafter, on 6 December 2017, in reference to the Eurozone. On that date, the Commission presented its policy package on “Completing Europe’s Economic and Monetary Union”,2 and in its Roadmap on “Further steps towards completing Europe’s Economic and Monetary Union”3, it defined unity as its priority alongside efficiency and democratic accountability. In making this statement, the Commission acts against differentiation as is also foreseen by the Treaties which prescribe that the euro is the currency of the EMU which the EU shall establish (Treaty of European Union 2012: art. 3–​4), even if it is doubtful whether this is still a realistic goal at this stage. The Commission hence appears to be called to strike a (difficult) balance between rejection and acceptance of differentiation in the EU, as is also confirmed by the post-​2019 EU elections situation. On the one hand, Brexit has triggered the creation of an apparently cohesive front of the EU-​27 vis-​à-​vis the United Kingdom during the negotiations of both the withdrawal agreement and the partnership agreement. On the other, the heterogeneity amongst member states on the finalité of the EU integration process and on their commitment towards the Union and its values has increased. Also, the divide between Eurozone and non-​Eurozone member states has further consolidated in the meantime as well as the acceptance that decisions affecting the EMU are taken by Eurozone countries only (Smoleńska 2020).The new appointed Commission and its President, however, have not seconded the “translation” of the increased heterogeneity into stable procedures and organization arrangements and have, instead, acted in support of unity. In her speeches, both as President-​elect and in her first State of the Union address, in 2020, Ursula von der Leyen did not directly hint at differentiation. She acknowledged the differences shaping the EU. At the same time, however, she advocated that “to move forward we must now overcome the differences that have held us back” (Von der Leyen 2020). Such an approach by the Commission to proceed together within a common policy framework was already evident in the flagship measures of the Green Deal.The pandemic and the EU’s response to date have further reinforced this trend on the part of the Commission and its plea for moving

219

Diane Fromage and Cristina Fasone

forward at the same pace (Kreilinger 2020), as is visible from the Next Generation EU initiative and the issuance of common debt it entails, to the regulation seeking to protect the EU’s budget from breaches of the rule of law, despite the controversy with Hungary and Poland surrounding its application. In fact, it would seem that whereas the gap between Eurozone and EU-​27 had been consolidated prior to the pandemic, it is now diminishing as a result of the adoption of these emergency measures to counter the negative economic consequences of the pandemic as well as a result of Brexit (Fromage 2020). It remains to be seen whether this approach by the Commission will survive the test bench of the enforcement of these measures adopted during the pandemic. It is generally unclear whether the member states will continue to act together or whether, on the contrary, the Eurozone will move forward on its own once the temporary instruments adopted in response to the pandemic have lapsed. Hence, even if its attitude is sometimes ambivalent, and even if in some instances, it may have itself been at the origin of further differentiation within the EU, overall, the Commission rather appears to support unity and uniformity over differentiation, in its most severe forms at least. This attitude could be explained by its quality as the EU institution in charge of veiling for the common, supranational, interest. As such, even if it does not act in full isolation of member states’ preferences and even if the personal preferences of the Commissioners or the Commission officials involved may play a role, given the observable continuity over time, it is in all likelihood rather its institutional position as it is anchored in the Treaties (that is its duty to “promote the general interest of the Union and take appropriate initiatives to that end” and to “exercise coordinating, executive and management functions” (art. 17–​1 TEU) that explains its attitude in our view. It could though also be related to its being always involved in the management of these initiatives as illustrated in the next section, since it is arguably easier and more comfortable a position for the Commission if all member states are on board than if only a limited number of them participate.

The Commission’s constant involvement in the management of initiatives of differentiation Like the preceding section, this part of the analysis also focuses primarily on the most serious forms of differentiation within the EU, that is enhanced cooperation, opt-​outs, inter se agreements and disintegration. However, it may be noted that the Commission plays a very important role in less serious forms of differentiation as well. It is indeed the Commission alone that may, for instance, decide to bring a case before the Court of Justice where a Member State has threatened unity within the EU by failing to fulfil its obligations under EU law (art. 258 TFEU as interpreted by the Court4). It is it, too, which monitors member states’ execution of their obligations in the field of economic and fiscal coordination and exercises discretion in so doing (Dawson 2019). Under the Cooperation and Verification Mechanism, the Commission is additionally in charge of monitoring the status of those States that entered the EU without fully complying with the accession criteria (i.e. Bulgaria and Romania) which, de facto, induces differentiation. Beyond the role it plays once a new State has joined the Union, the Commission is actually a key actor in both the accession and the “secession” procedures to the Union. It monitors candidate countries’ progress towards complying with the Copenhagen criteria (upon which membership is conditional), and as evidenced in the example of Brexit, it negotiates with the seceding Member State on behalf of the Union on the basis of a mandate of the European Council. The Commission’s role in the management of initiatives of enhanced cooperation is very important as well. It is the Council that authorizes a move towards enhanced cooperation (as

220

European Commission

opposed to an EU-​wide initiative, art. 20–​2 TEU). But, once initiatives of enhanced cooperation have been established, the Commission is, together with the participating member states, to “ensure that they promote participation by as many Member States as possible” (art. 328 TFEU), that is it is to act to reduce differentiation. Furthermore, it has to monitor these initiatives on whose developments it reports to the European Parliament and the Council. Most importantly, it assesses whether the necessary criteria for a Member State’s application for participation in an initiative of enhanced cooperation to be valid are met; it must provide a justification for its denial to do so where appropriate (art. 329–​1 and 331, TFEU). Moreover, both the Commission and the Council have to veil for the consistency between initiatives of enhanced cooperation and the policies of the Union as a whole (art. 334 TFEU), which implies that they are to keep these forms of differentiation in check. Probably for the very same reason, the Commission is also strongly involved in EU policy areas in which opt-​outs have been granted to some member states, such as the management of the monetary union, and the Area of Freedom, Security and Justice. For instance, at least every two years, it reports to the Council, together with the European Central Bank, on the progress towards meeting the criteria to adopt the common currency made by the member states with a derogation (art. 140–​1 TFEU). It is also called to propose the end of a derogation, i.e. the adoption of the Euro, to the Council (art. 140–​2 TFEU). Likewise, the Commission plays a key role in the management of the Schengen area, in which both some EU member states do not participate (opt-​outs) and other non-​EU States are involved (opt-​ins), making it therefore one of the paramount examples of differentiation within the EU. For example, participating States must inform the Commission if they intend to reinstate border controls, and it maintains the Schengen Information System, which is the database through which governments share information of common interest. Less expectedly perhaps in view of its quality as an EU institution, the Commission is equally involved in the implementation of the most recent inter se agreements concluded by (some of) the member states outside of the Treaty framework, most notably in the area of economic and fiscal cooperation (Treaty on Stability, Coordination and Governance [TSCG] and Treaty establishing the European Stability Mechanism). The legality of this involvement was, in fact, subject to judicial review by the Court of Justice, a fact that provided it with an opportunity to confirm the legality of such involvement whilst also defining its boundaries.The Court of Justice indeed established in its Pringle judgement that: the Member States are entitled, in areas which do not fall under the exclusive competence of the Union, to entrust tasks to the institutions, outside the framework of the Union, such as the task of coordinating a collective action undertaken by the Member States or managing financial assistance […], provided that those tasks do not alter the essential character of the powers conferred on those institutions by the EU and FEU Treaties.5 The Commission’s role is not trivial as is illustrated, for example, by the fact that it forms, together with the International Monetary Fund and the European Central, the “Troika” in charge of the monitoring of the economic performance of member states under assistance. However, its involvement in the implementation of these treaties is only logical in view of the strong intertwinement between them and the EU legal order. In fact, the TSCG was only resorted to because of the British opposition to any treaty amendment, but the measures it contains were originally meant to be enshrined in EU primary law, a development which, actually, should have followed within five years of its entry into force (art. 16 TSCG).

221

Diane Fromage and Cristina Fasone

Why the Commission as an institution must remain unaffected by differentiation within the EU Having now shown why it is that the Commission may be viewed as playing a key role in the management of differentiation, this section demonstrates why its structure and its institutional features, as well as its functions within the EU, argue against its becoming differentiated as well.

The structure of the Commission At first sight, the Commission, which must act to preserve the Union’s interest, appears (or should appear) “impenetrable” to the national logic and to member states’ interests in further integration or in disintegration. Articles 17(3) TEU and 245 TFEU) almost obsessively emphasize the independence from any government, institution, body, office or entity as an essential requirement for Commissioners-​to-​be and in office. No instructions may be taken by the Commissioners from the Members States and, conversely, the member states “shall not seek to influence them in the performance of their tasks” (art. 245(1) TFEU). Yet the relationship of the Commission with the member states severally or collectively, through the Council and the European Council, is an immanent feature of the institution starting from its appointment process (art. 17(7) TEU). The member states are the first point of reference for determining the composition of the Commission. The President, upon “suggestions made by Member States”, selects commissioners and many of those holding the position of Commissioner were previously members of national governments. For decades and until 2004, the “representation” of the member states within the Commission was differentiated (Oliver and Martenczuk 2018: 550). It comprised two nationals from the big countries –​France, Germany, Italy, Spain and the United Kingdom –​and just one from the remainder. Due to the Eastward enlargement and the risk of making the institution unworkable, Article 4 of Protocol no. 10 annexed to the Treaty of Nice established that the number of Commissioners would have been reduced to one per Member State (art. 213(1) of the then EC Treaty) and, after the accession of the 27th Member State, the number would have been further decreased to less than one per EU country. As is well known, the latter provision was never applied and all the attempts to reduce the size of the Commission, granting an “asymmetric representation” to the member states (see Macron 2017), albeit temporarily, or to differentiate the status of the Commissioners have failed to date (with the exceptions made below). Article 17(5) TEU, following the Treaty of Lisbon, provides that as from 1 November 2014, the Commission shall consist of a number of members equal to two-​thirds of the member states, including the President and the High Representative, through a rotation mechanism “reflecting the demographic and geographical range of all the Member States”. The provision has remained dead letter, following a commitment of the European Council prior to the second Irish referendum on the Lisbon Treaty of 2009,6 and then, since 2013, as a consequence of a European Council decision.7 Thus, the rule remains “one Commissioner per Member State”. The loss of EU membership, as in the case of Brexit, automatically leads to the removal of the Commissioner of the withdrawing State. However, the lack of appointment of a British Commissioner in the von der Leyen Commission, that took office on 1 December 2019 –​thus two months before Brexit occurred –​was an unprecedented decision (as the whole withdrawal process was, too) (Fabbrini and Schmidt 2019: 113–​114). In fact, it was the United Kingdom that did not nominate a commissioner despite a request of the President-​elect, who also consulted the Commission’s legal service to ascertain whether the Commission could legitimately work without a commissioner 222

European Commission

in the context of a further delay of the withdrawal, being made in this sense. Thus, in derogation of Article 17 TEU, the position remained vacant until its formal abolition. An issue that has been debated is whether the nationality of a certain Member State is not only a requirement to be appointed as Commissioner, but also for the allocation of portfolios within the Commission, which are assigned by the President. In theory, the distribution of the portfolios does not have to take into account the opt-​outs of the States of nationality of the candidates for the position of Commissioners. However, in practice, if one looks at the Commissioners in charge for the Euro –​the most prominent example of differentiation –​since the second Delors Commission, who negotiated the Treaty of Maastricht, the position has never been assigned to a national of a country outside the Eurozone. At the same time, with the notable exception of the draft of a Constitution of the EU presented to the European Parliament in 1994 (Herman 1994: art. 46), the nationality of a commissioner has never been considered as a condition to exclude her/​him from the Commission’s deliberations in areas subject to opt-​outs, enhanced cooperation or other forms of differentiation (Fasone and Fromage 2019). For example, at the time of negotiations of the Maastricht Treaty, when the United Kingdom obtained an opt-​out from the social policy field (amongst the others), formalized in Point 2 of the Protocol on Social Policy, thereby accepting to be excluded from the voting in the Council on the relevant legislative proposals, the possibility for the British Commissioners to continue to participate in the debates and votes in this field was not put into question, as British Commissioners are not UK representatives (Watson 1993; Brunazzo 2019: 17). All these elements show that notwithstanding its status as the supranational institution in charge of protecting the Union’s interest, the Commission’s composition is still permeable to national features. It is thus not its purely supranational nature that speaks against its becoming differentiated in response to the existing differentiation within the EU. In contrast, this is much more so due to a central principle of the Commission’s internal decision-​making, the principle of collegiality, which protects the Commission’s independence from the member states and its supranational character (Patrin 2018a).According to this principle, each member of the College of Commissioners is placed on an equal footing in the institution’s decision-​making (regardless of their nationality) and, as a consequence, Commissioners are collectively responsible for the decisions adopted and can be forced to resign altogether upon approval of a motion of censure: “as a body” according to the wording of Articles 17(8) TEU and 234(2) TFEU, since “as a body” they are subject to a vote of consent of the Parliament for the Commission’s appointment to be completed by the European Council. Notwithstanding this, it remains the case that coordination among Commission Directorate Generals (DGs) mostly runs smoothly, but not always, and that, at times, conflicts have arisen within the complex internal administrative apparatus supporting the work of the Commission due to the different policy agendas and interests the DGs may pursue (Wonka 2015). The importance of the principle of collegiality for the well-​functioning and for the legitimacy of the Commission also explains why attempts to differentiate the status and powers of Commissioners within the College have never been successful and it is unlikely and undesirable that they are proposed again in the future in response to the increasing differentiation in the EU. For example, in 2002–​2003, the Constitutional Convention drafting the Treaty establishing the Constitution for Europe discussed and ultimately rejected (see art. I-​26(6)) the idea to have the Commission composed of 13 Commissioners with full rights, in addition to the President and to the EU Minister for Foreign Affairs-​Vice-​President, plus a number of non-​voting Commissioners appointed by the President (Curtin 2009: 97–​98; Craig 2013: 92 ff.). It is certainly true that especially since the Treaty of Lisbon the Commission President (art. 17(6) TEU) and the High Representative of the Union for Foreign Affairs and Security Policy 223

Diane Fromage and Cristina Fasone

(art. 18 TEU), who is also a Commission Vice-​President, have a special position within the College of Commissioners. The President sets the guidelines for the Commission’s operation and the agenda, decides on its internal organization, appoints the vice-​presidents, in addition to the High Representative and can force any of the Commissioners to resign, including the High Representative, although in that case the European Council ultimately decides by qualified majority. Such a power may be used, in particular in cases of misconduct that risk impairing the legitimacy of the whole Commission. Besides triggering the adoption of a Code of Conduct for Commissioners (SEC(1999)1479), the developments that occurred with the Santer Commission led to endow the Commission President with a unilateral power to force the resignation of a single commissioner, whereas Article 217(4) TEC previously only foresaw a procedure whereby the President requested the resignation, but the College of Commissioners had to approve it, again, in the name of the principle of collegiality. Back then, the refusal of Commissioner Bresson to leave the office, upon request of the President, eventually led the entire Santer Commission to resign under the threat of a motion of censure (Oliver and Martenczuk 2018: 558).8 Indeed, it is the Commission President’s responsibility to ensure that the College acts “consistently, efficiently and as a collegiate body” art. 17(6), lit. b TEU. For her/​his part, the High Representative is subject to ad hoc appointment and dismissal procedures in which the European Council is the primary actor although the President of the Commission is involved as well (art. 18(1) TEU). He/​she has responsibilities in the area of EU external relations and in coordinating the Union external action but is bound by the Commission’s procedure as any other Commissioner (art. 18(4) TEU). No special voting rights are assigned to the President or to the High Representative: the Commission decides by a majority of its members (art. 250 TFEU), and not even the President is given a casting vote in the case of a stalemate. From this provision, back in the eighties, the Court of Justice derived the already mentioned principle of collective responsibility and collegiality of the Commission.9 With this regard and consistently with this approach, the proposals put forward over the last few years to create a “super Commissioner” in charge of the economic and financial affairs or specifically for the Euro area (Enderlein and Haas 2015; European Commission 2017; Macron 2017) have been met with scepticism (Karagiannis and Guidi 2014; Patrin 2018b). Although the proposals envisioned the creation of a European Finance Minister in slightly different ways,10 none of them has led to a reform in the end, partly because related proposals, like the one to create a budgetary instrument for the Euro area and that to establish a European Monetary Fund, have not seen the light either; partly due to the serious threat, which they would have posed to the principle of collegiality and to the institutional balance internal to the Commission, up to the point of requesting a Treaty reform in order to be pursued, according to some (Patrin 2018b; Fromage 2019: 68). The value of the institutional unity of the Commission (Ehlermann 1995: 23–​24; Fromage 2019: 67–​68) against internal differentiation has probably been reinforced also by other concurring factors. The College of Commissioners is not only “united” by the independence “beyond doubt” and the “general competence” of its members on EU affairs, but also by the “European commitment” of the Commissioners (art. 17(3) TEU). What this “European commitment” entails exactly remains to be determined. While this condition does not appear to preclude the appointment of candidates who have criticized the Union and have made pleas for more asymmetric integration, the selection of someone supporting withdrawals and disintegration, also as a consequence of implicit or explicit refusal of the values enshrined in article 2 TEU (e.g. democracy, rule of law etc.), seems questionable. This seems to be confirmed by the appointment process of the von der Leyen Commission. Although his nomination was finally rejected by the European Parliament’s JURI Committee due to concerns regarding his financial statements, in 224

European Commission

2019, the expected appointment of Laszlo Trocsanyi, a former Justice Minister of the Orban Government and elected MEP, as Commissioner for Neighbourhood and Enlargement was objected within and outside Parliament for his poor rule of law record as a Minister, and he was eventually replaced by Olivér Várhelyi. In sum, as it would create inequality among the Commissioners, any attempt by the Commission to operate on a differentiated basis, depending on the policy area in question and on nationality for example, would violate the principle of collegiality. It would also go against the proper exercise of the functions conferred upon it as shown subsequently.

The functions of the Commission The nature of the functions the Commission is conferred indeed runs in contrast to the perspective of an internal differentiation of the Commission to resemble the increasing and inevitable heterogeneity within the EU. The sight of those functions would be lost, and they could risk becoming meaningless. The exercise of these functions by a differentiated Commission could run in contrast to the Commission’s legitimacy. Indeed, this institution is perceived as an “honest broker”, a neutral guarantor of interstates compromises and a facilitator within several EU decision-​making procedures and in the enforcement of intergovernmental agreements (Thym 2018: 848–​858), and it must continue to be perceived as such or else it is arguably the very survival of the Union that is threatened. To reflect the EU’s variable geometry inside the Commission could impair its alleged neutrality from which the rationale for attributing its functions stems. Neutrality here does not imply that the Commission is seen as a-​political (Dawson 2019; Lupo 2018). Politics is indeed inherent in its work, but the Commission shall strive to keep its position impartial, non-​partisan and equidistant towards the national interests at stake when taking its decisions, and when negotiating and achieving compromises. Whether it is always able to deliver in this respect is another matter, but its present architecture and corollary of functions as designed by the Treaties hint to such a configuration for the Commission, which is, too, instrumental to protect the EU’s inter-​institutional balance. The Commission pursues the EU’s general interest, which is deemed to be unitary and consistent throughout the Union. With this regard, the Commission should be able to draw plural interests together. This role is exemplified by the function of external representation of the Union that is assigned to the Commission, with the exception of the common foreign and security policy and of specific cases foreseen by the Treaties (art. 17(1) TEU). As a consequence, on behalf of the Union, the Commission takes part in various international conferences on a wide range of topics, and it represents the Union in fora like the WTO panels and the International Tribunal for the Law of the Sea (see also art. 335 TFEU). The Commission is also responsible for the establishment of appropriate relations with the organs of the United Nations and its specialized agencies, the Council of Europe, the Organization for Security and Cooperation in Europe and the Organization for Economic Cooperation and Development (art. 220(1) TFEU). Moreover, the Commission negotiates international agreements on behalf of the EU –​again, the Brexit agreement and the partnership agreement with the United Kingdom being two clear ­examples –​except where they affect exclusively or primarily the common foreign and security policy (art. 218(3) TFEU), and it is given the same role in the crucial stage of the accession to the EU (art. 49 TEU), as mentioned previously. In a similar vein, as promoter of the EU’s general interest, the Commission can submit proposals to amend the Treaties, under the ordinary revision procedure (art. 48(2) TEU), it is also represented in the Convention, should it be established (art. 48(3) TEU), and participates in the simplified revision procedure (art. 48(6) 225

Diane Fromage and Cristina Fasone

TEU), where it has to be consulted by the European Council. The same reason can explain the conferral to the Commission of a sort of “monopoly” of the legislative initiative (art. 48(3) TEU). As guardian of the Treaties, together with the Court of Justice, the Commission is responsible for the proper application of EU law (art. 17(1) TEU), as well as of some intergovernmental agreements concluded amongst member states (see, e.g., arts 7 and 8 of the TSCG). Take the crucial role the Commission plays in initiating infringement proceedings, in particular in the administrative phase preceding the involvement of the Court of Justice, where needed (see art. 258 TFEU and above). How could the Commission be trusted to oversee the effective and uniform application of EU law on the part of the member states and to detect their violation of EU law, should it be thought to represent the interests of some EU countries only in a certain field? Similar considerations apply to its power, shared with other institutional actors, to activate both procedures under Article 7 TEU (paras 1 and 2) and to signal the clear risk of a serious breach by a Member State of the values referred to in Article 2 TEU or the existence of a serious and persistent breach of one or more of those values at national level. As an honest broker representing the Union’s interests (vis-​à-​vis the interests of the member states severally as well as sectoral interests of lobbies), the Commission, to some extent, leads the EU’s legislative procedures though not being a legislator. Even before the start of the legislative procedure, it is the Commission that collects and examines the reasoned opinions national parliaments issue on the ground of breaches of subsidiarity by legislative proposals falling beyond the EU’s exclusive competences taking the determinations it retains appropriate (Protocol no. 2 to the Lisbon Treaty). During the ordinary legislative procedure, the Commission is involved in every stage as a facilitator of an agreement between the Parliament and the Council, including in the Conciliation Committee (art. 294 TFEU). The Commission has the power to amend and withdraw its proposals (also on the ground of arts 289 and 293 TFEU) as long as the general interest of the Union (art. 17(1) TEU), the principle of sincere cooperation (art. 13(2) TEU) and the principle of inter-​institutional balance are respected. In particular, in the case of withdrawal, the decision has to be communicated to the Parliament and the Council at the earliest convenience and due reasons should be provided.11 Also, the Commission’s powers to adopt delegated acts (art. 290 TFEU) and implementing acts, whenever the measures require uniform conditions of implementation (art. 291 TFEU), and to exercise executive and management functions, especially in the field of competition and state aid, can be seen as confirmations of the Commission’s requirement to act as an honest broker amongst EU institutions, member states and different interests. The ability to mediate and to foster the finding of a compromise amongst this constellation of interests could hardly be achieved by a Commission that mirrors more or less stable clusters of national interests, as it would be in the event of the differentiation of the status of the Commissioners and of its internal structure. In between the role of honest broker and that of coordinator stands the exercise of the budgetary functions by the Commission. Indeed, the drafting of the multiannual financial framework (art. 312 TFEU), both of the draft Regulation and of the draft inter-​institutional agreement) and of the yearly budget of the EU (art. 314 TFEU) entails the reaching of a difficult compromise and coordination amongst interests that are oftentimes conflicting. Typically, member states are willing to reduce the size of the budget whereas the European Parliament wants to increase it, and then the interests of States that are net contributors confront those of the net beneficiaries. The same applies to the management and execution of the budget, in-​between the Union and the national levels. That of the EU budget, indeed, is possibly one of the first fields where the Commission has promoted the conclusion of inter-​institutional agreements since the eighties (Fasone and Lupo 2018). After the Treaty of Lisbon, Article 17(1)

226

European Commission

TEU) indirectly acknowledges this role of “coordinator” of the Commission who “shall initiate the Union’s annual and multiannual programming with a view to achieving inter-​institutional agreements”. The Commission also plays a primary role in coordinating certain policies, for example in the field of economic governance, employment and social policy. The coordination of national economic policies, an area in which the role of the EU has grown fast since the Eurozone crisis, the Commission has become the main interlocutor of national governments in procedures like those of the European and the National Semesters. For the Commission, it has not been easy to strike a fair balance amongst competing interests. For some, particularly for the member states from the south, the approach taken by the Commission has strengthened the austerity policies; for others, in particular for the so-​called frugal countries in the north, the flexibility pursued by the Commission in the implementation of the fiscal rules (see also above), especially from the appointment of the Juncker Commission onwards, has endangered the effectiveness of the EU’s economic governance (Menéndez 2020). In fact, the discretion the Commission has enjoyed in the coordination of economic policies “rendered it the target of attacks by national governments and interest groups that believe that it ought to exercise its powers more forcefully”, while, at the same time, there have been calls on the Commission “to treat the Member States more ‘fairly’—​especially from the Member States that are on the receiving end of those policies” (Markakis 2020: 504). Flexibility as a manifestation of differentiation in the implementation of EU law triggers problems of authority and legitimacy of the Commission, accused by some of being too lenient towards the austerity mantra, thereby favouring the political economic advocated by “frugal” EU countries, and by others of an unfair treatment amongst member states in favour of the so-​called PIIGS (Portugal, Ireland, Italy, Greece and Spain).

Conclusion As the preceding analysis has shown, even if the Commission has not remained indifferent to (growing trends of) differentiation within the EU, and even if it has been actively involved in its management, there are, however, very strong reasons why it should remain unaffected by those developments and, in particular, why its composition and its quality as the main promoter of the Union’s interests must not be altered by these trends. This is so despite its (alleged) tendency to push for more supranational integration. Put differently, future empirical research could examine why and how this preference for more Europe has not led to the Commission’s pushing for more differentiation where this could potentially have induced that more areas be governed at the EU level for some of the member states. Were the ongoing discussions on the reform of the accession procedure to lead to some changes being introduced in this field, an analysis of their consequences from the standpoint of differentiation and of the Commission’s role within it would also be necessary. Furthermore, it remains to be seen if and how the issue of differentiation is tackled by the Conference on the future of Europe.12 In any event, at a time where member states appear to disagree even about the definition of the core values that underpin the process of European integration (i.e. democracy and the rule of law), anything different would jeopardize unity within the EU and would probably lead to the end of the European integration experiment. Accordingly, if a group of member states pursuing further integration without the involvement of the whole of the EU-​27 were to refuse to confer tasks upon a supranational, EU-​wide, Commission, they would be better advised to conclude a separate agreement governed by its own institutional framework even though such an agreement

227

Diane Fromage and Cristina Fasone

risks marginalizing the overall influence of the Commission.This option is, however, only viable if sufficient safeguards for the overall coherence, unity and respect of the EU legal framework are put in place. In any case, it should be a solution of last resort in view of the additional fragmentation and complexity it would entail, especially against a background in which the EU is already particularly complex and difficult to understand for its citizens.

Notes 1 The occasion of the presentation of the White Paper was the Rome Summit of 25 March 2017, the celebration of the 60th anniversary since the ratification of the Rome Treaty. 2 European Commission (2017), “Policy Package on ‘Completing Europe’s Economic and Monetary Union”, 6 December 2017. 3 European Commission (2017), “Communication on Further steps towards completing Europe’s Economic and Monetary Union: a Roadmap”, COM(2017) 821 final, 6 December 2017. 4 Case C-​70/​99, Commission of the European Communities v. Portuguese Republic, ECLI:EU:C:2001:355, par. 17. 5 C-​370/​12, Thomas Pringle v. Government of Ireland and Others, ECLI:EU:C:2012:756, par. 158. 6 European Council (2009), “Presidency Conclusions of the meeting of 18–​19 June 2009”, Document 11225/​2/​09 REV 2, 10 July 2009, point I.2. 7 European Council Decision no. 2013/​272/​EU of 22 May 2013 concerning the number of members of the European Commission, OJ L 165/​98, 18 June 2013. 8 Case C-​432/​04, Commission of the European Communities v. Édith Cresson, ECLI:EU:C:2006:455, on the standards of conducts Commissioners must observe and to ensure that the general interest of the Community takes precedence over national and personal interests. 9 Case 5/​85, AKZO Chemie BV and AKZO Chemie UK Ltd v. Commission of the European Communities, ECLI:EU:C:1986:328, par. 30; Case C-​137/​92 P, Commission of the European Communities v. BASF AG and Others, ECLI:EU:C:1994:247, par. 64. 10 Macron, for instance, did not see the Minister appointed from within the Commission. 11 Case C-​409/​13, Council of the European Union v. European Commission, ECLI:EU:C:2015:217; Lupo (2018). 12 Joint Declaration on the Conference on the Future of Europe, Brussels, 5 March 2021, https://​data. consil​ium.eur​opa.eu/​doc/​docum​ent/​ST-​6796-​2021-​INIT/​en/​pdf.

Bibliography Avbelj, M. (2013) Differentiated Integration –​Farewell to the EU-​27?. German Law Journal, 14(1), pp. 191–​ 211. doi: 10.1017/​S2071832200001760. Avbelj, M. (2017) What Future for the European Union?. Center for Global Constitutionalism WBZ –​Berlin Social Science Center Discussion Paper SP IV 2017 –​802. Available at: https://​bib​liot​hek.wzb.eu/​pdf/​ 2017/​iv17-​802.pdf (Accessed: 1 February 2021). Brunazzo, M. (2019) The Evolution of EU Differentiated Integration between Crisesand Dilemmas. EUIdea Research Papers No. 1. Available at: www.iai.it/​sites/​default/​files/​euidea_​r p_​1.pdf (Accessed: 1 February 2021). Craig, P. (2013) The Lisbon Treaty: Law, Politics and Treaty Reform. Oxford: Oxford University Press. Curtin, D. (2009) Executive Power of the European Union: Law, Practices and the Living Constitution. Oxford: Oxford University Press. Cuvyers, A. (2017) Five Scenarios for Europe –​Understanding the EU Commission’s White Paper on the Future of Europe. VerfBlog, April 10. Available at: https://​verf​assu​ngsb​log.de/​five-​scenar​ios-​for-​ eur​ope-​unders​tand​ing-​the-​eu-​comm​issi​ons-​white-​paper-​on-​the-​fut​ure-​of-​eur​ope/​ (Accessed: 1 February 2021). Dawson, M. (2019) How can EU law contain economic discretion?’, in Mendes J. (ed.) EU Executive Discretion and the Limits of Law. Oxford: Oxford University Press, pp. 85–​106. De Witte, B. (2017) Variable geometry and differentiation as structural features of the EU legal order’, in De Witte, B., Ott A. and Vos E. (eds) Between Flexibility and Disintegration:The Trajectory of Differentiation in EU Law. Cheltenham: Edward Elgar Publishing, pp. 9–​27. 228

European Commission

Demertzis, M. et al. (2018) One Size Does Not Fit All: European Integration by Differentiation, Bruegel Policy Brief, 19 September. Available at: http://​brue​gel.org/​2018/​09/​one-​sized​oes-​not-​fit-​all/​ (Accessed: 1 February 2021). Ehlermann, C.-​D. (1995) Increased Differentiation or Stronger Uniformity. Robert Schuman Centre EUI Working Paper no. 95/​21: Italy. Available at: http://​diana-​n.iue.it:8080/​bitstr​eam/​han​dle/​1814/​1396/​ 95_​21.pdf?seque​nce=​1&isAllo​wed=​y (Accessed: 1 February 2021). Enderlein, H. and Haas, J. (2015) What Would a European Finance Minister Do? A Proposal. Jacque Delors Institute Policy Paper 145. Available at: www.institutdelors.eu/​wp-​content/​uploads/​2018/​01/​ ministrefinanceeuropeenjdi-​ben.pdf?pdf=​ok (Accessed: 1 February 2021). European Commission (2017) ‘Communication from the Commission to the European Parliament, the European Council, the Council and the European Central Bank on A European Minister of Economy and Finance’, COM(2017) 823, europa.eu. Available at: https://​eur-​lex.eur​opa.eu/​legal-​cont​ent/​EN/​ TXT/​PDF/​?uri=​CELEX:5201​7DC0​823&from=​EN (Accessed: 1 February 2021). Fabbrini, F. and Schmidt, R. (2019) The Extension of UK Membership in the EU: Causes and Consequences. European Journal of Legal Studies, Special issue ‘The Brexit Negotiations and the May Government’, pp. 87–​116. doi: 10.2924/​EJLS.2019.004 Fasone, C. and Fromage, D. (2019) The European Commission and the member states facing differentiated (dis)integration. The challenges of Brexit’, in Gänzle, S., Leruth, B., and Trondal, J. (eds) Differentiated Integration and Disintegration in a Post-​Brexit Era. London: Routledge, pp. 110–​126. Fasone, C. and Lupo, N. (2018) The union budget and the budgetary procedure’, in Schütze, R. and Tridimas,T. (eds) Oxford Principles of European Union Law. Oxford: Oxford University Press, pp. 809–​846. Fromage, D. (2019) Moving beyond ‘institutional unity’ within the EU? Euro area versus non-​Euro area representation in the EU institutions’, in Fromage, D. and De Witte, B. (eds) Recent Evolutions in the Economic and Monetary Union and the European Banking Union: A Reflection, Maastricht University Faculty of Law Working Paper Series 2019/​03, pp. 61–​73. Fromage, D. (2020) Towards increasing unity and continuing executive predominance within the E(M)U post-​COVID? Legal Issues of Economic Integration, Special issue ‘The future of EMU architecture post-​ COVID’, pp. 385–​ 408. Available at: https://​kluw​erla​wonl​ine.com/​Jou​r nal​Arti​cle/​Legal+​Iss​ues+​of+​ Econo​mic+​Inte​grat​ion/​47.4/​LEIE​2020​022 (Accessed: 19 April 2021). Hanf, D. (2001) Flexibility clauses in the founding Treaties: from Rome to Nice, in De Witte, B. et al. (eds) The many faces of differentiation in EU Law. Antwerp-​Oxford-​New York: Intersentia, pp. 3–​26. Herman, F. (1994) Second Report on the Constitution of the European Union. Report of the Committee on Institutional Affairs on the Constitution of the European Union A3–​0064/​94. Available at: www.cvce. eu/​content/​publication/​2005/​1/​6/​b198b297-​f3aa-​418d-​a495-​7b0cd9c85d71/​publishable_​en.pdf (Accessed: 1 February 2021). Juncker, J.-​C. (2014) A New Start for Europe: My Agenda for Jobs, Growth, Fairness and Democratic Change, Political Guidelines for the Next EuropeanCommission, July 15 . Available at: www.eesc.europa.eu/​ resources/​docs/​jean-​claude-​juncker-​-​-​political-​guidelines.pdf (Accessed: 1 February 2021). Juncker, J.-​C. (2018) State of the Union Speech 2018 –​The Hour of European Sovereignty, europa.eu. Available at: https://​ec.eur​opa.eu/​com​miss​ion/​sites/​beta-​politi​cal/​files/​soteu2​018-​spee​ch_​e​n_​0.pdf (Accessed: 1 February 2021). Karagiannis,Y. and Guidi, M. (2014) Institutional change and continuity in the European Union:The super-​ commissioner saga. Acta Politica, 49(2), pp. 174–​195.; doi: 10.1057/​ap.2013.21 Kreilinger, V. (2020) Tectonic Shifts in the EU’s Institutional System. LUISS School of Governance Working Paper 62. Available at: https://​sog.luiss.it/​sites/​sog.luiss.it/​files/​WP%2062_​T​ecto​nic%20shi​fts%20 in%20the%20EU’s%20in​stit​utio​nal%20s​yste​m_​0.pdf (Accessed: 1 February 2021). Leruth, B., Gänzle, S. and Trondal, J. (2019) Differentiated Integration and Disintegration in the EU after Brexit: Risks and Opportunities. Journal of Common Market Studies, 57(6), pp. 1383–​1394. doi: 10.1111/​ jcms.12957. Leruth, B., Gänzle, S. and Trondal, J. (2021) Differentiation in the European Union as a Field of Study, in Leruth, B., Gänzle, S. and Trondal, J. (eds) Routledge Handbook of Differentiation in the European Union. London: Routledge. Lupo, N. (2018) The Commission’s Power to Withdraw Legislative Proposals and its ‘Parliamentarisation’ Between Technical and Political Grounds. European Constitutional Law Review, 14(2), pp. 311–​331. Macron, E. (2017) ‘President Macron’s Speech at Sorbonne’, excerpts from the English translation, blogs. ouest-​, September 26. Available at: http://​intern​atio​nal.blogs.ouest-​fra​nce.fr/​arch​ive/​2017/​09/​29/​mac​ ron-​sorbo​nne-​verba​tim-​eur​ope-​18583.html (Accessed: 1 February 2021). 229

Diane Fromage and Cristina Fasone

Markakis, M. (2020) Differentiated Integration and Disintegration in the EU: Brexit, the Eurozone Crisis, and Other Troubles. Journal of International Economic Law, 23, pp. 489–​507. doi: 10.1093/​jiel/​jgaa004. Menéndez, A. (2020) The Sleep of Numbers Produces Monsters:The Case of ‘Numerical Rules’, in Caruso, C. and Morvillo, M. (eds) Governing With Numbers: Economic Indicators and the Budget Decision in the Constitutional State. Bologna: Il Mulino, pp. 93–​118. Oliver, P. and Martenczuk, B. (2018) The Commission, in Schütze, R. and Tridimas,T. (eds) Oxford Principles of European Union Law. Oxford: Oxford University Press, pp. 549–​580. Patrin, M. (2018a) The Legal Nature of the Principle of Colegiality: A General Principle of EU Law?. CERiM Online Paper no. 11. Available at: https://​pap​ers.ssrn.com/​sol3/​pap​ers.cfm?abst​ract​_​id=​3209​863 (Accessed: 1 February 2021). Patrin, M. (2018b) A European Finance Minister: form follows function, but is it legal? A legal analysis of the European Commission’s proposal to create a European Minister of Economy and Finance. ADEMU Working Paper Series 2018/​096. Available at: http://​ademu-​proj​ect.eu/​wp-​cont​ent/​uplo​ads/​2018/​05/​0096-​A-​ Europ​ean-​Fina​nce-​Minis​ter-​Form-​foll​ows-​funct​ion-​but-​is-​it-​legal-​1.pdf (Accessed: 1 February 2021). Piris, J.-​C. (2012) The future of Europe:Towards a Two-​Speed EU?. Cambridge: Cambridge University Press. Schmidt, V. (2019) The Future of Differentiated Integration: A ‘Soft-​Core,’ Multi-​Clustered Europe of Overlapping Policy Communities. Comparative European Politics, 17(2), pp. 294–​315. Smoleńska, A. (2020) Eurozone Reform and Differentiated Integration after Brexit. DCU Brexit Institute News, January 15 . Available at: http://​dcu​brex​itin​stit​ute.eu/​2020/​01/​euroz​one-​ref​orm-​and-​dif​f ere​ntia​ ted-​inte​g rat​ion-​after-​bre​xit/​ (Accessed 1 February 2021). Stubb, A. (2002) Negotiating flexibility in the European Union:Amsterdam, Nice and Beyond. Basingstoke: Palgrave Macmillan. Treaty of European Union (TEU). (2012). Consolidated version of the Treaty on European Union. Official Journal of the European Union, p. C326/​13.. Available at: https://​eur-​lex.eur​opa.eu/​resou​rce.html?uri=​cel​ lar:2bf14​0bf-​a3f8-​4ab2-​b506 fd71826e6da6.0023.02/​DOC_​1&format=​PDF (Accessed: 13 April 2021). Thym, D. (2018) Supranational Differentiation and Enhanced Cooperation, in Schütze, R. and Tridimas, T. (eds) Oxford Principles of European Union Law. Oxford: Oxford University Press, pp. 847–​883. Tuytschaever, F. (1999) Differentiation in European Union Law. Oxford: Hart Publishing. Von der Leyen, U. (2020) State of the Union Address at the European Parliamentary Plenary, europa.eu. Available at: https://​ec.eur​opa.eu/​com​miss​ion/​pres​scor​ner/​det​ail/​en/​SPE​ECH_​20_​1​655 (Accessed: 1 February 2021). Watson, P. (1993) Social Policy after Maastricht. Common Market Law Review, 30(3), pp. 481–​513. Available at: https://​hei​nonl​ine.org/​HOL/​Land​ingP​age?han​dle=​hein.klu​wer/​cmlr0​030&div=​31&id=​&page=​ (Accessed: 1 February 2021). Wonka, A. (2015) The European Commission’, in Richardson, J. and Mazey, S. (eds) European Union. Power and policy-​making, London and New York: Routledge, pp. 83–​105.

230

15 Differentiation and the European Court of Justice Sabine Saurugger and Fabien Terpan

Introduction In any democratic political system, the purpose of a court is to uphold the rule of law, i.e. to ensure that the law is respected by all, including the government, on the whole territory. Supreme courts and constitutional courts play an essential role in this regard: situated at the top of the domestic legal order, they guarantee the stability of the whole system. In other words, they centrally contribute to avoiding the differential impact of law on a given territory. The European Court of Justice (ECJ), whose role is to ‘ensure that in the interpretation and application of the Treaties the law is observed’ (Treaty on European Union (TEU) 2012: Article 19), bears similarities with supreme and constitutional courts. Indeed, the ECJ can be seen as the cornerstone of the EU’s normative system, defined as a coherent set of rules and governance methods that are (supposed to be) applied in a uniform manner on the territory of the EU. The Court has contributed consistently to the constitutionalization of the EU and has brought EU law close to a federal law (Capelletti et al. 1986; Weiler 1991; Stone 2004; Saurugger and Terpan 2017). However, while this legal understanding of the Court’s role is important, the uniform application of EU law has been questioned empirically (European Parliament 2015). The aim of this chapter is to analyze the place of the ECJ in the increasingly differentiated political system of the EU. If a constitutional court’s role is to guarantee the homogenous application of law in a political system, is increasing legal and judicial differentiation blocking its actions? Indeed, the Court is facing two main sources of differentiation, coming from both national courts and national governments. First, the constitutionalization process as defended by the ECJ, ideally leading to a uniform application of EU law in all member states, has faced resistance from national courts, and constitutional courts more specifically. While the ECJ does not allow for any limitations in the supremacy of EU law over domestic law, several constitutional courts still consider that member states’ constitutions are still situated at the top of the hierarchy of norms. To account for this complex situation, a number of authors have tended to describe the EU judicial system as one of constitutional pluralism (Walker 2002; Avbelj and Komarek 2012). This notion encapsulates the relation between two overlapping legal orders, where no clear hierarchy exists between the two. Contrary to the Kelsenian hierarchical model of norms, constitutional pluralism is a heterarchical system where each of the two legal orders remains autonomous and DOI: 10.4324/9780429054136-17

231

Sabine Saurugger and Fabien Terpan

claims for sovereignty in its own terms. Constitutional pluralism helps to explain why the ECJ, after having created the doctrine of supremacy in its 1964 Flaminio Costa vs. E.N.E.L. ruling (1964: Case 6–​64), had to accept the limitations posed by national constitutional courts, and why, in turn, national courts have increasingly accepted this doctrine without abandoning the idea that national constitutions can, in limited cases, block the application of EU law. Hence, while the coherent rule of law is the central value of the European legal and political system, constitutional pluralism has allowed to question the uniformity of law in the EU and hence introduced a specific form of legal differentiation –​judicial differentiation –​into the system. The second source of differentiation comes from national governments. Here, differentiation can be both vertical and horizontal (Leuffen et al. 2012; Leruth et al. 2019; Fossum 2019) and may either be constitutionalized at EU level or result from resistance at domestic level. Since the 1990s the member states acting as the ‘masters’ of the Treaties have included in the EU primary law different forms of vertical differentiation: indeed, the Treaties now provide for a variety of governance methods as opposed to the EU relying on the so-​called community method. Treaty provisions also lay down different forms of horizontal differentiation, or variable geometry, where only some of the member states actually participate in a policy, or sub-​policy, and third states may apply EU rules while remaining outside the EU system of governance. This evolution towards constitutionalized differentiation has become even more salient in recent years when leading political actors and institutions have suggested and decided differentiation as future scenario of European integration. At the same time, differentiation also results from a flexible application of EU common rules at domestic level (De Witte 2019), where national governments decide to overtly resist the uniform application of EU law. The use of the notion of ‘illiberal democracy’ can be seen as an attempt to oppose a process whereby EU legal integration would increasingly influence national normative and political systems. Overall, differentiation is supposed to accommodate the diverging preferences of national actors through flexible integrative means. Bending to the will of sovereign member states and accommodating the positions defended by national constitutional courts can be viewed as a way to strengthen the legitimacy and democratic foundations of the EU. However, differentiation in its different forms also has the potential to threaten the rule of law –​and consequently the democratic foundations of the EU –​because it increases diversity and potentially weakens the EU’s normative system. While the rule of law is supposed to ensure that people are equal before the law (Shklar 1987; Craig 1997; May 2014), differentiation creates exceptions to the common rules that potentially challenge the existence of a rule of law system. To answer the question whether the increasing legal and judicial differentiation of the EU blocks the ECJ’s actions, we will analyze, in ‘The ECJ challenged by national courts: homogenous application of EU law vs. constitutional pluralism’ section, when and why national courts have challenged the ECJ, leading to constitutional pluralism and a specific form of judicial differentiation. ‘The ECJ challenged by national governments: pushing constitutional pluralism even further?’ section will examine under which circumstances national governments have also undermined the uniform application of EU law through actions at both EU and national level. Finally, in the ‘Is the ECJ and the EU’s legal system adaptive enough to protect the rule of law against differentiation?’ section, we will discuss the way the ECJ, and the EU political system more generally, address the multiple challenges posed by legal differentiation. Using approaches from legal, sociological as well as network studies traditions, this chapter shows that while legal differentiation is indeed increasing, the ECJ adopts a nuanced strategy, based on a constant dialogue with national courts, in order to keep judicial differentiation at a minimum. 232

European Court of Justice

The ECJ challenged by national courts: homogenous application of EU law vs. constitutional pluralism The Court was created in 1952 with a view to controlling the High Authority of the European Coal and Steel Community (ECSC) and to protecting the sovereignty of the member states. After no more than a decade of functioning, it became a supranational judicial body independent from the member states. The Court has issued rulings, which have greatly influenced the European integration process. Though the ECJ was not established as a constitutional court, it has developed into a powerful and bold adjudicator.

EU constitutionalization and the homogenous application of EU law To do so, the Court has interpreted extensively its main task under the EU treaties, which is to ensure that in the interpretation and application of the Treaties the law is observed. Indeed, the ECJ is endowed with important prerogatives such as the interpretation of the treaties, and the validity and interpretation of acts of the institutions, bodies, offices or agencies of the Union. Most of all, the ECJ ensures that EU law is applied by both EU institutions and member states. The ECJ performs much the same function as the US Supreme Court with respect to issues regarding the division of powers between EU institutions, and between the EU and the member states. The judgements of the ECJ are binding on the member states, including their judicial institutions as well as their judicial bodies. Making use of its treaty-​based powers, the Court has developed audacious doctrines aiming at a homogenous application of EU law. Acting as a constitutional adjudicator, the European Court has interpreted Treaty provisions over time, often departing from the text and taking into account the objectives and the system of the treaties. This bold case law has resulted in the establishment of two major constitutional principles (direct effect and supremacy), the organization of horizontal and vertical powers, and the progressive inclusion of fundamental rights in EU law. The Treaties have been framed by the ECJ as the ‘constitutional charter’ of the EU (‘Les Verts’ v European Parliament 1986: Case 294–​83). Through a constitutionalization of the treaties, the ECJ has contributed to its own transformation into a constitutional court, a process that has been acknowledged by the academia. The constitutional role of the ECJ is based on the assumption that EU Treaties are constitutional by nature, even if they are not strictly called a ‘constitution’ (Stein 1981; Weiler 1991; Hartley 1996). The Treaties provide the various functions of a constitution (Sunstein 2001) organizing the hierarchy of legal norms; defining the conditions of exercise of political power; distributing competences among different levels of governance and administration; and protecting the rights and freedoms of individuals. Over time, these obligations in ‘primary law’ were supplemented by additional rules of ‘secondary law’, contained in regulations, directives and decisions, all of which were binding for and applicable to all member states. This collection of legal norms constituted the EU’s legal order marked by a sense of unity and uniform application across the territory of all member states. The ECJ insisted on the fact that EU legal rules were ‘fully applicable at the same time and with identical effects over the whole territory of the Community’ (Commission of the Europeans Communities v Italy 1972: Case 48–​71).

Constitutional pluralism Having said that, it is important to underline that the European Court is not (yet) clearly established as the authoritative constitutional interpreter of the legal system, due to competition with national constitutional courts. The European Court cannot repeal national laws that 233

Sabine Saurugger and Fabien Terpan

violate European law. Its judgement is simply a declaration that a violation of EU law has taken place. Although the ECJ ruling possesses the force of res judicata, it remains within the competence of the member states to remove national laws or practices that are incompatible with EU law. As Rosenfeld put it, ‘whereas the US federal government was able to send federal marshals to force resisting state officials to implement Supreme Court desegregation decisions, nothing comparable exists within the EU to back up the ECJ, if needed’ (Rosenfeld 2006: 628). Both the US Supreme Court and the European Court give binding decisions, but the ECJ does not have as many tools at its disposal as its US counterpart when it comes to enforcing decisions. EU judicial mechanisms are still inspired by the logic of international law. Yet, since the 1990s, the Commission can ask the Court to decide on financial sanctions in cases where its rulings are not complied with, and this makes the Court’s decisions more coercive than in the past. But in the EU, the authority of the European Court’s rulings mainly derives from the attitude of national courts and their willingness to ensure effective judicial cooperation with the ECJ. The coexistence of these two kinds of supreme rules (EU treaties and national constitutions) is often depicted by the legal doctrine as one of ‘constitutional pluralism’ (Walker 2002; Avbelj and Komárek 2012; Jaklic 2014), a concept that captures the lack of consensus over the hierarchy of norms, also called the conflict over the Kompetenz-​Kompetenz. In other words, it refers to the problem of which court –​the ECJ or the national constitutional court –​gets to decide where the EU’s authority ends and the member states authority begins (Beck 2013; Kelemen 2019), a decision highly dependent of the context in which law is developed (Selznick 2003; Mak and Taekema 2016).The limits that constitutional courts, and first of all the German Federal Constitutional Court (FCC) place on the supremacy of European law, led to a situation where national constitutional courts have, on the one hand, accepted the principle that EU law has supremacy over national law in areas where member states have conferred competence onto the EU, but where they argue, on the other hand, to retain Kompetenz-​Kompetenz. This term means that they can decide whether or not an area falls within the EU’s competence or whether, if the ECJ rules on the issue, it is considered to be ultra vires, hence outside its competence. In these areas the national courts argue that a state holds an inviolable ‘constitutional identity’. When the European Court rendered its first ground-​breaking judgements in Van Gend en Loos (1963: Case 26–​62) and Flaminio Costa vs. E.N.E.L. (1964: Case 6–​64) in the early 1960s, it ruled that the EC Treaty was establishing an autonomous legal order, which was prior to those of the member states and which could have (in part) direct effect on them. The German FCC agreed on this position by saying that the European Court was the jurisdiction of a sui generis community in the process of progressive integration, rendering legal rules that were neither national nor international law. It founded a domestic legal basis to adopt the Court’s positions in Article 23 of the Basic Law (Constitution of Federal Republic of Germany 1949), dealing with Germany’s participation in European integration, and argued that this article would not provide for any transfer of sovereignty, but would allow for an acknowledgment and implementation of the Court’s legal acts. Since the 1970s onwards, however, the FCC has had several occasions to clarify its position of the relationship between EU law and German law. In a series of rulings, it has shown reluctance to accept the European Court’s position on primacy, making it dependant on the strengthening of democracy and the rule of law at EU level. The first conflict between the two courts occurred in 1974 (Davies 2012) just a few years after Internationale Handelsgesellschaft (1970: Case 11–​70), where the European Court had reaffirmed the primacy principle. In Solange I, the FCC ruled in an export licence case, referred to it by a German administrative court that considered that a European act edict would be inapplicable because it was contrary to fundamental rights enshrined in German constitutional law. The European Court, however, had previously confirmed the legality of the regulation. In its 234

European Court of Justice

famous ‘Solange I’ decision, the FCC reserved the right to review the compatibility of European and German law in each individual case ‘as long as’ (solange) the integration process has not progressed so far that Community law receives a catalogue of fundamental rights decided by the parliament and of confirmed validity. Any German court could refer the case to the FCC if it ‘concerns Community law and insofar it conflicts with one of the fundamental rights of the Basic Law’ (BVerfGE 1974: Case 37–​271 2 BvL 52/​71). This position was reversed 12 years later by the ‘Solange II’ decision in 1986, most certainly because in the meantime the European Court had developed its jurisprudence on the protection of human rights. In a similar case, the FCC ruled that it would no longer review the Court’s jurisprudence by the standard of fundamental rights contained in the Basic Law and ‘no longer exercise its jurisdiction to decide on the applicability of secondary Community legislation’ as long as ‘the European Communities, in particular European Court case law, generally ensure the effective protection of fundamental rights (…) and insofar as they generally safeguard the essential content of fundamental rights’ (BVerfGE 1986: Case 73–​3392 BvR 197/​83).The FCC signalled herewith that it would be willing to trust the Court’s decisions as long as it continued to protect fundamental rights adequately. However, a year after the member states signed the Treaty of Maastricht, where several new competences were conferred upon the EU, particularly in the area of the Economic and Monetary Union, the FCC rendered another far-​reaching decision on EU integration (BVerfG 1993: 2BvR 2134; 2BvR 2159/​92), which made it clear that if European institutions were to adopt legislative instruments transgressing the rights conferred upon them by the treaties, it would not consider these instruments legally binding within the sphere of German sovereignty. According to the German Court, the EU as a Staatenbund (association of states) has only been transferred a very limited amount of sovereignty, a situation that preserved the FCC’s right to review whether legal acts issued by the European Court remained within the scope of EU powers or not (Thym 2009). As Grimm (2009) states, the FCC hereby ultimately moved itself into a situation of competition with the ECJ. The FCC ‘theoretically refused to recognize the primacy of Community law and the ECJ’s prerogative to delimit the competences of the Community in a given case’ (Stein 2001: 223). In 2009, the FCC returned to the stage with strong reservations with regard to the Treaty of Lisbon, in spelling out the constitutional limits of the integration process and in asserting a traditional view of sovereignty (BVerfGE 2009: 123–​267; Thym 2009), a view that was widely taken over by the Polish, Hungarian and Czech Supreme Courts (Beck 2013). It thus asserts its own jurisdiction of last resort in relation to any issues involving potential clashes between EU and German constitutional law. However, in its Honeywell decision in 2010 (BVerfG 2010: 2 BvR 2661/​06; see Payandeh 2011), the German Court refined its own doctrine in a way that was reminiscent of its evolution from ‘Solange I’ to ‘Solange II’. This decision illustrates the complex relationship between the ECJ and the German Court based on judicial flexibility, political sensitivity and mutual forbearance (Beck 2013). The decision is directly related to the ECJ’s controversial Werner Mangold ruling (2005: C-​144/​04). The latter was based on a referral from the Labour Court in Munich, which concerned the compatibility between German law (Employment Promotion Act 1996) and EU law prohibiting discrimination on the grounds of age. The European Court held in its judgement that Germany had to comply with the EU principle of non-​discrimination on the grounds of age, even before the Employment Equality Framework Directive (where this principle was stated) had to be implemented at national level. In so doing, the Court recognized prohibition of discrimination on the grounds of age as a general principle of EU law, regardless of the directive. In 2005, an almost identical Honeywell case was referred to the FCC. A plaintiff had challenged the European Court’s ruling in Werner Mangold (2005: C-​144/​04), arguing that 235

Sabine Saurugger and Fabien Terpan

it had created a new principle (prohibition of discrimination on grounds of age) that was not in the Treaties. The community of legal scholars expected the FCC to accuse the ECJ of having exceeded its competence to interpret European law in Mangold (Herzog and Gerken 2008). Instead, the FCC argued that an ‘ultra vires review by the FCC (i.e. a review of the compatibility of EU acts with fundamental rights) can only be considered if a breach of competences on the part of European bodies is sufficiently qualified’, which was not the case in Honeywell. The German court used this case to clarify its position in the debate on ultra vires (beyond the limits of the treaties) vs. intra vires (within the limits of the treaties) actions of the European institutions. It accepted a presumption that the Union would generally act intra vires but added that it could consider ultra vires a judicial review in the case of a ‘manifest’ violation of the principle of conferral –​upon which the EU only exerts the competences that were explicitly conferred by the treaties. In 2020, the German FCC, returned forcefully to its pre-​EMU crisis attitude as a fierce opponent of the ECJ (Saurugger and Terpan 2020). In this vein, the German court issued a judgement on May 5, 2020 (Public Sector Purchase Programme [PSPP] judgement) challenging the substance of the European Central Bank’s (ECB) monetary policy, and more specifically the PSPP, in a context where a new financial programme (Pandemic Emergency Purchase Programme [PEPP]) had been launched by the ECB, in March 2020, to counter the financial and economic risks for the Eurozone area due to the COVID-​19 outbreak. The judgement was not as such directly targeted at the PEPP, which has been initiated by the ECB, but it was issued in a context where monetary policy was presented by the ECB as a crucial tool to recover from the crisis and therefore clashed with the announcement made by the president of the ECB on the use of monetary programmes to provide assistance to the member states. The ruling was also made public in a context where the German government hesitated to approve an emergency programme in response to the COVID-​19 crisis, and de facto, if not de jure, to create the Euro-​ bonds it so fiercely opposed during the EMU crisis. In its judgement, the FCC disagrees with the ECJ ruling in Weiss and others (2017: C-​493/​ 17) and finds its interpretation ‘arbitrary’. According to the FCC, the ECJ has exceeded its judicial mandate (Scicluna 2008) due to its failure to exert a proper and sound judicial review of the ECB’s policy. The combination of the ‘broad discretion’ afforded to the ECB and the ‘limited standard of review’ of the ECJ is seen as a violation of the principle of conferral by which ‘the Union shall act only within the limits of the competences conferred upon it by the member states in the Treaties’ (TEU 2012: Article 2). According to the FCC, the ECB did not sufficiently assess the proportionality of the measures taken, in particular with regards their effects on ‘public debt, personal savings, pension and retirement schemes, real estate prices and the keeping afloat of economically unviable companies’. Furthermore, the ECJ failed to correctly review the proportionality of these measures. It follows from this that all German public authorities, including the Bundesbank, ‘may participate neither in the development nor in the implementation, execution or operationalisation’ of the ECB’s PSPP. In addition, ‘the Federal Government and the Bundestag are required to take steps seeking to ensure that the European Central Bank conducts a proportionality assessment’. The judgement of the Bundesverfassungsgericht declares both an ECJ’s ruling and a decision of the ECB as ultra vires (going beyond their actual powers) and not applicable in Germany. Hence, constitutional pluralism does not ensure the stability of the EU judicial system. On the contrary, the difficult relationship between the ECJ and the German FCC shows how judicial collaboration sometimes gives way to major disagreement, potentially leading to judicial differentiation. While broadly speaking collaboration prevails, a major destabilization of the EU judicial system is always a possibility, especially in the case of other national courts following the path opened by the 236

European Court of Justice

FCC. For example, the Czech Constitutional Court’s (CCC), in 2009, declared part of the ECJ’s Marie Landtová ruling (2011: C-​399/​09) to be ultra vires (or ‘beyond the powers’ of the ECJ) under Czech constitutional law –​a first in the EU.The ruling was, it later transpired, an attempt to discipline the Supreme Administrative Court, which had referred the Landtová case to the ECJ in the first place (Bobek 2014).Yet, this has resulted in a situation of judicial differentiation, that could be less and less isolated if national constitutional courts increasingly disagree with the ECJ.

The ECJ challenged by national governments: pushing constitutional pluralism even further? This binary reading of, on the one hand, a uniform application of EU law defended by the ECJ vs. the constitutional pluralism defended by national constitutional courts, and more precisely by the FCC, however, is empirically misleading. Indeed, challenges to the uniform application of ECJ rulings do not only come from national courts but also from national governments. The latter have departed from the notion of a European rule of law homogenously applied on the EU territory, where ‘government officials and citizens are bound by and act consistent with the law’, and law is set forth in advance, made public, is general, clear, stable and certain, and applied to everyone according to its terms (Tamanaha 2004: 3). They have introduced specific types of differentiation –​vertical and horizontal –​with the aim to offer the member states’ way out of a negotiating deadlock, and/​or preserve member states sovereignty.

Vertical differentiation as levels of centralization Vertical differentiation refers to the EU using alternative modes of governance in areas where the so-​called Community method used to prevail. The Community method has a strong supranational component, including significant participation of the Commission and the Parliament in decision-​making, majority voting at the Council, as well as the jurisdiction of the ECJ. A typical decision made through the Community method would imply a proposal made by the Commission, upon which the Council and the Parliament would co-​decide (ordinary legislative procedure), implemented at EU level by the Commission and at domestic level by relevant national authorities, with possible judicial control by both the ECJ and domestic courts. The Community method is mostly based on hard law –​regulations, directives, decisions –​having primacy and direct effect, and on the idea of a homogenous application of law. Even directives –​a specific type of European law which set out the goals that member states must achieve but leave it to them to decide by which means these objectives should be reached –​might not characterize as proper differentiation, as it creates differences in the member states but within the framework of the European rule of law. Although applicable to a large number of policy areas, the Community method has been challenged by the introduction of new modes of governance, in particular since the 1990s. Most of them are less integrated, protective of national sovereignty, based on soft law and thus mostly isolated from the Court (Terpan 2015). The Common Foreign and Security Policy (CFSP), which was created by the treaty of Maastricht in 1992 as a continuation of the European Political Cooperation launched 20 years earlier, can be seen as one of the most intergovernmental policy area of the EU. According to TEU (2012) Article 19 and Treaty on the Functioning of the EU (TFEU) (2012) Article 275 is out of the jurisdiction of the ECJ, with two possible exceptions with regard to sanctions and to the delimitation between CFSP and non-​CFSP matters. Similarly, the so-​called open method of coordination in the field of employment and macro-​economic policy, as well as other ‘new’ modes of governance used in education, research, health or social 237

Sabine Saurugger and Fabien Terpan

policy, mostly rely on soft law instead of hard law, incentives instead of constraint. Through guidelines, orientations, recommendations, EU institutions are supposed to shape the content of member states’ policies without having recourse to judicial control. And within the framework of the Economic and Monetary Union, a number of hard obligations are controlled by the Council, on the basis of a Commission’s proposal, instead of being adjudicated by the ECJ. In all these situations, the use of less integrated methods can be seen as an attempt to circumvent the Court in areas where the member states fear sovereignty losses. Admittedly, some policy areas initially based on the new modes of governance such as higher education (Chou and Gornitzka 2014; Gornitzka 2009), are getting closer to the Community method, However, other studies show the overall increase in soft management mechanisms in the EU (Cappellina et al. 2020), which confirms the challenge posed to the Court.

Horizontal differentiation (territorial application of law) In federal systems like Canada and the United States, horizontal differentiation –​understood as an important room of manoeuver allowed to the member states –​is an central aspect of the political system. However, all states are given the same degree of autonomy. The constitution establishes the division between federal and state competences. States have their own constitution as well as their own government and legislature. But there is no opting out situation. Within the EU, a member state may voluntarily choose to opt out from a specific policy area during the negotiation of a revision treaty (Europe ‘à la carte’) or may not be allowed to participate in a specific policy (multispeed Europe). Thus, Denmark, Ireland and the United Kingdom before Brexit were not willing to be part of the Schengen area while Bulgaria, Croatia, Cyprus and Romania are still kept outside. Similarly, the Euro area is open to all member states provided they meet certain macro-​economic criteria and ask to become a member. In addition to this, the European treaties provide for different mechanisms of differentiated integration, including ‘strengthened cooperation’ and ‘permanent structured cooperation’ in the field of defence. Although rarely implemented, these mechanisms create legal differentiation between participating and non-​participating member states. In all these situations, member states remaining outside a policy area are not placed under the jurisdiction of the ECJ. Interestingly enough, the reach of European law extends to European non-​member states such as Norway, Liechtenstein and Iceland, which are members of both the EFTA (European Free Trade Association) and the European Economic Area. These third states voluntarily accept to comply with EU rules without having contributed to their adoption. This creates paradoxical situations where the ECJ may affect non-​member states (EFTA members of the Schengen area) while being in no position to control certain member states (Denmark and Ireland as non-​ participants in the Schengen area; member states remaining outside the Eurozone). In addition to this, Iceland, Liechtenstein and Norway are indirectly influenced by the jurisprudence of the ECJ due to the fact that the EFTA court tends to mirror the legal interpretations made in ECJ rulings. This explains why the UK post-​Brexit is not willing to be placed in the same situation as EFTA members of the Schengen area. Territorial differentiation is not only treaty-​based but may also result from a refusal to abide by the law. A Member State may challenge common rules by adopting and applying norms that contradict EU law, or by refusing to comply with common rules. Justice at EU level can even be ‘contained’ at member states’ level, when national administrations voluntarily ignore the implications of ECJ’s rulings (Conant 2001; see also Martinsen 2015). At first

238

European Court of Justice

sight, non-​compliance is different from differentiation. However, it can be argued that a case of enduring non-​compliance creates a de facto situation that is close to differentiation. De facto differentiation is not specific to the EU. As Kelemen explains, ‘literature on comparative politics and American political development shows that many democratic federations have, de facto if not necessarily de jure, permitted differentiation in the rule of law’ (Kelemen 2019: 250). However, the rise of ‘illiberal democracies’ in Poland and Hungary (Bugarčić 2014), and to some extent in Bulgaria and Romania, poses a threat to the rule of law and more generally to the whole European legal system (Bogdandy and Ionnidis 2014). If de facto differentiation involves explicitly ‘permitting member governments to depart from the EU’s core rule of law principles, such as guaranteeing the independence of the judiciary, equality before the law and respect for fundamental rights’ (Kelemen 2019: 248), then the main tenets of European legal integration would be at risk (see also Chapter 5 in this volume).

Is the ECJ and the EU’s legal system adaptive enough to protect the rule of law against differentiation? In previous sections, we showed how legal integration and the uniform application of law in the EU are questioned by both national courts and national governments. Confronted with these dynamics, the Court has developed instruments to limit legal and judicial differentiation through its activism in fields where its competences are limited at best and through the establishment of the sociological instrument of judicial dialogue. In some rulings, however, in particular those dealing with the rule of law in Central and Eastern Europe, the ECJ J itself has contributed to a differentiated understanding of law abiding. This flexibility can be seen as a nuance to the main function performed by the Court, i.e. ensuring the uniform application of EU law in the EU.

The position of the ECJ: activism or acceptance? The Court limits the impact of differentiation when it seeks to get a grip on policy area that is not placed under its jurisdiction, for instance in the field of CFSP, and it does it through an extensive interpretation of its own competence. Apart from the two treaty-​based exceptions regarding to sanctions and competence delimitation (see above), we know that the ECJ has no jurisdiction over CFSP. However, the Court has interpreted this limitation restrictively in a series of cases. In particular, two rulings known as the ‘pirate’ cases (European Parliament v Council of the EU 2014: C-​658/​11 Mauritius; C-​263/​14 Tanzania) have established the right for the Court to control the way CFSP external agreements are adopted, including the information of the European Parliament (Saurugger and Terpan 2017, Terpan 2018). The Court has even gone as far as curbing national sovereignties in the field of security and defence. Despite the progressive establishment of the CSDP (Common Security and Defence Policy 2009) since the European Council of Cologne in 1999, and the inclusion of this policy in primary law through the Lisbon Treaty in 2009, the defence sector remains largely under member states’ control. Defence policy is a specific competence, based on the legal framework of the CFSP and kept apart from the more integrated internal market policies. While the ECJ is not competent with regard to CSDP, it does however have jurisdiction over procurement policy, and it is here where it can actively intervene in order to influence the uniform application of EU law in security/​defence fields. It does act against the inherent differentiation of the EU’s legal framework based on the capacity of member states to invoke Article 346 TFEU which allows

239

Sabine Saurugger and Fabien Terpan

the member states to escape EU procurement rules in referring to ‘essential security interests’. For decades, Article 346 TFEU has been interpreted as excluding the whole defence sector from the remit of EU law. The ECJ ruled on defence procurement in the cases of Commission vs. Italy of April 8, 2008 (C-​337/​05) and InsTiimi Oy of June 7, 2012 (C-​615/​10) (Terpan and Saurugger 2019). These rulings concerned two main issues: the scope of essential security interests and the specific products that are covered by the exception; the definition of measures that a member state ‘considers necessary for the protection of the essential interests of its security’ (TFEU 2016: Article 346). In both rulings Commission vs. Italy (2008) and InsTiimi Oy (2012), however, the Court has limited member states’ capacity to refer to Article 346 TFEU, which aims to protect the member states’ essential security interests in order to avoid the EU’s public procurement rules, and has thus decreased Member State’s capacity for differentiation, without actually ending it.The Court’s case law on this matter has established that this is actually a case-​by-​case derogation that is to be applied strictly in exceptional circumstances. While past discussions on Article 346 TFEU have often been focused on what kind of military products are covered, more recent case law shows that the key conditions for the application of this Treaty provision are necessity and proportionality. It is for member states to prove that the measures they take are necessary in order to protect their essential security interests, and that such an objective cannot be achieved through less restrictive means. Hence, differentiation in this field is still possible but has been embedded in a stricter definition of what ‘national security interests’ entail. The Court’s interpretations can indeed be considered activist, even if a careful analysis shows that this activism was combined with some element of restraint. These two examples are illustrations of cases, which triggered increased criticism as to the Court’s constitutionalization activities. The Court’s activism against legal differentiation is criticized by some commentators as over-​constitutionalization of the European political system (Grimm 2015; Schmidt 2018). However, the Court’s activism to decrease the system’s degree of differentiation cannot only be found in its rulings, of which the two above are only topical examples. It has also engaged activities to judge lower courts into an attitude that decreases the degree of judicial differentiation and creates a common level playing field.

Networking strategies We have seen from the above that the interpretation of the primacy of EU law –​a major element against future legal differentiation –​has not been met with a broad consensus, particularly not amongst domestic legal scholars (Alter 2010). European law and the ECJ’s doctrine tended, on the contrary, to be frequently ignored in the 1960s and 1970s. Those national judges who were not socialized to EU law were reluctant to refer cases to the European Court. Moreover, national high court rulings often contradicted the Court’s doctrine (Saurugger and Terpan 2017). An explanation for the incremental acceptance of ECJ’s rulings, and hence the decrease of judicial/​legal differentiation, in particular among higher courts, can be found in network approaches. Aware of potential conflicts, constitutional courts and the ECJ have established a series of exchanges or dialogue of judges (Burley and Mattli 1993; Dehousse 1998; Alter 2010; Stirn 2009). While visiting programmes for national judges have existed at the European Court since it was set up, the beginning of the 2000s saw an increasing number of formal meetings as well as a new practice of informal meetings and workshops (Leron 2014). The main function of these meetings is to anticipate and prevent potential conflicts. By this informal judicial dialogue, the ECJ can influence the articulation of EU and national legal orders. If there is a conflict, it must be kept in this extralegal sphere. 240

European Court of Justice

National judges have underlined the usefulness of this cooperation and even asked for it. One of the French members of the Conseil d’Etat, Jean-​Marc Sauvé (2013), argues that besides institutional or vertical judicial cooperation between supreme jurisdictions (through preliminary questions before the ECJ for instance), legal professionals should also put in place an informal, horizontal cooperation between administrative supreme judges of the EU. In 2000, a pan-​European Association of State Councils and Supreme Administrative Jurisdictions of the European Union was established by a group of judges, with the aim of fostering exchange programmes, to consider different public law approaches and see if they could be reconciled, to allow a convergence towards common practices. It has since fostered bilateral working meetings. New Member State judges have been visited older member states’ jurisdictions. Since 2008, this judge exchange programme has become more systematic. The association participates in the European Judicial Training Network (EJTN), created in 2000 with the financial support from the Justice Programme of the EU and conceived to be ‘the principal platform and promoter for the training and exchange of knowledge of the European judiciary’.1 Member states’ supreme courts have acted in a similar way and gathered in the Network of the Presidents of the Supreme Judicial Courts of the European Union. In April 2007, the network launched a mega-​search engine on national case law2 that allows for national case law search engines to be cross-​searched. It is a powerful tool for applied comparative law, and part of the welcome new trend of judicial collaboration in the EU (Leron 2014; Hübner 2016).

Increasing resistance, increasing differentiation? Despite these judicial and social strategies to decrease the uneven compliance with Court decisions, we observe increasing resistance from several EU governments, in particular the self-​proclaimed Hungarian and Polish illiberal democracies. Here, Kelemen talks about the ‘predictable abuse of constitutional pluralism by aspiring autocrats’. While constitutional pluralism might appear stable in a positive context where national courts are willing to cooperate, it ‘reveals its weaknesses when the political environment becomes inclement’, which is the case when the FCC issues challenging judgements, and even more when national judiciaries are ‘captured by autocratic governments’ (Kelemen 2019: 254; see also Blauberger and Kelemen 2017). This difficulty can be observed in the ECJ’s Celmer ruling (2018: C-​216/​18 PPU Reference for a Preliminary Ruling from the Thigh Court (Ireland) Made on 27 March 2018 Minister for Justice and Equality v LM). The ruling, made on a preliminary reference from the Irish High Court, concerned whether a crime suspect should be surrendered from Ireland to Poland in a context where serious doubts could be raised on the independence of Polish courts, and subsequently on whether the suspect could receive a fair trial in Poland. Indirectly, the case finds its origin in the decision of the Polish government controlled by the Law and Justice party (PiS), shortly after its 2015 election, to seize control of the Polish supreme court and to ‘pack it’ with PiS affiliated judges. The aim in further moves was to assert tight party-​political control over the selection and dismissal of judges and the management of the judiciary. As a result, the European Commission triggered in December 2017 the Article 7 TEU mechanism in response to the Polish governments’ attempt to question the rule of law and the independence of the judiciary. In July 2018, the Commission brought an infringement action specifically challenging the legality under EU law of the Polish governments judicial reform laws. In the Celmer case, these developments had a specific impact on the European arrest warrant (EAW), and more precisely on the execution of the EAW when it could lead to a breach of the rule of law. 241

Sabine Saurugger and Fabien Terpan

The EAW system is based on mutual trust between the member states, meaning that judges in one Member State are expected to surrender suspects when an EAW has been issued by judges in another Member State. In the Case of Minister for Justice and Equality v Celmer, the surrendering of the crime suspect to Polish authorities raised serious issues having regard to the situation in Poland, under the PiS government. The ECJ ruled that national courts had to consider the lack of actual independence of the judiciary in order to prevent a situation where specific individuals could suffer the risk of not receiving a fair trial (Kelemen 2019: 257). This can be seen as a self-​introduced differentiation of the rule of law in the EU by the ECJ, the latter opening the possibility for national judges to adjust the execution of EAWs to the situation of the rule of law in the issuing State. It would be too simple, however, to conclude that the Court itself has become a defender of differentiation and constitutional pluralism. While the Council has remained quite inactive towards the rise of illiberalism in Europe (Oliver and Stefanelli 2016), the Commission has brought several cases before the ECJ (Müller 2015; Mager 2016; Kochenov and Pech 2015, 2016). In a series of rulings related to the rule of law in Poland and Hungary, the ECJ did not pass up the chance to ensure compliance with EU law against national laws threatening the independence of the judiciaries (see, for example, cases related to the Polish and Hungarian laws on the retirement age of judges (Commission vs. Poland 2018: C-​619/​18; C-​192/​18; C-​66/​18; C-​286/​18)) or, more generally, against the failure to observe obligations arising under EU law (Commission vs. Poland 2021: C-​204/​21 R). Moreover, when confronted with a challenging judgement by a Member State constitutional court, such as the German FCC’s PPSP judgement of May 5, 2020, the ECJ reacted harshly in producing a press release –​a procedure that the Court usually does not use. The ECJ reminded the German court, and indirectly all national constitutional courts, that in order to ensure that EU law is applied uniformly, the Court of Justice alone–​which was created for that purpose by the Member States –​has jurisdiction to rule that an act of an EU institution is contrary to EU law. Divergences between courts of the Member States as to the validity of such acts would indeed be liable to place in jeopardy the unity of the EU legal order and to detract from legal certainty.3 In this respect, the Court can indeed be seen as a ‘supporter of “more Europe”, whatever it takes’ (Granger 2015: 209), both asserting judicial control, but displaying flexibility at the same time.

Conclusion The ECJ is facing two main sources of differentiation coming from both national courts and national governments. While the ECJ has strongly constitutionalized European law through its doctrines of direct effect and primacy, national courts have not accepted this evolution so easily, creating a situation of constitutional pluralism. To some extent, they still resist the ECJ’s attempt to ensure a uniform application of EU law. The Judgement of May 5, 2020, whereby the German FCC challenged the jurisprudence of the ECJ on the European Central Bank monetary programs, is emblematic of this continued resistance. At the same time, member states’ governments have introduced different forms of vertical and horizontal differentiation, challenging the ECJ’s role as the cornerstone of European integration. The introduction of new modes of governance has led to vertical differentiation, with the ECJ being kept aside of a number of ‘new’ policy areas. Horizontal differentiation results from the fact that all member states do not 242

European Court of Justice

apply EU law in its entirety. Legal exceptions allow them to escape the reach of EU law, or to benefit from opt-​outs, while third states such as Norway, Iceland and Lichtenstein have chosen to comply with a large number of EU rules, accepting an indirect influence of the ECJ. The most recent challenge to the uniform application of EU law is posed by illiberal democracies in Poland and Hungary. In these member states, the independence of the judiciaries as well as freedom of the press and fundamental rights is threatened. While in the Celmer ruling on the execution of EAWs the Court tends to acknowledge the existence of differentiation, in other cases, it has reacted to these developments though a number of rulings against the Polish and Hungarian governments, showing its continuing commitment towards the unity of European law. As any other supreme and constitutional court, the ECJ is supposed to decrease judicial differentiation in the EU political system. In practice, the Court provides audacious legal interpretations, sometimes depicted as judicial activism, in order to overstep its own limitations as seen in the cases related to CFSP. In addition, the Court engages in a social framing strategy, which through collective interaction with the national legal and judicial communities can lead to a more uniform acceptance of EU law and the Court’s rulings, but also does ‘whatever it takes’ (Granger 2015) to foster European integration, deferring to intergovernmental solutions, instead of supranational, when integration is at stake. We argue that a more sociological approach to supreme courts, including the ECJ, allows for understanding that their role in political systems is not only defined by law but is also one of agency (Terpan and Saurugger 2019). More precisely, the judges themselves frame a collective understanding of the Court and its position in a system increasingly described as differentiated integration or dis-​integration. It is here that we would need without any doubt a better understanding of the sociological and political mechanisms that, combined with a legal interpretation of the law, influence the Court’s approach to differentiation of the EU’s legal system.

Notes 1 www.refj.eu/​en/​About/​About-​EJTN/​ 2 www.reseau-​presidents.eu/​r pcsjue/​ 3 https://​curia.europa.eu/​jcms/​upload/​docs/​application/​pdf/​2020-​05/​cp200058en.pdf

References Alter, K. (2010). The European Court’s political power: selected essays. New York, NY: Oxford University Press. Avbelj, M., & Komárek, J. (Eds.) (2012). Constitutional pluralism in the European Union and beyond. Oxford and Portland, Oregon: Bloomsbury Publishing. Beck, G. (2013). The legal reasoning of the Court of Justice of the EU. Oxford and Portland, Oregon: Bloomsbury Publishing. Blauberger, M., & Kelemen, R.D. (2017). Can courts rescue national democracy? Judicial safeguards against democratic backsliding in the EU. Journal of European Public Policy, 24(3), pp. 321–​336. doi:10.1080/​ 13501763.2016.1229357. Bobek, M. (2014). Landtová, Holubec, and the problem of an uncooperative court: implications for the preliminary rulings procedure. European Constitutional Law Review, 10(1), pp. 54–​ 89. doi:10.1017/​ S1574019614001047. Bogdandy, A., & Ionnidis, M. (2014). Systemic deficiency in the rule of law: what it is, what has been done, what can be done. Common Market Law Review, 51(1), pp. 59–​96. Available at: https://​kluw​erla​wonl​ine. com/​api/​Prod​uct/​Cit​atio​nPDF​URL?file=​Journ​als\COLA\COLA​2014​003.pdf&cas​a_​to​ken=​niH4C​ U9fI​IcAA​AAA:DvPFQDPqeEcU_​RURMumYi6ZgxzLhmlU1XiKUibuRMTvb​EX9v​K2Pa​pw2l​ Y71s​QxMn​eWiJ​z4V-​w (Accessed: 20 April 2021). 243

Sabine Saurugger and Fabien Terpan

Bugarčić, B. (2014). Protecting democracy and the rule of law in the European Union: the Hungarian Challenge. London School of Economics (LSQS) Paper No. 79. Available at: www.semanticscholar.org/​ paper/​Protecting-​Democracy-​and-​the-​Rule-​of-​Law-​in-​the-​The-​Bugari%C4%8D/​74a0dbb67bd485c 652433c49400b767e6742eb4b?p2df (Accessed: 20 April 2021). Burley, A.M., & Mattli,W. (1993). Europe before the court: a political theory of legal integration. International Organization, 47(1), pp. 41–​ 76. Available at: www.jstor.org/​stable/​pdf/​2706882.pdf (Accessed: 20 April 2021). BVerfG (1993). 2 BvR 2134/​92, 2 BvR 2159/​92. eui.eu. Available at: https://​iow.eui.eu/​wp-​cont​ent/​uplo​ ads/​sites/​18/​2013/​04/​06-​Von-​Bogda​ndy-​Ger​man-​Fede​ral-​Con​stit​utio​nal-​Court.pdf (Accessed: 26 April 2021). BVerfG (2010). 2BvR2661/​ 06. bundesverfassungsgericht.de. Available at: www.bundesverfassungsgericht. de/​SharedDocs/​Entscheidungen/​EN/​2010/​07/​rs20100706_​2bvr266106en.html (Accessed: 26 April 2021). BVerfGE (1974). Cases 37, 271 2 BvL 52/​71 Solange II. utexas.edu. Available at: Beschluß utexas.edu. https://​law.ute​xas.edu/​transn​atio​nal/​fore​ign-​law-​trans​lati​ons/​ger​man/​case.php?id=​588 (Accessed: 26 April 2021). BVerfGE (1986). Case 73, 339 2 BvR 197/​83 Solange II-​decision. University of Texas School of Law. utexas.edu. https://​law.ute​xas.edu/​transn​atio​nal/​fore​ign-​law-​trans​lati​ons/​ger​man/​case.php?id=​572 (Accessed: 26 April 2021). BVerfGE (2009). 123,267 –​Lisbon Decision (Lissabon-​Urteil) –​act approving the treaty of Lisbon compatible with the Basic Law; accompanying law unconstitutional to the extent that legislative bodies have not been accorded sufficient rights of participation. Germanlawarchive.iuscom. Available at: https://​germa​ nlaw​arch​ive.iusc​omp.org/​?p=​11 (Accessed: 26 April 2021). Capelletti, M., Seccombe, M., & Weiler, J.H. (1986). Integration through law, Europe and the American Federal Experience. In Methods, tools and institutions (Vol. 1). Berlin, New York, NY: Walter de Gruyter and Co. doi.org/​10.1515/​9783110921540.3. Cappellina, B. (2020). EfSoLaw: a new data set on the evolution of soft law in the European Union. In ECPR virtual general conference 2020. Available at: https://​hal.archi​ves-​ouver​tes.fr/​hal-​03117​788/​docum​ ent (Accessed: 20 April 2021). Chou, M.H., & Gornitzka, Å. (Eds.) (2014). Building the knowledge economy in Europe: new constellations in European research and higher education governance. Cheltenham: Edward Elgar Publishing. Common Security and Defence Policy (2009). Fact sheets on the European Union. europarl.europa. Available at: www.europarl.europa.eu/​factsheets/​en/​sheet/​159/​common-​security-​and-​defence-​policy (Accessed: 26 April 2021). Conant, L. (2001). Justice contained: law and politics in the European Union. Ithaca, NY: Cornell University Press. Constitution of the Federal Republic of Germany (1949). Available at: www.bmi.bund.de/​EN/​topics/​ constitution/​constitutional-​issues/​constitutional-​issues.html#:~:text=​The%20current%20version%20 of%20the,the%20Federal%20Republic%20of%20Germany.&text=​ W hen%20reunification%20 took%20effect%20on,over%2060%20amendments%20since%201949 (Accessed: 26 April 2021). Craig, P. (1997). Formal and substantive conceptions of the rule of law: an analytical framework. Public Law, 467, 467–​487. Available at: www.taylorfrancis.com/​chapters/​edit/​10.4324/​9781315085302-​4/​formal-​ substantive-​conceptions-​rule-​law-​analytical-​framework-​paul-​craig (Accessed: 20 April 2021). Davies, B. (2012). Resisting the European Court of Justice:West Germany’s confrontation with European Law, 1949–​ 1979. Cambridge: Cambridge University Press. De Witte, B. (2019).The Law as Tool and Constraint of Differentiated Integration. Robert Schuman Centre for Advanced Studies Research Paper No. RSCAS, 47. doi:10.2139/​ssrn.3425914. Dehousse, R. (1998). The European Court of Justice: the politics of judicial integration (p. 5). London: Macmillan. European Court of Justice, Commission of the European Communities v Italian Republic (1972). europa. eu. https://​eur-​lex.eur​opa.eu/​legal-​cont​ent/​EN/​TXT/​?uri=​CELEX%3A6197​1CJ0​048 (Accessed: 26 April 2021). European Court of Justice, Commission vs. Italian Republic (2008). Case C-​337/​05. europa.eu. Available at: https://​eur-​lex.eur​opa.eu/​legal-​cont​ent/​EN/​ALL/​?uri=​CELEX%3A6200​5CJ0​337 (Accessed: 20 April 2021). European Court of Justice, Commission vs. Republic of Poland (2018). Case C-​619/​18. europa.eu. Available at: https://​curia.eur​opa.eu/​juris/​docume​nts.jsf?num=​c-​619/​18 (Accessed: 26 April 2021). 244

European Court of Justice

European Court of Justice, Commission vs. Poland (2021). Case C-​204/​21 R. europa.eu. Available at https://​curia.eur​opa.eu/​juris/​docum​ent/​docum​ent.jsf?text=​&docid=​248​261&pageIn​dex=​0&docl​ ang=​fr&mode=​req&dir=​&occ=​first&part=​1&cid=​93068 (Accessed: 20 December 2021) European Court of Justice, European Parliament v Council of the European Union (2014). Action for annulment—​decision 2011/​640/​CFSP—​legal basis—​common foreign and security policy (CFSP)—​ Article 37 TEU—​international agreement relating exclusively to the CFSP—​second subparagraph of Article 218(6) TFEU—​obligation to inform the Parliament immediately and fully—​Article 218(10) TFEU—​maintenance of effects. Case C-​658/​11. europa.eu. Available at: https://​eur-​lex.eur​opa.eu/​legal-​ cont​ent/​EN/​TXT/​?uri=​CELEX%3A6201​1CJ0​658&qid=​140372​9915​929 (Accessed: 26 April 2021). European Court of Justice, Flaminio Costa vs. E.N.E.L. (1964). Judgement of the court –​reference for a preliminary ruling: giudice conciliatore di Milano –​Italy. –​Case 6/​64. europa.eu. Available at https://​ eur-​lex.eur​opa.eu/​legal-​cont​ent/​EN/​TXT/​?uri=​CELEX%3A6196​4CJ0​006 (Accessed: 26 April 2021). European Court of Justice, InsTiimi Oy. (2012). Judgement of the court -​reference for a preliminary ruling from the Korkein hallinto-​oikeus -​Case C‑615/​10. europa.eu. Available at https://​curia.eur​opa.eu/​ juris/​liste.jsf?num=​C-​615/​10&langu​age=​FR (Accessed: 26 April 2021). European Court of Justice, Internationale Handelsgesellschaft mbH v Einfuhr-​und Vorratsstelle für Getreide und Futtermittel (I970). Judgement of the court-​. Reference for a preliminary ruling:Verwaltungsgericht Frankfurt am Main –​Germany. –​Case 11–​70. europa.eu. Available at: https://​eur-​lex.eur​opa.eu/​legal-​ cont​ent/​EN/​TXT/​?uri=​CELEX%3A6197​0CJ0​011 (Accessed: 26 April 2021). European Court of Justice, ‘Les Verts’ v European Parliament (1986). Action for annulment –​information campaign for the elections to the European Parliament. –​Case 294/​83. europa.eu. Available at: https://​ eur-​lex.eur​opa.eu/​legal-​cont​ent/​EN/​TXT/​?uri=​CELEX%3A6198​3CJ0​294 (Accessed: 26 April 2021). European Court of Justice, Marie Landtová v Česká správa socialního zabezpečení (2011). Judgement of the court –​reference for a preliminary ruling: Nejvyšší správní soud –​Czech Republic. Case 399/​09. europa.eu. Available at: https://​eur-​lex.eur​opa.eu/​legal-​cont​ent/​EN/​ALL/​?uri=​CELEX%3A6200​9CJ0​ 399 (Accessed: 26 April 2021). European Court of Justice, Minister for Justice and Equality v Celmer (2018). Case 216/​18 PPU –​europa.eu. Available at: https://​curia.eur​opa.eu/​juris/​liste.jsf?langu​age=​en&td=​ALL&num=​C-​216/​18%20PPU (Accessed: 26 April 2021). European Court of Justice, van Gend & Loos v Netherlands Inland Revenue Administration (1963). Reference for a preliminary ruling: Tariefcommissie –​Pays-​Bas. –​Case 26–​62. europa.eu. Available at https://​eur-​lex.eur​opa.eu/​legal-​cont​ent/​EN/​TXT/​?uri=​CELEX%3A6196​4CJ0​006 (Accessed: 26 April 2021). European Court of Justice, Weiss and Others (2017). Case C-​493/​17. europeansources. Available at: www. europeansources.info/​record/​case-​c-​493-​17-​heinrich-​weiss-​and-​others/​ (Accessed: 22 April 2021). European Court of Justice, Werner Mangold v Rüdiger Helm (2005). Judgement of the court –​reference for a preliminary ruling: Arbeitsgericht München –​Germany. europa.eu. Available at: https://​eur-​lex.eur​ opa.eu/​legal-​cont​ent/​EN/​TXT/​?uri=​CELEX%3A6200​4CJ0​144 (Accessed: 22 April 2021). European Parliament (2015). Trends in differentiation of EU law and lessons for the future. europarl. europa. Available at: www.europarl.europa.eu/​RegData/​etudes/​IDAN/​2015/​510007/​IPOL_​ IDA(2015)510007_​EN.pdf (Accessed: 24 March 2021). Employment Promotion Act (1996). Europa.eu. Available at: https://​ec.eur​opa.eu/​employ​ment​_​soc​ial/​soc-​ prot/​misso​c98/​engl​ish/​socp​rot/​d.htm (Accessed: 26 April 2021). Fossum, J.E. (2019). Europe’s triangular challenge: differentiation, dominance and democracy. EU3D research paper, (1). Available at: www.europarl.europa.eu/​thinktank/​en/​document.html?reference=​IPOL_​ IDA%282015%29510007, www.researchgate.net/​profile/​John-​ (Accessed: 20 April 2021). Gornitzka, Å. (2009). Networking administration in areas of national sensitivity: the commission and European higher education. In European integration and the governance of higher education and research (pp. 109–​131). Springer: Dordrecht. doi:10.1007/​978-​1-​4020-​9505-​4_​5. Granger, M.P. (2015). The court of justice’s dilemma—​between ‘more Europe’ and ‘constitutional mediation’. In Bickerton, C., Hodson, D., & Puetter, U. (eds), The new intergovernmentalism (pp. 209–​226). Oxford: Oxford University Press. Grimm, D. (2009). Defending sovereign statehood against transforming the union into a state. EuConst, 5, pp. 353–​373. doi:10.1017/​S1574019609003538. Grimm, D. (2015). Sovereignty: the origin and future of a political and legal concept. New York, NY: Columbia University Press. Hartley, J. (1996). Popular reality: journalism, modernity, popular culture (p. 32). London: Arnold. 245

Sabine Saurugger and Fabien Terpan

Herzog, R., & Gerken, L. (2008). Stop the European Court of Justice. EU Observer, September 10. Available at: https://​euo​bser​ver.com/​opin​ion/​26714. Hübner, D.C. (2016). The ‘National Decisions’ database (Dec. Nat): introducing a database on national courts’ interactions with European Law. European Union Politics, 17(2), pp. 324–​339. doi:10.1177/​ 1465116515613453. Jaklic, K. (2014). Constitutional pluralism in the EU. Oxford: Oxford University Press. Kelemen, R.D. (2019). Is differentiation possible in rule of law? Comparative European Politics, 17(2), pp. 246–​260. doi:10.1057/​s41295-​019-​00162-​9. Kochenov, D., & Pech, L. (2015). Monitoring and enforcement of the rule of law in the EU: rhetoric and reality. European Constitutional Law Review, 11(03), pp. 512–​540. doi:10.1017/​S1574019615000358. Kochenov, D., & Pech, L. (2016). Better late than never? On the European Commission’s rule of law framework and its first activation. Journal of Common Market Studies, 54(5), pp. 1062–​1074. doi:10.1111/​ jcms.12401. Leron, N. (2014). La gouvernance constitutionnelle des juges: l’institutionnalisation d’un nouveau mode de régulation du risque de conflit constitutionnel dans l’Union européenne, Doctoral dissertation. Paris, Institut d’études politiques. Leruth, B., Gänzle, S., & Trondal, J. (2019). Exploring differentiated disintegration in a post-​Brexit European Union. Journal of Common Market Studies, 57(5), pp. 1012–​1029. doi:10.1111/​jcms.12869. Leuffen, D., Rittberger, B., & Schimmelfennig, F. (2012). Differentiated integration: explaining variation in the European Union. Basingstoke, New York: Macmillan International Higher Education. Mager, A.M. (2016). Cracks in the foundations: understanding the great rule of law debate in the EU. Journal of Common Market Studies, 54(5), pp. 1050–​1061. doi:10.1111/​jcms.1240. Mak, E., & Taekema, S.H. (2016).The European Union’s rule of law agenda: identifying its core and contextualizing its application. Hague Journal on the Rule of Law, 8(1), pp. 25–​50. doi:10.1007/​s40803-​016-​0022-​1. Martinsen, D.S. (2015). An ever more powerful court?: the political constraints of legal integration in the European Union. Oxford: University Press. May, C. (2014). The rule of law: the common sense of global politics. Cheltenham: Edward Elgar Publishing Ltd. Müller, J.W. (2015). Should the EU protect democracy and the rule of law inside member states?. European Law Journal, 21(2), pp. 141–​160. doi:10.1111/​eulj.12124. Oliver, P., & Stefanelli, J. (2016). Strengthening the rule of law in the EU: the council’s inaction. Journal of Common Market Studies, 54(5), pp.1075–​1084. doi:10.1111/​jcms.12402. Payandeh, M. (2011). Constitutional review of EU law after Honeywell: contextualizing the relationship between the German Constitutional Court and the EU Court of Justice. Common Market Law Review, 48(1), pp. pp. 9–​38. Rosenfeld, M. (2006). Comparing constitutional review by the European Court of Justice and the U.S. Supreme Court. International Journal of Constitutional Law, 4(4), pp. 618–​651. doi:10.1093/​icon/​mol027. Saurugger, S., & Terpan, F. (2017). The Court of Justice of the European Union and the politics of law. Basingstoke: Palgrave Macmillan. Saurugger, S., & Terpan, F. (2020). Integration through (case) law in the context of the Euro area and Covid-​19 crises: courts and monetary answers to crises. Journal of European Integration, 42(8), 1161–​1176. doi:10.1080/​07036337.2020.1852233. Sauvé, J.M. (2013). Le rôle du Panel 255 dans la sélection du juge de l’Union. In Rosas, A., Levits, E., & Bot, Y. (eds), The Court of Justice and the construction of Europe: analyses and perspectives on sixty years of case-​law (pp. 99–​119). The Hague: TMC Asser Press. Schmidt, S.K. (2018). The European Court of Justice and the policy process: the shadow of case law. Oxford: Oxford University Press. Scicluna, N. (2018). Integration through the disintegration of law? The ECB and EU constitutionalism in the crisis. Journal of European Public Policy, 25(12), pp. 1874–​1891. doi:10.1080/​13501763.2017.1362026. Selznick, P. (2003). Law in context’ revisited. Journal of Law and Society, 30(2), pp. 30–​52. doi:10.1111/​ 1467-​6478.00252. Shklar, J.N. (1987). Political theory and the rule of law. In Hutchinson, A., & Monahan, P.J. (eds), The rule of law: ideal or ideology (pp. 1–​16).Toronto: Carswell. Available at: https://​phi​lpap​ers.org/​rec/​HUT​TRO-​14 (Accessed: 22 April 2021). Stein, E. (1981). Lawyers, judges, and the making of a transnational constitution. American Journal of International Law, 75(1), pp. 1–​27. Available at: www.pravo.unizg.hr/​_​download/​repository/​Eric_​Stein. pdf (Accessed: 22 April 2021).

246

European Court of Justice

Stein, E. (2001). International integration and democracy: no love at first sight. American Journal of International Law, pp. 489–​534. Available at: www.jstor.org/​stable/​pdf/​2668491.pdf (Accessed: 22 April 2021). Stirn, B. (2009). Le Conseil d’Etat et les juridictions communautaires: un demi-​siècle de dialogue des juges. La Gazette du Palais, 13(14), pp. 3–​7. Available at: https://​old.jurisd​octo​r ia.net/​pdf/​nume​ro6/​aut6_​L​ EPOU​TRE.pdf (Accessed: 22 April 2021). Stone Sweet, A. (2004). The judicial construction of Europe. New York, NY: Oxford University Press. Sunstein, C.R. (2001). Designing democracy, what constitutions do. Oxford, New York: Oxford University Press. Tamanaha, B.Z. (2004). On the rule of law, history, politics, theory. Cambridge: Cambridge University Press. Terpan, F. (2015). Soft law in the European Union –​the changing nature of EU law. European Law Journal, 21(1), pp. 68–​96. doi:10.1111/​eulj.12090. Terpan, F. (2018). The CJEU and the external powers of parliament, new activism or self restraint? In Vara, J.S., & Sánchez-​Tabernero, S.R. (eds), The democratisation of EU international relations through EU law (pp. 39–​59). Abingdon: Routledge. Terpan, F., & Saurugger, S. (2019). Assessing judicial activism of the CJEU the case of the court’s defence procurement rulings. Journal of European Integration, 41(4), pp. 543–​ 561. doi:10.1080/​ 07036337.2018.1537268. Thym, D. (2009). In the name of sovereign statehood: a critical introduction to the Lisbon judgment of the German Constitutional Court. Common Market Law Review, 46(6), pp.1795–​1822. Treaty of Function of European Union (TFEU) (2012). Consolidated version of the treaty of function of European Union. Official Journal of the European Union, europa.eu. Available at: https://​eur-​lex.eur​opa. eu/​Lex​UriS​erv/​Lex​UriS​erv.do?uri=​CELEX:120​12E/​TXT:en:PDF (Accessed: 26 April 2021). Treaty of the European Union (TEU) (2012). Consolidated version of the treaty on European Union Official Journal of the European Union, p.C326/​ 13. Available at: https://​eur-​lex.eur​opa.eu/​resou​ rce.html?uri=​ c el ​ l ar:2bf14 ​ 0 bf- ​ a 3f8- ​ 4 ab2- ​ b 506fd71826e6da6.0023.02/ ​ D OC_ ​ 1 &format= ​ P DF (Accessed: 26 April 2021). Treaty on the Functioning of the European Union (TFEU) (2016). Consolidated version of the treaty on the functioning of the European Union PART SEVEN –​GENERAL AND FINAL PROVISIONS, Article 346. Journal of the European Union, europa.eu. Available at: https://​eur-​lex.eur​opa.eu/​legal-​cont​ ent/​EN/​TXT/​?uri=​CELEX%3A12​016E​335 (Accessed: 13 April 2021). Walker, N. (2002). The idea of constitutional pluralism. The Modern Law Review, 65(3), pp. 317–​359. doi:10.1111/​1468-​2230.0038. Weiler, J.H. (1991). The transformation of Europe. Yale Law Journal, pp. 2405–​2483. Available at: www.jstor. org/​stable/​pdf/​796898.pdf (Accessed: 22 April 2021).

247

16 Third-​country participation in EU agencies Towards “condominio”?1 Sandra Lavenex

Introduction This chapter complements intergovernmental and supranational perspectives on differentiated integration by investigating the transgovernmental dynamics and shape of differentiated integration in European Union (EU) decentralised agencies. These secondary institutions in the EU system occupy a particular space that is relatively shielded from both the dividing dynamics of intergovernmental politics and from the unifying ambit of the supranational polity. As will be developed further below, the sector-​specific scope of EU decentralised agencies, their specialisation in specific regulatory tasks and their technocratic composition as hubs of national regulators have turned decentralised agencies into privileged fora for external differentiated integration vis-​à-​vis associated countries and beyond. The result is a varied polycentric patchwork of memberships and associations in EU secondary institutions with frequent ramifications to overarching international regimes. In sum, this transgovernmental web of differentiated integration echoes Philippe Schmitter’s vision of a “condominio” in which “instead of a Eurocracy accumulating organizationally distinct but politically coordinated tasks around a single centre, there could be multiple regional institutions acting autonomously to solve common problems and produce different public goods” (Schmitter 1996: 136). This chapter conceptualises and illustrates the transgovernmental layer of (external) differentiated integration in the multilayered EU polity. It first introduces the distinct modes of governance applying to transgovernmental bodies and EU agencies more specifically and identifies the formal and informal institutional features that make these transgovernmental bodies particularly permeable for flexible forms of differentiated integration. Section three on third-​ country participation in EU agencies documents this flexibility zooming into concrete examples of differentiated integration. It will be shown that third-​country participation in EU agencies has mostly developed incrementally and on a case-​by-​case basis driven by the mutual interest and interdependence between peer regulators, and that supranational actor’s efforts to guide and harmonise decentralised agencies’ external relations have rarely imposed limits on this functionalist outreach. Political limits have, however, been imposed when sector-​specific cooperation has become politicised in the context of overarching intergovernmental relations between the 248

DOI: 10.4324/9780429054136-18

EU agencies

EU and a third-​country, as, most recently, in the cases of Switzerland and the United Kingdom. Section four therefore explores the scope and limits of “condominio” between supranational control and national sovereignty, before we conclude.

Modes of governance and the flexibility of decentralised agencies As specialised organisations focused on the coordination of national regulators in a given policy area, EU agencies are typical instances of what Liesbet Hooghe and Gary Marks (2003) have referred to as “type II” institutions in the EU’s multilevel polity.Task-​specific governance in type II organisations contrasts with politically encompassing “type I” systems of governance that are based on territorial demarcations and cover a variety of policy areas, i.e. typically nation states but also the overarching EU polity (ibid.).While having their own structures and modes of operation,“type II” institutions are always embedded in overarching “type I” structures, with different degrees of flexibility and independence. The delegation of regulatory functions to technocratic executive bodies is a ubiquitous phenomenon in modern democracies and is at the centre of the notion of the regulatory state (Levi-​Faur 2011). This phenomenon has been found to be typical of the European multilevel and polycentric system (Majone 1996; Hooghe and Marks 2003), it, however, also exists at local (e.g. Ostrom 1990), national (e.g. Gilardi 2008) and international levels of governance (e.g. Keohane and Nye 1974; Slaughter 2004). While associated with particular modes of governance that are spelled out below, it should be noted that “type II” or technocratic, non-​majoritarian executive bodies are always embedded in overarching “type I” political structures –​therefore denoting compound rather than mutually exclusive modes of governance (see also Trondal and Jeppersen 2008). The modes of governance associated with “type II” institutions make them particularly interesting as venues for differentiated integration within a wider encompassing system such as the EU. In contrast to “type I” polities, “type II” jurisdictions are organised along functional lines resulting from the structure of interdependence of a particular problem; they are limited to specific tasks and sectors; they have intersecting, “polycentric” memberships across a large number of levels –​sometimes public and private –​, and they have flexible designs allowing them to adapt to changing situations (Hooghe and Marks 2003). The permeability of transgovernmental structures stems from their organisational features. They are based on horizontal ties between their members and have been described as hubs in a network of national experts (Eberlein and Newman 2008: 29). Integration does not consist in “vertical” transfers of competences to a higher, encompassing unit, but advances “horizontally” by linking competent bodies in national administrations in common regulatory activities. Policy-​making usually consists in the coordination of national regulations and practices.While an increasing number of agencies have been granted decision-​making authority, “soft law” is normally favoured over “hard law”, and integration also occurs through non-​legal means including the sharing of information, the development of shared curricula and operational cooperation in policy implementation. The horizontal structures also reflect in the composition of decentralised agencies. The management boards and subunits are normally composed of representatives of competent divisions in national ministries or of independent executive bodies which enjoy a certain degree of independence from central government and own expertise over the matter at hand. They are thus also those actors who not only contribute to the development of new policies but also those who will then implement agreed policies on the ground. In the case of the European Food Safety Authority (EFSA), the management board is not composed of representatives from national 249

Sandra Lavenex

administrations but by independent experts who are recruited on the basis of an open call and then appointed by the Council after consultation of the European Parliament, which gives this agency an even greater degree of independence from political actors. The emphasis on coordination, consensus and mutual learning lowers the hurdles for the participation of public officials from different “type I” jurisdictions and reduces the pressure for congruence. The experts and public officials involved enjoy a considerable degree of independence from their states’ central administration and are less subject to bureaucratic chains of command (Eberlein and Newman 2008: 32). While internally, these properties of “type II” governance relativise the importance of territorial boundaries, externally they allow member regulators to “follow function” and develop webs of foreign relations that blur the distinction between insiders and outsiders (Lavenex 2014; 2015a). These three characteristics (limited functional scope; the emphasis on task-​specific interdependence as rationale for inclusion; and the technocratic emphasis on expert-​led problem-​ solving rather than political representation) resonate with a functionalist vision of integration. According to David Mitrany (Mitrany 1965: 141), functionalist, task-​specific integration avoids the exclusivist tendencies of territorially based, encompassing political systems: “a federal system is bound to be closed and exclusive; a functional system is naturally open, as changes in membership can be absorbed without doing violence to policy and administration”. How does the transgovernmental layer of “type II” institutions such as decentralised agencies contribute to the wider edifice of European integration, and how far is this functionalist promise reflected in the differentiated inclusion of non-​member states?

Decentralised agencies as venues for external differentiated integration Interestingly, the progressive supranationalisation of decision-​making processes and the EU’s transformation from a primarily market-​related regulatory policy towards political system with “core state powers” (Genschel and Jachtenfuchs 2013) has gone hand in hands with a steep proliferation of decentralised agencies in all sectors of EU policy-​making. Egeberg and Trondal (2017) document an acceleration in the creation of new decentralised agencies over the last 15 years in terms of numbers but also in terms of powers and quality (ibid.: 679f., see also Ossenge 2016). Today, the EU counts a total of 36 agencies across market-​related and non-​ market-​related policy areas, including air safety, banking, border control, disease prevention, electricity, environment, food safety, medicines, migration, telecommunications, trademarks or external security. Decentralised agencies herewith form a constitutive layer of the EU’s institutional architecture. Broadly speaking, these networks of national regulators assist in the implementation but also in the development and specification of EU law and policy. Their tasks range from providing scientific advice for upcoming legislation and policy implementation, collecting and exchanging information, providing specific services, training national officials according to agreed curricula, offering capacity-​building and technical support to national authorities, carry out joint operations, and more generally coordinating national authorities. A good number of agencies have, however, also a regulatory mandate and define common standards and rules that are then valid (if not necessary always legally binding) for national authorities. If we apply the notion of “integration” proposed in this volume (Leruth et al. Introduction this volume), EU agencies make a significant contribution to European integration.They increase the “density” of interactions among national executives, deepen the “intensity” of these interaction based on commonly defined curricula, data, practices and regulations and modify the “character” of the relations among the participating countries by establishing a densely connected layer of transgovernmental cooperation. In consequence, participation in decentralised agencies increases 250

EU agencies

“consistency”, “interdependence” and “structural connectedness among the parts” –​and hence integration (Leruth, et al. this volume). Importantly, the type of integration promoted in decentralised agencies is complementary and to some extent also independent from the classic understanding of European integration “through law” (Dehousse and Weiler 1990;Weiler 1999).While the latter presupposes the vertical devolution of legal authority to common supranational institutions setting hierarchically binding legal standards which enjoy primacy over national law, integration through transgovernmental cooperation in EU agencies (as well as other EU secondary bodies such as committees or policy networks) works through the horizontal coordination of national regulations and practices (Mastenbroek and Martinsen 2018). At least formally speaking, transgovernmental integration is thus more protective of state sovereignty than the classic model based on supranational law. In other words Agencification’ of the EU administration may be regarded as a compromise between functional needs for the supply of more regulatory capacity at the European level, on one hand, and Member States’ reluctance to transfer executive authority to the European Commission, on the other Egeberg and Trondal (2017: 675), see also Kelemen (2002) Hofmann et al. (2019: 1) define agencies from a legal point of view “as bodies governed by European public law that are institutionally separate from the EU institutions, that have their own legal personality, that enjoy a certain degree of administrative and financial autonomy and that have clearly specified tasks”. As technocratic bodies set up with the task to promote effective solutions to genuinely transboundary problems such as consumer safety, environmental protection or migration, EU agencies depend on cooperation with interdependent states and organisations in order to fulfil their mandate.This functional need is recognised in the founding regulations of the vast majority of EU agencies. The importance of EU agencies’ external relations is reflected not only in increasing scholarly attention (Coman-​Kund 2018; Hoffman et al. 2019; Lavenex 2011; 2015a; Lavenex et al. 2021; Rimkute and Shyrokykh 2019) but also in manifest attempts by the EU central bodies, the Commission, Council and Parliament, to retain some control over these developments.

Third-​country participation in EU agencies The patterns of third-​country participation in EU agencies have been shaped by two distinct dynamics. The majority of decentralised agencies have developed international relations as a response to patterns of interdependence and have sought to formalise cooperation relations with their counterparts in those third countries with which they most closely interact in their daily functions. The second dynamic spurring the association of third-​country regulators is the EU’s overall foreign policy and the circles of association the EU has developed with its proximate neighbourhood. As a result, the external cooperation relations by EU agencies reflect a mix of highly diverse, sector-​specific patterns of functionalist interdependence on the one hand, and more territorially based political priorities on the other (Lavenex 2015a). While functionalist patterns of regulatory outreach predate the political guidance, the latter has intensified in the context of EU enlargement and the development of the European Neighbourhood Policy (ENP) from 2004 onwards. Today, the EU has adopted a number of policies guiding agencies’ external relations with priority countries under the common foreign policy, promoting 251

Sandra Lavenex

differentiated integration in particular of European Economic Area (EEA), candidate and ENP countries. Sector-​specific agreements granting access to EU agencies have been concluded with Switzerland as part of its bilateral agreements. Agencies’ external relations, however, also reach beyond the EU’s neighbours and include significant ties with peer regulators in countries such as Japan or the United States. Openness to third-​country cooperation is usually provided in the founding acts of EU agencies. The standard provision found in EU agencies’ constituent acts is that third countries shall be given the possibility of participating in these bodies if they “have concluded agreements… which provide for the adoption and application by these countries of Community law in the area covered by the basic act”. However, not all EU agencies are open to third state participation under their founding regulations, notable exceptions being the European Medicines Agency (EMA) and the European Union Intellectual Property Office (EUIPO). Specific provisions apply Iceland, Liechtenstein and Norway under the EEA agreement from 1992 and have been spelled out in subsequent years with the mushrooming of EU agencies from 1993 (see Table 16.1). EEA/​EFTA countries have been granted on a general basis participation “in the Administrative Board … [with] the same rights and obligations as EU Member States, except for the right to vote” in EU agencies (EEA 2012: §4.3). Considering that voting is not frequent in these boards, differentiated integration of EEA/​EFTA countries is thus far-​reaching at this transgovernmental level (and similar provisions apply to membership in EU programmes and committees).There is also one exception to this standard of full membership without voting rights which is the EFSA Advisory Forum in which EEA EFTA countries enjoy the same voting rights as the EU member states (Öberg 2019: 210). This far-​reaching integration is in line with these countries’ commitment towards the acquis communautaire under the EEA, including its dynamic development. The opening up of Eastern Europe in 1989 and the unfolding enlargement strategy brought a new group of countries in connection to the acquis communautaire. Candidate countries’ socialisation into EU policies and networks started with the opening-​up of certain EU programmes in the association agreements concluded with the Central and Eastern European countries in the beginning of the 1990s. Candidate countries’ participation in EU programmes was formally endorsed with the Copenhagen European Council Decisions in 1993. These provided that candidate countries should have access to the same programmes that had been opened EFTA states before (European Council 1993: 15). EU agencies came later onto the radar of the architects of EU enlargement in the Commission and the Council. The Commission included agencies with community programmes and committees into the pre-​accession strategy in 1999 with the proposal of a “framework approach” streamlining the Community procedures required to agree the conditions and modalities of the participation (European Commission 1999). Like for EU programmes and committees, participation in EU agencies is seen as a means to familiarise candidate countries with EU policies and to support the development of the legislative and administrative capacities required for adopting the EU acquis communautaire. Agencies are encouraged to offer participation possibilities, including observer status, to candidate countries in Management Board and expert group meetings. In other words, the differentiated integration of candidate countries in EU agencies is one instrument in the EU’s tool-​box of policy transfer based on learning, socialisation and capacity-​building in the context of EU enlargement. This is also supported by the EU Instrument for Pre-​Accession Assistance providing funding to EU agencies to help prepare candidate states for participation in certain agencies. In practice, this instrumentalisation of decentralised agencies as venues for pre-​accession policies has been less encompassing or coherent than Commission or Council documents suggest. Access by candidate countries to EU agencies’ management boards or secondary bodies has 252

EU agencies Table 16.1  EU Agencies, year of creation, function and legislative power Abbr.

Name

Year creation

Function

Legally binding acts?

ACER

2011

Supervision

Yes

2010

Cooperation

No

1994

Operational activities

No

1975

Information

No

2005

Information

No

CPVO EASA

Agency for the Cooperation of Energy Regulators Office of the Body of European Regulators for Electronic Communications Translation Centre for the Bodies of the European Union European Centre for the Development of Vocational Training European Union Agency for Law Enforcement Training Community Plant Variety Office European Aviation Safety Agency

1994 2003

Yes Yes

EASO

European Asylum Support Office

2011

EBA ECDC

2011 2005

ECHA

European Banking Authority European Centre for Disease Prevention and Control European Chemicals Agency

Certification Certification +​Scientific/​ technical expertise Information +​Cooperation +​operational activities +​ scientific/​technical expertise Supervision Information +​Cooperation

Yes

EDPB EEA

European Data Protection Board European Environment Agency

2018 1994

EFCA

European Fisheries Control Agency

2005

EFSA EIGE EIOPA

European Food Safety Authority European Institute for Gender Equality European Insurance and Occupational Pensions Authority European Labour Authority

2002 2007 2011

Scientific/​technical expertise +​ Registration Supervision Information +​Scientific/​ technical expertise Operational activities +​Cooperation +​inspection Scientific/​technical expertise Information +​Cooperation Supervision +​cooperation

No

European Medicines Agency European Monitoring Centre for Drugs and Drug Addiction European Maritime Safety Agency European Union Agency for Network and Information Security European Public Prosecutors’ Office European Union Agency for Railways European Securities and Markets Authority European Training Foundation

1995 1993

Information +​inspection +​ cooperation Scientific/​technical expertise Information

2002 2005

Cooperation +​Supervision Information

No No

2017 2004 2011

Operational activities Cooperation +​Supervision Supervision

Yes Yes Yes

1994

Information +​Cooperation

BEREC

CdT Cedefop CEPOL

ELA EMA EMCDDA EMSA ENISA EPPO ERA ESMA ETF

2007

2019

No

Yes No

No No No No No Yes

No No

No (continued)

253

Sandra Lavenex Table 16.1  Cont. Abbr.

Name

Year creation

Function

Legally binding acts?

EUIPO

European Union Intellectual Property Office European Agency for the Operational Management European Agency for Safety and Health at Work European-​Foundation for the Improvement of Living and Working Conditions Eurojust

1999

Registration

Yes

2012

Operational activities

No

1994

Information +​Cooperation

No

1975

Information

No

2002

Yes

European Union Agency for Law Enforcement Cooperation European Union agency for Fundamental Rights European Boarder and Coast Guard European GNSS Agency

1999 2007

Cooperation +​Operational activities Cooperation +​Operational activities Information

2016 2004

Operational activities Information

No No

EU-​LISA EU-​OSHA Eurofound

Eurojust Europol FRA FRONTEX GSA

No No

Source: Own compilation, function and legally binding acts taken from Vos (2018).

remained exceptional. The European Aviation Safety Agency (EASA) is among the most open ones and has provided for candidates to join the management board as observers from its establishment in 2002 (Lavenex 2015a; Lavenex et al. 2021). However, most other agencies either provide no access at all or only observer status in selected secondary bodies or advisory forums (such as, from 2011, the European Centre for Disease Prevention and Control [ECDC] advisory forum or, for the European Medicinal Agency EMA, in scientific and technical non-​product-​related meetings). Some exceptions apply to Turkey which has joined the European Environmental Agency EEA as member and invited to participate fully but without voting rights in the management board of the European Monitoring Centre for Drugs and Drug Addiction EMCDDA. In sum, candidate countries’ participation in EU agencies is therefore limited. Instead, candidate countries’ interaction with EU agencies is mainly based on concrete capacity-​building programmes and operational activities. These can, however, be quite substantive. An eminent example of far-​reaching differentiated integration in agencies’ operational activities is Frontex. The first working arrangements signed between Frontex and candidate countries regulated only exchange of information between the Frontex Risk Analysis Unit and the Border and Migration Departments of these countries. From 2019 onwards, the EU has been negotiating status agreements with the candidate countries of the Western Balkans that provide for joint border control teams composed of Frontex (member state) and third-​country officials sharing executive powers for border control and return operations on the territory of the third country. With the accession of the CEEC countries in 2004 and 2007, EU foreign policy attention has shifted to the neighbouring countries which would not be given a clear membership perspective. In the famous words by former EU Commission President Romano Prodi, the ENP launched in 2003 was to offer “everything but institutions”. Although membership in EU core decision-​making institutions such as the Commission, Council and Parliament was thus ruled 254

EU agencies

out, the offer of less encompassing forms of association did foresee access to EU secondary bodies, including agencies. The ENP has its legal basis in Article 8 of the Treaty on EU which provides that the EU “shall develop a special relationship with neighbouring countries, aiming to establish an area of prosperity and good neighbourliness, founded on the values of the Union and characterised by close and peaceful relations based on cooperation”. Article 8 also provides that the EU “may conclude specific agreements” with ENP countries, which “may contain reciprocal rights and obligations as well as the possibility of undertaking activities jointly”. Opening up EU agencies to the participation of ENP countries was envisaged as early as 2006 with the Commission’s Communication “on the general approach to enable ENP partner countries to participate in Community agencies and Community programmes” (European Commission 2006). This Communication emulated many elements of the enlargement policy, including the idea that “the participation of neighbouring countries [in EU agencies] may be in the interests of the enlarged EU and of the neighbouring countries” (European Commission 2006: 3). It also stressed that ENP countries’ participation in agencies would serve “the EU’s overarching goal of encouraging and supporting regulatory and administrative reform and institution building in neighbouring countries…in the overarching Community interest” (ibid.: 3). Notwithstanding these references to the role of decentralised agencies in foreign policy, the decision to open up towards a particular country rests with the EU agencies and not with the EU foreign policy apparatus. In practice, participation of ENP countries has evolved very little beyond cooperation targeted at promoting alignment with EU standards. The agency that has developed most meaningful cooperation with ENP countries is EASA.This is based on comprehensive air transport agreements with candidate and neighbourhood countries in view of a wider Common Aviation Area based on a parallel process of gradual market opening and regulatory convergence towards EU rules. Neighbourhood country regulators are associated to EASA and its information system on air safety of aircraft and of air operators (SAFA) through a harmonised system of inspections, information-​sharing database, standardisation visits, training programmes, joint inspection operations and coordination on intended regulatory changes. Partner countries partake in the coordination meetings of the SAFA programme and have a (technical) advisory role therein.With all other EU agencies, ENP countries’ participation is more ad hoc and targeted at specific capacity-​building and technical cooperation programmes. The exception to this rule is Israel, which has developed stronger ties with a number of EU agencies. A case in point is its cooperation agreement with the EMA which allows for mutual recognition agreements in relation to conformity assessment of medicinal products. Israel’s privileged position vis-​à-​vis other ENP countries underlines the importance of market integration and comparable levels of administrative capacity and regulatory expertise for stronger transgovernmental integration. For instance, EMA’s agreement with Israel recognises the country’s industrial standards as equivalent to European standards. The agreement also presupposes Israel’s capacity to implement EU legislation and to align so-​called good manufacturing practice standards, thus providing the basis for mutual recognition (Lavenex 2015a: 846). The discussion of the candidate and ENP countries shows that EU agencies primarily play a role as agents of EU external governance, i.e. “the projection of EU rules beyond its borders” (Lavenex 2004; Lavenex and Schimmelfennig 2009), including capacity-​building for implementing those rules. In so far as EU agencies establish links with peer regulators in these countries they do contribute to differentiated integration in the sense defined by Leruth et al. Introduction, this volume).They increase the “density” of interactions among national executives and deepen the “intensity” of these interaction based on commonly defined curricula, data, practices and regulations deriving from the EU acquis. While the “character” of the relations among the participating countries is more a one-​way street than technocratic cooperation 255

Sandra Lavenex

within the EU or with more similar highly developed regulators, one can also say that outreach by EU agencies contributes to an emerging layer of transgovernmental cooperation with public authorities in ENP countries. The fact that this layer has not led to more participation in the organisational structures of EU agencies, however, also points at important preconditions for transgovernmental integration. These include not only functional interdependence but also the existence of compatible bureaucratic structures, including competent authorities to liaise with EU counterparts, as well as the administrative capacity and capability to follow up on common action (Lavenex 2015b). The differentiated integration of candidate and ENP countries in EU agencies thus follows a primarily political logic driven by EU’s institutions overarching foreign policy ambition of avoiding “sharp edges” (Grabbe 2000) in and around Europe. While contributing to the density and intensity of interaction between these countries’ public administrations and the EU’s multilayered executive order, this integration operates one-​way based on the projection of EU rules and practices and finds its limits in the weakness of functional ties and interdependencies, including differences of administrative structures and capabilities. This situation contrasts with that of the EEA/​EFTA countries that have been fully associated with EU agencies as well as other Western countries that share stronger economic and regulatory ties with the EU, even if this does not always imply adaptation to the EU’s acquis communautaire. Switzerland occupies a special place in this context. On the one hand, Swiss authorities had developed transgovernmental links too peer regulators neighbouring EU countries in many policy areas already before the establishment of EU agencies (Lavenex 2009). On the other hand, Switzerland has progressively deepened its commitment to alignment with the EU acquis communautaire under its sectoral bilateral agreements with the EU (Lavenex and Schwok 2015). This creates a double, functionalist and political impetus for association to EU agencies. Contrary to the EEA/​EFTA countries however, which by way of their encompassing association to EU legislative integration have been granted generalised access, Switzerland’s bilateral exceptionalism necessitates negotiating separately each association to an EU agency. In recent year, Swiss-​EU relations have become increasingly politicised due to contrasting visions of the future of bilateral relations and the institutional architecture of an overarching institutional association agreement (Veuthey 2020). This has had repercussions on the functionalist drivers of differentiated integration in EU agencies. In the early years of “agencification”, Switzerland gained access in a relatively informal and permissive manner to EU agencies, based on the existing links between national authorities. For instance, and according to Swiss government sources, cooperation with the EEA started already in 1993 –​even before the agency was officially installed in 1994. Cooperation was informal and concentrated on technical and scientific cooperation in concrete projects. This cooperation became more formal in the context of the first round of bilateral negotiations in 2002 when Switzerland obtained the status of an associated country. This allowed Switzerland to send an expert to the EEA, including access to the management board. In 2006, Switzerland became a full member like the EEA/​EFTA countries and Turkey (Eidgenossenschaft 2010). Today, Switzerland is a formal member in two agencies dealing with the environment and air traffic (the EEA and EASA) and, in line with its encompassing association to the Schengen/​ Dublin acquis, three justice and home affairs agencies: Frontex, the European Asylum Support Office (EASO) and the European Union Agency for Law Enforcement Training (CEPOL). Moreover, it has comprehensive cooperation agreements with Europol and Eurojust (these two agencies do not foresee for third-​country membership). The wave of “agencification” in the EU has, however, motivated demand for accession to more agencies. De facto, Swiss authorities do participate in a variety of other transgovernmental networks operating under the ambit of EU 256

EU agencies

agencies, and in some agencies, it cooperates on an informal basis. For instance, Switzerland has enjoyed observer status in the EFSA Advisory body for several years, without having concluded a formal agreement (Ibid.: 69). Swiss experts have been cooperating in EFSA’s scientific panels and in ad hoc expert groups based on mutual interests and informal, often personal contacts.With the formalisation of transgovernmental cooperation in EFSA and the creation of new instruments such as the early warning systems RAPEX and RASFF, however, Switzerland has been gradually lost access to relevant fora and information. This has motivated a formal demand to conclude an association agreement with EFSA.Yet, this demand has been blocked due to overarching political controversies between EU central bodies and the Swiss government over the conclusion of an encompassing framework agreement which would put the various bilateral agreements under a common institutional framework, including commitment to progressive approximation to the EU acquis and formal supervision mechanisms (Lavenex and Schwok 2015;Veuthey 2020). A similar situation of informal functionalist integration constrained by overarching political contestations also applies to the field of health cooperation against infectious diseases. Switzerland has been an active player in transgovernmental health networks in Europe, including those dealing with epidemiological data or the exchange of best practices. It has also participated in the Early Warning and Response System (EWRS) integrated under the ECDC. With increasing formalisation of transgovernmental cooperation in the ECDC, however, Switzerland as feared becoming gradually disconnected from these networks, motivating the Swiss government to ask for formal association (Eidgenossenschaft 2010: 72). The case of Switzerland is illustrative of the tensions between encompassing political integration in Europe and decentralised, functionalist and transgovernmental integration à la “condominio”. While the “Swiss way” (Lavenex and Schwok 2015) has increasingly come under the focus of political integration, functionalist outreach persists with third countries that share high levels of functional interdependence and regulatory capacity but haven’t negotiated encompassing associations to the EU. A case in point are Australia, Canada, Japan and the United States which enjoy far-​reaching participation rights in several EU agencies based on bilateral agreements concluded with these agencies. Again, EASA provides for the most far-​ reaching cooperation with these more remote Western partners. Contrary to the arrangements concluded with candidate or ENP countries, EASA’s comprehensive bilateral agreements with the United States, Canada, Brazil, China and Japan allow for eye-​level cooperation based on the principle of mutual recognition, including for safety certificates (Hoekman and Sabel 2019; Lavenex et al. 2021). In some agencies, the United States has been granted access to secondary bodies, such as the possibility for US FDA staff members to participate as observers the EMA’s Pharmacovigilance Risk Assessment Committee (PRAC), the Committee on Human Medicinal Products (CHMP) on product-​related safety concerns or the Pediatric Committee meetings. Similarly, the United States, Canada and Australia cooperate with the EMA and competent subcommittees in joint training activities or in the mutual recognition of conformity assessments. Another agency which has traditionally had strong ties with the United States is the European Securities and Markets Authority (ESMA). Already its predecessor, the Committee of European Securities Regulators (CESR) enjoyed close regulatory cooperation with the US Securities and Exchange Commission (SEC). Created in the aftermath of the 2007/​2008 financial crisis, ESMA has concluded administrative arrangements with 32 important financial centres around the world, including the United States, Japan, Canada, Australia and Singapore (Lavenex et al. 2021: 8f.). With Brexit, the question of differentiated integration in EU agencies has also become pertinent for the United Kingdom. Seeking a closer association with the EU, the Brexit arrangement negotiated under Prime Minister Theresa May foresaw continued UK membership in most EU 257

Sandra Lavenex

agencies (UK 2018). With the choice for a looser EU-​UK Trade and Cooperation Agreement, the Johnston Government has decided against regulatory association to EU standards and has ordered resignation from all EU agencies. However, given the UK’s functional interdependence with the single market and also adjacent policies such as justice and home affairs or defence cooperation, the question of UK’s integration in the wider European transgovernmental networks, including their integration in EU agencies, is likely to stay on the agenda. For instance, the UK’s aerospace sector is the second largest in Europe and the third largest in the world. The UK government’s announcement in March 2020 that it would be withdrawing from EASA end of 2020 immediately raised wide criticisms on part of the aviation industry for fear of significant costs, diminished attractiveness for foreign investment and loss of influence in the EU agency as one of the world’s major regulators (Senior European Experts 2020). Similar concerns have been raised across sectors, including from the chemical industry regarding the European Chemicals Agency (ECHA), the financial sector regarding the European Banking Authority (EBA) and ESMA, pharmaceutics regarding the EMA or the fisheries sector concerning the European Maritime Safety Agency (EMSA), energy regulators regarding the Agency for the Cooperation of Energy Regulators (ACER) or justice and home affairs officials regarding participation in Europol and Eurojust. In sum, a recent report concludes that The costs of the UK’s withdrawal from EU regulatory bodies include the direct costs of setting up and skilling new UK regulatory functions, … and the direct costs to UK exporting businesses of obtaining dual authorisations. But there will also be indirect costs in terms of increased trade friction and duplication, possible transfer of operations from the UK to the EU, loss of influence in pan-​European (and sometimes global) standard setting and the potential loss of regulatory expertise Senior European Experts 2020: 6 For all these reasons, UK regulators are most likely to seek to maintain close links with their peers in EU member states and pertinent EU agencies, thereby sustaining the drivers for external differentiated integration.

Scope and limits of condominio The review of external differentiated integration in EU agencies above gives support to a functionalist vision of a decentred multi-​sited web of transgovernmental “type II” institutions that comes close to Philippe Schmitter’s notion of “condominio”. However, these institutions do not “float freely” and are to variable extents enmeshed in the wider system of territorially based, encompassing “type I” governance in the EU. The scope and limits of this “condominio” is circumscribed by the interplay between this layer of integration and the more federal mix of intergovernmental and supranational architectures in Europe. The case of Switzerland discussed above has indicated that when overarching diplomatic relations become politicised, supranational actors can seriously circumscribe the scope for flexible transgovernmental arrangements. The case of the United Kingdom is indicative of the incisive role national governments can play, and also of the “shadow of hierarchy” looming in the background of more soft law deliberations in decentralised agencies: indeed one of the reasons for the UK’s complete withdrawal from these institutions is that most EU agencies come under the jurisdiction of the ECJ, which runs counter the emphasis put on “recovering sovereignty” and “regulatory autonomy” emphasised in electoral politics (UK 2018).

258

EU agencies

Developments over the last two decades corroborate a tendency of supranational actors to gain stronger control over the decentred web of transgovernmental bodies. Up until the second wave of agencification in the early 2000, EU institutions refrained from interference with the work and evolution of decentralised agencies (Curtin and Dehousse 2012: 194). The impetus for stronger oversight came with the allegations of fraud and mismanagement that led to the fall of the Santer Commission. On the one hand, the Commission’s White paper on European Governance of 2001 (European Commission 2001: 24) emphasised the importance of delegating administrative tasks to agencies as non-​majoritarian bodies as part of a shift to new modes of more effective and legitimate governance (Curtin and Dehousse 2012: 195). On the other hand, this formal endorsement, coupled with the proliferation of agencies, motivated the desire for stronger supranational guidance of these bodies. In 2005, the Commission presented a draft for an interinstitutional agreement on European regulatory agencies (European Commission 2005). Neither the Parliament nor the Council saw the need for such a framework. Instead, the three EU bodies established a working group which, after lengthy discussions, adopted in 2012 the non-​binding Common Approach to EU agencies (Groenleer 2009: 110). The document establishes a number of general principles promoting the “coherence, effectiveness, accountability and transparency” of decentred agencies, including the principle that “agencies international relations should be streamlined” (Council of European Union 2012). In this view, the document provides that agencies which cooperate internationally should have “a clear strategy for those activities” which, together with “appropriate working arrangements with partner DGs in the Commission should ensure that the agencies operate within their mandate…and they are not seen as representing the EU position to an outside audience or as committing the EU to international obligations” (ibid.: §25). The document further stipulates that this strategy must be approved by the agencies’ management board. The agencies thus remain in the driving seat and retain to a large extent their autonomy, which is also reflected in the understanding that they do not represent the EU as such. However, closer coordination with the Commission is called for through the working arrangements with partner DGs and the provision that “an early exchange of information should take place on respective international activities between agencies, the Commission and the relevant EU Delegations, to ensure consistency of EU policy” (ibid.). These principles have also trickled down into the establishing acts of EU agencies. As a recent analysis shows, with the reform of their funding regulations, most EU agencies now require stronger coordination with the Commission and sometimes the Council and Parliament for concluding cooperation agreements with third countries, reaching from a looser requirement of “consultation” to the need for “approval” (Lavenex et al. 2021). As a general rule, agencies’ functionalist outreach has not relented with these increased coordination requirements. On the contrary, one can say that overarching EU political priorities such as the association of candidate and neighbourhood countries have come to supplement the decentred links to peer regulatory. However, the case of Switzerland also shows that supranational actors can put a brake on transgovernmental differentiated integration. The more the participation in EU agencies and associated subcommittees and networks becomes formalised under EU law, and the more this participation is identified as a stake in political controversies, the less scope there may be for flexible differentiated integration in “condominio”.

Conclusion This chapter has introduced cooperation in EU decentralised agencies as a hitherto under investigated space for differentiated integration in Europe. As anticipated by early functionalist

259

Sandra Lavenex

theories of integration and echoed in the literatures on multilevel and new modes of governance, the form and reach of public administrations follow their tasks and functions, leading to a differentiated web of transgovernmental ties and networks taking different territorial shapes across policy areas and sectors.This process of decentred transgovernmental integration has been spurred by the parallel devolution of regulatory authority to more or less independent technocratic bodies in liberal democracies and the EU. The waves of “agencification” in the EU system have built on previously established ties between national regulators, often involving also the cooperation with peers from non-​member states. As the discussion of third-​country participation in EU agencies shows, this transgovernmental layer of differentiated integration reflects primarily the patterns of interdependence predominating in the various sectors.The advent of new foreign policy priorities towards candidate countries and the countries of the ENP have added a more political rationale for transgovernmental integration. Supranational actors and in particular the Commission have encourage EU agencies to take an active role in the promotion of the acquis communautaire and the socialisation of candidate/​ENP countries’ administrations, including capacity building and training. In practice, however, these countries’ participation in EU agencies has remained limited, and only more advanced economies, including Israel and Turkey, have developed a meaningful access to some agencies. This contrasts with the full association (without voting rights) of the EEA/​EFTA countries and, in some cases, Switzerland, but also with the deep forms of cooperation based on the principle of mutual recognition that exists with peer regulators from non-​associated countries such as Australia, Canada, Japan or the United States in some agencies like the EASA, EMA or ESMA. In line with the definition adopted in this volume, these links amount to a form of differentiated integration They increase the “density” of interactions among national executives, deepen the “intensity” of these interaction based on commonly defined curricula, data, practices and regulations and modify the “character” of the relations among the participating countries by establishing shared regulatory regimes and blurring the distinction between “insiders” (from EU member states) and “outsiders” (from third countries). In consequence, participation in decentralised agencies increases “consistency”, “interdependence” and “structural connectedness among the parts” –​and hence integration (Leruth et al Introduction to this volume). As such, these patterns come close to Philippe Schmitter’s vision of a “condominio” based on functionally and territorially differentiated, sector-​specific hubs of regional integration (Schmitter 1996). That being said, our analysis also highlights supranational actors’ and in particular the Commission’s attempts to gain some control over agencies’ international relations. This reflects in the priorities given to candidate and ENP countries in EU overarching documents, new supranational guidelines and the introduction of new coordination requirements in the agencies’ funding regulations. As a general rule, these altered requirements have not constrained the functionalist patterns of differentiated integration in EU agencies. An exception, however, is Switzerland, for which formal accession to new agencies has been put on hold due to ongoing controversies over an institutional framework agreement. Limits to transgovernmental functionalist integration can, however, also stem from national politics, as shown in the case of the United Kingdom.Yet, whereas the UK government opted for a “hard” Brexit, including resignation from EU agencies, the deep interdependence of markets and societies is likely to demand for new forms of association also at this level of public authorities, and first reactions by affected industries and domestic regulators clearly point in this direction. To conclude, public administrations in Europe constitute a hitherto under-​investigated layer of differentiated transgovernmental integration that operates at arms lengths from centripetal supranational and particularist intergovernmental forces. While particularly salient in the EU’s 260

EU agencies

architecture, this phenomenon is not limited to the EU and its agencies but speaks to a more general feature of evolving global governance. With the progressive integration of markets and societies, expressed also in the increasing attention put on regulatory cooperation in preferential trade agreements, and, at the same time, the necessity to accommodate increasing diversity in Europe, differentiated integration à la “condominio” is likely to persist, with varied ties to supranational and national principals.

Note 1 Research funding by the H2020 Project “EU Integration and Differentiation for Effectiveness and Accountability” (EU-​IDEA) is gratefully acknowledged. This project involves the creation of comprehensive and systematic database of third country participation in EU agencies since their creation until today. Section three of this chapter partly draws on this collaborative work. The author thanks Signe Moe, Alexandre Veuthey as well as Benjamin Bertrand, Sarah Falcucci and Matis Poussardin for excellent research assistance and the volume’s editor for valuable feedback.

References Coman-​Kund, F. (2018) European Union Agencies as Global Actors: A Legal Study of the European Aviation Safety Agency, Frontex and Europol. London: Routledge. Council of the European Union (2012) Evaluation of European Union Agencies –​Endorsement of the Joint Statement and Common Approach.europa.eu. Available at: https://​data.consil​ium.eur​opa.eu/​doc/​ docum​ent/​ST-​11450-​2012-​INIT/​en/​pdf (Accessed: 26 April 2021). Curtin, D. and Dehousse, R. (2012) ‘EU Agencies: Tipping the Balance?’, in M. Busuioc, M. Groenleer and J. Trondal (eds) The Agency Phenomenon in the European Union. Emergence, Institutionalisation and Everyday Decision-​Making. Manchester: Manchester University Press. Dehousse, R. and Weiler, J.H.H. (1990) The Legal Dimension, in The Dynamics of European Integration, in W. Wallace and Weiler, J.H.H (eds). The Constitution of Europe. Cambridge: Cambridge University Press, pp. 99–​112. European Economic Area (2012) Reort on EEA EFTA States' participation in EU agencies, EEA Joint Parliamentary Committee Ref. 1118266 of 27 November 2012, Brussels. Eberlein, B., and Newman, A.L. (2008) Escaping the international governance dilemma? Incorporated transgovernmental networks in the European Union. Governance, 21(1), 25–​ 52. doi: 10.1111/​ j.1468-​0491.2007.00384.x. Egeberg, M., and Trondal, J. (2017) Researching European Union Agencies: what have we learnt (and where do we go from here)?. Journal of Common Market Studies, 55(4), pp. 675–​690. doi:10.1111/​ jcms.12525. Eidgenossenschaft, S. (2010) ‘Bericht des Bundesrates zum Verhältnis der Schweiz zu den europäischen Agenturen’, Bern, September. https://​scho​lar.goo​gle.com/​scho​lar?hl=​en&as_​sdt=​0%2C5&q=​Sch​ weiz​ e ris​ c he+​ E idgen ​ o sse ​ n sch ​ a ft+​ % 282 ​ 0 10%29+ ​ % E2%80%98Beri​ c ht+ ​ d es+​ B und ​ e sra ​ t es+ ​ z um+​ Verh%C3%A4lt​nis+​der+​Schw​eiz+​zu+​den+​europ%C3%A4isc​hen+​Agentu​ren%E2%80%99%2C+​ Bern.+​&btnG (Accessed: 26 April 2021). European Commission (1999) Communication of 20 December 1999 to the Council on “Participation of candidate countries in the Community programmes, agencies and committees”, COM(1999) 710 final. European Commission (2001) Enhancing democracy: a white paper on governance in the European Union. COM (2001) 428 final. Brussels. Europa.eu. Available at: https://​eur-​lex.eur​opa.eu/​Lex​UriS​ erv/​Lex​UriS​erv.do?uri=​COM:2001:0428:FIN:EN:PDF (Accessed: 26 April 2021). European Commission (2005) Draft Interinstitutional Agreement on the operating framework for the European regulatory agencies, COM (2005)59. Brussels. Europa.eu. Available at: https://​eur-​lex.eur​opa. eu/​Lex​UriS​erv/​Lex​UriS​erv.do?uri=​COM:2005:0059:FIN:EN:PDF (Accessed: 26 April 2021). European Commission (2006) Communication from the Commission to the Council and to the European Parliament on the general approach to enable ENP partner countries to participate in Community agencies and Community programmes’, COM (2006) 724 final, Brussels.Europa.eu. Available at: https://​ eur-​lex.eur​opa.eu/​Lex​UriS​erv/​Lex​UriS​erv.do?uri=​COM:2006:0724:FIN:EN:PDF (Accessed: 26 April 2021). 261

Sandra Lavenex

European Council (1993) Council Conclusions of the Presidency, SN 180/​ 1/​93 REV 1. Europa.eu. Available at: www.europarl.europa.eu/​enlargement/​ec/​pdf/​cop_​en.pdf (Accessed: 26 April 2021). Genschel, P., and Jachtenfuchs, M. (2013) Beyond the Regulatory Polity? The European Integration of Core State Powers. Oxford: Oxford University Press. Gilardi, F. (2008) Delegation in the Regulatory State: Independent Regulatory Agencies in Western Europe, Cheltenham: Edward Elgar Grabbe, H. (2000) The Sharp Edges of Europe: Extending Schengen Eastwards. International Affairs, 76(3), pp. 519–​536. doi: 10.1111/​1468-​2346.00149. Groenleer, M., (2009) The Autonomy of European Union Agencies: A Comparative Study of Institutional Development. Tilburg: Eburon Uitgeverij BV. Hoekman, B., and Sabel, C. (2019) Open plurilateral agreements, International Regulatory Cooperation and the WTO. Global Policy, 10(3), pp. 297–​312. doi: 10.1111/​1758-​5899.12694. Hofmann, H.C.H., Vos, E. and Chamon, M. (eds.) (2019) The External Dimension of EU Agencies and Bodies: Law and Policy, Edward Elgar Publishing. Hooghe, L., and Marks, G. (2003) Unravelling the central state, but how? Types of multi-​level governance. American Political Science Review, 97(2), pp. 233–​ 43. Available at: www.cambridge.org/​core/ services/​aop-​cambridge (Accessed: 26 April 2021) –​core/​ content/​ view/​ 8A5A618038BDE546E 5C582096F0B201F/​S0003055403000649a.pdf/​unraveling_​the_​central_​state_​but_​how_​types_​of_ multilevel_​governance.pdf. Kelemen, R.D. (2002) The politics of ‘Eurocratic’ structure and the New European Agencies. West European Politics, 25(4), pp. 93–​118.doi: 10.1080/​713601644. Keohane, R.O. and Nye, J.S. (1974) Transgovernmental relations and international organizations. World Politics, 27, p. 39. Available at: www.jstor.org/​stable/​pdf/​2009925.pdf (Accessed: 26 April 2021). Lavenex, S. (2004) EU external governance in wider Europe, Journal of European Public Policy, 11(4), pp. 680-​700. Lavenex, S. (2009) Switzerland’s flexible integration in the EU: A conceptual framework, Swiss Political Science Review, 15 (4), pp. 547-​575. Lavenex, S. (2011) Concentric circles of ‘EUropean’ integration: a typology of EU external governance relations, Comparative European Politics, 9(3), pp. 372–​393.doi: 10.1057/​cep.2011.7. Lavenex, S. (2014) The power of functionalist extension. How EU rules travel, Journal of European Public Policy, 21(6), pp. 885–​903.doi: 10.1080/​13501763.2014.910818. Lavenex, S. (2015a) The external face of differentiated integration. Third country participation in EU sectoral bodies, Journal of European Public Policy, 22(6), pp. 836–​853. doi: 10.1080/​13501763.2015.1020836. Lavenex, S (2015b) Experimentalist governance in EU neighbourhood policies: functionalist versus political logics, in J. Zeitlin (ed.), Extending EU Experimentalist Governance, Oxford: Oxford University. pp. 23–​49. Lavenex, S. Križić, I. and Veuthey, A. (2021): EU Boundaries in the making: functionalist versus federalist, Journal of European Public Policy, 28(3), pp. 427–​446. doi: 10.1080/​13501763.2021.1881586. Lavenex, S. and Schimmelfennig, F. (2009) EU rules beyond EU borders: theorizing external governance in European politics, Journal of European Public Policy, 16(6), pp. 791–​ 812. doi: 10.1080/​1350176 0903087696. Lavenex, S and Schwok, R. (2015) The Swiss Way.The Nature of the Swiss Relationship with the EU, in E.O. Eriksen (ed), The European Union’s Non-​Members: Independence Under Hegemony?. London: Routledge. p. 248. Levi-​Faur, D. (2011) Regulatory networks and regulatory agencification: towards a single European regulatory space, Journal of European Public Policy, 18(6), pp. 810–​829, doi: 10.1080/​13501763.2011.593309. Majone, G. (1996). Regulating Europe. Routledge. Available at: https://​books.goo​gle.no/​books?hl=​ en&lr=​ & id=​ 0 UG19​ R bCu​ B wC&oi=​ f nd&pg=​ PA47&dq=​ M aj​ o ne,+​ G.+​ ( 1996).+​ R eg​ u lat​ i ng+ Eur​ope.+​Routle​dge.+​&ots=​pO_​kJ0l​bm_​​&sig=​BL1p​CpbM​L6Ju​d2Qq​we4c​3L7u​sMg&redir_​esc= ​ y#v=​onep​age&q=​Maj​one%2C%20G.%20(1996).%20Reg​ulat​ing%20Eur​ope.%20Ro​utle​dge.&f=​f alse (Accessed: 26 April 2021). Mastenbroek E. and Sindbjerg Martinsen, D. (2018). Filling the gap in the European administrative space: the role of administrative networks in EU implementation and enforcement, Journal of European Public Policy, 25(3), pp. 422–​435. doi: 10.1080/​13501763.2017.1298147. Mitrany, D. (1965). The prospect of European integration: federal or functional. Journal of Common Market Studies, 4(2), pp. 119–​149. Available at: https://​hei​nonl​ine.org/​HOL/​Land​ingP​age?han​dle=​hein.journ​ als/​jcm​ks4&div=​11&id=​&page=​ (Accessed: 26 April 2021). 262

EU agencies

Öberg, M.L. (2019). Third Countries in EU Agencies: Participation and Influence, in H. Hofmann, C.H. Hartman, E.Vos and M. Chamon (eds). The External Dimension of EU Agencies and Bodies: Law and Policy. Cheltenham: Edward Elgar Publishing pp. 204–​221. Ossege, C. (2016) European Regulatory Agencies in EU Decision-​Making Basingstoke: Palgrave Macmillan. Ostrom, E. (1990). Governing the Commons: The Evolution of Institutions for Collective Action. Cambridge: Cambridge University Press. Rimkutė, D., & Shyrokykh, K. (2019).Transferring the acquis through EU agencies: the case of the European Neighbourhood Policy countries, in H. Herwig, E. Vos and M. Chamon (eds.) The External Dimension of EU Agencies and Bodies. Edward Elgar Publishing. Schmitter, P.C. (1996). Examining the Present Euro-​polity with the Help of Past Theories, in G. Marks, F. Scharpf, P.C. Schmitter, and W. Streeck (eds), Governance in the European Union, pp. 121–​50. London: Sage. Senior European Experts (2020) Brexit: UK Regulatory Agencies, Policy Brief. Senioreuropeanexperts.org. Available at: https://​senioreuropeanexperts.org/​ Slaughter, A.M., (2004) Sovereignty and power in a networked world order. Stanford Journal of International Law, 40, p. 283. Available at: https://​hei​nonl​ine.org/​HOL/​Land​ingP​age?han​dle=​hein.journ​als/​stani​ t40&div=​16&id=​&page=​ (Accessed: 26 April 2021). Trondal, J. and Jeppesen L. (2008) Images of Agency Governance in the European Union, West European Politics 31(3), pp. 417–​441. doi: 10.1080/​01402380801939636. UK (2018). The Future Relationship Between the United Kingdom and the European Union, Report by the UK Prime Minister to the Parliament. London: UK Government. Available at: https://​ass​ets.pub​lish​ing. serv​ice.gov.uk/​gov​ernm​ent/​uplo​ads/​sys​tem/​uplo​ads/​atta​chme​nt_​d​ata/​file/​786​626/​The_​Future_​ Relationship_​between_​the_​United_​Kingdom_​an​d_​th​e_​Eu​rope​an_​U​nion​_​120​319.pdf (Accessed: 26 April 2021). Veuthey, A. (2020) Negotiating privileged partnerships: EU-​ Switzerland relations and the joined-​ up approach in practice, CEPOB Policy Brief 10.20. Coleurope.eu. Available at: www.coleurope.eu/​ research-​paper/​negotiating-​privileged-​partnerships-​eu-​switzerland-​relations-​and-​joined-​approach. Weiler, J. (1999) The Constitution of Europe. ‘Do the New Clothes Have an Emperor?’ and Other Essays on European Integration, Cambridge: Cambridge University Press.

263

17 The Council of the European Union Organizational and social dynamics of differentiation Jeffrey Lewis

In today’s era of European integration, there are two distinct yet interrelated processes of differentiation affecting the European Union (EU) Council system.1 The first process is institutional, and its effects are observable in organizational dynamics or what can be called an “administrative phenomenon” (Leruth et al. 2019a: 1022). The second process involves the social dynamics of differentiation that includes new in-​g roup/​out-​g roup practices alongside a deeply conditioned habitus of how things are done that retains resilience. The net result of these differentiation dynamics is to cast a mature organizational culture into a more hybrid set of configurations with more status effects and “club goods” to generate more patterns of internal differentiation than at any time since the fateful “pillar” design of the Treaty on European Union (TEU) in the early 1990s. Where this is headed is no easy forecast, since some trends suggest “differentiated disintegration” and deconditioning acts that could recast a taken-​for-​g ranted sense of common enterprise as a more fractured or contingent sense of shared purpose.Yet, then again, there is a degree of buy-​in and engagement in the Council’s organizational culture that is akin to a consociational political system with a durable “system attitude” and “procedural code” even in the face of turbulence and serial crises (Lewis 2021). Differentiation post-​Maastricht has led to a more complex EU legislative process, though the ability for new areas of activity to produce “depoliticized modes of decision-​making” suggests ample creative agency in blurring the lines between high and low politics (Bickerton et al. 2015b: 313). Economic policy and budgetary with Reverse Qualified Majority Voting (RQMV) sanctioning, swaths of Justice and Home Affairs (JHA), including police and judicial cooperation, security, and even military procurement, all now fit this characterization. Such creative thinking will come in handy with future EU-​UK relations and the emergent forms of differentiation that may be necessary to find agreement on. Post-​Brexit differentiation will be many years in the making to establish new internal-​external constellations with variable levels of vertical and horizontal integration across policy sectors in UK-​EU relations.2 The patterns of opt-​out/​in and degrees of participation that one can image here makes today’s complex Schengen regime look like a new norm rather than a messy exception. Institutionally, the Council is a “most likely” case for the effects of differentiation, since it represents such a diverse group of member-​states and since territoriality and national interests are “baked” into the very organizational “cake” of the Council’s design. Organizationally, the Council affirms the observation by Leruth et al. 264

DOI: 10.4324/9780429054136-19

The Council of the European Union

(2019b: 1391) that there could be “many shades” of differentiation, including patterns of permanent self-​exclusion (Eriksen 2018) or quasi-​permanent opt-​out/​opt-​in dynamics as in the case of Denmark’s Maastricht carve-​outs. The remainder of the chapter is divided into four sections. First, the Council’s “procedural code” and “system attitude” is contextualized as a consociational-​like feature of the EU polity, essentially the institutional ecosystem in which differentiation dynamics unfold. Second, the chapter examines how differentiation affects the Council’s organizational dynamics and creates hybrid network configurations with more “club goods” and sub-​societies of informal and normative rulebooks, such as the contrasting cases found in financial and foreign affairs. This is followed by a section to address the social dynamics found in Council differentiation, illustrating how voting and consensus practices have evolved over time, including the “social power” of Council networks. A concluding section will consider the parametric conditions of the Council’s organizational culture, speculating on how institutional bricolage is a recurring feature, and what sorts of self-​reinforcing or self-​undermining patterns of behaviour and negotiation work to reproduce or undermine this over time.

The Council’s consociational “system attitude” Today’s differentiation fits a mould of being constantly negotiated within a deeply acculturated and path-​dependent decision-​making culture (see Leruth et al. 2019a; Cornut 2018). Johan Olsen’s characterization of the EU as a compound polity built on “imperfect processes” is apt, especially imagery of Council institutions as examples of “living institutions” best depicted as “social organisms that evolve over time as an unplanned result of historical processes” (Olsen 2010: 122). Combining the path-​ dependent thinking of historical institutionalism with [International Relations] IR practice theory is a fruitful analytic device to gauge the mix of stability and change in the Council’s organizational culture. Much of the Council’s institutional architecture and the practices that sustain it are forged and re-​forged during crises, a trend established as early as the 1960s (Ludlow 2006). And still, the organizational culture of the Council has an enduring quality that such change patterns do not seem to erode. If the Council system were equated to a “corporate culture” (Kreps’ 1990 term), then the process of consensus-​ seeking would be the most patently observable feature over time.There is now broad agreement that the Council rests on a distinct “culture of consensus” (Smeets 2015: 1) albeit one that varies internally by specialized configuration (Wallace and Reh 2015: 80) or as “sub-​societies” with distinct normative and behavioural expectations (Lewis 2019). The Council is not only a formal organization but also a social order with a “shared feeling of destiny” (Adler-​Nissen 2014: 175). “New Intergovernmentalism” theory recognizes this fundamental trait “post-​Maastricht” especially “how consensus building in informal settings…has become an end in itself ” (Bickerton et al. 2015a: 41). The roots of this culture are in the foundational era of European integration, originally what neofunctionalists observed as a “system attitude” or a “willingness to behave according to an accepted bargaining code keyed primarily to a determination to succeed” (Lindberg and Scheingold 1970: 242, 119). Weiler’s concept of EU “infranationalism” is helpful to understand a consensus-​seeking culture as an “underlying ethos” that is “managerial and technocratic” and dilutes a purely “national element” in the joint decision-​making process (Weiler 1999: 283, 272).3 The infranational dimension of the Council’s consensus culture is comparable in institutional durability and soft law functions as that in other “cooperative” styles of federalism. In describing EU infranationalism, Weiler mentions intriguing parallels to the inclusiveness of viewpoints that is hardwired into consociational polities or the accommodation style found in neo-​corporatist countries that avert a confrontational style of politics (Weiler 1999: 282–​285). 265

Jeffrey Lewis

Consider the comparison to a “spirit of accommodation” found in Lijphart’s (1968) seminal study of Dutch politics. Replace “Dutch” with “EU” and you find a description of Council-​style consensus-​seeking with at least family resemblance. In Lijphart’s (1968: 103–​104) words: Dutch politics is a politics of accommodation. That is the secret of its success. The term accommodation is here used in the sense of settlement of divisive issues and conflicts where only a minimal consensus exists. Pragmatic solutions are forged for all problems, even those with clear religious-​ideological overtones on which the opposing parties may appear irreconcilable, and which therefore may seem insoluble and likely to split the country apart. A key element of this conception is the lack of a comprehensive political consensus, but not the complete absence of consensus. Dutch national consensus is weak and narrow, but it does contain the crucial component of a widely shared attitude that the existing system ought to be maintained and not be allowed to disintegrate. The second key requirement is that the leaders of the self-​contained blocs must be particularly convinced of the desirability of preserving the system. And they must be willing and capable of bridging the gaps between the mutually isolated blocs and of resolving serious disputes in a largely nonconsensual context. From an organizational cultural perspective, the central concept here is that of a widely shared “system attitude,” namely that the system ought to be maintained, that the norms and principles of consensus-​seeking not erode into disuse, that the collective decision-​making process itself has value, and the obligation of finding results is a shared responsibility. A suggestive parallel to this foundational value is what Levitsky and Ziblatt (2018: 106–​111) identify as a democracy’s elemental norm for “institutional forbearance.” The ultima ratio resort to non-​consensual outcomes can likewise compare to the Council’s formal “shadow of the vote” under QMV or the July 2020 European Council’s conclusions on rule of law conditionality and future EU spending. The key aspect is the system attitude that the legislative process should succeed, that important national positions are not simply bigfooted or outvoted, that everyone finds themselves with instructions at times that are isolating or hopelessly unpersuasive, and so forth. The formal decision rule helps to hardwire this system attitude since the “shadow of the vote” compels a logic of explaining and justifying positions, listening and responding to each claimants demands, and providing group deliberated collective legitimation of outcomes.The policy segmentation of the Council’s work, into discreet networks of like-​minded policy specialists, also helps in moments of crisis to create an institutional “loose coupling” that “deter the failure of one component from reverberating across the entire system” (Ansell and Trondal’s terms 2017: 293). This helps explain how the contested mandatory asylum quota vote in 2015 did not leech into broader QMV practices in the Council as well as the containment of the mistrust generated by the frictions of the eurozone crisis within circles of finance and banking policymakers. The system attitude and norm of institutional forbearance are structural, cognitive safeguards against the kind of “constitutional hardball” and “norm erosion” patterns that scholars remind us can undermine a political order if the zone of acceptable political behaviour shifts far enough.4 Recent theory debates over how durable the European constitutional settlement is or if a stable endpoint is reachable typically underappreciate the degree to which the Council system and the policy networks that comprise it provide an organizational culture with a lot of “guardrail” stabilizing functions. Returning to this theme in the concluding section, it will be argued that the Council’s organizational culture has a characteristic resiliency that is consistent with Greif and Laitain’s (2004) notion of a “quasi-​parameter” which makes durable dispositions of “how we do things around here” less a calculative choice than a shared understanding of life inside of the Brussels bubble. 266

The Council of the European Union

The conditioning qualities for this system attitude receive reinforcement by everyday practices and the density of interactions in a club-​like environment of continuous negotiation. As Mai’a Cross (2007: 233) has described, “negotiations are essentially never-​ending.” But over the longer term, such parametric conditions can shift from being self-​reinforcing to self-​undermining, and the concluding section will offer some tentative examples of what the latter’s deconditioning acts might consist of.

Organizational dynamics of differentiation: Council hybridization The instruments of differentiation have led to more hybrid Council configurations. [Economic and Monetary Union] EMU and the eurozone display this most starkly. At the top of the food chain, the European Council evolved a distinct Euro Summit that first met in 2008 and took on a heightened significance in the context of the euro crisis. Since 2013, the Euro Summit meetings are governed by their own procedural rules that specify, for example, the need for statements to be adopted by consensus. There is also the possibility (thus far not exercised) for the heads of government to appoint a Euro Summit President that is different from the European Council President, presumably if the latter is from a non-​euro member-​state. The precedent of keeping Donald Tusk in a double-​hatted presidency role despite Poland’s euro outsider status is uncertain in future hypotheticals, since it may signify a new form of “club good” that the eurozone heads of government exercise in managing monetary union. If the heads of government are reluctant to create formal insider/​outsider roles in summitry networks, the economic and finance ministers have shown less compunction to do so. Since 2000 or 2001, the euro area economic and finance ministers have created out-​g roup status differences for the non-​euro members in a variety of institutional innovations. Holding Eurogroup meetings the day before Ecofin is now normal. The Eurogroup, itself an informal (unofficial) formation of the Council of the EU, presides over an autonomous preparatory network centred on the EWG (Eurogroup Working Group), which itself has an EWG of “alternates” acting in preparatory capacity as well as ad hoc working groups that report directly to them. One scholar describes this as “a veritable ‘exclusion zone’ with a parallel system of meeting forums that only include euro-​insiders” (Adler-​Nissen 2014: 94). This Council sub-​society has additional organizational infrastructure that networks with the [Economic and Finance Committee] EFC and the Commission administered EFC Secretariat. Some scholars, such as Majone, wonder how the common framework of the Treaties is impacted by EMU as an “autonomous policy domain” that is increasingly detached from other EU policy areas, including the SEM (Majone 2014: 1222). Others, such as Köllicker (2001), might stress the potential centripetal effects outsiderness to this policy domain might be subjected to in the long term. Echoing this view, Adler-​Nissen observes how economic policy coordination has intensified and deepened into a “relatively autonomous community” and has led to the “gradual forming of a common mindset” (2014: 82–​83). Another illustrative pattern of hybridity and organizational change is documented by Uwe Puetter’s research on the European Council’s new roles in tasking euro area coordination and surveillance reforms to the network of senior policy experts known as the “Van Rompuy task force” and assistance of the “Sherpa group” (most with backgrounds in the EFC) which has its own group president who reports directly to the European Council President (Puetter 2014: 179). In terms of organizational differentiation, the Council’s foreign affairs networks have transmogrified more than perhaps any other policy area. In the early decades, the foreign ministers who met as the General Affairs Council (GAC) were considered to be primus inter pares among the ministers’ networks. A key status marker was the unique participation right for foreign ministers to sit alongside their head of government during European Council summits (A 267

Jeffrey Lewis

practice discontinued since the 2002 Seville summit). As the EU’s agenda expanded, and more specialized sectoral Councils formed (such as the Internal Market Council in the 1980s), the foreign ministers gradually lost the horizontal and interministerial coordination capacities of earlier years (Gomez and Peterson 2001). A hallmark of foreign affairs networks that spans time is the mutual trust-​based consensus-​seeking practices and shunning contested public votes. The “general rule” is that the foreign ministers do not compose a “voting Council”; rather, “a constant background theme” to their work is “the search for consensus” (Westlake and Galloway 2006: 55). The reinforcing habits stem from an “implicit brief for tackling politically sensitive matters” (ibid.: 50). Given the early centrality of foreign ministries in overall EU-​level coordination, including the “communication infrastructure” linking Brussels and the capitals (Kassim and Peters 2001: 304), the foreign affairs networks were important seedbeds for consensus-​ seeking practices to disseminate within other organizational spaces in the Council and the national-​level administrations (Lewis 2015). In another precedent-​setting practice, the foreign ministers regularized a pattern of working in informal meetings, beginning with the so-​called Gymnich formula in 1974.5 Designed to accommodate the delicate sovereignty-​sensitivities over foreign policy cooperation under the fledgling European Political Cooperation (EPC) process, each rotating presidency organized a kind of retreat outside the public scrutiny of Brussels so that the foreign ministers could engage in policy coordination without a formal agenda or expectations of result. Well before there was a formal [Common Foreign Security Policy] CFSP, the practices of foreign policy coordination within EPC led to an “automatic reflex of consultation” that became durably reproduced and internalized by members (Nuttall 1992: 312). In Michael Smith’s account, “the EU foreign policy system…began to develop its own culture of cooperation involving standards of behavior, shared understandings and a common language” (Smith 2004b: 105). Of all the Council’s networked structures, the foreign policy realm is where some of the most creative intra-​organizational institutional engineering has taken place and the consensus-​driven practices show a pattern of “disaggregated sovereignty” that celebrates insulation and informality to avoid politicization. For example, since 2003, CFSP has significantly re-​engineered Common Security and Defense Policy (CSDP) in a process of “experiential learning” that rests on several decades of working to create a “common frame of reference among EU member states regarding the practice and purpose of European foreign policy” (Smith 2015: 112). A newer enhanced cooperation aspect with significant internal-​external differentiation potential is [Permanent Structured Cooperation] PESCO, or permanent structured cooperation in the fields of military and defence procurement. The first PESCO format of the [Foreign Affairs Council] FAC took place in March 2018 among the EU defence ministers. All 27 defence ministers were present at the meeting, yet only the PESCO members were involved in adopting legal acts, which included the initial list of 17 joint PESCO projects. PESCO now includes 25 of the EU27 as members (not Denmark or Malta) and has nearly 50 joint military projects in the works. Significantly, PESCO experiments with external differentiation by allowing non-​EU members to participate in defence projects on a case-​by-​case basis if that state “shares the values on which the EU is founded.”6 The administration of PESCO is handled by the [European External Action Service] EEAS and [High Representative] HR with involvement from the European Defence Agency (EDA) and EU Military Staff (EUMS) in coordinating and assessing projects. In preparing EU foreign policy, the Council has a dedicated network overseen by the Political and Security Committee (PSC), a group that is Brussels-​based and at the rank of ambassador. Although first established on an interim basis in 2000 as part of the evolving CSDP, the PSC ambassadors quickly seeded consensus practices in their everyday work, based on a “strong collective desire to achieve results” (Howorth 2011: 112). While the new PSC formation came to 268

The Council of the European Union

replace the older EPC-​era Political Committee (PoCo) of capital-​based Political Directors, what is striking is how the new Brussels-​based PSC format (meeting face-​to-​face on average two or more days per week) quickly routinized consensus practices that reflect a much older Council organizational culture. The general hybridization trends of permanent chairs –​from the FAC (at ministerial level) to the PSC, EFC, or Military Committee (at the preparatory level) –​signify the development of more executive-​like leadership (rather than rotational equality). Permanent chairs enhance a Council network’s institutional memory and can be viewed by others as a “more impartial mediator” (Juncos and Pomorska 2011: 1100). In policy areas where national opt-​outs apply, such as EMU, the status of eligibility for Chair –​say the Eurogroup or EFC or EWG –​is a new kind of “club good” that outsiders are disqualified from. Rather than the “collective good” provided by a rotating six-​month Council presidency which created a sense of parity within a group of asymmetrical powers and a shared sense of stakeholder responsibility to deliver a visible output of results, the internal selection of permanent chairs is an excludable “club good” for those on the inside and a different kind of status marker (Buchanan 1965; Majone 2014; Köllicker 2001). Less visible to the general public, we see the variable influence of the European Council’s “Sherpa” network organized and administered by the President’s chief of staff as chair of a virtual Council network that has autonomous links to the heads of government and reinforces patterns of “executive federalism.” On top of hybridization patterns, the overall density of institutionalized relations in the Council system is higher than ever before. One useful metric of this is the concept of “Brusselization,” or a gradual shift in authoritative decision-​making from capitals to Brussels (Juncos and Pomorska 2011).This institutional density is especially visible in post-​Lisbon foreign policy networks and the EEAS machinery (Leuffen et al. 2013: 195–​196). The EU permreps are hubs for “Brusselization” as more and more diplomats are posted there (rather than shuttling in/​ out from the capitals) and the scaled-​up number of permanent foreign affairs working groups under HR appointed chairs is a qualitative trend of institutionalization that can lead to or “spark” new social dynamics, including socialization to group norms and procedural methods (Juncos and Pomorska 2011: 1100). The correlation between Brusselization trends and the sheer complexity of internal-​external differentiation patterns is now hard to ignore; causally, we are witnessing an era of European integration with rapidly evolving and innovative changes in everyday practices. The next section will examine the social dynamics of differentiation further, including the value of practice theory in revealing in-​g roup/​out-​g roup boundary maintenance and how the “meaning context” of the Council’s procedural code can change over time.

Social dynamics of differentiation: the Council’s evolving practices At the practice level, the social dynamics of differentiation are a key element of today’s era of European integration. The Council system is most accurately viewed as a series of nested “sub-​ societies” rather than a monolithic politico-​administrative system, and each network of Council activity is embedded institutionally in a particular social field of cooperation (Lewis 2019). The advantage of seeing the Council system as a set of sub-​societies at the ministerial, preparatory, and expert levels is how it can reveal organizational culture as the contingent result of situated agents who are acting on beliefs and values that are drawn from “inherited traditions” (Bevir 2013). Compare the sense of joint enterprise in a preparatory network like the EFC or Coreper with the distributional if not zero-​sum politics of the [Special Committee on Agriculture] SCA for a long-​term contrast in social fields. New research on the PSC and the development of CSDP similarly shows that this policy area is “embedded in a particular social field of cooperation” 269

Jeffrey Lewis

(Svendsen and Adler-​Nissen 2019: 1422; Cross 2011; Howorth 2011). Another contrast can be made within the Council sub-​society for finance networks: often geared to informal deliberation and consensus-​seeking when in policy coordination mode (Puetter 2014), the budget and multiannual budget framework (MFF) negotiations generate more overt vote-​counting dispositions where the shadow of the vote is foreshortened as soon as hypothetical blocking minorities are eliminated (Lewis 2019: 152; Nicoll and Bryan-​Kinns 2006). In principle, the same national officials may exhibit different or even countervailing practices that are engendered by different social fields of Council negotiation. As “inherited traditions,” these practices reflect a social and institutional context that acculturates individual participants and new members to the club, but over time they may be modified or adapted to new circumstances. As in-​g roup, out-​group participation dynamics become more complex, the status-​enhancing or diminishing effects are likely to amplify. Denmark is a leading example of how outsiderness can be managed through constructive participation and compensation techniques at mutual responsiveness (Adler-​ Nissen 2011, 2014). Denmark’s opt-​ outs may make for an awkward partner in terms of uniformly applying the legal order of the acquis, yet as a member-​state they have unimpeachable credentials for playing by the informal procedural code. One way Denmark does this is running the Council presidency rotation with a club-​minded attitude.7 Denmark seems to quietly win a good share of “club rewards” in the form of top Council leadership jobs such as Niels Ersbøll’s long tenure as Council Secretary General (1980–​1994), or more recently, Jeppe Tranholm-​Mikkelsen (since 2015). Hungary may be at the other end of the status perk spectrum, especially in the current dysfunctions related to migration and asylum policy and the “Brussels’ can’t make us” signals that come from the Orbán government. Measuring “status” directly is inherently difficult, as invisible pecking order and social hierarchies are not matched by formal rules, size, or votes. Despite several decades of Council research, formal modelling approaches have not reached agreement on how voting power translates into actual influence.8 In general, formal modelling overlooks the “social power” of networks (Hafner-​Burton and Montgomery 2009). Practice theory can help redress the lacunae by studying what constitutes a “competent performance” (Adler and Pouliot 2011) in terms of informal and normative institutional contexts and the ways that “social capital” is literally steeped throughout everyday negotiations. Building off Bourdieu’s notion of “political capital,” Adler-​Nissen has developed an innovative concept to understand this social power of networks and how we measure influence, what she labels “diplomatic capital.” Rather than having fixed properties like voting rules, “diplomatic capital” is a composite form of influence and its meaning is constantly (re)negotiated and highly variable across different social contexts (Adler-​Nissen 2014: 160–​161). Diplomatic capital is a form of social currency in Council networks and it represents a dimension of invisible status hierarchies that run within and across the Council’s venues. Daniel Naurin’s (2007) “network capital index” shows this by using survey research to measure insider perceptions of which member-​states are considered more influential and trustworthy. From a sample of over 330 respondents, the results show considerable nuance in the “rankings” that do not easily match up with size, formal voting power, or length of membership. In status ranking, small may be beautiful. Diplomatic capital is captured in another expert survey that ranks the Netherlands high on the scale of “transaction power” for a “strong reputation as a like-​minded country that is responsive and easy to work with” (Janning 2016: 3). Likewise, a recent Politico composite ranking (“power matrix”) of national leaders and EU ambassadors with the most influence places the Netherlands, Denmark, and Sweden as highly as Germany, France, and Britain (Heath 2016). Analysing Britain’s slow decline in EU influence and increased marginalization since 2010, Rasmussen (2016: 717) notes “even large and significant countries can have their rank in the pecking order reduced if they 270

The Council of the European Union

breach the codes of conduct.” The salience of diplomatic capital is heightened in EU differentiation since the potential group deliverables of social influence and opprobrium are more markedly determined by patterns of participation and (self) exclusion. As Ian Johnston explains, “the rewards and punishments are social because only groups can provide them, and only groups whose approval an actor values will have this influence” (Johnston 2008: 25). Recent voting clusters and coalition politics affirm this kind of in-​g roup/​out-​g roup differentiation as well. Studies show that “coalition geometry” varies by social field with several different “centers” of influence across foreign, security, economic, and social policy contexts (Janning and Zunneberg 2017). In this view, informal coalitions and enhanced cooperation “consensus clusters” (ibid.’s term) in different issue areas will be a key to managing in-​g roup/​out-​g roup differences at the practice level. Post-​Brexit EU–​UK relations could face a steep differentiated integration learning curve. One could speculate the role-​playing re-​socialization dynamics may be more acute for UK negotiators who will face status diminution effects as “third country” representatives. The externally differentiated UK options for post-​Brexit opt-​in to EU policies are considerable, yet the UK’s “network capital” (Naurin and Lindahl 2010) and “diplomatic capital” (Adler-​Nissen 2011) will need to be recast. For almost any future relationship scenario, there is an “unequal and asymmetrical future partnership between the UK and EU” and “the UK will no longer be a peer in EU decision-​making, but a demandeur; no longer an insider but an outsider” (Murray and Brianson 2019: 1436). Anticipating the future “plus” negotiations or renegotiations for “bolt-​on” agreements in areas such as financial services or security cooperation, Simon Hix recognizes this will not feel like “taking back control” for the British, and possibly, “The EU will feel even more like a bully once the UK is outside the EU; just ask Switzerland and Norway” (Hix 2018: 23). Even before the 2016 referendum, qualitative evidence shows a decline in Britain’s pecking order status as they became more at odds with the Council’s “club rules” and others perceived aligning with the United Kingdom as “going against the community spirit, breaching community norms and alienating other countries” (Rasmussen 2016: 719). On the other hand, even where the United Kingdom was obstructionist, as in the vetoed Fiscal Compact, UK negotiators were valued to gain strategic information and knowledge about the positions of others, as the British were considered as “a special cadre of diplomats who are top of the range and of a special calibre” (ibid.: 716). As outsiders, a future compensatory mechanism for the UK’s brand of stigma management will be the deep knowledge and diplomatic skills they have earned a reputation for. To use the language of discursive institutionalism, while the “meaning context” of UK–​EU relations is undergoing constitutive change, the “background ideational abilities” of UK agents will give them an edge in finding their new place and status markers of standing in such new networked relationships (Schmidt 2008: 314). Brexit is also a social practices incubator for new institutional bricolage inside the Council’s sub-​structures. The Article 50 format of the Council (where the EU27 met without the United Kingdom in the room) was perhaps a social practice laboratory for developing “excludable network goods” such as joint security and defence cooperation in Europol or the EU-​sponsored Galileo satellite system. The path-​dependent trajectory of Article 50 practices has fascinating “competent performance” implications for the formal sanctioning rulebooks of internal EU politics. It would be ironic if Brexit did not cause other remaining members to emulate non-​ cooperation, withdrawal, or nonparticipation but instead routinized practices for group stigma imposition in the form of shunning, shaming, or financially penalizing that Article 50 formats might instil among like-​minded club insiders developing new “code of conduct” procedures (Article 7 sanctioning or the rule of law conditionality in the new MFF comes to mind). An unexpected Brexit contagion could be a more avuncular excludable network approach to 271

Jeffrey Lewis

dealing with recalcitrant members or behaviour inconsistent with EU values in internal differentiation administration. “Post-​Brexit” could come to evoke new forms of insider status, stigma management, and patterns of social influence rather than viable exit options or cherry-​picking forms of à la carte Europe. The high levels of mutual trust and sense of common purpose that are found throughout Council networks are designed to create club-​like environments of like-​minded officials. The emphasis in “new intergovernmentalism” on deliberative processes even in the most sovereignty-​ sensitive areas such as foreign policy, budgetary surveillance, and JHA presupposes an interaction context that cultivates such social trust and diplomatic capital. Likewise, new tentative initiatives that could extend QMV in principle to CFSP decisions presuppose a dynamic density of interactions in groups such as the PSC ambassadors. Citing one PSC official, Michalski and Danielson (2020: 336) refer to a “diplomat’s code of conduct” that unifies representatives “because around the table we deal only with people like us.” The “club-​like” composition of Council networks at all levels of political seniority and policy blends of specialist expertise have strong positive feedback effects on networked reputations and trustworthiness. Elinor Ostrom’s research convincingly shows how trust, reputation, and reciprocity are tightly coupled and positively reinforcing in all sorts of generic collective action situations (Ostrom 2007).When you add the institutional memory and context of Council repeat, interactive networks of policy making as well as the ritualistic aspect of what Bourdieusian scholars call the sens practique –​it is not this Wednesday’s Coreper meeting but the 2712th meeting of Coreper –​the Council’s brand of diplomatic capital under the habitus of the “Brussels bubble” comes into focus as an inseparable aspect of the qualitative institutional environment and culture of decision-​making. One way to assess this qualitative environment is to examine how the Council’s “meaning context” for QMV practices and consensus-​seeking habits evolves over time.9 There is a growing appreciation for how the era between the 1965 Empty Chair Crisis and the completion of the 1992 SEM Project involved a procedurally sophisticated evolution of majority voting practices (Novak 2011; Dinan 2015: 209). Indeed, the rising comfort level of the “shadow of the vote” from this period makes today’s constitutive principle for the “ordinary legislative procedure” (OLP) possible. A real-​time case involves the tentative discourse shifts in “European sovereignty” language and the principle of QMV in foreign policy decisions.The “meaning context” of negotiation under the formal shadow of the vote or veto is surprisingly understudied in EU Council research, though recent Political Sociology approaches and the IR practice turn has given this unspoken “how we do things around here” dimension a more visible standing (Adler-​Nissen 2016). Even insider accounts of the vaunted European Council summitry stress the intangible role of “collective improvisation” in resolving “events-​politics” crises (van Middelaar 2019). As Michael Mann demonstrates in his magnum opus on “social power,” the very nature of social relations can structure motivations and interests as well as “have emergent properties of their own” (Mann 1986: 5; Cornut 2018).10 The consensus-​seeking frame of the finance ministries’ networks looks heavily influenced by practices that depoliticize the annual Broad Economic Policy Guidelines (BEPG) process, including fiscal policy coordination and “peer review” surveillance process under the revised Stability and Growth Pact (SGP). The upgrades to the excessive deficit procedure anticipate a quicker recourse to the use financial sanctions, especially through an application of RQMV to Commission recommendations. Under the new RQMV rules, rejecting Commission sanction recommendations now requires a qualified majority of “noes” to do so (rather than a blocking minority; and abstentions count as votes in favour, not against) (Van Aken and Artige 2013: 153). Given the “shadow” of excessive deficit sanctions at least as a formal recourse implies that informal processes to explain and justify economic policies under a peer-​review process will 272

The Council of the European Union

also undergo evolution. In organizational culture terms, this will elevate social influence controls regarding compliance and could change the group willingness to sanction errant behaviour. For the euro area ministers, this represents a newer subculture of collective legitimation standards in what is clearly a very politically sensitive area of budgetary and fiscal sovereignty. Adler-​Nissen (2014: 85) explains how “processes of surveillance, benchmarking, best practice and peer pressure have facilitated cognitive convergence around a paradigm of sound economic policy and soft coordination.”To date however, the economic surveillance process relies on a peer review mechanism where “member states hardly ever criticize each other’s policies” (Kreilinger 2016: 1; Hodson 2011: 82–​83). In this light, the newer addition of an independent European Fiscal Board to advise the Commission on budgetary surveillance might enhance shared expectations for the peer review process.11 QMV practices display constant yet gradual change, consistent with the concept of “institutional bricolage.” Bricolage is a change process where actors can “reinterpret the meaning of rules and redeploy them under significantly altered circumstances” (Carstensen 2017: 140). This can entail novel interpretations of existing rules or redeploying them to solve new problems (ibid.: 143). During the “relaunch” era of European integration the 1980s and the legislative agenda of the “1992 Project,” Council QMV practices underwent significant repurposing. Perhaps little noticed at the time, but votes infrequently began from the late 1970s on, clustering around certain issues where formal QMV was an option, including the budget, [Common Agricultural Policy] CAP implementation, and technical trade issues (Hayes-​Renshaw and Wallace 2006: 268). As Council voting practices were refashioned, formal recourse to QMV gradually became a more effective deterrent to those with uncompromising positions. During the 1980s, “the vote was used in an abrupt way in the early years, without any real diplomatic niceties, like a ‘procedural guillotine,’ to use the words of one representative” (Novak 2011: 15). But over time, QMV practices evolved to become more “sophisticated” especially through the leadership capacities of the presidency to build compromise and allow those in the minority to “join the majority without losing face” (Novak 2011: 15). Through improvization, Council practices gradually repurposed the “shadow of the vote” in bricolage fashion over time. Bricolage techniques will be more important as EU differentiation becomes more internally and externally complex. We already see this in how voting standards now have a “normal” QMV rulebook as well as the new scope for RQMV in creating sanctioning and stigmatizing standards for aberrant behaviour (is there a cognitive linkage between excessive deficit penalties and rule of law infractions?). Because they exist as a social order, Council consensus and voting practices are never fixed or formally guaranteed, but contingent and evolving. The informal nature of consensus-​seeking practices means they are always subject to challenge and evolve through the reinterpretation of everyday acts. Sometimes consensus-​seeking fails, but a legitimate formal vote is politically unenforceable. A dramatic, publically charged example of this is the 2015 vote over mandatory asylum quotas in the JHA Council to relocate 120,000 refugees across the Union.12 In this instance, the Czech Republic, Hungary, Romania, and Slovakia voted against the measure and Finland abstained. The 14 September JHA Council failed to break the impasse over mandatory quotas as did a further week of intensive diplomacy by the Coreper II ambassadors who met right up until the re-​convened JHA Council on 22 September when a public vote was held. The timing of the decision to hold a vote was driven by pressure from the European Council which was scheduled to meet for an emergency summit on the migration crisis the following day and wanted the relocation issue settled beforehand. This example of contestation, beyond confirming that consensus-​seeking is no magic cure-​all for principled differences, shows how collective meanings can change or be made even in the bricolage practices of an outvoted 273

Jeffrey Lewis

minority contesting the legitimacy of outcome.The laggardly implementation of refugee quotas since this vote confirms the observation that Council actors try “not to push colleagues to the brink” when “isolation stemmed from a very real and very intractable domestic inhibition” (Hayes-​Renshaw and Wallace 2006: 268). In effect, holding a public vote over such a contested issue resulted in a legally but not politically binding outcome. The latest bricolage experiment in Council practices involves the rule of law conditionality that is increasingly tied not just to social influence, but financial club goods under the MFF. The five-​day July 2020 European Council, one of the longest summits on record, ultimately produced a blockbuster package deal with a compromise on the rule of law linkage that has multiple and ambiguous interpretations. At the end of 2019, the Finnish EU presidency, backed by a 2018 Commission proposal, was advancing a budget reform process where rule of law deficiencies would trigger a suspension of EU funds unless an RQMV stopped it. This “conditionality regime” is at least partially the product of frustration with the formal Treaty sanctioning under Article 7 violations that require unanimity and face pre-​emptive veto threats by would-​ be transgressors like Poland and Hungary. Of course, Poland and Hungary vehemently object to the new conditionality regime linkages to the budget, and in the run-​up to the July 2020 summit, European Council President Michel changed the sanctioning criteria from RQMV to QMV, in effect making a blocking minority easier to obtain (but not the unanimity rule that Poland and Hungary favour). The scope of rule of law conditionality was also narrowed under this proposed revision. The Italian delegation proposed a “reinforced qualified majority” procedure where a contested QMV outcome would be referred to the “next European Council” although (like the infamous Luxembourg Compromise does not specific what should happen if members do not all agree).13 After several more days of deadlocked discussions, a final agreement was reached that comes quite close to the Italian proposal. The European Council conclusions hold “the Commission will propose measures in cases of breaches for adoption by the Council by qualified majority” but in cases of a contested vote, “the European Council will revert rapidly to the matter.”14 In the immediate aftermath, Hungary and Poland demanded new proposals on conditionality at several Coreper II meetings of the EU ambassadors to clarify matters. They both insisted that only the European Council could ultimately decide on financial sanctions for rule of law violations, but the other delegations insisted this was not the only possible interpretation of the compromise. Charles Michel confirmed the ambiguity of the agreement: “It is very clear when you read the text, you have the words conditionality, rule of law, governance. And we vote by qualified majority” but in addition, “there is not yet complete clarity about the procedure.”15 The rule of law conditionality is a new potential “club good” mechanism that can produce internally differentiated forms of participation as well as material and immaterial status effects.The linkage, however, momentarily ambiguous it really is, between financial payouts from the EU budget to rule of law standards shifts the Council’s “coordinative discourse” (Schmidt’s term) of what appropriate and legitimate member-​states behaviour consists of. At the current moment, this may have a relatively narrow applicability as a form of conditionality (unlike the “fundamental values” basis for the big stick of Article 7 sanctioning), but like the long trajectory of QMV practices, the meaning context can and likely will evolve in bricolage fashion over time.The bricolage nature of the Council’s organizational culture illustrates the ongoing creative tension between an institution designed to represent territoriality (“national interests”) and the inherited traditions and club rewards that these social networks evolve over time. Seen in this light, the Council is a “most likely” case for differentiation effects since it operates at the epicentre of managing what New Intergovernmentalism calls the post-​Maastricht EU’s “persistent state of disequilibrium” (Hodson and Puetter 2019: 1157).

274

The Council of the European Union

Conclusions The Council’s social fabric contains within it a durable system attitude that practice theory connects to the outsize durability of consensus-​seeking as part of a habitus of cooperation and sens practique of what works (and what does not) inside the Brussels bubble (Adler-​Nissen 2016). Differentiation in today’s European integration process is becoming a more important mechanism in dispersing social influence and creating “club goods” within the EU27 and between the EU and non-​member participants (including the post-​Brexit UK). Whether differentiation ultimately creates “multiple Europes” or “classes” of membership is an open question, but paradoxically, it may generate significant reinforcing tendencies in the EU habitus. The good and appropriate behaviours of membership –​“normality” –​is reinforced and restored by patterns of differentiation and “the opt-​outers contribute to securing the doxa” (Adler-​Nissen 2014: 176–​ 177). The “new intergovernmentalism” theory’s emphasis on European integration’s need for “permanent consensus generation” (Bickerton et al. 2015a) presupposes such a habitus and doxa of taken-​for-​g ranted sensibilities about how things work in Brussels but is largely silent on the genealogy of where it comes from and what sustains it. The procedural sophistication of QMV practices, discussed above, offers an important illustration for why this genealogy matters. Treating the Council system as a “corporate culture” helps reveal both the timeworn “inherited traditions” and the constant bricolage adaptability that is possible in such an organizational ecosystem. While not yet fully resolved, the new rule of law conditionality linkage to the budget seems to confirm Lijphart’s “system attitude” is still at work –​a high stakes example of finding pragmatic solutions to apparently irreconcilable differences within a shared belief the system should be maintained and not be allowed to deprecate into “constitutional hardball.”16 The evidence of path dependence in the Council’s evolution and the reinforcing processes for consensus-​seeking practices are consistent with Greif and Laitan’s (2004) concept of a “quasi-​parameter” that again spotlights an intangible system attitude that Council politics inside the “Brussels bubble” hardwires into the doxa of integration in ways that today’s era of crises do not seem to diminish. But it needs mention that the EU’s consociational-​like system attitude and doxa are far from immutable. A deeply internalized style of cooperative negotiation should not literally be treated as taken for granted, since norm erosion, contestation, and delegitimation are always possible. Equating the Council’s consensus culture to a “quasi-​parameter” keeps one alert to the possibility that today’s self-​enforcing behaviours could gradually change or become less applicable in more situations, and institutional environments within the Council ecosystem might develop new reinforcing or undermining dynamics, including the “intentional selection of alternative behaviors” (Greif and Laitan 2004: 639). Today’s tangled complexities of differentiation –​both external and internal forms –​create new insider-​outsider status effects and make the invisible hierarchies of power and social influence more difficult to map out (Leuffen et al. 2013). The implication of different “classes” of membership is that a “system attitude” may become thinner or patchier among different Council networks, or even encourage some members to become more self-​isolated or marginalized rather than make the extra efforts and psychological compensatory behaviour reported by Adler-​Nissen in the opt-​out case of Denmark and, until recently, the United Kingdom (Adler-​Nissen 2014; Rasmussen 2016). And who knows at this point what “competent performances” could emerge from the traumas of Brexit? The Article 50 (EU27) meeting format is perhaps unknowingly a bricolage incubator for future areas of enhanced cooperation such as PESCO as well as sharpening the edges of group sanctioning skills that could apply to future Article 7 proceedings concerning “fundamental values.” As a

275

Jeffrey Lewis

mature organizational culture, what is most distinctive overall is how the Council combines a durable habitus with ever-​changing institutional hybridity. The Council system, as a meta-​network of distinct sub-​societies of transnational policy making is an advanced “community of practice” with a “system attitude” to make collective decisions according to club-​like rules and shared normative standards. In a world where the tumults of advanced democratic capitalism can remove the guardrails and erode the social norms of democratic stability, would it not be ironic if transnational EU governance helped “rescue” nation-​state legitimacy (in Europe) yet again? If so, the organizational culture of the Council system and such quasi-​parameters as the norm of institutional forbearance contributes a small but important element to that ongoing experiment beyond the nation-​state.

Notes 1 This chapter treats the European Council and Council of the EU (or “Council of Ministers”) as part of a singular “Council system” of decision-​making. While the European Council is a free-​standing EU institution post-​Lisbon, it still relies on Council networks for preparatory and follow-​up work and is still serviced by the General Secretariat of the Council (GSC). For a more developed conceptual treatment of the Council system as a “meta-​network” see Lewis (2019). 2 As Leuffen et al. (2013: 12–​18) explains, internal differentiation is where EU rules do not apply uniformly to all member-​states and external differentiation is the extension of EU rules to non-​member-​states. Vertical integration is the degree of centralization or depth of policy making, and horizontal integration refers to the territorial spread of a policy jurisdiction. See Leuffen et al. (2013), ­chapter 1 for a detailed mapping of the patterns and permutations. 3 For more on this theme, see Lewis (2015), Heisenberg (2007). 4 Stephen Levitsky and Daniel Ziblatt, “Is Our Democracy Wobbly?” New York Times, 28 January 2018. 5 Gymnich meetings get their name from the original 1974 session held at the Schloss Gymnich in Erftstadt outside of Bonn. 6 Based on a Finnish Council presidency proposal to the Coreper II ambassadors in 2019, this externally differentiated opt-​in concept is designed to keep close military ties with countries such as the United Kingdom (or more controversially for members such as France, the United States) but also rebuffs potential participants such as China or Turkey (Politico, 5 November 2019). 7 A classic illustration is Prime Minister Rasmussen’s declaration at the start of the 2002 presidency rotation that Danish domestic politics would effectively be put on hold for six months so Eastern enlargement negotiations could be completed. He is quoted as saying, “About 80 percent of time will be spent on EU matters…we consider it a national interest that we make it an effective presidency” (Financial Times, 28 June 2002). A quirkier recent example is Prime Minister Mette Frederiksen rescheduling her planned wedding around the July 2020 blockbuster European Council summit to decide on a new multiannual budget package (MFF) (Politico, 25 June 2020). 8 See Heisenberg (2008) for an excellent review of this research and the indeterminacy of findings. 9 For more on how a “meaning context” follows a deliberative and interactive discursive process, otherwise known as “discursive institutionalism,” see Schmidt (2008). For an analysis of the legacy status of the Luxembourg Compromise and how it’s “meaning context” has changed, see Lewis (2019, 2020). 10 Another suggestive and complementary line of research is Padgett and Powell’s (2012) generative concept of “autocatalysis” in social networks. 11 A suggestive parallel subculture was the initial EPC practice in foreign affairs of creating norms for “off limit” topics called domaines réservés, which were viewed by members as extremely sensitive, and hence, not open for group discussion. According to Smith, as EPC evolved and “its ambitions grew” the norm became gradually contested and the domaines réservés contracted (Smith 2004a: 123). Bickerton (2012: 35) seems to draw a similar comparison where he notes, “there are striking parallels between the way the EU’s foreign and security policy governance has evolved and that of the Eurozone.” Such learning effects may also shape the Council’s emerging rule of law conditionality (discussed further below). 12 Council Decision (EU) 2015/​1523 of 14 September 2015 establishing provisional measures in the area of international protection for the benefit of Italy and Greece, OJ L 239, 15.9.2015, pp. 146–​156. 13 Politico, Brussels Briefing, 20 July 2020. 276

The Council of the European Union

14 Conclusions of the Special Meeting of the European Council, 17, 18, 19, 20, and 21 July 2020, point 23. 15 Politico, Brussels Briefing, 27 July 2020. 16 Levitsky and Ziblatt attribute the concept of “constitutional hardball” to legal scholar Mark Tushnet and describe it this way: “playing by the rules but pushing against the bounds and ‘playing for keeps.’ It is a form of institutional combat aimed at permanently defeating one’s partisan rivals –​and not caring whether the democratic game continues” (Levitsky and Ziblatt 2018: 109).

References Adler, E. and Pouliot,V. (2011) ‘International Practices: Introduction and Framework.’ Cambridge Studies in International Relations, 119(1), pp. 3–​35. doi:10.1017/​CBO9780511862373. Adler-​Nissen, R. (2011) ‘Opting Out of an Ever-​Closer Union:The Integration Doxa and the Management of Sovereignty.’ West European Politics, 34(5), pp. 1092–​1113. doi:10.1080/​01402382.2011.591102. Adler-​Nissen, R. (2014) Opting Out of the European Union: Diplomacy, Sovereignty and European Integration. Cambridge: Cambridge University Press. Adler-​Nissen, R. (2016) ‘Towards a Practice Turn in EU Studies: The Everyday of European Integration.’ Journal of Common Market Studies, 54(1), pp. 87–​103. doi:10.1111/​jcms.12329. Ansell, C. and Trondal, J. (2017) ‘Coping with Turbulence.’ In Ansell, C., Trondal, J., and Ogard, M. (eds) Governance in Turbulent Times. Oxford: Oxford University Press pp. 285–​302. Bevir, M. (2013) A Theory of Governance. Berkeley, CA: University of California Press. Bickerton, C. (2012) European Integration: From Nation-​State to Member State. Oxford: Oxford University Press. Bickerton, C., Hodson, C., and Puetter, U. (2015a) ‘The New Intergovernmentalism and the Study of European Integration.’ In Bickerton, C., Hodson, C., and Puetter, U. (eds) The New Intergovernmentalism: States and Supranational Actors in the Post-​Maastricht Era. Oxford: Oxford University Press pp. 1–​48. Bickerton, C., Hodson, C., and Puetter, U. (2015b) ‘Conclusions: The Post-​Maastricht Period and Beyond.’ In Bickerton, C., Hodson, C., and Puetter, U. (eds) The New Intergovernmentalism: States and Supranational Actors in the Post-​Maastricht Era. Oxford: Oxford University Press pp. 304–​328. Buchanan, J.M. (1965) ‘An Economic Theory of Clubs.’ Economica, 32(125), pp. 1–​14. Available at: www. jstor.org/​stable/​pdf/​2552442.pdf (Accessed: 27 April 2021). Carstensen, M. (2017) ‘Institutional Bricolage in Times of Crisis.’ European Political Science Review, 9(1), pp. 139–​160. doi:10.1017/​S1755773915000338. Cornut, J. (2018) ‘Diplomacy, Agency, and the Logic of Improvisation and Virtuosity in Practice.’ European Journal of International Relations, 24(3), pp. 712–​736. doi/​pdf/​10.1177/​1354066117725156. Cross, M.K. (2007) ‘A European Epistemic Community of Diplomats.’ In Sharp, P. and Wiseman, G. (eds) The Diplomatic Corps as an Institution of International Society. Basingstoke: Palgrave Macmillan pp. 223–​245. Cross, M.K. (2011) Security Integration in Europe. Ann Arbor, MI: University of Michigan Press. Dinan, D. (2015) ‘The Political System of the European Union.’ In Magone, J. (ed) Routledge Handbook of European Politics. London: Routledge pp. 202–​218. Eriksen, E.O. (2018) ‘Political Differentiation and the Problem of Dominance: Segmentation and Hegemony.’ European Journal of Political Research, 57(4), pp. 989–​1008. doi/​pdfdirect/​10.1111/​1475-​6765.12263. Gomez, R. and Peterson, J. (2001) ‘The EU’s Impossibly Busy Foreign Ministers: ‘No One is in Control.’’ European Foreign Affairs Review, 6, pp. 53–​74. Available at: https://​kluw​erla​wonl​ine.com/​jou​r nal​arti​cle/​ Europ​ean+​Fore​ign+​Affa​irs+​Rev​iew/​6.1/​354​650 (Accessed: 27 April 2021). Greif, A. and Laitin, D. (2004) ‘A Theory of Endogenous Institutional Change.’ American Political Science Review, 98(4), pp. 633–​652. doi=​10.1.1.294.5684. Hafner-​Burton, E. and Montgomery, A. (2009) ‘Globalization and the Social Power Politics of International Economic Networks.’ In Kahler, M. (ed) Networked Politics: Agency, Power, and Governance. Ithaca, NY: Cornell University Press pp. 23–​42. Hayes-​Renshaw, F. and Wallace, H. (2006) The Council of Ministers. 2nd Edition. New York, NY: Palgrave Macmillan. Heath, R. (2016) ‘Power Matrix: Ranking Europe’s Leaders and Ambassdors.’ Politico, 28 April. Available at: www.politico.eu/​interactive/​power-​matrix-​charting-​the-​eu-​players-​by-​country-​european-​council-​ national-​capitals-​leaders-​ambassadors/​ (Accessed: 27 April 2021). Heisenberg, D. (2007) ‘Informal Decision-​Making in the Council: The Secret of the EU’s Success?’ In Meunier, S. and McNamara, K. (eds) The State of the European Union: Making History: European Integration and Institutional Change at Fifty. Oxford: Oxford University Press pp. 67–​87. 277

Jeffrey Lewis

Heisenberg, D. (2008) ‘How Should We Best Study the Council of Ministers.’ In Naurin, D. and Wallace, H. (eds) Unveiling the Council of the European Union: Games Governments Play in Brussels. Basingstoke: Palgrave Macmillan pp. 261–​276. Hix, S. (2018) ‘Brexit: Where Is the EU-​UK Relationship Heading?’ Journal of Common Market Studies, 56, pp. 11–​27. Available at: http://​epri​nts.lse.ac.uk/​89976/​1/​Hix​_​EU%E2%80%93UK-​relati​onsh​ip_​A​ccep​ ted.pdf (Accessed: 27 April 2021). Hodson, D. (2011) Governing the Euro in Good Times and Bad. Oxford: Oxford University Press. Hodson, D. and Puetter, U. (2019) ‘The European Union in Disequilibrium: New Intergovernmentalism, Post Functionalism, and Integration Theory in the Post-​Maastricht Period.’ Journal of European Public Policy, 26(8), pp. 1153–​1171. doi:10.1080/​13501763.2019.1569712. Howorth, J. (2011) ‘The Political and Security Committee: A Case Study in “Supranational Intergovernmentalism”.’ In Dehousse, R. (ed) The ‘Community Method’: Obstinate or Obsolete? New York, NY: Palgrave Macmillan pp. 91–​117. Janning, J. (2016) ‘The Netherlands: The EU’s Hidden Coalition Champion.’ European Council on Foreign Relations, 16 June. Available at: https://​ecfr.eu/​arti​cle/​commentary_​the_​netherlands_​the_​eus_​hid​den_​ coal​itio​n_​ch​ampi​on_​7​044/​ (Accessed: 27 April 2021). Janning, J. and Zunneberg, C. (2017) ‘The Invisible Web: From Interaction to Coalition-​Building in the European Union. European Council on Foreign Relations, Policy Brief, May 2017. Available at: https://​ ecfr.eu/​publ​icat​ion/​the_​invisible_​web_​from_​interaction_​to_​coalit​ion_​buil​ding​_​in_​the_​eu_​7​289/​ (Accessed: 27 April 2021). Johnston, A.I. (2008) Social States: China in International Institutions, 1980–​2000. Princeton, NJ: Princeton University Press. Juncos, A. and Pomorska, K. (2011) ‘Invisible and Unaccountable? National Representatives and Council Officials in EU Foreign Policy.’ Journal of European Public Policy, 18(8), pp. 1096–​1114. doi:10.1080/​ 13501763.2011.615197. Kassim, H. and Peters, G. (2001) ‘Conclusion: Co-​Ordinating National Action in Brussels: A Comparative Perspective.’ In Kassim, H., Menon, A., Peters, G., and Wright,V. (eds) The National Co-​Ordination of EU Policy:The European Level. Oxford: Oxford University Press pp. 297–​342. Köllicker, A. (2001) ‘Bringing Together or Driving Apart the Union? Towards a Theory of Differentiated Integration.’ West European Politics, 24(4), pp. 125–​151. doi:10.1080/​01402380108425468. Kreilinger,V. (2016) ‘Economic Policy Coordination in the EU: Linking National Ownership and Surveillance.’ Notre Europe, Jacques Delors Institute, 11 July. Available at: https://​ins​titu​tdel​ors.eu/​wp-​cont​ent/​uplo​ads/​ 2018/​01/​eco​nomi​cpol​icyc​oord​inat​ion-​kre​ilin​ger-​jdib-​jul​y16.pdf (Accessed: 28 April 2021). Kreps, D. (1990) ‘Corporate Culture and Economic Theory.’ In Alt, J. and Shepsle, K. (eds) Perspectives on Positive Political Economy. Cambridge: Cambridge University Press pp. 90–​143. Leruth, B., Gänzle, S., and Trondal, J. (2019a) ‘Exploring Differentiated Disintegration in a Post-​Brexit European Union.’ Journal of Common Market Studies, 57(5), pp. 1013–​1030. doi:10.1111/​jcms.12869. Leruth, B., Gänzle, S., and Trondal, J. (2019b) ‘Differentiated Integration and Disintegration in the EU after Brexit: Risks versus Opportunities.’ Journal of Common Market Studies, 57(6), pp. 1383–​1394. doi:10.1111/​jcms.12957. Leuffen, D., Rittberger, B., and Schimmelfennig, F. (2013) Differentiated Integration: Explaining Variation in the European Union. Basingstoke: Palgrave MacMillan. Levitsky, S. and Ziblatt, D. (2018) How Democracies Die. New York, NY: Crown. Lewis, J. (2015) ‘The Council of the European Union and the European Council.’ In Magone, J. (ed) Routledge Handbook of European Politics. Abingdon and New York, NY: Routledge pp. 219–​234. Lewis, J. (2019) ‘EU Council Networks and the “Tradition” of Consensus.’ In Bevir, M. and Phillips, R. (eds) Decentring European Governance. Abingdon and New York, NY: Routledge pp. 142–​170. Lewis, J. (2021) ‘The Council.’ In Riddervold, M., Trondal, J., and Newsome, A. (eds) The Palgrave Handbook of EU Crises. Basingstoke: Palgrave Macmillan pp. 239–​57. Lijphart, A. (1968) The Politics of Accommodation: Pluralism and Democracy in the Netherlands. Berkeley, CA: University of California Press. Lindberg, L. and Scheingold, S. (1970) Europe’s Would-​Be Polity. Englewood Cliffs, NJ: Prentice-​Hall. Ludlow, N.P. (2006) The European Community and the Crises of the 1960s: Negotiating the Gaullist Challenge. London and New York, NY: Routledge. Majone, G. (2014) ‘From Regulatory State to a Democratic Default.’ Journal of Common Market Studies, 52(6), pp. 1216–​1223. doi:10.1111/​jcms.12190.

278

The Council of the European Union

Mann, M. (1986) The Sources of Social Power. Volume 1, a History of Power from the Beginning to A.D. 1760. Cambridge: Cambridge University Press. Michalski, A. and Danielson, A. (2020) ‘Overcoming Dissent: Socialization in the EU’s Political and Security Committee in a Context of Crises.’ Journal of Common Market Studies, 58(2), pp. 328–​344. doi:10.1111/​ jcms.12945. Murray, P. and Brianson, A. (2019) ‘Rethinking Britain’s Role in a Differentiated Europe after Brexit: A Comparative Regionalism Perspective.’ Journal of Common Market Studies, 57(6), pp. 1431–​1442. doi:10.1111/​jcms.12961. Naurin, D. (2007) ‘Network Capital and Cooperation Patterns in the Working Groups of the Council of the EU.’ EUI Working Paper, RSCAS 2007/​14. Available at: https://​cad​mus.eui.eu/​bitstr​eam/​han​dle/​1814/​ 6762/​RSCAS_​2007​_​14.pdf?seque​nce=​3&isAllo​wed=​y (Accessed: 29 April 2021). Naurin, D. and Lindahl, R. (2010) ‘Out in the Cold? Flexible Integration and the Political Status of Euro Opt-​Outs.’ European Union Politics, 11(4), pp. 485–​509. doi:10.1177/​1465116510382463. Nicoll, W. and Bryan-​Kinns, M. (2006) ‘The Budget Procedure.’ In Westlake, M. and Galloway, D. (eds) The Council of the European Union. 3rd Edition. London: John Harper pp. 102–​112. Novak, S. (2011) ‘Qualified Majority Voting from the Single European Act to Present Day: An Unexpected Performance.’ Notre Europe, November.Available at: https://​link.sprin​ger.com/​chap​ter/​10.1007%2F978-​ 3-​030-​51791-​5_​13. Nuttall, S. (1992) European Political Cooperation. Oxford: Oxford University Press. Olsen, J.P. (2010) Governing through Institution Building. Oxford: Oxford University Press. Ostrom, E. (2007) ‘Collective Action Theory.’ In Boix, C. and Stokes, S. (eds) The Oxford Handbook of Comparative Politics. Oxford: Oxford University Press pp. 186–​207. Padgett, J. and Powell, W. (2012) ‘The Problem of Emergence.’ In Padgett, J. and Powell, W. (eds) The Emergence of Organizations and Markets. Princeton, NJ: Princeton University Press pp. 1–​29. Puetter, U. (2014) The European Council and the Council: New Intergovernmentalism and Change. Oxford: Oxford University Press. Rasmussen, M.K. (2016) ‘“Heavy Fog in the Channel. Continent Cut Off?” British Diplomatic Relations in Brussels after 2010.’ Journal of Common Market Studies, 54(3), pp. 709–​724. doi:10.1111/​ jcms.12319.Schmidt, V. (2008) ‘Discursive Institutionalism: The Explanatory Power of Ideas and Discourse.’ Annual Review of Political Science, 11, pp. 303–​326. doi:11.060606.135342. Smeets, S. (2015) Negotiations in the EU Council of Ministers: ‘And All Must Have Prizes’. Colchester: ECPR Press. Smith, M. (2004a) Europe’s Foreign and Security Policy: The Institutionalization of Cooperation. Cambridge: Cambridge University Press. Smith, M. (2004b) ‘Institutionalization, Policy Adaptation and European Foreign Policy Cooperation.’ European Journal of International Relations, 10(1), pp. 95–​136. doi:10.1177/​1354066104040570. Smith, M. (2015) ‘The New Intergovernmentalism and Experiential Learning in the Common Security and Defence Policy.’ In Bickerton, C., Hodson, C., and Puetter, U. (eds) The New Intergovernmentalism: States and Supranational Actors in the Post-​ Maastricht Era. Oxford: Oxford University Press pp. 111–​128. Svendsen, Ø. and Adler-​Nissen, R. (2019) ‘Differentiated (Dis)integration in Practice: The Diplomacy of Brexit and the Low Politics of High Politics.’ Journal of Common Market Studies, 57(6), pp. 1419–​1430. doi:10.1111/​jcms.12960. Van Aken, W. and Artige, L. (2013) ‘Reverse Majority Voting in Comparative Perspective: Implications for Fiscal Governance in the EU.’ In de Witte, B., Héritier, A., and Trechsel, A. (eds) The Euro Crisis and the State of European Democracy. Fiesole: European University Institute pp. 129–​161. Van Middelaar, L. (2019) Alarums and Excursions: Improvising Politics on the European Stage. Newcastle upon Tyne: Agenda Publishing. Wallace, H. and Reh, C. (2015) ‘An Institutional Anatomy and Five Policy Modes.’ In Wallace, H., Pollack, M., and Young, A. (eds) Policymaking in the European Union. 7th Edition. Oxford: Oxford University Press pp. 72–​112. Weiler, J.H.H. (1999) The Constitution of Europe: “Do the New Clothes Have an Emperor?” And Other Essays on European Integration. Cambridge: Cambridge University Press. Westlake, M. and Galloway, D. (2006) The Council of the European Union. 3rd Edition. London: John Harper.

279

Part 3

Policy-​based differentiation Stefan Gänzle, Benjamin Leruth, and Jarle Trondal

Whereas Part 2 has explored European Union (EU) institutions as base and reservoir for action capacity in context with differentiation, this part focuses on the policies that the EU ultimately produces. Just as the EU institutions have come to display growing variation over time in order to meet heterogenous demands, so do policies differentiate as well.Yet, the handling of differentiation in EU policies may ultimately amount to a rather intricate task. This part includes one of the most ‘communitarized’ policies, the EU’s Common Fisheries Policy (CFP), other chapters of this part will feature policies, which are more intergovernmental in their design, such as the Common Foreign and Security Policy; it also includes policies, such as climate policy, regulating global common goods. The example of Permanent Structured Cooperation (PESCO), the EU’s cooperation in defence, shows that differentiation not only has also become a core feature in policy areas that have been characterized for a long time by intergovernmental design but also brought supranationalism and intergovernmentalism closer together. Differentiation ultimately also makes the binary of supranationalism and intergovernmentalism increasingly obsolete. Naturally, the implementation of EU policies has always proceeded by the way of discretion and differentiation –​even before the launch of the Single European Market (SEM). The SEM, as succinctly demonstrated in Chapter 24, has traditionally been the policy field which was highly characterized by differentiation. In legal terms, the ‘vague’ formulation of directives, such as the new directive approach subscribed to since 1985 (Pelkmans 1987), has continuously opened up policy space for flexibility and differentiation. As differentiation has become a central modal aspect in the aftermath of Brexit, the European integration paradigm needs to be ‘complemented’ by the concept of disintegration –​as we have argued in the introduction to this volume (as well as in several contributions to Part I). Consequently, further research is needed to fully grasp this distinct type of differentiation in the EU –​and this part seeks to contribute to that objective. Although the EU has developed a wide range of instruments allowing to deal with differentiation at the policy level, the subsequent chapters illustrate that processes of differentiation and also disintegration pose significant challenges in specific policy sectors. It is perhaps not surprising that this applies in particular to those which had been most ‘common’, such as the CFP. In the run-​up to Brexit, the future of UK–​EU fisheries was amongst the most hotly discussed policy sectors –​despite a relatively low percentage of workforce employed in the policy area on both sides of the Channel. DOI: 10.4324/9780429054136-20

281

Stefan Gänzle, Benjamin Leruth, and Jarle Trondal

Chapter 18 Differentiated integration in EU energy market policy Recent EU integration in energy market policy, combined with significant heterogeneity across national energy sectors, begs the question of the extent to which and why there is differentiated integration of EU energy market policy. The chapter by Torbjørg Jevnaker assesses internal and external differentiated integration across EU energy market laws adopted during the years 2009–​2019 and examines whether member states have opted out of, and third countries have been induced to participate in EU rules in this policy area. Differentiated integration in the EU energy market policy is found where there is low interdependence and few externalities as well as in undeveloped energy systems. Differentiation is thus driven by demands from a relatively small yet coherent group of member states, which, due to low incidence of externalities, can be supplied by the group of member states that choose to proceed with integration. The same pattern is found in select third countries, although sovereignty-​sensitive concerns play a bigger role in accounting for whether they participate in (parts of) an EU energy market policy. Differentiated integration is low due to the high level of interdependence across countries, as well as due to a consensus orientation in negotiations on secondary EU legislation. The latter ensures that widespread resistance is accommodated via flexibilities (and by keeping overall integration at a lower level) rather than opt-​outs. Thus, not all heterogeneity translates into differentiated integration.

Chapter 19 Brexit and the Common Fisheries Policy: opportunities for multi-​level differentiated (dis)integration? The chapter by Arno van der Zwet, John Connolly, Christopher Huggins, and Craig McAngus examines differentiation from a multi-​level perspective, identifying and analysing contestation in (dis)integration goals between actors at different levels of governance. The differentiated integration literature has developed from focusing on the varied nature of integration between EU member states (internal differentiated integration) to considering the level of integration with non-​EU states (external differentiated integration). It is now compelled to also account the dynamics created by processes of disintegration. Within a multi-​level governance environment, differentiation opens up spaces for policy actors from networks at the sub-​state level. However, the scope of integration in these networks varies considerably from country to country as well as from (sub-​national) region to region. From a differentiated integration perspective, an analysis of the UK and Scottish governments’ divergent attitudes towards Europe offers an opportunity to examine narratives of disintegration as well as narratives that aim to establish collaboration in new or existing networks. The authors use the case of fisheries policy to highlight that there are divergent narratives that are shaped by domestic political environments but also by functional imperatives. Brexit has brought into focus the political importance of fisheries and the multi-​ level tensions that exist within the United Kingdom in terms of the future relationship with the EU. Overall, the chapter provides evidence of multi-​level tensions in relation to differentiated integration.

282

Policy-based differentiation

Chapter 20 The European Asylum Policy: core state powers, flexibility, and differentiated integration Constitutional differentiated integration and flexibility are two of the core characteristics of the Common European Asylum System (CEAS) as Foteini Asderaki and Eleftheria Markozani maintain. Their chapter finds the origins of these kinds of differentiation in the limitation of integration on the regulatory level and the politicization of the policy area. Constitutional arrangements of differentiated integration in CEAS include the opt-​ins, opt-​outs, or opt-​back-​ ins of the United Kingdom, Ireland, and Denmark from treaty provisions and asylum measures. Prior to Brexit, the United Kingdom had implemented a rather differentiated approach; choosing to opt out from most asylum measures of the 2013 CEAS revision but opting in to the Dublin Regulation and Eurodac rules. While Denmark has also secured a general opt-​out from AFSJ, the state opted in to the Dublin System. At the same time, the agreed structure of CEAS includes a significant level of flexibility regarding the implementation of EU rules. This flexibility has been the result of regulatory integration prevailing over capacity-​building and the limited impact of supranational institutions. Indeed, the implementation of EU rules are strictly delegated to national governments, resulting in the lack of harmonization and uniformity in the European Asylum Policy. The external shock of the 2015 refugee crisis has not altered this pattern yet; instead, it contributed to the increase of the sources of differentiation and flexibility, namely the politicization and the pre-​eminence of national over European solutions. As a result, both differentiated integration and flexibility have emerged as permanent features of EU Asylum Policy.

Chapter 21 Differentiated integration in EU climate policy EU climate policy is characterized by significant degrees of differentiated integration as Merethe Dotterud Leiren, Jørgen Wettestad, and Elin Lerum Boasson write. Although the two topics complement each other, differentiated integration and studies of EU climate policy have rarely been studied in conjunction. To address this gap, the authors develop a threefold conceptualization that makes it possible to explore differentiation during the course of a policy cycle: policy output differentiation, policy outcome differentiation, and policy impact differentiation. The examination of two key elements of EU climate policy legislation –​emissions trading and renewables policy –​shows that differentiated integration is an uneven and multi-​directional process that varies over time. The authors argue that differentiation is an important enabler for concerted action in EU climate policy and differentiation at the policy output and outcome stages may facilitate greater harmonization at the impact stage. Differentiation at the impact and outcome level of the three (ideal type) policy stages does not automatically reflect differentiation at the output policy stage; it is rather that output differentiation may facilitate both high and low levels of differentiation at these later stages.

283

Stefan Gänzle, Benjamin Leruth, and Jarle Trondal

Chapter 22 Differentiated integration in European external action Recent developments in European External Action and Foreign, Security and Defence have triggered ever more complex differentiation processes, layered on top of a set of already differentiated policies in Foreign, Security and Defence areas argues Thomas Henökl. This leads to an increasing diversification of policies directed at stronger intergovernmental coordination in a number of key government areas, namely external action, defence, migration and border management, development cooperation, climate change as well as internal and external security. It has been argued that these measures are necessary to address the simultaneous endogenous and exogenous pressures on the Union amounting to an existential political and legitimacy crisis for the EU, aggravated by the United Kingdom’s withdrawal from the EU. Taking a broad conception of EU External Action, including the above-​mentioned portfolios, this chapter dissects the cross-​sectoral involvement of institutions in the field.With an emphasis on the trend towards increased organizational specialization, it asks whether involving national and EU authorities bound together in the joint performance of these essential functions, amounts to the emergence of a ‘differentiated space of European external governance’ and how these developments resonate with the EU’s overarching strategic frameworks, the EU Global Strategy, and the European Neighbourhood Policy.

Chapter 23 PESCO: a formula for positive integration in European defence As in many other policy areas of the EU, differentiation, or what some have called the ‘negative starting point’ of integration, has always been the norm in EU defence policy find Steven Blockmans and Dylan Macchiarini Crosson. Driven by both endogenous and exogenous (f)actors, political leaders in the European Council are nevertheless mindful of the need for member states to cooperate in more structured ways to better protect their citizens against security threats. For this reason, a package of defensive measures has been developed with remarkable speed since 2016. PESCO is the most prominent innovation in this field. Given the high levels of politicization in defence, it is perhaps surprising that PESCO has produced the most inclusive expression of enhanced cooperation, even if it is the most flexible of the differentiated integration mechanisms provided by the Treaties. This is largely the result of a German push for inclusivity, which prevailed over a French desire for a higher level of ambition. Monitored by the EEAS and the European Defence Agency (EDA) and increasingly driven by the Commission’s Directorate General for Defence Industry and Space (DG DEFIS), which manages the European Defence Fund (EDF), PESCO is a formula that generates ‘positive integration’ by structuring the defence market in the EU. This chapter adopts an empirical approach and maps the varied clusters of member states lining up behind different types of projects. It observes a process of coagulation across the microcosm of PESCO, coupled with formal expressions of differentiated integration, both vertically and horizontally.

284

Policy-based differentiation

Chapter 24 The increasingly differentiated European single market? The chapter by David Howarth explores the range of differentiation in the EU Single Market and, specifically, the informal differentiation created through the varying transposition and implementation of EU Single Market law and EU Competition Policy, and legally entrenched differentiation. These varied forms of differentiation have in effect contributed to discrimination and undermined market integration. On the one hand, persistent soft, informal, and instrumental differentiation in the Single Market risks undermining the EU’s sense of coherence or purpose, fragmenting the Union’s legitimacy and authority. However, this differentiation can be understood as temporary. On the other hand, opt-​outs in the Treaty on EU, the use of potentially permanent derogation in at least two legislative packages, and the use of Enhanced Cooperation all involve legal developments that can be seen as undermining the Single Market and thus challenging the core constitutional commitment of EU membership. Increased differentiation is the result of deeper market integration and EU enlargement. However, increased differentiation through legislative and policy developments largely reflects German and French government preferences.

Chapter 25 Differentiation and social policy: a sustainable way forward? The notion of ‘European Social Model’, often linked to the creation of a common European social policy, has captivated scholars for decades argue Benjamin Leruth and Sven Schreurs. Although social policy remains a competence of member states, there have been some developments at the supranational level, especially since the global financial crisis and subsequent Great Recession of 2008–​2009. This is manifested in the discourse used by the European Commission and the successful ratification of four key directives setting EU-​wide minimum standards in the social and employment fields (such as the Work-​Life Balance Directive or the Transparent and Predictable Working Conditions Directive). The chapter investigates the potential for a differentiated European Social Model. It first offers a brief historical overview of attempts to transfer social policy competence to the European level, arguing that a window of opportunity was missed with the failure of harmonizing social policy in the early stages of European integration. It then discusses the current state of affairs, drawing on different European actors’ stances on establishing common European standards in the field, and emphasizes recent changes adopted by the European Commission to promote harmonization as a response to the Great Recession. Finally, this chapter discusses whether support for a supranational European social policy exists and whether the existence of a (vertically or horizontally) differentiated European Social Model would be sustainable.

Chapter 26 Differentiation in EU security and defence policy The chapter by Elie Perot and Stephan Klose advances three central claims about differentiation in the EU’s security and defence policy. First, this chapter argues that differentiation in EU 285

Stefan Gänzle, Benjamin Leruth, and Jarle Trondal

security and defence is best characterized as a phenomenon of differentiated cooperation, owing to the lack of true integration in this domain which remains a core prerogative of EU member states. Second, it suggests that the mechanisms of differentiated cooperation at work in the EU’s security and defence policy can be usefully divided into two broad categories: those that allow for negative differentiated cooperation and those that allow for positive differentiated cooperation –​i.e., opt-​outs vs. opt-​ins. Third, the authors argue that the emergence of these mechanisms at particular points in time has been driven by changes in the EU’s security environment, which have shaped member state perceptions of the roles NATO and the EU should play as frameworks for security and defence cooperation in Europe. Key findings This section discloses four major findings that tie all chapters together: first, all EU policies that have been presented here exhibit significant forms of variation in terms of both integration and also cooperation (see Chapter 269). Thus, differentiation has become the main feature of European ‘integration’ today, it is ‘here to stay’. Second, differentiation is a highly dynamic characteristic of the EU political order and, third, it is not exclusively constrained to EU members but does embrace partner states alike. Although it is too early to draw a definite conclusion, it seems, fourth, that disintegration –​such as in the case of ‘Brexit’ –​does not automatically lead to a radical cut or forms of ‘de-​Europeanization’; rather, it is that differentiated disintegration prevails as we have argued elsewhere (Gänzle et al. 2019). The latter key finding is put forth most compellingly by Arno van der Zwet, John Connolly, Christopher Huggins, and Craig McAngus who demonstrate that Brexit has not produced radical de-​Europeanization of fisheries, arguably one of the most Europeanized areas of UK public policy. Furthermore, they show that the institutional arrangements and specifics around the new agreement (e.g., access to waters and quota) maintain a strong degree of regulatory and institutional alignment and fisheries policy provides a case study of potential multi-​level differentiated integration (outside the scope of EU membership) –​and consequently differentiated disintegration. This may trigger important consequences for the United Kingdom as it is not unthinkable that Scotland aligns itself more to the EU in the future, whereas England remains apart –​provided British constitutional politics allows for it. David Howarth shows that EU Single Market law offers a variety of differentiation created through the varying transposition and implementation as well as legally entrenched differentiation. He finds that informal differentiation on both Single Market legislation and the implementation of EU Competition Policy did not increase significantly following the 2004 enlargement. Although differentiation certainly has its advantages for countries aspiring for it, its impact on Single Market legislation and policies needs to be carefully considered in terms of European integration more generally. Persistent soft, informal and instrumental differentiation risks undermining the EU’s sense of coherence or purpose, fragmenting the Union’s legitimacy and authority. Benjamin Leruth and Sven Schreurs demonstrate that while attempts to build a genuine Common European Social Policy have failed in the early stages of European integration, the Great Recession seems to have opened a window of opportunity for the European Commission to propose and implement new social policy directives in terms of setting minimum standards across the EU. However, these were met with some reluctance by some member states. They suggest that one way forward would be to use the enhanced cooperation mechanism to deepen integration in this field.The historical differences between member states (mirrored by the traditional ‘three worlds of welfare capitalism’) should serve as the basis for differentiated mechanisms

286

Policy-based differentiation

of cooperation in the field of social protection. Ultimately, they argue that the so-​called European Social Model is not a utopia, but much more needs to be done to reduce inequalities and disparities across member states. Similarly, Foteini Asderaki and Eleftheria Markozani pinpoint to European Asylum policy as an area where flexibility and constitutional differentiation have been prevalent for quite some time –​providing the United Kingdom and Denmark with constitutional differentiation since 1994. Even after the communitarization of the policy area, the limitation to regulatory integration was promoted to silence domestic politicization and sovereignty concerns by keeping national resources under the control of member states (see also Genschel and Jachtenfuchs 2014). The migration crisis of 2015/​16 eventually culminated in internal divisions and politicization over the redistribution consequences of the regulatory framework led to several attempts for redesigning the policy –​without success to date. In climate policy, as succinctly shown by Merethe Dotterud Leiren, Jørgen Wettestad, and Elin Lerum Boasson, differentiation appears as an uneven and multi-​directional process and indicates that differentiation is not a temporary, accidental or non-​systematic feature of EU policies but is essential and an enduring phenomenon (see Schimmelfennig et al. 2015). In energy policy, more broadly, Torbjørg Jevnaker shows that differentiated integration only prevails when there are low degrees of interdependence and few externalities (e.g., isolated, unconnected, small energy systems; notably, islands such as Malta and Cyprus) as well as in undeveloped energy systems (e.g., emergent systems).Yet, in cases of high levels of interdependence across countries as a consequence of growing physical and financial interconnection of energy systems, as well as due to a consensus orientation in negotiations on secondary EU legislation, uniformity of integration seems to prevail. Turning to the vast policy area of external relations,Thomas Henökl finds that differentiation has been a characteristic of European external action and its institutional design ever since.There are several dimensions of differentiation in the realm of EU external affairs: in legal, institutional, sectoral, geographic or geopolitical perspectives, producing various effects as for positive and negative differentiation and, consequently, the EU’s organizational architecture –​but also action capacity to perform effectively on the global stage. In the wake of crises affecting the European neighbourhood both in the East and the South, that is following the Arab Spring and the annexation of Crimea, the EU has sought to use its Global Strategy (European External Action Service 2016a,b) to better harmonize its differentiated institutional architecture at the interface of EU institutions, member, and partner states. Another new initiative, the PESCO in defence, PESCO, even constitutes a microcosm of formal and vertical differentiated integration as Steven Blockmans and Dylan Macchiarini Crosson maintain. In the almost 50 projects, placed below the primary law commitments to which 25 member states have signed up, membership and observer status in the projects are in flux. Furthermore, four years after its inception in 2017, four member states have come to a leading position, France, Italy, Germany and Spain (with Poland, Greece, the Netherlands and Belgium in a second tier). The factors defining the scope of these various collaborative clusters are primarily of systemic and economic nature, such as industrial cooperation, integrated supply chains, and the perception of NATO. Elie Perot and Stephan Klose argue that differentiation in the EU’s security and defence policy is best characterized as a phenomenon of differentiated cooperation, owing to the lack of true integration in this domain which remains a core prerogative of EU member states. Subsequently, the unveiled mechanisms of differentiated cooperation have presumably been driven by changes in the EU’s overall security environment. The authors ultimately remind us that even under the paradigm of differentiated integration (not to speak of disintegration), the integrative impulse may be absent to a significant degree, so that is more appropriate to speak of cooperation rather than integration.

287

Stefan Gänzle, Benjamin Leruth, and Jarle Trondal

References European External Action Service (2016a) Shared Vision, Common Action: A Stronger Union. A Global Strategy for the European Union’s Foreign and Security Policy. Available at: https://​eeas.eur​opa.eu/​ top_​stor​ies/​pdf/​eugs​_​rev​iew_​web.pdf (Accessed: 29 April 2021). European External Action Service (2016b) Roadmap for Implementation of the Global Strategy. Available at: http://​club.bru​xell​es2.eu/​wp-​cont​ent/​uplo​ads/​2016/​09/​feuil​lero​ute-​strat​egie​glob​ale@ue160​922. pdf (Accessed: 29 April 2021). Gänzle, S., Leruth, B. and Trondal, J. (eds.) (2019), Differentiated Integration and Disintegration in a Post-​Brexit Era, London: Routledge. Genschel, P. and Jachtenfuchs, M. (2014), Beyond the Regulatory Polity? The European Integration of Core State Powers, Oxford: Oxford University Press. Pelkmans, J. (1987), ‘The New Approach to Technical Harmonization and Standardization’, Journal of Common Market Studies 25 (3): 249–​269. Schimmelfennig, F., Leuffen, D. and Rittberger, B. (2015),‘The European Union as a System of Differentiated Integration: Interdependence, Politicization and Differentiation’, Journal of European Public Policy 22 (6): 764–​782.

288

18 Differentiated integration in EU Energy Market Policy Torbjørg Jevnaker

Introduction Energy was a central concern in the early steps of European integration, as illustrated by the European Coal and Steel Community of 1952 and the European Atomic Energy Community of 1957.1 Nevertheless, European Union (EU) integration of energy markets did not really begin until the 1990s. When the EU started adopting policies that sought to liberalize and integrate energy markets to improve economic efficiency, energy was a latecomer to the single market initiative. Although some cross-​border cooperation had existed previously, the overall situation was one in which there were separate national energy systems with high barriers to cross-​border trade. Energy markets were subject to diverse regulatory and technical arrangements across Europe, with different energy mixes stemming from differences in each country’s resource endowments (Eurostat 2020; Hancher and Hauteclocque 2010: 2; Herranz-​ Surrallés 2019: 2; Papsch 2021: 553). Over the course of the past decade, however, EU energy market policy has developed massively. Not only did the Lisbon Treaty codify the EU’s competence to adopt energy laws, but energy markets became subject to a growing amount of increasingly detailed EU rules, and an EU energy agency was set up. Moreover, the calls for an energy transition facilitated EU rules on several new domains of energy policy, as seen in the immense Clean Energy Package. While energy market policy originally focused on competition and trade, this was increasingly accompanied by attention towards reducing the energy sector’s greenhouse gas emissions which make a large share of overall EU emissions. EU climate objectives thus fuelled further integration(see Chapter 21). How was it possible to further integrate EU energy market policy despite the significant heterogeneity across member states? The picture makes it pertinent to look for differentiated integration. Differentiated integration has been conceptualized as a way of avoiding deadlock on the policy deepening and geographical widening of EU integration (Majone 2009: 218; Duttle et al. 2017: 407). By allowing member states to opt out, heterogeneity among member states can be accommodated (Duttle et al. 2017: 408). While early work on differentiated integration focused on the ‘constitutional’ level of the EU treaties, subsequent research expanded into differentiation in ‘everyday’ policies –​at the level of secondary EU legislation (Schimmelfennig and Winzen DOI: 10.4324/9780429054136-21

289

Torbjørg Jevnaker

2020). Such differentiation is not a new phenomenon but can rather be seen as a normal feature of the EU (Duttle et al. 2017: 422). This chapter asks to what extent and why is there differentiated integration of EU energy market policy? It provides an overview of EU energy market policy and assesses differentiated integration internally within the EU (across member states) as well as externally across select third countries. It examines the extent to which, and why member states have opted out from participating in EU rules in this policy area –​and the same for third countries that have been induced into participating. Subsequently, it discusses whether internal and external differentiated integration is driven by the same or different factors. Although several third countries cooperate with the EU in this policy area (Herranz-​Surrallés 2019; Hofmann et al. 2019), the chapter zooms in on the European Economic Area (EEA) countries because they cooperate with the EU within the same overarching institutional framework, which should make similar patterns across these third countries likely. Energy policy covers multiple sectors and affects various policy areas, as it concerns the production, transport, and use of energy (Herranz-​Surrallés 2019).The current chapter focuses on specific legislation adopted at the EU level to regulate energy systems and markets in Europe. This includes the successive energy market packages as well as infrastructure and market surveillance laws.With a focus on EU laws regulating the energy system, laws for promoting or ensuring the energy supply from specific energy resources are not covered here (but see Chapter 21). The delimitation to a narrow, but tightly related, set of rules enables a more focused comparison of the factors behind opt-​outs and opt-​ins across countries.

Conceptualizing and explaining differentiated integration What is differentiated integration? Definitions of differentiated integration point to variations in the level of integration, such as horizontally across countries or vertically across policy areas (see Chapter 1). The literature is mostly oriented towards horizontal differentiation, as suggested by the definition of differentiation as ‘territorially unequal formal validity of EU legal rules’ (Duttle et al. 2017: 410, italics in original; see also Schimmelfennig and Winzen 2020). Horizontal differentiation means that not all countries participate in EU policies, resulting in a level of EU integration that is differentiated across countries. Horizontally differentiated integration means that there is variation in the geographic extension of the application of EU laws, which can be assessed for a single policy area. Differentiation may entail full opt-​out from an entire policy area, or from parts thereof. Furthermore, internal horizontal differentiation means that not all EU laws apply in all member states (e.g., because individual member states have opted out). In contrast, external horizontal differentiation means that some EU laws apply to non-​member states, as third countries have been induced into participating in the EU policy (Holzinger and Tosun 2019; Schimmelfennig and Winzen 2020). Operationally, differentiation is ‘a provision that formally exempts at least one member state from applying a legal rule otherwise valid for all EU member states’ (Duttle et al. 2017: 410). Such opt-​outs may name specific member states or be formulated in more general terms. Opt-​ outs may be granted directly in the law or indirectly in the sense that an application and/​ or notification is required. Indirect opt-​outs are usually subject to regulatory approval at the national or EU level. Finally, opt-​outs may be temporary or permanent. This chapter measures opt-​outs for member states and third countries in terms of source, specificity, and time (Duttle et al. 2017: 410; Papsch 2021). Source refers to whether an opt-​out is directly provided within EU law (actual), or whether a follow-​up action, such as an application 290

EU Energy Market Policy

procedure, is required to be granted or make use of a derogation (potential). Specificity means that an opt-​out may be granted to a specific country referred to by name (specific), or it can be granted if it is available to everyone or those that qualify under general criteria (general). Time means that an opt-​out may be time-​limited (temporary) or not (permanent).

What explains differentiated integration? Differentiation in EU integration may broadly be explained in terms of supply and demand. Interdependence and heterogeneity among member states are demand-​side factors. Heterogeneity across member states can exist in terms of preferences, capacity and in their relative interdependence. While overall interdependence incentivizes cooperation and integration, heterogeneity triggers disagreement regarding integration. If integration-​sceptic member states cannot halt integration, they may seek opt-​outs. Examples of how heterogeneity creates demand for differentiation can be seen across richer and poorer member states: While richer member states seek more integration in regulatory standards, heterogeneity in level of economic development and wealth makes poorer member states interested in opting out.The same economic differences motivate poorer member states to push for more integration in EU funding and subsidies, from which richer member states may want to opt out. Other factors behind heterogeneity across member states in terms of preferences or capacity are country size (with big countries having lower autonomy gains from international cooperation than small ones), whether or not a policy area is considered ‘high politics’ (i.e., concerns regarding sovereignty-​sensitive issues of national security), and whether the national identity is more or less exclusive (Schimmelfennig and Winzen 2020: 26–​30). Supply-​side factors behind differentiated integration are critical mass within, as well as externalities across, the groups of insiders and outsiders (insiders being those seeking common policies, and outsiders being those preferring opt-​outs).Where integration steps are accompanied by opt-​ outs for some member states, a sizable insider group is expected to be necessary to make this compromise desirable. The outsider group also requires critical mass to get opt-​out provisions included. Moreover, weak externalities between insiders and outsiders are seen as necessary for sustaining differentiated integration, as strong negative or positive externalities would make it harder for insiders and outsiders, respectively, to accept differentiation (Schimmelfennig and Winzen 2020: 30–​33). When explaining differentiation at the level of secondary EU legislation, the institutional context needs to be taken into account. On the one hand, secondary legislation may be affected by primary legislation. Here, whether the treaty provides for uniform or differentiated integration in a policy area is relevant, as is the historical sequence in which the EU has gained competences in the treaty and consolidated its authority in that policy areas. While opt-​outs from policies related to the single market were more prevalent prior to the Maastricht Treaty, sovereignty-​motivated differentiation increased thereafter in step with the growing integration of core state powers. Thus, established paths or norms for uniform or differentiated integration at the level of primary legislation may filter down into secondary legislation (Schimmelfennig and Winzen 2020: 86–​91). On the other hand, secondary legislation is negotiated in a distinct setting. In contrast to the rare, high-​level negotiations for amending EU treaties, secondary legislation is negotiated in repeat, insulated arenas. This triggers norms of consensual decision-​making, where resistance is accommodated. Thus, Schimmelfennig and Winzen (2020) expect that the size of the integration-​sceptical group of member states will affect whether the adopted legislation is characterized by differentiated integration or not: resistance from many member states will result 291

Torbjørg Jevnaker

in flexibilities being introduced to the legislation, while resistance from a few member states will result in opt-​outs (Schimmelfennig and Winzen 2020: 84–​85). Widespread resistance leads to flexible, but uniform integration, while concentrated resistance results in differentiated integration. Moreover, as secondary legislation is negotiated in arenas populated by sectoral policy specialists, economic and capacity considerations are assumed to be more prevalent than identity or sovereignty concerns. Thus, poorer member states will seek opt-​outs from burdensome rules and richer member states will seek differentiation to uphold their own higher regulatory standards (Schimmelfennig and Winzen 2020: 86–​91). In sum, differentiated integration of the EU energy market policy is expected to be affected by the treaty-​level integration pattern for, and the historical sequencing of EU integration of, this policy area. Specifically, differentiated integration of the EU energy market policy will be the outcome of concentrated demand fuelled by economic and capacity concerns rather than identity or sovereignty concerns, combined with supply conditioned by low externalities. These expectations apply to internal as well as external differentiated integration (Schimmelfennig and Winzen 2020: 156).

Observing differentiation in the EU energy market policy This section first presents an overview of the EU energy market policy over time before turning to opt-​outs for member states and selective third-​country participation.

Development of the EU energy market policy While common European markets had already been established for many other products by the early 1990s, electricity and gas were mostly national monopolies.Vertically integrated companies (i.e., companies that are active in several parts of the value chain) controlled the production, transmission, and delivery of electricity to the end user. Moreover, a lack of cross-​border infrastructure and regulatory barriers entailed low levels of cross-​border trade. The EU has sought to integrate energy market policy by liberalizing the electricity and gas sectors within member states and facilitating cross-​border trade. The EU has adopted successive energy market packages, as well as specific regulations on market supervision and infrastructure development (see Table 18.1). A ‘package’ means that multiple laws are developed, negotiated, and adopted together (Skjærseth et al. 2016). These laws target multiple ‘gaps’ across member states, such as divergent regulatory and operational practices that constituted barriers to trade, inadequate regulatory supervision of cross-​border trade, and a lack of physical cross-​border infrastructure. The first and second energy market packages were mainly about opening electricity and gas sectors within member states for competition.The EU sought to avoid companies using income generated via the operation of their electricity or gas network –​a monopoly activity –​to subsidize their commercial activities of energy production and supply. The EU gradually developed stricter rules for unbundling network operation and commercial activities. Moreover, to ensure fair competition within a member state, the producers or suppliers who did not own the network for transporting electricity or gas should still be able to use the network to deliver energy to their customers (called third-​party access). Requirements for independent regulatory supervision also evolved over time (Eikeland 2011; Jevnaker, 2015). Finally, looser European cooperation among national network operators (the so-​called transmission system operator, TSO) and national regulatory authorities were initiated with the second package. In 2007, a strategic shift took place at a summit of prime ministers and presidents of EU member states, from which a European energy policy was advocated. This formed an important 292

EU Energy Market Policy Table 18.1  EU energy market laws Energy market law(s)

Description of content

Energy market packages: First energy market package (1996/​1998) Second energy market package (2003) Third energy market package (2009) Clean Energy Package (2018/​2019)

Rules to liberalize and integrate national energy markets. Over time, gradually higher requirements for separation between monopoly and commercial tasks (i.e., unbundling network operation from energy production/​supply); non-​discriminatory access to the network for third parties; market opening for all consumers; independent national regulatory authority; cross-​border cooperation among network operators, and among regulatory authorities, at European level; EU agency (ACER) with delimited conditional decision-​making powers (since 2009); regional coordination of network operation incl. regional coordination centres (RCC) (since 2019) Rules to prevent market abuse. Prohibition against abusive practices affecting wholesale energy markets (e.g., insider trade). European monitoring of wholesale energy markets by EU agency (ACER) via registration and collection of transaction data from market participants. National investigation and prosecution of market abuse. ACER may coordinate cross-​border investigation if requested by national authorities Rules to boost the refurbishment of existing energy infrastructure and the deployment of new energy infrastructure. Bottom-​up processes to identify high-​priority projects of common interest that would be subject to a more expedient national permitting procedure (i.e., one-​shop stop, shorter deadlines) and eligible for EU funds. National decision-​making on permits, EU decision-​making on application for EU funding

REMIT Regulation (2011)

TEN-​E Regulation (2013)

backdrop for the third energy market package adopted in 2009, which constituted a major change in the EU’s regulation of the electricity and gas markets. It entailed a clearer opening for competition in a common European energy market and a full separation between network operation and energy production/​supply as the main rule (Eikeland 2011). National regulatory authorities in energy were to be independent of national governments (including independent of national energy ministries). New centralised structures for cross-​border cooperation at the European level replaced previous voluntary and/​or horizontal network structures: an EU Energy Agency (ACER) was established together with European partner organizations for transmission network operators for electricity and gas, respectively (Jevnaker 2015). The purpose of this was to integrate the many national markets into a common European energy market to optimize the use of common energy resources and networks in Europe.This was to deliver overarching objectives of the EU’s energy market policy, namely competitive energy prices, security of supply, and sustainability. Until 2009, there had not been a separate treaty provision for adopting EU energy laws, yet several had been adopted via provisions on competition or environment. Art. 194 of the Lisbon Treaty introduced a treaty basis for energy, codifying the existing practice of energy as a shared competence between the EU and the member states. It permitted the EU energy policy, ‘in the spirit of solidarity’ to aim at (a) ensuring the functioning of the energy market; (b) ensuring the security of energy supply; (c) promoting energy efficiency, energy saving, and renewable energy; and finally (d) promoting the interconnection of energy networks (Art. 194(1), TFEU 2012). However, it also emphasized member states’ right to determine their energy mix based on ‘the 293

Torbjørg Jevnaker

conditions for exploiting its energy resources, its choice between different energy sources and the general structure of its energy supply’ (Art. 194(2), TFEU 2012). In the aftermath of the financial crisis, the absence of European rules to prevent the manipulation of energy markets gained attention. In step with growing cross-​border trade via integrated wholesale energy markets, and as financial transactions could be carried out via an energy exchange located in a different member state than that of the physical energy flows, it became clear that traded energy flows were not always within the regulatory remit of individual national regulatory authorities. Thus, in 2011, the EU adopted Regulation (EU) No 1227/​2011 on wholesale energy market integrity and transparency (the ‘REMIT’ regulation). It established rules prohibiting abusive practices that affect wholesale energy markets.While these were aligned with the rules applicable in financial markets, the specific characteristics of wholesale energy markets were also taken into account (Art. 1). Wholesale energy markets were to be monitored in order to detect, stop, and prevent trading based on inside information (inside trading) and market manipulation. This would be done by ACER in close collaboration with national regulatory authorities. ACER would collect, register, and analyse transaction data from market participants and share this with other relevant bodies. Investigation, prosecution, and rules on penalties remained in the hands of national authorities. However, ACER could alert the responsible national authorities and, if requested, could coordinate joint investigations. Despite the major changes in market regulation with the third energy market package, there was still little cross-​border infrastructure to trade across borders. This was seen as a barrier to market integration. New interconnectors were hampered by obstacles related to the granting of permits, regulatory issues, and financing. In 2013, the EU adopted Regulation (EU) No 347/​ 2013 on guidelines for trans-​European energy infrastructure (TEN-​E Regulation). This was to facilitate prioritized energy infrastructure projects that have a cross-​border dimension and contribute to amongst others market integration, sustainability, and security of supply (Art. 4). Priority projects, so-​called projects of common interest, would be subject to special rules and could get access to EU funding. Projects would first be shortlisted via regional groups, and these projects would then compete at the European level for attaining an elevated status as a project of common interest. Only member states and the Commission could make decisions within the regional groups, and no projects would be shortlisted without approval from the member state in which the project would be located. In order for a project to apply for EU funding, agreement among the affected member states (or their national regulatory authorities) on a cross-​border cost allocation was required –​i.e., how they would distribute the costs for the project and how such costs were to be recovered through tariffs. If the involved national authorities did not agree within a set deadline, ACER would decide on the cost allocation. Once applied for, the decision on EU funding would be made by an executive agency under the Commission. Although the TEN-​E regulation introduced rules for expedient national permitting processes for priority projects, decision-​making on permitting remained in the hands of the member states. In 2019, the EU adopted the Clean Energy Package (occasionally referred to as the fourth energy market package). Despite being an energy market package, it also contains EU laws for the restructuring of the energy sector in line with overarching climate and energy objectives. Such laws were previously part of the climate and energy package from 2009, and had been adopted separately from the third energy market package (Skjærseth et al. 2016). While the three energy market packages had mainly focused on introducing market-​based rules and competition while facilitating cross-​border trade, the Clean Energy Package also focused on energy transformation to mitigate climate change.This was combined with a retained 294

EU Energy Market Policy

emphasis on trade and efficient resource allocation; the package saw trade within and between countries as something that would continue to contribute to the more efficient use of the electricity system and reduce electricity prices. Efficiency, in turn, was conceptualized as contributing to reducing greenhouse gas emissions. The EU’s energy market policy was thus to help deliver the EU’s climate and energy goals for 2030.

Differentiation of the EU energy market policy This section presents the differentiation in the EU energy market laws adopted between 2009 and 2019. While derogations had been granted under previous electricity and gas market legislation (e.g., with regard to the accession process of new EU member states), these have since expired (Papsch 2021: 569–​570) and are therefore not presented here. Internal differentiation Starting with the laws in the energy market package, the Electricity Directive had a variety of opt-​outs and flexible provisions. The scope for derogating remained rather stable from the 2009 directive to its recast version of 2019 (see Table 18.2). Some member states could permanently opt out from unbundling the entity that operates their national transmission network (TSO), while opt-​outs from unbundling requirements were allowed where the entity that operates a local distribution network (distribution system operator, DSO) is small. The latter opt-​out is granted only as long as DSOs qualify as ‘small’, and in effect represents a sub-​national opt-​out.

Table 18.2  Opt-​outs in the electricity directive (directive 2009/​72/​EC –​recast as directive (EU) 2019/​944) Derogation (article in which derogation provided for)

Source

Specificity

Derogation from TSO unbundling (Art. 44(2) in 2009, Art. 66(3) in 2019)

Actual

Derogation from DSO unbundling (Art. 26(4) in 2009, Art. 35 (4) in 2019)

Actual

Specific (Cyprus, Permanent Luxembourg, Malta) General (DSOs Temporary (as long as with